Accelerator Basics, Rev. AF

COPYRIGHT 2005 VARIAN MEDICAL SYSTEMS
L FOR TRAINING PURPOSES ONLY 7

C-series Clinac Accelerator System Basics i
C-series Clinac
Accelerator System
Basics
Revision AF: September 2005
Copyright 2005 Varian Medical Systems
ii C-series Clinac Accelerator System Basics
Compiled and Edited by:
Bill Kirkness
Varian Medical Systems
Oncology Systems Customer Support
Education Department
WARNING!
Do not attempt to operate or repair the Clinac using the
descriptive material in this Course Manual. Refer only to
the speci fi c operati on and mai ntenance manual s sup-
plied with your equipment. Uninformed or careless oper-
ati on of the machi ne can expose the operator, pati ent
and other persons to hazards that can cause serious in-
jury or death.
Clinacs used for training purposes are specially config-
ured for that purpose. Speci fi cal l y, protecti ve screens
and covers have been removed from some areas where
dangerous vol tages may be present, and safety i nter-
l ocks may be di sabl ed. Al so, 3-phase pri mary power i s
present in the Power Distribution Chassis even when the
Cl i nac i s i n the Emergency Off state. Do not reach i nto
the Power Distribution Chassis without first turning off
the mai n ci rcui t breaker on the wal l , and do not reach
into any area of the Clinac until your instructor has told
you that it is safe to do so.
C-series Clinac Accelerator System Basics iii
Statement of Objectives
C-series Clinac
Maintenance Course
Thi s course i s provi ded for those personnel di rectl y respon-
si bl e for the mai ntenance of the Cl i nac. It wi l l cover Modul a-
t or and RF Syst em t heor y, beam gener at i on and cont r ol ,
motion control, dosimetry, position readouts, power distribu-
t i on, accessori es, comput er cont rol syst em, di agnost i cs,
interlocks and troubleshooting.
This material is for Training Purposes Only. For on-site oper-
at i on and/or mai nt enance procedures, ref er t o t he speci f i c
oper at i on and mai nt enance manual s suppl i ed wi t h your
equipment.
Statement of Policy
This material is the exclusive property of Varian Associates
and is loaned subject to return upon demand and with the
express condition that information contained herein not
generally known in the trade shall be treated confidentially
and shall not be reproduced, redistributed or used in any
manner, directly or indirectly, detrimental to Varians interests.
This material must be returned to Varian Medical Systems
immediately upon demand at any time.This material is
registered to:
Name: ________________________ Date: _________________
iv C-series Clinac Accelerator System Basics

Area
2
ONCOLOGY SYSTEMS
CUSTOMER SUPPORT
596 ALDER DRIVE
MILPITAS, CA 95035
(408) 321-9400
North
East
Area
1
West
Ed.
Dept.
Ed.
Dept.
Northern
Region
Service
Conf.
Room
Education
Department
Clinac
600C/D
Lab
Class-
room
6
Classroom
5
Clinac
2300C/D
Lab
E
l
e
c
t
r
i
c
a
l
B
r
e
a
k
e
r
s
Class-
room
7
B
o
o
k

R
o
o
m
Classroom 1
1a 1b
Cafeteria
Copy Room
Admin.
Offices
Storage
Areas
Classroom
2
Classroom
3
Conf.
Room
Open
8 am-
5 pm
Info.
Svcs
Copy
Room
Accounting
Support
Engineering
Logistics
Work
Shop
S.A.P.
Command
Center,
Help Desk,
Facilities
H.R.
Conf.
Room
Classroom
4
Clinac
23EX
Lab
Clinac
2100C
Lab
Shipping &
Receiving
Storage
Area
VAR S
Lab
i
Vision Lab
Main
Lobby
C-series Clinac Accelerator System Basics v
Revision History
Revision Date Description
A May 1996 Initial Version
B Feb 1997 General reformat and minor corrections.
Added Title Page and Table of Contents to each chapter.
C Aug 1997 Updated Chapter 1 from new Varian Safety Manual, including Lockout/Tagout
and Gantry Pin Procedures.
D Jul 1999 Added Klystron Theory section (formerly in High Energy Beam Delivery System
Manual) to Chapter 5.
E Jan 2000 Updated Title Page, added Table of Contents to this section, Tables of Illustrations
to all chapters with captioned illustrations.
F Jan 2002 General update. Increased font size for better legibility.
AA May 2002 Added abbreviation list. Converted from WordPerfect to Adobe FrameMaker result-
ing in some page and paragraph re-numbering. Added index.
AB Nov 2003 Minor corrections. Redrew illustrations with CorelDraw.
AC Aug 2004 General update for 2004.
AD Jan 2005 General update for 2005. Also reduced footer font size.
AE May 2005 Updated Chapter 1 from new Clinac Safety Manual. Removed Chapter 2, Controls
& Indicators and renumbered remaining chapters. Re-formatted and added
explanatory text to Chapter 2, Machine Physics.
AF September 2005 General update for consistency with Varian standards affecting notes, cautions
and warnings.
vi C-series Clinac Accelerator System Basics
Visual Cues This document uses the following visual cues to help you locate and iden-
tify information.
This symbol identifies comments about a specific task. Some
examples include notes, required data, messages, substitu-
tions and shortcuts.
CAUTION: Describes actions or conditions that can
result in minor or moderate injury or can result in
damage to equipment.
WARNING:Describes actions or conditions that can re-
sult in serious injury or death.
Italics: Used for emphasis, defining new terms, or book
titles.
Bold: Identifies menu commands, items you can select on
the screen, and buttons to press.
Abbreviations The following abbreviations are used throughout this manual:
Abbreviation Meaning
AMC Advanced Motion Controls, Inc.
BNC A standard coaxial cable connector developed by the Ber-
keley Nucleonics Corp.
CPU Central processing unit (today often used to mean
microprocessor)
CRT Cathode-Ray Tube Monitor
EEPROM Electrically Erasable Programmable Read-only Memory
EPROM Erasable Programmable Read-only Memory
EMI Electronic Measurements, Inc.
FPGA Field Programmable Gate Array
LED Light-emitting diode
MCU Microcontroller (a microprocessor especially designed for
control system applications)
MPU Microprocessor (unit)
PCB Printed Circuit Board (sometimes referred to as a card)
PSA Patient support assembly (Treatment couch)
PWM Pulse Width Modulation
SCR Silicon Controlled Rectifier or Thyristor
STD Pro-Log Corp. Control System Bus: Simple To Design
Table of Contents vii
Table of Contents
Chapter 1: Emergency and Safety ........................................1-1
1. Abstract........................................................................................... 1-5
2. Introduction..................................................................................... 1-7
2.1. Overview .................................................................................. 1-7
2.2. Operators and Treatment Personnel ......................................... 1-7
2.3. Maintenance and Service Personnel .......................................... 1-8
2.4. Customer Support .................................................................... 1-8
2.5. Related Publications................................................................. 1-9
3. Emergency Procedures..................................................................... 1-9
3.1. Terminating the Treatment Beam............................................. 1-9
3.2. Performing an Emergency-Off ................................................. 1-10
3.3. Using the Emergency Hand Pendant....................................... 1-10
3.4. Lowering the Treatment Couch in an Emergency .................... 1-11
3.5. X-Ray and Electron Beam Radiation....................................... 1-13
3.6. Induced Radiation in Accelerator Components........................ 1-14
3.7. Radio Frequency (RF) Radiation.............................................. 1-15
3.8. Electromagnetic Interference .................................................. 1-16
3.9. Sulfur Hexafluoride and Freon 12 Gases ................................ 1-16
3.10. Lead..................................................................................... 1-17
3.11. Beryllium............................................................................. 1-18
3.12. Dielectric Insulating Oil ........................................................ 1-18
3.13. Depleted Uranium (Low-Energy Clinacs) ............................... 1-19
3.14. Ozone and Oxides of Nitrogen (High-Energy Clinacs)............. 1-19
3.15. Implosion ............................................................................. 1-19
3.16. Electric Shock ...................................................................... 1-20
3.17. Service Precautions .............................................................. 1-20
3.18. Electrical Fire....................................................................... 1-21
3.19. Remote Movements............................................................... 1-21
3.20. LaserGuard .......................................................................... 1-21
3.21. On-Board Imager Precautions............................................... 1-22
3.22. High Dose Precautions ......................................................... 1-22
viii Table of Contents
3.23. Falling Parts or Accessories.................................................. 1-23
3.24. Deterioration of Plastic Parts from Radiation ........................ 1-23
3.25. Patient Fall from the Treatment Couch................................. 1-23
3.26. Treatment Couch Pinch Points ............................................. 1-24
3.27. High Temperature Surfaces.................................................. 1-24
3.28. Laser Beams ........................................................................ 1-25
3.29. Software Integrity................................................................. 1-25
3.30. Microwave Tube Operating Hazards ..................................... 1-26
4. Service and Maintenance Guidelines.............................................. 1-27
4.1. Clinac Accelerator Specifications............................................ 1-27
4.2. Lockout/Tagout Procedures ................................................... 1-28
5. Owner Guidelines.......................................................................... 1-48
5.1. Planning Operations .............................................................. 1-48
5.2. Radiation Protection Survey ................................................... 1-48
5.3. Safety and Emergency Training.............................................. 1-49
5.4. Routine Use ........................................................................... 1-49
5.5. Quality Assurance.................................................................. 1-50
5.6. Accidental Radiation Overdose ............................................... 1-50
5.7. Backup Interlocks.................................................................. 1-51
5.8. Emergency Beam Termination................................................ 1-51
5.9. Emergency Plan ..................................................................... 1-51
6. Appendix A: Venting Waveguide Gases .......................................... 1-52
6.1. Testing for Waveguide Arcing ................................................. 1-52
6.2. Prerequisites: Parts Ordering Procedure ................................. 1-52
6.3. Venting a Clinac With a Venting System................................. 1-53
6.4. Venting a Clinac Without a Venting System............................ 1-53
6.5. Retrofitting the Clinac With a New Venting System................. 1-55
6.6. Contents of 872031-01 and 872031-02 Kits........................... 1-55
Chapter 2: Machine Physics .................................................2-1
1. Introduction .................................................................................... 2-3
2. Definitions:...................................................................................... 2-3
Table of Contents ix
3. Kinetic Energy Relationships............................................................ 2-3
4. Rest Energy Relationships ............................................................... 2-4
5. Total Energy Relationships............................................................... 2-4
6. Conversion of Energy to Electron Volts............................................. 2-5
7. Measurement of Energy Change....................................................... 2-6
8. Example of Simple Acceleration ....................................................... 2-8
9. The Standing Wave Accelerator ........................................................ 2-9
10. Impedance Matching.................................................................... 2-11
11. Plotting........................................................................................ 2-12
12. Accelerator Equivalent Circuit...................................................... 2-13
13. Load Line Considerations............................................................. 2-16
14. Fill Time ...................................................................................... 2-16
15. Injection Timing........................................................................... 2-17
16. Electron Injection and Bunching.................................................. 2-18
17. Advances in Linear Accelerator Design for Radiotherapy............... 2-19
17.1. Electron Therapy.................................................................. 2-59
Chapter 3: Modulator Theory ...............................................3-1
1. Introduction..................................................................................... 3-5
2. Basic Concepts ................................................................................ 3-5
3. DeQing Principles ............................................................................ 3-7
4. Non-resonant Transmission Line Principles...................................... 3-8
5. Pulse Shape Definitions ................................................................. 3-16
6. Line Type Modulator Load Element Principles ................................ 3-17
7. Fault Conditions............................................................................ 3-18
8. Thyratron Theory........................................................................... 3-19
8.1. Introduction........................................................................... 3-19
8.2. Operating Notes on Hydrogen-filled Tubes .............................. 3-23
x Table of Contents
Chapter 4: RF Theory...........................................................4-1
1. Introduction .................................................................................... 4-5
2. Traveling Waves on Transmission Lines........................................... 4-5
3. Waveforms ...................................................................................... 4-8
4. Waveguides ................................................................................... 4-11
5. Resonant Circuits.......................................................................... 4-14
6. RF Transmission Theory................................................................ 4-16
7. RF Waveguide Design .................................................................... 4-17
7.1. Modes.................................................................................... 4-17
7.2. The TE
10
Mode ....................................................................... 4-18
7.3. Coupling ................................................................................ 4-18
7.4. Determining the TE
10
Dominant Mode of a Waveguide............ 4-19
8. Transmission Lines ....................................................................... 4-20
8.1. VSWR .................................................................................... 4-21
9. Vector Analysis 3dB Quadrature Hybrid...................................... 4-21
10. Circulators .................................................................................. 4-22
11. Klystron Theory........................................................................... 4-24
11.1. Theory of Klystron Operation................................................ 4-24
11.2. Associated Equipment.......................................................... 4-34
11.3. Power Supplies..................................................................... 4-35
11.4. Cooling ................................................................................ 4-39
11.5. RF Circuits .......................................................................... 4-45
11.6. Tuning ................................................................................. 4-49
11.7. Noise in Klystron Amplifiers ................................................. 4-51
11.8. Summary............................................................................. 4-52
Chapter 5: Ion Chamber Theory ...........................................5-1
1. Introduction .................................................................................... 5-3
2. Present Configuration...................................................................... 5-3
3. Efficiency ........................................................................................ 5-5
4. Upper Limit of Dose Range (Saturation) ........................................... 5-6
Table of Contents xi
5. Applied Electric Field ....................................................................... 5-7
6. Effects of Temperature and Pressure................................................ 5-7
7. Beam Opacity .................................................................................. 5-8
8. Inverse Square Law.......................................................................... 5-8
9. Insulation Materials......................................................................... 5-9
9.1. Mica......................................................................................... 5-9
10. Pulse Shape................................................................................. 5-10
11. Concluding Remarks.................................................................... 5-11
12. References ................................................................................... 5-12
Chapter 6: Vacuum Theory...................................................6-1
1. Introduction..................................................................................... 6-3
2. The Nature of Vacuum..................................................................... 6-3
2.1. What Is Vacuum?..................................................................... 6-3
2.2. What About Pressure?.............................................................. 6-4
2.3. How Is a Vacuum Produced? .................................................... 6-4
2.4. Different Types of Vacuum ....................................................... 6-4
2.5. Where Is Vacuum Used?........................................................... 6-5
2.6. Why Is Vacuum Needed? .......................................................... 6-5
3. Temperature .................................................................................... 6-6
4. Pressure .......................................................................................... 6-7
4.1. What is Gas?............................................................................ 6-7
4.2. Atmospheric Pressure............................................................... 6-7
4.3. Pressure Measurement ............................................................. 6-8
4.4. Partial Pressure........................................................................ 6-9
4.5. Vapor Pressure....................................................................... 6-10
4.6. Effects of Pressure.................................................................. 6-12
4.7. Pressure Ranges..................................................................... 6-12
5. Gas Particles.................................................................................. 6-13
6. Gas Laws....................................................................................... 6-13
6.1. Avogadros Law....................................................................... 6-13
xii Table of Contents
6.2. Boyles Law............................................................................ 6-14
6.3. Gas Expansion....................................................................... 6-14
6.4. Charles Law .......................................................................... 6-15
6.5. Gay-Lussacs Law .................................................................. 6-16
6.6. General Gas Law.................................................................... 6-16
7. Gas Flow....................................................................................... 6-17
7.1. Viscous Flow.......................................................................... 6-17
7.2. Molecular Flow....................................................................... 6-17
7.3. Mean Free Path...................................................................... 6-18
8. Conductance ................................................................................. 6-18
8.1. Conductance in Viscous Flow................................................. 6-19
8.2. Conductance in Molecular Flow ............................................. 6-20
9. Review of the Nature of Gases........................................................ 6-20
10. Ion Pump .................................................................................... 6-21
10.1. Components......................................................................... 6-22
10.2. How the Pump Works........................................................... 6-22
10.3. Vacuum System Use ............................................................ 6-26
10.4. Summary............................................................................. 6-26
11. Vacuum Gauges .......................................................................... 6-26
11.1. Thermocouple Gauge ........................................................... 6-26
Chapter 7: Glossary..............................................................7-1
Emergency and Safety 1-1
The material in this chapter is taken from the Varian Clinac Emergency and Safety
Manual, and is included in this chapter for reference to supplement the Safety Lecture
material presented during the Maintenance Training Course.
Emergency and Safety
1-2 Emergency and Safety
Table of Contents
1. Abstract: .......................................................................................................................... 1-5
2. Introduction: .................................................................................................................... 1-7
2.1. Overview:.................................................................................................................. 1-7
2.2. Operators and Treatment Personnel:......................................................................... 1-7
2.3. Maintenance and Service Personnel: ......................................................................... 1-8
2.4. Customer Support: ................................................................................................... 1-8
2.5. Related Publications: ................................................................................................ 1-9
3. Emergency Procedures: .................................................................................................... 1-9
3.1. Terminating the Treatment Beam:............................................................................. 1-9
3.2. Performing an Emergency-Off: ................................................................................ 1-10
3.3. Using the Emergency Hand Pendant: ...................................................................... 1-10
3.4. Lowering the Treatment Couch in an Emergency: ................................................... 1-11
3.5. X-Ray and Electron Beam Radiation: ...................................................................... 1-13
3.6. Induced Radiation in Accelerator Components: ....................................................... 1-14
3.7. Radio Frequency (RF) Radiation: ............................................................................. 1-15
3.8. Electromagnetic Interference:.................................................................................. 1-16
3.9. Sulfur Hexafluoride and Freon 12 Gases:................................................................ 1-16
3.10. Lead: .................................................................................................................... 1-17
3.11. Beryllium:............................................................................................................. 1-18
3.12. Dielectric Insulating Oil: ....................................................................................... 1-18
3.13. Depleted Uranium (Low-Energy Clinacs): .............................................................. 1-19
3.14. Ozone and Oxides of Nitrogen (High-Energy Clinacs): ............................................ 1-19
3.15. Implosion: ............................................................................................................ 1-19
3.16. Electric Shock: ..................................................................................................... 1-20
3.17. Service Precautions:.............................................................................................. 1-20
3.18. Electrical Fire: ...................................................................................................... 1-21
3.19. Remote Movements: .............................................................................................. 1-21
3.20. LaserGuard: ......................................................................................................... 1-21
3.21. On-Board Imager Precautions: .............................................................................. 1-22
3.22. High Dose Precautions:......................................................................................... 1-22
Emergency and Safety 1-3
3.23. Falling Parts or Accessories: .................................................................................. 1-23
3.24. Deterioration of Plastic Parts from Radiation: ........................................................ 1-23
3.25. Patient Fall from the Treatment Couch: ................................................................. 1-23
3.26. Treatment Couch Pinch Points: ............................................................................. 1-24
3.27. High Temperature Surfaces: .................................................................................. 1-24
3.28. Laser Beams: ........................................................................................................ 1-25
3.29. Software Integrity: ................................................................................................. 1-25
3.30. Microwave Tube Operating Hazards: ..................................................................... 1-26
4. Service and Maintenance Guidelines:.............................................................................. 1-27
4.1. Clinac Accelerator Specifications: ............................................................................ 1-27
4.2. Lockout/Tagout Procedures: ................................................................................... 1-27
5. Owner Guidelines: .......................................................................................................... 1-48
5.1. Planning Operations: .............................................................................................. 1-48
5.2. Radiation Protection Survey: ................................................................................... 1-48
5.3. Safety and Emergency Training: .............................................................................. 1-49
5.4. Routine Use: ........................................................................................................... 1-49
5.5. Quality Assurance:.................................................................................................. 1-50
5.6. Accidental Radiation Overdose: ............................................................................... 1-50
5.7. Backup Interlocks: .................................................................................................. 1-51
5.8. Emergency Beam Termination:................................................................................ 1-51
5.9. Emergency Plan: ..................................................................................................... 1-51
6. Appendix A: Venting Waveguide Gases: .......................................................................... 1-52
6.1. Testing for Waveguide Arcing: ................................................................................. 1-52
6.2. Prerequisites: Parts Ordering Procedure: ................................................................. 1-52
6.3. Venting a Clinac With a Venting System:................................................................. 1-53
6.4. Venting a Clinac Without a Venting System:............................................................ 1-53
6.5. Retrofitting the Clinac With a New Venting System: ................................................. 1-55
6.6. Contents of 872031-01 and 872031-02 Kits: ........................................................... 1-55
1-4 Emergency and Safety
Table of Illustrations
Figure 1.1. Console Dedicated Keyboard:............................................................................ 1-10
Figure 1.2. Emergency Pendant:......................................................................................... 1-11
Figure 1.3. Typical Lockout Devices: .................................................................................. 1-29
Figure 1.4. Main Circuit Breaker Lockout/Tagout: ............................................................. 1-30
Figure 1.5. Gantry Locking Pin with Lock and Tag:............................................................. 1-34
Figure 1.6. ETR or Exact Couch with Safety Braces:........................................................... 1-35
Figure 1.7. PSA Couch with Safety Braces: ......................................................................... 1-36
Figure 1.8. Air Hose Removed and Lockout Tag Attached: .................................................. 1-37
Figure 1.9. Handle Removed and Lockout Tag Attached: .................................................... 1-38
Figure 1.10. Water Valve with Handle Removed and Tag Attached: ..................................... 1-40
Emergency and Safety: Abstract 1-5
1. Abstract The Clinac Safety Guide (P/N 1104957-03) provides reference information about
safety precautions pertaining to C-Series Clinac accelerators.
European
Representative

Notice
Information in this document is subject to change without notice and does not
represent a commitment on the part of Varian. Varian is not liable for errors
contained in this document or for incidental or consequential damages in
connection with furnishing or use of this material.
This document contains proprietary information protected by copyright. No part of
this document may be reproduced, translated, or transmitted without the express
written permission of Varian Medical Systems, Inc.
FDA 21 CFR 820
Quality System
Regulations
(CGMPs)
Varian Medical Systems products are designed and manufactured in accordance
with the requirements specified within this federal regulation.
ISO 9001 and
ISO 13485
Varian Medical Systems products are designed and manufactured in accordance
with the requirements specified within ISO 9001 and ISO 13485 quality standards.
CE 0086
Varian Medical Systems products meet the requirements of Council Directive
MDD 93/42/EEC.
Trademarks
Clinac, Silhouette, LaserGuard, and VARiS are registered trademarks, and RMS,
PortalVision, On-Board Imager, iX, Trilogy, and EDW are trademarks, of Varian
Medical Systems, Inc.
All other trademarks or registered trademarks are the property of their respective
owners.
19952004 Varian Medical Systems, Inc.
All rights reserved. Printed in the United States of America.
Manufacturer: European Representative:
Varian Medical Systems, Inc.
3100 Hansen Way, Bldg. 4A
Palo Alto, CA 94304-1030, US
Varian Medical Systems UK Ltd.
Gatwick Road, Crawley
West Sussex RH10 9RG United
Kingdom
1-6 Emergency and Safety: Abstract
Important Notice
This manual includes information that is critical to ensure safe operation,
service, and maintenance of Varian Medical Systems medical linear accel-
erators.
This manual is designed to assist properly trained and equipped personnel
in safely completing work on the Varian Medical Systems medical linear ac-
celerators. Comprehensive knowledge of the standards of care required for
operation, service, and maintenance of the accelerator is needed to make
effective use of this manual; the manual is not a substitute for such knowl-
edge. Only properly trained personnel should operate, service, or maintain
any Varian Medical Systems medical linear accelerator.
Varian is not responsible for injuries or damages due to activities which do
not conform to generally accepted standards of care, or which are inconsis-
tent with specific provisions of this manual.
Emergency and Safety: Introduction 1-7
2. Introduction This chapter has been reproduced from the Varian Clinac

Safety Manual,
P/N 1104957-03, published 12/2004. Within this chapter, the original
document words Manual and Chapter have been changed accordingly.
2.1. Overview This safety manual provides information about emergency procedures,
safety precautions, service and maintenance guidelines, and owner guide-
lines that apply to the general operation and maintenance of the Clinac ra-
diotherapy accelerator and associated equipment.
2.1.1. Visual
Cues
This document uses the following visual cues to help you locate and iden-
tify information:
Note: This symbol identifies comments about a specific task.
Some examples include notes, required data, messages, sub-
stitutions, and shortcuts.
CAUTION: Describes actions or conditions that can result in
minor or moderate injury or can result in damage to equip-
ment.
WARNING: Describes actions or conditions that can re-
sult in serious injury or death.
Italics: Used for emphasis, defining new terms, or book
titles.
Bold: Identifies menu commands, items you can select on
the screen, and buttons to press.
Preparing Personnel
The Clinac is a sophisticated and potentially hazardous piece of equipment.
Uninformed or careless operation or service can result in poor perfor-
mance, equipment damage, serious and possibly fatal injuries.
The owner should require the following actions of each person who oper-
ates, maintains, or is otherwise associated with the Clinac:
Become thoroughly familiar with the material in this manual and the oper-
ating instructions detailed in your user documentation.
Become thoroughly familiar with and follow the emergency procedures and
safety precautions in this manual, as well as warnings and cautions con-
tained in this manual and in the other books in the documentation set.
Become thoroughly familiar with and follow the emergency and safety pro-
cedures established for local use by the owner.
2.2. Operators
and Treatment
Personnel
Varian Clinac medical linear accelerators are sophisticated and potentially
hazardous pieces of equipment. Unauthorized or careless operation or ser-
vice can result in poor performance, equipment damage, or serious and
possibly fatal injuries.
The owner must require the following actions of each person who operates,
maintains, or is otherwise associated with the Clinac:
1-8 Emergency and Safety: Introduction
! Learn the contents of this guide and the operating instructions de-
tailed in your user documentation.
! Follow the emergency procedures, safety precautions, warnings, and
cautions in this guide, and in all other related publications.
! Follow emergency and safety procedures established for local use by
the owner.
2.3. Mainte-
nance and Ser-
vice Personnel
Maintenance and service procedures are restricted to service personnel
who receive the appropriate maintenance training and are authorized by
the owner.
WARNING: Authorized service personnel must become
thoroughly familiar with and follow lock-
out/tagout safety procedures established
for local use by the owner during all service
and maintenance procedures. They are also
required to take all precautions necessary
to protect themselves, patients, and other
persons from injury, and to protect equip-
ment from damage.
To ensure safe operation and maintenance conditions for use of any medi-
cal linear accelerator, the owner is responsible for establishing emergency
and safety procedures. Use this manual, along with the warnings and cau-
tions in the other books in the documentation set as the starting point for
formulating local procedures.
2.4. Customer
Support
If you cannot find information in this user guide, you can contact Varian in
several ways:
! Help Desk Support
North American toll-free support: 1.888.827.4265
Global telephone support:1.702.938.4807
Global telephone support, Treatment Planning:
1.702.938.4712
! To order additional documents
From North America:1.800.535.5350
and press 1 for Parts on your touch-tone phone
Globally: 1.702.938.4700
! World Wide Web
If you have access to the Internet, point your browser to Oncol-
ogy Systems: http://www.varian.com
and then select Support.
! E-mail
Information Management Systems, Digital Imaging Manage-
ment Systems, and Delivery Systems:
onc.helpdesk@varian.com
Treatment Planning Systems: tps.support@us.varian.com
Brachytherapy Systems: brachy.support@varian.com
Emergency and Safety: Emergency Procedures 1-9
! United States mail
Varian Medical Systems, Inc.
3100 Hansen Way, Bldg. 4A
Palo Alto, CA 94304-1030, U.S.A.
! European representative
Varian Medical Systems, UK Ltd.
Gatwick Road, Crawley
West Sussex, RH102RG, England
Phone: +44-1293-531-244
2.5. Related
Publications
The following Varian publications provide further information about the
Clinac and related products:
! C-Series Clinac Instructions for Use (P/N 1102903)
! C-Series Clinac Technical Reference Guide (P/N 1106795)
! Exact Couch User Guide/Maintenance Manual for Couch and Couch
Top (P/N 1104201)
! C-Series Clinac Custom Coding Guide (P/N 1106292)
! MLC User Guide (P/N 1101351)
! MLC Systems and Maintenance Guide (P/N 1101018)
! Shaper User Guide (P/N 1101352)
! DMLC Implementation Guide (P/N 1105417)
! LaserGuard Clinical Reference Guide (P/N 100011555)
! Service Technical Bulletins (STBs)
! Customer Technical Bulletins (CTBs)
3. Emergency
Procedures
This section describes procedures recommended for use in the event of an
emergency. The owner is responsible for adapting these procedures locally
and establishing other emergency procedures.
3.1. Terminat-
ing the Treat-
ment Beam
To terminate the beam:
From the console dedicated keyboard (Figure 1.1), press the BEAM OFF
button and turn the DISABLE/ENABLE keyswitch to the DISABLE posi-
tion. Remove the key and place it in a secure location.
WARNING: In the event of an emergency during opera-
tion of the Clinac, terminate the treatment
beam immediately. Examples of an emer-
gency situation requiring termination of the
treatment beam include the following:
! Clinac fails to properly terminate a treatment
! Accumulated radiation dose displayed on the console
monitor exceeds the dose preset for the treatment
! Fire, smoke, or gas fumes are detected
! Power failure
1-10 Emergency and Safety: Emergency Procedures
WARNING: If the beam remains on, press the nearest
Emergency Off button.
3.2. Performing
an Emergency-
Off
Pressing an Emergency Off button immediately turns off power to all com-
ponents except the console video monitor and console computer. These re-
main powered on.
Emergency Off buttons supplied by Varian are located on the dedicated
keyboard, on both sides of the treatment couch, next to the auxiliary elec-
tronics chassis in the drive stand, and on both sides of the drive stand.
Other Emergency Off buttons may be installed by the owner of the facility.
For additional information about the location of Emergency Off buttons,
see Section 5.1.1 on Page 1-48.
To verify that the Emergency Off button has interrupted power, listen for
mechanical noises (for example, running motors or fans) in the treatment
room. If you hear evidence that power is still on, turn off the main facility
circuit breaker.
! If a patient is receiving therapy, remove the patient from the treat-
ment couch as soon as possible.
! Check the dose counters, and record the cumulative number of mon-
itor units received by the patient up to that time.
! Do not try to operate the accelerator until service personnel have re-
stored proper operation of the machine, including operation of the
emergency-off circuits.
3.3. Using the
Emergency
Hand Pendant
If couch motion controls do not function due to power failure or other emer-
gency, use the emergency pendant to lower the couch for safe removal of a
patient (Figure 1.2). You cannot use the emergency pendant to raise the
Figure 1.1. Console Dedicated Keyboard
couch. The emergency pendant is located in the drive stand of the Clinac. It
is battery-powered and remains functional after:
! Power failure.
! Emergency Off button has been pressed.
You can also use the emergency pendant to lower the treatment couch if
the couch fails to respond to a vertical motion control. The pendant is not
operational unless the emergency pendant switch is enabled.
Before using the emergency pendant, you must enable it by pressing an
EMERGENCY PENDANT ON/OFF switch. The emergency pendant and
emergency pendant on/off switch are located as follows:
! On standard Clinacs, the emergency pendant and EMERGENCY
PENDANT switch are mounted in the right side of the drive stand.
! On Silhouette high-energy models, the emergency pendant and
EMERGENCY PENDANT ON/OFF switch is mounted behind the door
to the immediate right of the accelerator.
3.4. Lowering
the Treatment
Couch in an
Emergency
You can use the emergency pendant to change the position of the couch so
that the patient can be removed from the treatment room, if other controls
do not function.
First, enable the emergency pendant:
1. On the console dedicated keyboard, turn the DISABLE/ENABLE
keyswitch to DISABLE.
2. Enter the treatment room and tell the patient to remain on the couch.
3. Open the right door of the drive stand and toggle the EMERGENCY
PENDANT ON/OFF switch to ON.
4. Remove the emergency pendant from its storage location inside the
drive stand and return to the treatment couch with the pendant. Do not
disconnect the pendant cable from the drive stand.
Figure 1.2. Emergency Pendant
Next, move the couch:
1. Check the longitudinal position of the couch to see if the patient can
leave safely.
2. If the couch top is in the longitudinal position you want, proceed to
step 3.
If the longitudinal position of the couch top does not allow the
patient to safely leave the couch:
A. On the emergency pendant (Figure 1.2), rotate the
NORM/OUT/DOWN switch to the OUT position.
B. Press and hold down the enable button on top of the emergency
pendant.
C. Continue to hold down the enable button while you manually
move the couch top to the desired position along the
longitudinal axis.
D. Release the enable button to reengage the longitudinal brake.
3. Check the height of the couch top to see if the patient can leave it safely.
If the couch top is at the height you want, proceed to step 4.
If the couch top is too high for the patient to leave safely, lower the couch
top:
A. On the emergency pendant, rotate the NORM/OUT/DOWN switch
to the DOWN position.
B. Press and hold down the enable button on top of the pendant.
While you hold down the enable button, the couch top lowers.
C. Release the button when the top reaches the height you want.
4. Help the patient from the couch and treatment room.
5. Rotate the NORM/OUT/DOWN switch to the NORM position and toggle
the EMERGENCY PENDANT ON/OFF switch to the off position.
6. Return the emergency pendant to its storage location in the drive stand.
7. Safety Precautions
This section identifies significant operating and maintenance hazards asso-
ciated with the Clinac. This information is provided to help you avoid or
control these hazards.
Note: The identification of potential hazards and suggested
safety precautions described in this manual are provided to
assist you in maintaining a safe workplace. Varian recom-
mends that you refer to applicable federal, state, and other
local standards for more information and specific safety
requirements.
Refer also to Related Publications on page 1-9 for additional
information.
3.5. X-Ray and
Electron Beam
Radiation
The Clinac can produce a lethal radiation dose in a very short time. Never
operate the X-ray or electron beam without adequate X-ray shielding.
Radiation exposure can cause serious illness or death, though not instan-
taneously. Radiation exposure may also cause certain types of cardiac
pacemakers to malfunction.
3.5.1. Person-
nel Precau-
tions
! Post signs warning all persons of the radiation hazard in the area.
! Permit no person other than the patient in the treatment room when
the treatment beam is on.
! When working on or near the machine, wear radiation monitoring de-
vices approved by the cognizant regulatory agency.
! Allow only trained, qualified personnel to operate or maintain the
machine.
3.5.2. DIS-
ABLE/ENABLE
Key Precau-
tions
! Before entering the treatment room, operators and service personnel
must turn the DISABLE/ENABLE keyswitch to the DISABLE posi-
tion. Remove the key and place it in a secure location.
! When entering the treatment room, block the door so that it cannot
close and enable the door interlock while you are inside.
! Insert the DISABLE/ENABLE key into the keyswitch on the dedicat-
ed keyboard and turn it to the ENABLE position only when necessary
to enable the beam-on condition. Immediately after the beam termi-
nates, turn the DISABLE/ENABLE keyswitch to the DISABLE posi-
tion.
! When powering down or placing the Clinac in standby, put the DIS-
ABLE/ENABLE and power keys in a secure key-storage enclosure.
Lock the enclosure to prevent unauthorized activation of the ma-
chine.
3.5.3. Treat-
ment Precau-
tions
! For each treatment plan, the physicist or dosimetrist needs to give
appropriate attention to the dose correction factors resulting from
scatter and attenuation through any object placed between or near
the radiation source and the patient. Typical objects include devices
such as shadow trays, centerspine attachments, centerspine and
side rails of the treatment couch, couch panels, tennis racket panels,
wedge and compensating filters, and shadow blocks.
! Operating personnel should monitor the accumulated radiation dose
readout continually during beam-on.
! Operating personnel should terminate the beam according to the lo-
cal emergency procedure if the beam fails to terminate on the preset
dose.
3.5.4. Cardiac
Pacemaker
Precautions
! Before you treat a patient wearing a pacemaker, contact the manu-
facturer of the pacemaker about possible hazards that may pertain to
treatment by a linear accelerator.
! Make arrangements to monitor the operation of a pacemaker worn by
anyone in the treatment room or in adjacent rooms.
! If you suspect EMI is affecting the operation of a pacemaker, stop op-
erating the Clinac immediately.
! Post warnings of the potential danger to individuals wearing a pace-
maker.
3.5.5. Radia-
tion Overdose
In the event a person is exposed or thought to be exposed to excess radia-
tion, the law in most localities of the United States requires the following
steps:
! Immediately notify the appropriate local, state, and federal authori-
ties.
! Request an investigation by a professional qualified in the detection
of radiation.
! Consult with medical experts in radiation treatment.
For more information, see Accidental Radiation Overdose on page 1-50.
3.6. Induced
Radiation in
Accelerator
Components
Copper, iron, lead, and tungsten components can become temporarily ra-
dioactive when irradiated by X-rays with energies above 10 MeV. Induction
of radioactivity is caused by the photoneutron or (c,n) reaction, which pro-
duces neutrons. When the product nucleus is left in an unstable state, it
becomes radioactive.
The primary radioactive components are those which absorb most of the X-
ray energy:
! Targets
! Collimator assemblies
! Compensating filters
! Other shielding and structural material surrounding the target
Except for copper, the other materials become only mildly radioactive fol-
lowing irradiation.
The greatest concentration of radioactivity occurs in the target (tungsten
and copper). Immediately after beam shutdown the target can contain up to
15% of the total activity produced in the accelerator. The target is approxi-
mately 20 times more radioactive than the next most active component.
Most of the measurable radiation streams out of the primary collimator and
jaw openings. The copper energy slits in the magnet system can also be-
come highly radioactive because they come in direct contact with the beam
and act as secondary targets. However, radiation from the target is well-
shielded and the radiation is collimated like the primary beam; it is easy to
avoid unnecessary exposure.
Some radioactivity is also induced in the tungsten collimator, the tungsten
or iron flattening filter, and the iron and copper in the magnet. Most of this
energy is induced in the shielding within 30 from the beam axis. The tung-
sten collimator and jaw system are also subject to high X-ray flux. Howev-
er, the relatively short half-life of the principal tungsten isotope (115 days)
limits the buildup of radioactivity.
Other components, such as the tungsten flattening filter and structural
parts, are 20 times less active than copper when exposed to the same X-ray
field.
3.6.1. Handling
Precautions
You can remove radioactive target assemblies safely, with minimal expo-
sure, if you follow safe well-planned work practices. Use of three basic prin-
ciplestime, distance, and shieldingresults in doses being As Low As
Reasonably Achievable (ALARA). To minimize exposure, follow these guide-
lines:
! Always allow the maximum time possible for radioactive decay to oc-
cur. Higher energies produce higher amounts of radioactivity. For ex-
ample, an 18 MV machine requires a longer cooling period than a
15 MV machine.
! Perform a radiation survey to assess the hazard. Using a dose rate
meter, open the jaws and survey the nearest accessible area beneath
the collimator.
! Plastic trays, wedges, and other accessories can become radioactive
following high-energy X-ray treatments. Inform personnel of the risk
of residual exposure and take appropriate precautions.
! Survey all potentially radioactive disposable components with a pan-
cake GM survey instrument prior to disposal. Radioactive compo-
nents require proper handling and disposal techniques.
3.6.2. Handling
the X-Ray Tar-
get
The X-ray target is handled only during major service or repair. The target
can become radioactive depending on the Clinac energy levels and how long
it has been since the target was irradiated. A radiation survey meter is the
only way to determine if the target is safe to handle.
Be sure to use a radiation survey meter to detect radioactivity before han-
dling the target.
3.7. Radio Fre-
quency (RF)
Radiation
The microwave power tube in the drive stand or gantry produces high levels
of microwave energy that is supplied to the linear accelerator through a
waveguide system. Since this energy does not appreciably penetrate metal,
the power tube body and the entire RF system are made of metal and de-
signed to attenuate or shield RF radiation.
Avoid exposure of personnel to RF radiation and prevent anyone from being
in the vicinity of open energized waveguides. Exposure to high levels of RF
radiation can result in serious bodily injury, including blindness.
3.7.1. General
Precautions
To minimize exposure of personnel to RF radiation:
! All input and output RF connections, waveguides, flanges, and gas-
kets must be tight to prevent RF leakage.
! Do not operate the magnetron or Klystron tube unless it is properly
attached to an appropriate energy-absorbing load.
! Cardiac pacemakers may be affected.
! If you suspect leakage of RF energy, do not attempt to operate the Cli-
nac until service personnel have verified proper operation of the ma-
chine.
! Do not expose any part of your body to an energized RF waveguide
system that is open or loosely bolted together, or to the window of an
energized magnetron or Klystron.
! Never look into or expose any part of the body to an open waveguide
while the tube is energized.
! Permit only service personnel with the appropriate training and expe-
rience to service or repair the magnetron or Klystron tube, the RF
waveguide system, or the energy-absorbing load of a Clinac.
3.8. Electro-
magnetic Inter-
ference
The low levels of electromagnetic radiation present around the Clinac when
the beam is on can cause electromagnetic interference (EMI) with cardiac
pacemakers and surrounding hospital equipment.
EMI can interfere with the operation of cardiac pacemakers or patient mon-
itoring equipment, possibly causing serious bodily injury or death. EMI
from other equipment, such as microwave hyperthermia or diathermy
equipment, can interfere with the Clinac integrated dose counters, which
could result in incorrect doses to patients.
3.8.1. General
Precautions
! Keep covers, doors, and panels on the Clinac closed during opera-
tion.
! If EMI appears to be interfering with the operation of the Clinac or
any equipment in the vicinity, cease operation of the Clinac immedi-
ately.
! To reduce EMI, reinstall all fasteners for covers, screens, and panels.
Screws in panels and covers must be correctly and completely in-
stalled. Check to make sure that they are tight.
! For EMI pacemaker precautions, see Cardiac Pacemaker Precau-
tions on page 1-13.
3.9. Sulfur
Hexafluoride
and Freon 12
Gases
Sulfur hexafluoride (SF
6
) and Freon 12 are colorless, odorless, nontoxic
gases used in Clinacs as a dielectric to prevent arcing in the RF waveguide.
These gases are stored as a liquid under pressure in a disposable metal
container in the drive stand. Freon 12 is present in older model Clinacs
only. SF
6
is now used in Clinac production.
The waveguide is a sealed unit. Exposure to breakdown products should
not occur unless waveguide integrity is compromised.
Note: Freon 12 is present in Clinacs 4/100 (prior to S/N 81),
6/100 (prior to S/N 457), and 600C (prior to S/N 66).
3.9.1. General
Precautions
WARNING: ! Improper handling of SF
6
and Freon 12 gas cylinders can
result in the rapid release of pressurized gas, causing se-
rious and possibly fatal injury.
! Contact with gas or liquefied gas may cause serious inju-
ries, including burns or frostbite.
! Handle gas cylinders with care. A ruptured gas cylinder
may become a projectile.
! Although these gases are nontoxic, they can act as an asphyxiant by
displacing oxygen.Avoid an inhalation hazard by venting in accor-
dance with standard venting procedures as described in CTB42,
Venting Waveguide Gases.
! High-voltage arcing in the RF waveguide may cause SF
6
and Freon 12
to break down into toxic compounds including hydrochloric and hy-
drofluoric acids; and chlorine, fluorine, and phosgene gases. Treat
used waveguide pieces as though they are contaminated and handle
with personal protective equipment (PPE). Always wear eye protection
and chemical-resistant gloves. For information about PPE, see Using
Personal Protective Equipment on page 1-45.
3.9.2. Gas Cyl-
inder Storage
and Handling
! Attach valve caps to all stored cylinders to prevent damaging the
stem and the possible release of stored energy.
! Store in a clean, dry, and well-ventilated area where the temperature
does not exceed 125 Fahrenheit (52 Celsius).
! Ensure that all cylinders are secured to the drive stand and that all
spares are secured with chains.
! Handle cylinders with care, avoiding any collisions.
! Do not refill or reuse damaged or dropped cylinders.
3.9.3. First Aid
for Gas Expo-
sure
! For inhalation, immediately remove the person to fresh air and seek
medical attention.
! In case of skin contact, flush with copious amounts of water. Treat
for frostbite if necessary.
! In case of eye contact, immediately flush with copious amounts of
water. Seek medical attention.
3.10. Lead Lead is present in the shielding, balance weights, and some wedge trays in
the Clinac.
! Handling of lead shielding, balanced weights, or wedge trays may
generate lead dust that can be inhaled or ingested.
! Exposure to lead dust can cause adverse health effects such as ane-
mia and gastrointestinal abnormalities.
! Severe overexposure to lead dust may cause neuromuscular dys-
function, paralysis, and birth defects.
3.10.1. Precau-
tions
! Observe proper hygiene and wear a disposable mask to avoid inhala-
tion and ingestion of lead-contaminated dust.
! Always wear gloves to protect against skin contact when handling
lead. The gloves worn for lead handling should not be used for other
purposes.
! When handling lead with leather gloves, wear latex or vinyl gloves un-
derneath the leather gloves. Lead will penetrate leather gloves that
have been used repeatedly for lead. Always remove gloves and store
them in a designated plastic storage bag before leaving the work area.
! After handling lead or lead-contaminated materials, wash your
hands and face thoroughly.
3.10.2. Lead
Handling and
Storage
! Do not use compressed gas to clean lead components.
! Do not use water or saturated wet cloths to clean lead. Excess mois-
ture may cause lead to oxidize. However, you can use a moist, lint-
free cloth to avoid airborne lead oxide particulates.
! Follow proper lifting practices when lifting and handling lead.
3.10.3. First Aid
for Lead Expo-
sure
! If lead residue contaminates the skin, wash well with soap and water.
! If lead is ingested or inhaled, contact a physician immediately for as-
sistance.
3.11. Beryllium Beryllium is present in the bend-magnet window of all high-energy Clinacs.
Normal service and maintenance procedures do not expose you to berylli-
um compounds. However, removal and manipulation of the window may
cause exposure to beryllium particulates.
! Beryllium can be absorbed through the skin.
! It may enter into the body through a cut or puncture, producing hard
lesions with central nonhealing areas.
! Beryllium poisoning may result in serious health problems including
neuromuscular dysfunction and paralysis. Beryllium has also been
known to cause cancer.
3.11.1. Precau-
tions
! Wear disposable gloves to avoid skin contact.
! Wear safety glasses to prevent eye contact.
! Observe proper personal hygiene after contact with beryllium.
3.11.2. First Aid
for Beryllium
Exposure
! If dust enters the eyes, flush with copious amounts of water.
! If cut with broken glass, treat with first aid. Wash with copious
amounts of water.
! Immediately seek medical attention.
3.12. Dielectric
Insulating Oil
Dielectric insulating oils are present in the pulse transformer, modulator,
rectifiers, and capacitors in all Clinacs.
! Dielectric insulating oil is harmful if swallowed. It is also a skin irri-
tant.
! Dielectric insulating oil may become hazardous over time by breaking
down and becoming infused with heavy metals.
! Dielectric oils can also produce aromatic hydrocarbons and carbon
monoxide upon combustion.
3.12.1. Precau-
tions
! Wear gauntlet-length rubber gloves when handling oils.
! Failure of an oil-filled component can result in smoke. If smoke is de-
tected, ventilate the area immediately. Excessive exposure to vapors
is moderately irritating to the eyes and mucous membranes, though
it does not pose a significant health risk.
3.12.2. First Aid
for Dielectric
Insulating Oil
Exposure
! In the event of eye or skin contact, flush the affected area with copi-
ous amounts of water.
! If ingested, seek medical attention.
3.13. Depleted
Uranium (Low-
Energy Cli-
nacs)
Depleted uranium is present in the X-ray head of older low-energy Clinacs.
New production machines no longer use depleted uranium. Depleted ura-
nium is a naturally occurring, processed radioactive material, and must be
handled with caution. A major portion of the radiation emanates as alpha
and beta particles, with only a small fraction being emitted as gamma rays.
External radiation levels are well below those generally considered hazard-
ous.
Depleted uranium is radioactive and must be handled with caution. Urani-
um oxide poisoning can result from oxidation of the depleted uranium com-
ponents, if the protective coating of plated components is damaged.
3.13.1. Precau-
tions
! Under no circumstances should anyone attempt to machine, file,
drill, scrape, scratch, or break the protective coating (plating) of the
depleted uranium components.
! When handling depleted uranium, prevent scratching the protective
coating by wearing heavy leather gloves without metal snaps or riv-
ets. This prevents scratching the protective coating.
! Some machines use depleted uranium alloyed with 0.75% titanium
and do not have the removable contamination problem associated
with earlier machines using unalloyed depleted uranium.
3.14. Ozone
and Oxides of
Nitrogen (High-
Energy Cli-
nacs)
The interaction of the Clinac electron or X-ray beam and air produces
ozone and oxides of nitrogen, although normally in negligible quantities
unless high dose rates and long exposures are experienced. Because of its
high radiolytic yield and chemical reactivity, ozone gas is the most toxic of
the gases formed from this interaction. Pure ozone is an unstable, faintly
bluish gas with a characteristically fresh, penetrating odor.
The average person can detect ozone at a concentration approximately
equal to the generally accepted threshold limit value of 0.1 ppm. Ozone af-
fects the respiratory system and irritates the eyes and all mucous mem-
branes. High ozone concentrations enhance the reactivity of combustible
materials.
3.14.1. Precau-
tions
! If ozone is detected, shut down the Clinac immediately and vacate the
treatment room.
! Allow sufficient time for normal room ventilation to exhaust the gas
before reentering the treatment room.
! Do not operate the Clinac again until service personnel have checked
room ventilation and verified machine operation.
3.15. Implosion Because of the internal vacuum in the magnetron or Klystron and the lin-
ear accelerator, the ceramic and glass windows that separate the vacuum
from the waveguide can shatter inward (implode) if struck with a hard ob-
ject or subjected to mechanical shock. Other components that can implode
are the thyratron tubes in the modulator and the cathode-ray tubes (CRTs)
in the console video monitors.
Flying debris from an implosion could result in bodily injury, including
cuts and puncture wounds.
3.15.1. Precau-
tions
When working with or near any part of the Clinac containing a vacuum,
personnel should take every precaution to protect their bodies from flying
debris produced by a possible implosion.
3.16. Electric
Shock
During operation, the normal voltages in some areas of the Clinac are over
25,000 Vdc. Pressing an Emergency Off button removes power from the
gantry, drive stand, and parts of the console. However, voltages up to 230
Vac (380 Vac at 50 Hz) still remain in the power line from the main facility
circuit breaker, the console computer, the console video monitor, the con-
sole printer, and the power line supplying these components.
With the main circuit breaker turned off, locked, and tagged, many high-
voltage capacitors in the Clinac continue to pose a potential shock hazard
until they are discharged. Several high-voltage capacitors can recharge
spontaneously to dangerous levels even after being discharged. Shorting
sticks attached to the drive stand, gantry, and the modulator cabinet are
designed to provide a convenient and safe means for discharging the high-
voltage capacitors.
Varian recommends you use shorting sticks to prevent inadvertent lack of
ground. The shorting sticks must be left hanging in place to prevent capac-
itors from recharging.
Contact with high-voltage circuits can cause serious injury or death.
3.16.1. General
Precautions
! Keep equipment covers, doors, and panels closed during operation.
! Do not bypass interlock switches on the Clinac doors and panels.
! Do not touch any component inside the Clinac unless you know it
does not present a shock hazard. The potential for electrical shock re-
mains until each circuit supplying the Clinac is turned off at the fa-
cility circuit breaker and high-voltage capacitors are shorted out and
remain shorted out.
3.17. Service
Precautions
! During service procedures, before you touch any high-voltage circuit
component, remove power from the circuit and follow appropriate
lockout/tagout procedures in accordance with OSHA 29CFR
1910.147 or equivalent local standards.
! Hire only properly trained and experienced service personnel with a
full understanding of electrical hazards and high-voltage safety prac-
tices.
! Ground the circuit with the nearest shorting stick. Keep the stick at-
tached to the circuit while you work on or near it. Use more than one
shorting stick to eliminate the possible loss of ground.
! Lockout/Tagout the system or subsystem that is being serviced.
! Interlocks that must be bypassed for maintenance purposes must be
restored immediately upon completion of the task. Never leave the
machine unattended with any interlock bypassed without first post-
ing appropriate warning signs.
! Permit only service personnel with the appropriate training and expe-
rience to remove the protective housing of laser-beam units, accesso-
ries, and power supplies. For additional warnings, refer to the docu-
mentation supplied by the laser unit manufacturer.
! Take special care and always turn off electrical current when working
in confined areas of the Clinac, for example, in the Silhouette service
bays located on either side of the stand. If any electrical components
in these areas are not turned off, inadvertent electrical shock could
occur.
! Adequate lighting is required in any area where work is performed. If
necessary, provide additional lighting. For example, additional light-
ing is required for pin-to-pin troubleshooting in the Silhouette cabi-
net.
! Turn off any electrical equipment before you move it.
3.18. Electrical
Fire
Any electrical equipment carries the risk of an electrical fire. An electrical
fire in the treatment room or console area could cause severe burns, as-
phyxiation, and other injuries or death.
3.18.1. Precau-
tions
! If you see smoke or smell a burning odor, follow the established local
emergency procedure for a fire.
! Never use water to fight electrical fires.
! Permit only personnel trained in fire-fighting procedures to attempt
to put out an electrical fire.
3.19. Remote
Movements
The Clinac can be configured to allow couch position corrections and large
couch rotation swings, as well as gantry and imager arm motions, from
outside the treatment room. Some treatments (for example, stereotactic
treatments) may require that the gantry move very close to the treatment
couch, which increases the risk of collision.
Although the gantry rotates at a maximum speed of only about one revolu-
tion per minute, substantial force is required to stop gantry motion
because of the inertia of its large mass.
In addition, if you have an imaging system (for example, On-Board Imager
or PortalVision), extended imager arms intrude into the space near the
treatment couch. 4D Console and On-Board Imager (OBI) software applica-
tions do not automatically retract the OBI arms. This presents danger of
collision.
For instructions on configuring motion limits, refer to the Clinac Technical
Reference Guide.
Use caution to prevent collisions:
! Retract all imager arms (OBI and/or PortalVision) away from the
couch when not in use.
! Always observe remote motions either directly, or using closed-circuit
monitors.
! Perform a dry-run (a pre-treatment execution of all motions prior to
first treatment, or after a Plan correction) to check for potential colli-
sions before delivering treatment.
! Stop motions immediately if you suspect a collision may occur. The
fastest way to stop motion is to release the Motion Enable bar.
If a collision occurs between the treatment couch and the gantry, call main-
tenance to inspect the system for damage. Do not operate the Clinac again
until the machine is fully checked by service personnel.
Never use the system for patient simulation until proper and safe operabil-
ity has been verified by maintenance personnel.
3.20. Laser-
Guard
LaserGuard is an optional collision detection system for the Clinac. Laser-
Guard uses an invisible laser beam shield to detect potential collisions
between the patient, treatment couch, and gantry. LaserGuard is a Class 1
laser product, and exposure for any length of time does not cause injury or
damage to the eyes.
LaserGuard does not replace existing safety systems or the vigilance of the
user. You must verify all patient and equipment clearances before perform-
ing any remote motion sequence.
Configuration of LaserGuard by unauthorized personnel can cause a
machine malfunction that may result in damage to equipment, bodily inju-
ry, or death. Only authorized personnel should configure LaserGuard.
Avoid divulging the configuration password to unauthorized individuals.
For detailed information, refer to the LaserGuard Clinical Reference Guide.
3.21. On-Board
Imager Precau-
tions
The Varian On-Board Imager (OBI) allows you to take kV setup fields with
two arms that you can rotate into position on either side of the treatment
couch. One On-Board Imager arm delivers kV beam and the other arm
transmits the image.
Take special care to ensure that a collision does not occur between the On-
Board Imager arms and the patient, operator, or surrounding equipment.
Such a collision can cause serious and possibly fatal injuries. Collisions
between On-Board Imager and the treatment couch or any other fixed
object can cause serious damage to both units, and serious or possible
fatal injury to anyone caught between them.
3.21.1. Precau-
tions
! Exercise extreme caution during any motion of On-Board Imager and
associated machine axes.
! Before taking kV setup fields, rotate the On-Board Imager and asso-
ciated machine axes to ensure that a collision will not occur during
kV beam delivery.
! Keep the patient under continuous observation whenever On-Board
Imager and other machine axes are moved. If a collision appears pos-
sible, stop motions immediately by pressing Emergency Off.
! Should the On-Board Imager arms collide with the treatment couch
or any other equipment, shut down the Clinac immediately by press-
ing Emergency Off. Do not operate the Clinac again until the machine
is fully checked by service personnel.
For details about On-Board Imager, refer to the On-Board Imager docu-
mentation.
3.22. High
Dose Precau-
tions
Depending on your Clinac software configuration, the Clinac allows you to
deliver high-dose stereotactic fixed and arc X-rays for single surgical treat-
ments and fractionated radiotherapy. It is important that you take appro-
priate precautions for delivering stereotactic treatments. When delivery
high-dose treatments, Varian recommends that you:
! Use appropriate collimation (such as paraffin blocks, or a Varian
MLC) to narrow the high-dose treatment field.
! Employ reliable patient safety and immobilization precautions for
stereotactic treatment:
Immobilize the patient as comfortably as possible, and briefly
explain the treatment procedure. This helps the patient to re-
main motionless on the treatment couch.
Remind the patient to stay on the treatment couch during
treatment until it is safe to leave. A patient should not leave the
treatment couch unless supervised by authorized personnel.
As with all treatments, perform a dry-run to check for potential
collisions before delivering treatment.
For detailed information about Stereotactic treatments, refer to the C-Se-
ries Clinac: Instructions for Use.
3.23. Falling
Parts or Acces-
sories
After prolonged use and its accompanying vibrations, the nuts, bolts, and
other fasteners on the MLC or accessory mounts can become loose and fall
from the machine. Similarly, wedge trays, shadow blocks, and other acces-
sories can fall from the treatment head, usually as a consequence of im-
proper installation.
A part or accessory falling on a patient, operator, or other person could re-
sult in serious injury.
3.23.1. Precau-
tions
Check all nuts, bolts, and other fasteners for tightness at least once every
six months.
! Take care to install accessories properly.
! Never install or remove accessories when a patient is on the treat-
ment couch. The only exception to this precaution is when an acces-
sory can be installed only after the field light establishes the treat-
ment area. In this case, use extreme caution.
! Do not exceed the weight capacity of the accessory mount when load-
ing shadow block trays.
3.24. Deteriora-
tion of Plastic
Parts from
Radiation
Some Clinac parts are made of plastic and can lose strength as a result of
age and prolonged exposure to radiation. Among these parts are the treat-
ment couch top panels, tennis racket strings, compensator trays, and
shadow block trays. Deterioration of plastic parts on the treatment couch
could cause a patient to fall. Deterioration of plastic parts attached to the
treatment head could allow an accessory to fall.
A patient fall from the treatment couch or an accessory falling on a patient,
operator, or other persons could result in severe injuries or death.
3.24.1. Precau-
tions
! Examine treatment couch panels monthly. Replace any panels that
are cracked or show any signs of degradation. Regardless of condi-
tion, replace all couch panels after every 1000 hours of beam opera-
tion or 5 years of use, whichever comes first.
! Inspect the strings on the tennis-racket-type panels periodically and
replace upon any sign of degradation.
! Examine all plastic parts periodically. Immediately replace any part
that is cracked, discolored, or shows any signs of degradation.
3.25. Patient
Fall from the
Treatment
Couch
Any of the following conditions could cause a patient to fall from the treat-
ment couch:
! Collision with the gantry
! Improper positioning or restraint of the patient
! Top panels weakened from radiation
! Degraded strings in the tennis-racket-type panels
! Sudden movements or stops of the treatment couch
! Overloading of the couch top
A patient fall from the treatment couch could result in severe injuries or
death.
3.25.1. Precau-
tions
! Lock the couch top into position before you place the patient on it.
! If you must rotate the couch top, do so before placing the patient on
the couch. After rotation, make sure that the top is latched securely.
Never apply weight to a couch top that is unlatched.
! Position the patient carefully on the treatment couch. Distribute the
patient's body uniformly over the couch top and secure with the re-
straining straps provided. For ETR and Exact Couches, the maxi-
mum safe patient weight is 400 lb. Other treatment couch weight
limits may be less; for maximum patient weight, refer to the docu-
mentation for your treatment couch.
! When the couch top is fully extended, do not exceed the maximum
weight of 300 lb. allowed on the end of the couch top. Do not allow
anyone to stand or sit on the end of a fully extended couch top.
! Avoid sudden movements or stops when adjusting the treatment
couch. Do not activate couch motions when brakes are released on
the couch top.
3.25.2. Gantry
Precautions
! Always keep the patient in sight when rotating the gantry. Stop the
rotation immediately if a collision appears possible.
! During patient setup, rotate the gantry from the hand pendant. Never
rotate the gantry from the console.
! Before beginning arc therapy, always perform a dry run and rotate
the gantry through its entire arc to make sure that it cannot collide
with the patient or equipment.
! Always keep the patient in sight when performing setup motions. For
remote motions, be sure that the camera angles provide adequate
views of the patient.
3.26. Treat-
ment Couch
Pinch Points
The lateral and longitudinal carriages and the couch top contain potential
pinch points during movement, particularly between the underside of the
couch top and the lateral carriage. Additional pinch points are exposed
when the protective bellows surrounding the lift mechanism must be low-
ered during maintenance procedures.
The couch carriages, couch top, and lift mechanism contain numerous
pinch points that could cause severe injury or death.
3.26.1. General
Precautions
When moving the couch's lateral and longitudinal carriages or the couch
top, make sure that all parts of your body are clear of the couch and that no
person or equipment can be caught in the moving mechanism.
3.26.2. Service
Precautions
If the bellows must be lowered for maintenance, install the safety braces
and remove all electrical power to the treatment couch before working in or
around the exposed lift mechanism.
Follow appropriate lockout/tagout procedures.
Be sure to keep hands and other body parts clear of potential pinch points.
3.27. High Tem-
perature Sur-
faces
Many Clinac components operate at high temperatures, including the
thyratron tubes in the modulator compartment of the drive stand, the
quartz-halogen lamps in the treatment head and their accessories, and the
transistor heat sinks in the power supplies.
Body contact with the surface of a component operating at a high temper-
ature can cause severe skin burns.
3.27.1. Precau-
tions
! Take care to prevent body contact with lamps, heat sinks, thyratron
tubes, or any other component that provides a sensation of warmth
upon approach.
! Tubes or porcelain components can be damaged by the oils and mois-
ture from your hand. When handling tubes or porcelain components,
wear gloves to prevent direct contact with skin.
! If you must handle a heat-producing component, allow a reasonable
cool-down period after turning off the power.
! Follow appropriate lockout/tagout procedures.
3.28. Laser
Beams
The optional localizer system uses up to five Class II laser beams to visually
locate the Clinac isocenter and to aid in setting up the patient. Each of
these beams is visible to the eye as a narrow line of bright red or green light.
WARNING: The radiation from one of these localizer la-
ser beams can cause permanent retinal
damage.
The optional LaserGuard collision detection system uses an invisible laser
beam shield to detect potential collisions between the gantry head and the
patient.
Note: LaserGuard is a Class 1 laser product. Exposure for any
length of time does not cause injury or damage to the eyes.
For more information about LaserGuard, see LaserGuard on page 1-21.
3.28.1. Precau-
tions
! Never stare directly into one of the visible green or red laser beams
from the localizer system. Advise patients of this precaution. The in-
visible LaserGuard laser beams are safe and do not cause retinal
damage.
! For additional warnings, refer to the documentation supplied by the
laser unit manufacturer.
3.29. Software
Integrity
Software and computer equipment included with the Clinac are installed
by Varian, and developed and tested exclusively for operation of the Clinac
system. This software remains the property of Varian and is licensed to the
user strictly for the purpose intended.
Modifying any of the software provided with the Clinac or using any other
software on the Clinac computer can seriously compromise the integrity of
data stored on the Clinac computer and can result in uncertain, unreliable,
and potentially hazardous system operation.
Any attempt to modify the Clinac computer, its operating system, or the
Clinac software, or to operate non-Clinac software on the Clinac computer
is considered by Varian to be a product alteration. This results in termina-
tion of the remainder of the warranty on a Clinac system.
To protect the hospital, its patients, and Varian from the potential conse-
quences of unauthorized modifications, software that has been altered
shall be removed by Varian.
3.29.1. Precau-
tions
! Do not modify the software supplied with the Clinac, including the
operating system and files placed on the fixed disk.
! Operation of the Clinac computer is restricted to Clinac software pro-
vided by Varian. No other software is allowed.
! Do not alter the configuration of the Clinac computer installed by
Varian personnel, including interior printed-circuit boards, peripher-
als, and switch settings.
3.30. Micro-
wave Tube
Operating Haz-
ards
Serious hazards exist in the operation of microwave tubes. Safe operating
practices require careful attention to hazards associated with microwave
tubes. Persons who work with microwave tubes or equipment that uses
them must protect themselves against serious injury. Careless operation of
microwave tubes can result in damage to tubes, the linear accelerator, or
other property; and potentially fatal injury.
The operation of microwave tubes involves one or more of the following haz-
ards:
3.30.1. Precau-
tions
Microwave tubes in the Clinac operate at voltages high enough to cause fa-
tal injury by electrical shock.
! Always break the primary circuits of the power supply and discharge
high-voltage circuits when direct access to the tube is required. Fol-
low appropriate lockout/tagout procedures.
! Keep a safe distance from the voltages encountered.
! Use grounded safety screens during tube operation.
3.30.2. Radio
Frequency
(RF) Radiation
Refer to Radio Frequency (RF) Radiation on page 1-15.
3.30.3. X-Ray
Radiation
Never operate high-voltage tubes without adequate X-ray shielding in
place.
3.30.4. Corro-
sive and Toxic
Compounds
External output waveguides, cathodes, and high-voltage bushings of mi-
crowave tubes are sometimes operated in systems that use dielectric gas to
impede microwave or high-voltage breakdown.
Release of dielectric gas in the presence of moisture and arcing can create
toxic and corrosive compounds. To prevent damage from dielectric gas
breakdown:
! When a leak in the dielectric system is detected or the waveguide is
disassembled for service, ventilate the area and avoid breathing any
fumes or touching any liquids that develop.
! Take precautions for highly toxic and corrosive substances before
permitting personnel to perform any work on or near the tube.
3.30.5. Hot
Water
Extreme heat occurs in the electron collector portion of microwave tubes
during operation.
Water channels used for cooling reach temperatures as high as the boiling
point of water (100 Celsius or above), and the hot water is under pressure
(typically as high as 100 psi). A rupture of the water channel or contact
with hot portions of the tube can scald or burn.
Emergency and Safety: Service and Maintenance Guidelines 1-27
Take precautions to prevent and avoid such rupture or contact.
3.30.6. Hot Sur-
faces
Portions of microwave and thyratron tube surfaces can reach extremely
high temperatures, especially the cathode insulator and cathode/heater
surfaces. All heated surfaces can remain hot for an extended time after the
tube is shut off.
To prevent serious burns, take care to avoid any bodily contact with these
surfaces, both during tube operation and for a reasonable cool-down period
after operation.
4. Service and
Maintenance
Guidelines
This section provides service and maintenance safety guidelines for the Cli-
nac and related systems.
Note: Lockout/tagout procedures are required for servicing or
maintaining the Clinac. Refer to OSHA regulation 29 CFR
1910.147, The Control of Hazardous Energy (Lockout/Tagout)
and 29CFR1910.333, Selection and Use of Work Practices and
other applicable state and federal occupational safety and
health regulations (domestic US), or other local (or interna-
tional) standards for additional information and requirements.
4.1. Clinac
Accelerator
Specifications
The following specifications must be observed for safe Clinac operation.
4.1.1. Electrical
Specifications
Electrical operating specifications:
! Type of protection against electric shock: Class I
! Degree of protection against electric shock: Type B
! Operation: The Clinac is classified as being suitable for continuous
connection to the supply main in the standby state and for specified
permissible loadings.
! The Clinac is not for use in the presence of flammable anesthetic mix-
tures.
! Electrical requirements:
Low Energy Clinac Series input voltage: 200 to 240 Vac 50 or
60 Hz 42 Amps max @208V; or 360 to 440 Vac 50 or 60 Hz 22
Amps max @ 400V.
High Energy Clinac Series input voltage: 200 to 240 Vac 50 or
60 Hz 125 Amps max @ 208V; or 360 to 440 Vac 50 or 60 Hz 65
Amps max @ 400V.
4.1.2. Environ-
mental Specifi-
cations
Environmental operating requirements:
! Humidity range: 15% to 80% relative humidity, non-condensing
! Temperature range: 60F to 80F (16C to 27C)
4.2. Lock-
out/Tagout
Procedures
Lockout/tagout is required when servicing or maintaining the Clinac or a
Clinac subsystem if unexpected startup or release of stored energy could
cause injury to personnel. All sources of hazardous energy to which per-
sonnel could potentially be exposed, including electrical, mechanical, and
1-28 Emergency and Safety: Service and Maintenance Guidelines
potential energy due to gravity or pressure, must be shut down and se-
cured by authorized personnel. Depending upon the work being performed,
it may be necessary to use multiple procedures to isolate all energy sourc-
es.
The rules and standards established in lockout/tagout procedures apply to
everyone associated with the maintenance and operation of the Clinac.
Failure to follow proper lockout/tagout procedures can result in serious
and possibly fatal injuries.
The lockout/tagout procedures in this section are organized according to
the type of energy source identified:
! Main Electrical Power
! Electrical Power to Subsystems
! Electrical Power to Facility Subsystems
! Gantry Locking Pin
! Patient Couch
! Air System
! Gas System
! Water System
4.2.1. General
Lock-
out/Tagout
Procedure
1. The general procedure for lockout/tagout is as follows:
2. Notify affected personnel that a lockout procedure is about to begin
and that the Clinac subsystem or facility power will be shut down
for service.
3. Locate all energy sources associated with the system or subsystem.
4. Operate the energy-isolating devices necessary to isolate the system
or subsystem.
5. Attach a lockout/tagout device to each energy-isolating device
involved. Typical lockout devices are shown in Figure 1.3.
6. Dissipate any stored energy.
7. Verify that the system or subsystem is isolated and deenergized.
WARNING: Do not remove, bypass, or ignore lockout/
tagout devices (Figure 1.3). Only the autho-
rized service person installing the lock-
out/tagout devices is permitted to remove
them. Unauthorized removal can result in
damage to equipment, and injury or death
to patients or personnel.
4.2.2. Releas-
ing Lockouts
This manual includes specific details of lockout/tagout procedures in the
section for each lockout/tagout procedure. General steps for releasing a
lockout are as follows:
1. Prepare the system or subsystem for operation. This includes
inspecting the system or subsystem to verify that all equipment
components are fully assembled and operational, that all safety
panels are in place, and that all tools and other nonessential items
have been retrieved.
2. Ensure that personnel are positioned safely.
3. Remove lockout devices and tags.
4. Notify affected personnel that the system or subsystem has been
released from lockout.
4.2.2.1. Releas-
ing Lockout
During Service
A lockout/tagout procedure is not required if service and maintenance per-
sonnel will not be exposed to the unexpected release of hazardous energy.
Some service or maintenance functions require equipment, or portions of
the equipment, to be turned on. Under these circumstances, the work must
be conducted with appropriate safeguards to prevent exposure to hazard-
ous energy sources.
Sometimes it may be necessary to remove lockout/tagout devices tempo-
rarily and turn on the machine or equipment for a limited time for testing
or positioning of equipment or components. Under such circumstances,
the work must be completed with appropriate safeguards to prevent expo-
sure to hazardous energy sources. As soon as possible, once the testing is
done, the authorized service personnel must again turn off the equipment
and resume lockout/tagout procedures.
Figure 1.3. Typical Lockout Devices
4.2.3. Main
Electrical
Power
When service is performed that requires removing main power to the Clinac
and where unexpected startup or release of stored energy could cause in-
jury to personnel, lockout/tagout the main electrical power to the Clinac.
WARNING: The activation of main electrical power to
the Clinac while someone is servicing the
Clinac can result in serious and possibly fa-
tal injury.
4.2.3.1. Lock-
out/Tagout
Main Power
To lockout/tagout the main power:
1. Notify affected personnel in the area that main electrical power to
the Clinac will be under lockout/tagout.
2. Place the Clinac in standby mode.
3. Turn off the main circuit breaker, switchbox disconnect, or other
energy isolating device that provides power to the Clinac.
4. Attach an appropriate lockout/tagout device and tag to the energy-
isolating device (Figure 1.4).
5. Verify that power has been removed.
4.2.3.2. Restor-
ing Main Elec-
trical Power
To restore main electrical power:
1. Check the Clinac and the immediate area around the machine to
ensure that nonessential items (such as tools or service equipment)
have been removed and that the machine is operationally intact.
Figure 1.4. Main Circuit Breaker Lockout/Tagout
4. Turn on the main power.
5. Notify affected personnel that service has been completed and that
the Clinac is ready for startup.
4.2.3.3. Electri-
cal Power to
Subsystems
When service is performed on a subsystem within the Clinac, where unex-
pected startup or release of stored energy could cause injury to personnel,
lockout/tagout the electrical power to the system at the applicable energy
source.
WARNING: The activation of electrical power to a sub-
system while someone is servicing the sub-
system may result in serious and possibly
fatal injury.
To lockout/tagout the power to a subsystem:
1. Notify affected personnel in the area that main electrical power to
the Clinac will be under lockout/tagout.
2. Identify and locate the source of electrical power to the individual
subsystem to be worked on.
Note: For detailed information, refer to the electrical sche-
matic diagrams in your product data book or service/systems
manual.
3. Turn off power to the subsystem by turning off the appropriate
energy-isolating device.
4. Attach appropriate lockout/tagout devices to the appropriate
energy-isolating devices.
5. Use shorting sticks to dissipate any remaining energy (see Using
Shorting Sticks on page 1-41).
4.2.4. Power to
Facility Sub-
systems
When service is performed on facility subsystems, where unexpected star-
tup or release of stored energy could cause injury to personnel, lockout/ta-
gout electrical power to the subsystem at the applicable energy source.
Typical facility subsystems include but are not limited to:
! Setup lights
! Laser positioning lights
! Closed-circuit television
! In-room monitors
WARNING: The activation of electrical power to a facil-
ity subsystem while someone is servicing
the Clinac or subsystem may result in seri-
ous and possibly fatal injury.
4.2.4.1. Cord-
and-Plug Sub-
systems
Lockout/tagout procedures do not apply to cord-and-plug subsystems, as
long as the cord to the subsystem is pulled and appropriate safety precau-
tions taken. Typical cord-and-plug subsystems include:
1. Clinac control console and display.
2. Multileaf collimator (MLC) console and display.
3. PortalVision console and display.
4. VARiS console and display.
5. Respiratory Gating, LaserGuard, and other additions to the Clinac.
4.2.4.1.1. Cord-
and-Plug Sub-
system Pre-
cautions
When servicing cord-and-plug subsystems, the following safety precau-
tions apply:
! Ensure that the electrical subsystem is unplugged.
! Plug must remain in the exclusive control of the person performing
service or maintenance work.
! If the plug is not under exclusive control, lock it out (with a plug cap)
and tag it.
To lockout/tagout the cord-and-plug subsystems:
1. Notify affected personnel that electrical power to the facility sub-
system will be under lockout/tagout.
2. Identify and locate the source of electrical power to the facility sub-
system to be worked on.
3. Turn off power to the subsystem by turning off the appropriate
energy-isolating device.
4. Attach appropriate lockout devices and tags to appropriate energy-
isolating devices.
5. Verify that power has been removed by measuring the absence of
voltage at the applicable subsystem.
To release the lockout on cord-and-plug subsystems:
1. Check the Clinac and facility subsystem to ensure that nonessen-
tial items (such as tools or service equipment) have been removed
and that the Clinac, Clinac subsystem, or facility is operationally
intact.
4. Turn on power to the facility subsystem.
5. Verify proper operation of the subsystem.
6. Notify personnel that service is complete and that the subsystem is
ready for startup.
4.2.5. Gantry You can prevent an unexpected gantry rotation during service by locking
out and tagging the gantry locking pin. Lockout/tagout is always required
when servicing the gantry drive system (for example, gantry motors, drive
chain, harmonic drive, and clutch).
With the gantry locking pin disengaged and the drive system components
intact, minor imbalances to the gantry due to the removal of fiberglass or
other components may result in slow gantry rotation. The speed of rotation
is directly related to the amount of the imbalance. If drive system compo-
nents are not intact (for example, if the drive chain is removed), gantry im-
balances may result in a more rapid rotation.
WARNING: A collision between the gantry and service
personnel could cause serious and possi-
bly fatal injuries. A collision between the
gantry and the treatment couch or any other
fixed object could cause serious damage to
both units, and serious or possible fatal in-
jury to anyone caught between them.
To lockout/tagout the gantry locking pin:
1. Notify affected personnel that the gantry will be under
lockout/tagout.
2. Ensure that the gantry rotation axis is unobstructed.
3. Rotate the gantry to one of the four locking positions (at 0, 90, 180,
and 270). Choose the angle that provides the most convenient
access for the maintenance or service to be performed.
4. Engage the locking pin.
5. Visually confirm that the gantry locking pin is engaged. While stay-
ing outside the arc of the gantry rotation, verify that the gantry is
locked in place by moving or rocking the gantry by hand, or by
operating the hand pendant.
Note: On low-energy Clinacs, the gantry locking pin can be
seen protruding approximately 1/4 in. to 1/2 in. past the inside
surface of the gantry sprocket hub, when engaged.
6. Attach appropriate lockout devices and tags to the gantry locking
pin (Figure 1.5).
To release the gantry lockout, see Releasing Lockouts on page 1-29.
4.2.6. Treat-
ment Couch
Lift
There are three types of Clinac treatment couches:
! PSA couch
! ETR couch
! Exact couch
When service is performed on the treatment couch, where unexpected en-
ergization, startup or release of stored energy could cause injury to person-
nel, block the couch with safety braces and tag it.
WARNING: Unexpected movement of the patient couch
can cause serious and possibly fatal injury.
Figure 1.5. Gantry Locking Pin with Lock and Tag
To lockout/tagout the treatment couch:
1. Notify personnel that the treatment couch will be blocked and
tagged.
2. For the PSA couch, raise the couch skirting. For the Exact or ETR
couch, lower the bellows.
3. Raise the couch sufficiently to allow insertion of the braces.
4. Insert the couch safety braces (Figure 1.6 and Figure 1.7).
5. Lower the couch until it is resting against the safety braces.
6. Attach lockout tags to the rear of the couch.
Figure 1.6. ETR or Exact Couch with Safety Braces
To release the lockout on the treatment couch:
1. Check the Clinac and the area around the machine to ensure that
nonessential items have been removed and that the machine is
operationally intact.
3. Remove the couch safety braces and lockout tags.
4. Notify affected personnel that service is complete and the patient
couch is ready for use.
4.2.7. Air Sys-
tem
When service is performed on the compressed air system (high-energy Cli-
nac models only), where the unexpected energization, startup, or release of
stored energy could cause injury to personnel, lockout/tagout the air sys-
tem.
CAUTION: The release of air pressure while someone is ser-
vicing the air system can result in injury.
Figure 1.7. PSA Couch with Safety Braces
1. To lockout/tagout the compressed air system:
2. Notify affected personnel that the air system will be under
lockout/tagout.
3. Turn off the air flowing to the machine.
4. Remove the valve handle or air hose (if present).
5. Attach an appropriate tag to the valve (Figure 1.8).
6. Release air pressure in the line.
7. Confirm that no air is flowing to the machine by checking that the
pressure gauge reads zero.
To remove the lockout:
ensure that nonessential items have been removed and that the
machine is operationally intact.
3. Remove the lockout tag.
4. Replace the valve handle or air hose.
5. Open the valve.
6. Verify that there are no air leaks.
7. Notify personnel that service is complete and that the system is
ready for use.
Figure 1.8. Air Hose Removed and Lockout Tag Attached
4.2.8. Gas Sys-
tem
When service is performed on the SF
6
or Freon 12 gas system, where the
unexpected energization, startup, or release of stored energy could cause
injury to personnel, lockout/tagout the gas system.
CAUTION: The release of gas pressure while someone is ser-
vicing the gas system can result in injury. Varian recom-
mends the use of an electronic sniffer device to detect gas
leaks.
To lockout/tagout the gas system:
1. Notify affected personnel that the gas system will be under
lockout/tagout.
2. Turn off the gas flowing to the machine.
3. Remove the valve handle (Figure 1.9).
4. Attach an appropriate lockout tag to the valve.
5. Release any stored gas pressure or other energy in the line.
6. Confirm that there is no gas flowing to the machine.
Figure 1.9. Handle Removed and Lockout Tag Attached
3. Remove the lockout tag.
4. Replace the valve handle.
5. Open the valve.
6. Verify that there are no gas leaks.
7. Notify personnel that service is complete and that the system is
ready for use.
4.2.9. Water
System
When service is performed on the water system, where the unexpected en-
ergization, startup, or release of stored energy could cause injury to per-
sonnel, lockout/tagout the water system.
WARNING: Release of water and water pressure on ex-
posed electrical components while some-
one is servicing the water system may re-
sult in serious and possibly fatal injury.
To lockout/tagout the water system:
1. Notify affected personnel that the water system will be under
lockout/tagout.
2. Turn off the water flowing to the machine.
3. Turn off the water pump.
4. Lockout/tagout the appropriate pump circuit breakers.
5. Confirm that there is no water flowing to the machine.
6. Remove the valve handle.
7. Attach an appropriate lockout tag to the valve (Figure 1.10).
8. When opening water system lines, use caution to prevent water
from dripping onto, or spraying into, electronic components.
3. Remove the lockout devices and tags.
4. Replace the valve handle, if applicable.
5. Open the valve, replace the locking pin, or otherwise restore equip-
ment to normal operating conditions.
6. If applicable, restore power to the system. For example, turn on the
appropriate pump circuit breaker to restart the water pump.
7. Verify that there are no leaks or other unexpected hazards.
8. Notify affected personnel that service is complete and that the sys-
tem is ready for use.
4.2.10. Shift
Changes
If maintenance on a Clinac extends beyond one shift, or if there is a person-
nel change, all additional personnel are required to place their locks on the
lockout device before they begin work on the equipment.
If maintenance on a Clinac requires multiple service personnel, all are re-
quired to place their locks on the lockout device.
4.2.11. Work-
ing with Con-
tractors
When outside personnel (contractors or others) will be working on the Cli-
nac, the on-site employer or owner and the outside contractors must in-
form each other of the lockout/tagout procedures they will use.
Figure 1.10. Water Valve with Handle Removed and Tag Attached
Authorized personnel responsible for performing the lockout must ensure
that any affected personnel understand and comply with appropriate lock-
out procedures in accordance with this manual and applicable regulations.
Coordinate lockout activities with all involved parties.
4.2.12. Using
Shorting
Sticks
When working on the modulator, gun deck, or VacIon subsystems, use
shorting sticks to dissipate any residual stored energy, and leave the short-
ing sticks in place to prevent the reaccumulation of charge.
Note: Varian recommends that you use more than one short-
ing stick to prevent the inadvertent loss of ground.
Verify that power has been removed by measuring the absence of voltage at
the subsystem.
To use shorting sticks:
have been removed and that the Clinac, Clinac subsystem, or facil-
ity is operationally intact.
3. Remove the shorting sticks (if in use), lockout devices, and tags.
4. Turn on power to the subsystem.
5. Verify proper operation of the subsystem.
4.2.13. Chemi-
cal Hazards
The Clinac contains substances that can be hazardous to your health. In-
formation located on container labels and material safety data sheets (MS-
DSs) are provided for your protection and safety. Always read the informa-
tion contained on labels and MSDSs provided by the manufacturer of haz-
ardous substances.
Failure to understand and identify the presence of hazardous substances
in the Clinac and its accessory products can result in serious injury.
This information is provided as general guidance only. Refer to OSHA reg-
ulation 29 CFR 1910.1000 and other applicable state and federal occupa-
tional safety and health regulations (domestic US), or other local (or inter-
national) standards for additional information and requirements.
The guidelines presented in the next section, Communication Guidelines
for Hazardous Chemicals, apply to all Clinac service personnel.
4.2.13.1. Com-
munication
Guidelines for
Hazardous
Chemicals
The owner must develop and maintain a written hazard communication
program. This program must accomplish the following:
! Inform employees about hazard-communication standards.
! Explain implementation of the program as it applies to their specific
workplace.
! Inform and train employees how to recognize, understand, and use
the labels and MSDSs provided by vendors or otherwise provided
with the Clinac.
! Inform and train employees about safety procedures for working with
hazardous substances.
4.2.13.2. Em-
ployee Re-
sponsibilities
Operators and maintenance and service personnel who come into contact
with the Clinac are required to read the labels and the MSDSs provided
with the Clinac. They are also required to follow pertinent instructions and
warnings.
4.2.14. Material
Safety Data
Sheets
(MSDSs)
Material safety data sheets provided with your Clinac contain the following
information:
! Identification of the substance, including the manufacturers name,
address, and emergency phone number
! Hazardous components, including the chemical ID, common names,
and worker exposure limits
! Physical and chemical characteristics, including boiling point, vapor
pressure, vapor density, melting point, evaporation rate, water solu-
bility, appearance, and odor under normal conditions
! Physical hazards, including safe handling methods
! Reactivity, indicating whether the substance is stable or not
! Health hazards, including information on: how the chemical could
enter the body, for instance through inhalation, skin contact, or by
swallowing; whether or not the chemical is considered a carcinogen;
signs and symptoms of exposure; and existing medical conditions
that could be aggravated by exposure
! Precautions for safe handling and use, including information on:
what to do if the substance spills or leaks, disposal of the substance,
equipment and procedures needed for cleaning up spills and leaks,
how to handle the substance properly, storage, and any other pre-
cautions
4.2.14.1. Con-
tainer Labels
Containers of hazardous chemicals are labeled with the following informa-
tion:
! Chemical name
! Manufacturer or importer information, including name, address, and
emergency contacts
! Physical hazards
! Health hazards
! Personal protective clothing, equipment, and procedures recom-
mended
! Storing or other special handling instructions
4.2.14.2. Haz-
ardous Sub-
stances in the
Clinac
Depending upon the Clinac model, the following hazardous substances
may be present:
1. Freon 12 is present in the waveguide system in Clinacs 4/100 (prior
to S/N 81), 6/100 (prior to S/N 457), and 600C (prior to S/N 66).
2. Sulfur hexafluoride (SF
6
) is present in the waveguide systems of all
Clinacs other than Clinacs 4/100 (prior to S/N 81), 6/100 (prior to
S/N 457), and 600C (prior to S/N 66).
3. Lead is present in the shielding and balance weights in all Clinacs.
4. Beryllium (99% pure) is present in the window of the bend magnet
vacuum chamber of all high-energy Clinacs.
5. Dielectric insulating oil is present in the pulse transformer in all
Clinacs.
6. Dielektrol III capacitor oil is present in the modulator, rectifiers, and
capacitors in all Clinacs.
7. Depleted uranium is present in the X-ray head of older low energy
Clinacs.
4.2.14.3. Other
Substances in
the Clinac
Depending upon the Clinac model, other substances, which may pose neg-
ligible health risks under normal operating conditions, may also be
present. These substances may include:
! Ethylene glycol
! Lubricants
! Paint (touch-up)
! Paint (black, aerosol)
! Asbestos
4.2.15. Pre-
venting Expo-
sure to Blood-
borne
Infections
Some bloodborne infections are caused by viruses carried in the blood-
stream. These viruses are usually transmitted by contact with infected
blood, although in some cases they may be transmitted by other bodily flu-
ids as well.
Individuals infected with bloodborne pathogens, such as the hepatitis B vi-
rus (HBV) or the human immunodeficiency virus (HIV), can develop serious
health problems that could ultimately prove fatal.
This information is provided for general guidance only. Refer to OSHA reg-
ulation 29 CFR 1910.1030, Occupational Exposure to Bloodborne Patho-
gens and other applicable state and federal occupational and health regu-
lations (domestic US), or other local (or international) standards, for addi-
tional information and requirements.
The guidelines presented in this section apply to all Clinac operators and
maintenance and service personnel.
Occupational situations that may present a risk of exposure to bloodborne
pathogens include:
! Periodic maintenance inspections (PMIs).
! Routine maintenance.
! Cleaning of the treatment couch and the collimator.
! Patient contact.
Since the Clinac and its accessories are located in a hospital environment,
the following occupational situations may apply:
! Service-related duties that bring you into contact with human blood
or other bodily fluids
! Service-related duties that bring you into contact with hypodermic
needles or other sharp instruments that might be contaminated with
infected blood
! Service-related duties that bring you into contact with materials
(towels, sheets, clothing) contaminated with blood or other bodily flu-
ids
! Potential occupational exposure to bloodborne pathogens
! Providing emergency first-aid assistance to coworkers
! Dealing with blood, unfixed tissue, or organs from humans (living or
dead); tissue cultures; or solutions that contain blood or other hu-
man tissue
! Handling experimental animals infected with HIV or HBV, or the
blood, organs, or tissues from these animals
4.2.15.1. Safety
Precautions
! Assume all bodily fluids are infectious and act accordingly. Avoid
contact with blood or other bodily fluids. Keep all cuts and scrapes
covered.
! Follow proper procedures when in contact with infected blood or oth-
er bodily fluids or contaminated materials. Avoid procedural short-
cuts, which may save you time, but place you at risk of becoming ex-
posed to potential bloodborne pathogens.
! Follow proper housekeeping procedures. Recommended housekeep-
ing guidelines include:
Clean and decontaminate all equipment and work surfaces
that have been contaminated with blood or other potentially in-
fectious materials. Use the disinfectant required by the employ-
er (or, if none is specified, with a solution of 5.25% sodium hy-
pochlorite [household bleach] diluted between 1:10 and 1:100
with water).
Remove and replace protective coverings, such as plastic and
foil, that may have become contaminated.
Always use mechanical means, such as tongs, forceps, or a
brush and dust pan, to pick up contaminated broken glass-
ware. Never use your hands to pick up broken glassware even
when wearing gloves.
Place all potentially contaminated wastes in designated dispos-
al containers.
Store sharp objects in a manner that ensures safe handling.
! Dispose of waste properly. Recommended waste disposal procedures
include:
Place all infectious waste in closable, leakproof containers or
bags that are color-coded and appropriately marked.
Place all sharp items in puncture-resistant containers for dis-
posal.
Double-bag infectious wastes if the outside of a bag is contam-
inated with blood or other potentially infectious materials.
Inspect and decontaminate bins, pails, and cans used for stor-
age or disposal of potentially infectious materials on a regular
basis.
! Observe proper hygiene. Infected matter can enter the bloodstream
through cuts and other openings in the skin, through the eyes, or
through the mucous membranes of the nose and mouth.
It is especially important to wash hands frequently, since they
are the most likely source of contact with infected blood or oth-
er bodily fluids. Wash carefully after any contact with body flu-
ids or other potentially infectious materials and after removing
gloves that may have been contaminated.
If blood splatters in the eyes, flush thoroughly with clean water
as soon as possible after the contact.
Wear appropriate personal protective equipment when the job
involves contact with blood or potentially infected material, or
when giving first aid.
4.2.16. Using
Personal Pro-
tective Equip-
ment
Personal protective equipment (PPE) prevents blood or other potentially in-
fectious material from coming into contact with the skin, eyes, mouth, mu-
cous membranes, or clothing.
This section details procedures for using PPE to protect yourself from pos-
sible hazards. Wear the appropriate PPE for the task you are performing.
This information is provided for general guidance only. Refer to OSHA reg-
ulation 29 CFR 1910.132-1910.139 and other applicable state and federal
occupational and health regulations (domestic US), or other local (or inter-
national) standards, for additional information and requirements.
4.2.16.1. Safety
and Service
Guidelines
These guidelines for using personal protective equipment apply to all Cli-
nac operators and maintenance and service personnel:
! Obtain PPE that fits well.
! Inspect and clean PPE before and after each use.
! Use the correct PPE for the situation. Depending on the hazards of
the job, you may need more than one type of PPE to protect yourself.
4.2.16.2. Eye
Protection
Safety glasses or goggles must be worn while servicing the Clinac where eye
hazards exist.
! Wear safety glasses fitted with side shields to provide maximum pro-
tection from flying particles or dust.
! Keep protective eyewear clean.
! Contact lenses are not a substitute for protective eyewear.
4.2.16.3. Foot
Protection
Shoes that provide protection for the whole foot, such as work shoes, are
recommended during service, maintenance, and installation of the Clinac.
For electrical service on equipment involving potentials above 20 kV, wear
safety-toe boots. Do not wear steel-toe boots.
4.2.16.4. Res-
piratory Pro-
tection
Masks should be worn whenever there is a risk of exposure to airborne
substances, such as dust or chemical vapor. The mask should cover the
nose and mouth.
4.2.16.5. Hand
Safety
Wear protective gloves while servicing the Clinac where hazards to your
hands exist, and follow these guidelines:
! Treat used waveguide pieces as though they are contaminated and
handle with personal protective equipment (PPE). Always wear eye
protection and chemical-resistant gloves. For information about PPE,
see Using Personal Protective Equipment on page 1-45.
! Leather gloves used for handling lead must be set aside and not used
for other purposes. With prolonged use, leather gloves can absorb
lead. Always wear disposable latex, vinyl, or nitrile gloves under
leather gloves that have been used to handle lead.
! Before servicing the Clinac, remove jewelry such as rings, bracelets,
and watches.
! Stay clear of pinch points and crushing hazards including, for exam-
ple, motors and gears in the collimator, gantry, and couch. Lock-
out/tagout the couch and gantry locking pin, when appropriate, to
prevent hand injuries.
! If you anticipate coming into contact with blood or other potentially
infectious materials, wear latex, vinyl, or nitrile gloves. Check items
for slivers, jagged edges, or burrs before lifting.
! Always wash your hands immediately after removing gloves.
! Dispose of single-use gloves in the proper containers. If gloves are
contaminated, do not dispose of them in regular trash.
Table 4-1 lists recommended types of gloves.
Note: For those with rubber or latex allergies, use vinyl or
nitrile gloves.
Table 1.1. Hand Protection
Operation Type of Glove Protect Against Notes
Handling lead Leather gloves with
latex, vinyl, or nitrile
gloves underneath
Lead particulates or
dust
Latex, vinyl, or nitrile
gloves prevent contamina-
tion from leaching of lead
through dedicated leather
gloves
Handling
depleted ura-
nium
Leather Low-level radiation
and uranium poison-
ing
Handling ion
chamber
Latex, vinyl, or nitrile Contaminated dust,
low-level radiation,
and moderate heat
Handling hot
thyratron
tubes
Leather Moderate heat
Handling
sharp or
rough objects
Leather Cuts and abrasions
Contact with
blood or other
bodily fluids
Latex, vinyl, or nitrile Bloodborne patho-
gens
Cleaning
couch or colli-
mator
Latex, vinyl, or nitrile Bloodborne patho-
gens
Handling
waveguide
parts
chemical-resistant
gloves, such as rub-
ber or neoprene
Chemical hazards Also wear eye protection
Handling
chemicals
Rubber, neoprene Chemical hazards
4.2.17. Heavy
Lifting and
Handling
There are a number of situations that may require lifting or handling heavy
objects during the service and maintenance of the Clinac.
Lifting and handling heavy equipment or objects improperly can cause per-
sonal injuries, including sprains, strains, fractures, wounds, and hernias.
4.2.17.1. Guide
lines for Heavy
Lifting
General guidelines for lifting and moving heavy equipment or objects safely
include:
! Dress for safety. Shoes with reinforced toes and nonslip soles may
help to reduce the risk of injuries.
! Think ahead. Plan a clear and unobstructed route to your destina-
tion.
! Examine the object. Judge its weight and stability. Look for sharp
edges. Decide how best to hold the object.
! Get a good grip. Use your palms and fingers. Wear only properly fit-
ting gloves.
! Get help if you need it. For objects over 20 kg (44 lb.), use suitable
handling devices and techniques. If you have any doubt about your
ability to move the object, ask for help or use a mechanical aid.
Practice proper lifting techniques to safely protect your back and spine:
! Stand close to the load with feet wide apart.
! Squat, bending at the hips and knees.
! As you grip the load, curve your lower back inward by pulling your
shoulders back and pushing your chest out.
! Be sure to keep the load close to your body.
! When you set the load down, squat, bending at the hips and knees.
Keep your lower back curved inward.
If you are unable to bend your knees easily or get very close to an object,
use alternative lifting techniques:
! Stand as close as you can.
! Brace your knees against a solid object.
! Bend at the hips, keeping your head and back straight.
! Lift slowly, using your legs, buttocks, and stomach muscles.
4.2.17.2. Pre-
cautions
When lifting or handling heavy objects (those weighing over 20 kg, or 44
lb.), obtain or use equipment appropriate for the task at hand. Also refer to
information provided during specific product safety training. Situations
that require lifting or handling heavy objects include, but are not limited to,
the following:
! Lifting or handling the turntable or collimator: Use appropriate A-
frame handling equipment.
! Lifting or handling the sled assembly: Use appropriate bracket-fix-
ture handling equipment.
! Lifting or handling the Klystron or Klystron solenoid: Use appropriate
handling equipment.
! Lifting or handling gas cylinders (high-energy Clinacs), e-rack power
supplies, RF drives, or fiberglass covers (canoes): Request assistance
when necessary.
1-48 Emergency and Safety: Owner Guidelines
4.2.18. Trip-
ping and Other
Hazards (Sil-
houette Edition
Clinac)
Tripping hazards exist when working in either service bay of the cabinet of
a Silhouette Edition Clinac. The RF side has a step-up plate so that the
hazard is decreased, but the modulator side has a stand-alone kick plate
where special care must be taken. Ignoring the size and location of this
plate can cause someone to trip and cause bodily injury.
Service personnel could bump their heads on the portion of the waveguide
that crosses the RF cabinet service bay. Before working in this area, Varian
recommends that you first remove this portion of the waveguide, and then
reinstall it when the work is completed.
5. Owner
Guidelines
This section provides guidelines for establishing emergency and safety pro-
cedures for Clinac operation and maintenance.
5.1. Planning
Operations
To prepare for the safety of patients and staff in the treatment room and
surrounding areas, the owner is responsible for installation planning and
installation of certain emergency and safety equipment.
5.1.1. Emer-
gency-Off But-
tons
Provide sufficient Emergency Off buttons in the treatment room and con-
sole area. These buttons should complement the Emergency Off buttons
supplied by Varian on the dedicated keyboard, on both sides of the treat-
ment couch, and on both sides of the drive stand. For additional informa-
tion about the location of Emergency Off buttons, refer to National Council
on Radiation Protection and Measurements (NCRP) Report Number 51.
5.1.2. Main Cir-
cuit Breaker
Install the main facility circuit breaker within 10 feet (3 meters) of the con-
trol console. The main circuit breaker for the Clinac should have an under-
voltage trip that causes the main breaker to trip if an Emergency Off button
is pressed.
5.1.3. Ventila-
tion and Tem-
perature Regu-
lation
Maintain adequate room ventilation. Heat and air conditioning should be
available to maintain the Clinac at room temperature (6570 F).
5.1.4. Fire
Extinguishers
Provide suitable fire extinguishers in the treatment room and near the con-
trol console. In the United States, the type of extinguisher must be ap-
proved for electrical fires by federal, state, and local codes and regulations.
5.1.5. Radio
Frequency-
Emitting
Equipment
Ensure that nearby rooms with RF-emitting equipment have proper room
shielding to prevent leakage and door interlock switches to prevent opera-
tion with doors open. Adopt operating procedures that minimize leakage
potential.
5.1.6. Emer-
gency Lighting
Provide automatic emergency lighting (with flashlights as backup) in the
treatment room and console area.
5.2. Radiation
Protection Sur-
vey
Before beam calibration and routine use of the Clinac, the owner must have
a radiation protection survey completed by a qualified radiation expert.
Emergency and Safety: Owner Guidelines 1-49
In the United States, the radiation survey report indicates if the installation
meets the recommended standards of the NCRP and the applicable local,
state, and federal regulations.
Outside the United States, the owner is responsible for compliance with the
applicable statutory and regulatory requirements.
Before routine use of the Clinac, the owner must:
! Have a qualified radiological physicist calibrate the dose rate and in-
tegrated dose measured by the transmission ionization chamber.
! Conduct checks at least daily during the first month of operation to
establish that the ionization chamber response is constant within
specified limits.
! Make constancy checks during the course of each day to compare
monitor response from the start to the end of the working day.
! Conduct daily, or at least weekly, calibration checks after the con-
stant output of the machine is established. Record all calibration
measurements in a log.
5.3. Safety and
Emergency
Training
Personnel who work with or near the Clinac must receive formal training on
the emergency and safety procedures adopted by the owner. Each person
should receive further training on a periodic basis. Training should include
at least the following items:
! Location and use of Emergency Off buttons
! Location and use of the main facility circuit breaker
! When to use lockout/tagout procedures as required by OSHA 29 CFR
1910.147 and/or other applicable state and federal occupational
safety and health regulations (domestic US) or other local (or interna-
tional) standards
! Local evacuation procedures for fire, smoke, or chemical fumes
! Location and use of the emergency lighting system (including backup
lighting such as flashlights)
! Procedure to remove a patient from the treatment couch in an emer-
gency
5.4. Routine
Use
The owner must establish operational and maintenance safety procedures
for routine use of the Clinac. Combine the following items with the other
safety precautions described in this section to reduce the likelihood of in-
jury to personnel and damage to the equipment.
5.4.1. Address-
ing Equipment
Malfunctions
Cease machine operation immediately if any equipment malfunction is de-
tected or suspected, and call service personnel to correct the problem.
5.4.2. Record-
ing Observa-
tions
Record any unusual machine behavior or other observations in the log
book when you perform the daily checkout procedure and throughout the
treatment day.
5.4.3. Securing
Keys
When the machine is placed in standby mode and left unattended, remove
the DISABLE/ENABLE and POWER keys from the equipment and deposit
1-50 Emergency and Safety: Owner Guidelines
them in a key storage enclosure. Lock the enclosure to prevent unautho-
rized activation of the machine.
5.4.4. Testing
Emergency-Off
Circuits
Test the emergency-off circuits at least once every three months to ensure
proper functioning.
5.4.5. Checking
Fasteners
Check all fasteners for tightness at least semiannually.
5.4.6. Posting
Signs
Post signs on doors to the treatment room and in the console area to in-
form:
! All persons that a radiation hazard exists in the area.
! Personnel to wear radiation monitoring instrumentation when they
enter the treatment room.
! Operators to have the DISABLE/ENABLE key in their possession
when they enter the treatment room and to make sure that the door
cannot close while in the room.
! Any person wearing a pacemaker to remain out of the area until the
effect of radiation and radio-frequency interference on pacemakers is
known.
5.5. Quality
Assurance
Because of the importance of precisely administered treatments in radia-
tion therapy, the owner should establish a comprehensive quality assur-
ance (QA) program for each radiation therapy facility. Sources of errors in
radiation therapy include:
! Tumor localization
! Patient immobilization
! Field placement
! Human errors in calibration
! Patient setup
! Equipment use
A program of periodic checks can minimize many of these errors.
References:
AAPM code of practice for radiotherapy accelerators: Report of AAPM Radi-
ation Therapy Committee Task Group, Medical Physics, Vol. 21, No. 7, July
1994
Comprehensive Quality Assurance for radiation oncology: Report of AAPM
Radiation Therapy Committee Task Group, Medical Physics, Vol. 21, No. 4,
April 1994
Physical Aspects of Quality Assurance in Radiation Therapy, AAPM Report
No. 13, American Association of Physicists in Medicine, May 1984
5.6. Accidental
Radiation
Overdose
The owner must establish the procedures to follow in case of an accidental
overexposure of a patient or personnel to radiation. Post these procedures
conspicuously in the console area.
Emergency and Safety: Owner Guidelines 1-51
In the event that a person is thought to be exposed to an excess of radia-
tion, the law in most localities of the United States requires the following
steps:
! Immediately notify the appropriate local, state, and federal authori-
ties.
! Request an investigation by a professional qualified in the detection
of radiation.
! Consult with medical experts in radiation treatment.
! Check the following references:
National Council on Radiation Protection and Measurement
(NCRP) Reports Number 38 and 102
Appropriate state regulations
5.7. Backup
Interlocks
The Clinac is designed to terminate the exposure when the total dose dis-
played on the console monitor equals the set dose. Should this normal
(DOS1) termination fail to occur, the following backup interlocks terminate
the beam:
! When the total dose exceeds the dose set by the operator (DSFA in-
terlock).
! At a fixed percentage of monitor units beyond the value set by the op-
erator or a fixed number of monitor units, whichever is less (DOS2 in-
terlock).
! When the numbers of monitor units (MU) from the primary and sec-
ondary dosimetry channels differ from each other by more than 5% or
2 MU, whichever is greater (DS12 interlock).
! Upon coincidence between the displayed time and the time value set
by the operator (TIME interlock).
! In dynamic therapy, when there is a disagreement between the in-
tended dose and position and the actual dose and position (DPSN in-
terlock).
5.8. Emer-
gency Beam
Termination
The operator must be aware of the progress of treatment at all times. If the
beam does not terminate correctly, the operator should take immediate ac-
tion (press the nearest Emergency Off button) and not wait for a backup
system to activate.
Beam termination by an interlock or means other than normal termination
may be a sign of a significant equipment malfunction. Varian recommends
that you suspend patient treatments after any abnormal termination until
qualified personnel determine that it is safe to continue using the Clinac.
5.9. Emer-
gency Plan
An emergency situation can arise at any time. The owner must establish
procedures for handling emergencies. Personnel operating or working
around the Clinac must be trained in these procedures.
! The emergency plan should include:
! Emergency procedures (modified for local conditions) described in
this manual
! Periodic testing of emergency equipment, including the emergency
pendant system, emergency lighting system, fire extinguishers, and
emergency-off circuits
1-52 Emergency and Safety: Appendix A: Venting Waveguide Gases
! Evacuation routes in case of an emergency situation, with the routes
posted near the control console
! Scheduling and content of periodic drills and training
! Identification of qualified personnel to be contacted in the event of a
fire, medical emergency, or other situation requiring outside help
! Procedures to restore operation of the Clinac following an emergency
! Procedures for using a hazardous spill kit
! Procedures for maintaining up-to-date copies of Clinac MSDSs in the
event of a hazardous substance leak or spill
6. Appendix A:
Venting
Waveguide
Gases
All Clinacs use either Freon 12 or SF
6
as a dielectric gas in the RF
waveguide system. Waveguide arcing causes Freon 12 to break down into
hydrogen chloride gas as well as hydrochloric and hydrofluoric acids. With
SF
6
, waveguide arcing produces hydrofluoric acids and other hazardous
fluoride products. These gases could be toxic and should not be inhaled.
The procedures for venting waveguide gases will reduce exposure to haz-
ardous gases that can be present when the RF waveguide system is purged.
Abnormal noises during BEAM ON often indicate RF arcing in the
waveguide. Specifically, listen for:
! Distinct or repetitive metallic dinging noise.
! Continuous racket or banging noises other than the normal high-volt
pulsing sound.
If venting the system is needed, follow the procedures. Varian recommends
reading the entire procedure (for the correct machine configuration) before
beginning work.
Varian can provide any of the service contained in this procedure. Call the
Regional Varian Customer Support Office (see Customer Support on
page 1-8) for details.
6.1. Testing for
Waveguide
Arcing
Use an oscilloscope to confirm if the RF is arcing in the waveguide. Deter-
mine if the following waveforms are tearing-off or have abnormally high
spikes.
! For Low Energy machines, observe Magnetron I, Forward Power, and
Reflected Power.
! For High Energy machines, observe Klystron I, Forward Power, Load
Power 1, Load Power 2, and Reflected power.
6.2. Prerequi-
sites: Parts
Ordering Pro-
cedure
If venting the Clinac is needed, some parts are required. Determine if there
is a venting device on the Clinac.
! All high energy Clinacs have a built-in venting device at the stand.
! All low energy Clinacs that use SF
6
also have the venting device at
the stand.
! All low energy Clinacs that use Freon 12 do not have the venting de-
vice at the stand.
If the Clinac has a venting device:
! Order the 872031-01 kit from Varian and perform the section Vent-
ing a Clinac With a Venting System on page 1-53.
Emergency and Safety: Appendix A: Venting Waveguide Gases 1-53
! A plastic bag may be used instead of the Cubitainer (listed in the ma-
terial list at the end of this procedure).
If the Clinac does not have the venting device and retrofit is not being per-
formed:
! Neither kit needs to be ordered. Complete the section Venting a Cli-
nac Without a Venting System on page 1-53.
! For this procedure, provide a large plastic bag to contain the gas, and
a rubber band to close the bag. A 30 50-gallon trash bag is ideal.
To retrofit the Clinac with the venting device, order the 872031-02 kit and
complete the section Retrofitting the Clinac With a New Venting System
on page 1-55.
6.3. Venting a
Clinac With a
Venting Sys-
tem
Clinacs with a venting device can be vented in two ways:
If the room has an exhaust vent, use flexible tubing to route the discharged
gas directly into the room exhaust vent.
1. If there is no exhaust vent in the room:
2. Order the 872031-01 kit to properly vent the gas.
3. Capture the gas into the plastic Cubitainer or a 3050 gallon trash
bag.
4. Be sure to close the bag securely with adhesive tape or a rubber
band.
5. Carry the bag to an exhaust vent or an unpopulated area outdoors,
and open it.
WARNING: For your safety and the safety of others, be
sure to discharge the gas away from people
or animals.
6. Discharge the container through at least 15 feet of flexible tubing to
allow a safe upwind breathing zone. Allow 510 minutes to dis-
charge all the gas.
6.4. Venting a
Clinac Without
a Venting Sys-
tem
WARNING: Read this entire section before performing
the steps.
Before you begin:
1. Decide where the gas will be vented: Into the ventilation system, or
outdoors away from people or animals.
2. Shut off the gas bottle on the Clinac.
6.4.1. Option 1
for Venting
Gas
This option may be the easiest since the relief can be used to control the
gas flow.
1. Rotate the gantry to either 90 or 270.
2. Remove the top gantry cover.
3. Locate the gas pressure relief valve at the RF system.
4. Wrap the opening of the plastic bag over the relief valve so that the
venting gases will fill the bag without escaping into the treatment
room; cover the valve handle as well.
5. Use one hand to secure the bag over the relief valve and the other
hand to carefully pull the valve handle.
6. Holding the bag shut, carefully remove it from the relief valve.
7. Twist the bag shut and seal it with the rubber band or adhesive
tape.
8. Carry the bag to an exhaust vent or an unpopulated area outdoors.
WARNING: For your safety and the safety of others, be
sure to discharge the gas away from other
people or animals.
9. Discharge the container through at least 15 feel of flexible tubing to
allow safe, upwind breathing zone.
10. Allow 510 minutes to discharge all the gas.
11. If the gas venting system is being added to the Clinac, see Retrofit-
ting the Clinac With a New Venting System on page 1-55.
6.4.2. Option 2
for Venting
Gas
1. Use one of the hose connections on the stand instead of the relief
valve.
2. Determine which hose connection will be used to vent the gas. The
connection chosen should allow enough room to comfortably hold
the bag on the hose as well as allow for inflation of the bag.
3. Test the hose connection selected:
4. Try to loosen the hose connection slightly and immediately retighten
it. This test helps ensure that full control over the hose connection
can be turned on and off when venting the gas into the bag.
5. Wrap the opening of the plastic bag over the hose so that the vent-
ing gases fill the bag without escaping into the treatment room.
6. Discharge the container through at least 15 feel of flexible tubing to
allow safe, upwind breathing zone.
7. Allow 510 minutes to discharge all the gas.
Emergency and Safety: Appendix A: Venting Waveguide Gases 1-55
6.5. Retrofit-
ting the Clinac
With a New
Venting Sys-
tem
1. Order the 872031-02 kit.
2. Complete the venting procedure in Venting a Clinac Without a
Venting System on page 1-53.
3. Assemble the mounting bracket, clamps, and venting manifold as
shown in drawing 1101708.
4. Find the filter trap at the pressure regulator in the stand.
5. On the Clinac, disconnect the yellow hose at the outlet of the filter
trap that routes to the gantry.
6. Connect the yellow hose to one end of the venting manifold.
7. Use the new yellow hose (supplied with the kit) to connect the filter
trap outlet to the open end of the venting manifold.
8. Mount the manifold assembly next to the pressure regulator.
6.5.1. Inspect-
ing the New
Venting Device
for Leaks
1. Slowly refill the gas system to operating pressure, 39 psi, for Cli-
nacs with 3-port circulators, and 32 psi for Clinacs with 4-port
circulators.
2. Perform a bubble test on the gas lines by applying the provided
Snoop on all the line joints and fittings.
To seal a leak, tighten the joints and connections or apply new Teflon tape.
6.5.2. Verify the
Gas Leak Rate
1. Set the gas pressure to operating pressure, 39 psi, for Clinacs with
3-port circulator, or 21 psi for Clinacs with a 4-port circulator.
2. Allow approximately ten minutes for the gas pressure to stabilize
and warm up.
3. Close the gas bottle main valve.
4. Turn the gas pressure regulator all the way down.
5. Write down the regulated pressure indicated on the regulator gauge.
6. Observe the pressure and compare it with the previous days
pressure.
If the leak rate exceeds 2 psi per day, call the regional Varian office for ser-
vice.
6.6. Contents
of 872031-01
and 872031-02
Kits
Order the 872031-01 kit if the Clinac already has the venting system. Or-
der the 872031-02 kit if the Clinac is being upgraded with the venting de-
vice.
*The Cubitainer and faucet are optional: A 3050 gallon trash bag can be used instead.
Table 1.2. Contents of 872031-01 and 872031-02 Kits
-02 -01 Part Number Description
1 1 88-301200-00 Cubitainer, 5 gallon*
1 1 88-301201-00 Faucet for the 5 gallon Cubitainer*
3ft 3ft 28-158988-00 Tygon tubing, 3/8 ID x 1/2 OD
1 28-628983-00 Cap, 1/4 flare
1 28-611673-00 Union, 1/4 flare x 1/4 MPT
1 27-109614-00 Valve, Shutoff
1 28-201311-00 Nipple, pipe, 1/4 inch
1 28-207010-00 Tee, 1/4 x 1/4 x 1/4 FPT
2 28-610922-00 Elbow, 1/4 Flare x 1/4 MPT
1 00-886892-02 Clamp, Tee Mounting
4 13-311160-00 Nut, KEPS, 10-32
1 00-834931-05 Hose Assembly, 20-inch
1 88-902021-00 Snoop
88-189795-00 Teflon Tape
1 00-886891-01 Bracket, Tee Mounting
Drawing Documents included with kit
Machine Physics 2-1
In this chapter, Accelerator Physics will be discussed to give the reader a greater under-
standing of how the Clinac treatment beam is produced.
Machine Physics
2-2 Machine Physics
Table of Contents
1. Introduction: .................................................................................................................... 2-3
2. Definitions:: ..................................................................................................................... 2-3
3. Kinetic Energy Relationships: ........................................................................................... 2-3
4. Rest Energy Relationships:............................................................................................... 2-4
5. Total Energy Relationships: .............................................................................................. 2-4
6. Conversion of Energy to Electron Volts: ............................................................................ 2-5
7. Measurement of Energy Change: ...................................................................................... 2-6
8. Example of Simple Acceleration:....................................................................................... 2-8
9. The Standing Wave Accelerator: ....................................................................................... 2-9
10. Impedance Matching: ................................................................................................... 2-11
11. Plotting: ....................................................................................................................... 2-12
12. Accelerator Equivalent Circuit: ..................................................................................... 2-13
13. Load Line Considerations: ............................................................................................ 2-16
14. Fill Time:...................................................................................................................... 2-16
15. Injection Timing: .......................................................................................................... 2-17
16. Electron Injection and Bunching: ................................................................................. 2-18
17. Advances in Linear Accelerator Design for Radiotherapy: .............................................. 2-19
Figure 2.1. The Basic Accelerator: ........................................................................................ 2-3
Figure 2.2. Electrostatic (DC): .............................................................................................. 2-8
Figure 2.3. Alternating Current (AC): .................................................................................... 2-8
Figure 2.4. Phase Velocity: ................................................................................................... 2-9
Figure 2.5. Forward Power Polarity:.................................................................................... 2-10
Figure 2.6. Reflected Power Polarity:................................................................................... 2-10
Figure 2.7. Summation of Forward and Reflected Power: .................................................... 2-10
Figure 2.8. Source vs. Load Impedance: ............................................................................. 2-11
Figure 2.9. Load Power vs. Load Resistance:....................................................................... 2-12
Figure 2.10. Power Calculation Parameters: ....................................................................... 2-12
Figure 2.11. Accelerator Equivalent Circuit: ....................................................................... 2-13
Figure 2.12. Equivalent Circuit Showing Resonant Components:........................................ 2-15
Figure 2.13. Load Line Showing Effect of Energy Slit: ......................................................... 2-16
Figure 2.14. Accelerator Equivalent Circuit: ....................................................................... 2-16
Figure 2.15. Fill Time: ........................................................................................................ 2-17
Figure 2.16. Fill Time Equivalent Circuit: ........................................................................... 2-17
Figure 2.17. Electron Injection Timing:............................................................................... 2-17
Figure 2.18. Effect of Energy Slit on Injection Timing: ........................................................ 2-18
Figure 2.19. Effect of Injection Timing on Target Current Waveform: .................................. 2-18
Figure 2.20. Velocity Vectors: ............................................................................................. 2-19
Machine Physics: Introduction 2-3
1. Introduction A basic electron accelerator consists of a vacuum chamber with two elec-
trodes connected across a voltage source. Since electrons carry a negative
charge, an electron entering the cavity between the plates will be attracted
toward the plate with the positive charge and repelled by the negatively
charged plate, as shown in Figure 2.1 below. In this example, with a 1-volt
battery, the electron will be accelerated by 1 electron volt (eV).
2. Definitions: Before proceeding, it is important for the reader to understand the differ-
ence between the terms energy and intensity, used to describe treatment
beam parameters. These can be defined in two ways:
1. Basic electron beam:
Intensity: The number of electrons passing a point within a
given unit of time (beam current).
Energy: The total energy possessed by each individual elec-
tron in the beam.
2. Treatment beam (electrons or photons):
Intensity: The amount of radiation delivered per unit of time
to a point.
Energy: The ability of the beam to penetrate.
The total energy possessed by an electron (or any other object) is the sum of
its kinetic and rest energies. These will now be defined and discussed in
some detail.
3. Kinetic En-
ergy Relation-
ships
Definition: Kinetic energy is that energy an object possesses by virtue of
its motion.
Statement: The KE (kinetic energy) of an object in joules = at ev-
eryday velocities.
Example 1: Take a laboratory size object. (e.g., a baseball which weighs
approximately 0.2 kilograms . 0.5 pounds moving at 30
meters/second . 60 miles/hour):
Figure 2.1. The Basic Accelerator
e
Vacuum
1 Volt
1 eV
1
2
---mV
2
KE
1
2
--- 0.2kG ( ) 30 m/sec ( )
2
=
KE 90 joules =
2-4 Machine Physics: Rest Energy Relationships
Example 2: Take an electron (e
) which weighs 9.1 10

31
kG moving at
30 meters/second:
Observation: Compare the kinetic energy of the baseball to that of the
electron:
Conversion:1 joule = 1 watt-second
1 kilowatt hour = 3.6 10
6
watt-seconds
4. Rest Ener-
gy Relation-
ships
Definition: Rest energy is that energy an object possesses by virtue of
its mass.
Statement: A. Einsteins equation states the total energy:
E
t
= mc
2
where c is the speed of light
(.3 10
8
meters/sec).
Example 3: Take the baseball again and calculate the rest energy (RE):
RE = (0.2 kG)(3 H 10
8
m/sec)
2
= 1.8 H 10
16
joules or
= 5 H 10
9
kilowatt hours
Example 4: Now calculate the rest energy of the electron:
RE = (9.1H 10
31
kG)(3 H 10
8
m/sec)
2
= 81.9 H 10
15
joules or
= 2.275 H 10
20
kilowatt hours
Observation: Compare the rest energy of the baseball to that of the elec-
tron:
5. Total Ener-
gy Relation-
ships
Definition: Total energy is that energy an object possesses by virtue of
its motion and its mass.
E
t
(total energy) = KE + RE
90 joules 4.09 10
-28
joules
vs.
(Baseball) (Electron)
2.5 10
-5
kilowatt hours 1.14 10
-34
kilowatt hours
vs.
KE
1
2
--- 9.1 10
31
kG ( ) 30m/sec ( )
2
=
KE 4.09 10
28
joules =
1.8 10
16
joules
vs.
81.9 10
15
joules
5 10
9
kilowatt hours 2.27 10
20
kilowatt hours
Machine Physics: Conversion of Energy to Electron Volts 2-5
Example 5: Take the baseball again and calculate its total energy (E
t
):
E
t
= 90 joules + 1.8 H 10
16
joules
= 18,000,000,000,000,090 joules
Example 6: Now calculate the total energy of the electron:
E
t
= 4.09 H 10
28
joules + 81.9 H 10
15
joules
E
t
. 81.9 H 10
15
joules
Observation: Almost all of the total energy of an object is due to its mass
at these velocities.
6. Conversion
of Energy to
Electron Volts
Definition: An electron volt is the kinetic energy an electron acquires by
being accelerated in a vacuum through a potential differ-
ence of 1 volt (See Figure 2.1 on Page 2-3).
Statement: The conversion factor to go from joules to electron volts (eV)
is:
1.6 10
19
joules = 1 eV.
An electron has a rest energy of 81.9 10
15
joules; therefore,
we must divide the rest energy by the conversion factor to ob-
tain the same thing in eV.
Example 7: Converting the rest energy of the electron:
Example 8: Converting the rest energy of the baseball:
Observation: The difference of equivalent energies in eV between the elec-
tron and the baseball:
1.12 10
29
MeV 0.511 MeV
vs.
RE
81.9 10
15
joules
1.6 10
19
--------------------------------------------------- =
5.11 10
5
eV =
0.511MeV =
RE
1.8 10
16
joules
1.6 10
19
--------------------------------------------- =
1.12 10
35
eV =
1.12 10
29
MeV =
2-6 Machine Physics: Measurement of Energy Change
Observation: In Example 2 the equivalent kinetic energy of the electron
moving at 30 meters/second or 60 miles/hour in electron
volts (eV) would have required a battery potential of 2 nano-
volts (See Figure 2.1 on Page 2-3.)
Example 9: Now let us calculate the total energy an electron possesses
by passing through a potential of 22 million volts:
E
t
= KE + RE
E
t
= 22 MeV + 0.511 MeV
E
t
= 22.511 MeV
7. Measure-
ment of Ener-
gy Change
Definition: Gamma ( ) is the ratio of total energy (E
t
) to the rest energy
(E
r
) of an electron.
however, as stated before, where m
o
= original mass
and
therefore substituting mc
2
for E
and canceling identical terms
Example 10: Calculating the effective increase in mass as a result of ac-
celerating an electron to typical linear accelerator values:
Observation: An acceleration of 22 MeV gives an equivalent increase in
mass of approximately 44 times.
The following equation is derived from Newtons second law of motion (as
applied to momentum):
Statement:
These equations are normally referred to as proof that an object with mass
can never attain the speed of light.
E
t
E
r
----- =
E
r
m
o
c
2
=
E
t
mc
2
=
mc
2
m
o
c
2
------------- =
m
m
o
------- =
4MeV 0.511MeV +
0.511MeV
----------------------------------------------------- 8.828 = =
10MeV 0.511MeV +
0.511MeV
--------------------------------------------------------- 20.569 = =
18MeV 0.511MeV +
0.511MeV
--------------------------------------------------------- 36.225 = =
22MeV 0.511MeV +
0.511MeV
--------------------------------------------------------- 44.053 = =
1
1
v
c
---

2
--------------------- =
1
1
v
c
---

2
---------------------

1
2
---
=
Machine Physics: Measurement of Energy Change 2-7
Needed: A way to express velocity as a ratio of the speed of light.
Definition: $ (beta) is the ratio of the velocity of an electron to the speed
of light.
$ has other meanings different from this one and is not to be
confused as it is defined here.
substituting $ for
Conclusion:
squaring both sides
solving for $
Exercise: Calculate the percentage of the speed of light (c) as a func-
tion of kinetic energy.
Example 11:
Conclusion: An electron accelerated to 22.511 MeV is traveling at ap-
proximately 0.9997 times, or 99.97% of, the speed of light.
1
1
v
c
---

2
---------------------

1
2
---
1
1 ( )
2
--------------------

1 2
=
v
c
---
1
1 ( )
2
--------------------

1 2
=
2 1
1 ( )
2
-------------------- =
1
1
2
-----

1 2
=
E
k
E
r
+
E
r
------------------- =
22MeV 0.511MeV +
0.511MeV
-------------------------------------------------------- =
44.053 =
1
1
2
-----

1 2
=
1
1
44.053
2
-----------------------

1 2
=
0.9997 =
2-8 Machine Physics: Example of Simple Acceleration
Observation:
8. Example of
Simple Accel-
eration
In a simple accelerator, more energy can be achieved in two ways: increas-
ing the accelerating voltage ( ), or increasing the number of accelerating
cavities, as shown in Figure 2.2 below.
When high electron beam energies are required, batteries are not practical;
however, since high-voltage AC power is easy to generate, it can be used to
generate high-strength electrical fields in the cavities, as shown in Figure
2.3 below.
Accelerated Energy Ratio of c
17 KeV .252 (X-ray tube)
100 KeV .548 (Klystron)
0.511 MeV .866 (Doubling rest energy)
4 MeV .993
6 MeV .997
10 MeV .998
18 MeV .999
Figure 2.2. Electrostatic (DC)
Figure 2.3. Alternating Current (AC)
V
a
Vacuum KE = (V + V ) eV
1 2
e
V
1
V
2
etc.
Vacuum
V
1
V
2
V
3
etc.
e

Machine Physics: The Standing Wave Accelerator 2-9
The best efficiency is achieved when the electron enters each cavity at the
time when is at maximum strength. Therefore, the electrical field must
propagate through the cavities at the same velocity as the electron. This is
called the phase velocity as shown in Figure 2.4 below. Accelerators using
this principle are called traveling wave accelerators.
Definition: = Phase velocity which equals
at 60 Hz = 3000 miles
at 3 GHz* = 10 centimeters
Conclusion: Thus a particle accelerator requires microwave frequency
generators and distances.
*Actually 2.856 GHz in high-energy and 2.998 GHz in low-energy Clinacs
9. The Stand-
ing Wave Ac-
celerator
Statement:
If the cavities are one-quarter of the of the AC field wavelength, at the time
when the first and fifth cavities are at maximum forward-acceleration po-
tential, the third cavity is at maximum reverse acceleration potential. By
the time the electron reaches the third cavity, the polarities will be oppo-
site. Thus, the electron acquires additional energy in every other cavity.
In a traveling wave accelerator, the forward power is absorbed in the final
cavity. In a standing wave accelerator, the power in the final cavity is al-
lowed to reflect back, creating a standing wave and doubling the field
strength in each cavity. However, in order for the reflected power to match
the phase of the forward power, one of the end cavities must be either half
as long, or one and one-half times as long, as the others. If all cavities are
the same size, the reflected power will cancel the forward power.
Also, in order to allow the forward and reflected power to travel between the
cavities, the connecting openings must be at least one-quarter of the wave-
length.
Figure 2.4. Phase Velocity
V
a
Vacuum
d
1
= 0
2
e
3

4
v
Distance
Time
------------------------
v
d
t
--- =
E
n
t
n
( ) cos =
2-10 Machine Physics: The Standing Wave Accelerator
Figure 2.5. Forward Power Polarity
Figure 2.6. Reflected Power Polarity
Figure 2.7. Summation of Forward and Reflected Power
Electric Fields
Forward Power
4
0 -270 -180 -90 0
+ - 0 0 +
Electric Fields
4
0 -270 -180 -90 0
+ - 0 0 +
Reflected Power
Electric Fields
4
0
0
-270
-270
-180
-180
-90
-90
0
0
+2E -2E 0 0 +2E
Machine Physics: Impedance Matching 2-11
10. Impedance
Matching
When power is being transferred from its source to its load, the most effi-
cient transfer occurs when the impedance of the load (R
l
) is equal to the im-
pedance of the source (R
o
). This explained in example 12, below:

substituting for
Example 12: Reference Figure 2.8 above.
Case 1: R
l
= R
o
Given:R
o
= 1 ohm
R
l
= 1 ohm
V
o
is constant
Case 2: R
l
< R
o
Given:R
o
= 1 ohm
R
l
= 0.5 ohm
V
o
is constant
Case 3: R
l
> R
o
Given:R
o
= 1 ohm
R
l
= 2 ohms
V
o
is constant
Figure 2.8. Source vs. Load Impedance
V
o
(Source)
I
l
R
o
(Source)
R
l
(Load)
I
l
V
o
R
o
R
l
+
------------------ =
P
l
I
l
2
R
l
=
P
l
V
o
2
R
l
R
o
R
l
+ ( )
2
-------------------------- =
V
o
R
o
R
l
+
------------------ I
l
P
l
1
1 1 + ( )
2
---------------------
1
4
--- 0.25 watts = = =
P
l
0.5
1 0.5 + ( )
2
---------------------------
0.5
2.25
------------- 0.222 watts = = =
P
l
2
1 2 + ( )
2
---------------------
2
9
--- 0.222 watts = = =
2-12 Machine Physics: Plotting
11. Plotting Figure 2.9, below, illustrates this effect.
Conclusion: For maximum transfer or power the load impedance must
equal the source impedance.
Needed: Expression to find actual maximum power.
Statement:
substituting for
and multiplying by
Figure 2.9. Load Power vs. Load Resistance
Figure 2.10. Power Calculation Parameters
P
1
R
1
.25
.222
.5 1 2
V
o
I
o
R
o
R
l
I
o
V
o
2R
o
---------- if R
o
R
l
= =
P
max
I
o
2
R
o
=
I
o
2
R
o
V
O
2R
o
----------

2
R
o
=
V
o
2R
o
---------- I
o
V
o
2R
o
----------

2
R
o
V
2
o
4R
o
---------- = R
O
P
max
V
2
o
4R
o
---------- =
Machine Physics: Accelerator Equivalent Circuit 2-13
12. Accelera-
tor Equivalent
Circuit
Figure 2.11, below, illustrates the relationships between the parameters.
Statements:
substituting for
Think about it!
rearranging terms
multiplying by n
combining and rearrang-
ing terms
dividing by and solv-
ing for
Figure 2.11. Accelerator Equivalent Circuit
R
o
I
o
V
o
R
a
I
2
1:n
I
ra
V
a
I
a
V
n
a
V
o
I
o
R
o
1
N
----V
a
+ =
I
2
I
ra
I
a
+ =
I
ra
V
a
R
a
------- =
I
2
V
a
R
a
------- I
a
+ =
V
a
R
a
------- I
ra
I
o
nI
2
=
I
o
n
V
a
R
a
------- I
a
+

=
V
o
n
V
a
R
a
------- I
a
+

R
o
V
a
n
------ + =
V
o
n
V
a
R
o
R
a
-------------

nI
a
R
o
V
a
n
------ + + =
nV
o
n
2
V
a
R
o
R
a
-------

n
2
I
a
R
o
V
a
+ + =
nV
o
V
a
n
2
R
o
R
a
------- 1 +

n
2
I
a
R
o
+ =
V
a
nV
o
n
2
R
o
R
a
------- 1 +
------------------------
n
2
I
a
R
o
n
2
R
o
R
a
------- 1 +
------------------------ = n
2
R
o
R
a
------- 1 +
V
a
2-14 Machine Physics: Accelerator Equivalent Circuit
Definition: (beta) is equal to the shunt impedance divided
by the source resistance times the turns ratio squared.
inverting
rearranging
restating the equation
for (see above)
substituting for
substituting for
multiplying by
restating the equation for (see above)
solving for
will be denoted as .
multiplying by n
combining terms
substituting for
multiplying by
R
a
n
2
R
o
------------- =
1
---
n
2
R
o
R
a
------------- =
n
2
R
o
R
a
------- =
V
a
nV
o
n
2
R
o
R
a
------- 1 +
------------------------
n
2
I
a
R
o
n
2
R
o
R
a
------- 1 +
------------------------ =
V
a
V
a
nV
o
1
--- 1 +

-------------------
n
2
I
a
R
o
1
--- 1 +

------------------- =
n
2
R
o
R
a
-------------
V
a
nV
o
1
--- 1 +

-------------------
I
a
R
o
------
1
--- 1 +

------------------- =
R
a
------- n
2
R
o
V
a
nV
o
1 +
--------------
I
a
R
a
1 +
------------- =
P
max
V
o
2
4R
o
---------- = P
max
V
o
4P
max
R
o
( )
1 2
= V
o
P
max
P
o
V
o
4P
o
R
o
( )
1 2
=
nV
o
n 4P
o
R
o
( )
1 2
=
nV
o
4P
o
n
2
R
o
( )
1 2
=
nV
o
4P
o
R
a
-------

1 2
=
R
a
------- n
2
R
o
nV
o
4P
o
R
a
( )
1 2
=
Machine Physics: Accelerator Equivalent Circuit 2-15
restating the accelerator voltage equation
substituting for
in accelerator equation
Statement:
substituting K
1
and K
2
in
accelerator equation
Statement: Where P
o
equals the input microwave power to the accelera-
tor and I
a
equals the electron accelerated beam current.
This is not the same as target current!
Statements:
The ratio of the accelerator impedance to the source imped-
ance is given by $. At $ = 1, the accelerator is matched. All of
the above calculations relate to an accelerator operating at
resonance.
V
a
nV
o
1 +
--------------
I
a
R
a
1 +
------------- =
V
a
4P
o
R
a
( )
2
1 +
----------------------------
I
a
R
a
1 +
------------- = 4P
o
R
a
( )
1 2
nV
o
K
1
4R
a
( )
1 2
1 +
--------------------------- =
K
2
R
a
1 +
-------------

=
V
a
K
1
P
o
( )
1 2
K
2
I
a
( ) =
P
cavity
V
a
2
R
a
--------- =
P
beam
V
a
I
a
=
Figure 2.12. Equivalent Circuit Showing Resonant Components
R
o
I
o
V
o
R
a
I
2
1:n
I
ra
V
a
I
a
V
n
a
2-16 Machine Physics: Load Line Considerations
where
Figure 2.12 on Page 2-15 defines the off-resonant conditions.
13. Load Line
Consider-
ations
14. Fill Time
See reference Figure 2.15.
Statement: where
Z
R
a
1 2Q
o
( )
2
+ [ ]
1 2
--------------------------------------------- =

o
o
---------------- =
Figure 2.13. Load Line Showing Effect of Energy Slit
V
a
(Energy)
+5%
-5%
Operating Point
3% Energy Slit
I
a
(Beam Current)
Figure 2.14. Accelerator Equivalent Circuit
R
o
I
o
V
o
R
a
I
2
1:n
I
ra
V
a
I
a
V
n
a
V
a
V
o
R
a
R
o
R
a
+
--------------------

1 e
t
R
EQ
C
---------------

= R
EQ
R
o
R
a
R
o
R
a
+
-------------------- =
Machine Physics: Injection Timing 2-17
15. Injection
Timing
for the steady state condition.
When capacitor current (I
c
) equals the desired value of I
a
, S1
is closed.
Figure 2.15. Fill Time
V
a
t
RF Power Level
Figure 2.16. Fill Time Equivalent Circuit
V
o
R
o
R
a
C
I
c
I
a
S1
V
a
V
o
1 +
-------------
I
a
R
a
1 +
------------- =
Figure 2.17. Electron Injection Timing
V
a
t
Magic time to close S2 for gun pulse timing
Desired RF Power Level
2-18 Machine Physics: Electron Injection and Bunching
Figures 3.18 and 3.19 illustrate how of Gun timing, RF loading and the En-
ergy Slit affect the target current.
16. Electron In-
jection and
Bunching
Statement: The wave field is initially moving faster than the electron. so
the electron appears to be moving backward with respect to
the wave field.
Injection energy (velocity) must be correct to deposit the elec-
tron at the proper point on the wave front. The percentage of
current captured is a function of the field.
Speed of Light Injection Equation:
The cos2 approaches 1 if bunching parameters are correctly
selected.
Figure 2.18. Effect of Energy Slit on Injection Timing
Figure 2.19. Effect of Injection Timing on Target Current Waveform
V
a
t
Injection too early
Injection too late No Injection
Desired RF Power Level
3% Slit
Injection too early
4.5 5 Sec
Optimal target current
Injection too late
(or bad light pipe)
cos
o
cos
2M
o
c
2
E
---------------------
1
1 +
-------------

1 2
=
Machine Physics: Advances in Linear Accelerator Design for Radiotherapy 2-19
In the highest energy modes only about 1/3 of the electrons
that enter the guide are captured. For lower energy modes,
this number is even less.
17. Advances
in Linear Ac-
celerator De-
sign for
Radiotherapy
This section has been compiled from an article by Dr. C. J. Karzmark, pub-
lished in 1984 by the Department of Radiology, Stanford University School
of Medicine, Stanford, California 94305, and is included in this manual for
reference only.
In this section the paragraph numbering scheme of the origi-
nal article is used and the bibliographic references are omit-
ted.
I. Introduction The microwave-powered electron linear accelerator, or linac, is becoming
the dominant radiotherapy treatment unit. In the U. S., linacs now com-
prise over one-half of all megavoltage treatment units in service and about
90% of newly installed units. Several technical advances, combined with
attention to how patients are most effectively set up and treated, have led to
continuing improvements in linac radiotherapy during the three decades
since their introduction in England and in the United States. A simplified
exposition of linac theory and operation can be found in a contemporary
reference by Karzmark and Morton. In it, the major modules of a medical
linac are identified, their principles of operation are described individually,
and then their collective functioning in providing a radiation treatment
beam is discussed. Additional aspects are presented in an earlier technical
review of the subject by Karzmark and Pering.
The microwave accelerator structure (sometimes referred to as the guide
or as the waveguide), the central component of linacs, consists of a linear
array of microwave cavities. Such structures have undergone extensive de-
Figure 2.20. Velocity Vectors
Wave field velocity vector
Electron velocity vector
Resultant with reference to wave field velocity
RF power amplitude
Region of bunched electrons
e
2-20 Machine Physics: Advances in Linear Accelerator Design for Radiotherapy

velopment and the result has been physically shorter accelerators having
higher energy gradients along the accelerating axis. Most treatment units
are isocentrically mounted, and many of these machines employ a horizon-
tally mounted accelerator structure and a beam bending system. Here, im-
proved beam transport magnet systems and treatment head designs help
ensure flatness uniformity of treatment beams and their stability with time
independent of gantry motion and position as well as of beam-limiting de-
vice orientation. Various methods have been developed to permit varying
the electron beam energy over a wide range. For example, in single-pass
(single traversal) linacs, the ratio of radio-frequency (rf) power to two por-
tions of the accelerator structure is varied so as to keep a constant energy
gradient in the first injection portion while providing a variable energy gra-
dient in the second portion. In two-pass linacs, the phase of the microwave
electric field encountered by electron beam is varied as it returns for the
second pass. In microtrons, the beam extraction path is shifted from one
orbit to another. By providing both high and low x-ray energies together
with a broad range of electron beam energies, a wide variety of treatment
plans can be implemented. The incorporation of digital electronics and
computer techniques has led to improved reliability together with desirable
monitoring, control and safety features. Human engineering improvements
have involved aesthetics, function, patient safety, and comfort, as well as
convenience for the radiotherapy treatment technologist. Several new re-
search directions suggest opportunities for continued improvement in lin-
ear accelerators for radiotherapy.
II. Microwave
Accelerator
Structures
We can understand how linacs accelerate electrons by first examining how
an electric E field wave pattern travels down a hollow cylindrical pipe, or
waveguide as it is called. Such waveguides, and their hollow rectangular
pipe counterparts, are used to transmit microwave power from an rf source
to an accelerator structure.
A. Introduction Waveguides replace conventional wires and cables which are inefficient for
transmitting power at microwave frequencies. Figure 2.21(a) shows the E
field pattern and charge distribution at one instant of time in a plane con-
taining the axis of a cylindrical waveguide. The accompanying magnetic H
field pattern, in this case circling around and orthogonal to the axis, is not
shown here or in later illustrations since it is not directly involved in the ac-
celerator process. Electrons injected along the axis would be accelerated by
the moving E field. Unfortunately, however, the velocity of this E field pat-
tern, the phase velocity of the wave exceeds that of light and is therefore
unsuitable for continuing acceleration of charged particles. The wave is
slowed by inserting washerlike discs into the waveguide, as shown in Fig-
ure 2.21(b), so that the wave stays in step with the accelerating electrons.
These discs divide the waveguide into a series of cylindrical cavities, the ba-
sic structure of a linear accelerator.
Virtually, all medical linacs operate at frequencies of approximately 3000
MHz in the so-called S band where typical accelerating cavities are about
10 cm in diameter and 2.5 to 5 cm in length. These cavities are arranged to
serve two purposes: (1) to couple and distribute microwave power between
adjacent cavities and hence, along the length of the structure; (2) to provide
an E field with suitable axial distribution for accelerating electrons. The
particular spatial configuration of E and H fields in a cavity is called the
mode and denoted by abbreviations such as TEM (transverse electric and
magnetic) field pattern as in a coaxial line or cavity or TM
010
(fundamental
transverse magnetic only field pattern), as in Figure 2.21(b).
By varying the aperture and length of the cavities initially traversed, the
continuum of injected electrons is concentrated into discrete bunches as
well as accelerated. This initial portion of the structure is called the
buncher and its cavities are non-uniform. Their inner diameter, aperture
diameter, and axial spacing vary to provide the increasing phase velocity E
field to accommodate the accelerating electron bunches. Since the electron
velocity is almost constant and near the speed of light beyond the buncher,
subsequent cavities are made uniform for further acceleration of the
bunches of electrons. Early buncher designs contained many cavities (see
Figure 2.22); later only several cavities were used and more recently, a sin-
gle half cavity. Improved understanding of bunchers has resulted in a re-
duction of electron injection voltage from the 100200 kV region to the 1
30 kV region. Typically, one-third of the injected beam is captured,
bunched, and accelerated. (See Ref 5 for an elementary description of
bunching action.)
B. Traveling
Wave Struc-
ture Linac
An early prototype traveling wave (TW) structure, cut in half along its cylin-
drical axis, is shown in Figure 2.22. The buncher section is on the left and
the uniform section is on the right. Figure 2.23 shows an elementary TW
accelerator with electrons and microwave power injected on the left. Accel-
erated electrons emerge on the right and the residual microwave power not
transferred to the electron beam or structure walls is absorbed in a load at
that end. The discs serve to separate the cylinder into a linear array of cou-
pled cavities which both transport the microwave power and accelerate
electrons down the structure.
Figure 2.21. (a) Spatial traveling wave electric E field pattern and
charge distribution at one instant of time along the axis of a smooth
cylindrical waveguide. (b) Spatial traveling wave electric E field pat-
tern and charge distribution at one instant of time along the axis of a
disk-loaded cylindrical waveguide. The direction of the electric E field
is reversed every half wavelength 8/2. The pattern repeats every
wavelength and there are four cavities per wavelength in the disk-
loaded structure. The direction of the E field also reverses every half
cycle in time.
8
8
8/2
8/2
(a)
(b)
+ +
+ +
+ +
+ +
+ +
+ +

+
+
+
+

The acceleration of an electron bunch in a TW linac is similar to the way a
boy on a surfboard, positioned just forward of the crest on a water wave of
velocity v
w
, moves forward as shown in Figure 2.24(a). Similarly, Figure
2.24(b) shows how two electron bunches of velocity v
e
, are accelerated by
the negative portion of the E wave moving at phase velocity v
p
. This E wave
is created by the charge distribution shown in Figure 2.24(c). For clarity in
Figure 2.24(c), the instantaneous axial E field patterns, such as depicted in
Figure 2.21(b), have been omitted. The boy and the electron bunches move
forward at the velocity of their respective wave motions, the phase velocity
v
w
and v
p
, respectively.
Traveling wave structures of 2B/4 design, in which one cavity in four con-
tains an electron bunch at any one time, dominated early linac structure
designs. Some later structure designs incorporated three cavities per wave-
length 8, a phase shift of 2B/3 rad (120) per cavity with one cavity in three
containing an electron bunch at any one time. More recently, standing
wave (SW) designs have appeared in which one axial cavity in two contains
an electron bunch at any one time.
Figure 2.22. Cutaway traveling wave accelerator structure; the
buncher section is on the left and the uniform section is on the right.
Figure 2.23. Short traveling wave accelerator structure. The rectangu-
lar waveguide on the left is the input coupler for conveying microwave
power from a source to the structure.
In general, TW structures have a larger frequency bandpass and hence, are
less sensitive to frequency change than comparable SW structures for the
same mode of operation. Since the buncher in TW linacs is adjacent to the
power feed at the gun end, the magnitude of the electric field in the early
cavities relatively independent of beam current so the buncher action is not
greatly affected by beam loading. As a result, bunching in TW structures is
efficient over a comparatively broad energy range, typically 35%.
C. Standing
Wave Struc-
ture Linac
We can convert the traveling wave linac just described to a standing wave
linac. We do so by arranging for both a forward moving (to the right) f and
backward moving (to the left) b E wave, each of which is reflected at both
ends of the structure. The two moving E field maxima and resultant wave
are shown in Figure 2.25 at three sequential times one-quarter cycle apart.
The resultant E field pattern is the sum of the forward and backward com-
ponents and ideally, in the absence of the I
2
R copper losses and beam load-
ing, is double the amplitude found in a compatible TW structure. Note that
every other cavity (e.g., #2 and #4) in Figure 2.25 has zero field at all times,
at times (e.g., t
1
, and t
3
), because both forward and backward waves are
each zero, at other times (e.g., t
2
), because they are equal to magnitude but
opposite in direction and hence completely cancel.
Figure 2.24. Traveling wave principle: (a) for a boy surfing on a water
wave advancing to the right, (b) for electron bunches occupying a sim-
ilar position on an advancing electric E field (for simplicity, E is
drawn upward), (c) the associated charge distribution which pushes (
charge) and pulls (+ charge) the electron bunches along the cylinder.
There are four cavities per wavelength 8 and one electron bunch every
four cavities.
Such zero-field cavities couple microwave power between cavities but play
no role in particle acceleration. They may be moved off axis to form a side-
coupled or bimodal SW structure, thus shortening the overall length for a
given energy gain. Figure 2.26 illustrates a cutaway section of a side-cou-
pled SW structure.
The apertures between cavities of TW structures involve a design compro-
mise. They should be large to transport microwave power efficiently be-
tween cavities but should be small enough to squeeze down the E field,
thus maximizing its value along the beam axis. This design conflict is re-
moved for the SW structure; axial acceleration cavities may be optimized
for electron acceleration and coupling cavities for microwave power trans-
port. The latter are made small for convenience, since to first order they
contain zero electric field and hence, can be of much lower Q than on-axis
cavities without entailing much I
2
R loss in their copper walls since almost
no wall current flows.
Figure 2.25. Standing wave electric E field patterns in an accelerator
structure for combined forward and backward waves at three sequen-
tial instants of time. Two traveling waves moving in opposite direc-
tions (f forward, b backward) generate such a standing wave. At time
t
2
, the E field is zero in all cavities. The pattern shown at time t
1
will
recur one-half cycle after time t3. Individual cavities are numbered
across the top.
b f
t
2
t
3
TIME PHASE
E FIELD MAXIMA
t
1
E
E
NEG. POS.
E
1 2 3 4 5
b
f
t
1
t + /2
1
t +
1
The axial apertures of SW accelerating cavities can be made small and
equipped with extended nose cones which increase their axial length.
They now function as zero-field drift tubes, and the electron bunch crosses
the cavity gap as E is near its maximum value (see Figure 2.26). Figure 2.27
illustrates the evolution of such an SW structure from a TW structure and
the associated E field pattern at one instant of time. The coupling cavities
in Figure 2.27(d) are staggered, in contrast to Figure 2.27(c), to reduce
asymmetries introduced by the coupling slots.
Figure 2.28 shows the time variation of the E field in an SW structure for
one complete microwave cycle. The pattern varies sinusoidally in magni-
tude but remains stationary in space just as a vibrating violin string. The
actual field axial distribution resembles more that of Figure 2.27 than Fig-
ure 2.25.
These somewhat rectangular shaped patterns can be represented by a sum
of spatial harmonics comprising a sinusoidal fundamental as in Figure
2.25 and higher order sinusoids which do not produce net acceleration of
synchronous electron bunches. The E field pattern recurs along the beam
axis twice as frequently as in TW structures; that is, an accelerating elec-
tron bunch is contained in every second cavity instead of every fourth cav-
ity. In SW structures, all the cavities tend to have the same electric field,
since the rf power bounces back and forth from each end of the structure
many times (typically 100 times) to fill the cavities with energy as expressed
in E and H fields. Hence, beam loading of any cavity contributes to the re-
duction of the E field in all cavities, including the buncher cavities. As a re-
sult, bunching in SW structures is efficient over only a narrow energy
range, typically 5%. Thus, to maintain effective bunching action in SW
structures, the input rf power must be increased with an increase in beam
current.
Figure 2.26. Cutaway bimodal or side-coupled standing wave accelera-
tor structure. The axial accelerating cavities are shaped for optimum
efficiency and the off-axis coupling cavities are staggered to reduce
asymmetries introduced by the coupling slots (courtesy of Los Alamos
Scientific Laboratory, Los Alamos, New Mexico).
Figure 2.27. Evolution of a standing wave accelerator structure from a
traveling wave structure and related E wave patterns along the axis.
The E field patterns below each structure show the spatial field along
the axis at the same time in the microwave cycle.
E FIELD MAXIMA
NEG. POS.
(a)
(b)
(c)
(d)
Figure 2.28. Axial E field pattern for one microwave cycle of time for
the bimodal structure depicted in Figure 2.27(d). This cycle includes
the sequential E field patterns shown. The corresponding phase angles
in degrees and radians are shown on the right. The pattern at time t9
is a repetition of that at time t1.
t
1
45
90
135
180
225
270
315
360
0 0
t
2
t
3
t
4
t
5
t
6
t
7
t
8
t
9
/2
3 2 /
2
D. Structure
Designs
Several new structure designs, in addition to those described, have been
developed recently. Two side-coupled structures of the Los Alamos (LASL)
design, such as illustrated in Figure 2.29(a) (and also Figure 2.26), may be
interlaced in the Varian design as shown in Figure 2.29(b).
By arranging the side cavity coupling to connect every other axial cavity as
shown in Figure 2.29(b) instead of adjacent cavities and feeding each group
of cavities with microwave power 90 out of phase, a remarkably short,
high-gradient structure results. It withstands high average E field gradi-
ents without breakdown along its length, in part because the ratio of the
peak E field at the cavity surface to average accelerating E field value on
axis of almost four in the LASL design is reduced to about 1.3 for the Varian
design. However, the latter is more difficult to manufacture and has higher
I
2
R dissipative copper losses associated with its larger ratio of cavity wall
area to cavity volume. Benguang et al. have recently studied the interlaced
structure with the aid of a model and a computer program. They conclude
that the structure is best suited for short straight-through design linacs.
Schriber has provided a useful comparison of SW and TW structures.
Figure 2.29. Cross section and end views of two SW structures: (a) side-
coupled Los Alamos (LASL) design; (b) side-coupled interlaced Varian
design. The sections AA, which include the cylindrical axes of sym-
metry, are shown on the left.
A
A'
(a)
A
A'
(b)
Figure 2.30 shows some structure variations which have been investigated
or adopted for the relativistic (constant velocity), section of the linac struc-
ture. The on-axis coupling designs, (a), (b), (c) and (d), are smaller in diam-
eter and entail simple machining and brazing operations than off-axis cou-
pling designs (e), (f), (g), and (h).
The constant impedance uniform TW structure [Figure 2.30(a)] character-
ized early structure designs. It provides a decreasing energy gain per meter
along its length as the microwave power is attenuated by the structure and
by transfer to the ever-more-energetic electron beam. The constant gradi-
ent non-uniform TW structure Figure 2.30(b)] provides a constant energy
gain per meter for a specific loading, i.e., beam current.
Figure 2.30. Linac structure variations showing E field maxima at one
instant of time: (a) TW constant impedance; (b) TW constant gradient;
(c) SW biperiodic with on-axis coupling cavities; (d) SW tri-periodic
with on-axis coupling cavities; (g) disk and washer cross section; (h)
disk and washer structure. The arrows represent the maximum E field
values at one instant of time. Structures (e) and (f) are shown in more
detail in Figures 3.29(a) and 3.29(b) respectively.
(a) (b)
(c) (d)
(e) (f)
(g) (h)
Here, the size of aperture varies regularly along the length, but the phase
velocity is constant. The 10,000-ft. Stanford Linear Accelerator Center
(SLAC) physics research accelerator is divided into 10-ft. sections of this
design. The bi- and tri-periodic SW structures, Figures 3.30(c) and 3.30(d),
respectively, are simple to manufacture and have smaller outside diame-
ters than the side-coupled variations [Figures 3.30(e) and 3.30(f)] but their
shunt impedance, a measure of efficiency, is usually lower and they have
lower energy gains for a given power and beam loading. The on-axis cou-
pling cavities of radiotherapy linac structures [Figures 3.30(c) and 3.30(d)]
can be thin since their resonant frequency depends only on their diame-
ter in the dominant TM
010
mode. They contain only rf coupling fields and
not accelerating fields and hence, can be of low-Q design. Power loss in
these cavities is small since the E field and associated wall currents within
them are small. Coupling between axial cavities is usually magnetic via pe-
ripheral slots [see Figures 3.26 and 3.27(b)] and axial apertures can be op-
timized for acceleration. The transverse magnetic modes (sometimes called
E modes), having only circular H field lines, are suited for such magnetic
coupling. Although most bi-periodic structures employ magnetic coupling
using peripheral slots, an on-axis coupled structure has been built and
tested at 3000 MHz. Schriber et al. have compared the performance of S
band, standing wave linacs having either on-axis or off-axis coupling. They
conclude that optimized on-axis coupled accelerating structures have
about a 20% higher effective shunt impedance (hence, higher energy gradi-
ent) and are an attractive choice. A disc and washer test cavity [Figures
3.30(g) and 3.30(h)] has been built and tested at 1350 MHz for use on a
proton linac where it compares favorably with the side-coupled design.
However, for 3000-MHz electron linacs, its relatively large diameter, diffi-
culties in supporting and cooling the inner washers, as well as the presence
of higher order spatial modes due to the washer supports, lower the shunt
impedance and mitigate against its use. Nevertheless, it is being studied for
use in a high-power (300 W) 2856-MHz linac or 36-orbit racetrack mi-
crotron intended to produce negative pi-mesons (pions) for radiotherapy.
E. Structure
Design Param-
eters and
Operating
Principles
Microwave cavities are efficient devices for accelerating charged particles.
An accelerating potential of about 1 MV can be established across the gap
of a 5-cm-long cavity with about 0.2 MW of pulse power loss in that cavity.
We characterize this efficiency by a figure of merit or quality factor, Q, de-
fined by
(1)
where is the resonant frequency and the frequency difference between
the half power points. The energy storage is in the E and H fields in the cav-
ity volume. The energy loss is to the cavity interior surface from the cur-
rents which flow there, giving rise to the E and H fields. Microwave cavities
for S-band accelerators may have unloaded values as large as 2 10
4
or
larger, but this value decreases with beam loading and external coupling to
adjacent cavities and back to the power source. Almost all medical linacs
operate in the microwave S-band at a frequency of approximately 3000
MHz. If a beam-loaded cavity has a Q of 10
4
, its half power is 300 kHz.
For stable output with less than 1% reduction in energy, an automatic fre-
quency control (AFC) system would be expected to limit frequency excur-
sions to 20 kHz. Using the same criterion, the dimensional tolerance for
achieving compatible resonant frequencies for a group of S-band cavities is
about 10
3
cm and precision machining is required. For copper, the temper-
Q f
0
f =
2f
Energy stored in cavity ( )
Energy loss per cycle ( )
----------------------------------------------------------------------- =
f
0
f
f
ature coefficient of expansion is such that a change in resonant frequency
of about 60 kHz/C results and temperature control of the cavity inner
surface of better than C is desired for the structure. Alternatively, the
AFC system can follow the resonant frequency of the structure as the
temperature of the structure changes. Electron energy is denoted by V in
this section to avoid confusion with the electric field E, and all powers are
peak values during the pulse unless stated otherwise.
The energy V, gained by an electron traversing a structure, is given by the
integral of the first spatial harmonic of the axial electric field E
z
, over the
length of the structure, L in meters, multiplied by the electron charge e. E
z

is equal to , where is its input maximum value and is the
phase factor, often near 90, and described in connection with Figure 2.32.
(Mev). (2)
A useful figure of merit for structures and one that establishes the power
needed for the requisite E field for a given energy gain is the shunt imped-
ance r.
(MS/m). (3)
Here, r is expressed as the square of the axial field divided by the power dis-
sipated per unit length. Expressed in megohms per meter, shunt imped-
ance values at S-band typically vary from 50 to 120 MS/m. Both Q and the
power dissipation, dP/dz, depend on the precise shape of the conducting
walls of the cavities and on the way current is distributed on them. For the
general case, it is useful to consider the maximum energy

gain for zero
beam current and then subtract the effect of a beam current i in reducing
E
z
.
(MeV). (4)
The constant K
1
is a voltage attenuation coefficient and is characteristic of
the particular structure. Strictly speaking, Eq. (4) defines the load line for a
TW structure. When impedance matched, TW linacs look like a non-reso-
nant, pure resistive load to the power source, and are characterized by a
straight voltage-current load line. The SW load line is curved and falls
slightly below that for the TW structure, except where it is tangent at the
beam loading for the correct impedance match. SW linacs look like a reso-
nant circuit with a reactive component absent only at one beam condition.
The zero current energy may be expressed by
(MeV). (5)
Here, is a constant typically slightly less than unity, r is in megohms per
meter, and is in megawatts copper loss to the total structure length L in
meters. Microwave input power is expended as
(6)
(MW), (7)
f
0
f
0
E
z
0
sin E
z
0

V e E
z
dz
0
L
=
r E
z
2
dP dz ( ) =
V
o
V V
o
iK
1
=
V
o
V
o
K
2
L r P
Cu
=
K
2
P
Cu
P
o
P
Cu
P
l
P
e
+ + =
P
e
V i =
where represents copper losses associated with the intense cavity wall
currents which give rise to the cavity E and H fields. The term represents
residual or reflected power dissipated in a load, and , the power ulti-
mately transferred to the electron beam. Linac structures are compared in
terms of values, the zero beam condition. Thus, for L = 1 m, = 2 MW,
and r = 50 MS/m, a value of 10 MeV is obtained. Electron energy is a
square root function of each parameter in Eq. (5), and there are linear
trade-offs between them. For example, we might halve the structure length
and double the peak power , leaving the zero beam current electron en-
ergy unchanged.
The effect of beam loading in a constant group velocity TW structure [Fig-
ure 2.30(a)] is described by Eq. (4), as a linear reduction in the rate of en-
ergy gain as the bunch proceeds down the length of the structure. We may
compensate for this reduction by progressively reducing the aperture size
which operates to reduce the group velocity and increase the stored energy
in later cavities. The result is a constant gradient structure as depicted in
Figure 2.30(b). The effect of beam loading in SW structures operates differ-
ently. Here, the effect is to reduce the average value of the E field in all cav-
ities.
A high shunt impedance denotes high efficiency in producing high-energy
gain at zero beam current and is a significant parameter when comparing
the performance of two structures under similar conditions of length, beam
loading, and power. At times, we may choose to optimize one parameter at
the expense of another. For example, the Varian structure depicted in Fig-
ure 2.29(b) has a peak to average axial E field ratio of 1.3 compared to 3.8
for the LASL structure of Figure 2.26(a), although the shunt impedance val-
ues do not differ greatly, for 83 MS/m for Varian versus 78 MS/m for LASL
structures. Hence, the Varian structure allows much higher average field
gradients without electrical breakdown, and a shorter linac structure can
be constructed. A 10-cm-long 4-MeV linac using the Varian structure
would require 1.8 MW of microwave power at zero beam current. Opera-
tionally, we require 0.8 MW of electron beam power; 4 MeV at 200 mA, dur-
ing the pulse to provide an adequate x-ray dose rate. Hence, total power in-
put to the structure would be 2.6 MW requiring about 3.2-MW magnetron
power considering transmission losses, a value considerably in excess of
that available from a typical 2-MW magnetron. At the present time, the
Varian structure is not commercially employed. We can construct a more
conservative 28-cm-long 4 MeV linac which requires only 0.75 MW of mi-
crowave power at zero beam current at 2-MW magnetron power at full
beam current using either it or the LASL structure. The tradeoff here is be-
tween structure length and microwave power. Large amounts of power are
expended in the normal metallic resistance of copper. Therefore, electron
linacs are operated on a pulse basis with a typical duty factor of 0.001.
Hence, the average power consumption in kilowatts is typically numerically
equal to the peak value in megawatts.
Individual microwave cavities are designed for operation in a particular res-
onant mode which connotes a specific geometrical pattern of E and H fields
within the cavity. Most common for electron linacs is the TM
010
mode
which has only transverse H fields, as well as the requisite axial E
2
fields.
We design the structure for operation in this, the dominant mode, and find
that other cavity resonant modes, such as the TM
010
or TM
011
, are rela-
tively far removed in frequency. However, the structure operating as a se-
ries of coupled cavities may also support additional modes at frequencies
near the resonant mode desired, e.g., TM
010
. These coupled cavity, opera-
tional modes consume energy, and usually do not contribute to beam ac-
P
Cu
P
l
P
e
V
o
P
o
V
o
P
Cu
celeration. This problem becomes more severe as the number of coupled
cavities is increased; the number of coupled cavity modes increases and
their frequency separation from the dominant resonant frequency TM
010

decreases. This problem is known as moding, and design efforts are di-
rected toward its reduction or suppression. For example, the tapered con-
stant gradient TW structure depicted in Figure 2.30(b) is less subject to
moding problems than the constant impedance structure of Figure 2.30(a).
The periodicity found in constant impedance structure is interrupted since
sequential constant gradient cavities differ slightly and coupled cavity mote
resonances are not reinforced along the structure. A mid-structure feed
point is one method of reducing moding in SW linacs.
Before an injected beam of electrodes can gain maximum energy, the struc-
ture must be filled with electromagnetic energy. Each structure design has
a characteristic fill time, typically about 1 Fs, which is determined by the
structure design and the velocity of propagation of rf energy down the
structure, the group velocity v
g
. TW linacs involve one-way propagation of
waves, and their fields build up in space during the fill time. A typical group
velocity in a TW linac is 0.01c where c is the velocity of light. SW linacs in-
volved two-way propagation of waves. Their fields build up in time to a con-
stant equilibrium value throughout the structure by the two waves bounc-
ing back and forth many times between the two ends of the structure dur-
ing the fill time. A typical group velocity for these waves is 0.05c in an SW
structure and there are typically ten bunches in a 1 m structure. Electrons
are injected into the structure with a delay approximating the fill time. The
precise time is chosen such that the unloaded E field has built up to about
the stable beam loaded value, thereby minimizing electron beam energy
spread, and optimizing capture and bunching of injected electrons.
III. Widely Vari-
able Energy
Linacs
Clinical considerations suggest the need for a wide variety of beam ener-
gies, particularly for electron therapy where tumors of different sizes ex-
tend to different depths adjacent to or near a body surface. A range from 2
MeV to more than 30 MeV electron beams has been employed.
A. Clinical
Need
Clinically, most tumors may be adequately irradiated with 46 MV x-ray
beams and treatment units having these energies comprise the majority of
linacs in use. However, thick body sections such as the lateral pelvis are
advantageously treated with 1025 MV x-rays. Marks has noted the advan-
tage of treating certain tumors with combined low- and high-energy x-rays.
Gale and Innes earlier cited the advantages of employing mixed high-en-
ergy x-ray and electron beams. Tapley and others have demonstrated an
advantage of adding an electron boost irradiation to some specific lesions
treated with megavoltage x-rays. It would appear from this perspective,
that a single versatile linac design could advantageously provide a low and
a high x-ray energy together with a wide variety of electron beam energies
without the necessity of moving the patient. Such requirements pose diffi-
cult design problems but several units have been developed recently to
solve them.
B. Technical
Difficulties
The major problem arises because of the wide energy limits inherent in
these specifications for both electrons and x-rays together with significant
differences in beam current requirements for the two modalities. For x-ray
dose rates comparable to those for electron therapy, the beam current re-
quirement may be 100 or more times greater than for electron therapy. The
overall energy gained by an electron is the integral of the axial E field en-
countered over the path length through the structure. The E field itself is
proportional to where is the microwave power delivered to the
copper walls of the structure; that is, after beam loading and other losses
are considered.
The major parameters in determining the range and magnitudes of electron
energy are the type and length of the accelerator structure, the available
microwave power, and beam loading as it depends on the beam current. Ei-
ther a klystron or magnetron may be employed as the microwave power
source. A klystron usually functions as an amplifier in conjunction with a
low-power oscillator. This combination is more stable than a magnetron os-
cillator and its higher output can be varied more readily over a wider power
range than can a magnetron. Although the klystron and associated elec-
tronics are more costly than a magnetron, the longer klystron life, some-
times ten times longer, may represent good value. Efficient capture of elec-
trons injected into a linac structure, i.e., the bunching action performed by
the first few cavity, is effective over a relatively narrow range of E field val-
ues and hence limited microwave power range. Outside this range, it may
be difficult to capture and bunch sufficient numbers of electrons to satisfy
the high-current requirement for low-energy x-ray therapy, the most de-
manding modality in this regard since x-ray production varies approxi-
mately as V
3
where V is the electron energy. The bunching action of the
conventional TW structure is less susceptible to beam loading than the SW
structure as explained in Sec. II.-B..
C. Early
Designs
In view of these considerations, one early approach in covering a wide elec-
tron and x-ray energy range was to employ a relatively long TW structure
whose bunching action encompasses the requisite beam currents and
beam energy range. Beam energy is varied by changing the microwave
power input but an increased energy spread of the beam also results be-
cause of poor electron bunching action as the energy range is increased.
Beam energy is a square root function of input power and therefore
changes slowly with changes in power. Klystrons function well down to half
power and magnetrons over a much narrower range. A 50% reduction in
klystron power achieves only a 30% reduction in beam energy. A typical en-
ergy requirement for electron therapy from 6 to 18 MeV would require a
ninefold decrease in power and is beyond the capability of either klystrons
or magnetrons themselves. Such an energy range can be accommodated by
varying the linac input power through the use of an additional microwave
attenuator and ancillary hardware.
A very early method of reducing the electron beam energy was to detune the
power source frequency from the structure resonant frequency . Since
the resonance curve, E
z
versus frequency, for these high-Q structures is
very steep, the axial electric field E
z
and beam energy change rapidly with
tuning. However, this method of beam energy reduction proved unsatisfac-
tory since operation on either of the resonance curve slopes is unstable.
The attendant complications include reflections from the impedance mis-
match, an increasingly non-uniform axial field E
z
, and reduced capture
and bunching of the injected beam from the gun. A detuned linac may give
rise to excessive x-ray leakage radiation originating from unexpected tar-
gets where a tuned beam would not normally impinge.
Another approach was to join two TW structures in cascade as shown in
Figure 2.31. The microwave power is divided between them and the power
P
Cu
P
Cu
f
0
level and phase made variable for the section. The electron bunches accel-
erated in the first section are stably located just forward of the advancing
wave crest as shown in Figure 2.24(b). By suitably adjusting the phase and
power to the second section, they may be further accelerated or de-acceler-
ated. Since the electron velocity is virtually independent of electron energy
in this section, a stable position of the bunch on the traveling wave is easier
to maintain. When de-accelerated in the second section, these bunches are
less stable in position on the wave as explained below so that both an at-
tenuator and phase shifter are used to optimize performance.
D. Phase Con-
siderations
The operation of the two-section linac can be understood in terms of the po-
sition of an electron bunch with respect to the phase of the microwave E
field being experienced. Figure 2.32 illustrates one cycle of the 3000-MHz E
field and several positions where an electron, or bunch of electrons, travel-
ing at a velocity v
e
, might be located and borne along at the phase velocity
of the wave v
p
. Typically, electrons are accelerated at positions 1, 2, and 3;
they are de-accelerated at positions 4, 5, and 6. In general, the phase, or
phase angle, refers to a position along the wave here expressed in degrees.
By convention, an electron at position 1 at 90 is said to be in phase and at
position 5 at 270 as out of phase.
The rate of energy gain or loss by an electron is determined by the electric
field E it is experiencing and therefore by its position along the wave, i.e.,
the phase, and the amplitude of the wave. By shifting the phase of the
wave, we move the electron bunch to a new position along the wave with a
higher or lower value of E. By keeping the phase constant and varying the
amplitude of the wave, the magnitude of the E field being experienced by
the electron can be altered.
The E field pattern recurs along the beam axis twice as frequently as in TW
structures; that is, an accelerating electron bunch is contained in every
second cavity instead of every fourth cavity. In SW structures, all the cavi-
Figure 2.31. Variable energy linac composed of two TW structures in
cascade.
KLYS-
TRON
LOAD
CIRCULATOR
POWER
DIVIDER
PHASE
SHIFTER
ELECTRON
GUN
ACCELERATOR
TW SECTION I
ACCELERATOR
TW SECTION II
ATTENUATOR
MODULATOR
POWER
SUPPLY
LOAD LOAD
EXIT
ELECTRON
BEAM
ties tend to have the same electric field, since the rf power bounces back
and forth from each end of the structure many times (typically 100 times)
to fill the cavities with energy as expressed in E and H fields. Hence, beam
loading of any cavity contributes to the reduction of the E field in all cavi-
ties, including the buncher cavities. As a result, bunching in SW structures
is efficient over only a narrow energy range, typically 5%. Thus, to main-
tain effective bunching action in SW structures, the input rf power must be
increased with an increase in beam current.
The overall energy gained is determined by the integral of the electric field
E over the path length of the electron through the structure. Electrons gain
maximum energy if they are concentrated in a spatially narrow bunch
along the axis at the wave crest position 1. They gain less energy but their
position on the wave is more stable and axial spread minimized if they were
located just forward of the crest at position 2, a position comparable to the
idealized electron bunches of Figure 2.24(b). This axial phase stability of an
electron bunch at position 2 is achieved as follows: An electron which leads
the bunch at position 3 on the wave encounters a smaller E field slowing it
down; if it lags the bunch at position 1 on the wave, it encounters a larger
E field speeding it up. Both effects operate in sequential cavities to ensure
a compact axial bunch and phase stable position as well as a constant ter-
minal energy of the exit beam. This phase stable bunch is often called the
synchronous bunch since it moves in synchrony with the wave.
The two-section linac illustrated in Figure 2.31 depends on first bunching
and accelerating a group of electrons located at or near position 2 in section
I to a preset energy. Then, by adjusting the amplitude and phase of the E
field in section II, we add to or subtract from this energy. In section II, the
energy is added to electron bunches at positions 1, 2, and 3 or subtracted
at positions 4, 5, and 6. The energy subtractive positions 4 and 5 are phase
stable, but position 6 is unstable in phase and tends to spread a low-energy
bunch spatially and, hence, energetically. However, for relativistic elec-
trons over a few MeV in energy, the effect is not large. Beam current losses
are more severe when operating linac section II in the energy subtractive
mode. These lost electrons, which impinge on the structure disks and
walls, are high energy and can give rise to significant anomalous leakage
radiation levels. Typically, a relativistic electron bunch would occupy a 20
Figure 2.32. Electron bunches shown in representative positions on
the microwave E field phase diagram. For simplicity, E is drawn up-
ward.
E+
M
I
C
R
O
W
A
V
E
E

F
I
E
L
D

C
Y
C
L
E
360 180 0
1
2
3
v
e
v
p
4
5
6
PHASE
ANGLE
E
phase angle interval; that is, somewhat longer than the idealized bunches
illustrated in Figure 2.32.
The feasible energy range of the two-section linac depicted in Figure 2.31
extends between the sum and difference values of the two individual struc-
ture energy gains. Representative electron differential energy spectra for
operation with several differences between sections I and II have been re-
ported by Lanzl. Since the first section is operated at constant power level,
the electron bunching action performs well. However, the necessity of two
microwave structures, two speed points, and additional high-power micro-
wave hardware are complicating factors. Moreover, TW structures are in-
herently less efficient than SW structure in that residual power is dissi-
pated in an external, or internal collinear load. The Varian Clinac 35 and
the AECL Therac Sagittaire treatment units are of the TW two-section de-
sign as is an early 550 MeV linac described by Carpender et al.
A hybrid variation consisting of a short backward wave TW structure fol-
lowed by a long SW structure is shown in Figure 2.33. Here, the backward
wave is magnetically coupled by slots placed peripherally in the disks be-
tween cavities. As in the side-coupled structure, the conflicting require-
ments of power transport versus electron acceleration are separated and
optimized independently. The short TW structure is designed for low-power
consumption and good buncher performance. Such backward wave struc-
tures have the unusual property that the phase velocity v
p
at which elec-
trons are bunched and accelerated, is opposite in direction to the direction
of power flow and group velocity v
g
, the velocity at which the structure fills
with energy. Therefore, this TW structure has the electron gun and power
feed on opposite ends of the structure. The klystron power is fed at the
beam output end of the TW structure. The residual power appearing at the
electron injection end is redirected to the center of the SW section. Electron
bunches entering the SW section are traveling at an almost constant veloc-
ity near that of light and little energy spread ensues. Standing wave struc-
tures may be fed at any point along their length, thereby offering more flex-
ibility in mechanical design. However, there is an advantage in feeding at
the mid-region of SW structures, in that absorption of energy in coupled
cavity modes is less troublesome.
Figure 2.33. Hybrid variable energy linac composed of a backward
wave TW section followed by an SW section.
KLYS-
TRON
MODULATOR
POWER
SUPPLY
ELECTRON
GUN
EXIT
ELECTRON
BEAM
LOAD
ATTENUATOR
PHASE
SHIFTER
CIRCULATOR
TW SECTION
(backward wave)
SW SECTION
(side cavity coupled)
E. Clinac 2500 A recent commercial entry, the Clinac 2500 illustrated schematically in
Figure 2.34, features a 1.9-m side-coupled SW structure, wherein one of
the side coupling cavities functions as an energy switch to provide two x-
ray beam energies of 6 and 24 MV. This side cavity, shown inset in Figure
2.34, is changed mechanically to alter the x-ray energy by inserting a
plunger with an attendant change of the E field amplitude or phase. The ac-
celerating E fields in the buncher are unchanged when switching energy
and hence, there is little change in energy spread of the beam. The Clinac
2500 can be viewed as a sophisticated two-section SW linac with an ex-
tremely compact power divider/phase shifter. A 5.5-MW klystron is the mi-
crowave power source. Six electron energies, 622 MeV, are provided.
Figure 2.35(a) illustrates two alternative methods of providing the high and
low x-ray energy modalities in the Clinac 2500 side-coupled linac structure
by employing two different locations of the energy switching side cavity. The
high x-ray energy modality depicted in Figure 2.35(b) is provided by oper-
ating all of the axial accelerating cavities of sections I, II, and III of the
structure approximately in phase (see electron bunch position 2 in Figure
2.32) with the energy switch plunger completely retreated. The average ac-
celerating field E
1
experienced is constant over the entire length of the
structure and the terminal electron beam energy is proportional to the area
under the average E
1
rectangle of Figure 2.35(b).
Figure 2.34. A two x-ray energy linac using an SW structure and an en-
ergy switching side cavity (Clinac 2500, courtesy of Varian Associ-
ates).
ENERGY SWITCHING
SIDE CAVITY
DUAL PHOTON
STANDING WAVE
ACCELERATOR
270 ACHROMATIC
BENDING MAGNET
DEMOUNTABLE GRIDDED
ELECTRON GUN
RF INPUT
COUPLER
ENERGY SWITCHING
SIDE CAVITY
COLLIMATOR
ASSEMBLY
ISOCENTER AXIS
PLUNGER
ACCELERATING
CAVITIES
2 1
In one method of providing the low-energy modality, the energy switching
side cavity is placed at location 1, as in Figure 2.35(a), and its normal
TM
010
mode (E field configuration) is unchanged. The phase is held con-
stant but the amplitude of the E field is reduced by inserting the plunger
half way into the cavity into dashed position #1 of Figure 2.34. Here, the
energy switch operates at location 1 in order to achieve the desired reduc-
tion in amplitude of the E field. Sections I, II, and III all operate in phase
but the amplitude of the E field is reduced from E
1
, in section I to E
2
in sec-
tions II and III as shown in Figure 2.35(c). The terminal energy is repre-
sented by the sum of areas under E
1
and E
2
.
Figure 2.35(d) illustrates an alternate method of providing a low-energy
modality wherein the energy switch is placed at location 2 in Figure 2.35(a).
Here, the normal TM
010
E field configuration of the energy switching side
cavity is changed to the TEM mode by inserting the plunger fully into the
cavity in the dashed position #2 of Figure 2.34. The effect of this is to pre-
serve the amplitude, but to reverse the phase in the axial accelerating cav-
ities which follow it by 180. The average electric E field now varies as
shown in Figure 2.35(d). As a result, the terminal energy is now the energy
gained in sections I and II, represented by the rectangle under E
1
minus
that lost in section III represented by the rectangle under E
2
.
F. Multiple-
pass Linacs;
Therac 25
Recently, two linac designs have been introduced in which the high-energy
requirement is achieved by passing the beam two or three times through
the same accelerating structure. Such multiple-pass linacs could be con-
Figure 2.35. Energy switching side-coupled SW structure and average
E field distributions for operating conditions described in the text.
ENERGY SWITCHING SIDE CAVITY
ELECTRON GUN
LOCATION 1
LOCATION 2
EXIT
ELECTRON
BEAM
I II III
SW SIDE-COUPLED STRUCTURE
(a)
(b)
(c)
A
V
E
R
A
G
E
F
I
E
L
D
A
V
E
R
A
G
E
F
I
E
L
D
E
1
E
1
E
1
E
2
E
2
A
V
E
R
A
G
E
F
I
E
L
D
(d)
strued as microtrons, but their geometry, construction, and isocentric
mounting are similar to that of conventional linacs.
Figure 2.36 illustrates one such unit described by Froelich. It employs ei-
ther one or three passes through a 0.7-m side-coupled SW accelerating
structure (2) and is magnetron powered.
An energy gain of 28 MeV/pass provides readily adjustable electron ener-
gies from 224 MeV and x-rays of 6, 12, and 20 MV. One distinctive feature
of this unit is the hollow cathode annular electron gun (1) which permits
passage of the returning electron beam in the three-pass configuration.
Such electron guns tend to have higher emittance than their axial counter-
parts; that is, beam angular divergence is greater and beam cross section is
larger for a given beam current. Another feature is the unique beam return
butterfly magnets (3) so-called because of the butterfly-shaped electron
trajectories through them. These magnets are constructed in three sections
to provide the 180 beam turnaround and have shaped pole edges to pro-
vide transverse focusing and isochrony so that the electron bunch is repro-
duced after traversing the magnet.
A somewhat similar commercial unit, the AECL Therac 25 shown in Figure
2.37(a), employs two passes of the beam, a hollow cathode gun and one re-
turn magnet. The 1.2-m SW structure is of the bi-periodic type shown in
Figure 2.37(c) [and Figure 2.30(c)] and incorporates thin pancake-like, on-
axis coupling cavities. It provides a 25-MV x-ray beam and eight electron
energies between 5 and 25 MeV. The Therac 25 is shortened because of the
double-pass design and fits into a shorter room than some high-energy
units. The unit is powered by a 2.6-MW magnetron. The beam energy is
varied by adjusting the distance of the 180 beam return magnet from the
accelerating structure over a short interval of about 2.5 cm. This alters the
phase of the accelerating E field encountered by the returning beam and
hence, the energy gain of the beam during the second pass as described for
Figure 2.32. The 2.5-cm adjustment, )x, of the four-sector, doubly achro-
matic, isochronous return magnet provides approximately a 0 180 phase
Figure 2.36. (a) A multiple-pass linac or shuttle microtron which in-
volves either one or three traversals of the electron beam through the
SW structure; (b) multiple-pass linac mounted in an isocentric gantry
(after Froelich)
2
3
1
3
0 20 40 cm
SCALE
(a)
(b)
0 0.5 1.0 m
SCALE
angle range of the E field for the return passage of an electron bunch (see
Sec. V.).
The treatment head consists of a conventional four-jaw movable collimator
suspended below a turntable that has three positions corresponding to the
three modes of operation: photon therapy, electron therapy, and field illu-
mination. The nominal 270 magnet system adjacent to the treatment head
incorporates two dipole magnet sectors. After traversing this 270 magnet,
the electron beam is scanned spirally at two fields per second on a scatterer
using a quadrupole magnet. Although not providing a low energy x-ray
mode, the Therac 25 provides high electron energies from a comparatively
short structure using a magnetron, a relatively inexpensive microwave
power source. The use of aluminum coated polyamide films (for example,
Kapton, DuPont Company, Industrial Film Division, Wilmington, Delaware
19898) in the monitor ionization chamber is expected to significantly pro-
long its useful life since these films appear to have good electrical proper-
ties and outstanding resistance to radiation damage.
IV. Microtrons The microtron is an electron accelerator which combines the principles of
the electron linac and the cyclotron.
Figure 2.37. (a) A double-pass linac incorporating a single reflecting
magnet; (b) half-cavity machined segment showing beam aperture and
coupling slots; (c) cross section of biperiodic cavity structure (Therac
25, courtesy of Atomic Energy of Canada, Limited).
270 BENDING MAGNET
PRIMARY COLLIMATOR
ELECTRON
GUN/INJECTOR
SYSTEM
(a)
(b)
BEAM APERTURE
ADJUSTABLE
COLLIMATOR
X-RAY
COUPLING SLOT
MACHINED SEGMENT
PANCAKE COUPLING
CAVITY
ACCELERATING CAVITY
(c)
x
REFLECTING MAGNET
DOUBLE PASS
ACCELERATING WAVE GUIDE
A. Introduc-
tion, Circular
and Racetrack
Designs
In the circular microtron, the electron gains energy from a microwave cav-
ity (sometimes called a resonator) and describes circular orbits of increas-
ing radius in a uniform magnet field. The cavity voltage, frequency, and
magnet field are so adjusted that after each transit through the cavity, the
electron gains sufficient energy so that its transit time in the magnet field
increases by an integral number of microwave cycles. An increase in energy
gain per orbit can be achieved by placing the cathode inside the cavity and
allowing the electron beam to be pre-accelerated before reaching the en-
trance hole of the resonant cavity for the first time. The microtron exhibits
axial phase stability characteristics similar to those of accelerating elec-
trons in linacs as described in Sec. III.-D., but with a narrower energy
spread.
Splitting the magnet into two D-shaped pole pieces and separating them
provide greater flexibility in achieving efficient electron injection and higher
energy gain per orbit through the use of multi-cavity accelerating struc-
tures. This configuration, called a racetrack microtron, consists of two
semicircular and two straight section orbits. One early racetrack microtron
designed for radiotherapy provided 1.5 to 15 MeV electron beams through
a common exit portal using from 1 to 6 accelerating orbits. The small beam
emittance of microtrons (product of beam radius and divergence) and min-
imal energy spread simplifies the beam transport system and encourages
the use of a single microtron for supplying several treatment rooms includ-
ing, for example, an intraoperative electron beam for an operating suite. An
additional advantage may be (depending on the energy and design param-
eters) attainment of high energies in a smaller system volume when com-
pared with linear accelerators, due to the cubical geometry characteristic of
microtron units. Depending upon their energy and design, microtrons may
require large heavy iron magnet poles and a high degree of field uniformity.
For the circular microtron, the required magnet volume, and hence cost,
grows as the energy E
3
. In recent years, modifications of the conventional
microwave accelerating structure and improvements in electron gun injec-
tion methods and magnet designs have opened up new possibilities for mi-
crotrons in radiation therapy.
B. Circular
Microtron
22 MeV
A novel 22-MeV circular microtron for radiation therapy, the Scanditronix
MM 22, shown in Figure 2.38, has been described by Svensson et al. Ex-
traction of the electron beam is made through a movable steel deflection
tube which can be moved to select any of the orbits between number 10
and 42, with all the electron energies emerging along a single axis. The de-
flection tube displaces any selected orbit downward by a constant distance.
As a result, the electron beam exits through the fixed horizontal extraction
tube as shown in Figure 2.38, instead of continuing through the resonator
cavity.
The electron energy gain per turn is approximately 535 keV and the energy
spread of the exit beam is about 35 keV [full width at half maximum
(FWHM)]. The magnet poles for this circular microtron are 1.8 m in diame-
ter, and a magnetic field uniformity of 1 part in 10,000 is required. Either
a klystron or magnetron can be used as the microwave power source. Two
x-ray energy options among 6 or 10 and 20 MV, are available and ten elec-
tron energies ranging from 222 MeV. A composite x-ray flattening filter is
employed to improve depth dose characteristics; high atomic number Z in
the center and low Z on the periphery. Two separated scattering foils are
employed for electron therapy; a high-Z primary foil of constant thickness
is used to spread the beam followed by a low-Z secondary foil of variable ra-
dial thickness to flatten the electron fields for sizes up to 40 cm in diameter
at the isocenter. Capability of uncoupling one x-ray and electron collimator
jaw along the beam axis permits easy abutting of x-ray, electrons, or x-ray
to electron treatment fields without divergence. In one configuration, a
Scanditronix MM 22 microtron feeds two isocentric treatment units and
also provides a research beam. Being stationary, this microtron is more
easily maintained, magnetron lifetime may be increased but electron gun
replacement may be more frequent as compared to a linac.
C. Racetrack
Microtron
50 MeV
In one newly constructed unit (Figure 2.39), a six-cavity accelerating struc-
ture rather than a single cavity is used between the separated pole pieces to
provide energy gains of 5 MeV per orbit. An exit beam energy ranging from
550 MeV for electrons and photons is provided. The beam energy is
changed by moving an extraction magnet in or out to select the appropriate
orbit for the desired energy as shown in Figure 2.39. The extracted beam
exits along a common beam line independent of energy. A thorough de-
scription of the development of the 50-MeV microtron prototype is provided
by Rosander et al.
Attaining desirable radiation beam characteristics for electrons and x-rays
becomes more difficult at higher energies. The microtron, by combining its
inherent small beam emittance and energy spread with a scanned pencil
beam such as described earlier, may provide one approach to improving
beam characteristics. Here, the scanned pencil beam of electrons could fa-
Figure 2.38. Cross section of a 22-MeV microtron illustrating the path
of electrons in the accelerator (left) and through an isocentric treat-
ment unit (right). The range of motion of the deflection tube from orbit
1042 is shown by arrows on the dashed line.
MOVABLE DEFLECTION TUBE
ELECTRON
BUNCHES
ELECTRON
GUN
RESONATOR
WAVEGUIDE EXTRACTED BEAM
cilitate the provisions of large flat fields of electrons and x-rays. The 50 MeV
as well as the MM 22 microtron designs involve moving either an extraction
magnet or a steel deflection tube, respectively, within the evacuated system
and this feature could present maintenance problems.
V. Beam
Transport
Magnet
Systems
The beam transport magnet system includes solenoids and steering coils
over the accelerator structure, together with focusing quadrupoles and
bending magnets after the accelerator structure, as well as associated
power supplies.
A. Introduction The system confines, steers, and guides the electron beam from injection to
the x-ray target or electron scatterer. For simplicity, magnet energizing
coils and magnetic return paths, where used, have been omitted in many of
the illustrations which follow. A fundamental design consideration for opti-
mal performance is matching the beam transport system capabilities to the
characteristics of the linac beam.
Most treatment units are isocentric, a circumstance necessitating bending
magnets in higher energy machines so that the correspondingly longer
structure can lie horizontally in energy machines. However, the introduc-
tion of short SW structures has led to straight-through isocentric 4and 6-
MV units having a 100-cm source-axis distance (SAD) with moderate iso-
center and room heights. No bending magnets are required, and the struc-
ture is so short that a solenoid need not be used to confine the beam. How-
ever, such straight-through designs let low-energy electrons strike the x-
ray target, so the peak electron energy must be slightly higher than in a
bent-beam machine to achieve the same x-ray penetration.
The beam transport systems of bent-beam medical linacs must adequately
respond to operating conditions which affect beam stability and symmetry.
Asymmetries arise when the primary x-ray lobe does not strike the flatten-
Figure 2.39. Racetrack microtron in 550 MeV range for electron and x-ray
therapy.
MAGNET MAGNET
movable
extraction magnet
electron bunches
linear accelerator
extracted beam
electron gun
ing filter symmetrically. This can be due to a change in the angle of inci-
dence of the electron beam on the target or due to the lateral displacement
of the position of the electron beam at the target (see Figure 2.40). These
changes can result from changes in beam energy because of a change in
microwave power or microwave frequency as well as from other sources
such as mechanical strains associated with motions of the gantry or tem-
perature variations as well as from the effects of stray magnetic fields. Such
changes occur more often in the bending plane than in the transverse
plane. Automatic frequency control and other feedback control devices,
based on sampling ionization chambers, have become more sophisticated
in limiting adverse response, (Sec. VI.-B.).
B. Definitions
and Conven-
tions
Magnets bend (deflect) and separate (disperse) particles of differing mo-
menta. These two actions are necessarily intermixed and we often wish to
emphasize one action, for example, deflection in bending a small beam of
electrons onto an x-ray target. Medical linacs may employ only one dipole
magnet but at times two or more magnets are used to better separate the
two actions. The electron beam which enters a beam transport system con-
tains a spectrum of particles which must be restricted in momentum (or
energy) range. Spatially, their trajectories may diverge from the central axis
and be laterally displaced from it. The task of the transport system is to
Figure 2.40. Effect on resultant x-ray dose distribution at D
max
of mis-
alignment between electron beam and flattening filter axis: (a) sym-
metrical flat field for a correctly aligned x-ray flattening filter, (b)
asymmetrical field for an angular divergence )N, and (c) asymmetrical
field for a radial displacement )r. At high energies, these effects can
be pronounced because of the very peaked filter and high attenuation
gradient as a function of its radius.
ELECTRON BEAM
TARGET
FILTER
(a)
(b) (c)
RESULTANT
DOSE AT D
max
r
bring this diversity of particles to a small and coaxially directed beam of
particles on the central axis of the x-ray target or the scattering foil.
The literature readily applicable to medical linac magnets is modest in ex-
tent. Some analyses make use of matrix methods and often differing con-
ventions, notations, and terminology are used. Many phenomena involved
with the transport of charged particles in magnetic fields can be described
with first-order theory in which quantities involving the products of two or
more small differentials of momentum or length are ignored. The theory is
limited to particle beams with small energy spread and small angular and
spatial spread. A general solution to the electron transport problem in-
volves a second-order differential equation. Its two orthogonal solutions are
sine- and cosine-like functions and all possible particle trajectories are
some linear combination of them. A number of definitions follow which are
useful in describing and understanding the performance of beam transport
systems by these as well as less sophisticated methods.
An electron of charge e and relativistic mass m moving with a velocity v at
right angles to and through a uniform field of strength B experiences a
magnetic force B
ev
perpendicular to its instantaneous direction of motion.
It gains no energy from such a static field but is constrained to travel in the
arc of a circle of radius D. The requisite centripetal force for such motion,
mv
2
/D, is provided by the magnetic field such that: mv
2
/D = B
ev
or BD = mv/
e. The quantity BD measures the stiffness of the beam and is called the
magnetic rigidity. For example, a 15,000-G dipole field will bend a 25-MeV
electron beam with a radius of 5.66 cm. A 5-MeV electron beam can be bent
with the same radius by a field of 3240 G.
In addition to supplying the centripetal force to bend the beam through an
angle, the bending magnet system provides forces to focus the beam. These
focusing forces can be provided by tilting the pole faces with respect to each
other to curve the field lines throughout the magnet or by tilting the pole
edges to employ the curved fringe field lines at the magnet entrance and
exit. Tilting the pole faces produces a radial gradient n of the magnetic field
defined by
. (8)
Some magnets provide such a non-uniform transverse field in which the
magnetic field increases (n < 0) or decreases (n > 0) with bending radius of
the particle. However, most dipole magnets provide a uniform field trans-
verse to the bending plane and are characterized by a field gradient value of
n = 0. In either case, the radial focusing (often analyzed in terms of the hor-
izontal or axial motion) can be separated to first order from the transverse
(vertical) focusing. The strong focusing properties of quadrupoles may also
be employed.
Electrons whose momenta differ from a central value P
0
are deviated from
the central orbit by a magnet. This phenomenon is termed energy disper-
sion. The term achromatic describes systems which bring particles of differ-
ing energies, which were originally paraxial, to the same focus. In a zero
dispersion system (sometimes called singly achromatic), the particles di-
verge after reaching this focal point. In an achromatic system (sometimes
called doubly achromatic), the particles do not diverge but remain parallel
after reaching this focal point. Double focusing refers to an ability to focus
both the transverse and radial image components at the same point along
the central trajectory. A double-focused image is called stigmatic. The term
energy focus is sometimes used for the momentum focus associated with
n

B
----
B
x
------- - =
dispersion and the term spatial focus is used for the radial and transverse
focusing associated with the divergences and lateral displacement of parti-
cles with respect to the entering central ray. Divergence refers to the angu-
lar departure )n of the beam from a central axis or trajectory, displacement
refers to the lateral positional departure )r from this axis. The latter two
quantities are illustrated in Figure 2.40. The return magnets of multiple-
pass linacs or microtrons must be isochronous; that is, the length of time
required to traverse the magnet is independent of electron energy. This en-
sures that the electrons comprising the bunch are not further separated in
time and hence, in phase and energy on subsequent passes. A central orbit
reference trajectory is defined herein by particles of momentum P
0
which
enter the transport system axially and are deflected along a central orbit of
radius D
0
passing through the median plane between poles [see Figure
2.43(b)].
It is convenient in examining beam transport to combine two main charac-
teristics of a beam, its radius and its divergence, into a single parameter,
the phase space area, which is termed emittance. For a constant velocity
beam and with the reasonable proviso of no coupling between coordinate
motions, the phase space is invariant in time as the beam moves through
the transport system. Defining z along the central orbit reference trajec-
tory, we can separate the transverse perturbation motion into small )x and
)y displacements with respect to this coordinate. We define momentum co-
ordinates P
x
and P
y
which are proportional to their divergences from the z
axis. thus: x = dx/dz and y = dy/dz. The ensemble of particles moving
through the system has a constant phase space for both x and y and can be
defined at any position z in terms of the maximum excursions of their dis-
placement and divergence coordinates at that position. This phase space
can be expressed as an ellipse whose shape varies as the beam progresses
through the system but whose area remains constant. The individual par-
ticles oscillate about the equilibrium orbit as representative points in phase
space. We call the four-dimensional space the emittance g, with
(9)
An important corollary is that if we focus the beam to a smaller diameter,
its divergence will increase and vice versa, but its emittance will remain
constant. We can also reduce the diameter of a beam, and consequently the
current and emittance, by intercepting a portion of it. Acceptance is a com-
plementary term to emittance and refers to a characteristic of an equip-
ment, a measure of its ability to accept and transport a beam rather than a
property of the beam. For maximum efficiency, there must be a good match
between the emittance of the entrant beam and the acceptance of the
transport system, its counterpart. Unless the acceptance of the beam
transport system is at least as large as the emittance of the entrant beam,
some of the particles will not be transmitted. Beams characterized by small
values of emittance are more readily transported over long distances.
Most isocentric treatment units employ a nominal 90 or 270 beam-bend-
ing magnet with the accelerator guide structure mounted approximately
horizontally in the gantry. The 270 magnet systems are usually achro-
matic and the resulting isocenter height is acceptably low. However, when
a 90 magnet system is made achromatic, two or more magnets are in-
volved. The beam path through it becomes long and the accelerator struc-
ture axis may lie well above the x-ray target. This can raise the isocenter
height excessively in a high-energy machine. The sections which follow de-
scribe specific magnet systems.
xxyy
2
=
C. 90 Magnets Figure 2.41 illustrates the effect of a simple 90 dipole magnet on the exit
beam for entrant beams having an energy spread E about E
0
or a radial
displacement )x or a divergence )n. These aberrations, which are easiest to
understand individually in the radial plane of simple 90 magnets can
appear in combination and are also present in the transverse plane of mag-
nets. They may, in part, be corrected in more complex magnet systems.
Typically, the beam energy spread is restricted to 10% or less about its
central value. The trajectories of the low, central, and high-energy compo-
nents in Figures 3.40(a) and 3.41(a) are denoted by 1, c, and h, respective-
ly.
A single 90 bend magnet such as shown in Figure 2.42(a) is not achromat-
ic. It can bend a mono-energetic beam on axis to a point at the x-ray target
but the spread of energies of the actual beam results in a spread of such
focal points at the x-ray target as shown. This spreading effect can be min-
imized by reducing the bending radius of the magnet, restricting the emit-
tance and the energy spread of the beam, stabilizing the operation of com-
ponents which affect beam energy or by incorporating a second magnet to
provide focusing. However, even with these precautions, changes of energy
as well as variation of the angle and position of the entrant electron beam
will produce detrimental asymmetries in the exit beam and treatment field
in the non-achromatic 90 systems more readily than in the achromatic
270 systems. The principal effect of an energy change of the entrant beam
in a non-achromatic 90 magnet system is a lateral displacement at the tar-
get although a secondary effect could be the appearance of new leakage
foci. To be effective, energy controlling slits must be located near the output
of a 90 magnet, but here they tend to become enlarging the effective focal
spot size. A uniform field 90 achromatic bending magnet system suggested
by K. L. Brown is described by Penner. It consists of a single quadruple
magnet positioned in the plane of symmetry between two 45 deflection
magnets. The magnet system is achromatic and double focusing but its
design requires a high isocenter. The performance of the 90 bent-beam
Clinac 6 treatment unit has been reported by Horsley and its magnet
described by Avery.
Figure 2.41. Effect of simple 90 dipole magnet deflection system on
entrant beams which are (a) axial and nondivergent with E = E
0
)E, (b)
nondivergent with E constant but parallel and displaced )X above and
below the entrant center line, and (c) axial with E constant but diverg-
ing )n above and below the equilibrium orbit. The electron beam en-
trant center line is denoted by c. The electron trajectories shown were
constructed using straight-line entry and exit paths connected tan-
gentially to arcs of constant radius when the electron is in the dipole
magnet field.
ELECTRON BEAM
MAGNET
POLE
c l h
(c) (b) (a)
x
x
c c c
c c
E
0
E E
0
+E E
0
D. 270 Mag-
nets
Figure 2.42(b) illustrates a 270 triple-focus, zero dispersion (singly achro-
matic) uniform field magnet wherein a 10% beam energy spread is
brought to a single focal point on the target, in part, by choice of the angle
of the entrant and exit pole faces. The higher energy component is deflected
through a circle of larger radius, the lower energy component l through a
circle of smaller radius, but both converge on the target at the same point
on the central energy trajectory c. However, a change in mean energy of the
beam from the accelerator structure will result in a change in mean angle
of the beam at the target and hence to asymmetry in the x-ray field.
Figure 2.42. Nominal (a) 90 and (b) 270 bending dipole magnets with
representative electron orbits. Trajectory c corresponds to the central
energy E
0
, trajectories l and h correspond to E
0
10% and E
0
+10%, re-
spectively. One type of 270 achromatic magnet is shown in (c) and (d)
and another type in (e) and (f).
ELECTRON
BEAM
TARGET l c h
MAGNET POLE
ELECTRON
BEAM
TARGET
(a) (b)
MAGNET
POLE
MAGNET
SECTORS
TARGET
MAGNET
POLE
(c) (d)
ELECTRON
BEAM
MAGNET POLES
MAGNET POLES
SECTION d -d
1 2
(e) (f)
a
2
d
1
d
2
a
1
TARGET
ELECTRON
BEAM
d
1
d
2
a
1
a
2
S
1
S
2
Y X
Z
l c h
Z
X
S
1
S
2
h
c
l
MAGNET POLE
SECTION
d -d
1 2
SLITS
B = B = 0
X Y
B = G
Z X
n
Magnet edges contribute important focusing properties. In one approxima-
tion, the transverse field can be assumed to begin and end abruptly at an
edge near where the beam enters and where it leaves the magnet. A beam
of particles entering normal to such an edge will be radially converged. The
edge may be angled other than at 90 with respect to the particle beam and
particles will be more rapidly converged (focused) or less rapidly converged
(defocused) in the radial plane. The fringing magnetic field which exists be-
yond the pole edges, provides additional focusing for an angled pole edge.
Here, a radial field component exists which interacts with the azimuthal
particle velocity to produce a transverse (vertical) force. This force will be fo-
cusing (toward the central orbit plane) or defocusing (away from it) depend-
ing on whether the pole edge angle, with respect to the beam, is greater or
less than 90. If the edge tilt produces focusing in the transverse plane,
then it produces defocusing in the radial direction, and vice versa. As in the
case of magnetic quadrupoles, transverse focusing and radial defocusing
go together and vice versa. The pole edge angles chosen in Figures 3.42(b)
and 3.42(c) provide this edge focusing effect. Cross has shown that some
judicious choices of angles can result in two-directional focusing in both
the radial and transverse directions.
In order to minimize distortions in radiation field flatness due to changes in
beam energy, it is desirable to employ an achromatic magnet system. Fig-
ure 2.42(c) illustrates one way to build a 270 achromatic magnet system
by combining uniform and non-uniform field regions in the same magnet. It
focuses a range of entrant momenta, and an input configuration of lateral
displacements and angular divergences of the electron beam, to an opti-
cally similar configuration at the output focal plane. Its entrant pole face
angles, a
1
, and a
2
, can be adjusted for optimal radial and transverse focus-
ing of each mono-energetic bundle of rays at the d
1
d
2
plane. In addition,
two adjustable pole sections of the magnet, shown in section view d
1
d
2

[Figure 2.42(d)] provide an adjustable radial field gradient with n > 0. This
radial gradient is controlled by adjusting angle a
3
[Figure 2.42(d)] to re-con-
verge the different energy bundles of rays into a single spot at the x-ray tar-
get, the distribution of rays in this spot being the same as in the beam cross
section which enters the magnet. The Siemens Mevatron treatment units
employ this type of bending magnet. Establishing and adequately preserv-
ing the electron beam position and direction at the x-ray target or scatter-
ing foil to maintain a symmetrical radiation field become more difficult for
higher energy linacs which operate over a wide range of energies. High-en-
ergy treatment units incorporate as many as six discrete magnets in their
beam transport system. Because of iron hysteresis and saturation effects, a
programmed sequence of magnet current changes may be employed to re-
liably and accurately establish or change a given beam energy or modality.
A three element, double focusing, 270 achromatic magnet system has
been described by Hutcheon and Heighway. It employs an input quadru-
pole doublet and a 180 uniform field magnet separated by a short drift
space from a 90 uniform field output magnet to provide a nominal 270
system. The system minimizes the magnet dimensions in the direction op-
posite to the exit trajectory, facilitating a lower isocenter height. It is suit-
able for electron beam energies from 5 to 50 MeV, an overall energy spread
of 20% and a total bending angle between 230 and 290. The AECL Therac
25 treatment unit employs this type of bending magnet.
E. Mirror 270
Magnet
The 270 magnet illustrated in Figures 3.42(e) and 3.42(f) is a design of
Enge. It is sometimes called a loop or pretzel magnet but more often an
achromatic magnetic mirror since particles which traverse it appear to be
reflected from the yz surface at the entry point. It is employed in the
Brown-Boveri Dynaray treatment units. Typically, its non-uniform field in-
creases with x having an n = 1 exponent value near the edge and n = 0.8
elsewhere and angle a
1
= a
2
= 45. The xy proportions of the trajectory
loops are constant independent of momentum, i.e., the loops have the same
shape, and all particles exit at the origin independent of momentum, i.e.,
there is no dispersion and the deflection is achromatic. This magnet is ca-
pable of faithfully focusing a wide range of momenta as limited by slits S
1

and S
2
placed on the plane of symmetry [see section d
1
d
2
in Figure
2.42(f)]. The Varian Clinac 2500 employs a modified loop magnet wherein a
simple two-step gap replaces the smoothly varying gap along the x axis
[Figures 3.42(e) and 3.42(f)]. The beam trajectory in the modified loop mag-
net has been described by Tronc. The Dynaray-4, employing a 270 mir-
ror magnet is described by Sutherland.
F. Three-sec-
tor 270 Mag-
net
The illustrations comprising Figure 2.43 describe the nominal 270 magnet
system used in the Varian Clinac 18 treatment unit and are shown
mounted in the treatment head in Figure 2.45. Figure 2.43(a) is a cross-
sectional view of this magnet in the median bending plane of the central or-
bit. It incorporates three uniform field magnet sectors, or pole sets, M
1
, M
2
,
and M
3
, with short drift tubes connecting them. A magnetic shunt between
poles (not shown) provides a relatively magnetic field free region for passage
of the particles between sectors.
The emittance of the beam entering the magnet system is typical of the out-
put beam of an electron linac in terms of cross-sectional are, divergence,
and energy spread. The performance of the system is analyzed with respect
to a particle which enters along the central axis with reference momentum
P
0
and whose central orbit reference trajectory is shown as a heavy dashed
line in Figures 3.43(a) and 3.43(b). It has been shown by Brown that the
properties of such a system are completely determined to second order by
specifying five representative trajectories or paths relative to the reference
trajectory. Spatial departures from the reference trajectory by particles of
reference momentum P
0
are separated into orthogonal radial and trans-
verse components for initially axial but divergent trajectories S
x,y
and for
initially parallel but displaced trajectories C
x,y
. The momentum trajectories
D
x
dispersed from the reference trajectory in the median plane are for par-
ticles initially axial and with momenta within )P of reference momentum
P
0
.
Two of these trajectories, shown in Figure 2.43(a), depict divergence from
the central orbit (S
x
) and lateral displacement from the central orbit (C
x
) in
the median plane. Figure 2.43(b) is a simplified view similar to Figure
2.43(a) depicting trajectory D
x
, of momentum dispersed particles initially
on the central axis. The trajectories of all particles through the system are
symmetrical about a plane of asymmetry located at 135 midway along and
normal to the reference trajectory. As shown, the energy selection slits S
1

and S
2
which are placed at )P about radius D
0
of the reference trajectory
intercept all particles except a narrow momentum band )P centered about
P
0
. An end view of this magnet system in the transverse plane is shown in
Figure 2.43(c). The entering angular divergence S
x
and lateral displacement
C
x
in the beam cross-section from the accelerator are reproduced at the
target plane as shown in Figure 2.43(a) with no significant increase in their
magnitude. Figure 2.43(d) illustrates focusing properties in the transverse
plane due to the fringing fields of the shaped pole faces along and near the
reference trajectory and depict angular divergence S
y
and lateral displace-
ment C
y
. Again, these transverse divergences and displacements are repro-
duced at the target with no significant increase in their magnitude. The en-
tering radial and transverse slop and displacement perturbations are equal
in magnitude and remain uncoupled in the transport system to first order.
Since their dimension in the transverse plane is equal to that in the bend-
ing plane, the focal spot is circularly symmetric. This system is achromatic
since both the spatial dispersion and its derivative are zero at the output
plane.
The energy determining slits shown in Figures 3.43(a) and 3.43(b) are
sources of leakage radiation from stopped electrons. However, this
bremsstrahlung is directed away from the isocenter and not contributory to
patient exposure. A magnetic analysis of the electron treatment beams
from a Clinac 18 gave energy dispersion values of 0.4 to 0.7 MeV FWHM
over the range 618 MeV, respectively. These dispersion values include the
effect of scatter from the linac thin window and 1 m of air. A description of
magnetic and threshold techniques for energy calibration of high-energy
radiations has been given by Lanzl.
Figure 2.43. Nominal 270 achromatic bending magnet system: (a:
cross-section view in the radial (bending) plane of the orbit; (b) simpli-
fied view in same plane showing momentum dispersion and the use of
energy selection slits to limit the momentum to a narrow band P
about the central reference momentum P
0
; (c) transverse section end
view of the magnet system; (d) central orbit trajectories (simplified)
along transverse section.
MAGNET
COIL
MAGNET
SECTOR
PLANE OF
SYMMETRY
SLITS
CENTRAL ORBIT
REFERENCE
TRAJECTORY
(a)
TARGET
M
2
S
2
BEAM
APERTURES
(c)
MAGNET COIL
VACUUM
CHAMBER
SLITS
(b)
PLANE OF
SYMMETRY
CENTRAL ORBIT
REFERENCE
TRAJECTORY
ENTRANT
TRAJECTOR
Y
n
n
ENERGY SELECTION
SLITS
(d)
PLANE OF
SYMMETRY
CENTRAL ORBIT
REFERENCE
TRAJECTORY
N N N
S S S
S
Y
C
Y
M
3
S
1
M
1
S
X
C
X
M
S
S
M
M
D
+
+
0
M
3
M
1
C
Y
S
Y
M
1
M
2
M
3
Typically in this magnet system, the energy defining slits are set for 3%
transmission and are sometimes placed between dipoles M
1
and M
2
rather
than at the plane of symmetry as shown in Figure 2.43(a). The energy servo
system, fed by signals from the ionization chamber monitor system de-
scribed in Sec. VI.-B., controls the input microwave power level so as to
maintain constancy of beam energy. The inherent open-loop stability of the
system confines the energy spread to within 1% during a typical treat-
ment, and to within 5% during a treatment day. The closed-loop gain con-
fines the energy spread to within 0.1% during a treatment, and to within
1% during a treatment day.
G. Other Mag-
nets
Panofsky and McIntyre describe an achromatic beam translation system
involving two sector magnets. Their objective is to translate the beam with-
out energy dispersion, in order to dispose of unwanted low-energy elec-
trons. More recently, a system involving two magnetic quadrupoles acting
as magnetic mirrors have been combined with two microwave cavities to
translate, chop, and bunch an electron beam. Here, use is made of an ear-
lier discovery that a magnetic quadrupole with a rectangular aperture can
be provided by current sheets bounded by iron rather than the shaped iron
pole faces of more conventional designs with their angle aperture limita-
tions. The multiple-pass linacs described earlier incorporate unusual 180
turnaround magnet designs which were described there.
Two early radiotherapy beam transport systems not cited in the preceding
sections merit attention. Both are isocentric in design and involve rotation
of the magnet system about a fixed horizontal linac axis. In one system, the
electron beam from a single-section 45-MeV TW linac is first momentum
analyzed by a 45 magnet followed by energy defining slits. The beam is
then redirected by a 135 magnet so as to be perpendicular to the axis of ro-
tation. The magnet system can rotate 45, with respect to the zenith,
about the horizontal linac and patient axis. The second magnet system is
incorporated in the AECL Therac Sagittaire, a two-section TW linac. It in-
volves a similar but more sophisticated magnet system incorporating addi-
tional magnets and permitting 360 rotation around the recumbent patient.
H. Scanned
Pencil Beams
Most beam transport systems provide as output a small pencil beam of
electrons which are incident on an x-ray target or scattering foil in a fixed
position and direction. In an alternative design, the nominal 90 bending
magnet is preceded and followed by a scanning magnet as shown in Figure
2.44.
The two scanning magnets are placed orthogonal to each other. The elec-
trons still impinge on the x-ray target at a fixed point but their angle of ar-
rival is varied by the two scanning magnets in order to sweep the direction
of the x-ray lobe. By suitably varying their magnetic fields, a raster scan of
the x-ray lobe direction can be generated similar to that used in a television
picture tube. Alternatively, a spiral scan can be generated by appropriate
variation of scan magnet currents. An x-ray scanning system incorporating
a quadrupole magnet has been described. By retracting the x-ray target, a
scanned electron therapy beam is obtained. Such scanned electron beams
provide electron treatment fields of good uniformity, low x-ray contamina-
tion and could be programmed for the arbitrarily shaped fields character-
istic of electron therapy. The x-ray lobe scanning feature can largely com-
pensate for the unflatness arising from the angular distribution of
bremsstrahlung production and the variation of energy with angle across
the field. A thin flattening filter can then be used thereby avoiding signifi-
cant changes in the photon energy spectrum in the middle of the field rela-
tive to the edges of the field. This makes it easier to provide large uniform x-
ray fields at high energies. The scanning feature, however, adds complexity
and cost and dosimetry is more difficult. For example, monitor ionization
chamber collection efficiency may be significantly impaired with attendant
non-linearities of the dose monitor. Isodose measurements in a phantom
are more complex because of the scanned nature of the beam.
Scanned beam systems are interlocked so that if scanning becomes inoper-
ative, no significant hazard arises from the stationary air-scattered, high-
energy beams. This pencil beam or electrons is minimally scattered by the
vacuum window of the structure and the intervening air. For low electron
energies, it may be as broad as 10 cm at l00-cm distance, but narrows with
increasing energy.
A study of the scanned electron beam from a Therac 20 accelerator has
been carried out by Pfalzner and Clarke. This unit employs a scanning qua-
drupole with sawtooth modulated currents of 0.615 Hz and 4 Hz, respec-
tively, in the x and y directions. They conclude that the electron beams of
the Therac 20 do not differ appreciably from those of the nearly mono-ener-
getic 22-MeV MM-22 microtron beams, as reported by Svensson, Brahme
et al. A high-energy scanned pencil beam of electrons was used earlier by
Carpender et al. on a 550 MeV linac. This two-section linac was equipped
with a novel magnet system for scanning the pencil beam of electron.
VI. Treatment
Head
The characteristics of electron and x-ray treatment beams are strongly
influenced by the design of the x-ray treatment head.
A. Introduction A representative design is illustrated in Figure 2.45. In addition to a bend-
ing magnet, fixed shielding, and large movable collimators, the head con-
tains the x-ray target, flattening filter, and in some cases, dual electron
scattering foils, often mounted on a large carousel. Included also is a field
Figure 2.44. Scanned pencil beam provided by a three-magnet system
(adapted from Brahme)
BENDING
MAGNET
ELECTRON
BEAM
SCANNING
MAGNETS
(a) (b)
light with a sizable 45 mirror for illumination of large field sizes and optical
distance indicator optics together with an increasingly large diameter and
complex ionization chamber assemblies for monitoring and control. Acces-
sibility for service often becomes difficult, but a recent swing away colli-
mator design may significantly improve access. An earlier Therac 20 head
design allows one of the magnet half yokes and associated shielding to be
hinged for easy access.
The desire for larger field sizes, improved beam characteristics and conve-
nient, functional accessories has led to a number of studies and improve-
ments in treatment head design. The trend from 46 MeV to higher energy
x-ray treatment beams, larger field sizes and the increasing use of accesso-
ries have resulted in an increase in treatment head size in terms of its
height above the accelerator structure axis (radially away from isocenter)
and its diameter below it. The former stems from the incorporation of the
larger 270 bending magnets, thereby increasing the height of both the
treatment head and the isocenter. The increase in diameter stems primarily
from the requirement for large fields. The large treatment head diameter
may interfere with optimal placement of the treatment beam in some ana-
tomical regions such as lateral breast. Increased length of the treatment
head toward the isocenter may limit the use of outboard accessories for
short SAD units, such as the Clinac 4, as well as increase the amount of
Figure 2.45. A representative linac treatment head (Clinac 18) with
major components identified.
BENDING MAGNET ASSEMBLY
ELECTRON ORBIT
FLATTENING
FILTER
SCATTERING
FOILS
DUAL IONIZATION
CHAMBER
FIELD DEFINING
LIGHT
RANGE FINDER
X-RAY TARGET
(RETRACTABLE)
COLLIMATORS
scatter radiation reaching the patient from the head. Large, heavy, movable
collimator jaws are needed for megavoltage x-ray beams with field sizes
variable from zero area to as much as 40 40 cm at 1 m from the source.
High-energy isocentric machines incorporate large bending magnets. These
must be shielded extensively to attenuate radiation arising from their en-
ergy selection slits and elsewhere. A high-density, high atomic number ma-
terial such as uranium (usable up to about 6 MeV), tungsten, or lead are
frequently used to conserve space and to rapidly attenuate x-rays. When
tungsten or lead are employed for collimators, head shielding, x-ray targets
or flattening filters above 10 MV, photoneutron production increases rap-
idly. In some machines, low-density hydrogenous shielding materials have
been incorporated in the head for neutron shielding, further increasing its
size.
Some 46 MeV units employ short accelerating structures in straight-
through beam designs in both isocentric and non-isocentric treatment
units. However, at 8 MeV and above, longer accelerating structures are re-
quired and are usually incorporated into bent-beam, isocentric units. At
such higher energies, x-ray beam flatness becomes increasingly sensitive
to angular and spatial misalignment of the x-ray lobe with respect to the
axis of the flattening filter. The effects on flatness can be described in terms
of displacement )r and divergence )n of the electron beam with respect to
the axis of the filter as illustrated in Figure 2.40. Such misalignments can
result from small changes in beam energy but are less likely to do so for
achromatic magnet systems which may incorporate narrow energy defining
slits. They are more likely to result from mechanical strains which occur
from the stress of gantry rotation, from beam stopper extension or retrac-
tion, from temperature changes of mechanical and electronic components,
as well as from the presence of nearby ferromagnetic materials. The effect
of such small-energy, displacement and divergence changes on an electron
beam traversing a simple 90 magnet system is illustrated in Figures
3.41(a), 3.41(b), and 3.41(c), respectively. Achromatic magnet systems tend
to limit the effect of such changes in entrant )E, )r and )n changes on the
output beam image and hence, field symmetry and stability. Their output
phase space image is a faithful reproduction of their input phase space im-
age with unity magnification.
Beam energy stability of a few percent is required to ensure satisfactory
constancy of symmetry in bent-beam linacs. Under normal operation, the
energy interval selected by the bending magnet slits is centered about the
current maximum as illustrated inFigure 2.43(b). If the energy of the beam
entering the bending magnet from the accelerator structure shifts, the pri-
mary effect is a reduction in output. A secondary effect is a change in beam
symmetry in the radial plane, the sense of the change depending on which
side of the current maximum the new distribution is centered. Similar, but
far less severe, stability problems apply to electron therapy. Naylor and
Chiveralls have examined the variations in x-ray beam flatness and calibra-
tion with gantry angulation and with time for an 8-MV unit equipped with
a 90 bending magnet. They conclude that such variations are confined to
a few percent when averaged over four-week periods, an interval pertinent
in treatment. Padikal et al. describe a method for assessing the stability of
symmetry with gantry angle rotation. In an early study, Naylor and Will-
iams call attention to the need for frequent symmetry, dose monitor, and
beam energy checks of linac treatment units. Characteristics of the Meva-
tron 15-MV photon beam have been described by Paul et al. The x-ray and
electron beam dosimetry characteristics, as well as neutron measurements
for a Mevatron 80, are described in a series of recent reports. The treatment
head design for this 20-MeV unit has been described earlier. Similarly, the
x-ray beam characteristic of the Clinac 18 has been reported by Connor et
al., and those of the Therac 20 by Patterson and Shragge Design consider-
ations involving the treatment head and pertinent to electron and x-ray
therapy, neutron production considerations, together with ionization
chamber, dosimetry, and beam steering aspects, follow.
B. Ionization
Chamber,
Dosimetry,
and Beam
Steering Sys-
tem
The monitor ionization chamber of a contemporary, high-energy linac is
constructed of several plates or electrodes, whose areas may be divided into
sectors so as to serve two different monitoring purposes: (1) dosimetry of
the x-ray and electron treatment beams, and (2) monitoring of the intensity
distribution of the radiation field. The resulting signals can be used in au-
tomatic feedback circuits to steer the beam through the accelerator, bend-
ing magnet, and onto the target (or scatterer) in order to ensure beam flat-
ness and symmetry. How these needs are satisfied in one representative
treatment unit, the Varian Clinac 18, is described below and illustrated in
Figure 2.46, a simplified diagram of the ion chamber, dosimetry, and beam
steering system. The transmission ionization chamber, shown in the treat-
ment head diagram of Figure 2.45, subtends the entire useful beam and
provides two independent outputs for the dual dosimetry monitor. Located
just below the flattening filter or scattering foil, it samples the flattened x-
ray or scattered electron beam. This ionization chamber, shown in more de-
tail in Figure 2.46, consists of three polarizing plates and two collecting
plates, the latter divided into four collecting sectors, with the sectors defin-
ing four distinct laminar collecting volumes. A 500-V polarizing voltage is
used with a plate spacing of 1 mm. The two inner D-like sectors provide sig-
nals for both dosimetry and steering and the two outer arclike sectors for
steering only. The dosimetry system monitors and displays readings related
to the quantity and uniformity of the useful beam of radiation. Sutherland
early recommended that integrated dose be based on monitoring only the
central portion of the field.
Most treatment fields are small, and he found that monitoring the entire
beam profile can result in axial calibration errors as large as 3%. In one
construction technique, the collecting plate sectors are formed by vacuum
deposition of a thin metallic coating on defined areas of an insulating lam-
ina of mica Additional grounded metallic coatings, not shown in Figure
2.46, surround the collecting areas and serve as guard rings to minimize
leakage currents over the insulation. Ionization chambers may be sealed to
the outside air making them free of the need for temperature and pressure
corrections.
The electron beam center line through the linac structure and 270 bend-
ing magnet is shown in Figure 2.46. This center line establishes the angu-
lar orientation of the semicircular collecting plates with respect to the ra-
dial and transverse coordinate planes of the beam through the bending
magnet as shown. In general, the upper collecting electrode of Figure 2.46
is concerned with signals pertinent to the radial plane, the lower collecting
electrode to the transverse plane signals. Semicircular plates A and B are
oriented to provide signals related to the radial plane. Their signals are first
amplified via A
1
, and A
2
, then summed via A
3
to provide a console indica-
tion of dose rate and of integrated dose via integrator #1 for MU1 channel
as shown. Similarly, semicircular plates C and D are oriented to provide
signals related to the transverse plane. They feed MU2 channel via amplifi-
ers A
5
and A
6
summing amplifier A
7
and integrator #2. The two dose chan-
nels are completely independent, either can terminate the preset exposure
with the second channel lagging the first by a constant 40 monitor units.
As shown in Figure 2.46, two groups of four steering coils are used in a
servo-feedback system using signals from the ion chamber sectors to con-
trol and limit the lateral displacement (position) and angular divergence
(angle) of the electron beam in the radial and transverse directions. Both
groups of steering coils provide small angular deflections of the electron
beam. The angle steering coils shown on the left in Figure 2.46 are located
immediately adjacent to the 270 bending magnet entrant aperture. The
position steering coils, shown on the right in Figure 2.46, are located at a
distance from this aperture and at the end of the accelerating structure.
The principal effect of their small angular deflection is to provide a lateral
displacement or position correction at the bending magnet entrant aper-
ture. These coil groups are energized by error signals generated if the elec-
tron beam strikes the target, or scattering foil at an angle or at a position
that produces an asymmetrical x-ray or electron beam, as shown in Figures
3.40(b) and 3.40(c), respectively.
Figure 2.46. Five electrode ionization chamber with simplified block diagram of dosimetry
and beam steering system. The radial and transverse coordinate planes of the bending
magnet orbit are identified in the upper left.
Angle R
Steering
Coils
Angle T
Steering
Coils
Target
-500V
P.S.
Position R
Steering
Coils
Position T
Steering
Coils
Buncher T
Steering
Coils
Buncher R
Steering
Coils
Accelerator
Guide
E
le
ctro
n
B
e
a
m
A
A
1
A
4
A
13
A
14
A
8
A
7
A
3
A
2
A
9
A
10
A
11
A
12
A
5
A
6
A - B
E - F
G - H
C - D
B
F
E
G
H
C
D
A + B
C + D
MU1
MU2
SYM1
SYM2
(A - B) + (E - F)
(C - D) + (G - H)
Dose Rate
Meter
Integrator
Integrator
Two of the beam position (lateral displacement) steering coils are connected
to limit the radial displacement, and two the transverse displacement of the
electron beam from the field flattening filter axis as shown in Figure 2.40(c).
The beam angle steering coils are located near entrant dipole M1 of the
270 magnet system shown in Figure 2.43(a). Two of them are connected to
limit the radial angular divergence, and two the transverse angular diver-
gence of the electron beam from the field flattening filter axis as shown in
Figure 2.40(b). Not shown is a third group of four coils positioned around
the beam at the buncher end of the linac structure. These buncher coils
also control motion in the radial and transverse planes, steering the elec-
tron beam leaving the gun onto the center line of the structure.
Signals from peripheral plates E and F are amplified by amplifiers A
9
and
A
10
. Their difference signal (EF) from A
13
feeds the radial position steering
coils. The radial position steering coils control the radial component of lat-
eral displacement of the electron beam with respect to the flattening filter
axis as shown in Figure 2.40(c). Amplified signals from semicircular plates
A and B are subtracted in difference amplifier A
4
which feeds the radial an-
gle steering coils. The radial angle steering coils control the radial compo-
nent of angular divergence of the electron beam with respect to the field
flattening filter axis shown in Figure 2.40(b). The angle and position asym-
metry signals from amplifiers A
4
and A
13
, respectively, are combined to
provide a visual display of radial plane beam asymmetry at the console.
They are set to provide an operational radial asymmetry limit beyond which
the beam is turned off. In a similar manner as shown in Figure 2.46, but
not described here, the transverse position and transverse steering coils
are connected to amplifiers and ion chamber sectors so as to set servo-con-
trol, limit, and display the electron beam asymmetry in the transverse
plane. The symmetry meter can be switched to indicate either radial or
transverse symmetry. It and the associated steering coils are connected to
use all signals from the collecting electrodes. The buncher, beam position,
and beam angle steering amplifiers are each provided with six programma-
ble gain and balance controls corresponding to the one x-ray and five elec-
tron modes of operation.
17.1. Electron
Therapy
Since the requisite beam current for electron therapy is typically several
hundred times less than for x-ray therapy, most shielding problems, in-
cluding neutron attenuation, are comparatively insignificant even at the
higher energies. One is concerned clinically with providing wide, flat elec-
tron fields with modest penumbra, a relatively low surface dose, deep pen-
etration to 80% depth dose for each selected energy, rapid falloff of dose
with depth on the distal side of the depth dose curve as well as a low con-
taminating x-ray background. Achieving and assessing these electron
beam characteristics have been the subject of a number of studies. Usu-
ally, electron applicator cones are attached to the treatment head with the
x-ray collimator jaws set to provide fields a few centimeters wider. There is
accompanying improvement in electron field flatness from this procedure,
preferentially at shallower depths. A too wide setting may create exposure
problems outside the treatment volume as in the case of pregnant patients
for one particular linac. Schneider has evaluated such leakage radiation for
another model of linac, but measurements by Jones suggests that the leak-
age may be a machine-specific characteristic which should be evaluated by
individual users. Often, a lead or Lipowitzs alloy (for example, Cerrobend,
an alloy of Cerro Copper and Brass Company, 16,000 St. Clair Avenue,
Cleveland, Ohio 44110) cutout defines the final field size and shape and is
placed in the end of the applicator at or near the patients skin surface. The
dosimetry of such shaped small and irregular fields has been described for
electrons from 4 to 10 MeV. Electrons scattered off the applicator walls im-
prove flatness at the periphery of the field at a shallow depth with less pen-
etrating electrons and hence, poorer flatness at greater depth. A continu-
ously variable electron beam collimator has been described by Robinson
and McDougall, and at least one manufacturer makes such an option avail-
able. Most studies find that a single, high atomic number scatterer will ad-
equately flatten small fields up to about 10 cm in diameter for low electron
energies up to about 10 MeV. Additional steps are taken at higher energies
and for larger fields. Providing several different scatterer thicknesses, often
in a carousel, facilitates optimization of beam characteristics for different
energies. A dual-foil scattering system, with a few centimeters or more be-
tween the two foils, significantly improves electron beam flatness charac-
teristics, particularly above 15 MeV and for fields 15 cm in diameter and
larger. The first high atomic number (Z) foil is selected to minimize energy
loss for a given scattering distribution and the second foil made of a low-Z
composite, thicker on axis, functions more as a field flattening absorber
preferentially scattering electrons peripherally. Hence, the electron applica-
tor in such systems primarily serves to define the field size and only affects
flatness secondarily. A thorough analysis of the dual-foil scattering system
has been provided by Mandour and Harder.
Scanning the treatment field with a pencil beam of electrons is an alternate
approach to electron therapy. It offers advantages at higher energies of 25
50 MeV where bremsstrahlung contamination from scattering foils be-
comes significant. Earlier, Rozenfeld et al. developed a scanned pencil
beam system for arbitrarily shaped fields defined by a full size template, in
lieu of using scattering foils. The isocentrically directed beam, up to 50 MeV
in energy, employs a complete, non-achromatic magnet system. It com-
bines a linear translation of the magnet up to 21 cm along the direction of
the gantry axis with an indexed rotation of the gantry to provide 5-mm
spacing between scan lines at the skin surface. A treatment is completed in
one scan series covering the field, and different electron energies can be
used in different portions of the field. This scanned beam system employs a
large, non-achromatic beam transport system with attendant stability
problems of the treatment beam. Earlier, Briot et al. investigated the
scanned electron beam of a Sagittaire 35-MeV treatment unit. They con-
clude that an adjustable outboard metallic collimator, which could be at-
tached to the x-ray jaws, improved the depth dose and dose gradient at the
edges of the field. Others compared the Siemens betatron and Sagittaire
linear accelerator electron beams. Bell and Waggener have described a
method for rapid determination of the energy of electron treatment beams
from medical linacs.
Microtrons, having a wide energy range, have come to be employed for elec-
tron therapy. A controversial aspect of linac versus microtron treatment
beams concerns dependence of the characteristics of the electron beam
depth dose on the energy spread of the electron beam. Brahme and Svens-
son contend that the microtron beam has a depth dose significantly im-
proved (sharper buildup and steeper falloff regions) over that of the linear
accelerator and attributed, in part, to the narrower energy spread of the mi-
crotron beam. Fregene suggests that a small difference in energy spread,
which is of the same order of magnitude in the microtron and linear accel-
erator beams, should not lead to appreciable differences in depth dose at
10 MeV. Some of the disagreement between observers might be explained
on the basis that measurements were performed on different accelerators
having different treatment heads (different collimators, scatterers, geome-
try, etc.) using different phantom materials and dosimetry techniques. Re-
cent measurements by Johnsen et al. of electron beams for a single linac
operated at nominal peak energies of 6 and 12 MeV appear to have resolved
this controversy. Based on magnetic analysis of the beams, they compare
narrow energy spread beams (0.1 and 0.2 MeV FWHM) with broad (0.8 and
1.2 MeV FWHM) for 6 and 12 MeV, respectively. They conclude that, when
consistent measurement techniques are used for the beams tested on the
same beam-collimator system, the electron depth dose characteristics are
not significantly affected by these relatively large changes in the width of
the accelerators energy spectrum. Earlier, Bjarngard et al. found no signif-
icant difference between the Mevatron XII electron depth dose curves and
those from a microtron.
C. X-ray Ther-
apy
A number of studies have focused on x-ray target selection, beam penetra-
tion (e.g., percent depth dose at 10 cm or depth of 50% depth dose), and
flatness aspects. Podgorsak et al. examined the effects of different atomic
number targets and flattening filters on small 10-cm-diam fields 100 cm
from the target for energies of 25 MeV. They found that x-ray output on the
central axis does not depend significantly on the Z of the targets. However,
an Al target gives a more penetrating beam, although high-Z targets emit
more radiation at large angles. They also found that an Al filter hardens the
beam and a high-Z filter softens it. They recommend a thick Al target and
flattening filter above 15 MeV for the most penetrating beam. Below 15
MeV, a high-Z target and low-Z filter are recommended. At 25 MeV, an Al
flattening filter is 25 cm in length, an impractical size to incorporate in
most linac treatment heads. Ideally, one wishes flattened fields for all field
sizes, at all depths, an impossible requirement because of energy changes
and scattering in the phantom. One early 4-MV filter design provided satis-
factory flatness at 10-cm depth but resulted in excessive dose at shallow
depths near the edges of large fields. Invariably, flattening filter choice in-
volves a compromise in order to achieve uniform small and large fields over
a range of depths. Two flattening filters are provided in some treatment
units to optimize beam flattening with the changeover at about 10 10 cm
fields, or an additional filter may be attached to the accessory ring for as-
suring large field flatness.
McCall et al. examined linac depth doses using a semi-empirical analytic
depth dose model correlated with experimental measurements. They find
that an Al filter at 25 MeV produces a great deal more beam hardening on
axis than do Ni and W and consequently, there is a larger variation of E
with production angle for Al, a significant effect for large field sizes. Al is a
good flattener, but the penalty is significant energy spread across large flat-
tened fields. For the Clinac 35, operating at 25-MV x-ray energy, they rec-
ommend a Cu target with an Fe filter containing a W conical insert for an
optimum combination so as to minimize energy variation with angle and re-
strict beam hardening on the centerline.
Other important aspects of a good target-flattener system are that the flat-
tener must not become too radioactive in operation, and neutron produc-
tion must be minimized. Thus, other good properties notwithstanding, cop-
per filters are not commonly used above 10 MeV since the gamma-ray dose
rate from 9.76 min Cu-62 activation becomes very high. Iron, on the other
hand, has essentially the same absorption properties as copper, and the in-
duced activity is much weaker.
Flattening filters may absorb 50%90% of the central axis photon intensity.
Brahme et al. recommended scanned photon beams at high energies point-
ing out that the flatness and intensity problems are greatly alleviated.
Hence, some radiation shielding problems (including those for neutrons)
may be as much as 25 times less for high-energy scanned photon beams
than for a heavily filtered unscanned beam. Methods of measurement and
characteristics of leakage radiation for a Clinac 18 treatment unit have
been described by Lane et al.
The spectral change problem associated with flattening filters was cited
earlier by Hansen et al. who identified its importance for a 4-MV linac
beam. Later, Larsen et al. developed a calculative program for filter design
and applied it to a 4-MV beam. Their program sums the primary and scat-
tered components in an iterative manner to fit the dose profile. Jones
points out that the dominant effect is selective hardening of the beam by
the flattening filter, and that for thin targets the energy increases with dis-
tance off axis. More recently, Hanson and Berkley measured the off-axis
quality change for 4 to 10 MV beams and suggested a technique to correct
for the effect in treatment planning calculations.
Megavoltage x-ray beams exhibit an increased depth dose in the buildup
region and a shift of dmax toward the surface as the field size is increased.
Recent investigations indicate that these effects are largely due to low-en-
ergy electron contaminants originating in components of the treatment
head, primarily the flattening filter. Treatment heads can be equipped with
electron filters to attenuate this component. Reductions of 10%20% in
surface dose and an increased depth of d
max
from 2.5 to 4.5 cm at 25 MV
have been observed.
Moyer has identified systematic patient x-ray dose errors for 4 and 10-MeV
linacs associated with elongated rectangular collimator settings. This colli-
mator-exchange effect, which depends on which movable collimator pair
forms the larger or smaller field dimension, may be as large as several per-
cent for highly elongated rectangular fields.
D. Neutron
Leakage and
Contamination
A significant number of neutrons are produced by high-energy x-ray
beams. For most relevant materials, the neutron production threshold oc-
curs at 810 MeV, rises rapidly and then plateaus above 20-MeV photon
energy. Neutrons which originate in the primary collimator, target, and flat-
tening filter contaminate the useful beam. Others are filtered through the
treatment head, some are generated in the patient and most are multiply
scattered by barriers comprising the room. These neutrons, together with
x-ray leakage, affect the patient as well as those outside the treatment
room. McCall and Swanson provide a thorough description of neutron
sources and their characteristics originating in linac treatment heads. They
find that high-Z materials do not significantly alter the neutron fluence but
do substantially reduce the average energy of the transmitted spectrum.
The principal source at 25 MeV is the primary collimator which contributes
one-half or more of the fluence, often followed by the target and flattening
filter, in that order. Depending on the material, energetic neutrons can in-
duce radioactivity in the treatment head and patient support components
of the linac as well as the treatment room barriers. Such radioactivity could
constitute an appreciable source of exposure, particularly for the technolo-
gist (see Sec. VI.-C.). With respect to neutron production, a recent mea-
surement showed that a tungsten flattening filter five radiation lengths
thick produced two and one-half times as many neutrons as an equivalent
steel filter at 25 MV, although ratios as high as seven times have been pre-
dicted.
Recent examination of neutron biological effectiveness has led to a scrutiny
of neutron protection requirements by regulatory agencies. This scrutiny
and the associated need to verify neutron production data, as well as to es-
tablish appropriate measurement, calculation, and dose reduction tech-
niques were the subjects of a recent conference. A consensus appears to be
developing wherein proposed neutron leakage requirements might reason-
ably be met in contemporary high-energy linacs. Often, the shielding pro-
vided for x-ray leakage is sufficient for neutron leakage as well, and satis-
fies the total leakage limitation of 0.1% measured in rads. Rawlinson and
Johns note that the energy imparted outside the useful beam due to scatter
can be more than 20 times greater than the energy due to leakage. Some
simplifications in assessment have resulted from defining two pertinent
measurement surfaces for protection purposes. One is a plane circular pa-
tient surface of radius 2 m centered on and perpendicular to the axis of the
beam at the normal treatment distance and relates to patient protection. A
second cylinder-like complex surface is defined by all points at 1 m from
the path of the electrons between the electron gun and the target or elec-
tron window and relates to room shielding. A recent report of AAPM Task
Group No. 21 on neutrons from high-energy x-ray medical accelerators
provides a careful reasoned, quantitative analysis of the problem together
with recommendations regarding risk to the radiotherapy patient. The re-
port concludes that the principal risk of carcinogenesis is extremely small,
and the implementation of more restrictive regulation is unnecessary and
would be counterproductive.
Acknowledg-
ments
Several commercial treatment units are cited herein to illustrate design
principles and variations of them. Their selection in no way constitutes an
endorsement of a particular equipment. The choice, at times, depended on
the availability of detailed information and numerical data relating to the
design. I am grateful to those who have responded to my requests. Many in-
dividuals have contributed to my understanding of medical linacs and to
the preparation of this review. Their generous efforts have often facilitated
my exposition of specific topics. Several have critiqued portions or all of this
manuscript. I am grateful to all of these individuals: L. Atherton, A.
Brahme, K. Brown, E. Dally, P. Fessenden, D. Goer, R. Jean, S. Johnsen, P.
La Riviere, R. Levy, R. McCall, A. McEuen, G. Meddaugh, N. Pering, T.
Smith, E. Tanabe, L. Tauman, J. Ting, and W. Turnbull. I am particularly
grateful to Craig Nunan whose broad background and critical appraisal
have added immensely to my perspective. Pieter Huismans talents have
provided many of the illustrations. Juanita Clacks outstanding secretarial
skills, patience, and untiring efforts are much appreciated and have greatly
simplified my task.
Modulator Theory 3-1
This chapter will discuss the theory of non-resonant series line type modulators used in
Varian C-series Clinacs.
Modulator Theory
3-2 Modulator Theory
Table of Contents
1. Introduction: .................................................................................................................... 3-5
2. Basic Concepts: ............................................................................................................... 3-5
3. DeQing Principles:............................................................................................................ 3-7
4. Non-resonant Transmission Line Principles: ..................................................................... 3-8
5. Pulse Shape Definitions: ................................................................................................ 3-16
6. Line Type Modulator Load Element Principles: ............................................................... 3-17
7. Fault Conditions: ........................................................................................................... 3-18
8. Thyratron Theory: .......................................................................................................... 3-19
8.1. Introduction: .......................................................................................................... 3-19
8.2. Operating Notes on Hydrogen-filled Tubes: ............................................................. 3-23
Modulator Theory 3-3
Figure 3.1. Resonant Charging Circuit:................................................................................. 3-5
Figure 3.2. Voltage and Current Waveforms of Capacitor C1: ................................................ 3-6
Figure 3.3. Resonant Charging Circuit with Series Charging Diode: ...................................... 3-6
Figure 3.4. Voltage and Current Waveforms on C1 with Series Charging Diode: .................... 3-7
Figure 3.5. Addition of DeQing Switch Circuit: ...................................................................... 3-7
Figure 3.6. DeQing Waveforms: ............................................................................................ 3-8
Figure 3.7. Distributive Transmission Line: .......................................................................... 3-8
Figure 3.8. Finite (Lumped) Network: .................................................................................... 3-9
Figure 3.9. Resonant Charging with PFN: ............................................................................. 3-9
Figure 3.10. The Pulse Discharge Circuit: ........................................................................... 3-10
Figure 3.11. PFN Discharge Waveform Formation: .............................................................. 3-11
Figure 3.12. PFN Discharge Waveforms vs. Load Impedance: .............................................. 3-13
Figure 3.14. Typical Line Type Modulator for Linear Accelerator Applications:..................... 3-14
Figure 3.13. Basic Line Type Modulator Circuit: ................................................................. 3-14
Figure 3.15. Pulse Shape Definitions, Theoretical: .............................................................. 3-16
Figure 3.16. Pulse Shape Definitions, Practical: .................................................................. 3-16
Figure 3.17. Equivalent Circuit of Magnetron as Load:........................................................ 3-17
Figure 3.18. Magnetron Voltage-Current Characteristics:.................................................... 3-17
3-4 Modulator Theory
Figure 3.19. Conventional Industrial Thyratron. (a) Anode. (g) Control Grid, (k) source of
electron emission: .............................................................................................................. 3-19
Figure 3.20. Critical Grid Potential in Relation to Anode Voltage:........................................ 3-20
Figure 3.21. Reference Points Associated with the Interpretation of Pulse Shapes: .............. 3-22
Figure 3.22. Schematic Diagram of Line Discharge Circuit: ................................................ 3-23
Figure 3.23. Charging Voltage Waveforms: ......................................................................... 3-24
Figure 3.24. Circuit for Recovery Time Measurement:......................................................... 3-27
Figure 3.25. Methods of Providing Negative Grid Voltage Swing: ......................................... 3-28
Figure 3.26. Waveforms with Pulse Transformer Trigger Drive: ........................................... 3-30
Figure 3.27. Inductive Overswing with Bias: ....................................................................... 3-30
Figure 3.28. Inductive Overswing without Bias: .................................................................. 3-30
Figure 3.29. Waveforms Normally Seen During Modulator Adjustments:............................. 3-31
Figure 3.30. Schematic Diagram of Tetrode: ....................................................................... 3-33
Figure 3.31. Single Pulse Drive for Tetrode: ........................................................................ 3-33
Figure 3.32. Simplified Single Pulse Drive: ......................................................................... 3-34
Figure 3.33. Double Pulse Drive for Parallel Operation: ...................................................... 3-35
Figure 3.34. Single Pulse Drive for Parallel Operation: ........................................................ 3-35
Figure 3.35. Anode Circuit Using Separate PFNs: .............................................................. 3-36
Figure 3.36. Arrangement for Adjustment of Current Sharing:............................................ 3-36
Figure 3.38. Thyratrons in Series: ...................................................................................... 3-37
Modulator Theory: Introduction 3-5
1. Introduction The material in this chapter should be read and thoroughly understood by
the student before proceeding to the specific modulator descriptions in the
High and Low Energy Clinac Beam Delivery System course manuals.
2. Basic Con-
cepts
Consider the circuit of Figure 3.1 below.
When switch S1 is closed, current will begin to flow through inductor L1 to
charge capacitor C1. Initially the reactance of L1 will limit current flow re-
sulting in a voltage drop equal to the battery voltage appearing across L1.
As time passes, current does begin to flow. This results in the buildup of a
magnetic flux or field in the core of L1 (storage of energy).
Eventually, the charge across C1 approaches a value equal to the battery
voltage E
dc
. The current through L1 at this point is maximum and both L1
and C1 have stored energy. Current in the circuit begins to decrease be-
cause there no longer exists a voltage difference between C1 and the bat-
tery. This causes the magnetic field in L1 to collapse.
The collapsing field produces a continuation of current flow in L1 that cre-
ates a voltage source that adds to the battery voltage. C1 now begins to
charge higher than the battery voltage E
dc
, eventually approaching 2E
dc

when all the stored energy in L1 has been transferred to C1.
There now exists a voltage difference between the battery voltage E
dc
and
the capacitor voltage 2E
dc
. This causes current flow to reverse in the circuit
resulting in the extra stored energy in C1 now causing a reverse current to
flow in L1. A magnetic field is again developed in L1 in the reverse direction
and L1 now has an excess of stored energy.
This cycle of events will continue until all the original stored energy in L1 is
dissipated in the IR losses of the circuit. As a result of this cyclic reaction,
a damped oscillation of current will flow between L1 and C1. The oscillation
has a resonant time period determined by the value of L1 and C1 defined by
the relationship:
Figure 3.1. Resonant Charging Circuit
B1
S1
L1
C1
E
dc
TC
1
2 LC
------------------ =
3-6 Modulator Theory: Basic Concepts
It can be seen by the waveform of Figure 3.2 that the capacitor C1 will even-
tually have a charge voltage equal to the battery voltage E
dc
after several cy-
cles have occurred. This resonant charging condition can be put to use to
allow C1 to maintain the stored energy in L1 by adding a diode in series be-
tween L1 and C1 as indicated in Figure 3.3.
The sequence of events in the circuit of Figure 3.3 is exactly the same as the
circuit of Figure 3.1 except that when the capacitor C1 reaches a voltage
equal to 2E
dc
the diode CR1 will block the reverse flow of current back
through L1. The waveforms of Figure 3.4 indicate these events.
There are three major advantages to the use of the circuit in Figure 3.3:
a. The power source voltage required is only half that of the
capacitor stored value.
b. The efficiency of the transfer of energy is raised from approxi-
mately 50% to almost 100%.
c. It is possible to regulate the voltage charge on the capacitor.
Figure 3.2. Voltage and Current Waveforms of Capacitor C1
Figure 3.3. Resonant Charging Circuit with Series Charging Diode
2 LC
E (Voltage)
I (Current)
Resonant Time Period
2E
dc
E
dc
0
t
2
S1
L1
C1
E
dc
B1
Modulator Theory: DeQing Principles 3-7
3. DeQing
Principles
Regulation of the charge voltage on Capacitor C1 can be accomplished by
the circuit of Figure 3.5. Consider switch S2 and resistor R1 shown in par-
allel with L1. It was stated that the current through L1 was maximum when
C1 was charged to E
dc
and zero when charged to 2E
dc
(refer to Figure 3.4).
If the switch S2 is closed at some time after the current in L1 has reached
maximum value and the charge voltage on C1 has reached E
dc
, the stored
energy in L1 will be routed through S2 and be dissipated in resistor R1.
By controlling the exact time when S2 is closed, any charge voltage level
from E
dc
to approximately 2E
dc
can be placed on C1 resulting in regulation
of the stored energy in C1. The shorting circuit consisting of S2 and R1 has
been identified by the term DeQing circuit.
The term DeQing has been created to identify, in a simple manner, the ac-
tion of the circuit composed of S2 and R1 across L1. The word Q or Q
Factor is defined as the ratio of the reactance to resistance of an inductor,
capacitor, or resonant circuit. With reference to Figure 3.5, it can be seen
that when switch S2 is closed, R1 is placed in parallel with L1. This lowers
the Q of the inductor. Thus, the circuit is called a DeQing circuit, etc.
Figure 3.4. Voltage and Current Waveforms on C1 with
Series Charging Diode
LC Time Period
E (Voltage)
I (Current)
2E
dc
E
dc
0
t
Figure 3.5. Addition of DeQing Switch Circuit
S1
S2 R1
L1
C1
E
dc
B1
3-8 Modulator Theory: Non-resonant Transmission Line Principles
Observe the current and voltage waveforms of Figure 3.6. The current
waveform indicates an abrupt drop to zero when switch S2 is closed. The
energy still remaining in the inductor L1 is the area under the curve indi-
cated by the shaded area under the dashed line. This energy is dissipated
by resistor R1.
4. Non-reso-
nant Transmis-
sion Line
Principles
Non-resonant transmission lines are lines that are either infinitely long or
terminated in some impedance. A uniform transmission line has what is
called a characteristic impedance. This is the impedance that would be
measured at the end of such a line if it were infinitely long.
The importance of this characteristic impedance lies in the fact that if any
length of line is terminated in an impedance of this value, then all the en-
ergy flowing along the line will be absorbed at the termination and none is
reflected back along the line. A result of this is that the input impedance of
any length of transmission line terminated in its characteristic impedance
is equal to that characteristic impedance.
Transmission lines such as coaxial cables can be represented as a network
of distributive series inductive and parallel capacitive elements as dia-
gramed in Figure 3.7. This distributive network can be further simplified or
simulated by constructing a finite number of series and parallel elements
in the form of the network diagramed in Figure 3.8. This network is com-
monly called a PFN (Pulse-Forming Network) when used for energy storage
purposes in a line type modulator.
Figure 3.6. DeQing Waveforms
E (Charged Voltage Level of C1)
I (Current through R1)
I (Charging Current to C1)
2E
dc
E
dc
0
t
LC Time Period
Figure 3.7. Distributive Transmission Line
IN
OPEN
CIRCUIT
Modulator Theory: Non-resonant Transmission Line Principles 3-9
The development of a pulse-forming network that simulates a transmission
line is a problem of network mathematics. No network having a finite num-
ber of elements can exactly simulate a transmission line which in reality
has distributed rather than lumped parameters.
The pulse-forming network serves a dual purpose of storing exactly the
amount of energy required for a single pulse and of discharging this energy
into a load in the form of a pulse of specified shape. The required energy
may be stored either in the capacitances or in the inductances, or in com-
binations of these circuit elements.
Networks in which the energy is stored in the electrostatic field of the ca-
pacitors are referred to as voltage-fed networks. Networks in which the en-
ergy is stored in the magnetic field of the inductors are referred to as cur-
rent-fed networks. Networks of the voltage-fed type are generally used in
practice because only with this type of network can gaseous switches such
as thyratrons be used to switch the energy.
The network of Figure 3.8 can be defined as a voltage-fed network where
the stored energy is in the capacitors. If the capacitor C1 of Figure 3.5 is re-
placed by the network of Figure 3.8 to form Figure 3.9 and the circuit
switch S1 is closed, the same sequence of events occur as originally hap-
pened in Figure 3.5. Now the current begins to charge each capacitor of the
network through the series inductance of each section of the network. The
result is that each capacitor will charge to 2E
dc
at a charge time that is in-
creasingly longer for each capacitor down the network.
However, the difference in charging time of each network section is relative-
ly insignificant to the total charge time of the entire circuit. This is due to
the large charge time defined by the charging choke L1 and the total value
of all the capacitance within the network. The PFN inductors have no sig-
nificant effect during the charge cycle because of their small inductance
value compared to the large inductance of L1.
Figure 3.8. Finite (Lumped) Network
Figure 3.9. Resonant Charging with PFN
IN
OPEN
CIRCUIT
S1
S2 R1
CR1
E
dc
L1 L2 L3 L4
C1 C2 C3 C4
PFN
B1
Since the application for which these modulators are usually designed re-
quires the application of an essentially rectangular pulse of energy, the
open-ended transmission line (PFN) considered in the foregoing example
may be taken as the starting point in a discussion of the discharge circuit.
Figure 3.10 represents a simple discharging circuit consisting of a PFN,
which is in a charged state of 2E
dc
, a switch S1, and a load resistance R1
which has the same impedance as the PFN characteristic impedance.
When S1 is closed the following events occur:
1. C1 begins to discharge through L1 into the load resistor R1. Ini-
tially, all the charge voltage across C1 will be developed across L1.
As time passes, current flows through L1 and begins to dissipate in
load resistor R1. The time required is dependent on the values of
L1C1 and R1. Eventually the voltage across R1 will reach E
dc

because the impedance of the load resistance R1 is equal to the
impedance of L1 and C1.
2. At this point C1 will stop discharging and C2 will begin to dis-
charge through L2 and L1 into R1. Similarly, C2 will stop
discharging at E
dc
and C3 will begin to discharge. This sequence of
events continues until the last capacitor C4 begins to discharge.
When C4 reaches E
dc
, it will continue to discharge to zero. Then C3
will discharge to zero. Eventually all capacitors will discharge to
zero and all the energy will have been dissipated in R1.
The above sequence of events results in a rectangular current pulse of en-
ergy being supplied to R1 with a duration which is twice the transmission
time (aggregate discharge times) of the network. Reference Figure 3.11.
From the foregoing discussion, as illustrated by Figure 3.11, it can be seen
that if the load impedance is equal to the line impedance (Rl=Zo), that is if
the line is matched to the load, the current out of the line consists of a sin-
gle rectangular pulse of 2 time periods (The time to discharge each capaci-
tor down the line and then back). These conditions satisfy Ohm's law where
half the voltage appears across the load but for twice as long thus dissipat-
ing all the stored energy in the network.
Figure 3.10. The Pulse Discharge Circuit
S1
R1
E
dc
L1 L2 L3 L4
C1 C2 C3 C4
PFN
Figure 3.11. PFN Discharge Waveform Formation
Voltage
E
C1
E
C1
I
C1
I
L1
T
1
T
2
I
C3
I
L3
I
C3
I
L3
I
C1
I
L1
I
C2
I
L2
I
C4
I
L4
I
C2
I
L2
E
C3
E
C3
E
C2
E
C2
E
C4
t
Current
t
Current
Current
Discharge
t
t
2T
The effect of mismatching the load is to introduce a series of steps into the
transient (rectangular) discharge.
As can be seen in Figure 3.12, these steps will all be of the same sign when
the mismatch is Positive (the load resistance is greater than the line imped-
ance) Rl>Zo, and of alternate sign when the mismatch is negative (the load
resistance is less than the line impedance) Rl<Zo. A simple explanation of
these steps can be made by considering the voltage division that takes
place between the line impedance (Zo) and the load impedance (Rl).
Example: If the load is 2 times the line impedance C1 will only dis-
charge 1/3 of its total charge during the first discharge period, then
1/3 of the remaining charge during the return discharge period. So
after 2 periods of time the line will still have an appreciable charge
remaining. This results in still another cycle of discharge events
which will continue until all the energy in the line is dissipated in
the load. The significant thing here is that it will take a longer time
to discharge the line and the output will not be a single rectangular
pulse but a staircase of voltage and current.
Example: If the load is 1/2 times the line impedance, C1 will now
discharge more than half its stored value during its first discharge
period. It will then attempt to discharge beyond zero during the
return period, thus leaving the network now charged to the oppo-
site sign () for the second discharge cycle, etc. This results in a
damped train of changing sign pulses that also will take longer to
discharge all the stored energy in the network.
Thus the line can be considered as having reflections that traverse the line
to the open end in one time period which are completely reflected there,
and travel back to the load end in the second time period, where they ap-
pear as positive or negative steps depending upon the mismatch ratio. The
reflections continue in this manner with constantly diminishing amplitude
until all the energy initially stored in the line is dissipated in the load.
If the discharge switch and load resistor of Figure 3.10 are added to the cir-
cuit of Figure 3.9, the circuit of Figure 3.13 is developed which is a basic
line type modulator with a resistive load termination.
Figure 3.12. PFN Discharge Waveforms vs. Load Impedance
(+)E
L
(+)E
L
(+)E
L
0
0
0
()
R = Z
Matched
L O
2T
2T
2T
4T
4T
6T
6T
8T
8T
R = 2Z
Positive Mismatch
L O
R = Z
Negative Mismatch
L O
In the following discussion refer to Figure 3.14.
1. A 10 KV DC high voltage power supply has been substituted for the
battery.
2. Switches S2 and S3 have been replaced with high voltage high cur-
rent type thyratron tubes.
Figure 3.13. Basic Line Type Modulator Circuit
S3 R2 (Load)
S1
S2 R1
CR1
E
dc
L1 L2 L3 L4
C1 C2 C3 C4
PFN
B1
Figure 3.14. Typical Line Type Modulator for Linear Accelerator Applications
3-Phase
High Voltage
Power Supply
Main Switch
Trigger Gen.
System
Clipper Current
Fault Monitor
System
Compensated
Voltage Divider Circuit
3000:1 Ratio
DeQing Switch
Trigger Gen.
System
HVPS O/C
Fault Monitor
System
10 KV
L1 CR1
20 KV
Current Toroid
C1 C2 C3 C4
R1
R2
CR2
HVPS Current
Monitor Circuit
R4
R5
C5
C6
De-Spiking
Network
V1
V2
T2
T1
R3
CR3
Pulse Transformer
Turns Ratio = 1:11
Klystron
Equivalent Circuit
End
Clipper
Circuit
6.7V
C5
R4
CR
L
R
L
C
L
PFN
3. The load resistor R2 has been replaced by a pulse transformer T1
and the equivalent circuit of klystron (magnetron in low energy Cli-
nacs), Rl, CRl and Cl.
From the foregoing discussion on impedance matching, it is clear that the
PFN must be terminated with a load impedance equal to its characteristic
impedance. The impedance associated with high power line type modulator
used for klystron-driven accelerators will be typically around 12.5 ohms
and for magnetron-driven accelerators, 25 ohms. (The klystron will have an
impedance of 1400 ohms and the magnetron 400 ohms.). The pulse trans-
former will require a step up turns ratio of approximately 1:11 for klystrons
and 1:4 for magnetrons because the impedance transformation of a pulse
transformer changes as the square of the turns ratio of its windings.
Note the new components:
1. A high voltage power supply current monitor resistor R2 and a fault
detector system has been placed in the return path of the power
supply. The voltage across the resistor will be proportional to cur-
rent flow from the power supply. If the current exceeds a specific
limit the monitor system will turn off the high voltage power sup-
ply. Thus, any failure of components in the charging circuit will be
detected.
2. The DeQing thyratron is controlled by a DeQing trigger system
which receives input information on the DC charge level of the PFN
through a compensated voltage divider circuit R4,C5R5,C6 that
has a voltage division ratio of 3000 to 1.
3. The main switch thyratron is controlled by a trigger generator sys-
tem. This will enable repetitive operation of the modulator circuit at
any desired PRF (pulse repetition frequency) within the design limi-
tations of the components of the charge circuit.
4. A diode CR3, resistor R3 and a current toroid T2 have been added
across the end of the PFN. The diode is polarized so current will
only flow through the resistor when the charge on the PFN goes
negative. The toroid will detect any current flow and the clipper cur-
rent fault monitor system will turn off the high voltage power supply
thus protecting the components in the discharge circuit.
5. A resistor R4 and capacitor C5 have been placed in series across (in
parallel with) the primary of the pulse transformer to form a de-
spiking network. In a practical modulator the pulse output of the
PFN is of the order of 1 to 5 microseconds duration. The rise and
fall time of the pulse will usually be in tenths of microseconds. The
pulse transformer must be designed to transmit these very short
pulses. However, during the rise and fall time of the pulse, the
transformer will tend to appear as a very high impedance which will
cause overshooting of the pulse voltage due to mismatched condi-
tions. This can cause serious problems with the klystron or
magnetron as well as the pulse transformer. The de-spiking net-
work will appear as a matched impedance during the fast rise and
fall time of the pulse due to the low reactance of C5 effectively plac-
ing resistor R4 across the PFN during these times. R4 will be equal
in value to the characteristic impedance of the PFN.
3-16 Modulator Theory: Pulse Shape Definitions
5. Pulse Shape
Definitions
A pulse can be considered as a trapezoidal shape with a finite rise and fall
time and a defined width and peak amplitude as indicated by Figure 3.15.
Generally the rise time is defined as the time between 10% and 90% of the
rising portion of the waveform and the fall time is likewise defined as 90%
to 10% of the fall portion of the waveform. The pulse width is considered to
be defined as the width at the 50% amplitude point. Additionally, pulses of
a practical nature have definitive values of overshoot and droop as well as
reverse tails as indicated in Figure 3.16.
Figure 3.15. Pulse Shape Definitions, Theoretical
Figure 3.16. Pulse Shape Definitions, Practical
Rise Time Fall Time
50% Pulse Width
Flat Top Width
0%
100%
Rise Time Fall Time
% Overshoot
50%
Reverse Tail
Pulse Width
% Droop
0%
90%
10%
100%
Modulator Theory: Line Type Modulator Load Element Principles 3-17
6. Line Type
Modulator
Load Element
Principles
The output pulse of a line type modulator can be applied to either a passive
element such as a resistance that will dissipate the pulse energy in the
form of heat, or an active device such as a klystron or magnetron that con-
verts the pulse energy into a form of high frequency electromagnetic ener-
gy. A magnetron will be utilized as the active load element for our discus-
sions. The circuit elements of Figure 3.17 define the equivalent circuit of a
magnetron as a load element of the modulator.
The behavior of a magnetron as a load element of the modulator is equiva-
lent to that of an ideal diode CRL, that is a diode that has a linear EI (volt-
agecurrent) characteristic, in series with a battery of voltage Vs whose po-
larity is in opposition to the applied voltage pulse of the modulator. For cir-
cuit analysis it is possible to represent such a load as a resistance RL in
series with the diode CRL and battery of voltage (Vs). The stray capacity of
the magnetron can be represented as a capacitor CL in parallel across the
network.
The effect of the equivalent circuit of Figure 3.17 can be defined by the volt-
agecurrent characteristic of Figure 3.18.
Figure 3.17. Equivalent Circuit of Magnetron as Load
Figure 3.18. Magnetron Voltage-Current Characteristics
Pulse Output
Magnetron Oscillator
+
Line Type
Modulator
R
L
C
L
V
S
CR
L
Current
Theoretical
Actual
Voltage (V )
S
0
3-18 Modulator Theory: Fault Conditions
The magnetron does not begin to conduct current until the applied pulse
voltage approaches the effective value of the bias voltage Vs. At this point
the magnetron begins to conduct heavily with only the small additional rise
in the applied voltage. During the rise time of the pulse the stray capaci-
tance of the magnetron C
L
tends to limit the rate of rise of the applied volt-
age to a realistic value. When the modulator pulse begins to fall the stray
capacitance C
L
tends to prevent the rapid fall of the pulse.
7. Fault Condi-
tions
Whenever a magnetron arcs its internal impedance drops toward zero,
causing a reflected low impedance to appear across the PFN. This results in
the circuit action as defined under negative mismatch conditions. The PFN
discharges negative and at this point the thyratron will stop conducting.
This results in a large amount of energy being left in the PFN capacitors
stored as a reverse charge. Referring to Figure 3.14, the end clipper diode
CR3 conducts this energy through resistor R3 where it is all dissipated as
heat.
This condition must not be allowed to continue. Some form of monitor sys-
tem will be used to detect this current flow and turn off the high voltage
power supply, thus preventing serious damage to any components.
Whenever the magnetron misfires, that is, does not conduct during a pulse,
the result is a reflected high impedance mismatch at the PFN, resulting in a
slow rate of discharge of the PFN.
The charge circuit normally has a charge time several orders of magnitude
longer than the discharge time. Typical values would be charging time 1
millisecond and discharge time of 4 to 6 microseconds. The charging choke
L1 appears as an open circuit during the short discharge time while the
thyratron is conducting thus isolating the power supply from the thyratron
which would look like a short to the power supply.
Whenever the PFN is mismatched positive it tends to discharge very slowly,
resulting in possible continuous firing of the main thyratron. This results
in excessive current being monitored by the power supply current monitor
resistor R2. The HVOC (High Voltage Over-Current) fault monitor circuit
will turn off the high voltage power supply, thus protecting the components
from a major failure.
The preceding discussion has dealt with only the key factors in the opera-
tion of a line type modulator driving a magnetron. Most of the principles
discussed also apply to klystron systems, except that the klystron is basi-
cally an amplifier tube and behaves accordingly, e.g., there are no problems
associated with misfiring and internal arcing within the tube is abnormal
and cannot be tolerated. There are many other aspects of its operation that
must be dealt with to fully understand any given system or accelerator ma-
chine performance. This is only a preliminary exercise.
Modulator Theory: Thyratron Theory 3-19
8. Thyratron
Theory
The following information is compiled from The Operation and Use of Hy-
drogen-Filled Tubes by R. E. Lake, B.Sc. and H. Menown, M.Sc., and edited
for this manual.
8.1. Introduc-
tion
The conventional thyratron with mercury vapor or rare gas filling was first
introduced in the early 1920s, and the basic geometry of the industrial
type of tube has changed very little since then. The tube consists essential-
ly of an anode (a), control grid (g), and source of electron emission (k), as
shown in Figure 3.19.
With a positive voltage on the anode, the tube will remain in a nonconduct-
ing state if a suitable voltage (usually negative) is applied to the grid. This
voltage depends on the anode voltage, and for every value of anode voltage
there is a critical grid potential (Figure 3.20). The electrons leaving the
cathode are prevented from reaching the grid/anode space by the potential
barrier at the grid. As the grid voltage is made less negative, or anode volt-
age more positive, the field due to the anode attracts an increasing number
of electrons from the cathode. These collide with gas atoms and when the
resultant electrons receive sufficient energy and sufficient collisions occur,
cumulative ionization takes place and the tube fires through. The voltage
across the tube then drops to a value (10-15V for mercury and rare gases),
which is dependent on the application and the nature and pressure of the
gas filling. The current passed is then largely determined by the external
circuitry.
Figure 3.19. Conventional Industrial Thyratron. (a) Anode.
(g) Control Grid, (k) source of electron emission
3-20 Modulator Theory: Thyratron Theory
During breakdown the negative potential on the grid attracts positive ions
and these form a sheath through which the grid potential cannot penetrate.
Thus, any increase in the negative value of the grid supply voltage has no
effect on the current passing. The tube will return to its non-conducting
state after removal of anode voltage for a sufficient time (known as the re-
covery time) to allow the charged particles to disappear. The voltage on the
grid then returns to its original value, and a positive voltage can be reap-
plied to the anode without conduction taking place. The tube, therefore,
acts as an electronic switch which may be closed by the application of a
positive signal to the grid but which can only be opened by the removal of
anode voltage for a minimum time.
The grid is almost always a far more massive affair than is found in vacuum
tubes since it has to withstand recombination heating and bombardment
without allowing a rise of temperature sufficient to cause primary grid
emission. The apertures in the grid may consist of perforations of various
shapes and sizes or may even be a single large hole or an annulus. Some-
times, a baffle in the form of a disc is attached to the cathode side of the
grid with a small spacing between the baffle and the grid proper. Such baf-
fling modifies the characteristic of the tube so that the potential of the grid
in the non-conducting state may be either zero or positive. At the same
time, this baffle helps to prevent deposition of cathode material on the grid
proper. In this case ionization must occur between grid and cathode near
the grid apertures before breakdown to the anode can take place.
The first reported use of hydrogen as a filling for gas tubes was in 1936.
Nothing further was done until the 1940s when the rapid growth of radar
required the development of a switch tube Able to operate at higher fre-
quencies and with more reliable triggering than the mercury tubes then in
use. (Pulse generators Radiation Laboratory Series p. 335 et seq. K. J.
Germeshausen. Published by McGraw Hill.)
Figure 3.20. Critical Grid Potential in Relation to Anode Voltage
With mercury, the maximum voltage that can be developed across the tube
is about 30V. At a higher voltage, positive ion bombardment of the cathode
becomes very intense and cathode destruction rapidly occurs. In pulse op-
eration this is invariably about 100V so that destruction of the cathode is
inevitable. With hydrogen this critical voltage is at least twenty times great-
er due to the much lower mass of the hydrogen ion. The recovery time of the
hydrogen tube is an order of magnitude less than a similar mercury or rare
gas tube because of the higher mobility of the proton.
The chemical activity of hydrogen immediately introduces the problem of
gas clean-up. To overcome this, all the materials used for the tube struc-
ture have to be of a very high degree of purity which is closely controlled.
The processing of all the components used and of the tube itself during the
exhaust and filling schedule is also very critical. In addition, tube geometry
has a considerable influence on the rate of gas clean-up for given operating
conditions.
The common method of improving the life of hydrogen-filled tubes is by
means of a hydrogen replenisher or reservoir. This consists essentially of a
controlled weight of (usually) titanium hydride which is maintained at a rel-
atively constant temperature within the tube envelope. Under these condi-
tions titanium, hydrogen and the hydride are in thermal equilibrium and
any variation of the amount of hydrogen within the tube envelope causes
an adjustment of the equilibrium conditions. By this means the pressure
within the envelope is maintained relatively constant. The hydride is usu-
ally contained in a metal cylinder which may be closed at one or both ends
depending on the cylinder material. The capsule is heated by placing it in
series or parallel with the cathode heater or by means of a separate supply.
The latter arrangement is sometimes convenient if the thyratron is required
to operate over a wide power range, and allows for a less critical design of
the reservoir. It is obviously less convenient from the user's view point.
Hydrogen thyratrons have been manufactured in the United Kingdom since
about 1951. Development in recent years has concentrated on the multi-
electrode structure, which offers advantages over the conventional triode
form. This is discussed in more detail in a later section.
The particular nature and application of hydrogen tubes has resulted in
the use of terms and definitions peculiar to these devices. The more impor-
tant and less familiar of these are given below.
8.1.1. Peak
Forward Grid
Drive
The peak positive value of the unloaded grid pulse with respect to the cath-
ode. This must be added to the bias voltage (if any), to determine the am-
plitude of the output pulse from the trigger generator.
8.1.2. Recov-
ery Impedance
The ratio of the potential difference between the instantaneous potentials
appearing at the thyratron grid and the on-load grid bias voltage to the grid
current, at the same instant during the recovery period. This can be ex-
pressed as:
8.1.3. Recov-
ery Time
The time interval between the cessation of anode current and the instant
when the grid regains control under specified anode and grid circuit condi-
tions.
Inst. grid potential bias voltage ( )
current
----------------------------------------------------------------------------------
8.1.4. Grid 2
Pulse Delay
The time interval between the voltage pulses on the terminals of grid 1 and
grid 2 with the thyratron removed, measured at the 26% level on the lead-
ing edge of each pulse. It is essential that the grid pulses overlap.
8.1.5. Anode
Delay Time
The time interval between the 26% point on the leading edge of the unload-
ed grid pulse and the instant when anode conduction occurs. (In tetrodes
the grid 2 pulse is used as reference.)
8.1.6. Anode
Delay Time
Drift
The change in anode delay time over a specified period of time as a result of
continued operation of the thyratron under specified conditions.
8.1.7. Time Jit-
ter
The pulse-to-pulse variation of the instant when anode conduction occurs
referred to the 26% point on the leading edge of the unloaded grid pulse. (In
tetrodes the grid 2 pulse is used as reference.)
8.1.8. R.M.S.
Current
Normally computed as:
The reference points associated with the interpretation of pulse shapes are
shown in Figure 3.21.
Figure 3.21. Reference Points Associated with the
Interpretation of Pulse Shapes
peak current ( ) mean current ( )
Pulse Duration
Time of Rise Time of Fall
Spike
Amplitude
Spike
Duration
Amplitude
70.7%
26%
8.2. Operating
Notes on
Hydrogen-
filled Tubes
The most important application for hydrogen thyratrons is in line-type
pulse generators for radar transmitters and particle accelerators where
high powers have to be switched rapidly and with precise timing and where
small size, high efficiency and tolerance of ambient temperature variations
are obvious advantages.
A growing use is to be found for thyratrons as protection devices for Single
Shot use, e.g., to discharge condenser banks in thermonuclear research.
8.2.1. Thyra-
trons in Line
Type Modula-
tors
The basic circuit for the DC charging line type modulator is shown in Fig-
ure 3.22, together with the simpler relationships most often used.
Although simple in diagrammatic form, this circuit has many components
which are distributive in nature and interact with each other in different
ways which depend upon their relative values. The modulator circuit must
be able to cope with the situations that arise from switching on high voltage
at a low value through normal full load operation to mismatch or short cir-
cuits produced by the RF load tube. (For a fuller treatment see Pulse Gen-
erators Radiation Laboratory Series.) Those features of the circuit which
most affect the performance and reliability of hydrogen thyratrons will now
be discussed.
Figure 3.22. Schematic Diagram of Line Discharge Circuit
E = DC power supply voltage
LC = charging choke inductance
LN = total network inductance
CN = total network capacitance
ZN = network impedance
f = pulse repetition frequency
tp = pulse width
epy = thyratron peak anode voltage
epx = thyratron peak inverse voltage
I = thyratron mean anode current
ib = thyratron peak anode current for matched load
ZL = load impedance
PL = peak load power
f =
L
C
L
N
E
Z
L
C
N
Z
N
L
N
C
N
------- =
t
p
2 L
N
C
N
2C
N
Z
N
= =
i
b
epy
Z
N
Z
L
+
------------------- =
epy 2E epx + =
P
L
epy
Z
N
L +
----------------

2
Z
L
epy
2
4Z
L
----------- = =
I i
b
ft
p
=
1
L
C
C
N
( )
1 2
------------------------------ epx
Z
N
Z
L
Z
N
Z
L
+
------------------- epy =
8.2.2. The
Charging Cir-
cuit
The pulse forming network (PFN) is charged to approximately twice the DC
supply voltage in a time which equals half the period of the resonant circuit
formed by the charging choke inductance and network capacitance. After
the thyratron has fired and the voltage has collapsed, sufficient time must
be made available for the thyratron to recover. This means that the anode
voltage must be kept below cathode potential until the thyratron has recov-
ered.
If the thyratron is triggered at the instant the network voltage reaches a
maximum then the condition of resonant charging is achieved, see Figure
3.23(a) If this triggering is delayed then the peak voltage will commence to
fall unless a charging diode is present, see Figure 3.23(b). (The same effect
is produced by a low value of charging choke inductance for a given trigger-
ing frequency.) If triggering frequency is increased or inductance increased
then a condition of linear charging is approached, Figure 3.23(c). Therefore,
in order to make most time available for recovery of the switch tube, the
charging choke inductance should be calculated for the highest repetition
frequency that will be used.
Where the repetition frequency is high (greater than 5000 p.p.s.) or the
duty ratio is high (of the order of 0.01), the time available for recovery be-
comes a very important factor. (See section on Recovery.) In these cases, to
enable the switch tube to recover, it may be necessary to delay the charging
of the PFN by using a triggered charging diode or a saturable reactor.
The power supply should have good regulation to avoid any increase in an-
ode voltage under conditions of interrupted triggering, and should also be
free from appreciable overshoot when 'snap start conditions are required.
8.2.3. Charg-
ing Diodes
The charging diode must be rated to withstand a peak inverse voltage
(p.i.v.) of not less than the full network voltage since at the end of its charg-
ing period the anode may swing down to near earth potential.
The mean current requirements of high-power modulators may preclude
the use of vacuum tubes. In such cases, a gas tube may be used which of-
ten takes the form of a triggered diode such as the FX297.
When it is necessary for the hold-off period to be long, the grid of the charg-
ing diode may be triggered through a suitably insulated pulse transform-
er but for conditions not far from resonant a high resistance potential di-
vider will be satisfactory. The inclusion of an anode inductor helps to pro-
tect the charging diode from the initial spike of charging current produced
by the self capacity of the charging reactor, and also from the full effect of
the inverse voltage swing which occurs after the pulse forming network is
Figure 3.23. Charging Voltage Waveforms
epy epy epy
Supply
Voltage
Supply
Voltage
Supply
Voltage
T T T
TIME TIME TIME
T
epy
= charging period
= thyratron peak voltage
(a) (b) (c)
charged. When delayed triggering of the charging diode is required in order
to effect an increase of time available for recovery of the modulator switch
tube, care must be taken in the design of the trigger circuit. It is possible
for triggering to be caused by RF generated in the modulator when the
switch tube fires, due to the ease with which the charging diode is trig-
gered. This will give prefiring before the delayed trigger voltage appears. A
low impedance drive circuit will help to overcome this.
8.2.4. Thyra-
tron Anode
Circuit
Associated with the thyratron anode in conventional arrangements are the
pulse forming network and inverse diode, and the most important parame-
ters to be considered are rate of rise of current through the thyratron
(di/dt) and inverse voltage at the anode (epx) after the main pulse.
A high rate of rise of current will increase leading-edge heating and also the
rate of gas clean up in the tube. The latter is particularly important in non-
reservoir tubes. The mean rate of rise of current as shown on an oscillo-
scope trace is not necessarily a good indication, since the peak value may
be considerably higher than this due to stray capacitance from the network
to ground, and from charging and inverse diode heater transformers. For
this reason the three-terminal network shown in Figure 3.22 is preferred
since the bulk of these stray capacitances are removed from the thyratron
anode.
The end inductive section of the network should be mounted close to the
thyratron anode, and will have a minimum value . If di/dt is in
amperes per microsecond, then LA is calculated directly in microhenries.
It must be realized that the mean current through the thyratron is the sum
of the DC power supply current and the inverse diode current (if any).
8.2.5. Inverse
Diode Circuit
The inverse diode or clipper tube is normally found in high-power equip-
ment to protect both the load and thyratron. In the event of a short circuit
in the load the circuit impedance is halved and the peak current through
the thyratron is doubled. With no inverse diode in the circuit the thyratron
anode voltage swings negative to a value approaching the peak positive val-
ue and the succeeding charging cycle starts from there. This cycle will pro-
vide a positive voltage tending to double the previous value and, if the short
in the load persists, this amplification will continue until limited by circuit
losses or component failure. The inverse diode circuit must dissipate the
energy involved. Depending on the modulator conditions, the resistance of
this circuit may lie between 4Z
n
and about 40Z
n
. The lower value is to be
preferred for fast removal of inverse voltage but higher values may be nec-
essary to permit recovery of the switch tube. A value of around 40Z
n
will
limit the increase in peak positive voltage to approximately 1% following a
short circuit where the modulator duty ratio is 0.001.
An inductive overswing circuit usually results in the inverse voltage rating
of the thyratron being exceeded, due to the time of application being longer
than in the resistive case. The primary inductance of the pulse transformer
may be sufficient to produce this effect, particularly If the anode of the in-
verse diode is connected directly to earth.
Any excess of inverse voltage in amplitude or time may result in arc-back
and overheating of the grid structure, with the possibility of distortion in
extreme cases.
The rise of negative voltage appearing on the thyratron may be very rapid
and the amplitude of the spike is approximately the same whether a diode
is used or not. However, as soon as the diode conducts, the voltage rapidly
L
A
epy
di dt
-------------- =
collapses to a value determined by the inverse diode circuit impedance.
When pulse transformer loads are used the effect is to delay the rise of the
inverse voltage and thus tend to reduce the possibility of arc-back in the
switch tube. Vacuum diodes are able to withstand higher voltages than gas
tubes but present a higher impedance and are more likely to suffer failure
in the event of a prolonged period of reduced load impedance. For this rea-
son the FX297 is widely used, preferably triggered by a simple pulse trans-
former fed from the main current pulse, or alternatively by a capacity divid-
er.
8.2.6. The Trig-
ger Circuit
The values of jitter and anode delay time drift quoted in tube data sheets
are extreme values to cover variation during life and are measured under
conditions of minimum trigger amplitude, drive current and rate of rise of
voltage. In practice, the trigger signal applied to the thyratron grid should
be from a low impedance source and should have a high rate of rise of volt-
age and a pulse amplitude sufficient to cause rapid ionization of the gas in
the grid-cathode space. This will minimize jitter and anode delay time drift.
A trigger pulse duration very much in excess of the minimum value speci-
fied on the tube data sheets is wasteful and may, in certain circumstances,
inhibit recovery of the thyratron after the main current pulse has been
switched. In tetrodes, excessive current to grid 1 or excessive delay in the
application of the grid 2 pulse may lead to a deterioration in tube perfor-
mance. Any modulation superimposed on these parameters or on the bias
supply will show up as jitter on the main current pulse.
The minimum trigger pulse amplitudes quoted in data sheets refer to cath-
ode potential, and the value of any negative bias used should be added to
this figure to give the required minimum unloaded pulse amplitude from
the trigger generator. This amplitude must always be checked at the thyra-
tron socket with the tube removed.
At the instant of firing, the grid potential rises rapidly. A voltage spike of the
order of 20ns duration and an amplitude which is an appreciable fraction
of the thyratron anode voltage, may be observed on the oscilloscope trace of
the grid waveform. This spike can cause breakdown in the trigger unit. The
amplitude of this spike increases as the rate of rise of current in the tube is
increased. A series resistor, placed adjacent to the grid terminal and aided
by the stray capacity, may provide a filter against the spike. A filter network
is sometimes necessary, but should be kept as small as possible since it
will degrade the grid firing pulse front and thereby increase anode delay
time drift. Alternatively a nonlinear resistor may be used in parallel with
the trigger unit output.
8.2.7. Recov-
ery
At the end of the current pulse, a plasma exists throughout the tube which
presents a short circuit to positive anode voltages. Therefore, the circuit is
arranged so that the thyratron anode is held at a slight negative potential
until recovery is complete. The tube has recovered when re-application of a
positive anode voltage does not cause further conduction. During the re-
covery period the plasma decays with time as recombination of ions and
electrons mainly on electrode surfaces. Since there are relatively close
spacings in the tube between anode and grid, and also in the grid, the plas-
ma density drops rapidly in this region with a time constant of the order of
a microsecond. The grid-cathode plasma decays much more slowly because
of the wider gaps involved and retains the inherent nature of a plasma by
having approximately equal numbers of electrons and ions. Recovery is
complete when the grid cathode plasma has shrunk away from the grid ap-
ertures so as not to come under the influence of any applied anode voltage,
but deionization is not complete until all the ions have disappeared from
the tube. It should he noted that a tube has recovered long before it has
deionized.
Negative bias applied to the grid decreases recovery time. The bias voltage
does not pull ions out of the discharge but when the grid potential goes
negative with respect to the cathode, the anode field penetration is reduced.
Grid bias may be applied via a resistor or an inductor. Both of these must
inherently retard the application of negative bias but they are necessary to
enable a forward drive pulse to be applied economically. The trigger unit
output stage may be a cathode follower, a blocking oscillator or a pulse
transformer. Care is needed in the design of the associated circuitry since
not only must it apply a positive pulse to the grid but it must also be capa-
ble of passing several amperes of deionization current which follow the
pulse. A paper by Malter and Johnson (RCA. Rev., June 1950) shows the
change in plateau length and exponential decay of positive ion current to
the grid of a small thyratron. The wave shape is shown in Figure 3.24.
Figure 3.24. Circuit for Recovery Time Measurement
r
s
E
s
rRg
2 F
E
cc
G
r
i
d

V
o
l
t
a
g
e
Cathode
Potential
Bias
Potential
Exponential Decay
Time
Grid Waveform after Current Pulse
g
150V
Figure 3.25. Methods of Providing Negative Grid Voltage Swing
2 F
2 F
Bias persisting because
of capacitor
of capacitor
Exponential
decay
G
r
i
d

V
o
l
t
a
g
e
(a)
Cathode
Potential
Inductive Swing
Time
G
r
i
d

V
o
l
t
a
g
e
(c)
Cathode
Potential
Inductive Swing
Time
G
r
i
d

V
o
l
t
a
g
e
(b)
Cathode
Potential
Inductive Swing
Time
G
r
i
d

V
o
l
t
a
g
e
G
r
i
d

V
o
l
t
a
g
e
(d)
(e)
Cathode
Potential
Cathode
Potential
Inductive Swing
Time
Time
The published data shows recovery curves for various values of bias and re-
covery resistance. All these measurements have been made in a specially
developed circuit which presents a purely resistive impedance between grid
and bias (if any) during the recovery period (Figure 3.24). It will be noted
that the zero bias recovery time is independent of recovery resistance.
These recovery times are much longer than those achieved from a pulse
transformer drive since the energy stored in the pulse transformer causes
an inductive swing of the grid which produces an effective bias in the grid
and facilitates recovery Figure 3.25(a). This improvement may also be
achieved by the insertion of an inductance of around 6mH directly between
grid and cathode Figure 3.25(b). Figure 3.25(c) and Figure 3.25(d) show
similar effects when a capacitor is included as a virtual bias source. Some
advantage may be gained from the longer persistence of the bias and con-
trol of positive overshoot. Cathode follower outputs can also produce ap-
preciable bias, especially at high repetition frequencies (e.g., 50kHz), from
the energy stored in the coupling capacitor (Figure 3.25e).
When a bias is applied via an inductor, a voltage (e = L di/dt) builds up due
to the deionization current more negative than the bias and providing the
subsequent swing back to the bias level is slow enough, a reduction in re-
covery time will be obtained. In the case of bias applied through the sec-
ondary of a pulse transformer excessive damping may inhibit recovery (Fig-
ure 3.26). A saturable inductance provides a relatively high impedance
during the forward grid pulse but by suitable design this will saturate at
the end of the main anode current pulse and present a low impedance to
the grid and facilitate recovery (Figure 3.27). Again in the absence of any
bias supply the forward pulse drive may be used to produce the bias source
(Figure 3.28).
Recovery impedance is defined as where e
g
is the value of the grid
voltage at any instant during the recovery period, i
c
is the value of the grid
current at the same instant and E
cc
is the grid bias supply voltage.
The published recovery characteristics of thyratrons show that decreasing
the resistance in series with the bias supply or increasing the bias voltage
reduces the recovery time. From inspection of these a value of voltage and
resistance may be found which will guarantee recovery in a particular cir-
cuit for which the time spent negative by the anode is known.
The bias supply must of course be capable of passing a large current with-
out appreciable voltage drop for the duration of the recovery period and
should, therefore, be shunted by a suitable capacitor (between 0.1 and 10
F depending on requirements). It must also be capable of recharging the
capacitor rapidly and should be free from ripple which would cause jitter of
the output pulse.
e
g
E
cc
i
c
-------------------
The modulator engineer will, however, always have a simple check on his
design by using an oscilloscope to display the grid wave shape See Figures
4.29(a) and 4.29(b). A long plateau indicates that either the thyratron is
still conducting because its anode is being held slightly positive or the grid
bias is being applied through a very high impedance. The appearance of in-
stabilities on the grid wave shape as the high voltage is increased is also in-
dicative of insufficient time allowed for recovery of the switch tube. Further
increase of high voltage invariably results in trip outs under these circum-
stances.
Figure 3.26. Waveforms with Pulse Transformer Trigger Drive
Figure 3.27. Inductive Overswing with Bias
Figure 3.28. Inductive Overswing without Bias
Normal pulse transformer driven circuit
G
r
i
d

V
o
l
t
a
g
e
Exponential decay
+ inductive bias
Plateau as deionization current is
limited by circuit impedance
If R is too great, exponential decay
Steady bias level
R
Time
Bias Voltage
2 F
Inductive swing + switching
action of saturation
G
r
i
d

V
o
l
t
a
g
e
Time
Bias Voltage
2 F
of capacitance
G
r
i
d

V
o
l
t
a
g
e
Time
2 F
8.2.8. Thyra-
tron Heater
Voltage
Although not strictly a circuit element, the effects of heater voltage varia-
tions are worthy of mention. A tolerance of 7.5V is normally allowed on
heater voltage but every endeavor should be made to keep within closer
limits. This applies particularly to tubes which include a hydrogen reser-
Figure 3.29. Waveforms Normally Seen During
Modulator Adjustments
AMPS
VOLTAGE
CATHODE CURRENT PULSE
Start of
current
pulse
Arc level
Plateau
TIME
CATHODE CURRENT PULSE
GRID VOLTAGE WAVEFORM
Bias level
Arc level
Plateau
Thyratron anode
ceases to conduct
voir in series or parallel with the cathode heater. Any variation of cathode
heater voltage thus affects gas pressure as well as cathode temperature.
Anode delay-time drift will be affected by changes in heater voltage. A large
proportion of the observed jitter is caused by the field from the cathode
heater and in cases where jitter requirements are to be reliably less than
1.0ns. DC heater supplies should be used.
8.2.9. Cooling The current passing through the tube and in particular the rate of rise of
current are dependent on a sufficient availability of ionized particles. The
discharge path between cathode and anode represents the region of high
temperature and thus there will normally be a distribution of gas density
within the envelope, the density being highest at the coolest spot.
If the envelope is cooled artificially, then a higher than normal gas density
will occur at the envelope giving a low density in the discharge region This
may limit the rate of rise of current demanded by the circuit. thereby in-
creasing the dissipation within the tube, and accelerating the rate of gas
clean-up.
8.2.10. Mount-
ing
Most thyratrons may be mounted in any position although the size of the
larger tubes generally dictates a base-down position. Care should be taken
that strong RF or magnetic fields are kept away from the thyratrons. These
could cause ionization within the tube envelope seriously affecting the
hold-off capabilities of the tube and increasing the recovery time. Some gas
clean-up may also occur.
For medium- and high-power tubes the anode connector should preferably
be of large surface area.
8.2.11. Warm-
up Time
The time quoted on tube data sheets is the minimum necessary for the
cathode to reach operating temperature and for the gas pressure to reach a
minimum value where reservoirs are used. If trigger pulses are applied be-
fore the expiry of the warm-up time then grid/cathode breakdown may be
observed, but this does not mean that the cathode temperature or gas pres-
sure are high enough for full power operation.
If the ambient temperature before warm up is very low (say below 20 C)
then some increase in warm-up time may be necessary.
8.2.12. The
Tetrode Thyra-
tron
The advantages offered by the tetrode over the triode in pulse modulator
use are sufficiently important to warrant special mention. The reduction of
firing time variations in thyratron circuits is always foremost in the minds
of both tube designer and users and, while much can be done by careful
circuit design, the variations obtained with triode thyratrons may still be
excessive for certain applications.
Firing time variations fall into two main classes: (a) long-term variations,
both repetitive and cumulative. The first of these is largely due to the
warming up of tubes and components each time an equipment is used. The
second, a much slower variation is due to gradual reduction of gas pres-
sure during the life of the thyratron; (b) short-term or transient variations
due to various causes including interference from the A.C. heater field.
Variations due to all these causes are reduced when tetrode thyratrons are
employed. Tetrode thyratrons have two grids which are pulsed successively
with a delay of about 1.0 s. Grid 1 has a positive characteristic with re-
spect to the cathode as in the triode, and grid 2 is given a negative charac-
teristic with respect to grid 1 (Figure 3.30). Grid 1 performs its priming
function very much as in the triode while grid 2 performs what may be
termed a gating function when ionization by the grid 1 pulse is well under
way. This results in very rapid and precise takeover by the anode. The neg-
ative bias applied to grid 2 assists recovery in the inter-pulse period and
thus makes the tetrode thyratron eminently suitable for use in high p.r.r.
precision radar equipment.
In order to switch pulses which are successively accurate in time over a
long period the two grids should be pulsed from separate sources with the
grid 2 drive delayed by about 1.0 s.
When only one grid drive pulse is available, this may be made use of as in
Figure 3.31.
Figure 3.30. Schematic Diagram of Tetrode
Figure 3.31. Single Pulse Drive for Tetrode
Anode
Pulse input
100V bias
G2
G1
5k
1k
1k
0.1F 0.001F
A more simple drive circuit is shown in Figure 3.32. It must be stressed,
however, that some of the inherent advantages of the tetrode may be lost by
the use of these circuits but they will be satisfactory for many applications.
An increase in time jitter up to about 5.0ns may result and anode delay
time drift may be as much as 0.1 s as against 0.02 s.
These three circuits all depend on transferring the discharge from the grid
1 electrode to grid 2 and thence to the anode. The grids should not be sim-
ply strapped together, since this may result in a discharge to the grid 1
electrode which would not necessarily transfer to grid 2 and would result in
erratic firing.
8.2.13. The
Parallel
Operation of
Hydrogen
Thyratrons
In common with all gas-filled devices, thyratrons cannot be connected di-
rectly in parallel without some form of impedance in series with each tube.
In repetitive pulse applications, the triode thyratron does not provide a suf-
ficiently precise pulse-to-pulse triggering facility for parallel operation to be
successful. The tetrode tube does not have this defect.
In theory, any number of tetrodes may be parallel connected, although in
practice six is a convenient maximum. Separate pulse-forming networks,
each with its own charging diode, are recommended, since fault conditions
will result in each thyratron discharging its own network and this will not
involve as much energy as in the case of one large network. Further, the
triggering requirements are less complicated as will be shown below. The
following circuits show suitable arrangements for two tubes, but they may
be extended for additional tubes as required.
Figure 3.33 shows a suitable circuit where two driving pulses are available
each pulse being split between the two thyratrons. Figure 3.34 shows an
arrangement using a single driving pulse which automatically provides a
delay on grid 2 to each tube. Figure 3.35 shows the general arrangement of
the anode circuit.
Figure 3.32. Simplified Single Pulse Drive
Pulse input
100V bias
G2
G1
5k
1k
1k
0.1F
Where the use of a single network cannot be avoided, it is essential that
each tube has an anode inductor of the same value and arrangements
should be made for adjustment of the current sharing between tubes. This
is most conveniently achieved using the circuit of Figure 3.34, and making
R2 variable. This provides adjustment of the delay to grid 2 on each tube.
Figure 3.36 shows suitable component values. This arrangement has been
successful in the parallel operation of six CX1140 tubes to provide a pulse
output power of 75MW.
When several thyratrons are being used, a check should be made on the
triggering supplies by removing one of the thyratrons from its socket to en-
sure that the amplitude of the available voltage pulse at that socket is
greater than the minimum specification requirements when all the other
tubes are conducting between grids and cathode.
Figure 3.33. Double Pulse Drive for Parallel Operation
Figure 3.34. Single Pulse Drive for Parallel Operation
Pulse 2
100V bias
I/P
Pulse 1 I/P
G
2
G
2
V
2
V
1
G
1
G
1
R
2
R
2
R
1
R
1
1k
1k
1k
1k
Pulse I/P
R
3
R
3
G
2
G
2
V
2
V
1
G
1
G
1
0.1F 0.001F
0.001F 0.1F
R
2
R
2
R
1
R
1
The rate of rise of current rating of the assembly is the sum of the individ-
ual tube ratings, and this often provides a capability in excess of a single
larger tube.
The driving pulse should have a fast rising front and be of near maximum
rated amplitude to minimize anode delay time and jitter. This applies par-
ticularly to the circuit of Figure 3.33, due to the effect of capacitive loading
on the leading edge. This effect may be considerably reduced as shown in
Figure 3.37 by the use of a variable delay line in series with R2 instead of
the 0.001 F capacitor for applications requiring minimum delay time and
jitter. The values of series resistors given will, in general, provide satisfac-
tory performance, but if other values are used, the grid 1 current must be
kept within the specified limits. With separate networks and separate iden-
tical anode inductances, current sharing between tubes is automatic, and
small differences in delay time between tubes do not affect the perfor-
mance.
Figure 3.35. Anode Circuit Using Separate PFNs
Figure 3.36. Arrangement for Adjustment of Current Sharing
G
2
G
2
G
1
G
1
V
2
V
1
PFN
PFN
Charging
diode
Charging
choke
Anode
Inductor
Anode
Inductor
R
L
I/P
Pulse input
100V bias
G2
G1
5k
1k
5k
0.1F 0.001F
8.2.14. The
Series
Operation of
Hydrogen
Thyratrons
Where extremely high voltages have to be switched, thyratrons may be op-
erated in series without the necessity of insulated trigger supplies. The gen-
eral arrangement is shown in Figure 3.38 with suggested component val-
ues.
Tetrodes are preferred since they generally have a lower anode delay time
than triodes. Furthermore, in the arrangement shown, it is essential that
the switch tube should fire with low current triggering. a property not nor-
mally found in triodes.
The operation is as follows. Tube (1) at the lower voltage is triggered nor-
mally, using separate pulses to each grid, or a single pulse and the conven-
tional RC arrangement. When this tube fires, the voltage at its anode falls
and thus the voltage at the cathode of tube (2) falls. Capacitor (C1) between
grid 1 and grid 2 of tube (2) helps to maintain the resulting high potential
between grid 1 and cathode and thus the tube fires to grid 1, and thence to
grid 2 resulting in complete breakdown.
Figure 3.37. Using a Delay Line to Minimize Time Jitter
R
4
R
2
R
1
R
3 0.1 F
2 F
Bias
60 to 100V
1
Delay Line
S Hydrogen
thyratron
(tetrode)
Figure 3.38. Thyratrons in Series
C
3
C
2
Tube 2
Tube 1
HV
C 100
1
F
10 M
10 M
The insulated heater transformer for tube (2) will have capacitance to
ground (C2) which, if not minimized, will require an equal compensating
capacitor (C3) to be placed in parallel with tube (2). This divider, assisted by
the tube interelectrode capacitance, will automatically provide equal volt-
age distribution between tubes. Added capacitance should be minimized
since this will discharge through the tube, and must be connected to the
network side of the anode inductors.
To avoid the presence of isolated elements, a resistive divider is preferable.
The value of this should not be so low as to cause a serious drain from the
power supply, and should not be so high as to introduce (with the various
capacitances) time constants at variance with the PRF required.
8.2.15. Hydro-
gen and Deu-
terium
The advantages of hydrogen over mercury and the rare gases have already
been mentioned in Part 1 of this article. However, an increasing use is
made of deuterium as the gas filling in place of hydrogen. Chemically the
gas is similar to hydrogen, so that conventional hydrogen reservoirs oper-
ate equally well with deuterium. The Paschen breakdown curve, however,
shows that deuterium is capable of higher hold-off voltages than hydrogen
for the same pressure. The greater mass of the ion, however, means less
mobility and so recovery time is increased (by a factor of for the same
geometry).
At the same time, surface recombination effects are reduced and the arc
loss is lower. For this reason, deuterium is mostly used in high-power
tubes, where recovery time is usually of less importance than hold off volt-
age and dissipation.
8.2.16. Single
Shot and
Crowbar Appli-
cations
The increasing use of high-power klystrons, traveling wave tubes and other
such devices has demanded that some protection should be given in the
event of an internal flash-over or similar fault. This protection may be pro-
vided by a gas-filled spark-gap or thyratron across the main high voltage
supply, which, when triggered by the fault current, effectively short circuits
the high voltage until the main circuit breakers operate.
Since the crowbar device may have to remain quiescent for many thou-
sands of hours until suddenly required, it is desirable that it should con-
sume no power, and essential that it should function with complete reli-
ability when needed.
The spark gap has an advantage over the thyratron in its zero power con-
sumption on standby. However, for both spark gap and triode thyratron,
there is no simple method of ensuring that the crowbar circuit is in a state
of readiness, e.g., the device may have become leaky during a long
standby period.
With tetrode thyratrons the switching action may be controlled completely
by the second grid and it is thus possible to run a discharge continuously
to grid 1 without affecting the peak hold-off voltage under DC conditions
provided bias is applied to the grid. It has been found that a current of
about 100mA continuously to grid 1 in no way affects the life or perfor-
mance of the tube, and that this grid 1 current may then be used as a sim-
ple monitor, e.g., by lighting an indicator bulb, to show that the crowbar
circuit is in a state of readiness.
It is possible for a reversal of high voltage to take place following crowbar
action and before circuit isolating elements have responded. If the trigger
voltage to the crowbar tube has been removed, then high voltage will be re-
applied to the load as this voltage becomes positive again. It is thus desir-
able that the trigger voltage waveform should be a train of pulses rather
than a single pulse.
The peak forward voltage given in the data sheets for these thyratrons is
the rating for repetitive pulse work, and is normally reduced for DC holdoff
applications. At the same time the peak current may be considerably in-
creased provided the repetition frequency is not greater than 1 per 10 sec-
onds.
2
RF Theory 4-1
This chapter will discuss microwave RF system theory, including microwave technology,
in order to familiarize the student with the application of this technology to the func-
tional operation of the Clinac RF System.
RF Theory
4-2 RF Theory
Table of Contents
1. Introduction: .................................................................................................................... 4-5
2. Traveling Waves on Transmission Lines:........................................................................... 4-5
3. Waveforms: ...................................................................................................................... 4-8
4. Waveguides: ................................................................................................................... 4-11
5. Resonant Circuits: ......................................................................................................... 4-14
6. RF Transmission Theory: ............................................................................................... 4-16
7. RF Waveguide Design: .................................................................................................... 4-17
7.1. Modes:.................................................................................................................... 4-17
7.2. The TE10 Mode:...................................................................................................... 4-18
7.3. Coupling:................................................................................................................ 4-18
7.4. Determining the TE10 Dominant Mode of a Waveguide: .......................................... 4-19
8. Transmission Lines: ....................................................................................................... 4-20
8.1. VSWR: .................................................................................................................... 4-21
9. Vector Analysis 3dB Quadrature Hybrid: ..................................................................... 4-21
10. Circulators: .................................................................................................................. 4-22
11. Klystron Theory:........................................................................................................... 4-24
11.1. Theory of Klystron Operation: ............................................................................... 4-24
11.2. Associated Equipment: ......................................................................................... 4-34
11.3. Power Supplies: .................................................................................................... 4-35
11.4. Cooling: ................................................................................................................ 4-39
11.5. RF Circuits: .......................................................................................................... 4-45
11.6. Tuning:................................................................................................................. 4-49
11.7. Noise in Klystron Amplifiers: ................................................................................. 4-51
11.8. Summary:............................................................................................................. 4-52
RF Theory 4-3
Figure 4.1. Propagation of a Step Wavefront on a Transmission Line:.................................... 4-5
Figure 4.2. Infinitely Long Line: ............................................................................................ 4-6
Figure 4.3. Finite Line Terminated by its Characteristic Impedance: ..................................... 4-6
Figure 4.4. Traveling Waves on an Open-end Line:................................................................ 4-7
Figure 4.5. Traveling Waves on a Shorted-end Line: .............................................................. 4-9
Figure 4.6. Voltage and Current Waves on a Shorted-end Line: ........................................... 4-10
Figure 4.7. Parallel Strip Lines:........................................................................................... 4-11
Figure 4.8. Waves on a Parallel Strip Transmission Line: .................................................... 4-12
Figure 4.9. Electric Field Distribution: ................................................................................ 4-13
Figure 4.10. Electric and Magnetic Fields in a Closed Rectangular Waveguide: ................... 4-13
Figure 4.11. Electric and Magnetic Fields of the Dominant TE10 Mode in a Rectangular
Waveguide: ......................................................................................................................... 4-14
Figure 4.12. Tuned Resonant Circuit: ................................................................................. 4-15
Figure 4.13. Resonant Circuit Having Two Natural Frequencies: ......................................... 4-15
Figure 4.14. WR 284 RF Waveguide: ................................................................................... 4-16
Figure 4.15. RF Waveform Divided into -Wavelength Intervals:......................................... 4-16
Figure 4.16. RF Rectangular Waveguide Dimensions: ......................................................... 4-17
Figure 4.17. The Transverse Electric (TE) Mode: ................................................................. 4-18
Figure 4.18. The Transverse Magnetic (TM) Mode:............................................................... 4-18
Figure 4.19. Loop Coupling:................................................................................................ 4-19
Figure 4.20. Probe Coupling: .............................................................................................. 4-19
4-4 RF Theory
Figure 4.21. Frequency Determining Parameters: ............................................................... 4-19
Figure 4.22. Characteristic (Surge) Impedance: .................................................................. 4-20
Figure 4.23. PFN Parameters:............................................................................................. 4-20
Figure 4.24. 3dB Quadrature Hybrid Vector Analysis: ........................................................ 4-21
Figure 4.25. 3-Port Circulator: ........................................................................................... 4-22
Figure 4.26. Low Energy Clinac 4-Port Circulator and Phase Diagram: ............................... 4-22
Figure 4.27. High Energy Clinac 4-Port Circulator and Phase Diagram (Used in Shunt Tee
Clinacs):............................................................................................................................. 4-23
Figure 4.28. High Energy Clinac 4-Port Circulator and Phase Diagram: .............................. 4-23
Figure 4.29. Triode Vacuum Tube Amplifier: .......................................................................... 4-25
Figure 4.30. Sectional View of a Klystron: .............................................................................. 4-26
Figure 4.31. Generation of Alternating Current in a Cavity:.................................................... 4-28
Figure 4.32. High-Power Four-Cavity Klystron: ...................................................................... 4-30
Figure 4.33. High-Power Four-Cavity Klystron, Simplified: ..................................................... 4-31
Figure 4.34. Effect of Tuning on Klystron Performance:.......................................................... 4-32
Figure 4.34. Effect of Tuning on Klystron Performance:.......................................................... 4-32
Figure 4.35. Klystron Power Supply Connections: .................................................................. 4-35
Figure 4.36. Typical Liquid Cooling System for a Klystron Amplifier: ...................................... 4-42
Figure 4.37. Typical RF Circuitry for a Klystron Amplifier: ..................................................... 4-46
RF Theory: Introduction 4-5
1. Introduction The material in this chapter should be read and thoroughly understood by
the student before proceeding to the specific RF system descriptions in the
High and Low Energy Clinac Beam Delivery System course manuals.
2. Traveling
Waves on
Transmission
Lines
Figure 4.1 has been drawn to show the mechanism by which a step wave-
front (pulse) travels along a real transmission line.
When switch S1 is closed, the battery with a terminal voltage of E
dc
is ap-
plied to the input of the transmission line. The voltage E
dc
does not appear
instantly at all points along the line. Instead, a wave of voltage progresses
along the line. The farther from the battery a given point on the line is, the
later the time at which the line voltage at that point jumps from 0 to E
dc
. A
current wave also travels along the line exactly in step (in phase) with the
voltage wave. Current flows away from the battery in the top conductor and
returns to the battery in the bottom conductor. Plus and minus signs are
placed on the conductors at points where voltage exists between the con-
ductors, and magnetic flux lines are shown encircling the conductors wher-
ever current is flowing.
The reason that the line cannot charge all at once is due to the existence of
series inductance along each conductor as well as shunt capacitance
across the conductors. The voltage wave can progress only as fast as the
line current can carry a charge to the wavefront to produce the change in
voltage. The current wave can progress only as fast as the voltage can de-
velop across each short section of conductor at the wavefront, thus starting
current in that corresponding section of line inductance. The voltage and
current waves must move in phase with each other along the line. The rate
of travel (propagation) along the transmission line is proportional to the
square root of the inductive and capacitive reactance per unit length of line.
Figure 4.1. Propagation of a Step Wavefront on a Transmission Line
S1
i = I
e = E
E
dc
i = I
q
q
e = 0
i = 0
e
i
0
0
S
S
E
I
T LC =
4-6 RF Theory: Traveling Waves on Transmission Lines
The characteristic impedance of the line is proportional to the square root
of the quotient of the inductive and capacitive reactance per unit length of
line.
Thus far, only the start of a wave from the input terminals of a line and the
propagation along a line of infinite length have been discussed. The effects
that occur when a wave reaches the end of a finite length of line must also
be considered.
In Figure 4.2 a 100-volt battery has been connected through a 1K series re-
sistance to an infinitely long transmission line with the same characteristic
impedance as the series resistance.
The circuit of Figure 4.3 is the same except that the line has been short-
ened to a length equal to 1 microsecond of propagation time and terminat-
ed in its characteristic impedance of 1K ohms. In both circuits, a 50-volt
wave, and 50-mA wave travel along the line from the input terminals after
the switch is closed.
For the first microsecond of time both lines have the same propagation
characteristics, because the remainder of the infinitely long time is equiva-
lent to a 1K resistance also. However, the circuit of Figure 4.3 will develop
a steady-state condition as soon as the wave reaches the 1K termination re-
sistance. Once in the steady state, 50 volts appears across the conductors
of the line, and a 50-mA current will continue to flow through the conduc-
tors and terminating resistance, as expected when two 1K ohm resistors
are connected in series across the 100 volt battery.
The circuit of Figure 4.4 (A) uses the same 1 microsecond line segment with
the end left unterminated. When the switch is closed, the line first appears
as a 1K impedance connected in series with the 1K resistors. This is be-
cause the end terminal conditions do not affect the wavefront before it
reaches the end of the line. A 50-volt, 50-mA wave starts along the line from
the input terminals. The current must be zero at the end of the line at all
times due to the open circuit. To maintain this zero current condition when
the wavefront reaches the end of the line, a second 50-mA current wave
must start back along the line toward the battery. There must also be a re-
Figure 4.2. Infinitely Long Line
Figure 4.3. Finite Line Terminated by its Characteristic Impedance
R
z
L
C
--- - =
S1 i
d
e
d
1K
R = 1K
C
100V
i
e
S1 i
d
e
d
1K
R = 1K
C
100V
1 S Travel Time
i
e
RF Theory: Traveling Waves on Transmission Lines 4-7
flected voltage wave associated with the reflected current wave. This is be-
cause the characteristic impedance of the line also relates to the quotient of
the voltage and current wave.
The voltage-wave amplitude is 50 volts thus causing the total line voltage to
rise to 100 volts. These traveling waves are indicated in Figure 4.4 (B) and
corresponding time variations of e and i are shown by Figure 4.4 (C).
Figure 4.4. Traveling Waves on an Open-end Line
R
c
also
E across line ( )
I across line ( )
---------------------------------- =
0
0
A. Open End Transmission Line
B. Traveling Wave
C. Time Waveforms
e
e
d
i
i
d
s
s
s
s
t
t
S1 i
d
e
d
1K
R = 1K
C
100V
i
e
50V
50V
50V
50mA
100V
100V
0
0
t = S
t = 1 S
0 2 S
2 S
e
d
i
d
0
50mA
50mA
4-8 RF Theory: Waveforms
3. Waveforms The incident (forward) wave of Figure 4.4 (A), in traveling from the battery
to the open end of the line, charges the line capacitance from 0 to 50 volts
because of the 50-mA current supplied by the battery. When the wave
reaches the open end, no capacitance remains to be charged. The 50-mA
current is, however, maintained by the line inductance and thus charges
the capacitance near the open end of the line to twice the battery voltage.
This voltage is produced across the line near the open-end in a direction to
reduce the line current toward zero.
This increase of voltage and decrease of current near the open end of the
line corresponds to a reflected wave traveling back toward the battery. As
the reflected wave progresses back to the battery, the entire line is charged
to 100 volts and the total current on the line is reduced to zero. Thus, a
static condition is reached with the line at 100 volts, and no further chang-
es occur in the circuit.
In the circuit of Figure 4.5 on Page 4-9 a short circuit is placed at the end
of the section. As a result, the voltage at the short circuit must always be ze-
ro, and at the instant the 50-volt, 50-mA incident wave reaches the short
circuit, a reflected voltage wave of 50 volts amplitude and of a reversed po-
larity starts toward the battery. A current wave of 50 mA amplitude is also
associated with the reflected voltage wave, and the current in this wave
flows in the same direction as the incident current wave.
After the reflection is over, a current of 50 mA flows from the battery into
the moving wavefront, and a current of 100 mA flows from the wavefront to
the short circuit. Thus, the line capacitance at the wavefront is discharged
by the net current of 50 MA, and the voltage across the conductors drops to
zero. When the reflected wave reaches the battery, the line voltage is zero at
all points and the current is 100 mA. No further changes occur because
100 mA is exactly the current drawn by the 1K resistor connected in series
with the shorted line section across the 100 volt battery.
A fact of fundamental importance is that a wave of any shape can be prop-
agated along a transmission line without any change of shape or magni-
tude. The voltage wave is always accompanied by a current wave of similar
shape. Thus for RF frequencies a sine wave generator rather than a battery
can be connected to the transmission line, and accordingly the same func-
tions occur with the sine wave as occurred with the battery.
When a line is terminated in an impedance that is different from the char-
acteristic impedance (mismatched) the voltage and current on the line are
no longer the result of a single wave traveling from generator to load. In-
stead, the total voltage and current are the algebraic sums of two waves
traveling in opposite directions.
As an example, consider the diagram of Figure 4.6 on Page 4-10 in which a
sine wave generator with an internal resistance of RL is connected through
a switch S to a transmission line of impedance RZ and terminated in a
short circuit. The behavior of the line after the switch is closed is similar to
that of a shorted line with a battery as the source (reference Figure 4.5), ex-
cept that sinusoidal waves of voltage and current travel along the line.
In order that the voltage at the short circuit may be zero at all times, the
voltage reflected back at the short circuit must be equal in magnitude to
the incident voltage, but reversed in polarity.
The generator has an internal resistance which is equal to the line imped-
ance RZ. This causes the line to behave with respect to the reflected waves
exactly as if the line were terminated in its characteristic impedance result-
ing in steady state conditions when the reflected wave reaches the input
terminals.
RF Theory: Waveforms 4-9
The incident and reflected voltage and current waves are shown by Figure
4.6 on Page 4-10 (B) and (C) respectively. The resultant sum of these waves
is indicated by Figure 4.6 (D). The total voltage at any point on the line var-
ies as the individual waves move in opposite directions along the line.
At time (t1), an instant slightly later than that for which Figure 4.6 (B) and
(C) are drawn, the peak of the incident voltage wave is at the short circuit
end, and the component voltages have equal magnitude and opposite po-
larity at each point on the line. Thus, the voltage would be zero everywhere
on the line as indicated by the line marked t1 in Figure 4.6 (D).
Figure 4.5. Traveling Waves on a Shorted-end Line
0
0
A. Shorted Transmission Line
B. Traveling Wave
C. Time Waveforms
e
e
d
i
i
d
s
s
s
s
t
t
S1 i
d
e
d
1K
R = 1K
C
100V
i
e
50V
50V
50V
50mA
0
0
t = S
t = 1 S
0 2 S
2 S
e
d
i
d
0
50mA
50mA
1 S Travel Time
100mA
100mA
4-10 RF Theory: Waveforms
At other instants, cancellation of the individual waves does not occur. For
example, at time (t2) a quarter period later than time (t1) a zero value of the
component waves in Figure 4.6 (B) and (C) is at the short circuit end, and
the total voltage marked by the curve (t2) in Figure 4.6 (D) is a sine wave of
twice the amplitude of either component wave. Looking at additional in-
stants of time (t3) and (t4) a quarter period later, etc., is indicated by the
curves (t3) and (t4) respectively in Figure 4.6 (D).
The total voltage and current patterns of Figure 4.6 (D) are defined as
standing waves. By performing additions at other intermediate times to the
quarter period instants, it may be shown that the total voltage and current
have a sine wave distribution along the line with zeroes at the short circuit
and half-wave intervals from the short circuit. The amplitude and polarity
of this sine wave varies as the voltage at each point changes sinusoidally
with time. The points of zero voltage are called voltage nodes, and the
points of maximum voltage are called antinodes.
Figure 4.6. Voltage and Current Waves on a Shorted-end Line
i
+
i
A. Circuit Diagram
B. Traveling Waves of Voltage
C. Traveling Waves of Current
D. Standing Waves
e
i
R
S
S
e
+
wave e
wave
e
e
t ,t
1 3
t
2
t
4
i
i
t
1
t ,t
2 4
t
3
wave wave
2e
+
m
2i
+
m
RF Theory: Waveguides 4-11
Referring back to Figure 4.1 on Page 4-5, it should be noted that a magnetic
flux or field is associated with the inductance of a transmission line and
that an electric field is associated with the capacitance of the line. Thus,
the energy in terms of volts and amperes can be translated into units of
electric field flux and magnetic field flux. The latter two conditions are to be
found within an RF field that is propagated through free space.
4. Waveguides A waveguide performs the same function at microwave frequencies as a
two-wire transmission line does at lower RF frequencies. The waveguide
can be considered to possess all the same principles of operation discussed
in the preceding sections. However, the transmission of energy is dealt with
in terms of the electric and magnetic fields rather than in terms of voltage
and current. A special form of transmission line, called a parallel strip line
and pictured in Figure 4.7, can be used to define the electric and magnetic
field intensities related to the voltage and current waves on a transmission
line.
The line of Figure 4.7 (A) is composed of two parallel flat-plate conductors
of width (w) and spacing (h). The line is considered perfect with no losses
with zero resistance in the conductor plates and zero conductance in the
dielectric (air) between the plates. Incident and reflected waves of voltage
and current can be propagated along the line in a like manner to the two
wire transmission line of Figure 4.1 on Page 4-5.
Consider that sinusoidal waves are traveling along the line and let the volt-
age be defined between the plates and the current along the plates be de-
noted by e and i respectively.
The two conductors can be considered as plates of a capacitor. The charge
on the capacitor appears on the inner surfaces of the plates, and an electric
field is developed between the plates because of the voltage (e). The charge
and field flux at any cross section point increases and decreases in phase
with the voltage wave at that section. The pattern of the electric field is in-
dicated by the solid arrows of Figure 4.7 (B). If (w) is much larger than (h),
the electric field will be uniform except at the edges of the strip plates.
The current in the conductor plates develops a magnetic field that encircles
each conductor. The direction of the field is related to the direction of the
instantaneous current on the conductor plates by the right-hand rule and
is indicated by the dashed line arrows of Figure 4.7 (B). Because of the
plane geometry of the conductors, the magnetic field is uniform in most of
the region between the conductors. Because the conductors have no resis-
tance, the current will flow entirely on the surface of the conductors. Thus,
the magnetic field as well as the electric field is prevented from entering the
conductors. Figure 4.8 shows the resultant waves in the strip line.
Figure 4.7. Parallel Strip Lines
h
w
i
i
Direction of
Propagation
e h
i (out)
i (in)
w
(A) Pictorial Representation (B) Cross Section
e
4-12 RF Theory: Waveguides
The waves in the parallel-strip line are a very simple example of many
forms of electromagnetic waves that can take place. When the conducting
boundaries have a more complex shape, a great variety of more complex
field patterns can be generated. The waves of the parallel strip line may be
described as fundamental or simple transverse linearly polarized waves. A
wave of this type for which both the electric and magnetic field vectors lie in
the transverse plane of the conductors (the transverse plane is the plane
that is tangent to the conductor surface and perpendicular to the direction
of propagation of the wave on the conductor) is defined as the TEM wave or
mode. If only one field of vectors lies in the transverse plane the wave is ei-
ther a transverse electric (TE) or transverse magnetic (TM) wave.
It is possible to take the parallel strip line and develop a waveguide struc-
ture from it. If the frequency of the wave to be propagated along the strip
line is high enough so that it has a half wavelength value that is equal
to or less than the width of the plates, there will exist across the plates an
electric field distribution as shown in Figure 4.9 on Page 4-13.
Figure 4.8. Waves on a Parallel Strip Transmission Line
(A) Pictorial Representation
(B) Voltage and Current Waves
(C) Electric and Magnetic Field Waves
Direction of
Propagation
i
i
e
i
E
H
e
2
RF Theory: Waveguides 4-13
The magnetic field will be as indicated before at right angles to the electric
field as referenced in Figure 4.7 (B).
If there is no voltage or electric field present at the sides of the plates, we
could connect them as shown in Figure 4.10.
Now the magnetic field will be contained within the walls of the structure.
We now have a waveguide transmission line that has all the same charac-
teristics and follows the same laws of propagation as a plane transmission
line.
It is possible to propagate several different types of electromagnetic waves
within a waveguide. Each wave is characterized by a different electric and
magnetic field configuration. Associated with each wave type is a cutoff fre-
quency below which, for a particular size guide, the propagation becomes
impossible. In general the larger the guide, the lower the cutoff frequency of
each type of wave.
The wave with the lowest cutoff frequency is defined as the dominant wave
or mode. To transmit the dominant mode, a rectangular guide must have a
width of at least half the free space wave length of that wave. The height is
not critical and is usually made about one half the width.
In waveguides, the electric and magnetic fields are confined to the space
within the guide walls. Thus, all power is transmitted along the guide with
no radiation nor dielectric losses of any practical importance.
The electric field (voltage) is defined across the height or short dimension of
the guide as in the parallel strip line and the magnetic field current is de-
fined parallel to the opposite walls of the waveguide as identified by Figure
4.11 on Page 4-14.
Figure 4.9. Electric Field Distribution
Figure 4.10. Electric and Magnetic Fields
in a Closed Rectangular Waveguide
(B) (A)
2

2
4-14 RF Theory: Resonant Circuits
Note that the letters TE define the electric field as transverse and the sub-
scripts 1 and 0 indicate the number of half cycle variations of the elec-
tric field along the width and height dimensions of the guide in accordance
with industry convention of identification.
5. Resonant
Circuits
Circuits composed of lumped inductive and capacitive elements can be
made to resonate at any desired frequency by selection of the values of in-
ductance and capacitance used. To increase the frequency, the size of the
inductance and capacitance must be made smaller. At extremely high fre-
quencies the electrical size as well as the physical size of the components
becomes so small that the stray inductance and capacitance of the circuit
begins to affect the resonant frequency of the circuit. This results in differ-
ent construction techniques being employed to raise the frequency higher.
In the UHF region parallel wire or coaxial cable transmission lines are used
in place of the lumped components. At microwave frequencies resonant
cavities are used.
All forms of resonant circuits, whether lumped components, transmission
lines or cavities, have certain natural frequencies of oscillation. A natural
frequency of oscillation is that which can be sustained by the circuit (as-
suming it has no losses) once started that will continue indefinitely even
though there is no internal connection to a power source. Each natural fre-
quency of oscillation is called a natural mode or resonant mode. The term
mode indicates the manner of oscillation. Various modes may occur de-
pending on the nature of the design and excitation parameters.
Figure 4.12 on Page 4-15 shows a simple parallel resonant circuit consist-
ing of a single inductor L1 and capacitor C1.
Assume the capacitor C1 is charged before switch S1 is closed. When S1 is
closed, the circuit will oscillate at some discrete frequency dependent on
the fact that both L1 and C1 must have equal reactance. The frequency can
be determined by:
Figure 4.11. Electric and Magnetic Fields of the Dominant TE
10
Mode
in a Rectangular Waveguide
B
B
A A
(A) TOP VIEW
Direction of propagation
8
2
SECTION B-B
(B) END VIEW
SECTION A-A
(C) SIDE VIEW
f
0
1
2 LC
------------------ =
RF Theory: Resonant Circuits 4-15
Both voltage and current vary sinusoidally with time, and the mode pattern
may, therefore, be described by specifying the phase and magnitude of the
voltage and current during the oscillation. Because the reactance of L and
C are equal at only one frequency, the circuit of Figure 4.12 has only one
mode of oscillation.
The circuit of Figure 4.13 has two natural modes of oscillation. (Note the
circuit is drawn to resemble a two-section transmission line shorted at one
end and open-circuited at the other).
Below the frequency (f
0
) at which L1 and C1 and also L2 and C2 resonate,
the reactance of the parallel combination of L2 and C2 is inductive. The L2
and C2 combination in series with L1 may, therefore, be considered an in-
ductance that resonates with C1 at some frequency less than (f
0
). Similarly,
the L2 and C2 combination and C1 in series is less than C1 and resonates
with L1 at a frequency greater than (f
0
).
From the foregoing discussion a shorted transmission line can be consid-
ered to have many frequencies at which it could oscillate dependent on how
it is excited.
Resonant cavities can be considered as an extension of the transmission
line into a single physical metal walled chamber fitted with suitable devices
for admitting and extracting electromagnetic energy.
One example of a resonator cavity is a rectangular box that may be thought
of as a section of rectangular waveguide closed at both ends by conducting
plates. Because the end plates appear as short circuits for a wave traveling
along the waveguide, the cavity is analogous to a transmission line section
with a short at both ends.
Resonant modes will occur at frequencies for which the distance between
the end plates is a multiple of half the guide wavelength. Higher order
waves as well as the dominant wave may give rise to several resonant fre-
quencies, and thus a great variety of resonant modes of operation are pos-
sible.
Figure 4.12. Tuned Resonant Circuit
Figure 4.13. Resonant Circuit Having
Two Natural Frequencies
S1
C1 L1
L1 L2
C1 C2
4-16 RF Theory: RF Transmission Theory
The cavity formed by closing the ends of a section of waveguide is only one
of many cavity configurations that can be used to generate microwave en-
ergy. By appropriate choice of cavity shape, advantages such as ease of
tuning, compactness, simple mode spectrum, and high Q may be obtained.
6. RF Trans-
mission Theo-
ry
The RF waveguide used on our machines is called WR 284 or JAN RG
48/U. It has a usable frequency range (S Band) of 2.6 to 3.95 GHz in the
TE
10
mode and its dominant TE
10
mode is 2.5 GHz.
RF waves travel 186,000 miles/sec or 300 10
6
meters/sec in air. There-
fore, the wavelength (8) of one cycle of a particular frequency, measured
crest to crest, is equal to:
Where:
8 = the wavelength of one cycle in meters.
300 10
6
= the distance in meters covered at a speed 186,000
miles/sec.
To convert from length in meters to inches divide by 0.0254. To convert
from length in inches to meters multiply by 0.0254.
Figure 4.14. WR 284 RF Waveguide
Figure 4.15. RF Waveform Divided into -Wavelength Intervals
2.875" I.D.
1.312"
I.D.
300 10
6
f (in Hz)
----------------------- =
300
f (in MHz)
-------------------------- =
4

2

4
RF Theory: RF Waveguide Design 4-17
RF waves travel somewhat slower in waveguides or parallel lines, depend-
ing on material and conditions, as much as 15% slower. Therefore, the dis-
tance measured crest-to-crest will be shorter compared with the same fre-
quency in air.
Example:
A quarter-wavelength of a particular frequency on any transmission
line is calculated using the formula:
Where:
L = in feet.
K = Transmission line constant (1.000 for air, usually 0.975 for
parallel line to 0.85 for air dielectric coax).
f = Frequency in MHz.
7. RF
Waveguide
Design
Figure 4.16 illustrates the design of a typical rectangular RF waveguide.
Dimension (a) is usually of the frequency to be transmitted, in air.
The low frequency cutoff of the guide is roughly twice its width converted to
wavelength (TE
10
dominant). Dimension (b) is determined by the power to
be generated and is usually 0.2 to 0.5 times the wavelength in air.
Bends required in waveguides should be made so that the radius is no
shorter than 2 wavelengths.
7.1. Modes There are basically two methods of transmitting energy through a
waveguide, the TE and TM mode. The TE mode is most commonly used be-
cause:
! It is easy to excite.
! It is plane polarized.
! It is each to match to a radiator.
! Its cutoff frequency is dependent on only one guide dimension (a)
therefore easy to design.
4 ( )
L
246 K
f
------------------- =
4 ( )
Figure 4.16. RF Rectangular Waveguide Dimensions
a
b
1.3 2
4-18 RF Theory: RF Waveguide Design
In the TE mode, the electric field lies in transverse planes that contain the
X and Y axes, and the E lines are parallel to the Y axis and perpendicular
to the Z axis of the guide, as shown in Figure 4.17.
The magnetic field is composed of closed loops that lie in the transverse
plane that contains the X and Y axes and are wholly transverse to the
guide's Z axis, as shown in Figure 4.18.
7.2. The TE
10

Mode
The digits after the mode specify the following:
1. The first digit states the number of variations in the wide or
transverse part of the waveguide.
2. The second digit states the number of variations in the narrow
section of the waveguide.
7.3. Coupling There are two methods of extracting energy from a waveguide, loop and
probe coupling. In loop coupling, a small loop is inserted into the guide that
cuts the magnetic lines (H lines) and therefore acts as a transformer, as
shown in Figure 4.19 on Page 4-19.
In probe coupling, a small antenna is inserted into the guide parallel with
the electric field (E lines), as shown in Figure 4.20 on Page 4-19.
Figure 4.17. The Transverse Electric (TE) Mode
Figure 4.18. The Transverse Magnetic (TM) Mode
E lines
Magnetic lines
M lines
E lines
2
2
RF Theory: RF Waveguide Design 4-19
7.4. Determin-
ing the TE
10

Dominant
Mode of a
Waveguide
Figure 4.19. Loop Coupling
Figure 4.20. Probe Coupling
Figure 4.21. Frequency Determining Parameters
a
a = 2.875"
4-20 RF Theory: Transmission Lines
The TE
10
dominant frequency is 2.05 GHz.
8. Transmis-
sion Lines
Transmission lines can be classified as resonant or non-resonant. Reso-
nant lines are used primarily for impedance matching, phase shifters, in-
verters, wave filters and chokes. Non-resonant lines are lines that are ei-
ther infinitely long or terminated in its characteristic impedance. The volt-
age and current waves move in the same phase with each other and all of
the energy is absorbed by the load.
Where the term represents the number of sections; as this number ap-
proaches infinity then this term will approach zero.
The PFN in the Clinac is a transmission line, so let us calculate its imped-
ance.
2
--- 2.875 inches =
2 2.875 inches =
5.750 inches =
(meters) 0.0254 10
2
5.750 =
14.6 10
2
=
f (MHz)
300
(meters)
------------------------- =
f
3 10
2
1.46 10
1
--------------------------- =
f 2.05 10
3
=
f 2.05 GHz =
Figure 4.22. Characteristic (Surge) Impedance
Figure 4.23. PFN Parameters
Z
0
characteristic impedance =
Z
1
total inductive reactance of line 2fL = =
Z
2
total capacitive reactance of line
1
2fC
------------- = =
Z
0
Z
1
Z
2
Z
1
2
-----

2
+ =
Z
1
Z
2
-----

2
C = 18 F
total

L = 36.5 H
total

RF Theory: Vector Analysis 3dB Quadrature Hybrid 4-21
As explained in Chapter 3 of this manual, the load resistor(s) on the despik-
ing network can be used as a temporary load while troubleshooting the
modulator. The load resistance, 12.5 ohms, would match the impedance of
the generator (PFN line).
8.1. VSWR VSWR is the ratio of the effective voltage at a loop to the effective voltage at
a node. It is also equal to the ratio of the characteristic impedance of the
line to the impedance of the load, or vice versa.
To measure VSWR, a probe can be inserted in a slot cut in the waveguide
that acts as an antenna and is excited by the E lines that flow parallel to it.
Since the line current flows parallel to the slot, the effective resistance of
the waveguide is not appreciably reduced by the presence of the slot.
9. Vector Anal-
ysis 3dB
Quadrature
Hybrid
4.24 shows a vector analysis of the input and output signals of the 3dB
quadrature hybrid.
Figure 4.24. 3dB Quadrature Hybrid Vector Analysis
V2
Vi = incident signal
Vr = reflected signal
V1
Port 1 Output
Port 2 Output Phase Shift V1 V2
V2
Vr (90)
Vr (180)
Vr (180)
Vr (270)
Vr (270)
Vr (90)
Vi (90)
Vi (90)
Vi (90)
Vi (90)
2
Vi
2
V1
2
Vr
2
V2
0V
2V
+2V
0V
0
90
270
180
Vi
2
2 2
2
Vi
2
Vr
2
V1
Vi
2
2
2
2 2
2
4-22 RF Theory: Circulators
10. Circulators When the RF Power first reaches the accelerator structure all of it is ini-
tially reflected back toward the source. Once its amplitude has stabilized, if
its frequency is equal to the resonant frequency of the accelerator it begins
to flow into and resonate within the accelerator. At the end of the pulse, as
the amplitude begins to decrease, the power is again totally reflected. In or-
der to separate the forward and reflected power, a circulator is employed.
There are two types of circulators used in Varian Clinacs, 3-port and 4-
port. All Clinacs being manufactured as of this writing use 4-port circula-
tors. Many older low-energy Clinacs use 3-port circulators, which are sim-
pler in design. Figures 5.25 and 5.26 show the Low Energy Clinac 3-port
circulator and 4-port circulator respectively.
Figure 4.25. 3-Port Circulator
Figure 4.26. Low Energy Clinac 4-Port Circulator and Phase Diagram
(Configured as 3-Port)
From Magnetron To Accelerator
To Water Load
Cutaway View Top View
Ferrite Beads
From Accelerator
From Water Load
To Magnetron
Magnet
RF Power Path
Blocked off
To Accelerator
Magnets
To Water Load
NOTE: Circles indicate where phase rotation occurs
Ferrite Phase Shifter 3dB Magic Tee 3dB Quadrature Hybrid
1 1
1
2 2
2 3 3 4 4
1
3
2
4 1
3
2
4 1
3
2
4
Drawn: Bill Kirkness, 05/96
Redrawn: Bill Kirkness, 10/03
From Magnetron
180 Phase Rotation
90 Phase Rotation 90 Phase Rotation
3 4
RF Theory: Circulators 4-23
Figure 4.27 shows the High Energy Clinac 4-port circulator used with the
shunt tee power attenuator. In 1996, Varian eliminated the shunt tee and
began operating the klystron in its linear mode, thus removing the need for
the fourth port. On these machines, port 3 of the Magic Tee is blocked, and
the RF power from the Klystron is applied to port 4 of the 3dB Quadrature
Hybrid, as shown in Figure 4.28.
Figure 4.27. High Energy Clinac 4-Port Circulator and Phase Diagram
(Used in Shunt Tee Clinacs)
Figure 4.28. High Energy Clinac 4-Port Circulator and Phase Diagram
(Configured as 3-Port for KLM Clinacs)
From Klystron
To Accelerator
Magnets
To Water Load #2
1 1
1
2 2
2 3 3 4 4
1
3
2
4 1
3
2
4 1
3
2
4
To Shunt Tee
180 Phase Rotation
3 4
Blocked off
To Accelerator
Magnets
To Water Load
1 1
1
2 2
2 3 3 4 4
1
3
2
4 1
3
2
4 1
3
2
4
From Klystron
180 Phase Rotation
3 4
4-24 RF Theory: Klystron Theory
11. Klystron
Theory
This discussion is primarily for those engineers and technicians with little
or no knowledge of microwaves who may find themselves suddenly con-
fronted with operating and maintaining a high power microwave transmit-
ter using klystron amplifier tubes. It is assumed that the reader is familiar
with the theory and operation of triode RF amplifiers and general electronic
circuit theory. We will discuss a simplified theory of operation for klystron
amplifier tubes, the associated equipment required to make a klystron am-
plifier tube operate properly, some of the things that must be protected
from malfunctions, and some generalized operating procedures associated
with this type of equipment.
After reading this information, the reader probably could not design a
klystron amplifier; however, he should be able to understand better what is
going on in the equipment and the reasons behind some of the operating
instructions presented in a typical klystron amplifier Instruction Manual.
Knowing and understanding this simple theory should help the operator to
realize why certain functions are built into a klystron amplifier, and the
why of certain operating instructions that he may receive. We feel that
this understanding is important to a man who must operate or service a
fairly complicated and expensive piece of electronic equipment.
A little history: The klystron amplifier was in its infancy even at the end of
World War II, although reflex klystrons had been developed (for use as local
oscillators in radar receivers) to a fairly high degree of sophistication by the
end of the war. Since World War II the klystron amplifier has undergone a
spectacular evolution. It has become one of the most widely used devices
for the amplification of microwave signals, particularly for high power ap-
plications.
Klystron amplifiers currently in production cover microwave frequency
ranges from UHF to 100 GHz, or higher; outputs range from a few milli-
watts to many megawatts peak and more than 100 kilowatts average; pow-
er gains vary from 3 to 90 dB; and sizes vary from extremely small tubes
that can be held easily in the palm of the hand to tubes that are more than
12 feet long.
The uses of klystron amplifiers cover almost every microwave application,
from low level signal generators, to giant radar equipment and huge trans-
mitters for deep space communication and command. Some of the equip-
ment is complicated and quite expensive. A serious shortage of engineers
and technicians trained to operate and maintain this equipment has re-
sulted from the rapidity with which this equipment has been developed and
produced.
Many engineers and technicians, experienced on lower frequency equip-
ment using conventional vacuum tubes, have required retraining to oper-
ate this microwave equipment. We hope this information will help slightly
in the training process.
11.1. Theory of
Klystron Oper-
ation
The basic theory of klystron amplification is quite simple. In fact, the
klystron amplification principle can be readily explained by an analogy with
a simple triode RF amplifier. Obviously there are some differences (which
will be explained), and these differences are what make a klystron amplify
at microwave frequencies whereas a triode will not. First, let us consider
the basic theory of operation of a simple triode amplifier.
Figure 4.29 shows a simplified diagram of a triode amplifier with resonant
circuits at both the input and the output. Such resonant circuits restrict
the bandwidth of the amplifier and increase the gain. Such an amplifier
might be part of an intermediate frequency amplifier circuit typically used
at frequencies from 10 to 100 megacycles.
RF Theory: Klystron Theory 4-25
A triode radio tube consists of three elements: a cathode that emits a
stream of electrons, a grid that stands in the path of the stream, and a plate
that attracts the electrons and catches them after they pass through the
grid. The grid acts as a valve, opening or closing the passage of electrons
according to the voltage on it. The RF input signal comes to the grid as a
weak alternating current, oscillating at the RF frequency. The oscillating
voltage thus applied to the grid modulates the flow of electrons across the
tube at the RF frequency. The electron stream then delivers, at the plate, an
alternating current that reproduces the weak signal on the grid with ampli-
fication. This alternating current at the plate flows through the resonant
plate circuit and excites alternating voltages across it; these voltages con-
stitute the RF output from the amplifier.
Now the time it takes an electron to cross the tube is in the order of a bil-
lionth of a second. This transit time is short compared with the cycle of a
long radio wave (around a millionth of a second); therefore the electron is
slowed or speeded by the voltage on the grid at a given moment of the RF
cycle. The flow of electrons, therefore, can follow the voltage fluctuations
on the grid. In microwaves, however, the oscillations are so rapid (i.e., the
cycle is so short) that the voltage on the grid may go through several com-
plete oscillations while an electron travels across the tube. In other words,
the grid voltage changes too fast and produces only chaos among the elec-
trons. The grid voltage can no longer impose its signal pattern on the elec-
tron flow.
There are other reasons why the conventional triode tube fails in the micro-
wave range, but this is the most fundamental one the simple fact that the
transit time of an electron from cathode to plate is long compared with the
time of one cycle of the microwave signal.
The klystron tube makes a virtue of the very thing that defeats the triode
the transit time of the electrons. What it does is to modulate the velocity
of electrons so that, as they travel through the tube, they sort themselves
into groups and arrive at their destination in bunches. These bunches de-
liver an oscillating current to the output resonant circuit of the klystron.
Figure 4.30 on Page 4-26 shows a cutaway representation of a typical
klystron amplifier. Schematically it is very similar to a triode amplifier in
that it includes an electron gun, resonant circuits, and a collector (which is
roughly equivalent to the plate of a triode). In fact, the klystron amplifier
consists of three separate sections the electron gun, the RF section and the
collector section.
Figure 4.29. Triode Vacuum Tube Amplifier
RF Input
RF Output
Resonant grid circuit
Resonant
Plate
Circuit
Let us consider, first, the electron gun structure: As in the triode, the elec-
tron gun consists of heater and cathode, a control grid (sometimes), and an
anode. Electrons are emitted by the hot cathode surface and are drawn to-
ward the anode that is operated at a positive potential with respect to the
cathode. The electrons are formed into a small, dense beam by either elec-
trostatic or magnetic focusing techniques, similar to the techniques used
for beam formation in a cathode ray tube. In some klystron amplifiers a
control grid is used to permit adjustment of the number of electrons that
reach the anode region; this control grid may be used to turn the tube com-
pletely on or completely off in certain pulsed-amplifier applications.
The electron beam is well formed by the time it reaches the anode. It passes
through a hole in the anode, passes on to the RF section of the tube, and
eventually the electrons are intercepted by the collector. The electrons are
returned to the cathode through an external power supply (not shown on
Figure 4.30). It is evident that the collector in the klystron acts much like
the plate of a triode as far as collecting of the electrons is concerned. How-
ever, there is one important difference; the plate of a triode is normally con-
nected in some fashion to the output RF circuit, whereas, in a klystron am-
plifier, the collector has no connection to the RF circuitry at all.
From the above discussion it is apparent that the klystron amplifier, as far
as the electron flow is concerned, is quite analogous to a stretched-out
triode tube in that electrons are emitted by the cathode, controlled in num-
ber by the control grid, and collected eventually by the collector.
Now let us consider the RF section of a klystron amplifier. This part of the
tube is physically quite different from a triode amplifier. One of the major
differences is in the physical configuration of the resonant circuit used in a
klystron amplifier. The resonant circuit used with a triode oscillator, at low-
er frequencies, is generally composed of an inductance and a capacitor,
while the resonant circuit used in a microwave tube is almost invariably a
metal-enclosed chamber, known as a cavity resonator.
Figure 4.30. Sectional View of a Klystron
Heater Cathode Anode First Cavity
(Buncher)
Drift
Tube
Intermediate
Cavity
Last Cavity
(Catcher)
Collector
Electrons
Control
Grid
Buncher
Gap
RF
Input
Bunch of
Electrons
RF
Output
Catcher
Gap
Electron Gun RF Section Collector
A very crude analogy can be made between the resonant cavity and a con-
ventional L-C resonant circuit. The gap in the cavity (see Figure 4.30) is
roughly analogous to the capacitor in a conventional low frequency reso-
nant circuit in that alternating voltages, at the RF frequencies, can be made
to appear across the cavity gap. Circulating currents will flow between the
two sides of the gap through the metal walls of the cavity, roughly analo-
gous to the flow of RF current in the inductance of an L-C resonant circuit.
Since RF voltages appear across the sides of the cavity gap it is apparent
that an electric field will be present, oscillating at the RF frequency, be-
tween the two surfaces of the cavity gap.
When a cavity is the correct size, it will resonate to microwaves of a certain
frequency. The cavity can be tuned to various microwave frequencies by ad-
justing its size by some mechanical means. A crude analogy to the cavity
resonator would be a glass goblet that resonates at a certain pitch depend-
ing on the level of the water in it, i.e., the size of the air cavity in the goblet.
As shown in Figure 4.30, electrons pass through the cavity gaps in each
resonator, and pass through cylindrical metal tubes between the various
gaps. These metal tubes are called drift tubes. In a klystron amplifier the
low-level RF input signal is coupled to the first resonator, which is called
the buncher cavity. The signal may be coupled in through either a
waveguide or a coaxial connection. The RF input signal will excite oscillat-
ing currents in the cavity walls, if the cavity is the correct size (that is,
tuned to the right frequency). These oscillating currents will cause the al-
ternate sides of the buncher gap to become first positive, and then negative,
in potential at a frequency equal to the frequency of the RF input signal.
Therefore, an electric field will appear across the buncher gap, alternating
at the RF frequency. This electric field will, for half a cycle, be in a direction
that will tend to speed up the electrons flowing through the gap; on the oth-
er half of the cycle the electric field will be in a direction that will tend to
slow the electrons as they cross the buncher gap. This effect is called ve-
locity modulation, and it is the mechanism that permits the klystron am-
plifier to operate at frequencies higher than the triode.
After leaving the buncher gap, the electrons continue toward the collector
in the drift tube region. Ignore for the moment the intermediate resonator
shown in Figure 4.30, and let us consider the simple case of a two-cavity
klystron amplifier. In the drift tube region the electrons that have been
speeded up by the electric field in the buncher gap will tend to overtake
those electrons that have previously been slowed (by the preceding half of
the RF wave across the buncher gap). It is apparent that, since some elec-
trons are tending to overtake other electrons, clumps or bunches of elec-
trons will be formed in the drift tube region.
If the average velocity of the electron stream is correct, as determined by
the original voltage between anode and cathode, and if the length of the
drift tube is proper, these bunches of electrons will be quite completely
formed by the time they reach the catcher gap of the last cavity (which is
called the catcher). This results in bunches of electrons flowing through
the catcher gap periodically, and during the time between these bunches
relatively fewer electrons flow through the catcher gap. The time between
arrival of bunches of electrons is equal to the time of one cycle of the RF in-
put signal.
These bunches of electrons will induce alternating current flow in the metal
walls of the catcher cavity as they pass through the catcher gap. If the
catcher cavity is of correct size (tuned to the proper frequency) large oscil-
lating currents will be generated in its walls. These currents cause electric
fields to exist, at the RF frequency, within the catcher cavity. These electric
fields can be coupled from the cavity (to the output waveguide or coaxial
transmission lines) resulting in the RF output from the tube.
It is not particularly obvious why a bunch of electrons, passing through the
catcher gap, should generate an oscillating RF current in the walls of the
catcher cavity. Fortunately, a qualitative explanation is easy to under-
stand. Refer to Figure 4.31 which shows the catcher cavity at three instants
of time as a bunch of electrons flow across the catcher gap. The electrons
are shown passing the catcher gap, traveling from left to right. To simplify
the explanation, we have shown grid wires across the gaps; the grid on the
left side of the gap is labeled No. 1, while the grid on the right is labeled No.
2. Since the grid wires, and the cavity walls, are made of high-conductivity
metal, such as copper, a large number of free electrons will be present in
the metal.
In Figure 4.31, as the bunch of electrons approaches Grid No. 1 the free
electrons in Grid No. 1 will be repelled since negative charges repel each
other. This will tend to cause these electrons to flow from the grid wires into
the cavity walls and around the cavity walls toward Grid No. 2. This is
shown by the current flow path in Figure 4.31A. The result is that Grid No.
2 will tend to accumulate a surplus of negative charges, whereas Grid No.
1 will have a scarcity of negative charges present. Figure 4.31B shows the
instant when the bunch of electrons is between Grids 1 and 2. At this in-
stant the electrons in the bunch are repelling free electrons in both grids
equally, and the net current flow around the cavity walls is essentially zero.
Figure 4.31C shows the instant just after the electron bunch has passed to
the right of Grid No. 2. Remember that Grid No. 2 has accumulated an ex-
cess of free electrons already and these free electrons would tend to redis-
tribute themselves back toward Grid No. 1 even if the electron bunch was
not present. However, the electron bunch further repels the excess free
electrons in Grid No. 2 and tends to push these free electrons back to-
ward Grid No. 1. As the electron bunch moves further to the right, the elec-
trons will redistribute themselves to essentially an equilibrium condition
during the time between bunches.
The process of course repeats every cycle of the RF wave because a bunch
of electrons comes past the catcher gap in a time equal to the interval of one
cycle of the RF wave. Since the resonant cavity is a high-Q circuit the oscil-
lating currents tend to be essentially sinusoidal although the bunches of
electrons arrive in short bursts. The situation is quite analogous to striking
a pendulum one blow for each cycle of its oscillation; this will cause the
pendulum oscillation to build up although the driving force is not continu-
ously applied. Another analogy is a Class C triode amplifier where bursts of
Figure 4.31. Generation of Alternating Current in a Cavity
Direction of
Electron Flow
Grid 1 Grid 1 Grid 1 Grid 2 Grid 2 Grid 2
Direction of
Electron Flow
A B C
Electron
Bunch
current generate essentially sinusoidal voltages in the plate resonant cir-
cuit.
RF power can be taken from the output (catcher) cavity by coupling to the
oscillating current flowing in the cavity walls (or to the electric fields inside
the cavity which are generated by these oscillating currents). If the amplifi-
er is functioning properly, the oscillating current in the catcher cavity will
be considerably larger than the oscillating currents in the buncher cavity;
consequently, amplification has taken place. When the bunches of elec-
trons pass through the output gap in the catcher cavity, they deliver energy
to this cavity which causes currents to flow in the cavity walls. Since the
electron beam is delivering energy to the cavity, it is slowed in velocity;
therefore the beam arrives at the collector with less total energy than it had
when it passed through the input cavity. This difference in electron beam
energy is approximately equal to the RF energy delivered from the output of
the cavity.
It is appropriate to mention here that the velocity modulation effect does
not form perfect bunches of electrons. There are some electrons which
come through out-of-phase. These electrons show up at the last gap be-
tween the bunches. The electric field, at the time these out-of-phase elec-
trons come through, is in a direction to accelerate them; so some few elec-
trons will actually have their velocity increased as they come through the
output gap. The electrons reaching the collector therefore have a wide
spread of energy. Some of them (the out-of-phase electrons) may have ve-
locities almost twice as high as the average electron velocity; other elec-
trons (the in-phase, useful electrons) will be materially slowed and will ar-
rive at the collector with a velocity much less than they started with.
In the previous discussion we have considered only a two-cavity klystron
amplifier, having neglected the intermediate cavity shown on Figure 4.30.
Klystron amplifiers have been built (to our knowledge) with as many as sev-
en cavities, i.e., with five intermediate cavities. The effect of the Intermedi-
ate cavities is to improve the bunching process; the result is to increase
amplifier gain, and to a lesser extent, the amplifier efficiency. Adding more
intermediate cavities is roughly analogous to adding more stages to an I-f
amplifier, i.e., the gain of the overall amplifier is increased, and the overall
bandwidth is reduced, if all stages are tuned to the same frequency. The
same effect occurs with the klystron amplifier. However, it is well known
that the bandwidth can be increased, and the gain reduced, by stagger-
tuning an I-f amplifier. This analogy carries over to the klystron amplifier.
A given klystron amplifier tube will deliver high gain and narrow bandwidth
if all the cavities are tuned to the same frequency; this is called synchro-
nous-tuning. If the cavities are tuned to different frequencies the gain of
the klystron amplifier will be reduced and the bandwidth may be apprecia-
bly increased; this is called stagger-tuning. Most klystrons which feature
relatively wide bandwidth are stagger-tuned. The appropriate method of ac-
complishing stagger tuning is discussed in more detail later.
Figure 4.32. High-Power Four-Cavity Klystron
Figure 4.33. High-Power Four-Cavity Klystron, Simplified
Collector
Water
Circuit
Collector
Pole Piece
Tuning
Diaphragm
Water
Circuit
Magnetic
Circuit
Drift
Tube
Focus
Coils
Anode
Pole Piece
Output
Window
Electron
Bunch
Output
Iris
Output
Cavity
(Catcher)
Third
Cavity
Second
Cavity
Input
Cavity
(Buncher)
Input
Loop
Anode
Heater
The klystron is not a perfect linear amplifier; that is, the RF power output
is not linearly related to the RF power input at all operating levels. Another
way of stating this is that the klystron amplifier will saturate, just as a tri-
ode amplifier will limit if the input signal becomes too large. In fact, if the
RF input is increased to levels above saturation, the RF power output will
actually decrease. Figure 4.34 shows the plot of typical klystron amplifier
performance for various tuning conditions. The RF output is plotted as a
function of the RF input.
Curve A of Figure 4.34 shows typical performance for synchronous tuning.
Under these conditions the tube has maximum gain. The power output is
almost perfectly linear, with respect to the power input, up to about 70 per
cent of saturation. However, as the RF input is increased beyond that
point, the gain decreases and the tube saturates. As the RF input is in-
creased beyond saturation, the RF output decreases. The reason for this
decrease in output is quite interesting. Remember, in our previous discus-
sion, that the electron bunches were formed by the action of the RF voltage
across the buncher cavity gap. This RF voltage speeded up some electrons
and slowed other electrons, resulting in formation of bunches in the drift
tube region. Obviously this speeding up and slowing effect will be increased
as the RF drive power is increased. The saturation point on Figure 4.34 is
reached when the bunches are most perfectly formed at the instant they
reach the output (catcher) gap. This results in the maximum power output
condition. When the RF input is increased beyond this point, the bunches
are most perfectly formed before they reach the output gap, i.e., they form
too soon in the drift tube region. By the time the bunches have reached the
output gap they tend to debunch because of the mutual repulsion of the
electrons, and because the faster electrons have overtaken and passed the
slower electrons. This causes the power output to decrease.
Figure 4.34. Effect of Tuning on Klystron Performance
RF Input
R
F

O
u
t
p
u
t
A
B
C
D
Saturation Point
Envelope of
Saturation Peaks
Curves B, C, and D illustrate a phenomenon of klystron amplifiers which is
difficult to explain theoretically, but which should be recognized by person-
nel operating these amplifiers. It turns out that, if we start with a multi-
cavity klystron which is synchronously tuned, and then tune the next-to-
the-last cavity to a higher frequency, we find that the gain of the amplifier
is reduced but that the saturation power output level may be increased.
This effect is shown by curves B and C. Curve B represents a small amount
of detuning of the next-to-the-last cavity, and curve C represents even more
detuning of that cavity. Note that the gain of the tube has been reduced (it
takes more RF input to obtain a given RF output), and that the saturation
output power is higher than obtained with synchronous-tuning (curve A).
As stated previously, this stagger-tuning also results in a wider bandwidth
for the amplifier. Many klystron amplifiers are operated in this fashion be-
cause it enables one to obtain more power output, with the same beam
power input, and therefore increases the efficiency of the tube; of course
this can only be done if enough RF drive power is available to operate under
the stagger-tuned condition. As one might intuitively expect, we can go too
far with this stagger-tuning, and the saturation output will eventually drop.
This is illustrated by curve D of Figure 4.34.
Figure 4.30 does not show one very important item which is usually re-
quired for high power klystron amplifier operation. This is an axial magnet-
ic field, i.e., one which is parallel to the center line of the klystron. In
klystron amplifiers which are physically long it is quite difficult to keep
the electron beam formed properly during its travel through the RF section.
Since electrons are negatively charged particles, they tend to repel each
other; this causes the beam to spread in a direction perpendicular to the
axis of the tube. If this occurs, the electrons will strike the drift tubes and
be collected there, rather than passing through the drift tubes to the collec-
tor. To overcome this beam spreading an axial magnetic field is used. The
action in the magnetic field is to exert a force on the electrons to keep them
going in the correct direction during their transit through the RF section.
The magnetic field may be developed by a permanent magnet or by one, or
more, electromagnet coils. A permanent magnet is generally used on tubes
which are physically small or of medium power rating. Unfortunately, the
size and weight of a permanent magnet become excessive for long or high
power tubes, making it necessary to use electromagnets. In some large
tubes several, separate, magnet coils are used; the current in each coil is
individually adjustable to optimize the magnetic field shape. The magnetic
field is normally terminated as quickly as possible after the catcher cavity
so that the beam can spread before it hits the collector. This tends to
spread the electron beam interception over a large surface on the collector;
this minimizes collector-cooling problems which would result if the beam
remained concentrated at the time of interception.
Even with an axial magnetic field some electrons will go astray and not re-
main in the main electron beam. These electrons will be intercepted by the
anode or the klystron drift tubes. In high power tubes is it particularly im-
portant to minimize the number of these stray electrons because they gen-
erate heat when they strike the drift tubes. In high-power klystrons this
heating can be a very severe problem because drift tubes are difficult to
cool. Temperatures can become high enough to melt the metal in the drift
tubes and destroy the tube.
The collector is normally insulated from the RF section of large klystron
amplifiers to permit separate metering of the electrons intercepted by the
drift tubes, and those intercepted by the collector. The e1ectrons intercept-
ed by the RF section are normally called body current, while those elec-
trons intercepted by the collector are normally called collector current.
Obviously, the sum of the body current and the collector current is equal to
the total current in the electron beam which is normally called beam cur-
rent. Klystron amplifier specifications will quite often place a maximum
limit on allowable body current.
The previous discussion (describing the general theory of klystron opera-
tion) implied that klystron amplifiers normally have actual metal grid
structures across the gaps in the resonant cavities. Many low power
klystrons do indeed have wire-mesh grids. However, most high-power
klystrons do not have actual grids across the gaps, because such grids
would intercept sizable amounts of the electron beam. It is very difficult to
cool grid structures, and large beam interception would cause the grids to
melt, destroying the tube. Fortunately, by proper design, the klystron can
be made to work efficiently without actual grid wires across the gaps. The
absence of these grids does not change the operating principles discussed
previously, but it does have a secondary effect on the klystron performance.
It turns out that, if the electron beam has a very small diameter compared
with the size of the drift tubes, the beam does not couple strongly to the
gaps and therefore it does not react as strongly with the klystron cavity.
Therefore, the performance of a klystron amplifier, which does not have
gridded gaps, can sometimes be improved by permitting the electron beam
to be as large as possible (while keeping the body current down to the max-
imum specified for the tube).
The size of the beam can be somewhat controlled by the magnetic field
strength. We therefore find that the klystron performance can sometimes
be improved by adjusting the magnetic field in a way which does not result
in the minimum possible body current condition, i.e., by adjusting the field
so that the beam shape is somewhat larger than the minimum obtainable.
In gridless-gap klystrons therefore, best operation may be obtained with a
body current which is not the minimum obtainable; however, body current
must be kept within the maximum specified for the tube.
Body current usually increases with RF input level which might be expect-
ed since RF causes electron bunches to form. The dense electron concen-
tration in the bunch causes the electrons to repel each other, and the di-
ameter of the bunch may become larger than the diameter of the beam
without the bunches. Consequently, some of the electrons in the bunch
may be lost to the drift tubes, and the body current may increase.
11.2. Associ-
ated Equip-
ment
In the preceding sections, we have discussed the basic theory of operation
of the klystron amplifier tube. Considerable additional equipment is re-
quired for a complete amplifier system. First, we will need power supplies to
deliver the voltages and currents required for the klystron and for the elec-
tromagnets. In high-power systems we will need various types of cooling to
get rid of the power supply energy which is not converted into RF output
power. We will need various RF circuit components to control and measure
the RF input to the klystron tube, and to measure the RF output from the
tube. For testing we may need a dummy load to dissipate RF output when
it may be inconvenient, or impossible, to radiate. We will need a large col-
lection of meters and protective devices to monitor performance and to pro-
tect operating personnel (and the equipment itself) in case of equipment
malfunction or operator error. This associated equipment will be discussed
in this section.
11.3. Power
Supplies
Figure 4.35 is a simplified diagram showing the power supplies used in a
typical klystron amplifier. In most klystron tubes the anode and RF section
of the tube are connected inside the vacuum envelope. These parts are nor-
mally called the tube body, and they are generally operated at ground po-
tential as shown in Figure 4.35. Operating the tube body at ground is con-
venient because the input and output connections (either waveguide or co-
axial) are then at ground potential; this makes it easy to connect into the
rest of the system. Also, this keeps the cavity tuners at ground potential,
eliminating any danger to personnel who are tuning the tube.
The beam power supply, shown in Figure 4.35, generates the voltage re-
quired to accelerate the electrons and form the electron beam. It must also
deliver the beam current required for the klystron tube itself. As shown, the
positive end of the beam supply operates at (nearly) ground potential,
whereas the negative output from the supply is the high potential point in
the system.
The design details for beam power supplies vary widely depending upon the
application of the power amplifier. However, in general, they employ fairly
conventional circuits. They usually include means of adjusting the ac volt-
age to the primary of the power transformer, either an auto-transformer
(such as a Variac), an Inductrol, or perhaps an ac generator whose output
is varied by adjusting the dc field control. Beam supplies incorporate a
step-up transformer, a rectifier circuit, and an LC filter. Either solid-state,
hardtube, or gaseous diodes are used in the rectifier circuit. Tube rectifiers
normally have a lower initial cost; however, they require periodic replace-
ment. Solid-state rectifiers, particularly for high voltage and high current,
are usually more expensive initially; however, their reliability is excellent
after the initial design and debugging.
Filter design is quite conventional; the amount of filtering depends upon
the allowable ripple for the system. In some special cases, extremely low
ripple and extremely good beam voltage regulation is required. For low- and
medium-power systems this can often be achieved by electronic regulation
of the dc output voltage. The circuits are, technically, fairly conventional,
but they may become quite complicated and expensive for the medium-
power systems.
Figure 4.35. Klystron Power Supply Connections
HEATER
SUPPLY
BEAM
SUPPLY
GRID SUPPLY
OR PULSER
ELECTROMAGNET
SUPPLIES
Current
Limiting
Resistors
Beam
Current
Meter
Beam
Current
Overload
Body
Current
Overload
Body
Current
Meter
Collector
Current
Meter
Cathode
RF Input RF Output
Collector
Direction of Electron Flow
Crowbar
For extremely high-power systems electronic regulation has not proven
practical to date. For these systems it is normal to obtain the primary pow-
er from a motor generator set. The inertia of the large MG set effectively
smooths out variations in the incoming line voltage; and the ac output
from the generator can be quite easily regulated by a feedback system to
the generator field. Brute-force filtering is used to achieve the allowable rip-
ple.
Beam voltage and beam current metering are always provided, as well as
beam current overload protection. Some systems have beam over-voltage
protection to protect the tube against the possibility of power-supply run-
away (rare), and against operator error (more common). Since the beam
supply is the source of most of the energy in the system, it is usually turned
off when any malfunction occurs in the system; this will be discussed in
more detail later.
A variable voltage beam supply is usually provided so the tube can be op-
erated at whatever power level is desired (within maximum ratings). It is
also desirable to have low beam voltage capability; this is useful when a
new tube is installed and initial adjustments are being made. Some sys-
tems have a feature that automatically starts the beam voltage at a low val-
ue when the supply is first turned on; the voltage then slowly increases un-
til it reaches a preset level; and it may regulate to that level for changes in
ac line voltage. The voltage automatically runs down to a low level when the
supply is turned off.
For high power systems it is normal to have some series resistance between
the beam supply and the klystron cathode; this limits the tube current to
some finite value in case the tube should arc from cathode to ground. With-
out some limiting resistance the peak current during an arc could be very
high and might destroy the cathode surface; with current limiting resis-
tance a tube can often be cleaned up even if it is somewhat gassy or if it
arcs on initial turn-on. Most klystron amplifiers include a getter, and
some include a VacIon vacuum pump. A getter will absorb a limited
amount of gas that may accumulate after long storage periods, and there-
fore may permit a slightly-gassy tube to clean up and be perfectly service-
able (if the tube is not damaged during initial start-up). The VacIon pump
operates continuously and will absorb a tremendous amount of gas and
may allow a tube to continue in service for its normal life even if it has a
small vacuum leak.
Some high power amplifiers use a crowbar system to discharge the beam
supply very quickly in case of an internal klystron arc, or other high-voltage
fault condition. Most crowbar systems consist of a triggered spark gap con-
nected across the power supply. Circuits are provided which will trigger the
gap in case of excessive body current, arcs in the output waveguide (to be
discussed later), and (sometimes) loss of magnetic field. When the trigger
gap is fired, the main spark gap breaks down. This discharges the energy
stored in the beam power supply very quickly and prevents damage to the
klystron or associated equipment. Crowbars are normally used only on very
high voltage, or very high-power systems where the amount of stored ener-
gy could cause serious damage. They are normally not used in equipment
operating at less than 20-kilowatts average-power; those systems can be
adequately protected by simple current-limiting resistors (which are much
less complicated and much less expensive than a crowbar system). A crow-
bar system will normally operate in a few microseconds and therefore will
limit the amount of destructive energy delivered to the tube to a very low
value. Conversely, it may take several seconds to turn off and discharge a
large beam power supply because of the long time required for overload re-
lays and ac contactors and filter capacitors to discharge.
The heater supply furnishes power for the klystron heater, which heats the
cathode that emits the electrons for the electron beam. Most klystrons have
an indirectly-heated cathode, i.e., the cathode is heated simply by being
close to the heater windings. A few klystrons have bombarded cathodes.
In tubes of this type a voltage is applied between the heater and the cath-
ode (with the cathode positive). The heater and cathode then function as a
conventional diode. The heater becomes hot enough to emit electrons.
These electrons are drawn to the cathode by the bombarder voltage.
When they strike the cathode they liberate their energy (to the cathode) as
heat, just as the plate of a diode is heated by the electrons striking it. Bom-
barded cathodes have been largely displaced by recently developed im-
pregnated cathodes, although a few bombarded-cathode tubes are still in
production.
The heater supply is normally a rather simple unit. It may deliver either al-
ternating or direct current. For many applications, an ac supply is ade-
quate; it consists simply of a variable auto-transformer, a step-down trans-
former, and appropriate voltage and current metering. In a few systems
that require extremely low-noise performance, dc supplies are necessary.
The heater supply must be insulated to withstand the full (negative) beam
voltage potential. Meters are normally used to show the heater voltage or
current or both. Since these meters must operate at a high negative poten-
tial, they may sometimes give incorrect readings due to the large electro-
static fields present; special care must be taken in the design of metering
circuits to prevent these false indications. Heater voltages are normally ad-
justable to take care of individual tube-to-tube variations and compensate
for variation of incoming ac line voltage. A normal klystron heater presents
very nearly a short-circuit to the power supply when the heater is cold (first
turned on). Therefore, it is normal to use some type of current limiting in
the heater power supply. Many klystron specifications require that this ini-
tial surge current be limited to 150 per cent of normal operating current.
Protective circuits are often used to turn off the beam power supply if the
heater supply fails, since some tubes will be damaged if the beam voltage is
on while the heater and cathode are cooling after a heater supply failure.
Some high power klystrons require that the cathode assembly be cooled by
an air blower. An airflow protective interlock is normally included to turn
off the heater and beam voltage supplies if the klystron blower ceases to op-
erate. The entire electron gun section of some very-high-voltage tubes is
immersed in oil, for insulation and cooling.
Some klystron amplifiers have a grid (or modulating-anode, which per-
forms the same function) to control the number of electrons in the electron
beam. Such grids are often used in pulsed systems to turn the tube either
full-on or full-off; a few systems employ grid modulation for transmission of
intelligence. In most gridded klystron tubes the grid is never allowed to go
positive with respect to the cathode, as this might cause undue grid inter-
ception and result in burnout of the grid element. A grid power supply is re-
quired in those tubes that have grids. These power supplies and pulsers
may take many forms depending upon the system application and will not
be discussed in detail. It is important to note, however, that the grid power
supply must be insulated for the full beam voltage. Fortunately, most
klystron amplifiers designed for communication service do not use grids.
The collector of most high power klystrons is insulated from the body of the
tube. This allows separate metering and overload protection for the body
current and for the collector current which would be impossible if the col-
lector and the body were connected internally. In most systems the collec-
tor and body operate at very nearly the same potential; any potential differ-
ence is normally only the difference in voltage drop across the various me-
tering circuits.
Figure 4.35 shows three electromagnet coils, apparently wrapped around
the body of the klystron. Some klystrons are indeed made with the electro-
magnet coils physically a part of the tube itself. However, in most systems
the electromagnet coils are separate from the tube, and the klystron is in-
serted into the electromagnet structure. In Varian klystrons the electro-
magnet is designed physically to support and center the klystron tube in
the correct position; no physical adjustment of the electromagnet coils is
provided or required for correct operation. Many modern klystron amplifi-
ers have only one electromagnet coil and therefore require only one power
supply; others may have as many as six separate coils, requiring one power
supply for each coil.
Electromagnet power supplies are usually simple. They are dc supplies us-
ing conventional rectifying techniques. Voltage variation is normally ac-
complished with an auto-transformer on the input. The supplies are usu-
ally well filtered so that the output current contains relatively small ripple
components. Ripple on the electromagnet current may cause the electron
beam in the klystron to wander slightly, at the ripple frequency; this can
cause undesirable amplitude and phase modulation of the RF output sig-
nal. Voltage and current metering is normally supplied for each of the elec-
tromagnet power supplies. If an electromagnet power supply should fail,
the electron beam would almost certainly spread, and the total beam cur-
rent would be intercepted on a small section of the drift tube. In most cas-
es, this would cause the drift tube to melt and permanently destroy the
tube. Therefore, klystron amplifier equipment normally has under-current
protection in each of the electromagnet coil circuits. When the magnet cur-
rent falls below a predetermined level, the beam supply is turned off to pre-
vent damage to the klystron. Redundant protection is provided by the body-
current overload circuits, which also turn off the beam supply in case of
magnet current failure or misadjustment.
Figure 4.35 shows the method normally used to monitor body current, col-
lector current, and beam current separately; the diagram shows the most
frequent arrangement where the klystron body operates at ground poten-
tial. In many systems separate monitoring of collector current is not done
since the collector current and total beam current are normally almost
equal. It is quite unusual, in a relatively high-power klystron amplifier sys-
tem, to allow the body current to exceed 10 per cent of the beam current,
because high body current usually means low efficiency and increases the
danger of burning out drift tubes in the klystron. In very-high-power
klystrons the body current is often limited to 1 or 2 per cent of the total
beam current. Over-current protection is almost always supplied both for
body current and beam current. If a tube arcs internally, the arc will always
occur between cathode and anode. When this occurs the body current im-
mediately becomes excessive, tripping out the body current overload relay.
If an arc occurs, the beam current is also much higher than normal, and
the beam current overload will also trip out. In fact, almost any high-volt-
age system fault (such as an insulation breakdown from high voltage to
ground) will cause excessive current through the body current meter and
overload relay.
Because of the possibility of extremely high currents flowing under fault
conditions, the protection of the body current and beam current meters
(from burnout) presents a somewhat difficult problem. This problem is nor-
mally solved by using very high-current solid-state rectifiers, connected
back-to-back, across the meters. Sometimes adding a small resistance or
inductance in series with the meter is necessary. Surge capacitors are nor-
mally placed across the combination. Connecting the rectifiers back-to-
back is necessary because fault conditions often cause oscillating currents
to flow through the meters.
11.4. Cooling Most low power klystron amplifiers are air cooled, while all high power
k1ystron amplifiers are liquid cooled. At the present state-of-the-art, air
cooling can be used up to RF output levels of about one kilowatt, CW. How-
ever, we find a few special cases where liquid cooling is employed with
tubes having a power output as low as 10 watts; these tubes are used in
special applications that are beyond the scope of this bulletin.
Remember that the main source of power (and therefore heat) in a klystron
amplifier package is the beam power supply. The power generated by the
beam supply must go somewhere; part of it is converted to RF power; the
remainder eventually shows up as heating somewhere in the klystron. The
klystron cooling must be adequate to handle the entire beam power be-
cause, if no RF output is being generated (either due to low RF input power,
or detuning of the klystron tube) then all of the beam power is dissipated in
heat somewhere within the tube. As discussed previously, most of the elec-
trons in the beam eventually end in the collector. When they strike the col-
lector, their energy is dissipated and turned into heat. The small fraction of
the beam lost to the drift tubes also generates heat. Klystron amplifiers are
normally somewhere between 30 and 50 per cent efficient. It is obvious,
therefore, that a tube rated at 10 kilowatts output must be designed to dis-
sipate between 20 and 33 kilowatts depending upon its efficiency. A tube
rated at 100 kilowatts must be able to get rid of about 250 kilowatts as
heat. It is obvious therefore that very advanced cooling techniques are nec-
essary. The power levels involved can melt a hole in the drift tube, or in the
collector, in a small fraction of a second if the cooling system fails and ad-
equate protective devices are not provided.
There are other but smaller sources of heat in a klystron amplifier system.
The heater must be hot to heat the cathode for electron emission. This heat
will be conducted and radiated to the exterior surfaces of the electron gun
assembly, and must be dissipated. Large tubes require a blower on the
electron gun assembly to get rid of this heat. The electromagnet will gener-
ate a considerable amount of heat; the power generated by the focus coil
power supply is all dissipated in the electromagnet. Large electromagnets
are almost always liquid cooled. If the cooling liquid for the electromagnet
fails for any reason, the focus coil power supply must be shut off quite soon
or the magnet will burn out; the beam voltage must also be removed (pref-
erably before turning off focus coil supply) to protect the tube against ex-
cessive body current, as discussed previously.
Earlier in this discussion we described how electron currents oscillate back
and forth in the metal walls of the resonant cavities. Although these cavi-
ties are made with very high-conductivity metal (usually copper), the metal
does present a finite resistance to these oscillating currents; therefore, heat
will be generated in the cavity walls. The amount of heat generated can be
quite sizable in high-power, high-frequency tubes. For instance, consider
the case of a 20-kilowatt CW, X-band klystron amplifier. In this tube, ap-
proximately 1 kilowatt of heat is generated by the circulating RF currents in
the output cavity. Since the cavity is approximately a 1-inch cube, it is ap-
parent that removing this kilowatt of heat is a formidable problem. Cooling
the cavity tuners is particularly difficult. Tuners normally incorporate some
type of metal bellows arrangement, to permit changing the cavity size and
still maintain the vacuum envelope of the tube. Metal bellows are thin
structures, normally more lossy than the remainder of the cavity walls;
therefore, a large amount of heat is often generated in the tuner assembly.
Removal of this heat is a serious problem in high power tubes, and water
cooling is invariably necessary.
Another problem associated with cavity heating is not immediately appar-
ent. Remember that the resonant frequency of the cavity depends upon its
physical size. The cavities are made of metal that expands as it gets hot;
this effect tends to change the resonant frequency of the cavity and to de-
tune the tube. As the tube detunes, the power output will drop; then the RF
heating decreases and the tube will tend to come back in tune. If this
problem was not considered in the initial tube design, it would be quite
possible to design a tube that would never settle down; it would be con-
tinually unstable in its operation. This situation indeed exists in some
tubes which use external cavities. These external cavities are cooled by
air rather than by liquid, and the cavity tuning is seriously affected by the
ambient air temperature. All high-power Varian klystrons are liquid-
cooled, including the cavities and the tuners. The cavities are maintained
at a stable temperature by controlling the temperature of the cooling liquid,
and thermal-detuning is no problem.
Drift tube heating is a serious problem in very high-power klystrons, and in
medium-power, high-frequency, klystrons. The drift tubes that are inside
the vacuum envelope are physically quite small, and it is difficult to remove
the heat by conduction to the region outside the vacuum envelope. In some
high-power tubes, it is actually necessary to bring the cooling liquid inside
the vacuum envelope, and around the drift tubes, to remove the heat from
the drift tubes.
In recent years it has become necessary to cool the waveguide in some
high-power, high-frequency systems. RF currents circulate in a waveguide
that is carrying power, just as in the cavity walls of the klystron. An X-band
waveguide carrying 5 kilowatts CW becomes too hot to touch in normal am-
bient air. Fortunately, the waveguide can be cooled easily by soldering cop-
per tubing along the sides of the waveguide and running cooling liquid
through the tubing.
Most klystron amplifiers have a dummy load to dissipate the RF power dur-
ing adjustment and test (when it may be undesirable to radiate). All high-
power dummy loads are cooled, usually by liquid. In many loads the RF en-
ergy is dissipated in the cooling liquid itself, since water, oil, and ethylene-
glycol (the normal cooling liquids) are quite lossy at microwave frequencies.
In other types of dummy loads the RF energy may be dissipated in a solid,
lossy material. Some of these lossy-material loads can be cooled by air
blasts (for low- and medium-power applications); higher power versions are
liquid cooled.
We have discussed the various sources of heat in a klystron amplifier sys-
tem, to impress the reader with the fact that an expensive klystron can be
destroyed in a matter of seconds if the cooling system fails. A well-designed
system uses many protective devices to prevent this from happening. The
moral is: Check the operation of these protective devices periodically, and
never short-circuit the protective interlocks.
Systems that use blowers for cooling will usually have an airflow switch. If
the blower fails, the switch will open and remove power from the appropri-
ate power supplies. Systems employing liquid cooling normally distribute
the liquid into several paths, since the flow requirements are quite dissim-
ilar. A well-designed amplifier system will have a low-flow interlock in each
of the various paths. If one liquid-cooling circuit becomes plugged, the flow
interlock will open and remove power from the system. Liquid-cooling sys-
tems also include pressure gauges and pressure switches, temperature
gauges and over-temperature switches. Many systems have pressure or
flow regulators. Some systems include devices that will sound an alarm be-
fore trouble actually occurs; sometimes the situation can be corrected
without shutting down the equipment.
In a liquid-cooled system, it is obviously necessary to pump the cooling liq-
uid through the various parts of the tube, and the other equipment that is
generating heat. The liquid becomes hotter as it is pumped through these
channels, and it is then necessary to get rid of the heat that has gone into
the liquid. Some type of heat exchanger is required. Most systems use a liq-
uid-to-air heat exchanger, which consists of a radiator, and a blower which
blows air through the radiator this system is very similar to that on an au-
tomobile. The hot liquid passes through the radiator and heats the radiator
surface. The air blows across the radiator surface and removes the heat
from the radiator. Therefore, the liquid that exits from the radiator is cooler
than the liquid that entered the radiator. In some other systems, primarily
those used on shipboard, a liquid-to-liquid heat exchanger may be used. In
this device, two liquid cooling paths are involved. One path carries the cool-
ant pumped through the klystron amplifier. The other path may carry sea
water. Heat is transferred from the klystron cooling liquid to the sea water,
and the sea water is dumped back into the ocean. Liquid-to-liquid heat ex-
changers are smaller than liquid-to-air heat exchangers, and they are also
quieter, since no blower is required.
Refer now to Figure 4.36, a diagram of a typical klystron amplifier liquid-
cooling system. The right half shows the method of distributing the cooling
liquid to each of the individual channels. The cool liquid enters the high-
pressure manifold at the top of the drawing. From the high-pressure man-
ifold, the liquid passes through valves used to adjust the flow (in each indi-
vidual channel) to the desired level. The liquid then passes through the
component of the system that requires cooling, such as the klystron collec-
tor, body, etc. Flow meters are normally incorporated in each of the individ-
ual channels to monitor the flow and to make it easy to adjust the flow to
the desired level. Liquid flow-switches are placed in each individual chan-
nel, so that the appropriate power supplies will be turned off if the flow in
that channel falls below the minimum required level. The liquid then goes
through shutoff valves into the low-pressure manifold. Both manifolds
are often fitted with temperature gauges and pressure gauges, over-tem-
perature interlocks, and over-pressure interlocks.
Figure 4.36. Typical Liquid Cooling System for a Klystron Amplifier
M
P M
D
e
-
i
o
n
i
z
e
r
L
i
q
u
i
d

L
e
v
e
l
A
l
a
r
m
N
i
t
r
o
g
e
n
T
a
n
k
P
u
m
p
P
o
w
e
r
S
u
p
p
l
y
M
i
s
c
.
C
o
o
l
i
n
g
P
r
e
s
s
u
r
e
R
e
g
u
l
a
t
o
r
K
l
y
s
t
r
o
n
C
o
l
l
e
c
t
o
r
K
l
y
s
t
r
o
n
B
o
d
y
M
a
g
n
e
t
D
u
m
m
y
L
o
a
d
W
a
v
e
g
u
i
d
e
T
e
m
p
e
r
a
t
u
r
e
G
a
u
g
e
O
v
e
r
-
t
e
m
p
I
n
t
e
r
l
o
c
k
P
r
e
s
s
u
r
e
G
a
u
g
e
O
v
e
r
-
p
r
e
s
s
I
n
t
e
r
l
o
c
k
T
e
m
p
e
r
a
t
u
r
e
G
a
u
g
e
O
v
e
r
-
t
e
m
p
I
n
t
e
r
l
o
c
k
P
r
e
s
s
u
r
e
G
a
u
g
e
O
v
e
r
-
p
r
e
s
s
I
n
t
e
r
l
o
c
k
A
i
r
-
f
l
o
w
A
l
a
r
m
F
i
l
t
e
r
B
l
o
w
e
r
T
e
m
p
e
r
a
t
u
r
e
O
p
e
r
a
t
e
d
B
y
-
p
a
s
s
H
e
a
t
E
x
c
h
a
n
g
e
r
(
R
a
d
i
a
t
o
r
)
B
y
-
p
a
s
s

L
i
n
e
H
i
g
h
-
p
r
e
s
s
u
r
e
M
a
n
i
f
o
l
d
L
o
w
-
p
r
e
s
s
u
r
e
M
a
n
i
f
o
l
d
D
r
a
i
n
D
r
a
i
n
F
l
o
w
A
d
j
u
s
t
i
n
g
V
a
l
v
e
s
F
l
o
w
M
e
t
e
r
s
L
o
w
-
f
l
o
w
I
n
t
e
r
l
o
c
k
s
S
h
u
t
-
o
f
f
V
a
l
v
e
s
In Figure 4.36 we have shown the waveguide cooling in series with the mag-
net cooling. Individual system arrangements vary considerably. For in-
stance, in some systems it is possible to put the klystron body cooling in se-
ries with the magnet cooling; in other systems the waveguide cooling may
be in series with the RF dummy load, etc. The important thing to note is
that each of the individual channels must have provision for adjustment
and measurement of the amount of liquid flowing, and must be provided
with low-flow interlock switches for protection. A pressure regulator is of-
ten installed somewhere in the system. In Figure 4.36 it is shown between
the high-pressure and low-pressure manifolds.
The middle of Figure 4.36 shows provisions for cooling the power supply,
and any number of other things that may be liquid-cooled in a typical sys-
tem. The hot liquid passes from the low pressure manifold, usually through
a filter, and into the heat exchanger.Figure 4.36 shows a liquid-to-air heat
exchanger consisting of a radiator and a blower. After the liquid has been
cooled in the heat exchanger, it goes into the coolant-storage and de-aer-
ator tank.
The de-aerator tank deserves some discussion. Bubbles have a tendency to
form in this type of liquid cooling system. Cool liquid will tend to pick up air
bubbles if the system is open to the air at any point. As the liquid is heated,
these bubbles tend to come out of solution. They will tend to collect in
high parts of the system and may cause difficulty in filling the system in
the first place. A fairly small bubble-content in the cooling liquid can seri-
ously diminish the cooling efficiency, and may even cause damage to the
equipment. Furthermore, air bubbles cause undesirable oxidation of the
metal parts of the system. Fortunately, it is quite easy to remove the bub-
bles. The de-aerator tank is fitted with baffles.
The liquid enters the tank and passes (rather slowly) around and through
the baffles in the tank. The bubbles will be released and rise to the surface,
rather than remaining in the liquid system. With a de-aerator of this type,
it is quite easy to remove bubbles from the system and keep the liquid bub-
ble-free. The air can be bled from the tank, and the tank refilled with liquid
if necessary. A low-liquid level switch is normally used in the coolant stor-
age tank. This switch can be connected to ring an alarm that alerts operat-
ing personnel to the situation; it may sometimes be connected to shut
down the equipment in the event the liquid falls below the safe level.
The liquid passes from the de-aerator tank into a de-ionizer. If free ions are
allowed to build up in the cooling liquid, they may eventually cause damage
to the collector and body channels in the klystron. The seriousness of this
problem varies widely from system to system, and from tube to tube. Some
systems require that ions be kept to a very low level. The de-ionizer, nor-
mally built with a replaceable cartridge, will remove ions from the liquid
and prevent this problem from occurring. The liquid passes from the de-
ionizer into the pump, where it is pumped into the high-pressure manifold.
One or more filters are normally included in the cooling system to remove
any accumulation of foreign material. In some klystrons the cooling pas-
sages are very small and can be plugged easily by dirt or sludge in the cool-
ing system.
Figure 4.36 shows a nitrogen tank that can be connected to pressurize
the de-aerator tank. This requires some explanation. Some klystron ampli-
fiers may be at a higher elevation than the heat exchanger and pump por-
tion of the cooling system; a typical case would be a klystron amplifier on a
large parabolic antenna, with the heat exchanger on the ground. These sys-
tems are quite hard to fill because it is difficult to remove the air from the
system initially. The problem can be solved with the pressurizing arrange-
ment shown in Figure 4.36. The storage tank is first filled as full as possi-
ble. The drains on the manifolds are opened, and pressure is applied to the
storage tank from the nitrogen bottle. The nitrogen pressure forces the liq-
uid out of the tank and up to the manifolds. The air in the lines escapes
through the drain valves. When the lines and manifolds are full, the drains
are closed. The nitrogen tank is valved off and the storage tank may be
filled to the top. The pump can then be started and the remaining air
trapped in the system will normally be picked up by the coolant and deliv-
ered to the de-aerator tank.
Figure 4.36 shows a bypass line around the heat exchanger radiator, and a
temperature operated bypass valve. This is the system that is often used to
control the temperature of the coolant liquid. In the previous discussion,
we pointed out that klystron tuning may be changed somewhat by the tem-
perature of the coolant, since this can change the physical size of the cavi-
ties. Fortunately, temperature control is done easily by bypassing some
coolant around the radiator. In systems where the heat exchanger is fairly
close to the klystron tube, the temperature valve may be simply a bimetal
mechanism that senses the temperature of the liquid as it leaves the radi-
ator. If the liquid is too hot, the valve closes partially and causes more of
the total liquid to flow through the radiator. Conversely, if the temperature
is too cold, the valve readjusts itself to cause more of the liquid to go
around the radiator via the bypass line.
If the klystron amplifier is a long distance from the radiator, a temperature
sensor is normally placed at the high-pressure manifold. This temperature
sensor operates the bypass valve. Since the temperature sensor is a long
way from the temperature controller, and it takes an appreciable length of
time for the liquid to travel this distance, a proportional controller may be
necessary to keep the system from hunting.
Some systems use motor-operated louvers (in the air stream between the
blower and the radiator) for temperature control, rather than the bypass
arrangement. Either system, of course, can only control the temperature to
some point above ambient since the liquid leaving the radiator will always
be somewhat hotter than the temperature of the air blowing through the ra-
diator. Most systems are designed to control the liquid to a temperature be-
tween 10 and 20F higher than the maximum-expected ambient tempera-
ture at the particular location. It is fairly easy to hold the liquid tempera-
ture within +5 F. This close temperature control results in very stable
klystron tuning.
Some discussion of cooling liquids is appropriate here. Distilled water is the
best all-round liquid for cooling klystron amplifiers. Some very-high-power
amplifiers specify that only water can be used. Normal tap water usually
has a large mineral content, and causes scaling of the klystron cooling sur-
faces. Scaling reduces heat transfer and may eventually completely close
the cooling channels. If this occurs, the tube will be seriously damaged.
Unfortunately, water freezes at an inconveniently high temperature. Many
low- and medium-power klystrons permit the use of ethylene-glycol and
water as the cooling liquid. The cooling efficiency of ethylene-glycol and wa-
ter is not as good as pure water. Furthermore, ethylene-glycol reacts with
certain types of metals and hoses that might be used in the system; there-
fore, special care must be taken in designing a system that is to use ethyl-
ene-glycol. Only nonferrous metals should be used in a cooling system for
a klystron amplifier.
Some very large tubes permit only water to be used for the coolant. This
complicates the design of the cooling system, since care must be taken to
protect it from freezing. This is no problem if the system is continually op-
erated, but becomes a very serious problem if the system is shut down in
cold weather. Some large systems are designed with immersion heaters in
the coolant tank. If the klystron is shut down for any reason, these immer-
sion heaters are turned on, and the pump is left running to keep the cool-
ant circulating and prevent freezing.
Additional information on klystron cooling is contained in Varian Applica-
tion Engineering Bulletin No. 17.
11.5. RF Cir-
cuits
We have discussed the theory of klystron amplifier operation, power supply
requirements, and cooling requirements. Now let us consider what is nec-
essary to get RF into, and out of, the klystron amplifier tube.
Figure 4.37 on Page 4-46 shows the RF components typically associated
with a klystron power amplifier. We will not consider the RF exciter (or
driver), since we are only discussing the amplifier portion of a complete
transmitter. The RF input signal from the amplifier is derived, of course,
from an RF exciter. This signal is shown on the left of Figure 4.37.
The RF input signal normally goes through a ferrite isolator so that a con-
stant RF load impedance is presented to the exciter. The input cavity of a
well-designed klystron amplifier normally presents a low VSWR, to the in-
put signal, at the resonant frequency of the cavity; but the VSWR increases
very rapidly for frequencies slightly off the resonant frequency of the cavity.
It is desirable to isolate these high VSWR's from the exciter; the ferrite iso-
lator accomplishes this function.
After the ferrite isolator, the input signal is normally applied to a variable
attenuator. The attenuator is used to adjust the input signal level so that
the amplifier may operate at saturation, or at lower-than-saturation levels
if this should be desired. It may be desirable to monitor the amount of RF
input power being applied to the tube. This is normally done with a direc-
tional coupler and some sort of RF power monitor. This monitor may be a
simple crystal detector and meter, or it may be a thermistor and RF power
bridge arrangement.
The input coupler can also be used to help tune the first cavity of the
klystron to resonance. Many klystrons have a coarse-tuning indicator that
allows them to be set approximately to frequency. However, this is not true
for all tubes, and it may be quite difficult to get them on frequency when
they are first put into a system. The first cavity tuning can be done easily by
using the input coupler as a reflected-power coupler. To do this, put the
input monitor on the opposite arm from that shown in Figure 4.37. When
this is done, the monitor will show the power that is reflected from the
first cavity of the tube, back toward the RF exciter.
We stated earlier that the cavity presents a low VSWR at its resonant fre-
quency; but if it is not tuned to the frequency of the RF input signal, the in-
put power will be mostly reflected from the first cavity. If we now set up to
monitor this reflected power, and then tune the cavity, the reflected power
will decrease and go through a null when the cavity is tuned to the frequen-
cy of the input signal. This simple procedure allows one to find the first
cavity and tune it to resonance. After this is done, the RF input monitor is
connected to again monitor forward input power.
Figure 4.37. Typical RF Circuitry for a Klystron Amplifier
F
e
r
r
i
t
e
I
s
o
l
a
t
o
r
K
l
y
s
t
r
o
n
R
F

I
n
p
u
t
M
o
n
i
t
o
r
W
a
v
e
g
u
i
d
e
A
r
c

S
e
n
s
o
r
L
o
w
-
p
a
s
s
F
i
l
t
e
r
C
o
n
t
r
o
l
U
n
i
t
D
u
m
m
y
L
o
a
d
R
F
S
a
m
p
l
e
B
a
c
k
-
P
o
w
e
r
M
e
t
e
r
H
a
r
m
o
n
i
c
F
i
l
t
e
r
B
a
c
k
-
P
o
w
e
r
C
o
u
p
l
e
r
F
o
r
w
a
r
d
-
P
o
w
e
r
C
o
u
p
l
e
r
S
a
m
p
l
i
n
g
C
o
u
p
l
e
r
I
n
p
u
t
C
o
u
p
l
e
r
C
r
y
s
t
a
l
S
w
i
t
c
h
V
a
r
i
a
b
l
e
A
t
t
e
n
u
a
t
o
r
R
F
S
w
i
t
c
h
T
o

A
n
t
e
n
n
a
R
F
I
n
p
u
t
T
o

B
e
a
m
S
u
p
p
l
y
L
o
w
-
p
a
s
s
F
i
l
t
e
r
R
F

O
u
t
p
u
t
M
o
n
i
t
o
r
T
h
e
r
m
i
s
t
o
r
Between the input coupler and the first cavity, the diagram shows a crystal
switch. This is a very fast-acting device that will insert between 20 and 30
dB attenuation in the RF input line. It is used primarily to remove RF input
from the tube quickly in case of arcs in the output waveguide; this will be
discussed in more detail later. The crystal switch is normally biased to have
low RF attenuation. This completes the discussion of the components asso-
ciated with the RF input to the tube.
Let us now consider the RF output components. The first item normally
found in the RF output circuit is a waveguide-arc sensor. Waveguide arcs
are a troublesome problem in high power CW systems, particularly those at
the higher frequencies where waveguide sizes are quite small. Typically,
waveguide arcs will occur at power levels above 5 kilowatts at S-band, and
above 1 kilowatt at X-band. Although the cause of this arcing is not com-
pletely understood, one of the most plausible theories is that ions build up
due to thermal ionization from heating of contaminants within the
waveguide, or from local area heating at small discontinuities within the
waveguide. This ionization builds up in a CW system until the dielectric
strength of the gas in the guide is sufficiently reduced to cause a sustained
arc to form. Apparently, this buildup of ionization does not occur as readily
in pulse systems because the ions have time to disperse during the interval
between the pulses. In any event, once the arc is formed in a CW system, it
almost invariably travels toward the source of RF power. If the arc is al-
lowed to reach the output window of the klystron, local heating will occur
and the window may be destroyed very quickly. Since the arc presents an
effective short-circuit to the waveguide, a very high VSWR exists, and it is
quite common to start a secondary arc in other points in the waveguide
feedline or at the output window of the klystron.
Experience with very high-power CW amplifiers at X-band indicates that
the arc should be quenched in a few microseconds to prevent damaging the
tube. This short time precludes removing the RF power by de-energizing
the power supplies due to the long time lag of the relays and contactors in-
volved in power supply shutoff. Removing RF drive is the fastest method of
quenching the waveguide arc. This removes the RF output power and the
arc disappears. Since the arc causes a bright light in the normally dark
waveguide interior, a good way to sense the arc is to look into the
waveguide with some light sensitive device, such as a photoelectric cell or a
solar cell. Solar cells have proven superior to photo cells for this service, be-
cause they respond more quickly to the presence of light, and because they
are less affected by temperature. When an arc occurs, the sensor will devel-
op a voltage, which can be used (with follow-up control circuitry) to change
the bias on the crystal switch. This inserts a large amount of attenuation in
the RF input to the klystron; the RF output falls 20 or 30 dB, and the arc is
extinguished. Additional circuits may be used to remove the beam voltage
from the klystron if desired.
The next component shown in the RF output circuitry is a backpower di-
rectional coupler. This coupler is connected to monitor power reflected
from a mismatch in the antenna or the dummy load. Klystron amplifier
specifications typically require the load VSWR to be below 1.2. Excessive
reflected power causes high voltage gradients in the waveguide and excess
heating in the tube. It may also tend to detune the last cavity and lower the
output. Backpower monitoring is normally done with a directional coupler
and some type of RF power meter. As shown in Figure 4.37, this may be a
simple crystal detector and meter, or it may be a thermistor and an RF
power bridge. The backpower coupler can be arranged for redundant
waveguide-arc protection. If an arc occurs between the coupler and the an-
tenna, a very high VSWR will be present, and the reflected (back) power will
rise suddenly. This can be sensed by the crystal detector, which can trigger
the control unit of the waveguide arc-detector system. The control unit
then changes the bias on the crystal switch, increases its attenuation, and
removes the RF output. In this fashion, waveguide arcs occurring far from
the light sensor can be detected. Again the beam power supply may be
turned off if desired. A variable attenuator is often inserted between the
backpower coupler and the detector to permit convenient adjustment to the
desired operating level. The backpower meter is often of the type that in-
cludes upper-limit contacts. If the backpower slowly increases to an exces-
sive level, these contacts will close and can be used to turn off the beam
power supply.
The next component in the RF output circuit is the forward power direc-
tional coupler. This coupler monitors the power being delivered to the an-
tenna (or to the dummy load). The power indicating device can, again, be ei-
ther a simple crystal detector and meter, or it may be a thermistor and an
RF power bridge. A variable attenuator is normally included between the
coupler and the power monitoring device to permit convenient adjust-
ments. In some systems the power meter has both upper- and lower-limit
contacts. These contacts can be arranged to ring alarms if the power out-
put varies excessively.
Many power amplifier systems use a third directional coupler to sample the
RF output. Such a sample may be used to monitor noise performance of the
equipment, to check distortion in the output signal, etc.
The RF output is normally applied to an antenna. However, a dummy load
is very useful for absorbing, and accurately measuring the power being
generated. A dummy load is also handy for initial adjustment and tune-up
when it may be undesirable to radiate. An RF switch is incorporated in
some amplifiers to permit connecting the klystron easily either to the an-
tenna or to the dummy load. This should be done only when the drive has
been removed from the tube.
A harmonic filter is sometimes included in the RF output circuit. Depend-
ing upon the type of tube and the operating conditions, the output of the
klystron may be rich in harmonics. It is common for the second and third
harmonics to be only 20 dB below the fundamental. In some situations ra-
diation of this harmonic power causes objectionable interference. The har-
monic energy can be removed by a harmonic filter in the system. This is
simply a low-pass filter that absorbs the harmonic power, while passing the
fundamental.
Figure 4.37 shows low-pass filters in each of the directional coupler output
arms. Directional couplers have the unhappy characteristic that they
usually couple harmonics more strongly than the fundamental. An appre-
ciable amount of harmonic energy in the RF output may cause incorrect
readings in the RF power meters. Simple low power, low-pass filters remove
this harmonic energy from the RF power monitors and prevent this inaccu-
racy. It is particularly important to have a low-pass filter in the forward
power coupler, because the ratio of harmonic energy to fundamental energy
is quite dependent upon the way the klystron tube is tuned. If one is watch-
ing the forward power meter while tuning the tube, and the harmonics are
not suppressed by a filter, it may be difficult to tune the tube correctly.
This discussion has covered the microwave components usually used with
a high-power klystron amplifier. The components discussed allow all of the
important parameters to be monitored. In addition, some of these compo-
nents provide features necessary to protect the tube and operating person-
nel.
11.6. Tuning Tuning a klystron amplifier is very simple, when one understands the prin-
ciples. Let us consider first the steps to tune the klystron to the synchro-
nous-tuned condition. This is the simplest tuning adjustment. Remember
that it results in highest-gain and narrowest-bandwidth operation of the
tube. Many klystrons have a dial arrangement that allows adjustment of
the cavities to approximately the right frequency before applying power to
the tube. However, some tubes have no integral tuning indicators. Most
tubes, when delivered, are tuned to some frequency, indicated by test data
accompanying the tube. So, at least, one knows where the tube is tuned
when new, and which direction to go for the desired frequency. The instruc-
tions also give the direction of tuner rotation to raise, or lower, the cavity
frequency. All these things help tune the tube the first time.
Let us consider the most difficult case, a tube with no built-in tuning indi-
cators and no indication of the present tuning. In addition, the driver is
fixed-frequency so that its frequency cannot be matched to that of the tube;
one must tune the tube to the driver. The tube is installed in the transmit-
ter; the exciter is operating and delivering power; the cooling has been
turned on, and voltages are applied to the tube. Everything is working, but
there is no power output, because the tube is not tuned to the frequency of
the exciter. How can you get power from the tube? It is simple. First, you
must adjust the first cavity frequency. When we were discussing Figure
4.37, we mentioned that you could find the first cavity tuning by reconnect-
ing the RF input directional coupler to read the power reflected from the
tube. This is done by moving the RF input power monitor to the reflected-
power arm of the coupler. (Don't forget to put the termination on the for-
ward power arm of the coupler). You are now set up to monitor the power
reflected from the first cavity of the amplifier. This power will be minimum
when the first cavity is tuned to the driver frequency. When you start tun-
ing a tube, it is a good idea to keep some mental notes on how far you've
gone, so you can return to the starting point. A simple way to do this is
count turns as you rotate the tuning tool. So, suppose you begin tuning
the first cavity, rotating the tuner in a clockwise direction, and counting
turns as you go. Look for a significant dip in the RF input monitor that is
measuring the reflected power.
You may find some small dips on the way, but look for the major one, which
will be the correct tuning point. Most tubes are equipped with tuner stops
to prevent damage to the tuner mechanism. Suppose you rotate the tuner
clockwise and do not find any major dips in the reflected power all the way
to the tuner stop. Then, return to the starting point, counting turns as you
go. Then continue counterclockwise (again counting turns from the original
position) until you find a significant dip in the reflected power. Minimize the
reflected power by tuning. Now you can leave the first cavity alone for a
while, since it will be almost on resonance. Reconnect the RF input monitor
to read forward power.
Next, tune the output cavity. Ignore any intermediate cavities for the mo-
ment. You may still have no measurable reading on the output meter. In-
crease the RF drive to the highest level you can get from the exciter. Now
you are ready to tune the output cavity. Although you may have tuned the
first cavity counterclockwise, it does not necessarily follow that the other
tuners should be turned the same direction. Refer carefully to Operating
Instructions or to markings on the tube itself. Start tuning the output cav-
ity, counting turns as you go. With luck, you will soon bring the output cav-
ity to resonance, and will see an indication on the RF output meter. Maxi-
mize this reading by tuning the last cavity of the tube. Once you see any
reading on the RF output meter, the rest is simple. You know you have the
first and last cavities in tune (or almost).
If you have a three-cavity tube, now adjust the middle cavity. Determine
the tuning direction from the instructions or tube markings, or carefully try
one direction and then the other. Simply tune the cavity to maximize the
power output reading. However, once you approach a sizable output, it is a
good idea to reduce the RF drive to be sure that you do not inadvertently
saturate the tube. Tune the middle cavity to maximize the power output
reading. Now reduce the drive to a low level, so that the power output meter
is far below full power (less than 30 per cent of full-power). Retune the in-
put cavity to resonance, then retune the output cavity to resonance, then
retune the middle cavity to resonance. The tube is now synchronously-
tuned. Synchronous-tuning is always done with low RF input power. Now
increase the RF drive until the tube saturates (or to whatever power level
you may wish to use).
The procedure for a four-cavity tube is very nearly the same. First, tune the
input and output cavities to resonance, then the intermediate cavities one
at a time. Cavities are often numbered, number 1 being the input cavity
and the number 4 the output. For synchronously-tuning a four-cavity tube
(after you have some power output), reduce the drive to a low level and tune
in the sequence 1-4-3-2; i.e., first tune the input cavity, then the output
cavity, then the next-to-the-output cavity, and then the next-to-the-input
cavity. The tuning of the fourth cavity is normally quite broad, whereas
the tuning of the first, second, and third cavities is quite sharp. The rea-
son is that the output cavity is fairly low-Q compared with the other cav-
ities.
You have now learned to tune a klystron to the synchronous-tuned condi-
tion. This is always the first tuning condition even if you want to stagger-
tune the tube later. There are several methods of stagger-tuning, two of
which will be discussed briefly. Remember, in Theory of Operation we stat-
ed that stagger tuning could be used to obtain more power than synchro-
nous-tuning. This is done simply by adjusting the third cavity to a higher
frequency. The detailed steps are approximately as follows: First, tune the
tube synchronously at low RF input; then increase the RF drive until the
tube is saturated. Now, leave the drive alone and detune the third cavity in
the high-frequency direction. The Operating Instructions for the tube will
tell you whether this is clockwise or counterclockwise. As you detune the
third cavity, the output will decrease, because more drive is needed for sat-
uration. Continue detuning until the power output has dropped approxi-
mately 6 to 10 dB. Now increase the RF drive power until the tube again is
operating at saturation. You will find that this new saturated output is
higher than the output obtained with the tube synchronously-tuned. You
may be able to squeeze a little more out of the tube, but probably not
much. You can detune the third cavity still farther and then increase the
drive to see if you get more power output than before. The power output
maximum is normally quite broad; you will be able to detune the third cav-
ity considerably either side of this point without making an appreciable
change in output. Eventually you may become limited by the amount of
power available from the exciter.
A second type of stagger-tuning should be mentioned. This is stagger-tun-
ing to achieve a desired bandwidth characteristic over the passband. The
problem is similar to broadbanding an IF amplifier by stagger-tuning. You
will need an RF driver that can be swept in frequency rapidly (electronical-
ly-swept), and whose power output is constant during sweeping. You will
need to sample the RF output with a crystal detector. Apply the crystal de-
tector output to an oscilloscope so that you can see the passband of the
tube. The X-axis of the oscilloscope sweep must be synchronized with the
RF input sweep voltage. Again you will start with the synchronous-tuned
condition and probably with the tube operating at saturation. To broad-
band the tube, it is usually best to detune the third cavity to the high fre-
quency side, and to detune the second cavity to the low frequency side (as-
suming a four-cavity klystron). The first and fourth cavities are normally
left tuned to the center of the passband. You may wish to adjust the RF in-
put power periodically, as you detune the klystron, to keep the tube oper-
ating near saturation. You will find that the bandwidth of the tube is larger
when you are operating at saturation than below saturation. The details of
the broadband tuning that you may wish to accomplish are beyond the
scope of this note. We have only indicated the equipment that is necessary
and the general procedures to be followed.
11.7. Noise in
Klystron Am-
plifiers
Volumes have been written about noise in microwave systems; obviously,
we can only touch the very high points in this discussion. Noise is any-
thing that causes the RF output signal to be different from the RF input
signal. We have already mentioned that the output may contain harmon-
ics. This is primarily because the RF output cavity is excited by bunches
of electrons that come through the output gap once every cycle. These
bunches essentially kick the output cavity and cause oscillating cur-
rents to flow in it. Since the driving force on the output cavity is not con-
tinuous, but rather occurs in quick kicks, it is intuitively evident that the
output current may not be purely sinusoidal; therefore, it will contain
harmonic components. This situation is quite analogous to a class-C tri-
ode amplifier in which the plate current flows in bursts, and sets up oscil-
lating currents in the resonant plate circuit. Class-C amplifiers are also
rich in harmonics for the same reason. In general, the harmonic output
from klystron amplifiers is largest (percentage-wise with respect to the
fundamental carrier power) when the tube is operating at saturation, or is
being over-driven beyond saturation. Harmonic content decreases (per-
centage-wise) when the tube is operated below saturation. As discussed
previously, harmonics can be reduced in the output by using harmonic
filters.
Another source of distortion is non-linearity of the klystron. If the RF in-
put signal is amplitude-modulated, the RF output may not perfectly fol-
low the RF input. This can result in distortion, becoming worse as the
tube is driven closer to saturation on the peaks of the RF input signal. In
general, klystron amplifiers should not be used to amplify amplitude-
modulated signals if the RF output is driven higher than about 0.7 of the
saturated level. Between 70 and 100 per cent of saturation, considerable
distortion can occur.
A klystron amplifier will generate a certain amount of white noise, just as
any other electron tube. White noise occurs primarily because an electron
beam is never perfectly homogeneous. The number of electrons will vary
slightly with time, primarily due to shot noise at the cathode surface; this
variation shows up as random noise in the RF output. A certain amount of
noise may also be generated by electrons striking the drift tubes. These are
the electrons that create the body current. The body-current interception
may be slightly random; this again will perturb the electron beam and
cause a small amount of random noise to appear in the output.
You should understand one interesting effect about klystron amplifiers. In-
tuitively, one would think that the output of an amplifier cannot possibly
be quieter than the input signal. In certain cases, the klystron amplifier
can, indeed, have an output that is quieter than its input. Consider a tube
operated at saturation; this is the normal situation when the intelligence is
being transmitted by frequency-modulation of the carrier. And suppose
that the output of the RF exciter is fairly noisy with amplitude-modula-
tion. The klystron amplifier has the desirable property that it will suppress
amplitude-modulation of the input signal, if the tube is being operated at
saturation. An examination of the output-vs.-input curves shown in Figure
4.34 on Page 4-32 will explain how this happens. It is obvious that, with
the amplifier operating at saturation, rather large changes in the amplitude
of the RF input signal will cause no change in the amplitude of the RF out-
put signal. In some systems this effect is very noticeable, and it is not un-
common to find that the AM noise from the exciter can be suppressed by 10
to 20 dB, simply by operating the amplifier at saturation.
Additional information on noise characteristics of klystrons is given in Vari-
an Application Engineering Bulletins Numbers 11 and 18. Definitions of
AM and FM noise and methods of measurement are discussed. Equations
for computation of noise caused by power supply ripple are derived and ex-
amples are given for typical conditions.
11.8. Summary This bulletin has attempted to familiarize you with the basic principles of
klystron amplifiers and the equipment usually associated with these tubes.
Precautionary measures and safety devices have been described in consid-
erable detail in order to explain their importance, and to convince you that
cheating them for expediency will very likely result in expensive damage
to tubes, equipment, or personnel. Tuning procedures described are those
used for Varian klystrons but are generally applicable to similar tubes. Too
much cannot be said in favor of studying Operating Instructions for equip-
ment and tubes thoroughly before applying power to your transmitter.
Ion Chamber Theory 5-1
This chapter covers the rudimentary concepts of ion chamber characteristics and is writ-
ten for an intended audience of engineers, test personnel, and manufacturing personnel,
wishing to learn more about Varian's ion chamber.
Ion Chamber Theory
5-2 Ion Chamber Theory
Table of Contents
1. Introduction: .................................................................................................................... 5-3
2. Present Configuration: ..................................................................................................... 5-3
3. Efficiency: ........................................................................................................................ 5-5
4. Upper Limit of Dose Range (Saturation): ........................................................................... 5-6
5. Applied Electric Field: ...................................................................................................... 5-7
6. Effects of Temperature and Pressure: ............................................................................... 5-7
7. Beam Opacity: ................................................................................................................. 5-8
8. Inverse Square Law: ......................................................................................................... 5-8
9. Insulation Materials: ........................................................................................................ 5-9
10. Pulse Shape: ................................................................................................................ 5-10
11. Concluding Remarks: ................................................................................................... 5-11
12. References: .................................................................................................................. 5-12
Figure 5.1. Basic Ion Chamber Components and Characteristics:......................................... 5-3
Figure 5.2. Simplified Dosimetry/Steering System Block Diagram: ....................................... 5-4
Figure 5.3. Geometric Layout of Electrode and Signal Plates:................................................ 5-4
Figure 5.4. Fraction of Ions Collected as a Function of Dimensionless Variable 1/U
p
: ........... 5-6
Figure 5.5. Upper Limit of Dose Range: ................................................................................ 5-6
Figure 5.6. The Different Regions of Operation of Gas-filled Detectors:.................................. 5-7
Figure 5.7. Diagram Illustrating the Inverse Square Law: ..................................................... 5-8
Figure 5.8. Diagram showing the Derivation of the Pulse Shape V
R
(t): ............................... 5-10
Figure 5.9. Output Pulse Shape: ........................................................................................ 5-11
Ion Chamber Theory: Introduction 5-3
1. Introduction Gas filled ion chambers are a type of radiation detector. The normal opera-
tion mode is based on collection of the charges through the application of
an electric field. These charged particles are created by direct ionization of
the gas within the ion chamber. After a neutral molecule is ionized, the re-
sulting positive ion and free electron produce the basic electrical signal de-
veloped by the ion chamber.
Figure 5.1 illustrates the basic principles of an elementary ion chamber. A
volume of gas is enclosed within an electric field. At equilibrium, the cur-
rent flowing in the external circuit will be equal to the ionization current
collected at the electrodes, and a sensitive ammeter can be used to mea-
sure the ionization current. Notice on Figure 5.1 that current output is con-
stant after the knee of the curve, especially at low dose rates.
2. Present
Configuration
The present monitoring ionization chamber of the high energy Clinac 1800
and Clinac 2100C is constructed of several plates or electrodes. The pur-
pose of the ion chamber is twofold: 1. to monitor the beam position, and 2.
to monitor beam intensity. These needs are satisfied as illustrated in Figure
5.2, a simplified diagram of the ion chamber, dosimetry, and beam steering
systems.
This ionization chamber consists of two collecting plates sandwiched be-
tween three polarizing plates. The two collecting plates are oriented 90 in
relation to each other to allow inplane and crossplane beam symmetry
monitoring. The collecting plates are divided into four sectors, each defin-
ing a distinct laminar collecting volume.
Figure 5.1. Basic Ion Chamber Components and Characteristics
Electrodes
High Irradiation Rate
Low Irradiation Rate
Gas Enclosure
I
V
V
I
5-4 Ion Chamber Theory: Present Configuration
Figure 5.2. Simplified Dosimetry/Steering System Block Diagram
Angle R
Steering
Coils
Angle T
Steering
Coils
Target
-500V
P.S.
Position R
Steering
Coils
Position T
Steering
Coils
Buncher T
Steering
Coils
Buncher R
Steering
Coils
Accelerator
Guide
E
lectron
B
e
a
m
A
A
1
A
4
A
13
A
14
A
8
A
7
A
3
A
2
A
9
A
10
A
11
A
12
A
5
A
6
A - B
E - F
G - H
C - D
B
F
E
G
H
C
D
A + B
C + D
MU1
MU2
SYM1
SYM2
(A - B) + (E - F)
(C - D) + (G - H)
Dose Rate
Meter
Integrator
Integrator
Figure 5.3. Geometric Layout of Electrode and Signal Plates
Electrode Plate Signal Plate
Ion Chamber Theory: Efficiency 5-5
The two inner D-like sectors provide signals for both dosimetry and beam
angle monitoring. The outer two arc-like sectors are for beam position mon-
itoring only. The collecting plate closest to the target monitors the radial
plane, the collecting plate closest to the patient monitors the transverse
plane.
A polarizing voltage of 500 volts is used with a plate spacing of 1mm. A good
rule of thumb for dielectric breakdown is 24,000 V per in. So at 0.040"
(mm) we should be able to hold off 1000 V.
3. Efficiency Assume the plates to be a distance d apart and held at a potential differ-
ence v. The positive plate is on the left and the negative plate is on the right.
The positive ions will occupy the space to the right as they are pulled to the
negative plate, while the negative ions will occupy the space to the left as
they are attracted to the positive plate.
If no recombination occurs, the charge collected per second, i.e., the cur-
rent, is:
(Equation 117)
where: Q
c
is a charge per unit volume per second.
d is the plate separation.
A is the area.
Recombination occurs when + and ions meet and recombine before the
charged ions make it to the collecting electrode. When recombination is
taken into consideration, an efficiency factor must be multiplied to Equa-
tion 117. For pulsed radiation, whose duration is short compared with the
collection time, the collection efficiency or fraction of charge collected for
pulsed radiation, f
p
, may be expressed by:
(Equation 118)
where: U
p
is the dimensionless parameter:
(Equation 119)
where: k is a gas constant.
q is the charge in esu produced per cm
3
per second.
v is the voltage potential between the plates in volts.
d is the separation of the plates.
The actual current or charge collected per second with recombination
taken into consideration is Equation 117 multiplied by the efficiency
(Equation 118).
(Equation 120)
From Equation 120 it can be seen that if the distance d is moving because
of unstable signal plates, the effect on the current signal is enormous.
Equation 120 also suggests that we want:
a. small plate separation, and
b. high voltage.
I Q
c
d A =
f
p
100 U
p
1 U
p
+ ( ) ln =
U
p
k
v
-- d
2
q =
I Q
c
d A f
p
=
5-6 Ion Chamber Theory: Upper Limit of Dose Range (Saturation)
4. Upper Limit
of Dose Range
(Saturation)
A constant dose sensitivity throughout the dose range provides a linear re-
sponse (i.e., reading vs. dose, r vs. D). Saturation is manifested by a de-
crease in the dose sensitivity. If pushed to the limit, this function will be-
come zero and then become a negative value. Figure 5.5 illustrates a dou-
ble-valued dose response function resulting from a decrease in dosimeter
sensitivity at high doses.
One factor that can add to saturation in an ion chamber is recombination.
An ion chamber is said to be saturated to the degree that ionic recombina-
tion is present. Saturation is minimized by ensuring that a large electric
field exists everywhere within the ion chamber. Increasing the ion collect-
ing potential generally helps, but is limited by electrical breakdown of insu-
lators. For a more detailed explanation, refer to Attix
1
page 281.
Figure 5.4. Fraction of Ions Collected as a Function of
Dimensionless Variable 1/U
p
Pulsed Radiation
F
r
a
c
t
i
o
n

C
o
l
l
e
c
t
e
d

f
P
F
r
a
c
t
i
o
n

C
o
l
l
e
c
t
e
d

f
P
Dimensionless Variable 1/U
P
10.0
0.9
0.8
0.7
0.6
0.5
0.4
0.2
0.1 1.0 10 100 1000
10.0
0.98
0.96
0.94
0.92
0.90
0.88
0.86
P
N
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

V
d
X
+
V
+
X
X
Figure 5.5. Upper Limit of Dose Range
Reading
r
= 0
Negative
slope
Double-valued Function
= sensitivity Slope =
dr
d g D
dr
d g D
Dg Dose
Ion Chamber Theory: Applied Electric Field 5-7
5. Applied
Electric Field
The different regions of applied electric fields (volts per meter) are shown in
Figure 5.6. Region I has very low values of applied voltage, 0 to 100 V for a
plate spacing of 0.040". The electric field is insufficient to prevent recombi-
nation of the original ion pairs. Region II shows the normal mode of opera-
tion for ionization chambers. Region III is the region of true proportionality,
and represents the mode of operation for proportional counters. Increasing
the applied electric field even further introduces nonlinear effects as shown
in region IV. If the applied electric field is increased even further, we enter
the Geiger-Mueller region.
The present ion chamber configuration uses an Acopian (or equivalent)
power supply of 500V dc. Experimental data shows that the ion chamber
output is constant from 300 600 volts. Good voltage regulation is essen-
tial, since HV fluctuations induce current to flow in the electrometer input
circuit from the capacitive coupling. Also, we previously mentioned that re-
ducing the plate separation increases the collection efficiency and signal to
noise ratio.
6. Effects of
Temperature
and Pressure
Whenever absolute ionization measurements are made, corrections must
be applied to account for the change in density of the gas (with pressure
and temperature). The mass of a given volume of air at temperature T and
pressure P is related to its mass at 0C and 760mm Hg by:
(Equation 121)
Figure 5.6. The Different Regions of Operation of
Gas-filled Detectors
Pulse
Amplitude
(log scale)
2 MeV
1 MeV
Ion
Saturation
Proportional
Region
Limited
Proportional
Region
Geiger-
Mueller
Region
I
II
III
IV
V
Applied Voltage
m(T,P) m(0C, 760mm)
273.2K ( ) P
mm
( )
273.2K T + ( ) 760mm ( )
----------------------------------------------------------- =
5-8 Ion Chamber Theory: Beam Opacity
The first bracketed term corrects for the expansion of the gas with in-
creased temperature. The second term corrects for changes due to pres-
sure.
If the ion chamber was calibrated at another temperature, the appropriate
(absolute) temperature must be used. This correction formula assumes
that the ion chamber cavity is not sealed and that pressure inside the
chamber is atmospheric. This is not the case with any Clinac ion chamber;
it is hermetically sealed with a weld. The changes in mass in a sealed ion
chamber should be nearly zero.
A worst case scenario would be an unsealed ion chamber. A 5F change in
temperature (2.7C) represents a 0.9% change in mass. A 10F change in
temperature (5.6C) represents a 2.0% change in mass. The effects of pres-
sure changes are not as dramatic.
7. Beam Opac-
ity
In dealing with a compound or a mixture of molecules, it is sometimes con-
venient to describe the mixture by an effective atomic number, Z. The con-
cept is useful in dealing with ion chambers. The effective atomic number, Z,
of a mixture may be defined by:
(Equation 122)
where: a
1
to a
n
are the fractional numbers of electrons per gram
belonging to materials of atomic numbers Z
1
to Z
n

respectively.
m has the experimental value of 3.5 for air.
8. Inverse
Square Law
The ion chamber is not located at isocenter where one wishes to measure
dose. The actual photon fluence, , (number of photons per unit area) seen
by the ion chamber is quite high as illustrated below.
In general it follows that (Equation 123)
This is a simple statement of the inverse square law. In Varian's High En-
ergy Clinacs f
2
= 100cm, f
1
= 25cm. The calculated resulting ratio of these
two distances squared is 16; which means that the photon fluence is 16
Z a
1
Z
1
m
a
2
Z
2
m
a
n
Z
n
m
+ +
m
=
Figure 5.7. Diagram Illustrating the Inverse Square Law
P
a
a b
b
f
1
f
2
B
------
B
f
2
2
f
1
2
------ =
Ion Chamber Theory: Insulation Materials 5-9
times higher at the ion chamber than at isocenter. This is only approximate
and does not take back scatter and secondary electron fluence into consid-
eration. Experimental data shows that the photon fluence is 20 to 50 times
higher at the ion chamber than at isocenter.
9. Insulation
Materials
Polystyrene, polyethylene are excellent insulators for ion chambers. Teflon
is more readily damaged by radiation and should be avoided. Charged-par-
ticle beams incident on a thick insulator will build up charge wherever the
particle stops at the end of their path. When large insulating plastics are ir-
radiated to high doses by electron beams, the charge buildup due to
stopped electrons may cause electric fields strong enough to influence the
paths of the primary electrons in the ion chamber.
9.1. Mica Mica is the name given to a group of minerals of related similar physical
properties characterized chiefly by perfect basal cleavage (they can split
readily in one direction). The ASTM visual quality classifications for mica
are as follows:
V-1 clear
V-2 clear and slightly stained
V-3 fair stained
V-4 good stained
V-5 stained, A quality
V-6 stained, B quality
V-7 heavy stained
V-8 black dotted
V-9 black spotted
V-10 black stained
In practice, first quality is equivalent to V-3 and second quality to V-4.
Varian presently purchases V4 mica through Spruce Pine Mica Co. The mi-
cas are complex silicates of aluminum with potassium, magnesium, iron,
sodium, lithium, fluorine and traces of other elements. The mica we use in
our application is Muscovite, H2KAl3(SiO4)3. Other types of micas are: Phl-
ogopite, Biotite, Lepidolite, Paragonite, Zinnwaldite.
The mica thickness specification was dropped from 0.010" to 0.007"
+0.003" in January 1988 because of the dwindling mica supply. The 0.003"
reduction in thickness constitutes a 50% stiffness change. The unstable
mica assemblies resulted in changes in symmetry and dose calibration.
There is also suspicion that other changes occurred since we had instabil-
ities even in chambers with selected thick mica. This phenomenon is pretty
well understood now and engineering changes have been implemented to
minimize the mica ion chamber problem, allowing manufacturing yields to
go up from 40% to 90%. This is only a short term solution because the mica
supply is still limited. The gold is silkscreened on with thicknesses ranging
from 0.0001" to 0.0007".
Shortly after this document was published, Varian switched to
DuPont Kapton as the insulating material for all High En-
ergy Clinacs except for the 2500 and 2500C.
5-10 Ion Chamber Theory: Pulse Shape
10. Pulse
Shape
The pulse shape depends on the configuration of the electric field and the
position at which the ion pairs are formed. Assume a parallel plate ion
chamber with electric field intensity E
f
, voltage V across the electrodes and
separated by a distance d.
(Equation 124)
A further simplification assumes that all ion pairs are formed at an equal
distance x from the positive electrode. This situation is sketched in Figure
5.8.
The pulse shape is most easily derived on arguments involving the conser-
vation of energy. The energy required to move the charges from their origin
must come from the energy originally stored across the capacitance C. This
energy is where V
0
is the applied voltage.
After a time t, the ions will have drifted a distance v
+
t toward the cathode,
where v
+
is the ion drift velocity. Similarly, the electrons will have moved a
distance v
t toward the anode. Both of these motions represent the move-

ment of charge to a region of lower potential (dV). This energy is equal to
Q(dV) for both ions and electrons, where Q is the total charge and dV is the
change in electric potential. The charge Q = n
o
e, where n
o
is the number of
original ion pairs and e is the electron charge. Conservation of energy can
be written:
The signal voltage is measured across R in Figure 5.8 and will be denoted
as V
R
. The following equation describes the initial portion of the signal
pulse and predicts a linear rise with time:
(Equation 126)
Figure 5.8. Diagram showing the Derivation of the Pulse Shape V
R
(t)
for the Ion Chamber Signal
Original
stored
energy
=
Energy
absorbed
by ions
+
Energy
absorbed
by electrons
+
Remaining
stored
energy
= + + (Equation 125)
E
f
V d =
v t
+
V
O
C R V
R v t
x
d V
ch
1 CV
0
2
1 2 CV
o
2
n
o
eEv
t n
o
eEv
+
t 1 2 CV
ch
2
V
R
n
o
e
dC
-------- v
+
v
+ ( )t = rising portion
Ion Chamber Theory: Concluding Remarks 5-11
The decaying portion of the signal pulse is dependent on the position x at
which the electrons were originally formed within the chamber. The pulse
reflects only the drift of the electrons and will have an amplitude given by:
(Equation 127)
If the collection circuit time constant were very large (RC>>t+) the maxi-
mum amplitude of the signal pulse would be:
(Equation 128)
Many of the details which are omitted in the preceding discussion can be
found in the theoretical books on ionization chambers by Attix
2
and Knoll
3
.
11. Conclud-
ing Remarks
The ion chamber converts flux (MUs/minute) through the ion chamber
into a more easily measurable quantity: a current signal. Factors affecting
this conversion process include:
! collector and polarizing plate spacing
! chamber saturation
! voltage regulation of the polarizing plates
! density of the plate substrate
! temperature and pressure (although to a lesser degree than the
above-mentioned factors).
Figure 5.9. Output Pulse Shape
V
D
n
o
ex
ed
----------- = decaying portion
V
max
n
o
e
c
-------- = maximum amplitude
V
R
V =
max
n e
c
o
t
t
+
t
V
elec
5-12 Ion Chamber Theory: References
In Varian's application, the current signal generated is a pulsed signal. The
basic equations for the pulse shape were presented (in terms of voltage in-
stead of current, Thevenin's theorem was used).
Discussions of manufacturing concerns include the effects of the inverse
square law governing radiation attenuation. The photon fluence level is 20
to 50 times more intense at the Ion chamber versus isocenter (because of
scatter and secondary electron fluence). Also, use of Teflon as a high volt-
age insulating material should be avoided.
Again, the primary objective of this report is to inform. If you require more
detailed information on any of the topics discussed, please consult the Ref-
erences list below, as well as the endnotes referenced in the text of this
chapter.
12. References H. Johns and J. Cunningham, The Physics of Radiology, Thomas, Illinois
(1983).
B. B. Rossi and H. H. Staub, Ionization Chambers and Counters, McGraw-
Hill, New York (1949).
D. H. Wilkinson, Ionization Chambers and Counters, Cambridge Univ.
Press, Cambridge (1950).
1. F. Attix, Introduction to Radiological Physics and Radiation Dosimetry, John Wiley & Sons,
New York (1949)
2. Ibid.
3. G. Knoll, Radiation Detection and Measurement, John Wiley & Sons, New York (1979)
Vacuum Theory 6-1
In order to understand the Clinac vacuum systems, a basic knowledge of vacuum theory
is required. This chapter will provide the reader with sufficient information to be able to
perform service and maintenance on these vacuum systems.
Vacuum Theory
6-2 Vacuum Theory
Table of Contents
1. Introduction: .................................................................................................................... 6-3
2. The Nature of Vacuum: .................................................................................................... 6-3
2.1. What Is Vacuum?: .................................................................................................... 6-3
2.2. What About Pressure?: ............................................................................................. 6-4
2.3. How Is a Vacuum Produced?: ................................................................................... 6-4
2.4. Different Types of Vacuum:....................................................................................... 6-4
2.5. Where Is Vacuum Used?: .......................................................................................... 6-5
2.6. Why Is Vacuum Needed?: ......................................................................................... 6-5
3. Temperature: ................................................................................................................... 6-6
4. Pressure:.......................................................................................................................... 6-7
4.1. What is Gas?: ........................................................................................................... 6-7
4.2. Atmospheric Pressure: .............................................................................................. 6-7
4.3. Pressure Measurement: ............................................................................................ 6-8
4.4. Partial Pressure: ....................................................................................................... 6-9
4.5. Vapor Pressure: ...................................................................................................... 6-10
4.6. Effects of Pressure: ................................................................................................. 6-12
4.7. Pressure Ranges: .................................................................................................... 6-12
5. Gas Particles: ................................................................................................................. 6-13
6. Gas Laws: ...................................................................................................................... 6-13
6.1. Avogadros Law: ...................................................................................................... 6-13
6.2. Boyles Law:............................................................................................................ 6-14
6.3. Gas Expansion: ...................................................................................................... 6-14
6.4. Charles Law: .......................................................................................................... 6-15
6.5. Gay-Lussacs Law: .................................................................................................. 6-16
6.6. General Gas Law: ................................................................................................... 6-16
7. Gas Flow: ....................................................................................................................... 6-17
7.1. Viscous Flow: ......................................................................................................... 6-17
7.2. Molecular Flow: ...................................................................................................... 6-17
7.3. Mean Free Path: ..................................................................................................... 6-18
8. Conductance:................................................................................................................. 6-18
8.1. Conductance in Viscous Flow: ................................................................................ 6-19
8.2. Conductance in Molecular Flow: ............................................................................. 6-20
9. Review of the Nature of Gases: ....................................................................................... 6-20
10. Ion Pump: .................................................................................................................... 6-21
10.1. Components: ........................................................................................................ 6-22
10.2. How the Pump Works: .......................................................................................... 6-22
10.3. Vacuum System Use: ............................................................................................ 6-26
10.4. Summary:............................................................................................................. 6-26
11. Vacuum Gauges:.......................................................................................................... 6-26
11.1. Thermocouple Gauge: ........................................................................................... 6-26
Vacuum Theory: Introduction 6-3
1. Introduction In the first part of this chapter, we will introduce you to vacuum:
! What it is
! How it relates to pressure
! How it is produced
! The different types of vacuum
! Where it is used
! Why we need it
You will also learn about temperature as a factor in vacuum work and the
types of pressure and how it is measured.
Finally, we will discuss some basic concepts used in vacuum work. These
are:
! The effects of pressure
! Pressure ranges in vacuum systems
! Some basic laws about the behavior of gases
! Some types of gas flow
! How we measure the work done by vacuum systems
2. The Nature
of Vacuum
The word vacuum comes from the Latin vacua, which means empty.
2.1. What Is
Vacuum?
Actually, vacuum is only partially empty space. In a vacuum, some air and
other gases have been removed from a contained volume. This volume is
usually called the work chamber. It separates the vacuum from the outside
world.
6-4 Vacuum Theory: The Nature of Vacuum
A more practical definition for vacuum is what exists in any contained vol-
ume where there is less gas than there is in the surrounding atmosphere.
We will see that these gases exert a force on the surface area of the contain-
er. This force is called pressure. We can measure the pressure in the cham-
ber by comparing it with the atmospheric pressure on the outside. In this
way, we can find out how much gas is left in the vacuum.
2.2. What
About Pres-
sure?
Pressure is defined as force per unit area. Gases are composed of small par-
ticles. These gas particles are in constant motion. As these particles move
around in space, they hit objects. When they hit something, they exert a
force, or pressure. We can take a unit of area and measure the number and
intensity of particle impacts on that surface. The result is a pressure mea-
surement.
2.3. How Is a
Vacuum Pro-
duced?
A vacuum is made by removing air and other gases from the work chamber.
We remove the air and other gases by using special pumps, called vacuum
pumps.
There are many, and very different, kinds of vacuum pumps. Some of them
actually remove the gases. Other pumps trap the gases or change their
form. In any case, the pumps job is to take as many gases out of circula-
tion as necessary.
2.4. Different
Types of Vac-
uum
There are different degrees of vacuum, called rough vacuum, high vacuum,
and ultrahigh vacuum. Which one is used depends on the application. As
the chambers below show, the better (or higher) the vacuum is, the less air
and gas are present.
Vacuum Theory: The Nature of Vacuum 6-5
2.5. Where Is
Vacuum Used?
Vacuum is used for many products and processes. Some of them are:
*as in Varian Clinac Linear Accelerators
2.6. Why Is
Vacuum
Needed?
We use a vacuum when we need a space that is very clean. It must be free
of gases that can interfere with what we want to do.
Let us take iron, for example. When iron is left out in air, it reacts with the
gases in the air, and the result is rust. This would not happen in a vacuum.
GOOD
ROUGH VACUUM
BETTER
HIGH VACUUM
BEST
ULTRA-HIGH VACUUM
Table 6.1. Uses of Vacuum
Rough Vacuum High Vacuum Ultrahigh Vacuum
Food processing Tube processing Space research
Evaporation Heat treating Materials research
Freeze drying Integrated circuit manufacture Metallurgy
Distillation Decorative coating Physics research
Sputtering Particle acceleration* Surface analysis
Electrical conduction
(neon lights)
Chemistry research Molecular beam epitaxy
E-beam welding
Vapor deposition
Ion implantation
Insulation (thermal)
RUSTY CLEAN
VACUUM SYSTEM WALL
6-6 Vacuum Theory: Temperature
Another example is television. If gases are not removed from a TV tube, the
electrons are blocked from reaching the screen no picture!
The easiest way to define clean is to say that everything is contaminated,
or dirty, to some degree. It is a matter of how much contamination is
present. The less contamination, the cleaner something is.
Let us look at some of this contamination that we are trying to remove. At-
mospheric air is a mixture of gases. Over 99% of atmosphere is nitrogen
and oxygen. All other gases make up less than 1%.
Water vapor, another common gas, is not listed above because the amount
changes with atmospheric pressure and temperature. Water vapor, which
varies from 0.6% to 6% by volume, is one of the biggest sources of vacuum
contamination, or dirt.
3. Temperature We have mentioned temperature already in our discussion. Most of us are
familiar with the Fahrenheit (F) and the Celsius or Centigrade (C) scales
of temperature measurement. In the world of vacuum, we are also con-
cerned with the absolute temperature as well.
Temperature is a qualitative measurement of energy. The hotter something
is, the more energy it contains. Or, if we want to get rid of gases, we could
pump the energy out of them until they become frozen. That is, we have
lowered the temperature of the gases.
Calculations of heat and energy do not work well in the Celsius and Fahr-
enheit scales because of the negative numbers. This is where the absolute
or Kelvin scale comes in. Let us compare some temperatures and conver-
sion factors.
Table 6.2
Gas Percent by Volume
Nitrogen 78.08
Oxygen 20.95
Argon 0.93
Carbon Dioxide 0.03
Neon 0.0018
Helium 0.0005
Krypton 0.0001
Hydrogen 0.00005
Xenon 0.0000087
Table 6.3. Temperature Scales
F C K Reference
212 100 373 Boiling point of water
32 0 273 Freezing point of water
321 196 77 LN2 temperature
437 261 12 Cold head temperature
459 273 0 Absolute zero
Vacuum Theory: Pressure 6-7
Conversion factors:
Now let us discuss some information about gases.
4. Pressure Earlier we defined pressure. Now, we will explain the kinds of pressure vac-
uum is concerned with. We will also describe how we measure pressure.
First, let us look at what a gas is.
4.1. What is
Gas?
What is a gas? It is a state of matter where the individual particles are free
to move in any direction and tend to expand uniformly to fill the confines of
a container. The gas particles are very small and freely moving. Some, like
hydrogen and oxygen, are very reactive and easily form stable chemical
compounds with other gases or elements. Other gases, such as helium and
argon, are inert. These are sometimes known as the noble (inert) gases.
They do not tend to form compounds.
All gases have mass and are thus attracted to the earth by the force of grav-
ity. This ocean of gas we call air has weight. This weight pushing on the
earths surface is called atmospheric pressure. By definition, pressure (P) is
the force (F) exerted on some particular area (A), such as a square inch,
square foot, or square centimeter. Put into mathematical terms,
(Pressure = Force per Unit Area)
At 45 N latitude and at sea level, the average pressure exerted on the
earths surface is 14.69 pounds per square inch (absolute), or 14.69 psia.
When the temperature is 0C, this 14.69 psia is called a standard atmo-
sphere (1 std atm). Gas behavior is usually described with reference to
standard conditions of temperature and pressure (stp).
4.2. Atmo-
spheric Pres-
sure
We use several different pressure scales. Here are four readings, all at stan-
dard conditions:
14.7 psia = 760 torr = 1 std atm = 101,325 pascal
The average atmospheric pressure at sea level (45 N latitude) is 14.7 psia,
760 torr, or 101,325 Pa. Vacuum processes are usually done at pressures
much lower than atmospheric pressure.
Atmospheric pressure changes with distance above sea level (altitude) and
changes in our weather.
C
5
9
--- F 32 ( ) = K C 273 + =
F
9
5
---C

32 + = K
5
9
--- F 32 ( ) 273 + =
P
F
A
--- =
6-8 Vacuum Theory: Pressure
A way to measure the force exerted by the atmosphere was developed in the
mid-1600s by Evangelista Torricelli. It consisted of balancing a fluid of
known weight against the weight of air. The first fluid used was water. Lat-
er, mercury was used. The measurement was made using an instrument
called a barometer. We have named a pressure unit, torr, in Torricellis
honor.
4.3. Pressure
Measurement
There are several different scales for pressure measurement. Millimeters of
mercury, torr, and microns are all commonly used. Pascal (Pa) is the metric
unit for pressure measurement and is the international standard.
The following table shows some of the common scales. The values for these
scales are all listed at the same pressure one standard atmosphere (1 std
atm).
Table 6.4. Average Pressure at Various Altitudes
Altitude
(Ft)
Pressure
(Torr)
Altitude
(Ft)
Pressure
(Torr)
1,000
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
787.87
760.00
732.93
706.66
681.15
656.40
632.38
609.09
586.49
564.58
543.34
522.75
11,000
12,000
15,000
20,000
25,000
30,000
40,000
50,000
60,000
100,000
120,000
140,000
502.80
483.48
429.08
314.51
282.40
226.13
141.18
87.497
54.236
8.356
3.446
1.508
Source: U.S. Standard Atmosphere, 1962 (NASA)
Vacuum
Water
1 Atm
Vacuum
Mercury
1 Atm
33.9 ft
406.8 in
760 mm.
30 in.
1 in of water = 0.36 lb.
Weight of water = 406.8 in. 0.36 = 14.69 lb.
3
3
Note: Mercury is 13.56 times heavier than
water, so the mercury barometer will
be 13.56 times shorter;
i. e., 406.8 in/13.56 = 30 in.
The Barometer
Here is a table for the equivalent values for one torr and one millitorr
(mtorr).
A conversion table and equivalents for the different measurement scales
are provided in the Appendix.
4.4. Partial
Pressure
The total pressure of a mixture of gases is the sum of each of the individual
gas pressures in the mixture. This is known as Daltons Law of Partial Pres-
sure. Each individual gas pressure in a mixture is called a partial pressure.
At standard conditions (760 torr, 0C), each gas exerts a pressure relative
to its percent of the total volume: for example:
N2 = 78% = 0.78 760 = 593 torr.
Table 6.5. Pressure Equivalents
Atmospheric Pressure (Standard)
0
14.7
760
760
760,000
101,325
1.013
1013
psig (gauge pressure)
pounds per square inch (psia)
mm of mercury
torr
millitorr or microns
pascal
bar
millibar
Table 6.6
One Torr = One Millitorr =
1/760 atmosphere 1/1000 TORR
1 mm of mercury 1/1000 mm of mercury
1000 microns or millitorr 10
-3
torr or 1 millitor
10
3
microns or millitorr 0.001 torr
133 Pascal 0.133 Pascal
Table 6.7. Partial Pressures of Gases Corresponding
to Their Relative Volumes
Gas (Air) Symbol
Percent by
Volume
Partial Pressure
Torr Pascal
Nitrogen N
2
78 593 79,000
Oxygen O
2
21 159 21,000
Argon Ar 0.93 7.1 940
Carbon Dioxide CO
2
0.03 0.25 33
Neon Ne 0.0018 1.4 10
2
1.8
Helium He 0.0005 4.0 10
3
5.3 10
Krypton Kr 0.0001 8.7 10
4
1.1 10
1
Hydrogen H
2
0.00005 4.0 10
4
5.1 10
2
Xenon Xe 0.0000087 6.6 10
5
8.8 10
3
Water H
2
O Variable (5 to 50 torr
typically)
Variable
4.5. Vapor
Pressure
When a liquid or solid becomes a gas, we call that process evaporation. The
gas produced, we call a vapor. It, of course, exerts a pressure. This pres-
sure we refer to as the vapor pressure for that particular material. The act
of turning the gas back into a liquid, we call condensation. When a solid
evaporates to a gas directly, we call that process sublimation.
In general usage, vapors are gases that tend to condense back to the liquid
state at moderate temperatures and pressures. All substances have a char-
acteristic saturation vapor pressure that varies directly with temperature.
The lower the temperature, the lower the vapor pressure. This is true for all
substances.
Water deserves special attention because of its behavior in the vacuum sys-
tem. It is present in air as a gas in relatively large quantities. In the vacuum
system, it is hard to remove condensed water vapor from surfaces at room
temperatures.
Acetone has the highest vapor pressure of the liquids on this list. It evapo-
rates the fastest of those substances on the list. It releases the most gas
into the chamber in a given length of time. High vacuum pump oil is the
least volatile liquid on the list. It will take the longest time to evaporate.
When gases become cooled sufficiently, they liquefy and/or freeze. These
curves give the vapor pressure for selected gases when they are liquids or
solids. In the illustration below, curves to the right of the vertical dotted
line (77K, 196C) indicate low vapor pressures at this temperature. Curves
to the left show high vapor pressures at this temperature, which is the boil-
ing point of liquid nitrogen.
Table 6.8. Vapor Pressure of Water at Various Temperatures
Temperature in C Pressure in Torr
100.0 (Boiling) 760
50.0 93
25.0 24
0.0 (Freezing) 4.8
40.0 0.1
78.5 (Dry Ice) 5.0 10
4
196.0 (LN
2
) 1.0 10
24
Table 6.9. Vapor Pressures of Some Liquids
Liquid
Vapor Pressure in Torr
@ 20C (68F)
Benzene 74.6
Ethyl Alcohol 43.9
Methyl Alcohol 96.0
Acetone 184.8
Turpentine 4.4
Water 17.5
Carbon Tetrachloride 91.0
High Vacuum Pump Oil 1.0 0
7
Gases at the left side of the chart have high vapor pressures at extremely
low temperatures.
Note: Vapor pressure of all gases is the same at the boiling
point in atmosphere (760 torr) even though they boil at differ-
ent temperatures.
All materials have a vapor pressure, even though it may be very small. Note
that, for some of these materials, their vapor pressure may be high enough
to be a problem in some vacuum systems.
(BOILING POINT)
V
A
P
O
R

P
R
E
S
S
U
R
E

(
T
O
R
R
)
TEMPERATURE (K) E
VAPOR PRESSURES OF COMMON GASES
272 270 260 250 200 100 100 0
1 2 3 4 5 10 20 30 4050 100 200 400
10
11
10
10
10
9
10
8
10
7
10
6
10
5
10
4
10
3
10
2
10
1
10
0
10
1
10
2
He
H
2
Ne
N
2
O
2
NO
CO
2
H O
2
10
3
TEMPERATURE (C)
Cd
Zn
Pb
Cu
Ti
Fe
Mo
W
2
0
0
4
0
0
6
0
0
8
0
0
1
0
0
0
1
5
0
0
2
0
0
0
3
0
0
0
4
0
0
0
5
0
0
0
V
A
P
O
R

P
R
E
S
S
U
R
E

(
T
O
R
R
)
10
9
10
7
10
5
10
3
10
1
10
1
10
3
10
11
ABSOLUTE TEMPERATURE (K)
4.6. Effects of
Pressure
Before the air and other gases are pumped from the work chamber, con-
stant, high-speed motion makes the particles bump into each other and
into the chamber walls. This activity develops a total actual (absolute) pres-
sure of 14.7 pounds per square inch (psia). As we have already seen, 14.7
psia is the average atmospheric pressure at sea level. Therefore, the pres-
sure is the same inside and outside the chamber.
As air is pumped out of the chamber, pressure drops. However, we can nev-
er remove all particles from the chamber.
After most of the free-moving gas (sometimes called the volume gas) is re-
moved, there are still other sources of gas entering the system. Gases come
off of surfaces in the vacuum system or out of the materials inside the work
chamber. This is called desorption or outgassing.
Vacuum systems can implode because of the external atmospheric pres-
sure, causing the walls to collapse inward.
4.7. Pressure
Ranges
These are the pressure ranges generally used in vacuum work:
Rough (low) vacuum759 to 1 10
3
torr (approx.)
High vacuum1 10
3
torr to 1 10
8
torr (approx.)
Ultrahigh vacuumLess than 1 10
8
torr
Vacuum Theory: Gas Particles 6-13
5. Gas Parti-
cles
Lets talk about the nature of the gases that exert this pressure. They are
made from naturally occurring chemical elements. These elements are the
building blocks of earthly matter. The smallest identifiable part of an ele-
ment is one of its atoms.
An atom has a dense center portion known as the nucleus. This nucleus
has particles called protons and neutrons. The protons have a positive
electrical charge. Neutrons are neutral. The number of protons and
therefore the electrical charge in the nucleus is different for each ele-
ment. If the atom has more or less than its normal number of neutrons, it
is called an isotope of the element and is unstable.
Under normal conditions, the nucleus is surrounded by a number of elec-
trons. Electrons have a negative electrical charge. The number of elec-
trons balances the positive charge and this makes the atom electrically
neutral.
Neutrons and protons weigh approximately the same and make up the
bulk of the atom. The atoms of the different elements have different num-
bers of protons and neutrons. They thus have different masses. This
means they have different weights (masses). They are classified by their
atomic mass or weight. We call this atomic mass units or amu.
Molecules simply consist of one or more atoms joined together. with defi-
nite chemical and physical characteristics.
Molecules are likewise classified by their molecular weight (or mass). This
is simply the sum total of the individual atomic weights that make up the
molecule. Some of the elements usually exist as gases. Some of these, like
hydrogen, nitrogen and oxygen, travel as molecules with two or three
atoms bound together. Some gases are composed of more than one ele-
ment, such as water (H
2
O). For instance, the atomic weight of hydrogen
(H) is 1 amu. Its molecule is made up of two hydrogen atoms (H
2
) so its
Molecular weight or mass is 2 amu.
The atomic weight of oxygen is 16 amu. Thus, the molecular weight of
water (H
2
O) is 18 amu. That is the mass of two hydrogen atoms plus the
mass of one oxygen atom (1 + 1 + 16).
Under certain conditions, an atom or a molecule can become electrically
charged. It is then referred to as an ion. This process will be considered in
more detail in the discussion on Ionization.
6. Gas Laws Lets look at what happens to gases as we use them in our vacuum sys-
tem. We first assume that gases are perfect and in general, they are. So
we can apply some laws to their behavior Lets look at some of these
laws.
6.1. Avo-
gadros Law
Under the same conditions of pressure and temperature, equal volumes of
all gases have the same number of particles (molecules, actually). We call
this a mole. One mole of any gas has 6.023 1023 particles, under stan-
dard conditions (760 torr, 273K), occupies 22.4 liters, and weighs one
molecular weight.
We know this as Avogadros Law.
1. How many particles would be in a standard liter?
2. How many in a standard cubic centimeter?
6.023 10
23
particles
22.4l
--------------------------------------------------- 2.69 10
22
particles l =
6.023 10
23
particles
22.4l
---------------------------------------------------
10
3
l
1cc
------------- 2.69 10
25
particles/cc = =
6-14 Vacuum Theory: Gas Laws
6.2. Boyles
Law
Boyles Law, P
1
V
1
= P
2
V
2
, or original pressure times original volume equals
new pressure times new volume. Reduce the volume by half, the pressure
is doubled. This equation predicts new pressure or new volume whenever
the other is changed by any amount, providing that the temperature re-
mains the same.
6.3. Gas Ex-
pansion
Gas expands tremendously under vacuum (from Boyles Law). This hap-
pens to gas absorbed in fingerprints and dirt in general.
BOYLES LAW
P V = P V
1 1 2 2
VOLUME =
10 LITERS
PRESSURE
= 50 TORR
VOLUME =
5 LITERS
PRESSURE
= 100 TORR
SAME
NUMBER
OF GAS
MOLECULES
NOTE: TEMPERATURE HELD CONSTANT
Vacuum Theory: Gas Laws 6-15
Water and solvents are also sources of large gas loads. The large volumes
these materials produce are a major part of outgassing.
Suppose you have a chamber which has a volume of 100l at a pressure of
1 10
4
torr. If 1 std cc of gas is suddenly added, what will be the pressure?
Lets use Boyles Law.
Note that we are really calculating a new pressure, not a new volume. Also,
the partial pressure of the gas we are adding will add to the gas pressure al-
ready there.
P
1
V
1
= P
2
V
2
For the gas we are adding to the chamber:
760 torr 1 cc = P
2
100l
Solving for P2 and converting cubic centimeters to liters:
Now the total pressure in the container is the sum of the pressure there (1
10
4
torr) plus the pressure from the gas we added (7.6 10
3
torr).
We see that the 1 cc of gas at atmospheric pressure contributed much more
to the pressure in the chamber than the gas already there!
6.4. Charles
Law
Lets look at what happens to the volume of gas as we change the temper-
ature. As we cool a gas, its volume gets smaller. If we heat the gas, its vol-
ume increases. We call this Charles Law. The equation looks like this:
Charles Law states that if the absolute temperature is doubled, the volume
of gas is doubled providing that the pressure is unchanged.
P
2
760 torr 1cc
100l
----------------------------------
10
3
l
1cc
------------- =
P
2
7.6 10
3
torr =
P
total
P
chamber
P
2
+ =
1 10
4
torr 7.6 10
3
torr + =
7.7 10
3
torr =
V
1
T
1
------
V
2
T
2
------ =
6-16 Vacuum Theory: Gas Laws
6.5. Gay-Lus-
sacs Law
If Charles Law is examined carefully, a more specific relationship develops:
If the temperature of a volume of gas at 0C is changed by 1C, the volume
will change (plus or minus) by 1/273 of its original value. This is Gay-Lus-
sacs Law. Thus:
Rearranging this equation gives us:
Lord Kelvin used this relationship to develop the absolute temperature
scale.
6.6. General
Gas Law
We can combine these laws to get a general gas law (Boyles and Charles
combined):
The general gas law combines pressure, volume, and temperature in a sin-
gle equation.
The temperature in Charles Law and the general gas law is
expressed in the absolute scale, or degrees Kelvin; to convert
from C to K add 273 to C. Thus: 100C + 273 = 373K.
VOLUME =
10 LITERS
TEMPERATURE
= 100 K
TEMPERATURE
= 200 K
NOTE: PRESSURE HELD CONSTANT
VOLUME =
20 LITERS
HEAT
1 ATM 1 ATM
AMOUNT OF GAS DOES NOT CHANGE
CHARLES LAW
V = V
T = T
1 2
1 2
SAME
NUMBER
OF GAS
MOLECULES
V V
o
C
273
-------- -

V
o
+ =
V V
o
1
C
273
-------- - +

=
P
1
V
1
T
1
------------
P
2
V
2
T
2
------------ =
Vacuum Theory: Gas Flow 6-17
7. Gas Flow Since we want to move gas molecules out of the vacuum chamber, we
should know how gas flows. Of the many types of gas flow, we will discuss
two kinds: viscous flow and molecular flow. Both types of flow have to do
with how tightly molecules fill a space.
7.1. Viscous
Flow
Generally, gas molecules occupying a space at a pressure greater than
1 10
2
torr act very much like a fluid, so this is called viscous flow. In the
viscous flow range, the molecules are constantly bumping into each other.
The molecules are so closely packed together that as our vacuum pump
moves some of them out of the chamber. others will rush to fill up that
empty space.
In viscous flow conditions, molecular movement is predictable. When a
molecule is hit or hits a surface, we can predict its movement after impact
with reasonable accuracy.
Because the molecules are tightly packed and move predictably, we can use
smaller diameter hoses and tubulations for rough pumping operations.
Viscous flow conditions will generally allow us to move great quantities of
molecules per unit time from one place to another.
7.2. Molecular
Flow
Molecular flow occurs when the molecules are so far apart that they no
longer have any influence on each other. Their motion is strictly random.
This occurs at low pressures where fewer molecules are present.
Depending on the pressure, a gas molecule might travel inches, feet, or
even miles before it strikes another molecule. This means we cant depend
on molecular interaction to push or start a flow pattern.
In the molecular flow range, molecular movement is unpredictable. This is
why we have such large inlets in high vacuum pumps.
The use of large inlets increases the probability that one of these randomly
moving molecules will move into the pump.
6-18 Vacuum Theory: Conductance
In molecular flow, the molecular motion is explained by kinetic theory,
which uses statistics (chance) to describe the condition.
The difference between viscous flow and molecular flow does not depend
upon the pressure alone. It also depends upon the dimensions of the vacu-
um container (pipes, chamber, etc.). Basically, it depends upon the mean
free path and whether it is longer or shorter than the container dimensions.
Lets take a look at what is meant by mean free path.
7.3. Mean Free
Path
As we lower the pressure in the vacuum chamber, the amount of space be-
tween the gas particles increases. The particles bump into each other less
frequently. The average distance a particle moves before it bumps another
particle is the mean free path.
At atmosphere, the mean free path is extremely short, about two millionths
of an inch. Under vacuum, fewer molecules remain, and the mean free path
is longer. Its length depends on the number of molecules present, and
therefore on the pressure. The mean free path for air can be estimated from
the relationship:
From this, we can see that as the pressure gets lower, the mean free path
gets longer. Likewise, as the pressure gets lower, there are fewer molecules
of gas present, so there is less chance of them running into each other.
In 1 cc of gas at standard conditions (760 torr at 0C), there are about
3 10
19
gas molecules and the mean free path is about 2 10
6
cm (a few
millionths of an inch). At 1 10
9
torr, there are about 4 10
7

molecules/cc, and the mean free path is about 30 miles or 50 kilometers.
The number of molecules per unit volume (in this example cubic centime-
ters) is called the gas density.
8. Conduc-
tance
When we talk about moving a gas through a vacuum system, we use the
term conductance. Conductance is the ability of an opening or pipe to allow
a given volume of gas to pass through in a given time. It is expressed in
such units as liters per second, cubic feet per minute or cubic meters per
hour.
In molecular flow, a good conductance path is wide and short. It has few
turns, thus allowing free gas flow. In viscous flow, these conditions are not
so important. This is because the molecules tend to push one another
along under the influence of a pressure difference.
Table 6.10. Molecular Density and Mean Free Path
7.6 10
2
Torr (atm) 1 10
3
Torr 1 10
9
Torr
# mol/cm3 3 10
19
(30 million trillion)
4 10
13
(40 trillion)
4 10
7
(40 million)
MFP 2 10
6
in. 2 in. 30 mi.
Mean Free Path 5 10
3
torr cm = ( )
P
torr
------------------------------------------------------------------------------------------
Vacuum Theory: Conductance 6-19
In the molecular flow range, a 1-in
2
opening has a 75 l/sec conductance.
The pump speed, in this case 400 l/sec, is really 75 l/sec as far as the
chamber is concerned because the molecules must go through the hole be-
fore they can be pumped. To improve system performance, the conduc-
tance must first be improved. (Make the hole bigger!)
To repeat: In the molecular flow range, a pump works only when molecules
migrate into the pump by chance.
8.1. Conduc-
tance in Vis-
cous Flow
The volume of gas that can flow per unit of time through a pipe under vis-
cous flow conditions is related to the fourth power of the pipe diameter and
is inversely related to the length of the pipe.
For example, if you use a pipe with a diameter twice that of the pipe pres-
ently being used, it will allow 2
4
or sixteen times as much gas to flow
through it, assuming that the length of the pipe is the same.
Now lets compare this to molecular flow conditions.
CHAMBER
400 /SEC PUMP l
1 IN.
OPENING (75 /SEC)
OR 11.6 /SEC CM
2
2
l
l
MOLECULAR FLOW
6-20 Vacuum Theory: Review of the Nature of Gases
8.2. Conduc-
tance in Molec-
ular Flow
The volume of gas that can flow per unit of time through a pipe under mo-
lecular flow conditions is related to the cube of the diameter and inversely
to the length of the pipe.
Using the same pipe as in the viscous flow example, doubling the diameter
of the pipe will, at most, allow 2
3
or eight times the flow for the same length
of pipe.
In either case of viscous flow or molecular flow, making the pipe shorter will
increase the flow of gas through the pipe. Please note that these are gross
statements that are subject to all kinds of qualifications.
There is another region where we approach molecular flow, but the flow is
not really viscous either. This region is called the transition range. There is
another set of calculations to be used for the transition range, but we will
not discuss them in this text.
9. Review of
the Nature of
Gases
Before continuing, a few basic facts about the atomic and molecular nature
of gases should be reviewed. These facts will be useful for understanding
how ion pumps operate. Let us review them briefly here.
An atom is the smallest particle of matter that can exist and still retain the
basic characteristics of the material or element from which it came. Mole-
cules are simply one atom or two or more atoms joined together; many
gases exist as molecules.
Atoms and molecules normally have an equal number of protons (positively
charged particles) and electrons (negatively charged particles). The neu-
trons in the nucleus contribute to the weight (mass) of the atom, but not to
the charge. The atoms are thus neutral, or electrically balanced. If this bal-
ance is upset, useful work can be produced. If we remove electrons from
the atom, we have made a positively charged atom or molecule we call an
ion. This process of creating ions is called ionization. We can put these oth-
erwise useless charged particles to work because we can direct their motion
using a magnetic or electrical field.
Vacuum Theory: Ion Pump 6-21
10. Ion Pump Let us now make our ion pump by connecting two electrodes to a high-volt-
age supply. Electron flow will be from cathode to anode as in this drawing.
Ions will carry current from anode to cathode. Fewer ions than electrons
will be produced so that we can say that the current through the pump is
the ion current.
In this drawing, a free electron is attracted to a positively charged anode.
On the way to the anode, it collides with a neutral atom, ionizing it. Now
two electrons are free to continue toward the anode, increasing the proba-
bility of still further ionization. The positively charged ions are then accel-
erated toward the negatively charged cathode. They may strike the cathode
with such force that they stick to the cathode material, and are thereby
pumped. As one gas molecule is driven into the cathode, one or more mol-
ecules of the cathode is usually released from this surface. This process is
called sputtering.
The ion pump is also a gas capture pump. It is not designed to pump heavy
gas loads. For this reason, it is not generally used alone in high-production
applications. Instead, it is more often used in research and analytical ap-
plications where there is no need to repeatedly and rapidly cycle the work
chamber to atmosphere. When combined with a Titanium Sublimation
Pump (TSP), it also provides adequate pumping for these applications.
Ion pumps are clean operating devices. They are electronic devices which
use no moving parts or oils. It is possible to achieve pressures in the 10
11

torr range, with overnight bakeout of the system. The bakeout process
drives residual gas off walls. This gas is then pumped by the ion pumps.
In research and analytical applications, the ion pumps cleanliness, bake-
ability, low power consumption, vibration-free operation and long life make
it the pump of choice for most ultra-high vacuum uses.
Ion pumps come in various sizes. A small appendage ion pump is used not
for pumping down, but for maintaining vacuum conditions in operating de-
vices such as transmitting tubes.
Larger pumps can be used to evacuate small chambers, or several can be
connected in parallel with other ion pumps to pump down larger chambers.
ELECTRON
SOURCE
CATHODE ANODE
( + ) ( )
( + ) ( )
POWER
SUPPLY
6-22 Vacuum Theory: Ion Pump
10.1. Compo-
nents
A basic ion pump cell consists of two titanium cathodes and an anode. All
are placed between the poles of a strong permanent magnet.
10.2. How the
Pump Works
10.2.1. Pump
Operation
The magnetic field forces the free electrons to travel in long helical paths in-
stead of straight lines. This increases the probability of collision with mol-
ecules on their way to the positively charged anode. This, in turn, increases
the ionization probability, and therefore the amount of useful pumping ac-
tion that can be performed by the pump.
Because of the action of the magnetic field, the electrons do not easily come
in contact with the anode. As a result, a cloud of electrons is formed in the
anode area. This electron cloud becomes fairly stable during pump opera-
tion. The electron density is high enough for efficient ionization of gas mol-
ecules. Therefore, a hot filament electron source is not needed. So, the
name for this process is cold cathode discharge.
The positively charged ions, which are relatively heavy particles, are accel-
erated into the negatively charged titanium cathodes. This impact causes
sputtering, or chipping away of the titanium cathode material.
Sputtered titanium deposits onto the internal structure of the pump. There
it is available for chemical combination with gas molecules to convert them
to solids. Thus we have the needed pumping action.
ANODE
CATHODE
ELECTRON
MAGNET
AN ELECTRON IS

BROKEN
FREE FROM A GAS
MOLECULE. THE GAS
MOLECULE THEN BECOMES
A POSITIVE ION.
NEUTRAL GAS
MOLECULE
In addition, a second pumping action takes place. Some of the ionized mol-
ecules strike the cathodes with enough force to become buried in them.
This burial prevents them from recombining and becoming a free gas again.
Still another pumping process occurs in the case of hydrogen, which diffus-
es directly into and reacts with the cathode plate. Also, neutral molecules
in the anode regions can literally be buried or plastered over by the sput-
tered cathode material. Complex molecules may also be split in the dis-
charge to smaller, more readily pumped molecules.
ANODE
TITANIUM
CATHODES
SPUTTERED ATOM
FROM CATHODE
MAGNET
POSITIVE ION
MAGNET
( ) ( )
( + )
6-24 Vacuum Theory: Ion Pump
There is a problem with the pump design we have described (also called a
diode configuration). Some of the buried molecules can be released again
into the vacuum system. This re-release can be caused by heating of the
cathodes or reduction of cathode material due to sputtering. It can also be
caused by a molecule or atom being physically separated from the sput-
tered film.
10.2.2. Pump-
ing Character-
istics of
Different Con-
figurations
Ion pumps are available in different design configurations. Each design has
its own special pumping characteristics.
In the diode pump, as we have seen, the ions strike the cathode plate and
react with the sputtered titanium.
The triode pump, which is a variation on the diode pump, improves inert or
noble gas pumping.
Titanium cathodes are in the form of grids. Ions sputter titanium onto the
pump walls. This angled impact sputters more titanium than in the diode
model and thus furnishes more material for argon or noble gas burial. Be-
cause of the electrical arrangement of the pump components, the glow dis-
charge that happens in starting the diode pump is typically confined in
the triode pump. As a result, the triode pump can be started at slightly
higher pressure.
TITANIUM
CATHODE
PLATES
CONTROL UNIT
CONTROL UNIT
MAGNET
MAGNET
MULTICELL ANODE
MULTICELL ANODE
PUMP WALL FORMS
THIRD ELECTRODE
IN TRIODE PUMP
SPUTTER CATHODES
TITANIUM VANES
ION PUMP SCHEMATICS
S N
S N
(B) TRIODE ION PUMP (A) DIODE ION PUMP
MAGNETIC
FIELD
MAGNETIC
FIELD
ANODE
ANODE
TITANIUM ATOMS
S
P
U
T
T
E
R
I
N
G
TI SPUTTER
CATHODE
OBLIQUE IMPACT CAUSES
MAXIMUM SCATTERING
ENTRAPMENT OF
BURIED ARGON IONS
ARGON IONS
(B) TRIODE ION PUMP (A) DIODE ION PUMP
PUMPING MECHANISMS
SPUTTERED
TITANIUM,
ARGON ATOMS
BURIED HERE
PUMP WALL
ARGON
V+
+
VB
+ +
10.2.3. Other
Characteris-
tics
The ion pump is self-regulating. At the higher pressures, where much ion-
ization takes place, more current flows. At low pressures, less current
flows. This characteristic current drain can be used to measure the pres-
sure, or degree of vacuum achieved with the pump. This feature eliminates
the need for an ion gauge on the system.
Ion pumps are long-lived; the lower the pressure, the longer the life. Once
they begin pumping, they quickly lower the pressure to the long-life region.
As long as they are not pumping against a leak, they will last for years. Ide-
ally, ion pumps should be started at pressures approaching 10
5
torr. At
higher pressures, the plasma discharge that is generated minimizes pump-
ing speed and reduces cathode life. A more common and practical ap-
proach is to sorption rough the pump to less than 10
2
torr before applying
the ion pump power. At very low pressures, the time taken to begin the ion-
ization process may be excessively long.
Table 6.11. Typical Diode Pump Service Life
Pressure (Torr) Life (Hours)
10
3
20
10
4
200
10
5
2,000
10
7
200,000
(over 20 years of
constant operation)
Table 6.12. Life (Pumping N2 at 10
4
Torr)
Type Life
Triode 35,000 hours approx. 4 years
Diode 50,000 hours approx. 6 years
PUMP CURRENT
P
R
E
S
S
U
R
E

(
T
O
R
R
)
10
2
10
3
10
4
10
5
10
6
10
7
10
8
10
9
1A 10A 100A 1mA 10mA 100mA 1Amp
6-26 Vacuum Theory: Vacuum Gauges
10.3. Vacuum
System Use
Ion pumps are typically used in systems which demand ultra-clean, ultra-
high vacuum. This type of vacuum system is pumped to high vacuum or
lower pressure and then kept in that condition for long periods of time. A
load-lock chamber is often built on the system to allow access to the cham-
ber without bringing the chamber back to air.
Typical uses are for electron microscopes, mass spectrometers, and surface
analysis, to mention a few.
Very little maintenance can be performed on ion pumps other than an oc-
casional bakeout. When pumping eventually deteriorates to the point
where operating pressures can no longer be attained, pump replacement or
sometimes anode/cathode assembly replacement is necessary.
10.4. Summary We have discussed the pressure ranges of vacuum pumps and the major
types of pumps in each range. By now, you should be familiar with the dif-
ferent types of vacuum pumps what their major components are and how
they work. You have also learned how they are placed in vacuum systems
and some general maintenance information.
Lets go on now to gauges. These are major vacuum components that tell
you what is going on inside your vacuum system.
11. Vacuum
Gauges
To transfer heat by convection, we need massive numbers of molecules
flowing. Your hot-air furnace heats by convection. Some gauges use this
principle between 760 torr and 2 torr but are generally less accurate in this
pressure range.
To transfer heat by radiation, we need light energy. Not the kind of light
that you see, but typically infrared light. The heat you feel when standing in
front of a fireplace is mostly the radiated heat. No gas molecules need be in-
volved; that is, radiation is independent of the number of gas molecules
present. Radiated heat is the only way to transfer heat inside of a vacuum
system at high vacuum. There are insufficient molecules present to provide
heat transfer by either conduction or convection. Now, lets go on to discuss
gauges that depend on heat transfer to work.
11.1. Thermo-
couple Gauge
Vacuum Theory: Vacuum Gauges 6-27
The thermocouple, or TC, gauge is another rugged, simple instrument. It is
used to measure pressures in the rough vacuum range. It does its work
well under less than ideal conditions. The TC gauge measures temperature
and converts it to a pressure reading. Many modern thermocouple gauges
have been modified to use convection as well as conduction principles. This
effectively extends their useful range to atmosphere. It is typically consid-
ered as a very approximate device. Lets take a look at how it works.
11.1.1. How
the Gauge
Works
A thermocouple gauge consists of a gauge tube and control unit. Within the
gauge tube is a heated filament. Spot welded to the filament is a thermo-
couple that measures the temperature of the hot wire. The meter is cali-
brated in pressure units, not in temperature.
At atmospheric pressure, there will be many molecular collisions with the
heated filament. The gas molecules conduct heat away from the filament. The
amount of heat removal can be related to the amount of gas in the chamber.
At higher pressures, with lots of molecules, much heat will be conducted
away from the wire. Therefore, the wire will be at a lower temperature (cooler).
When we pump away the gas, there are fewer molecules to collide with the
wire. The wire is therefore at a higher temperature (hotter).
TO VACUUM
SYSTEM
TC GAUGE
TUBE
FILAMENT
THERMOCOUPLE
CONTROL UNIT
FILAMENT
THERMOCOUPLE
METER
(+) ()
THERMOCOUPLE GAUGE PRINCIPLE
6-28 Vacuum Theory: Vacuum Gauges
There is not a linear relationship between wire temperature and pressure,
so the pressure scale on your TC gauge is not linear. The gauge stops re-
sponding at about 1 millitorr (10
3
torr) because the heat loss through ra-
diation is now the largest factor. The heat lost through radiation is also
constant. Therefore, the gauge reads zero. Compared to other gauges, the
TC gauge has a slow response time. This is because the wire must have
time to heat up or cool down as the pressure changes.
Some newer gauges speed up the response time by operating the gauge at
constant temperature and measuring the change in current required to
hold the temperature constant.
11.1.2. Mainte-
nance
If the sensing unit, or gauge head, gets dirty, it may be cleaned with an ap-
propriate solvent. Most people will simply discard the TC gauge and install
a new one in its place.
Whenever you clean or replace a TC gauge, it should be adjusted to read
the proper values. To do this, you expose the gauge head to a pressure of
10
4
torr or less and adjust the control unit to read zero on the pressure
gauge. If for some reason you cannot obtain a pressure below 10
4
torr,
then install a good gauge and set the system gauge to read the same pres-
sure.
Please check the operation manual for your particular unit for adjustment
instructions, because they do vary in detail.
Glossary 7-1
This chapter is a glossary of terms commonly used in radiotherapy, and is intended for
reference use only.
Glossary
7-2 Glossary
Table of Contents
A:......................................................................................................................................... 7-3
B: ........................................................................................................................................ 7-5
C: ........................................................................................................................................ 7-8
D: ...................................................................................................................................... 7-12
E:....................................................................................................................................... 7-15
F: ....................................................................................................................................... 7-18
G: ...................................................................................................................................... 7-19
H: ...................................................................................................................................... 7-21
I:........................................................................................................................................ 7-22
J: ....................................................................................................................................... 7-24
K:....................................................................................................................................... 7-24
L: ....................................................................................................................................... 7-25
M: ...................................................................................................................................... 7-25
N: ...................................................................................................................................... 7-27
O: ...................................................................................................................................... 7-28
P: ....................................................................................................................................... 7-29
Q: ...................................................................................................................................... 7-31
R:....................................................................................................................................... 7-32
S:....................................................................................................................................... 7-35
T: ....................................................................................................................................... 7-38
U: ...................................................................................................................................... 7-40
V:....................................................................................................................................... 7-40
W: ...................................................................................................................................... 7-41
X:....................................................................................................................................... 7-42
Glossary: Absorbed Dose 7-3
Absorbed Dose See Dose, Absorbed Dose.
Absorbed Frac-
tion
A term used in internal dosimetry. It is that fraction of the photon energy
(emitted within a specified volume of material) which is absorbed by the
volume. The absorbed fraction depends on the source distribution, the
photon energy, and the size, shape, and composition of the volume.
Absorption The process by which radiation imparts some or all of its energy to any ma-
terial through which it passes. (See also Compton Effect, Pair Production
and Photoelectric Effect.)
Absorption Coefficient: Fractional decrease in the intensity of a beam of x
or gamma radiation per unit thickness (linear absorption coefficient), per
unit mass (mass absorption coefficient), or per atom (atomic absorption co-
efficient) of absorber, due to deposition of energy in the absorber. The total
absorption coefficient is the sum of the individual energy absorption pro-
cess (Compton effect, photoelectric effect, and pair production).
Linear Absorption Coefficient: A factor expressing the fraction of a beam
of x or gamma radiation absorbed in unit thickness of material. In the ex-
pression , I
0
is the initial intensity, I the intensity of the material
x, and is the linear absorption coefficient.
Mass Absorption Coefficient: The linear absorption coefficient per cm. di-
vided by the density of the absorber in grams per cu. cm. It is frequently ex-
pressed as /p, where is the linear absorption coefficient and p the ab-
sorber density.
Self-Absorption: Absorption of radiation (emitted by radioactive atoms) by
the material in which the atoms are located; in particular, the absorption of
radiation within a sample being assayed.
Accelerator A machine that accelerates electrically charged atomic particles to high ve-
locities. Electrons, protons, deuterons, and alpha particles can be acceler-
ated to nearly the speed of light for use in nuclear research. Types of accel-
erators include the betatron, cyclotron, linear accelerator, and
synchrotron.
Achromatic A type of bend magnet in which particles of differing energies are brought to
the same focus.
Actinic A type of radiation that is capable of producing a chemical change.
Activation The process of inducing radioactivity by irradiation.
Activity The number of nuclear transformations occurring in a given quantity of
material per unit time.
Adsorption The adhesion of one substance to the surface of another.
Alpha Particle A charged particle emitted from the nucleus of an atom having a mass and
charge equal in magnitude to those of a helium nucleus, i.e., two protons
and two neutrons. Aluminum Equivalent: The thickness of aluminum af-
fording the same attenuation, under specified conditions, as the material in
question.
I I
0
e
x
=
7-4 Glossary: Alternating Current (A.C.)
Alternating Cur-
rent (A.C.)
Electric current that flows for a given length of time in one direction and
immediately flows in the opposite direction for the same length of time. It
usually consists of 60 complete cycles per second.
Alveoli The terminal air sacs of the lung.
Ampere (A) A unit of electrical current or rate of flow of electrons.
Amplification As related to radiation detection instruments, the process (gas, electronic,
or both) by which ionization effects are magnified to a degree suitable for
their measurement.
Amplifier, Lin-
ear
A pulse amplifier in which the output pulse height is proportional to an in-
put pulse height for a given pulse shape up to a point at which the amplifier
overloads.
Amplifier, Pulse An amplifier, designed specifically to amplify the intermittent signals of a
nuclear detector, incorporating appropriate pulse-shaping characteristics.
Analyzer, Pulse
Height
An electronic circuit which sorts and stores the pulses according to height.
Angstrom Unit
()
One angstrom unit equals 10
8
cm.
Anion Negatively charged ion.
Annihilation
(Electron)
An interaction between a positive and a negative electron in which they
both disappear; their energy, including rest energy, being converted into
electromagnetic radiation (called annihilation radiation).
Anode Positive electrode; electrode to which negative ions are attracted.
Arc Therapy Radiation therapy in which the source of radiation is moved through a lim-
ited arc about the patient during treatment. In this way, a larger dose is
built up at the center of rotation within the patients body than on any area
of the skin. Multiple arcs may be used. Synonymous with arc treatment
and rotation therapy.
Ataxia The inability to coordinate muscular movements.
Atom Smallest particle of an element that is capable of entering into a chemical
reaction.
Atomic Mass (u) The mass of a neutral atom of a nuclide, usually expressed in terms of
atomic mass units. The atomic mass unit is one twelfth the mass of one
neutral atom of carbon 12; equivalent to 1.6604 10
24
gm.
Atomic Number
(Z)
The number of protons in the numbers of a neutral atom of a nuclide. The
effective atomic number is calculated from the composition and atomic
numbers of a compound or mixture. An element of this atomic number
would interact with photons in the same way as the compound or mixture.
Atomic Weight The weighted mean of the masses of the neutral atoms of an element ex-
pressed in atomic mass units.
Glossary: Attenuation 7-5
Attenuation The process by which a beam of radiation is reduced in intensity when
passing through some material. It is the combination of absorption and
scattering processes and leads to a decrease in flux density of the beam
when projected through matter.
Attenuation
Coefficient
A general term used to describe quantitatively the reduction in intensity of
a beam of radiation as it passes through a particular material.
Attenuation Coefficient, Compton: The fractional number of photons re-
moved from a beam of radiation per unit thickness of a material through
which it is passing as a result of Compton effect interactions.
Attenuation Coefficient, Linear: The fractional number of photons re-
moved from a beam of radiation per unit thickness of a material through
which it is passing due to all absorption and scattering processes.
Attenuation Coefficient, Pair Production: That fractional decrease in the
intensity of a beam of ionizing radiation due to pair production in a me-
dium through which it passes.
Attenuation Coefficient, Photoelectric Effect: That fractional decrease
in the intensity of a beam of ionizing radiation due to photoelectric effect in
a medium through which it is passing.
Attenuation
Factor
A measure of the opacity of a layer of material for radiation traversing it; the
ratio of the incident intensity to the transmitted intensity. It is equal to I
0
/I,
where I
0
and I are the intensities of the incident and emergent radiation, re-
spectively. In the usual sense of exponential absorption (I = I
0
e
t
), the at-
tenuation factor is e
t where t is the thickness of the material and is the

absorption coefficient.
Auger Effect The emission of an electron from the extranuclear portion of an excited
atom when the atom undergoes a transition to a less excited state.
Autoradiogra-
phy
Record of radiation from radioactive material in an object, made by placing
the object in close proximity to a photographic emulsion.
Autotrans-
former
A transformer with a single wrapping or winding of wire, with both ends of
the wire attached to the primary alternating current.
Avalanche The multiplicative process in which a single charged particle accelerated by
a strong electric field produces additional charged particles through colli-
sion with neutral gas molecules. This cumulative increase of ions is also
known as Townsend ionization or Townsend avalanche.
Average Life
(Mean Life)
The average of the individual lives of all the atoms of a particular radioac-
tive substance. It is 1.443 times the radioactive half-life.
Avogadros
Number
(Avogadros
Constant) (N
A
)
Number of atoms in a gram atomic weight of any element; also the number
of molecules in a gram molecular weight of any substance. It is numerically
equal to 6.023 10
23
on the unified mass scale.
Backplane A printed circuit board with connectors for inserting other printed circuit
boards.
Back Pointer A linear accelerator accessory used to identify the central axis of the radia-
tion beam.
7-6 Glossary: Backscatter
Backscatter The deflection of radiation by scattering processes through angles greater
than 90 degrees with respect to the original direction of motion.
Barriers, Protec-
tive
Barriers of radiation-absorbing material, such as lead, concrete, and plas-
ter, used to reduce radiation exposure.
Barriers, Primary Protective: Barriers sufficient to attenuate the useful
beam to the required degree.
Barriers, Secondary Protective: Barriers sufficient to attenuate stray ra-
diation to the required degree.
Beam A unidirectional or approximately unidirectional flow of electromagnetic ra-
diation or of particles.
Useful Beam (Radiology): Radiation that passes through the aperture,
cone, or other collimating device of the source housing; sometimes called
primary beam.
Beam Axis A geometric line from the target (or source) outward along the geometric
center of the beam.
Beam Hardening The process of eliminating the low-energy photons from a beam of x-rays.
This process changes the quality of the beam in such a manner that the av-
erage energy of the beam increases.
Beam Quality The spectral energy distribution of the radiation beam. Beam quality affects
the penetration of the beam through tissue and the relative absorption of
the energy in different types of tissue.
Beam Shaping The use of special blocks, wedges, compensators, and other devices to cre-
ate a treatment beam of the geometric proportions required for a treatment
plan beyond the capabilities of the collimator.
Becquerel (Bq) The new special unit of activity. One becquerel equals one nuclear disinte-
gration per second.
Bend Magnet
Assembly
A beam transport system for guiding the electron beam from the linear ac-
celerator structure to the x-ray target or electron scattering foil.
Beta Particle Charged particle emitted from the nucleus of an atom, with a mass and
charge equal in magnitude to that of the electron.
Betatron A magnetic induction accelerator which makes use of a varying magnetic
field to accelerate electrons. Electrons are injected into a toroidal vacuum
chamber which is between the poles of an iron-core magnet. The rate of
change of the magnet flux and magnetic field at the orbit radius are related
to maintain a constant radius for the accelerating electrons.
Biologic Effec-
tiveness of Radi-
ation
(See Relative Biologic Effectiveness.)
Blood Dyscrasia Any persistent change from normal of one or more of the blood compo-
nents.
Glossary: Bone Marrow 7-7
Bone Marrow Soft material which fills the cavity in most bones; it manufactures most of
the formed elements of the blood.
Bone Seeker Any compound or ion which migrates in the body preferentially into bone.
Brachytherapy Therapy at short distances with beta or gamma radiation. Implantation or
placement therapy with needles, inserts, or other such applications con-
taining radioactive materials. Useful in the treatment of various diseases.
Bragg-Gray Prin-
ciple
The relationship between energy absorbed in a small gas-filled cavity in a
medium to energy absorbed (in the medium) from ionizing radiation. The
relationship is expressed as E
m
= W J
g
S
g
m
, where E
m
= energy/mass
absorbed in the medium, W = average energy needed to produce an ion pair
in the gas, J
g
= number of ion pairs/mass formed in the gas, and S = ratio
of the stopping power for secondary particles in the medium to that in the
gas.
Branching The occurrence of two or more modes by which a radionuclide can undergo
radioactive decay. For example, Ra
C
can undergo or
decay,
64
Cu can
undergo
,
+
, or electron capture decay. An individual atom of a nuclide
exhibiting branching disintegrates by one mode only. The fraction disinte-
grating by a particular mode is the branching fraction for that mode. The
branching ratio is the ratio of two specified branching fractions (also
called multiple disintegration).
Bremsstrahlung Secondary photon radiation produced by deceleration of charged particles
passing through matter.
Buildup The increase in absorbed dose with depth below the surface in a material
irradiated by a beam of photons or particulate radiation. Buildup may be of
two kinds:
Electron Buildup: This is due to the production by the incident radiation
of increasing numbers of forward-moving high-energy electrons increasing
with depth until a maximum electron fluence rate has been reached. This
effect gives rise to the phenomenon of skin sparing and is most marked
for photon energies greater than about 400 keV. The effect is not noticeable
for x-ray photons generated by potentials of less than 400 kV. For high-en-
ergy beams, this process is more important.
Photon Buildup: Multiple photon scattering in the superficial layers of the
phantom, which may lead to an increase in absorbed dose for a short dis-
tance. This effect is observed particularly with photons generated by poten-
tials of 50 to 150 kV and large field sizes.
Buildup Factor The ratio of the intensity of x or gamma radiation (both primary and scat-
tered) at a point in an absorbing medium to the intensity of only the pri-
mary radiation. This factor has particular application for broad beam at-
tenuation. Intensity may refer to energy flux, dose, or energy absorption.
Buncher The input resonant cavity in a klystron or linear accelerator.
Burial Ground
(Graveyard)
A place for burying unwanted radioactive objects to prevent escape of their
radiations, the earth or water acting as a shield. Such objects must be
placed in watertight, non-corrodible, containers so the radioactive material
cannot leach out and invade underground water supplies.
7-8 Glossary: Calibration

Calibration Determination of variation from standard, or accuracy, of a measuring in-
strument to ascertain necessary correction factors.
Calorie (cal) Amount of heat necessary to raise the temperature of one gram of water
1C (from 14.5 to 15.5C).
Cancer Any malignant neoplasm (Popular usage).
Capillary A small, thin-walled blood vessel connecting an artery with a vein.
Capture, Elec-
tron
A mode of radioactive decay involving the capture of an orbital electron by
its nucleus. Capture from a particular electron shell is designated as K-
electron capture, L-electron capture, etc.
Capture, K-Elec-
tron
Electron capture from the K shell by the nucleus of the atom. Also loosely
used to designate any orbital electron process.
Capture, Radia-
tive
The process by which a nucleus captures an incident particle and loses its
excitation energy immediately by the emission of gamma radiation.
Capture, Reso-
nance
An inelastic nuclear collision occurring when the nucleus exhibits a strong
tendency to capture incident particles or photons of particular energies.
Carcinogenic Capable of producing cancer
Carcinoma Malignant neoplasm composed of epithelial cells, regardless of their deriva-
tion.
Card Cage A chassis or frame that holds printed circuit boards.
Carrousel An rotating assembly in the treatment head that places various elements,
such as flattening filters and scattering foils, into the beam path.
Catalyst A substance which alters the velocity of a chemical reaction (positive cata-
lysts increase velocity) yet may be recovering practically unchanged after
the reaction has occurred.
Cataract A clouding of the crystalline lens of the eye which obstructs the passage of
light.
Cathode Negative electrode; electrode to which positive ions are attracted.
Cation Positively charged ion.
Cell (Biological) The fundamental unit of structure and function in organisms.
Cells, Somatic Body cells, usually with two sets of chromosomes, as opposed to germ cells,
which have only one set.
Central Axis The straight line passing through the center of the target (or source) and
the center of the final collimator. Concentric with the beam axis.
Chamber, Cloud A device for observing the paths of ionizing particles. It is based on the prin-
ciple that supersaturated vapor condenses more readily on ions than on
neutral molecules.
Glossary: Chamber, Ionization 7-9
Chamber, Ion-
ization
An instrument designed to measure a quantity of ionizing radiation in
terms of the charge of electricity associated with ions produced within a de-
fined volume. (See also Condenser r-Meter.)
Air-Wall Ionization Chamber: Ionization chamber in which the materials
of the wall and electrodes are so selected as to produce ionization essen-
tially equivalent to that in a free-air ionization chamber. This ionization is
possible only over limited ranges of photon energies. Such a chamber is
more appropriately termed an air-equivalent ionization chamber.
Extrapolation Ionization Chamber: An ionization chamber with elec-
trodes whose spacing can be adjusted and accurately determined to permit
extrapolation of its reading to zero chamber volume.
Free-Air Ionization Chamber: An ionization chamber in which a delimited
beam of radiation passes between the electrodes without striking them or
other internal parts of the equipment. The electric field is maintained per-
pendicular to the electrodes in the collecting region. As a result, the ionized
volume can be accurately determined from the dimensions of the collecting
electrode and the limiting diaphragm. This ionization chamber is the basic
standard instrument for x-ray dosimetry within the range of 5 to 1400 kVp.
Standard Ionization Chamber: A specially constructed ionization cham-
ber from which other ionization chambers can be calibrated.
Thimble Ionization Chamber: A small cylindrical or spherical ionization
chamber, usually with walls of organic material.
Tissue Equivalent Ionization Chamber: An ionization chamber in which
the material of the walls, electrodes, and gas are so selected as to produce
ionization essentially equivalent to that characteristic of the tissue under
consideration. In some cases it is sufficient to have only tissue equivalent
walls, and the gas may be air, provided the air volume is negligible. The es-
sential point in this case is that the contribution to the ionization in the air
made by ionizing particles originating in the air is negligible, compared to
that produced by ionizing particles characteristic of the wall material.
Chamber,
Pocket
A small, pocket-sized ionization chamber used for monitoring radiation ex-
posure of personnel. Before use, it is given a charge, and the amount of dis-
charge is a measure of the radiation exposure.
Charger-Reader An auxiliary device used for establishing a particular voltage level in an ion-
ization chamber and subsequently for evaluating that voltage level.
Charge, Space The electric charge carried by a cloud or stream of electrons or ions in a
vacuum or a region of low gas pressure, when the charge is sufficient to
produce local changes in the potential distribution. It is of importance in
thermionic tubes, photoelectric cells, ion accelerators, etc.
Cerenkov Radia-
tion
Blue light emitted when a charged particle moves in a transparent medium
with a speed greater than that of light in the same medium.
Circuit, Antico-
incidence
A circuit with two input terminals which delivers an output pulse if one in-
put terminal receives a pulse, but delivers no output pulse if pulses are re-
ceived by both input terminals simultaneously or within an assignable time
interval.
Circuit Breaker A switch that automatically interrupts an electrical circuit upon sensing an
abnormal flow of electrical current.
7-10 Glossary: Circuit, Coincidence
Circuit, Coinci-
dence
An electron circuit that produces a usable output pulse only when each of
two or more input circuits receives pulses simultaneously or within an as-
signable time interval.
Circuit, Inte-
grating
An electron circuit which records the total number of ions or events col-
lected for a given time from which an average value for the number of ions
or events per unit time can be found.
Cladding (Clad) An external layer of material applied directly to nuclear fuel or other mate-
rial to provide protection from a chemically reactive environment, to pro-
vide containment of radioactive products produced during the irradiation
of the composite, or to provide structural support.
Clinical Pertaining to the observed symptoms and cause of disease.
Coincidence The occurrence of counts in two or more detectors simultaneously or within
an assignable time interval. A true coincidence is one that is due to the in-
cidence of a single particle or of several genetically related particles. An ac-
cidental, chance, or random coincidence is one that is due to the accidental
occurrence of unrelated counts in the separate detectors. An anticoinci-
dence is the occurrence of a count in a specified detector unaccompanied
simultaneously or within an assignable time interval by a count in other
specified detectors. A delayed coincidence is the occurrence of a count in
one detector at a short, but measurable, time after a count in another de-
tector. The two counts are due to genetically related occurrence, such as
successive events in the same nucleus.
Collimator A device for confining the elements of a beam within an assigned solid an-
gle. Sets of metal blocks, fixed and movable, in the treatment head that
limit the treatment field to the desired size.
Collimator Rota-
tion Readout
A display that indicates the degrees of rotation of the collimator about the
central axis.
Collision Encounter between two subatomic particles (including photons) which
changes the existing momentum and energy conditions. The products of
the collision need not be the same as the initial systems.
Elastic Collision: A collision in which no change occurs either in the inter-
nal energy of each participating system or in the sum of their kinetic ener-
gies of translation.
Inelastic Collision: A collision in which changes occur both in the internal
energy of one or more of the colliding systems and in the sums of the ki-
netic energies of translation before and after the collision.
Compensator A slab of material placed in the treatment beam to compensate for uneven-
ness of machine output or body contour.
Compton Effect An attenuation process observed for x or gamma radiation in which an in-
cident photon interacts with an orbital electron of an atom to produce a re-
coil electron and a scattered photon of energy less than the incident pho-
ton. (See also Absorption, Pair Production, and Photoelectric Effect.)
Condenser r-
Meter
An instrument consisting of an air-wall ionization chamber together with
auxiliary equipment for charging and measuring its voltage. It is used as an
Glossary: Console Group 7-11
integrating instrument for measuring the exposure of x or gamma radiation
in roentgens (R). (See also Chamber, Ionization)
Console Group The Clinac operator interface and system control units, typically including
a keyboard, display terminal, control console, console computer, and
printer.
Contactor A heavy duty relay used to control high-power electrical circuits.
Contamination A foreign substance dispersed where it is undesirable: for example, un-
wanted electrons in the photon beam or unwanted photons in an electron
beam.
Contamination,
Radioactive
Deposition of radioactive material in any place where it is not desired, par-
ticularly where its presence may be harmful. The harm may be in vitiating
an experiment or a procedure, or in endangering personnel.
Control Console The Clinac system control unit, which includes operator controls for start-
ing and stopping an exposure.
Controlled Area A defined area in which the occupational exposure of personnel (to radia-
tion) is under the supervision of the Radiation Protection Supervisor.
Control System A coordinated group of components designed to exert a directing influence
on other components. A system of apparatus for automatically controlling
an accelerator by a servo system that adjusts the control elements to main-
tain the flux level near a desired value.
Corpuscle A blood cell.
Corpuscular
Emission, Asso-
ciated
The full complement of secondary charged particles (usually limited to elec-
trons) associated with an x-ray or gamma ray beam in its passage through
air. The full complement of electrons is obtained after the radiation has tra-
versed sufficient air to bring about equilibrium between the primary pho-
tons and secondary electrons. Electronic equilibrium with the secondary
photons is intentionally excluded.
Coulomb (C) Unit of quantity in current electricity. A quantity afforded by 1 ampere of
current in 1 second flowing against 1 Ohm of resistance with a force of 1
Volt.
Count (Radia-
tion Measure-
ments)
The external indication of a device designed to enumerate ionizing events. It
may refer to a single detected event or to the total number registered in a
given period of time. The term often is erroneously used to designate a dis-
integration, ionizing event, or voltage pulse.
Spurious Count: In a radiation counting device, a count caused by any
agency other than radiation.
Counter, Gas
Flow
A device in which an appropriate atmosphere is maintained in the counter
tube by allowing a suitable gas to flow slowly through the sensitive volume.
Counter, Geiger-
Muller
Highly sensitive, gas filled radiation measuring device. It operates at volt-
ages sufficiently high to produce avalanche ionization.
7-12 Glossary: Counter, Proportional
Counter, Propor-
tional
Gas-filled radiation detection device; the pulse produced is proportional to
the number of ions formed in the gas by the primary ionization particle.
Counter, Scintil-
lation
The combination of scintillator, photomultiplier tube, and associated cir-
cuits for counting light emissions produced in the phosphors.
Counting, Coin-
cidence
A technique in which particular types of events are distinguished from
background events by coincidence circuits, which register coincidences
caused by the type of events under consideration.
Counting Rate-
meter
An instrument that gives a continuous indication of the average rate of ion-
izing events.
Cross-Sectional
Area (of an x-ray
beam)
An area in the plane of the beam perpendicular to its direction of travel.
Curie The former special unit of activity. One curie equals 3.7 10
10
nuclear
transformations per second. Several fractions of the curie are in common
usage. (Abbreviated Ci)
Microcurie: One-millionth of a curie (3.7 10
4
disintegrations per second).
Abbreviated Ci.
Millicurie: One-thousandth of a curie (3.7 10
7
disintegrations per sec-
ond). Abbreviated mCi.
Picocurie: One-millionth of a microcurie (3.7 10
2
disintegrations per
second or 2.22 disintegrations per minute). Abbreviated pCi; replaces the
term Ci.
Cutie Pie An ionization chamber device commonly used for detecting radiation expo-
sure rate.
Cyclotron A particle accelerator in which charged particles receive repeated synchro-
nized accelerations or kicks by electrical fields as the particles spiral out-
ward from their source. The particles are kept in the spiral by a powerful
magnet.
Daughter Synonym for Offspring. (See Decay Product.)
Deadman
Switch
Synonymous with Motion enable switch.
Decay, Radioac-
tive
Disintegration of the nucleus of an unstable nuclide by spontaneous emis-
sion of charged particles and/or photons.
Decay Constant The fraction of the number of atoms of a radioactive nuclide which decay in
unit time. Symbol: . (See also Decay Curve and Disintegration Constant.)
Decay Curve A curve showing the relative amount of radioactive substance remaining af-
ter any time interval.
Decay Product A nuclide resulting from the radioactive disintegration of a radionuclide,
formed either directly or as the result of successive transformations in a ra-
dioactive series. A decay product may be either radioactive or stable.

Glossary: Decrement Lines 7-13
Decrement
Lines
Imaginary lines drawn through parts where the absorbed energy (radiation
dose) is a certain percent of the energy absorbed at the same depth along
the central axis of the radiation beam.
Delta Ray Any secondary ionizing particle ejected by recoil when a primary ionizing
particle passes through matter.
Densitometer Instrument utilizing a photocell to determine the degree of darkening of de-
veloped photographic film.
Density (Physi-
cal)
The mass per unit volume of a substance. Usually kg/m
3
or g/cc
l
. (Sym-
bol: p)
Depth Dose A radiation dose at some specified depth in tissue relative to the dose at a
fixed reference point on the beam axis. It is usually expressed as a percent-
age of surface dose.
DeQing A circuit in the modulator cabinet that regulates the size of dc voltage
pulses delivered by the high-voltage power supply to the pulse forming net-
work.
Detector An instrument capable of registering the presence of radiation. The two
common modes of operation for a detector are:
Mean-Level or Integrating: The average effect of the radiation is cumu-
lated over time.
Pulse-Type: Individual radiation interactions are separated or resolved in
time.
Detector, Radia-
tion
Any device for converting radiant energy to a form more suitable for obser-
vation. An instrument used to determine the presence, and sometimes the
amount, of radiation.
Deuterium (D) A heavy isotope of hydrogen with one proton and one neutron in the nu-
cleus.
Deuteron An isotopic form of hydrogen in which the nucleus contains one proton and
one neutron. When deuterons are substituted for the common form of hy-
drogen in the water molecule, the substance is known as heavy water.
Directly Ioniz-
ing Particles
Charged particles such as alpha or beta particles which cause ionization of
an atom without any intermediate interaction taking place.
Disintegration,
Nuclear
A spontaneous nuclear transformation (radioactivity) characterized by the
emission of energy and/or mass from the nucleus. When numbers of nuclei
are involved, the process is characterized by a definite half-life.
Disintegration
Constant
The fraction of the number of atoms of a radioactive nuclide which decay in
unit time; in the equation , in which N
0
is the initial number of
atoms present, and N is the number of atoms present after some time, t.
(See also Decay Constant.)
Dose A general form denoting the quantity of radiation or energy absorbed. For
special purposes it must be appropriately qualified. If unqualified, it refers
to absorbed dose. (See also Maximum Permissible Dose.)
N N
0
e
t
=
7-14 Glossary: Dose
Absorbed Dose: The energy imparted to matter by ionizing radiation per
unit mass of irradiated material at the place of interest. The former unit of
absorbed dose is the rad. One rad equals 100 ergs per gram. The new unit
of absorbed dose is the gray. One gray equals 1 joule per kilogram. (See
also Rad and Tissue Dose.)
Cumulative Dose (Radiation): The total dose resulting from repeated ex-
posures to radiation.
Depth Dose: The radiation dose delivered at a particular depth beneath the
surface of the body. It is usually expressed as a percentage of surface dose.
Dose Distribution: The variation of dose in any region of an irradiated ob-
ject.
Dose Equivalent (H): A quantity used in radiation protection It expresses
all radiations on a common scale for calculating the effective absorbed
dose. It is defined as the product of the absorbed dose and certain modify-
ing factors. (The former unit of dose equivalent is the rem. The new unit of
dose equivalent is the Sievert [Sv].)
Exit Dose: Dose of radiation at surface of body opposite to that on which
the beam is incident.
Integral Absorbed Dose (Volume Dose): A term used mainly in radiation
biology to mean the total energy absorbed by an individual or other biolog-
ical object or phantom during exposure to radiation. It is frequently ob-
tained by integrating the absorbed dose with respect to mass throughout
an irradiated region. It may be stated in joule or kilogram gray.
Maximum Permissible Dose Equivalent (MPD): The greatest dose equiv-
alent that a person or specified part thereof shall be allowed to receive in a
given period of time.
Median Lethal Dose (MLD): Dose of radiation that would be required to
kill, within a specified period, 50% of the individuals in a large group of an-
imals or organisms; also called LD
50
.
Midline Absorbed Dose: The absorbed dose calculated or measured for a
point in tissue and at the midline or center of the biological specimen,
i.e., for the point lying equidistant from the exterior points on the speci-
men. The designation is for dosimetric purposes and implies no particular
biological significance for the midline location.
Percentage Depth Dose: The ratio expressed as a percentage, of the ab-
sorbed dose rate at a point, at depth along the beam axis, to the absorbed
dose rate at a fixed reference point in the beam axis.
Permissible Dose: The dose of radiation that an individual may receive
within a specified period with expectation of no significantly harmful result.
Skin Dose (Radiology): Absorbed dose at center of irradiation field on skin.
It is the sum of the dose in air and scatter from body parts.
Surface Dose: The absorbed dose delivered by a radiation beam anywhere
the radiation passes through the superficial layer of the phantom or pa-
tient.
Threshold Dose: The minimum absorbed dose that will produce a detect-
able degree of any given effect.
Tissue Dose: Absorbed dose received by tissue in the region of interest, ex-
pressed in rads. (See also Absorbed Dose and Rad.)
Glossary: Dose, Protraction 7-15
Dose Meter, Integrating
Ionization chamber and measuring system designed for determining total
radiation administered during an exposure. In medical radiology the cham-
ber is usually designed to be placed on the patients skin. A device may be
included to terminate the exposure when it has reached a desired value.
Dose, Protrac-
tion
A method of administering radiation by delivering it continuously over a
relatively long period at a low dose rate.
Dose Rate Absorbed dose delivered per unit time.
Dose Ratemeter Any instrument that measures radiation dose rate.
Dosimeter Instrument to detect and measure accumulated radiation exposure. In
common usage, a pencil-size ionization chamber with a self-reading elec-
trometer, used for personnel monitoring.
Dosimetrist An individual with training and knowledge in treatment planning. Under
the supervision of a qualified radiological physicist, a dosimetrist makes
dose calculations and assists in calibration and verification of dose distri-
bution within the patient.
Dosimetry The calculations, measurements, and other activities required for deter-
mining the radiation dose to be delivered.
Dosimetry Inter-
lock
A machine condition is identified in which the ability of the Clinac to deliver
or measure dose may be impaired.
Dosimetry, Pho-
tographic
Determination of cumulative radiation dose with photographic film and
density measurement.
Drive Stand The stationary unit on the Clinac that holds the gantry.
Dual Dosimetry The use of two independent signals from the Clinac dosimeter proportional
to the integrated dose. The primary channel is programmed to terminate
the beam at the dose set by the operator. If the beam continues, the sec-
ondary channel is programmed to terminate the beam at a preset number
of monitor units beyond the set dose.
Dyne The unit of force which, when acting upon a mass of one gram, will produce
an acceleration of one centimeter per second per second.
Eddy Current An induced electric current circulating wholly within a mass of metal. Such
currents are converted into heat, and thus cause serious waste.
Efficiency
(Counters)
A measure of the probability that a count will be recorded when radiation is
incident on a detector. Usage varies considerably, so it is well to ascertain
which factors (e g., window transmission, sensitive volume, energy depen-
dence) are included in a given case.
Electricity One of the forces of nature developed by chemism, magnetism, or friction;
also said to be electrons in motion.
Electrode A conductor used to establish electric contact with a nonmetallic part of a
circuit.
7-16 Glossary: Electrolyte
Electrolyte A substance capable of conducting an electric current and being decom-
posed by it.
Electromag-
netic Interfer-
ence (EMI)
Electromagnetic radiation that can degrade the performance of a radiother-
apy accelerator or other nearby electronic equipment.
Electromag-
netic Radiation
Transport of energy through space as a combination of an electric and mag-
netic field: for example, visible light and x-rays.
Electromag-
netic wave
A wave produced by the oscillation of an electric charge.
Electrometer Electrostatic instrument for measuring the difference in potential between
two points. Used to measure change in electric potential of charged elec-
trodes resulting from ionization produced by radiation.
Electromotive
Force
Potential difference across electrodes tending to produce an electric cur-
rent.
Electron A stable elementary particle having an electric charge equal to 1.60 10
19
C and a rest mass equal to 9.1091 10
31
kg.
Secondary Electron: An electron ejected from an atom, molecule, or sur-
face as a result of an interaction with a charged particle or photon.
Valence Electron: Electron that is gained lost, or shared in a chemical re-
action.
Electron Affin-
ity
The tendency of a neutral atom to attract a free electron to itself.
Electron Beam
Therapy
Treatment by electrons accelerated to high energies in a linear accelerator.
Primarily used for lesions situated at or near the surface.
Electron Equi-
librium
A condition established in a standard ionization chamber whereby the
number of electrons entering a specified volume equals the number of elec-
trons leaving that volume.
Electron Gun A structure that injects electrons into the linear accelerator.
Electron Volt A unit of energy equivalent to the energy gained by an electron in passing
through a potential difference of 1 volt. Larger multiple units of the electron
volt are frequently used: keV for thousand or kilo electron volts; MeV for
million or mega electron volts. (Abbreviated: eV, 1 eV = 1.6 10
19
J.)
Electroscope Instrument for detecting the presence of electric charges by the deflection
of charged bodies. It has two metallic leaves hanging at the end of a very
slender vane. When like charges are placed on the leaves, they move apart
or repel. As the charge is reduced, the leaves move closer together until
they are finally side by side when the charge has been reduced to zero.
Electrostatic
Field
The region surrounding an electric charge in which another electric charge
experiences a force.
Element A category of atoms all of the same atomic number.
Glossary: Emergency-off Circuit 7-17
Emergency-off
Circuit
An electrical circuit in the Clinac that causes all high power to be removed
from the system whenever an emergency off button is pressed.
Energy Capacity for doing work. Potential energy is the energy inherent in a mass
because of its spatial relation to other masses. Kinetic energy is the en-
ergy possessed by a mass because of its motion; SI units: kg m
2
S
2
or
joules.
Binding Energy: The energy represented by the difference in mass between
the sum of the component parts and the actual mass of the nucleus.
Excitation Energy: The energy required to change a system from its
ground state to an excited state. Each different excited state has a different
excitation energy.
Ionizing Energy: The average energy lost by ionizing radiation in produc-
ing an ion pair in a gas.
Energy Depen-
dence
The characteristic response of a radiation detector to a given range of radi-
ation energies or wavelengths compared with the response of a standard
free-air chamber.
Energy Fluence The sum of the energies, exclusive of rest energies, of all particles passing
through a unit cross-sectional area.
Energy Flux
Density (Energy
Fluence Rate)
The sum of the energies, exclusive of rest energies, of all particles passing
through a unit cross-sectional area per unit time (energy fluence per unit of
time).
Energy
Imparted
See Dose, Integral Absorbed Dose.
Entrance Port The area on the surface of a patient or a phantom on which a radiation
beam is incident.
Enzyme A biological catalyst of great specificity for a particular substance or a par-
ticular group of closely related substances which generally activates or ac-
celerates a biochemical reaction.
Epidermis The outermost layer of cells of the skin.
Epilation (Depil-
ation)
The temporary or permanent removal or loss of hair.
Epithelium A term applied to cell that line all canals and surfaces having communica-
tion with external air; also, cells specialized for secretion in certain glands
as the liver, kidneys, etc.
Erg Unit of work by a force of one dyne acting through a distance of one cm.
Unit of energy which can exert a force of one dyne through a distance of one
cm; ergs units: dyne-cm or gm-cm
2
/sec
2
.
Erythema An abnormal redness of the skin due to distension of the capillaries with
blood. It can be caused by many different agents: heat, drugs, ultraviolet
rays, and ionizing radiation.
Erythrocyte A red blood corpuscle.
7-18 Glossary: Eugenics
Eugenics The science which deals with the influences that improve the hereditary
qualities of a race or breed.
Excitation The addition of energy to a system, thereby transferring it from its ground
state to an excited state. Excitation of a nucleus, an atom, or a molecule
can result from absorption of photons or from inelastic collisions with other
particles.
Exit Dose The absorbed dose at the point where the beam axis emerges from the pa-
tient.
Exposure A measure of the ionization produced in air by x or gamma radiation. It is
the sum of the electric charges on all ions of one sign produced in air when
all electrons liberated by photons in a volume element of air are completely
stopped in air, divided by the mass of the air in the volume element. The
former special unit of exposure is the roentgen. The new special unit of ex-
posure is C kg
1
.
Acute Exposure: Radiation exposure of short duration.
Chronic Exposure: Radiation exposure of long duration.
Fallout Radioactive debris from a nuclear detonation, which is airborne or has
been deposited on the earth. Special forms of fallout are Dry Fallout,
Rainout, and Snowout.
Field A plane section of the beam perpendicular to the beam axis.
Field Block A solid object of attenuating material used to shape a treatment beam.
Field Light A light system that illuminates an area on the patients surface identifying
the area of therapy beam entry.
Field Size The size of an area irradiated by a given beam, usually measured by one of
the following conventions: geometric field size, which measures the geomet-
ric projection on a plane perpendicular to the central axis, or physical field
size, which measures the area included within the 50 percent maximum
dose isodose curve at the depth of maximum dose.
Film Badge A pack of photographic film that measures radiation exposure for person-
nel monitoring. The badge may contain two or three films of differing sen-
sitivity and filters to shield parts of the film from certain types of radiation.
Film Ring A film badge in the form of a finger ring.
Filter (Radiol-
ogy)
Primary: A sheet of material, usually metal, placed in a beam of radiation to
absorb preferentially the less penetrating components. Secondary: A sheet
of material of low atomic number (relative to the primary filter) placed in the
filtered beam of radiation produced by the primary filter.
Filtration,
Inherent (X-
rays)
The filter permanently in the useful beam; it includes the window of the x-
ray tube and any permanent tube or source enclosure.
Fissile (Fission-
able) Material
Any material readily fissioned by slow neutrons, for example,
235
U,
239
P.
Glossary: Fission 7-19
Fission The splitting of a heavy nucleus into two roughly equal parts (which are nu-
clei of lighter elements), accompanied by the release of a relatively large
amount of energy and frequently one or more neutrons. Fission can occur
spontaneously, but usually it is caused by the absorption of gamma-ray
photons, neutrons, or other particles.
Flattening Filter A cone-shaped attenuator placed in the x-ray beam to achieve uniform in-
tensity over the specific treatment field at a specific depth and a specific en-
ergy.
Fluence The number of particles passing through a unit cross-sectional area.
Fluorescence The emission of radiation of particular wavelengths by a substance as a re-
sult of absorption of radiation of shorter wavelength. This emission occurs
essentially only during the irradiation.
Fluorescent
Screen
A sheet of material coated with a substance (such as calcium tungstate or
zinc sulfide) which will emit visible light when irradiated with ionizing radi-
ation.
Fluorography
(photofluorogra-
phy)
Photography of image produced on fluorescent screen by x or gamma radi-
ation.
Fluoroscope A fluorescent screen, suitably mounted with respect to an x-ray tube for
ease of observation and protection, used for indirect visualization (by x-
rays) of internal organs in the body or internal structures in apparatus or
in masses of material.
Flux Density
(fluence rate)
The number of particles passing through a unit cross-sectional area per
unit of time. (Fluence per unit of time.)
Focal Spot (X-
rays)
The part of the target of the x-ray tube struck by the main electron stream.
Fractionation A technique of administering radiation therapy in multiple doses over a
number of days or weeks to achieve a maximum therapeutic ratio.
Frequency In harmonic motions, the number of cycles, revolutions, or vibrations com-
pleted in a unit of time. (See also Hertz.)
Function Key
Assignments
A bar on the bottom line of the screen indicating the command currently
assigned to each function key on the keyboard.
Fusion, Nuclear Act of coalescing two or more atomic nuclei.
Gamete Either of the two germ cells (sperm or ovum).
Gamma-ray Short wavelength electromagnetic radiation of nuclear origin (range of en-
ergy from 10 keV to 9 MeV) emitted from the nucleus.
Gantry The entire rotating unit of the Clinac that emits the treatment beam. The
upper part of the gantry includes linear accelerator, and the lower part con-
tains a counterweight or a retractable beam stopper.
7-20 Glossary: Gantry Rotation Readout
Gantry Rota-
tion Readout
A display that indicates the degrees of rotation of the gantry about the iso-
center.
Gas Amplifica-
tion
As applied to gas ionization radiation detecting instruments, the ratio of the
charge collected to the charge produced by the initial ionizing event.
Geiger Region In an ionization radiation detector, the operating voltage interval in which
the charge collected per ionizing event is essentially independent of the
number of primary ions produced in the initial ionizing event.
Geiger Thresh-
old
The lowest voltage applied to a counter tube for which the number of pulses
produced in the counter tube is essentially the same, regardless of a limited
voltage increase.
Geiger Tube An ionization type radiation detector with a very high sensitivity for pho-
tons in the energy range 10 to 1000 keV.
Gene Fundamental unit of inheritance which determines and controls hereditary
transmissible characteristics. Genes are arranged linearly at definite loci
on chromosomes.
Genetics The branch of biology dealing with the phenomena of heredity and varia-
tion.
Generator
(Cow)
A device in which a daughter radionuclide is eluted from an ion exchange
column containing a parent radionuclide long lived compared to the daugh-
ter.
Genetic Effect
of Radiation
Inheritable change, chiefly mutations, produced by the absorption of ioniz-
ing radiations. On the basis of present knowledge these effects are purely
additive; recovery does not occur.
Genetically Sig-
nificant Dose
That absorbed dose equivalent which, if received by every member of the
population, would be expected to produce the same total genetic injury to
the population as the actual absorbed dose equivalent received by various
individuals.
Genotype The fundamental hereditary (genetic) constitution of an organism.
Germ Cells The cells of an organism whose function is reproduction.
Given Dose The applied dose delivered by one beam in a complete treatment or in a
treatment session.
Glove Box An enclosure used for working with radionuclides particularly those in the
form of powders and volatile liquids.
Gonad A gamete-producing organ in animals; testis or ovary.
Gray (Gy) The new unit of absorbed dose equal to 1 joule per kilogram in any me-
dium. (See Absorbed Dose.)
Grenz Rays X-rays produced at voltages of 5 to 20 kVp, intended primarily for surface
therapy.
Glossary: Ground State 7-21
Ground State The state of a nucleus, atom, or molecule at its lowest energy. All other
states are excited.
Half-Life A general term used to describe the time elapsed until some physical quan-
tity has decreased to half of its original value. Here the concept of half-life
will be applied to radionuclides.
Half-Life, Biologic: The time required for the body to eliminate one-half of
an administered dosage of any substance by regular processes of elimina-
tion. Approximately the same for both stable and radioactive isotopes of a
particular element.
Half-Life, Effective: Time required for a radioactive element in an animal
body to be diminished 50% as a result of the combined action of radioactive
decay and biologic elimination.
Half-Life, Radioactive: Time required for a radioactive substance to lose
50% of its activity by decay. Each radionuclide has a unique half-life.
Half-Value Layer
(Half Value
Thickness)
(HVL)
The thickness of a specified substance which, when introduced into the
path of a given beam of radiation, reduces the exposure rate by one-half.
Hand Pendant A hand held control device that allows the operator to adjust the treatment
couch, collimator, and gantry for a patient.
Hardness (X-
rays)
A relative specification of the quality of penetrating power of x-rays. In gen-
eral, the shorter the wavelength the harder the radiation.
Health, Radio-
logic
The art and science of protecting human beings from injury by radiation,
and promoting better health through beneficial applications of radiation.
Heat Exchanger A cooling device that uses city water to carry off the heat generated by cer-
tain Clinac systems.
Heel Effect The cathode end of the x-ray tube has a slightly visible tendency to make a
more dense image than has the anode end.
Heredity Transmission of characters and traits from parent to offspring.
Hertz Unit of frequency equal to 1 cycle per second. (See also Frequency.)
Heterogeneous
Radiation
Beam of x-rays consisting of many x-rays of different wavelengths.
Highlight A reverse-video bar shown on the monitor screen to identify a choice in a
window or an input space.
Hysteresis A lagging or retardation of the effect. The magnetization of a piece of iron or
steel due to a magnetic field that is made to vary through a cycle of values;
lags behind the field.
Effective half life
Biological half-life Radioactive half-life
Biological half-life Radioactive half-life +
------------------------------------------------------------------------------------------------------------------ =
7-22 Glossary: Immunity
Immunity The power which a living organism possesses to resist and overcome infec-
tion.
Implant (Radiol-
ogy)
Encapsulated radioactive material embedded in a tissue for therapy. It may
be permanent (seed) or temporary (needle).
Indirectly Ioniz-
ing Particles
Particles that cause ionization to occur only after an intermediate interac-
tion producing a charged particle has taken place.
Infrared Radia-
tion
Invisible thermal radiation whose wavelength is longer than the red seg-
ment of the visible spectrum.
Input Space A reverse-video box on the monitor screen that provides a space for the op-
erator to enter input to the system.
Integral Dose A measure of the total energy absorbed by a patient or object during expo-
sure to radiation.
Integrating Cir-
cuit
An electronic circuit that records the total number of ions or events col-
lected for a given time from which an average value for the number of ions
or events per unit time can be found.
Intensification
Factor
The quantity of intensification, expressed numerically as light energy, of
the applied source of energy when it passes through the screen emulsion.
Intensifying
Screen
Sheet of cardboard or other substance coated with fluorescent material,
placed in contact with the film in radiography. The x or gamma rays excite
the fluorescent substance. The light thus emitted adds to the radiation ef-
fect on the film and produces an image of greater density for a given expo-
sure. Sheets of thin lead may be used in industrial radiography and radia-
tion therapy with very high energy radiation. In this case, the increased
effect is due largely to secondary electrons and x-rays emitted by the lead.
Intensity Amount of energy per unit time passing through a unit area perpendicular
to the line of propagation at the point in question.
Interlock A electrical circuit or mechanical device to prevent operation or application
of power until the circuit or device is placed in a certain state.
International
Commission on
Radiological
Protection
(ICRP) An international organization, founded in 1928, and supported fi-
nancially by the World Health Organization (WHO), the International
Atomic Energy Agency (IAEA), the United Nations Environment Program
the International Society of Radiology, and others, which operates under
rules approved by the International Congress of Radiology. Members of
ICRP are selected from nominations submitted to it by the National Delega-
tions to the International Congress of Radiology and by the ICRP itself. The
International Executive Committee of the Congress approves the selec-
tions.
International
Commission on
Radiation Units
and Measure-
ments
(ICRU) An international organization, founded in 1925, with the principle
objective of development of internationally acceptable recommendations re-
garding: (l) quantities and units of radiation and radioactivity, (2) proce-
dures suitable for measurement and application of these quantities in clin-
ical radiology and radiobiology, and (3) physical data needed in the
application of these procedures, the use of which, tends to assure unifor-
mity in reporting. The ICRU works closely with the ICRP in its consider-
Glossary: International System of Units (SI) 7-23
ation of and recommendations for the radiation protection field. Finan-
cially, ICRU is supported by the United States National Institutes of Health
and many national and international societies, foundations, and compa-
nies.
International
System of Units
(SI)
The standard metric system of measurement adopted in 1975 for world-
wide use. SI units commonly used in radiotherapy include the gray (mea-
sures absorbed dose), sievert (measures the dose equivalent), coulomb per
kilogram (measures exposure), and the becquerel (measures the disintegra-
tion rate of a radionuclide).
Inverse Square
Law
1. A rule relating two physical entities by a particular proportionality con-
stant. This constant is one divided by the square of some other physical
quantity, usually the distance between the two physical entities. 2. A for-
mula for the relationship that the intensity of radiation is inversely propor-
tional to the square of the distance from a point source.
Ion Atomic particle, atom, or chemical radical bearing an electric charge, either
negative or positive.
Ion Chamber See Monitor ion chamber.
Ion Pair Two particles of opposite charge, usually referring to the electron and pos-
itive atomic or molecular residue resulting after the interaction of ionizing
radiation with the orbital electrons of atoms.
Ionization The process by which a neutral atom or molecule acquires a positive or
negative charge.
Primary Ionization: (l) In collision theory: the ionization produced by the
primary particles as contrasted to the total ionization, which includes the
secondary ionization produced by delta rays. (2) In counter tubes: The to-
tal ionization produced by incident radiation without gas amplification.
Secondary Ionization: Ionization produced by delta rays.
Specific Ionization: Number of ion pairs per unit length of path of ionizing
radiation in a medium, e.g., per centimeter of air or per micron of tissue.
Total Ionization: The total electric charge of one sign on the ions produced
by radiation in the process of losing its kinetic energy. For a given gas, the
total ionization is closely proportional to the initial ionization and is nearly
independent of the nature of the ionizing radiation. It is frequently used as
a measure of radiation energy.
Ionization Den-
sity
Number of ion pairs per unit volume.
Ionization Path
(Track)
The trail of ion pairs produced by an ionizing radiation in its passage
through matter.
Ionization
Potential
The potential necessary to separate one electron from an atom, resulting in
the formation of an ion pair.
Ionizing Event Any occurrence of a process in which an ion or group of ions is produced.
Ion Pump A vacuum pump that maintains the high vacuum in the accelerator by re-
moving gas molecules.
7-24 Glossary: Irradiation
Irradiation Exposure to radiation.
Isobar One of two or more chemical elements that have the same mass number
but different atomic numbers.
Isocenter The intersection of the gantry axis of rotation and the collimator bearing
axis.
Isodose Curve A line connecting points of equal radiation doses.
Isodose Chart A graphic display containing a series of isodose curves that maps out the
relative intensities of a radiation field in a phantom or patient.
Isomers Nuclides having the same number of neutrons and protons but capable of
existing, for a measurable time, in different quantum states with different
energies and radioactive properties. Commonly, the isomer of higher energy
decays to one with lower energy by the process of isometric transition.
Isotones Nuclides having the same number of neutrons in their nuclei.
Isotopes Nuclides having the same number of protons in their nuclei, and hence the
same atomic number, but differing in the number of neutrons, and there-
fore in the mass number. Almost identical chemical properties exist be-
tween isotopes of a particular element. The term should not be used as a
synonym for nuclide.
Stable Isotope: A nonradioactive isotope of an element.
Joule The unit for work and energy, equal to 1 Newton expended along a distance
of 1 meter (1J = 1N 1M)
Kerma (kinetic
energy released
per unit mass)
The kinetic energy of charged ionizing particles liberated per unit mass of
specified material by uncharged ionizing particles such as photons and
neutrons. Kerma is measured in the same units as absorbed dose, joule per
kilogram (J/kg
-1
) and its special name is gray (Gy). Kerma can be quoted
for any specified material at a point in free space or in an absorbing me-
dium. Since air kerma and tissue kerma differ by less than 10% over a wide
range of photon energies, these two may be considered equal in magnitude
for radiation protection purposes. In this respect, air kerma means air
kerma in air. Kerma is independent of the complexities of geometry of the
irradiated mass element, and permits, therefore, specification for photons
or neutrons in free space or in an absorbing medium and hence has a wider
applicability than exposure.
Keylock Switch
or Keyswitch
A rotary switch that requires a special key to operate. Similar to an ignition
switch on a car.
Kilo Electron
Volt (keV)
One thousand electron volts, 10
3
eV.
Kilovolt (kV) A unit of electric potential difference, equal to 1000 volts.
Kilovolt Con-
stant (kVcp)
The value in kilovolts of the potential difference of a constant potential gen-
erator.
Glossary: Kilovolt Peak (kVp) 7-25
Kilovolt Peak
(kVp)
The maximum value in kilovolts of the potential difference of a pulsating
potential generator. When only half the wave is used, the value refers to the
useful half of the cycle.
Laboratory Mon-
itor
(See Survey Meter.)
Laser A device for transforming incoherent light of various frequencies into a very
narrow, intense beam of coherent light.
Latent Image Development occurring between the time of exposure of a film to radiation
and the processing of that film.
Latent Period The period or state of seeming inactivity between the time of exposure of
tissue to an injurious agent and response.
LD
50
(Radiation
Dose)
Dose of radiation required to kill, within a specified period, 50 percent of
the individuals in a large group of animals or organisms. Also called the
Median Lethal Dose.
Lead Equivalent The thickness of lead affording the same attenuation, under specified con-
ditions, as the material in question.
Lesion A hurt, wound, or local degeneration.
Leukemia A disease in which there is great over-production of white blood cells, or a
relative over-production of immature white cells, and great enlargement of
the spleen. The disease is variable, at times running a more chronic course
in adults that in children. It can be produced in some animals by long con-
tinued exposure to low doses of ionizing radiation.
Linear Accelera-
tor Structure
A linear series of adjacent cylindrical microwave resonant cavities (called
the guide) in which charged particles are accelerated by applying a high-
frequency voltage during the particle transit inside the structure.
Linear Energy
Transfer (LET)
The quotient of d
E
by d
L
, in which d
L
is the distance traversed by a particle
and d
E
is the average energy loss in d
L
due to collisions with energy trans-
fers less than some specific value. Simply, it is a conventional expression
for energy deposition measured along the track of an ionizing particle.
Gamma and x-ray photons generate low LET electron tracks. Natural alpha
particles and fast neutrons and protons give high-LET tracks.
Localization
Films
X-ray films taken with radiopaque markers to define the tumor position rel-
ative to external markings.
Major Interlock A machine condition is identified that could damage the Clinac if not cor-
rected.
Mass The material equivalent of energy; different from weight in that it neither in-
creases nor decreases with gravitational force.
Mass Numbers The number of nucleons (protons and neutrons) in the nucleus of an atom.
(Symbol: A)
7-26 Glossary: Maximum Permissible Dose (MPD)
Maximum Per-
missible Dose
(MPD)
(See Dose.)
Mean Free Path The average distance that particles of a specified type travel before a spec-
ified type (or types) of interaction in a given medium. The mean free path
may thus be specified for all interactions (i.e., total mean free path) or for
particular types of interaction such as scattering, capture, or ionization.
Mean Life The average lifetime for an atomic or nuclear system in a specified state.
For an exponentially decaying system the average time for the number of
atoms or nuclei in a specified state to decrease by a factor of e (2.718).
Mega Electron
Volt (MeV)
One million electron volts, 10
6
eV.
Meson One of a class of medium-mass, short lived elementary particles with a
mass between that of the electron and that of the proton. Examples: Pi me-
sons (pions) and K mesons (kaons)
Metabolism The sum of all physical and chemical processes by which living organized
substance is produced and maintained and by which energy is made avail-
able foe the uses of the organism.
Metastasis The transfer in the body of malignant neoplastic cells from the original or
parent site to one more distant.
Micron () Unit of length equal to 10
6
meters.
Microwave Radio waves in the frequency range of approximately 1000 megahertz and
upward.
Milliampere A unit of current. Generally the current flowing between the filament and
anode of an x-ray tube is stated in this unit.
Milliroentgen
(mR)
A submultiple of the Roentgen, equal to one one-thousandth of a Roentgen.
Minor Interlock A machine condition is identified that prevents Clinac beam-on, but the
condition is normally user-correctable.
Modulator A system in the Clinac that generates a succession of short pulses of high
current and voltage for operating the klystron.
Molecule Smallest quantity of a compound which can exist by itself and retain all
properties of the original substance.
Molybdenum
Breakthrough
This term refers to the amount of parent nuclide, molybdenum, contained
in an eluted sample of its offspring
99m
Tc. 37 kBq (1 Ci) of Mo is allowed
per 37 MBq (l mCi) of
99m
Tc eluate. However, no more than 185 kBq (5 Ci)
of Mo are allowed per patient dose.
Momentum The product of the mass of a body and its velocity; SI units, kg m s
l
.
Monitor A cathode-ray tube device used to view data produced by a computer.
Glossary: Monitoring 7-27
Monitoring Periodic or continuous determination of the amount of ionizing radiation or
radioactive contamination present in an occupied region.
Area Monitoring: Routine monitoring of the radiation level or contamina-
tion of a particular area, building, room, or equipment. Some laboratories
or operations distinguish between routine monitoring and survey activities.
Personnel Monitoring: Monitoring any part of an individual, his breath, or
excretions, or any part of his clothing.
Monitor Ion
Chamber
A special radiation measuring device in which the collected electrical
charge from ionization in a gas filled cavity is taken to be proportional to
some parameter (for example, dose or exposure) or radiation field.
Monitor Unit
(MU)
A unit of radiation exposure. A table for the conversion of monitor units
into units of absorbed dose (gray or rad) can be generated by a dose cali-
bration of the machine by a qualified physicist.
Monoenergetic Having only one energy associated with it.
Monte Carlo
Method
A method permitting the solution by means of a computer of problems of
particle physics, such as those of neutron transport, by determining the
history of a large number of elementary events by the application of the
mathematical theory of random variables.
Motion Enable
Switch
A safety switch that allows motion of certain motorized functions only so
long as the operator continues to press the switch.
Multiple-port
Treatment
Directing more than one radiation beam toward the tumor from different
angles for the purpose of increasing the dose without irrevocably destroying
normal tissue.
Mutation Alteration of the usual hereditary pattern, usually sudden.
National Council
on Radiation
Protection and
Measurements
(NCRP)
This committee was granted a United States Congressional charter in
1964. It is operated as in independent organization financed by contribu-
tions from government, scientific societies, and manufacturing associa-
tions.
Natural (Napier-
ian) Logarithms
A system of logarithms using the base e.
Negative Ion Negative charged ion; commonly termed anion.
Neoplasm Any new and abnormal growth, such as a tumor; neoplastic disease refers
to any disease that forms tumors, whether malignant or benign.
Neutrino A neutral particle of very small rest mass originally postulated to account
for the continuous distribution of energy among particles in the beta decay
process.
Neutron An electrically neutral or uncharged particle of matter existing along with
protons in the atoms of all elements except the mass 1 isotope of hydrogen.
The isolated neutron is unstable and decays with a half-life of about 13
minutes into an electron, proton, and neutrino. Neutrons sustain the fis-
sion chain reaction in a nuclear reactor.
7-28 Glossary: Newton
Epithermal Neutron: A neutron having an energy between 0.5 and 1O
5
eV.
(Sometimes the energy range is given as 0.5 to 100 eV and the energy range
from 100 eV to 10
5
eV is called Intermediate.)
Fast Neutron: A neutron having an energy above 0.1 MeV (105 eV).
Thermal Neutron: A neutron having an energy of about 0.025 eV which
corresponds to a velocity of 2200 m/s.
Newton The unit of force that, when applied to a 1 kilogram mass, will give it an ac-
celeration of 1 meter per second per second. (1 N = 1 kg 1m 1s
2
).
Nomogram Conversion scale between two sets of units.
Nonionizing
Radiation
Radiation that does not cause ionization when it interacts with matter.
Nuclear Reac-
tion
See Reaction, Nuclear.
Nuclear Reactor A device by means of which a fission chain reaction can be initiated, main-
tained, and controlled. Its essential component is a core with fissionable
fuel. It usually has a moderator, a reflector, shielding, and control mecha-
nisms.
Thermal Nuclear Reactor: A nuclear reactor in which the fission chain re-
action is sustained primarily by thermal neutrons. Most existing reactors
are thermal reactors.
Nucleon Common name for a constituent particle of the nucleus. Commonly applied
to a proton or neutron.
Nucleus
(Nuclear)
That part of an atom in which the total positive electric charge and most of
the mass is concentrated.
Nuclide A species of atom characterized by the constitution of its nucleus. The nu-
clear constitution is specified by the number of protons (Z), number of neu-
trons (N), and energy content; or, alteratively, by the atomic number (Z),
mass number A = (N + Z), and atomic mass. To be regarded as a distinct nu-
clide, the atom must be capable of existing for a measurable time. Thus nu-
clear isomers are separate nuclides, whereas promptly decaying excited
nuclear states and unstable intermediates in nuclear reactions are not so
considered.
Occupational
Exposure
The exposure of an individual to ionizing radiation because the occupation
of the individual includes duties or activities that necessarily involve the
likelihood of exposure.
Ohm The unit of electric resistance.
Oncology Preferred name for tumor treatment. (See Therapy, Radiation Therapy.)
Offspring Synonym for Daughter. (See Decay Product.)
Operating Soft-
ware
The integrated collection of programs used by the Clinac system computer
to interface the system with the operator and control the machine.
Glossary: Organ 7-29
Organ Group of tissues which together perform one or more definite functions in a
living body.
Osmosis The passage of pure solvent from the lesser to the greater concentration
when two solutions are separated by a membrane which selectively pre-
vents the passage of solute molecules, but is permeable to the solvent.
Osmotic Pertaining to osmosis.
Owner A person or organization having title to or administrative control over one
or more radiotherapy installations.
Ozone A gas produced by an electrical discharge in ordinary oxygen or air. Pure
ozone is an unstable, faintly bluish gas with a characteristically fresh, pen-
etrating odor.
Pair Production An absorption process for x and gamma radiation in which the incident
photon is annihilated in the vicinity of the nucleus of the absorbing atom
with subsequent production of an electron and positron pair. This reaction
only occurs for incident photon energies exceeding 1.02 MeV. (See also Ab-
sorption, Compton Effect, and Photoelectric Effect.)
Parent A radionuclide that, upon disintegration, yields a specific nuclide either
directly or as a later member of a radioactive series.
Password A number or word used to gain entry to a certain program or a restricted
part of a program.
Path, Mean Free Average distance a particle travels between collisions.
Patient Support
Assembly (PSA)
See Treatment couch.
Penumbra The region, at the edge of a radiation beam, over which the absorbed dose
rate changes rapidly as a function of distance from the axis. It may be de-
fined geometrically and dosimetrically.
Geometric Penumbra: That region in space which could be irradiated by
primary photons or particles coming from part of the source only. By anal-
ogy, the transmission penumbra is the region irradiated by photons or par-
ticles which have traversed part of the thickness of the collimator, i.e., at its
outer edge.
Geometric Penumbra Width: The width of the geometric penumbra in a
plane perpendicular to the beam axis at any distance of interest from the
source. It is a geometrical concept only and is calculated from the expres-
sion
W = c(SSD + d SCD)/SCD
in which c is the source diameter (or effective diameter), SSD + d is the dis-
tance from the source to point of interest, and SCD is the distance from the
source to the edge of the collimator.
Physical Penumbra: This is a dosimetric concept; the physical penumbra
width is the lateral distance between two specified isodose curves at a spec-
ified depth.
7-30 Glossary: Periodic Table
Periodic Table An arrangement of chemical elements in order of increasing atomic num-
ber. Elements of similar properties are placed one under the other, yielding
groups and families of elements. Within each group, a gradation of chemi-
cal and physical properties exists but, in general, chemical behavior is sim-
ilar. From group to group, however, a progressive shift of chemical behavior
occurs from one end of the table to the other.
Permeable Affording passage or penetration.
Personnel Moni-
tor
A dosimeter (usually a film badge, thermoluminescent device, or ionization
chamber) used for determining the exposure to an individual. Such moni-
toring is required for all persons who are radiation workers.
Phantom A volume of material approximating as closely as possible the density and
effective atomic number of tissue. Ideally a phantom should behave in re-
spect to absorption of radiation in the same manner as tissue. Radiation
dose measurements made within or on a phantom provide a means of de-
termining the radiation dose within or on a body under similar exposure
conditions. Some materials commonly used in phantoms are water, per-
spex polystyrene, Masonite, pressed wood, and beeswax.
Phosphores-
cence
Emission of radiation by a substance as a result of previous absorption of
radiation of shorter wavelength. In contrast to fluorescence, the emission
may continue for a considerable time after cessation of the exciting irradi-
ation.
Photoelectric
Effect
Process by which a photon ejects an electron from an atom. All energy of
the photon is absorbed in ejecting the electron and in imparting kinetic en-
ergy to it. (See also Absorption, Compton Effect, and Pair Production.)
Photon A quantity of electromagnetic energy (E) whose value in joules is the prod-
uct of its frequency (v) in hertz and Plancks constant (h). The equation is: E
= hv. (See also Radiation.)
Photosynthesis The production of carbohydrates by green plants in the presence of sun-
light through the agency of chlorophyll.
Physics, Health A science and profession devoted to the protection of man and his environ-
ment from unnecessary radiation exposure.
Pig A lead-lined container used for storing radionuclides.
Plancks Con-
stant
A natural constant of proportionality (h) relating the frequency of a quan-
tum of energy to the total energy of the quantum.
Plateau As applied to radiation detector chambers, the level portion of the counting
rate-voltage curve where changes in operating voltage introduce minimum
changes in the counting rate.
Polycythemia A disease characterized by overproduction of red blood cells.
h
E
v
--- 6.6256 10
34
J s = =
Glossary: Port Film Exposures 7-31
Port Film Expo-
sures
A radiograph taken with the patient interposed between the machine portal
and an x-ray film. The purpose is to demonstrate that the treatment field
on the patient adequately encompasses the treatment volume and at the
same time avoids adjacent critical structures.
Positive Ion Positively charged ion; commonly called cation.
Positron Particle equal in mass to the electron and having an equal but positive
charge.
Potential Differ-
ence
Work required to carry a unit positive charge from one point to another.
Potential Ioniza-
tion
The potential necessary to separate one electron from an atom, resulting in
the formation of an ion pair.
Power, Stopping A measure of the effect of a substance upon the kinetic energy of a charged
particle passing through it.
Printed Circuit
Board (PCB)
A sandwich-like set of insulated boards onto which circuits are etched and
various components are soldered.
Proportional
Region
Voltage range in which the gas amplification is greater than one, and in
which the charge collected is proportional to the charge produced by the
initial ionizing event.
Proton Elementary nuclear particle with a positive electric charge equal numeri-
cally to the charge of the electron and a rest mass of 1.67474 10
27
kg.
Pulse Forming
Network (PFN)
An electrical circuit in the modulator that supplies accurately shaped
pulses of high voltage necessary for klystron operation.
Pulse Height
Selector
A circuit designed to select and pass voltage pulses in a certain range of
amplitudes.
Pulse Repeti-
tion Frequency
(PRF)
A signal applied to the thyratron tube that causes the capacitors in the
pulse-forming network to discharge. This provides a nearly flat high-power
dc pulse to the electron gun and the klystron of the accelerator.
Purpura Large hemorrhagic spots in or under the skin or mucous tissues.
Quality (Radiol-
ogy)
The characteristic spectral-energy distribution of x radiation. It is usually
expressed in terms of effective wavelengths of half-value layers of a suitable
material; e.g., up to 20 kV, cellophane; 20 to 120 kVp, aluminum; 120 to
400 kVp, copper; over 400 kVp, tin.
Quality Factor
(Q)
The linear-energy-transfer-dependent factor by which absorbed doses are
multiplied to obtain (for radiation protection purposes) a quantity that ex-
presses, on a common scale for all ionizing radiations, the effectiveness of
the absorbed dose.
Quantum An observable quantity is said to be quantized when its magnitude is, in
some or all of its range, restricted to a discrete set of values. If the magni-
tude of the quantity is always a multiple of a definite unit, that unit is
called the quantum (of the quantity). For example, the quantum or unit of
orbital angular momentum is h, and the quantum of energy of electromag-
7-32 Glossary: Quantum Theory
netic radiation of frequency v is hv. In field theories, a field (or the field
equations) is quantized by application of a proper quantum mechanical
procedure. This quantization results in the existence of a fundamental field
particle, which may be called the field quantum. Thus, the photon is a
quantum of the electromagnetic field, and in nuclear field theories the me-
son is considered the quantum of the nuclear field.
Quantum The-
ory
The concept that energy is radiated intermittently in units of definite mag-
nitude called quanta, and absorbed in a like manner.
Quenching The process of inhibiting continuous or multiple discharge in a counter
tube which uses gas amplification.
Quenching
Vapor
Polyatomic gas used in Geiger-Muller counters to quench or extinguish av-
alanche ionization.
Rad The former unit of absorbed dose equal to 0.01 joule per kilogram in any
medium. (See also Absorbed Dose and Tissue Dose)
Radiant Energy The energy of electromagnetic radiation, such as radio waves, visible light,
x and gamma rays.
Radiation (l) The emission and propagation of energy through space or through a ma-
terial medium in the form of waves; for instance, the emission and propa-
gation of electromagnetic waves, or of sound and elastic waves. (2) The en-
ergy propagated through space or through a material medium as waves; for
example, energy in the form of electromagnetic waves or elastic waves. The
term radiation or radiant energy, when unqualified, usually refers to elec-
tromagnetic radiation. Such radiation commonly is classified, according to
frequency, as hertzian, infrared, visible (light), ultraviolet and x-ray or
gamma-ray. (See also Photon.) (3) By extension, corpuscular emissions,
such as alpha and beta radiation, or rays of mixed or unknown type, as
cosmic radiation.
Annihilation Radiation: Photons produced when an electron and a
positron unite and cease to exist. The annihilation of a positron-electron
pair results in the production of two photons, each of 0.511 MeV energy.
Background Radiation: Radiation arising from radioactive material other
than the one directly under consideration. Background radiation due to
cosmic rays and natural radioactivity is always present. Background radi-
ation may also be due to the presence of radioactive substances in other
parts of the building or in the building material itself.
Characteristic (Discrete) Radiation: Radiation originating from an atom
after removal of an electron or excitation of the nucleus. The wavelength of
the emitted radiation is specific, depending only on the nuclide and partic-
ular energy levels involved.
External Radiation: Radiation from a source outside the body the radi-
ation must penetrate the skin.
Internal Radiation: Radiation from a source within the body (as a result of
deposition of radionuclides in body tissues).
Ionizing Radiation: Any electromagnetic or particulate radiation capable
of producing ions, directly or indirectly, in its passage through matter.
Glossary: Radiation Beam 7-33
Leakage (Direct) Radiation: All radiation coming from the source housing
except the useful (primary) beam.
Monochromatic Radiation: Electromagnetic radiation of a single wave-
length, or radiation in which all the photons have the same energy.
Monoenergetic Radiation: Radiation of a given type (e.g., alpha, beta,
neutron, gamma) in which all particles or photons originate with and have
the same energy.
Primary Radiation: The useful beam of an x-ray tube.
Scattered Radiation: Radiation that, during its passage through a sub-
stance, has been deviated in direction. It may also have been modified by a
decrease in energy.
Secondary Radiation: Radiation resulting from absorption of other radia-
tion in matter. It may be either electromagnetic or particulate.
Radiation Beam The flow of therapeutically useful radiation energy through a defined area.
Radiation Pro-
tection Survey
An evaluation of the radiation hazards in and around an installation.
Radiation
Oncologist
A physician who has received specific training and experience in therapeu-
tic radiology.
Radiation Sick-
ness
(Radiation Therapy): A self-limiting syndrome characterized by nausea,
vomiting, diarrhea, and psychic depression, following exposure to appre-
ciable doses of ionizing radiation, particularly to the abdominal region. It
usually appears a few hours after irradiation and may subside within a
day. It may be sufficiently severe to necessitate interrupting the treatment
series or to incapacitate the patient.
(General): The syndrome associated with intense acute exposure to ioniz-
ing radiations.
Radiation Ther-
apist
An individual who has received specific training in radiation therapy tech-
nology and who is certified by a recognized specialty board as being compe-
tent in radiation therapy technology.
Radiation Ther-
apy
Treatment of disease with any type of radiation.
Radioactivity The property of certain nuclides of (l) spontaneously emitting particles or
gamma radiation or (2) emitting x radiation following orbital electron cap-
ture or (3) undergoing spontaneous fission.
Artificial Radioactivity: Man-made radioactivity produced by particle
bombardment or electromagnetic irradiation, as opposed to natural radio-
activity.
Induced Radioactivity: Radioactivity produced in a substance after bom-
bardment with neutrons or other particles. The resulting activity is natu-
ral radioactivity if formed by nuclear reactions occurring in nature, and
artificial radioactivity if the reactions are caused by man.
Natural Radioactivity: The property of radioactivity exhibited by more
than 50 naturally occurring radionuclides.
7-34 Glossary: Radiobiology
Radiobiology That branch of biology which deals with the effects of radiation on biologi-
cal systems.
Radio Fre-
quency (rf)
Any frequency at which coherent electromagnetic radiation of energy is
possible. Usually considered to denote frequency above 150 kilohertz and
extending up to the infrared range.
Radiography The making of shadow images on photographic emulsion by the action of
ionizing radiation. The image is the result of the differential attenuation of
the radiation in its passage through the object being radiographed.
Radiological
Physicist
An individual who devotes the majority of occupational time to the physics
of radiology, including therapeutic radiological physics, diagnostic radio-
logical physics, and medical nuclear physics.
Radiology That branch of medicine which deals with the diagnostic and therapeutic
applications of radiant energy, including x-rays and radionuclides.
Radionuclide A nuclide that displays the property of radioactivity.
Radiopharma-
ceutical
A pharmaceutical compound that has been tagged with a radionuclide.
Radioresistance Relative resistance of cells, tissues, organs, or organisms to the injurious
action of radiation. The term may also apply to chemical compounds or to
any substances.
Radiosensitivity Relative susceptibility of cells, tissues, organs, organisms, or any living
substances to the injurious action of radiation. Radioresistance and radi-
osensitivity are currently used in a comparative sense, rather than in an
absolute one.
Radiotherapy
Accelerator Ser-
vice Technician
An individual with the following minimum qualifications: training equiva-
lent to an associates degree in electronics, training in servicing and main-
taining the machine, and a demonstrated understanding of the emergency
and safety regulations adopted by the owner of the accelerator.
Range The depth in any material measured from the entrance of an ionizing par-
ticle to the stopping position of that particle after it has lost all of its energy.
Rare Earth Any of the series of very similar metals ranging in atomic numbers from 57
through 71.
Reaction
(Nuclear)
An induced nuclear disintegration, i.e., a process occurring when a nu-
cleus comes in contact with a photon, an elementary particle, or another
nucleus. In many cases the reaction can be represented by the symbolic
equation: X + a Y + b or, in abbreviated form, X(a,b) Y. X is the target nu-
cleus, a is the incident particle or photon, b is an emitted particle or pho-
ton, and Y is the product nucleus.
Recombination The return of an ionized atom or molecule to the neutral state.
Recovery Rate The rate at which recovery takes place after radiation injury. It may proceed
at different rates for different tissues. Differential recovery rate: Among
tissues recovering at different rates, those having slower rates will ulti-
mately suffer greater damage from a series of successive irradiations. This
Glossary: Regenerative Process 7-35
differential effect is considered in fractionated radiation therapy if the neo-
plastic tissues have a slower recovery rate than surrounding normal struc-
tures.
Regenerative
Process
The process by which damage or destroyed cells are replaced by new ones
of the same type.
Relative Bio-
logic Effective-
ness (RBE)
The factor used to compare the biologic effectiveness of absorbed radiation
doses (i.e., rads) due to different types of ionizing radiation; more specifi-
cally, it is the experimentally determined ratio of an absorbed dose of radi-
ation in question to the absorbed dose of a reference radiation required to
produce an identical biologic effect in a particular experimental organism
or tissue.
This term should not be used in radiation protection. (See also
Quality Factor.)
Rem A special unit of dose equivalent. The dose equivalent in rem is numerically
equal to the absorbed dose in rad multiplied by the quality factor, the dis-
tribution factor, and any other necessary modifying factors.
Repair The partial or complete restoration of functional integrity in cells following
damage caused by radiation. Operationally, repair means that after irradi-
ation a cell responds as though it had received a smaller dose than under
conditions in which damage is more fully expressed. The ability to observe
repair implies, therefore, that a comparison is made with a treatment of ref-
erence. Full repair indicates that cells respond as though they had not been
previously irradiated. (Repair embraces processes sometimes referred to as
bypassing of damage, shedding of damage, compensating for damage, elim-
ination of damage, and/or the specific biochemical reversal of damage.)
Resolving Time,
Counter
The minimum time interval between two distinct events which will permit
both to be counted. It may refer to an electronic circuit, a mechanical indi-
cating device, or a counter tube.
Rest Mass The intrinsic mass of any physical entity; the mass possessed by that entity
apart from any motion it may have.
Roentgen (R) The special unit of exposure. One roentgen equals 2.58 10
4
coulomb per
kilogram of air. (See also Exposure.)
Roentgenogra-
phy
Radiography by means of x-rays.
Roentgenology That part of radiology which pertains to x-rays.
Roentgen Rays X-rays.
Scattering Change of direction of subatomic particles or photons as a result of a colli-
sion or interaction.
Coherent Scattering: Scattering of photons or particles in which definite
phase relationships exist between the incoming and the scattered waves.
Coherence manifests itself in the interference between the waves scattered
7-36 Glossary: Scattering Coefficient, Compton
by two or more scattering centers. An example is the Bragg scattering of x-
rays and neutrons by the regularly spaced atoms in a crystal, for which
constructive interference occurs only at definite angles, called Bragg an-
gles.
Compton Scattering: The scattering of a photon by an electron. Part of the
energy and momentum of the incident photon is transferred to the electron,
and the remaining part is carried away by the scattered photon.
Elastic Scattering: Scattering caused by elastic collisions and, therefore,
conserving kinetic energy of the system. Rayleigh scattering is a form of
elastic scattering.
Incoherent Scattering: Scattering of photons or particles in which the
scattering elements act independently of one another; no definite phase re-
lationships exist among the different parts of the scattered beam. The in-
tensity of the scattered radiation at any point is obtained by adding the in-
tensities of the scattered radiation reaching this point from the
independent scattering elements.
Inelastic Scattering: The type of scattering that results in the nucleus be-
ing left in an excited state and the total kinetic energy being decreased.
Scattering Coef-
ficient, Comp-
ton
That fractional decrease in the energy of a beam of x or gamma radiation in
an absorber due to the energy carried off by scattered photons in the
Compton effect. (See also Compton Absorption Coefficient.)
Scattering Foil In electron beam therapy, a thin metal plate used to disperse the electron
beam before it passes through the collimator jaws. The function of the scat-
tering foil is to flatten intensity over the field, analogous to the function of
the flattening filter with x-rays.
Scintillation
Counter
An instrument that detects and measures ionizing radiation by counting
the light flashes (scintillations) induced by the radiation in certain materi-
als.
Sealed Source A radioactive source sealed in an impervious container which has sufficient
mechanical strength to prevent contact with and dispersion of the radioac-
tive material under the conditions of use and wear for which it was de-
signed.
Series, Radioac-
tive
A succession of nuclides, each of which transforms by radioactive disinte-
gration into the next until a stable nuclide results. The first member is
called the parent, the intermediate members are called daughters, and
the final stable member is called the end product.
Shield A body of material used to prevent or reduce the passage of particles or ra-
diation. A shield may be designated according to what it is intended to ab-
sorb (as a gamma-ray shield or neutron shield), or according to the kind of
protection it is intended to give (as a background, biologic, or thermal
shield). The shield of a nuclear reactor is a body of material surrounding
the reactor to prevent the escape of neutrons and radiation into a protected
area, which frequently is the entire space external to the reactor. It may be
required for the safety of personnel or to reduce radiation enough to allow
use of counting instruments for research or for locating contamination or
airborne radioactivity.
Glossary: Sievert 7-37
Sievert The new special unit of dose equivalent. The sievert equals the absorbed
dose in gray times the quality factor for the radiation in question. (Symbol:
Sv)
Sigmoid Curve S-shaped curve, often characteristic of a dose-effect curve in radiobiological
studies.
Simulation Use of a simulator to determine the various treatment field outlines and
orientations to be used during radiation therapy.
Simulation
Films
X-ray films taken on a simulator with the same field size, target-to-skin dis-
tance, and orientation as a therapy beam.
Simulator A radiation generator operating in the diagnostic x-ray range that can di-
rect a radiation beam toward a patient with parameters imitating those
proposed for therapy, and providing direct x-ray fluoroscopic visualization
and roentgenographic images. The simulator x-rays do not contribute to
the therapy dose required.
Skin Dose Absorbed dose at the center of the irradiation field on skin. It is the sum of
the dose in air and scatter from body parts.
Softness A relative specification of the quality or penetrating power of x-rays. In gen-
eral, the longer the wave length the softer the radiation.
Somatic Effects Effects that may become evident in the irradiated individual.
Source Synonymous with Target.
Source Axis Dis-
tance (SAD)
Synonymous with Target axis distance (TAD).
Source Film Dis-
tance (SFD)
Synonymous with Target-film distance (TFD).
Source Surface
Distance (SSD)
The distance measured along the beam axis, from the front surface of the
source to the surface of the irradiated object Synonymous with TSD.
Specific Activity Total activity of a given nuclide per gram of a compound, element, or radio-
active nuclide.
Specific
Gamma-ray Con-
stant
For a nuclide emitting gamma radiation, the product of exposure rate at a
given distance from a point source of that nuclide and the square of that
distance divided by the activity of the source, neglecting attenuation.
Spectrum A visual display, a photographic record, or a plot of the distribution of the
intensity of radiation of a given kind as a function of its wavelength, energy,
frequency, momentum, mass, or any related quantity.
Split Course A course of radiotherapy delivered in two parts separated by a rest period of
several weeks.
Standard Something established as a measure; a model to which other similar things
should conform.
7-38 Glossary: Sterility (Biological)
Standard, Radioactive: A sample of radioactive material, usually with a
long half-life, in which the number and type of radioactive atoms at a defi-
nite reference time are known. It may be used as a radiation source for cal-
ibrating radiation measurement equipment.
Sterility (Biolog-
ical)
Temporary or permanent incapability to reproduce.
Sublethal Dam-
age
Cellular damage, the accumulation of which, may result in lethality.
Survey, Radio-
logic
Evaluation of the radiation hazards incident to the production, use, or ex-
istence of radioactive materials or other sources of radiation under specific
conditions. Such evaluation customarily includes a physical survey of the
disposition of materials and equipment, measurements or estimates of the
levels of radiation that may be involved, and sufficient knowledge of pro-
cesses using or affecting these materials to predict hazards resulting from
expected or possible changes in materials or equipment.
Survey Meter
(Laboratory
Monitor)
A detection instrument used to monitor an area for unsuspected radiation
or to search for a lost radiation source or contamination.
Syndrome The complex of symptoms associated with any disease.
Target A metal plate placed in the beam of high-speed electrons to produce x-rays.
For electron therapy, the target is retracted from the beam.
Target Axis Dis-
tance (TAD)
The distance measured along the central axis from the center of the front
surface of the target to the isocenter.
Target Film Dis-
tance (TFD)
surface of the target to an x-ray film.
Target Skin Dis-
tance (TSD)
surface of the target to the surface of the irradiated object.
Target Theory
(Hit Theory)
A theory explaining some biological effects of radiation on the basis that
ionization, occurring in a discrete Volume (the target) within the cell, di-
rectly causes a lesion which subsequently results in a physical response to
the damage at that location. One, two, or more hits (ionizing events within
the target) may be necessary to elicit the response.
Tenth-Value
Layer (Tenth-
Value Thick-
ness) (TVL)
The thickness of a specified substance which, when introduced into the
path of a given beam of radiation, reduces the kerma rate by ten.
Therapy Medical treatment of a disease.
Brachytherapy (therapy at short distances): The treatment of disease
with sealed radioactive sources placed near, or inserted directly into, the
diseased area.
Contact Radiation Therapy: X-ray therapy with specially constructed
tubes in which the target-skin distance is very short (less than 2 cm). The
voltage is usually 40 to 60 kV.
Glossary: Thermionic Emission 7-39
Radiation Therapy: Treatment of disease with any type of radiation.
Rotation Therapy: Radiation therapy during which either the patient is ro-
tated in front of the source of radiation or the source is revolved around the
patient. In this way, a larger dose is built up at the center of rotation within
the patients body than on any area of the skin.
Teletherapy (therapy at long distance): The treatment of disease with
gamma radiation at a distance from the patient.
Thermionic
Emission
Release of electrons from the cathode filament by heat.
Thermolumines-
cent Dosimetry
A method of determining dose by exposing certain phosphoric materials to
radiation, and then heating the materials and measuring the light emitted.
The luminescence is proportional to the dosage delivered.
Threshold, Pho-
toelectric
The quantum of energy hv0 that is just enough to release an electron from
a given system in the photoelectric effect. The corresponding frequency, v
O
,
and wavelength, 8
O
, are the threshold frequency and wavelength respec-
tively. For example, in the surface photoelectric effect, the threshold hv0 for
a particular surface is the energy of a photon which, when incident on the
surface, causes the electron to emerge with zero kinetic energy.
Thyratron A gas-filled electron tube in which the grid controls only the start of a con-
tinuous current, giving the tube a trigger action. A thyratron is used to dis-
charge the pulse-forming network in the modulator.
Tissue Equiva-
lent Material
Material made up of the same elements in the same proportions as they oc-
cur in a particular biologic tissue. In some cases, the equivalence may be
approximated with sufficient accuracy on the basis of effective atomic num-
ber.
Townsend Ava-
lanche
(See Avalanche.)
Track Visual manifestation of the path of the ionizing particle in a chamber or
photographic emulsion.
Transformer An electrical device for increasing or decreasing the incoming voltage.
Transmutation Any process in which a nuclide is transformed into a different nuclide, or
more specifically, when transformed into a different element by a nuclear
reaction.
Transport Group This is any one of seven groups into which normal form radionuclides are
classified according to their radiotoxicity and potential hazard in transpor-
tation.
Transport Index The number to be placed on a package label to designate the degree of con-
trol to be exercised by the carrier during transportation and indicating the
following: (l) the highest radiation absorbed dose equivalent rate in micro-
Sievert per hour at three feet from any accessible external surface of the
package, or (2) for Fissile Class II packages only, the number calculated by
7-40 Glossary: Treatment Beam Parameters
dividing the number 50 by the number of similar packages that may be
transported together.
Treatment
Beam Parame-
ters
The data required for complete specification of an individual treatment
beam, including radiation energy, field size, use of wedges and blocks, ori-
entation with respect to patient, and prescribed exposure time, dose, and
distance.
Treatment
Couch
The Clinac unit that supports the patient during therapy.
Treatment Field A plane section of a beam, perpendicular to the beam axis, as defined by
the collimator.
Treatment Head The section of the Clinac gantry from which the treatment beam exits. The
treatment head includes the carrousel, collimator, an ionization chamber,
the range finder and field-defining light, and other supporting components.
Treatment Plan An ensemble of radiation beams or sources designed to produce a pre-
scribed dose pattern in and for the patient; includes spatial and temporal
distributions.
Treatment Plan-
ning
A complex process carried out prior to the administration of radiation ther-
apy. The planning process usually includes such items as tumor localiza-
tion, treatment volume determination, contour preparation, and treatment
dose determination to prescribe the dosage pattern required.
Treatment
Room
An enclosed space specially designed and dedicated for patient treatment
by a radiotherapy accelerator.
Treatment Type Standard and specialized therapies available on the Clinac 2100C, includ-
ing fixed x-ray, fixed electrons, arc x-ray, port film exposures, arc electrons,
total body x-ray, total body electrons, and high dose rate total skin elec-
trons.
Tritium (T) The hydrogen isotope with one proton and two neutrons in the nucleus.
Tube, Photomul-
tiplier
An electron multiplier tube in which the electrons initiating the cascade
originate by photoelectric emission.
Umbra The region within the beam receiving the full strength of the primary x-ray
or gamma-ray photons.
Uncontrolled
Area
An area not under the authority of the Radiation Protection Officer and not
subject to restriction due to the presence of radiation.
Valence Number representing the combining or displacing power of an atom; num-
ber of electrons lost, gained, or shared by an atom in a compound; number
of hydrogen atoms with which an atom will combine, or which it will dis-
place.
Van De Graaff
Accelerator
An electrostatic machine in which electrical charge is carried into the high
voltage terminal by a belt made of an insulating material moving at a high
speed. The particles are then accelerated along a discharge path through a
vacuum tube by the potential difference between the insulated terminal
and the grounded end of the accelerator.
Glossary: Velocity 7-41
Velocity The ratio of displacement to the time required for this displacement. Time
rate of motion in a given direction and sense. Average velocity equals the to-
tal distance passed over, divided by the whole time taken.
Video Display See Monitor
Volt The unit of electric pressure or electromotive force; the force necessary to
cause 1 ampere of current to flow against 1 Ohm of resistance.
Voltage The potential difference, in volts, between two different points in an electric
circuit or between two different electrodes.
Voltage, Operating: As applied to radiation detection instruments, the
voltage across the electrodes in the detecting chamber required for proper
detection of an ionizing event.
Voltage, Starting: For a counter tube, the minimum voltage that must be
applied to obtain counts with the particular circuit with which it is associ-
ated.
Volume, Sensi-
tive
That portion of a counter tube or ionization chamber which responds to a
specific radiation.
Watt The unit of power equal to 1 joule per second (1W = 1J/s).
Waveguide Metal pipe of rectangular or circular cross section that transfers radio-fre-
quency energy to the accelerator.
Wavelength Distance between any two similar points of two consecutive waves ( ). For
electromagnetic radiation, the wavelength is equal to the velocity of light (c)
divided by the frequency of the wave (v), = c/v. The effective wavelength
is the wavelength of monochromatic x-rays that would undergo the same
percentage attenuation in a specified filter as the heterogeneous beam un-
der consideration.
Wave Motion The transmission of a periodic motion or vibration through a medium or
empty space. Transverse: Wave motion in which the vibration is perpendic-
ular to the direction of propagation. Longitudinal: Wave motion in which
the vibration is parallel to the direction of propagation.
Wedge Filter A tapered block of attenuating material, designed to produce wedge shaped
isodose curves.
Window (1) A plate, usually made of ceramic or glass, placed across the microwave
waveguide to separate the air in the waveguide from the vacuum in the lin-
ear accelerator and the klystron. (2) A plate, made of thin aluminum or be-
ryllium, through which electrons are extracted. (3) A rectangular portion of
a monitor screen used for input from the operator or to display information
to the operator.
1V 1J C
1
=
7-42 Glossary: X-rays

X-rays Penetrating electromagnetic radiations whose wavelengths are shorter than
those of visible light. They are usually produced by bombarding a metallic
target with fast electrons in a high vacuum. In nuclear reactions, it is cus-
tomary to refer to photons originating in the nucleus as gamma-rays, and
those originating in the extranuclear part of the atom as x-rays. These rays
are sometimes called roentgen rays after their discoverer, W.C. Roentgen.
Some of these definitions are reproduced with permission of
the U.S. Department of Health Education and Welfare Public
Health Service Food and Drug Administration Bureau of Ra-
diological Health. from the Radiological Health Handbook rev.
ed. 1970.
Index 8-i
This chapter contains an index of terms used in the C-series Clinac Accelerator System
Basics manual.
Index
8-ii Index
Index
Numerics
3dB Quadrature Hybrid ................................................................ 4-21
A
Advances in Linear Accelerator Design for Radiotherapy ............... 2-19
B
Beam Trans port Magnet Systems ................................................. 2-44
C
Circulators ................................................................................... 4-22
D
Definitions, Intensity vs. Energy ..................................................... 2-3
E
Electron Injection and Bunching ................................................... 2-18
Emergency and Safety .................................................................... 1-1
Emergency Off Button .................................................................. 1-10
Enabling emergency pendant ........................................................ 1-11
F
Fill Time ....................................................................................... 2-16
G
Gas Flow, defined ......................................................................... 6-17
Gas Laws ...................................................................................... 6-13
Gas, defined ................................................................................... 6-7
Index 8-iii
I
Injection Timing ............................................................................ 2-17
Ion Chamber Theory ....................................................................... 6-1
Ion Pump, defined ......................................................................... 6-21
K
Karzmark, Dr. C. J. ....................................................................... 2-19
Kinetic Energy Relationships ........................................................... 2-3
Klystron Theory ............................................................................ 4-24
L
Load Line Considerations .............................................................. 2-16
M
Machine Physics ............................................................................. 2-1
Microtrons .................................................................................... 2-42
Microwave Accelerator Structures ................................................. 2-20
Modulator Theory ........................................................................... 3-1
N
Nature of Vacuum ........................................................................... 6-3
P
Pendant, emergency ...................................................................... 1-11
Pressure, defined ............................................................................ 6-7
R
Resonant Circuits ......................................................................... 4-14
Rest Energy Relationships ............................................................... 2-4
RF Theory ....................................................................................... 4-1
8-iv Index
RF Transmission Theory ............................................................... 4-16
S
Standing Wave Accelerator ............................................................. 2-9
T
TE10 Mode ................................................................................... 4-18
Temperature, defined ...................................................................... 6-6
Thermocouple Gauge .................................................................... 6-26
Thyratron Theory .......................................................................... 3-19
Total Energy Relationships ............................................................. 2-4
Transmission Lines ........................................................................ 4-5
V
Vacuum Gauges ........................................................................... 6-26
Vacuum, defined ............................................................................ 6-3
Vapor Pressure, defined ................................................................ 6-10
W
Widely Variable Energy Linacs ...................................................... 2-33

Accelerator Basics, Rev. AF

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Accelerator Basics, Rev. AF

Загружено:

Авторское право:

Доступные форматы

COPYRIGHT 2005 VARIAN MEDICAL SYSTEMS

L FOR TRAINING PURPOSES ONLY 7

) which weighs 9.1 10

COPYRIGHT 2005 VARIAN MEDICAL SYSTEMS

2-20 Machine Physics: Advances in Linear Accelerator Design for Radiotherapy

2-22 Machine Physics: Advances in Linear Accelerator Design for Radiotherapy

t toward the anode. Both of these motions represent the move-

t where t is the thickness of the material and is the

7-8 Glossary: Calibration

COPYRIGHT 2005 VARIAN MEDICAL SYSTEMS

7-42 Glossary: X-rays

Вам также может понравиться