PDF Cover Page

Engineering and Intuition Serving the Soul of Music Perkins Electro-Acoustic Research Lab, Inc.
86008, 2106 33 Ave. SW, Calgary, AB; CAN T2T 1Z6
Ph: +. 1. 403. 244. 4434 Fx: +. 1. 403. 245. 4456
Web: http: // www. pearl - hifi . com
E-mail: custserv@pearl - hifi . com
Inc.
Please note that the links in the PEARL logotype above are live
and can be used to direct your web browser to our site or to
open an e-mail message window addressed to ourselves.
To view our item listings on eBay, click here.
To see the feedback we have left for our customers, click here.
This document has been prepared as a public service . Any and all trademarks
and logotypes used herein are the property of their owners.
It is our intent to provide this document in accordance with the stipulations with
respect to fair use as delineated in Copyrights - Chapter 1: Subject Matter and
Scope of Copyright; Sec. 107. Limitations on exclusive rights: Fair Use.
Public access to copy of this document is provided on the website of Cornell Law School
at http://www4.law.cornell.edu/uscode/17/107.html and is here reproduced below:
Sec. 107. - Limitations on exclusive rights: Fair Use
Notwithstanding the provisions of sections 106 and 106A, the fair use of a copyrighted work, includ-
ing such use by reproduction in copies or phono records or by any other means specied by that section,
for purposes such as criticism, comment, news reporting, teaching (including multiple copies for class-
room use), scholarship, or research, is not an infringement of copyright. In determining whether the use
made of a work in any particular case is a fair use the factors to be considered shall include:
1 - the purpose and character of the use, including whether such use is of a
commercial nature or is for nonprot educational purposes;
2 - the nature of the copyrighted work;
3 - the amount and substantiality of the portion used in relation to the copy-
righted work as a whole; and
4 - the effect of the use upon the potential market for or value of the copy-
righted work.
The fact that a work is unpublished shall not itself bar a nding of fair use if such nding is made
upon consideration of all the above factors
Verso Filler Page
AN 23.1.1 - Page 1 ! !
I
T SEEMS TO ME that somewhere along the way
almost everyone who considers himself an
audiophile has dreamed of owning fabulous-sound-
ing, high-power amplication. Unfortunately the
retail cost of such amps is so high that only a very
few people are ever able to realize such aspirations.
As with anything worth owning, the cost to design
and build good electronics is unavoidably high. Add
to that the prot margin that a retailer must secure
in order to make his business worthwhile and the
cost of great amps is in the vicinity of a decent car or
a down payment on a house.
During the 1950s and the early 60s there were all
kinds of kits available to the home builder and
although none of them was particularly high pow-
ered by todays standards, a few were then consid-
ered to be monsters. They were eagerly purchased,
constructed and enjoyed by the literal thousand.
How many Dynaco Stereo 70s are out there, giving
pleasure to this day?
With the demise of the tube in mainstream
audio during the late 60s, the rise of the transistor
and the onslaught of i nexpensi vely produced,
aggressively marketed and heavily hyped J apanese
equipment, the needs of the hobbyist were somehow
forgotten. Marketi ng became the name of the
game well, you know the rest.
The fact that many capable people would much
prefer to build a thing with their own hands than
buy it at retail seems to have been obscured almost
to the point of obliteration by self-interested manu-
facturers and their wallet-hovering marketeers.
The high-end audio-market has matured to a
state where superb products are sold for enormous
sums but not a single high-end kit is to be found.
I n spite of the considerable cost of the parts
required to build it, I believe there is a real market
for an honest-to-God, high-power amplier kit. My
plan is to make available all of the unique and
difcult-to-source parts required, such as the trans-
former set, painted and silk-screened chassis, pc-
boards, power supply caps, tubes, I so-sockets, I so-
mounts for the output tubes, etc. The sourcing of the
more usual parts such as resistors and capacitors
will be left to the individual builder. To make life as
easy as possible we will supply pre-printed order
forms made out to certain suppliers such as Digi-
Key, Sonic Frontiers and others so that all that is
required to get the common parts is to send the form
to the supplier along with the necessary funds.
I would like to form an SC280 Builders Club and
publish a newsletter every so often containing infor-
mation, tips, mods and different circuit ideas. I fully
expect that some of these kits will be bought by peo-
ple with design ideas quite different from mine, it
should be truly interesting to see what other people
come up with. With enough people building these
amps and tinkering away at them, it should be pos-
sible to offer a kit and associated circuitry that is
simply the best available.
Having invested several years work in the devel-
opment of the circuitry presented in this Audio
Note, I think the current design is a well-balanced
approach to the many inter-related problems such
an amplier presents. The power supplies in partic-
ular are well tried and sonically proven not only by
myself, but by other designers who arrived at the
same design conclusions independently and in a
couple of cases, several years before I did.
This effort has received input from many people
and is by no means solely my work. The part for
which I am responsible is the way in which all of the
various ideas have been assembled, the electrical lay-
out and the mechanical, thermal and aesthetic design.
While I was working on this and several other
tube-related projects, I undertook a massive and
thorough-going literature search that now runs to
something on the order of about a nine hundred arti-
cles. Many times Id be thinking about this amp and
dream up what was sure to be a new idea only to later
nd that someone had done it in 1956 or some such!
Prior art notwithstanding, I can say with a rea-
sonable amount of condence that no one has pro-
duced a commercial offering that uses quite the
same topology presented here. Given the cost of this
circuit and the fact that high-end audio is only really
about ten years old, its not too surprising that this
could be the case. There was essentially no market
Notes on aWork in Progress
The PEARL SC 280March, 90.
Hand-Builders of Fine Music-Reproduction Equipment Precision Electro-Acoustic Research Laboratory.
2106 33 Ave. SW, Calgary, AB; CAN T2T 1Z6
Ph: +.1.403.244.4434 Fx: +.1.403.244.7134
Web: http://www.pearl -hifi .com
E-mail: custserv@pearl -hifi .com
AN 23.1.1 - Page 2 ! !
in 1955 for a $1000.00 amplier.
I have invested hundreds of hours of work into
this project so far and have tried to design the cir-
cuits to be as simple, reliable and trouble-free as
possible. Tube-aging effects are essentially nullied
in the phase-inverter/driver stages by the use of a
form of feedback known as active error-correc-
tion
4
whereby the differential ampliers are made
permanently self-balancing. The output stage uses
two, manual bias-balance controls that operate in
conjunction with a third control that allows the
operating point of the amplifier to be continuously
varied over a wide range.
The tube types have been chosen on the basis of
linearity, sonics and on-going availability. The output
stage uses EL34s, while the front-end uses 3 6CG7s,
1- 5687, and two FETs. The EL34, a pure pentode, is
renown for its sweet upper registers and good, if not
overpowering bottom end response. Unlike the 6550,
this tube is not a high perveancehigh current-out-
putdevice and so must be run at a fairly high plate
voltage and rather lower plate current for a given
amount of power output. While the 6550 certainly
belts out the bottom end, the treble is often a little
edgy and this being a distortion form I cannot stand,
Ive chosen to run more EL34s at higher plate voltage.
The EL34has been produced in some European plants
for at least 30 years and one can reasonably suppose
that the bugs are out of the production processes by
now. The 6550 is presently made only in China. With
the recent demise of the Sylvania plant in Emporium,
Pa. and the GE (MPD) plant in Owensboro, Ky., there
isnt really a source of what I consider to be good
sounding, consistent and reliable 6550s. Im told that
while the Chinese items are sonically quite acceptable,
reliability can be an issue. There are apparently about
30-odd plants in China, all of which market through
the same agency, so one never really knows the origin
of a given lot of tubes. I am not out to belittle the Chi-
nese tubes, so if anyone has more accurate informa-
tion than the info I have, please ll me in!
I hope that this offering strikes a responsive
chord within some of you as I feel that building and
listening to a pair of these amps will be a richly
rewarding experience in every way. I d like to give
credit to some of the people who have helped with
the design of this amp because without their assis-
tance and encouragement this project would not
likely have been undertaken. I n no particular order,
thanks to:
Norman Crowhurst, for 20 years of the best
technical writing from which its been my
pleasure to learn.
J ack Senecal , ex of Preci si on Fi del i ty
presentl y consul tant to K rel l . We have
enjoyed many discussions relating to tube
equipment design over the years.
Chris Paul, for much help with circuit analysis
and for putting before me the active plate load
circuit used in this amplier.
Ed Logan, Logan Labs, for telling me how to
build a tube, power-output stage, and provid-
ing many useful and hard-to-come-by tidbits
of information.
Bruce DePalma, ex of Dynaco during the PAS,
Stereo 70 and Mklll era. Bruce pointed out the
value of running EL34s at high plate voltage
and low current and provi ded some very
timely encouragement.
Ken Stevens, Convergent Audio Technology:
weve enjoyed hours of lively and enlivening
discussion relating to tube audio in general
and to the operation of triode output stages in
particular.
Ed Meitner, Museatex and Meitner. Ed pro-
vided me with a lot of insight into dielectrics
and many of other peculiar phenomena, he
also coined the unforgettable phrase I nca
Firebottles.
the hundreds of people who wrote the articles
that constitute archive library I have built up
over the course of my literature search.
THE SI NGLE CHANNEL 280
Perhaps the single most daunting aspect of own-
ing a large tube amplier is the near inevitability of
ongoing and costly tube replacement. Hand in glove
with this is the steadily diminishing sonic quality
shown by aging tubes. The reason for owning tube
gear in the rst place is that it sounds better than
anything else and this fact makes tube deterioration
doubly frustrating.
I own and use an enormous amount of equip-
ment of many types; a complete and well-equipped
woodworking facility with a large table saw, jointer,
thickness planer, wood lathe, radial saws, polishing
lathe, veneer press and a large tool-room metal-
worki ng l athe; an equal l y compl ete and wel l -
equipped electronics-acoustics laboratory with dual
channel fast Fourier analyser, scopes, voltmeters,
microphones, oscillators, anechoic chamber, and
lastly, an office/design room set up with a big, Mac-
intosh desktop publishing system, a jillion megs of
application software, back-up hard drive, CD ROM,
scanner, printer, high-speed photocopier, Cerlox
binding equipment, etc.
I f the equipment I buy is not of extremely high
quality and reliability I soon nd myself in position
where I am constantly repai ri ng one thi ng or
another and chewing up a lot of time doing so.
The design of this amplier meant that I had to
AN 23.1.1 - Page 3 ! !
solve the reliability problem if I was ever going to be
able to tolerate owning it. I worked this problem
over in my mind for several years and intuitively it
always seemed to me that the very high tempera-
tures at which tubes typically operate had to be a
factor in the short working-life so often observed.
Through several bits of good fortune and a cou-
ple of co-incidences I found conformation that high
bulb temperature was denitely a problem and went
ahead to develop the line of tube coolers that PEARL
now markets. Working on the basis that whatever I
developed had to be readily and simply retro-ttable
to all kinds of existing equipment without resorting
to forced air cooling, a number of effective products
have been developed. See our Audio Note 1.3, Tube
Coolers and Equipment Reliability for the full scoop
on this.
I n any convective cooling system there is only so
much power available to move air. The effective-
ness of most such cooling or heating arrangements
can generally be increased by forcing air through
the system with a fan. I n a tube amp the trick is to
provide a balanced ow of air over all the tubes so
that they operate at very nearly the same tempera-
ture. This has been accomplished in the SC280 by
the use of a pressurized pan-type chassis. I t is basi-
cally a 25" wide x 14" front-to-back x 4" deep chassis
with the tubes and transformers mounted on the 25"
x 14" top surface. See Fig. 2 for a sketch of this. Low
noise fans are compliantly mounted on the bottom,
inside the chassis and quietly create a positive pres-
sure within the chassis. This pressure nds relief
through a number of clear acrylic plastic chimneys
seated in suitably sized holes punched in the top
deck of the chassis. With coolers tted, the tubes,
both front-end and output, reside within a directed
air ow guided by the chimneys. I n this manner,
extremel y reasonabl e bul b temperatures are
achieved and very long tube life can be rightly antic-
ipated. As an example of the effectiveness of this
arrangement is as follows:
I f an EL34 is simply plugged into a socket, to
which only lament power is supplied and
allowed to come up to temperature, the bulb
hot spot will run about 100C. I f the tube is
Air flow
PC board
C om pliant
m ounting
C om pliantly m oun-
ted sub-chassis
C lear acrylic
plastic chim ney
C hassis
50 75 100 125 150 175 200 225 250
122 167 212 257 302 347 392 437 482 527 F
275 C
1 2 3
Fig. 1.The effectiveness ofthe cooling system used in the SC 280
is show n above.
C urve 1 illustrates the bulb tem perature w ith only the lam ent run-
ning.The tube is dissipating 9 w atts of pow er into free air and is not
tted w ith a cooler nor placed in a forced-air chim ney.
C urve 3 show s the free-air bare-bulb tem perature and substantial gra-
dient w hen the tube is operated at its m axim um rated plate+screen dissipa-
tion of28 w atts.The total pow er radiated by the envelope is 37 w atts includ-
ing the lam ents contribution.Again,neither cooler nor chim ney is used.
C urve 2 show s the bulb tem perature and low gradient achieved
w ith the fan-cooler-chim ney arrangem ent used in the 280. O perating
conditions are the sam e as for curve 3;for clarity the cooler is not show n
in the illustration.The hot-spot tem perature is reduced by a w hopping
230 F.
AN 23.1.1 - Page 4 ! !
then triode connected and run at its maximum
rated screen-plus-plate dissipation of 28watts,
it will be radiating an additional 9 watts from
the lament, for a total of 37 watts of energy to
be sunk into the atmosphere. The bulb hot-
spot will then be about 250C. and a tempera-
ture gradient of 100C. will be present from the
hot spot to either end. I f the cooler-chimney-
fan arrangement is then tted, the bulb hot-
spot temperature will drop, in about 5 min-
utes, to around 130C. I n other words the tube
can be made to run nearly as cool under con-
ditions of atout-maximum dissipation as it
does under conventional operation with noth-
ing more than the lament lit! See Fig. 1
During the 50s and 60s the people at Mclntosh
produced thousands of ampliers that ran with very
low standing currents in the output tubes (essentially
Class B) and consequently, low bulb-temperatures.
These amps were famous for their incredibly long
and reliable tube performance. Other instances of
cooling vs. usable life exist by the hundred.
One situation that was instrumental in getting
me going in the cooler business has to do with some
tube-type acoustic analysis gear made by Messrs.
Bruel & Kjaer during the60s and 70s. I have a num-
ber of their level recorders that I bought for good
prices because they wouldnt run reliably. I t took me
several years of ddling to develop a series of mods
that would solve the problems in such parts as
capacitors and resistors. But it was not until I began
to pay serious attention the operating temperature of
the tubes that the problems really began to go away.
The gear was originally tted with bright J AN
tube shields (see Audio Note 1.3, Appendix 2: Figs. A
& B) and one day I decided that these had to go. I
removed not only the shields but the shiny-base
retaining-sockets and installed a powerful but some-
what noisy fan to force an air blast of about 1000
ft/min directly onto the hottest running tubes. Being
a maximum of 4" from the fan, these tubes run at a
far lower average temperature than was previously
the case. Whereas I was getting about 1 year of inter-
mittent use from a given level recorder before tube
failure, I have units in the eld that have been run-
ning at least 8 hrs a day, 5 days a week for over two
years without so much as a hiccup. Lowering bulb
temperature extends tube life!
I have talked with a few people who are under the
mistaken impression that the cathode coating in a
tube is somehow used up by the basic operation of
the tube. While the coating is chemically quite fragile
and subject poisoning by spurious gas within the tube,
damage due to ion bombardment, stripping due to
excessive demands for current or cold-cathode starts
with full B+applied; there is no fundamental reason
why a tube cannot be made to run for years on end.
This was the thinking of the people at Bell Labs
and Western Electric when they developed tubes of
such high reliability that they were installed and per-
manently sealed into trans-Atlantic telephone cables
as repeater ampliers. One cant be fooling around
replacing tubes in 2500# of water! Admittedly these
tubes were handbuilt laboratory specimens and sub-
jected to the most rigorous QC procedures possible.
The fact remains however, that with sufcient care
in construction and use, cathodes can be made to
last virtually indenitely.
During my literature search, I came across some
information regarding various life-extending mea-
sures taken by the folks who designed, built and ran
the 18,800 tube Electronic Numerical I ntegrator and
Computer.
1
Changes in operating procedures show
results quite quickly with a machine of such size and
complexity as a test bed.
Most people who have worked with tubes for any
length of time have wondered about the effect of
high-inrush-lament-current on tube life. The ENI AC
people wondered about this too and did a good deal
of work on the matter. Their conclusion was that a
cycle of daily turn-on; turn-off had no effect on the
frequency with which tubes failed. The following is a
quote from ref. 1:
A series of experiments was started, using 100
tube samples, to determine whether a graduated
application of voltage would materially reduce
the rate of failure when heaters were turned off
and on. Such experiments showed no difference
between the group of tubes to whi ch full
lament voltage was applied at once and those
receiving a gradual application of voltage.
A number of measures were implemented that
did show an increase in tube life:
the operation of laments rated for 6.3V at
6.0V. See Audio Note 1.3, Tube Coolers and
Equi pment Reli abi li ty, Fi g. 1 for further
details.
operation of plate circuitry so as to reduce
plate dissipation to
1
8 of its design centre rat-
ing. Such low-power operation lowers bulb
temperature dramatically
the heater-to-cathode potential was limited to

50V, with all heaters biased with respect to

the cathode voltage.
2
In the late 40s, the concept of bulb cooling hadnt
been discovered and wasnt used in the ENIAC project.
As far as possible, the foregoing measures have
AN 23.1.1 - Page 5 ! !
been applied to the present amplier although power
tubes cannot be run at signicantly lower lament
voltages without a noticeable reduction in current-
output capability.
I ts quite unusual to see the output tubes run
with DC lament supplies becauseI guessmost
people think of induced-hum as the problem and
think that an output stage, with its large signals, will
not be adversely effected.
The output-tube laments in this amplier run
on DC for a very good reason: a distinct buzzing/tin-
kling sound can sometimes be heard from power-
tubes when their lament are energized by AC yet
when DC is applied the tube goes silent. This occurs
because AC-current ow through the folded up
length of insulated lament wire within the cathode
initiates a motor action that causes the lament to
contract, accordion-style, with every voltage swing
away from zero, positive or negative. As the voltage
returns to zero the lament expands to its original
shape. Thereby, it acts as a little 120Hz. shaker
within the cathode and almost certainly stimulates
guitar string resonances within itself, cathode res-
onances and probably several grid resonances. This
shaker mechanism gets things going within the tube
to such an extent that plainly audible sounds are pro-
duced. With the internals of the tube set into reso-
nance and thereby into relative motion, adverse
effects upon any musical signal amplied by the tube
can easily be imagined. This mechanism is silenced
DC is used to energize the laments.
The problems caused by acoustically induced
microphonic output have been given a lot of thought
and measures have been taken to ensure that such
coloration is kept to a minimum. I so-sockets and
compliant pc-board mountings are used throughout
and the output tubes are mounted in groups of three
on separate, compliantly mounted sub-chassis.
All of the power supply circuitry is built on con-
ventional PC boards while the signal-path circuitry is
hard-wired using a specially developed ground-
plane, hybrid construction. Normal double-sided pc-
board material is etched on one side to carry the
lament currents and on the other to provide a
ground-plane grid in the usual manner. Provision is
made for solder-type turret terminals, to which all
signal path connections are made. The PC board acts
as nothing more than a gloried terminal board and
provides the convenience of PC board construction
while maintaining the sonic superiority of point-to-
point hard-wiring.
BASI C NOTES ON THE CI RCUI TRY
Referring to the block diagram, the amplier is
seen to be comprised of three sections of ampli-
cation. Three sorts of feedback are employed, local,
loop and active common-mode error. True absolute
phase swi tchi ng i s remotely selectable and the
amplier may be driven from single-ended or bal-
anced sources. The output stage uses amply AC
bypassed solid-state regulation of the standing cur-
rent in each individual output tube, ensuring that the
amplier will hold its DC balance regardless of the
match of the DC characteristics of the output tubes
used. The front end is completely self-balancing
both AC and DCand maintains its balance regard-
less of tube aging or match.
THE FRONT END
All balanced, push-pull, tube-type output stages
must be driven by a pair of antiphase signals. One of
these drives the push side while the other drives
the pull si de of the output stage. The phase
inverters task is to provide such a signal pair and to
be able to derive it from a single-ended source. I t
must do so in an accurately balanced fashion with
regard to both amplitude and phase.
Ampliers using only a single pair of tubes in the
output stage often require that these signals be
slightly unbalanced in amplitude while being exactly
180 antiphase. This is done to account for side-to-
side gain variations in the output tubes and is usually
necessary to achieve balanced currents in the half-
primaries of the output transformer. With a push-
pullparallel output-tube conguration, reasonably
careful matching will ensure that the average-tubes
acting in the push and pull halves exhibit very similar
transfer curves. Each half of the push-pull congura-
tion can then be driven by a signal that is an exact
mirror-image of the other. As long as Class A opera-
tion is maintained, local feedback in the output stage
in combination with the high rejection of power-sup-
ply-induced common mode signals exhibited by a
good output transformer, will effectively mop-up
most of the residual, side-to-side imbalances.
Literally scores of circuits have been developed
that are pretty much capable of mirroring a single-
ended signal into a pair of equal amplitude, phase
opposed signals. The great majority of these however,
suffer from one defect or another and I have spent
several years working on high-precision, phase-split-
ti ng techni ques. The result of thi s effort i s The
Actively Plate-Loaded, Self Balancing, Ultra-Linear,
Differential Cascode, Phase-I nverter/Driver Stage. This
circuit solves virtually all of the problems encoun-
tered and does so using only four dual-triodes, two J -
FETs, one positive and one negative power supply. I t:
is DC coupled from input to output.
is exceptionally quiet.
has very high common mode rejection from
both input and feedback sources.
AN 23.1.1 - Page 6 ! !
exhibits very wide bandwidth, low distortion
both harmonic and I M.
treats even-order distortion as a common-
mode signal and substantially reduces it.
has a low output impedance.
will easily swing 200V p-p.
can be driven from any single-ended source-
i mpedance up to 100K$ wi th vi rtually no
effect on output balance.
The application of feedback is always an issue
that must be carefully considered. I n this case, feed-
back applied to V1
b
and V2
b
has the interesting effect
of linearizing the operation of the tubes (V1
a
and V2
a
)
that are upstream from the point at which feed-
back is applied. The mechanism by which this hap-
pens will be explained in Audio Note 15.0, The Cas-
code for Audio Amplication; New Developments.
THE OUTPUT STAGE
For purposes adequate to this brief introduction,
the operation of the output stage is covered in Audio
Note 2.1 & 2.1.1, A Little I nput on Audio Output
Transformers & Update. The subject will be dealt
with in considerable depth in the following, upcom-
ing Audio Notes:
3 - Basic Transformer Action.
16 - The Transformer-Coupled Output-Stage: The
Basics of Push-Pull Operation.
17 - Cathode Coupling; Optimizing the Output-
Stage.
18 - Class Distinctions; Modes of Amplier Opera-
tion.
20 - Feedback From the Output-Stage; An Exami-
nation of the Options.
21 - Triode vs. Pentode; The Grass is Greener in the
Middle!
Experienced builders will notice the unusual con-
nection of the output-tube suppressor grids. I discov-
ered that by swinging the suppressor in phase with the
plate, but at a fraction of its peak voltage, a worthwhile
quantity of secondary emission could be eliminated. A
suppressor grid looses it effectiveness as the plate volt-
age swings towards the suppressors zero potential and
the grid must be swung negative to account for this.
Sonically the result is quite gratifying.
I t turns out that the grounded, center-tapped sec-
ondary provides an ideal source of dynamic bias for
the suppressors. To achieve proper phasing the grids
are tied to a point thatusing negative cathode-feed-
backis the cathode of the opposing tube of the
push-pull pair.
This cross coupling effects only tubes that have a
physical suppressor grid connected to pin 1 and does
not effect the operation of beam power tubes such as
KT66, 77, 88, 90, 6CA7, 6L6, 5881, 6550, 8417, etc. These
tubes suppress secondary emission by the creation of a
virtual cathode between the screen and plate and dont
rely on a physical grid; more on this later.
MAY 1991 ADDENDUM
There is some confusion regarding the 6CA7/EL34
as this tube is produced in both a two-grid, beam-
tetrode and a three-grid pentode version. Yet, no dis-
tinction is made between them and either device can
be found labeled with either number.
For those of us who concern ourselves with the
soni c characteri sti cs of tubes, thi s haphazard
labelling presents a problem because a pure pentode
and its beam-tetrode equivalent are not at all the
same beast from a sonic standpoint. The tetrode
exhibits signicant grid-current problems in the 10
to 0 control-grid-volt region, whereas the pentode
version does not and is therefore sonically superior
in capacitively coupled, xed-bias or self-biased out-
put stages. Look for more on this in Audio Note 23.2.
The pentode can be di sti ngui shed from i ts
beam-tetrode look-alike by counting the pairs of
grid-support side-rods protruding through the top
support mica and anking the cathode tubewithin
which resides the lament. With one pair of side
rods required per grid, its a simple matter to count
the rods and thereby derive the internal structure of
the tube.
Two pairs of side rods are seen in the beam-
tetrode, which I label a 6CA7. Three pairs are found
in the pentode which I call an EL34. I have created
this connotative nomenclature based on the follow-
ing thinking:
the pentodeEL34has not, to my knowl-
edge, ever been produced in North America so
I have chosen to use the non-North American
number to designate the non-North American
tube
the beam-tetrode6CA7is the version pro-
duced on this continent and others as well, so
I have given it the number used in the Ameri-
can system.
REFERENCES
1 Tube Failures in ENI AC, F.R.Micheal, Electronics, October
1947.
2 The Heater-Cathode Leakage Problem, M. Horowitz, Radio TV
News, publication date unknown, copies available from
PEARL.
3 Technical Reference Manual, Lamptronix Co. Ltd., Crystal
Lake, I ll, USA 60014.
4 The Application of Active Error-Feedback, J .R. Macdonald,
Proc. I .R.E., J uly 1955
AN 23.1.1 - Page 7 ! !
SC 280 Power Amplier
Fig. 2 - Mechanical Layout - Plan View
Fig. 2. The PEA R L SC 280 M O N O B LO C K is show n in plan view.
N ote the sym m etry of the layout. Fans w ithin the chassis force
cooling air up through all of the cooler-chim neys resulting in the
sort of tube cooling perform ance show n in Fig.1.A ll pow er-utility
AC is contained w ithin a sealed,bulkheaded com partm ent directly
below the high-isolation pow er transform er.O nly isolated, ltered
and initially recti ed AC is allow ed into the rest ofthe chassis.
The rem otely-controlled absolute-phase-sw itching circuitry is oper-
ated from an um bilical that runs from the listening position back to
either one of the am pli ers.The pair of am pli ers is connected by a
second um bilical, ensuring that the absolute phase of both am ps
sw itches sim ultaneously w ith rem ote sw itch actuation.
H igh Isolation Pow er Transform ers
T1 and T2
O utput
Transform er
AC Sw itching Fuses, etc. Inputs & O utputs
Rem ote A bsolute-Phase C ontrol
C onnections, B + & Screen Supply
Fuses
1/2 O utput Stage 1/2 O utput Stage Front End
14"
25"
M ain B + and Front End
Inductors
Superseded
A
N

2
3
.
1
.
1

-

P
a
g
e

8
!
!
S
C
2
8
0

P
o
w
e
r

A
m
p
l
i

e
r
F
i
g
.

3

-

B
l
o
c
k

D
i
a
g
r
a
m
XLR
W
i
d
e
-
B
a
n
d

O
u
t
p
u
t
-
T
r
a
n
s
f
o
r
m
e
r
Screen.
Regltr.
1
/
2

S
c
r
e
e
n
1
/
2

S
c
r
e
e
n
1
/
2

P
r
i
m
a
r
y
1
/
2

P
r
i
m
a
r
y
1
/
3

S
e
c
.
1
/
2

T
e
r
t
i
a
r
y
1
/
2

T
e
r
t
i
a
r
y
O
u
t
p
u
t

S
t
a
g
e

B
+
S
i
g
n
a
l

C
o
m
m
o
n
T
o

L
o
a
d
EL34
O utput Stage
V5 to V16
3dB Screen
Local N FB
3dB Screen
Local N FB
6dB C athode
Local N FB
6dB C athode
Local N FB
12dB Loop N FB
Front-End
V1, V2, V3
C om m on-M ode
Feedback
A ctive
C urrent
Source
Front-End
B + Regulator
RC A
Signal
C om m on
Rem ote A bsolute-
Phase C ontrol
U m bilical to the
O ther A m plifier
Input
Sw itching
(O ptional)
Rem ote
A bsolute-
Phase
Sw itching
12dB Loop N FB
V4a, b
Front-End
B - Regulator
C athode-C oupled, U ltra-
Linear O utput Stage
Self-B alancing, U ltra-Linear D ifferential
C ascode, Phase Inverter/D river Stage
1
/
6

S
e
c
.
1
/
3

S
e
c
.
1
/
6

S
e
c
.
8
$
8
$
4
$
4
$
Superseded
A
N

2
3
.
1
.
1

-

P
a
g
e

9
!
!
S
C

2
8
0
.
1

P
o
w
e
r

A
m
p
l
i

e
r

S
c
h
e
m
a
t
i
c
s
F
i
g
.

4

-

P
o
w
e
r

S
u
p
p
l
y

B
l
o
c
k

D
i
a
g
r
a
m
Primary Voltage-Tap
Selector Block
Dual-Choke Input
B Rectifier & Filter
Floating,
Raw DC
B+Supply
Floating,
Raw DC
B Supply
F
r
o
n
t
-
e
n
d

F
i
l
a
m
e
n
t

S
u
p
p
l
y
D
r
i
v
e
r

S
t
a
g
e

B

P
o
w
e
r

S
u
p
p
l
y
110VAC
Nominal
110VAC
Nominal
Power Transformer T1; Main B+, Screen Supply
& Output Stage Filament Supply
Manual Soft-Start
Switch
110240VAC
5060Hz.
Power Transformer T2
Front-End B Supply
Choke Input B+Rectifier
& 750V, 260 J oule, Filter Capacitor
400VDC Screen
Regulator & FIlter
C apacitor Input, O utput-
Stage Filam ent Rectifier
O utput-Tube Filam ent
Regulator, 1/Tube
Power Transformer T2;
Front End Filament Supply
V1, V2
Filament
Supply
V4
Filament
Supply
B
i
a
s
e
d

t
o

+
2
0
0
V
Floating
Output
Floating
Output
Floating
Output
Outptut
Stage B+
Signal
Common
Screen
B+
Front-End
B Regulator
Front-End
B+Regulator
B
i
a
s
e
d

t
o

+
7
5
V
B
i
a
s
e
d

t
o

2
0
0
V
V3
Filament
Supply
Superseded
AN 23.1.1 - Page 10 ! !
SC 280 Power Amplier
Fig. 5 - Front End - Component Values
V1, V2, V3 - 6C G 7, 6FQ 7
V4 - 5687, 7044, 7119
Q 1,2 - Toshiba 2SK 147 - B L
M atched pair; R 43, R 55 set
to yield 225V across V3a & V3b
Signal
G round
V4b
9
7
6
+ Input
- Input
Signal
G round
Front-End B -,
-240V N om inal
R 1
47K $
R 3
47K $
R 19
470$
470$
R 27
82.5K $
R 17
V1a V1b
V2b
33$
R 21
1
3
5
13
2F - 600V
Polyprop.
R 33
100$
2
3
1
2
3
V4a
3
1
2
1.0M $
R 31
C 1
V2a
Front-End B + ,
+ 465V N om ina
7
D S
G
D S
G
V3a
V3b
R 25
33$
100K $
R 23
158K $
R 15
330$
R 29
9
5
11
365K $
R 49
33$
R 35
33$
R 37
R 39
33$
R 41
33$
33$
R 57
10K $
R 45
R 53
10K $
100K $
R 47
100K $
R 51
33$
R 59
R 43
xx$
xx$
R 55
1
15
14
16
1/2
Tert.
1/2
Tert.
Q 1
Q 2
C 2
10f - 200V
Polyprop.
2
3
1
6
7
8
6
7
8
6
7
8
To the O ut-
put Stage
To the O ut-
put Stage
Superseded
AN 23.1.1 - Page 11 ! !
SC 280 Power Amplier
Fig. 6 - Front End - DC voltages
V1, V2, V3 - 6C G 7, 6FQ 7
V4 - 5687, 7044, 7119
Q 1,2 - Toshiba 2SK 147 - B L
Signal
G round
V4b
9
7
6
+ Input
- Input
Signal
G round
Front-End B -,
-240V N om inal
V1a V1b
V2b
1
3
5
13
2
3
1
2
3
V4a
3
1
2
V2a
Front-End B + ,
+ 465V N om inal
7
D S
G
D S
G
V3a
V3b
9
5
11
1
15
14
16
1/2
Tert.
1/2
Tert.
Q 1
Q 2
C 2
2
3
1
6
7
8
6
7
8
6
7
8
To the O ut-
put Stage
To the O ut-
put Stage
0
0
0
0
0
0
0
00
0
2.4
2.4
100 365
8.3
8.3
0
0
XX
XX
100
225
125
0
0
0 82
158
3.3
480
90
150
Superseded
AN 23.1.1 - Page 12 ! !
SC 280 Power Amplier
Fig. 7 - EL 34 Output Stage - Component Values
1
2
5
6
12
11
15
14
+650V from
Output Stage
B+Regulator
+400V from
Screen
Regulator
1
/
2

P
r
i
m
a
r
y
1
\
2

S
c
r
e
e
n
1
/
2

T
e
r
t
i
a
r
y
1
/
3

S
e
c
3
4
7
8
11
9
16
15
1
/
2

P
r
i
m
a
r
y
1
\
2

S
c
r
e
e
n
1
/
2

T
e
r
t
i
a
r
y
1
/
3

S
e
c
60V from
Bias Reg
From the
Driver
Stage
13
12
10
10
1
/
6

S
e
c
1
/
6

S
e
c
4$
Nom
8$
8$
V5
V6
V7
V8
V9
V10
V11
V12
V13
V14
V15
V16
11
11
5
9
R73 150K
P1
20K
R61
22k
C3
5F
100V
C15
10F
200V
R75 150K
P2
20K
R63
22k
C5
5F
100V
C17
10F
200V
R77 150K
P3
20K
R65
22k
C9
5F
100V
C19
10F
200V
R79 150K
P4
20K
R67
22k
C11
5F
100V
C23
10F
200V
R81 150K
R69
22k
C13
5F
100V
C25
10F
200V
R83 150K
R71
22k
C21
10F
200V
P5
20K
P6
20K
C7
5F
100V
R85 10R
R109 2k2
R155 22R
R133
10R
R87 10R
R111 2k2
R157 22R
R89 10R
R113 2k2
R159 22R
R91 10R
R115 2k2
R161 22R
R93 10R
R117 2k2
R163 22R
R95 10R
R119 2k2
R165 22R
R135
10R
R137
10R
R139
10R
R141
10R
R143
10R
R121 2k2
R97 10R
R145
10R
R167 22R
R123 2k2
R99 10R
R145
10R
R169 22R
R125 2k2
R101 10R
R147
10R
R171 22R
R127 2k2
R103 10R
R149
10R
R173 22R
R129 2k2
R105 10R
R151
10R
R175 22R
R131 2k2
R107 10R
R153
10R
R177 22R
CW
4$
Nom
Superseded
AN 23.1.1 - Page 13 ! !
1
2
5
6
12
11
15
14
+650V from
Output Stage
B+Regulator
+400V from
Screen
Regulator
1
/
2

P
r
i
m
a
r
y
1
\
2

S
c
r
e
e
n
1
/
2

T
e
r
t
i
a
r
y
1
/
3

S
e
c
3
4
7
8
11
9
16
15
1
/
2

P
r
i
m
a
r
y
1
\
2

S
c
r
e
e
n
1
/
2

T
e
r
t
i
a
r
y
1
/
3

S
e
c
60V from
Bias Reg
From the
Driver
Stage
13
12
10
10
1
/
6

S
e
c
1
/
6

S
e
c
4$
4$
8$
8$
V5
V6
V7
V8
V9
V10
V11
V12
V13
V14
V15
V16
11
11
5
9
R73
P1
R61
C3
C15
R75
P2
20K
R63
C5
C17
R77
P3
20K
R65
C9

C19
R79
P4
R67
C11
C23
R81
R69
C13
C25
R83
R71
C21
P5
P6
C7

R85
R109
R155
R
133
R87
R111
R157
R89
R113
R159
R91
R115
R161
R93
R117
R163
R95
R119
R165
R121
R97
R167
R123
R99
R169
R125
R101
R171
R127
R103
R173
R129
R105
R175
R131
R107
R177
28.6V
0.0V
28.6 to
60V
0.0V
.40V
0.0V
0.0V +30mV
+30mV
+30mV
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
.40V
.40V
.40V
.40V
.40V
28.6 to
60V
.40V
.40V
.40V
.40V
.40V
.40V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
0.0V
28.6V
28.6 to
60V
0.0V
28.6V
28.6 to
60V
0.0V
28.6 to
60V
28.6 to
60V
28.6V
28.6 to
60V
0.0V
28.6 to
60V
28.6V
28.6 to
60V
0.0V
28.6 to
60V
28.6V
28.6 to
60V
0.0V
28.6 to
60V
0.0V
50mV
R
135
R
137
R
139
R
141
R
143
R
143
50mV
50mV
50mV
50mV
50mV
50mV
50mV
R
143
50mV
R
145
50mV
R
147
50mV
R
149
50mV
R
151
Superseded
SC 280 Power Amplier
Fig. 8 - EL 34 Output Stage - DC Voltages
AN 23.1.1 - Page 14 ! !
Front-end C onnection B oard
Shielded Enclosure
K
1
A bsolute Phase Rem ote C ontrol U m bilical
H and-held Sw itch
RC A
Single-ended
Input
3 Pin XLR
B alanced
Input
A m plifier Inter-connecting U m bilical
5 Pin
XLR
4 Pin
XLR
4 Pin
XLR
Front-end
Input-tube
G rids
V1
V2
6VD C
Signal
G round
33$
33$
33$
33
$
33$
M ounted on
C hassis
5 Pin
XLR
4 Pin
XLR
4 Pin
XLR
SC 280.1 Power Amplier Schematics
Fig. 9 - I nput Switching & Remote-Absolute-Phase I nverting Control