Underground Infra

UNDERGROUND INFRASTRUCTURE OF URBAN AREAS
SELECTED AND EDITED PAPERS FROM THE INTERNATIONAL CONFERENCE ON UNDERGROUND INFRASTRUCTURE OF URBAN AREAS, WROClAW, POLAND, 22-24 OCTOBER 2008

Underground Infrastructure of Urban Areas

Editors

Cezary Madryas, Bogdan Przybyla & Arkadiusz Szot

Faculty of Civil Engineering, Wroclaw University of Technology, Wroclaw, Poland

o ~~~~F~~~~~"P

Boca Raton London New York Leiden

CRC Press is an imprint of the

Taylor &; Francis Group, an informa business

A BALKEMA BOOK

Taylor & Francis is an imprint of the Taylor & Francis Group, an informa business

© 2009 Taylor & Francis Group, London, UK

Typeset by Charon Tee Ltd (A Macmillan Company), Chennai, India

Printed and bound in Great Britain by Cromwell Press Ltd, Towbridge, Wiltshire

All rights reserved. No part of this publication or the information contained herein may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, by photocopying, recording or otherwise, without written prior permission from the publishers.

Although all care is taken to ensure integrity and the quality of this publication and the information herein, no responsibility is assumed by the publishers nor the author for any damage to the property or persons as a result of operation or use of this publication and/or the information contained herein.

Published by: CRC PresslBalkema

no. Box 447,2300 AK Leiden, The Netherlands e-mail: Pub.NL@taylorandfrancis.com

www.crcpress.com - www.taylorandfrancis.co.uk - www.balkema.nl

ISBN: 978-0-415-48638-5 (Hbk) ISBN: 978-0-203-88229-0 (eBook)

Underground Infrastructure of Urban Areas - Madryas, Przybyla & Szot (eds) @ 2009 Taylor & Francis Group, London, ISBN 978-0-415-48638-5

Table of Contents

Preface VII

Scientific CornmitteelReviewers IX

Sponsors XI

Problems of trenchless rehabilitation of pipelines situated under watercourses 1

T. Abel

Building on underground space awareness 9

JB.M. Admiraal

New challenges in urban tunnelling: The case of Bologna Metro Line 1 15

G. Astore, S. Eandi & P. Grasso

Numerical analysis of the effect of composite repair on composite pipe

structural integrity 27

A. Belzowski & P. Strozyk

Repair of RC oil contaminated elements in case of infrastructure 37

T.z. Blaszczynski

Modelling the behaviour of a micro-tunnelling machine due to steering corrections 45

W. Broere, J Dijkstra & G. Arends

Trenchless replacement of gas and potable water pipes with new PA 12 pipes

applying the pipe bursting method 55

R. Buessing, A. Dowe, C. Baron & M. Rameil

Experiences with Polyamide 12 gas pipes after 2 years in operation at 24 bar and

new possibilities for HDD 67

A. Dowe, C. Baron, W. Wessing, R. Buessing & M. Rameil

Simulation researches of pump-gravitational storage reservoir and its application

in sewage systems 75

J Dziopak & D. Slys

New developments in liner design due to ATV-M 127-2 and case studies 83

B. Falter

Concrete - durable composite in municipal engineering 97

Z. Giergiczny. T. Puzak & M. Sokolowski

Fly ash as a component of concrete containing slag cements 107

Z. Giergiczny & T. Puzak

Rehabilitation of road culverts on the equator. Implementation of innovative

open cut and jacking/relining trenchless solutions 115

JM. Joussin

Urban technical infrastructure and city management 129

W. Kaczkowski; K. Burska, H. Golawska & K. Kasprzak

V

Maintenance of drainage system infrastructure in Butare Town, Rwanda 141

A. Karangwa

Contact zone in micro tunneling pipelines 149

A. Kmita & R. Wr6blewski

Effect of variable environmental conditions on heavy metals leaching from concretes 155

A. Kr61

Design of the pipelines considering exploitative parameters 165

A. Kuliczkowski, E. Kuliczkowska & U. Kubicka

Management of sewer network rehabilitation using the mass service models 171

C. Madryas & B. Przybyla

Utilizing the Impact-Echo method for nondestructive diagnostics of atypically

located pipeline 183

C. Madryas, A. Moczko & L. WYsocki

Selected problems of designing and constructing underground garages in

intensively urbanised areas 193

H. Michalak

Material structure of municipal wastewater networks in Poland in the period of

2000 to 2005 203

K. Miszia-Kruk; M. Kwietniewski, A. Osiecka & 1 Parada

Two HDD crossings of the Harlem River in New York City lP. Mooney Jr: & lB. Stypulkowski

Preliminary design for road tunnels on Trans-European V c Corridor motorway, section Mostar North - South Border (Bosnia and Herzegovina)

l. Mustapic, D. Saric & M. Stankovic

Mapping the underworld to minimise street works C.D.F. Rogers

Assumptions for optimization model of sewage system cooperating with storage reservoirs

D. Sfys & 1 Dziopak

Curvature jacking of centrifugally cast GRP pipes U. Wallmann & D. Kosiorowski

213

225

237

249

257

Relining with large diameter GRP pipes U. Wallmann

269

Underground infrastructure of historical cities as exceptionally valuable cultural heritage

M. Wardas, M. Pawlikowski, E. Zaitz & M. Zaitz

275

Author Index

287

VI


Preface

In most cases, towns have grown up on the basis of industrial capital and relevant rules of industrialization. It has caused that such towns are malfunctioning, expensive, not ecological with all resulting consequences impeding everyday life of their inhabitants. Thus, public expectations are that portions of towns subjecting to modernization and also expansion of such towns would be progressed with utmost consideration for residential comfort by adapting newly grown town infrastructure to social, spiritual and cultural needs resulting from changed style of life and continuously changing scale of values. Creating urbanized space of such features is one among fundamental tasks that need to be undertaken to fulfil the expectations specified above. This task is also resulting from the necessity of unifying the towns and adapting them to standards becoming popular due to globalization process. New projects of modernization and expansion of towns having been now coming into being must be distinguished by better-than-before use of town space through stereoization of development, i.e. development of tower-block housing and underground structures. To meet this condition a higher level of integration of infrastructure systems among which the following equipment is distinguished:

• equipment related to communication services for the town;

• equipment connected with power management, water supply and sewage disposal system, waste disposal and management;

• communication and information related systems which, assuming the need of control, also in respect to the remaining infrastructure systems, create the basis of urban management system.

The most important is however to work out engineering solutions that would be the basis for creating integrated structures. The fundamental assumption for such studies must be creation of urbanized space enabling:

• to exchange energy between systems/equipment and to use town heat from some structures, such as for example communication tunnels, sewage system or power systems, etc.

• self-filling in of the water supply system (by waste water treatment),

• to rise the safety of town inhabitants both in respect of natural threats (flood, seismic and paraseismic quakes, etc.) and external risks (terrorist or war actions),

• to use the profits resulting from stereoization of town, i.e. temperature, humidity and acoustic conditions other than those existing over the ground,

• to release the ground space from some functions (first and foremost the communication related functions) which shall be mainly used for residential and recreational purposes,

• to renovate the historical, cultural and ecological environment of city centres.

Thus, researches, planners and investors must focus their attention on making better use of underground space as the potential to improve town communication, on expanding centre capacity by moving many commercial and service function underground, and also on modernization and integration of underground system to improve their functionalities and to create conditions for construction and operation of other underground structures.

Should the specified targets be reached, a package of administration regulations preferential for underground construction would be necessary, such which affects, but are not limited to the rules of crediting, subsidizing or commissioning the best solutions.

According to experience gained to date in developed countries we can state that the underground space would be, and often already is, used without any limitations to generally all purposes (except residential function, which in this way could get more space on the ground). However this would

VII

not mean that the topic has been exhausted and related problems resolve. Just the opposite. As it results from the materials included in this paper, this subject is still topical and many related issues need to be resolved. Hence, I hope that this paper would arise interest and inspiration for further examinations in persons engaged in widely understood shaping of underground infrastructure of urbanized areas.

At the end of this preface, I would like to express special acknowledgements to institutions and companies which logos and names are included in this book as their financial support was a decisive factor allowing its publication.

Main editor Cezary Madryas

VIII


Scientific Committee/Reviewers

HanADMlRAAL, President of Dutch Group ITA-AITES, The Netherlands GerardARENDS, Delft University of Technology, The Netherlands

Rolf BIELECKI, President EFUC, Germany

Bert BOSSELER, Wissenschaftlicher Leiter des IKT, Germany

Jozef DZIOPAK, Rzeszow University of Technology, Poland Bernhard FALTER, University of Applied Science-Munster, Germany Kazimierz FLAG A, Cracov University of Technology, Poland Piergiorgio GRASSO, Vice-President of ITA-AITES, Italy

Wojciech GRODECKI, President of Polish Group ITA-AITES, Poland Eivind GR~V, Vice-President of ITA-AITES, Norway

Alfred HAACK (D), STUVA, Germany

lens HOLTERHOFF, President ofGSTT, Germany

Jozef IASICZAK, Poznan University of Technology, Poland Martin KNIGHTS, President of ITA-AITES, UK

Andrzej KULICZKOWSKI, Kielce University of Technology, Poland Darius: lXDZBA, Wroclaw University of Technology, Poland Cezary MADRYAS, President of PSTB, Poland

HerbertA. MANG, Technische Universitiit Wien, Austria

Dietmar MOLLER, Universitdt Hamburg, Germany

Harvey PARKER, Past-President of ITA-AITES, USA

Anna POLAK, University of Waterloo, Canada

Chris ROGERS, University of Birmingham, UK

Anna SIEMTNSKA - LEWANDOWSKA, Warsaw University of Technology, Poland Ray STERLING, Louisiana Tech. University, USA

Markus THEWES, RUB - Bochum, Germany

Roland W. WANIEK, President of IKT, Germany

Andrzej WICHUR, University of Science and Technology Krakow, Poland

IX


Sponsors

PLATINUM SPONSORS

HERRENKNECHT AG

HOBAS System Polska Sp. Z 0.0

HOBAS@;

GOLD SPONSORS

INFRAS.A

GRUPA.PBG

REHAU Sp. Z 0.0.

(JREHRU

Unlimited Polymer &Jlutions

SILVER SPONSORS

Amitech Poland Sp. Z 0.0.

ITECH

Gorazdze Cement S.A.

ERGCEMENTGroup

KWH Pipe (Poland) Sp. Z 0.0.

A

KWH

PIP.:::

XI

OTHER SPONSORS

BEWA - Systemy Oczyszczania Sciekow

Dolnoslaska Okregowa Izba Inzynierow Budownictwa

SIKA Poland Sp. z 0.0.

Systemy Oczyszczania SCiekow

Book supported by

Foundation for Polish Science

XII


Problems of trenchless rehabilitation of pipelines situated under watercourses

T.Abel

Wroclaw University of Technology, Poland, KAN-REM Sp. z 0.0. Wroclaw, Poland

ABSTRACT: All types of collisions of pipelines with water race are frequently solved by elaborating sewer trap constructions. Such constructions are often encountered in sewerage systems when it is necessary to overcome an obstacle and for this reason by main collectors of large dimensions are used. Channels located in close neighborhood to surface water should be constantly monitored, since every damage or failure of the under-river pipeline or channel leading surface water may cause very serious consequences. Municipal and industrial wastes, when in contact with surface water may quickly result in contamination and ecological catastrophe. Surface water disturb water and sewerage balance and in extreme cases, with high pressures applied to the sewer trap construction, they may produce a very quick propagation of the damage and finally, construction disaster. Due to a very specific construction of a sewage trap (most frequently located under watercourses of Rother fix hindrances) it is impossible to repair or replace a pipeline network that is created by sewage trap in a traditional, dig technology. Owing to the development of civil engineering and the use of trenchless technologies, restoring the original condition of pipelines and assuring their safe exploitation are highly feasible. In the paper, examples of sewer traps will be provided and finished projects of sewerage systems rehabilitation, shown. The first structure to comment on will be a waste trap on the main drain of DN1200 located in the region of Pulawy (southeast of Poland) under Kurowka River. The sewer trap is composed of three steel pipeline networks of DN600. Rehabilitation works consisted in making short relining with PEHD modules. The second example will be a steel sewer trap of DN2200 located under discharge water of Thermal-electric power station in Konin (central Poland). Regeneration was made in Maxi- Trolining technology. The aim of this paper is to present trenchless methods of sewer trap structures rehabilitation and explain in detail all technological process as well as by-pass methods and materials solutions applied in pipelines and sewer traps entrance chambers.

1 INTRODUCTION

As a result of intensive development, routes of sewage systems forming underground part of urban infrastructure are, in many places, situated in direct vicinity of other objects, particularly in urban areas.

Engineering objects with which collision of sewage systems may occur are, first of all, road and rail tunnels, communication arteries and railway lines laid in towns in trenches, navigable waterways, water and heat supply mainlines and even buildings. The second group of obstacles that can be encountered while laying sewage systems are natural obstacles such as water courses and ravines.

Depending on the depth of laying sewers as well as differences in grade lines, the ducting may pass above routes of engineering objects and other terrain obstacles or below them.

1

Laying a sewer above an obstacle can be performed on an aqueduct or in a vault passing over the engineering structure. When laying the sewer under an engineering structure, the following three cases can occur:

• the sewer can run under the obstacle without change in shape or dimensions of the section. The bottom of the engineering structure intersects the ducting vault.

• the section of the sewer enters the object structure only in its top part. In this case, it is necessary to change section of the sewer to a lowered one while maintaining drop of the channel bottom and speed in the channel possibly unchanged.

• if the sewer enters into the structure of the colliding object with its whole section, a sewer trap should be planned under the obstacle.

2 ASSESSING THE TECHNICAL STATE OF DAMAGED SEWER TRAP PIPE

The starting point of technology selection for rehabilitation of trap pipes, as structural element transferring definite load, is precise assessment of their technical state. In accordance with ATVDVMK M 127, three different technical states of damaged pipe can be distinguished. Depending on the kind of technical state, various loads act on the executed shell (grouting). Assessment of the technical state should be performed on the basis of TV camera inspection results. It should be realized that such mode of examining the structural state of sewer is not always sufficient. Especially in the case of trap construction of concrete and reinforced concrete pipes, very dangerous vitriol corrosion of concrete often occurs as a result of which it transforms into gypsum causing lowering in load capacity of the whole structure, even leading to its loss and occurrence of state of emergency. Level of corrosion cannot be assessed by optical examination but only by testing of samples taken. If justified, before rehabilitation, point repairs should be undertaken using robots. In such case, the recommended solution is to increase the strength of rehabilitating shell over the whole length.

3 TRAP CONSTRUCTIONS AS SPECIAL CASES DURING REHABILITATION OF SYSTEM

Traps are constructions consisting of one or more pipes whose operation occurs under pressure. Trap constructions occur most frequently as objects made from pipes of cast iron, steel or steel in concrete or reinforced concrete casing.

In case of repairing sewage systems, engineering objects as are trap constructions generate complications for application of trenchless technologies.

Performing repairs of sewer traps using trenchless technologies requires conducting individual analysis of the case and drawing up a project, and particularly planning the technology for carrying out the works.

Part of the trenchless methods of renovating the system does not find application in sewer traps. Characteristic features for structures surmounting terrain obstacles, causing narrowing of capabilities for application of solutions, are first of all:

• very frequent changes in direction of laying trap lines in profile (figure 1).

Location of trap lines at large angle disables application of most close-fitting technologies due to their limitations regarding susceptibility at arcs and occurrences of deformations that can be avoided. When using unconstrained linings in case of occurrence of even minimum change in direction of sewer route due to introduction into the pipeline of rigid pipe modules of lengths from 0.6 to 6.0 m, the trap arrangement completely disables execution of repairs in the abovementioned technologies. Technologies of unconstrained linings constitute a very good solution in case of occurrence of trap line arrangement as straight lengths (figure 2).

• diameter of trap pipes. In case of passage pipelines, i.e. of diameters larger than 1000 mm, it is possible to utilize close-fitting technologies enabling formation of the insert directly in the

2

Figure 1. Changes in direction of laying trap lines in profile.

sewer by the fitters thanks to which their precise execution is ensured. For non-passage pipes, due to absence of the possibility of direct intervention inside the sewer, the range of applications of close-fitting technologies narrows considerably. Lack of possibility of controlling execution of lining in every sensitive point (as in the case of passage pipes) forms a serious obstacle and aspect against the use of close-fitting linings.

• trap construction in sewage system is most frequently applied in case of surmounting an obstacle in the form of water courses. Such a situation very often disables execution of by-pass type system enabling working on a section cut-off from utilization. Most trenchless techniques of repair are applicable only on inactive systems. Lack of facility for pumping over medium conducted through the pipe meant for rehabilitation defines to a certain extent the group of technologies that are possible for application.

All the above-mentioned conditions constitute the group of factors which, occurring simultaneously, limit to a very small group of trenchless technologies that are possible for application.

4 REVIEW OF TRENCHLESS TECHNOLOGIES FINDING APPLICATION IN REHABILITATION OF SEWER TRAPS

Linings used in trenchless methods of renovating sewage system trap constructions can be divided into close-fitting and unconstrained. The former (viz. in situ form) are methods consisting in making linings inside the existing pipe, whereas the latter consists in installation inside the section under repair, of pipes or modules of smaller dimensions than its inside diameter allows.

Close-fitting methods are sleeves of technical fabric saturated with resins as well as polyethylene sleeves. An example of unconstrained linings, in case of sewer traps, is method of relining with short modules.

3

, '

'., ........

< .:

. ...

.. -'~. I~' '... • ,....; ....

~:~._._: ~_ ~~~.~~:.J_~~.;~ .:_'~ ~.~~._._~~. ::~; ~: ~-_~ .: ._: _ w. .. ss. :._~. _ ... : ~.~:. .. \.. _:: _~.,.~~ __ ~_~. ~ _ :._:~ _:::_ .. s- ~:. _.!~ ... ,. ~~_ .~~ __ ~~_,~~ _ .:», .: .c.:

; ~ I"

t-!

1

I

Figure 2. Trap line arrangement as straight lengths.

4.1 Sleeves of technical fabrics

Technologies from group of close-fitting technical fabrics consist in inserting into the sewer a resinous shell - sleeve of technical fabric saturated with resin which, after filling with e.g. hot water or hot air, gets hardened and adheres closely to the old sewer structure.

4

At present, there are several variations of technical-fabric sleeve technologies, differing from the mode of introducing the shell, kind of medium used causing pressure in the shell, kind of agent hardening the shell.

In certain situations, technical conditions make the CIPP sleeve to become the only rational solution. They could include, for example, deformation of the section. Considered here are such deformations as, for example, imperfections of the cross section. In other situations, application of CIPP sleeve is not possible, for example, in the case of loss of load capacity and break-down of structure of the trap construction utilized. At present, due to decrease in utilization of water and consequently in reduction in quantity of sewage, in many cases, reduction in cross section is a positive operation since, thanks to this, flow speed increases resulting in improved self-cleaning of the whole system and reduction in maintenance costs.

4.2 Polyethylene sleeves

The basic example of this type of technology is Trolining system. It is a trenchless system for reconstructing combined sewage system and also sanitary, rainwater and industrial drainage systems as well as other pipelines both gravitational and also pressure, fabricated from various materials. This system belongs to the group of GIPP (grouting in place pipe) methods. Technologies of closefitting linings are characterized by eliminating to a minimum the need for carrying out earth works as are indispensable in case of classical technologies of underground infrastructure repairs; this facilitates reduction in communication difficulties. An important matter is also limitation of noise, dust and other nuisances. Deserving special attention however is the radical shortening of the period of work duration.

In contrast with linings of technical fabrics saturated with resins (differing in fabric material and kind of resin and particularly the mode of hardening), TROLINING consists of polyethylene insert made from special kind of sheet/foil and filling layer.

Large-size pipes are renovated using segments of PEHD panels stiffened with solid formwork, instead of "sleeves". In case of infiltration of groundwater, panels are also used on the outside, adhering directly onto the renovated construction, protecting the new structure against external influences. The free space between the insert and the pipe walls is filled with concrete. Reinforcement can first be installed in it. The quantity of reinforcement, wall thickness and class of concrete are the deciding factors regarding rigidity of the repaired construction. These quantities are defined through precision calculations. The repair system of large-size pipes is also perfectly suitable for rehabilitation of inspection chambers.

4.3 Example of execution

Discharge water trap in Konin Thermal-Electric Power Station. The object is located in Konin at Rybacka Street in northern part of the town. The trap construction has the task of carrying water from the discharge duct of Patnow Thermal-Electric Power Station under the duct leading water to Konin Power Station.This object belongs to the system of ducts and traps whose task is to carry cooling water of Patnow - Konin Power Plant Complex.

The ducting is made from smooth St3S steel pipes with longitudinal seam and wall thickness of 20 mm together with external and internal anticorrosion protection. Length of each trap line is 32.5 m (figure 4).

Assessment of technical state of the pipeline to be repaired showed very large corrosion cavities in places in pipe walls. Measurements taken with thickness gauge showed that wall thickness in certain places had reduced from 20 mm to 7.1 mm.

From structural analysis conducted, it was concluded that in case of reduction of wall thickness to 5 mm, the trap construction would lose its strength.

Technical state determined of the pipeline indicated the necessity for immediate reinforcement of the middle line of the trap. Deterioration of the technical state and the resulting loss of load capacity could threaten a construction catastrophe.

5

PE-HO Unlit

~'s!ing pipe

InJectot

Figure 3. GIPP system, TROLINING system.

Figure 4. Water trap in Konin.

Internal forces were calculated for two load situations:

• case I - all loads occur,

• case II - no water load in duct (pipeline is emptied of water).

After dimensioning the construction, results were obtained ensuring adequate technical parameters for reinforced concrete layer of 150mm doubly reinforced with 11114 rods in spacing of 100mm.

Structural and strength analysis for case II (trap construction emptied of water) indicated that for a pipe of such large loss in wall thickness (even up to 7.1 mm), its emptying of water is inadmissible since it could lead to breakdown.

In view of the above, it was planned to provide a preliminary reinforcement by installing rings made of rolled 160 mm channel sections. The rings were installed by divers before pumping out water from the ducting (figure 5).

6

existing pipe

PE-HD liner

structural concrel.e

reinforcement strengthening rings IPE 160

Figure 5. Preliminary reinforcement.

4.4 Relining with short modules

For rehabilitation of gravitational pipelines by means of short pipe modules, the modules utilized are of slightly less outside diameter than inside diameter of the renovated pipe, e.g. for renovation of DN 300 sewer, pipe modules of outside diameter 280 mm or 250 mm can be used. Rehabilitation consists in successive joining of consecutive pipe modules and simultaneous sliding the lining so assembled into the interior of the old pipeline. The modules have total length from about 0.6 m. This enables carrying out work inside the reinforced concrete chamber/pit and hence it is possible to rehabilitate the full length of trap construction without performing any earth work whatsoever. Available in the market are different methods of inserting renovation modules into the pipeline to be rehabilitated. Some firms propose pushing/jacking in the modules by means of hydraulic actuators/jacks; other propose pulling them in by means of winches.

4.5 Example of execution

Sewer trap carrying communal sewage from the town of Pulawa to the wastewater treatment plant. The object is located on the route of the main drain of diameter 1200 mm.

The trap consists of 3 lines of diameters 600 mm each, connected to the main drain through inlet and outlet chambers/pits. The trap construction lengths carry the sewage under the Kurowka River which constitutes a small tributary of the Vistula River. It was planned to perform rehabilitation using short relining technology. PEHD modules oflength 1 m were used. The installation technology consisted in pulling in the modules by means of a hydraulic winch. The works were carried out under difficult winter conditions with sewer in operation. Sewage was transferred by means of a log stop provided through active part of the trap construction. This was possible thanks to division of the pit into cells. Before proceeding to execute the short relining, all the lines of the trap were subjected to hydrodynamic cleaning. After rehabilitation, the new pipeline obtained was of slightly less diameter, the modules installed being of outside diameter 580 mm.

5 SUMMINGUP

Due to the specifics of trap system operation which very often operates at 100% capacity, any method of rehabilitation used must not lead to deterioration of the hydraulic conditions. The repair must ensure improvement of hydraulic conditions, increase in load capacity and prolongation of

7

pipeline durability. Rehabilitation of the said objects must be carried out with maximum accuracy and precision due to inability of monitoring the lengths of trap construction at operational stage. Breakdown of sewer trap, in contrast with the whole sewage system, may carry more serious and dangerous consequences with it. As a result of sewer trap breakdown, damage can be caused to structures in its close proximity which very often include the objects for which provision of the sewer trap was necessary. Damage to a communication artery or water course may create a direct hazard to people.

As can be seen, the consequences of no repairs or repairs carried out with errors may lead to several failures of structures. Sewer traps constitute a top-class challenge for any contractor undertaking their rehabilitation. Thanks to the possibility of applying the above-mentioned technologies for repairing such constructions, every designer has the capability of selecting the appropriate technology and planning out the rehabilitation procedure so as to acquire long-term effect and protection of the construction against breakdown.

REFERENCES

Sewage system. T. 1, Sewage system and pumping station. (In Polish). Kanalizacja. T. 1, Sieci i pompownie.

Blaszczyk, Waclaw, 1983.

ATV-DVWK-A127P Instructions - Structural analysis of sewers and sewage system pipes. (In Polish).

Wytyczne ATV-DVWK-A127P - Obliczenia statyczno-wytrzymalosciowe kanal:6w i przewod6w kanalizacyjnych.

ATV-DVWK-M 127P Auxiliary Materials "Structural analysis for technical rehabilitation of sewage system pipes by introducing liners or by the assembly method". (In Polish). Material:y Pomocnicze ATV- DVWK - M 127P "Obliczenia statyczno-wytrzymalosciowe dla rehabilitacji technicznej przewod6w kanalizacyjnych przez wprowadzenie liner6w lub metoda montazowa",

PN-841B-03264 "Concrete and reinforced concrete structures. Basic principles of designing". (In Polish).

PN-841B-03264 "Konstrukcje betonowe i zelbetowe. Podstawowe zasady projektowania".

Technical materials of MIs KAN-REM Sp. z 0.0. (In Polish). Material:y techniczne firrny KAN-REM Sp. z 0.0.

Technical materials of MIs TROLINING GmbH. (In Polish). Material:y techniczne firrny TROLINING GmbH. Problems of trenchless rehabilitation of sewage system pipes. (In Polish). Problemy bezodkrywkowej odnowy przewod6w kanalizacyjnych - Prof. Andrzej Kuliczkowski, 2004.

Rehabilitation of Konin trap construction - Execution project. (In Polish). Naprawa syfonu koniriskiego - Projekt Wykonawczy.

Testing and acceptance of CIPP technical sleeves and their durability - Andrzej Kolonko, Trenchless Technology 3/2007. (In Polish). Badania i odbiory techniczne rekawow CIPP a ich trwalosc - Andrzej Kolonko, Inzynieria Bezwykopowa 3/2007.

8


Building on underground space awareness

IB.M. Admiraal

Centre Applied Research Underground Space - CARUS, Gouda, The Netherlands

ABSTRACT: Worldwide the demand on available space in urban areas is growing. The awareness that the use of underground space can offer a solution is often lacking. This paper will discuss the latest developments in the field of underground space planning. It will highlight why planning should be considered in order to avoid spatial conflicts which will be detrimental to the use of underground space. The paper will also discuss the sustainable use of the underground. Given the many benefits the use of the underground offers to life on the surface, it is often deemed to be sustainable. The author will argue that this is not necessarily always the case. A balanced decision is needed when considering the use of underground space in which the underground as a living organism should be considered.

I INTRODUCTION

With the worldwide search for more space in urban areas, the use of underground space is seen as a valid solution. The awareness that this is the case is in practice not widespread. In many cases the development of underground space is autonomous with far reaching effects. One of these being that further development is severely hindered and can only take place at great depth. Tunnelling is seen as one of the methods which can be applied to use the underground space.

To prevent the autonomous development becoming common practice, a vision on the use of underground space needs to be developed at a local level. This in turn can facilitate planning the use of underground space, which should avoid conflicts between resource demands and also lead to multi-functional use of underground space.

In many cases the underground space is a living organism. This will require balanced decisions on the use of underground space. Only in this way can the use of underground space be deemed to be part of sustainable development.

2 UNDERGROUND SPACE AWARENESS

2.1 The worldwide quest for urban space

As the world population keeps growing, mega-cities are growing bigger and bigger. It is however not only the growth of the world population that leads to the development of mega-cities. The UNHABITAT programme has stated that as of mid-2007 more than 50% of the world population lives in cities. This means that the population shift from rural areas to urban areas is also contributing to this growth. One of the most common aspects of mega-cities is the struggle for space to accommodate all functions required to maintain liveability but also mobility.

Cities can not survive without infrastructure. Infrastructure to allow its population to move, but also the infrastructure to provide the city with its power and water. Utilities also require space to be accommodated. With the adverse affects of climate change, sewer systems need to cope with

9

ever increasing amounts of rainwater. All this needs to be taken into account. The worldwide quest for urban space requires radical new insights into land use. Multiple land use is often seen as a new way to cope with the ever increasing demand. The use of underground space must be seen as a valid option within this context.

2.2 Sustainable development and climate change

In The Netherlands a urgency agenda was published in 2007 by leading scientists and research programmes, calling on the nation to take to heart sustainable development and to climate proof the country. The so-called 'Urgenda' provides an action plan for the coming 40 years. The basis assumption is that The Netherlands will need to change more rapidly in the coming 50 years than in the past 500 years in order to cope with all the challenges the country is faced with.

These challenges are both in the social cultural arena as in the civil engineering field. One of the statements in the Urgenda, is that within 15 years, intensive use of underground space will be common in The Netherlands.

For the authors of the Urgenda it is a given fact that underground space use will playa vital role in the sustainable development of the country and its climate proofing, i.e. ensuring that the adverse effects of climate change are mitigated.

2.3 Underground space development

Given, as show above, the role underground space use can play within the context of the worldwide quest for more urban space, there is a paradox which needs to be addressed. This paradox being that on the one hand many countries and cities already practice an intensive use of underground space whereas on the other hand there seems to be a worldwide ignorance to the fact that underground space can playa vital role in alleviating the spatial shortages at surface level.

This paradox is one of the reasons for setting up the ITA Committee on Underground Space - ITACUS by the International Tunnelling and Underground Space Association, ITA-AITES. Worldwide many Western cities are finding the need to go deeper and deeper into the underground as the top layers are already congested with various functions. Users of transport systems need to be transported to great depth as the top layers are used for utility systems. This fact arises from the autonomous development of underground space without any form of coordination and no vision by city authorities on multiple use of underground space.

The author has often publicly stated that the goal to achieve intensive underground space use in The Netherlands within the next 15 years is unreachable if the current practice of uncoordinated use of underground space remains. The autonomous development will lead to a chaos in the underground which will make future development impossible.

Another problem which arises from the lack of awareness, is the simple fact that development of for example infrastructure will take place without even considering the possibilities of underground space use. This can lead to using contemporary methods which often give a suboptimal results as reported by the author (Admiraal, 2004).

Awareness of the possibilities which underground space use has to offer is therefore needed on a large scale. Not only to ensure that this use is considered right from the start of development of cities, but also to ensure that once development takes place, it is done in a coordinated way. The role of ITACUS will be, to provide a platform for a worldwide dialogue on the use of underground space. A dialogue which will consider the use of underground space within the context of societal needs, environmental concerns, sustainable development and the climate challenge.

2.4 Underground space use

There seems to be confusion in practice on what the use of underground space entails. Often tunnelling is seen to be the sole use of underground space. This is true in so far that tunnelling is a method which allows for various functional uses of underground space. Transport Use and

10

Production Use of the underground call for tunnels to make this possible (Admiraal, 2006). There are however many other uses which all compete for space in the underground. This fact in itself requires a balanced decision on how to develop underground space (Parriaux, Blunier, Maire & Tacher, 2008). A further complexity is added when we require this development to be sustainable as will be discussed later in this paper.

As Parriaux and others point out, underground space can be modelled as consisting of four different resources: space, water, geo-material and geo-energy. All these resources can be used, but these uses can conflict with various results. These results can vary from pollution of drinking water to transportation projects not being carried out. There is need to consider the use of underground space in its entirety and not limit it to tunnelling.

3 PLANNING BASED ON VISION

3.1 Action without vision

A Japanese proverb states that: 'Vision without action is a daydream, action without vision a nightmare'. As stated above, with rapid autonomous development of underground space, a nightmare situation can arise as city planners discover the chaos which exists underground. In The Netherlands and in China the need for creating a vision on the use of underground space as the basis for a planned development is understood and put in practice. Although the scale is still limited, interesting results can be reported. The city of Zwolle in The Netherlands was the first to develop a vision on the use of the underground.

One of the interesting results of this vision was the identification of polluted groundwater under a new urban development area. This has lead to the idea to combine the application of heat-cold storage with the cleaning-up of contaminated ground water over a period of 10 years. In this way the environment is served in two ways: the ground water is decontaminated and the carbon footprint for the development area is reduced as no gas or electricity is required to heat the houses in winter or cool them during summer.

In the city of Shanghai in China, a pilot project is being carried out whereby for new developments of the city, underground space must be included in the planning of the development. It is evident that this situation will lead to a coordinated development in which an optimal use of underground space is ensured. China is a prime example of a country where the use of underground space is more and more seen as part of urban development. Vision and planned action is paramount for a controlled development of underground space.

3.2 Conflicts between resources

The main problem which can arise from not planning the use of underground space can be best explained by two examples. As natural energy resources are seen to be limited, the search for alternatives is also a worldwide event. One of the most promising in this area is the application of geo-thermal energy systems. The application of these systems does however require vertical pipes to be inserted into underground space, often hundreds of meters deep.

The use of these systems is very popular and a rapid autonomous deployment of these systems is observed both in Germany and The Netherlands. The downside to this is that these systems can become a serious obstacle for future development of underground space in the horizontal plane. It is not unthinkable that future alignments of underground mass rapid transport systems is severely hindered by the presence of these systems.

Moreover, as observed by Parriaux, Tacher & Joliquin (2004), when a decision needs to be made for an underground mass rapid transport system or a geothermal application, the first is mostly chosen. In any case, no consideration is given to the possibility to combine these functions. Although not deemed to be feasible from an engineering perspective on the moment, there is no reason why in future this should not be the case. The point being made here is that without planning these situations can not be identified and therefore innovation as mentioned is not stimulated.

11

A second example stems from the conflict which arises from seeing underground space as a unlimited reservoir of space versus underground space as a nature reserve which needs to be preserved at all cost. This conflict arises from the fact that in more Deltaic regions, given the specific character of the subsoil, it is deemed to be the supporter of life on the surface, given the many natural processes and systems that exist below the surface. In one case this conflict has lead to a tunnel project being scrapped in The Netherlands for fear of changing the character of the Naarder Lake, a lake deemed to be part of an area of outstanding natural beauty.

That the fear in itself is not unfounded can be demonstrated with the adverse effects of water inflow in tunnels as experienced in Norway (Grev, 2008). Although the situation in Norway and The Netherlands are not comparable, the lack of understanding that the conflict between these two approaches exists, lies at the basis of the decision taken not to carry out the project.

A balanced-decision making framework for the use of underground space can avoid these conflicts. Parriaux and others, are working on the development of such frameworks. The implementation of these frameworks can however only take place when there is a general awareness of the virtues of the use of underground space, this awareness is translated into a vision which in turn makes planning possible.

3.3 Planning methods in practice

Various methods are in use regarding the planning of underground space. As reported by the author (Admiraal, 2006) research in The Netherlands has come up with practical methods which identify areas which are most likely to prove worthwhile for underground space development. These areas are identified by taking various aspects into account. Other methods try to approach the problem from a theoretical side and use a systems approach on which decisions can be based (Parriaux et al, 2008). The most common development is to focus on areas for development rather than on individual cases.

The so-called area development approach often incorporates a dialogue based model in which all interested stakeholders are involved. Planners then use the outcome of this dialogue to develop a stakeholder based vision on which the planning can be based. In this approach the interests of all parties are taken into account and the conflicts as observed above can be avoided as all interests are weighed-up.

An interesting tool for analysis is the so-called 'layered approach' to area development. In this approach an area is deemed to consist of three layers: habitation, networks and the underground. The method has been developed by the Ministry of Housing, Spatial Planning and the Environment in The Netherlands. The fact that the underground is recognised as an entity to be considered in future planning issues is a positive development.

The city of Arnhem in The Netherlands is actively using this approach for the planning of developments with positive results in terms of underground space use. One of the results being the combination of functions underground. A prime example is the development of an underground car park in combination with an utility tunnel which incorporates an underground waste refuge collection system. In other cities these developments would eventually have taken place on an autonomous scale. In this example the combination of these functions through planning based on vision makes it an excelling best practice example.

4 THE UNDERGROUND AS LIVING ORGANISM

The use of underground space takes place in different regions of the world with varying geological conditions. Mega-cities are most commonly found within 50 km of the sea. This makes that a lot of mega-cities and therefore underground space development takes place in Deltaic regions. These Deltaic regions have in common that the underground consists of soft soils being old river deposits. The regions also have in common that they are often very fertile, providing land for crop growth.

12

In Deltaic regions, the underground can be modelled as being the supporter for life on the surface as mentioned before. The time scale at which processes take place in the underground varies dramatically from life on the surface. Compare to appreciate this the time it took to produce coal, gas and oil as natural resources with the lifespan of an average building. The argument being made is that we often are not aware of the effects of manmade interventions in the subsoil. A prime example being the contamination of drinking water aquifers through storage of waste in the subsoil. The Netherlands is still facing a massive clean-up operation to this effect.

The use of underground space as part of sustainable development must consider the above when decisions are taken. It clearly illustrates that autonomous development of the underground not only can lead to resource conflicts as mentioned in the case of transport versus geothermal energy. It can also lead to developments which may prove to be non-sustainable.

In general it is felt by the author that resource conflicts can be avoided and a sustainable development of underground space can be achieved through planning. In the case of the city of Zwolle, the combination of a geothermal application with decontamination of groundwater, clearly shows what can be achieved through vision development.

5 CONCLUDING REMARKS

The use of underground space is seen to be vital in a world where more than half of the population now lives in urban areas. Further concentration in mega-cities is a trend which can not be stopped. The use of underground space can contribute to sustainable development, maintaining liveability and preparing the world for the impact of climate change. Creating awareness on the use of underground space in this respect is vital. Furthering the development of visions on urban underground space use and rational use of the underground space through planning techniques is essential.

The future of underground space use is furthermore governed by the ability to combine functions, e.g. combination of transport functions with water management. This requires a dialogue within city authorities across policy boundaries. The results can however be very positive as is demonstrated by the Stormwater Management and Road Tunnel project which is now operational in Kuala Lumpur, Malaysia.

The ability to plan the use of underground space in combination with a multi-functional use will determine the future of underground space use as a valuable contributor to the worldwide quest for more urban space and a sustainable development of mega-cities. It will also show that society can really not afford not to use underground space.

REFERENCES

Admiraal, IB.M. 2004. Developing a knowledge infrastructure for underground space in Indonesia. Proceeding 7th Joint Meeting ITA-COB, Stichting COB, Gouda, The Netherlands.

Admiraal, IB.M. 2006. A bottom-up approach to the planning of underground space. Tunnelling and Underground Space Technology, Volume 21, Issues 3-4, Pages 464-465.

Grev, E. 2008. Water control in Norwegian tunnelling. Proceeding South American Tunnelling 2008. Brazilian Tunnelling Committee - CBT, Sao Paulo. Brazil.

Parriaux, A, Tacher, L. & Joliquin, P. 2004. The hidden side of cities - towards three-dimensional land planning.

Energy & Buildings, Volume 36, Pages 335-341.

Parriaux, A, Blunier, P, Maire, P. & Tacher, L. 2008. The urban underground in the deep city project: for construction but not only. Proceedings Underground space challenges in urban development, Stichting COB, Gouda, The Netherlands.

13


New challenges in urban tunnelling: The case of Bologna Metro Line 1

G. Astore, S. Eandi & P. Grasso Geodata SpA, Turin, Italy

ABSTRACT: The Line 1 of Bologna Metro is 7 km long with 12 stations which crosses the entire city from the Fiera District to the Maggiore Hospital. The system adopted is a light rail tramway operated with 34-m long, single-unit vehicles. The line is designed to be completely underground and its construction involves the use of all the available tunnelling technologies: cut and cover, TBM and NATM. The portion of the vertical alignment in the city centre is very deep to avoid damage to buildings and to allow the line to underpass the new High Speed Railway tunnel near the Central Railway Station, where it is foreseen an interchange with the metro. The Piazza Maggiore Station is the most complex and important in the entire line and represents a great challenge, for designers in particular, because at this station site the horizontal alignment has a turn of 90° which has to be built completely by conventional tunnelling techniques.

1 INTRODUCTION

1.1 The Bologna Case

Located in northern part of Italy, Bologna, a medium-size city with about 400.000 residents, has an ancient historical centre and can be considered as the heart of Italian roads and railway network. In fact, its unique position, being in the centre of a crossroad linking north to south and west to east of Italy, creates a huge demand for public transportation systems.

As a matter of fact, to improve and change the city layout, at least four infrastructural works will be built in 10 years time: a new high speed railway station; a people mover connecting railway station to airport, an important line for trolley bus called (Civis); and an enforceable Metro line.

In the planned network, the metro Line represents one of the most complex works because of its length (7 km) and depth, alignment, and difficulties in constructing the civil works.

The main constrains are the presence of old-built city centre with historical monuments and the possibility of the archeological findings, entailing the metro line to be completely underground.

Geodata, leader of a team consisting of other design firms, has developed for the Commune of Bologna the Final Design of the whole metro line, which includes the civil works, the E&M installations and the tramway track system, besides the geological and environmental studies.

1.2 Description of the Metro System and the main project data

Bologna will be equipped with a semi-automatic tramway system with drivers aided by ACC (Automatic & Centralized Control system), a system which allows for monitoring and tele-control of trains and subway traffic, and by ATP (Automatic Train Protection), a system for controlling the speed and distance between trains as well as for managing the mobilization of trains, once they have arrived at the station.

Each station is to be equipped with platform screen doors that separate the platform from the train. These screen doors represent a relatively new technological addition to many metro systems

15

Figure 1. Schematic plan of the entire line 1 from Fiera (right on top) to Maggiore Hospital (left).

around the world, with some platform doors added to the existing systems later. They are widely used in Asian and European metro systems.

The modem low-floor rolling stock will run with 2' headway during peak hours in the underground section, elsewhere with 4' to 6' maximum frequency.

The principal characteristics of the line are:

• Total length = 7800 m

• Semi-automatic tramway system (Driver aided by ACC IATP technology)

• Underground stations = 12

• Surface stations = 1

• Ventilation shafts = 11

• TBM(Tunnel Boring Machine) tunnel length = 5600 m

• Cut & Cover tunnel and U -shaped section length = 800 m

• Conventional tunnel length = 200 m.

2 GEOLOGICAL AND GEOTECHNICAL SETTING

2.1 Geology

The geology of the top layers may be outlined through dividing the alignment into three sections, each of which has peculiar characteristics in terms of the depositional stratigraphy.

The first section (from Michelino station to FS station) and the third one (from Saffi station to Maggiore Hospital station), corresponding respectively to the eastern and western side of the route, are characterized by alternate presence of gravel-sand sedimentary layers from river channel and variable bands of silt and clay from floodplain. The intermediate section concerning the historical town centre (from FS station to Saffi station), by contrast, consists almost entirely of fine soils. Coarse sedimentary bodies are almost absent and the main stratigraphic markers are represented by paleosols, passing through soil bands, rich of organic matter on the top, to over-consolidated bands with a lot of carbonate concretions.

This stratigraphic layout is consistent with the well-known geomorphologic framework of the Bologna valley. In particular, in this case the extremes of the alignment cross alluvial cones, while the central section passes through predominantly fine inter-cone areas.

16

The soil-layers' attitudes reflect approximately the complex geological reality of the proj ect area, characterized almost exclusively by alluvial deposits which, by their lenticular geometry, show high vertical and lateral variations.

2.2 Geotechnical setting and hydrogeological regime

The design geotechnical model of the project has been developed on the basis of historical data and new survey campaigns carried out in 2007 in the course of its design development.

The depth of boreholes is directly linked to the depth of the stations and ventilation shafts, in order to provide a reliable definition of design parameters along the entire line.

During the site investigation many in-situ geotechnical tests were carried out such as Standard Penetration Test (SPT) and also undisturbed soil samples have been obtained for laboratory tests. Other in situ tests carried out were: cone penetration tests with piezocone (CPTU) to measure porewater pressure; in situ dissipation tests for evaluating the coefficient of horizontal consolidation and horizontal hydraulic conductivity; Seismic Dilatometer Marchetti Test (SDMT) for defining the dynamic properties of the soils. Finally all boreholes were equipped with a piezometer in order to measure the hydraulic head in the aquifers along the entire route.

In the first 40 m it can be found a multilevel groundwater aquifer. It consists in four different levels, partially saturated and locally under pressure, named respectively, from bottom to top: SUPI, SUP2, SUP3 and SUP4. Groundwater levels SUPI, 2 and 3 are located in sands and gravels, while SUP4 is included in sands, limes and silts, more superficial (-6 to -7 m from surface). In any case, SUP3 and SUP4 can not be easily distinguished, particularly where the soils change from one type to another.

Tunnel and stations do not touch the deepest groundwater level (SUPI), but they are affected mainly by SUP3 and SUP4

The Geotechnical units are reported below.

• Geotechnical unit A - Gravel layers

This unit is formed by lens of coarse sand, gravely sand, gravel with sand, and sandy gravel, and clasts up to approximately 8 em, as well as rare pebbles. The thickness of the soil lens are 6 to 8 m (NSPT = 15-50, in no case refusal). It is largely present in the Michelino-FS station section (Fig. 2) and partially in the section between Malvasia and Maggiore Hospital. The unit has good permeability and hosts major the aquifers of Bologna (SUPI, SUP2, SUP3)

• Geotechnical unit B - Cohesionless sands

This unit consists of uncemented sands, from coarse to fine, sometimes silty, predominantly saturated soils. SPT values do not show particular granulometric differences. The coarse sands are less compacted than the fine ones (NSPT < 10, until 2).

• Geotechnical unit C - Fine cohesive soils

This unit includes fine soils with a cohesive behaviour, mainly silty clay and clayey silt with peat trails. It is the unit that is most intercepted by the metro alignment, mainly in the section between FS Station and Saffi station.

Table 1 shows the geotechnical properties of the three units defined for the design.

3 TUNNEL CONSTRUCTION TECHNIQUES

3.1 Main constraints and adopted solutions

A very complex and time-consuming work in a historical, urbanized area is always a challenge for designers. In the case of Bologna Metro line 1, different and difficult items were considered in planning and designing the underground works such as tunnelling under water table in very poor ground conditions, underpassing of buildings particularly at historical places, protecting archeological findings related to the Roman era, siting of stations, exits and ventilation shafts

17

"""-

1MIoI1'illlllllu..N1III~ .-~~ .-r. •

. ~ ~

111:111;.~~~,~

,=-'IIiIIIiIIIIIII .. ~~

QI"I.!.~,CIlIOJPlJ 0--

.-.-.-.-.'-.-.-.----.---.-.-.-.-.---.~ . ."..._ .... _

0- ....

Figure 2. A part of the geological profile near Bolognina Station.

Table 1. Geotechnical parameters
Unit Yn[kN/m3] ({J~ [0] c'[kPa]
A 18-20 30-36 0
B 18-20 26-30 0
C 19-20 22-28 0-20 Cu[kPa]

E'[MPa]

50-150

40-80 20-40 10-40

both in city centre and in commercial and congested areas (at north-east of the town), minimizing construction-site areas, and managing intensive surface traffic.

In the light of the above constraints, the solutions adopted for the project are:

• the entire line for its 7.8 km length, is conceived to be underground with a configuration of single tube, double track and the designed solution provides systematic recourse to the mechanized tunnelling;

• external superficial tunnels are designed as cut and cover sections in order to realize connections between the depot and the Michelino station at the north side and between the Ospedale Maggiore station and the Malvasia station at the west side; and

• for special situations, short conventional tunnels are foreseen, like, for example, in the Piazza Maggiore station.

Specifically, the running tunnel will be realized using an EPB Shield with a diameter equal to 9.80m, which can ensure (Fig. 3):

• an internal tunnel diameter of 7.90 m (functional minimum = 7.80 m, plus 10 em of tolerance);

• tail void of 15 em;

18

..... .....

",

/ / I

I I I

I

Figure 3. TBM tunnel typical section (left) and EPB operational scheme (right).

• final lining thickness of 35 em with a nominal length of 1.4 m;

• each ring of precast segments is tapered to negotiate the curved tunnel alignment;

• the joints between segments are to be sealed using either neoprone or hydrotite gaskets.

The key objectives contributing to the above choices:

• altering as little as possible the original stress state of the soils;

• avoiding unnecessary, extra-excavations in order to control the excavation-induced effects on the surface (subsidence).

The EPB excavation mode can provide continuous support to the tunnel face, with the soils excavated by the cutting head accumulated under pressure in the excavation chamber and then extracted by a rotating conveyor.

Geodata has successful experience in tunnelling with EPB- TBMs in Bologna because in 2006 Geodata was involved in the construction studies and technical assistance during the works of two parallel tunnels (9,4m diameter and 6.112m in length each) of the Bologna high speed railway line (especially Lot 5 of urban penetration in the quarter S. Ruffillo in the south of the city and the new central station). Both tunnels were realized by EPB- TBMs. The success was reflected by the tunnel daily production rate and the solutions to prevent damage of buildings.

In designing the Metro Line 1 Geodata has used the technical and environmental known-how learned directly from the high-speed rail tunnel. In particular, back-analysis of the observed settlements have been made to determine the parameter values (Vp e k, O'Reilly and New,1992), which are necessary for input to the subsidence prediction and building risk assessment.

3.2 Risk analysis of buildings and soil improvement design

Potentially, buildings affected by tunneling are as many as 400, of which 35 are underpassed directly by line or station tunnels. In such situations, a comprehensive, detailed survey in-situ was carried out to collect and organize critical building data. The crucial parameters have been managed using a GIS system in order to faciliate the assessment of potential building damages due to underground works: excavation of tunnels, stations and shafts.

The evaluation of critical buildings revealed that where expected settlements are not compatible with prescribed safety limits, the design had to apply soil improvement: principally jet-grouting (Fig. 4) or compensation grouting in the very critical area like Bolognina.

19

-6.00

Jet-grollting.

PerforQCJ:ione Q VUQlo

Jet-9JOilting. I

Figure 4. Soil improvements: tunnel crown completely grouted (left) or grouted wall to protect building edges (right).

Figure 5. Typical station - assonometric view (left) and internal rendering of stairs connecting atrium with mezzanine.

4 TYPICAL STATIONS

4.1 Functional and architectural layout

The stations are the connections between the surface and the running tunnel and between the town and the line; with this point of view the functional layout of a station is defined in order to:

• reduce the length of the paths to and exit from the platforms;

• create a wide, bright space where passengers can easily recognize the right directions;

• locate access where pedestrian flows are most significant;

• minimize the volume of the entire station.

In particular, the architectural image chosen for each station is built on the purity of the volumes, in which are avoided scarcement and blind spots; the clarity of functional space, where stairs and elevators are always visible to users; and the usage of finishing and furnishing elements are chosen for simplicity and elegance.

With the above design principles in mind, 9 out of the 12 underground stations are designed according to a typological scheme, which is 18 m wide and 42 m long, and lies at an average depth varying from 15 m to 25 m. The platforms are separated from the tracks by platform screen doors.

20

Figure 6. Piazza Maggiore Station plan view and detail ofTBM passage.

4.2 Constructive method

Typical stations are to be built with the top-down variant of the cut and cover method. This solution permits to reduce construction times and worksite areas, In fact, the surface areas above the cover slab can be returned to the city for realizing parking, viability or construction depot.

The principal construction phases are: diaphragm walls built by hydromill; construction of concrete cover slab; excavation under the cover; construction of bottom slab; building other internal concrete works from bottom to the top.

The stations of Piazza Maggiore, Riva Reno and FS differ from the typological scheme. In the next section a description of the Piazza Maggiore station will be given, which is no doubt the most significant piece of work of the whole Bologna metro project.

5 PIAZZA MAGGIORE STATION

5.1 General problem

The Piazza Maggiore Station is located at a historical square of Bologna where two major roads cross the town centre: via Indipendenza and via Rizzoli, with a lot of critical and historical buildings and a lack of space to for construction sites. For all these constraints the Piazza Maggiore Station itself can be considered as a project in the metro project.

The station is 150 m long and positioned on a curve (with 25 m radius), the station platforms on the two sides shall be staggered in order to optimize the unusual shape and permit an easier train stoppage.

This station has 4 principal constituting elements (Figs 6, 7):

• Platform tunnel on a curve to be built with conventional tunnelling method;

• Large access shaft, used also as construction shaft;

21

ISECTION A-AI·

EXISTING ATRIUM

Access AND CONSTRUcnON SHAFT

PLATFORM CONVENTIONAL I TUNNEL

SOil GROUTING

Figure 7. Piazza Maggiore Station longitudinal section (A-A section in Figure 6) .

• Platform access tunnels to connect shafts with platforms;

• Existing underground atrium. This structure will be upgraded to create a new atrium for passengers, but during construction it will be used as storage area to minimize demand for surface area.

The 25 m curve is not a problem for the tramway alignment itself, because trains will approach the curve slowly, leaving after stopping at the platform. However, in the construction phase it has to permit the TBM to pass through that become a fundamental point in the entire work. Generally, a TBM of the required size can not excavate curves with a radius of curvature less than 200 m. Furthermore, in this particular case there is also not enough space to extract the TBM from a shaft located at one end of the station and lower it down at the other. For these reasons it is decided to create a conventionally-excavated platform tunnel with a wide cross-section shape to permit the shielded TBM to pass through the already-excavated station space.

The TBM passing phase can be detailed as follow: shield machine enters from the north (via Indipendenza) in the already-excavated tunnel; backup are dismounted; the shield is moved through the curve with a special trolley system to the new start position aligned to via Rizzoli; shield starts excavation in the new E- W direction with back-up re-connected to the shield.

5.2 Tunnel and shaft calculation

Such a complex work needs impressive studies and calculations to check the structural solidity and minimize risks in all construction phases, being in the town centre.

Construction phases have been studied for a long time with experts in conventional and mechanized tunnelling, ground improvements, underground works, etc. in order to be sure that such a work can be affordable. Principal construction phases have been used to define: preliminary and definitive lining, extension and type of soil improvements, internal tunnel shape (minimum to move TBM in curve), connection tunnels principal parameters (Fig. 9).

The circular shaft, from which curved platform tunnel and access tunnels should be constructed, will be constructed first. Being 20 m in diameter and 39 m deep with diaphragm walls of 45 m long, the circular shaft will be excavated using cut & cover top-down method. The shaft has intermediate slabs, every 4.8 m, to create support for stairs and to stabilize the diaphragm walls themselves.

Bottom slab has an arched invert shape in order to reduce bending moment and to transfer bending forces stemming from compressive ground water forces on the diaphragms.

22

Figure 8. Rendered assonometric view of the Piazza Maggiore Station.

CgmtrudjOD ptw,e 1:

J,t-'lfOU1;"9 r""" ... ,f ... er "0;"9 $pO" I ..... ~ "~~

Cc!"lf$truetjcn pl'!q;", J~

bcorolion ,,( plotlo,m eu rm;l "M<I 1'"'11 sIlopo)

CooS!ruCIi>n '",,0$1' 2:

[XCa:vtlOOn or ItoII cormeciian hoJl'! nel used es :si:art o,",mbor 10< pI.Uo,m e ..... d t.,",,]

CgfIstM''m1 oJl'hlJ~ _2:

[""""tion ot pIol' .. m ","""d lLmnol ;,,,,,,

Figure 9. Major construction phases of the Piazza Maggiore Station.

23

_0(01

'5.'1~ 103m ~'~~5.t1~ aUl~r(nq'''.4~

"'"

lko.-"" U""..cr'..sd__;.1t(I;r iJ:;«:Piiriitia

l.iIItOIi~~.h:!. :~.~USA.

Figure 10. 2D and 3D analysis performed in access shaft design.

Figure 11. Conventional platform tunnel FEM analysis. Model and settlements result.

Both 2D Finite Difference Method (FDM) using FLAC and 3D Finite Element Method (FEM ) using ANSYS have been used to evaluate bending moments, shear and axial forces on the diaphragm walls, creating an axial-symmetric model with 3D analysis to simulate the creation oflarge openings in the structure (Fig. 10).

For the curved platform tunnel, the top-heading and bench invert technique is considered to be more suitable given its big cross-section area of 170 m2. This large section requires some special attentions in calculating the internal structures and in evaluating the settlement effects on surface buildings.

This conventional tunnel is to be built in soils with very poor geotechnical characteristics and close to buildings and monuments. Thus, the construction is conditional to intense ground improvement. The techniques to be used are jet-grouting and soil freezing. Jet-grouting is used where there is enough space to work without large interference to surface activities, while soil freezing is used where surface works are not allowed. Compensation grouting is also foreseen in some particular situations like the under-passing of a critical historical building or to prevent settlements in the more critical sections of tunnel, where curve is at most.

Like for the shaft, the tunnel is also designed using 2D (PHASE 2) and 3D numerical analyses (FLAC3D), modelling the excavation sequences. A series of PHASE 2 analyses was carried out to identify the more suitable solution in terms of the construction steps, and the type and extent of ground improvement (Fig. 11).

24

6 CONCLUSIONS

The Bologna Metro line I represents a real challenge for designers. As a multi-constraint tunneling condition in urban area, the design of the Bologna metro project demanded special attentions to link properly the obstacle components of the projects, namely, conservation of a historical town, preservation of buildings and monuments, and minimization of the superficial impacts. All these constraints led to a very complex construction work. The job difficulty at Piazza Maggiore Station brought about also some challengeable design criteria, of which the most noticeable are: a staggered platform solution on curve, a new technical approach in using EPB TBM, a special solution to pass through the curve with a very small radius. The project also demanded for an extensive use of the available ground improvement techniques: freezing, jet-grouting, compensation grouting, etc. The solutions illustrated in this paper should be valuable for those who have to design similar underground works in analogous urban conditions.

REFERENCES

Amorosi A. & Farina M., 1994. Stratigrafia della successione quaternaria continentale della pianura bolognese mediante correlazione di dati di pozzo. 1st European Congress on Regional Geological Cartography and Information Systems, Bologna (Italy), June 13-16, 1994. Volume 5, 16-34.

Amorosi A. & Farina M. 1995. Large-scale architecture of a thrust-related alluvial complex from subsurface data: the Quaternary succession of the Po Basin in the Bologna area (northern Italy). Giornale di Geologia, 57/1-2,3-16.

Geodata S.p.A., 2007. Metrotranvia di Bologna - Final design documentation. Guglielmetti et. al., 2007. Mechanized Tunnelling in Urban Areas. Taylor & Francis.

Marchionni V. & Guglielmetti V. 2007 EPB- Tunnelling control and monitoring in a sensitive urban environment: the experience of the "Nodo di Bologna" construction (Italian High Speed Railway system), ITA-AITES World Tunnel Congress 2007 "Underground Space - the 4th Dimension of Metropolises" Prague, 5-10 May, 2007.

Grasso P. & Guglielmetti v., 2008 High-speed Railway Underground-Crossing Bologna, Italy, Workshop on "Tunnels in densely populated urban areas" - Professional Association of Civil Engineers of Catalonia Barcelona, 7 April 2008.

25


Numerical analysis of the effect of composite repair on composite pipe structural integrity

A. Belzowski & P. Strozyk

Wroclaw University of Technology, Faculty of Mechanical Engineering, Wroclaw, Poland

ABSTRACT: The paper deals with the problem of assessing the structural integrity of a composite reinforcement used to repair a nonpressure sewerage piping system with assembly damage. An analysis of the strain criteria applied to assess the structural integrity of polyester-glass composites was carried out. The technical aspects of the repair of the damage are discussed and a numerical model of the pipe section under repair is presented. The calculations made indicate the composite repair structural and material solutions' potential for further improvements and measures which are less effective in this regard.

1 INTRODUCTION

Polymer composites formed from glass reinforced plastics are characterized by relative lightness and stiffness and good resistance to environment action and caustic substances. Polymer composites are now commonly used to build all kinds of piping systems, such as:

• urban infrastructure networks - pressure piping systems for water supply, sewers without pressure, sewage treatment plant fittings, etc.;

• process plants, including storage tanks for petroleum derivatives and caustics, cooling circuit piping systems, flue gas pipelines in power plants, and so on.

The minimum service life of urban infrastructure piping systems is 50 years (EN 1796, 2006), (EN 14364, 2006). The life of 20--30 years, 40 years and 10--30 years is assumed respectively in the petrochemical industry, nuclear power plants (Le Courtois, 1995) and chemical-resistant tanks.

Polymer composites are susceptible to chemical and physical ageing and mechanical degradation (Bollaert & Lemascon 1999), (Tuttle, 1996) which adversely affect the material properties. As a result, the composite's elasticity modulus and strength during the anticipated service life may decrease by as much 10--60%. The changes in the composite properties are gradual but they can be predicted on the basis of accelerated ageing tests (Belzowski, 2005). Long-term test procedures for composites are usually modelled on the tests described in (ASTM D2992, 1991). Tubular specimens (minimum 18 pieces) are subjected to pressures which should result in a failure within 10000 h (14 months). By extrapolating the simple regression determined from the coordinates of the failure points one can estimate the strength (ASTM D2992, 1996) or stiffness (PN-EN 1120,2000) and (Farshad & Necola, 2004) of the material over its whole service life. The above procedure is considered to be a reliable but expensive way of predicting changes in the properties of composites. Irrespective of gradual degradation, most of the structures can be accidentally damaged (e.g. by impacts), which may additionally reduce their life.

Discontinuities in the protective layers (PL) on the inner surface of pipes and tanks in structures exposed to the action of corrosion factors pose a threat to their durability. PL discontinuities can be technological defects or service damage. Protective layers should protect the glass fibres constituting the structural reinforcement against the corrosive effect of liquids filling the system. PL discontinuities may drastically reduce the durability of the piping system.

27

2 REPAIRS OF PIPING SYSTEMS

Repairs on composite pipes, chemical resistant tanks and so on are carried out in order to remove:

• technological defects,

• local damage caused during transport or assembly and by incidental service overloads (e.g. accidental impacts),

• deterioration in the properties as a result of long-lasting material degradation.

The descriptions of repairs based on the composite reinforcement technology, found in the literature, focus on a few applications:

a. Repairs of chemical-resistant, composite (often high-risk) process facilities for storing dangerous caustic substances, etc. The principles of carrying out such repairs are described in (ASME RTP-l, 2000). The criteria of qualifying a device for repair and the permissible repair range (limited to 3-10% of the inner surface) are quite stringent.

b. The repair of the inner surface of whole sections of sewer pipelines by producing a new composite shell inside the old worn out conduit. Mainly preimpregnated sleeves hardened inside the pipeline by means of elevated temperature or UV radiation are used. The achievement of proper ring stiffness of the new shell can be the repair effectiveness criterion.

c. Repairs of steel and composite piping with local corrosion damage causing leakage, by making external sealing composite rings. Chemical and petrochemical industry process pipelines are repaired in this way. The principles of designing, carrying out and evaluating a repair are described in (ASME PCC-2, 2006). Besides repairs to high-risk pipework, the codes also cover repairs to low-risk pipework but with its diameter limited to 1000 mm. Repairs on leaking places in steel pipelines are discussed in (ABA Tech, 2005).

From an analysis of the standards and publications devoted to piping system repairs using composite-based technologies the following conclusions emerge:

• Most attention is devoted to high-risk piping system repairs.

• The criteria of qualifying damage for such repairs, defined in the standards, are quite stringent.

• The considered techniques of repairing local damage usually do not take into account the specificity of pipelines laid directly in the ground.

• The planning of repairs and their realization and technical acceptance include expert assessments of the damage extent and the effectiveness of the measures taken.

The above mentioned publications on repairs aim to ensure high professionalism in this regard.

This is understandable since most of the repairs are done on chemical resistant components of highrisk process plants. Professionalism is essential here as evidenced by descriptions of dangerous failures of unprofessionally repaired facilities (Belzowski, 2004), (Myers et al. 2007).

This paper presents a numerical analysis of the way in which assembly damage to a buried sewerage piping system without pressure was carried out, focusing on the comparison of the numerically calculated strains in selected points of the area subjected to repairs with the criteria for dimensioning composite components used in the construction of various piping systems. The possibility of improving the material-structural solution used in the repair is assessed. The pipeline section with damage was made from DN1400mm polyester-glass pipes with ring stiffness SN = 10000N/m2• The wall thickness was about 34 mm. Because of the improper application of the force damage in the form of cracks, chips and spalls appeared in the pipes as they were being shifted. The damage was analyzed in (Belzowski & Strozyk, 2008).

3 STRAIN CRITERIA OF DIMENSIONING COMPOSITES

The design of chemical resistant composite piping systems is based on the assumption that a 0.2- 1.0 mm long crack propagating across wall thickness represents dangerous damage. In the literature

28

Table 1. Allowable strain and safety factor values for chemical-resistant tanks and pipelines.

Standard

Strain constraints

Other dimensioning criteria

BS 4994

NFT 57-900 ASTMD3299 AD2000 Merkblatt and

WUDT-UC-UTSI01:10.03

8d::::: min(O.2%, 0.18B)*

8d::::: 0.2% and e.i ::::: 0.1 % ** L

8d s 0.1 % for cylindrical part under test pressure: 8j_ ::::: 0.2% in

direction .L to UD reinforcement

and 8 ::::: 0.35% for CSM reinforcement 8d::::: 0.1 % in wound components

8d :::::0.25%

8d = 0.09%,0.12%,0.15%,0.18% depending on temperature and environment

ASMERTP-1 EN 13121

BS 7159

8 = 10 for hand laminating 8::::4

LCU1.3****

* 8d - allowable strain for composite, 8B - failure strain for resin ** 8 .L - strain perpendicular to UD reinforcement fibres

*** 8 - load or stress safety factor

**** LCL -long-term strength for failure probability of 2.5%

it is referred to as the first ply failure (FPF) criterion. This crack size approximately corresponds to a single ply of typical chemical resistant (polyester-glass) laminates. Assuming a safety factor of 1.5 for the so defined damage threshold (Eckold, 1985), the criterion of allowable linear stress ed, which usually amounts to 0.1-0.25% (Table 1), was introduced into the design standards. This strain limitation protects the structure against the development of transverse cracks which would facilitate the infiltration of liquids from the inside of the piping to the structural layers containing glass fibre E which is susceptible to corrosion. Since the failure strain of polyester-glass composites predominating in such applications amounts to about 1.5-2.5%, relatively high safety factor values are obtained as a result. The actual strain safety factor values assumed for chemical-resistant process plants mostly amount to about 6--10. The load safety factor values are similar.

Interesting conclusions emerge from (AEA Tech, 2005) devoted to repairs ofleaking metal pipes.

For one of the categories of composite repairs on high-risk equipment, defined there as Class 3, the allowable strain for a repair life of 2 years amounts to 0.30%. For a repair life of 10 years and 20 years Cd = 0.27% and ed = 0.25% are respectively assumed. The allowable strain for composite repairs on low-risk equipment is respectively: 0.40% for a repair life of 2 years, 0.32% for a repair life of 10 years and 0.25 for repair life of 20 years. The above strain values show the conservatism of the authors of (AEA Tech, 2005) in the assessment of repair durability. The assumed allowable strains for composite repairs are not much different from the ones used in the design of high-risk chemical-resistant equipment (Table 1). The highest assumed allowable strain amounts to 0.40% and it applies to low-risk equipment repairs with a life of 2 years. The above analysis takes into account only the strain criteria used in the design of composite repairs, neglecting the calculations based on the long-term strength values of the laminates.

The initial safety factor for pipes is defined as a ratio of the relative deflection of the pipe along its diameter (Figure 1), reached until the first symptoms of failure, to the allowable deflection. According to standard EN 14364, a composite pipe with ring stiffness SN10000N/m2 should withstand deflection amounting to minimum 15% of its mean diameter. For an allowable deflection of 6% the value of this safety factor in new pipes amounts to at least 15/6 = 2.5. It was assessed that the pipes would meet this safety factor requirement also after a repair. It was also found (through numerical calculations) that linear strain c in selected points of the reinforced area considerably exceeded the allowable values assumed in the various codes and studies concerning repairs: (AEA Tech, 2005), (ASME PCC-2, 2006). Moreover, one cannot ignore the fact that the required service life of the pipeline is 50 years and it much exceeds the repair durabilities adopted in (AEA Tech, 2005) and (ASME PCC-2, 2006), which amount to 2-20 years.

29

Figure 1. Ring stiffness test. Deflectiony = 0.03 x dme is used to determine stiffness SN. Allowable deflection amounts to 0.06 x dme. Symbol dme denotes mean diameter.

Similar disturbing conclusions emerge from the investigations described in (Farshad & Necola, 2004). On the basis of tests lasting up to 1000h the regression line for pipes DN500/SNI0000 loaded as in Figure 1, but in addition immersed in sulphuric acid with a concentration of 5%, was determined. The acid solution is used to model the effect of sewage on structural materials. It was found that hoop strain at the moment of failure decreases from 2.1 % in the short-term test to 0.5% after 1000 h and to as little as 0.16% after 50 years. This is below the values assumed for the design of chemical-resistant piping systems (Table 1) in the standards which are highly conservative as regards the choice of allowable strain values.

From the above investigations and analyses one can conclude that the maximum linear strains in the circumferential direction of the investigated pipes, loaded as in Figure 1, should not substantially differ from the values assumed in the design of chemical-resistant facilities. Although the corrosive effect of the water filling the analyzed pipeline is generally less harmful than that of sewage, studied by (Farshad & Necola, 2004), the required service life is very long (50 years).

The technical aspects of repairs and the results of calculations for different material-structural designs of the composite repair are presented below.

4 DAMAGE AND REPAIR - TECHNICAL ASPECTS

Depending on the extent of damage, several repair methods were proposed. The analyses presented here concern cases of quite extensive damage repaired through local composite reinforcements inside the pipe. The extent of damage in directions tangent to the inner surface mostly amounted to 100 mm (Figure 2), but in some cases it was larger (e.g. 100 x 250 mm in Figure 3).

According to the ASME RTP-1 criteria for high-risk chemical-resistant facilities, the damage did not qualify for repair because of its extent. Also the criteria of acceptance for repairs of damage to low-risk facilities, described in ASME PCC-2 were exceeded. On the basis of its long operating experience the manufacturer of the pipes proposed its own procedures for assessing the damage and carrying out repairs.

The damage affected the protective and structural plies of the pipes. The repair consisted in removing the damaged material (Figure 4), filling in the cavities with a polyester filler (UP) and making an inner reinforcing ring composed of five layers of polyester laminate with glass CSM with a mass of 600 glm2 (Figure 4). The ring width towards the axis of the pipe was 400 mm. Finally, a protective layer made of resin was formed.

30

I~ ~~'.

~

•

Figure 2. Damage at pipe's end (37) and small impact damage distant from edge (17). Dark gap between pipes is about 15 mID wide.

Figure 3. Cracks and chips in laminate.

Figure 4. 100 x 200 x 15 mID cut-out left after removal of cracked material.

Figure 4 shows that the void left after the removal of the material extends deeply into the structural layers, disturbing their integrity. An expert assessment showed that the pipe's initial load capacity and stiffness in the circumferential direction could be recovered by means of the inner reinforcing ring. It was assumed that in a pipeline which does not carry axial pressure the axial load capacity and stiffness of the pipes are of secondary importance.

31

Figure S. Longitudinal section of pipe wall and denotations of individual layers.

Table 2. Denotations and structure of pipe wall layers and reinforcement variants.

Layer denotation

Layer function

Structure and reinforcement

Ml M2 M3 M4 MS M6 M7 M8 M9

outer protective outer structural pipe core

inner structural inner protective resin filler

repairing ring - variant A repairing ring - variant B repairing ring - variant C*

UP resin with sand

UP resin with staple fibre UP resin with sand asinM2

UP resin

UP resin

UP resin with CSM

glass fabric WR + UP M7/M8/M9/M8/M7/MS + UP

* In variant C the following system of reinforcement layers (moving from the contact with the pipe towards the ring inner surface): M7: 2.2mmIM8: O.84mm1M9: 1.4mm1M8:

O.84mm1M7: 2.2mm1MS: O.Smm was adopted.

5 MATERIAL PROPERTIES AND PIPE MODEL

Figure 5 shows the structure of the wall of the repaired pipes. Thickness twas 34 mm. Layers M1, M2 ... are described in tables 1, 2. It follows from the structure of the reinforcement that most of the considered layers can be regarded as isotropic in volume (M1, M3, M5, M6) or in plane (M2, M4, M7). Only layer M8 and the laminate in variant C, representing improvements in the composite repair, are moderately anisotropic. The values of Poisson ratio v and Young's modulus (E) of the layers are give in Table 2.

First the 300 mm long undamaged pipe section was modelled in ANSYS. Using SOLID 45 elements with 8 nodes the system of layers M11M21M31M41M5 was modelled starting from the outside and moving towards the inner surface (Figure 5). The layers are described in Tables 2 and 3. There were about 36000 elements.

The aim of the calculations was to match the thickness and properties of the particular layers so as to obtain the actual pipe ring stiffness close to the nominal one (SN10000 N/m2). In the ring stiffness test, the ring cut off the pipe is loaded as in Figure 1 until deflection 0.03 x dme = 42 mm is reached in the case of the considered pipes. It was assumed that the allowable pipe deflection is 0.06 x dme = 84mm (Madryas et al. 2002, EN 14364). When the properties of layers M1-M5 given in Table 3 and the load twice as heavy as the one corresponding to the rated ring stiffness were substituted the maximum deflection of 75 mm was obtained. The ring stiffness of the pipe model is about 11100 N/m2 and it is higher than the nominal one (10000 N/m2). The stiffness of new pipes is as a rule higher than the rated stiffness.

32

Table 3. Thickness of layers and adopted proper mechanical properties.
Layer Layer thickness Young's modulus* Poisson
symbol (mm) (GPa) ratio**
M1 1.9 12.8 0.23
M2 204 8.8 0.3
M3 19.6 10.5 0.23
M4 7.7 8.8 0.3
M5 204 3.0 0.3
M6 10.1 3.0 0.4
M7 7.25 8.0 0.2
M8 7.25 18.0 0.13
M9* 0.84 35.0 and lOA 0.281 and 0.083 * For the unidirectional layers (UD) axial and transverse Young's moduli are given; the same applies to the Poisson coefficients. Longitudinal elasticity modulus G = 3.2 GPa was assumed for layers M9.

Place of repair

Figure 6. View of reinforcing ring model.

The extreme hoop strains for the deflection of 75 mm amounted to: emax = 0.607 x 10-2 and emin = -0.598 x 10-2• Assuming failure strain under bending eB = 1.6 - 2.2 x 10-2 (according to the manufacturer specifications) one can find that the safety factor amounts to 0.606/1.6 = 2.64 and it is close to the one required by standard EN 14364. Numerically determined strain emax = 0.607 x 10-2 exceeds the strain values assumed for high-risk chemical-resistant facilities (Table 1), but the analyzed pipeline does not belong to this class of equipment.

6 MODELLING OF COMPOSITE REPAIR

6.1 Inner ring reinforced with glass mat

Then a model of a situation similar to the one shown in Figure 3 (200 x 200 mm cavity with its depth equal to the sum of layer M4 and M5 thicknesses, located in the pipe's highest place) was created inANSYS. Under the load as in Figure 1, the cavity is the area of the highest bending moment. The cavity was filled with a material having the properties of layer M3. Then a reinforcing ring with width b = 400 mm and thickness t 1 = 7.25 mm, made of a material with the properties of layer M7 (Tables 2, 3, Figure 6) was introduced into the pipe. In Table 2 this is denoted as variant A (ring W7/fillerW3).

According to standard ASME RTP-l, five layers reinforced with CSM with a mass of 600 g/m2 have the above thickness, which corresponds to the structure of the rings formed during the repair. The protective layer formed from resin and surfacing veil, laid as the last one, was not taken into account in the calculations.

33

~. 1!X • .unH:::D II(;C" Iit:M

I\N

""'" ~I 20~1 OO,Ol"U

Figure 7. Locations of selected analyzed points in repair area.

Certain simplifications were made in the analyzed model of the pipe under repair, i.e .

• the influence of the end-of-pipe location of the repaired damage was neglected and

• the loading scheme used for determining ring stiffness (Figure 1) was adopted.

At a pipe deflection of 55 mm, the hoop strains in the most stressed points of the ring, denoted as 4 and 8 in Figure 7, amounted to: C(4) = 0.494 X 10-2 and C(8) = 0.564 X 10-2•

The highest hoop strains e in material M3 filling the cut -out left after the damage had been removed amounted to 0.342 x 10-2•

When filling material M3 was replaced by the less stiff filler M6 (variant AI), the hoop strain in point 4 increased by 18.4% (up to 0.585 x 10-2) and in point 8 by 2.8% (up to 0.580 x 10-2). Hoop strain also increased in the filling - its maximum value in material W6 was 0.395 x 10-2• Even though the stiffness of the pipe after the repair is considerably higher than the required one, the calculated hoop strain values in the ring are quite high - amounting to about 0.5-0.6 x 102•

6.2 Inner ring reinforced with glass fabric M8

Similar calculations were done for variant B (ring M8/filling M3 as in table 2). The main difference with variant A is that the ring is reinforced with material M8 with a higher stiffness than that of the previously used M7. In comparison with the results obtained for variant A, hoop strain decreased in the ring's points 4 and 8 down to respectively 0.356 x 10-2 and 0.416 x 10-2. As a result of the replacement of M7 by M8, the value of Cmax in the ring decreased by (0.564 - 0.416)/0.564 = 26%.

Hoop strain cmax in the filling in variant B is lower than in variant A (previously it amounted 0.342 x 10-2 and now it is 0.247 x 10-2).

Preserving ring reinforcement, M8 filling material M3 was replaced by M6 (variant Bl). Similarly as in variants A and AI, this resulted in an increase of hoop strain Cmax in the filling area from 0.231 x 10-2 to 0.247 X 10-2 (by 7%). Strains cmax in the ring also increased.

6.3 Inner ring reinforced with laminate containing UD layer

Then reinforcing ring structure variant C described in Tables 2 and 3 was calculated. This ring structure takes into account the general principles of designing and repairing chemical-resistant facilities. The introduction of a UD layer with a circumferential orientation did not result in any

34

Table 4. Strains and deflections before and after increase in ring thickness.
Reinforcement Ring thickness Strain8max Strain8max Deflection
variant (mm) in ring in filling (mm)
A 7.25 0.564 x 10-2 0.342 x 10-2 55
Al 7.25 0.585 x 10-2 0.395 X 10-2 55
B 7.25 0.416 x 10-2 0.231 X 10-2 44
B1 7.25 0.422 x 10-2 0.247 X 10-2 44
Ax 1.5 10.90 0.529 x 10-2 0.263 X 10-2 46
Al x 1.5 10.90 0.538 x 10-2 0.287 X 10-2 46
B x 1.5 10.90 0.360 x 10-2 0.152 X 10-2 35
B1 x 1.5 10.90 0.362 x 10-2 0.157xlO-2 35 significant reduction of strain in the ring and in the filling in comparison with variants A and B. In point 4 strain amounts to e = 0.381 x 10-2 while in point 8 to e = 0.445 x 10-2• In the area filled with material M3 strain amounts emax = 0.248 x 10-2•

6.4 Effect of increased ring thickness

In the considered reinforcing ring structure variants (A, B, C), pipe stiffness in the reinforcement area was sufficient, but the maximum strain, amounting to 0.4-0.6 x 10-2, was relatively high (section 3 in the paper). Therefore it was decided to determine what effect an increase in wall thickness by 50% will have on hoop strain. In discussed variants A, AI, B, Bl the ring thickness was 7.25 mm. Now it was assumed to be 1090mm. The obtained calculation results are shown in Table 4.

7 CONCLUSIONS

The analysis of the literature on piping system repairs by means of polyester-glass composites shows that strains in the repaired components of low-risk piping systems should not significantly diverge from the allowable strain values assumed for high-risk chemical-resistant equipment (table 1). Allowable values e = 0.25-0.3 x 10-2 are proposed.

The stiffness of glass CSM is insufficient for it to be used in the structural layer of such rings reinforcing the repair area. Glass fabric provides more effective reinforcement, significantly reducing hoop strains.

The elastic properties of the material filling the cut-out left after damage removal affected only slightly the hoop stress and pipe deflection values.

The introduction of a UD layer into the ring structure did not result in a substantial reduction in hoop stress. This may be ascribed to the fact that a very small quantity of the UD reinforcement was used.

When the ring thickness was increased by 50%, the reduction in hoop stress was smaller than expected. It seems that hoop stress can be further reduced by increasing the ring's thickness or the UD reinforcement fraction in its structure.

The obtained results show that the optimization of such composite repairs has great potential.

The paper does not cover problems relating to possible improvements in repair techniques, the effectiveness of the latter or durability assessments.

REFERENCES

AD 2000-Merkblatt. 2000. Pressure vessels in glass fibre reinforced thermosetting plastics. ABA Technology. 2005. Design of Composite Repairs for Pipework.

35

ASME PCC-2 Repair standard. 2006. Non-metallic composite repair systems for pipelines and pipework. ASME RTP-1. 2000. Reinforced thermoset plastic corrosion resistant equipment.

ASTM D 2992. 1996. Standard Practice for Obtaining Hydrostatic or Pressure Design Basis for Fiberglass" (Glass-Fiber-Reinforced Thermosetting-Resin) Pipe and Fittings.

ASTM D3299. 1995. Standard Specification for Filament-Wound Glass-Fiber-Reinforced Thermoset Resin Corrosion-Resistant Tanks.

Belzowski A. 2005. Koncepcje oceny wytrzymalosci dlugotrwalej polimerowych kompozyt6w konstrukcyjnych. V Konferencja KOMPOZYTY POLIMEROWE.: 5-37. Politechnika Warszawska.

Belzowski A. 2004. Badanie uszkodzefi zbiornika z larninatu wzmocnionego wl6knem szklanym. Raport SPR, nr 9. Politechnika Wrodawska.

BelzowskiA. & Strozyk p. 2008. Assessment of repair reinforcement of polyester-glass fibre pipe. Composites 8 nr 2: 179-184. Polish Society for Composite Materials.

Bolleart F. & Lemascon A. 1999. Analyse de defaillance pieces plastiques, elastomeres ou composites. Guide

Pratique. CETIM, France.

BS 4994. 1987. Specification for design and construction of vessels and tanks in reinforced plastics.

BS 7159. 1989. Code of practice for design and construction of glass-reinforced plastic piping systems. ECKOLD G. 1985. A design method for filament wound GRP pressure vessels and pipework. Composites, V.

16, Nr 1: 41-47.

EN 1120. 2000. Plastics piping systems - Glass-reinforced thermosetting plastics (GRP) pipes and fittings - Determination of the resistance to chemical attack from the inside of a section in a deflected condition.

EN 1796. 2006. Plastic piping systems for water supply with or without pressure. Glass-reinforced themosetting plastics (GRP) based on unsaturated polyester resin (UP).

EN 13121. 2001. GRP tanks and vessels for use above ground.

EN 14364. 2006. Plastic piping systems for drainage or sewerage with or without pressure. Glass-reinforced thermosetting plastics (GRP) based on unsaturated polyester resin (UP). Specification for pipes, fittings and joints.

FARSHAD M., NECOLA A. 2004. Strain corrosion of glass fibre-reinforced plastics pipes. Polymer testing, 23: 517-521.

Le Courtois T. 1995. PWR Composite Material Use: A Particular Case of Safety-Related Service Water Pipes.

Proc. of Enercomp 95: 835-843. Technomic Pub., Montreal.

MADRYAS C., KOLONKO A. WYSOCKI L. 2002. Konstrukcje przewod6w kanalizacyjnych. Oficyna Wyd.

Politechniki Wrodawskiej, Wrodaw.

MYERS TJ., KYTOMAA H.K. SMITH T.R. 2007. Environmental stress-corrosion cracking of fiberglass:

Lessons learned from failures in the chemical industry. 1. of Hazardous Mater. 142: 695-704.

NFT 57900. 1987. Reservoirs et appareils en matieres plastiques renforcees,

Tuttle M.T. 1996. A framework for long-term durability predictions of polymeric composites. Progress in Durability Analysis of Composite Systems: 169-176. Balkema, Rotterdam.

URZ1\D DOZORU TECHNICZNEGO. 2003. Urzadzenia cisnieniowe. Stale zbiorniki cisnieniowe z tworzyw sztucznych wzmocnionych wl6knem szklanym. WUDT-UC-UTSI01.

36


Repair of RC oil contaminated elements in case of infrastructure

T.Z. Blaszczyriski

Civil and Environmental Engineering Faculty, Poznan University of Technology, Poznan, Poland

ABSTRACT: RC elements in case of underground and network infrastructure, are subjected to the activity of different agents, i.e. water, chemicals and oil-products. Because of it so important is their durability. Indispensable are then suitable repairs, because the possibility of the underground exchange of elements is comparatively difficult and expensive. In the relationship with the environment which can appear in objects of the infrastructure, repair methods must be resistant to the activity of different environments, and must also be prepared to the repair of concrete surfaces with the different state: moisture and oiling. Comparing the influence of various oil products on compressive strength of concrete, leads to the conclusion that there are large differences in effects. The repair of oiled structures is technologically difficult. The paper will present research works on some nano- and modern technologies in case of oiled RC structures.

1 INTRODUCTION

In view of environmental influences and structural solutions the separate repair procedure often is also required. If one takes under consideration that concrete is a product of simple technology and complicated knowledge which begins to be only mastered, then right prognoses are, that XXI century for the building construction can be in large measure the age of repairs, rehabilitations and demolitions. The concrete is relatively cheap material in construction, however its repair during use of the property is very costly. Especially depressing and expensive, and sometimes even operable impossible, is the repair of already completed repair. Repairs became a civilisation-wide problem (Kucharska 2001).

The acid content of crude oil products might be affected at the oil/water interface as a result of bacterial activity. Aerobic oil oxiding bacteria (OBB) produce acetic acid, which can reduce the pH level of water to 5. In addition, anaerobic sulphate reducing bacteria (SRB) also operate to produce H2S, which can be converted to H2S04• This latter process has often been encountered in sewers. Research at Imperial College has indicated that the inorganic acid content (Onabolu at all 1985) of the crude oil products is unlikely to increase as a result of bacterial activity.

That kind of environment create three different corrosion mechanism: biological, chemical and physico-chemical (Fig. 1). The less known is physico-chemical process, which take place in case of oil environment with low neutralisation number.

2 MINERAL OILS INFLUENCE ON CONCRETE

Long term laboratory experiments have been conducted to assess the changes of physicomechanical characteristics of oil contaminated concrete. The compressive strength was determined from 100mm cubes, according to EN 12390-1:2001, for concrete type C20/25 as most commonly used for industrial RC structures in Poland (5 specimens, Sfc = 0.84-2.87, Vfc = 2.25%-6.99%). The average 28 day compressive strength of concrete was!cm = 29.8 MPa. The water-cement ratio was 0.59 and aggregate-cement ratio was 6.70.

37

Figure 1. An example of oil/water environment in case of RC structure.

60

50 40 30 20 10

o;-:=;~~~

048

12 16 20

24 28 32 36 48

tz [months] 60 64 68 72

Samples M-40 TU-20

Figure 2. Variation of concrete C20/25 compressive strength during the period of exposure to H70, TU20 and M40 oils (Blaszczyriski 2002).

Concrete was impregnated with the most commonly used industrial oils of different kinematic viscosities namely turbine oil TU20 (81 mm2/s), machine oil M40 (211 mm2/s) and hydraulic oil H70 (383 mm2/s). These oils have low neutralisation numbers with values between 0.05 and 0.075 mgKOHlg. The oils was first applied to concrete 2 months after casting, subsequently the specimens were examined every 4 or 12 months during total period 72 months. The control specimens (samples) were additionally examined after 28 days and 2 months (Fig. 2). Time of oiling tz is started in age of two months with an average compressive strength of concrete fem = 37.35 MPa.

The results clearly show, that as a result of the influence of the oils used, the different degree of decrease of concrete compressive strength (comparing to control samples): from 55% for oil H70 to almost no influence for oil M40. Oils H70 and TU20 affected the analysed concrete compressive strength!cm the most.

Contamination of concrete by hydrocarbons gives an almost new material, which behaves differently. The results of the stress «()e) - strain (ee) relation in non-oiled state and after 12-months of oiling by mineral oil TU20 for concrete C20/25 in function of the longitudinal strains are different. The non-linear behaviour of strength and strain variations depend on the contents of hydrocarbon and its type. It can be noticed that the strain eel, corresponding to the maximum stress, is lower for oil saturated concrete than for non-oiled concrete (Fig. 3 - Blaszczyriski 2006).

38

45
40
35
30
8!. 25
6
o 20
b
15
10
5 ~ ~
-
/ \
-: - \
", ~ non-oiled
J .......
oiled
/Y
E
f
° 0,0

0,5

1,0

1,5

e C [1,0]

2,0

2,5

3,0

Figure 3. (Fe - Be diagram for non-oiled and oiled by oil TU20 concrete C20/25.

~ 100
2.....
N
U.
r::::
0
U
(I) 50
II)
"0
~
(5 0.05 mm 0.2 mm 0.5 mm 1.0 mm

a,

~

Crack width 0.3 mm

~ 100

u.N r:::: o

~

II)

"0

~

(5

50

O~--~-r+_--+--r-r_+_;~ 1

O+-----~--~-----+----~----+--;~ o

234510203040 Time of oiling tz [days]

0.05

0.10

0.15 0.20

0.25

Distance from crack a, [m.]

Figure 4. Percentage of section oiling (TU20): at in case of different crack width, bl in section grow away from crack (Blaszczyriski 1983).

3 OIL PRODUCTS PENETRATION INTO RC ELEMENTS

During the research on the influence of mineral oil penetration into crack's zones, it was shown that oiling kinetics in case of smaller cracks is greater (Fig. 4.a). Besides of this also intensive oiling of all the crack region occurred (Fig. 4.b).

This mechanism is a basic mechanism of the oil products penetration into RC elements. The permeability through porous material, like concrete, can be defined as the flow of liquid through capillary channels, pores aerial and others (about different sizes, often joint with itself by the net of microcracks) (Matti 1976). The knowledge of concrete permeability is important, because it exerts the influence on the resistance of aggressive liquid environment, and consequently decides - along with the reactivity parameter - about its durability. The concrete permeability in the greater degree is relative to pores structures than to the general porosity (Kagimoto 2000).

39

100
~ 80
OJ 60
~
·0
'5 40
..r:::
C.
Q)
Cl 20
0 1\ '"
\ ","on oiled
\ partialy -,
oiled
\ \
r- oiled --
I 4.5 9.1

85.7

~ 100
OJ
~
·0
'5 87
..r:::
C.
Q)
o
75 - //
/ /
/
TU20 M40 ( H70 1198

30 60 90 30 60 90 30 60 90 Time of oiling tz [dni]

Kinematic viscosity 11k [mm/s2]

Figure 5. Mineral oil kinematic viscosity influence on depth of section oiling: a/ concrete after 13 months of oiling (Manns 1977), bl cement mortar after 3 months of oiling (Blaszczyriski 1983).

The permeability of so heterogeneous material surely is relative to its internal structure, however insignificant is also viscosity of interfering liquid TJk. what is represented in Fig. 5 (Blaszczyriski 1983, Manns 1977).

4 REPAIR TECHNOLOGY OF MINERAL OIL CONTAMINATED RC STRUCTURES

There can be many causes of deterioration in a concrete structure. Concrete repair is a specialist activity requiring fully trained and competent personnel at all stages of the process. Up until now there has been no common European Standard in this field. Often simple "patch and paint" strategies have been employed as short term cosmetic repairs which have failed to address the root cause of the problem. This can, and has, lead to dissatisfaction from building and structure owners.

The new European Norm EN 1504 will standardise repair activities and provide an improved framework for achieving successful, durable repairs and satisfied clients.

Importantly, this Norm (expected date of full implementation: 31.12.2008) will deal with all aspects of the repair process including:

• definitions and repair principles,

• the need for accurate diagnosis of causes before specification of the repair method,

• detailed understanding of the needs of the client,

• product performance requirements, test methods, material production control and evaluation of conformity,

• site application methods and quality control of works.

All of this especially difficult in case of oiled structures. Today we have got many modem technologies even nanotechnologies. Nanotechnology does not mean nano-sized particles. The better understanding of cement hydration has allowed us to improve the quality and density of the nanostructures in cement paste. This reduces micro-defects in the systems and improves bond between the cement matrix and the aggregate, and, the cement mortar with the substrate. Physical properties such as tensile strength are improved to reduce the possibility of cracking. This is the basis of applied nanotechnology in cement systems.

The purpose of all range of superficial treatments is the assurance of suitable adhesion of repair material. This decides often about the efficiency of the repair. For structural repairs is recommended, and according to ENV 1504-9 even required, the usage of bond layers.

For the purpose of detailed analysis of above problems two series of tests was done. In the first series of tests the efficiency of crack repair methods was taking into account. In the second series the efficiency of surface protection method was tested. Before using any repair treatment on oiled

40

Figure 6. Oiled concrete surface treated with modern paste based on solvent with absorptive solid material: a/ oiled surface, bl first treatment, cl second treatment, d/ last treatment.

Figure 7. Oiled concrete surface in case of split samples treated with modern paste based on solvent with absorptive solid material: a/ splited sample, bl surface after treatment, c/ surface after steel-brushing.

Figure 8. Split test in case of concrete cylinders: a/ oiled concrete before bonding, bl bonded with epoxy resin.

concrete surface the proper surface preparation is necessary. This preparation should give clean surface without any oil particles on it.

4.1 Surface preparation

In case of oiled surface cleaning is necessary. There was used two types of cleaning one typical with the emulsifying medium and the second modem with the paste based on solvent with absorptive solid material, treated on the surface three times (Fig. 6). After each treatment concrete was cleaned with steel-brushing.

The same treatment was done in case of split sample, necessary for crack repair testing (Fig. 7).

4.2 Cracks repairs

Crack repair tests was done based on cylinders (160 x 160 mm) using split test. After 6 years of oil TU20 influence cylinders was splited and bonded by two kinds of resin: epoxy and polyurethane (both the most popular in crack repairing). After week time the bonded cylinders was also splited (Fig. 8).

41

2,5
'i? 2
n,
~
..c: 1,5 o Sample
"5
c:: • Epoxy resin
~
iii o Polyurethan resin
~
·iii
c::
Q) 0,5
I-
° Figure 9. Split test results of oiled concrete cylinders bonded with two kinds of resins.

Figure 10. Examination of analysed layer adhesion to concrete surface.

Figure 9 is presenting the average split tensile strength of all kind of analysed elements (6 samples each): oiled cylinder, oiled cylinder bonded with epoxy resin and oiled cylinder bonded with polyurethane resin.

The split test results clearly show, that in case of both resins crack in oiled concrete could be repaired with success. Better connection is of course in case of epoxy resin (20% of original split tensile strength loss), but the crack connection made by polyurethane resin gives only 38% of original split tensile strength loss. Comparing the cost of both these resins satisfactory effect for polyurethane resin is visible.

4.3 Surface protection

Apart from crack repairing the surface protection is also necessary, both in case of new structures or cleaned and repaired. The variety of surface protection materials is high. For testing procedure paintwork based on copolymer vinyl-acetyl-ethylene, epoxy and polyurethane coat, was used. For the purpose of surface repairing two kind of cement repair material was checked. One ordinary and one nanomaterial. Both were single component. The nano repair layer was very high strength, high modulus, fibre reinforced, shrinkage compensated, expansive, structural repair mortar.

On prepared samples with pull-off test adhesion of analysed layer to oiled (TU20) and clean concrete surface was examined. The method of research consisted in the measurement of minimal force applied perpendicularly to sample layer (Fig. 10).

42

I Dclean concrete .oiled concrete I I D clean concrete .oiled concrete I

2,5 3
'iii' 'iii'
n, n, 2,5
~ 2 ~
II) II) 2
II) II)
Q) 1,5 ~
.....
ti ti 1,5
OJ OJ
C C
~ ~
ra 0,5 ra
c. g- 0,5
Q)
C/') C/')
0 0
a b a b c
a - nano system a - epoxy coat
b - ordinary system b - vinyl- acetyl-ethylene copolymer
c - polyurethane coat Figure 11. Adhesion (separation) stress results: a/ repair layer, bl paint layer.

Obtained results testify hereof that values of bond force to oiled concrete surface are lower aside from all used surface materials (Fig. 11).

The required adhesion, measured by pull-off test, for not structural repairs amounts above 0,5 MPa, and for structural repairs must be above 1,5 MPa. Comparing all results is clear, that in case of repair materials only single component nano-material has still adhesion above this value. In case of paintworks all of them have value of oiled surface adhesion still higher, then in above technical recommendations.

5 CONCLUSIONS

The results clearly show the different degree of decrease of concrete compressive strength (comparing to control samples): from 55% for oil H70 to almost no influence for oil M40. Oils H70 and TU20 affected the analysed concrete compressive strength/em the most. Contamination of concrete by hydrocarbons gives an almost new material, which behaves differently. The non-linear behaviour of strength and strain variations depend on the contents of hydrocarbon and its type.

The split test results clearly show, that in case of both used resins crack in oiled concrete could be repaired with success. Better connection is of course in case of epoxy resin (20% of original split tensile strength loss), but the crack connection made by polyurethane resin gives only 38% of original split tensile strength loss. Comparing the cost of both these resins satisfactory effect for polyurethane resin is visible. Apart from crack repairing the surface protection is also necessary, both in case of new structures or cleaned and repaired. In case of repair materials only single component nano-material has still adhesion above the technical recommendations. All of tested paintworks have value of oiled surface adhesion still higher then necessary.

REFERENCES

Blaszczyriski T., Kozaczewski J., Nowakowski B. 1983. About the mineral oils influence on physical and bearing concrete features, Proceedings of XXIX Polish Scientific Conference, Poznan - Krynica, 11-16, (in Polish).

Blaszczyriski T. 1994. Durability analysis of RC structures exposed to a physico-chemical environment, Proceedings of the Third Kerensky Conference, Singapore, 67-70.

Blaszczyriski T. 2002. Some effects of crude oil environment on RC structures, Foundation of Civil and Environmental Engineering, (2): 7-14.

43

Blaszczyriski T. & Scigallo 1 2006. Ultimate bearing capacity assessment of RC sections under mineral oil exposure, Archives of Civil and Mechanical Engineering: 41-56.

Kagimoto H., Sato M., Kawamura M. 2000. Evaluation of degree of ASR deterioration in concrete and analysis of pore solutions, Concrete Library International, (36): 480-493.

Kucharska L. 2001. Failures and damages of concrete structures and its development, Proceedings of XX Structural Failure Conference, Szczecin-Miedzyzdroje, 89-118, (in Polish).

Manns W & Hartmann E. 1977. Zum Einfluss von Mineralolen auf die Festigkeit von Beton, Schriften-reiche des DAfStb, Ernst, H. 289.

Matti M.A. 1976. Some properties and permeability of concrete in direct contact with crude oil. Ph.D. Thesis, University of Sheffield.

Onabolu O.A., Khoury G.A., Sullivan P.lE., Sterritt R. 1985. Inorganic acid contents of NS oil: effect of anaerobic bacterial activity in tanks, Petroleum Review, (1): 42-45.

44


Modelling the behaviour of a micro-tunnelling machine due to steering corrections

W. Broere

Delft University of Technology, Delft, The Netherlands A. Broere BY, Amsterdam, The Netherlands

1. Dijkstra & G. Arends

Delft University of Technology, Delft, The Netherlands

ABSTRACT: Micro-tunnelling often encounters restrictions expanding into new areas. In the Netherlands, and more general in soft soils, one of the problems is the controlled boring of curves. To better understand the behaviour of the tunnel boring machine in such conditions, an analytical model has been developed, that takes translation and rotation of the TBM into account. The model takes the sub grade reaction and the stiffness of the soil into account as major parameters describing the soil behaviour. It is shown that the sub grade reaction in the inner and outer curve of the TBM differs substantially and a sub grade reduction factor is introduced in the model to deal with this effect. A first derivation of the subgrade reduction factor was made comparing field measurements with model parameter variations. In this paper a full three-dimensional finite element simulation is presented, using large deformation analysis to model the movement of the TBM through the soil, which has been used to study the behaviour of the micro-tunnel machine in more detail and derive the sub grade reduction factor.

1 INTRODUCTION

In most of the Netherlands the upper layers consist of alluvial soil deposits, where the stiffness of the soil is low. The groundwater level is generally very high and locally reaches the ground surface. This combination has adverse implications for the drilling techniques used when construction tunnels using the micro-tunnelling technique. In such poor soil conditions, the control and steerability of the micro-tunnelling machine can become problematic (see e.g. Oreste et al., 2002). One aspect, the boring of curves, is described in more detail in this paper.

In the past several borings were executed in soft soils without any significant problems, or only with limited problems (Broere et al., 2007). Recently however, problems occurred on a project where a curve was introduced in the trajectory coinciding with the transition from very soft to stiffer soils. At this location the concrete pipe snapped when the tunnel boring machine (TBM) had just entered the stiffer soils. A second boring, with a greater curvature, was successfully completed at the same location. This event raised questions concerning the actual behaviour of the TBM, the concrete pipes and the coupling forces between the pipes. To increase understanding of the behaviour of the TBM in soft soils, an analytical model was formulated that describes the behaviour of the TBM during the boring of curves in soft soils.

In this model the behaviour of the soil is modelled as a sub grade reaction modulus. To correctly predict the TBM behaviour, a reduced sub grade reaction modulus is needed in the inner curve as compared to the outer curve. The necessary subgrade reduction factor has been derived first based on field measurements at the aforementioned project.

45

Translation

+ Rolation

- - .....

, .. , , , , , " " .----__".:..:._--,

.... ..... _ •• L..,_ .......

sa Total movement TBM

/ SI!!e~ing~ireC;tiO irBM

- _Af?' Direction ofnnovennelll n/(.ifP' . TBM

to

Oire!;UOIl 0 r the jack 10 rces

Figure 1. Movement of the TBM is split into a rotational and translational mode. These are combined to calculate the reaction w of the TBM to a steering action a.

In order to study the soil behaviour around the TBM in more detail and derive the sub grade reduction factor, independently from the field measurements, a finite element analysis has been used that models the movement of the TBM through the soil during a curved boring.

2 ANALYTICAL MODEL OF A MICRO-TUNNELLING MACHINE IN CURVES

In order to model the behaviour of a micro-tunnel machine during the boring of a curve in (very) soft soil, an analytical model has been developed by Broere et al. (2007). In this model the movement of the TBM through the soil is split in a translational mode and a rotational mode, as shown in Figure 1. For both modes the forces acting on the TBM are determined and the torque resulting from these forces is calculated. Supposing the soil is elastic, the forces can be superposed and the reaction of the machine can be determined from moment equilibrium. In soft soils often the reaction of the machine co is less than the intended steering action ex.

Of the forces acting on the machine, the jack forces, the normal forces between soil and shield and the face support are the most important. Both the jack forces and the face support are considered as given, what is left is to determine the normal forces in the soil. The linear-elastic soil behaviour is characterized by a subgrade reaction modulus k. For a given steering angle ex the normal force Fsoil and the reaction to can then be calculated. See Broere et al. (2007) for full details of the model derivation.

When the results of the model were compared to field measurements, it became clear that a different subgrade reaction modulus was needed at the inner and outer curves of the TBM. This is understandable as the TBM will excavate a slightly oval-shaped hole in order to make a curve. On one side the soil will not be excavated but rather displaced sideways, whereas on the other side an over-excavated zone exists where there is no direct contact between TBM and soil and some relaxation of the soil will occur. In this area the soil will have a lower subgrade reaction modulus, which is included in the model by introducing a sub grade reduction factor Cf for this area.

Based on field observations in soft soils and realistic input parameters, a possible range of Cf = 0.15 to 0.35 was determined. Cf = 0.25 was selected for further parameter studies of the

46

3

t2.5

-n

2+------------------------------------------

1.5

a=2°

~------

a=1°

0.5

a=0.5°

o

2000

4000

6000

8000 k

10000 12000 14000

k [kN/m3] -----7

Figure 2. Reaction of the TBM to as a function of the subgrade reaction modulus k and steering action a.

subgrade reaction modulus k and the magnitude of the steering angle a. The results are given in Figure 2, which shows that in stiff sand layers (k = 10,000kN/m3 or higher) the TBM reacts well to desired steering action. In softer (clay) layers (k =2000 - 4000kN/m3) the machine does not follow the desired steering angle completely, but can still be controlled. Only in even softer clay and peat layers the machine will hardly react to any steering actions taken.

As the subgrade reduction factor Cf = 0.25 was determined for soft to very soft soils and has a significant influence on the results, a more detailed determination of this value is warranted. A numerical simulation of a curved boring is therefore made to derive the subgrade reduction factor independently.

3 NUMERICAL SIMULATION

The curved boring of a micro-tunnel machine is simulated in the numerical framework FEAT, capable of a full Eulerian large strain analysis as well as traditional Lagrangian finite element calculations (TOCHNOG, 2007). In the Eulerian scheme the material flow and the mesh are decoupled. This allows for extremely large deformations to be modelled, whilst the calculations remain numerically stable. This approach has been used previously to model the jacked installation of piles (Dijkstra et al., 2007).

The TBM, with a front and back part of the machine, and the first three tunnel segments are included in the simulation.

In modelling the excavation and steering process, a two-step approach is taken. First, an Eulerian calculation is made in which a tunnel boring machine on a straight alignment is simulated. In this stage a forward movement of at least the length of the finite element mesh is simulated, to make sure that steady-state conditions are reached. This stage yields a proper stress and strain field around the TBM and tunnel and is used as the starting point for the next phase.

In the second phase the front part of the TBM is rotated with respect to the back part until a 1 degree rotation is reached. This simulates the actual steering of the TBM.

Additionally, a reference calculation is made in which the Eulerian inflow of the soil is skipped and only the initial conditions are set and the steering phase is simulated.

47

Figure 3. Mesh and boundary conditions for the numerical simulation.

- 20m ~"" 0.Q1 m/s

3.1 Details of the numerical schematisation

In order to create a finite element mesh that includes the proper boundary conditions, a two-step approach is used. First, an axi -symmetric mesh is created, consisting of first order quadrilateral elements. In this mesh, the tunnel and TBM are situated on the axis of symmetry. In order to prevent mesh problems, a small gap is maintained around the axis. All elements, the soil as well as the TBM and tunnel, are represented by volume elements. No plate elements of any kind are used.

In this mesh, the TBM is 6 meters long and 2.2 meters in diameter. In the middle, separating the front and back part of the machine, a slice of 10 em thickness is included that will be used to model the steering jacks. Behind the machine 15 meters of tunnel with an outer diameter of 2.1 meters and inner diameter of 1.8 meters is modelled. This approach results in a tail void of 5 em behind the TBM. The mesh is extended 10m from the axis of symmetry and 5 m in front of the TBM. This ensures boundary conditions are sufficiently far removed from the tunnel.

Subsequently, the mesh is rotated around the symmetry axis to create a fully three-dimensional mesh consisting of 36 10° wedges. The intentional gap on the axis is remeshed and merged with the full mesh at this stage. Following this approach, a numerically stable mesh as shown in Figure 3 is obtained.

As shown in Figure 3, the outer boundaries of the mesh are considered fixed. The same hold for the tail end of the last tunnel segment. At the right, in front of the tunnel face, soil flows into the mesh at a steady velocity of 0.01 mls. Soil leaves the mesh at the left boundary. In order to (somewhat crudely, but effectively) model the excavation process, soil is also removed from the mesh at the tunnel face, such that soil that would have flowed into the tunnel is removed from the calculation. Finally, in order to keep the calculation stable, a distributed force of 50 kN/m2 is used on the boundaries where soil flows out of the mesh.

For the discretisation in time Euler backward time stepping is used, because of its high numerical stability. For the Eulerian formulation, where mesh and material state are decoupled, the convective terms (also known as state parameters) need to be transferred through the mesh. This is done by a Streamline Upwind Galerkin method. In this method material and state parameters are calculated in the nodes.

48

D_lining = 2.1 m

Figure 4. Close-up of the boundary conditions around the TBM.

Table 1. Material parameters for the soil.
Name Symbol Value Unit
Dry volumetric weight y 17 kN/m3
Cohesion c 1 kPa
Friction angle ¢ 35 a
Dilatation angle 1/1 0 a
Young's modulus E 30 MN/m2
Poisson's ratio v 0.3 After 30 m of material flow (i.e. 30 meter of forward movement of the TBM and tunnel has been simulated), the steering action is simulated. In this stage the material flow is stopped. The steering action of the TBM is simulated by the thin slice separating the front and back of the machine, similar to the steering jacks in an articulated shield (see Figure 4). An asymmetric velocity field is applied to a single wedge of this slice, extending this wedge by 10mm at a rate of 0.1 mm/s. Given the small deformations and in order to improve accuracy, the integration scheme is switched to a normal Lagrangian scheme. Therefore, the material and state parameters are calculated in the integration points and need to be interpolated from the nodes before the start of this calculation phase.

Initial stress conditions are set to the horizontal effective stress representative for a 10m deep tunnel, although these will be completely replaced at the end of the first calculation phase and before the steering action is modelled. The influence of pore water is not modelled in the current calculation.

In the reference calculation, the Eulerian phase is skipped. The same boundary conditions are used, except for the equilibrium force on the outflow boundaries. These are replaced by a fixed displacement boundary condition, which are sufficiently far away from the area of interest as not to influence the result.

3.2 Constitutive model

The soil is modelled using the Mohr-Coulomb model, with parameters listed in Table 1. A tension cut-off was used to prevent tensile stresses in the soil. Also, to prevent problems with large volumetric strains occuring during large deformations, dilatant behaviour of the soil was prevented, setting 1/1 = o.

49

( [ t.

-

"t.~)!IOO

Ccll'Ilaur Fro otlllmen !'IrIl!U~ ...wI.

Figure 5. Vertical total stress at the end of the Eulerian inflow phase.

Both the TBM and the tunnel lining are modelled as a linear elastic material with Poisson's ratio v of 0.2 and Young's moduli E of 23 GPa for the concrete lining and 200 GPa for the steel of the TBM.

4 NUMERICAL MODEL RESULTS

4.1 First (Eulerian) inflow phase

The vertical total stress at the end of the first phase, i.e. after 30 meters of forward movement of the TBM is given in Figure 5. Figure 6 shows the plastic shear strains. In these figures, a cross-section through the soil is made, although the full tunnel is plotted, in order to highlight the results at the interface between the soil and the tunnel.

A large stress increase is seen in the soil directly in front of the tunnel face, which peaks at the front thick edges of the TBM. This increase is due to the local boundary conditions and should not be considered a correct representation of the excavation process. However, compared to a closed face where all soil would flow around the TBM it is a significant improvement of the soil behaviour.

It can also be observed in Figure 5 that stress increases as the soil flows around the TBM and decreases again at the tail end of the TBM. This is due to the distinct tail void modelled in the analysis, which causes a partial unload of the soil as it flows into the tail void.

Similar behaviour can be observed in the shear strain plot (Figure 6). Large strains occur where the soil is forced around the TBM and the friction between the TBM and the soil causes further shear strain to develop. Once the soil is free to expand into the tail void and relax the magnitude of the strain decreases. Again, the tail void dominates the observed behaviour of the soil around the TBM.

4.2 Steering action phase

In Figure 7 the normal component of the total stress acting on the tunnel, at the end of the steering action, is shown. At this point the front of the TBM is rotated by 10 compared to the back side, with

50

{

(

,.'

,

rJj~~O(ICI

00tI~. F.lIt ClfM.llet'Il"'~ P"1~ .. '1IPl-

Figure 6. Plastic shear strains at the end of the Eulerian inflow phase.

T1ep '~I)(I.

CCtclOu:r FII ar 1I1:1Stn .1tI'U~ '5iI~

Figure 7.

-

Normal stress on the tunnel at the end of the steering phase (for the calculation with a first Eulerian inflow phase).

51

I

-

I

"

~IIXI;

COiMaut Fill C(jriglt ~ •• I~

Figure 8. Normal stress on the tunnel at the end of the steering phase (for the reference calculation without inflow phase).

Table 2. Normal stresses acting on the TBM at the end of the steering action and derived local subgrade
reduction factors. (Coordinates refer to distance from steering jack or TBM articulation point.)
Normal stress (kPa) Normal stress (kPa) Local subgrade
Outer curve Inner curve reduction factor
Full Full Full
Coordinate (m) Reference Analysis Reference Analysis Reference Analysis
+1.6 -91 -790 -23 -583 0.26 0.75
+1.1 -189 -657 -21 -779 0.11 1.19
+0.6 -311 -930 -32 -1100 0.10 1.18
-0.6 -289 -1940 -135 -1068 0.48 0.55
-1.1 -54 -2789 -64 -1097 1.19 0.39
-1.6 -287 -2080 -16.5 -1061 0.06 0.51 the jack extension (or outer curve) modelled at the lower side of the TBM in this figure and the resulting curvature upwards. Figure 8 shows results for the reference calculation where phase one was skipped and no soil inflow was modelled. In both cases the tunnel and TBM are not shown for improved clarity of the stress scale. When the reference case (without inflow of soil) is compared to the complete calculation some differences can be observed. In the reference case two distinct zones with a stress increase below the tunnel can be discerned, which act as support zones to carry the reaction of the rotating TBM. In the full model only a single support zone has developed. Clearly, the proper initial stress state calculated from the first inflow phase has a significant influence on the failure mode.

52

In order to derive the impact on the sub grade reaction modulus for the analytical model, the normal stresses acting on the TBM on the outer curve are compared to those on the inner curve. Some key numbers are given in Table 2, where the coordinate refers to the distance from the jack position or articulation point.

The subgrade reduction factor is clearly affected by the stress history, given the differences between the reference analysis and the full analysis. Strong local fluctuations in the local subgrade reduction factor occur, but overall the results from the reference analysis are lower than the range derived for the analytical model from field observations, whereas the results for the full analysis are somewhat higher.

5 CONCLUSIONS

During the boring of curves in very soft soils the steerability of the TBM may become problematic or even completely impossible. An analytical model that can determine the reaction of the TBM based on a given steering action has been developed. This model needs a sub grade reaction modulus on both the inner and outer curve side of the TBM to yield proper results. The subgrade reaction modulus on the inner side is reduced with respect to the value on the outer curve.

The local subgrade reduction factors derived from the full numerical analysis indicate that a single value for the subgrade reduction factor may not be appropriate. It would be better to use separate factors for the front and back part of the TBM.

The conclusion from the earlier paper by Broere et al. (2007), that the analytical model needs improvements, still stands, but the presented numerical approach alone is not sufficient to reach the required level of accuracy and thereby verify the analytical model in a wide range of soil conditions.

In order to reach that goal the numerical model needs several improvements. These include a further refinement of the mesh, a more detailed modelling of the soil excavation at the tunnel face and a continued inflow of soil during the steering action of the TBM. Although these are all possible, they would increase the computational effort beyond that of a common desktop PC.

An alternative approach would be to slightly adapt the analytical model to allow for a variable reduction factor along the TBM, with different values for the front and back part of the TBM on both the inner and outer curve side.

REFERENCES

Broere, W, Faassen, T.E, Arends, G. & van Tol, A.E 2007. Modelling the boring of curves in (very) soft soils during microtunnelling. Tunnelling and Underground Space Technology, 22(5-6), pp. 600--609.

Dijkstra J., Broere, W & van Tol., A.E 2007. Numerical simulation of the installation of a displacement pile in sand. In Numerical Models in Geomechanics, Taylor & Francis, pp. 461-466.

Finite Element Application Technology. TOCHNOG Professional User's Manual Version 4.2. (http://www.feat.nll)

Oreste, P.P., Peila, 0., Marchionni, V. & Sterling, R. 2002. Analysis of the Problems Connected to the Sinking of Micro-TBMs in Difficult Ground. Tunnelling and Underground Space Technology, 16 Suppl. 1, pp. 33-45.

53


Trenchless replacement of gas and potable water pipes with new PA 12 pipes applying the pipe bursting method

Reinhard Buessing

SB Projektentwicklung, Herdecke, Germany

Andreas Dowe & Christian Baron Evonik Degussa GmbH, Marl, Germany

Meinolf Rameil

Tracto- Technik GmbH & CoKG, Lennestadt, Germany

ABSTRACT: Pipe Bursting offers a real technical and cost -efficient alternative to the open trench method; it is also very effective when compared to other alternative pipe rehabilitation methods. Pipes made from Polyamide 12 (PA 12) extend the area of application for trenchless technologies, pipe bursting in particular. Polyamide 12 pipes showed that problems like point loads and crack growth resistance can be solved. A field test in Dortmund showed that the handling of PA 12 pipes is as easy as the pipe cracking with PE.

1 INTRODUCTION

Pipe replacement by open trenching can involve traffic impairment, noise- and emission pollution when breaking open the surface. Also there is a risk of damaging existing underground pipe, soil and groundwater intervention and higher storage, transport and soil removal costs. These drawbacks can be almost completely avoided by replacing pipes using trenchless technologies. According to estimates of Advantica, the pipe bursting method has become the most widely applied trenchless pipe replacement method with more than 50,000 km worldwide. With the pipe bursting method, the defective or under dimensioned pipe is cracked with a burst head and then displaced into the surrounding soil. This creates space for the new pipe directly proceeding, of same or greater diameter. The pipe bursting technique allows the replacement of defective pipelines in the same path without any substantial influences on soil and groundwater. Open trenching which requires the breaking up and repairing of valuable surfaces is thereby almost entirely eliminated. In addition, this eco- friendly technique helps to cut down costs considerably. The pipe bursting distances usually have a length of approx. 100--200 m.

In general, there are two methods to choose from:

1. hydraulically operated static pipe bursting with ladder shaped Quicklock rods

2. pneumatically operated dynamic pipe cracking with piercing tools or rammers

The proportion of pipes that need to be rehabilitated or renewed will increase. The current situation in Germany: The length of the transportation network of water pipes is approximately 460.000 km (Wagner, v.: Sanierung von Wasserdruckleitungen - Eignungspriifung fur Druckliner, WWT Heft 9/2003) and 270.000 km for the gas supply lines (Bundesverband der Deutschen Gas- und Wasserwirtschaft e.v. (BGW): Jahrbuch Gas- und Wasser 2004, Oldenburg Industrieverlag GmbH, Munchen, 2004). There are as we11160.000km of house-connections for water and 80.000 km for gas.

55

In order to continuously ensure a fully functional mains and distribution network, an annual rehabilitation rate of at least 2% should be achieved. This rate of renewal would correspond to a service pipe life of 50 years. That is why the network-operators are looking for innovative methods which offer increased productivity combined with increased economical savings. The Pipe Bursting technology for trenchless pipe renewals is just such a method.

1.1 Pipe materials for transportation network of water and natural gas in Germany

Grey cast iron was used with lead joint sockets until the end of the 1940's and with screwed and bolted flange joints up until the mid-60's. There are still many of these fragile grey cast iron pipes which are likely to leak and break and these now have to be renewed as a matter of priority.

Ductile iron without sufficient coated protection was installed from the mid-60's to the early-70's. Steel pipes were installed with jute fibre/tar coating and joints until the mid 40's, then with a more simple bitumen coating until the end of the 50's, and after that with a double protective layer of bitumen but often the pipes were not laid on a granular bedding until the end of the 70's and since then with PE-coatings and with/or without protective surface coatings.

Asbestos cement- and lead water pipes need to be replaced just as soon as possible because they are considered harmful to peoples' health.

Plastic pipes (PVC) with bonded sockets are leaking and PE pipes are often not of sufficient carrying capacity and therefore need to be upsized and renewed simultaneously.

1.2 Choosing the proper method

The most important question for the network operators is: do pipes have to be repaired, rehabilitated or renewed/replaced? To repair the pipes is feasible if the repair measure ensures a reliable and lasting result. Another option is pipe rehabilitation, but rehabilitation methods sometimes do not meet the required standards in terms of pipe carrying capacity, flow volumes and pipe life durability. Problems also occur if an annular gap is left between the new replacement and the host pipe. Also because of additional pipe cleaning and the need for encrustation removal, the overall costs may be comparatively high. The third opportunity is trenchless renewal.

A pipe replacement is necessary:

- when repair or renovation is technically or economically inappropriate

- when the hydraulic capacity needs to be improved by a greater pipe diameter

- when repair or renovation offers only a short term solution with a pipe replacement being

inevitable

- when there is request for a long lasting pipe durability and or a higher product life span

- when the static loading capacity of the defective pipe would be otherwise negatively affected

The renewal of pipes with traditional open trenches is consistent with traffic jams, noise, dust and not liked ecological effects as well as risks of damage to existing buried pipes cables and telecommunication lines, disruption of the groundwater flow and the need to remove previously undisturbed soil, storage and transport of the excavated soil and its associated high disposal and tipping costs. All this can be avoided to a large extent by using a trenchless method.

With the Pipe Cracking method the effected or under capacity pipe is fragmented by the profile bursting blades and the broken pieces are displaced by an expander into the surrounding soil. Thus a sufficient channel space is preformed for the attached new pipe with the same or even bigger diameter that follows the blade/expander. The Pipe Cracking method permits renewals of damaged pipes by using the existing pipeline course without intrusion into the soil or disturbing the groundwater. The work method does not cause surface subsidence damages and by using a filler for the annular gap between the new and the burst pipe the bedding can be improved. As opposed to the open trench method where the open trenching and reinstatement of the road and pavement surfaces are largely unnecessary and thus pipe cracking is environmentally friendly and in most cases more economical.

56

1.3 The development of the Pipe Bursting technology to date

Pipe Bursting has a long history and could not have developed as successfully as it did without soil displacement hammers and ramming technology. The idea of the Pipe Bursting technology, using the existing, damaged pipe in order to accommodate a new pipe has its origins at British Gas. As early as the 1980's they commonly used adapted soil displacement hammers and pneumatic pipe ramming machines to trenchlessly lay pipes and cables. In Great Britain a contractor named D J Ryan & Sons had already taken out the first patents (Fraser: The Pipe Bursting Options, Conference documents "Asian Trenchless Tech 1994")

Even in the first years of the Pipe Bursting technology pneumatically and hydraulically propelled systems existed. The pneumatic machines were known as PIM-machines (Pipe Insertion Method). The first hydraulic machines were systems where the expander radially extended using a hydraulic force and thus shearing apart the pipe and the fragmented parts displaced into the soil around it. In Germany the machines were known, amongst others, by the name of "KM-Berstlining". This expanding technique though did not become generally accepted in Germany. First users in Germany were companies like "Kanal-Mi.iller" (KM-Berstlining), DIGA (PIM, Grundocrack) and Brochier (Grundocrack). In the USA the method known as "cracking" was introduced successfully more than 20 years ago. Ever since then it has been used very successfully on a daily basis.

In the UK the first applications of the Pipe Bursting method were limited to grey cast iron gas pipes. The method rapidly spread allover Europe already including operations in the water and sewage sectors. In those times Polyethylene pipes were already used as new product carrying pipes. Mostly in the early years plastic protection tubes were installed first because PE pipes with protective coating have been not available. Nowadays the Pipe Bursting method is mainly used to replace pressure pipes (gas and water networks ND 80 to 1000) and for sewage pipes (ND 150 to 1000). In the USA even sewers up to 1200 mm in diameter have been replaced.

In the mid-90s the Dynamic (Pneumatic) Pipe Cracking was complemented by a new method - the so-called Static (Hydraulic) Pipe Bursting. With static bursting ladder shaped rods without screwed threads were introduced. These new rods could easily be connected and disconnected and have had many practical advantages. The functionality of screw threads can easily be impaired by dirt and therefore have to be kept clean at all times. Often the screw threads are damaged on the jobsite and thus slow down the connection and thus the pushing and pulling process. In addition to that the tensile strength and pulling forces may be impaired. In contrast the QuickLock system provides an absolutely safe and reliable connection - for the tractive pushing and pulling forces. The new QuickLock rod joints are also flexible and can follow the radius of the existing curved offset pipe bends. Time efficiency is another key advantage - the QuickLock rods can be driven into the host pipe about 40% faster than conventional screwed rods because the ladder shaped design allows a fast and easy rod string assembly. In contrast to the screwed rods where friction drive clamp jaws transmit their force on the rods from outside, the QuickLock rods have positive locking brake and drive fingers that prevents any slippage of the rods when under traction load. Also the tiresome labour intensive screwing and unscrewing process is no longer necessary. A hydraulically operated rig pushes or pulls the rods into the host pipe. Once they appear in the arrival pit, an up sizing cone is connected and the new pipe then attached to it. By pulling back the rods the bursting and installation process begins. A traction bursting force of up to 2.500 kN (250 metric tons) is available.

1.4 Main options of the Pipe Bursting

Pipe Bursting is the trenchless replacement of pipelines by using the host pipe's line and level. With a dynamic or hydraulic bursting device the host pipe will be broken, fragmented and then displaced into the surrounding soil. Protection tubes can also be installed with this method. During this work process the new pipe with the same or even larger diameter is installed. For gas- and sewer pipes most of the time PE pipes are used. They are flexible and able to adjust well to the host pipeline course. Smooth outer but welded pipe joints guarantee an easy entry into the new burst line. Prior to the beginning of the burst process the host pipes have to be disconnected and the house services

57

Figure 1. Dynamic pipe cracking method.

Figure 2. Dynamic cracking hammer GRUNDOCRACK with bladed pipe cracking head and attachment for winch rope and expander at the rear end for pressure pipes.

have to be exposed at their connection points on the main. As Pipe Bursting is a really fast pipe renewal method some network operators do not install a temporary bypass fluid or gas supply.

1.5 The dynamic pipe cracking method

After the constant tension winch is set up at the pit and the winch boom has been anchored, the cable is pulled through the host pipe until it arrives at the launch pit. At the same time, the cracking hammer is prepared and the new pipe, usually a PE or PA 12 pipe, is attached to it. This means, firstly single pipe sections are butt-welded together and attached to the rear end of the cracking hammer by welding to an adapter. The bladed pipe cracking head is attached to the front nose of the cracking hammer which itself is connected to the winch cable.

The bursting machine together with the attached PA 12 or PE pipe is pulled into the pit by means of the cable winch until it comes up against the host pipe to be renewed. Then, by starting the compressor and the impact cracking machine via its control unit the pipe cracking process starts. The forward motion and the directional stability are maintained by the traction force of the cable winch.

The pipe cracking hammer breaks the host pipe by means of its bladed head and radially displaces the fragments of the pipe into the surrounding soil. At the same time it expands the soil to accept the new pipe with an equal or bigger diameter. Once the bursting machine and the expander cone reach the arrival pit they can be removed.

1.6 The Static Pipe Bursting method

The Static Pipe Bursting obtains its required force for the bursting, displacing and installation process hydraulically and transmits it via the ladder shaped QuickLock rods, which are absolutely safe and reliable for the thrust and traction forces. First the hydraulically powered rig has to be braced inside the arrival pit. After that the burst rods with their precursory guide-rod are pushed into the host pipe. In the starting pit the guide-rod has to be exchanged by the burst tools (burst head and expander with swivel, roller-blade, etc). The new PA 12 or PE pipe is affixed to a pipe pulling adapter and attached to the burst tools. By pulling back the burst rods towards the arriving pit the

58

Figure 3. Static pipe bursting method.

host pipe will be broken and its fragments displaced by the expander into the soil right behind the burst head/expander. The bore for the new pipe can be the same or a larger diameter.

2 LIMITATIONS

There are only few limitations (Rameil, M.: Handbook of Pipe-Bursting Practise, Vulkan Verlag Essen 2007) for the use of the Pipe Bursting method:

- at this time, the application range is limited to circular existing pipes,

- if necessary, the host pipe has to be taken out of service during the bursting process (not necessary

e.g. for mains, drainage pipe replacement, etc.),

- the course of the host pipe must be usable for the new pipeline (e.g. inclination). Heavily encrusted pipes must be cleaned so that bursting rods can be pushed through (static bursting) or respectively the winch rope can be pulled in when (dynamic/pneumatic pipe cracking),

- the soil surrounding the host pipe must be displaceable,

- house connections have to be installed using pits. This, however, guarantees a professional and

safe integration,

- sharp bends, flanged joints of steel and ductile iron pipes, etc., depending on selection of bursting tools require intermediate pits,

- pipe slumps (sags) cannot be removed however may be reduced,

- a minimum distance has to be kept away from existing parallel or crossing pipes as well as an

adequate cover depth.

However, the application of the pipe bursting method also offers all advantages of a modem, trenchless installation method: Pipe bursting is the installation of new, industrially produced pipes, which may be compromised when installed by open cut methods,

- pipe bursting gives a considerable reduction in excavation and road works (almost no traffic disturbance, no annoyance of the public or noise and dust pollution, reduced construction time and the reduction of indirect costs),

- high daily output up to 150 m gives a cost-effective replacement and considerable cost-savings compared to open cut,

- almost any pipe material available for trenchless installation methods can be installed by pipe bursting, e.g. plastic (PE, PA 12), ductile iron (DIP), steel, glass reinforced plastic (GRP) and even vitrified clay (VCP) and polymer-concrete pipes (PCP),

- pipe bursting allows the replacement of almost any host pipe material (some with limitations),

- pipe bursting can be applied for any kind of pipe damage as long as the bursting rods can be

pushed in (static bursting) or the winch rope can be pulled in (dynamic/pneumatic pipe cracking), - no reduction of pipe diameter, up-size of pipe diameter is also possible,

59

- preparation of the host pipe, like high-pressure cleaning, removal of debris and blockages are

not necessary (but may possibly be necessary for specific reasons),

- considerably less danger of unintended ground settlement compared to open cut,

- applicable for pipes in sloped areas and areas with trees, shrubbery and the like,

- pipe bursting can also be used for the replacement of laterals,

- pipe bursting is controlled and described by worldwide standards, norms and regulations.

3 MATERIALS FOR THE NEW PIPES

New pipes made of HD PE with a coating (e.g. SLM 2.0) are particularly suitable. They are more impact-resistant as the PE without coating, sufficiently flexible and align well to the course of the host pipeline. This material is mainly used for gas networks from 0.5 to 4 bars. Evonik Degussa GmbH knew from experiences in the gas pipe project about the excellent characteristics of Polyamide 12 (PA 12). After a lot of testing they decided to make a pipe cracking. After a lot of testing they decided to make a pipe bursting test with this material. All results turned out satisfactory. Now the first PA 12 pipes especially for the use in water networks are produced and will be installed soon. During this conference it may be possible to present the first results. Some details about the outstanding results of PA 12 material tests.

3.1 Crack growth resistance of PA 12

During the feasibility study of PA 12 for gas applications several tests were observed in ace with international standards. Considering the sensitivity of former PE materials to slow crack growth (SCG) and its influence on the installation technologies Evonik paid much attention on this issue. Because of the high resistance to stress cracking PA 12 is used for air brake tubing systems since decades. Therefore it could be assumed that SCG would not be an issue for PA 12. This is also indicated by the long-term hydrostatic pressure test in ace with IS09080. Whereas PE is known to have a second branch at the time to failure curves this is not investigated by tests on PA 12 pipes. Even after 16,000 test hours at 80°C no stress cracking initiated failures are detected.

Additionally the SCG behaviour of PA 12 was investigated with the Notch Pipe Test in ace with ISO 13479 and ISO 22621-1. Also the PENT test in ace with ASTM F1743 was investigated. The results are shown in Table 1. At 80°C and 20 bar no failures occur at the Pipe Notch Test after 2000 h. Even after an increase of the notch depth from 20 to 30% of the wall thickness no failures occur. The minimum requirement in ace. with ISO 22621-1 is 500 h without failure with a notch depth of 20%.

The PENT test is very well established in the US gas society and PE materials are tested at 80°C and 2,4 MPa stress. Presently, the requirements within ASTM D2513 for PE materials require PENT time to failure of 100 hours. However, no such requirements are in place for Polyamide materials. Two replicates of the PA 12 material were tested in accordance to ASTM F1743 requirements at an increased stress of 4.8 MPa. The results of the testing indicated that there were no failures with any of the specimens after 1000 hours. The testing was discontinued after 1000 hours.

Table 1. Results on slow crack growth resistance of VEST AMID LX9030.

Test

Standard

Requirement

LX9030pipe

Pipe Notch Test 20% notch depth 30% notch depth PENT

ISO 13479 & 22621-2

>500h

>2000h >2000h >1000h

ASTMF1743

>100h

60

3.2 Effects of secondary stresses on PA 12 pipes

In addition to characterizing the SCG performance characteristics and influence of surface scratches, additional tests at GTI were performed to characterize the influence of secondary stresses. The main motivating factor for performing these tests was the fact that under actual field conditions, the piping systems are subjected to the combined effects of both internal pressure and other secondary stresses including rock impingement, earth loading and bending. Often, these secondary stresses, not internal pressure, are the root cause of many in-service field failures. Additionally the point load resistance is highly discussed relating to the various trenchless technologies and sand-bed free installation of thermoplastic pipes.

Therefore, comprehensive long term sustained pressure tests were performed at elevated temperatures to characterize the effects of various types of secondary stresses. It is important to note that these tests are not a part of either the ASTM or ISO standard. The test methodology is an extension of previous research performed by Dr. Charles Bargraw- DuPont and further refined by Dr. Michael Mamoun - Gas Technology Institute to study the performance characteristics of older generation PE materials. For the case of the rock impingement, the intent is to evaluate the performance of pipe materials subjected to indentations by a 13 mm rock. For the case of the earth loading, the typical safe deflection limit that is specified is 5%. For the case of the bending strain, the typical bend radius limits for a pipe specimen without any joints or appurtenances is 20 times the outside diameter.

Six (6) 2-inch SDR-Il LX9030 pipe specimens were placed in appropriate test rigs to simulate the effects of rock impingement, earth loading, and bending strain. The entire test assembly was placed under long term sustained pressure testing at 20 bar at 80°C. For all of the various types of secondary stresses which were evaluated, the results of the testing demonstrated that there were no failures after 1000 hours of testing, as presented in Table 2 below.

In summary the test results show that PA 12 has a unique resistance to SCG and secondary stresses including point loading. Therefore it could be assumed that PA 12 pipes can be installed without sand-bedding and with trenchless technologies even without additional protection layers.

At a usual gas pipe bursting site in Dortmund we made a test how the PA 12 material will resist compared to PE. This test showed that for welding the same equipment, with different welding parameters can be used. The employees have been able to weld PA 12 without any problems. The PA 12 pipe has been connected with the PE pipe through an inside connection because it is not possible to weld them together. The bursting process has been done like usual. Not a single problem has been recorded. PA 12 is as easy to use as PE. At the last pit the PA 12 pipe and a sample of PE has been taken out for the microscopic analysis in the laboratory.

3.3 Results of the microscopic analysis

The pictures show very well the difference in the damage of the surface of both materials.

Table 2. Test results of VEST AMID LX9030 pipes investigating secondary stresses Field test in Dortmund.

Secondary Stress

Test Criterion

Results

Rock Impingement

13 mm Indentation

Test Pressure = 20 bar Test Temperature = 80°C

5% Deflection of Outside Diameter Test Pressure = 20 bar

Test Temperature = 80°C

20 times OD

Test Pressure = 20 bar Test Temperature = 80°C

Test Time> 1000 hour with no failures

Earth Loading


Bending Strain


61

Beside the above reported excellent resistance to SCG and point loads the PA 12 shows also a high scratch resistance. Summarizing these performances it can be assumed that Evonik Degussa found a promising solution for a pipe bursting material which can be used in pipes with high pressure.

3.4 Bursting tools for pressure pipes

When static pipe bursting was introduced in the middle of the 1990s, it became possible to replace ductile host pipe materials like steel and ductile iron by splitting. Up until then, it had only been possible to burst brittle host pipe materials applying dynamic/pneumatic pipe bursting. Steel pipe bursting was limited and ductile iron pipes could not be burst at all.

Meanwhile, different splitting and bursting tools were made available to select from, depending on the material of the host pipe to be replaced. Cast iron pipes, but also other brittle pipe materials (fibre or asbestos cement) can easily be burst into fragments by unbladed bursting heads and then displaced. Here, the tapered angle of the bursting head which induces powerful radial loads into the host pipe has a great influence on the destructibility of the host pipe. The fragments generated should be as small as possible to achieve an even load distribution covering the complete circumference of the new pipe, thus preventing twisting and misalignment of the old pipe.

Differing from brittle host pipe materials are steel or plastic as well as ductile iron pipes which require splitting techniques that cut the host pipes and expand them open while also requiring

Figure 4. PA 12 and PE pipes connected by an mechanical joint.

Figure 5. Typical pit on the pipe bursting site in Dortmund.

Figure 6. The PA 12 and PE pipes connected with the pipe bursting head.

Figure 7. The site is ready to start the pipe bursting process.

62

specialist tooling to break joints and pipe repair clamps. Particularly, within the area of joints and repair clamps (e.g. steel pipes with bell-and-spigotjoints or not welded steel pipes or ductile iron flanges and fittings), the capacity limits of the pulling rig are often reached before they can be cut when using conventional splitting techniques. To split steel pipes, roller blade cutters (fig. 12) are used, with a splitting line of roller wheels that attack the host pipe and joints mainly in the lower third of their circumference. The roller blades are arranged to achieve optimal splitting geometry. This is the only way to minimise the required splitting and pulling forces. It is of the greatest importance that the split edges of the host pipe are bent outwards to keep them from damaging the new pipe.

Meanwhile, in order to split ductile iron pipes, so-called "roller blade trailers" are used. The roller blades are not arranged in a single holder with minimal space between them, but instead every single roller blade is arranged in its separate roller blade holder. Usually, the space between the single roller blade holder has the length of a rod between. The diameter of the roller blades fitted is steadily increased. This arrangement leads to an even distribution of the splitting forces. When the blade holders run through the host pipe, the first profiled roller blade perforates the pipe so the following smooth roller blades can evenly split it.

In order to cut PE pipes or defective inliners, special hook shape splitters with 2 or 4 blades are used. These special splitters prevent the host pipe from buckling and slide easily through the host pipe when splitting.

Figure 8. The pipe bursting head is cutting the existing cast iron pipe.

Figure 9. The burst host pipe has very sharp edges.

Figure 10. Intermediate pit for the house connection. Also here the sharp edges of the host pipe can be seen.

Figure 11. The PA 12 and PE pipe pieces for the microscopic analysis.

63

PE pipe

PA J2 pipe

Figure 12. New Roller Blade to cut 1,000 mm steel pipes.

4 CONCLUSION

Pipe bursting with its different variations and decades of applied experience is a comprehensive trenchless renewal option for a large proportion of the worlds damaged fluid, gas supply and sewer pipelines. This amounts to enormous potential for economical savings when carrying out the inevitable pipe rehabilitation programs. Especially, in times of critical financial situations, the cities and municipalities can hardly afford to simply give away this potential as a result of technological

64

conservatism. All in all, we can surely look to pipe bursting to be the preferred technique as more than 50,000 kilometres of supply and sewer pipelines worldwide have already been "cracked", "burst", "split" and replaced with new pipes, particularly, since with the new PA 12 material also gas and water pipes operated under higher pressure can now be installed. As a result in many countries a steadily increasing number of contractors are now successfully using this technology. In combination with the PA 12 pipes, which are resistant to issues like point loads and crack growth resistance gives pipe bursting the chance to enter the market in higher pressure pipes f.e. in water distribution networks.

65


Experiences with Polyamide 12 gas pipes after 2 years in operation at 24 bar and new possibilities for HDD

A. Dowe & C. Baron

Evonik Degussa GmbH, Marl, Germany

W. Wessing

E.ON RuhrgasAG, Essen, Germany

R. Buessing

SB Projektentwicklung, Herdecke, Germany

M. Rameil

Tracto-Technik GmbH & CoKG, Lennestadt, Germany

ABSTRACT: For more than 20 years PA 12 is evaluated as a pipe material for gas distribution and for more than 10 years PA 12 pipes have been used for low pressure installations. In recent years PA 12 is being investigated for operation pressures above 10 bar. Since 2002 ISOrrC 138 is elaborating standards for PA 12 gas installation systems with operation pressures up to 20 bar. In 2007 the first three parts of the ISO 22621 were published. Evonik Degussa GmbH, Marl, Germany, as one of the four manufacturers of PA 12 and E.ON Rubrgas are cooperating on a test installation of a 60 m pipe system (110 mm SDR-ll) in the technical center ofE.ON Rubrgas in Dorsten, Germany. The system with various connections was installed in October 2005. After 2 years operating at a pressure of 24 bar first samples were taken and investigated. The authors will give a report of the test installation and the experiences after 2 years. The paper will include also an introduction of the installation technology "Horizontal Directional Drilling" (HDD). In combination with the exceptional resistance of Polyamide 12 pipes against point loading and slow crack growth HDD offers new possibilities.

1 PA 12 - A HIGH PERFORMANCE THERMOPLASTIC MATERIAL

Due to the specific carbon-amide group in the polymer chain polyamides (PA) have strong intermolecular actions which induce high mechanical strength, high melting temperatures and chemical stability. The "long chain" PA 12 with 11 carbon atoms between carbon-amide groups has the lowest water absorption of all commercial available PA and represents the best compromise in thermal and mechanical properties. That is why PA 12 is the material of choice for some challenging applications relied on already for decades, e. g. in the automotive industry for fuel lines of passenger cars or for airbrake tubings of trucks.

Compared with medium and high density polyethylene in use for low pressure gas supply, "long chain" polyamides like PA 12 provide "naturally" superior performance due to their described chemical structure (table 1). Besides PA 12 only PA 11 is commercially available with almost identical properties. PA 11 supplied by one manufacturer is based on a planted feed stock, castor bean, while PA 12 is synthesized from butadiene, a crude oil by-product, in a complicated multi-step process. Evonik Degussa is the only company of the four PA 12 suppliers who is fully backintegrated to butadiene.

67

Table 1. Basic properties of PA 12 VS. PE.
Property Unit PA 12 PE 100 PE80
Melting temperature °C 178 130 126
Tensile strength at yield MPa 45 20-23 17-19
Tensile elongation at break % 200 >800 >800
Flexural Modulus MPa >1200 950 700
Charpy impact strength kJ/m2 No break 30 20
Hardness, Shore D 74 63 58
Permeability (23°C, mm3/bar/day)
Methane <0.005 0.24
Hydrogen <0.01 0.7
(data for PA 12 pending) Table 2. Maximal allowable operation pressures for PA 12 and PE based gas pipes with considering a safety factor of 2 and a SDR-l1.

MOP in bar

PE80 PE 100 PA 12

8 10 18

4 7 12

2 NEW VESTAMID GRADE FROM EVONIK DEGUSSA, PA 12 DESIGNED FOR HIGH PRESSURE GAS DISTRIBUTION

Controlling molecular weight and intermolecular interacting forces is an Evonik Degussa core competence to design PA 12 extrusion grades with optimized processing and performance properties. Evonik PA 12 is sold under the registered trade mark VESTAMID™ L. When Evonik started the activities for large pipe applications new high molecular PA 12 grades for extruding pipes with bigger sizes was designed.

For various tests and for test installations pipes with dimensions up to 300 mm have been manufactured so far using these new grades at several different pipe extrusion companies without any problems.

2.1 VESTAMID LX9030 - a material of choice for high pressure gas supply

There are two major design criteria for thermoplastic materials used in gas supply: the Maximum Required Strength (MRS) and Rapid Crack Propagation (RCP). Due to the intermolecular forces in PA 12 the hydrostatic strength of PA 12 pipes is much higher than for PE-HD, even for the optimized grade PE 100. Long term hydrostatic strength investigations according to ISO or ASTM standards have proven that in a 50 years extrapolation PA 12 has got a MRS of 18 MPa. Therefore PA12 SDR-ll systems are able to operate at pressures up to 18 bar using a safety factor of 2 for gas (table 2). In contrast to PE PA 12 is naturally resistant against stress cracking in general and slow crack growth and passes easily all the relevant tests originally created for this weakness of PE.

For the time being resistance to RCP at low temperatures is a matter of concern and discussion in the gas utilities. In the ASTM territory RCP is under consideration. RCP data have to be provided

68

but they are not limiting the listings for pipe materials from the Plastic Pipe Institute for use in the USA. In the European standards and also adopted in the ISO standard the maximum operation pressure (MOP) for a pipe material is limited from two sides, the hydrostatic strength of the material and the MOP derived from a RCP test at O°C. Due to that a test institute performed a RCP full scale test on 1l0mm and 6-inch SDR11 pipes ofVestamid LX9030 according to ISO 13478. The initiated crack arrested up to 30 bar respectively 25 bar internal pressure.

2.2 Technical positioning ojVESTAMID LX9030 between PE and steel

Steel has a long history in gas installation and it is still the only option for real high pressure gas transportation at 50 bar or higher. Reinforced plastic pipe constructions might be feasible to serve systems with pressures up to 45 bar (Wessing, Grass, 2006).

Welding of steel is a well established technology and weld quality control procedures are in place. Also with respect to third party damages steel has naturally a high stability. On the other side steel pipes are rigid, heavy and corrosive, which makes them principally not attractive for installations.

In Europe and USA the gas distribution networks are operating at pressures of up to 25 bar, with a big share around 16 bar. PE100 as the latest development of High Density Polyethylene received the approval for operating pressures up to lObar however this comprises only a small share of distribution networks. PA 12 pipe systems are technically able to carry gas at pressures up to 18 bar and withstand elevated temperatures up to 80°C, even in a longer term. This would fit quite well to serve distribution networks running at a pressure of 16 bar (Dowe, Baron, Buessing, 2007).

3 CO-OPERATION OF EVONIK AND E.ON-RUHRGAS ON EVALUATING PA 12 FOR HIGH PRESSURE GAS SUPPLY

3.1 First high pressure test installation in 2005

Evonik Degussa and E.ON Ruhrgas decided in 2005 to set up a high pressure test installation on the E.ON Ruhrgas Technical Center site in Dorsten. Evonik had to provide 60 m of 110 mm SDR-11 VEST AMID pipe on a coil and some straight pipes for assembling a system including butt fusion and electro fusion joints. Evonik also had to provide electro-fusion end caps. For the extrusion and coiling of the requested pipe Evonik choose the company Egeplast in Greven. Although having PA 12 the first time on their production line Egeplast was able to extrude the pipe within the tolerances and with excellent appearance without any problem. The online coiling on a 2.5 m diameter drum was running smooth without cranking the pipe on the coil at all.

For the development of electro-fusion fittings and end-caps Evonik co-operated with company Friatec in Mannheim, one of world leading companies in electro-fusion fittings. Using the VESTAMID gas pipe material they manufactured the required components for the test installation in a perfect manner.

Before the test installation on the E.ON Ruhrgas site a burst pressure test was carried out with a 3 m test pipe including a butt fusion and an electro-fusion joint and electro-fusion end-caps. The pipe sample burst at a pressure of 94 bar in a tough crack at the main tube, not at a fitting or joint.

For the test installation the coiled tube was unrolled in the field without any mechanical stretching tool. Only at few points needed some heating assistance to be stretched for applying the connections. Butt fusion was carried out with standard equipment and slightly adjusted temperature profiles. For the electro-fusion fittings standard power generators from Friatec were used. The adjusted fusion conditions were read from a bar code adhered to the fittings. The system with butt fusion and electro fusion joints and two electro-fused end caps was installed and sealed in a 500 mm steel pipe. For 72 hours 36 bar natural gas was applied for checking the tightness of the system. Then the pressure was lowered to 24 bar.

Additionally a similar pipe to the burst pressure sample was installed in parallel. This 3 m long pipe section was pressurized to 36 bar.

69

Table 3. Properties after 2 years of test installation.
Virgin pipe After 2 years After 2 years
Properties Standard from stock at 24 bar at 36 bar
Tensile test at 23°C ISO 527
Strain at yield [MPa] 38 39 40
Elongation at yield [%] 13 12 14
Strain at break [MPa] 49 46 49
Elongation at break [%] 315 287 293
Water content [%], across wall
Outer 0.69 0.63 0.52
Centre 0.46 0.27 0.34
Inner 0.56 0.28 0.31 3.2 Observations after 2 years of operation at 24 and 36 bar

Both installations run very successful. Except an external leakage at a manometer after 8000 h caused by storm Kyrill no further distinctive features were observed. After 2 years (l8000h) of operation, the 36 bar pipe sample was depressurized and taken. The following burst pressure test did show a burst pressure of 86 bar. This is a reduction of 8 bar compared to the virgin pipe sample tested prior to the installation. The investigation of the water content in the pipe material did show a increase from 0,1 % to about 0,4% as expected. PA 12 absorbs a maximum of 1,4% of water if fully immersed and 0,8% at 50% humidity at 23°C. This absorption causes a softening of the material.

It must be emphasized that the investigation of the MRS value in ace. with ISO 9080 was done at saturated pipe samples. Therefore and because the test is done fully immersed in water this softening effect is already fully considered at the investigation of the MRS of the material.

Also a pipe sample including a butt fusion joint and a electro fusion joint was cut after depressurization the 24 bar installation. Beside tensile tests on dog bones prepared from the pipe samples the water content was tested. At table 3 the results of the tests in comparison to virgin pipe samples are shown. A hydrostatic pressure test in ace. with ISO 22621-2 at 80°C and 20 bar was also set up. This test is still pending and results will be shown at the conference.

3.3 Expert report by TUVNord and SKZ Wuerzburg

With the very good experiences of the test installations and the good test results in ace. with international standards Evonik assigned a review of all available data to the German TUVNord and SKZ Wuerzburg. As a result an expert report was set up in co-operation of TUVNord and SKZ which stated the general suitability of VESTA MID LX9030 for high pressure gas pipe applications up to 18 bar. It is also mentioned at the report that pipes made of VEST AMID LX9030 are as safe as steel pipes if installed following the recommendations of the report.

4 TRENCHLESS TECHNOLOGIES CONSIDERING PA12 PIPES

4.1 Crack growth resistance of PA 12

During the feasibility study of PA12 for gas applications several tests were observed in ace with international standards. Considering the sensitivity of former PE materials to slow crack growth (SCG) and its influence on the installation technologies Evonik paid much attention on this issue. Because of the high resistance to stress cracking PA 12 is used for air brake tubing systems since decades. Therefore it could be assumed that SCG would not be an issue for PA 12. This is also indicated by the long-term hydrostatic pressure test in ace with IS09080. Whereas PE is known to

70

Table 4. Results on slow crack growth resistance of VEST AMID LX9030.

Test Standard Requirement LX9030pipe
Pipe Notch Test ISO 13479 & 22621-2
20% notch depth >500h >2000h
30% notch depth >2000h
PENT ASTMF1743 >100h >1000h have a second branch at the time to failure curves this is not investigated by tests on PA12 pipes. Even after 12,000 test hours at 80°C no stress cracking initiated failures are detected.

Additionally the SCG behavior of PA12 was investigated with the Notch Pipe Test in ace with ISO 13479 and ISO 22621-1. Also the PENT test in ace with ASTM F1743 was investigated. The results are shown in table 4. At 80°C and 20 bar no failures occur at the Pipe Notch Test after 2000h. Even after an increase of the notch depth from 20 to 30% of the wall thickness no failures occur. The minimum requirement in ace. with ISO 22621-1 is 500h without failure with a notch depth of 20%.

The PENT test is very well established in the US gas society and PE materials are tested at 80°C and 2,4 MPa stress. Presently, the requirements within ASTM D2513 for PE materials require PENT time to failure of 100 hours. However, no such requirements are in place for Polyamide materials. Two replicates of the PA12 material were tested in accordance to ASTM F1743 requirements at an increased stress of 4.8 MPa. The results of the testing indicated that there were no failures with any of the specimens after 1000 hours. The testing was discontinued after 1000 hours.

4.2 Effects of secondary stresses on PA 12 pipes

In addition to characterizing the SCG performance characteristics and influence of surface scratches, additional tests at GTI were performed to characterize the influence of secondary stresses. The main motivating factor for performing these tests was the fact that under actual field conditions, the piping systems are subjected to the combined effects of both internal pressure and other secondary stresses including rock impingement, earth loading and bending. Often, these secondary stresses, not internal pressure, are the root cause of many in-service field failures. Additionally the point load resistance is highly discussed relating to trenchless technologies and sand-bed free installation of thermoplastic pipes.

Therefore, comprehensive long term sustained pressure tests were performed at elevated temperatures to characterize the effects of various types of secondary stresses. It is important to note that these tests are not a part of either the ASTM or ISO standard. The test methodology is an extension of previous research performed by Dr. Charles Bargraw - DuPont and further refined by Dr. Michael Mamoun - Gas Technology Institute to study the performance characteristics of older generation PE materials. For the case of the rock impingement, the intent is to evaluate the performance of pipe materials subjected to indentations by a 13 mm rock. For the case of the earth loading, the typical safe deflection limit that is specified is 5%. For the case of the bending strain, the typical bend radius limits for a pipe specimen without any joints or appurtenances is 20 times the outside diameter.

Six (6) 2-inch SDR-11 LX9030 pipe specimens were placed in appropriate test rigs to simulate the effects of rock impingement, earth loading, and bending strain. The entire test assembly was placed under long term sustained pressure testing at 20 bar at 80°C. For all of the various types of secondary stresses which were evaluated, the results of the testing demonstrated that there were no failures after 1000 hours of testing, as presented in Table 5 below.

In summary the test results show that PA12 has a unique resistance to SCG and secondary stresses including point loading. Therefore it could be assumed that PA 12 pipes might be suitable

71

Table 5. Test results of VEST AMID LX9030 pipes investigating secondary stresses.

Secondary stress Test criterion Results
Rock Impingment 13 mm Indentation Test Time> 1000 hour
Test Pressure = 20 bar with no failures
Earth Loading 5% Deflection of Outside Diameter Test Time> 1000 hour
Bending Strain 20 times OD Test Time> 1000 hour
Test Temperature = 80°C Figure 1. Surface of the PA 12 pipe (left) vs. PE 100 pipe (right) after HDD trial.

for installations wihtout sand-bedding and with trenchless technologies even without additional protection layers. To solve remaining questions to this issue Evonik is running additional tests like field installations and laboratory tests at Hessel Ingenieurtechnik GmbH.

4.3 First directional drilling trial at Tracto-Technik Gmb&CoKG (IT) test centre

To investigate the behavior of PAl2 pipes during the directional drilling technology (HDD) ofTT a 4m test sample of a 1l0mm SDR-ll VESTAMID LX9030 pipe was drilled at TT test field in Lennestadt, Germany. To enable a direct comparison to PE a standard PE100 pipe with the same dimension was drilled in parallel. Both pipes were drilled thru a depth of 4 m and a distance of 60 m. The drilling diameter was 135 mm and the soil consists of a pebbles - sand mixture. After pulling out the pipes a visual investigation of the outer surface showed that the PA12 pipe has only very tiny scratches compared to the PE pipe (Fig. 1). A closer look on the surface by microscopy and semi electron scanning (SEM) proved this result (Fig. 2).

Additionally hydrostatic pressure tests were set up in ace, with ISO 13479 and ISO 22621-2 to show the excellent performance of the pipe material even after the drilling trial. Results will be presented at the conference.

The excellent SCG and point loading behavior combined with the minimized notch sensitivity underlines that PA12 is a promising candidate for trenchless technologies like directional drilling and pipe bursting as well.

72

Figure 2. Depth of the scratches after the trial, LX9030 (left) vs. PE100 (right).

5 CONCLUSION

Summarizing the field test experiences and the laboratory tests in ace with international standards it can be concluded that PA12 is a suitable material for high pressure applications. Therefore PA12 extends the well known benefits of PE pipe systems to the pressure range of steel pipes up to 18 bar at gas applications. The test results also underline that along with the traditional way of installation also trenchless technologies are suitable. The outstanding SCG and point loading resistance of PA12 is one of the strength of this promising material for high pressure applications.

REFERENCES

Dowe A., Baron Ch., Buessing R. 2007. New VESTAMID® (Polyamide 12) grades to manufacture products with large geometrical dimensions for oil and gas applications. Conference book. Plock: Technical Conference Risk Management for Pipeline Operation 2007

Wessing W, Grass K., Kanet, 1., Capdevielle, J-P. 2004. Novel PE gas supply system for a maximum operation pressure of 16 bar, Conference book. Vancouver: Gasresearch Conference 2004

Bayer, H.-1.2006. HDD-Practise Handbook. Vu1kan Verlag Essen 2006

Ramei1, M. 2007. Handbook of Pipe-Bursting Practise. Vulkan Verlag Essen 2007

73


Simulation researches of pump-gravitational storage reservoir and its application in sewage systems

1. Dziopak & D. Slys

Department of Infrastructure and Sustainable Development, Rzeszow University of Technology, Poland

ABSTRACT: Low accessibility of municipal building sites and growing demands in the field of volume efficiency and reduction of net-work depth below storage reservoirs limits the range of economically reasonable applications of storage reservoirs. This fact brought to the beginning of the researches directed to elaboration of new constructions of storage reservoirs of limited building surface. The paper deals with the hydraulic systems' solutions of pump-gravitational waste water retention. The range and results of simulation researches of such type of sewage systems' objects are also presented.

1 INTRODUCTION

The problem of waste water transport by sewage systems and waste water management in urban areas is actually very significant in the light of the increase of proofed surfaces in catchments and frequency of extreme atmospheric precipitation occurrence. The development of urban areas influences on the reduction of green and non-hardened surfaces, from which surface flows could feed ground waters. At the same time the demands and standards concerning the preservation and drainage of urban areas are becoming more rigorous. The effect of this phenomenon is the growth of storm water flows discharged by sewage systems.

The growth of storm water and waste water quantity and their great variability in time negatively impacts the processes of waste water treatment and surface water quality and demands the expansion of existing sewage infrastructure. Modem sewage systems requires, then, the usage of efficient methods of waste water flows regulation at the stage of their transport by sewage systems and before waste water treatment plant.

The problem of waste water flow control in sewage systems is the subject of many qualitative and quantitative researches.

The researches are based on the usage of modem soft-ware instruments for simulation of sewage systems' functioning. Interesting researches in the field of simulation of storm waters' influence on sewage system and waste water treatment plant, acting jointly with storage reservoir, were carried out by Calabro and Viviani (2006).

Calabro (2001) confirmed the significant influence of storage reservoir on contamination's removal in the process of sedimentation, that is important in the case of objects located on canals, discharging storm water to recipients and before primary sedimentation tanks. The author also showed that first rain portion in different catchments has different content and influences the efficiency of storm water reservoirs, but under definite flow intensity the increase of storage reservoir's volume do not impacts significantly on the suspended substances' removal efficiency.

Waste water flow control is very important in the case of sewage systems with significant contribution of combined sections, that brings to great variability of waste water flows and negative impacts on waste water treatment plants (Diaz-Fierros et al. 2002). In such cases significant improvement of storm water quality and flows stabilization are obtained by application of different hydraulic schemes, where the main role is played by storage reservoirs (Huebner & Geiger 1996).

75

2 HYDRAULIC SCHEMES' CONCEPTION FOR PUMP-GRAVITATIONAL STORAGE RESERVOIRS

The main problem in sewage systems' projecting is the necessity of obtaining of significant volumes of storage reservoirs in the areas of high level of urbanization and low availability of building surfaces. This problem concerns reservoirs located within waste water treatment plants and city areas.

Taking into account the significance of this problem for the development of sewage infrastructure, the authors are carrying out the researches of storm water accumulation processes in storage reservoirs with pump-gravitational hydraulic scheme of waste water retention. The usage of pump methods of transport enables the constructing of reservoirs of high level of storm water accumulation without of necessity of excavation of chambers and canal, located below the reservoirs. In some solutions there is a possibility of shallow location of inflow canal of reservoir. The result of the researches is the elaboration of few reservoirs constructions which can be classified in following ways:

- pump-gravitational reservoirs with upper pump accumulation chamber KAW,

- pump-gravitational reservoirs with lower pump accumulation chamber KAD,

- hybrid pump-gravitational reservoirs.

All solutions of pump-gravitational reservoirs with upper pump chambers enable the usage of open constructions of pump chambers and their building in embankments. So, these solutions are especially useful as the reservoirs within waste water treatment plants, where they can play the role of waste water pump station of high accumulation capacity.

Storage reservoirs with lower pump chamber enable the building of such objects on their surface as garages, storehouses, parking etc.

Hybrid reservoirs, owing to storied scheme of accumulation chambers, have the most limited building surface, so their application field are mostly the areas of high investment level and low availability of building grounds.

3 HYDRAULIC AND MATHEMATICAL MODELING FOR CHOSEN SOLUTION OF PUMP-GRAVITATIONAL RESERVOIR

Hydraulic modeling of storage reservoirs is the base of mathematical description of their functioning in sewage systems (Dziopak & Stys 2007, Stys & Dziopak 2006) and of simulation programs (Dziopak & Slys 2007, Slys 2006).

The idea of hydraulic modeling of storage reservoir is based on separation of characteristic phases of its functioning in sewage system and definition of marginal conditions for reservoirs' filling by waste waters in each chamber and the range of characteristic flows. Figure 1 illustrates the hydraulic scheme of pump-gravitational reservoir of GPWT type and characteristic parameters, used in elaborated model of reservoir functioning.

Taking into account the complexity of hydraulic processes during waste water accumulation, synchronism of their occurrence in few chambers simultaneously and probable complexity of route of inflow function and the resulting curvilinear functions, describing the processes inside the reservoir, as well as the variability of outflow, the general mathematical model of reservoir functioning can be described by Equation 1.

dh

- = QA . FrI. dt - Qt . Fi1 • dt - Qp . Fi1 • dt - Qr . Fi1 • dt dt

dHw -1 -1 -1

-- =Qp.Fw .dt-Qw·Fw .dt-Qz·Fw -dt

dt

dhs -1 -1 -1

- = Qt· Fs . dt + Qw . Fs . dt - QO . Fs . dt dt

76

(1)

Figure 1. Hydraulic scheme of GPWT type storage reservoir and characteristic parameters of hydraulic model (QO - waste water outflow from storage reservoir; Qw - waste water outflow from accumulation chamber KAW to steering chamber; Qt - reduced waste water flow from transport chamber KT to steering chamber KS; Qr - waste water outflow from transport chamber KT through emergency overflow; QA - waste water inflow from sewage system to storage reservoir; Qp - capacity of pump system transporting waste water from KT to accumulation chamber KAW; QZ - waste water outflow from accumulation chamber KA W through emergency overflow; KA W - gravitational accumulation chamber; KS - steering chamber; KT - transport chamber with pump system; hsnor- level of filling in steering chamber KS in the period of dry weather with reference to comparative level; hs - temporal level of filling in steering chamber KS with reference to comparative level; hsmax - maximum level of filling in steering chamber KS in the period of storm water inflow or in the period of accumulation chamber KAWemptying with reference to comparative level; hz - average bottom elevation of steering chamber KS with reference to comparative level; Hg - elevation of upper edge of outlet; He - average elevation of accumulation chamber's bottom with reference to comparative level; ha - elevation of emergency overflow edge of transport chamber KT with reference to comparative level; ho -level of switching of waste water pump transport system from transport chamber KT to steering chamber KS and waste water treatment plant with reference to comparative level; hOmin - level of switching off the waste water pump transport system from transport chamber KT to steering chamber KS and waste water treatment plant with reference to comparative level; hi - average elevation of transport chamber KTbottom with reference to comparative level; Hw - temporary filling of accumulation chamber KAW with reference to comparative level; h - temporary filling of transport chamber KTwith reference to comparative level; hr -level of switching of waste water pump transport system from transport chamber KT to accumulation chamber KA W with reference to comparative level; hr min - level of switching off the waste water pump transport system from transport chamber KT to accumulation chamber KA W with reference to comparative level; Hpa - elevation of emergency overflow edge of accumulation chamber KAW with reference to comparative level).

Minimal marginal values for filling of storage reservoir's chambers are: h ~ hi, Hw ~ He, hs~Hz.

4 SIMULATION PROGRAM SIMTANK

On the base of hydraulic and mathematical model of GPWTtype storage reservoir, software instrument was elaborated that enables the simulation of hydraulic processes in storage reservoir of this type for any route of waste water inflow function.

77

ME 8.----.-----.-----.-----.----.-----.-----.-----.----.-----.--.

~ , , , , ..... , , , , , ,

~ 7 ····~~~·~r·~·~~~·~t-.·~~·~·~'·~~·~~/(·::····:l:·:\:·······1·············1·············1·············1·············1·····

~ ::: J :....: Td= 50:min: : :

~ 6 nn ... n...l ...... nn ... l ............. l ....... ,/:·:".l. ... n ... nnLn .... ><' .... n .... ; ....... n ... .L ... nn .... Lnn ...... Ln

c:: o

~ -s E 5

:::J

~ £ 4

Q)

I

.s Cl

c:: 3

-=

...

Q)

~ 2

20

40

60

80

100

120

140

160

180

200

time t, min

Figure 2. The level of filling Hw and waste water volume Vzb max, stored in accumulation chamber, for rain duration Td = 20, 35 and 50 min. The rest simulation parameters: average annual precipitation H = 720 mm; probability of calculation precipitation occurrence p = 50%; catchment's area F = 125 ha; surface flow coefficient '11 = 0,8; coefficient of domestic waste water dilution at overflow nrp = 6; coefficient of domestic waste water dilution in transport chamber nrz = 2.

Calculation program SIMTANK makes possible the dynamic simulation and analyzing of hydraulic processes in storage reservoir during waste water inflow.

Simulation program SIMTANK considers the order and sequence of reservoir's functioning stages in accordance with hydraulic model of its functioning.

While operating, the program makes the analyses for demanded parameters of marginal conditions and calculated parameters. At the moment of achievement of marginal conditions for particular phases of reservoir's functioning, the transition between consecutive stages of its functioning takes place. Such conditions are: characteristic waste water levels in reservoir; waste water inflow's duration and intensity. Figure 2 presents the results of exemplary simulation of filling process in accumulation chamber KA W for the rains of different duration and different outflow functions in the shape of triangle.

One of the most important stages of storage reservoirs' calculation is the estimation of the route of function of waste water inflow to storage reservoir. The modern calculation instruments in the form of hydrodynamic catchments models enable the simulation of waste water flow functions' routes for any precipitation intensity, that is very useful in the case of project parameters' selection for storage reservoirs.

In the case of the lack of information about precipitation and the lack of possibility of real waste water flow diagrams' usage, the substitutive waste water flow diagrams can be used. In this situation the most significant calculation parameter is duration of calculation rain Tdm. The process of this parameter's selection for substitutive diagrams were described by Dziopak in his publications (1997).

78

'0
Cl 1000
c
·u
ID
-~ c
0.--
._ E
o "
-e: 800
-~~
._
ID--
> 0
-.;:::: c:
co ID
.... 1Il
C ID
ID ._ 600
III ID
~Cl
0. CO
ID ._
._ 0
aU)
C
0
.~ 400
::::l
Cl
200 1200

.
.
--p=50% .
.
r--- - -p=20% .
---p=10% .
.
.
.
.
.
. "
. "
. '"
"
I--'
_,
"
""
" ~
""
~
~ V 500 600 700 800 900 1000 1100 Average annual precipitation H, mm

Figure 3_ The influence of average annual precipitation H on duration of representative rain Tdm for project" ing of storage reservoir under following calculation data: unit discharge of domestic waste water flow qs = 2 dm3/s ha; catchments area F = 200 ha; surface flow coefficient III = 0,6; coefficient of domestic waste water dilution at overflow nrp = 4.

Figure 3 presents the results of calculations of rain duration Tdm, appropriate for storage reservoir projecting of GPWT type in dependence of the following hydrological parameters of catchments: average annual precipitation H and probability of the rain for the projecting of storage reservoir p.

Rain duration Tdm depends on the number of hydrological and hydraulic parameters as well as on hydraulic capacity of waste water treatment plant, the way of catchment's usage and water consumption. Figure 4 presents the results of simulation of rain duration Tdm appropriate for projecting of storage reservoir, located between storm overflow and waste water treatment plant in dependence on the level of surface sealing in catchment's area.

5 CONCLUSIONS

The exploited gravitational sewage systems are under constant expansion and modernization. The main problems for technical solution are: hydraulic overloading of sewage nets, surface water protection against contamination discharged by sewage systems and control of waste water flow to waste water treatment plants. These problems can be solved successfully with the help of modern efficient constructions of reservoirs for periodic retention of waste water.

Taking into account the merits and imperfections of existing constructions of storage reservoirs within storm and combined sewage systems, the researchers of new group of solutions of pumpgravitational scheme of waste water retention have been carried out.

Such type of storage reservoirs makes possible to store waste water in accumulation chambers of significant heights and limited grounds for building, that is especially important for urban areas.

79

800
750
700
'0
Cl 650
r::::
·u 600
Q)
·er:::: 550
0..-
.... E
o -
..... e: 500
. ~~
Q) .: 450
> 0
. .;::::: c:
ctl Q) 400
+-'11)
r:::: Q)
Q) ....
II) Q) 350
~Cl
o.ctl
Q) .... 300
.... 0
000
r:::: 250
0
.~ 200
....
::::l
0
150
100
50
.
.
.
.
--H=700mm /
- -H=800 mm .
••• H= 900 mm . /
.
. 1 /
. 1 /
. ~
. /
. 1
.' I /
~
. 1/
.
•• J '/
.'//
.' / V
~/ 0,3

0,4

0,5

0,6

0,7

0,8

Waste water flow coefficient for catchment's area t!J

Figure 4. Influence of waste water flow coefficient for catchment's area \11 on rain duration Tdm, appropriate for storage reservoir projecting for the following calculation data: probability of calculation rain occurrence p = 50%; domestic waste water unit inflow qs = 1,5 dm3/s ha; coefficient of domestic waste water dilution at overflow nrp = 5; catchment's area F = 500 ha.

Simultaneously, owing to the idea of multi -chamber reservoirs' use, which are equipped with transport chamber steering the operation of accumulation chambers, the high efficiency of waste water accumulation was achieved. Equipping of multi-chamber reservoirs by pump transport schemes and outlet schemes enables the options in chambers' hydraulic forming, that influences on the broadening of the field of their application, economically and practically motivated.

The elaborated calculation programs SIMTANK, serving for the purposes of simulation of GPWT type storage reservoir operation, are the modem researching and projecting instruments for detailed investigation of waste water accumulation processes in such types of reservoirs, located before waste water treatment plants. The use of SIMTANK programs in project processes for selection of main hydraulic parameters, chambers' capacity and geometry, as well as installations acting jointly with them, enables the wide usage of pump-gravitational constructions of reservoirs, especially for the purpose of waste water treatment plants' hydraulic unloading.

REFERENCES

Calabro, P. S. & Viviani, G. 2006. Simulation of the operation detention tanks. Water Research 40: 83-90. Calabro, P. S. 2001. Cosmoss: conceptual simplified model for sewer system simulation. Urban Water 3(1-2): 33-42.

Diaz-Fierros, E, Puerta, 1., Suarez 1. & Diaz-Fierros, V. 2002. Contaminant loads of CSOs at the wastewater treatment plant of a city in NW Spain. Urban Water 4: 291-299.

80

Dziopak, 1. 1997. Multi-chamber storage reservoirs in the sewerage system. Technical University of Czestochowa. Czestochowa.

Dziopak, 1. & Siys. 2007. Modelowanie zbiomikow klasycznyclt i grawitacyjno-pompowych w kanalizacji.

Rzeszow University of Technology. Rzeszow,

Huebner, M. & Geiger W 1996. Characterisation of the performance of an off line storage tank. Wat. Sci. Tech. 34(3-4): 25-32.

Siys, D. 2006. Simulation model of gravitation-pump storage reservoir. Environment Protection Engineering 32(2): 139-146.

Siys, D. & Dziopak, 1. 2006. Simulation of Trough-Flow Chamber Operation in Storage Reservoirs. Ecological Chemistry and Engineering 13(10): 1143-1155.

81


New developments in liner design due to ATV-M 127-2 and case studies

B. Falter

University of Applied Sciences, Munster, Germany

ABSTRACT: The German Code ATV-M 127-2 published in 2000 has proved itself a helpful guideline to find the optimum wall thickness of any liner material, e.g. CIPP or stainless steel sleeves. Many rehabilitation projects in different European countries have been performed successfully using this code. The code differentiates between three host pipe states: State I for untight sewers without cracks, state II for sewers with longitudinal cracks but a stable soil pipe system and state III for cracked pipes with larger deformations and considerable risk to collapse in the near future. According to the code stress, deformation and stability tests are necessary. For many practical cases charts with stress factors and imperfection reductions allow to design without a computer. The paper reports about the progress in liner design since the 1 st edition of the code. Additional clauses have to be introduced into the 2nd edition for non circular geometries (e.g. for horseshoe and rectangular profiles) and for new applications (e.g. railway crossings).

1 INTRODUCTION

The critical water pressure equation is based on a Glock formula (Glock 1977) for elastic rings encased in a rigid boundary. This equation was used by many researchers and enhanced by reduction factors to describe the real situation of the host pipe (e.g. deformations and annular gap). In a few codes the critical pressure of an unsupported ring (Timoshenko 1961) is used and increased by a support factor K; for example K = 7 for good and 4 for poor installation conditions (WRclW AA 2000).

The critical soil pressure is treated less often in research papers. The reason might be that experiments on the broken pipe soil-system with an overburden (e.g. Watkins 1988) show a conservative behaviour without a collapse. On the other hand in such experiments uniformly distributed pressures are applied onto the sandbox surface more often than concentrated wheel forces.

2 DESIGN CONCEPT OF THE CODE M 127-2:2000, 1ST EDITION

2.1 Host pipe state

The German liner design concept is based on the differentiation of three host pipe states. The state I and II (without and with longitudinal cracks) must be calculated only for groundwater acting as a pressure on the outside of the lining. In case of state III an additional calculation for soil and traffic loads is prescribed.

2.2 Buckling pressure, imperfection reductions and proof of stability

Due to the Code ATV-M 127-2 the buckling load for the water pressure Pa valid for all host pipe states is evaluated regarding three kinds of imperfections, cf. Figure 2:

a) Local imperfection Wv = 2% which must be chosen according to the relevant buckling mode

b) Annular gap Ws = 0.5% for CIPP caused by shrinkage of the liner material

c) Global imperfection WGR,v ::: 3% caused by the deformation of the cracked host pipe

83

(e)

Figure la-c. Host pipe state I (a), II (b) and III (c) as defined in the German Design CodeATV-M 127-2.

wy=2%

Figure 2a-c. Local imperfection Wy (a), annular gap Ws (b), global imperfection WOR,y (c), minimum values.

For the imperfections in Figures 2a-c reductions factors K for the buckling load are given in the Design Code. The factors depend on the depth of the imperfection and the rils: ratio describing the slenderness of the liner construction and the character and the size of the host pipe damages.

For the critical water pressure Pa of a circular lining the following formula has been developed (Falter 1993).

crit Pa = Ky,s . aD • SL

(1)

where Ky,s ~ Ky' ««: KOR,y is the common reduction factor for all imperfections due to Figures 3a-c; aD = 2.62· (rL I sdO.8 is the snap through factor (Glock 1977); and SL =Ed12 . (SL I rd3 is the ring stiffness of the liner (EL = long-term Young modulus).

The minimum values for Wy, Ws and WOR,y to be applied are given in Figures 2a-c and 3a-c. The main problem in the practice is the correct assumption of the global imperfection WOR,y, Usually the value of WOR,y has to be evaluated from a video screen which results in sometimes different opinions of the engineers about this issue.

For non circular linings a computer evaluation of the critical water pressure is necessary or a substitute radius on the safe side has to be taken for rL in Equation 1. It is strongly recommended to calculate the substitute radii by separate non linear analyses; sometimes the amount and the location of the local imperfection must be varied to get proper results. The correct location of this kind of imperfection is usually the region with the smallest curvature, e.g. the invert for horseshoe profiles; the region beneath one springline for egg shaped profiles.

2.3 Proof of stresses and strains

The section forces M and N caused by groundwater pressure Pa are calculated by means of dimensionless factors lnpa and llpa given in appendixA4 ofM 127-2 for the diameters ND 200 -ND 600, cf. Figure 4.

Mpa = lnpa . Pa . rl with lnpa due to Code M 127-2, appendix A4

(2a)

84

1,0 .-----,-------,------r-----=---

1 ~::

0,7 I__------<=~'io?_o;;::_----"",-.I:::-____._._"""'-~--=p'=_-- 0,6 I__----..~_f"o;;:---""'o,"""""..t:-~"""_:::_+_---=-+-==_-i:; 0,5 I--~_P_o.~_..::_~a!i"""""'+""""" ........ d----=

0,4 I__--_f"o;;:---::""""'=+""....e'-":::O"~.=.._cl--;:::::.._

0,3 I---+---=""""=+""""~",_"'"""=="'"t- ....... ~"",. 0,2 I---+------t-=='~=_t_-=",,---+~:::::::::~ 0,1 I---+---+--____:+=~-+---_';

0,0 °

1,0
1 ~::

0,7
0,6
'Q 0,5
0,4
0,3
0,2
0,1
2 4 6 8 0,0 ° 2 3 4 5
ws/rL·100% 1,0 .-------:::::--..,------r------r-----r------,

f~::E~~~

0,7

0,6 t=::t~j::S~:::::~~~

0,5

~0,41--~L_+-~.--+----r---::~~~~~ ~0,31--;+-+-~~-+----r---+-~~

0,2 I__~.----+____#L..-_+_--_r--_+-----'~ 0,1 I__-__",__~O<""--_+_--_r--_+-----J 0,0 '-- __ ...J..._ __ --'- __ --'- __ ----'- __ ----'

°

2

4

6

8

10

Figure 3a-c. Reduction factors Kv, Ks and KGR,v for the buckling pressure due to local imperfection Wv, annular gap Ws and global imperfection WGR,v, source: ATV-M 127-2:2000.

0,12 .------,------,--------,,----.--------.--------,

Host pipe state II (3 Yo ovalisat on) Host pips state I

0,10 I-----+----+-------jl----+----+------J

t ND 600

0,08 I-----+----+-s--=d-m-m-.,.j] f-=-1-9.-'--o .: ,-..'--1----+------1

10 12.5

E~ 0,06 ". . .. ' . ~.<. 15

~: .. 'r>: ;;::;;;;: ~~ .: ·· .. f2.5~

0,04 '''_

0,02 1__---+------J----+---_+---+_-----1

0,0°1

2

3

4

5

6

7

max hw above the invert [m] --__.

Figure 4. Invert bending moment factors for liners ND 600, source: M 127-2, Appendix A4.

Npa = npa . Pa . rL with npa ~ -0.8 to - 1.1 The stresses are calculated using Equations 3a, b:

(2b)

N M

a, = A + aki • W with aki = 1 + sL/3rL, aka = 1 - sL/3rL (3a)

N M .2 2· 3

(J"a = A - aka' W A = sonmm jmm and W = sd6mmm jmm (3b)

The resulting tensile stress is compared with the ultimate flexural strength of the lining material, reduced for long-term behaviour. As the pressure stresses due to Equation 3b are a bit larger an additional material pressure test leads in many cases to a more economical design.

85

For host pipe state III the soil and the traffic loads qy are applied to the pipe's crown and the structural safety of the total system of soil, cracked host pipe and lining is analysed. The axial force and bending moment factors nq and mq are given for ND 200 to ND 600 in appendix A5 of Code ATV-M 127-2.

Elaborate experiments have been necessary to prove the design formula, referring to national and international research, cf. chapter 4.3.

2.4 Discussion

The stress proof in chapter 2.3 is the most important test. The factors 1npa calculated non linearly with appropriate imperfections contain the risk of stability failure. In many cases computer analysis is necessary as the validity of the cross section factors is restricted to a range of material properties. The stability proof in chapter 2.2 has the advantage of general validity for all diameters and all material properties. This proof can be used for approximate evaluations of the wall thickness without computer programs. Deformation analyses are less important and used sometimes for judgement of serviceability.

3 EXTENSIONS OF THE DESIGN CODE M 127-2, 2ND EDITION

3.1 Additional reductionfactorsfor the buckling load

The regular values for most rehabilitation situations with CIPP-liners are 2% local imperfection, 0.5% annular gap and an arbitrary global imperfection. For this constellation only one chart with reduction factors Ky,s for the critical pressure is necessary, cf. Figure 5.

3.2 Simultaneous calculation of water and soil pressure

In the Design CodeATV-M 127-2:2000 the load cases Pa and qy are treated separately. Subsequently they must be combined by an interaction formula. New developments show that an enhanced numerical model is able to cover both load cases in one step. For host pipe state III without water

0,9
0,8
0,7
1 0,6
0,5
III
oj
0,4
0,3
0,2 I I
r-.. WV I'L=2.0%
-- ~ I
r-..... r-- 'LI SL = 1O wsl 'L = 0.5%
h:-- ~ r-- --- r--
r--.
r---- ---= F::::::: ::::::: hi- -- r--
-- 2 r---- I-- - -
r---- r-- - ~ ~ --
-- -- ~ -- r--- -
r--- -- r-- --- -- - ~
~ r--- ,_ -
r-- -- --
--- r-- 100 0,1

0,0

°

2

3

4

5

6

7

8

9

10

Figure 5. Reduction factors Ky,s of the buckling pressure, 2% local imperfection, 0.5% annular gap and arbitrary global imperfection (Falter 2003).

86

table a double symmetry of the system could be assumed; in the case of a load combination this is however no longer possible, cf. Figure 6a,b.

3.3 Lining offlexible sewers

Sometimes the future integrity of the old pipe is called into question. In the Design Code ATVM 127-2 it is assumed that the host pipe has enough strength to support the lining in the radial direction. If the absence of any host pipe structure is expected in the future new analysis problems arise: The liner without host pipe. The main stresses are now to be expected in the liner crown, cf. Figure 7a.

3.4 Soil bedded lining (host pipe fully detriorated)

Sometimes the future integrity of the old pipe is called into question. In the Design Code ATVM 127-2 it is assumed that the host pipe has enough strength to support the lining in the radial direction. If the absence of any host pipe structure is expected in the future new analysis problems arise: The liner without host pipe. The main stresses are now to be expected in the liner crown, cf. Figure 7a.

3.5 Host pipe-soil system

In order to prove the host pipe soil-system the equilibrium of a rigid circular ring with four excentric hinges is investigated, cf. Figure 8. For regular cases with elastic soil behaviour the load deflection curve is described by the following equation:

~ • TJs • (PGy - TJs/3 - TJ)

(4)

where PGx and PGy = horizontal and vertical distance of the hinges from the pipe wall centre; TJ = deflection of the pipe's crown to the inside, ~ = outside deflection of the springline;

(a)

(b)

Figure 6a,b. Load cases groundwater Pa and soil pressure qv - a system without horizontal symmetry, (a) bending moments of the liner, (b) contact forces of the liner versus the host pipe (Linerb 2008).

(a)

(b)

(c)

Figure 7a-c. Bending moments (a), contact forces (b) and deflections (c) of a circular lining subjected to groundwater-host pipe neglected (Linerb 2008).

87

Underground Infra

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

Underground Infra

Загружено:

Авторское право:

Доступные форматы

UNDERGROUND INFRASTRUCTURE OF URBAN AREAS

Вам также может понравиться

Underground Infra

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

Underground Infra

Загружено:

Авторское право:

Доступные форматы

﻿UNDERGROUND INFRASTRUCTURE OF URBAN AREAS

Вам также может понравиться

UNDERGROUND INFRASTRUCTURE OF URBAN AREAS