Sewer Leakage -
Detection and Cures
by Mike Jones, mike@newtony.demon.co.uk
Project undertaken at various locations within the Wessex Water region
Abstract
This project examines the topic of sewer survey and rehabilitation from the perspective of leaking sewers. It is divided into four sections, Background, Identification, Cures and Recommendations, and uses tables to list techniques and costs. Case studies are listed in the Appendices.
Subject:Sewer leakage
Associated subject(s):Sewer survey, Sewer rehabilitation
Keywords: sewer, pipe, rehabilitation, leakage, infiltration, exfiltration, monitor, survey, CCTV, Sonar, Drainage Area Study, Drainage Area Plan, lining, pipe-bursting, stabilisation, micro-tunnelling
Index
Section |
Page |
Foreword |
4 |
Background |
|
Historical |
5 |
Cause |
6 |
Cost benefit |
7 |
Identification |
|
Techniques |
8 |
Methods |
11 |
Cures |
|
Techniques |
12 |
Methods |
16 |
Recommendations |
|
Infiltration |
17 |
Exfiltration |
19 |
The Future |
19 |
Appendices |
|
1 - Case Study A - Wilton Infiltration studies |
20 |
Case Study B - Martinstown |
21 |
Case Study C - Bransgore |
21 |
2 - Case Study D - Bourne Valley |
22 |
Case Study E - Salisbury Sewers |
22 |
Case Study F - Wilton Infiltration Remedials |
23 |
3 - References |
24 |
Tables, Pictures and Flowcharts |
Page |
Table 1 - Identification techniques |
10 |
Table 2 - ĪCuresā |
14 |
Table 3 - Cost of relaying in urban areas |
15 |
Table 4 - Cost of soft linings |
15 |
Table 5 - Cost of Īone-offā repairs |
15 |
Pictures 1 & 2 - Typical ĪUā lining unit |
Not included |
Flowchart 1 - Method selection |
18 |
Picture 3 - Typical CCTV unit |
Not included |
Picture 5. Pipe bursting unit |
Not included |
Picture 6. Heading required after pipe burst |
Not included |
Picture 7. Sanipor chemical being added to sewer |
Not included |
Report produced by Mike Jones, January-December 1998
Foreword
"There are 295,000 km of public sewers in England and Wales, of which 40% have structural defects that could allow leakage".......Paul Hayward, The Drain Trader, June/July 1998.
"Pollution of groundwater by raw sewage leaking from sewers is a serious and growing problem".......Joan Walley MP, Chairwoman of the Campaign for the Renewal of Older Sewerage Systems (CROSS), writing in Water Services. October 1998.
In recent years, there has been much interest in leaking water pipes, both by industry and public. Focus for this interest has largely been water supply pipes. The emphasis on sewerage however has nearly always been on serious structural defects. With the concerns of environment, cost reduction and legislation, companies are now starting to consider in more depth the problem of leaking sewers.
The aim of this project is to review methods and systems for both detecting and stopping infiltration and exfiltration in sewers, particularly in regard to sustainability and cost benefit. I have divided the report into four sections: background, identification, cures and conclusions. I have drawn on existing publications and historical data from works I have carried out. This is supplemented by current information from sewerage undertakers, contractors, suppliers, consultants, research being carried out by my own company (a water plc) and works I have undertaken over the past 18 years. I have also touched on the history of sewerage in Britain, and looked at potential scenarios in the future.
One area I have not covered is inflow. Americans talk a great deal about I&I (Infiltration and Inflow), and both are regarded in the same light. Many countries (and companies) have spent much time and money on removing inflow from foul sewers. Inflow however is not related to leaking pipes, but is merely a function of the particular drains connected. Recent research showing that 50% of pollution reaching watercourses derives from roads may reverse the trend of separate sewerage systems. I therefore believe inflow to be such a large subject that it demands a paper of its own.
As this is a short report I am unable to cover all areas fully, particularly details of renovation techniques, and formulae for flow calculation. I have thus made reference to sources for these details at appropriate points in the text. I would however recommend readers to acquaint themselves with the texts I have listed as references, particularly The Sewerage Rehabilitation Manual (WRc) and its periodically issued CD-ROMās of the latest techniques. I would especially recommend Sewers - Rehabilitation and New Construction, by Read and Vickridge, which gives an excellent all round view of sewer Īmattersā, and is very readable!
I would like to acknowledge the help of the following:
Background
Historical
Sewers have been in existence almost as long as people have lived in dwellings. It was long thought that sewers in the Indus valley, circa 2500 BC were the oldest, but excavations in the Orkney Islands (Skara Brae) have shown primitive drainage systems dated about 3000 BC. Roman civilisation was well known for its sanitation and yet in Britain it was not until the late 18th century that any consideration was given to sewerage. The Īmodernā sewerage networks begin at that time and I will examine this with reference to a town I know, Salisbury, whose sewerage history is typical of many in Britain.
Salisbury was once known locally as ĪLittle Veniceā. This was because of the multitude of canals that channelled the rivers (Avon, Naddar and Bourne) through the town. In common with many cities these had become little more than open sewers. The 1847 Water Clauses Act was the first legal document to mention the word Īsewerā, and from the definition given some of Salisburyās canals could be considered sewers. From this time City Councils were increasingly being asked to consider sanitation: the 1848 Public Health Act further reinforced this issue.
One of Salisburyās residents, a Dr Andrew Middleton, was one of the pioneers in research into water borne diseases. In 1850 he pressed Salisbury City Council to provide a proper sewerage system: up to that point the Council had only ever provided road drainage. To reinforce his arguments Dr Middleton called for a Inspector from the Health Board, a Mr Rammell, who made recommendations in his October 1850 report for a sewerage network. There were many objections, but cholera in the City reinforced the need, and sanction was given. The first pipes were laid in 1852, under the direction of Rammell (the Inspector?) and Lister, Consulting Engineers: the original plans of these are still available! In July 1853 the City Engineer, John Pannell, accepted the new sewerage system. The sewers ran to three river outfalls, Bugmore. Summerlock stream and Fisherton Mill, which are still visible today. The original ditches were then cleaned and filled. The many artefacts found in the ditches can be viewed in the Pitt-Rivers collection at the City Museum.
In 1900 plans for a Sewage Treatment Works at Bugmore were drawn up. Work started in 1901 under the direction of Binnie and Partners, and was completed in 1904. Further sewers were added to the main network in 1905. Other than local network additions for new development, little work was then done on the system until the 1960ās. Between 1962 and 1964 a new Sewage Treatment Works, designed by Sir William Halcrow and Partners and the City Engineers, was built at Petersfinger and this is still in use, largely unaltered, today. New interceptor sewers was also added to the network. The total cost of the project was £1,200,000, of which £880,000 was for the Treatment Works. In the 1980ās a Drainage Area Study (DAS) was commissioned from Halcrowās and from this a Drainage Area Plan (DAP) in 1989. A rehabilitation scheme was undertaken in 1991 to 1992, a notable feature being the very small amount of work that was required on the Victorian sewers.
My own experience, which includes 9 years of CCTV survey of sewers, has generally reinforced the Salisbury DAP findings in that nearly all of the Victorian sewers seen were sound. Newer sewers however built between the years 1920 and 1960 can have many defects. This could well be due to the lack of skilled workers after the two World Wars. As an ex-Resident Engineer, involved in a number of sewer rehabilitation schemes, I am also amazed at the number of what can only be called Ībodgingā repairs to defects. These include concrete patches to broken pipes (see Appendix 2, Case Study E for an example), pipes within pipes, and in one case a galvanised bucket used to make a connection!
Cause
There are many sources of sewer leakage. All are structural defects, including cracks, fractures, joint displacement, deformation, collapse, reverse gradients and unsealed connections. These defects are the result of one of more of the causes listed below.
The most common cause must be poor initial laying technique. Considering the lack of knowledge in ground engineering at the time, Victorian sewers demonstrate that good craftsmanship is often underestimated in these days of mechanisation. These sewers are however very dependent on the support of the surrounding ground. Poor backfilling can lead to leakage and eventually structural failure. Later sewers, particularly those sewers laid in the 1920ās to 1960ās often suffer from poor workmanship, and the engineering may also be suspect. Even modern sewers which are generally well engineered are sometimes poorly laid, the most common faults being failing to remove temporary laying supports (usually bricks) and poor connections (often simply a Īholeā knocked in the pipe).
Another very common cause is third party damage. Sewers suffer far more damage than other services as, unlike others that cause immediate disruption when damaged, Īsewers donāt biteā. Damage is often covered over without repair and forgotten until a bus disappears down the hole created! The mechanisms for these double decker bus holes (DDBās - so called as they are measured by the number of buses they could contain) are well explained in both the Sewerage Rehabilitation Manual and the Read and Vickridge book. All mechanisms follow a basic pattern of infiltration or infiltration/exfiltration cycles through holes gradually weakening the soil, fines being washed into the sewer and eventually voids being created. This mechanism is initially slow but, as debris is built up in the sewer channel, a swirl effect begins which washes progressively more and more soil and sewer fabric away, leading rapidly to collapse.
A third cause is service damage, which includes external damage from items such as leaking water mains. Brick sewers are very susceptible to wear, spalling and mortar loss, due to heavy flows, scouring and sewer gases (largely hydrogen sulphide). Pitchfibre pipes progressively lose structural strength. Concrete pipes sometimes scour, although this is not common, but do suffer greatly from hydrogen sulphide damage. Clay and plastic pipes are generally not susceptible to service damage if correctly engineered and laid. Recent research (John Reynolds et al, work for Southern Water, as yet unpublished) has however shown progressive failure of the rubber sealing rings on sewers laid in 1940ās and 1950ās. These worrying findings need further research as a matter of urgency, as this defect could affect a significant proportion of the sewer network.
The last cause of leakage I call operational damage. This category includes roots, siltation and blockage. Roots can cause structural damage as well as opening joints. Siltation can cause scour (although this is rare) but more often leads to blockage that can cause excessive pressure stressing the system and hence producing leakage. Damage is sometimes caused in maintaining the sewer network, particularly from the use of inappropriately high water pressure during jetting. This form of maintenance must I believe always be accompanied by CCTV inspection, albeit on a random sample basis.
In all cases leakage is usually seen as a detail of the problems associated with defective sewers. Most sewer collapses are however the result of infiltration/exfiltration cycles, so I believe that sewer leakage is the most significant item to be identified when inspecting sewers.
Cost Benefit
All work on assets, even just appraisals, will cost money. Owners of the assets will quite rightly ask those proposing works what are the costs and what are the benefits. In order to assess this, we need first to address the risks associated with leaking sewers.
Infiltration causes four main problems. The first and probably most apparent cause is unnecessary pumping and treatment costs. The second is the risk of structural collapse as the finer material from the surrounding soil is washed into the sewer by the ingress of the groundwater. This in turn reduces the support to the pipe, increasing stress locally, and causing failure. The finer material washed into the sewer itself creates the third problem by gradually silting the sewer network, including pumping stations and treatment works (increased pump wear). The last problem is the reduction in capacity of the network. All these problems can be costed for any given location, and by using net present value (npv) calculations, a true cost benefit can be shown (formulae, Sewerage Rehabilitation Manual).
The problems caused by exfiltration however are much more difficult to value, except in a very few cases. Where the sewer is by a groundwater abstraction source, the risk of contamination can be quantified, and costed (formulae, Reliability of sewers in environmentally sensitive areas, Report No 44 - various authors, CIRIA, 1996). At most locations however contamination of the soils is difficult to measure, let alone undertake a cost benefit analysis. Other risks associated with exfiltration, such as local hardening of pipe bedding producing uneven loadings and stress, and loss of support following infiltration/exfiltration cycles, are equally hard to quantify. However with groups such as the Campaign for the Renewal of Older Sewerage Systems (CROSS) applying political pressure, there may soon be changes in legislation causing asset holders to reassess their priorities.
All the above costs are often classified as Īdirectā, being those which directly affect the statutory undertakers profits. ĪIndirectā costs can however be significant. These include traffic congestion, damage to buildings, underground services, other roads used for diversions, loss of amenity and trade losses. Until very recently these were given a low economic priority by statutory undertakers. This situation is now changing under both legal and social pressure. The New Roads and Streetworks Act has given far greater powers to Highway Authorities. Recently they have started to exercise these powers and indeed in many areas local Inspectors are expected to cover their departments costs purely by the fines they levy! Traders and organisations are much more aware of the powers they have to recover any losses, and since privatisation, are keen to pursue Īrichā Water Companies. The greatest impact has however been in Public Relations. Poor P.R. can have a very marked effect on companyās share prices, and this factor is becoming increasingly important in cost benefit analysis. The impact of CROSS and similar groups on the P.R. of Water Companies may in the end be more significant than any changes in the law.
Identification (of leaking sewers)
Techniques
All identification techniques follow the same basic pattern of starting with Īthe macroā, generally a catchment, and ending with Īthe microā, sometimes down to an individual defect. ĪMacroā techniques all involve both flow measurement and statistical techniques. ĪMicroā techniques are varied, from flow measurement to CCTV survey to manual observation.
Table 1 (on page 10) shows a list of current techniques. The cost (£) per Population Equivalent (PE) I feel is a much more useful figure than cost per metre at appraisal stage. This is because when undertaking Īmacroā work the exact extent of the network may not be known, whereas population in any area can be obtained from census or Water Company billing data. Although The Sewerage Rehabilitation Manual and update CD ROMās give explanations of the techniques, I felt they may benefit from additional descriptions, as some of the practical disadvantages may not be apparent.
Statistical
This heading covers the full range from Dry Weather Flow (DWF) to modern stochastic computer models. For sewer leakage work all can only ever give a very Ībroad brushā picture, and thus should only be used for a general view. The use of rain hydrographs interposed on flow data can quickly show catchments where base flows are high, of course presuming that flow data is available. The big disadvantages therefore are cost and time, as most systems use deterministic models that require many flow and sewer surveys to be useful. They are however the best starting point for identifying problem areas.
The very latest computer programs, based on stochastic methods, do however show great promise. Accuracy of within 10% has been obtained from very little base data (flows at key points of a catchment). As these programs are at present experimental I will outline the philosophy of the system:
Flow Monitoring
Flow monitors (or meters) are only ever useful as part of a statistical modelling exercise, as their accuracy is insufficient to gauge minor variance between manholes. Data on water use, and hence discharge to sewers, is at present mostly statistical, and thus not accurate enough for micro studies. A few strategically placed monitors however can quickly include or exclude parts of a system from a study. When combined with network modelling a very accurate overall picture of a catchment can be obtained at a reasonable cost. The various types of monitor are listed in Table 1.
Dye Dilution
A system first used in the 1960ās, this relies on injecting a Īdyeā (nowadays often Rhodamine) into the system at a known dilution, and measuring the dilution rate downstream (formulae: Existing Sewer Evaluation and Rehabilitation: ASCE/WPCF). Obviously this can act as a form of monitoring where infiltration is suspected, but by adding a flow monitor at the point of sampling, exfiltration can be gauged. This second technique is currently under trial: it is too early to comment on its effectiveness.
CCTV
A very useful Īmicroā tool for infiltration, its use is very limited in exfiltration as it relies totally on the visual record. Whilst visually it is easy to identify defects which may leak, defective joints are almost impossible to spot. It is also almost impossible to survey the parts of the sewer under the flow, unless the effluent is very clear. However at present CCTV remains the best tool for micro studies. CCTV has also been combined with a sonar unit on partially surcharged sewers to give a Īcompleteā picture of the sewer both above and under flows (Insight Surveys - work for Severn Trent plc). Until recently a major limitation of CCTV has been in storing and reviewing videotapes. Images are now being encoded in MPEG format on CD ROMās, along with meterage data, so that searching for a particular section of sewer now takes seconds instead of tens of minutes. This format allows full linkage of location, manhole and defect data, and images from other systems (such as sonar).
Sonar
Sonar can give a fairly accurate picture of the profile of the pipe wall, and surrounding soil. The results however are very much open to interpretation, and a very skilled operator is vital. It can however show flow regimes under water, so can be used to spot infiltration into surcharged pipes.
Current Sonde
A very new system, developed in Germany, the sonde emits a current perpendicular to itself (and the pipe) which increases in current when the sonde passes a leak point. The main disadvantages are that the read-out (on computer) is difficult to interpret, and that there is no difference between a leak and a connection on the screen.
Ground Probing Radar
Until recently GPR was of little use as each type of soil responded to a different frequency of radar, hence selection of the correct frequency was critical. A new system, developed initially to trace plastic land mines, uses multi-frequency radar, and hence works in a variety of ground conditions. At present however this system has had very few trials on sewers, and its effectiveness is difficult to gauge.
Infrared Thermography
Carried out by aeroplane overflying the area, or specially equipped vans, good results have been obtained. The technique is however susceptible to environmental conditions (particularly rain!) and is prohibitively expensive for general use.
Dye and Smoke testing
A technique much used in America where records are reasonably accurate, it is of less use in European sewers as all connections must be known prior to tests (cellars full of smoke do not lead to good P.R.!).
Air Pressure tests
The standard test for new sewers, until very recently it was of little use on older sewers as it was almost impossible to seal laterals, etc., against air leakage. New robotic techniques have enabled the insertion of stoppers Īin sewerā. This has made air testing of older sewer feasible and cost effective, the limitation being that only the main sewer is tested and none of the laterals or manholes.
Water tests
The only testing system which gives 100% accurate results for both infiltration and exfiltration, this is unfortunately very labour intensive and disruptive. It is therefore important to target the lengths for testing by other methods.
Manual Surveys
These simply involve opening manholes in a progressive manner at periods of low flows (night) and noting any inexplicable clear flows (interim stoppers may be needed). Probably the most cost-effective method of tracing infiltration, its use in exfiltration studies is very limited.
Method |
I/E 1 |
Ma/ Mi 2 |
Advantages |
Disadvantages |
Cost 3 ö£/ PE |
Calc of min flow |
I/E |
Ma |
Quick where figures known, per capita contribution not required |
Based on assumptions, which may not be true for a location |
0.5 |
Calc of DWF |
I/E |
Ma |
Only needs inlet work monitoring |
Multi-catchment method, per capita estimating is difficult |
0.5 |
Map & Model ö deterministic |
I/E |
Ma |
Reasonably accurate figures for whole catchment - used with macro flow (monitor) surveys |
Needs accurate flow, sewer and manhole survey to give necessary accuracy, hence costly |
5 |
Map & Model ö stochastic |
I/E |
Ma |
Accuracy to within 10% claimed - used with macro flow (monitor) surveys |
Still under trials, so overall effectiveness difficult to gauge |
1 |
Monitor ö Īdrop-inā weir |
I/E |
Ma/Mi |
Reasonably accurate Īinstantaneousā system. |
Can cause blockage if left in sewer. Labour intensive. |
0.1-1 Ma/Mi |
Monitors ö Manning dipper |
I/E |
Ma/Mi |
Simple, cheap |
Unreliable, and accuracy inadequate for meaningful assessment of infiltration. |
0.1Ma/ 0.5Mi |
Monitor ö pressure transducer |
I/E |
Ma/Mi |
Widely used for open channel, accurate if used with gauged weir |
Difficult to set up accurately in sewers, with structural work (weirs) required |
0.3Ma/ 2Mi |
Monitor ö magnetic |
I/E |
Ma/Mi |
Fairly accurate for clean flows in open channels with velocities < 0.1m/s |
Probes easily fouled with grease, open channel only |
0.4Ma/ 3Mi |
Monitor ö Doppler |
I/E |
Ma/Mi |
Fairly accurate for open channels with velocities > 0.1m/s, and depths < 600mm |
Open channel only, will not work < 0.1m/s. Less reliable in very clear flows. |
0.6Ma/ 8Mi |
Monitor - pulse Doppler |
I/E |
Ma/Mi |
Very accurate for sewers > 600mm and flows > 200mm. Will accommodate flows in any direction (compound) |
Large diameters and deep flows only, costly to install |
0.1Ma/ 8Mi |
CCTV survey |
I/E 4 |
Mi |
Visual evidence of infiltration, hence very accurate |
Expensive, and no ability to detect leaking joints |
4 |
Sonar survey |
I/E |
Mi |
Gives evidence of pipe profile, including externally, and flow Īmovementsā |
Profile only recorded when stationary, hence slow and costly, needs very expert interpretation |
8 |
Current Sonde |
I/E |
Mi |
Very effective at checking Ītightnessā of pipe |
Difficult to distinguish between defects and laterals, very new and largely untried in UK. Relies on Īexpertā operators for interpretation. |
8? |
Ground Probing Radar |
I/E |
Mi |
Can profile Īvoidsā down to a depth of 4m |
Relatively untested and expensive for I/E alone |
15 |
Infrared Thermography |
E |
Ma |
Aerial surveys can locate unknown leaks |
Very expensive, and sensitive to environmental conditions |
? |
Dye & smoke testing |
I/E |
Mi |
Simple and accurate |
Lengths almost need to be identified first, flow diversion required |
3 |
Dye Dilution |
I/E |
Ma/Mi |
Simple and reasonably accurate |
ĪDyesā suitable for heavy flows (Rhodamine, isotopes) can be environmentally sensitive |
2? |
Manual Survey |
I |
Ma/Mi |
Suprisingly efficient with experienced staff, and quick |
Rough estimate only: needs follow up (CCTV, Sonar, Sonde, Water test) |
1 |
Water test |
I/E |
Mi |
Very accurate |
Flow diversion required, hence very costly and disruptive |
5 |
Air Pressure test |
I/E |
Mi |
Accurate for Īperfectā pipes |
Flow diversion required, pipes which Īfailā may not leak! |
3 |
For description of techniques, please consult The Sewerage Rehabilitation Manual , 1 Infiltration/Exfiltration applicable
and associated update CDās. 2 Macro/Micro applicable
3
Where known: date reference for prices is August 19984
Only useful for exfiltration from defectsAll costs include for Īreasonableā ancillary works, although not major cleaning, etc.
Table 1 - Identification Techniques
Method
Most techniques need to be combined to give the best results. After the work in the late 1970ās and early 1980ās by WRc and others, most Water Companies have implemented Drainage Area Studies (DASās) and Drainage Area Plans (DAPās). The DAS is a combination of desk study, gathering all records of the system and collating, ground surveys, such as manhole and flow surveys, and CCTV. In the DAP the data from the DAS will be analysed and the system graded. A full description of the grading classification system can be found in the Sewerage Rehabilitation Manual: note that CCTV operators should be trained in the identification and classification of defects. After further surveys (CCTV, etc.) recommendations can be made based on the survey reports and computer modelling. Strategies for DASās and DAPās can be found in the Sewerage Rehabilitation Manual. A common strategy, much used throughout the world, would be:
Many other options and combinations are possible and three case studies are listed in Appendix 1. My own preferred methods are shown in Flowchart 1, and below is a possible option using stochastic computer models, which is considerably cheaper than the method shown above.
This system has two fewer steps and eliminates the need for an expensive manhole and sewer survey (see Page 8 for a description of stochastic computer programs).
Cures
Techniques
Choice of technique depends on the nature of the problem, but in most locations more than one type may be required. Table 2 (page 14), which shows a list of current systems, gives my own amended list of the techniques being used. The cost (£) per metre is derived both from CIRIA and WRc figures, and my own data. Once again, although The Sewerage Rehabilitation Manual and update CD ROMās give explanations of the techniques, I felt they may benefit from additional descriptions, as some of the practical disadvantages may not be apparent. I have also included some additional costing data in Tables 3-5. Note that all lining systems use robotics to recut laterals, and the costs for these are included.
Chemical grouting
Although an effective Īcureā, there is a risk of pollution incidents as the chemicals are not inert. The sewers must also be free of any structural defects above Grade 2 (grading of sewer defects, Sewerage Rehabilitation Manual).
Stabilisation - sodium silicate
Only one system is currently available (Sanipor). Up to Grade 3 it is very effective and although seemingly costly compared to linings it is the only system that seals leaks in sewer, laterals and manholes in one Ītreatmentā. The chemicals are also inert, which can be a distinct advantage in old sewers where cross-connections to storm sewers can be very common!
Cementitous and resin injection
These systems either use Īpigā devices that seal either side of the leakage point and then pressure inject the
sealer, or use robotic units to inject at the leakage source (Amcrete, Seerseal, etc.). An effective technique for point source problems, but can get very costly if used at more than a few locations. There is also a risk of structurally damaging the pipe (see Case Study D).
Pipe bursting
An effective technique to maintain or upsize the existing pipe, it can fail if there are obstructions surrounding the pipe (see Case Study E). Care must also be taken to avoid Īheaveā damage to surrounding structures.
Soft lining - continuous
Insituform is the only softlining system currently operating in Britain, and it is effective but sometimes more costly than other linings. Care must also be taken to ensure the chemicals used to cure the liner do not pollute watercourses.
ĪUā lining - continuous
Most ĪUā liners (all derivatives of the gas industry rolldown techniques) are heated to soften (by water or steam), pressurised to form the sewer shape, and then cooled to form a hard lining. One reported benefit is the ability to reheat and remove the liner, however I have never seen this effectively carried out. Some new ĪUā liners are using UV radiation to cure. A very cost-effective technique provided the defects are not excessive and the laterals and manholes are not leaking excessively.
ĪUā lining - patches
An effective and cheap way of dealing with isolated defects and leakage, but overall price is proportional to the number of repairs required so careful costing must be undertaken.
Spiral winding
Due to problems with the joints on the windings this system has largely been superseded.
Hard lining
One main disadvantage is the large lead-in trench required. If sections (SnapLok or similar) are used, these can be avoided, but then an operator is required in a manhole, with consequent confined spaces procedures. Care must also be taken to avoid flotation of the liner if grouted.
Man entry - pointing, segments and in situ coatings
As Health and Safety legislation (quite rightly) becomes more onerous, these techniques are becoming increasingly costly. Safe systems of work do however exist, and with an experienced team these techniques can still be the best for larger sewers. The systems used include in situ and precast gunite, ferrocement, precast concrete and GRP segments, etc.
Local repairs - excavation
Often used for isolated defects as direct costs for these can be quickly assessed, the trend of transferring indirect costs to the statutory undertakers is starting to make this option less attractive.
Relaying
Again due to the trend of transferring indirect costs is making this option less attractive, although of course relaying is the best option in overall Īlife costsā terms.
Tunnels
Tunnels are notoriously difficult to cost, can have serious health and safety risks and are generally far more expensive than other systems. It is therefore better to consider them as a Īlast resortā if no other option is viable.
Micro-tunnelling
The latest micro-tunneller can install down to 300mm pipes, but the systems do have large set up costs. Micro-tunnelling is therefore most effective when there are many metres of sewer to be replaced within a proximity, particularly if the sewers are deep. Care must be taken however if groundwater is present as gauging the correct balancing pressure in a sealed head tunnel boring machine (TBM) is very difficult.
There are good guides and formulae in the Sewerage Rehabilitation Manual, and other publications listed, that can help select materials for any location. However my own works have shown that correct installation is far more critical than minor differences between similar materials. This is particularly true of linings where failure to cure or fix can lead to early failure. I would therefore recommend that, having chosen the appropriate type of cure for a location, the material selection is left to the contractor. A tight specification should however be provided in all contracts.
Please note that this (and the previous) section is meant as a general guide only. Anybody wishing to undertake these works would be advised to consult The Sewerage Rehabilitation Manual, and other specialist literature. This is particularly true where tunnels are being considered, as the complexities of the geotechniques involved require very specialist advice.
I will also place a Īword of cautionā in this section. Sewers to be rehabilitated are generally chosen as the result of a DAS and DAP. The survey results will determine the location and choice of technique. However it is all too easy for details of the survey to be entered incorrectly. Manhole numbers can be transposed, chainages recorded incorrectly, sewer details such as size and material can be wrong, etc.
I would therefore recommend a second survey of lengths proposed for rehabilitation. It can be rather embarrassing to reline or replace a perfectly sound section of sewer! It is far better to find errors before expensive constructional work commences.
Method |
Size range (mm) |
Advantages |
Disadvantages |
Cost 1 - £/m 2 |
Chemical grouting |
<625 |
100% effective, seals laterals as well as main pipe. |
No structural strength (i.e. sewer must have no structural defects). Overpumping required. |
70 - 160 |
Stabilisation - sodium silicate |
<625 |
100% effective, seals laterals as well as main pipe. Gives structural strength. |
Will only Īfixā up to grade 3 defects. Overpumping required. |
70 - 200 |
Cementitous injection |
<475 |
95% effective. Only seals where needed. |
Does not seal laterals. Can burst pipe. Slow if many defects. |
400 each 3 |
Resin (epoxy) injection |
>100 |
95% effective. Only seals where needed. |
Does not seal laterals. Slow if many defects. |
500 each 3 |
Pipe bursting - continuous |
>125 |
Almost 100% Can enlarge pipe. |
Costly if many laterals, as each must be open cut re-laid. Potential for head to get stuck (see case study E). |
120 - 1000 |
Pipe bursting - sections |
150 - 450 |
Almost 100% Can enlarge pipe. |
Costly if many laterals, as each must be open cut re-laid. Potential for head to get stuck (see case study E). |
200 - 350 |
Soft lining - continuous |
>200 |
80% unless additional sealing carried out on cut connections |
May require overpumping. Requires fairly substantial setups at manholes. |
100 - 500 |
ĪUā lining - continuous |
>75 |
80% unless additional sealing carried out on cut connections |
May require overpumping. Can fail to adhere (and hence seal) if infiltration heavy. |
50 - 500 |
ĪUā lining - patches |
100 - 450 |
100% over patch area |
May require overpumping. Can fail to adhere (and hence seal) if infiltration heavy. |
700 each 3 |
Spiral winding |
150-900 |
80% unless additional sealing carried out on cut connections |
May require overpumping. May require lead-in trench. |
70 - 450 |
Hard lining - close fit |
150 ö 900 |
80% unless additional sealing carried out on cut connections |
May require overpumping. Requires lead in trench. |
60 - 500 |
Hard lining - grouted |
150 ö 900 |
80% unless additional sealing carried out on cut connections |
May require overpumping. Requires lead in trench.. Liner can Īfloatā on grout |
80 - 500 |
Hard lining - grouted - sections |
150 ö 450 |
80% unless additional sealing carried out on cut connections |
May require overpumping. |
60 - 250 |
Man entry - segments |
>900 |
100% if connections resealed. |
May require overpumping. Costly, and requires confined spaces entry |
200 - 500 |
Man entry - in situ coating |
>900 |
100% if connections resealed. |
May require overpumping. Costly, and requires confined spaces entry. Infiltration must be stopped whilst coat cures |
450 |
Pointing |
>900 |
Up to 100%, simple |
Confined space entry required. Costly |
Per ? |
Local repairs - excavation |
Any |
Up to 100% |
Costly, disruptive. May require overpumping. |
1000+ each |
Relaying |
Any |
100% effective |
Very costly and disruptive |
250-2000 |
Tunnels |
Large |
100% effective |
Very costly and difficult |
?? |
Micro-tunnelling |
300-1200 |
100% effective |
Cost entirely dependent on type of ground and water table |
?? |
For description of techniques, please consult The Sewerage Rehabilitation Manual and associated update CDās.
All costs include for Īreasonableā ancillary works, although not major cleaning, etc.
Table 2 - ĪCuresā
Table 2 gives a good estimate of Īnormalā prices for rehabilitation works, but some will be outside the scope of this list. Tables 3 to 5 give some particular examples of costings that have been derived from works undertaken by my company.
Dia (mm) |
Rate (£/m) (+/- 20%) |
Dia (mm) |
Rate (£/m) (+/- 20%) |
Dia (mm) |
Rate (£/m) (+/- 20%) |
225 |
246 |
600 |
360 |
1200 |
830 |
300 |
250 |
750 |
425 |
1350 |
1020 |
375 |
260 |
825 |
480 |
1500 |
1320 |
450 |
280 |
900 |
550 |
1800 |
1980 |
525 |
320 |
1050 |
640 |
Table 3 - Cost of relaying in urban areas
Table 3 gives costs for relaying in urban areas. These include for all the necessary liaisons, traffic control and P.R. that this type of work requires.
Dia (mm) |
Mobilisation (£ +/- 20%) |
Lining/m (£ +/- 20%) |
Connections ( £+/- 20%) |
225 |
3,000 |
28 |
80 |
300 |
3,000 |
38 |
80 |
450 |
3,800 |
80 |
80 |
575 |
4,500 |
100 |
90 |
750 |
5,000 |
250 |
90 |
Table 4 - Cost of soft linings
Table 4 has typical costings for soft linings over various diameters.
Method |
Cost (each - £ +/- 20%) |
Patch lining |
3,500 |
Resin injection |
5,700 |
New manhole |
8,000 |
Relay short length (approx. 3m) |
10,000 |
Table 5 - Cost of Īone-offā repairs (costs include mobilisation, administration, etc.)
Table 5 is a comparison of costs for Īone-offā repairs in urban areas.
Ancillary Works
These would include cleaning and overpumping, and typical costs for these have been included in the prices shown in Table 2. If flows or siltation are excessive then extra may need to be added, but such items can only be assessed on an individual basis. Overpumping requires analysis of the flows to gauge the size and type on pump required. If close to houses the pump(s) need to be acoustically shielded. Thought must be given to the need for standbys. These will be required if the rehabilitation process cannot be interrupted once started, and the sewer is unable to store the flows for the time required. Storm flows must be considered in all calculations. Sewer cleaning is now almost exclusively carried out by jetting as manual cleaning requires confined spaces procedures, and dragging is labour intensive and risks damage to the sewer fabric. To minimise damage I would recommend the Īhigh flow/low pressureā method of jetting. Work should comply with the ĪSewer Jetting Code of Practice (WRc 1997)ā. Disposal of material moved must comply with "Duty of Care" imposed by the Environmental Protection Act 1991.
Method
Method(s) applied will depend on:
a. Size of the sewer.
b. Location (under structure, in road, class of road, urban/rural, etc.).
c. Flow.
d. Defect(s).
e. Nature of sewage.
f. Percentage of sewer capacity already used.
The size of the sewer will determine whether remote techniques or man-entry is more appropriate. Location will determine whether excavation is possible, or no-dig techniques must be used. If flows are heavy this may lead to overpumping problems, and also limit the choice of techniques as a Īquickā system, or overnight working (and hence quiet systems) may be needed. The type of defect(s) will determine whether rehabilitation or replacement is appropriate. The nature of the sewage must be known in order to choose the most appropriate materials (e.g. concrete pipes would not be appropriate if the sewage were septic). Lastly the percentage of sewer capacity already used will determine whether linings could be used (i.e. reduction in capacity) or some form of replacement is required.
My own recommended strategies are shown on Flowchart 1.
Contract Strategy
There has been much discussion as to the best form of contract to use. ICE 6th or ICE Minor Works 2 with a Bill of Quantities is much favoured as they give more surety of price, provided items have been included in the tender. The IChemE ĪRed bookā has been used with some success for rehabilitation works, particularly linings, as the lump sum process approach can suit this type of work. There are significant risks however if Īunknownsā are encountered. The cost reimbursable IChemE ĪGreen bookā is very useful where the scope of work is largely unknown. The Engineering and Construction Contract (ECC - formerly NEC) with its various option however probably is the best vehicle for this type of work. The Gainshare Target Cost options C and D are very useful as they provide both client and contractor with the opportunity to make savings resulting from efficiencies. As a Contracts Engineer however I feel that the choice of contractor is far more significant as much of the work remains Īvagueā until commenced, and so the opportunities for claims are great. A co-operative contractor who wishes to enjoy a continued working relationship with a client is much more likely to look for solutions rather the claims! It is also very important that the client (and engineer/project manager working for him) ensure that everything possible is done to Īsmooth the contractors pathā. This would include ensuring liaisons and notices with third parties (HAās, local traders, etc.) are completed prior to works commencing.
Recommendations
In undertaking this project I have been struck by the similarity of all publications on the subject of sewer rehabilitation. This is however not surprising as there is consensus on best practice throughout the industry: my own work also leads me to the same conclusions. Undoubtedly the definitive publication is the Sewerage Rehabilitation Manual, although all the other publications listed at the end of this paper have considerable merit. Such is the rate of progress within the rehabilitation industry however that some of the techniques and systems listed in older publication are no longer being used. I would therefore recommend that anyone undertaking this type of work keep up to date with methods by reading trade publications and attending trade fairs such as the ĪNo-Digā shows.
Where this paper differs is in approaching the subject of sewer rehabilitation from the perspective of leaking sewers. I have therefore split my recommendations into three sections, infiltration, exfiltration and Īthe futureā.
Infiltration
I have shown in this paper the wide range of techniques both for finding and curing infiltration. From both practical work, and experimentation, I feel there is a definite path to be taken in all cases. I have set out my recommendations in Flowchart 1 on page 18.
The logic behind the steps listed in my proposals in Flowchart 1 are:
Flowchart 1 - Method selection
Exfiltration
Exfiltration, whilst as Īeasyā to cure as infiltration, is far harder to identify. My present recommendation is to first identify possible problem areas by statistical methods, and then to monitor down to progressively smaller catchments. This is despite my reservations about the accuracy of monitoring. To identify actual lengths that are leaking, the best method seems to be water tests, although the current sonde system is showing promise. At present however exfiltration work is generally ruled out on cost benefit grounds, and seems likely to remain this way unless legislation, or public pressure, forces changes on the industry.
The Future
Leaking sewers seem likely to become a significant part of the Water Industry Regulatorās (OFWAT) AMP 4 and ĪKā review. ĪKā is the amount above, or below, inflation that water rates can be charged and is set every 5 years by the regulator. This is based on the Īworkā he believes the water company needs to do to maintain its assets and is set by the water companies Asset Management Plan (AMP). I believe that both statutory undertaker and regulator will make far greater use of computer modelling over the next few years. The accuracy of the computer models will be enhanced both by the improvements in computer software, and the increased knowledge of water consumption gained through the far greater use of water meters on domestic properties. This in turn is likely to lead the regulator towards the German system of Īcatchment provingā, where the German water companies have to prove that sewers are not leaking in a particular catchment. ĪProvingā catchments in Britain will be a mammoth undertaking, and will also need an advance in exfiltration tracing techniques.
Identification techniques are thus likely to be improved significantly. Whilst new systems may appear, the most likely developments will be in the combining of current systems. A combination of CCTV, current sonde and GPR would be a powerful tool, provided it was not prohibitively expensive! There will also be advances in cure techniques, and Īno-digā will become Īthe normā. Robotic systems will be improved, particularly for Īspotā repairs. The non-mechanical systems (such as chemical treatments) will be developed and used far more widely as they remove the risk of excavation to recover robotics and failed liners. Micro-tunnelling will be used far more in preference to excavation as indirect costs will be transferred to the sewer owner.
I believe however that the most significant factor will be the public perception of sewer leakage. The focus of the media has up to now largely been on water pipe leakage, but this is likely to turn onto sewers. Water companies and other owners will thus make sewer leakage a major factor in their forward planning.
Appendix 1 - Identification Case Studies
Case Study A - Wilton Infiltration Studies
Wilton near Salisbury, Wiltshire, suffered a long history of infiltration, the system being fully surcharged in winter. Written records of this problem date back to 1947.
Two studies were carried out jointly by Wessex Water and WRc in 1980 and 1983. This was a traditional Map and Model (deterministic) backed by strategic monitoring. 10 monitors were used, mostly of the magnetic Īringā type, although some manual measurements were made, and 3 tipping bucket rainguages. A report on the second survey was produced by WRc (Sewage flow measurement - Wilton, A D Thorpe).
As someone involved in the second survey, I was well placed to examine the relative merits of the methods used. The Map and Model was reasonable, if labour intensive. There were many problems with the reliability of the monitors, and some errors in the summation of the data. The most useful work was the manual night surveys. This consisted of opening manholes in a progressive manner, working downstream. If flows were too high, interim stoppers were used. Any lengths where significant increase, or largely clear water, were noted, and then CCTV surveyed. In this manner, over 4 nights, 20 l/s of infiltration was identified. This contrasted with the Īmicroā flow monitoring work which tended only to give a Īmacroā picture. A scheme to relay 4 lengths of sewer, approximately 200m, was instigated, which eliminated the infiltration found.
Further infiltration was found, but this was over a much larger area. Much related to leaking joints, and laterals also were found to be defective. Connections to the river were suspected, but never traced. Due however to the quantity of defective lengths identified, and the traffic sensitivity of the area, further constructional work was ruled out at that time on cost benefit grounds. A full DAP was however undertaken (Wilton Drainage Area Plan, Wessex Water).
Case Study B - Martinstown
The heavy pumping figures gave initial indications of problems in this area thus identification was by a Īvariantā of minimum flow calculation. As a result of this it was decided to fully CCTV the whole catchment.
Fresh from the work at Wilton, the team sent, which included myself, decided to first carry out a manual survey. This proved very advantageous as much of the infiltration was found from two sources. The first of these was a small section of damaged pipe (about Grade 4 - see the Sewerage Rehabilitation Manual for a description of the grading system). The second was an unsealed ĪMarleyā manhole on a private development. These two defects accounted to approximately 65% of all the estimated infiltration in the catchment, so both were corrected. Further lengths were identified as suspect in the manual survey, and all these were CCTVād. However, as the defects found were mostly leaking joints, on balance of cost no further work was carried out at that time (mid 1980ās). There is a current investigation on this catchment utilising manual surveys, water tests and CCTV.
Case Study C - Bransgore
On the edge of the New Forest, Bransgore was long known to suffer from surcharging in the winter months. Pumping figures were about six timeās minimum flow. I was part of the team sent to do a full study including CCTV. Fortunately this was undertaken in the winter months when problems are easier to spot and again we started with a manual survey. A great deal of Īclearā water was found in the lower reaches of the system. The sewers were then CCTVād and a number of defective joints were found. The amount of infiltration identified though was not enough to account for the excess flow or surcharge.
The sewers however were also showing signs of surcharge in the upper parts of the system. These parts were far too shallow for any infiltration. It was only in discussions with local residents that the large part of the Īproblemā was discovered. The catchment was largely new build on the side of small hill. The ground was fairly impervious, but little provision for land drainage had been made on the boundaries of the site. In heavy rain therefore water was flowing across fields and into private gardens. In order to prevent flooding in their properties, residents were opening foul sewer manholes to dissipate the water! This was leading to surcharge in storm conditions. New land drainage was installed around the catchment, and the defective sewers were relined.
Appendix 2 - Cure Case Studies
Case Study D - Bourne Valley
The Bourne valley has suffered from infiltration for many years. It is currently subject to a study. In 1994 however it was decided to trial a new technique in the top village in the catchment, Newton Tony. As a resident in the village I was not surprisingly Īco-optedā into the team!
The primary cause of leakage in the system was radial cracks at the centre of many pipes. The clayware pipes, laid early in the 1970ās, were from an experimental batch produced at 2m long rather than the usual 1.65m. It was believed that the pipes were unduly stressed during transport as they were only supported at two points. The technique chosen for the repair was cementitous injection, namely Amcrete. This worked reasonably well until one pipe that had, due to the excessive infiltration, lost its bedding. The pressure from the Īpigā burst the pipe and a very large quantity of Amcrete was then required to make good. The process was effective but laborious. I also have reservations about the long-term effectiveness of this solution as only those pipes found defective at that time were Īfixedā. If the remaining pipes are Īstressedā than further cracks are likely in the future.
Case Study E - Salisbury sewers
Although not a Īleakageā project, I felt one aspect of this work was worth including due to its salutary nature.
A section of pipe was found to have Grade 5 defects. This pipe ran under a dual carriageway, but was also near to capacity. It was therefore decided to use pipe bursting. The particular system chosen was of the driven type, utilising a guided steel tube, welded in sections to burst open the old sewer. The HDPE lining is then inserted in the oversized hole created. Erring on the side of caution, trial holes were dug either side of the dual carriageway to determine the pipe surround: in both cases this was soil.
Initially the burst went well, but after halfway progress was very slow. After a few hours we decided to halt the exercise as we were concerned that an obstruction may have been encountered. A short heading was made and unfortunately the pipe was found to be surrounded in concrete.
Quite why a sewer with Grade 5 defects was exposed prior to the building of the road, and then surrounded in concrete without repair, no one is sure. We were however then left with a very difficult problem. Picture 6 shows the only practical solution, a heading! The eventual cost of this repair rose from £7,000 to £25,000. I would therefore recommend that pipe bursting is used only where there is no practical alternative
Case Study F - Wilton Infiltration Remedials
In 1994 we returned to Wilton to undertake Īcuresā for the main town centre. I was fortunate to be Resident Engineer. As the investigations found defects to be numerous, and connections were also suspected, it was decided to use stabilisation with sodium silicate, namely Sanipor. This was the first time this technique had been used on such a complex network, so the whole project was in the nature of an experiment. On some sewers however, soft lining was carried out, and some Grade 4 and 5 defects were repaired by spot excavation.
The combination of techniques proved to be a great success. The āSaniporā system proved not only to be effective in curing leaks, it was also very helpful in finding rogue connections, as the method uses water tests. Indeed by this method we discovered, and then capped, the long suspected connections to the rivers. Picture 7 shows the second chemical being added.
With so many very sensitive rivers nearby there were many concerns about pollution. The Sanipor chemicals are nominally inert, but there always remained a suspicion this may not prove totally true. In the event a large Koi carp pond was contaminated via an unknown storm sewer next to the foul being treated. This storm sewer in turn fed the carp pond. Whilst the contamination looked unsightly, and we removed the fish whilst cleaning, no fish were in fact harmed.
Appendix 3 - References
Title |
Author |
Publisher |
Date |
The Sewerage Rehabilitation Manual (and associated update CD ROMās) |
Various authors |
WRc publications |
1984 onwards |
Reliability of sewers in environmentally sensitive areas |
Report No 44 - various authors |
CIRIA |
1996 |
Control of infiltration to sewers |
Report No 175 - various authors |
CIRIA |
1997 |
Sewers - Rehabilitation and New Construction |
Geoffry Read and Ian Vickridge |
Wiley |
1997 |
Existing Sewer Evaluation and Rehabilitation |
Various authors |
ASCE (New York) and WPCF |
1983 |
Sewage flow measurement - Wilton |
A D Thorpe |
WRc publications |
1982 |
Water and Environmental Management magazine |
Various articles |
||
Sewerage Renovation Federation |
Various articles |
||
The Drain Trader magazine |
Various articles |
||
Water Services magazine |
Various articles |
I have also researched many other publications, which are too numerous to list. I am fortunate to have access to a good technical library, but can recommend the ICE library in London as a good source. This and many other sources are best contacted on the internet.
There you will find lots more information, including the Septic Tank page, Sewers of London history and a link to the notorious Thomas Crapper page.