47 Can we build sewerage systems that don’t leak?

Background

In Episode 26 and Episode 39 I discussed the problems of infiltration into sewers (and I also swept under the carpet the problem of exfiltration out of sewers).?But how bad is the problem, how does it arise and what can we do to prevent it?

In short, can we build sewerage systems that don’t leak - in either direction?

As I set out in Episode 26, infiltration into sewers causes a lot of additional costs and a lot of service impacts.?Of particular concern at present in the UK is its contribution to the discharge of untreated sewage over storm overflows.?This problem can occur both on combined sewerage systems, which were intended to have storm overflows and on separate foul systems that were not, but where infiltration flows now require them.

The problem is particularly bad on the porous chalk catchments where in winter the groundwater table can rise above the sewers causing large long-term inflows.?This can lead to storm overflows operating continuously for weeks or even months.

Previously the Environment Agency has allowed temporary discharges from these systems to alleviate flooding, but they are now reluctant to allow these as they argue that sewerage systems that suffer from excessive infiltration are not fit for purpose under the Urban Wastewater Treatment Directive.?A recent article on infiltration reduction in Norway repeated this often-held view that, “It is widely acknowledged that most I&I (Infiltration & Inflow) is caused by ageing infrastructure that requires maintenance or replacement.”?However, the definitive WRC Sewerage Rehabilitation Manual challenges this stating “Infiltration often enters pipelines through joints or lateral pipes, many of which may appear to be structurally sound.?During periods of low groundwater table, such joints would appear perfectly adequate, the lack of watertightness not being obvious in CCTV surveys”.

Design of new sewers in the UK (based on Ofwat’s mandated document Sewers for Adoption now replaced by the Sewerage Sector Guidance Document) accepts that sewers will experience groundwater infiltration, with a standard design allowance of 10% of the peak flow rate. ?For foul sewers (where the design peak flow is usually six times average Dry Weather Flow (DWF), that allows for infiltration equal to 60% of design residential DWF.?

It appears to be common for recently built housing developments, where the sewers have been built and tested in line with standard practice, to suffer infiltration much greater than the 60% of DWF design allowance.?Two examples are shown in the table below.

So how does this happen?

Testing sewers

The standard for testing gravity sewers is BS EN 1610 “construction and testing of Drains and Sewers”.?The focus is on testing for exfiltration with an air pressure test or a water pressure test but less focus on testing for infiltration with only a simple measurement of current inflow.

Air pressure test

This test is carried out on the sewers and the laterals up to the boundary demarcation chamber. ?There is no requirement for a pressure test for drains within the curtilage.?The air in the sewers is pressurised to 100 mm of water (0.01 bar) and needs to drop to no less than 75 mm over 5 minutes.?Bearing in mind that either internal surcharge pressures or external groundwater pressures could be several metres of water, this is not a particularly informative test.

Water test for gravity sewers

If the sewer fails the air pressure test (or if optionally required) then a water pressure test is carried out.?This pressurises the sewer to a minimum of 1.2 m head of water and the loss of water measured over 30 minutes.?The loss needs to be no more than 0.5 litres per linear metre of sewer per metre of nominal diameter.

For a typical length of 150 mm diameter sewer in a housing development plus the length of the lateral this would be 72 l/d compared to daily DWF of about 350 l/day.?This is 20% of DWF.

Infiltration test

The third test is to measure the flow out of the new sewers coming from infiltration. ?This has the same permitted flow rate as the water pressure test.?The permitted infiltration for a brand new sewer is therefore already a third of the long term design allowance.?I am not sufficiently familiar with testing practice to know whether this part of the testing includes the drains within the property curtilage.?

There is no guarantee that the test is carried out under typical, never mind worst, groundwater conditions.?In fact, if it is carried out as soon as the sewer is complete it is likely to have very low groundwater levels as the excavation would have been dewatered to allow the sewer to be laid and then backfilled with selected dry fill.?

What is going wrong

So UK practice is to test thoroughly for risk of exfiltration and have poor testing for infiltration.?But aren’t the two risks equivalent?

For sewer defects such as holes and cracks, the risk of infiltration and exfiltration flows would be roughly the same given the same pressure difference in the two different directions.?However this seems to not be the case for leaking joints.

Work by Professor Richard Fenner has shown that just slight misalignment or damage to the flexible joints used between sections of pipe can create a “one-way valve” that allows infiltration into the sewer but does not allow leakage out.

This has also been explained for plastic pipes by Phillip Clisham who has kindly provided the diagrams below.?These show the mechanism that internal pressure will tend to close the joint so that there is little or no exfiltration.?However external pressure will tend to open the joint allowing infiltration.?

There is a similar effect on clay pipes although the mechanism is less obvious.

The future

We are currently building sewerage systems that the EA says are not fit for purpose as they allow too much infiltration.?What can we do to change.

In other jurisdictions including the USA and Australia it is common to carry out air vacuum testing of sewerage systems to test for the risk of infiltration.?This testing often includes the manholes as well as the sewer pipes themselves.

The tests typically use a vacuum of 0.33 bar with a loss of vacuum of no more than 10% of this over two minutes.?his test does not require a high groundwater table to show infiltration and so is more reliable in identifying potential problems.?However, the test may still not be rigorous enough to ensure low infiltration rates.

So the steps to fix this:

1. Require vacuum testing of all new sewers
2. If the sewers fail the tests, then improve pipelaying standards.?Other jurisdictions seem to be able to pass so we can too.
3. Monitor long term infiltration on sewers that pass the new test.?If there are still problems, then make the test more rigorous.
4. If necessary change to jointless construction methods in areas with a risk of high groundwater levels.