This is an archival version of the original KnowledgePoint website.

Interactive features have been disabled and some pages and links have been removed.

Visit the new KnowledgePoint website at


Pipe material for developing a spring water source


We are developing a spring water source to supply a small community in Kenya.  The distance is 6000metres and height of 480 metres. How do I determine the proper pipe material to use to handle water pressure and the pipes will be laid over the ground surface in forest land?


Many thanks for this advice. Its made decision making precise for a successful project. Thanks you RedR!

RedR TSS gravatar imageRedR TSS ( 2013-11-11 10:04:05 )

9 Answers


I recommend this book and look at WaterAid note attached


There are quite a few factors to consider


Pipes should be buried. If at surface, iron pipes are recommended because they are stronger and unlike PVC do not deteriorate in sunlight, but using GI pipe would be ridiculously expensive except for short gulley crossings, etc.


The size of pipe depends on the volume water required at times of peak demand. If given a schematic of the pipeline with altitudes and flows a friendly pipe manufacturer may advice you on pipe size. They certainly will have head loss charts for different pipe sizes that will enable you to do this. Otherwise input the data into an online tool like this one


Consider need for break pressure boxes (to avoid pressure exceeding pipe limits) , placement of outlets at high points to break air pockets and the placement of storage tanks (which will also act as a break pressure tank)


Good luck


Ian Neal


There is quite a bit of information about piped water supply schemes available from the World Bank Water and Sanitation website, such as:

I seem to remember a manual they produced about gravity water supply schemes in Malawi, but this does not seem to be available any more. There is also quite a lot of information about spring protection and piped schemes elsewhere on the internet.

Regarding the pressure and material issues there are a number of factors to consider besides the pressure class of the pipe. Generally it would be better to control pressures in the system as this will reduce losses and with a large head difference there is the possibility of creating substantial negative pressures in the pipe, for instance if there is a burst at the lower end. I visited Tuum in Kenya many years ago and the water points higher up the mountain only admitted air during the day because of the high demand lower down. My advice would therefore be to install one or more break pressure tanks to reduce the pressure in the system so that it is no more than necessary to provide sufficient head at all standpipes and connections. Depending on the complexity of the distribution system some simple modelling may be required and there is free software available for this from the US EPA:

Laying pipes on the surface is not generally recommended, plastic pipes degrade and are easily damaged, steel pipe is better but expensive and difficult to lay and the Outward Bound School in Chimanimani, Zimbabwe had their entire water supply pipe stolen during the night!




Pipe material depends on what is available.

As far as pressure is concerned I would just put in a number of break pressure tanks (BPT) to keep pressure within the limit of the available pipe.  As you are probably aware 480m pressure in the village will blow the taps apart, so probably best to keep the pressure in any tap down to a maximum of about 30m/3 bar pressure. 


Pipe over ground will need restrained joints or pipes that have end restraint.  Building restraints over movable pipe joints (e.g. standard push fit spigot/socket) would not be impossible but may be impractical.  The brief description suggests either steel pipe with threaded joints, or polyethylene (PE) pipe with butt fusion joints, electrofusion joints or, if pipe is small enough (I suggest 90mm diameter and below), then compression joints.   Note for PE pipe you can use a combination of the jointing methods.


As long as it is readily available in country I would probably start by looking at PE pipe. Some observations:

·         PE pipe will be able to flex to the ground surface so less work to get it flat.  You will still need to make sure you avoid air traps.  If you get these then just install a small 25mm plastic air valve in the high spot and a washout valve in the low spot (these would need to be regularly checked).

·         Black coloured pipe usually has good resistance to UV light (50+ years, check with supplier for resistance in Kenya). 

·         PE pipe can have pressure rating up to 16 bar (about 160m), there are also lighter weight pipes.  So within a pipe run (between pressure break tanks) you could vary the pressure rating of the pipe to suit.  This will save weight and pipe cost.

·         So using PE pipe with maximum 16 bar pressure you could have something like 150m below spring have BPT, next 150m down have second BPT, third 150m down have main store tank leaving 30m of pressure to feed into community taps.  You would need to do the friction calcs to work out the details.

·         Within each 150m height drop you could have top third from SDR21 or 26 (max 6 or 8 bar) PE pipe, middle third of SDR17 (max 10 bar) PE pipe and bottom third of SDR11 (max 16 bar) pipe.  On smaller diameter pipes you may only have the choice between SDR11 and 17 pipe.

·         PE pipe usually has additional short term pressure resistance to withstand occasional water hammer pressures.  If in doubt put in pressure dampers or pressure release valves to deal with the water hammer pressures.

·         Pipe can be fixed in place with posts or straps bolted to rocks.  However PE pipe will expand and contract so you need to build in areas where there is flexibility (usually around sweeping bends) to allow the PE pipe to move. 

·         You will also need to make sure that movement areas do not cause rubbing or abrasive damage to pipe.


Over ... (more)


Sounds to me like galvanised pipe if it is simply laid on ground to give some measure of security, with break pressure tanks to keep static pressure within pipe capacity and to give taps a chance if the pipeline supplies tap stands or similar.

Design will also have to look at flow velocity which risks to be very high (danger of physical erosion of pipe) given fall of nearly 1:12, also need to keep the pipes flowing full if at all possible to avoid air locking (ignore this risk at your peril!), so might have to put in more break pressure tanks than necessary simply to keep static pressure within pipe capacity.

Re air locks, laying pipe to a constant fall so far as possible rather than simply on a straight line horizontally may be important if the fall is not constant.

Difficult to say much more without further details of anticipated pipe flow and profile ground between source and delivery point, and whether to tap stands or other.


Best regards

Tim Foster


My first advice would be don’t do it!  Pipes laid on the surface would be subject to mechanical damage from passing people, animals (do you have elephants where you are?) and vehicles as well as more vulnerable to anyone along the 6km route that might want to gain access to water through an illegal connection.  I guess, however, that you are limited by cost or time so if you do have to lay the pipes on the surface then it would be most advisable to use strong pipe such as galvanised iron.  Again there is a cost implication so you may be limited to plastic pipes, in which case use a thick walled HDPE pipe.  At all costs avoid the use of PVC pipe which becomes brittle and cracks when exposed to sunlight.  If you are using High Density Polyethylene (HDPE) then ensure that it is at least buried at any point where it would be more subject to damage such as where it crosses tracks.  Also try to ensure that no sharp stones or other edges are pressing against the pipe.  If it is any comfort, I have seen surface laid HDPE for the complete water supply system on the island of Rodrigues which is part of Mauritius.


You have a 480m head difference between the top and the bottom of the pipe which is equivalent to 48 bars of pressure at the downhill end, ignoring head losses due to friction along the 6km of pipe.  The head losses will depend on the flow rate and the pipe size and you should size the pipes to comfortably handle the flow from the spring.  Tables of flow rate/pipe size/head loss are available in various publications such as “Engineering in Emergencies” as well as from the manufacturers and on-line.   Once you have an estimate of the maximum head likely to be experienced along the pipeline you can specify the pipe by the Nominal Pressure Class PN which I believe goes up to PN20 or 20bar (200m vertical column of water) of pressure.  The Pressure Class is related to the wall thickness of the pipe. If your pressure is still too high you might need to include break pressure tanks or a pressure reducing valves along the pipe run.    


Don’t forget that you will also need air valves to allow air to escape from high points along the pipe run and wash-out valves at low points, especially where there is likely to be a build-up of sediment. 


Best regards




If the pipeline is going to be laid on the surface of the ground then galvanised steel pipe is the only option. Plastic pipes laid on the ground are subject not only to animal damage, but also to sabotage by those wishing to feed their goats etc, or to get water for other purposes. Galvanised steel pipe is more expensive than plastic and also costs more to install as you have to have a joint every 6m or so. 


The simplest way to deal with the pressure is to have break-pressure tanks to limit the pressure building up in the pipeline. The number you would need would depend on the profile of the pipeline. Ideally you would have the initial section feeding a lower level tank without any valves and simply run the excess water to waste in an existing steam course. This would give you a lower pressure in your pipe without the need for any expensive and difficult to maintain pressure control valves.


John Cosgrove


In response to the TSS Water Pipe Query, detailed guidance is included in "Engineering in Emergencies" (RedR/Practical Action Publicaitons, 2nd edition).


See section "13.4 Water transmission and distribution", page 344, and subsequent sections:


Section 13.4.1 Pipes - guidance on different pipe materials, pressures, pipe laying, pipe fittings.


Section 13.4.2 Valves and tapstands - including service pipes


Section 13.4.3 Pipeline design - including guidance on gravity flow pipeline design which would be required for a spring fed gravity pipeline of 6km in length. This section importantly includes guidance on break pressure tanks, avoiding air blocks, wash outs and additional considerations for spring fed schemes.


480m is a significant height drop and quite rightly the design needs to be carefully checked to avoid both excessive pressures and negative pressures, according to the terrain. Clearly close engagement is required with not only the community who is to benefit but also the people across whose land the pipe will pass, if different, as pipes laid above the ground are vulnerable to being damaged.


Jan Davis


Just a further couple of points to add about above ground pipes:
1. They are vulnerable to both sabotage and theft hence my comment above about engaging with the communities across whose land the pipeline crosses. If they are able to benefit then they are likely to look after and watch over the pipeline. Hence, something to consider in the design are tap points along the route serving potential users, either domestic supply or animal watering.

2. Be very careful when commissioning the pipeline. Catastrophic failure of joints can occur when a very hot steel (or plastic) pipleine is suddenly filled with cold water. Choose the time of day, or night, to commission gradually and monitor closely.



That's a heck of a long way to bring water to a small community.

What about rainwater harvesting, wells, well-jetting, rivers and infiltration galleries? What about using existing sources but improving the water quality through biosand filters or other means (settling, filtration, chlorination)?

If you really want to pipe water over 6kms then it would be good to know the volume available, the needs of the population and institutions.

Would extracting the volume of water required cause problems to farmers, habitat or other downstream users?

I was closely involved with a project in the DRC where we initially ran a 25mm dia, mdpe (alkathene type) pipeline buried where possible in a shallow channel through a forest but over a much shorter distance and with only around 75m head. We kept it flowing downhill and where we couldn't bury it we encased it in concrete. People set fires in forests for all sorts of reasons - including getting rid of snakes - so you have to protect plastic pipe. Even when buried, the pipeline and break pressure boxes will need to be inspected daily. Break pressure points are needed if you expect to need to stop the flow but these provide means of public access to a community water source - do the water source points and break presssure points present potential security risks?

One year later we installed 125mm dia. HDPE SDR11 piping in 6m lengths butt welded via a hot-plate trimmer, welder and clamping machine and installed a pico-hydro system that is still running after over 20 years (with one rehabilitation) providing power for lighting, fridges and surgery to hospital buildings and worker's houses. You could make power at each break pressure point but it is useless unless there is a local demand.

Local distribution was performed using the previously installed 25mm mdpe piping and Talbot Talflo taps at several tapstands. We used several pressure regulators to reduce the pressure at some of the tapstands.

If only water is required it is tempting to examine what could be done with an open pipe design - how far could the pipe run with only friction creating the pressure before a break point would be needed? This may allow a much cheaper pipe to be selected. It would however be susceptible to blockages causing bursts higher up....

Don't forget to mark where the pipe is - undergrowth will cover it up, small trees need to be discouraged from taking hold and damaging the pipe and any break tanks need to be examined on a very regular basis.

Water from resurgent high mountain sources tends to also have entrained particle which over years prove to be quite abrasive. The hdpe and mdpe pipes we used have survived well but the brass/bronze valves and bronze Pelton wheel suffered.