Treadle pump photo, Fig. 9, in How Treadle Pumps Work, U.N. FAO, original source: cited in this article Treadle Pump Capacities for Human-Powered Water Delivery
Treadle pump capacity, sources, system design discussions

InspectAPedia tolerates no conflicts of interest. We have no relationship with advertisers, products, or services discussed at this website.

Water well treadle pump types, models, pumping capacities & applications:

This article discusses a community water supply system design using human-operated treadle pumps to deliver pressurized water from a water reservoir tank to pressure tanks and on into people's homes in Central America. For a community without electrical power and not near a running stream, adapting a treadle pump originally used for irrigation may provide a solution. The authors invite comments and suggestions from readers by email or by using the page bottom comment box.

This article series compares the pumping capacity in gallons per minute well pumps of different models and designs. At page top, a treadle pump in operation in Niger, a Bielengberg pump design, illustrated in How Treadle Pumps Work, U.N. FAO, cited below in this article

Green links show where you are. © Copyright 2017, All Rights Reserved.

What is the Pumping Capacity in Gallons per Minute for a 1-Line Shallow Well Jet Pump?

The following document, still in rough draft form, is presented to invite readers to offer suggestions, reports of experience, or product recommendations for delivering water in a rural community without electrical power and where elevated water storage tanks are deemed not feasible.

Reader Question:

David paul
2015/12/16 10:47 AM


Hello. I live and work in a very poor central american country. We are trying to set up a system for a very remote and extremely poor community (no electricity, roads etc..) to bring water to all the huts. For many reasons (technical. cultural, cost, etc) that would be long to explain here, we can not use gravity tanks higher up or solar so we are stuck with hand powered options. Pumping the water to a reservoir placed on a ground is not the problem as many human powered solutions exists (helps by a local windmill manufacturer when wind is available) for an insignificant cost and nearly free of maintenance. Our problem is to bring the water to each hut.
The solution we are currently working one involves two pressure tanks and a main reservoir. The system needs to deliver about 750 gallons of pressurized water a day. As I said earlier, pumping the water to the main reservoir on the ground is already solved BUT the pressurized part NOT.

Our system explained:
- A holding main reservoir is placed above ground
- 2 x80 us gallons expansion tank mounted en serie and depressurized are just below.
- the expansion tanks are gravity fed by the main reservoir (?)
- the pressure is built why an hydrostatic hand powered pump connected to the expansion tanks.

Because the hydrostatic pump can deliver very high pressure with few effort (women and kids will action it) but has a very poor flow rate (0.5 gallons/mn) we just want to use it to pressurize the tanks once they are mostly water gravity fed by the main reservoir.

Having to manually pump this amount of water several times a day into the expansion tanks is too much work and time for this community so this why we want to find a way to fill them by gravity and pressurize them by hand with an hydrostatic pump.

Do that system makes sense to you?
How long will it take to pressurize the tanks with one or two hydrostatic pumps in the eventuality that the tanks are already gravity filled with water?

Thank you very much in advance for your inputs.


I'll be glad to try to help with this question - I expect we'll need to exchange more information. I'm not a hydraulic engineer but I am familiar with some of the issues.

I would look first at the assumption that you can't use elevated tanks to both store and deliver water. Typically the elevation needs be no more than about 2 meters above the roof of the home or above the ceiling of the highest point of use of water.

I agree with your view that pumping this volume of water by hand is not reasonable and give my own analysis in the italics below.

The time and effort to pump water to delivery pressure in large tanks will be significant. Keep in mind that your tank needs to have enough internal volume not just to store water but to store the pressurized air that then pushes water out and into the delivery system. And you need a lot of air. A small air volume means that the ability of even a large reservoir of water to deliver water to points of use will be small too - perhaps just a few minutes of flow.

The Engineering Toolbox gives several helpful if rather technical calculations. For example, the air pressure drop in a piping and tank system depends on volumes and pressures - at

A seat of the pants guesstimate of pumping time will depend on volumes being delivered and desired flow rates you want to see at the point of use.

Your 0.5 gpm pressure that you describe, is actually a flow rate of 1/2 gallon per minute. It' s not telling us pressure.

Suppose we *could* put water into a holding tank at an entering flow rate of 1/2 gpm.

To put 750 gallons into an empty tank would require 325 minutes of pumping - ASSUMING - that the tank remained at 0 psi internal pressure all during that time, which it wont. In fact the pressure in the tank will increase steadily as the incoming water compresses the originally-empty air volume in the tank into an ever smaller space left by the incoming water.

Let's use a 1500 gallon tank as an example. When you've pushed 750 g of water into the tank you've doubled the original air pressure in the tank from "0" (actually it's at one ATM of pressure) to two ATM or to an increase from the original 0 on the pressure gauge to 14 psi on the pressure gauge.

That's not much delivery pressure but it's enough to get water to the first point of use.

As users draw water, the pressure in the tank falls.

When you've drawn half of your 750 gallons (325g) the pressure is down to 7 psi. Quite weak, poor flow, but maybe usable. The people at the end of the draw-down cycle will be standing around quite a while waiting for their bucket or container to fill with water from the tap.

So I'd guess that in daily use people would need more than 325 minutes of pumping to keep a water delivery system functional. So we need 5 1/2 hours of pumping each day to fill the tank and more hours of pumping (probably) to keep it functional.

We (you and I) need to review my reasoning with a hydraulics engineer but I think I've got this basically right.

So I agree that the hand pumped volumes and pumping method you described would not be reasonable.

But I add that pressurizing smaller pressure tanks to give flow is counter-intuitive as follows - correct me if I've misunderstood:

To get water from the reservoir into a pressure tank the incoming water must be at pressure higher than the pressure already in the pressure tank.

An internal bladder type pressure tank, even a big commercial one, typically isn't going to give more than 50 gallons of water draw down before the tank has to be re-filled and re-pressurized. If we took the drawl down water out until the pressure tank drops to 0 psi and then started pumping water back in, we are back to the identical total pumping time as my previous analysis.

In other words if we are pumping by hand and using 750g of water a day, the total pumping time is not reduced by using intermediate pressure tanks, and no water would flow out of the gravity system or reservoir tank until pressure in the intermediate pressure tanks was below the outlet pressure of your gravity system - which if they're all at the same level ground area is 0 psi.

Question 2: how do we fill tanks by gravity? The water source has to be higher than the tank.

But not always. Is there a nearby running stream? If you have running water we could look at a hydraulic water ram - aka kinetic water ram to do the pumping for us.
See KINETIC WATER RAM if you think this might work.

Question 3: what are the elevations involved between water source and point of use? If there is not enough stream flow to use a hydraulic water ram can we use siphons to move water from a stream or lake into a delivery conduit or pipe?

As its publication may prompt other readers to offer a helpful suggestion, if you give me your OK I will include a redacted (to respect your privacy) version of this discussion online

Reader follow-up: calculating vertical lift, treadle pump output, pressures and the water head

... there are a few points I need to be cleared about and because as well I like challenges. To start with, there is no flowing water around and I know a ram pump would be of many use to fill the main reservoir placed on the ground.

The original pumping to the main holding 750 gallons reservoir(s) is not a real problem. We have at our disposal a local made treadle pump that can pump 100L/mn to the reservoir. We can even have 2 of them coupled and pumping at the same time. On the top of that, the treadle pump(s) would be part of a windmill system. Now that does not solve the pressure issue, far from it.

About the pressure tanks: Please tell me if I am wrong and let me give you more data. The huts are at about 10 feet from the ground + 3 feet to reach the higher water exit (shower). The distance between the huts is approximately on average 35 feet + the 17 feet of vertical distance.

If we decide to use 2x120 gallons pressure tanks, effectively the volume useable will not be 240 gallons but maybe 1 third of it. Let say 80 gallons of pressurized water available. That amount of water is enough for half the community water needs assuming the pressure is sufficient to use all of it before a re-fill.

Let's consider that the pressure tanks are filled with an independent treadle pump pumping water from the reservoir into the pressure tanks at a rate of 60L/mn (15 gallons) until the pumping start to be too difficult or the air inside the pressure tank oppose too much resistance. I know treadle pumps put some pressure as well but don/t know the amount of it in comparison with the rate flow they produce this is why I assumed a flow rate of 60L/mn instead of 100L/mn.

Let's suppose now (need input on that) that 70 gallons are pushed into the expansion tanks with reasonable ease. I understand the pressure in the tanks will be very low at that stage and the treadle pump would have to fight against the air inside the tanks and the exponential pressure been built but the pumping time should be no more than 70/15= 5mn and (X) psi into the tanks. Now, if we use an hydrostatic pump (the same one used to test pipeline) to hand build the pressure into the tanks as proposed earlier:

Once you help me answer those questions, then I can calculate how many time the overall pumping has to be done per day, how many people would be involves and how much time in total that will takes. I am not expecting or even want a high psi at the water exit (shower) but yes that the flow last the longest possible.

About using rooftop tanks and gravity pressure systems instead of a treadle pump system:

There are various reasons why we can not use gravity pressure. Here are some of them:

On the top of that we want the system to be in the center of the "village" as it can act as an educational elements and secure the cleaning process that can be done by all members of the community (specially children). As well we want the community to really federate around the system has it will bring them a better quality water and better health. We want to use chlorine tablets to treat the water.

Why we need to clean the water reservoir tank weekly

Because bacteria grows very quickly in these latitudes. If we want the water to be drinkable, the level of chlorine can not be high. For ex a swimming pool here if not filtered for 2 days starts to be grow green EVEN with the chlorine. Of course a swimming pool is opened to sun ray and that helps bacterial growth when a tank is closed to UV ray. But even so, once a week is what we want to achieve to be safe. Normally the water they use for drinking and washing is below ground (7 feet or so) but this is not still water. And even then there are some water diseases that affect the community. Their organism is more resilient than ours so they usually cope with it but the all idea is to bring better hygiene and that mean easy cleaning.

I am talking about the holding tank NOT the pressure tanks.


Interesting; where we live in Mexico people store water in rooftop tanks and never clean them; maybe we're all just immune.

Water potability is another whole discussion in which I'm also interested. Thanks for the explanation.

Reader follow-up:

Maybe the water that arrives into those Mexico tanks are already heavily city treated (UV, Chlorine, etc). Our water come directly from underground and can be contaminated by dead animals. Chlorine kills a lot of the bacteria (depending on level of chlorine and time) but we can have some resilience so we believe that implying a cleaning protocol helps and act as well like an educational process specially if the water is still.

The tanks on the roof in Mexico are supposed to be built in various "capas" and with a material that stop UV and if on the top of that the water is already city treated by various process, the risk is far less. And even so, it is recommended to clean them. The reservoir we are planning to built will be built with cement (easy transportable by horse) and local stones. Again for transportation purposes.


Yes in the community where we live you're quite right. Out of town the use of cisternas including ones below ground is very common as is the collection of rooftop rainwater; And most hand dug pozos are nowhere near potable by health standards.

I've got quite a bit of info on methods of making water potable if any of that is wanted.

I wonder if for your system a ceramic filter or a solar distillation for drinking water is plausible. Probably the solar still would be a too-big operation and the ceramic filter too slow for high volume drinking water.

Calculations for Discussing Use of a Treadle Pump & Pressure Tanks for Water Delivery for Home Use

I realize that treadle pumps have long been in use for irrigation systems. Here we're exploring using a treadle pump system to deliver water for cooking, bathing, drinking in a community where there is no electrical power and where other restrictions argue against gravity-operated rooftop water tanks.

This is a sort of "thinking out loud" piece. Sorry, if I had more time it'd be more concise.

Not knowing your treadle pump specs I looked at a 102mm diameter piston -
- max lift is 5 meters and industry standard is 4 meters (or meters as we say)

and at

for more about these pumps. The FAO gives a bit more lift in their estimates and gives a lesson in hydraulics for non-engineers, explaining head pressure etc.

David Paul Adds:

Another approach to this is to have two 40 Kg operators facing each other work the pump. The increases the available pressure on the pistons to 80 Kg. One operator has their weight discounted for standing between the fulcrum and the piston, and the other has their weight `bonused' because they are standing on the far side of the piston.

If a 50mm set of cylinders is fitted, the maximum suction+lift combination is 30 metres (100 feet). The output is proportionally reduced so with a 5 metre suction and a 25 metres push, you could expect to pump about 1800 litres per hour.

I think that suction lift of water should translate directly to pressure increase in a tank. I.e. lifting water 4m up out of the ground should be the same as pushing water 4m up above ground and should be the same as pushing that water into a tank against air pressure increasing in the tank as we pump. (I'm not a hydraulics engineer so I don't really know what I'm talking about).

David Paul Adds:

Extremely interesting data. With our local treadle pump artisan, we can maybe implement the 2 persons treadle pump! Thank a lot!

Figuring out the delivery head pressure for a treadle pump water system: how much pressure and flow can a treadle pump deliver?

FAO calls the output side the "delivery head" and says that the pump doesn't know if it's lifting water out of the ground or lifting water above the ground: the total lift capacity of any pump is divided between the well lift (water out of the ground) and the delivery lift (lifting water above the pump).

If we assume that the suction lift out of the tank is 0m, that is water wants to run out of your tank into the treadle pump by gravity alone, and if we assume that the gravity tank is not adding any pressure boost of its own to the system, then all of your (treadle) pump output can be given to the delivery head. That's the pressure we can achieve as we push water into a pressure tank OR as we lift water up to somewhere.

Actually when the gravity feeding reservoir tank is filled it will add some boost pressure to the pressure treadle pump step, probably reducing the time and effort needed to pressurize the add-on pressure tanks - let's call them "lift tanks" for this application.

But we can't directly convert pump lpm (or gpm) into pressure without calculating other effects: head, friction losses etc.

David Paul Adds:

Agreed, but the reservoir ahs been just above the pressure tanks and the treadle pump just nearby to the pressure tanks: I don't think the losses from head or friction losses will be significant.

If your LPM numbers for a pump are horizontal movement without considering lift or head pressure, that's not going to be the same.

We need 17 ft. of lift; or we need to overcome 17ft of head pressure in the system.

This table will help us

20 ft. of head is at 8.66 psi (measured at the bottom of the column which is where our pump resides).

I talk about this measurement in nauseating detail at

So from what you've said, we figure that when you've pushed enough water into the pressure tank, starting from 0 psi, to reach 9 psi (8.66 in Danspeak), water will start to come out of the shower head in one hut at the elevations you described.

David Paul adds:

Just talked to my local artisan manufacturer. He confirmed that a treadle pump with a succion/lift of 2 meters no produce more than 8 psi ;-(. So your calculations are right!

Boyle's Law & Water Lift Tank Pressurization

Using P1xV1 = P2xV2 to describe the volume and pressure of air in the tank (Boyle's law) to look at the pressure change in the air cushion above the water in a 120L tank, at sea level I think we can calculate how much water we have to push into the tank to get it from "0"on the gauge (really 1 ATM pressure or 14 psi but the gauge will say 0) up to 9 psi as follows:

14psi x 120L = (14+9)psi x ?L
(solving for volume)
1736 = 23 x ?
1736/23= ? L
75= L = the volume of water we push into the tank to get to 9 psi on the gauge and to lift the water head you described.

If I did this right I'm showing that when we've pumped 75 L we've reached 9 psi in the tank and we will start to see water coming out of the shower head.

If a pump were pumping at 60 Lpm we would reach the lowest functional pressure in less than 2 minutes of pumping IF there were no reduction in the actual pump output to overcome head pressure. We both are betting there will be a reduction in pump output as pressure in the tank rises.

You estimated 5 minutes of pumping to push 70 gallons (250L) into two 80-gallon pressure tanks ("Lift tanks") or 35 gallons into one of the tanks.

I calculated that before considering the head pressure we pumped 75L in a bit over a minute of pumping time (before allowing for resistance from any head pressure) to get the tank up to 9 psi and start delivering water.

So I think your estimate is safe. That is in 5 minutes we ought in theory to be able to get to between 9 and say 20 psi in the tank. That's enough to get water delivered and between 15 and 25 that's perfectly functional flow for taking a shower.

David Paul adds:

15 psi should be enough working pressure for our system.

Unfortunalty, in order to get a reasonable pressure of 20 to 30 psi into the tanks, pumping more time will not make more pressure because the effort that a person would have to put on the treadle will be high as he or she will have to fight against the air into the pressure tanks. We are trying with him to think how to drastically reducing the effort by adapting a treadle pump able to deliver 30 or 40 psi without being penalized with a long time pumping and low flow.

If you have time, try to Google Super Lario pumps. The SL models claimed to be able to pump up to 110L/mn and give a 50 psi pressure.

DF adds:

Super-Lario pumps, also used in marine applications and capable of pumping seawater and "aggressive liquids" are widely sold world wide. We include references later in this article. We need to review these specifications as I'm not sure about the duty cycle nor the extent to which the design is intended for handling large daily water volumes. From the product literature this product line from Gianneschi are intended for bilge pumping, fire fighting, and handling diesel and oil as well as seawater.

Treadle pump must overcome head pressure:

The fly in that ointment is that these treadle pumps have a limit to what pressure they can produce no matter what. In lift the pumps I looked at were say 4 meters or 13 ft or about 6-7 psi.

I'm guessing that we can take the maximum lift specification (if there is one) for your particular treadle pumps and that will tell us the maximum pressure we can produce in the tank in any amount of pumping.

David Paul adds:

We can increment the pressure by reducing the cylinders in the treadle pump. That will as well reduce the volume of water pumped per minute but it seems that the pumping time is not really the problem here, building the pressure is. I am contacting my local artisan maybe he has more inputs on the pressure matter. 6-7 psi seems to me very low for those pumps but let's wait later until he gives me more inputs on that.

Treadle pump water system must handle draw down cycle and multiple users, flow rates, distances etc.

If we like that number we also need to look at the draw-down cycle time: how long can we run water until it's not functional - the answer is we can run water until the pressure drops to below 9 psi - an amount we calculated (I think) as 75L

Let me know what you think of this reasoning.

That won't be wonderful but it's a start. Some installations or users may need more pressure to get a decent shower flow and more pressure to serve multiple fixtures at once; I suggest we also want to use or invent a low flow shower head for bathing; other flow rates, say for drawing water for other uses are less demanding.

David Paul adds [paraphrasing]:

Here we don't need more pressure to obtain a functional shower for bathing but we definitely will need more pressure and a greater flow rate to acocmodate multiple simultaneous users.

If necessary, we can as well manually reduce the flow at the shower tubes with a simple adjustable valve tap or buy for cheap a saving water shower head (save 30%of water). That should give us no more pressure but more time using the available pressure. But you are right, the most demanding is the shower for pressure and volume.

DF: Yes but some shower head designs that mix air into the outgoing water stream can generate a better bathing flow at low supply pressure. Just turning a supply valve partly closed reduces flow rate but it's not the same effect as can be achieved by a low-flow shower head design even if that's a DIY product.

Variations on Water Lift Tanks for a Treadle Pump System

David Paul adds:

Using the lift tanks

If instead of adding water to the pressure tanks, air is added. The first step will then be to totally depressurized the 120 gallons pressure tanks. Then treadle pump water into the tanks until reaching half of their volume capacity or more. At that stage no resistance is made by air as the tanks are depressurized by creating a vacuum. Then hand pump air instead of water (a pedal air pump can reach 4 bars or 60 psi). Will that work better and easier? The only draw back is that once the tanks are emptied of the water, the air would have to be vacuumed again to allow the easy water re-fill of the tanks.

I attached as well the BINDA catalogue water hand pump (very neat hand pumps). On the price quoted on the list, they can offer us 50% for this project. They claimed any Super Lario pump models can reach 50 psi

Correction: No vacuum would need to be done. The Pressure tanks can be gravity fed (a sort of valve system can be mounted on the air inlet pressure tank letting the air flows out while the tank are gravity fed) and then re-pressurized the tank adding air to until the desire pressure.

Other Notes

A one-line jet pump can typically raise water from depths of just a few feet (or "0" depth) to about 25 feet in depth, and at water delivery rates of 4 gpm up to as much as 25 gpm depending on the variables we list below the well pump capacity tables shown.

A nice example table of shallow well 1-Line Jet Pump Capacities for 1/2 hp, 3/4, and 1 hp shallow well pumps is provided in the Water Ace Jet Pump Installation Manual and excerpted below to illustrate the factors that determine well pump capacity. Both of the charts below are for one-line jet pumps produced by Water Ace. 1-Line jet pumps intended for shallow well use and made by other manufacturers can be expected to have similar capacities.

Deep well pumping capacity - Water Ace Co. Deep well pumping capacity - Water Ace Co.

[Click to enlarge any image or table]

The Water Ace charts (shown in part above) make clear that the capacity of a one-line shallow well jet pump to deliver water at a given flow rate varies by these factors:

  1. the depth of the well (the bottom scale in the two charts)
  2. the pump horsepower (the color codes indicate pump model and horsepower HP variation)
  3. The well pump model (the right hand table is for the company's more powerful well pump series of 2-line jet pumps)
  4. The condition of well piping, including pipe diameter, length, number of bends or elbows
  5. The presumption that the entire piping system has no leaks

Permission requested, Water Ace Corp. Aug 2010 - Pentair Pump Group.

Watch out: Safety warnings are throughout any pump manufacturer's instructions. Because some pump models are capable of developing internal pressures of more than 100 psi, if your building piping, pressure relief valves, safety controls, wiring, and plumbing are not properly installed, very dangerous conditions including electrical shock, tank explosion, and leaks or floods can occur.

What is the Pumping Capacity in Gallons per Minute for a Treadle Pump

A two-line jet pump can typically raise water from depths of 30-feet to 80-feet, and at water delivery rates of 4 gpm (gallons per minute) (for a 1/2 hp 2-line jet pump serving an 80 foot deep well) to 16 gpm (for a 1 hp 2-line jet pump serving a 30 foot deep well).

Well Pump Capacity Charts Plot Pumping Head vs GPM Flow Rate Curves vs Pump Horsepower (or PersonPower)

If for theoretical argument we exclude other snafus like clogged piping or a low-flow-rate well, the lift distance or head against which the pump has to work is the most significant factor as pump lift capacity varies by pump horsepower and type and location.

Treadle Water Pump References & Treadle Pump Sources


Continue reading at WATER PUMP CAPACITIES TYPES RATES GPM or select a topic from closely-related articles below, or see our complete INDEX to RELATED ARTICLES below.




Suggested citation for this web page

TREADLE PUMP CAPACITIES & USE at - online encyclopedia of building & environmental inspection, testing, diagnosis, repair, & problem prevention advice.


Or use the SEARCH BOX found below to Ask a Question or Search InspectApedia


Frequently Asked Questions (FAQs)

Click to Show or Hide FAQs

Ask a Question or Search InspectApedia

Use the "Click to Show or Hide FAQs" link just above to see recently-posted questions, comments, replies, try the search box just below, or if you prefer, post a question or comment in the Comments box below and we will respond promptly.

Search the InspectApedia website

Comment Box is loading comments...

Technical Reviewers & References

Click to Show or Hide Citations & References

Publisher's Google+ Page by Daniel Friedman