Bicycle Water Pump

Zambia farm.jpg

From Pete:
Where on your website can people leave comments? Where are my comments from last month? I am impressed with the work and effort you put in and also the transparency with which you presented your process to the group. Below are comments I have for you and for future project people:

In your initial problem statement, I don’t think that one of the problems is that treadle pump requires too much energy and gives you very little output. Likely the efficiency (energy_out/energy_in) is just as good for a treadle pump as for any other pump. The challenges lie in how the pump is used and the difficulty in transport. Additionally, I don’t think that there is a spring in the treadle pump as you have indicated in your description. I think that there is a chain or pulley mechanism that pulls one piston up when the other gets pushed down. You also state that the treadle pump cannot be adjusted for different elevations changes (not elevation gradients as you indicate… it’s not the elevation gradient that determines the necessary pressure, but the total change in elevation.) Do we know this for sure? Maybe this would be a good source of inquiry for a project? This is of course not a shortcoming in your project. However, for future projects, it may be a good idea to just get a treadle pump and see what we can learn from it.

You refer to several “failures”. I myself wouldn’t call them failures. In particular, how did you “fail” to build a rotary pump? I don’t recall you trying to build a rotary pump. I recall that you considered it and decided it was outside the scope of your project…. This seems like a success to me.

Later on, you state:
The community of Zambia will be able to purchase a bicycle powered water pump based on the design we build in class. This will allow the community to have portable water access that can be stored in a safe place when not in use. “
Are you sure that your project resulted in people using these pumps? This seems a little premature and overstated to me. I think it may be more correct that you have started a process that may lead to someone using a pump that may work for a community in Zambia. I think that this statement is a statement of your ambition. I also think that this ambition and focus on the end result and unmet expectations is often the source of conflict in a group.

Why did you assign a 60% efficiency to the rotary pump?.. what elevation gain did you assume in the 0.5 liters per second calculation? Can you put that calculation on an auxiliary website? Can you measure this pumping rate in the shop?

Again, although I have provided a considerable amount of feedback and criticism, I am very pleased with your participation and what you learned and accomplished.

Problem Statement

Current methods of extracting water from either wells or the ground and transporting it are insufficient in many parts of the world, specifically Africa. Current technology includes the treadle pump, however, this devices requires a high energy input with little useful output and is heavy, difficult to transport, and easily stolen. Many places in Africa, villagers walk miles each day for clean and drinkable water and are unable to irrigate any crops that could potentially feed hundreds.

Working with Seeds of Hope International Partnership, its obvious that a new technology is needed to pump water from a low elevation to a higher one, a technology that is easy to use, easy to transport, and can be made from materials commonly found in or easily brought to Africa. The bicycle water pump is a device that can accomplish all these goals, with research and trial and error the best available pump will be created.


Zambia location in red. Source:

Seeds of Hope International Partnerships, Zambia currently assists in distributing treadle pumps, among
many other useful technologies, to the local community peoples to assist them in watering their agriculture. Seeds of Hope was founded in 2003 to help create a foundation for service in Africa that would have a continuing level of continuity within the community and build relationships with the local people. Through their efforts, it has been noted that much time and effort could be saved through the use of a water pump to transport groundwater for agricultural uses. Originally, the locals used buckets. To assist, Seeds of hope brought in the treadle pump. However, using the treadle pump is much like working out on a stair-master at the gym, so the locals tend to shy away from using it after a short while. This is where the need for a bicycle powered water pump is derived from. The locals already have bikes and are conditioned from riding them daily, so pumping water will ideally transformed from exercise into productivity.


Current Technology

The current treadle pump (shown at right) is an effective means of transporting water from low elevation to high elevation. They work by having a person put their weight on a lever, which pushes down on a piston full of water. This forces water out the effluent hole under pressure while closing the influent hole. As the person releases the lever, a spring inside the piston forces the chamber to expand, drawing water into the piston. The effluent hole closes and the influent hole opens, allowing the pump to draw water from a hose. However, the pump has many flaws that leave much room for improvement. The locals use the treadle pump to move large quantities of water uphill to drip irrigate their land. However, this causes excessive fatigue and cannot be adjusted to accommodate for different pressures required for different elevation gradients. Also, they are very heavy and are difficult to transport so the must be left in the field. This leads to them getting stolen.

Pump Designs

The design of the pump is a large factor in the properties it has. Below are a list of possible pump designs, how they work, and their potential properties.

Piston Pump

1. Piston Pump: This pump is identical to the one currently being used in the treadle pump. Water flows in through a one way valve as the piston expands. When the piston contracts, the water is forced into a second one way valve and into another hose at high pressure. This pump is very basic and reliable but it is difficult to manufacture due to the necessity of an almost perfect seal between the cylinder liner and piston. If this were to be mounted to a bike, it would also require a method of transforming the circular motion of the pedals into a back and fourth motion of a piston (like the crankshaft of a car engine). However, it still functions at low speed when high pressure is needed. It also is capable of sucking water as well as pushing it out, so it may not need to be primed. Visual is shown to the right.

centrifugal pump 2.gif
Centrifugal Pump

2. Centrifugal Pump: This is a very basic pump where water enters a chamber and is flung around at high speed. The high speed pushes the water towards the outer walls of the pump where the water escapes into a hole at pressure. This pump is capable of moving a lot of water but it is not as effective at creating high pressure. High pressure would be created by spinning the inner wheel and the water at very high speeds and would quickly wear out a human. This pump must be primed as it cannot create much suction. Visual is shown to the left.


Gear Pump

3. Gear Pump: A gear pump works by forcing water between the teeth of two gears and spinning those gears quickly to push the water out of the pump. This is a very basic design that could be maintained in the village but it is not as efficient as the others and cannot pump at high pressures or volumes. This design is usually used for very small pumps and cannot create much suction. Visual is shown to the right.

Rotary/Geometry Pump

4. Geometry: This pump is basically a piston pump wrapped around in a circle. The geometry of the wheel on the inside and the shape of the cylinder interact in such a way that it causes suction and compression at different places. This is a very complex design and requires precision machining to create the correct shape. Visual is shown to the left.

Attaching to the Bike

Attaching the pump to the bike is a critical part of our design. We want to make it in such a way that the bike can still be used as a means of transportation and the pump can be attached and detached easily, no matter what shape the frame is. Below are a list of possible mounting solutions for each of the different pumps.

1. Piston Pump: This design requires changing the rotational motion into linear motion. This can be done by attaching a wheel or crank arm of some sort with an offset pin to attach the piston to.
-It is possible to mount the pistons to the pedals of the bike so that the pedals move up and down and compress the piston. However, this is hard to prevent from interfering with normal pedaling and requires a mounting point on the frame.
-Also, a second wheel could be attached to the back wheel so that the second wheel takes force from the wheel and then converts it to
linear motion. This is very easy to remove and since tires are usually similarly sized, it can be adapted to different frames. However, this design relies on friction and may lead to slipping or inefficiency.
-Redesigning the back wheel to be switched out for the pump would utilize many of the existing components. However, designing this in such a way that the bike is still usable is difficult.

2,3,4. All these pumps accept rotary motion, which is abundant on a bicycle.
-Mounting a gear to attach to the chain would be a way to tap energy from the chain but it requires a mounting point on a bicycle.
-Tapping into the rear wheel is a nice way of mounting the pump and the speed and resistance can be controlled by using the gear shifters. It’s more controllable than mounting to the chain as it can be controlled by both the front and rear derailleurs.
-The back wheel could be redesigned, which is probably the most durable option but again has problems with using the bike as a bike.
-It would be possible to attach a friction wheel to the back wheel but there is no advantage of this system over attaching it to the chain.

Project Goals

1. Establish a working relationship with Seeds of Hope International Partnerships.

  • Seeds of Hope International Partnerships (SHIP) is a holistic christian community development organization with a focus in integrated water, sanitation, and hygiene (iWASH). SHIP has been designated as an iWASH expertise training center. The specific center we are communicating with is located in Zambia. We have created this website to not only inform people of our project, but to communicate our findings and suggestions for improving the pump with SHIP. We hope, by expressing our successes and failures in this form, SHIP can take our findings and continue perfecting our pump design. Hopefully at this point, this pump can be sold to the community.

2. Design an efficient and easy to use water pump that can be directly hooked up to bicycles already in use by the Zambian community members without rendering the bicycle useless as a transportation method.
Photo on 2013-03-05 at 12.29.jpg

  • We obtained a bike from Seeds of Hope from the collection of donated bikes they were preparing to send to Zambia. We chose a bike in poor condition which would most likely represent the common bike used in Zambia.
  • This bike is very rusted and has 21 gears. There is no bike stand.
  • The gear cluster located at the center of the pedals is important to the function of the pump. The pump power is derived from the pedaling of the bike. There is a chain circling the outer most sprocket at this location that also loops around the sprocket on the rotary pump. A bike with a single sprocket would not not be able to hook up to the pump in this way. An optional attachment method could be at the back tire gear cluster. However, this attachment would require removing the back tire to insert the extra chain.

2. Find a locally appropriate design solutions/materials that draw upon the local resources of Zambia and can be made in country at a relatively low cost. (current treadle pumps run ~$100)

  • We initially built a piston pump out of cheap plumbing materials from home depot. This pump cost about $30 and was successful pumping water, but can not be attached to a bike.
  • We purchased a rotary pump for $25. This specific pump is used commonly to pump oil, anti-freeze, or gasoline. Because we are pumping water and not oil, the pump experiences high friction. We experimented with various lubricants to reduce the friction.
  • We have not come up with a way to easily build a rotary pump from local materials. This would be a good project for SHIP.


As a group, we basically all had the same intentions coming into the class. We wanted to take a problem and engineer a solution. In this case, make a water pump that is powered by a bike without having to take apart the bike. However, this was not the sole purpose of the class. Our group experienced some personality clashes in the beginning of the quarter. After conversing about the issues we were experiencing with each other’s communicating styles, our work environment became much more enjoyable. This was a great gain from our project. Another gain we enjoyed was becoming more comfortable with using tools in a shop environment.

We did have some failures along the way. We made a functioning piston pump, but the pump could not be attached to the bike. We ran into many roadblocks when we were deciding how to attach the pump to the bike. We failed to build our own rotary water pump. Instead we purchased a pump and attached it to the bike. The piston pump absorbed about 12 hours of work and was leaky, difficult to stroke, and was too large. The rotary pump took 6 hours to attach to the bike. The only portion of the rotary pump we built ourselves was the gear that attached to the arm of the pump. We would have liked to build our own rotary pump but were unsure of what materials to use. We attached the rotary pump to the bike with zip ties.



  • The community of Zambia will be able to purchase a bicycle powered water pump based on the design we build in class. This will allow the community to have portable water access that can be stored in a safe place when not in use. Because of the bicycle power, the pump is off-grid and utilizes the strong conditioned muscles of the community members preforming an exercise they do regularly.


As a group we hope to become comfortable with the divergent/convergent thought process. Our group is comprised of both of these types of thinkers: divergent (Sheila) and convergent (Pearce, Mark, Shauna). It is our goal to realize the positive feed back this type of thought process brings to the group, even though the thought styles will most likely clash at certain points in the project. Once we are able to recognize the strengths each individual brings to the table, a cohesive group will emerge. Therefore, it is not just the goal of manufacturing a bicycle powered water pump that we have, but also to learn how to communicate and cooperate in a group of diverse thinkers.


We built two working pumps: one piston pump and one rotary pump. We did not get the chance to perfect the pumps, but below are some performance notes.

Piston Pump

  • High friction using rubber, wood, and flip flop material.
  • Seal was decent if it was pumped at high speed (air didn’t have a chance to leak)
  • Performance increased remarkably with vegetable oil as lubricant.

Rotary Pump

  • Approximately .5 liters per second.
  • 19 tooth gear on the pump connected to the 50 tooth chainring
  • Humans can produce about 100 watts, we assumed a 60% efficiency and calculated the amount of water that a human could lift and started at a gear ratio that matched that.
  • The chainring was old, the chain was very old, and the 19 tooth fix-gear sprocket was brand new. None were from the same manufacturer nor were they the same size or wear. This lead to major issues with chain skip.
  • This was actually mounted on a bike using hose clamps. This lead to some issues with rotation, so a solid mount would fix this problem.

Design Process

Initial Decision Matrix

Picture 1.png

This decision matrix was based on the idea that the pump would be designed and built from scratch. The piston pump was chosen based on simplicity, effectiveness and feasibility. Attempt #1, described below, depicts our attempts to do so.

Attempt #1

Piston Style Pump Design:

Initial pump produced out of pipe with two one way valves attached to the closed bottom. The piston was made out of a plunger attached to a circular piece of wood and a rubber sandal. The wooden handle of the plunger, was to act as the piston arm and would be attached to the bike in some fashion. Pictures of the finished pump are shown below.

Unfortunately, we made the mistake of several other “save the world” engineers and failed to consider the feasibility of this pump for the customer. We realized this pump style is too heavy and bulky to be efficiently attached to the bike. Although we are still proud of our creation, we are now looking into other designs that will meet the needs of the PEOPLE more closely.


Secondary Decision Matrix

Picture 2.png

Picture 3.png

The secondary decision matrices dismissed the idea of creating the pump from scratch, and took into further consideration the attaching to the bike aspect. The first table compares the piston and rotary pumps and their ability to be attached to the bike. The second table considers the various ways the pump can be attached to the bike. Attempt #2, described below, depicts our decisions and attempts to do so.

Attempt #2

Tera Pump – Rotary Pump

Rotary Style Pump:

A cast aluminum rotary barrel hand pump was purchased from Tera Pump for $24.99. The pump features a suction head of up to 5′, a lift distance of up to 16’5″, and is capable of pumping 7 gallons per 100 revolutions. Pump is pictured to the right.

Rotary Pump with Gear and Chain Attachment

In order to attach this style pump to the bike, the rotating arm was removed from the pump and replaced with a gear to attach to an additional chain. Pictured to the left. The additional chain is connected to the gear on the pump and on the gear with the pedals. As the pedals are turned the chain turns, causing the rotary pump action.The pump is attached to the bike in the location shown in the picture below.

Pump Location

Bamboo Stand

Stand with Bike

Bike Stand:

A bike stand is needed to hold the rear end tire off the ground as the user pedals. Collaborating with the Bamboo Constructiongroup, a bike stand was made out of the strong and cheap material, bamboo! Pictured to the right is the stand during construction and the final product with the bike in place.

What to Do Next?

Now that we have a working bicycle powered water pump, there are a few improvements that can be made. Firstly, constructing a rotary pump out of local materials would erase the need to buy one. Secondly, this would allow you to gear the design towards the purpose of pumping water and not towards pumping oil, like our current pump. Since the water that will be pumped into the filter is not pure, it would be beneficial to design the pump to accept these impurities without clogging the pump. Lastly, more design effort should be put into the bamboo bike stand because currently it is not reliable under more than 150 pounds. Please consult the following decision matrix for possible design improvements.

This decision matrix determined the best next step to take is to improve the bike stand and test lubricants. However, it would be wise to experiment with bike connections and to try to build a rotary pump. The past two options would be more appropriate for ten weeks of work during a quarter lab. Thank you for taking the time to review our project website.

Team Members

Pearce Swerdfeger: Fourth year Environmental Engineer graduating in Spring 2013. From Phoenix, AZ and enjoys being outdoors and being active!
Sheila Magladry: Fourth year Environmental Engineer graduating in Spring 2013. Interested in helping the impoverished people of the world. Wants to travel and loves adventure. Never a dull moment with this girl!
Shauna Falvey: Fourth year Environmental Engineer graduating in December 2013. I would love to travel the world!
Mark Mekkittikul: First year Computer Engineer from the Bay Area. Highly involved with Engineers Without Borders and dreams of making a positive impact on the world.