Impeller or screw pump powered by pedaling a bicycle

[snowboarder1531]

Hello,

History:

We have been spending a litle over a year designing a bicycle powered pump for irrigation in developing countries (

http://www.developingworldtechnologies.org).

Previously we were using a positive-displacement diaphgram pump, where we used a crankshaft and chain drive interface to transfer the rotational power from a bicycle into a back and forth motion to power two pumps 180 degrees out of phase. We continually get a jerk during the pedaling, so now we are looking to change the pump from a back and forch positive displacement pump, to a rotational impeller or screw pump.

What we are looking for:

An impeller or screw pump that is self-priming.
Works adequately at 1/10th hp output (power that a human can sustain)
Generate a decent flow rate
Generate some suction and delivery head

Any suggestions?

I appreciate any input you can provide! Thanks!

Brendan

 

[waross]

Tell the jerk to smarten up and fly right!! Sorry, the devil made me do it.
Have you considered a gear pump? Another option that may work well with a bicycle is a bucket wheel elevator type pump.
Yo will always have an uneven power production with a bicycle as the pedals pas top dead center. A bicycle coasts through this part of the pedal arc much better than a pump.
Are you using an arrangement where the pump pistons or diaphgrams are basically directly below the pedal crank? The power output from a bicycle crank will be very cyclic. Direct action at the optimum angle may end up to be the smoothest option.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[insideman]

The type of pump to use depends heavily on the head that needs to be developed, the flow, and the rpm. In short: Specific Speed. Head and flow depend on what you need, rpm can be adjusted by gear or belt drives. The irregular effort you mention can be smoothed by using a flywheel.

My guess is that you would be best served by a centrifugal machine. These can be fairly simple, to be produced by "backwoods" foundries and machine shops. Please give us a representative sample head that you would need. Then we can figure what flow can be had for 1/10 hp input.

I am a retired mechanical engineer, dying of boredom. Save me!
Let me help.

 

[zeusfaber]

Another possibility might be a flexible vane pump (like a Jabsco engine coolant pump, or one of those things you attach to an electric drill when you're feeling lucky)

A.

 

[itsmoked]

Diaphragm pumps are what you should continue with. They are extremely easy to maintain. They can be cleared of debris easily. They can be understood by the uneducated much faster than a centrifugal(magic)pump. They also create a healthly vacuum and are fairly human power rated.

Note that they're the hands down standard for sailing vessels as emergency, human powered pumps, pumping trash and anything else that a damaged boat bilge can dish up.

If you're having trouble with the cyclical loading do as insideman suggests, add a flywheel.

If this thing has to be hauled from place to place and you are concerned by a flywheel's weight, you can alleviate that by having the flywheel geared high. The energy or smoothing provided by the flywheel goes up with the square of its rotational speed.

Alternatively make it as a reservoir and fill it with water once it reaches the day's location. You would need a water filled flywheel to be radially partitioned to work correctly.

Got any pictures?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[JoeySoap]

Another vote for a flywheel from here too! Good luck.

 

[JoeySoap]

Sorry, I meant to add......... I'd avoid a centrifugal pump because its perfomance will simply depend far too much on the speed of pedalling. A recprocating pump will pump just fine whether you pedal quickly or slowly. Only the rate of pumping will vary with pedalling speed. HTH.

 

[Compositepro]

Human and animal powered pumps have been around for thousands of years and those designs have been little improved upon. Two good designs are the water wheel and the Archimedes screw. Compact pumps will be much less energy efficient due to friction losses.

 

[snowboarder1531]

Hello,

Thanks for all of the great responses!

I should have posted our youtube videos earlier, so you could see our current designs:

http://www.youtube.com/watch?v=xkhcOmWd_5g

http://www.youtube.com/watch?v=1Ch5veTdfoY&feature=related

The delivery head we are looking for is 30 feet (minimum) and a suction head of 10 feet. The higher the flow rate the better, as this pump is intended for irrigation purposes for farmers in developing countries. We traveled to Malawi in Africa, and during the dry season there is a large need for improved irrigation. They typically use buckets to retrieve water from shallow wells (holes dug in the ground, that can be up to 10 feet deep). Most of the farms are a set-up to support a village that has up to 70 people (10 families), so it's a lot smaller than most american farms. These farms can have a bit of elevation gain (i've seen up to 40 feet on a farm) and as you can imagine, carrying buckets of water for 12 hours a day is not only inefficient, but back-breaking labor. There are some pumps called "treadle pumps" designed for these applications, but they are awkward to use, difficult to transport and leak large amounts of water (plus we've heard horror stories about the ethics of the company which sells these treadle pumps in africa).

We initially worked through a couple different pump designs. Our initial designs were with an impeller pump, but the pump wasn't a good fit for the amount of power a human can produce (was very difficult and tiring, plus had mediocre performance). Any recommendations you could make for centrifugal/drill/impeller pumps that can deliver 10 feet of suction head, 30 feet of delivery head are greatly recommended!

We ended up switching to diaphragm pumps, and we have a push-pull irrigation pump designed around one of Itt-Jabsco's diaphgram pumps that works well. In developing countries of Africa, bicycles are the main mode of transportation, so we have been working to design a pump that works in conjunction with a bicycle. The issue we have is the non-continuous load, which makes pedaling jerky (please see the youtube videos above). The fly wheel idea is a good one, but I wonder how it could effect the simplicity, overall cost and robustness of the design. We are designing this pump to have as few moving parts as possible to limit the chance of any quality issues (as you can imagine, repair shops are hard to come by in much of the developing world) and to be sell for roughly 100 USD.

All of your responses and ideas are greatly appreciated!

One new idea we have is to make an irrigation pump that would work very similar to a stair-master at a gym. There would be two levers coming out, each lever would be connected to a diaphragm below it (so two diaphragm pumps in parallel). Each time the farmer steps down on one lever, it would lift up the other lever (and close one diaphgram pump, while opening the other). If we could design the pump to fold up into a backpack as well, it would make the whole design very transportable.

Any thoughts on this design? I look forward to all of your responses!

We recently formed a non-profit organization called Developing World Technologies, and you can view our website at

http://www.developingworldtechnologies.org.

Brendan

 

[zdas04]

The flywheel could replace your crank disk, just make that disk larger and heavier. A segmented disk with water chambers would work to improve the transportability, but the chambers (more is better than less) are crucial or you will use up all of your energy in fluid friction within the flywheel.

Looks like a workable solution, much better than a centrifugal or any other dynamic pump. I'd stick with a PD pump, and few of them are going to develop the head that the diaphram pump will.

David

 

[waross]

I suspect that there is a lot of wasted energy in that setup. Gearing tends to lose efficiency.
I really like the Stair-master idea. Put the diaphragms directly under the foot boards. You may want to make the diaphragm mounting movable closer or farther from the foot board pivots to find the most comfortable position. The optimum mounting position may change with the strength of the operator and or the head required.
I think that the direct action possible with a stair master type design will be the most efficient.
Portable? Put a couple of wheels on it and it may be towed by a bicycle or moved by hand like a wheel barrow.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ccjersey]

I'm not real familiar with a stairmaster, but I suspect the leverage adjustment would be related to the weight and height of the operator and pump discharge head required.

Maybe the operator just chooses where to stand along the length of the levers to match the length of stroke/weight.

A standard stair is around 7 inches rise.

 

[racookpe1978]

Hmmmn.

What about using three (smaller) pumps at offset 120 degrees from the "operational" load?

That way, while the pedal power (as noted above) would be low at the top and bottom of each pedal rotation, the load (the pumps) would be drawing or pushing a little bit at those directions and the output be smoother.

Alt, and cheaper, would be two smaller pumps - each drawing at a point in the bicycle "circle" so the "suction stroke" is at the pedal top and bottom, but the "push" "pressure stroke" is at the 90 and 370 "power" positions.

 

[racookpe1978]

Maybe that 370 degree point should be moved to the 270 degree point ..... 8<)

 

[waross]

Cheap, easy to build and efficient.

http://www.youtube.com/watch?v=LqIgLLmhq_E&feature=related

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[EdStainless]

The issue with the treadle pumps isn't the group building them. It is freelancers that are 're-distributing' them.

Stay with PD. It is more robust.
The price point is is about $20. After all someone has to get enough improvement in yield over a season to pay for it off of one of these small farms.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube