Help me build a bicycle powered water pump.

April 9, 2008 12:37 PM

Help me build a bicycle powered water pump.

So I got drafted into putting together a human powered water pump for the neighborhood Green Fair. I tried to tell them I'm a civil engineer, not mechanical, but to no avail. So I thought I'd ask y'all for suggestions.

Anyway, the organizer is interested in a demonstration pump similar in concept to the Playpump, so that the local kids can pump water from Bucket A to Bucket B. Eventually the pump will end up at the community garden to be used to pump water from the rain barrel onto the plants.

So the basic requirements are:

1. Relatively cheap (hopefully the bike will be donated)
2. Somewhat durable. It won't be used 24 hours a day, but should be able to stand up to rowdy kids and semi-regular use in the garden.
3. Reasonably efficient, so that a normal person pedalling at a moderate rate can move 3-4 gallons in a few minutes.

It'll be drawing out of a rainbarrel so it won't need to generate a lot of suction, nor will it need to blast the water out of the hose.

I was thinking of trying to salvage a waterpump from a car and fitting it with a sprocket. Is this a reasonable idea? Any other suggestions for cheap, nonelectric pumps? Will the gear ratio be enough to get any volume out of the pump or should I try something different? Has anyone built anything like this before?

posted by electroboy to Grab Bag (21 answers total)

If you have to use a bike, something along the lines of Archimedes' screw is fairly simple.

For more practical applications, ditch the bike in favor of a treadle pump. You can build treadle pumps with ordinary bike air pumps. If you still want to use the bike, you could attach one end of an arm to some point on the wheel away from the center and the other end of the arm to the end of the pump, kind of like a locomotive wheel. The arms converts the rotation of the wheel into linear motion (a crank) that drives the pump piston.

Be aware however, that you can get a tremendous amount of power out of a bike wheel, especially a geared bike like a street 10-speed. Just think, that system can move a 200lb human many miles per hour from a standing stop. A gallon of water weighs about 8lbs.
posted by Pastabagel at 1:04 PM on April 9, 2008

There seems to be some work on pedal power going on at Humboldt University - link to start.

I've been interested in pedal power for some time, but unfortunately I didn't really study engineering (got the calculus and physics under my belt along side future engineers, but that's it). I've looked around the web some, though it's been a few years, and haven't found the resources I'd wish for. Humboldt was the main one. Another is Sustainable Village products (a quick check just now didn't turn up anything definitely for your project, but you might have better luck if you mail them).

Good luck! I'd really, really love to hear more about your project, and I hope you publish your results somewhere on the web. I'll happily put up your report on my own site if you don't have anywhere else -- but I bet your local press will cover your efforts.
posted by amtho at 1:09 PM on April 9, 2008

It might or might not work for your purposes, but I've always liked the simplicity of rope pumps. I've seen plans for pretty snazzy ones, although I've never built one.

And you have already googled "bicycle water pump"? Because there are a lot of websites out there with photos and schematics on them (even on youtube). There are also all the appropriate technology-style books on village water systems; many of those include bicycle water pumps. The plans on this website are pretty typical for the genre. The Appropriate Technology Sourcebook is a classic, though I don't know how up to date it is.
posted by Forktine at 1:21 PM on April 9, 2008

If you consider the typical gearing on a contemporary bike, you'll have a maximum gear ratio of about 4:1 (48:12). Comfortable pedalling speed for a inexperienced cyclist is about 60 rpm, so you can get, at most, 240rpm without adding further gearing. Automotive water pumps run in the thousands of rpm, so I think that they would be less than ideal.
posted by ssg at 1:23 PM on April 9, 2008

I've been reading about the play pump, and I'm not sure that's the right way to go. The play pump and the pumps I linked to are for drawing water from an underground well, so you need a machine that can do a lot of work to counteract gravity and lift the water up. In the examples on the website, the water is actually pumped up to a water tower. Getting water out of the tower is easy. Just open the spout. Gravity is doing all the work.

In your case, you have a rain barrels sitting on the ground ( I assume the rain barrel is not below ground). When the barrell is full, the water is already at an elevated height. All you need is a tube to get it out of the rain barrel on to the ground. Moving the water back and forth between two barrels is actually a different problem.

So for providing water to the garden, all you need to do is raise the barrel off the ground, so that the height of the rain barrel is higher than the garden, put a hose or pipe in the bottom leading to the garden, and put a spout on the end. When the spout is open, the water will rush out at a pressure equivalent to the amount of water in the barrel. A full 50-gallon drum will produce a decent amount of pressure.
posted by Pastabagel at 1:26 PM on April 9, 2008

Here is a better link to a bicycle powered water sprinkler.
posted by Pastabagel at 1:39 PM on April 9, 2008

Response by poster: At the risk of trying to micromanage the discussion, I should point out that the primary purpose of the project is to demonstrate a human powered pump built using technology familiar to kids. Using it to pump around rainwater isn't strictly required and it will probably just be used to keep kids busy while their parents are working in the garden.

The bicycle powered sprinkler is definitely along the lines I'm looking at, but much much too expensive.
posted by electroboy at 1:55 PM on April 9, 2008

This small pump has a shaft that could easily be connected to a bicycle wheel.
posted by Daddy-O at 2:00 PM on April 9, 2008

The car water pump is a bad idea for several reasons, not the least of which is that half the housing is (typically) the side of the engine block. IOW, you can't just unbolt it and use it elsewhere, because you're leaving a substantial part of it behind.

I like the idea of a piston pump. I'm not sure you even need your bicycle frame to have any wheels - you might be able to attach the piston rod directly to the crank, next to one of the pedals. I just did a little sketch, and I think you could build a simple pump from not much more than some PVC pipe and a couple of rubber balls. I don't have any way to post a sketch at the moment, but email me if you're interested.
posted by jon1270 at 2:14 PM on April 9, 2008

What about some kind of reverse water wheel, with cups on a conveyor chain of some sort. The cups would get tilted and spilled into the new container when they arrived in the proper position. You'd have to use some kind of down-gearing to give slower speed but greater force going from the pedals to the conveyor chain.
posted by amtho at 2:25 PM on April 9, 2008

Something a little beefier than the Sears pump: a rotary barrel pump. Buy one, and replace the hand crank with a shaft extension which you connect to your bicycle PTO.
posted by bricoleur at 2:27 PM on April 9, 2008

Response by poster: Yeah, I'm not married to the auto waterpump idea, but so far it seems to be the most economical option, since old ones are more or less free. This is off of a 4-wheeler, and seemed like a possibility. From what I could find, these put out about 30-50 gpm at operating speeds, which I'm guessing is around 1000-5000 rpm.

I was resisting the piston pump idea because I couldn't visualize how to translate the rotary motion into a reciprocating action.
posted by electroboy at 2:30 PM on April 9, 2008

Stirrup pump + see saw would have even fewer parts, because the motion's already reciprocal.

Failing that, some kind of linkage that converts rotary motion to reciprocal? Jon's rubber balls would be good substitutes for valves if the pipe is transparent.
posted by Leon at 2:33 PM on April 9, 2008

...rubber balls would be good substitutes for valves if the pipe is transparent.

I had the same thought, though clear PVC puts you into special-order territory (as opposed to big-box store) and probably triples the price. The "super balls" you can get from a candy machine would make great valves.

Do you have a budget for this project?
posted by jon1270 at 2:56 PM on April 9, 2008

I think bricoleur has it if you want bicycle power and an easy time building this thing. Here's what I would add: get a bicycle trainer. Either a stationary bike or one of the trainers that attaches to a real bike. The trainers have a friction device and usually a flywheel, so remove the friction device and replace the flywheel with a pulley. Attach another pulley to the barrel pump, run a belt, and you're golden. Just make sure that pump is very secure.
posted by backseatpilot at 3:21 PM on April 9, 2008

Response by poster: I'm thinking around $50, aside from the bicycle. How does the superballs as valves work?
posted by electroboy at 4:00 PM on April 9, 2008

Basically the balls sit loosely on top of circular openings and control the direction of water flow. When water tries to move through the hole in the direction that pushes the ball out of the way, it does so easily. When water tries to move in the direction that pushes the ball against the opening, the ball gets in the way.

The piston pump works by first sucking some water past one valve into a chamber as the piston is pulled out of that chamber, and then pushing the water out of that chamber, past a second valve and into a pipe or hose, on the return stroke.
posted by jon1270 at 4:08 PM on April 9, 2008

I did something sort of like what backseatpilot is suggesting a while back with a car alternator. The problem is, I need to power the coil at less than 12 volts or the back wheel slips on the training stand's roller and that's after lots of fussing. For something that's going to be out in the open where anybody can mess around with it, I'd look for a more one size fits all approach that can't be easily disabled by someone with good intentions (or bad intentions).

I think Leon's see-saw idea is good for a couple reasons. Simple construction, hard to misadjust and nothing really theft-worthy. I'd put a release valve in the plumbing so if whoever was doing watering didn't want water it would just cycle back into the tank if the kids still wanted to play on it.
posted by Kid Charlemagne at 4:15 PM on April 9, 2008

A design similar to the rope pump: Loop of rope goes over a cranked wheel, down to water, and back up through a pipe. Every few inches the rope is knotted and holds a disc cut out of available material such as tire rubber, which helps to lift water up the pipe to a spout. I suspect some kind of funnel rig might help the discs enter the pipe more smoothly, but if the rope is long enough it should hang relatively straight and not be a problem.
posted by eritain at 4:19 PM on April 9, 2008

Bilge pump. Build or buy. I used a purchased one for about a year, to pump from a shallow well (less than six feet to water) into a 55 gallon drum; from the drum, the water would siphon out to the garden rows.
posted by jet_silver at 4:51 PM on April 9, 2008

Response by poster: Ok, here's a quick update. The guy in charge of the Green Fair turned up with this barrel pump (although made of heavy-duty polyethylene, not metal). So now the plan is to epoxy a sprocket onto the crankshaft.

However! Now he wants to power it with two bicycles. He's really hung up on this idea of cooperative experiences. Any thoughts on how to power it using two bikes?
posted by electroboy at 6:54 AM on April 15, 2008