I am here in Zambia. I have a fast flowing river at a safari camp and I want to use this to pump water to a header tank some 30ft above the river. I am looking to create a tapered box with one end 100cm x 50cm ish and the other end tapering to a 40mm pipe. The box would be about 2m long and have a fall within the river of about 20cm (river bed gradient). So the water flows into the wide end of my box and the pressure will push water up through my 40mm pipe and into my tank. And this will happen as long as the river flows which so far is several million years. My question therefore is will this work. Or am I wasting my time. The river flow is probably 1m per second.

  • 1
    $\begingroup$ Have you considered a water ram? $\endgroup$
    – Jon Custer
    Mar 10, 2019 at 15:33
  • $\begingroup$ if you suck your pipe to give an initial "pull" then it will always work $\endgroup$
    – xray0
    Mar 10, 2019 at 15:55
  • $\begingroup$ What Jon said en.wikipedia.org/wiki/Hydraulic_ram They do make a little bit of noise, though. $\endgroup$
    – PM 2Ring
    Mar 10, 2019 at 16:06
  • $\begingroup$ As I understand it, a hydraulic ram obtains energy from a vertical drop of water. I suppose that if you can build a protective structure one might be able to set up a water ram at a level that corresponds to the bottom of the river. That would require a formidable protective structure, which would be costly to build. $\endgroup$
    – Cleonis
    Mar 10, 2019 at 17:16
  • $\begingroup$ What is the speed of the water? Under ideal conditions, a conservation of energy approach would imply that you can turn the water's kinetic energy into gravitational potential energy to determine the maximum height that you can get. Without knowing any more than you posted, my guess is that you can't get a 30' height out of your contraption. $\endgroup$ Mar 10, 2019 at 19:57

2 Answers 2


You describe a contraption without moving parts.
An inlet, and then the cross section tapers down. I expect that such a contraption will not have any lifting capacity.

I expect: even if you would dam an entire river, with an inlet somewhere in that dam, and the outlet at some height above the rim of the dam, water would still not flow out of the outlet. When the height of the outlet is above the rim of the dam I expect that the water will climb to the rim of the dam, and then overflow the dam. That is: I expect that the taper will not make any difference.

(I expect: if you don't dam the entire river the back pressure will make the water flow around the contraption rather than through it.)

I don't see any way to avoid building a contraption with moving parts.

One way, I'd say, would be to have a paddle wheel in the water, to harvest mechanical energy, and have an Archimedes screw attached to the same axle as the paddle wheel. The longer the Archimedes screw, the higher the water is lifted. (The paddle wheel doesn't have to be perpendicular to the water, hence the axle can be tilted, allowing the Archimedes screw to lift the water.)

Technically that is a single moving part, but of course in all it would be an elaborate build, and the axle must turn in bearings. As we know, bearings are subject to wear, requiring maintenance.

[Later edit, one hour after first posting]

In the comment section the possibility of a water ram was suggested. I commented that as I understand it a water ram obtains energy from a vertical drop of water.

You describe that the river is fast flowing; I assume that means that geographically the land slopes down quite a lot.

Now, a very shallow slope can be enough to sustain significant flow, here is a story to illustrate that:
Nowadays water is pumped, but in Roman times water flow for supplying cities was gravity flow. There is a place in France where the Romans obtained water for a town from a spring 50 kilometers or so away from that town, with a total elevation difference of 17 meters. So the engineers building the water conduit to the town had to maintain a very shallow slope very evenly. This conduit supplied the town with more than enough water.

So I wonder: if the landscape has quite a lot of slope, might it be possible to construct a long conduit, at a more shallow slope than the landscape, from a distance upstream. The idea would then be to have a flow of water, at the location of the camp site, at a higher elevation than the level of the river. That height difference can then be used to keep a water ram cycle going. The reason you need that height difference: the used water that flows out of the water ram must have a place to go. If the used water just backs up then the water ram cycle will stop. So the used water must flow out at the same level as the river surface, allowing it to flow away.

  • $\begingroup$ +1 for the Archimedes screw. $\endgroup$ Mar 11, 2019 at 12:51

According to the Bernoulli equation, a fluid stream traveling at velocity v can support a column of water of height $\frac{v^2}{2g}$ (above the water surface). For 1 m/sec, that translates into 0.05 m, or 5 cm. That's quite a bit less than 30 ft. And that would only be if the flow up the column is zero. You would need a stream velocity in excess of 15 m/s to really make this work.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.