Human-powered water desalination


I saw an RFP for a human-powered water desalination unit (liter scale), and I’m hoping someone can help me understand how crazy of a screw + gearing system one would need to generate sufficient pressure (800-1000 psi). Any takers? LMK…


I’m not sure I’d go with screw/gearing to get the pressure. A $40 hydraulic bottle jack from Harbor Freight will get you 40,000 pounds of force. Use that to push a 6" ( 15cm) diameter piston and you’ll have pressure to spare. a 7" piston, with 20T of force, will just give you slightly over 1000 PSI (an 8" piston will give you just under 800 PSI).


Thanks for the suggestion – I think the sponsor is also concerned about weight/portability, so that’s why I was trying to think of a potentially lighter weight (composite materials, etc) solution. I guess one could still employ something like a bottle jack design, with different materials.


I think what I’m trying to say is that 1000 psi isn’t hard to do, that there are common, cheap, tools that’ll do way more.

For instance, an M20 screw is a fairly hefty machine screw, 2 cm in diameter, with a rather coarse thread of about a 10th of an inch (2.5 mm). The bolt cross section is about 1/2 square inch. It is rated to be tightened to a torque of 500 Nm, or about 350 ft-lb of torque. If you put a handle on it that was 1 ft long, you’d get a force multiplier of about 60 times for downward force (6.28 inches of travel for the handle for 0.1 inches of travel for the screw). To get 1000 psi using just that screw and handle alone, you’d need to apply 33 lb/force (or so) on the handle. Use the bolt to drive a piston with a 1 square inch bore, and you’d just need 15 lb/force or so.

The M20 screw isn’t designed for that application, and probably isn’t be best choice. It also doesn’t have a lot of travel.

From your brief description, I’m envisioning a contraption that is basically a tube with a piston at one end and an RO membrane at the other. Fill it with water, apply 1000 psi pressure with the piston, and pure water drips out the other end. You want it to be light and portable.

Unfortunately, standard PVC pipe is out, because 1000 psi exceeds the maximum pressure rating for Schedule 40 or Schedule 80 PVC. A 2" diameter Schedule 80 carbon steel pipe gives you a pressure rating of about 5000 psi, so it’s good for this application, as long as the weight isn’t too much.

Since RO “wastes” about 4 times the water it purifies (wastes in this case means turns into a more concentrated brine), you’d want the volume of the vessel to be at least 5 liters, or you’ll have to empty the waste and refill it several times to get a liter. This may not be a problem.

If we go with a screw idea, I’d use a threaded rod fixed to the piston instead of a screw, and use a captive nut pushing against a thrust bearing. Turning the nut will drive the rod/piston, and give you lots of mechanical advantage.

Or, for a different approach, take that $40 bottle jack, and disassemble it. At its core, it’s a hand pump to push hydraulic fluid from one reservoir to another at high pressure. drain out the hydraulic fluid, remove the pump itself, and use it to pump water at high pressure into your desalination chamber. It’s probably not the best pump for the application, because it’s designed to work with lubricating clean hydraulic oil, not corrosive dirty brine.

But alternatives do exist: I was able to find a hydraulic tool manufacturer who makes a line of pumps and jacks for use in medical, pharmaceutical, food, and mining applications where water is a better choice than hydraulic fluid for safety reasons. They have a hand operated water pump that can do up to 72 MPa of pressure (10,000 psi), well more than you need. It is also 5 kg and probably dreadfully expensive, but it validates the idea that such pumps can and do exist. I found a different site that makes stainless steel hand hydraulic pumps that can be used with water up to 1000 PSI. It’s worth looking at non-hand hydraulic pumps as well, especially ones which are shaft-driven. You can mount a hand-wheel on the shaft, and it becomes hand-powered, and probably cheaper than a designed hand pump (which tend to use levers, not wheels, to drive them).


haha, AWESOME!!! Thanks, Buddha – you rule!