Wednesday, September 2, 2020

Internal gear pumps for small-scale, middle-tech irrigation


Recorded rainfall for Charlotte, Michigan minus the potential Evapotranspiration. This chart suggests the amount of water that plants must mine from the soil OR the amount that must be supplemented with irrigation.

I have been giving some thought to "sustainable", off-grid pumps to pull water out of the ground to irrigate orchards and gardens.

There are many applications where the intermittency of solar or wind are a huge problem. Irrigation less vulnerable to intermittency than most applications because water can be stored as soil-moisture or in tanks.

My technical problem is that my water table is down about fifty feet and a cheap "suction" pump can only pull about twenty feet.

One option would be solar panels and a 24V diaphragm pump. hile efficient, that setup is orders of magnitude more difficult to produce than what the local manufacturing base can replace.

Old tech

The issue with old-tech is that the mechanicals of reciprocating pumps are heavy and prone to rust and breakage. They also have volume limits although they can lift fantastic distances. Another issue with sucker-rod pumps is that half of the high-wear components are deep underground.

Middle tech

Section A-A

There is a class of positive displacement pumps called internal gear pumps that have some interesting characteristics.

For one thing, as a positive displacement pump, you can apply pressure to the intake and it becomes a motor.

Another interesting characteristic is that it fits very neatly inside a pipe since its profile is very nearly round in cross-section.

Since it is a positive displacement pump, it can run slowly and still deliver usable amounts of flow. There are still some issues with internal leakage but compared to an impeller-type pump, it can still deliver water over a much, much greater range of power inputs.

This type of pump is sometimes used as an oil pump in internal combustion motors because it can be sleeved around a crank journal or any other rotating shaft.

They are typically made from sintered metal, extrusions or ultra-high-weight-polyethylene. They CAN be made from bronze or cast-iron, zinc/aluminum foundry alloy or even, conceivably, from glass.

The blue rectangles are internal gear pumps. Blue lines are water lines. Thickness of line indicates relative flow-rate. Black line joining two blue rectangles is common shaft. In reality, it might be smart to put the "motor" below the "pump"

The picture in my head is that a reservoir feeds an internal gear pump driven by a windmill. It is not necessary to have the windmill directly over the well.

The high-pressure water flows through a small diameter, high pressure line that runs down the bore-hole. The high-pressure water drives an internal gear pump/motor which spins a common shaft that drives a larger volume (i.e. longer) internal gear pump. 

Every turn of the shaft pushes more water up the return line than came down the high-pressure feed line. Some of the return water replenishes the water in the reservoir, the rest runs to a tank for further distribution.

Ignoring friction losses and kinetic energy for the moment, 300 psi water at the drive pump can can push up to twelve times as much return volume at 25 psi (every foot of head equates to about 0.45 psi) Since water from the well is the driving fluid, leakage of "toxic" working fluids is not an issue.

Since the set-up is only intended to be used in the growing season, it can be drained before freezing weather can ice it up and burst lines and reservoirs.


  1. The only issue I see is the inefficiency of the gear pump (and motor) and the delivery lines. Lots of losses there. It'll work, but not efficiently. Your energy losses might easily be over 70%. Positive displacement piston pumps or vane pumps might work better. The losses in the water lines that drive the underground pump are workable, but will still be high. There is a reason that hydraulically driven machinery has a cooler on the line. All those losses come out as heat. And there are a lot of losses due to friction, compression and turbulence.

    Good thoughts though. Thinking outside the box.

  2. Suction lift stops working at around 30 vertical feet.
    If you can get the pump closer to the water table you can make something work.

    1. My water has significant amounts of dissolved CO2 and I think I would run into cavitation issues before 30 feet.

    2. Yes, lowering the suction pressure would cause gases in solution to come out, but I was thinking more along the lines that by the time you draw a vacuum that will lift 30 feet, you are either at, or near a perfect vacuum, and that near perfect vacuum is what limits the suction lift.
      If every thing is perfect you could get to around 34 feet, but nothing in the real world is that perfect.
      I think that what B said about the total losses will cause your planned system to not do what you want it to do.
      I do not have the math skills to figure out if the windmill driven pump would work to drive the deep water pump.

    3. I apologize. I became focused on the suction lift and didn't really grasp that you were proposing a system of pumps and motors that used water as the working fluid.

  3. how about a ram pump ,we used one to lift water up to storage tank

    1. Ram pumps work on the same principle as a hammer. The water starts flowing in a pipe and a flapper-valve shuts. The momentum of the flowing column of water is like the head of the hammer. The pulse burps some water via a reed-valve into the feeder pipe.

      So what I see missing is the large mass of water that is flowing and can be suddenly stopped.

      This is my initial look-see as to why a ram-pump might not be the best choice in this situation.


Readers who are willing to comment make this a better blog. Civil dialog is a valuable thing.