I want to verify the pump curve of a regenerative turbine pump. The manufacturer doesn't supply any pump curve of it.

This is the test setup.

enter image description here

With the discharge valve shut off (plugged) and water line opened. I measured the following discharge pressure of 28 psi.

enter image description here

This is the inlet or suction pressure of 14.3 psi.

enter image description here

Is it correct to deduct the discharge pressure from the suction pressure to arrive at the head dead pressure? It is 28 psi - 14.3 psi = 13.7 psi.

So 13.7 psi is the pump dead pressure right?

I made a lot of measurements with pressure gauges and water flow rates to get some idea of the pump curve which the pump doesn't have

This is the pump specs. 8 is the meter in heads or about 11.7 psi.

enter image description here

enter image description here

This is my bathroom minipump connected to a heater and shower with 2 gauges label by A and B. The green arrow is the water pipe entry to the bathroom with 1/2 diameter hole. It can produce 10 liter per minute (2.64 gallon per minute) when directly measured with a container without all the hoses which are 1/4" in diameter (note all tankless heaters use the same 1/4" flexible hoses).

Without the pump turn on. I'm getting 4 liters per minute (about 1 gallon per minute) of water in the shower. With pump turn on. I'm getting 6 liters per minute (about 1.6 gallon per minute of water in the shower).

With and without the pump turned on. The gauge in the bathroom pipe entry labeled in A read 13 psi.

Without the pump turned on. The pressure gauge at B (after pump) reads 10 psi. With the pump turned on. The pressure gauge (after pump) reads 20 psi.

Do the actual test data above support the pump specs shown above?

From my existing water of 6 liter/minute (1.5 gallon per minute). I want it to be at least 2.5 gallon per minute (or 9.4 liter per minute). So i need to understand pump curve to be sure i got the right one.

  • $\begingroup$ “8 is the meter in heads”? Sorry I am confused, was it your translation? $\endgroup$ – Solar Mike Oct 4 '20 at 13:33
  • $\begingroup$ Its the rated lift of 8 meters. The 10 meter is maximum lift. $\endgroup$ – Jtl Oct 4 '20 at 13:58

To build a pump cure (especially when you don't have many data points), first look at an existing pump curve for the pump technology you are inspecting. Here is one for a Regenerative Turbine Pump.

Your 13.7 psi dead head pressure is correct based on your data. This is one point on the graph.

You have also evaluated one running point: 1.6gpm. In this condition the pressure was increased from 14.3 psi to 20psi. 20psi-14.3psi = 5.7psi differential. This is also a point on the graph.

Generally a pump curve is evaluated with lots more points, and going from atmospheric pressure suction to a gate valve that can be adjusted. This lets you easily move across the curve and doesn't require any math or second pressure measurement. The last point (at minimum) you need is the unrestricted flow. With an atmospheric pressure intake and an atmospheric pressure outlet, measure the flow.

I have sketched a table and rough plot. Once you have your data you should plot it in a spreadsheet with a trend line. What I have drawn is just a guess and not to scale.

enter image description here

Another thing to consider is that your shower head nozzle also has a pressure flow curve. Don't mix the two. It can be evaluated exactly the same way though by measuring pressure going into the nozzle (the other side is atmosphere). Then record the flow and pressure into a table. Repeat this for various pressures/flows and graph.

  • $\begingroup$ But WITHOUT the pump turned on, there is already a pressure drop between the inlet and outlet due to the long hose and impeller due to turbulence. The inlet is 14.3 psi. But d pump outlet is just 10 psi. This s wit water running in the faucet. When I turned on the pump. The outlet pressure is 20 psi. Therefore isn't the pressure differential supposed to be taken at the outlet, that is, 20psi - 10 psi = 10 psi,so shouldn't the differential pressure or increase be from 10 psi to 20 psi measured from outlet instead of from inlet to outlet of 14.3 psi to 20 psi producing 5.7 psi as you mentioned? $\endgroup$ – Jtl Oct 4 '20 at 21:07
  • 1
    $\begingroup$ The internal flow resistance in the pump is something to consider in your design if you will be operating with and without the pump. However, when building a pump curve you just consider differential pressure across the pump and flow. A pump curve is independent pipe resistances and plumbing configuration. If you have large resistances in you test setup, then the numbers will be skewed. $\endgroup$ – ericnutsch Oct 5 '20 at 1:42
  • $\begingroup$ Ok. I measured the unrestricted flow with nothing connected to the discharge side of the pump and the flow rate is 3 gallon per minute pump on (compared this to 5.7 psi at 1.6 gallon per minute). This would create end points of the pump curve. Thanks. But why did you mention that "With an unrestricted atmospheric pressure intake and an atmospheric pressure outlet, measure the flow"? My inlet is not atmospheric pressure but from the water company with 14.3 psi pressure". $\endgroup$ – Jtl Oct 5 '20 at 1:53
  • 1
    $\begingroup$ Your original question is about figuring out the pump curve. Even though one was not provided to you, a pump curve can be made by anyone with the pump. The curve is system independent; and that is what makes it useful and comparable to other pumps. If you buy a different pump, it will have a curve that is independent of your system, and you can compare them. $\endgroup$ – ericnutsch Oct 5 '20 at 2:31
  • 1
    $\begingroup$ Well. I put the inlet pressure gauge right at pump and its much closer now in preasure for inlet and outlet. So you are right. Its the piping and plumbing resistances. Anyway i realized by changing to bigger pump. Ill starve the main so id be satisfied with this pump. Thats it. I think thats enough plumbing and physics to learn for now. Thanks very much for all the assistances. $\endgroup$ – Jtl Oct 5 '20 at 5:19

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.