Finding NPSHr - Varying Pump Height 1

[justomi]

Hi,

I have been given the task by my company to find the NPSHr for our pumps.

The way I have been carrying out the test procedures is by raising the pumps above the vessel it is connected to and to record the TDH as the inlet pressure decreases.

The results I get are bizarre as I expected the TDH to stay steady and then drop as the machine is lifted. But what has happened is that the TDH rises as the machines reaches the top of the vessel, and then drops as the machine rises higher then the tank and liquid level.

So I have recorded the NPSHr to be when there is a 3% drop in TDH from when the TDH is at its peak. But the 3% TDH drop value when raised above the vessel is the same as the TDH value when the machine is on the ground. So this has confused me.

Is there a reason for why this happens??

I have attached a diagram of my set up.

 

[LittleInch]

That's because your end point hasn't changed. Hence the head available at the inlet goes down, but relative to the end point the outlet head doesn't change. The head across the pump remains the same and any reduction in head / pressure caused by raising the pump is matched exactly the reduction in head at the outlet caused by raising the pump because the end point hasn't raised the same amount ( or at all)

The way pumps suppliers do it is to throttle the inlet until your Total Developed Head reduces to lower than 3%.

That way the only thing to change is the inlet head, not anything else.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[eduvillela]

Hello, justomi

let me ask a few questions first. How excaclty do you measure the TDH? Are there pressure gauges? I'm asking that because I didnt notice them at your sketch.
By raising the pump, you end up changing the pump line geometry, and hence you cannot calculate the head loss at suction line anymore. So it becomes harder to calculate the NPSHR. I guess that this changing in the line geometry might be causing this problem you're facing.

 

[1gibson]

Put a pressure gauge right at the pump inlet, and fit some sort of clamp on the flexible hose to slowly crimp it down. This may or may not work depending on the size and type of the hose, but I assume this thing is pretty small if you just lift the whole pump, hoses attached.

 

[justomi]

Little Inch,

Thanks for your reply! The reason I tried this method is because in the literature it describes two methods, sump test method changing the height and the suppression test method using throttle valves. I have tried the suppression test before by throttling the inlet but it was very difficult to keep the flow rate constant with a valve on the outlet and the results were not accurate. What arrangement would you suggest for the sump test where you vary the machine height? Thanks

eduvillela,

Sorry I didn't put them on the diagram. I have put pressure gauges on the inlet and outlet of the machine. I have measured the the TDH by taking the difference.

 

[Artisi]

The easiest way to establish the NPSHr is to review the original pump curves that would / should have been supplied with the pumps or spelt out on the pump data sheets or contact the pump manufacturer. NPSHr is not a hit-and-miss arbitrary bunch of numbers but established by carefully controlled testing in a certified test facility.
Trying to test on-site without calibrated gauges, flow measuring instrumentation, control valves, etc. is just a waste of time leading to meaningless results -- as you have already discovered.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[justomi]

Hi Artisi,

I do think it can be carried out on site as I do have calibrated gauges and flow measuring instrumentation. I am just not sure whether my arrangement/method is correct. If you could offer any suggestions that would be helpful. Thanks.

 

[LittleInch]

OK, to make sense of this you need to do a proper sketch and show us where the water discharges to. and the relevant elevations of it all.

The issue you might have is that you're trying to do this inside an existing system.

On a test rig, you normally need to throttle the outlet to keep the pump inside the performance curve. Hence if you throttle the inlet valve, you can open the outlet control valve to compensate and hence stabilize flow. this becomes much more difficult if you have fixed discharge piping providing all the back pressure.

However I fully agree with Artisi, you would be much better off finding the serial nos / types of your pumps and tracking down the manufacturers pump curves. Far easier and more accurate.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

Suggest you Google "NPSHr testing of pumps" there is a wealth of knowledge there for the reading. To start a new discussion on NPSHr testing is just reinventing the wheel.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)