finding flow from head 1

[Sambrose]

What methods are there for determining the flow from the change in pressure accross a multi stage centifugal pump? This data will be used for determining efficincy f the pump.

 

[Sambrose]

Just thought you might want some more data. 170m Head of pressure. Pipe diameter 500mm.

 

[PUMPDESIGNER]

Are you sure you want to use pressure differential?
Accuracy will always be low, very low sometimes, and sometimes your method totally fails with no hope at all.

Some of the problems:

1 - Changes in pump rpm due to load and voltage differences. This is one of the minor variables.

2 - Most pumps operate with some cavitation caused by low flow rates, insufficient NPSH, and incipient cavitation. Cavitation intereferes with flow and pressure causing variation from the "norm".

2 - You will be forced to do one of two things to calibrate your method: use the pump curve as published by the manufacturer which typically has high variations, or calibrate your pressure differential with a flow meter.

3 - If you have a relatively "flat curve" pump you cannot even begin to produce anything accurate except at specific and small areas of the curve out near and beyond BEP where the curve usually dives downwards (pressure varies with flow more).

4 - If you have what is called an "unsteady" curve you will have portions of the curve where you cannot possibly predict the flow by pressure differential. These areas may be near shut off, in the middle of the curve, at BEP and also beyond BEP.

Flow Meters are not subject to any of the above problems, they just simply tell you the flow rate with any possible output you may desire available on the market, assuming the flow meters is correctly chosen and installed.

The differential method could be made to work on specific pumps, but I personally would not spend the time because of the terrible inaccuracy even though it could be done easily on a spreadsheet by comparing the flow to pressure data obtained from your calibration or from the manufacturer's curve. But how will you handle the fact that most pumps will have the same exact pressure differential for many different flows, especially at the low flow end of the curve, or at any of the "unstable" portions of an unstable pump curve?

The pressure differential method is similar to the "Amp" or current method. At certain flow rates amperage can be used to predict flow, but at other areas of the flow range amperage is useless.

PUMPDESIGNER

 

[vpl]

This question was asked by Ivan69 a couple weeks ago. Please look at thread407-79853.

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.