Monitoring pump operations

[kbs]

When monitoring normal pump operations (primarily hydronic centrifugal pumps), which gives a better indicator of a pump "wearing out", motor amperage readings or pressure diff across the pump? By wearing out, I mean impellor deterioration, seals tightening up or packing glands overtightened, etc.

 

[monaco8774]

KBS

the problem is flow, if you dont know what the flow is then high amps could be caused by running down the pump curve and increasing flow without any wear on the impeller and reduction in pump pressure differential could be caused by the same thing. Best indication in my book if you dont have flow indication is to do a pump shut in head test, for a single stage impeller this is not a problem for the pump and gives a guaranteed point on the pump curve, then you just look at the deterioration in pump differential pressure.

 

[kbs]

monaco877

I appreciate your help. For single stage pumps, would a shutoff head test done at regular intervals (during PM) show these problems developing? Also, once the shutoff head test gives me a point on the curve, can I use the press diff to tell flow from this?

 

[pmover]

continuously operating any centrifugal pump in a poor hydraulic region MAY eventually result in accelerated failure of rotating or wetted parts of the pump. a properly instrumented centrifugal pump (inlet/outlet pressure gauges) and having accurate pump performance curves for the pump/pumpage will give any end user the means of determining pump performance.

a shutoff head test is merely one indicator of overall head developed at zero flow. obtaining motor data (voltage and current) will also further aid in motor performance. yes, the pressure differential, knowing the fluid properties, and having accurate pump performance curves should enable the end user to reasonably determine pump performance.

a suggestion, part of the overall health of any rotating equipment should also include monitoring the bearings (i.e. vibration monitoring). not only should process conditions be periodically monitored, but if the pump or machinery is of critical service, some type of vibration monitoring should also be conducted. not to mention having a spare unit available.

good luck!
-pmover

 

[Ashereng]

Heat measurement are also very common.

A "worn out" pump may also run hotter?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[kbs]

Thanks for your input. Vibration and thermal monitoring is quickly becoming a very useful tool in our PM system.

 

[godlo]

When you are doing pump monitoring you need to know the power, and the way in which you do that, is you need a three phase measurement. If you measure using your amps and you have a three phase it wont give you the exact average. You need to measure the suction and discharge pressure, balance flow. From that you can generate a curve where you can see your flow. Also when commissioning a pump you need to do a curve so you would have something to compare your current conditions with the initial one.

Hope that helps on how you would go about monitoring a pump

 

[kbs]

godlo

Please forgive my ignorance, what is meant by "balance flow"? If possible, can you expand on "creating your own curve" with a pump? Our pumps (approx 120 or so) are not large output, ranging from single phase 1/4hp to 30hp 3-phase. However, we are upgrading our pools and buildings and I would like to readdress the planned/preventive maintenance work on the new replacement systems to be less reactive maintenance and more predictive. To do this, I think better monitoring of performance is key. In the past, most of the pumps would be torn down annually to "check the state" of the parts. I am hoping through monitoring performance we could cut out "fixing what isn't broken".
Long winded, but I appreciate the input.