Centrifugal pump anomaly

[roricaja23]

I have a system that uses a 480V 3ph motor driving a submerged centrifugal pump. The system is flushing at the moment and the discharge goes to 8 different bearings. The pump when deadheaded is right on with the pump curve. When all of the valves are open allowing maximum flow, the pump also matches the pump curve. As the valves are closed one by one, there for restricting the flow, the pump discharge pressure increases as expected. The anomaly here is that the amp draw on the motor increases as well. I am dealing with bearings at different elevations but the kicker is that the last valves to be closed at those at the highest elevation which should even more so decrease the flow and decrease the amp draw on the motor. Any ideas?

 

[impeller1]

You can probably check the Hydraulic Power curve for the pump. On some curves there is a very small area of the curve where the power required actually increases as the flow decreases for a small range. This would require more current draw from the motor.

 

[SteveWag]

I am assuming a vertical pump with impeller(s) supported by a line shaft, ? If so as flow decreases, motor current should decrease. I have seen low flow-high head conditions result in line shaft stretch and the impellers contacting the fixed bowel assembly, with an increase in motor current. But this does not equate with your operation at shutoff. Did you measure shut-off current?
Steve

 

[CH5OH]

sounds like a vertical deep well pump....
is it an open impeller design?
sometimes, you have to look at internal leakages to explain kW disappearing

 

[roricaja23]

The pump has a close type impeller with a balanced chamber design for the control of the axial thrust. It is not deep well, this is a lube oil supply where the pump shaft is approx. 4 ft long.If i had internal leakage, i would not hit the curve when the system is deadheaded. The unit was tested at our facility and performed as you would think. We simulate the head with a offskid gate valve. Now that the unit is in the field, the bearings are controlling the flow. The current when deadheaded was measured in the field and equates to the pump curve BHP. The pressure as well is right on the curve in head. I plotted the data from the field and as the flow is restricted, the pressure increases along with the amp draw so I dont think that I am hitting a sweet spot on the curve. Things that make you go HUH.

 

[BigInch]

Could be a symptom of poor motor efficiency when the motor is only partially loaded. Maybe you have way too much motor for what you actually need there.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[SNORGY]

Take a look at the pump efficiency coincident with the characteristic as well.

Regards,

SNORGY.

 

[Artisi]

You need to establish the flow rate thru'the pump at the various settings of the valves - this is the only way you will make sense of what is going on.

 

[BigInch]

Very true, a low flow to a high valve, might have the same or more power requirement as a high flow to a low valve. But, since the OP says they are tracking the pump curve, you can reasonably presume the flow is also what is indicated on that curve and hydraulic power can be calculated. Problem is that the unknown (to us anyway) efficiencies do not allow us to calculate the corresponding amperage. We can only assume the OP knows the efficiency of both pump and motor and has accounted for those effects. What I assumed is that he may know the efficiency of the pump, since he apparently has the pump curve, but does he really know the efficiency of the motor, or has he only assumed 0.94 or something, and does he really know the efficiency of the motor at reduced load, if reduced load is an issue. A very lightly loaded motor might have only 60% efficiency, or even less. To try to do this with only amp readings, you must also have good knowledge of the motor efficiency curve at all load ratios. (Which is incidently the same reason that many VFDs fail to save expected energy levels at low rpm and flows).

So far the story here seems to be that he's on the backside of the power curve, which would typically indicate he's operating out of the normal range of something.

So, roricaja what say you?

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[roricaja23]

I appreciate everyone input, it gave me many avenues to consider. My recommendation to this customer is that he is in a flush configuration right now and that is not what the system was designed for. I am going to have him proceed as normal until the cleanliness criteria is satisfied. If the condition is present after all pressure regulators, orifices and misc. components are restored, we will pursue corrective action. There are to many unknowns to pinpoint a cause. Thanks again all.

M.N.Ball
Process Engineer
CSSBB

 

[itsmoked]

When you clamp an ammeter around a cable feeding an induction motor you are not reading just the current attributed to the work being done. You are reading the vector sum of the current actually doing work AND the magnetizing current necessary to excite the rotor.

As you reduce the load on an induction motor the power factor gets worse. Often MUCH worse. This greatly increases the motor's excitation current demand. This causes the clamp-on ammeter to see little decrease and occasionally, even an increase - as is happening here.

If you clamp-on and see the current reduce as you lower a motors load fine! You see what you expect, though it's not really correct. If you reduce the motor's load and see little reduction or an increase, then that particular motor's power factor changes more than the typical and you need to pitch the ammeter and switch to a power meter. Specifically a Watt meter.

Get a hold of a Watt meter roricaja23 and repeat your test. Tell us what you see then.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MartinSr00]

Pumps don't a priori have rising horsepower curves from shutoff. I don't remember exactly how this goes, but my sense of it is that very high specific speed pumps (axial flow, propeller, etc.) may have horsepower curves that rise as you move from duty point to shut-off.

This may not be an anomaly at all.

 

[BigInch]

Wouldn't think so on a lube oil pump.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[CH5OH]

you should really perform a kW measurement on the motor,
for small size motors, cos phi is everything but constant.
so with a rising amp meting and a falling cosinus phi,
the kW might be falling

 

[itsmoked]

roricaja23; Do yourself and your customer a favor and tell them the real problem as I stated and CH5OH just repeated. Otherwise your customer will be wasting his time.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ccfowler]

roricaja23,

You may want to check your pump curve and adjust for shaft speed variations. Commonly, the pump curve is generated for a constant shaft speed, often a synchronous speed such as 1800 rpm for the 60 Hz part of the world. Since you are presumably using an ordinary induction motor, the "slip" (which, of course, varies according to the load) can have enough of an effect to make some of the numbers look a bit off.

Was the pump curve that you are using established by a special test by the manufacturer, or is this a generic curve for this pump series? If it is just a generic curve, some modest variations in performance can be expected.

Similarly, how well do you know the motor's actual performance characteristics. Are you working from nameplate data, or do you have actual performance test data.

 

[Artisi]

Have you checked for mechanical fault, you say as the pressure increases so does the power, is the impeller being thrust into the case due to high axial load due to insufficient clearance - worth a check.

 

[itsmoked]

The point is being missed here! Theories over, "what's causing a power increase", are pointless against the fact the POWER HAS [!]NEVER[/!] BEEN MEASURED!

Furthermore the OP has left the room...

Keith Cress
kcress -

http://www.flaminsystems.com

 

[BigInch]

me too.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[vpl]

Thank you Keith

Patricia Lougheed

******

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