A few thoughts on your post. First, the rated efficiencies used by most motor manufacturers are the worst case values. Specifically, not every motor produced is efficiency tested, the rated values are based on testing and design criteria and are conservative values that all of their motors will meet or exceed when produced. That being the case, it is more likely that you will find a motor that exceeds rated efficiency rather than failing to meet it.

Second, it would appear that you are using motor amps, pump curves, and measured pump parameters (pressure/flow ?) to determine pump efficiency. I know enough about pump systems to realize that it is a very complicated topic beyond my expertise. I do know that the pump curve alone is insufficient in that the system head (suction and output) and system static and dynamic losses must be considered. With respect to the motor (more comfortable ground for me), to determine motor input power you need to measure motor terminal voltage as well as amps. Amps alone cannot be used to determine motor power output as any change in terminal voltage results in a corresponding change in current to maintain constant power output.  I hope this helps.

rhatcher,

I do know that these pump sizes were used in other customers plants with no problems.  I realize that the motor voltage needs to be used to determine power but our customer will not open or pig tail the enclosure to find out the true volts. The equation I used to calculate motor power was:

Power = (1.73 x amps x volts x P.F. x Eff.)/746

Water horse power was taken at the TDH and flow of the pump, we verified the flow by pitot. The pressure diff.'s were determined by gauges.  I'm not sure I trust the pressure gauges but it's all I have to go by for now.

You said your expetise is electric, any comments or suggestions?

Some one told me that if the incoming power to the transformer had one leg out of phase we could have problems.  Any comments?

Please let me know if there is anyways to determine if it is an electrical fault.

Stan K.

It sounds like you are measuring all of the important pump parameters although I wonder:
- Are you pumping water? Most pump curves are for water, a fluid with a specific gravity other than 1.0 would require an adjustment to the pump curve. Note that adjustments may be required for salt water or brackish water.  
- Doesn't rpm factor into one of the curves used to evaluate a pump system? If so, you may be able to use that data to more completely evaluate the pump performance. As well, this data can be useful for evaluating the motor loading

With respect to the electric side, I think that it is very unlikely that one leg of your transformer is "out of phase" if you mean that it does not have the proper 120 degree phase separation. A shorted leg or a leg with a lower primary supply voltage would be more likely, but this would be reflected in unbalanced phase currents (did you measure all three phases and use the average value?). I do recommend that the motor terminal voltage should be measured as this info is required if you want to do more than estimate the motor's power output. Most facilities do not have voltages at the nominal values. If you allow for a possible +/- 10% variation on the motor terminal voltage, how does that affect your final calculation for pump efficiency?

Finally, as implied by my last statement, the accuracy of your calculations will be limited by the accuracy of your input data. I know that this is an obvious statement but you should give serious considerations to the implications in your case. Do the above exercise again except allow for a possible +/- 10% variation in the readings of the pressure gauges that you don't trust. Finally, combine the effects of the worst cases (ie lowest final pump efficiency) for voltage variation and pressure inaccuracy and see where the results leave you. Either of these variations is possible in your system and the effect of one or both (or any other inaccuracies) may account for the appearance of loss of efficiency in the pump.  

I agree with everything rhatcher said. Calculating efficieny of a motor in the field is a tough job. If you didn't measure voltage, then you also didn't measure power factor.  Did you assume a value for power factor?  (If you assumed 1.0, that would explain why you calculated an efficieny of 0.8).  

You'll find a boat-load of motor efficiency calculation info at DOE's motor challenge site http://www.oit.doe.gov/bestpractices/motors/

If you want more specific and useful help from this board, I would recommend that you post your calculations so we can see what you have done.  

A brief overview of the process of evaluating the efficiency from my standoint:  Measure electrical input power precisely using a suitable instrument which measures voltage, current and power factor (while you're at it you'll be able to check for voltage imbalance to confirm your suspicions).  Measure pump output power based on flow-rate, differential pressure, and fluid density.  The difference between pump output and motor input is due to the combined effect of motor and pump inefficiency.  It can be tough to tell which one is causing the problem.  I know you had pump efficiency data but you also had motor efficiency data.... why did you assume it was the motor and not the pump that was causing the problem. Pump performance can degrade for a number of reasons.... first and foremost due to wear ring problems. Also sometimes unexpected recirculation paths, debris in the system, or possibly lots more.  

Be skeptical of your meausrements... double-check them with other plant indications or test equipment when you can.  For example your on-site reliability folks may have the capability to do non-intrusive ultrasonic flow measurement which is a fairly accurate double check of your flow indications.  

Measure the speed of your motor with a strobe (or from high-resolution vibration data).  Compare to expected speed from speed torque curve (using 2PI*torque*speed=power).  If that curve is not available, it's usually fairly accurate to assume that slip varies linearly with power.... so you'll have zero slip at no-load, and nameplate slip (=[syncronous RPM-nameplate RPM]/syncrounous RPM) at full load.  Rotor bar problems may cause a motor to run at an RPM lower than we would expect based on the input power.  Easy to check for rotor bar problems using current signature analysis.

Suggestions: In addition to the above postings:
1. Check voltage and current harmonic contents. If Total Harmonic Distortion (THD) of power supply voltage is below 5% then it is in compliance with industry standards (IEEE Std 519). If not some THD mitigating means should be considered.
2. The power factor of motor load receiving power from harmonically contaminated power supply is somewhat lower causing the efficiency to be lower.
3. The voltage harmonic distortion may come from some close-by nonlinear loads, e.g. large ac motor drives, large AC-DC converters, etc.