Over current issue of a motor

[Electrometrical]

Hi,
We have 450HP, 400RPM, 4160V, 16 pole SCIM. This motor is directly coupled with a make-up gas compressor. Its nameplate says FLA as 81.5A. However, this motor never run at this FLA rather it always run at 88A. Manufactur was recently contacted and he said that the measured FLA was 88.9A @ 0.586 pf. Manufacturer has no satisfactory answer on why the PF is too low at full load. Relay was also set (115%) as per 88A not 81.5A. Two years ago, this motor was also re-wound. Recent routine monitoring shows that FLA is 94A @ 100% load and 85A @ 50% load. Original design had Class B but after re-wound it has now Class H insulation.

Concerns were raised that why it is drawing more than rated FLA and secondly would be safe to run continuously at 94A? Currently we do not RTD connected to show any temperature data.
Other than this high current issue this motor used to have no performance related issues. Bearing and other mechnical checks shows there are no mechnical issue with this motor. Another interesting finding was NLA which was measured as 77A and the manufacturer says that it is because of high number of poles. There is a discussion on replacing this machine or keep running it. How you see it? are there any test or calculations you recommend to do?

Any comments?

Thanks in advance.

 

[electricpete]

It is certainly correct that slower motors tend to have lower power factors, but this seems extreme to me.

You mention motor was rewound 2 years ago, did it act differently before that?

Are the currents from all 3 phases balanced?

Do you happen to have a winding diagram?

You might try a surge test since this can reveal a misconnection of the coils or groups.

Recent routine monitoring shows that FLA is 94A @ 100% load and 85A @ 50% load

How did you determine the load?

=====================================
(2B)+(2B)' ?

 

[PHovnanian]

How do the motor terminal voltages look? Power factor (and currents) go up as voltage goes down.

 

[Electrometrical]

There seems no difference in performance before and after the rewinding.

Voltage seems ok. I did not check current imbalance but I will do it soon. Previous history though does not indicate any sign of imblanace

I do not have winding diagram but certainly I can check for it.

We have checked for CT saturation by taking measurement at motor but current were still the same.

Similarly, the gas composition were checked if molecular weight or composition is changed but there is no change as well.

I do not know how operations check % load and determine if it is 100% or not. I can ask them and let you know.

Another question is that it is firm fact that pf of this motor is surprisingly very low. So if we install a capacitor bank in addition to imroving pf, would it lower down FLA too?

 

[electricpete]

Here is one datapoint. We have a 324 rpm (60hz, 22 pole) 13.2kv induction motor purchased in early 1980's. Datasheet attached.
FLA - 153A. NLA = 61A, LRC = 990A.
Slide 3: Pf at 100/75/50/25 is 80.8%/76.0%/68.4%/48.8%
Slide 4 shows efficiencies at same data points.

So, my motor which is slower speed than yours has far higher full-load power factor (80.8%) than yours (0.586). Of course it is larger motor which may tend to be slightly higher efficiency and power factor but to my knowledge there were no special requirements imposed on power factor during purchase.

In fairness, motors are unique. Who's to know. If you don't specify it, you take what you get. It sounds like maybe the OEM didn't expect the low power factor they got since they nameplated the motor for 81.5A, but full load amps ended up being 88A... presumably due to low power factor.

Since it is a recip machine, for completeness, I'll mention that if you have torsional oscillation occuring, the curretns and voltages may not be sinusoidal and the concept of power factor gets a little tricky and some instruments can be fooled. One example shown here:
thread237-249262

=====================================
(2B)+(2B)' ?

 

[electricpete]

So if we install a capacitor bank in addition to imroving pf, would it lower down FLA too?

It will lower the current seen upstream of the cap (perhaps at your protection), but it will not lower the actual current of the motor. So if you are at your 100% load point, the motor presumably still has 94A and operating above its FLA... whatever concerns you have for overheating are not changed by adding a cap.


=====================================
(2B)+(2B)' ?

 

[waross]

I wonder if the air gap is too great. That may account for low power factor and the lower power factor would cause the full load current to be increased.
Occasionally someone thinks that it will be a good idea to clean up a dirty rotor and the result is an increase in the air gap.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Electrometrical]

Both the manufacturer as well as the company who re-wound the motor suspect that large air gap is causing high magnetizing current (i.e. NLA) and because of this the motor is running at higher FLA. Since re-wounding didnt result into any significant change in FLA thus it is hard to grasp that re-wounding have alter the magnetic wedge design or alter the rotor etc. The motor is exhibiting the same behaviour since it was first installed in seventies. So, if we suppose that the airgap was large as per peculier design of the motor then whatever we have is basically how the motor is designed. The question is that do we keep running it and would it be safe / reliable or replace it (over $110K job!).

Another question is that can we do something to power supply system or to the motor itself to lower down the FLA?

Second is that would it be safe to operate the motor at 94A while its FLA is 89. Service factor is not known but I assume that it would 1.0 (original design). However, since after re-winding its original Class B insulation was replaced with Class H insulation thus, I am tend to believe that it could now be considered to have a service factor of 1.15; hence, my calculation it may run to have FLA up to 96A.

Windings have RTD but unfortunately, we have no means to measure the temperature at MCC since they (RTD) are not connected and will not be connected in near future due to operations' constraints. Can temperature noted at motor with the help of contactless temperature gun would provide some meaningful temperature?

 

[waross]

What are your ambient temperature conditions. In a cool climate you can safely run some overload but very hot days may negatively impact the life expectancy of the insulation.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[electricpete]

Yes, the very high NLA would rule out torsional oscillations assuming the machine was uncoupled at the time of the measurement. Also would tend to rule out leakage reactance factors and focus on magnetizing reactance.

There are a lot of factors that can affect magnetizing reactance. One assumes that the motor OEM is particularly conscious on a slow speed motor to take steps to counteract the natural tendency for low power factor. Most likely a low speed small motor may be a fractional slot design. For fractional slot design, I thought it would be interesting to examine to see the effective distribution factor. Very low distribution factor (and pitch factor, collectively referred to as winding factor) could have a pretty big role in power factor because the magnetizing reactance is proportional to the 2nd power of winding factor. It is the 2nd factor (rather than 1st) because if we are examining effectiveness of magnetizing current in causing stator voltage, we need winding factor to convert magnetizing current to flux, and we need winding factor again to convert flux to voltage. That is one reason why I asked for winding diagram.

But again that (low distribution factor of possible fractional slot winding) is only one of many factors. The more data you provide (turns per coil, dimensions etc), people here could try to analyse it to see if the low power factor is truly expected. Although I'm sure some would question how productive that is.

I suspect the example I provided (0.8 pf at full load for 22-pole motor) was on the high side compared to other slow motors. I'd be interested to hear if anyone else has data on slow speed motor full load power factors.

Another question is that can we do something to power supply system or to the motor itself to lower down the FLA?

As already discussed, adding caps would be treating a symptom, doesn't help the motor.
VFD doesn't sound like much of an option either. Theoretically, if you run the machine above base speed with voltage limited to nameplate, you can get same horsepower output with reduced magnetizing current. However that same horsepower corresponds to lower torque at higher speed... don't know if your load is adjustable to accommodate that. Also we're talking big, big $ for mv vfd. so I think it is pretty much out of the question, but just included here for completeness.

Second is that would it be safe to operate the motor at 94A while its FLA is 89. Service factor is not known but I assume that it would 1.0 (original design). However, since after re-winding its original Class B insulation was replaced with Class H insulation thus, I am tend to believe that it could now be considered to have a service factor of 1.15; hence, my calculation it may run to have FLA up to 96A.

You're right, the Class H insulation bought you back quite a bit of margin. And possibly low ambient can be credited as Bill mentioned. But also the fact that original nameplate was 81A should be factored into the equation......I know you said the OEM is telling you FLA is 88A, but is he also saying it still meets class B temperature rise on those conditions (full load, 88A)? And further was there an actual temperature rise recorded during factory tests (if less than Class B, that is margin you might also claim back). Actually if you have full factory test data to post, that provides the best opportunity for someone to double-check your thought process. However as much as we sharpen our pencils, in the end I believe the limits are not exact and there is lots of room for judgement.

Windings have RTD but unfortunately, we have no means to measure the temperature at MCC since they (RTD) are not connected and will not be connected in near future due to operations' constraints. Can temperature noted at motor with the help of contactless temperature gun would provide some meaningful temperature?

Temperature gun will give some data. It probably won't tell much about the winding, but it may tell a little bit about the bearings (which also have thermal considerations).

You should be able to go directly to an auxiliary terminal box on side of the motor (assuming it is separate from main terminal box) and get access to measure the winding RTD's from there.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Both the manufacturer as well as the company who re-wound the motor suspect that large air gap is causing high magnetizing current (i.e. NLA) and because of this the motor is running at higher FLA.

What is the airgap dimension?

=====================================
(2B)+(2B)' ?

 

[electricpete]

A few more datapoints:

100hp, 650rpm 460vac motor. full-load PF = 0.858

http://www.reliance.com/pdf/pdf/aced/L2226A-V.pdf

250HP, 650rpm, 460vac motor, full-load PF = 0.791

http://www.reliance.com/pdf/pdf/aced/L2818A-V.pdf

=====================================
(2B)+(2B)' ?

 

[Electrometrical]

After re-winding, with regard to airgap, we were told by the re-winding company:

* The air gap flux density is nearly 70 kl/in2 and our usual limit is about 65 kl/in2.

* Attached, please find Stator Winding.

* Load adjustment to control FLA is not an option since as per operations this compressor has to run at 100% load in order to meet production targets.

 

[Electrometrical]

Attached, please find the factory test data sheet for further discussion.

Ambient temperature are not high (on average Min. -10 and Max +20 degree C).

Regarding your question of how do we determine about % loading of compressor; operations indicated that they have pressure / flow guages that help them to determine how much would be the load on the compressor.

Regarding the possibility of winding temperature, I have asked maintenance guys to check and I will revert as soon as I get some information.