75 HP or 100 HP

[MWeb7]

I have a vacuum system with a 75 HP pump running on 480 with a SF of 1.15. FLA is 86.

The motor is running overloaded at/above the service factor load (~99-100+). We had a contractor look at the pump and give us three recommendations:

Rewind the motor to a 1.25 SF
Buy a new motor with a 1.25 SF
Buy a 100 HP motor to handle the load

We have had the motor fail many times and recent testing resulted having the largest difference in resistance between T leads be about 0.72% (readings in mOhms: 83.2, 82.6, 82.8). Resistance to ground is over 200Mohms for each T lead. The motor is only a few years old and has run for 12 hours every 36 hours. Is 0.72% a big enough difference to be concerned?

I am weary about buying a new motor/rewinding a motor for a higher SF just to run it above the FLA rating. Would this cause potential problems such as the current ones?

 

[electricpete]

The resistive imbalance that you identified is not cause for concern imo.

You identified repetitive failures, but I would recommend to keep an open mind to the cause – may have nothing to do with loading.


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

 

[Compositepro]

Vacuum systems often see highly variable loads. Have you considered that a simple process change may solve the overloading problem?

 

[X49]

The service factor is a measure of periodical overload capacity. Your motor is not intended to be run higher than 86A continuously or for extended periods of time. A motor running continuously at a service factor greater than 1.0 will have reduced life. In your case, you're running at a service factor greater than 1.15 at times.

In my opinion, the proper solution to your problem would be to replace the existing motor with a 100hp motor. This may mean replacing the existing starter and cable, and is the most expensive solution, but should greatly extend the life of your motor (assuming thermal issues are your only problem).

 

[MWeb7]

Thanks for the responses. In our application the pump runs continuously at a fixed speed and goes into a bypass when the vacuum supply is met (which is not often). I do not believe we have much, if any variation in the load. We have gone through several ideas to reduce the load/operation, just not possible in our application.


I hear contradictory statements with regards to the SF - and I do not have a lot of knowledge when it comes to motor design. Are there motors that are designed for better cooling so that they can handle operation at the service factor? i.e. if they wind it in such a way to have a 1.25 SF, are they going to design it to have better cooling than a motor designed to have a 1.15 SF?

X49 - thanks for the recommendation - At this point that does sound like the best approach. I will have to check on the thermals of the motor and see if I am exceeding rated rise temp. What type of thermal impact would I have say I am running 10°C below the rating versus 20°C below - would that impact life expectancy? Or should all temperatures below the rating keep the life expectancy in the same general range?


Sorry for all the questions! Thanks again.

 

[rbulsara]

100 HP motor...based on what you indicate.

Rafiq Bulsara

http://www.srengineersct.com

 

[rbulsara]

I would however look at adequacy of the cooling or even enhancing it. If the failures were due to overload/heat.

Rafiq Bulsara

http://www.srengineersct.com

 

[itsmoked]

The life will greatly change over a 10C difference. Chemical reactions typically double their rates for every 10 degrees. Your failures are ultimately due to chemical changes in the insulation.

Is the motor in a hot poorly ventilated space?
Does direct sunlight ever fall on it?
Is the motor's case clean?

The voltage running to it will be a huge determining factor of its running temperature. You do NOT want the voltage to be low as that will will greatly increase I²R losses which will contribute directly to the heating problem. If the voltage is more than a few volts low from the nameplate you should probably get that corrected. If your facility is like a lot of places, during the hottest part of the day the voltage is the lowest which compounds the heating issues.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

The term "Service Factor" was for years very loosely defined and only recently was it semi-officially defined in the NEMA MG-1 design specifications to which North American motors are made. But outside of the US, motors don't even have any Service Factor.

That semi official spec now allows for the motor to operate at a 10deg C additional rise when running into the SF. The generally accepted rule-of-thumb is that for every 10deg C increase in rise, insulation life is cut in half (probably based on what Keith posted above. So although the specs do not specifically state a finite value for decreased motor life when running into the SF, it implies that it can be expected to be lower. It also mentions that efficiency, power factor, torque and speed ratings are to be expected to be other than as stated as well.

In other words, use SF at your own peril. It is meant to allow for occasional overloading conditions or temporary changes in voltage/frequency outside of excepted limits. But because the definition does not specifically state a time limit, many OEMs tend to use it and allow the end user to suffer the consequences, knowing they won't occur until the warranty expires.

By the way, I am trying to imagine how a motor can be "rewound" for a higher SF. They can use a higher insulation rise on the winding material, but I'm not convinced that necessarily increases the SF of the motor. You are not changing the amount of iron or the cooling method, all you are doing is allowing everything to run hotter and survive a little longer under more adverse conditions. But your efficiency and power factor may go so far out that it would be cheaper in the long run to use the 100HP motor in terms of operating costs.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[dpc]

I agree with jraef - the intent of the service factor is to cover temporary overloads and other variables. It's still a 75 hp motor. Also, the efficiency and power factor will not be at nameplate values at the SF amps.

I would definitely go with a larger motor.

David Castor

http://www.cvoes.com

 

[MWeb7]

Thanks for all of this input. All of it is very helpful.

 

[LionelHutz]

It should be a 100hp motor based on your description of the operation.

 

[Tmoose]

I'm interested in hearing about the "failures".

Dan T

 

[edison123]

SF is a marketing hogwash, imo, practiced only in US. IEC motors don't have them.

If your motor winding is failing is due to high temp (typically winding burn-out), time to replace it with right HP motor.

Muthu

http://www.edison.co.in

 

[electricpete]

S.F. is defined in NEMA MG-1, so it is not purely marketing.

I'll agree the interpretation of exactly what it buys you is not straightforward.

The op situation is not well defined but if there were a concern for loading I would be inclined to go with 100hp if it was available in the correct frame size.

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

 

[DickDV]

If the larger frame for the 100hp won't fit, a good motor shop just might be able to wind 100hp into the 75hp frame using Class H insulation or some other such trick.

A good motor shop can often work outside of NEMA limitations to get you what you need, in a pinch.

No, you won't be able to buy a commodity replacement motor so you will have to think thru what level of spares you require.

 

[ScottyUK]

It's one thing taking care of the insulation temperature class but squeezing more magnetic flux into a core is another thing entirely. If this is a fairly modern motor I'd be surprised if there's enough margin between operating flux and saturation to squeeze the extra power out. An old motor will almost certainly be more generous with the iron. If you're operating voltage is toward the bottom end of the range that works in your favour too.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[DickDV]

Skogs, I'm sure you are aware that a typical NEMA Design B motor has about 220% short-term peak overload torque available. That tells me that the iron is capable of at least that level of flux.

Converting some of that short-term capacity into continuous capacity is what I've seen motor shops do with custom designs. Usually, this is accomplished by increasing the thermal capacity of the motor and by reducing losses which result in heat.

As always, there are limits to how far this can go.

 

[ScottyUK]

Dick, it's me not Skogs!

I'm not intimately familiar with NEMA motors, being in IEC land, but generally I'd say that NEMA designs generally have more 'fat' built in than IEC ones. Perhaps it would work with a NEMA motor. With an IEC motor I'd expect that the core would be well on the way to saturation with the currents associated with peak torque and the iron would get damned hot quite quickly. I'm sure someone must have done a more in-depth study of this - maybe they're reading this...?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[electricpete]

I think everyone here is correct. As Dick mentioned, the main concern related to magnetics in increasing load beyond nameplate is whether you have margin to breakdown torque.

As Scotty alluded, there are small local saturation of leakage reactances (primarily the teeth) as current goes up. I don't think this is a huge factor because:
1 - the magnetizing reactnace does not go further into saturation... it goes further out of saturation because of increased voltage drop accross the magnetizing reactance.
2 - if magnetizing reactance does not go into saturation, we are not going to draw more magnetizing current. This is one of the big factors which normally causes heating due to saturation. Without this, all that is left is hysteresis and eddy losses. I'd say the eddy losses increase proportionally to the square of flux...(so 10% increas in flux and 20% increase in heating), while the hysteresis loss get bounded at some point (the area inside the b-h curve gets pinched off).
3 - Whatever increased heating occurs in the teeth is localized and can be transferred to cooler parts of the core by conduction which is a very good heat transfer mechanism.
4 - Temperature of degradation of core laminations is ridiculously higher than temperature of damage of electrical insulation on the windings.
5 - Putting it all together, there is not a concern for localized heating, just a concern for the contribution of that localized heating to the overall temperature of the machine. If there is any concern for global temperautre increase associated with an uprate, it is probably more associated with the bearings than the core.

At least that's my opinion thinking it through for the first time. But I'm not claiming to be any expert.

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