Heat Generation in a High Speed PM Electric Motor

[450exc115]

Howdy,

I am the lead engineer working on a system that uses a high speed PM motor to drive a vacuum pump and blower. I've been working on the cooling requirements for this unit and have been struggling on how much internal cooling Air to supply for the rotor, magnets, and bearings. Calculating the bearing cooling is relativly easy but I have yet to find information on how much heat is generated by the magnet and rotor. I'm just looking for a design rule of thumb. Example, it's a 5KW motor and the losses are .75KW. The losses are assumed heat generation, xx% from the stator and xx% from the rotor.

Any advice or insite to this would be great!

Thanks,
Tom

 

[Clyde38]

Tom,
I'm not sure you'll find a design "rule of thumb". You could calcuate all of the stator losses (I^2R, eddy current, hysteresis etc.) and subtract from the total loss of your machine. Rotor only losses in a PM motor can vary. Do you have magnets that are susceptable to eddy current loss? Any way, most of the loss should be in the stator (I think). You can obtain software to help simulate this (FEA) or Motor-Cad (

http://www.magsoft-flux.com/).

 

[Clyde38]

I may have misunderstood your problem. I was assuming that you had a PM Brushless DC motor:

"how much internal cooling Air to supply for the rotor, magnets, and bearings."

Can you elaborate?

 

[450exc115]

Sorry about that. It is a 2 pole brushless PM motor. Current plan for the magnets is a laminate rare earth magnet that we have used in a bunch of other high speed motors. Unfortunately I can't get into the nity gritty details of the project due to the nature of the project (customer and our technology0 but I will look into the predicted stator losses to back out the rotor losses. Typically our motors are 90% effecient, I just need to know what % of those losses needs to be cooled internally and externally.

Tom

 

[Skogsgurra]

My experience is that the rotor losses are quite high. Plastic bound ferrite powder rotors may help. The PWM causes high rotor losses. A filter or a sine inverter does a lot to reduce rotor losses.

It is rather important what you mean by high speed, some say that 10 kRPM is high, others say 100 kRPM. I have run designs at around 80 kRPM and used water cooling for the stator. The intense air circulation in the motor makes for the rotor cooling, so you do not need to supply any air. Only increase water flow to remove the rotor heat.

BTW, we used ceramic ball bearings. The centrifugal forces were too much for the steel balls.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[sreid]

Rotor losses in a brushless PM motor (especially with a sine drive) are small. This is because there are no I^2R losses and since (ideally) the rotor flux is constant, there are no iron losses.

 

[Skogsgurra]

Agree. But only if you use a sine drive or a rather good filter. PWM has killed rotors of the designs I have worked with.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[450exc115]

I venture to say it is a true High speed motor.. We build turbo machinery using centrifical rotor technology. This particular application is going to run close to 160K RPM and be about 4.5KW.
Bearings are no problem as that is one of our signature specialties. Bearing DNs in the 3 million are not out of our experence.

Tom