Motor Damage Curves

[OhioAviator]

Hi All,

Our company employs large squirrel-cage and wound-rotor motors in the 3500 HP to 7000 HP range, at 4,160 volts and approx. 600 RPM. These motors are from Siemens Allis, Schorch, Helmke, and GEC Alstom. I have been trying to obtain thermal-damage curves for these motors from either the manufacturers or their U.S. representatives, but have thus far been unsuccesful, though I will keep trying.

Can anyone suggest time-current plotting points to use in order to allow me to produce "generic" thermal damage curves (assuming this is possible)? Alternatively, can anyone provide me with a formula or formulae that I can use to produce "generic" thermal damage curves for these motors? Are there better approaches to obtaining this info? My ultimate goal is to improve the settings of our Multilin 469 relays to more closely match the thermal/mechanical capabilities of our motors.

Thanks for your help!

 

[GGOSS]

Hello OhioAviator,

Many manufacturers of MV motors build their machines to suit the intended application and therefore each can be significantly different from the other.

I have had some success in obtaining the sort of information you are looking for just by using different terminology eg;

1. Motor Thermal Withstand curve.
2. Motor Overload curve.
3. Or values for Locked Rotor Current and Locked Rotor Time.

or by explaining what I need in detail.

As indicated above many of these motors are specials. Therefore I would refrain from attempting to produce curves myself. Equally important, when contacting manufacturers or suppliers, it would be beneficial to have serial numbers available.

Hope that helps.

Regards,
GGOSS

 

[gsimson]

The manufacturers can supply the thermal image curve for new motors only. Even if the curve is available, it may not be of much use as the thermal capacity of the motor undergoes changes due to the total no. of starts, dust accumulation in the winding, moisture in the winding, efficiency reduction due to increase in magnetic losses.
More accurate relays have been developed for protection but the finding methods for actual themal characteristics are yet to be developed.
It is alright if we can set app. values in the relays which can protect the motor. Ofcourse the application is much more important to select the curve

 

[OhioAviator]

Hi GGOSS & GSimson:

Thank you for your replies. I suspected there was really no way to come up with a 'generic' thermal-damage curve for our motors, but I thought I'd ask the question anyway.

As for my quoting motor serial numbers... yes, I have done that but so far no luck. Like I said, I'll keep trying.

Our application is automobile shredders. These are a high HP (obviously) application with very high inertial shock loads, similar to large crusher application. We keep tripping out the motors, I feel, prematurely and I'd like to try and squeeze a bit more HP out of the motors without damaging them thermally and/or mechanically.

Again, thanks!

 

[jbartos]

Suggestion: Since the motor is an ac machine similar to transformer with damage curve having somewhat similar direction and location as the transformer damage curve, it could be approximated, if not available from the manufacturer. Any reasoning that the motor damage curve will change in time is justified; however, so will transformer damage curve, generator, and cable damage curves. They are being available and plotted on the protective relay coordination charts.
The motor thermal damage will be originated from RI**2 in Watts heat dissipation in the stator winding. Therefore, the Watts need to be converted to heat and temperature increase. The temperature increase and maximum temperature of the stator winding are limited by the winding temperature insulating Class, e.g. 220degC. Electrical consulting engineers should be able to reasonably estimate the motor damage curve. However, the quality of the produced motor will be an important factor not to be overlooked. Perhaps, this is why the motor damage curves are hard to obtain. I had a hard time to obtain a specified damage curve (according to MasterSpec) for the larger generator.

 

[DougMSOE]

OhioAviator,
I know your problem VERY well having worked with this type, size and application for 5 years.
Do Not use the typical transformer damage curve mentality. These motors have a much higher and tougher duty than most as you well know.
What I have done to produce the "damage curves" is to use the RTDs in the stator and a separate method to calculate the thermal damage curve in the rotor circuit if a wound rotor. There are published curves for time, temperature and life of electrical insulation. The rest is some simple integration.
I have used this on some of your competition's motors and other heavy motor applications, it has worked well.
If you are interested call me 920.453.2195
Doug

 

[Barry1961]

This is a bit off the path, but have you tried limiting the shock loads seen by the motor with a flywheel/torque limiter arangement?

Barry1961

 

[jbartos]

Suggestion to the previous posting: The flywheel/torque limiter is a good idea. However, not all applications can be covered by it, e.g. ship applications where the motors tend to be lighter than normal (else the ship might sink).

 

[OhioAviator]

Hi Everyone,
I suppose perseverance (sp?) pays off. I received a thermal-damage curve for our 3500 HP Siemens-Allis motors today. I'll keep working on the other manufacturers.

Thanks, all, for your suggestions and comments.

 

[gsimson]

For such tough applications, the almost true picture of thermal capacity can be checked only by RTDs installed in the winding. Certainly the RTDs should be available for these large motors. By recording & analysing the temperature curves during starting/ running/ idle, along with Currenr & time, the real thermal capacity of the motor can be inferred.
Based on the inference, the relay can be set

 

[OhioAviator]

Hello Gsimson & all:

I wholeheartedly agree that the best way to thermally model a motor for maximum protection & performance is to use multiple stator RTDs in the Multilin, and that's what we're in the process of doing. The RTDs were never brought into the Multilin. We're doing that now.

On a related issue, at what point in the motor's speed vs torque curve does the motor switch from being thermally limited by the stator to being thermally limited by the rotor (in the case of a squirrel cage motor)? In other words, as the motor slows down under increased load, does the rotor heat become more of a limiting factor in the motor's thermal capabilities?

Thanks!

 

[GordS]

IEEE 620 Guide for Construction and Interpretation of Thermal Limit Curves for Squirrel cage Motors Over 500 hp states that the locked rotor condition should be represented at about 300-400% of current. A motor overload condition should be plotted from rated operating current to the current corresponding to breakdown torque. The standard isn't written in terms of "stator limiting" of "rotor limiting".

 

[jbartos]

Suggestion: Reference:
Padden, L. K., and Pillai, P., “Simplifying Motor Coordination Studies,” IEEE Industry Applications Magazine, vol. 5, no. 2, pp. 38–52, Mar./Apr. 1999.
for:
Motor starting, and capability curves