Calculating motor winding temperature 4

[charlierod]

I'm evaluating motors operating conditions in a factory

Does anyone know a method to calculate motor winding temperature (hot spot) based on motor operating current, voltage and ambient temperature?
If so, which motor parameters do i need? are those parameters easy to obtain from manufacturer, tests or nameplate data?

Thanks for your comments

 

[jbartos]

Suggestion to the original posting: Try Resistance Method that is very precise. Visit

http://www.iprocessmart.com/leeson/leeson_motor_temperature_article.htm

for more info

 

[charlierod]

Thanks everybody for your help. Before reading your posts i've already obtained such on line papers which i have found really helpful. I still don't understand the model wich uses R1/Ro as a capacitance value. Can anyone explain it?

Also in the document "INDUCTION MOTORS: PARTI-Analysis", the authors made some aproximations that i cannot realize (Rs = Ro/5, Xs = X1/2 and the way they find the value for Xo). I think those estimations are valid for the motor they use in the calculations but are not standard values. What's your opinion?

I'm thinking about applying the resistance method but i got two problems. First, we got motors with nominal powers up to 200 HP and this implies the use of a special kind of ohmmeter (wich i'm afraid is not available at the company and they won't be willing to buy one until i show the method works at least for smaller motors). Second, motor terminals are not readily available so most of measurements will have to be performed in the motor control center (corrections to be made to account for feeder resistance). Any comments about this?, Does anyone have experience in applying the resistance method?

Again, thanks for your help. I find really interesting all these motor thermal issues.

 

[charlierod]

I forgot to say, one of the papers mentioned above states protection based on RTDs is not reliable due to slow reponse. Can someone expalin this?

 

[edison123]

While RTD's may be slow to respond to transient overcurrents (since they rely on transfer of heat from winding copper by conduction, which involves time factor), they are time tested and accepted winding protection as per all international standards.

 

[jbartos]

Comment on charlierod (Electrical) May 7, 2004 marked ///\\I'm thinking about applying the resistance method but i got two problems. First, we got motors with nominal powers up to 200 HP and this implies the use of a special kind of ohmmeter (wich i'm afraid is not available at the company and they won't be willing to buy one
///Normally, it is tacitly assumed that whoever has 200HP motor and associated hardware or production means will have a smaller amount of money reserved for the necessary maintenance, testing, measurements, etc.\\ until i show the method works at least for smaller motors).
///The ohmmeter that is more accurate is called:
Digital Low Resistance Ohm (DLRO) meter. Visit

http://www.avointl.com/cae/products...58&ProductSubGroupCode=P701&ModelFamily=DDLRO

for a sample.\\Second, motor terminals are not readily available so most of measurements will have to be performed in the motor control center (corrections to be made to account for feeder resistance). Any comments about this?, Does anyone have experience in applying the resistance method?
///Precise motor feeder leads impedance calculations are suggested, either by hand or by computer.\\

 

[charlierod]

jbartos:
I totally agree with you but that's life, some human beings have a strange way of reasoning ...let's see what happens.

 

[BobM2]

Interesting way to measure resistance on DC motors. My question, even if you turn off the power to the motor there can be a residual magnetic field from the saturated iron. This would produce an emf in the rotating armature wires. How can you measure resistance with this unknown emf in the coils?

 

[electricpete]

Bob - residual magnetism should not affect a dc measurement.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[aolalde]

The temperature rise of a motor is a function of the generated losses, the motor mass and the heat removed by the cooling system; you need a model that includes these factors to calculate it.

 

[BobM2]

The output from the armature leads due to the residual magnetism is actually DC. The motor we use is a universal motor (AC changed to DC before I can get at it at the lead wires on the brushes).

I'd like to not use a model but actually measure the temperature if possible. The idea of measuring the resistance is still appealing.

 

[electricpete]

Bob - It sounds like maybe you want to measure resistance while rotating? In that case you may see dc voltages as you say. I don't understand any reason for measuring resistance while rotating. While stationary there should be no reason for residual magnetism to interfere with dc resistance measurment.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[BobM2]

Yes, I want to measure it while the rotor is rotating. The motor powers a winch with a limited duty cycle. Some users overload it or run it too long. My hope was to use the triac to turn off the power to the motor for a cycle and make the measurement, then power it back up again. If done quick enough the user wouldn't notice the power being off for the 15 milliseconds or so.

 

[Marke]

Hello BobM2

Your idea of turning off the power to the motor for a cycle or so in order to make a measurement could result in major problems and damage.

Open transition switching causes very high current and torque transients at reconnection. This can break rotor bars, couplings and do a lot of damage to the driven load.

The problem is that while the motor is being driven, there is a rotating magnetic field in the stator which is where the torque is derived from. When the stator is disconnected, the motor behaves like a generator, generating voltage out its terminals. The voltage and frequency are dependent on the motor speed. When the motor is reconnected, the generated voltage will not be synchronised to the supply, so effectively, you can be closing on to a much higher voltage than is usual. The generated voltage drops away with time and this dcay dpends on the time constant of the rotor circuit, but typically, you need to disconnect the motor for more than 0.5 seconds to eliminate it.

The other problem that you would have, is that while the motor is spinning and the rotor field has not decayed, there will be voltage across the terminals preventing you from making DC measurments.

It sounds to me like the best solution is to have some form of thermal model calculating the effective temperature from the line current. This is done in many of the protection relays, and some of the better soft starters on the market today. This can be pretty good at providing the sort of protection that you need.
Why not invest in a small electronic protection relay that includes a thermal model. There must be provision to program into the relay, the rated current of the motor, and the motor start time constant. (Locked Rotor Time)

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[AC5DC]

Depending on the need for a fast response to overheating, I would take another look at RTD's, thermistors or bimetallic switches which can be purchased to respond to the maximum temperature you want to winding to reach. This temperature can be below the temperature rating of the insulation system to allow for a hot spot.

It seems like a much simpler solution than attempting to measure the resistance on a motor while running.

Ed