Motor Thermal time constant

[stratford]

Dear all. On site we have one 6.3 kV motr, 1090kW. This motor is vertical mounted, drivinf one main cooling water pump of our power station. The pump had a very big failure on the bearings and the motor was left running for 20 minutes with currents higher than nominal. the motor has SIPROTEC protection relay 7SJ63. The thrmal time constant is set 80 minutes but this is not the value of the motor manufacturer. I think to decrease it to 40 min at least. Is that a good idea?

 

[dpc]

The thermal time constant setting in the relay may be more related to how long it takes the motor to cool down, not heat up. These relays attempt to replicate the thermal characteristics inside the motor.

If you think the relay should have tripped on overload, you need to first verify the Full Load Amp setting and the minimum pickup setting of the relay (which I'm not really familiar with)

Motor stator RTDs might be a better indicator of *long-term* overloading of the motor than the thermal model, although they will be too slow to help with short term locked-rotor type heating.

 

[jraef]

Yes, the thermal time constant has to do with the cooling rate of the motor, therefore the reset time of an overload trip.

 

[stratford]

dpc the motor RTD's during the fault did not exceed the values of 110 oC which is OK for temp.rise B Ins.Class F windings. But the thing is that the pump impeller had one faulty bearing and the noise and damage created was amazing. From SIPROTEC manual the thermal time constant has to do with the time integration that the overload protection uses to trip a motor in on overload situation.

 

[electricpete]

Normally, standard motor relaying practice is based on protecting the motor (not the pump).

If you want to tweak the time constant or overload setting based on your experience, that's your perogative. Just consider possible possible scenario's that create temporary where you don't want to trip and make sure you set high enough/long enough to avoid tripping. For example if a sister pump trips, does this pump run at overload until the sister is recovered?

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

 

[electricpete]

Let me try that second paragraph again

If you want to tweak the time constant or overload setting based on your experience, that's your perogative. Just consider possible scenario's that create temporary overload where you don't want to trip and make sure you set high enough/long enough to avoid tripping. For example if a sister pump trips, does this pump run at overload until the sister is recovered?

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

 

[electricpete]

If these are axial flow pumps like our power station pumps, then power drawn by the motor goes down as flow goes up, so the scenario I posed would not be a concern.

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

 

[stratford]

electricpete The pumps are axial flow indeed. One other strange thing with these pumps is the very low pwoer factor of the motors, they run at 0,71 with 0,73 nominal. While we made a test run with the motor uncoupled from the pump then we saw that the power factor was 0,027. Anyway i will do this modification from 80 to 40 minutes. There is an equation that gives the time that the relay will trip according to overload current, current before overload and thermal time constant.