Resistance Thermometer Induced Currents

We have two 11MW rated compressor drive motors with resistance thermometers embedded in the windings to measure windings temperature. These are in turn connected to a measuring system such that on a high temperature the drives are tripped / shutdown. During a normal start-up of the machines they are prone to tripping. The maintenance engineer on the electrical side has concluded that this is being caused by induced currents / voltages into the resistance thermometer which cause an apparent high temperature trip? Has anyone experienced such a problem? The proposed solution is to delay the drive tripping for several seconds to allow normal operating speed to be achieved where things then seem to operate normally.

There are several types of resistance temperature detectors used in motor windings. These include PTC thermistors and RTDs (Resistance Temperture Detectors). RTDs are the most common.

You don't state if the measurement system is a PLC or a special purpose controller like a MultLin. I have done a number of applications using winding RTDs wired directly to the PLC inputs without problems. It is possible that you are getting EMI (ElectroMagnetic) induced currents. These are most likely coming from the wiring and not from the winding. It may be tht you aren't using shielded cable with the shields propoerly grounded at one end. It may be that you have the RTD wiring in the same conduit as power or AC control wiring, which will cause problems.

I'd check the wiring issues first. There isn't anything wrong with using a time dealy on start-up, as long as it is limited to a few seconds and as long as your high temperature alarm is latching so it holds itself on when the motor is stopped, until it is manually reset.

I have a generator with embedded thermocouples in windings. They were connected to
a thermocouple input card ( AB Flexlogic ). Signal was very noisy causing intermittent shutdowns. Solution was to replace thermocouple cards with analog input cards
( 4-20ma ), installed signal conditioners ( thermocouple input to current output ).
This solved the problem.
The delay solution can work also, we call them spike timers.