Yes, you are right. I was talking about electronic overloads. I should of made that more clear.
Some types of thermals also require ambient temp compensation in one form or another.
It's an area I haven't thought of much before. I'm guessing that any digital relay will have true rms characteristics for the overload. I'm guessing that any solid state non-digital relay would have probably not have true rms detection of current.
If heavy harmonics are present, the true rms detector is better than not since it takes some account of harmonic heating... but even the thermal (true rms) relay is not going to properly compensate for the complex resulting heating effects in the motor. (for example increased hysteresis and eddy current heating, rotor skin effect for high-frequency). Perhaps the digital guys try to model these more accurately. I'm sure most competent experienced relaying engineers don't pay much attention to these small distinctions.... at some point you have to say it's close enough.
I would be suspicious of protective devices claiming to be "true RMS" in the face of the sort of hash they would see on the output of a PWM drive. One would have to investigate quite a bit to determine claimed frequency range, sample rate, etc.
I think a much better approach is to install an embedded thermal detector in the motor, either an RTD or a thermistor.
Interesting comment about the temperature. In our plant all the large motors have rtd's but none have trips from rtd. Until now I never wondered why we don't trip on the more direct indication of temperature (we do alarm though). Perhaps relay engineers are more familiar with ct's. Perhaps rtd's are less reliable.... would electronic switch actuated by processing circuitry.... a little more complex than a simple relay contact. Although a simple temperature switch certainly could spec'd into the motor.
I guess the meaning of TRMS as far as current detection would make some differences as far as using a CT or a shunt(dont think anyone uses this for AC motor protection). I think electricpete pointed out in a previous thread about the motors inductive properties suppressing some of this hash so maybe measuring TRMS current for motor protection is wasted money? It might be the way to go if there is some horrible distortion, I suppose. But it would seem that most installations would not benefit from this if the price was any more than a standard RMS device.
I am not sure many people trip a motor based on temperature only or at all. It would seem this would be the best way to protect from an overload conditon. A lot of electronic OL mfg's don't offer this in a standard overload device. It is usually an add-on. I do know of motor mfg's who embed these in their motors.
I think alot of people don't trip on temp is because of the cost of the electronics to do this(at least RTD's). It is not that high but the development time, etc, is longer. And you have to worry about ambient temp so there is another measurement. About four years ago I think I remember stand alone RTD monitoring boxes being $150 per RTD. This didn't include a suitable relay for a motor control circuit or a user adjustable trip point. I am sure the cost has come down some since then though.
Another important factor in selecting a TRMS electronic device to be used on hacked up waveforms is the crest factor.
There must be a point where sampling the higher frequency components would not add much to the actual trms value. Given a carrier freq of say 20kHz on a PWM drive, is it necessary to sample frequencies as high as this? My guess is no but doing so would produce more accurate results. Also, the most energy is concentrated in the lower harmonics so sampling anything above the 7th harmonic say, would not produce that much more accuracy. I can't recall where the SIGNIFICANT energy is in the harmonics on a PWM (isn't it the third, fifth, and sometimes the seventh?).
Electricpete, Most digital overloads are not TRMS devices. But there are some available in a variety of price ranges. They involve DSP's or are performing complex math (at least for a microcontroller) or they may use an off the shelf IC like maxims line of TRMS converters. The limitation with these seems to be the crest factor (5 is highest I have seen). I haven't checked into the bandwidth of these either but I would imagine it is well above the highest energy concentration of the waveforms of most drives.
For years we have specified thermistors in all motors 25 hp and above that are driven by AFDs. These are wired to shut down the motor. This function can be provided in the AFD. We also provide standard thermal overloads (non-electronic).
Most of the new digital motor protection relays do accept RTD inputs that can be used for tripping. We see this used in many facilities.
The embedded temperature detector is much superior to any overload relay. An RTD or thermistor system will provide protection against overheating from any cause, including single-phasing, ventilation problems, high ambient, etc.
Suggestion marked ///\\\ to buzzp (Electrical) Apr 11, 2002
For motor protection, how many people look for true RMS vs. RMS devices(for voltage and current protection)?
///In many instances, the true RMS are mandatory. With ever more increasing application of electronic circuits and signal processing, the true RMS are also possible to obtain by electronics rather then over the thermal element. The complex RMS converter produces the RMS result for any AC waveform. Visit
etc. for more info.\\ Or do you care?
///The manufacturers do care, e.g. motor, meter, etc. manufacturers and whoever is highly dependent on the manufacturer product proper applications or implementations.\\
