Motor Protection 1

[tunelabguy]

I have been asked to design a custom motor protection circuit for use in the compressors of an airborne air conditioning system. For reasons that I do not completely understand, my client does not believe there is an off-the-shelf solution. The power is 3-phase, 400Hz. Apparently packaging is an issue, as well as the 400 Hz. The only functions that my client knows he wants are:

1. configurable thermal modeling for intelligent delayed overcurrent detection.

2. missing phase detection with a much shorter delay.

I plan on using a microcontroller to read current transformers on each of the three phases. Only the amplitudes will be significant - not the instantaneous AC current.

The goal is to provide something that is a little better than a simple breaker or slow-blo fuse.

My question is this: What other features are useful in a motor protection circuit that can be implemented without additional inputs? My client has hinted that maybe ground fault interruption might be nice, but that would require an additional transformer. Anything that requires tapping directly into the motor voltage raises a safety issue or requires expensive galvanic isolation. Even though the application is in an aircraft, the air conditioning system is not a critical safety issue. These are freight haulers, not passenger jets. The main thing is these compressor motors cost $10,000 each and shutting them down is a lot better than burning them up.

Robert Scott
Real-Time Specialties
Embedded Systems Consulting

 

[jraef]

Just to save you a bunch of time and effort, the old Multilin (now GE) 269 Motor Protection Relays were good for 400Hz applications. Their newer versions, the 369 and 469 are not, but the 269 is still available if you act now.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

With no further inputs!! ? I don't see what else you are going to get from anything.. There is only so much information in the motor current.

Personally for this 'gold plating' I would lobby for more inputs.

1) Measure the voltage of every phase! (Use an isolator IC to run the data across and I don't mean an opto.)

2) Measure the temperature of the compressor.

3) Measure the head pressure and the suction pressure.

Compressors die from: (in the order of likeliness)
1) Low Voltage
2) Phase Loss
3) Slugging
4) Excessive head pressure.
5) Short cycling
6) Low Suction Vacuum(Pressure)
7) Low charge
8) Starting component failure
9) Condenser fan failure
10) Clogged condenser
11) High Voltage
12) Loss of lubrication (poor plumbing)
13) Refrigerant contamination

With the voltage you can run POWER calcs.

With more inputs you can monitor for far more things that can kill the compressors. PLUS help the diagnostition in fixing any problems that do occur.

With just the paltry current coils you can do phase loss and short cycling.


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[dpc]

jraef,

That's interesting - I knew the 469 would not work at 400 Hz, but never knew the 269 was an option.

 

[jraef]

I used to use them on GSE* generators for Boeing. From what I've heard that is one of the reasons why GE has not yet pulled the trigger on the 269 even though everything else it does is in the 369. The 369 design team, in their infinite wisdom, stopped at 300Hz. Typical case of engineering not speaking to sales.

*GSE = Ground Support Equipment, the catch-all term for the compressors, generators, A/C units etc. that hook up to the aircraft when it parks at the gate.

 

[waross]

Temperature is more important than current to a hermetic compressor. The compressor motor is cooled by the refrigerant passing over it. If for some reason the temperature of the suction gas does not drop as soon and/or as much as it should under normal conditions, rated current may be too much current. This condition may be related to high head pressures.
I feel that temperature monitoring of a hermetic compressor motor is more important than current monitoring.
Keith, don't almost all of your issues result in heating?
I guess that with loss of lubrication, heating tells you that it's too late, but even with loss of lubrication, an over heat shutdown gives you a chance to avoid rewinding the stator.
respectfully

 

[itsmoked]

Yeah Bill. A lot of the items come down to temperature deaths.

3) Slugging
4) Excessive head pressure.
8) Starting component failure
12) Loss of lubrication (poor plumbing)
13) Refrigerant contamination

Don't however. Though #4 can go either way. Promptly snapped connecting rods or holed pistons verse a slow-heat death.

Unfortunately the problems are not easily dealt with by case temp measuring nor even internal temp measurement because refer comps live in such a nasty realm. They can run at obscene temperatures for brief periods that are considered 'normal' merely withdrawing from the units lifetime bank account. So if you just watch the temp you may be false-alarming constantly. It's kind of considered in the industry, that the 'head pressure is king' for most of the info.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com