Medium voltage motors - differential protection (87)

you refer to the American Naional Standards Institute relay numbering convention and to 'Medium Voltage'

Both of these suggest you are after a US-centric answer, and this may not be the best forum, as our (UK) centric rules are rather different in the detail - the physics is the same, but if you open the box it is sometimes hard to believe ?.

Certainly we do have systems that can trip off the primary side of transformers at 11kV or whatever if there is a load side fault on the 400v LV side, and the telemetry circuits that do so  need to be both fail-to safe, and testable. However there are other differences that may or may not matter - our HV circuits never carry neutral for example, and signalling may be voltage or current.

I am probably not helping, but may explain the lack of responses.

Firstly the differential relay doesn't have to be in the same place as the breaker.  This is often the case for transformer differential protection, where the circuit breaker can be tens of km away.  If the breaker is a long way away then the relay can be housed locally and the trip signal transmitted over distance to the breaker.  Transmitting a trip signal over a long distance is easier than extending CT leads over a long distance.


But for a motor it is likely that the circuit breaker is fairly local to the motor.

The CT lead to the relay is electrically just a load - a resistance with a current passing through it.  If you calculate what current you expect, and you know the wire resistance, then you know the VA that the CT has to provide.  Order a CT with an adequate rating.  If that becomes expensive or difficult to house then reduce the VA load (usually called the VA burden in the case of CTs, but it's just a load) by increasing the wire size to reduce the resistance.

Hi John,


thank you for your reply, which is very helpful.


For a transformer you mentioned that "the relay can be housed locally and the trip signal transmitted over distance to the breaker." I believe that over distance means withouth cables. Am I right?


For a motor it seems to be that the circuit breaker is generally installed on a panel local to the motor itself. This means that, in practice, there might be 2 breakers in series: one from the main MV swirchboard protecting the cable downstreem and one protecting just the motor. I wondered if in practical exaples distances of only 25m or 50m could avoid duplicating MV breakers as explained above. But I believe you already gave a good answer which is related to the CT burden.


Thank you again for your help.

vici,

I do not understand over distance to mean without cables. I believe it just means that the relay and the breaker are in two separate locations and there is no immediate limitation on that separation or the length of the connecting cables. Practical matters will start to intrude if you make it hundreds of metres but as most trip signals are either Under-voltage or Shunt-trip it is not a signal transmission issue.

As Mike said, this may not be the best forum, and I would have suggested "Ask the Community", but as you are based in the Home Counties I doubt if you are after a US-Centric answer. I have used the various ANSI relay numbering convention as it is fairly international and if you are dealing with the big international boys (e.g. Siemens, ABB, etc.) they will be using this numbering. The real reason for not having it on this forum is that it is not really covered by BS 7671 (as far as I know - no doubt someone reading this will chip in and say "But in clause....").

My feeling here is that you need to look at the purpose of the protection and think about what fault(s) you are trying to protect against. One of the main faults you are considering with differential protection (if my understanding is correct) is a fault - generally short circuit - on the connecting cable. The protection is therefore trying to shut off the energy feeding in to the fault. You therefore need the CB protecting the cable and can't dispense with it. However the CB at the motor is not actually protecting the motor but rather isolating the motor so that it doesn't feed power into the cable fault. Large MV motors are often synchronous machines with subtransient reactances of around 10-15% and with significant inertia, so can feed quite a lot of energy into the fault. Therefore there needs to be a means to prevent this, which in the situation you describe is a breaker at the motor. With synchronous machines another possible means is immediately shutting off the excitation, which would do away with the extra breaker, but many machines now are permanent magnet machines which means you can't kill the field, and leaves the CB as the most practical solution.

Hope this helps.

Alasdair

Alasdair,


thank you for your suggestion and apologies if i did't select the right section of the forum. I will maybe post it in "Ask the Community" later then. My question was meant more for applications in the UK. Regarding differentail protections (87), generally and at a very high level, these are used for internal faults. For example an internal transformer fault can be detected with the differential protection. The same concept applies for internal faults to motors or even to sections of power lines. The differential protection will exculde the faulty machine or the faulty section from the rest of the system.


Thank you again,