Time OverCurrent relay with voltage restraint for arc-flash study

[tem1234]

Hi,

I have a Basler BE1-51/27R which protect a generator. I have never use this kind of relay. How can you determine the time to trip for this relay for an arc-flash study? The 51 curve change with the voltage. Do you have to approximate the voltage drop?

Thanks

 

[davidbeach]

Generally a 51/27R is backup protection against an uncleared fault on the system, so there should be something else that will clear the fault before the 51-27R. But if not, you will have to get the voltage at the generator VT from your fault study and use that to determine the tripping time of the relay. Just hope that your fault study correctly models the generator voltage decay.

 

[dpc]

These relays are a pain to coordinate. Typically, you can plot a curve at 100% voltage, 0% voltage, and maybe 50% voltage to get an approximation of how the relay will respond.

Traditionally, the pickup of a 51V (at 100% voltage) is set between 150% and 200% of the generator full load amps. At 0% volts, the pickup should be well below the generator full load amps.

 

[AMBMI]

dpc,

I agree that they are difficult to coordinate.
Usually I set the 51 trip threshold at 110% In generator.
The last model I used the voltage restrain was activated for U < 80% Un. I set the 50 trip threshold at 300% of the generator rated current with a 0.25 s delay. In this way the primary protection (87) of the generator had time to operate. No way to get something useful from the grid following overcurrent (50) protection: it was set at 8.5 generator In with again 0.25 s delay, for sure it was no tripping for the generator fault contribution (T'' = 0.02s T' = 0.300s X''=0.1 X'= 0.17 X =1.6)

 

[tem1234]

Thanks for the reply

As i can see, the tap setting of the curve decrease until 25% max of the 100% curve if the voltage drop under 25 %. In the simulation, the voltage drop much more than this if a 3 phase bolted fault happens. So i think that the 25 % value is ok.

I agree that this must be hard to coordinate

 

[dpc]

The voltage-controlled relays are easier to deal with than the voltage-restrained.

 

[slavag]

I'm agree with all posts above.
We usually used voltage contolled function, as Dave recommended,it'is more, more easy.
with voltage setting about 0.7-0.8Unom
51V or 51/27 function used against: Phase fault protection of generators with rapidly decaying fault current such that a normal time overcurrent function could reset before its delay had expired.
51V or 51/27 function used as back-up protection of 87GT ( diff protection of block generator transformer).
Regards.
Slava

 

[rbulsara]

tem1234:

I do not believe 51V setup could apply to arc flash study. As davidbeach indicated it is a "back up" over current protection and for close-in faults. It generally takes "several seconds" to operate which is way beyond the 2 sec window of relevance for an arc flash analysis. So the end result will be "too dangerous" to work live!

Only part of its function or curve so to speak would be applicable when the full voltage is available. The pickup at full voltage is generally set above 100%, more typicaly 150% of FLA. For that purpose you can treat it as straight 51 relay with pick up set at the FV setting.

 

[tem1234]

the 51V is the only overcurrent protection for the generator(strange? there's no 87G, but a 32R-inverse power).

there are 50/51 OC protection for the distribution network, but the problem is when the plan is feed by the emergency generator only (with no utility), the fault current is very low and the 50/51 protection take a century to operate. The 51V is the only protection that can operate in a decent time, it is almost the 2 second limit if i take the 100% voltage curve.

So i'll recommend an over protective equipment on this case, and recommend too to add 50/51 setting group for the case of no utility to reduce the risk. Any other idea?

Thanks

 

[tem1234]

Just to clarify

this is for the HV side of the transformer, at 25 kV. The fault current at his voltage is only 400 A. Event if i take the 2 second limit, it gave me cat 3 since the fault current is low.

There is an other 51V on the low side of the transformer, but it is slower.

 

[slavag]

Hi Tem1234.
Now it's clear. You dont need any 51V ( only on the LV side of step-up transformer as back-up).
You need relay ( as you wrote ) with two group of settings .
1. for parallel operation with grid.
2. island mode.
You can change settings with .. for example.. aux contacts of main CB.
I have another Q, what with your E/F protection?
What is type of your step-up trafo, what is grounding/earthing type?. Usually it's problem.
Regards.
Slava

 

[dpc]

For an arc-flash calculation, I would just ignore the 51V or compute at 100% voltage, since the actual voltage will be impossible to predict under arcing conditions.

Since these are generally used as backup protection, the clearing times tend to be very long anyway.

 

[tem1234]

Hi slavaq

The transformer is Yd, with reactance of 30 ohms on the grounding path of the wye side. There are 50G/51G protection on it with pickup at 60 A.

The generator is HRG of 5 ampere, with a 59G protection to detect earth fault.

Thanks

 

[slavag]

Hi Tem1234.
You need some digital relay with two setting groups, it's all. for arc-flash calculation see Dave's recommendation ( one post above).
Or second set of simple relays with logic like to : contact of main CB disconnect power supply to second set of relays.( or something like to this).
Regards.
Slava

 

[tem1234]

Thank you guys for your help