Nuisance tripping? 1

[KVuorio]

We have a case where is some nuisance tripping of protective devices during the start of big motors (several 3,45kV 800kW induction motors). When starting big 3,45kV motors the feeder circuit breakers in 690V MCC are tripping occasionally? We have been investigating the problem and have been able to eliminate some possible causes. The voltage quality is good enough (THD approx 2…3%), voltage dip in starting conditions is approx 5…6%, this is also good enough. Also the protective devices are not tripping due to wrong short circuit and overload settings.


We have come to a conclusion that the nuisance tripping may have something to do with ground currents. There are several transformers in the system, 13,2/3,45kV transformers are low resistance grounded through 400 Amps NRG, 13,2/0,69kV transformers are high resistance grounded through 2 Amps NRG. There is also one 690/400V transformer which is solidly grounded (is there any restrictions combining high resistance grounding and solidly grounded systems?).

In the 690V switchgear there is IPC DSP MKII ground fault protection system. System led’s are indicating normal state (NORMAL lamp lit, PHASE ALARM lamps ON steady, All other indicators OFF). Now when I push the FEEDER GROUND CURRENT pushbutton I have a reading of 100% in two feeders? What does this situation tell me, do I have a solid one phase ground fault? And do I have a solid one phase ground fault in another feeders same phase too? But why DSP is not showing me any alarm, instead it is showing normal condition?

 

[RRaghunath]

I am afraid it requires much more detailed study (of more data than indicated) to find the problem.

What I do not understand is how reliable the earth fault protection in a feeder can be, when the fault current magnitude is limited to 2A. I think the current based sensing is not suitable in such high impedance grounded system and cannot be wired for trip. My experience is change in capacitive leakage in the cables is sufficient to cause operation of these relays, especially in systems with extensive cabling.

Let's know when you finish (hope you care).

Thanks.

 

[KVuorio]

I was also thinking that what if the system capacitive leakage is way more bigger than 2A, then it is possible that you can experience malfuntions?

On the other hand the DSP MKII-system will not trip before it also senses ground voltage. The phase-to-ground voltage of faulted phase has to go 50% below the nominal and also the voltage of other two phases phase-to-ground has to rise 30%.

Maybe the situation now is that the system capacitive leakage current is bigger than 2A (which has been set to DSP-unit as an 'nominal' ground current) >> somewhere in the system is a high resistance fault which is enough for 2A ground current >> the fault is so high resistance that it will have only little affect on voltages >> the ground fault protection system is not working properly?

We havent been designing the system and we dont know what is the capacitance leakage current of this system. Maybe we should try to estimate cable lenghts etc.? Atleast then we would know if the system has been designed wrong!

 

[KVuorio]

Hi All,

We have finally found the problem, it was quite simple (thats how all the faults seem to be after yoy have found them).

In 690V switchgear there was four pcs. 690/400V transformers for 400V small (22kW) VFD's.

Those transformers were auto-transformers eg. there is no separate windings for primary and secondary and due to that there is no galvanic isolation between pri-sec.

Transformer was Y-connected and also the starpoint had been wired to transformer terminals.

There was a jumper cable which connected the starpoint directly to the ground! From that jumper cable we measured 15 to 20A current in each device.

The DSP unit trips instantaneously if it detects 80A current through cable current sensors.

Original problem was that when big motors in 3,45kV side were starting the other motors in 690V switchgears were stopping.

Actually the breakers feeding the MCC's were tripping. The tripping has nothing to do with the starting of 3,45kV motors. When 3,45kV motor is started there are also other motors in process starting according to interlockings. Now when those 690/400V transformers were 'leaking' almost 80A together it needed only little more capacitive current from starting motors/feeders to make breakers tripping.

This is quite hard to explain because english is not my mother language but i hope you understand what i mean.

I would like to thank rraghunath for the answer which was the only reply.

BR
Kalle Vuorio

 

[KVuorio]

Forgot to mention that we removed those jumper wires and now system is working fine!

 

[RRaghunath]

Probably the current you measured in the neutral (most of it atleast) is of triplen harmonics and not the zero sequence type. This is because of the VFDs downstream.