Utility Ground Fault shutdown my MV switchgear 1

[living2learn]

Working in a sugar factory in Texas. Utility is 13.2kv 3w+g. My switchgear utilizes sel 351 relays. It has a utility input, load breaker, tie breaker that connects another MV breaker to a transformer that has a 1000kw 480v transformer. The generator step up transformer is a 1000kva 13.2kv grounded wye : 480v delta.

A fault occurred on the utility side between C phase and ground. The utility breaker opened after 18 cycles and ALL of my MV breakers tripped and locked out only after 3 cycles. Only the main relay has voltage sensing.
Q1) How can all of my breakers trip and lockout?
Q2)can my system supply enough ground fault current? The secondary is open where is the source?
Q3) My gen transformer secondary is 480 delta and the gen breaker is opened how can ground fault current flow from the primary grounded wye to the utility fault?
Q4) Can 0 sequence current be pulled from the grounded wye and thus trip the feeder breaker? Would disconnecting the neutral to ground connection on the 13.2kv side prevent 0 sequence current from flowing? I still don't understand how it can flow if gen is not connected??

I hope these questions makes sense and I know I asked the same question a lot just hoping that it paints a better picture.

 

[davidbeach]

Draw out the sequence diagrams. Nothing needs to be connected to the delta for the grounded wye to supply significant amounts of zero sequence current.

 

[jghrist]

Can 0 sequence current be pulled from the grounded wye and thus trip the feeder breaker?

Yes.

Would disconnecting the neutral to ground connection on the 13.2kv side prevent 0 sequence current from flowing?

Yes, but disconnecting the neutral would leave your 13.2 kV system ungrounded when running on the generator with the utility disconnected. This is probably not acceptable.

What relay tripped the breakers during the utility fault? Do you have bus differential? If so, how are the CTs connected? What are the CT connection settings of the relay? Were there any relay event reports that can be downloaded?

 

[living2learn]

What relay tripped the breakers during the utility fault?
It was a SEL-551 Relay and all of the breakers where tripped and locked out.

Do you have bus differential? No.

Were there any relay event reports that can be downloaded?
I am going out there today so I will hopefully grab them. THANKS!

Solutions.

All of my cable is 100% rated. Can I install a low impedance 100A/10 sec resistor on the grounded wye of the 13.2kv transformers? I believe 133% rated cables is able to ride a ground fault for an hour and 100% is only required for a minute or less, 173% for ever.

I remember reading somewhere that you don't want high impedance grounding (5A, forever) for 15kV, but only 5kV and below? Is that true and why can't I find any standard?

Also if I had wye-wye PT's I could tell which way the ground fault is going and coordinate the relays that way?

It is kinda funny we spend all this time and money trying to maintain realibilty and boom a ground fault on the utility takes me down hard. I have to coordinate my GF settings with the utility so I don't take them down. How can it go both ways?

What could have prevented this?
Impedance grounding?
WYE-WYE PT's and directional overcurrent?

 

[Kiribanda]

living2learn,
It looks like your primary WYEGND-DELTA transformer has fed to the ground fault in the Utility line eventhough your gen bkr is opened which is normal.You should note that your transformer high side shall be WYEGND because it fulfills the effectively grounded requirement during a Utility ground fault when the generator is on line.My advice is to calculate your ground fault contribution for a Utility ground fault and set your transformer overcurrent and ground fault settings little SLOWER so that Utility will first trip the line from their end and make the line dead.Then calculate your gen contribution when your gen is on line and set your gen phase & ground elements accordingly.Do you have intertrip provision?Could you please upload your SLD with the relay scheme?

 

[odlanor]

Which functions were activated of SEL-351 of all MV breakers?
It seems operation of 81, 59 or 27 relay.

 

[rcw retired EE]

Are the outgoing feeders and tie breaker connected by cables with zero sequence CT's around the cables? If so, check the routing of the cable shield grounding wires through the CT.

If your cable shields are grounded at both ends, utility ground fault currents for any voltage can flow though the shields and ground wires if they are a lower impedance path than through the earth. If the wires are not routed correctly at the donut ZSCT's, the CT's will send false ground fault signals to the relays.

That may explain why all your breakers tripped.

 

[living2learn]

they utilize a residual connection not a pure zero sequence.

 

[living2learn]

The relay only operated on ground fault!

 

[DTR2011]

If the feeder relays do not have voltage functions, not tripped by LOR, have you considered a problem in the DC System? How / is the battery system grounded? Have you checked for DC ground faults?

Has the main ever been opened previously with all feeders closed?

Was the plant ever properly commissioned?

If you can't rule out relay settings issues / grounding issues, you may have to chase these issues down.

 

[davidbeach]

they utilize a residual connection not a pure zero sequence.

And what does a residual connection measure if not pure zero sequence?

 

[living2learn]

This is what I meant by residual vs. zero sequence.

From the post below

"You can obtain a much lower setting with a sensitive ground fault protection (zero sequence ground fault protection) than with residual ground fault protection.
A setting of 1%-2% Full Load Current is obtainable with sensitive ground fault protection while a setting of lower than 10% Full Load Current (residual ground fault protection) normally result in nuisance trips."

http://www.eng-tips.com/viewthread.cfm?qid=156182

 

[davidbeach]

It's all zero sequence. The reason you can get more sensitivity from a core balance CT (some times referred to as a zero sequence CT) is that it can have a much lower ratio, say 50:5, while the residual connection, which is also only zero sequence, has to use CTs that are sized for load, usually many times larger.

 

[jghrist]

If you have a four wire system, the core-balance CT (around all wires) will measure ground fault current only, excluding neutral return current. This allows sensitive ground fault protection where the neutral is grounded only at the source. If the neutral is grounded at more than one point, this can lead to false trips because some of the neutral current can flow through the ground.

 

[waross]

Your GSU will be seen as a four wire wye:delta from the primary side. With the low impedances typical of distribution transformers you will see heavy circulating current in the delta and a transfer of power from the healthy primary phases to the faulted primary phase. The fault may be miles away. This transfer of power will be accompanied by heavy primary neutral currents. The current will be limited by about three times the PU impedance of the transformer. The four wire wye:delta will also try to correct primary voltage unbalances and primary phase angle errors. A moderate primary voltage unbalance may cause more than full load current to circulate in the delta winding. Again accompanied by primary neutral current.
This effect may be involved in your event.


Bill
--------------------
"Why not the best?"
Jimmy Carter