Earth fault protection 1

[vennivivi]

Hi,

My transformer, ONAN 3,0 MVA 11 k V/ 400 v , used to trip on earth fault protection. I have got a zero sequence CT wired to my relay with a setting of 0,1 A instantaneous.We had four trips in the last week. The transformer has been sucessfully megerred, PI > 5 , and di-electric test also ok. After a trip, the transformer is powered and run fine for hours.
I have also got an overcurrent earth fault relay which used summation current for earth fault protection. The relay saw IDMT on one phase.
Any guess on the setting.

Grundig

 

[slavag]

Morning Grundig.
What is a CT size.
When you install this CT on HV, LV or on the neutral of trafo.
0.1A secondary?
Please send more information.
0.1A seems very, very low
Regards.
Slava

 

[vennivivi]

Hello Slavag,

Thanks for your prompt reply.The zero sequence CT is on the outgoing cable to the 11 k V side of the transfo. The ratio of the zero sequence CT is 20/1 and the 0,1 A is the secondary current.

Regards.

Grundig

 

[slavag]

Hi Grundig.
It's very, very low setting. is 2A primary.
What is a your 11kV network, solid grounding, low or high impedance or unerthed/ungrounded?
Please check shield connection of power cable.
Before few days we had topic on the issue. please see the attached.
thread237-211758

Next problem point is CT 20/1A, is not so good CT, for this small currents error is about 10%. prefferd 70/1 or 100/1A
Regards.
Slava

 

[vennivivi]

Hi Slavag,

Thanks again for your quick reply. Our 11 k V network is high-impedance grounded where the maximum earth fault current is limited to 10 amps and that`s the reason why we used a 20/1 CT.

Grundig

 

[slavag]

Ok, Grundig.
Next Q's, what is distance between trafo HV side and this CT and are your trafo neutral earthed. Seems as capacitance+resistance current in case of some faults in the networks, outside faults.
I recommend in this systems use directional earth fault function 67N , but with 70/1A--100/1 ( with relay's setting 1%) CT.
Regards.
slava

 

[cranky108]

With such a low setting, are you sure your not seeing an induced current from another source?

 

[vennivivi]

Thanks Slavag and Cranky108. The transformer is a delta connected on the 11 k V side and our network is a TN-C one.

Grundig

 

[slavag]

Hi Grundig.
From my point of view, you must use directionl earth/ground fault protection, 67N or up your setting, both current and time. Your setting is very low. As Cranky108 pointed, possible generation from another source (for example LV trafo side). Yes, resistor limmeted earth/ground fault up to 10A, but don't forget about capacitive currents of cables.
About IDMT on the residual principle. Possible, but I think isn't sensetive enough for high imoedance systems, min. setting is 10%-15% of phase CT size. Better add residual voltage protection 59N as back-up protection.
Regards.
Slava

 

[Marracuen]

How can I protect from earth faults the delta side of a transformer that is delta/ star with the star side directly grounded?

 

[AMBMI]

Marracuen,
You can use a "simple" 50N element. Obviously it will see only faults at the primary side (delta side). The secondary side earth faults are producing a zero sequence current which is "trapped" by the delta connection of the primary side.
It means that you can use a 100 ms or a 200 ms time delay for your 50N at the primary side without taking care of the time delay of the 50N elements at the secondary side.
By the way to come back to the "original" Vennivivi problem I think that a 20/1 CT could be fine. Also the 0.1 secondary setting is just fine the only problem is the time delay: you cannot use a no delay element, the element must have a 100-200 ms delay. It should avoid spurious trips. Depending upon the network configuration 67N elements could be used but only if the fault capacitive contribution of the branch with the transformer is greater than 2 A (primary)

 

[slavag]

Hi.
Two reasons, why I'm never use toroid/window/ring type CT 20/1A-50/1 A as source of 50N/51N for small currents and aspeacialy for the 67N function and especialy for the ungrounded and compensated systems.
1. Size of CT!!!
2.see attached
Regards.
Slava.
BTW, if used 67N, possible put current also 1A, 0.5 A primary with residual voltage block 10-20V secondary. No unwanted trips!!!!

 

[AMBMI]

Thanks Slavag!
Now I understand why usually the core CT are something like 100/1.
Please could you post the document you showed us as a picture?

Recently I saw a project where they were going to use a measurement CT (ratio about 100/1) to feed the 50N element and the max grnd short circuit current was 630 A (light R system)!!!
I wrote them that it was probably not the best solution but they kept it!

 

[slavag]

I'm sorry AMBMI, I cannot send this document, I wrote it for (this document base on reseach's documents of few relays mnf.) salesmans and it for internal use only.
But not once, I attached to here part of this documents ( with technical issue only).
Sorry again.
Regards.
Slava