CT grounding 1

[mirilovic]

After reading this thread thread240-191665 and this article:

http://www.netaworld.org/files/ItemFileA877.pdf,

I need to ask one question to get clear something to myself.

Article mentioned above, recomends grounding of CT's in protective relay house. I also found same recomendation in some other sources.

Why is better to ground CT in relay house then on one secondary terminal of CT itself?

 

[slavag]

Hi.
We always use only one rule of thumbs.
1. grounding CT's only in one place , near to CT as possible. At the first place, where you can easily connect or disconnect this grounding.
Regrds.
Slava.

 

[davidbeach]

Our standard is to ground the CT circuit at the first terminal block inside the control house.

 

[stevenal]

From IEEE C57.13.3:

The instrument transformer secondary circuits are generally grounded at one point. The circuits shall
preferably be connected to ground at the first point of application of the circuit (switchboard or relay
panel). This practice provides maximum protection for personnel and connected equipment at the
switchboard, which is the location where overvoltages are likely to be experienced.

 

[521AB]

SLAVA: "grounding CT's only in one place , near to CT as possible".

I fully agree with that.

 

[jghrist]

SLAVA: "grounding CT's only in one place , near to CT as possible".

I fully agree with that.

You left out the second part of slavag's statement.

At the first place, where you can easily connect or disconnect this grounding.

This place is normally the "first point of application of the circuit (switchboard or relay panel)." Also, as indicated in IEEE C57.13.3, this is where personnel are more likely to be exposed to any high voltages.

 

[521AB]

Well, I have omitted it because I thought it was enough to say "as close to the CT as possible".
That's means in the switchyard. IEC practice.

I have talked right now with some colleagues about your post, jghrist, and they commented that your post is "typical ANSI". Am I right?

"We" do not earth at cubicle location. Of course "we" do earth the cubicles, but this is another issues. From the CT, we get 4 cables: L1,L2, L3 and N. L1, L2, L3 enter the relay input CTs. The output of the input CTs are connected together and this is where we connect "N", which goes back to the main CT, which is earthed close to the CT.

I hope I have managed to explain this..

but of course, forgetting that "everybody believes he's doing right", it should be interesting to know the motivations from both proposals, if you can remember them. because for me it has just been accepted as "normal practice".

 

[stevenal]

Should the CT secondary become energized at primary voltage levels, you will have a ground potential rise at the ground location. If this location is remote from the control building, this potential can be transfered via the secondary wiring to the building, putting instruments and humans at risk. Probably an unlikely occurrence, but so easy to fix by grounding at the control building.

 

[slavag]

I would like add why I wrote first place or of course first terminal. I think we told about same.
For example, if its generator, grounding usually connected at the junction box near to generator neutral point and on the ( if I remeber right) on the glad box for the stator Ct's.
If its GIS, usually is LCC
If its outdor swg, first CT junction box.
Regards.
Slava

 

[jghrist]

Should the CT secondary become energized at primary voltage levels, you will have a ground potential rise at the ground location. If this location is remote from the control building, this potential can be transfered via the secondary wiring to the building, putting instruments and humans at risk. Probably an unlikely occurrence, but so easy to fix by grounding at the control building.

As a "typical ANSI" guy, my first inclination is to agree. But look at the two attached sketches. With the CT grounded at the relay panel, the touch voltage would be I·(Rlead + Rstinger). With the CT earthed at the breaker, the touch voltage would be I·Rstinger only.

 

[jghrist]

Here's the ANSI sketch.

 

[stevenal]

jghrist,

Looks like you assumed that GPR at both locations is the same. I was looking at the transfered potential that is there because of unequal ground potential rise.

 

[scottf]

I think we may be mixing a few subjects here with the talk of GPR.

The reason that CT secondary circuits need to be grounded is to give the secondary winding some form of ground reference to prevent transferred voltages via capacitance coupling between the secondary winding and the primary winding. From this perspective, where the ground is applied, really doesn't make a difference.

From a current measurement accuracy perspective, it's very important to only ground at one point. From a strictly theoretical perspective, to best guard against EMI, it's best to run both leads all the way to the control house and then ground one there. That way there is opposing current throughout the entire wiring run. However, it's not commonly done that way except in EHV and UHV applications.

 

[etronics]

STEVENAL refereced IEEE C57.13.3:

We also follow this convention from the IEEE. It is important to have only 1 ground in the system. and the ground should be at the first point in the relay bldg. We do not ground at the breaker.

I would add, we follow this same guide for CCVTs/PTs. I find a lot of situations where many people are not familiar with the IEEE spec. on this.

The thing I always keep in mind is that you are grounding first and FOREMOST for personell protection.

 

[jghrist]

Looks like you assumed that GPR at both locations is the same. I was looking at the transfered potential that is there because of unequal ground potential rise.

In most situations, the GPR is nearly the same at any point in the station. There is significant voltage drop in the grid wires only where the station is very large and the soil resistivity is low.

 

[stevenal]

Proper ground grid design goes a long way in mitigating the transfered potential I described. When it comes to safety, I like to have a belt along with my suspenders, especially if the belt is free.

The transfered potential is my way of explaining how the first point of application provides for "maximum protection for personnel." as the standard states. Anyone have other ideas on this? Or is IEEE full of excrement?

This is not a new exercise for me. Older versions of the ANSI metering standards had the CT secondary grounded as close to the CT as possible. The newer versions have aligned with IEEE, but habits die hard and I had to go further than "IEEE says so" in defending the change.

 

[slavag]

As close to the CT as possible.
Many times people understand it not so right and connect groundung on out of CT's and on each phase.