Grounding/Bounding - Artic condictions

Status
Not open for further replies.

Lars

Member
Location
Arctic
I have used this forum before to nearly the same question, but by now I know a little more about my problem, so now I?ll try again.

In my job as an Electrical Technician I have come upon a serious problem and I hope this forum will be able to help me out.

A heat tracing hot wire had a simple ground failure, which cut off a 20A breaker along with a 100A main breaker in a 480V installation, probably because of a very high short circuit current (more than 3500A).

That was not the end of my problems. The entire high-voltage circuit (4,160V) was also cut off due to a HV ground failure.

The problem with a HV ground failure caused by a LV failure has happened before, somewhere else on this site.

The location is under Artic conditions and the grounding/bounding system, on which the 480V neutral point is mounted, is connected to a copper plate placed in seawater.
The HV grounding is connected to the same grounding/bounding system, and that might create a lot of strange paths for fault current to flow on, but that makes no sense to me. How can a high-voltage circuit identify a low-voltage ground failure?

The HV cable shield does not go through the doughnut CT, so that cannot have caused the failure.

It is an ordinary d-y transformer and the high-voltage circuit has been controlled for failures.

The relay is a Westinghouse type (50/51N instantaneous relay) being calibrated every 2 years.

What could be the reason for the above HV ground failure?

I will be patiently waiting in anticipation of hearing from you.

Thanks,
Lars
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Re: Grounding/Bounding - Artic condictions

Lars said:
A heat tracing hot wire had a simple ground failure, which cut off a 20A breaker along with a 100A main breaker in a 480V installation, probably because of a very high short circuit current (more than 3500A).
By "ground failure" do you mean ground fault? As in, an ungrounded conductor accidentally touched something 'bonded', and tripped the 20A breaker?

In short (no pun intended ha ha ugh), are you saying that a ground fault not only tripped the 20A branch circuit circuit breaker, but also the 100A feeder, and also the 4160V source?

I am very simple: but I don't see a grounding problem. I don't see a bonding issue. If everything is connected, then it's connected. From the fact that breakers are jumping all over each other to trip under a ground fault, it sounds as though your bonding system is spectacular.

It seems as though you have a selective coordination issue. As in, you need your breakers to trip in the correct order, but the available fault current is so high that all the breakers trip with the same current. In this, I can be of little help to you.

What help I can be is to look at this link I just found when searching google for "selective coordination".

Let us know if that helps. :)
 

Lars

Member
Location
Arctic
George

I thank all of you for your response to my little problem.

That?s right, the ground fault (of course not failure) tripped the 20A branch circuit breaker, the 100A feeder, and the 4160V source.

I grant you that I have a selective coordination problem between my 20 and 100 A. low-voltage breakers, which I can live with.

But I can?t live with the high-voltage ground fault - I can?t see how high-voltage circuits identify a low-voltage ground fault - That makes no sense to me.

I suspect that the particular arctic bonding system is a part of the problem, but someone must have some knowledge I don?t have about this problem - that?s why this thread is placed in the bounding/grounding part.

Thanks,
Lars
 

suemarkp

Senior Member
Location
Kent, WA
Occupation
Retired Engineer
If you're drawing excessive current from the secondary (low voltage) side of the transformer, doesn't that imply that the primary will also have an excessive current? The primary and secondary are indirectly connected, but they affect each other. I am surprised that the primary current is enough to trip that breaker though.

Would adding some more inductive reactance to the transformer provide enough of a delay so the secondary breakers trip first?
 

eric stromberg

Senior Member
Location
Texas
Lars,
Is there any way you could post a one-line diagram of this system?

Is the 4160 Volt system high-resistance grounded?

Eric
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
I'd be interested to see what the available fault currents are at each point, from the plans or whatever source you can glean that info from.

In the meantime, have you put some sort of time-delay fuse on the supply to the 4160V to try to compensate?
 

Lars

Member
Location
Arctic
Suemarkp
Eric Stromberg
Georgestolz

The high-voltage relay is a type Westinghouse CO-6 50/51 with Amp.Tap=0.5 and TimeDial=1. The CT is range 600/5A

The short circuit relay will react if the limit of current is exceeding the sets of current and time.
We have made Protective Device Coordination of all HV relays and the subsequent equipment.

As far as I know, the Westinghouse relay also has an instantaneous ground fault accessory, which reacts on perhaps a little currant.

I am informed that the HV fault was a ground fault. A secondary low-voltage ground fault should not have been sensed by the high-voltage protective relay as a ground fault ? perhaps as an over-current or a short-circuit fault, but the selectivity between those types of faults HV/LV should be 100%

THAT?S why I?m so confused.

Eric Stromberg:
I?m sorry, but this is a military installation, for which reason I?m not allowed to distribute any drawings (I have asked), but you can compare the installation with a single wire installation (no ring circuit).
I don?t think we have high-resistance grounding, which the below LineToGround data perhaps will make probable.

Georgestolz:
The below is a small part of the Protective Device Coordination Study based on our ESDA Advanced Analysis Program for calculations of short circuits.
The ground fault happen lesser than 30 ft. from the transformer (last calculation).
No, I have not put any sort of time-delay fuse on the supply ? I don?t dare because the fault is a ground fault.


HV feeder 4.16 KV (plant)
Three Phase Bolted Fault - Asym. - ?-Cyc = 24,394A
Line to Ground Fault - Asym. - ?-Cyc = 2,923A

HV feeder 4.16 KV (Primary Transformer)
Three Phase Bolted Fault - Asym. - ?-Cyc = 2,171A
Line to Ground Fault - Asym. - ?-Cyc = 1,133A

LV feeder 0.48 KV (Secondary Transformer)
Three Phase Bolted Fault - Asym. - ?-Cyc = 10,880A
Line to Ground Fault - Asym. - ?-Cyc = 11,935A

Again I thank all of you ? please let me know if any of you are getting wiser.

Lars
 

MJJBEE

Member
What is the instantaneous setting of the 4160 kV relay. It looks like the 4160 relay and the 100 A breaker are not coordinated properly also what is the transformer size?
 

Lars

Member
Location
Arctic
I thank you for the interests you have shown my little problem.

The subject is not a real heat tracing but a kind of heating wire for a door, connected to the light installation, which is probably the reason for the missing GFPE.

The settings of the Westinghouse CO-6 50/51 relay: Amp.Tap=0.5 and TimeDial=1. The CT is range 600/5A.

The 4.16/0.48 KV transformer is a 1000 KVA - %R=0.8 - %X=5.69

It is still my opinion that the real problem is not a coordination problem, but a to me unknown phenomenon. To the best of my knowledge, a high-voltage ground fault relay SHOULD NOT detect a low-voltage ground fault.

Maybe I am wrong, in that case I am very interested to hear your opinion.

Thanks
Lars
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Perhaps you can post a "verbal" one line of this? I can see bits and pieces, and they're not coming together very easily.

As in

1. 4.16kV supply xxxxA (24,394 AFC)
2. 4.16kV Panelboard xxxxA
3. 4.16kV feeder xxxxA (
4. 4160V - 480/277 XFMR (1000kVA, 10,800AFC)

And so forth. That would help a lot, and I imagine with a clear picture someone might seize on the problem.

(I agree with your idea that the HV GFP shouldn't notice a LV GF. I'm still thinking it was an overload to the HV and that's why the HV tripped.)
 

MJJBEE

Member
The way I understood it your one line looked something like this:

cordination11sf.png


If this is right your coordination looks like this:

cordination6ff.png


As you can see the coordination is not properly done also it depends on what type of breaker you have I assumed Cutler Hammer breakers but you could have any number of different breakers.
 

Lars

Member
Location
Arctic
Thanks a lot to both of you. I?ll try to use MJJBEE?s drawing, which is probably better than my own ?verbal? one line drawing.

The UTIL-1 = 4 generators ea. 4.16kV/3MW.

BUS-1 is placed right next to the transformers and from here the HV Cable Feeder (4/0-5kV) is outgoing.

The Cable is protected by a Westinghouse relay 50/51 CO-6. (Mechanical type ? from around 1965)
Settings: Amp.Tap=0.5 and TimeDial=1.
CT is range 600/5A.
BUS-1 - Three Phase Bolted Fault - Asym. - ?-Cyc = 24,394A

In the end of this cable (16000 ft.) a GOULD A055C1D0R0 CL-14 200E fuses is placed in the switchgear (primary transformer)
Three Phase Bolted Fault - Asym. - ?-Cyc = 2,171A

(There is an existing Protective Device Coordination Study between the relay and the HV fuses)

The 4.16/0.48 KV transformer is a 1000 KVA - %R=0.8 - %X=5.69

Secondary the transformer (0.48KV)
Three Phase Bolted Fault - Asym. - ?-Cyc = 10,880A
Line to Ground Fault - Asym. - ?-Cyc = 11,935A

That is what I have of exact calculations.

From secondary transformer to the main panel BUS-2, (contains a 600A Main breaker and a 100A breaker for the power panel) - about 20 ft. (very large MCM)

From Main Panel to Power Panel BUS-3 (contains a 20 A breaker for ?heat tracing?) about 20 Ft. (large AWG).

From Power Panel to Phase/Ground fault about 30 Ft. ? AWG 8.

My predecessor has measured the fault spot Phase/ground with a FLUKE Loop Impedance Meter - About 3,500A

That was a complicated story. I wonder if you understand just a part of it, but I?ll hope the best.

Thanks
Lars
 

MJJBEE

Member
First of all I think that I should show you the one line as I think you have described it.

cordination13kx.png



Tell me if I'm totaly off base but I think this is how you described it. Secondly it looks like you do have a coordination problem between the high and low voltage at about 3 kA this is a common mistake when doing coordination studies look at the plot below.

cordination7re.png


That green line is your CO-6 relay and the green shaded area is your 100 A breaker. To fix this problem you need to make some changes to your protection scheme. It probably means you need to change the relay setting on your CO-6. You do have a coordination issue that is why your 4160 feeder tripped.
 

Lars

Member
Location
Arctic
MJJBEE

I notice that you have described the circumstances to my problem correct.

I agree with you that its looks like I have a coordination problem between the 100A breaker and the CO-6 in the area around 700 to 5000A.

The fault happened after the 20A breaker. As I can see on your drawing, there will be time selectivity between the 20A breaker and the CO-6. Do you agree on that?

The coordination problem between the 20A and the 100A breaker is expected at high level short-circuits, but I can hardly live with the high-voltage interruption.

By the way, when I?m corresponding with experts.
I?m normally working in a Scandinavian country with other systems and components, that?s why the different HV relay types seems a little strange to me.
Were the HV relays are placed, there are 3 relays CO-8 ? 51. These are over-current relays, 1 ea. phase I suppose.
Furthermore there is the CO-6-50/51 instantaneous relay ? I suppose that is a combined relay for short-circuit and ground fault.

If it is:
1. What current will get the relay to trip when the Amp.Tap=0.5 - CT=600/5A?
2. What is the duration before reaction when the TimeDial=1?
3. What is the current and duration in case of ground fault?

Or can any of you tell me where on the Internet I can get that kind of information.

Thanks again. Your assistance so far has been highly appreciated.
 

MJJBEE

Member
Lars said:
The fault happened after the 20A breaker. As I can see on your drawing, there will be time selectivity between the 20A breaker and the CO-6. Do you agree on that?

I agree with your statement. It should have worked that way. However it may be that your breaker didn't trip the way it was supposed to or the breakers you have don't have exactly the same curves. There should be time selectivity even with the opening time for your breakers.

When you mention the other relays do you know for sure which one tripped the breaker?

The current that will trip the CO-6 relay is 60 A at 4160. The duration is based upon the fault current and follows that curve. Also is the transformer delta-wye or wye-wye?
With the setup I have shown the fault currents look like this


faultcurrent3ou.png
 

Lars

Member
Location
Arctic
MJJBEE

As far as I am told, a ground fault in the heat tracing was interrupting both the 20A breaker and the 100A main breaker placed in the main panel.

An Electrical Technician first shut down the 20A breaker and then the 100A main breaker ? And at this time both breakers and the HV circuit interrupted.

I have had a correspondence with SIEMENS about the 20/100A relays, and they informed me about coordination problems larger than 2 kA ? But this does not explain why the CO-6 relay tripped.

Could the reason be that the 20A breaker is defective? Maybe that could be the explanation.

If so, do any of you know if any gear to check out the magnetic release function on this older breaker exists?

That would be a nice solution to a complex problem.

Thanks
Lars
 

MJJBEE

Member
Lars said:
MJJBEE

Could the reason be that the 20A breaker is defective? Maybe that could be the explanation.

If so, do any of you know if any gear to check out the magnetic release function on this older breaker exists?

It could be that you have a defective breaker. I would send it out to a breaker shop to test it if it were larger. At that size I may just replace it and then test my fix.
 
Status
Not open for further replies.
Top