High Impedance Ground

Status
Not open for further replies.

wiigelec

Member
Location
Red Desert

Anyone test their high impedance ground system by connecting the load side of a spare bucket to the MCC grounding lug then closing in the breaker?
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)

Anyone test their high impedance ground system by connecting the load side of a spare bucket to the MCC grounding lug then closing in the breaker?

Yes works great. Check the size of the resistor. Only two I have seen are 28 ohm or 56 ohm, cooresponds with 5A or 10A ground currents.

20A TM CB bucket works great. As I'm sure you knew, just connect one phase of the CB loadside to ground. #12 or #14 is fine. Leave the other two phases open.

Measure the current with a clamp-on ammeter. Turn on the pulser and one can see the current pulse on the clamp-on.

ice
 

eric9822

Senior Member
Location
Camarillo, CA
Occupation
Electrical and Instrumentation Tech
We built a box with a switch, a current limiting 7.5A 100KA interrupt rated fuse, and a really long cord. Connect lead one to any phase of a spare disconnect and lead 2 to ground and then move somewhere far away. Typically we only perform this test on new installations or if we had to repair the ground fault detection system. We also do this when the load on the substation is low or non-existent so we can turn the trip settings down. It is critical that you ensure you don't already have a ground fault prior to conducting this test. This confirmation must be made independently of the ground fault detection system you are going to test. If a ground fault already exists you will cause a phase to phase fault if you pick the wrong phase to connect to ground. What you suggest will work but you will be right there if something goes wrong.
 

eric9822

Senior Member
Location
Camarillo, CA
Occupation
Electrical and Instrumentation Tech
Actually when I saw this in our maintenance plan I thought it was absurd, but if others are doing the same thing...

I just wouldn't do it the way you described in the original post. We had a ground fault detection system fail a few months ago because a PT fried. Of course we had a ground fault and nobody noticed for who knows how long because the only thing people were looking at was the idiot light and ignoring the ground fault ammeter. We have changed our rounds procedures to look at both the idiot light and the ammeter. Personally I would not want to induce a ground fault as part of a regular maintenance procedure since there are other ways to determine if the detection system is working provided it was designed properly. Newer substations have built in test functions if you order them that way. I just finished commissioning a new Eaton substation with a built in test function and the pulser function that ice described.
 

BJ Conner

Senior Member
Location
97006
High reisitance ground systems.

High reisitance ground systems.

The theory behing high resistance ground systems was (is) your system can have a line to ground fault and stay in operation.
The systems have ground detection systems that tell you the system is grounded but not where.
GE (and others) use a system that locates the ground by pulsing ground currrent by changing the ground resistance slightly.
This is done when a ground is detected. The system is usually timed, you have time to find and isolate the fault, if your to slow the system shuts down.
Ge supplies a pulse detector that lets you find and isolate the fault.
No pictures of the detector in here but it's loop with a handle, about the size of a ping-pong paddle.

http://www.geindustrial.com/publibr...NR=Installation and Instruction|GEI-72116|PDF
 

SG-1

Senior Member
I just wouldn't do it the way you described in the original post. We had a ground fault detection system fail a few months ago because a PT fried. Of course we had a ground fault and nobody noticed for who knows how long because the only thing people were looking at was the idiot light and ignoring the ground fault ammeter. We have changed our rounds procedures to look at both the idiot light and the ammeter. Personally I would not want to induce a ground fault as part of a regular maintenance procedure since there are other ways to determine if the detection system is working provided it was designed properly. Newer substations have built in test functions if you order them that way. I just finished commissioning a new Eaton substation with a built in test function and the pulser function that ice described.

The Eaton Medium Voltage HRG test functions only check the health of the control/pulsing circuit, not the detection circuit. A horn could be added to back up the idiot light.
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
My comments here are limited to 480V Y HRG. 13.8kv is somewhat different. I have never seen a 208Y or 240D with an HRG

Actually when I saw this in our maintenance plan I thought it was absurd, but if others are doing the same thing...

... Personally I would not want to induce a ground fault as part of a regular maintenance procedure ...
I wouldn't either. Commissioning - yes. And, possibly some troubleshooting shooting scenerios - yes. But not for a regular PM.

... Newer substations have built in test functions if you order them that way. I just finished commissioning a new Eaton substation with a built in test function and the pulser function that ice described.
I don't think I have ever seen an HRG installation that didn't have test functions (However, I'm sure someone, somewhere has been or is being dumb enough to do that) .

My experience includes work on early 60s installations (I didn't put them in - just worked on them **) Pulsers used a clock motor - one of those little 1.5" canned, geared, motors with cam and microswitch. Individual phase ground detection could be three lights. Alarm functions from a relay across a transformer across the grounding resisitor

Newer ones are electronic timers and switches, electronic phase voltage and fault current detection. Startco makes a good one as well as GE and Eaton. Personal experience is the newer ones are not particularly any more reliable than the old style. Both work fine (until they don't) :)

... GE (and others) use a system that locates the ground by pulsing ground currrent by changing the ground resistance slightly. This is done when a ground is detected. The system is usually timed, you have time to find and isolate the fault, if your to slow the system shuts down. ...]
Most installations I have seen are 58 ohm, 5A continuous resistors. The pulser shorts out half of the resistor. Fault current goes from 5A to 10A - maybe 2 sec ON, 2 sec OFF. No automatic start on the pulser - Acutuated with a front panel test switch. With one fault, systems will stay energized forever. And they should. That is the whole point of a 480Y HRG system.


... Ge supplies a pulse detector that lets you find and isolate the fault. ...]
This is the one part of the system that has not changed at least as far back as 60s installations. They still function the same, and they even look the same. I bought one a few years ago and it looked exactly like the ones that came with the 60s installs.

The device is a clamp on ammeter, analog meter, 50A range, with a clamp big enough for a 3" RSC. They work great for seeing a fault current pulsing from 5A to 10A and back - unless one gets unlucky and the fault current is coming back on the same conduit.

ice
 

eric9822

Senior Member
Location
Camarillo, CA
Occupation
Electrical and Instrumentation Tech
The Eaton Medium Voltage HRG test functions only check the health of the control/pulsing circuit, not the detection circuit. A horn could be added to back up the idiot light.

I guess I need to do some research. That test switch wasn't covered very well in the documentation we recieved so we tested it the old fashion way. It has a horn as well as a set of dry contacts for remote alarming.

I don't think I have ever seen an HRG installation that didn't have test functions (However, I'm sure someone, somewhere has been or is being dumb enough to do that) .ice

I guess we were dumb enough to do that, none of our older installations have test functions built-in.
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
... I don't think I have ever seen an HRG installation that didn't have test functions (However, I'm sure someone, somewhere has been or is being dumb enough to do that). ...

... I guess we were dumb enough to do that, none of our older installations have test functions built-in.
Interesting.

First I would say careful with the "we". I've read a lot of your posts and I highly suspect, "you" were not part of the "we" that decided not to put in any test functions.

Second:
No test functions at all? really? No pulser shorting out part of the grounding resistor? No alarm relay across the grounding resistor? Not even phase to ground lights?

Yuk. You got really unluckly inheriting that thing.

ice
 

eric9822

Senior Member
Location
Camarillo, CA
Occupation
Electrical and Instrumentation Tech
Interesting.

First I would say careful with the "we". I've read a lot of your posts and I highly suspect, "you" were not part of the "we" that decided not to put in any test functions.

Second:
No test functions at all? really? No pulser shorting out part of the grounding resistor? No alarm relay across the grounding resistor? Not even phase to ground lights?

Yuk. You got really unluckly inheriting that thing.

ice

You are correct on your first point but the detection system is pretty good IMHO. The detection system consists of a PT across the resistor that is tied to a 59G overvoltage relay which is then tied to an 86G Lockout relay that turns off a local indicator (lamp on means good, lamp off means bad) and opens a set of contacts that triggers an alarm at our control system. We also have a CT around the grounding conductor that is connected to a local ammeter showing ground fault current.

As far as testing we don't have pulser functions. Ground current is held at a constant value and we go hunting with ammeters until we find the problem. It works and we usually find the problem pretty quickly. To be honest I had never heard of a pulsing system until we were looking to purchase this new sub. I lead a sheltered life since I have been in the same facility for a long time.
What exactly is the "old fasion way" of testing/

ice

We induced a ground fault using our previuosly discussed ground fault testing box.
 
Last edited:

wiigelec

Member
Location
Red Desert
So the important things to check here are the alarming using the test button on the resistor system, and verifying ground connection with visual inspection and clamp-on ground resistance tester which should see the value of the grounding resistor...
 
Status
Not open for further replies.
Top