corner ground

[mickeyrench]

I was doing some pm's today at one of our stations and i took voltage readings and what i found was a surprise. The voltage at L1 - ground was 495 volts and L2- ground was 495volts and L3-ground was 0 and L1-L2 was about 495volts .
Is there a different way to meg L3 being it is grounded?
At another location we have a ground detection system and L1 AND L2 lights were bright and L3 was dim. I believe L3 has a ground somewhere , if so why doesn't the cb trip out? I went outside to the service conduit that enters the building and goes to the poco transformer than down to us. Outside I found a large tree growing around the conduit coming into the building. It appears to have bent the conduit some. Could this be the cause of the ground fault? I think if i open a cb 1 at a time i will locate the ground . both services are 3 wire 480 volt. Only sure about one being a corner ground service.

thanks for any input

 

[augie47]

I was doing some pm's today at one of our stations and i took voltage readings and what i found was a surprise. The voltage at L1 - ground was 495 volts and L2- ground was 495volts and L3-ground was 0 and L1-L2 was about 495volts .
Is there a different way to meg L3 being it is grounded?

If you are megging the line itself, for an insulation check, you would need to unground it, I believe. Since it is a grounded conductor, I know no reason you would want to meg it.

At another location we have a ground detection system and L1 AND L2 lights were bright and L3 was dim. I believe L3 has a ground somewhere , if so why doesn't the cb trip out? I went outside to the service conduit that enters the building and goes to the poco transformer than down to us. Outside I found a large tree growing around the conduit coming into the building. It appears to have bent the conduit some. Could this be the cause of the ground fault? I think if i open a cb 1 at a time i will locate the ground . both services are 3 wire 480 volt. Only sure about one being a corner ground service.

thanks for any input

The first one may be "accidently" grounded since you have no indicators, although they were not required in earlier Codes.
Opening one circuit breaker at a time should help locate the ground unless you have more than one.

 

[LarryFine]

I was doing some pm's today at one of our stations and i took voltage readings and what i found was a surprise. The voltage at L1 - ground was 495 volts and L2- ground was 495volts and L3-ground was 0 and L1-L2 was about 495volts .
Is there a different way to meg L3 being it is grounded?

First of all, I'd ascertain that it's an intentional grounding of L3. The lack of ground-detection here suggests it is, and L3 was the traditional phase to ground.

Simple answer: temporarily remove the bond to perform the test. As with any grounded conductor, there should be no other bond on the load side of the disco.

At another location we have a ground detection system and L1 AND L2 lights were bright and L3 was dim. I believe L3 has a ground somewhere , if so why doesn't the cb trip out?

Because there is no other supply conductor grounded. That's the main reason for using an ungrounded system: so the first fault (unintentional grounding) does not trip the breaker.

Draw the 3-ph supply and a simple load. Ground one conductor and try to trace a circuit (conductive pathway) that uses the ground to allow a tripping current to flow.

I went outside to the service conduit that enters the building and goes to the poco transformer than down to us. Outside I found a large tree growing around the conduit coming into the building. It appears to have bent the conduit some. Could this be the cause of the ground fault?

Highly unlikely, unless it's kinked. Opening the main and testing voltages to ground again should show you which side of the main is the fault located.

I think if i open a cb 1 at a time i will locate the ground . both services are 3 wire 480 volt. Only sure about one being a corner ground service.

That's a good method. You might turn on each circuit individually just in case there happens to be more than one fault, coincidentally on the same phase.

In the future, by the way, I recommend using a low-impedance tester instead of a voltmeter for most troubleshooting, unless you need to know the specific voltage.

 

[brian john]

If you are megging the line itself, for an insulation check, you would need to unground it, I believe. Since it is a grounded conductor, I know no reason you would want to meg it.

When isolated from the bond you meg it to verify insulation and that this conductor is free from additional grounds.