Ungrounded Panelboard

[dcheser]

We are working in a plant where we have found several 120/208 VAC 3-phase, 4-wire panelboards that are ungrounded (no ground conductor or ground bus). Should we at a minimum have an electrician bond the panel(s) to the building ground grid or leave it as is?

 

[infinity]

Is there a metallic conduit feeding the panel? If is it may be fine as is.

 

[dcheser]

Yes, The Panel(s) are fed with Rigid Metal Conduit (RMC) and all loads from the panel are also run in RMC.

 

[petersonra]

Yes, The Panel(s) are fed with Rigid Metal Conduit (RMC) and all loads from the panel are also run in RMC.

If so, it already has an equipment grounding conductor as RMC is allowed to be used as an EGC.

 

[dcheser]

If so, it already has an equipment grounding conductor as RMC is allowed to be used as an EGC.

Is there anything in the code that we can reference with respect to this?

 

[infinity]

Is there anything in the code that we can reference with respect to this?

250.118 lists equipment grounding conductors and RMC is on the list so there is no code issue with your installation.

 

[ActionDave]

Is there anything in the code that we can reference with respect to this?

As well as being code approved it is also worth noting that RMC is a better equipment ground than any copper wire you could stuff in it.

 

[user 100]

As well as being code approved it is also worth noting that RMC is a better equipment ground than any copper wire you could stuff in it.

FWIU, during a fault with rmc/emt with a wire type egc pulled, the bonded pipe will actually carry more of the fault current than the wire. All that extra conductive surface makes a great low r path.

 

[alan232]

Are bond bushings required when using a conduit as the EGC? I have not been able to find any requirements in the NEC.

 

[infinity]

Are bond bushings required when using a conduit as the EGC? I have not been able to find any requirements in the NEC.

With regards to your question in the OP no not required.


Is the voltage over 250 volts to ground? If yes then maybe, if no then probably not required with a possible exception.

 

[2Broke2Sleep]

FWIU, during a fault with rmc/emt with a wire type egc pulled, the bonded pipe will actually carry more of the fault current than the wire. All that extra conductive surface makes a great low r path.

You should check out the GEMI software. It's free. Has to do with exactly what you are talking about.

 

[cuba_pete]

I think this guy was just in my building. He didn't think that it was cool that we had no EGC's.

dcheser, you in WA lately?

 

[jtinge]

As well as being code approved it is also worth noting that RMC is a better equipment ground than any copper wire you could stuff in it.

We have an installation that is pre 1970 and metal conduit was routed from the unit substation in concrete below grade to the building panelboards. The RMC was used as the EGC's. We have had several instances while replacement of several panelboards and MCC's and the supply feeders where we had to abandon the feeders because the conduit had deteriorated and the old feeders could not be removed. Given the probability that other feeder conduits from this era may have also corroded or deteriorated, can these still reliably be considered adequate EGC's. The fear being that the conduits cannot be relied upon to be a continuous low impedance fault path back to the source. We are currently planning to upgrade the supply unit substation and the question about needing to add EGC's for the load side feeder conduits has come up. How can we determine the adequacy of the existing feeder conduits as EGC's?

 

[jaggedben]

... How can we determine the adequacy of the existing feeder conduits as EGC's?

Spool out a wire above ground from one end to the other, terminate it to the enclosure at one end, then measure ohms from enclosure to wire at the other end? Subtract the known resistance of the wire you spooled out, compare the resulting ohms to that of a would-be copper EGC. If they're equal or less, you're okay.

You need half decent estimates of the length of conduit and spooled out wire, but if the conduit is still good it should exceed the requirement by a significant factor so precision is not essential. Underestimate the distances to be conservative.

Maybe someone will suggest an instrument that does it by itself, but that would work. I suppose if you could temporarily take the lines out of service you can use one of the existing conductors instead of spooling one out.

I suppose a simpler way to start would be to measure ohms from grounded conductor to enclosure at the panelboard. It won't necessarily give you enough accuracy if the RMC is borderline, but if you get an open line to start off then you know you have a problem.

 

[jtinge]

Thank you, sounds doable. We could initiate a task to document EGC conditions as described during a weekend planned outage.


Sent from my iPhone using Tapatalk