grounding at detached garage

Status
Not open for further replies.
don_resqcapt19 said:
[FONT=&quot] I think that the point that Gmack is trying to make is the same one that Bennie made a number of times before his passing. The impedance of the fault return path for a second building with a 3 wire feeder is almost always less than that of a feeder with a 4 wire service. The feeder neutral will almost always be larger than the EGC for that feeder. If you have a 3 wire feeder and a bonding jumper at the second building the ground fault current is carried on the neutral back to the source. If you have an EGC that current is on the EGC back to the service main bonding jumper and then on the neutral to the source. The second path has a higher impedance and will change the opening time for the OCPD. Assuming that the wire sizes in Table 250.122 are adequate, the trip time change should not be of any consequence.
Don
[/FONT]
Click to expand...

Yes Don, fine. But the shunt occurs at first building GC [distance]

But with a parallel "extra" path the others problems can and do occur.

Take a HO or "maintenance worker landing a "neutral on a ground bus at second building.

Or: neutral separation and load using 4wire EGC and branch EGC for "neutral"

A MBJ solves many issues IMO.

What advantage is there with a 4wire.

Lets install that 4wire in IMC.

How many parallel paths do we have?
 
Gmack said:
Yes Don, fine. But the shunt occurs at first building GC [distance]
Click to expand...

This is the same mistake in thinking I had on this subject a while ago.

The point that the bonding jumper is placed is irrelevant.

The distance remains the same.

The 'distance' is from the fault to the source regardless of the bonding point.

As far as the "what if's" the NEC is written assuming people will follow the rules.
 
Stick

Better get the electrical police right on this case
Click to expand...

If that's what it takes. ;) :D

For what it's worth I have a CT J-man license and work in CT occasionally.

I have seen more than a few irregularities.

Here on this forum we discuss the rules, in the real world rules are often ignored.
 
georgestolz said:
His or her boss. :)

Don't worry about the debate, it's already begun. :D
Click to expand...

LOL

You all have fun. I've got to go to town, been watching the Deer in the front yard waiting to leave.

Roger
 
Gmack said:
What advantage is there with a 4wire.
Click to expand...
If the neutral at the remote structure opens, then none of the equipment becomes energized from the open neutral. The neutral is seperate from the grounding conductors, so that is the advantage.

Gmack said:
Lets install that 4wire in IMC.

How many parallel paths do we have?
Click to expand...
Do you refuse to use metallic conduits with EGCs normally (inside a single structure), due to the two paths connected in parallel? More paths for the EGC mean lower resistance. 250.148 effectively requires parallel paths when the chance arises. :)
 
iwire said:
Stick



If that's what it takes. ;) :D

For what it's worth I have a CT J-man license and work in CT occasionally.

I have seen more than a few irregularities.

Here on this forum we discuss the rules, in the real world rules are often ignored.
Click to expand...

I didn't want to start an uproar, I was only stating how we do a remote panelboard install, but I will find more info about this policy... I'm sure everyone would find irregularities in every state...
 
stickboy1375 said:
One more post about 3 wire and 4 wire feeds to remote panelboards, what about the effects of GFI's and AFI's? if any?
Click to expand...
There's nothing special regarding remote panelboard feeder options and AFCI/GFCI's.
 
Bob,
The point that the bonding jumper is placed is irrelevant.
The distance remains the same.
The 'distance' is from the fault to the source regardless of the bonding point.
Click to expand...
While the physical distance is the same, when you use a 3 wire feeder with a bonding jumper at the second building, the inpedance of the fault return path for a ground fault at the second building is less that of a 4 wire feeder with the bonding jumper only at the main service disconnect. This is true in the vast majority of the cases where the grounded conductor of the 3 wire feeder would be larger than the EGC of the 4 wire feeder.
Don
 
don_resqcapt19 said:
the inpedance of the fault return path for a ground fault at the second building is less that of a 4 wire feeder with the bonding jumper only at the main service disconnect.
Click to expand...

Don I am well aware of this, I already posted about that in this thread.

I don't believe that is what he was saying.

It seemed to me that he was saying the closer the bond point the better regardless of the conductor sizes.
 
iwire said:
Don I am well aware of this, I already posted about that in this thread.

I don't believe that is what he was saying.

It seemed to me that he was saying the closer the bond point the better regardless of the conductor sizes.
Click to expand...

No. Don got it right.

There were other issues as well.

I did mention a "full sized" GC.
 
iwire said:
Well strangely I agree.

In certain circumstances for remote buildings I believe a 3 wire feeder with it's larger grounded conductor for a fault path is better.

However if there are other bonded metallic paths between the buildings than my opinion quickly changes.

In 2008 the NEC will eliminate the 3 wire option all together, which IMO is ridiculous as long as we get 3 wire services from the POCO.

If a service is safe than a 3 wire feeder to a remote building must also be safe.
Click to expand...

Yes, strangely "we" agree.
 
georgestolz said:
Gmack,

While I agree that in many cases the three-wire feeder is preferable, there are certain things you've said I take issue with.

That's a sloppy use of words, and can lead to confusion. An outside feeder to a structure is not the same as a "service." The NEC has defined "services" and there are certain rules that do not carry over to feeders to remote structures fed from the same service.


Why are you concerned about a fault's interaction with any building's Grounding Electrode System?
Click to expand...

1] First I will respond to an earlier post where you stated that a open neutral 4wire had the advantage because building #2 would be in effect "isolated".

A] Well where do you think all that "floating" neutral current is?

B] It is sitting righ back at you in building #2 GES and all branch EGC's, complements of your EGC run with your 4wire.


My answer:

Go back and read 1] A&B.

Potential is very serious business in our line of work.

It would be "sloppy" to not be concerned.

Your question as quoted:

"Why are you concerned about a fault's interaction with any building's Grounding Electrode System?
 
Gmack said:
.
Slow down.

We can run a 3 wire feeder without a EGC at a separate building if no parallel paths are present.

Example: PVC. Then treat the separate building as a service and bond everything at main and install GES as usual.
Click to expand...
Yes but it is not recommended, and not allowed in Washington State since Aug 1973. And this allowance for a this practice will be (likely) deleted in the 2008 NEC.
 
This sure seemed to be a HOT topic... but really if it were so easy to tell which way was better than why are we even talking about this, there must be some point that we are all missing... and why is one way better than the other? Is it only under certain conditions? I've never had a problem running 3 wires to a remote panel board, so unless someone can PROVE without a reasonable doubt why one way is better (other than some book that some group of people wrote (nec)) than I don't have a problem with it... (Is there anyone else from connecticut on this forum? And have talked to Bob Nuzzi?)
 
Last edited:
tom baker said:
Yes but it is not recommended, and not allowed in Washington State since Aug 1973. And this allowance for a this practice will be (likely) deleted in the 2008 NEC.
Click to expand...

Even if it is deleted, does not mean a state will adopt it...
 
Stickboy, the problem is Parallel Neutral paths as has been stated many times in this thread. If there are no parallel paths there is no problem.

Imagine if you will a Coax between the two structures for example. If this coax is commonly bonded (through whatever means) to the Grounded Conductors of each of these structures and the Grounded Conductor is lost at the second building, we have the coax shield trying to carry the unbalanced neutral current back to it's source, not a good option is it?


Roger
 
tom baker said:
Yes but it is not recommended, and not allowed in Washington State since Aug 1973. And this allowance for a this practice will be (likely) deleted in the 2008 NEC.
Click to expand...


It will not improve anything.

Under sized EGC's cannot do the job of a full size GC as we are relating to here.
 
Status
Not open for further replies.
Top