No ground at garage sub panel

Status
Not open for further replies.
dhalleron

dhalleron

Senior Member
Location
Louisville, KY
Occupation
Master Electrician/Senior Fire Alarm Technician
Sorry so long a post. I’ve looked over the forum and I understand the correct way to do this. Can anyone help explain the dangers? I think most likely there will never be a problem, but that one time could happen.

We are taking over an installation where the owner is going to sue the previous electrician and the inspector is working on getting his license revoked. This guy did not use bushings or connectors on metal boxes and Romex, he did not ground the boxes but he did ground the receptacles that do not have self grounding straps. He installed Romex under a deck to a weatherproof box in a wet location. He ran USE cable from underground inside a wall with no conduit and into the bottom of the sub panel.

With all that said, my main issue is something the inspector is planning to allow and my boss with 30 years experience does not seem to fully understand the danger. My boss is good at what he does and normally does things neat and correct. The main reason it is planning to be allowed is that concrete has already been poured for a nice sidewalk and patio that would prevent the cable to be easily replaced.

The feeder from the house to the garage about 50 feet away is installed using underground USE with three conductors - two hots and one neutral. The sub panel has no equipment ground connected directly to the main panel. It does have two ground rods installed. The neutral is floating on its own neutral bar. The branch circuit grounds all tie to the ground bar that is only grounded by the driven ground rods.

If there is a direct hot to ground fault I think the current will need to go to the ground bar through the earth ground and 50 feet back to the main panel before it can meet back up with the neutral and cause enough current to flow to trip the branch circuit breaker in the garage. If the dirt is dry it might not trip. If someone installs a pool or somehow gets in between the garage and the house in the dirt, they could become part of a fault circuit back to the main panel.

If this situation was going to be allowed, would it at least not be best to bond the neutral and grounds in the subpanel? At least then it wouldn’t depend on the dirt all the way back to the main panel. I know this still would not be correct.

I know many garages are wired wrong and have not had a problem. This tie since the owner is planning on suing the last electrician; I want to try to make sure we are in the clear with our work. Thoughts?
 
Last edited:
Garage is detached 50 feet from house with PVC water pipe connecting to house with no metal connections such as gas pipe.

The sidewalk is really too wide and goes all the way up to the garage wall to drive a pipe under. The only way I see to fix the wiring will involve busting concrete and a trench.

I know the correct way is to run the equipment ground. I just want to hear from people that can describe the dangers of leaving it without a ground connecting to the house panel or prove to me that it can be configured as safe without the ground to the house (which I don?t think can be done).
 
Thanks Brian John. I know an extra is in order, but this whole job is an extra for us. I know my boss is not going to run the ground and I think the inspector is going to allow it. I know it does not meet code, I don't think it is safe and I don't like it.
 
dhalleron said:
I know it does not meet code, I don't think it is safe and I don't like it.
Click to expand...

If this job was permitted under the 2005 NEC or before it does in fact meet code to use the neutral as the grounding means to a detached structure.

See 225.32(B)(2), there is no danger, it is exactly the same as the service feeding the main panel.

However this was changed under the 2008 and if the job is under the 2008 then a separate EGC and Neutral must be run.
 
Thanks iwire. I guess I'm not sure if it met 2005 code. Maybe I've been unsure about sub panels in a separate structure. I'll read up on it.

If it did meet code, I still think the neutral would have to be bonded to the ground at the garage exactly as it is at the house. How else can a ground fault properly trip the breaker?

I've drawn it out on paper and the current would have to go from the hot breaker to ground in the garage somewhere, to the driven ground and all the way back to the house panel through the dirt to complete the circuit with the neutral.
 
dhalleron said:
If it did meet code, I still think the neutral would have to be bonded to the ground at the garage exactly as it is at the house. How else can a ground fault properly trip the breaker?
Click to expand...

You are 100% correct, you would have to treat it exactly like a service to meet all the NEC requirements :smile:
 
Thanks for the replies.

I'm going to go with Bob iwire's post. I'll review my codes and see which one we are required to meet and if we can go with the 2005 we will bond the grounds and neutrals at the garage panel and I'll feel better about the install.

I sure need more than my annual continuing education classes to keep my master electrician license in Kentucky since I do this only part time. So many details are lost in my mind over time. No matter what, I still know or care more than the electrician we replaced on this job.
 
That's the kind of electrician we refer to as "knows just enough to be dangerous."

dhalleron said:
If this situation was going to be allowed, would it at least not be best to bond the neutral and grounds in the subpanel?
Click to expand...
Absolutely, as long as, as Bob pointed out, the 2005 (or previous) NEC was in effect when the permit was pulled. Otherwise, it wasn't an option.

I'll review my codes and see which one we are required to meet and if we can go with the 2005 we will bond the grounds and neutrals at the garage panel and I'll feel better about the install.
Click to expand...
Unless Kentucky has weird rules, what should matter is, again, which code cycle was in use at the time the permit was pulled.


Um, a permit was pulled, wasn't it?
 
dhalleron said:
Thanks for the replies.

I'm going to go with Bob iwire's post. I'll review my codes and see which one we are required to meet and if we can go with the 2005 we will bond the grounds and neutrals at the garage panel and I'll feel better about the install.

I sure need more than my annual continuing education classes to keep my master electrician license in Kentucky since I do this only part time. So many details are lost in my mind over time. No matter what, I still know or care more than the electrician we replaced on this job.
Click to expand...

Sounds good. The last thing you want to do is not connect the EGCs back to the electrical source SOMEHOW. So if your under the '05 and meet the requirments then bond the neutral. If you don't connect the EGC back to the source you might hurt someone.
 
In reality, this is no different than the main panel setup in the house. The house is fed by a 3-wire service from the POCO, and the POCO grounds the neutral. This creates a ground loop from the POCO to the house, and your subpanel has the very same type of ground loop (assuming that you do ground the neutral, which you should) between the house and the garage. No one seems to question the existence of a ground loop from the POCO. At least on your subpanel, the normal current through the ground rod should be a lot lower than it is for the house's ground rod.
 
Rick we did and I THOUGHT I explained it to you IN MOST*1 Case's it is a non issue due to the voltage level and the resistivity of the soil. Not saying it can't exist just saying I know of no cases

Can you give any case's you have had where this has resulted in problems?

In my expierence I have not come across any issues resulting from the utility and main service bonds (in residential). I have seen plenty of cases in commercial installations but in these cases there is a concrete/rebar connection or metallic conduit.
 
Brian, what I stated above still stands; if there is no concern on the POCO side, there should equally be no concern on the subpanel side.

That being said, I have given the common water pipe further thought, and I do agree that a significant portion of the ground current can be from turning this water pipe into a "super neutral" or large node. Assuming there aren't any high impedance connections in the water system, all of the adjoining homes have their neutrals at relatively equal voltages with respect to the transformer. The result is that each home will attempt to have the same voltage drop across their own neutral line to the transformer, and the current will divide proportionally depending on the length of each run and other resistive factors.

Ohhh, wait a minute. While I was typing this, I just realized how I could either Prove or Disprove the notion of this ?super neutral,? but it required me to kill the mains to my house.

If this idea of a super neutral were correct, then the house with the lowest internal neutral current, would have the highest percentage of leach current taken from all of the neighbor?s houses. I was just about to say that if you had an unoccupied house on the block, then it should still show a significant current flowing through the ground and neutral--that's when I decided to cut the main on my house. Ohm?s law dictates that if there is no current flowing, the voltage drop is zero, so this house would have the lowest potential of all surrounding houses, and therefore, the surrounding houses would be sending some neutral current to the unoccupied house because their neutral, by Ohm?s law, would be at a higher voltage than both the transformer and the unoccupied house.

So, I went downstairs and recorded my current through the ground. Before I shut everything off in my house, it was sitting at 3.5 amps in the water line, and near zero through the ground rod.

After turning off the computer, TVs and other things, the current dropped to 2.5 amps in the water pipe, and the ground rod of course remained at zero.

When I shut off the main breaker, the current in the water pipe also dropped to zero.

If I was getting stray current from this ?super neutral? then I should have continued see a significant current flow through the water line. It is simple Ohm?s law. By the way, my load center is the closest to the transformer of all houses feeding from this transformer, so by all rights, my house should be taking the largest of all of the leach currents.
==========================================

So if you are still wondering why there is so much current flowing through the water pipe as opposed to the ground rod, it is because the water pipe is like a huge ground rod, with very low resistance, and it is also in closer proximity to the ground rod of the transformer. The current flowing through the water pipe (and ground rod) is in fact from a ground loop.
 
When I shut off the main breaker, the current in the water pipe also dropped to zero.
Click to expand...

I still have current on the neutral when the main breaker at our house is turned off. I have seen it as high as 7 amps, usually it's only 2 or 4. Sometimes it's 0.

I want to play with it more, I just haven't gotten a round tuit yet.

On EDIT: Our panel is only about 300 feet from a medium sized substation. It seems I live on a capacitor as the soil is really sand, nearly white beach sand.
 
Last edited:
K8MHZ said:
I still have current on the neutral when the main breaker at our house is turned off. I have seen it as high as 7 amps, usually it's only 2 or 4. Sometimes it's 0.
Click to expand...
I can accept this. As I started to say above, the existence of a super neutral is not out of the question.
K8MHZ said:
On EDIT: Our panel is only about 300 feet from a medium sized substation. It seems I live on a capacitor as the soil is really sand, nearly white beach sand.
Click to expand...
However, before automatically assuming you have a super neutral condition, I would give a little more thought and investigation into this substation and possibly how the sandy soil might contribute.

I would have to give it more thought about what the possibilities might be, but the existence of this substation could elevate the voltage of the water line, and with poor ground conditions, your ground rod would remain at a more ambient voltage. I am just thinking off the top of my head, so I don't know if this is legitimate or not.

If you have a way of isolating the various parts of your system, it might help explain this better.......

Oh, I just got sidetracked mid-thought. If the water line is getting an induced voltage from the power line, the return path would conceivably be back through your neutral, but that does not automatically prove the existence of a super neutral.

I am not saying that you (or anyone else) doesn't have a super neutral, but unfortunately in your case, you threw in one more variable that cannot be dismissed.
 
Rick you have proved nothing until you.

1. Megger all branch circuit neutrals to prove there is no contributing factors from downstream shorts on neutrals to your current on the water line.
2. Are sure your neighbors are even connected to you through the water lines or at the transformer.
 
Forgot:

if there is no concern on the POCO side, there should equally be no concern on the subpanel side.
Click to expand...

Pre 2008 NEC there was no concern as long as there was no metallic connection there was no concern with parallel paths with the neutral and EGC.
 
Status
Not open for further replies.
Top