Dangers of bonding grounds and neutrals

[ThunderWizard]

I have always been confused about a specific part of this topic. Hopefully you guys can help me understand this better.

Scenario 1:

Service with disconnect (Main panel) feeds to interior panel (sub panel). Grounds and neutrals must be bonded at service and separated at sub panel.

Scenario 2:

Service without disconnect, feeds to interior panel with first point of disconnect (Now making the interior panel the main panel?). Grounds and neutrals now become bonded at the interior panel.

My question is, why does scenario 2 not pose the same dangers as scenario 1 would if the interior panel also had grounds and neutrals bonded in scenario 1?
Or would they have the same risks but that is considered the safest for that specific situation?

I guess i'm just confused on why the first point of disconnect is where we bond grounds and neutrals regardless of where that first point of disconnect is, rather than always bonding it only at the service regardless of whether there is a disconnect there or not.

 

[infinity]

In scenario 2 the neutral is bonded via the main bonding jumper because that is where the service disconnect is. All metal parts upstream of the service disconnect are bonded to the neutral. in scenario 1 the service disconnect is at the meter so the neutral is bonded via the MBJ there. Both are code compliant and one is no safer than the other.

 

[CoolWill]

We only want to do it at one point so we don't have neutral current taking a parallel path through ground wires and metal conduits etc.

 

[infinity]

We only want to do it at one point so we don't have neutral current taking a parallel path through ground wires and metal conduits etc.

That's correct with one general exception. Neutral current on a metal conduit that contains service conductors is code complaint.

 

[kwired]

I guess i'm just confused on why the first point of disconnect is where we bond grounds and neutrals regardless of where that first point of disconnect is, rather than always bonding it only at the service regardless of whether there is a disconnect there or not.

That first point of disconnect is the "service disconnect" regardless of whether it is inside or outside. Other rules govern how far inside it possibly can be.

Then more recently NEC required most dwellings to have an "emergency disconnect" which may or may not also be the service disconnect, but wherever the service disconnect is located is where bonding the grounded conductor stops and anything beyond that is bonded by the EGC. Reason being we don't want normal operating current of the grounded conductor to take alternate paths through items not intended to be carrying current in normal operating conditions. Having such currents can lead to voltages being present on those items and increases risk of shock.

 

[ThunderWizard]

I do understand that we need the bond to be able to clear a ground fault.

But in scenario 2, aren't we creating that parallel path on the neutral and ground in the feeders?

Wouldnt it make more sense even in that situation, to only bond at the service equipment regardless of whether there is a disconnect there or not?

 

[infinity]

But in scenario 2, aren't we creating that parallel path on the neutral and ground in the feeders?

Scenario 2:

Service without disconnect, feeds to interior panel with first point of disconnect (Now making the interior panel the main panel?). Grounds and neutrals now become bonded at the interior panel.

In scenario 2 how is there a parallel path? The conductors to the panel with the service disconnect are service conductors not a feeder.

 

[ThunderWizard]

That first point of disconnect is the "service disconnect" regardless of whether it is inside or outside. Other rules govern how far inside it possibly can be.

Then more recently NEC required most dwellings to have an "emergency disconnect" which may or may not also be the service disconnect, but wherever the service disconnect is located is where bonding the grounded conductor stops and anything beyond that is bonded by the EGC. Reason being we don't want normal operating current of the grounded conductor to take alternate paths through items not intended to be carrying current in normal operating conditions. Having such currents can lead to voltages being present on those items and increases risk of shock.

Ok thanks that actually makes a lot of sense. I think I was missing the point that it was still the "service disconnect" regardless of where it's located.
So would the feeders from the service equipment to the interior panel in scenario 2 still be considered "service conductors" and not "feeders"?

 

[kwired]

I do understand that we need the bond to be able to clear a ground fault.

But in scenario 2, aren't we creating that parallel path on the neutral and ground in the feeders?

Wouldnt it make more sense even in that situation, to only bond at the service equipment regardless of whether there is a disconnect there or not?

There is no feeder your service conductors just happen to extend to the inside of the building.

There is still some issues at times with voltage drop on the service grounded conductor causing so called "stray voltage" on all the items you ultimately have bonded to it at the service equipment and beyond. But going to be kind of hard to get away from that with most utilities commonly using a MGN type distribution system plus the fact that NEC doesn't apply to what is ahead of the "service point". In general for a single phase 120/240 service you are getting three conductors from the utility one of them is grounded but likely will carry current in a majority of applications.

 

[ThunderWizard]

In scenario 2 how is there a parallel path? The conductors to the panel with the service disconnect are service conductors not a feeder.

Oh thanks that was part of my confusion then. I was seeing those as feeders. I was thinking of when i've seen 4 wire SER's from the service equipment to the interior panel in this situation and it sounds like the 4 wire would be redundant since they're still service conductors and meant to be bonded at that point anyway

 

[infinity]

Ok thanks that actually makes a lot of sense. I think I was missing the point that it was still the "service disconnect" regardless of where it's located.
So would the feeders from the service equipment to the interior panel in scenario 2 still be considered "service conductors" and not "feeders"?

The terminology helps to make things as clear as possible. I know many of us have been taught about the first means of disconnect but now with EM disconnects that term is pretty obsolete.

 

[ThunderWizard]

There is no feeder your service conductors just happen to extend to the inside of the building.

There is still some issues at times with voltage drop on the service grounded conductor causing so called "stray voltage" on all the items you ultimately have bonded to it at the service equipment and beyond. But going to be kind of hard to get away from that with most utilities commonly using a MGN type distribution system plus the fact that NEC doesn't apply to what is ahead of the "service point". In general for a single phase 120/240 service you are getting three conductors from the utility one of them is grounded but likely will carry current in a majority of applications.

Thanks for that explanation. That was definitely part of my confusion how bonding at one point eliminated the risk while the other point created the risk. What i'm understanding is that risk is always there its just minimized by bonding it only at the end of the service conductors.

 

[kwired]

Thanks for that explanation. That was definitely part of my confusion how bonding at one point eliminated the risk while the other point created the risk. What i'm understanding is that risk is always there its just minimized by bonding it only at the end of the service conductors.

And when you have a transformer (separately derived system) you start all over with that new system in similar fashion. You bond either at the source itself or at first disconnecting means but keep grounded and equipment grounding conductors separated beyond that point - for same reasons to keep grounded conductor current from taking unintended paths.

 

[wwhitney]

In scenario 2 the neutral is bonded via the main bonding jumper because that is where the service disconnect is.

A question occurs to me: is there any safety benefit to the NEC's requirement, in the case of a service, that the MBJ be at service disconnect, and that we aren't allowed to put it anywhere else?

GECs can be bonded to the grounded conductor anywhere between the service point and the service disconnect. So would there be a downside if the MBJ could similarly be located anywhere between the service point and the service disconnect? The bonding rules would be that on the service side of the MBJ, you bond to the grounded service conductor; on the other side of the MBJ, you bond to an EGC.

Cheers, Wayne

 

[infinity]

So would there be a downside if the MBJ could similarly be located anywhere between the service point and the service disconnect?

I don't see a downside. The meter neutral is typically factory bonded to the enclosure so if you called that the MBJ and just installed an EGC from that point on electrically there is little difference.

 

[kwired]

A question occurs to me: is there any safety benefit to the NEC's requirement, in the case of a service, that the MBJ be at service disconnect, and that we aren't allowed to put it anywhere else?

GECs can be bonded to the grounded conductor anywhere between the service point and the service disconnect. So would there be a downside if the MBJ could similarly be located anywhere between the service point and the service disconnect? The bonding rules would be that on the service side of the MBJ, you bond to the grounded service conductor; on the other side of the MBJ, you bond to an EGC.

Cheers, Wayne

In systems with MGN distribution I don't think the grounding electrode at the service really matters much and you will not see much difference in regards to which location you connect it to, or in many cases if you even connect one as the MGN network is what is giving you a pretty stable earth reference and your local electrode at your service is just another drop in the bucket of electrodes in the system.

That said I haven't really ran into an inspector yet that won't tell you you are going to either electrocute someone or burn the place down if you don't have a grounding electrode, even when the utility has a huge network of them including the one on a pole just a few feet away from where you have your service disconnect.

 

[wwhitney]

In systems with MGN distribution I don't think the grounding electrode at the service really matters much

Did you mean to quote somebody else? Grounding electrodes are incidental to my post about the MBJ. The only relationship is that you can only have an MBJ on a grounded service, so there does have to be a grounding electrode somewhere connected to the grounded service conductor.

Cheers, Wayne

 

[ThunderWizard]

Alright its making more sense but I still don't think I fully grasp it.

So even for scenario 2, wouldn't it be better to have the MBJ at the meter base and establish our equipment ground there instead of bringing our bond point into the main panel?

In my mind that would create less opportunity for unwanted current to flow through EGC's from the main panel.

 

[ActionDave]

Alright its making more sense but I still don't think I fully grasp it.

So even for scenario 2, wouldn't it be better to have the MBJ at the meter base and establish our equipment ground there instead of bringing our bond point into the main panel?

In my mind that would create less opportunity for unwanted current to flow through EGC's from the main panel.

In scenario 2 all the neutral current is flowing on the neutral wires and all the fault current will flow on the EGC conductors back to the MBJ. From the MBJ back to the meter the neutral and EGC are the same conductor. The only difference between scenario 1 and scenario 2 is the distance the Neutral and EGC are the same conductor.

 

[CoolWill]

Alright its making more sense but I still don't think I fully grasp it.

So even for scenario 2, wouldn't it be better to have the MBJ at the meter base and establish our equipment ground there instead of bringing our bond point into the main panel?

In my mind that would create less opportunity for unwanted current to flow through EGC's from the main panel.

It would be better, yes, but the electric utilization industry isn't always in sync with the electric distribution industry. The utilities insist on bonding their neutrals to their meter enclosures. And the NEC insists that we bond the neutral at the service disconnect. And neither is going to give.

If we could just start our EGC from the meter neutral, then we wouldn't need the MBJ and all of our neutrals would be insulated from ground. BUT! There is no NEC requirement for a meter at all. So, Maybe we could insulate the neutral at the meter and use a MBJ at the disconnect and take an EGC back to any meter that might be installed? The utility is NOT going to rely on us to ground their equipment. So we just live with it.