Understanding the need for a floating neutral in a sub-panel.

Think of potential as a difference between two objects. If two surfaces are at the same voltage, there is no potential between them.

Also remember that the electrons are trying to return to the source, not the eartth.

If the neutrals were bonded to ground in a sub panel you would split the neutral current between the neutrals and grounds creating a shock hazard on parts like the enclosure that should not be energized.

Everywhere I look the information says there is zero potential between a 120v conductor and ground. Obviously they are not at the same voltage. This is a major point of misunderstanding for me.

You mean they’re trying to return to the facility that produced them, hundreds, maybe thousands of mile away? Why would they do that? Electrons are not salmon for god’s sake.

Right, and how is that different from what would happen on a service panel where the neutrals are bonded to ground, the only difference being that it has a main disconnect installed, which I can’t see makes any difference at all to electrons, since this all takes place on the load side of the disconnect, and the neutral and grounding systems are bonded together, which is different from the sub-panel a foot away…why?

The current gets onto the sub-panel, the current gets onto the main panel, why are you going to get shocked with a bonded sub-panel and not with a bonded service panel? Does it have something to do with the distance between the two panels?

You’re not going to get shocked by the subpanel if you bond the neutral and everything else is correct. It’s conditions such as those depicted in the graphic in post #4 that may present a hazard.

The neutral is bonded to ground at a service panel so that if a grounding conductor were to contact the case there is a path to trip the breaker.

Roger Frazee had some great diagrams on this site that may help you understand. Google will bring up some of them. I will try and search when not on the tablet.

Kenton, think of a sub panel as nothing more than an additional circuit derived from the service equipment - like a receptacle or some other type of electrical outlet.

The conductors all serve the same purpose - 1 or 2 ungrounded (hot) conductors, 1 grounded (neutral) conductor and one equipment grounding conductor (ground).

Then ask yourself the same questions, based on an outlet rather than thinking about a sub panel. Let me know if that helps at all.

Kenton, I haven’t found the diagrams I have been looking for yet. Will keep trying.

The books are wrong about zero potential. You have 120 volts of potential. Think about a squirrel on a wire. It is fine until it touches a grounded surface and then the electrons will start to flow. You can also think about a hole in the side of a bucket. as long as the water is above the hole there is potential and the water will flow out. Once the water is the same level as the hole the potential is gone.

The source the current is trying to return to is the transformer the voltage was derived from.


Kenton, you may want to take this on-line course. It explains what you’re confused about very well, IMO.

I took it last year for my state CE’s. It’s a recorded classroom powerpoint format.
Electrical Inspection Tune Up (online class only)
(with a focus on proper grounding and bonding of service and sub panels)

Also here’s an illustration from Mike Holt.


When the current faults to noncurrent carrying metal parts of electrical equipment it is carried back to the service on the equipment grounding conductor where it is bonded with the grounded neutral conductor, through the neutral conductor back to the transformer causing a high current draw which causes the breaker to trip or the fuse to blow.

You can have the metal raceway be the path to the grounding electrode and the breaker will not trip. Bond the grounded or neutral if you wish to call it that, to the electrode and the breaker will trip the second that it is turned on or would I be more correct to say it would trip in .033 of a second.
Ohm’s law proves this as I pointed out in post 11

Should the remote panel or sub panel if you prefer to call it that be bonded to the neutral then from that point back to the neutral at the service would be a parallel path with part of the current flowing on the neutral and part on the equipment grounding conductor. All exposed noncurrent carrying metal would be energized. Should someone complete a third path the returning current would travel on all three paths and this would not be good for the human completing this third path.

At the transformer the neutral is connected to earth. This is why it uses the past tense in its name, “GroundED” conductor as outlined in Article 200 of the NEC. The electrician connects the grounded neutral conductor to earth again at the service and the four reasons outlined in post 11 from 250.4(A)(1) gives these four reasons. Should that person be standing on earth or something in contact with earth be it a concrete floor or the resistance of water through plumbing then current will flow through that person through earth up the grounding electrode both at the service and the transformer which is really bad for that person. At .006 (6 milliamps) a GFCI device will open under the description given above but should it be on a 15 amp breaker then as much as 20.25 amps could flow on those three paths for as long as 1 hour 59 minutes and 59 seconds before the breaker tripped and still fit the standard outlined by UL. This much current being split between the three paths for that amount of time would definitely be fatal for that person. See the diagram of a single phase transformer below where the midpoint or neutral point is connected to earth.

single phase transformer.jpg

Now there is a classic DAHMIK, don’t ask me how I know :neutral:

Good morning Kenton. Give me a call again today at 1 and we can talk about this. I am still just catching up from being a new grandpa so it will take me a few days to get the pictures we discussed.

Wow! Guys, I am just overwhelmed at the response! I had to just get away from this for a few days, and after coming back to find all your posts, I am just blown away! I will go back through each of these posts. Thank you SO MUCH for taking the time to post!

the bonded neutral question:

There are two grounding electrodes (GE), one at the pole, and one at the service.

  1. Current returning from branch circuits goes to a bonded neutral in the Dist. panel (improper);
  2. Due to a fault, current gets onto the panel and some electrons flow across B to the grounding bus bar;
  3. Current travels on parallel circuits back to the service panel where the neutral and grounding bus bars are bonded together by the enclosure;
  4. Current flows from the grounding bus bar to the neutral bus bar; and
  5. From there flows back to the pole.

Current is flowing across the panel at both A dnd B. Why can a person get shocked by touching the Dist. panel and not the Service panel?

Joseph Whitt wrote:
When the current faults to noncurrent carrying metal parts of electrical equipment it is carried back to the service on the equipment grounding conductor where it is bonded with the grounded neutral conductor, through the neutral conductor back to the transformer causing a high current draw which causes the breaker to trip or the fuse to blow.
Isn’t that what I have drawn here? What creates the high current draw that makes the breaker (Main Disconnect?) trip? Isn’t there pretty much the same amount of current flowing from the Service panel to the pole © as was flowing into the Dist. panel from the branch circuits (D)?

What one does not want is a parallel path for fault current to flow from the remote panel to the service disconnect. We want all that faulted current to flow on the EGC so the impedance will always be low which will cause the overcurrent device to operate quicker. The grounding electrode plays no role in this circuit at all.

A general rule of thumb for circuit breakers is they will carry 135% of their rating for up to two hours, 300% for two minutes, and 600% for two cycles or .033 seconds.

Should the remote panel be bonded to the neutral then the neutral and the bonded EGC and everything in the system that is grounded by the EGC will be carrying current on parallel paths. This will be the normal current load on the branch circuits and does not require a fault in order for the EGC to be carrying this normal current. Now someone comes in contact with exposed metal and completes the third parallel path through earth then that person will be in dire need of medical attention.

If the equipment grounding conductors only gets bonded in the service disconnect then all fault current will go back to the transformer via the neutral conductor and there is no parallel paths for current to flow.

IMO that’s not entirely true. The person in contact with the earth and in parallel to the other paths will only be exposed to at best a few volts. The voltage drop across the person/earth will be very low because the earth is a poor conductor and almost all of the current will flow on the other two low impedance paths. That’s why you don’t get shocked when you touch a metal meter enclosure, which is bonded directly to the neutral, while standing on the ground.

That third parallel path will see “current” based on the amount of resistance of that parallel path.

We must remember that it is not the amount of voltage that does damage to the human but the amount of current.
With the meter base just as with the service equipment the neutral is in direct contact with the metal which causes a very low resistance therefore not much current flows when we touch the exposed metal but when we start bonding remote panels causing the EGC to be in parallel with the neutral then the resistance of the third path can cause much damage to someone touching exposed metal.

I am sure that you have heard of people getting a tingle from touching their range and kitchen sink at the same time or some sort of equipment. This is a completion of a current path that allows that person to feel that current flow.

In my career I have been bitten by poorly installed meter bases especially around farm equipment where aluminum SE was used and the conductors was corroded in the terminals.
Now consider a bonded remote panel and the neutral being lost from the feeder that is supplying that panel, yes in my career I have run into this and let that remote panel be in an unfinished basement and the pain felt upon contact is not very pleasant at all.

In my career I have even seen the EGC used as the return path on things such as motion detectors but the NEC addressed this issue by requiring a neutral to be installed in areas where these items are to be used. Just one and not much of a problem but let them start adding up and the problem can be really bad.

What we try to do as electricians is to comply with 250.6(A)
**250.6 Objectionable Current. **
(A) Arrangement to Prevent Objectionable Current. The grounding of electrical systems, circuit conductors, surge arresters, surge-protective devices, and conductive normally non–current-carrying metal parts of equipment shall be installed and arranged in a manner that will prevent objectionable current.

Early on in my career I stopped using a metal raceway between the meter pan and the power outlet of temp poles simply due to the shock hazard involved with people touching them. A temp pole is treated badly and some even gets destroyed during their use. Ever been shocked by a temp pole?
By removing the metal raceway and something happening to the neutral everything becomes a series 240 volt circuit instead of the metal raceway becoming the neutral.

I agree with you that in a perfect installation one can walk up barefooted and touch their meter base without concern but let something go wrong such as the neutral being lost at the transformer and touching that meter base might not be a good idea.

I agree that having a bonded remote panel and touching something while standing on carpet and not much will happen but be on concrete barefooted and it won’t be so pleasant. I hope that I am not implying that someone will be hurt each and every time but what I am implying is that the potential is there when we allow objectionable current to flow in places it is not supposed to.

Been there done that well explained … Thanks … Roy

Uhhhhh, yes, electricity is dangerous, especially when components of a system are damaged or deteriorated. This holds true of any electrical system, not just a bonded neutral point.

I thought the whole point of bonding the neutral is to keep the grounding conductor from becoming a parallel path, to keep current OFF the grounding conductor. If we want current to flow on the grounding conductor, all we have to do is bond the neutral and ground at the distribution panel.


I am in hopes that you will forgive my statement here but alas I need to respond. In your question various terms are expressed that could (and probably will) add to confusion with regards to the important differences in Grounding and Bonding. The lack of the term Bonding(Bonded) in your question and subsequent responses from members leave me slightly concerned that the demarcation of the two concepts is fuzzy.

Here is what I propose, see if you can get Nick and Ben to consider a recorded “Talk” or live nachi show and I will explain all the important details of Grounding and Bonding for the members…for free of course. While I am extremely busy traveling around the country I will always make time for my NACHI Family.

Also there are very critical changes in the latest editions of the NEC that over time home inspectors will encounter that throws current bonding and grounding logic upside down with regards to remote distribution panels (sub-panels as you all call them). I have and will continue to support NACHI and NEMA supports your education as well…our members know that you interact with their products and I am here to ensure that interaction is beneficial for everyone.