Cause of Burned Neutrals

Status
Not open for further replies.
Looking closely at the close-up image, it seems that the heat source is the bus, and not the conductors. They are too evenly damaged for it to be coincidence that every neutral was equally overloaded.

I agree with those who believe the main lug is more likely the source. Perhaps the metal that connects the bus to the main lug is the culprit. It is not visible in the images, but it's gotta be there.

Perhaps a voltage reading between the bus and the neutral conductor while under load is in order. Do we know if there could be a triplen-harmonics issue happening?
 
Last edited:
LarryFine said:
Looking closely at the close-up image, it seems that the heat source is the bus, and not the conductors. They are too evenly damaged for it to be coincidence that every neutral was equally overloaded.

I agree with those who believe the main lug is more likely the source. Perhaps the metal that connects the bus to the main lug is the culprit. It is not visible in the images, but it's gotta be there.
I disagree. Note the condition of the insulation on the eighth branch neutral down from the top... its insulation appears to have disintegrated and fallen off near the connection (either that or it was an extra long stripping of insulation :D )... and the conductors both above and below seem to gradually get less charred. This is a sign of a loose connection that has done quite a bit of arcing. Being the bus bar is the largeest of the conductors about the connection, the heating effect has spread vertically. Note how the charring has not made it to the first bend in the top (close up pic) and bottom-one-in-the-group (away pic) branch neutrals.
 
I disagree back. If you look at the less-close picture, the 12th screw seems to be the most blackened. However, I looked even closer in a graphic program. The most oxidation seems to be concentrated around the bracket that holds the bus (and conducts the total neutral current).

Look again at the two studs to the left of the bus. The larger is holding the neutral-conductor lug, and the smaller is unused, but appears to be the same piece of metal that is the bracket itself. It's rather dark. Remember, the bracket is what conducts the neutral current.

If you look at a piece of bus, there are two rows of holes: the upper row is where the conductors land, and the lower row is where the self-tapping screws that hold the bus to the bracket are screwed. I believe this bracket is more discolored than the bus itself.

Additionally, the bottom-most wires are at least as charred as the upper ones, if not more so. Look how far around the first bend they are blackened. These wires are too evenly burned for one of them to be the source, in my opinion.

If indeed triplen harmonics are contributing to the heat, it would be concentrated in the neutral bus, and not the individual circuits, unless each and every wire is subject to such currents. Remember that third-order harmonics add in-phase, and this damage looks like an overloaded bus.

Plus, this much damage spread out this evenly is indicative of long-term overload, and not a short-term event. It takes a long time for this much oxidation to occur, along with a steady supply of heat. A localized heat source capable of this much energy would do a great deal of damage in a small area.

My conclusion is that the metal parts, the bus itself along with the bracket, have been subject to current beyond their design, and for a long time. I may be incorrect about the harmonics, but I believe I am correct that the bus and its bracket are the source of the heating.

augie47 said:
I would be curious as to the connection between the neutral bar and the main neutral lug. The bar containing the #12s obviously has been hot...the heat may have generated from the bar connection to the neutral also.

supergeek said:
I was thinking of that too. There are just 2 screws holding the long neutral bar to an 'L' shaped piece of copper where the main lug bolts on. I wondered if that connection was bad and if that could be the reason all the #12's burned up. When it gets changed out, I'm going to take a good look at that connection and see how tight everything is.

I believe these posts are heading in the right direction. The OP mentioned that the building is under renovation, and the measured neutral load is now under 20a. This is obviously not what was going on when the damage occured. We need to know what the loads were when the building was in full use.

A large non-linear loading, such as computers, fluorescent lighting, etc., is more likely to have created the third-order harmonics current than anything else, except for plain old resistance in the bus's mounting bracket.

If looseness of the individual conductor screws was the problem, we would expect to see greater conductor damage for the amount of insulation damage. Was there ever complaint of voltage level, power quality, or EMI interference issues? Arcing would likely have melted the individual wires.

My brain is screaming at me to end this now, so I'll see your responses tomorrow. Good night.
 
It seems like the main lug could be the problem

It seems like the main lug could be the problem

Hi everyone, thanks so much for all the information. I really appreciate the ideas and suggestions.
Today we changed out the bad neutral assembly. We discovered that the neutral had apparently been shortened because it was routed diagonally across the back of the panel (unless it was originally cut too short accidentally.) The other conductors feeding the panel were routed in a normal manner, parallel to the top of the enclosure.
This makes me think there was a problem with the main neutral connection in the past. Maybe it was burned up and someone shortened it and reconnected it. So if only the main neutral was repaired, I guess it's possible that the bus had heat damage and continued to deteriorate.
When the neutral bar was removed, all the connections seemed tight. If they were really damaged, though, they might have just been siezed up.
Still not sure of the original cause of the problem, though. With the building having new tenants, there's no way to tell what was originally fed from the panel. The lighting is primarily 277 volt, and when the panel was shut down it seemed like just random receptacle circuits were affected. There's nothing in the old panel schedule that looks suspicious, either.
The theory we came up with was that maybe the original neutral lug didn't get fully tightened. Then it could have overheated and caused the damage. So the neutral was trimmed back and relanded but nothing was done to the branch circuit neutrals.
We checked the phasing on the branch circuits; everything looks fine. Most conduits have one set of A B C and a neutral and a few have 2 sets. All the blacks land on A, reds on B, etc. Now of course things could have gotten mis-wired in the various j-boxes, that would take a lot more investigating!
So anyway, does this thoery sound feasible? Thanks for all the help.
 
It is difficult to trouble shoot just through pictures , but here's my 2 cents. Have you checked to verify that the neutral feeding the bar is actually a neutral and not an ungrounded conductor. I ran across this situation once on a job where the lights kept burning out when the rheostat brought them up to 75-80% power. When terminating the wire someone crossed C-Phase and the Neutral condictors making the Neutral bar hot and C-phase was the Neutral. Like I said just my 2 cents.
 
ecirplr said:
It is difficult to trouble shoot just through pictures , but here's my 2 cents. Have you checked to verify that the neutral feeding the bar is actually a neutral and not an ungrounded conductor. I ran across this situation once on a job where the lights kept burning out when the rheostat brought them up to 75-80% power. When terminating the wire someone crossed C-Phase and the Neutral condictors making the Neutral bar hot and C-phase was the Neutral. Like I said just my 2 cents.


How is that possible without shorting out the panel?
 
SuperGeek
My 2 cents if you do not mind.

I believe the original problem was the supply neutral.
1. it is in very good condition relative to the branch circuit neutrals.
2. The antioxident inhibitor in the threads of the neutral lug is very unusual. I believe it was put there so the person who made the repair would be able to thread the lug more easily as it may have been distorted during the original incident.

What is puzzling is that it would seem they/he forgot to repair the branch circuit neutrals.
 
Pierre C Belarge said:
How is that possible without shorting out the panel?
The neutral bus is isolated from ground in subpanels. Therefore, landing an ungrounded feeder conductor on the neutral bus will not be apparent until the panel is energized. And even at that time, if voltages are not confirmed (many electricians consider theirselves to be infallible in this regard, so this precautionary measure is often not performed), such a crossing of wires is not apparent until an unusual and/or unexpected event occurs, perhaps several. Actually had this happen on a job... and I must add I was not involved with the pulling/landing of the feeders or energizing the panel :grin:
 
LarryFine said:
I disagree back.
Your prerogative... and it seems now we'll never know which of us is [more] correct.

I'd just like to point out that judging which conductor got the hottest by equating it with the insulation turning the blackest is misleading, especially when considering nylon-jacketed, PVC-coated wire such as THHN. My knowledge in this regard comes from having worked at a plastics extrusion factory in my adolescent years and includes experience with many types of PVC and nylon.

First off, PVC in general has a much lower "melting" point than nylon. I put emphasis on "melting" because PVC doesn't really melt to a liquid or liquid-like state. Rather it becomes more formable at moderate temperatures, relatively speaking. On the other hand, nylons used for coating have a near syrup-like consistency when heated to above moderate temperatures (say roughly 100-150?F higher than extrudable PVC at 300-350?F). At higher than normal processing temperatures PVC will char whereas nylon just becomes more runny.

Getting back to overheated wire, as the temperature increase to above its rating, PVC or rather the plasticizers contained therein "blacken". As the temperature increases more, the nylon melts and runs off while the PVC chars and its plasticizers evaporate. At sustained high temperatures, the resulting PVC turns to dull black and gradually becomes a dark gray. After prolonged heating, what's left of the PVC disintegrates.

(Though a generalized summation, I'm open to correction on any or all of the above :D )
 
Smart $ said:
The neutral bus is isolated from ground in subpanels. Therefore, landing an ungrounded feeder conductor on the neutral bus will not be apparent until the panel is energized. And even at that time, if voltages are not confirmed (many electricians consider theirselves to be infallible in this regard, so this precautionary measure is often not performed), such a crossing of wires is not apparent until an unusual and/or unexpected event occurs, perhaps several. Actually had this happen on a job... and I must add I was not involved with the pulling/landing of the feeders or energizing the panel :grin:


Smart$
What about the phase to phase connections?

I have witnessed a 4000amp, 480v service that had just what you say occur. The "C" phase and the "Neutral" were transposed. When they enegized the main service disconnect, the bottom line was $285,000.00 of repair...
I did get many pictures of that incident about 1/2 after it occurred. It was quite intense.
 
Pierre C Belarge said:
Smart$
What about the phase to phase connections?
I don't understand what it is you are asking here...

I have witnessed a 4000amp, 480v service that had just what you say occur. The "C" phase and the "Neutral" were transposed. When they enegized the main service disconnect, the bottom line was $285,000.00 of repair... ...It was quite intense.
I bet it was.

Connecting service equipment with an ungrounded and grounded conductor crossed is quite different than a subpanel. The neutral bus is in most cases required to be bonded to ground in service equipment... whereas in a subpanel it is the exact opposite: the neutral bus is required to be isolated from ground. In the case of a subpanel, a neutral-ground fault must be present upon energizing to have "an incident".

I should note here that proper panel energizing procedure dictates that all branch circuits be open and proper connection, voltages, etc. be verified before energizing any branch circuit. The issue we are discussing is predominant among the reasons for doing so!

Regarding the incident you witnessed, exactly what do you mean when you say, "When they enegized the main service disconnect..."? I'm of the impression the fireworks started when the poco energized the service, prior to the main disconnect switch/device being closed.
 
short neutral

short neutral

Pierre C Belarge said:
SuperGeek
My 2 cents if you do not mind.

I believe the original problem was the supply neutral.
1. it is in very good condition relative to the branch circuit neutrals.
2. The antioxident inhibitor in the threads of the neutral lug is very unusual. I believe it was put there so the person who made the repair would be able to thread the lug more easily as it may have been distorted during the original incident.

What is puzzling is that it would seem they/he forgot to repair the branch circuit neutrals.
is the evidence I was missing. This was definitely a repair. The original neutral was missed in the install for tightening. the repair saw the copper wasnt burned on the branches so he let a sleeping dog lie. This whole neutral buss was probably replaced durring the shutdown.
 
Smart $ said:
Regarding the incident you witnessed, exactly what do you mean when you say, "When they enegized the main service disconnect..."? I'm of the impression the fireworks started when the poco energized the service, prior to the main disconnect switch/device being closed.




The utility meggered the line side of the service disconnect.
They then powered up their side.

The EC then closed his main (4000A), that is when the fun started.
BTW: the 4000a main service fuse did not open, the utility primary did...
That took more time (cycles) hence all of the damage to equipment inside of the building. All electronic equipment in the building was destroyed, plus some enclosures and conductors.
 
must have been

must have been

Pierre C Belarge said:
The utility meggered the line side of the service disconnect.
They then powered up their side.

The EC then closed his main (4000A), that is when the fun started.
BTW: the 4000a main service fuse did not open, the utility primary did...
That took more time (cycles) hence all of the damage to equipment inside of the building. All electronic equipment in the building was destroyed, plus some enclosures and conductors.
Some explosion. You never forget the sound of the air ionizing and the plasma cloud forming just before the arc blast. I have witnessed this a number or times (not due to fault of mine of course). One where I made the local papers was a swan trying to fly between a horizontal to vertical primary span on Main Street . Instant dinner and a 12 foot diameter fireball until the primary cutouts blew. The swan actually got up and flew back after a half hour dog and pony show involving the local police animal rescue groups and a dozen people trying to catch the swan by hand. It was hilarious but not for the swan
 
Pierre C Belarge said:
The utility meggered the line side of the service disconnect.
They then powered up their side.
I'm suprised meggering did not catch the cross-wiring. Checking C phase to ground should have indicated some possibility of a fault. Perhaps the poco only meggered phase conductor to phase conductor, which would not have revealed the fault. There's a chance meggering C-phase to neutral would have indicated some possibility of a fault, but this relies on low resistance parallel grounding in the area, such as through a metal underground water piping system and other consumer's services.

The EC then closed his main (4000A), that is when the fun started.
This is where the EC messed up. They should have verified their connections by way of a voltage check.
 
Last edited:
Burnt Neutrals

Burnt Neutrals

I could be a 100 reasons, but my bet is Lightning, because if was no real loose connection and system was properly grounded.
 
Electric38

Electric38

Just a guess...
Since the nuetral bar has only over heated at the top, I would check the feeder neutral tightness. If this connection were loose, it would tend to create heat mostly at the point of highest resistance (the loose connection). This could explain why the wires coming into this area appear to have endured a similar high temperature. Even if this connection was found to be tight, it could be an indication that the problem was previousely corrected. It could be helpful at this point to check the wires for carbon tracking. Cleaning them may help lessen the chance of an insulation breakdown.

Hope this helps.
 
Status
Not open for further replies.
Top