Loose neutral damage

[ptonsparky]

My brother called me about an issue his neighbor was having. Classic neutral issue. Bright/dim lights. Told him to pass on the suggestion that neighbor should not delay in getting it fixed.

I've seen a lot of items damaged over the years but never have seen an actual house fire because of it. Anyone see it?

 

[Dennis Alwon]

I am trying to see how a loose neutral can actually cause a fire. Even if the neutral is on the load side and not utility side I don't see it with everything in jb's etc.

Most cases putting 240v to a 120v piece of equipment just burns it out

 

[infinity]

I am trying to see how a loose neutral can actually cause a fire. Even if the neutral is on the load side and not utility side I don't see it with everything in jb's etc.

Most cases putting 240v to a 120v piece of equipment just burns it out

Could putting 150 or 160 volts on a 120 volt piece of equipment start a fire?

 

[paulengr]

Electrical fires are usually the result of either arcing (forcing electricity through air which is an insulator makes lots of heat), usually involving another issue such as water or oil or gas, or some kind of short. In both cases it is almost never an isolated fault. It is in conjunction with something else going wrong.

Plus you are assuming all wiring is enclosed and done exactly according to Code. Apparently you haven’t seen what creative homeowners and hacky electricians can do. Your first mistake is assuming everything is in boxes. Ever heard of Western Union splices? Those are a step up. If you can just twist them wires together so they touch that should work. Did you know duck tape is a substitute for electrical tape? Or that uninsulated split bolts make great terminal blocks? How many thousands of photos of critters, bubbas, and outright dumb things do you need to see to get the idea that if it can be done, it has been? No need for lab testing…the Petri dish of human stupidity does a great job on its own.

With voltages above 300 V most of the time once an arc starts you get a sustaining arcing condition. The air heats up enough that it becomes conductive and arcs become self sustaining until a breaker or fuse opens or it eats up enough of the conductors to open up a huge air gap. The longest sustained arc recorded in a lab at 240 V so far is 0.08 seconds. This was at 20 kA and 1/4” bus gap with bus bars in a box-barrier configuration which tends to sustain arcs. That doesn’t sound like much but it was enough to kill an electrician in Georgia in 2009 from burn injuries.

UL generally requires equipment to be tested to 200% of rated voltage plus 1,000 V in a hipot test. The equipment doesn’t have to ever function again , just not short out. So not an issue on 150 V. But there are exceptions here too. About 15 years ago UL started requiring a fuse on surge strips. Surge arresters tend to short out once they are used up. They had a nasty habit of starting carpet fires before UL required short circuit protection. So an extended surge or a series of them that pushes a cheap old $10 surge strip can cause a fire and many of them have. Yes the breaker SHOULD trip but many of them didn’t. You can Google this and see for yourself. Also lots of laptop and similar load shorts causing fires.

Nothing magic about a neutral except it has close to zero voltage referenced to ground. It’s still a phase conductor and has current carrying capabilities. It can still arc and get hot from loose or shorted faults.

 

[Dennis Alwon]

Could putting 150 or 160 volts on a 120 volt piece of equipment start a fire?

Let's say I have never seen a fire from overvoltage of any kind except lightning strikes but that didn't necessarily have anything to do with the wiring.

Sure it is possible but I bet it's rare.

 

[ptonsparky]

There probably is not a member here that has not witnessed creative wiring by whomever.

One of our customers lost an entire building due to their wiring. (We took pictures after the 1st fire call.) Hi winds reignited a not quite out fire that destroyed what was left.

Still not caused by lost neutral over voltage.

 

[Knuckle Dragger]

My brother called me about an issue his neighbor was having. Classic neutral issue. Bright/dim lights. Told him to pass on the suggestion that neighbor should not delay in getting it fixed.

I've seen a lot of items damaged over the years but never have seen an actual house fire because of it. Anyone see it?

I've seen plugin surge protectors, contractol circuit boards burn up.

 

[FionaZuppa]

N's are CCC's, so a loose N can create hi-ohm condition, get some soot and heat at the junction.
Most things have an inrush current (magnitude makes no diff, start-up is 'always' more than run), so a bouncing N means the inrush has more cycles, happens every time the N conducts at the weak junction. I have had to fix many switches and outlets where sub simply rushed through and did not tighten screws correctly, some of them you can here a slight buzzzzzzzzz because of the poor contact.

Junctions are usually in some type of box (should be, but in MX many times not). Fire causing I not sure, surely possible in rare circumstances.

A loose N on the service side, panel, means perhaps some of the ckt's N's CCC (plural) runs in earth over to neighbor's panel, which means additional ohms in that path, hence a bright/dim light. GFCI's don't even see this issue, ckt amps on each CCC are equal, etc.

I thought somewhere here on MHF there was a similar thread, but talking about copper vs older AL wire. Bottom line, loose CCC's is bad.

 

[robertd]

The problem with a loose or lost neutral is you no longer have a 120V, 120V system. You have a 240V system and the neutral "floats" around, so some loads on one phase see less than 120V but loads on the other see more than 120V. That's the problem. Even if the entire building meets current code you can still have devices, appliances or other loads start on fire. Put more then 120V on a 120V item made in Chine that came from Amazon or Ebay and which couldn't pass a UL standard, and you have a fire.

 

[Gene B]

Cable TV wiring could cause a fire, but only if not bonded. If bonded, the outside coax could melt but the inside should be ok.

 

[goldstar]

My brother called me about an issue his neighbor was having. Classic neutral issue. Bright/dim lights. Told him to pass on the suggestion that neighbor should not delay in getting it fixed.

I've seen a lot of items damaged over the years but never have seen an actual house fire because of it. Anyone see it?

Usually, when you lose the neutral ahead of the power distribution panel, everything down stream that's electronic or has a digital display is toast as Dennis pointed out. When you have a loose neutral on a branch circuit (as paulengr pointed out and depending on where it is in the circuit) the wire could heat up to the point where the insulation melts off and the wire glows. I'm sure we've all seen that in our experiences and that could cause a fire IMHO.

 

[FionaZuppa]

The problem with a loose or lost neutral is you no longer have a 120V, 120V system. You have a 240V system and the neutral "floats" around, so some loads on one phase see less than 120V but loads on the other see more than 120V. That's the problem. Even if the entire building meets current code you can still have devices, appliances or other loads start on fire. Put more then 120V on a 120V item made in Chine that came from Amazon or Ebay and which couldn't pass a UL standard, and you have a fire.

Huh?
240v CT, if you lose the N that does not mean 240v is on the device. One side of the device is on a CCC (N) that does not make it back to the CT.

MWBC's might have an issue, like one device no worky, but both 'on' the 240v is across two 120v devices works a-ok.

 

[jaggedben]

Huh?
240v CT, if you lose the N that does not mean 240v is on the device. One side of the device is on a CCC (N) that does not make it back to the CT.

MWBC's might have an issue, like one device no worky, but both 'on' the 240v is across two 120v devices works a-ok.

For any circuit closed circuit that shares the lost neutral with loads on the other leg, the voltages seen by each load will be inversely proportional to the impendence of the loads on their leg, since the loads on different legs will be in series at 240V. So it's certainly possible some loads will see close to 240V. This is pretty basic ohms law.

If the neutral is lost at the service or feeder level, then 2-wire circuits on the panel served by that neutral are not immune.

 

[FionaZuppa]

For any circuit closed circuit that shares the lost neutral with loads on the other leg, the voltages seen by each load will be inversely proportional to the impendence of the loads on their leg, since the loads on different legs will be in series at 240V. So it's certainly possible some loads will see close to 240V. This is pretty basic ohms law.

If the neutral is lost at the service or feeder level, then 2-wire circuits on the panel served by that neutral are not immune.

I hear ya, but what BC's are using shared N. MWBC is one, any others?

And in a MWBC, you need two items to be "on", which forms a virtual CT on the shared N between them, so across each device is still just 120v.

If you lose the xfmr CT at the panel, how does a non-MWBC two-wire BC see 240v ?

 

[Hv&Lv]

I hear ya, but what BC's are using shared N. MWBC is one, any others?

And in a MWBC, you need two items to be "on", which forms a virtual CT on the shared N between them, so across each device is still just 120v.

If you lose the xfmr CT at the panel, how does a non-MWBC two-wire BC see 240v ?

Not sure where your going here. You lose a N at say the point of delivery, it is entirely possible to see 240V on one leg if there is nothing running on the other leg.
Go over Jaggedbens post again.

It happens more than you can imagine. Tree falls, tears neutral in half, the hot legs still together, can’t run equipment long like that.

No MWBC involved

 

[ptonsparky]

Not sure where your going here. You lose a N at say the point of delivery, it is entirely possible to see 240V on one leg if there is nothing running on the other leg.
Go over Jaggedbens post again.

It happens more than you can imagine. Tree falls, tears neutral in half, the hot legs still together, can’t run equipment long like that.

No MWBC involved

Well, it is MW, just not a BC.

 

[mopowr steve]

My brother called me about an issue his neighbor was having. Classic neutral issue. Bright/dim lights. Told him to pass on the suggestion that neighbor should not delay in getting it fixed.

I've seen a lot of items damaged over the years but never have seen an actual house fire because of it. Anyone see it?

Yes

 

[mopowr steve]

Happens probably more often than one thinks,

Actually had it happen to one of our jobs. Then within a couple years of that I witnessed 2 other incidents. The common problem with all these incidents was the cable vision/internet coax shielding taking on the current returning to transformer. Phone lines with shielding also. Especially homes not connected to a community metallic water line EGC.
So if anyone here ever runs across some telling them that the coax on the side of the house looks melted, this is a tell-tale sign that the neutral is compromised.

So between connections oxidizing to tree limbs wearing thru an overhead triplex neutral, one can see how often this occurs.

 

[ActionDave]

I hear ya, but what BC's are using shared N. MWBC is one, any others?

And in a MWBC, you need two items to be "on", which forms a virtual CT on the shared N between them, so across each device is still just 120v.

If you lose the xfmr CT at the panel, how does a non-MWBC two-wire BC see 240v ?

From the power company transformer to the service disconnect is one big MWBC.

 

[FionaZuppa]

From the power company transformer to the service disconnect is one big MWBC.

Right, so if the N is lifted there then no xfrmr "pole" has a path back to the CT, hence the 120v ckts won't run. So why would any device on a 120v ckt see 240v and fry? 120v BC's are two CCC's, one hot (black) and one neutral (white), and you lift the neutral, put a volt meter across that receptacle and you get zero. The N does not magically connect to the other end of the CT xfrmr. What am I missing?