EC&M Forensic Study for August...

[JoeTroubleShoot]

http://ecmweb.com/fire-amp-security/electrical-forensics-case-service-cable-arcing-incident

Just curious if anybody read this article. After I read it, I thought maybe they missed what another possible cause of the fire was. It's seems more possible that the neutral burned clear towards the transfomer and welded to a hot leg towards the house giving the homeowner 240v on half of the 120v circuits. That would explain them being part off when the handy man showed up and not notcing the arcing until the whole service eventually burned down.

I've seen it happen a few times before and it had the exact same effect. You end up reading 240 to ground on one leg, zero to ground on the other, and 240 leg to leg. Ground rod plus regular factors did not create enough current to trip the primary fuse and randowm fires broke out throughout the house. Just curious if anybody else thinks this might have been over thought or if it was thought about and ruled out.

 

[ActionDave]

I just read it this week and thought about starting a thread like yours. The final report sounded fishy to me.

 

[readydave8]

I did not understand the answer, which seemed to be saying that problem with service drop was putting more than 600 volts to house?

 

[Lectricbota]

I did not understand the answer, which seemed to be saying that problem with service drop was putting more than 600 volts to house?

Same here. I did have a little trouble following that one.

 

[ronaldrc]

Same here. I did have a little trouble following that on


The potential voltage that can be applied to the electrical system of this house can the 4.16 Kv or 23.kv
what ever the primary is in the worse case if you loose all grounding and the grounded primary connection and the primary cutout does not open.

Don't forget that Pole mounted transformer has a #4 hard drawn copper jumper from the primary to the secondary.

The utility systems will never own up to this danger.

Play the first video Moores mill road.

http://home.comcast.net/~ronaldrc/wsb/Videos.htm

Ronald

 

[GoldDigger]

I just read it this week and thought about starting a thread like yours. The final report sounded fishy to me.

The idea that the output impedance of a transformer is primarily or even significantly inductive seemed ridiculous to me.
Motors have inductance, primary windings of unloaded transformers form an inductive load, but the source impedance of a transformer secondary supplying less than rated load should be negligible. Unless you get into very high frequencies where the primary connections back toward the utility do not represent a low impedance.

 

[ActionDave]

The idea that the output impedance of a transformer is primarily or even significantly inductive seemed ridiculous to me.
Motors have inductance, primary windings of unloaded transformers form an inductive load, but the source impedance of a transformer secondary supplying less than rated load should be negligible. Unless you get into very high frequencies where the primary connections back toward the utility do not represent a low impedance.

You sound like a someone who knows why it sounded fishy. I like your opinion....I think. We do agree don't we?

 

[GoldDigger]

You sound like a someone who knows why it sounded fishy. I like your opinion....I think. We do agree don't we?

So far, yes.
It seems much more likely that overvoltage from a neutral problem killed both systems since they were probably the only sensitive loads online at the time. The lights could well have gone out because half of them were getting zero volts and the other half burned out immediately. With the house destroyed, the forensics on that might have been overlooked or unavailable.
The only argument against the neutral hypothesis that comes to mind would be if the A/C were purely 240V in the first place. But that is not likely for a window shaker.

 

[ActionDave]

So far, yes. ...

Cool. Do I get a key to the Smart Guy's Restroom now?

 

[GoldDigger]

Cool. Do I get a key to the Smart Guy's Restroom now?

It's down the block at the 7-11.

 

[steve66]

The final report sounded fishy to me.

I thought so too. I'm not sure I'm sold on the idea that the neutral connected to the other hot but I think it is worth considering. I also believe Golddigger was right that the evidence would have been there if this happened.

I did not understand the answer, which seemed to be saying that problem with service drop was putting more than 600 volts to house?

Same here. I did have a little trouble following that one.

Due to inductance on the wires, any current flowing through the wires can't stop instantly. If there is a brief open circuit (maybe a bad connection at the service conductors), the inductance will create as much voltage as necessary to keep the current flowing. Basically, the inductance creates enough voltage to make the current arc through the open circuit. The current and voltage then decay to zero over some time.

If the wind is blowing, and the open circuit keeps opening and closing, the inductance will keep generating short high voltage pulses.

Don't try this - its a really bad idea, but you can see the same thing happen if you have ever watched someon connect jumper cables to a battery, and short the cables together. The sparks are the high voltage kick as the current decays to zero. The much longer wires on a service feed means the generated voltage can be much higher, and the acring can last a longer duration.

Turning current to an inductor on and off is also the way they make tasers that create high voltages from a 9 volt battery.

So I don't have any doubt about the service conductors ability to generate high voltages. I just find it unlikely that the voltage would pierce through wire insulation with enough kick left to start a fire.

 

[GoldDigger]

Due to inductance on the wires, any current flowing through the wires can't stop instantly.

However, if the source and return wires for the intermittent circuit are close together (like inside SE cable or a raceway or NM) rather than being separate conductors with inches or more of separation, the inductance will be far less than for a single isolated wire of the same length. We do not know the configuration of the service wires to this house.

 

[junkhound]

Agree the final report is BS, there were NO numbers (e.g. L*di/dt = arcing voltage) to back up the claims.

The "jumper cables to a battery, and short the cables together" statement also does not stand up to facts. The sparks are the high voltage kick as the current decays to zero" statement also shows a lack of the scale of of L*di/dt inductive kickback. Say the indctance of the jumper cables is 2 uH, and current is 500A. Break time of a hand movement switch is typically 1 ms or so. V= 2u*500/1m = only 1 volt from inductive kickback.

Say xfmr was 25 kVA, percent Z is typically < 2%, equates to about 5V. Irated = 100A, so L=5/(377*100)=132uH
Typical short circuit current about 3 kA.

Arcing of 60 Hz continues for about 5-6 ms, so di=say 5 ms. V=132u*3000/5m = only 80 Volts.

So, basic premise the report is BS, no numbers to back it up cause it was just BS as investigator not astute enough to run the numbers.

What was the cause?
Well, since there were 2 separate arcing events, we can surmise that there were not 2 separate situations of wire contact or insulation failure that would allow a 320Vac arc (240+80). 1500V is the UL dielectric withstand test, so likely > 1500 V was present.


Where did the 1500 V come from:
Let us assume that there was an open neutral at the pole.
What I have seen accurately reported in the past is that a squirrel caused the fire.
Also have a reference to a paper somewhere where a squirrel destroyed an entire substation!! (cannot find that ref right off)
I did google squirrel shorts and the first hit was
http://articles.courant.com/2012-10...mall-fire-circuits-fire-marshal-steven-waznia

Squirrel out on the pole, open secondary neutral, squirrel gets fried between 4 kV (or whatever on primary) and the secondary, momentarily applies 4 kV to the service drop.


Probably no one looked for the dead squirrel, the crows probably had it picked clean by the time the fire was out?

 

[kwired]

I just read it this week and thought about starting a thread like yours. The final report sounded fishy to me.

As do most of these reports they publish. I personally believe there is more of an agenda behind every one of these stories then there is a desire to find the truth in each case. Notice how he started pushing AFCI's in the story?

We also had two fires in two separate areas of the house and then suddenly we throw in a kind of by the way problem with the service drop:?

Wouldn't you think if service drop melted and fell away from the house it would have had some significant attention early in the investigation, and probably early in the story line?

I once submitted a photo to them for their code violations column. They published it, but changed the story behind it, which I was not impressed with at all. Sure it was a better story for reading, unless you actually knew the truth of what happened. They claimed in my submission that an electrical inspector approved the install. I never even mentioned any inspection in my submission. Needless to say the executive director at the State Electrical Division contacted me after he was made aware of the story wanting to know which one of his inspectors approved that installation, had to tell him none of his inspectors even seen the install, and that for whatever reason they thought their story was better for publishing than mine was. I still read their real life stories, but do not actually buy into the information given being entirely true since that incident. If it were a paid subscription magazine I would not buy it.

 

[steve66]

The "jumper cables to a battery, and short the cables together" statement also does not stand up to facts. The sparks are the high voltage kick as the current decays to zero" statement also shows a lack of the scale of of L*di/dt inductive kickback. Say the indctance of the jumper cables is 2 uH, and current is 500A. Break time of a hand movement switch is typically 1 ms or so. V= 2u*500/1m = only 1 volt from inductive kickback.

I think you are way off on dt. It would be very close to zero - almost instantaneous. One instant the cables are connected, the next instant they are not. Even if we use 1 microsecond, we now have 1000 volts, not 1V. Full current flows until the cables are not connected, and then the voltage has to be enough to jump a short air gap.

Say xfmr was 25 kVA, percent Z is typically < 2%, equates to about 5V. Irated = 100A, so L=5/(377*100)=132uH
Typical short circuit current about 3 kA.

Arcing of 60 Hz continues for about 5-6 ms, so di=say 5 ms. V=132u*3000/5m = only 80 Volts.

The voltage is determined by the time it takes to break the connection, not the time the arc is sustained.

 

[junkhound]

Try this:
Put one lead of a storage scope on a battery terminal and the other lead on the jumper cable clamp. Open the circuit.
Read the scope. What voltage do you see? Then talk about 'instantaneous'.

 

[steve66]

Try this:
Put one lead of a storage scope on a battery terminal and the other lead on the jumper cable clamp. Open the circuit.
Read the scope. What voltage do you see? Then talk about 'instantaneous'.

I don't just happen to have a storage scope laying around, and i'm not going to short out a good battery either.

Anyway you look at it, the voltage is going to have to be enough to jump an air gap. That's why there are sparks.

One volt is not going to jump an air gap. The only way to do that is to create enough voltage to ionize the air.

 

[kwired]

I don't just happen to have a storage scope laying around, and i'm not going to short out a good battery either.

Anyway you look at it, the voltage is going to have to be enough to jump an air gap. That's why there are sparks.

One volt is not going to jump an air gap. The only way to do that is to create enough voltage to ionize the air.

The spark that jumps the gap between conductors is one thing, the sparks that fall to the floor are molten metal pieces that were once a part of one of the conductors.

 

[JoeTroubleShoot]

Glad I wasn't the only one who thought this article sounded strange. I think they have it all wrong. I would be willing to bet there was no high voltage involved at all. Just a plain old broken neutral and equipment caught fire, not the wire.

I have seen hundreds of secondary faults in various locations from the panel to the transformer and I have never seen arcing cause a fire in something on the load side of the arc unless the neutral gets broken or hot legs get burned together. Just wanted to throw out the possibility that the neutral was welded to the hot and burned clear back to the pole because it's something for guys to look out for in the future. Arcs almost always blow secondaries apart, but on the very rare occasion, you can get 240 across half the house with a welded neutral . I agree that it could have been just the normal old fashioned burned open neutral too. The only reason I can think of besides that is that the hot legs burned together and one of them burned clear. That would make it possible to overload a neutral on a MWBC in the house.

They make it sound like the company paid because of this investigation, but I highly doubt it. I've been through a couple cases where the company was 99.9% sure it was the customer's problem, but the fact that we were near the house in the last year made them pay without even investigating. If this were a criminal lawsuit, I would bet the utility would have had a field day with this investigation. Also, that service was not new. You can tell by the tape on the connectors and the fact that nobody has used connections like those on in a very long time.

Just curious, if the smart guy bathroom is at 7-11, then where is the dumb guy bathroom?

 

[steve66]

The spark that jumps the gap between conductors is one thing, the sparks that fall to the floor are molten metal pieces that were once a part of one of the conductors.

Yes, i can agree with that. But some of the sparks seem to travel from one cable to the other after the connection is broken. That can't be bits of molten metal.

And at least we all agree the high voltage sounds kind of iffy starting the fire.

Glad I wasn't the only one who thought this article sounded strange. I think they have it all wrong. I would be willing to bet there was no high voltage involved at all. Just a plain old broken neutral and equipment caught fire, not the wire.

Most of these forensic investigations sound a little fishy to me. There doesn't ever seem to be any proof, just some circumstantial evidence. I think an open neutral sounds like another good possibility.