Coax cable caught fire

[nickelec]

Has anyone ever come across this before I had a customer call me up and tell me that the coax cable going into there cable box caught fire. I've had issues personally with plugging in a coax line and seeing as arcing due to after investigation I found the nutreal was not bonded after bonding the nutreal problem went away I'm heading over there in a little while any ideas on where to start what to test I know that the nutreal is bonded at this location for sure.

Sent from my SM-G950U using Tapatalk

 

[nickelec]

I forgot to mention that this same customer not too long ago was telling me that he notices that his lights dim periodically without him really turning on any major loads, that to me seems like a Poco issue.

Sent from my SM-G950U using Tapatalk

 

[gadfly56]

Has anyone ever come across this before I had a customer call me up and tell me that the coax cable going into there cable box caught fire. I've had issues personally with plugging in a coax line and seeing as arcing due to after investigation I found the nutreal was not bonded after bonding the nutreal problem went away I'm heading over there in a little while any ideas on where to start what to test I know that the nutreal is bonded at this location for sure.

Sent from my SM-G950U using Tapatalk

A former work colleague of mine had an issue with a bad/intermittent neutral at his house. The shield on the coax was carrying about 7 amps as a result. I can imagine Bad Things will happen if you push this up much.

 

[kwired]

Sounds like a good chance there is neutral current on the shield, you need to determine if it is the customer's neutral gone bad, or if it is something elsewhere. If his lights are dimming without him changing any load the problem could be further upstream on utility conductors. Also could result in his coax is carrying neutral current from more then just his house.

 

[nickelec]

I guess the only way to to determine if the coax is carrying current would be to but an amp probe on ? In normal operation should the coax have any voltage?

Sent from my SM-G950U using Tapatalk

 

[kwired]

I guess the only way to to determine if the coax is carrying current would be to but an amp probe on ? In normal operation should the coax have any voltage?

Sent from my SM-G950U using Tapatalk

There has to be some voltage to do what it is intended to do. but the amount of "normal" current would probably be in the mA ranges I would guess.

 

[JFletcher]

Coax catching fire is a sure sign the service neutral is toast or nearly so (maybe a strand or two isnt burnt thru).

eta: this is how it happens, excellent article that explains everything concisely:

https://www.electrical-forensics.com/Open-Neutral/Open-Neutral.html

 

[JFletcher]

Coax can arc/spark lightly if your phone service is thru the cable and someone rings the house while you're connecting it (100V ring voltage), tho that will not be the cause of a burned cable.

Coax is grounded back to the service panel via the 1st splitter. Cable Co shares POCO ground on pole. Split phase service, neutral carries imbalance of current back to the xfmr. Normally the neutral is in good condition and carries all of that current despite the parallel path. Lose the neutral (and w/o a metal water pipe bond) now all the house's neutral current goes thru the neutral ground panel bond, onto the ground system, to which the coax is tied to via the #14 solid bare ground wire that's usually there at that 1st splitter, across the outer braids/shield of the coax back to the common xfmr/Cable Co ground to the xfmr, and that thin foil and braids is gonna carry maybe 10A before it smokes. That's also why the coax doesnt smoke after that splitter (on the customer side); the #14 ground wire is carrying the imbalance of the current and a free air #14 wire is good for probably 30A. The choke point is that splitter and the coax shielding.

The article I cited doesnt quite mention the complete pathway as above.

eta: for this exact reason is why I do not like tying the coax to the building's GES. A separate rod at the cable demarc like they did 30 years ago will shunt a lightning strike just as much as tying to the GES will (probably better given the extremely short run of #14 to the local rod). If you lose a neutral you will not fry the cable. Continental Cable Co did every house here in the 80s like that and I've never seen problems of any kind with that setup.

 

[JFletcher]

Missed the edit time; the chart on that page show the current going thru your coax thru the neigbors coax thru their panel bond, triplex and service neutral. If that were the case, the neighbor's coax line would be melted along with yours/the house in question.

 

[hbiss]

Coax can arc/spark lightly if your phone service is thru the cable and someone rings the house while you're connecting it (100V ring voltage), tho that will not be the cause of a burned cable.

Ah, no. If you have phone service from the cable company the ring voltage is generated by the EMTA (cable modem) in the building where the phone wiring plugs into. The only thing on the coax is a digital RF signal that carries video, internet and voice.

I think this has become enough of an issue that it should be addressed. What I would like to see is cable companies using a ground block that isolates the shields. I do believe they make these. The drop shield would be grounded out on the street and the house side would be bonded to the GES. No continuity between the two except for RF. That would stop neutral current from using the coax shield as a parallel path.

Maybe time for the NEC to make some changes to Art. 820.

-Hal

 

[JFletcher]

Ah, no. If you have phone service from the cable company the ring voltage is generated by the EMTA (cable modem) in the building where the phone wiring plugs into.

-Hal

Okay, well I've been tingled a few times terminating coax at the NID - still must have been an incoming call signal the modem and backfeeding the ring voltage to me then? It didnt happen every time I touched the coax.

In any event, I've seen a tiny arc there and gotten a little zap.

"What I would like to see is cable companies using a ground block that isolates the shields."

Me too, and they do make those; coax power isolators.

 

[nickelec]

So I got to the customer house L1 to nutreal 110
L1 to ground 90v
i put in a call to poco isee no visible sings of a bad nutreal in customer equpiment


Sent from my SM-G950U using Tapatalk

 

[nickelec]

Correction 90v nutreal to ground

Sent from my SM-G950U using Tapatalk

 

[JFletcher]

Correction 90v neutral to ground

Sent from my SM-G950U using Tapatalk

What is L2 to neutral and ground? and Line-Line? You should have 0V neutral to ground; look for a failed neutral in other places like the meter or service drop, weatherhead, or pole

 

[ggunn]

In a related story...

A friend bought a small travel trailer that was wired for 120V which connected to "shore power" through an extension cord. It was wired by the previous owner.

Everything seemed to work fine until he ran a CATV line out to the trailer. In the course of moving some wires around, a coax connector happened to touch an aluminum window frame. It caused a big arc and melted a divot out of the window frame. It seems that the previous owner had got line and neutral reversed when he wired the double male extension cord he used to connect the trailer to power and in the trailer he tied what he thought would be neutral to the trailer frame.

My friend is lucky that no one got hurt.

 

[hbiss]

Okay, well I've been tingled a few times terminating coax at the NID - still must have been an incoming call signal the modem and backfeeding the ring voltage to me then? It didnt happen every time I touched the coax.

Because of exactly what we are talking about here. Neutral current on the coax. It isn't even remotely possible* that the POTS outputs of the modem could affect the coax input.

*Unless some idiot was trying to use coax for phone wiring and it got connected to the CATV cabling- or something like that.

-Hal

 

[Coppersmith]

double male extension cord

The infamous "suicide cord". hmy:

 

[kwired]

Coax can arc/spark lightly if your phone service is thru the cable and someone rings the house while you're connecting it (100V ring voltage), tho that will not be the cause of a burned cable.

Coax is grounded back to the service panel via the 1st splitter. Cable Co shares POCO ground on pole. Split phase service, neutral carries imbalance of current back to the xfmr. Normally the neutral is in good condition and carries all of that current despite the parallel path. Lose the neutral (and w/o a metal water pipe bond) now all the house's neutral current goes thru the neutral ground panel bond, onto the ground system, to which the coax is tied to via the #14 solid bare ground wire that's usually there at that 1st splitter, across the outer braids/shield of the coax back to the common xfmr/Cable Co ground to the xfmr, and that thin foil and braids is gonna carry maybe 10A before it smokes. That's also why the coax doesnt smoke after that splitter (on the customer side); the #14 ground wire is carrying the imbalance of the current and a free air #14 wire is good for probably 30A. The choke point is that splitter and the coax shielding.

The article I cited doesnt quite mention the complete pathway as above.

eta: for this exact reason is why I do not like tying the coax to the building's GES. A separate rod at the cable demarc like they did 30 years ago will shunt a lightning strike just as much as tying to the GES will (probably better given the extremely short run of #14 to the local rod). If you lose a neutral you will not fry the cable. Continental Cable Co did every house here in the 80s like that and I've never seen problems of any kind with that setup.

I have to agree the ring voltage would be produced by whatever equipment converts to a POTS signal. If they just stepped voltage up - every one that had a phone connected would get the same signal and they all would ring. A data signal has to be sent to the conversion equipment that gives it a command to produce a ring signal on it's output.

As far as being connected to the POCO grounded conductors as well as to your grounding system in the house - this does put the cable shield in parallel as you said. But the current doesn't just flow on the neutral just because that is where it is intended to flow. Current flows through all available paths. How much flows through each individual path when there is more then one path depends on resistance of each path, least resistance path carries the most but all paths carry some current.

 

[ggunn]

The infamous "suicide cord". hmy:

I call them "Aggie Extension Cords". I went to both LSU and UT.

 

[goldstar]

I had a problem like this several years back. The CATV company in my area will allow 1A or less on their coax. So, if you clip your amp-probe around the entire incoming coax that's the reading you should have. If you have more, then it is more than likely a bad neutral either at or before the distribution panel. In addition, if the HO sees some of their lights dimming there's probably another area in the house where the lights age getting brighter and also quite possibly, any appliance that has a digital display is now toast.