'sparks' when connecting gas pipe to furnace

Did you check the neutrals at any other houses fed by this transformer and metallic water lines?
When I met the POCO lineman there he checked connections and load tested, since the problem goes away when the Emergency Disconnect is shut off the problem seems to be between the Emergency disconnect and the Service disconnect in the house,
Service disconnect is original in the SQ-D QO panel on the opposite end of the house, I have narrowed it down when I shut off all L2 breakers and create large neutral load with all L1-N breakers on, current is flowing parallel with the SE cable across the house via gas and water pipes. Same if I shut off all L1 breakers and switch on the L2. I am going to go back and look for a buried splice in the SE cable, part of the basement is finished and I cant see the entire cable as its is routed around the finished area into a crawl space. The gas pipe goes thru the finished part a shorter direct route out to the meters.
I am not the installer of the service but its new, there is new ground rod at the EM disconnect.
 
Rule out possibility Neutral to ground fault both L1 and L2 sides
Putting both circuit conductors and grounded conductor together in clamp ammeter
 
There is 2 amps on the gas pipe
The incoming gas piping is normally isolated by either non metallic piping if it is newer or even on old metallic underground they usually had isolating unions at the gas meter so should be no path through incoming gas piping, but since you mentioned bonding for CSST piping and jumping over to the metallic water piping that is also an electrode, that allows current to flow on interior piping. Might just be parallel current to what is flowing on the GEC. Metallic water pipe distribution is prone to have stray neutral current on it and is not considered objectionable current.

If you remove bond jumper between the water pipe and gas piping and connect gas piping bond more directly to the service equipment you lessen the chances of parallel paths for any current on the GEC. Maybe open that connection temporarily just to see what changes that makes.
 
What's between the two panels that both have G-N bonds?
Underground service lateral comes up into a meter main with a 200A breaker, there appears to be a ground rod with a #6. It has a factory N-G bond in this disconnect, appears standard.
The meter main is labeled 'Emergency Disconnect Not Service Equipment'.
The gas meter is about 6' away and there is a CSST manifold also connected to the new ground rod with a #6.
The HVAC guy (that I work for) says the installation instructions of the CCST require the fitting to be grounded.
From the service a 4/0-4/0-2/0 SE cable (AL) enters the crawl space. There is no EGC in said cable.
The cable goes thru a crawl space (disappears I cant follow it) pops into the basement and lands in the Service panel a older Square D. I can say it does not take a direct route due to the house construction.
There are several #6 and #4 bare ground wires exiting this service panel, the #4 hits the water meter jumps around it as is typical (they are in the basement here).
Not sure what the other #6 goes to probably another ground rod buried.
The neutral and ground are bonded in this panel.
I am not too familiar with these 'Emergency Disconnect Not Service Equipment' devices any time I had to do something like this I would have ran a 4 wire SE-R cable and had a one ground electrode system outside then ran new #4 bare to the water meter in the basement etc.
I would greatly appreciate it if you could point me to the code sections on how grounding is correctly done on one of these 'Emergency Disconnect' ?
The way I was always taught is you have one GES outside at the meter main and a 4-wire, if this is legal like it is it seems to have been a great time saver for the installer.
I believe this was done under the 2020 code.
Thanks for your help
 
Last edited:
My opinion em disconnect also service disconnect, service meter there
Service panel then feeder panel , no ng bond should there.
So label to change
 
Underground service lateral comes up into a meter main with a 200A breaker, there appears to be a ground rod with a #6. It has a factory N-G bond in this disconnect, appears standard.
The meter main is labeled 'Emergency Disconnect Not Service Equipment'.
The gas meter is about 6' away and there is a CSST manifold also connected to the new ground rod with a #6.
The HVAC guy (that I work for) says the installation instructions of the CCST require the fitting to be grounded.
From the service a 4/0-4/0-2/0 SE cable (AL) enters the crawl space. There is no EGC in said cable.
The cable goes thru a crawl space (disappears I cant follow it) pops into the basement and lands in the Service panel a older Square D. I can say it does not take a direct route due to the house construction.
There are several #6 and #4 bare ground wires exiting this service panel, the #4 hits the water meter jumps around it as is typical (they are in the basement here).
Not sure what the other #6 goes to probably another ground rod buried.
The neutral and ground are bonded in this panel.
I am not too familiar with these 'Emergency Disconnect Not Service Equipment' devices any time I had to do something like this I would have ran a 4 wire SE-R cable and had a one ground electrode system outside then ran new #4 bare to the water meter in the basement etc.
I would greatly appreciate it if you could point me to the code sections on how grounding is correctly done on one of these 'Emergency Disconnect' ?
The way I was always taught is you have one GES outside at the meter main and a 4-wire, if this is legal like it is it seems to have been a great time saver for the installer.
I believe this was done under the 2020 code.
Thanks for your help
Confusion how to ground em disconnect to delete in next code cycle
 
Update is the homeowner got her General Contractor (GC) ( the EM disconnect was done by his electrician when they buried the service for an addition) and the HVAC guy over for a meeting this morning, the HVAC guy showed the GC how the parallel neutral current flows on every metal system thats bonded in the basement and outside even when everything is to code and is saying this could cause a gas explosion. (He's a former firefighter)
The GC is going will warranty it, he is going to talk to his electrician about it. They requested a bid from me to 'fix it',
I said I would do it my way:
Replace the SE with 4-wire SE-R
Run a #4 bare from the water pipe out to the meter main,
Clamp the other existing grounds in the basement to the water pipe.
Separate neutrals on grounds in the old QO panel.

I cant think of any other way to 'fix' the parallel neutral current other than have a single neutral ground bond for the structure what do you all think?
 
Last edited:
Hey gang here is a odd one I got a call from a HVAC guy said his son (the tech) was replacing a gas furnace and saw sparks when he connected the gas pipe. He did not undo the gas line after he connected it and he left the gas off, none of the breakers or fuses blew, its a old building. My guess is static electricity, what would else cause sparks but not trip a breaker?
Thanks
Ground-loop.
Electricity flowing the path of all resistance. If you have a 0.1 ohm cable and you parallel a 0.2 ohm cable, the latter will carry 1/3 of the total current. If you assume that there's an outlet right at the panel, ground is connected to neutral at this point. If you also connect the ground and neutral at the receptacle end of the extension cord, current will flow in the green too. If you jump the green and neutral on the load end of the extension cord while it's heavily loaded, it will likely spark a bit.

When there's a voltage difference between the panel's bonding point and where the gas line bonds on the utility end, current will flow when the two are joined.

Path of least resistance is only true for gas discharge arc. So, if you parallel two fluorescent lamps together, they won't glow together, but only one will strike.
 
Last edited:
Wrong way original installation. The correct way do it explain here
protonsparky right—- what that guy has in there he could make it the emergency disconnect and the inside the service disconnect. We got rid of that rule, which is nice at the same time. Sucks cause I’ve had situations where an emergency disconnect would’ve been nice since I didn’t need that fourth wire since it’s not the service disconnect.
But there’s nothing special with that video. he should’ve fought that 16 inches and not been a chicken and back down—
 
Top