Meltdown. What happened?

[electrofelon]

A building where I did a service upgrade a few months ago had a fault in a feeder feeding a transformer. Presumably at about the same time, the 2 C phase conductors broke/melted off the utility transformer secondary terminals. Here are the details: 1000 amp 120/208 overhead to underground service. 6 service disconnects. One service disconnect is a 150 amp breaker feeding 3phase 3 wire #1 CU in 2" emt which feeds a 45 KVA transformer. The transformer installation is old and incorrectly done. Its a 240 delta to 208 wye used in reverse. The XO (now the primary side) had a neutral landed on it. I had to rework the conduit and pull out the conductors as part of the service upgrade and I did not reconnect the neutral when I hooked it back up. The secondary is ungrounded and has no ground fault detectors. The transformer feeds three AC compressors. The load is not very much, probably 50-60 amps worst case. So I get called in an find a hole melted through the 2 inch EMT, and one phase out. I look up the pole and the c phase conductors are just hanging there. The fault on the EMT looked like it was quite dramatic: There was a 2 inch long slot burned in and there is black spot on the ceiling and bits of black charred stuff everywhere. It looks like perhaps it started as an overload, or the fault took a while to clear. The fault was right at a coupling where leads to be think there was a burr on the pipe or nick in the insulation, but it is odd as it was a big pipe for three number ones and only 6 feet of conduit - not like it took any sort of effort to pull those conductors in and damage them. Then the conductors burning off the transformer terminal at the same time seems strange too. Any ideas as to the chain of events here? Ill post some pictures in a bit.

 

[mgookin]

How much vibration is there when those compressors are on?
And does the EMT absorb any of that vibration?

 

[electrofelon]

How much vibration is there when those compressors are on?
And does the EMT absorb any of that vibration?

No the fault was far from the transformer, just 8 feet out of the mdp.

 

[JFletcher]

The transformer let its smoke out! Dunno really, but you said the xfmr was old and kinda rigged. Saw one last year on a dollar store; they lost one phase. Fuses blown. Find out a tree root had grown up thru the pad, into the xfmr, and shorted one phase. ofc it was pouring rain when the POCO replaced it. I dont know enough about xfmrs to even hazard a wag at this one. Interesting tho.

 

[mgookin]

Was that 6' chunk of 2" touching building steel?

 

[big john]

I'm guessing the reason the wires in the customer transformer burned was because that winding was also contributing current to a short conductor already very overheated by a serious fault.

And because aluminum oxidizes more easily when hot, it created a self-propegating resistance until the termination failed.

 

[electrofelon]

Was that 6' chunk of 2" touching building steel?

No. It was on strut screwed into wood framing.

I'm guessing the reason the wires in the customer transformer burned was because that winding was also contributing current to a short conductor already very overheated by a serious fault.

And because aluminum oxidizes more easily when hot, it created a self-propegating resistance until the termination failed. .

So just to clarify, we had a fault on the transformer primary circuit (150A 208 3 Ph) and the conductors up on the pole were melted off the secondary lug on the utility transformer. I dont see how our fault (150A breaker) would have caused enough fault current for a long enough duration to melt the 2 2/0 CU up on the pole. They must have had a real bad connection up there already and/or maybe our 150 amp breaker is defective. All effected conductors were copper.

I havent yet reconnected the transformer to see if everything is ok with it, although visually it looks fine.

 

[shortcircuit2]

Just heard report of a squirrel that had gone down a 4" riser pipe on a street pole and gnawed though wires underground in the pipe where it was setting up its winter bungalow...

 

[electrofelon]

Just heard report of a squirrel that had gone down a 4" riser pipe on a street pole and gnawed though wires underground in the pipe where it was setting up its winter bungalow...

yeah the lineman who hooked us back up a few hours ago was just telling me about all the critter problems they have.

Here are the pictures as promised. Maybe there was an arcing fault that went on for a while and ate away a the pipe until the insulation failed on the other conductors and then the breaker finally tripped?

 

[GoldDigger]

The conduit damage looks a lot like arcing between the tubing and the connector. Possibly the result of a ground fault sending high current through the conduit EGC. Was there any alternate EGC path?
Not sure whether this was an early or late event in the failure cascade.

 

[electrofelon]

The conduit damage looks a lot like arcing between the tubing and the connector. Possibly the result of a ground fault sending high current through the conduit EGC. Was there any alternate EGC path?
Not sure whether this was an early or late event in the failure cascade.

There was no wire EGC. The coupling was firmly tightened and the emt was fully seated in it or very close to it. Just seems suprising to me that that much damaged happened before the breaker tripped, or is that not suprising given the right conditions and an arcing type fault?

 

[GoldDigger]

I suspect that the duration of the arc was up in range of seconds rather than cycles.

 

[kwired]

I'm confused at the source arrangement. Is the reverse fed 240-208 transformer part of premises or something the utility controls? If utility controls it why do they have that instead of typical medium voltage stepping down to your service voltage?

If there is a ungrounded system mixed in there somewhere, I would be looking into failure of grounding/bonding to be a contribution to this event, possible some (grounded conductor) current was flowing somewhere it shouldn't have been - possibly for some time to start setting up the catastrophic failure that ended up happening, or even a ground fault on a high resistance path.

 

[electrofelon]

I'm confused at the source arrangement. Is the reverse fed 240-208 transformer part of premises or something the utility controls? If utility controls it why do they have that instead of typical medium voltage stepping down to your service voltage?

If there is a ungrounded system mixed in there somewhere, I would be looking into failure of grounding/bonding to be a contribution to this event, possible some (grounded conductor) current was flowing somewhere it shouldn't have been - possibly for some time to start setting up the catastrophic failure that ended up happening, or even a ground fault on a high resistance path.

The reverse fed 240-208 transformer is customer owned running three AC compressors. The service is pole mount bank of 3 X 50 KVA (as a side note the line crew said the primary was 24 something KV and a very odd voltage. Said they are one of a handful of utilities that use it) and 120/208 secondary. The service is down the pole and underground. I was suspicious of a "weak" neutral as that fault didnt seem to clear very quickly, but clamp meters shows even distribution through the parallel neutrals and no significant voltage imbalance under load, or significant current on EGCs and MBJ's and GEC's. The service entrance conductors are pretty long however - almost 200 feet. The first 100 feet is 4 1/0 AL to a hand hole which is the service point and then 2 2/0 AL another 100 feet to and up the pole.

 

[kwired]

Think I have a better picture now of what you have, still a little confused on exactly what and where damage was.

But if you are back feeding a delta-wye transformer - if that damaged section you have a picture is in the supply to that transformer I sort of suspect it was overloaded neutral that should have never been connected that caused that. If there was additional damage outside on the service lines (which is what I think you had, it may or may not be related - there should have been overcurrent protection in the grounded conductors so maybe just a failure of parallel components putting too much load on remaining components here?

 

[jaggedben]

...But if you are back feeding a delta-wye transformer - if that damaged section you have a picture is in the supply to that transformer I sort of suspect it was overloaded neutral that should have never been connected that caused that. ...

Except that he said he didn't reconnect the neutral. Evidently from the picture of the three conductors, he didn't even re-run the neutral in the EMT. Which seems fine, AFAIK.

 

[highlegdelta]

Is X0 still bonded to the can? I know you said you removed the neutral that was tied into the transformer primary, but if you didn't disconnect the bonding jumper between X0 and the transformer housing then you will likely see a pretty hazardous amount of current on that EGC (2" EMT). Because all three phases would be contributing to this, it might not have been enough to trip that 150A breaker.


Sent from my iPad using Tapatalk

 

[electrofelon]

Think I have a better picture now of what you have, still a little confused on exactly what and where damage was.

But if you are back feeding a delta-wye transformer - if that damaged section you have a picture is in the supply to that transformer I sort of suspect it was overloaded neutral that should have never been connected that caused that. If there was additional damage outside on the service lines (which is what I think you had, it may or may not be related - there should have been overcurrent protection in the grounded conductors so maybe just a failure of parallel components putting too much load on remaining components here?

Except that he said he didn't reconnect the neutral. Evidently from the picture of the three conductors, he didn't even re-run the neutral in the EMT. Which seems fine, AFAIK.

Is X0 still bonded to the can? I know you said you removed the neutral that was tied into the transformer primary, but if you didn't disconnect the bonding jumper between X0 and the transformer housing then you will likely see a pretty hazardous amount of current on that EGC (2" EMT). Because all three phases would be contributing to this, it might not have been enough to trip that 150A breaker.

Right I disconnected the neutral a few months back, but highlegdelta is correct that there was a XO to case bonding jumper and I made the mistake of not disconnecting it -I didnt even notice it (Actually it just so happens I noticed this today while putting things back together and did disconnect it). But, wouldnt that just mean the unbalanced current would be flowing back on the EMT? The only imbalance on the loads served by the transformer would be a few single phase fan motors, hardly enough to heat that pipe up no? I have a theory: What would happen if I lost a phase serving that transformer with the XO (connected to the EMT EGC?

 

[highlegdelta]

What would happen if I lost a phase serving that transformer with the XO (connected to the EMT EGC?

If the only imbalanced load was very small then I'm sure you're right. You probably lost that C phase from the utility creating a large imbalance and putting a load across the X0 which returned on the EMT. A good example showing the need for a EGC wire, IMO.


Sent from my iPad using Tapatalk

 

[electrofelon]

If the only imbalanced load was very small then I'm sure you're right. You probably lost that C phase from the utility creating a large imbalance and putting a load across the X0 which returned on the EMT. A good example showing the need for a EGC wire, IMO.

So with an open phase and the XO connected to the neutral (via the EGC/EMT), what would be the magnitude of the current on the EGC? I am thinking that the utility phase failure happened first and somehow caused the feeder to burn up, not the other way around so just trying to come up with a plausible explanation.