Disconnect Melting

[JoelHunter1102]

Good afternoon all,
I have a situation that is stumping me.

My coworker installed a disconnect (100A 3ph 208v) for a client and the disconnect had the C leg melt, without blowing the fuse. So we replaced it, it happened again. Replaced it for a 3rd disconnect, and this time upsized the disconnect to a 200A heavy duty disconnect, and used fuse reducers to put new 100A fuses in.
Lo and behold, it lasted longer than the others, but ALSO melted the C leg without tripping the fuse... it took a couple of years to do so, but still ended up with the same result.

After the 1st time we ran a bunch of calculations and made sure wire sizing/distance/etc was all good. After the 2nd time we hired an outside engineering team to come in and do some studies, they deemed everything was fine. We have monitored it ourselves and keep coming up with the equipment pulling about 68A per leg...so not enough to trip the fuses, but somehow still melting the disconnect.

We assumed weak internal connections in the disconnect itself (faulty from the factory), but at this point after three disconnects met the same end, my coworker and I are a bit dumbfounded.

Anyone have any thoughts?

Thanks in advance!

 

[retirede]

Sabotage?

 

[LarryFine]

Do you have a thermal camera? I would want to pinpoint the hot spot.

 

[Tyler C]

Any external sources of heat nearby?

 

[jim dungar]

Which lugs or fuse clips are failing, the line or the load?
What is your system, how many wires, what voltages?

 

[ModbusMan]

What's the connected gear? Sounds to me like a harmonics issue of some kind since your load is within spec, multiple disconnect replacements (plus the upsized one) pretty much rules out a factory defect. External heat sources seem unlikely since it's always C phase.

 

[Tyler C]

External heat sources seem unlikely since it's always C phase.

Potentially an external heat source is located to the right side of the disconnect, so it always melts C.

 

[jim dungar]

Potentially an external heat source is located to the right side of the disconnect, so it always melts C.

It would have to substantially hot to melt the C phase components versus melting just the fuse. They have been replacing the entire switch.

 

[ptonsparky]

Thermal camera or do FOP across the connection.
I would be inclined to think poor termination techniques. No torque wrench applied. Did they cut the damaged conductor back to good? Or reuse? Same tech?

Need pictures. Before & after of course

 

[synchro]

...
I would be inclined to think poor termination techniques. No torque wrench applied. Did they cut the damaged conductor back to good? Or reuse?

I'm also thinking that the conductor is likely causing the problem. If they cut off the obviously damaged conductor, perhaps there is oxidation or other corrosion that goes further up the conductor. Is it copper or aluminum?

 

[Deltaforce]

A preventive monitoring option may also tried, IR scan instead:
A simple wireless temperature probe or IoT thermocouple setup be mounted inside the disconnect enclosure
monitor continuously critical hot spots, log data, and automatically send alerts to a smartphone or computer when a set temperature limit is reached, cost-effective preventive option for repeated disconnect overheating.