Breakers and buss appear to be overheating on a long-term basis

[Pants]

My 125A/240V service meter-main/load center is about ten years old, an Eaton MBE1224B125BTS, which (aside from the main common-trip 125A pair) has 12 buss lugs available. Mainly it's used for light residential loads. But since I also do some hobby-level shop fabrication metal work, I run a 50A common-trip for a large air-compressor. Due to lack of slots in those days, I cheated a bit by ganging that load with an EWH that draws 20A, just measured with clamp-on. The compressor draws 25A per leg, according to my measurement this AM. The heater is on a timer and runs in the late afternoon, and the air-compressor is of course intermittent based on my usage and having it on to begin with. So total 45A on that 50A breaker seems to be the worst-case, yes?

I have a second 50A common-trip that's used on relatively rare occasions for a plasma-cutter or a welder, both of which present varying loads, and are really very-rarely used.

Finally, I had a 20A common-trip that was, until recently, used to charge our EV (15A max) - maybe a year old. I just pulled that one last night, and found that it had smoky brown deposits on one of the two female slots. (photo shows it and the 50A breaker)

I've replaced 50A breakers three times now - not sure what they were feeding when they failed. Last night and on at least one prior occasion, it was the compressor/heater breaker. When I tried to pry it out of the bus last night, it fell apart at the buss slot, leaving the female/springy part on the buss (photo), and releasing a bunch of cracked black case-plastic into the panel. Similar thing happened on two other occasions, though I can't swear whether it was the same load combination.

The bussing itself now looks like something from a fire-inspector's training manual (photo). The black plastic insulator is warped and melted in numerous locations, corresponding to where I've hopscotched around in years past, looking for clean metal without the smoked insulation. Evidently there's enough heat generated to leave a damp-looking brown sooty deposit around the blades in those locations, but the distorted half-melted plastic is really what stands out.

Now, I'd have given up on this and just replaced the load-center, but this is a unitized meter-can/breaker panel. I'd be at risk with the local electrical utility if I were to go pulling the meter out (and I'd be working live, which I'd prefer not to do) to replace the whole nasty affair. So I'd much rather replace the bussing - separately.

Through an unbelievably bureaucratic process that took around a month, I got in touch with the right POCs at Eaton. I was thinking they're big enough that they'd just send me the damned replacement bus, and would perhaps share with me whether there had been such problems reported with some of them. Nothing like that happened. No, they require that I

• buy the new bus from one of their local distributors
• return the damaged one to the same distributor
• that distributor then has to UPS the damaged part to them in Illinois or wherever - can't come from me
• IF - and only if - Eaton inspects and then agrees that there was some problem that they're responsible for, they will credit the local distributor for the UPS charge, and separately refund me what I paid for the replacement part.

It's BYZANTINE. Since none of the local distributors wanted to take any of this risk on (understandably), they declined to get the replacement part for me.

But I'm really not sure that this is a defect, and if so, whether it might just as well be with the CH breakers (or whatever I've gotten to replace them in the past.) If they're passing more current than they, or the bussing, can handle without the drama, then why don't they ever trip?

Am I breaking a standard rule in terms of derating for my loads?

 

[Megawatt]

View attachment 118805View attachment 118806View attachment 118807My 125A/240V service meter-main/load center is about ten years old, an Eaton MBE1224B125BTS, which (aside from the main common-trip 125A pair) has 12 buss lugs available. Mainly it's used for light residential loads. But since I also do some hobby-level shop fabrication metal work, I run a 50A common-trip for a large air-compressor. Due to lack of slots in those days, I cheated a bit by ganging that load with an EWH that draws 20A, just measured with clamp-on. The compressor draws 25A per leg, according to my measurement this AM. The heater is on a timer and runs in the late afternoon, and the air-compressor is of course intermittent based on my usage and having it on to begin with. So total 45A on that 50A breaker seems to be the worst-case, yes?

I have a second 50A common-trip that's used on relatively rare occasions for a plasma-cutter or a welder, both of which present varying loads, and are really very-rarely used.

Finally, I had a 20A common-trip that was, until recently, used to charge our EV (15A max) - maybe a year old. I just pulled that one last night, and found that it had smoky brown deposits on one of the two female slots. (photo shows it and the 50A breaker)

I've replaced 50A breakers three times now - not sure what they were feeding when they failed. Last night and on at least one prior occasion, it was the compressor/heater breaker. When I tried to pry it out of the bus last night, it fell apart at the buss slot, leaving the female/springy part on the buss (photo), and releasing a bunch of cracked black case-plastic into the panel. Similar thing happened on two other occasions, though I can't swear whether it was the same load combination.

The bussing itself now looks like something from a fire-inspector's training manual (photo). The black plastic insulator is warped and melted in numerous locations, corresponding to where I've hopscotched around in years past, looking for clean metal without the smoked insulation. Evidently there's enough heat generated to leave a damp-looking brown sooty deposit around the blades in those locations, but the distorted half-melted plastic is really what stands out.

Now, I'd have given up on this and just replaced the load-center, but this is a unitized meter-can/breaker panel. I'd be at risk with the local electrical utility if I were to go pulling the meter out (and I'd be working live, which I'd prefer not to do) to replace the whole nasty affair. So I'd much rather replace the bussing - separately.

Through an unbelievably bureaucratic process that took around a month, I got in touch with the right POCs at Eaton. I was thinking they're big enough that they'd just send me the damned replacement bus, and would perhaps share with me whether there had been such problems reported with some of them. Nothing like that happened. No, they require that I

• buy the new bus from one of their local distributors
• return the damaged one to the same distributor
• that distributor then has to UPS the damaged part to them in Illinois or wherever - can't come from me
• IF - and only if - Eaton inspects and then agrees that there was some problem that they're responsible for, they will credit the local distributor for the UPS charge, and separately refund me what I paid for the replacement part.

It's BYZANTINE. Since none of the local distributors wanted to take any of this risk on (understandably), they declined to get the replacement part for me.

But I'm really not sure that this is a defect, and if so, whether it might just as well be with the CH breakers (or whatever I've gotten to replace them in the past.) If they're passing more current than they, or the bussing, can handle without the drama, then why don't they ever trip?

Am I breaking a standard rule in terms of derating for my loads?

The panel obviously has lost its ability to conduct electricity, having said that I wouldn’t recommend replacing the inside buss. Just replace the whole panel and draw a permit first so the power company can disconnect the power. Some states allow licensed electricians to cut the power lose but I never recommend that. Be safe and good luck. Maybe you can ship Eaton the burnt panel and maybe get reimbursed.

 

[Pants]

The panel obviously has lost its ability to conduct electricity...

I like that, but can't tell whether it's entirely on the sarcastic side, or...if that's really a thing. Then again, you did mention the possibility that Eaton would cover the replacement, so...gotta Google "load center lost mojo" and see what comes up.

I called the utility about a related issue around a year ago, wanting to temporarily drop the service-entrance to allow for removal of a huge overhanging tree. They offered not even a TOUCH of humanity, telling me I'd need a permit, and a licensed electrician. I almost didn't bother trying to get quotes (electricians here can cherrypick their projects all day), but one guy said something between $5K and $6K. (and yes, I will risk electrocution for that.)

 

[Tom256]

I am in the UK so not familiar with USA codings so take this for what it is and, as such, will not go into heavy detail. You asked about points to consider for de-rating.

A distribution assembly has an 'InA' rating (european symbol) - InA is the maximal permissable load that an assembly can distribute and this Eaton unit will have such a rating.
In simple terms you need to know this figure and set it against the demand as a first step. This then further needs rating against the protective device upstream of the distribution assembly.
(Aside from the obvious melting of the BUS (Aluminium) I see more evidence which support my statement that such might be worth investigating).

Keeping in mind that small overloads for a long duration will shorten the life of any wiring system and components. Cables, assemblies and all electrical connections must be capable of withstanding the currents and the temperatures that accumulate from the sum of those currents.
Small overloads of a long duration must be designed out. There is a UK regulation that states as such and I guess USA code as similar.

Example: with the compressor the starting current can be many times the rated current and frequent stop / starts can produce a cumulative heating effect. This starting surge current is often many times the circuit rating but if infrequent and of short duration no unacceptable heating occurs. However, when frequent the design MUST take into account the higher duty cycle.
Likewise de-rating may need to be applied to breakers running at rated current when next to each other. De-rating factors are applied to breakers running full capacity situated side-by-side.

Heating effects can become cumulative and protective devices will not necessarily provide protection for currents that just exceed the rating of the device yet the assembly feels it all.

EDIT: Does this panel have any generation attached - solar and such like?

 

[Pants]

I am in the UK so not familiar with USA codings so take this for what it is and, as such, will not go into heavy detail. You asked about points to consider for de-rating.

A distribution assembly has an 'InA' rating (european symbol) - InA is the maximal permissable load that an assembly can distribute and this Eaton unit will have such a rating.
In simple terms you need to know this figure and set it against the demand as a first step. This then further needs rating against the protective device upstream of the distribution assembly.
(Aside from the obvious melting of the BUS (Aluminium) I see more evidence which support my statement that such might be worth investigating).

Keeping in mind that small overloads for a long duration will shorten the life of any wiring system and components. Cables, assemblies and all electrical connections must be capable of withstanding the currents and the temperatures that accumulate from the sum of those currents.
Small overloads of a long duration must be designed out. There is a UK regulation that states as such and I guess USA code as similar.

Example: with the compressor the starting current can be many times the rated current and frequent stop / starts can produce a cumulative heating effect. This starting surge current is often many times the circuit rating but if infrequent and of short duration no unacceptable heating occurs. However, when frequent the design MUST take into account the higher duty cycle.
Likewise de-rating may need to be applied to breakers running at rated current when next to each other. De-rating factors are applied to breakers running full capacity situated side-by-side.

Heating effects can become cumulative and protective devices will not necessarily provide protection for currents that just exceed the rating of the device yet the assembly feels it all.

EDIT: Does this panel have any generation attached - solar and such like?

Thanks, Tom.
In reverse order -
No, no generation incoming, other than from the utility.

Derating, in general: if it's so dependent on variable such as the distance between near-capacity breakers, surge current, etc then I'd believe that "correctly" assembling a panel like this would be nearly impossible - and that by, say, IEEE convention the industry's EEs would have already applied an appropriately-conservative derating to all components. Otherwise, I can only IMAGINE the kind of inspector-vs-contractor arguments that would take place at each new panel's commissioning... I agree that surge-current could play a part in the compressor-startup period, but I don't think that's an issue in my setting (and I didn't cover this because it was already a wall of text) because my compressor-motor is actually three-phase, interfaced with a VFD that I set up to a very gentle start. So at least from what I can see (on an analog meter) the startup current rises very gently, more or less corresponding with the ramp-up speed that I once programmed when doing the VFD setup. So if there's a spike, it's very short-duration (I do need to go back and try startup directly from the breaker, where I have the best means to monitor current).

Busses: I've not gone so far as to file into them to check, but I think they'd not aluminum but rather tin-plated brass or similar - or I'd have seen a lot of oxidation on them, given the steamy climate we often have here. Also to clarify, it's the (black and transparent) plastic insulation between legs and against the panel's housing that seems to be doing the melting and burning, and not the current-carrying parts. But there's no question that there's overheating going on at SOME point during use, given the repeated breaker-baking. I tend to think that it's from local resistance at the breaker-to-bus connections, wherein the current wouldn't escalate, but the temperature certainly would. But why would that happen with multiple breakers? Some metallurgical problem with the bus? Would Eaton know about some batch problem like this, or would it have been lost in a sea of nationwide installations, wherein failures were not reported, occurring years afterwards, and handled by electricians who just want to get things back in order?

Then there's the truly bizarre potential for things like.... a rat urinating while perched atop this (exposed/sheltered) panel, and said urine having trickled on to the busses, remaining forever as a salty, semi-conductive curse? Ugh.

What I think might be the identical panel came up on ebay, and both since it would minimize the considerable effort involved in swapping out for a new panel, and because it would give me a pretty fair apples-to-apples long-term comparison, I think I'm gonna go for that option.

But wow, will I ever feel like a glutton for punishment if the same damned thing happens over the coming years.

 

[Tom256]

Thanks, Tom.
In reverse order -
No, no generation incoming, other than from the utility.

Derating, in general: if it's so dependent on variable such as the distance between near-capacity breakers, surge current, etc then I'd believe that "correctly" assembling a panel like this would be nearly impossible - and that by, say, IEEE convention the industry's EEs would have already applied an appropriately-conservative derating to all components. Otherwise, I can only IMAGINE the kind of inspector-vs-contractor arguments that would take place at each new panel's commissioning... I agree that surge-current could play a part in the compressor-startup period, but I don't think that's an issue in my setting (and I didn't cover this because it was already a wall of text) because my compressor-motor is actually three-phase, interfaced with a VFD that I set up to a very gentle start. So at least from what I can see (on an analog meter) the startup current rises very gently, more or less corresponding with the ramp-up speed that I once programmed when doing the VFD setup. So if there's a spike, it's very short-duration (I do need to go back and try startup directly from the breaker, where I have the best means to monitor current).

Busses: I've not gone so far as to file into them to check, but I think they'd not aluminum but rather tin-plated brass or similar - or I'd have seen a lot of oxidation on them, given the steamy climate we often have here. Also to clarify, it's the (black and transparent) plastic insulation between legs and against the panel's housing that seems to be doing the melting and burning, and not the current-carrying parts. But there's no question that there's overheating going on at SOME point during use, given the repeated breaker-baking. I tend to think that it's from local resistance at the breaker-to-bus connections, wherein the current wouldn't escalate, but the temperature certainly would. But why would that happen with multiple breakers? Some metallurgical problem with the bus? Would Eaton know about some batch problem like this, or would it have been lost in a sea of nationwide installations, wherein failures were not reported, occurring years afterwards, and handled by electricians who just want to get things back in order?

Then there's the truly bizarre potential for things like.... a rat urinating while perched atop this (exposed/sheltered) panel, and said urine having trickled on to the busses, remaining forever as a salty, semi-conductive curse? Ugh.

What I think might be the identical panel came up on ebay, and both since it would minimize the considerable effort involved in swapping out for a new panel, and because it would give me a pretty fair apples-to-apples long-term comparison, I think I'm gonna go for that option.

But wow, will I ever feel like a glutton for punishment if the same damned thing happens over the coming years.

As I said UK , just food for chewing down on in consideration.
Did this start with the installation of the VFD?

I take it this is single phase to three phase and it sounds like you were monitoring current at the load side of the VFD. Is that correct?

Happy thanksgiving day - over here it's just piggin freezing