Several arc fault breakers in panels

Status
Not open for further replies.
We have filed a U.L. Field Report and have requested that they observe these conditions. But, if the engineer works something out quickly with Cutler Hammer on switching the breakers, this condition may simply go away............
This DOES need more attention!
 
Strange that Eaton has changed their mind about comming out to check it out and then the engineer changes his mind...sounds like Eaton got ahold of him and paid him off...LOL
 
earshavewalls said:
Yes, they contain electronics and microprocessors that generate heat.
Click to expand...

More accurately, it's the power supplies they contain, that powers those microcontrollers, that generates the heat.
 
BarryO said:
More accurately, it's the power supplies they contain, that powers those microcontrollers, that generates the heat.
Click to expand...


I would say both.

Processors generally make their own heat as well.

Just check out the fan and heat sinks used in a PC.

Now I am sure the chip in an AFCI is not as powerful but it must make some heat.
 
Final action of CH

Final action of CH

Well, here we go.......Eaton (Cutler Hammer) sent a technician to the site of the problem. The technician agrees that this is an unacceptable situation. It appears that Cutler Hammer will be swapping out regular breakers for the arc fault breakers-----straight across.

I have been in contact with U.L. and they were preparing to send their field team to the site to run site testing, but now it appears that that will not be happening.

I am encouraging U.L. to run their own testing based upon the conditions as they were here to check the terminal temperature ratings and possible deration of conductors because of the ambient temperature levels in the panels.

Our problem will go away here, but look out for this in your neighborhood......
COMING SOON.......THE 2008 NEC.......LOCK AND LOAD!
 
Thanks ears. Please keep us posted on any new info. It looks like the AFCI saga is far from over.
 
What about thermal trip curves ?

What about thermal trip curves ?

earshavewalls said:
... The only potential problem mentioned on the phone to me was that, at these temperatures, the breakers COULD trip slightly sooner than they were designed to trip ...
Click to expand...
Just finished read thru this very interesting thread; and while the terminal temp ratings were discussed quite a bit, I was surprised that the issue of early tripping was only briefly mentioned.

I wonder how CH/Eaton defines "slightly" early ???

The current way thermal breaker trips are spec'd (i.e. the current v. time-to-trip curves) generally assume that the breaker starts out "cold" at 40 deg in "free-air" (providing some convective cooling) and then the load is applied.

The OPs breakers were starting out at well over twice that temp w/ little-to-no "free air" ... I wonder what the actual trip curves for his breakers would look like ... i.e. could a "40 deg" 20A breaker starting out at 86 deg even carry it's "rated" 20 amps ??? What about when the normal start-up currents are thrown on-top ???

Obviously the implications for a 2008 across-the board AFCI implementation (i.e. full panels of AFCIs) continues ... maybe we'll see a new NEC chart to "derate" multiple AFCIs within an enclosure similar to the derating scheme for CCCs within a raceway ... ?
 
Last edited:
Well, Nova, the CH/Eaton representative kept telling me that the breakers would still meet the UL listing requirements as long as they operated at 80% of their capacity.
I gave up on waiting to hear anything really constructive or truly informative from them a long time ago. CH has been dancing all around this issue and has yet to admit, directly, that they have NOT tested these breakers under these conditions. They continued their mantra of 'They meet UL'.
I hope someone with influence sees this string and maybe does something to have a bit more research done before we make the electrical manufacturers as rich as the oil tycoons. It's funny how FAST the whole arc fault thing is moving, isn't it? Shows what a lot of money and lobbying as well as some strategically placed CMP representatives will do for an industry.
 
Ears,
Yes, these are three-phase 120/208, 42 circuit panels with about 36 spaces taken up by AFCI breakers. They are 2-pole breakers but have been properly wired and the proper number of neutrals have been used and distributed properly.
Everything is correctly installed per the manufacturer's installation instructions and to Code.
Click to expand...
I wonder if the fact that these AFCIs are 2 pole AFCIs on a 3 phase system is part of the problem? With that application the neutral current does not cancel like in a 3 wire system. If both hot legs are loaded to 16 amps, the neutral will also be carrying 16 amps. There will be 1/3 more I^2R heating produced when this breaker is installed on a 3 phase system with balanced loads then when it is on a single phase system.
Don
 
A lot of questions arrise concerning the 'why' of this installation. Why are these breakers heating up SOOO much. The possibility of the neutral causing the heat is definitely a possibility, but the problem is:
THESE BREAKERS HAVE NOT BEEN TESTED UNDER THESE CONDITIONS.

How can a product be brought to market without sufficient testing? The same way drugs are approved by the FDA........with $$$$$$$?

I move for complete re-testing of all arc fault products!!!!

This time, take into consideration ALL of the potential problems, not just the obvious. That means, do ALL of the research needed to be certain these products can be used safely, in code compliant ways, without putting customers through the headaches that we are having with this product line now.
 
don_resqcapt19 said:
Ears,

I wonder if the fact that these AFCIs are 2 pole AFCIs on a 3 phase system is part of the problem? With that application the neutral current does not cancel like in a 3 wire system. If both hot legs are loaded to 16 amps, the neutral will also be carrying 16 amps. There will be 1/3 more I^2R heating produced when this breaker is installed on a 3 phase system with balanced loads then when it is on a single phase system.
Don
Click to expand...


Don, I was wondering if a 3 pole afci will be manufactured for 4-wire mwbc's. It seems like the design of the current breakers are for residential type dwellings (typical 3-wire services). The definition of a dwelling keeps getting broader and ends up covering commercial applications as well. These may have 3 phase service equipment.
 
yanici said:
The definition of a dwelling keeps getting broader and ends up covering commercial applications as well. These may have 3 phase service equipment.
Click to expand...

While many multi dwelling units do indeed have 3 phase services the panels for each unit are typically single phase.

But give the manufacturers some time, I am sure once the AFCI rules for dwellings are all inclusive we will see a push for non-dwelling unit AFCIs necessitating 3 pole AFCIs.:rolleyes:
 
earshavewalls said:
but the problem is:
THESE BREAKERS HAVE NOT BEEN TESTED UNDER THESE CONDITIONS.
Click to expand...

Actually we do not know that.

I am willing to bet they have been tested in many ways.

But finding an issue through testing only helps if you can do something about the problem.

They are trying to shoehorn 10 pounds of stuff in a 5 pound bucket.

It's all about the money
 
iwire said:
... I am willing to bet they have been tested in many ways ...
Click to expand...
I agree.

I'm still wondering how that CH rep determined that a 40-deg tested breaker might trip only a "little early" in 86 deg ambient conditions.

I looked thru CH's breaker trip curves and most -- if not all -- seem to be based on the standard 40 deg "cold start" test w/o providing any adjustment factors for ambient temp. (some did cite that the curves were valid up to 55 deg)

I did find one breaker model (not the same as the OP's) which stated that above 85 deg ambient it would automatically trip regardless of the actual load.
 
Last edited:
iwire

I also believe that the AFCIs have been tested in many configurations. But, I also believe that they have NOT been tested installed in panels as these have. Else, why would CH/Eaton be in such a hurry to change them out. It is logical that Eaton is covering a up a serious flaw in the AFCI overall designs, not just by them but by all of the mfgs. This flaw is well expounded by NoVA above. If the AFCI is already at 85 deg C and all the tests are based on a 40 deg C start temperature, then the test are flawed. IMO Also, the AFCIs at the top of the panel will be receiving the brunt of the heat and will, most probably, fail first. Another point is that UL did not send a evaluation team out. Why? Was it because UL also dropped the ball by not testing the AFCIs as they are not installed? All we can do is discuss this with our fellow electricians and factory reps. If the inspectors that lurk and post here would bring this up to their higher ups that could help as well. I agree with a previous poster that indicated that this product is being fast-tracked into the market on rocket sled power. Was there not a post on Mike's forum about a test that showed that a AFCI did not perform as specified even though the unit was alone and in excellent environment? Money is always the suspect.
.
 
Last edited:
Bayouamps said:
... then the test are flawed ...
Click to expand...
Thanks; I think I'd say that the 40-deg test was "invalid for the conditions."

While I completely understand how UL developed the "40 deg / free air" methodology as a way to accurately compare apples to oranges in a standardized way ...

... any testing regime should attempt to approximate the expected operating environment.

If new devices are introduced which significantly change that environment, it would seem that a new UL testing methodology should also be be developed which better approximates that new environment.

Jim
 
Status
Not open for further replies.
Top