Overcurrent Protection in Cold Enviroments

[Tom Shock]

As an electrical contractor in WI and MN I have now had two cases of circuit breakers carrying
up to as much as twice the rated amperage. No, these breakers are not orange in color and are
sold by a reputable manufacturer. Both cases involve breaker panels located in un-heated
garages during very cold conditions. One of the cases I took an amperage reading of 26 amps
on a 15 amp breaker. The people were having a x-mas party with all the lights on including
outside x-mas decorations. Someone discovered an unusually warm outside facing sheetrock
wall and called the fire department.
Has any one else experienced any thing like this and how should this be addressed?

 

[jim dungar]

As an electrical contractor in WI and MN I have now had two cases of circuit breakers carrying
up to as much as twice the rated amperage. No, these breakers are not orange in color and are
sold by a reputable manufacturer. Both cases involve breaker panels located in un-heated
garages during very cold conditions. One of the cases I took an amperage reading of 26 amps
on a 15 amp breaker. The people were having a x-mas party with all the lights on including
outside x-mas decorations. Someone discovered an unusually warm outside facing sheetrock
wall and called the fire department.
Has any one else experienced any thing like this and how should this be addressed?

Breakers trip based on how hot they get. If they are installed in cold areas they can carry more load than if they were in a hot environment. Most manufacturers provide curves to adjust the current carrying capacity based on ambient (similar to the adjustment factor in the NEC for conductors).

this is the first one that came up using Bing.
http://static.schneider-electric.us/docs/Circuit Protection/0100DB0101.pdf

 

[Tom Shock]

The BIG picture

The BIG picture

I understand how breakers and fuses trip due to heat. My point is that up here in the "Frozen
Tundra" we have many cases where overcurrent protection is installed outdoors or in cold
enviroments i.e. rooftop equipment, solar combiner boxes, garage mounted panelboards, etc.
Has anyone ever really looked into this and studied the affects of overcurrent protection in
places where it gets quite cold. Maybe someone with some fire investigation experience would
have some insight.

 

[GoldDigger]

I understand how breakers and fuses trip due to heat. My point is that up here in the "Frozen
Tundra" we have many cases where overcurrent protection is installed outdoors or in cold
enviroments i.e. rooftop equipment, solar combiner boxes, garage mounted panelboards, etc.
Has anyone ever really looked into this and studied the affects of overcurrent protection in
places where it gets quite cold. Maybe someone with some fire investigation experience would
have some insight.

Looking at the specific graphs in the Square D link from jim, we see that the effect is there. Although maybe not a factor of two at least enough that combined with breaker-to-breaker variations and the normal tolerance of limited overcurrent, they might all add up to 26 amps on a 15 amp breaker.

The solution based on the Square D description is to use only electronic CBs in extreme temperatures instead of thermal magnetic.
One saving grace is that in those same cold conditions the ampacity, of exposed conductors at least, will also be increased.

 

[PetrosA]

...
One saving grace is that in those same cold conditions the ampacity, of exposed conductors at least, will also be increased.

That doesn't help if the panel's in an unheated garage and the wiring is in insulated spaces I would think the more rational option to using electronic breakers would be to mount the panel in a conditioned environment similar to the wiring it protects.

 

[Tom Shock]

Existings Circumstances

Existings Circumstances

Peter,
I get all of that. How do we deal with hundreds of thousands of existing installations where an
Overcurrent device is located in a tempurature environment that is much lower than the
Conductors are in?
Tom

 

[Smart $]

Peter,
I get all of that. How do we deal with hundreds of thousands of existing installations where an
Overcurrent device is located in a tempurature environment that is much lower than the
Conductors are in?
Tom

Install a thermostatically-controlled heater in the panelboard...???

 

[iwire]

Peter,
I get all of that. How do we deal with hundreds of thousands of existing installations where an
Overcurrent device is located in a tempurature environment that is much lower than the
Conductors are in?
Tom

There is nothing to deal with, this is not something new.

 

[kwired]

I would expect a 15 amp breaker to carry 26 amps for some time before tripping .

I would also expect 14 AWG to carry 26 amps for quite some time before reaching any dangerous temperatures. The 90 deg C rating of 14 AWG conductor is 25 amps, that is where we start applying adjustment factors to it.

What was the cause of the excess heat in the wall in the OP? Probably not a conductor alone, maybe a bad connection involved?

 

[Tom Shock]

Heat in Wall

Heat in Wall

Kwired,
The heat in the wall was from the "home run" 14-2 NM wire running through an insulated wall.
Because this outside wall was warm to the touch, one could assume that this breaker was
carrying a high level of amperage for a long time. Maybe up to half of the homes in the
Midwest have their panels in the garage and most of them are not heated. The other thing that
added to this situation is there was no "duct seal" in the short piece of conduit to the meter
socket backed up to this panelboard (many electricians forget how important this is and
inspectors can also miss it). The breaker was also located on the lowest position and only
had a breaker above it so heat dissapation was greater than would be if "sandwiched" by
other breakers. This is not an issolated case. I experienced almost the same thing this winter.
the 14 guage conductor from a 15 amp breaker(not orange) melted away it's insulating
covering and shorted to ground finally tripping this breaker. This panel was also in a cold
garage. I suspect this condition is more prevalent than we would like to admit. Thank you for
you responces. I will experiment with this and post my results. Maybe a youtube video!

 

[kwired]

Kwired,
The heat in the wall was from the "home run" 14-2 NM wire running through an insulated wall.
Because this outside wall was warm to the touch, one could assume that this breaker was
carrying a high level of amperage for a long time. Maybe up to half of the homes in the
Midwest have their panels in the garage and most of them are not heated. The other thing that
added to this situation is there was no "duct seal" in the short piece of conduit to the meter
socket backed up to this panelboard (many electricians forget how important this is and
inspectors can also miss it). The breaker was also located on the lowest position and only
had a breaker above it so heat dissapation was greater than would be if "sandwiched" by
other breakers. This is not an issolated case. I experienced almost the same thing this winter.
the 14 guage conductor from a 15 amp breaker(not orange) melted away it's insulating
covering and shorted to ground finally tripping this breaker. This panel was also in a cold
garage. I suspect this condition is more prevalent than we would like to admit. Thank you for
you responces. I will experiment with this and post my results. Maybe a youtube video!

I understand what you are getting at, but have a hard time believing you are getting conductors as hot as you say you are. If the conductor only gets hot for a few inches from a termination it is because of resistance in that termination, and not that the conductor was overloaded.

I also find it hard to believe that a 14 AWG NM cable carrying 26 amps and buried in typical construction would get hot enough to be noticed at the wall surface.

Will overcurrent devices work in cold environments, certainly. Will the response to light overloading be longer - most likely.

What about environments where we have overcurrent devices in moderate temperature and conductors in high ambient temperatures or the other way around?

Do circuit breakers compensate somehow for ambient temp? Without further research I kind of suspect they do, or else code would likely make us find ways of dealing with this. I have seen motor overloads that claim they do.

 

[iwire]

Do circuit breakers compensate somehow for ambient temp? Without further research I kind of suspect they do, or else code would likely make us find ways of dealing with this. I have seen motor overloads that claim they do.

No, they do not, they will trip earlier in hot ambient temps.

 

[A/A Fuel GTX]

So, the consensus is that placing OCPD's in a cold environment is not illegal but a poor design choice?

 

[iwire]

So, the consensus is that placing OCPD's in a cold environment is not illegal but a poor design choice?

That is not my postion, my postion is that it is nothing to worry about.

Here is a quote from the link Jim provided.

Square D circuit breakers can be applied in ambient temperatures within the range of -10?C to 60?C (14?F to 140?F). This document provides guidelines to follow when adjusting for ambient conditions.

There is so much margin of safety in the conductor ampacity tables I think it would be a very rare installation that could be a problem.

 

[jim dungar]

That is not my postion, my postion is that it is nothing to worry about.
...
There is so much margin of safety in the conductor ampacity tables I think it would be a very rare installation that could be a problem.

A circuit that is overloaded in the cool months will probably 'nuisance trip' in warmer months.

 

[PetrosA]

I'm not saying it's something to freak out about, but I do think it's worth considering at the design stage. 14 degrees F isn't that cold for some regions, and I could certainly imagine people using extra space heaters in lower temps and creating overload conditions in the winter that wouldn't exist in the summertime. I can also easily imagine a 14/2 NM buried in insulation getting pretty warm if carrying 26A for an extended period of time. It may not cause a fire, but it could degrade the insulation on the wire. Probably more of an issue in an older house with fewer circuits than a newer one with more circuits...

I guess I'll have to build an experiment now to test it

 

[templdl]

Peter,
I get all of that. How do we deal with hundreds of thousands of existing installations where an
Overcurrent device is located in a tempurature environment that is much lower than the
Conductors are in?
Tom

It's all a crap shoot. It has always amazed me how wire is protected when there are so many variables.UL489 listed breakers are calibrated at 40degC and often placed in 20-25degC environment. As such you start off behind the 8 ball to begin with and try to create the wires as required by the NEC as best as you can often based upon an educated guess of what you think the ambient temperature the the wire is run through which is entirely subjective in nature which is not as objective as the deratings for number of cables on a raceway which follows actual rules and after all of this place the breaker in an ambient temp which is far less than even 20degC with a breaker that is calibrated at 40degC?
At lower ambient temperatures it takes a greater current to deflect the bithermal element in a physically thermal magnetic breaker to trip the breaker. The suggestion of using an electronics trip breaker is a good one but electronic trip are not available in smaller frame breakers. As I recall even in larger frame breakers they may not be available less than 70a.
When you thing about it this only is an issue with OLs and does not involve short circuit protection which for the most part is not an issue.
But it is of my opinion that it is amazing that with all things considered we can protect wire from overloads as there are so many variables.

 

[kwired]

There is so much margin of safety in the conductor ampacity tables I think it would be a very rare installation that could be a problem.

I think that is the best thing to say on this issue.

The temp we design things around and actual temps reached are generally not the same

Run a 12AWG with a continuous 16 amp load on it and it doesn't get anywhere near 60 or 75 deg C, it may be warmer near the breaker, but remember the breaker does have thermal device within and it sinks some of its heat into the conductor. The conductor will warm up but if it reaches 75 deg C is hot enough you will not be able to touch it for very long without getting burned. 60 deg C you may be able to hold on to a little longer but still is pretty hot for human flesh.