CFL & Fire

Status
Not open for further replies.

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
50323-1746 EDT

Can you provide more information? For example: location, orientation, what in the CFL started the fire, how did the bulb fire propagate from the bulb to the attic.

.
 

czars

Czars
Location
West Melbourne, FL
Occupation
Florida Certified Electrical Contractor
The forensic fire investigator spent 6 hours investigating the cause of the fire. Here is the situation.

The fire appeared to start in the vicinity of the CFL attic light. No other electrical (AC or battery) equipment, combustible material or wiring was in the same area except for the attic light switch and the power cable to the switch. The attic light switch was mounted on a 4 square box below the light. The light was mounted via a porcelain socket on a standard hex ceiling box. The ceiling box was mounted, so the the CFL was horizontal, on a vertical frame member about 20" below the roof.

The investigation revealed that all the wiring was proper. It was presumed that the switch was in the off position and the the CFL had not been inadvertently left on. For some reason, the switch apparently failed and provided power to the CFL. There was no evidence of fire in the 4 square switch box, but there was evidence of overheating on the conductors where they connected to the switch. The switch itself was consumed. There was no evidence of fire in the hex ceiling box and the CFL and socket were also destroyed in the fire. It was concluded that the CFL caused the fire since there was nothing else in the vicinity that could.
 

Strathead

Senior Member
Location
Ocala, Florida, USA
Occupation
Electrician/Estimator/Project Manager/Superintendent
The forensic fire investigator spent 6 hours investigating the cause of the fire. Here is the situation.

The fire appeared to start in the vicinity of the CFL attic light. No other electrical (AC or battery) equipment, combustible material or wiring was in the same area except for the attic light switch and the power cable to the switch. The attic light switch was mounted on a 4 square box below the light. The light was mounted via a porcelain socket on a standard hex ceiling box. The ceiling box was mounted, so the the CFL was horizontal, on a vertical frame member about 20" below the roof.

The investigation revealed that all the wiring was proper. It was presumed that the switch was in the off position and the the CFL had not been inadvertently left on. For some reason, the switch apparently failed and provided power to the CFL. There was no evidence of fire in the 4 square switch box, but there was evidence of overheating on the conductors where they connected to the switch. The switch itself was consumed. There was no evidence of fire in the hex ceiling box and the CFL and socket were also destroyed in the fire. It was concluded that the CFL caused the fire since there was nothing else in the vicinity that could.


Would an arc fault breaker have prevented it? I ask because I am not a fan of them but am continuously trying to find justification.
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
What is there inside a light switch that could fail in such a way to send power to the light?

I sense a strong desire on the part of the investigator to close this out, resulting in some serious hand-waving as to root cause.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
What is there inside a light switch that could fail in such a way to send power to the light?
Well, a snap switch could fail to open when been turned off one last time. :)
A failure that causes continuity when actually starting from the open position is harder to imagine.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Well, a snap switch could fail to open when been turned off one last time. :)
A failure that causes continuity when actually starting from the open position is harder to imagine.
Is it possible that the last time the switch was turned off it retained something resistive across the contacts which became a glowing series fault and started the fire? I know that's a reach...
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
Why would they "presume" the switch was in the off position? It would be more logical to presume it was on.

I have seen a CFL suddenly fail with a large flash, a puff of smoke, and maybe even a bang.
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
My first hunch would be the switch was inadvertently left "on" and that a loose connection on the switch or within the switch was victim to the high inrush currents of the cfl ballast. Untill it's heat gain was large enough to ignite nearby combustibles.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150325-1734 EDT

Suppose the CFL was a 20 W device. If the CFL was undamaged and as approximation was considered something of a resistive load, and at the switch there was a variable resistance as a loose connection. then the maximum power that could be dissipated at the switch would be 10 W. In a metal box and probably in a plastic box this would never start a fire.

I would suspect much higher currents as a result of a short in the CFL, and that the switch had been left on.

.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
My first hunch would be the switch was inadvertently left "on" and that a loose connection on the switch or within the switch was victim to the high inrush currents of the cfl ballast. Untill it's heat gain was large enough to ignite nearby combustibles.

I agree with the switch being left on, but CFL inrush would be a fairly low current for a fairly short duration.

I don't quite understand how the switch, CFL lamp, and socket can be "consumed" or "destroyed", with no evidence of fire inside the boxes.

My best guess would be the CFL ballast shorted. I would assume the CFL would have some type of internal overcurrent protection, but that may have failed in a shorted mode - possibly from the heat of being left on in a hot attic. Then, apparently the current was high enough to start a fire by heating at the points of highest resistance - the switch contacts, and either the lamp ballast itself, or maybe the lamp-socket connection.

Yet, the current was low enough not to trip the branch breaker, or the branch breaker is also bad. IMO, the investigator should have tested the branch breaker also.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
I agree with the switch being left on, but CFL inrush would be a fairly low current for a fairly short duration.

I don't quite understand how the switch, CFL lamp, and socket can be "consumed" or "destroyed", with no evidence of fire inside the boxes.

My best guess would be the CFL ballast shorted. I would assume the CFL would have some type of internal overcurrent protection, but that may have failed in a shorted mode - possibly from the heat of being left on in a hot attic. Then, apparently the current was high enough to start a fire by heating at the points of highest resistance - the switch contacts, and either the lamp ballast itself, or maybe the lamp-socket connection.

Yet, the current was low enough not to trip the branch breaker, or the branch breaker is also bad. IMO, the investigator should have tested the branch breaker also.
FWIW, I had a CFL in a lamp that started flickering and making a zzzzzttt noise. I turned it off and reached in to unscrew it, and I burned my hand on the base. That incident and this story is causing me to have concerns about the CFLs I have in recessed cans in the ceiling. CFLs in general have not (for me, anyway) lasted anywhere near as long as they are touted to.
 

gadfly56

Senior Member
Location
New Jersey
Occupation
Professional Engineer, Fire & Life Safety
FWIW, I had a CFL in a lamp that started flickering and making a zzzzzttt noise. I turned it off and reached in to unscrew it, and I burned my hand on the base. That incident and this story is causing me to have concerns about the CFLs I have in recessed cans in the ceiling. CFLs in general have not (for me, anyway) lasted anywhere near as long as they are touted to.

Double thumbs up on that!:thumbsup::thumbsup: You may note that the lifetimes touted for CFL's have dropped from 23,000 hours to 4,000 hours. I eagerly wait to see what the real-life results will be for LED's.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Double thumbs up on that!:thumbsup::thumbsup: You may note that the lifetimes touted for CFL's have dropped from 23,000 hours to 4,000 hours. I eagerly wait to see what the real-life results will be for LED's.

The LED's I have installed are holding up well so far.
 

Sierrasparky

Senior Member
Location
USA
Occupation
Electrician ,contractor
Double thumbs up on that!:thumbsup::thumbsup: You may note that the lifetimes touted for CFL's have dropped from 23,000 hours to 4,000 hours. I eagerly wait to see what the real-life results will be for LED's.

Where do you see this reduction in claimed Life?
 
Status
Not open for further replies.
Top