10watt LED in a non IC rated can light

If I use a 10 watt LED in a non IC rated can (yes insulation will be covering the can) is this really an issue. This is for my own home and if I were ever to move I would remove the can lights. I have a sloped ceiling that is tight to my rafters and obviously has insulation. What I need is an IC rated, angle, remodel can but no one makes one. I want to install the NON IC rated angle can and use LED to solve the heat issue. Its my understanding that the only difference between IC and non IC rated is thier capacity to dissipate heat. In an IC rated can, the heat trip is set lower because it will be covered in insulation and wont dissipate heat as well. A NON IC rated can has a higher heat trip and can handle more heat( more wattage bulb) because it can dissipate from the top. I know what the UL approved method is. I know that if this was a costumers house I wouldn't do it. All That being said, will using LED ( that has no chance of ever getting hot) be safe. I know its not the right thing but will it be safe.......
 

GoldDigger

Moderator
Staff member
If I use a 10 watt LED in a non IC rated can (yes insulation will be covering the can) is this really an issue. This is for my own home and if I were ever to move I would remove the can lights. I have a sloped ceiling that is tight to my rafters and obviously has insulation. What I need is an IC rated, angle, remodel can but no one makes one. I want to install the NON IC rated angle can and use LED to solve the heat issue. Its my understanding that the only difference between IC and non IC rated is thier capacity to dissipate heat. In an IC rated can, the heat trip is set lower because it will be covered in insulation and wont dissipate heat as well. A NON IC rated can has a higher heat trip and can handle more heat( more wattage bulb) because it can dissipate from the top. I know what the UL approved method is. I know that if this was a costumers house I wouldn't do it. All That being said, will using LED ( that has no chance of ever getting hot) be safe. I know its not the right thing but will it be safe.......
Probably just as safe as putting no lamp in the fixture at all and turning the heat in the house up to 110F. :)
But I will never admit saying that.
 

gar

Senior Member
140318-1944 EDT

A 12 W LED is about equivalent to a 60 W incandescent. Take one of your cans and attach a temperature sensor to the can socket end. Pack a thick layer of insulation, 12" possibly, all around the can.

Put the LED in the socket and run until the temperature sensor stablizes. Record the temperture. Do the same with a 60 W incandescent. Is the LED test a safe temperature?

Some guesstimates. A 60 W incandescent radiates both visible and IR energy. But suppose 80% of the input power remains in the fixture. So that might be the equivalent of the heat from a 48 W resistor inside the can.

Assume the LED is much more efficient in light output vs input energy and only 50 % of the input energy remains in the can, or 6 W. This is 1/8 of wasted power in the can compared to the incandescent, and thus would result in about 1/8 of the temperature rise of the can surface.

Try the experiment and see what the results really are.

I have a 12 W Cree in an aluminum reflector with a couple vent holes. The socket area is about room temperature.

.
 
If I use a 10 watt LED in a non IC rated can (yes insulation will be covering the can) is this really an issue. This is for my own home and if I were ever to move I would remove the can lights. I have a sloped ceiling that is tight to my rafters and obviously has insulation. What I need is an IC rated, angle, remodel can but no one makes one. I want to install the NON IC rated angle can and use LED to solve the heat issue. Its my understanding that the only difference between IC and non IC rated is thier capacity to dissipate heat. In an IC rated can, the heat trip is set lower because it will be covered in insulation and wont dissipate heat as well. A NON IC rated can has a higher heat trip and can handle more heat( more wattage bulb) because it can dissipate from the top. I know what the UL approved method is. I know that if this was a costumers house I wouldn't do it. All That being said, will using LED ( that has no chance of ever getting hot) be safe. I know its not the right thing but will it be safe.......
I did just that the other day. Looks like ill be joining you down below for the afterlife ;) I always found it strange that light fixtures never give a larger allowance for a cfl or led bulb, makes no sense. Also strange that LEDs have these huge cast aluminum housings like they will be generating all this heat when they dont.
 

dfmischler

Senior Member
Location
Western NY
Also strange that LEDs have these huge cast aluminum housings like they will be generating all this heat when they dont.
The goal of the heatsink is to keep the junction temperature in the LEDs low. There is not as much heat overall as in an incandescent bulb, but there is enough heat in those little tiny junctions to melt them if the heat isn't spread out.

I have begun using the Cree A19 lamps. I see they now have a 1600 lumen bulb (AKA "100 watt equivalent") that takes 18 watts and has a 10 year warranty. I have contacted Cree about power factor on these bulbs but I haven't heard back yet.
 

gar

Senior Member
140319-0648 EDT

A Cree BA19 has a measured power input of 5.4 W at 123 V and a PF of 0.97 using a Kill-A-Watt EZ with 0.1 W resolution below 100 W. This bulb is rated 6 W and 50 mA at 120 V on its label. Power factor calculated from rating is 1.00 . The free air surface temperature of the finned heatsink area is about 130 F. with room ambient 70 F.

Even if no energy was going into radiated light there is no more than 5.4 W into the socket region.

.
 

kwired

Electron manager
Location
NE Nebraska
As has been mentioned the components of the LED are heat sensitive and must be kept below certain design temperature, just happens it is not the kind of heat levels that happen with an incandescent lamp.

Same with CFL's. Too much heat is going to cause damage to components, but that max heat level is lower than the operating temp of an equivalent incandescent lamp.

Why aren't some luminaires marked differently - is probably more complex than it seems it should be. The ability for air to circulate and get rid of heat easier I'm sure is a contributing factor. The luminaire likely doesn't heat up to a fire safety concern, but still is warm enough to affect performance of the installed lamp.
 

Electric-Light

Senior Member
Take a 10W resistor and a 10W light bulb. Let's say the light bulb gives off 0.5W in visible light while the LED gives off 2.5W

LED depends almost entirely on conduction and convection to dissipate heat. A light bulb on the other hand sheds a good portion of energy as infrared. This is why a 40W computer CPU needs a fan, but a 40W light bulb can release heat in an enclosed case.

What they call "thermal management" is a euphemism for limp mode. It keeps it from self destructing or severely reducing life(of power supply or LEDs) by throttling down the power which reduces the lumen output.

Trying to run at full power and lacking heat dissipation is the dominant cause of premature ballast failure, which is why you'll sometimes come across fixtures plagued with ballast burn outs. Incandescent fixtures are designed with incandescent in mind.

If you were to remove the reflective foil insulation, the heat will be absorbed by the fixture rather than being reflected and spat out the front and likely start a fire. The foil is not particularly effective at helping with heat removal for LEDs.
 
Last edited:

gar

Senior Member
140330-0914 EDT

green2012:

First, don't do what you suggest in your first post.

Second, do not use a screw-in LED in any can type (enclosed) fixture. LEDs and CFLs of the screw-in type are going to overheat in an enclosed space with no substantial air flow. The problem is worse than I might have imagined.

Results of several experiments.

The test bulb was a 10 W Utilitech 0424722 LED I previously mentioned. The can fixture was a cheap one from Lowes. Temperature was measured with a Simpson 389 thermistor instrument. The temperature sensor is about 1/4" dia. and 1" long. The sensor on the bulb was in a heat sink flute near the glass bulb. The can sensor was at the middle top of the can. The bulb sensor was held on with Scotch 33 tape and this would somewhat modify the heat transfer characteristics.

Ambient air temperature for all tests was 68 F. Adequate time was allowed, hours, for the measured temperature to reach steady-state conditions.

Can tests. The bulb was mounted in the adjustable position closest to the open end. The can is fully enclosed except for the one open end. 6" dia. and 8" tall.

Test 1.
The bulb was in an open socket in free air vertically oriented.
129 F on bulb.

Test 2.
Open end of can on bench, no insulation around the outside, and not much internal air flow within the can.
180 F on bulb, 91 F on can top.

Tests 3 and 4 can mounted vertically with open end down and not obscured.

Test 3.
3.5" fiberglass insulation around outside of can, but not over the top, and not uniformly tight around the can.
153 F on bulb, 86 F on can top.

Test 4.
3.5" fiberglass insulation laid loose over the top of test 3.
183 F on bulb, 120 F on can top.

If a glass diffuser was on the open end I would expect the LED temperature to be higher.

If you plan to use recessed or enclosed LED lights, then use fixture assemblies where the whole assembly including the LED has been designed as a package, and rated for the application.

.
 

gar

Senior Member
140330-2136 EDT

Some additional tests.

In an effort to reduce my electrical energy use I have been using more local lighting in place of my 8' Slimlines. One way I use the low power bulbs is to buy a socket and cord with an 8" aluminum reflector from Home Depot, and locate this somewhere near my work spot. This is less illumination, but workable.

New tests, same bulb and can.

Test 5.
Temperature rise of LED in the 8" reflector. There are three 3/8" holes near the socket for ventilation.
145 F for the bulb heatsink temperature. This compares with 129 F when the bulb was completely open with the same socket but no reflector.

Test 6.
Back to the can fixture with no insulation and the open end open and pointing down.
Bulb is a 60 W incandescent.
116 F on the can outside top. This is a rise of 116-68 = 48 F over ambient.

Test 7.
Same as 6 except the bulb is the 10 W LED.
83 F on the can outside top. This is a rise of 83-68 = 15 F over ambient.
167 F on bulb heat sink.

The ratio of the outside temperature rises is 48/15 = 3.2 .
The ratio of the input powers is 60/10 = 6.0 .

Over a limited range one can approximate the temperature rise by the ratio of power input. However, as the temperature rise is greater the cooling effect should improve, and thus a linear assumption is not a precise solution. The temperature rise ratio with the incandescent is much lower than predicted by the power input ratio. With more power input there may be better heat transfer. To test this I need to put 60 W into resistors inside the can. Resistors are at a much lower temperature than an incandescent bulb filament and therefore will radiate less infra-red energy. This test will need to wait for another day. As Electric-Light pointed out an incandescent radiates a lot of its input power as infra-red radiation.

.
 

Sierrasparky

Senior Member
Location
USA
Occupation
Electrician ,contractor
If I use a 10 watt LED in a non IC rated can (yes insulation will be covering the can) is this really an issue. This is for my own home and if I were ever to move I would remove the can lights. I have a sloped ceiling that is tight to my rafters and obviously has insulation. What I need is an IC rated, angle, remodel can but no one makes one. I want to install the NON IC rated angle can and use LED to solve the heat issue. Its my understanding that the only difference between IC and non IC rated is thier capacity to dissipate heat. In an IC rated can, the heat trip is set lower because it will be covered in insulation and wont dissipate heat as well. A NON IC rated can has a higher heat trip and can handle more heat( more wattage bulb) because it can dissipate from the top. I know what the UL approved method is. I know that if this was a costumers house I wouldn't do it. All That being said, will using LED ( that has no chance of ever getting hot) be safe. I know its not the right thing but will it be safe.......

You probably already did your installation however there are slope ceiling IC airtight cans. Elco makes them and there are some others.
 

mwm1752

Senior Member
Location
Aspen, Colo
If you were to follow building/energy code Airtite would be included due to your install in the thermal envelope. Inherently Protected marked luminaires area also IC rated. I could never pass the installation and in your own home is your responsibility especially doing work without inspections. The NEC relies on workmanship to be upheld by professionals doing non inspection required type work. This question can only be answer by you IMHO
 

Fulthrotl

~~Please excuse the mess. Sig under construction~~
If I use a 10 watt LED in a non IC rated can (yes insulation will be covering the can) is this really an issue. This is for my own home and if I were ever to move I would remove the can lights. I have a sloped ceiling that is tight to my rafters and obviously has insulation. What I need is an IC rated, angle, remodel can but no one makes one. I want to install the NON IC rated angle can and use LED to solve the heat issue. Its my understanding that the only difference between IC and non IC rated is thier capacity to dissipate heat. In an IC rated can, the heat trip is set lower because it will be covered in insulation and wont dissipate heat as well. A NON IC rated can has a higher heat trip and can handle more heat( more wattage bulb) because it can dissipate from the top. I know what the UL approved method is. I know that if this was a costumers house I wouldn't do it. All That being said, will using LED ( that has no chance of ever getting hot) be safe. I know its not the right thing but will it be safe.......
well, you could cut a hole for the light, and use a stick to push the insulation away far enough
to meet the listing requirements, yes?

when i ran can lights in my house, i used all 4", with airtight IC rated housings, but as i
had ceiling access, it wasn't an issue. i've got some old first generation halo architectural
grade 6" LED's that are listed to put into any can, in my garage. massive heat sinks,
take up the entire volume of the housing, but they are for a flat lid.

i believe elite has a CFL angle rough in that is IC rated, in their commercial line.
i put 4 of them in a lobby a little while ago, in a 4/12 pitch roof.
 

broadgage

Senior Member
Location
London, England
Use of an LED retrofit lamp as described in the O/P can not possibly be a fire risk, with a total loading of only 10 watts spread over a significant surface area there is simply not enough heat input to raise the fixture or surrounding woodwork etc. to a dangerous temperature.

It is however entirely possible that the LED retrofit lamp will get hotter than the design temperature and therefore suffer premature failure. The lamp wont catch fire as they are made of non flammable materials and incorporate a thermal fuse, but it may have a reduced life.

In practice the room temperature, operating cycle and qaulity of the thermal design within the lamp will determine if this reduction in life be trivial or significant.
 
Last edited:

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Electrical Contractor
We install LED 10.5 and 11 watt bulbs all the time in IC cans. The only difference in the non ic can is a different setting for the thermal protection. I agree that fire is not an issue.

Use an IC can with a regressed eyeball trim. I see no reason to install non IC cans in that situation-- not worth worry about
 

Electric-Light

Senior Member
big problem industry wide for CFLs and LEDs is that they aren't cut out to stay cool enough for good life for the applications they market for. LED fixtures or fixture like drop ins tend to be more reliable as it is designed to handle its own heat while pressed up against basically con-conducting surface
 

Electric-Light

Senior Member
Something like CREE CR6 is closer to a partial fixture than a "bulb". It is very different.

LED "bulbs" do not have any thermal connection at the front due to design limitations and the type of fixture dictates their life. CREE CR6 looks like a cowboy hat with a socket on the top.

The inner walls of the hat and the brim makes a very nice looking, gap free trim. The inner sidewall is rather tall at about 1 1/2" and the brim is about an inch wide. The overwhelming difference is that this hat is not just a reflector. It is actually made of aluminum and wraps around to the back and solidly bonds to LED heat sink. So even if the parts not visible to the eyes are fully insulated, the white reflector and brim has full contact with ambient air.

Many know I'm far from an LED advocate. The CR6 however is a bit different. It's closer to an engineered product than a drop-in bulb and it has a low flicker comparable to high quality fluorescent ballast and has a luminaire efficacy of 60-65 lm/W. This is unthinkable with CFLs. Luminiaire efficacy is in the range of 30-40 lm/W with CFL cans due to lamps only getting only 65-70lm/W and utilization factor of the fixture which drops the luminaire efficacy that low.

Cans are much more troublesome in commercial lighting than linear lamps due to concentrated heat from lamps and they're plagued with ballast failures. They're basically F32T8 2 lamp fixtures condensed into a can size and the ballast do not have a nice big heat plate like in a linear system. Quad and triple lamps are generally only rated at 10,000 to 12,000 hours which is 1/3 to 1/4 of T8
 

kwired

Electron manager
Location
NE Nebraska
big problem industry wide for CFLs and LEDs is that they aren't cut out to stay cool enough for good life for the applications they market for. LED fixtures or fixture like drop ins tend to be more reliable as it is designed to handle its own heat while pressed up against basically con-conducting surface
And the price for the ones designed for the application make consumers take the chance on the lesser priced unit. They usually last long enough that any estimated life is forgotten about when they do fail and they are simply replaced with no questions asked.

If you have seen the cost of the replacement halogens that are still acceptable to the DOE they are high enough to make one seriously consider CFL's or LED's even if they will only last for half their estimated life.
 
Top