help with heat problem

[Axkiker]

Hey all I have a problem

Im trying to develope a sealed lighting fixture which holds 6 k2 leds running at 1A.

So as you can see heat is an issue. I have been experimenting with different plastic composits that are developed to dissipate heat however none have been able to dissipate the amount of heat these little buggers produce. I have even gone as far as adding an aluminum heat sink inside the housing hopeing it would spread the heat throughout the case better. Still too hot. These housings must be as small as possible and I am unable to add external fins etc to help with heat.

Any ideas as to what I may be able to do.

 

[snarfer]

why can't you make the whole thing aluminum?

 

[Stage X Tuning]

carbon fiber composites work best. A CF heat sink works 2x better then an aluminum heat sink. and if you make the hole housing out of it you should be fine but with out knowing the size and thickness i cant say for sure.

 

[Axkiker]

why can't you make the whole thing aluminum?

That would be ideal but its not practical for mass production.

 

[Axkiker]

carbon fiber composites work best. A CF heat sink works 2x better then an aluminum heat sink. and if you make the hole housing out of it you should be fine but with out knowing the size and thickness i cant say for sure.

Tell me more about carbon fiber. I am familiar with carbon fiber being used kinda like fiberglass. However I didnt know it had good heat dissipation properties.

I have been using an epoxy with alot of aluminum filler to help dissipate the heat. Maybe adding carbon fiber or a carbon powder would help.

give me som insight.

 

[Illum]

you could drill a hole behind your enclosed fixture wide enough to fit an aluminum "stool" in place and thermal epoxy one side to your enclosed array and the other to an outside heatsink, which an occasional breeze should drop the heatsink temperature a bit.

can you provide a pic of your "fixture" so we can better understand how the whole thing is set up?

 

[R33E8]

Woah hold on before you waste money on a CF heatsink.. Where did you get this information from?

As far as I know, graphite is only highly thermally conductive "in plane" and has a very low thermal conductivity perpendicular to the plane. CF is just a bunch of woven strands of graphite (it's actually called something else which I can't remember right now..).. I doubt you could get good thermal conductivity even in plane with CF.. Also, the resin that holds it together adds more thermal resistance to the system.
Resource: http://hypertextbook.com/physics/thermal/conduction/
look at "carbon, graphite (⊥)"(out of plane) and "carbon, graphite (∥)" (in plane)

I would say your best bet is to go full copper or aluminum.. Copper would make little difference over aluminum but if you are running at the limit I guess every little bit will help.. What are the size constraints of fixture you are making? If it's too small then it would be virtually impossible to get the heat out fast enough without exotic materials (diamonds.. graphene.. carbon nanotubes.. ect..)

Have you thought of using what ever holds the fixture as a thermal path? Like if it's hanging from the ceiling, you could use a solid aluminum rod connected to the heat sink to help bring more heat away and it may look aesthetically pleasing.

 

[snarfer]

Carbon fiber will not help you. Although yes it is possible to construct an extremely efficient heatsink with carbon fiber, it requires pitch-based carbon fiber, which is extremely expensive and only used in aerospace applications. I actually obtained samples of this material and was not impressed. Thermal conduction per weight is very good. But it is more than 100 times more expensive than an equivalent aluminum solution, let alone plastic.

If you are interested, it is called also called pyrolytic carbon.

Another related solution used for electronics cooling is to use a graphite heat spreader. It is available in sheet form, with an adhesive on one side, like a sticker. Again it is very expensive and inferior to aluminum, because it conducts heat only in two dimensions.

Since evidently it is possible for you to use an aluminum heatsink inside the product, but not as part of the actual case, I would suggest that you take a look at the construction of mass-produced items with similar thermal characteristics.

One obvious item is the ubiquitous power brick. Many sealed power bricks, such as for laptop computer supplies, must dissipate about 18 watts, which is what your LEDs dissipate, without exceeding 85 C inside the case.

I recently took one apart and found that the interior of the plastic case was entirely lined with folded copper.

That said, I do not believe that it is impossible to make a mass-produced item with an aluminum, or partially aluminum, case. It would be ridiculous to attempt to document the number of examples of such mass-produced items.

 

[Axkiker]

Carbon fiber will not help you. Although yes it is possible to construct an extremely efficient heatsink with carbon fiber, it requires pitch-based carbon fiber, which is extremely expensive and only used in aerospace applications. I actually obtained samples of this material and was not impressed. Thermal conduction per weight is very good. But it is more than 100 times more expensive than an equivalent aluminum solution, let alone plastic.

If you are interested, it is called also called pyrolytic carbon.

Another related solution used for electronics cooling is to use a graphite heat spreader. It is available in sheet form, with an adhesive on one side, like a sticker. Again it is very expensive and inferior to aluminum, because it conducts heat only in two dimensions.

Since evidently it is possible for you to use an aluminum heatsink inside the product, but not as part of the actual case, I would suggest that you take a look at the construction of mass-produced items with similar thermal characteristics.

One obvious item is the ubiquitous power brick. Many sealed power bricks, such as for laptop computer supplies, must dissipate about 18 watts, which is what your LEDs dissipate, without exceeding 85 C inside the case.

I recently took one apart and found that the interior of the plastic case was entirely lined with folded copper.

That said, I do not believe that it is impossible to make a mass-produced item with an aluminum, or partially aluminum, case. It would be ridiculous to attempt to document the number of examples of such mass-produced items.

Hummm I have never heard of a power brick??? I will have to see what I can find and take a look. Maybe I can get some ideas.

You are right that it would be possible to mass produce the cases out of aluminum. The reason its not feasable for me is because it would take too long for me to manually machine them by hand. If I had a CNC outfit it would be possible but at that point probably too expensive.

I was screwing around last night and actually had great results. I have found a way to actually incorporate a larger heat sink inside the composite housing. This way the heat sink dissipates the heat though out the entire fixture and not just the bottem. I have a few more items to work out but I think I have my problem fixed.

 

[Stage X Tuning]

Woah hold on before you waste money on a CF heatsink.. Where did you get this information from?

As far as I know, graphite is only highly thermally conductive "in plane" and has a very low thermal conductivity perpendicular to the plane. CF is just a bunch of woven strands of graphite (it's actually called something else which I can't remember right now..).. I doubt you could get good thermal conductivity even in plane with CF.. Also, the resin that holds it together adds more thermal resistance to the system.
Resource: http://hypertextbook.com/physics/thermal/conduction/
look at "carbon, graphite (⊥)"(out of plane) and "carbon, graphite (∥)" (in plane)

I doing some experiments with CF heat sinks and you can buy special resin systems for heat sink applications. The heat sink are made in layers so that the its in plane with fins and there is not as much thermal resistance as you think. You come into contact with the CF more then the resin system. Will post photos of the heat sink later.

Carbon fiber will not help you. Although yes it is possible to construct an extremely efficient heatsink with carbon fiber, it requires pitch-based carbon fiber, which is extremely expensive and only used in aerospace applications. I actually obtained samples of this material and was not impressed. Thermal conduction per weight is very good. But it is more than 100 times more expensive than an equivalent aluminum solution, let alone plastic.

most people think it is expensive, but its not. i have both 2x2 Twill 3K Carbon Fiber and Plain weave 3K. im thinking of getting the 1K CF which is easier to use. ill get a photo of it getting torched.

 

[IMSabbel]

This kind of carbon fiber wont help at all.
The resin kills it.

Thermal resitance can be handled the same way electrical resistance is. Even if the conductivity of the fibres were infinity, the few 100 um of resin on top are such a high thermal resitance in series that the whole thing would be a lot worse than metal.

Also, kinda fail to see how a metal heatsink would be impossible. There are many ways to extract heat through a small crossection. Just recycling mass market cpu heatsinks available for a few bucks would be possible (just cut a square inset in the back of you fixture, but in solid part of heatsink, epoxy the gap.)

 

[Stage X Tuning]

This kind of carbon fiber wont help at all.
The resin kills it.

Thermal resitance can be handled the same way electrical resistance is. Even if the conductivity of the fibres were infinity, the few 100 um of resin on top are such a high thermal resitance in series that the whole thing would be a lot worse than metal.

Also, kinda fail to see how a metal heatsink would be impossible. There are many ways to extract heat through a small crossection. Just recycling mass market cpu heatsinks available for a few bucks would be possible (just cut a square inset in the back of you fixture, but in solid part of heatsink, epoxy the gap.)

Have you ever worked with CF? if not then dont speak on things you do not know about. They make resin systems that are thermally conductive. I would love to hear why "this kind of CF wont help at all."?

 

[Oznog]

I have worked with CF. I do a lot of work with pultruded graphite (CF) shafts. Those have continuous, linear fibers with a minimum of resin between them.

And I can tell you this, from experience- it's a very poor heat conductor down the shaft. I'd use hot-melt glue and you can accidentally heat the fibers to the point where the resin breaks down and it's not gonna get warm even 2" away. That's down the continuous fiber and the resin doesn't even enter into the picture in that direction. Those fibers are good thermal insulators not conductors.

Your plan sounds... wrong. A sealed plastic housing simply cannot dissipate heat. So sealing it, and making it plastic, was a bad idea not a good idea. Your specs are basically physically unworkable. Change the specs.

What you might also not be aware of is that any LED will become less efficient and suffer permanent lumen loss from operating at elevated temps. 6 K2's @1A/ea requires a LOT of performance heatsink to keep the die temps "good".

 

[Axkiker]

I have worked with CF. I do a lot of work with pultruded graphite (CF) shafts. Those have continuous, linear fibers with a minimum of resin between them.

And I can tell you this, from experience- it's a very poor heat conductor down the shaft. I'd use hot-melt glue and you can accidentally heat the fibers to the point where the resin breaks down and it's not gonna get warm even 2" away. That's down the continuous fiber and the resin doesn't even enter into the picture in that direction. Those fibers are good thermal insulators not conductors.

Your plan sounds... wrong. A sealed plastic housing simply cannot dissipate heat. So sealing it, and making it plastic, was a bad idea not a good idea. Your specs are basically physically unworkable. Change the specs.

What you might also not be aware of is that any LED will become less efficient and suffer permanent lumen loss from operating at elevated temps. 6 K2's @1A/ea requires a LOT of performance heatsink to keep the die temps "good".

The material I am working with is not a typical plastic. In fact its not really plastic at all I just used the "plastic" term as a general description since its what the stuff reminds me of.

Its more of a 2 part kinda epoxy which is designed to dissipate heat.

As of last night I had 6 k2s in a housing running and was able to hold my finger on the heatsink between a set of leds. Im still trying to fine tune the mix but right now everything is working out well.

 

[snarfer]

I held one end of the sample of high thermal conductivity carbon fiber over the open flame of a gas stove, and didn't feel the slightest warmth in the other end, three inches away. And this stuff was supposed to have higher thermal conductivity than copper.

Well what is this material you are using called and where do you get it?

The real test would be to put a thermocouple on the LED, and another on the heat spreading material, so you could see the temperature difference. Everything else is just guessing.

 

[Axkiker]

I held one end of the sample of high thermal conductivity carbon fiber over the open flame of a gas stove, and didn't feel the slightest warmth in the other end, three inches away. And this stuff was supposed to have higher thermal conductivity than copper.

Well what is this material you are using called and where do you get it?

The real test would be to put a thermocouple on the LED, and another on the heat spreading material, so you could see the temperature difference. Everything else is just guessing.

Yes I would love to put a thermocouple on the led. I am trying my hardest to find a way to accuratly check the junction temp. I just dont have the equipment.

so right now its the ole can I hold onto the led without burning myself. LOL


the material im using is simply a casting epoxy with alot of aluminum powder added to it. Im sure it doesnt dissipate heat as well as a true piece of aluminum but it does seem to do it fairly well. So well that the entire case heats up fairly uniformly and keeps all the leds running. However as you said until I find a way to check junction temp its all guessing

 

[Bimmerboy]

As of last night I had 6 k2s in a housing running and was able to hold my finger on the heatsink between a set of leds.

At 1A?

Unless I'm missing something, that sounds like the heat is not making it from the emitters to the heatsink. The K2's are fairly tough, but I'm not so sure they'll withstand these conditions for long.

 

[Axkiker]

At 1A?

Unless I'm missing something, that sounds like the heat is not making it from the emitters to the heatsink. The K2's are fairly tough, but I'm not so sure they'll withstand these conditions for long.

The housing which contains the heat sink is also water cooled..

 

[StevelKnievel]

carbon fiber composites work best. A CF heat sink works 2x better then an aluminum heat sink. and if you make the hole housing out of it you should be fine but with out knowing the size and thickness i cant say for sure.

I would love to see some quantitative data on the materials you're working with to make CF heatsinks. Can you post some data sheets or pics of a heatsink? What is the thermal resistance of the end product? There are a bunch of folks who would be very interested in the development of thermally conductive composites and if you have a solution that can come close to the conductivity of aluminum at a lighter weight or higher strength then let's hear about it. Data...Measurements...:thumbsup:

 

[R33E8]

The housing which contains the heat sink is also water cooled..

Haha well that makes a big difference.. Have you ever thought about casting or forging aluminium? If you are doing mass-production then it is very cost effective.. Just send it out to china and get it done for like $1 a piece..:thumbsup: