Cooling LEDs?....Bigger volume of metal, or bigger surface of metal? 3

[grigson]

Hello,

We have four narrow rectangular LED PCB's, each one comprising three 1W LEDs. These are double sided PCB's with thermal vias to the bottom layer.

Each PCB is mounted on one of the four faces of a vertical, solid aluminium bar...so we have light shining North, East, South & West from the aluminium bar, so to speak.

Now, unfortunatley, the LED PCB's totally cover the vertical faces of the aluminium bar, so unfortunately the only bare metal surface of aluminium that can contribute to heat transfer by convection is the top 1 inch square part of this vertical aluminium bar.

As you can see, this big aluminium bar is mostly a waste of time, since most of its vertical sides are covered by the LED PCB's themselves....which impedes convection from the metal sides of the aluminium bar.

Anyway, do you believe that a better heatsinking solution would be to simply remove the solid , vertical aluminium bar, and have the four LED PCB's standing vertical, with their copper clad bottom layers providing the convection surface?

(...i.e more metal surface area but far less metal volume)

(The bottom copper is obviously covered in solder-mask, as it must be to prevent oxidation of the copper......do you think that the solder mask layer will ruin the effective convection of heat from the bottom of these PCB's.?)

Is there any other material (other than solder mask) that we can cover the bottom copper layer with to aid convection more?....what about covering it with solder?,

-eg as in some EMS's will cover bare copper pads with solder (as in HASL = Hot Air Solder Layering).




Anyway, to summarize, for cooling LEDs, what do you choose.......A Big volume of metal, or much, much smaller volume of metal but more metal surface area?

 

[zeusfaber]

Would usually think volume to soak up transient heat and area to dissipate steady state power.

A.

 

[OperaHouse]

What about a hollow tube.

 

[grigson]

Operahouse


.....the option of having four copper clad PCBs with no aluminium in the middle does effectively form a hollow metal tube......the metal being provided by the backs of the PCBs.

Otherwise, a hollow aluminium tube actually costs more than a solid one, since it takes more work to get the extrusion shape.

 

[VEBill]

As zeusfaber summarized...

An increased volume of metal provides increased thermal mass that can absorb increased amounts of heat for a given temperature rise. This is good for intermittent, low duty cycle, heat loads. For continuous duty cycle, thermal mass by itself only delays the inevitable.

Radiating the heat away occurs on the surface into free space or other sinks, and this mechanism thus requires both surface area and temperature rise. There will be other heat sink mechanisms such as convection.

Ideally, if this is a non-cheap product, then the controller could measure the temperature and self-protect itself (perhaps even with visual feedback such as flashing).

 

[zeusfaber]

Would your vertical bar look silly if you put a hat on it? The other thing massive bars are good for is conducting heat away to more effective radiators or convectors somewhere else.

A.

 

[grigson]

the product is enclosed in a plastic and perspex case, so thers no convenient "somewhere else" for the aluminium bar to conduct the heat to.


Our application is constant power, with no transients.

VE1BLL...

You appear to be saying what i am not brave enough to say.........that in fact the big bar of aluminium is , for us, a waste of time, and the four PCBs would be better off without it, and would be cooler if theyre just allowed to just convect heat away from there back copper planes?

It takes a brave person surely , to say that in this case, simply throwing out the aluminium bar is going to make the leds run cooler?

Should we throw away the aluminium bar?

 

[IRstuff]

What are the thermal environment specs for this? The bottom line for passive cooling for your 12W is 5314 in²°C. Assuming you want to keep the heatsink to ambient delta to 20°C, you'd need 266 in² of heat sink area. It's therefore clear from your description that no one has done any systems engineering on it, given that your aluminum bar only has about 13 in² of surface area at all.

TTFN
faq731-376

 

[MiketheEngineer]

Try what ever you think might help. Fairly cheap solution and will prove your theories one way or another. I am a big proponent of "Just try it" and see what happens.

 

[IRstuff]

I didn't see you enclosure posting when I posted. You're FUBAR. The fact that you've wrapped the thing with an insulator just compounded the problem. Just think about it this way; a 12-W candelabra bulb is too hot to touch, and that's with the bulb able to radiate 30% of the heat away. LEDs are supposedly more efficient. At the minimum, you'll need to run a fan to blow air through it.

TTFN
faq731-376

 

[MacGyverS2000]

[qu9ote grigson]It takes a brave person surely , to say that in this case, simply throwing out the aluminium bar is going to make the leds run cooler?[/quote]
No one said anything of the sort. From the sound of things, you have enclosed the entire shebang in plastic, so it doesn't matter if you attach the LEDs to aluminum or Jell-O. Eventually, the entire system will be limited in heat dissipation by the plastic enclosure. As mentioned earlier, when you first turn them on they'll run cooler, but eventually the bar will heat to ambient temp (i.e., what's inside the plastic), and then you're stuffed.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[MiketheEngineer]

Just like CFLs - just don't grab the base of these "cool" running light bulbs. I learned the hard way!!

GOD - I hate those things!!

 

[grigson]

its till worth convecting heat into the internal ambient, so that the warmed air can flow against the perspex enclosure and transfer heat to it, which will then conduct through the perspex, and then convect from the outer perspex surface to the external ambient.

So we need to get an internal convection current going, -i'm asking is the best way to do this to simply throw out the aluminium bar and let the surface areas of the backs of the led pcb's convect heat to the internal ambient?

 

[IRstuff]

Again, no, although it might be worthwhile making the aluminum bar hollow with fins on the interior. Are there ANY openings in the perspex? I think you're dreaming if you think your going to conduct all heat through the enclosure. You've essentially wrapped the assembly with thermal insulator.

What's the temperature rating of the LEDs?

TTFN
faq731-376

 

[MacGyverS2000]

I think I tired of this one earlier than prior threads... little to no critical physics thinking going on here in the design, and from the looks of the going path I don't expect that to change. I'm out...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[grigson]

LEDs temperature rated to 120C junction.....they are 1W LEDs as i said, but actually its flashing, so, in fact, each of the twelve leds is dissipating 0.6W.

We cant put holes in the enclosure as condensation would get in.

Our margins are too tight and we cant afford hollow aluminium bar.

I always thought for convection, surface area of bare metal was the most importnat thing.
-so why not remove the alu bar?.....if it is removed, there is more metal surface available for convection.

 

[OperaHouse]

It sounds like you have the boards already so you must have done a test and have actual trst data. You didn't like my hollow tube idea. The reason I mentioned this is the JUNCTION TEMPERATURE. You need something to get the heat away from the junction. Foil on a circuit board is not going to do that. If the heat can be transfered quickly from the junction the entire assembly can run realy hot and survive. Anyway, it seems like you already have this data or you wouldn't be here. It seems bad form not to have provided it.

 

[LionelHutz]

I saw the first problem within your first paragraph. Thermal pads (used when making thermal vias) are lousy way to tranfer heat.

the product is enclosed in a plastic and perspex case, so thers no convenient "somewhere else" to conduct the heat to.

Fixed it for you.

There's a reason why you commonly see high power LED bulbs with external heatsinks.

 

[IRstuff]

Your copper surface area is not as large as you think it is because it's too thin to properly spread the heat. Again, if you make the aluminum bar hollow, say, only 0.1" thick, you can get some more heat spreading.

It would have been substantially information had you mentioned the average power a trifle bit earlier. However, that's all the more reason to have the thermal mass and conductivity of the aluminum to help you. Aluminum is WAY more conductive than air, so running Cu-->air is worse than Cu-->Al-->Al-->air, given the useful area.

TTFN
faq731-376

 

[OperaHouse]

I just finished a little animation project that used 3W LED star mounts. Those are quarter size aluminum disk with ultra thin circuit board bonded to it. For proof of concept I just attached the disk to the case with hot melt glue. That little disk dispersed the heat so well I never went to a different bonding method.