Star v Emitter, Rebel, et al.

[cadstarsucks]

seems to be, the only way to handle a rebels heat is in 3D ... that is, a heatsink with fins, yielding enormous surface area while providing a small footprint.

if the rebel emitter is rice-grain sized, we'll want to pack a lot of them in a small area.  trying to give 9 sq. in. of flat pcb space to each one would result in really low density, and a really big project.

Well I was going to work something up a bit ago but got off on other things.  I did find convenient copper bar stock at enco at $14 for 1/8"x1/2"x6'.  Combined with the proper interface to a heat sink that should work well.

As to the spring jig, I would not recommend it.  The Rebel is an alumina (ceramic) substrate with a soft silicone lens.

Dan

[minster9]

  I pretty much have to agree with DIY concerning the 3D heatsinks. I'm feeding a Lamina Atlas 5.7 watts that's mounted on a heat sink made for it by the same mfr. It's a radial design with fins radiating from a solid core about 3/4" dia.,extruded aluminum,it's about 2" dia.x 1.5" tall. It gets warm,but not hot enough to cause discomfort when laying back of finger on it. I'd like to have one of those digital infrared portable thermometers w/the built-in laser; be so handy when working with leds or especially computers. Cheapest I can find them is 70 bucks( a previously mentioned outfit has them on sale). If my quarterly bonus next week is decent,I might just buy one. Anyhow,when figuring surface area for finned heat sinks,does each face of a fin count as half of total area of the fin? And is heat dissipation reduced because each face shares the same material?

[cadstarsucks]

I pretty much have to agree with DIY concerning the 3D heatsinks. I'm feeding a Lamina Atlas 5.7 watts that's mounted on a heat sink made for it by the same mfr. It's a radial design with fins radiating from a solid core about 3/4" dia.,extruded aluminum,it's about 2" dia.x 1.5" tall. It gets warm,but not hot enough to cause discomfort when laying back of finger on it. I'd like to have one of those digital infrared portable thermometers w/the built-in laser; be so handy when working with leds or especially computers. Cheapest I can find them is 70 bucks( a previously mentioned outfit has them on sale). If my quarterly bonus next week is decent,I might just buy one. Anyhow,when figuring surface area for finned heat sinks,does each face of a fin count as half of total area of the fin? And is heat dissipation reduced because each face shares the same material?

There is no doubt that 3D heat sinks are best... the thermal resistance depends on surface area, the bulk metal getting to the surface area and the air flow over it.

http://www.elexp.com/tst_r100.htm with out the laser is $30... I could have sworn I saw one with for $20 but I do not remember where.

Dan

[cadstarsucks]

:)Thanks for clearing things up a bit,cadstarsucks. I think the star-type units offer a major advantage simply because their thermal pad is so much bigger & thicker to begin with. Using vias on sandwiched pc board to get to the heat sink on the other side,blech!

That is actually what they recommend, vias can be quite effective, particularly if they go through the wave afterward and get filled with solder.   

I might actually be getting them in house.   A customer is asking for a lot of light so I suggested rebels running at low power instead of what he asked for: putting 24 T1s in the package he now has 6 in. 

I am supposed to be getting the spec on the T1 LED he has and am expecting it is maybe 0.5Lm each for 12Lm total out of 24 LEDs.  If I put 0.25W into a string of 3 40Lm/W rebels I can 10Lm without heat sinking, which I have no space for anyhow.