Heatsink for microboost

[Pontiacg5]

What parts of the microboost driver actually crate the heat? It's the little black chip on the back next to the coil right? (I'm pretty sure this is a FET, am I right?)

 

[magonegro]

Hehe, I usually put my finger on and then switch on the circuit for some seconds, never had troubles detecting the hot chip with this advanced system :crackup:

edit: For sure you have done that sometime, just kidding, no offence...

 

[ped]

If you have any old xbox360's lying around, the RAM heasink pads on the underneath of the board are made of a heat conductive gel, which when pressed will mold itself to the various parts on the microboost, you can then heatsink the other side.

 

[Pontiacg5]

If you have any old xbox360's lying around, the RAM heasink pads on the underneath of the board are made of a heat conductive gel, which when pressed will mold itself to the various parts on the microboost, you can then heatsink the other side.

I've got that part worked out, but I just need to know where to put it. I'm almost 100% positive that little black 8 pin IC is a FET, and that is the thing all the current runs through so it is the part that gets hot. I just wanted some confirmation, but it looks like its hard to come by.

I guess I'll have to use the finger method!

 

[AJ_Dual]

I was just thinking of warapping the whole microboost in this:

Aavid Thermalloy - 161000F00000G - Chemicals & Adhesives - Chemicals - Allied Electronics Mobile

with a small cut-out for the pot so I can adjust it, and leaving the battery and diode contacts clear so I can re-solder connections if I ever need to.

Main thing right now is I'm on Kenom's Kryton Groove #3 group buy with a pressed in 445nm diode. I need to see the inside of the host before I can decide how to sink the driver.

I got that idea from this thread: http://laserpointerforums.com/f51/heatsinks-driver-52743.html

Although I see you participated in that one too.

 

[Pontiacg5]

There is really no reason to wrap the whole driver, the whole thing does not make heat, only one part does. Wrapping the whole thing with that stuff you linked to would probably be worse then leaving it open to the air inside the host.

There is not too much you could stick the driver to inside the groove, everything is round in there.

 

[AJ_Dual]

I see your point. Finger test, of what gets hot, then a gob of that thermal epoxy then I guess. :/

 

[Pontiacg5]

you need more than thermal epoxy, you need something to put on the other side of the epoxy.

I tested it tonight and was rather surprised. At 1A the little FET gets hot, but the pot does too. The coil and some resistors get warm, but nothing alarmingly warm.

I'll be making up a heatsink to sit over the FET at least, I think the pot should be ok if it gets warm.

 

[AJ_Dual]

you need more than thermal epoxy, you need something to put on the other side of the epoxy.

I tested it tonight and was rather surprised. At 1A the little FET gets hot, but the pot does too. The coil and some resistors get warm, but nothing alarmingly warm.

I'll be making up a heatsink to sit over the FET at least, I think the pot should be ok if it gets warm.

That's what I meant. Stick the other side of the epoxy to the body of the Kryton somewhere, or the back of the pill/head.

OTOH, I don't like the sound of something that permanent. So the sticky heat-pad tape/goo/foam used for RAM and IC's in computers might be better.