Looking for help with improving chip resistor thermal management

[Siwastaja]

I don't see how the heatsinks will fall off the thermal pads.

They will, thermal pad material seems surprisingly sticky, but it isn't glue. Unless you actually brought one with glue, but the datasheet would spell it out clearly.

Thermally conductive glue is available, though.

[gjpmho]

I don't see how the heatsinks will fall off the thermal pads.

They will, thermal pad material seems surprisingly sticky, but it isn't glue. Unless you actually brought one with glue, but the datasheet would spell it out clearly.

Thermally conductive glue is available, though.

How can you be that certain? The heatsinks are on a flat and level surface... The datasheet doesn't mention mounting. https://www.tglobaltechnology.com/wp-content/uploads/2019/07/TG-A6200-4.pdf

I contacted T-Global, the manufacturer, and they say it should be OK as long as I don't move the amp. 

I did try the same setup on the phono preamp PCB which is oriented vertically. The thermal pad is sticky enough to hold the heatsink at a 90 degree inclination. In that case, it is obvious mounting the heatsink is necessary.

[Siwastaja]

Part number of the exact thermal pad? Some are adhesive and suitable for mounting of lightweight heatsinks. It seems you have bought this, based on the manufacturer response.

If you got that kind of stuff by accident, then it was good luck, keep buying the same. 

If it is the normal stuff and currently gravity is keeping it in place, it will be OK if you keep the thing upright and move it carefully. But it's quite an assumption! At some point, someone will handle it more roughly.

Having some mounting pressure is beneficial for good thermal contact, though. If only lightly touching, gaps of air can remain.

[gjpmho]

Part number of the exact thermal pad? Some are adhesive and suitable for mounting of lightweight heatsinks. It seems you have bought this, based on the manufacturer response.

If you got that kind of stuff by accident, then it was good luck, keep buying the same. 

If it is the normal stuff and currently gravity is keeping it in place, it will be OK if you keep the thing upright and move it carefully. But it's quite an assumption! At some point, someone will handle it more roughly.

Having some mounting pressure is beneficial for good thermal contact, though. If only lightly touching, gaps of air can remain.

I thought I added the exact part in an earlier post:
https://www.digikey.com/en/products/detail/t-global-technology/TG-A6200-30-30-1-0/11201361

As stated, I am not moving or selling the amp. Mounting the heatsinks is too difficult IMO. I don't have enough clearance and don't want to venture drilling holes in the PCB. It is duly noted that this arrangement isn't ideal, and I will check on the heatsinks if/when I move the amp in the future. I am the sole owner.

[Siwastaja]

Sorry I missed the link earlier.

Note that the material is rated to be used with at least 10psi of mounting pressure, and thermal conductivity will improve with more pressure.

A randomly picked heatsink ( https://www.cuidevices.com/product/resource/hsb09-212115.pdf ) with mounting surface area of 21x21mm = 0.68 square inches, weighs 10.1g = 0.0223 pounds. By gravity mounting, this is just 0.033 psi.

Anyways, it's just fine for carefully handled one-off, and even if thermal conductivity suffers, it's still likely a significant improvement. Me too have temporarily used thermal pads during prototyping without actual mounting.

[gjpmho]

Sorry I missed the link earlier.

Note that the material is rated to be used with at least 10psi of mounting pressure, and thermal conductivity will improve with more pressure.

A randomly picked heatsink ( https://www.cuidevices.com/product/resource/hsb09-212115.pdf ) with mounting surface area of 21x21mm = 0.68 square inches, weighs 10.1g = 0.0223 pounds. By gravity mounting, this is just 0.033 psi.

Anyways, it's just fine for carefully handled one-off, and even if thermal conductivity suffers, it's still likely a significant improvement. Me too have temporarily used thermal pads during prototyping without actual mounting.

Thanks, I missed that part about 10psi mounting pressured. I am surprised how well it works being under pressured. I also calculated a very similar value: 0.02 psi. Using 8x8x8mm^3 cube as the heatsink and finding the mass (~1 g) based on the density of aluminum - 2.7 g/cm^3.  surface area is 0.1 sq inches.

[gjpmho]

I didn't want to worry about the heatsinks falling off or the thermal pads going bad and shorting the resistors.
 A user on diyaudio recommended placing the resistors on a perf board mounted to the device output heatsinks! I also replaced several of the surrounding components. As it turns out, a couple of the MLCC terminals fell off as I was desoldering them. The excessive heat was wearing down components! Also DC offset is more stable. Thanks again for all the help. Moving the resistors as close possible to the vented lid seems like a solid solution. Here's the photo.