best coatings for heatsink to radiate heat

mds82 · Jan 6, 2014

Hi All, I'm not sure if this has been covered already so i wanted to start a new thread about it. I've been doing a lot of research on different coatings to aluminum heat sinks to help radiate heat to keep temperatures as low as possible.

I have been running tests with a 4 XP-G's, 700ma each on a small cpu style heat sink with just passive heat sinking, no fan. So far i have been able to run 3 tests, identical configurtions but with different coatings on the aluminum heatsink

1 - Bare aluminum - highest temperature, 200F
2 - grey anodized - 2nd best, 183f
3 - Avian Black-S - best at 178 F

What i am looking to find out is what other coating can i put on the metal heatsink ( without changing anything else) to help passively radiate out heat and keep temperatures the lowest.

Steve K · Jan 6, 2014

Isn't a flat black surface associated with the best radiation? That seems to be what commercial heatsinks use too.

Anodization seems pretty popular with heatsink manufacturers too, partly because it forms an insulating layer, which is often beneficial. i.e. the transistor or IC case is usually connected to one of the terminals, so the anodized layer lets you keep the voltage off of the heatsink (and whatever the heatsink is bolted to) without having to add a separate insulating layer between the transistor and heatsink. The insulating layer just adds to the thermal resistance and reduces the effectiveness of the heatsink.

mds82 · Jan 6, 2014

Well a "black body" is best, but that doesn't necessarily mean a flat black surface. the Avian Black-S is by far the blackest thing i have ever seen and it does offer better temperatures compared to the anodized piece. I was reading that Kyrlon flat White paint also does a great job. i think i will run a test with that soon to.

monkeyboy · Jan 6, 2014

God job on doing this.

I seem to remember an old thread by CPF member newbie who performed similar experiments and came to the same conclusions. The more tests the better.

DIWdiver · Jan 6, 2014

Well, there's black, and then there's black. What you want is black in the IR region where heat would be radiated. Anodization doesn't have an inherent color; the color is added with dyes. You want a dye that's black in the IR region. That's not necessarily the same as black in the visible region. I know that an iodine solution can be nearly black to the eye, but nearly transparent to IR. So you can't tell by looking at a heatsink if it has a good surface finish.

The oxide has poor conductivity, both electrically and thermally. But it's so thin and easily damaged that it is not generally considered a reliable electrical insulator. You'd have to anodize and handle the part very carefully, or use a very thick and tough anodizing, to make it useful for electrical insulation. That would increase costs and thermal resistance. In extreme cases, you want to remove the oxide from the thermal path from your source into the heatsink to minimize thermal resistance.

I read somewhere that the amount of heat radiated from a heatsink is pretty small compared to the amount that is transferred by convection. I don't know how true that is, so test results like this are great. It appears from your tests that radiation can account for at least 1/6th of the total dissipation in a passive system.

Another thing to consider in radiation is visibility. With two parallel fins with small separation, much of the energy radiated from one fin will impinge upon the other, and get re-absorbed. In this case much of what's 'visible' to one fin is the other fin. If the two fins are separated, either by distance or by angle, much more 'cold background' is visible to each fin. This improves heat dissipation, not because radiation is increased, but because re-absorption is decreased.

Pinarello · Feb 23, 2017

Newby flasholic with a lot of questions, of course. Speaking about various coatings, I wonder whether the modern ceramics could be used. Ceramic made great passes in last few decades. Also, can ceramic be used in other parts of the flashlight, for a heat dissipation f.e. Or I am not enlightened yet?
Flasholic "...long as I can see the light..."

ssanasisredna · Feb 23, 2017

mds82 said:
Hi All, I'm not sure if this has been covered already so i wanted to start a new thread about it. I've been doing a lot of research on different coatings to aluminum heat sinks to help radiate heat to keep temperatures as low as possible.

I have been running tests with a 4 XP-G's, 700ma each on a small cpu style heat sink with just passive heat sinking, no fan. So far i have been able to run 3 tests, identical configurtions but with different coatings on the aluminum heatsink

1 - Bare aluminum - highest temperature, 200F
2 - grey anodized - 2nd best, 183f
3 - Avian Black-S - best at 178 F

What i am looking to find out is what other coating can i put on the metal heatsink ( without changing anything else) to help passively radiate out heat and keep temperatures the lowest.

Practically most basic paints are almost as good as specialized coatings with emissivity of 0.93 and higher common. There are many charts on the web with emissivity numbers.

DIWDiver, how important radiation is all comes down to:

- Area of the emissive surface
- How hot the emissive surface is w.r.t. the surrounding (Note, when outside, up is often very cold)
- ... and those two above in combination with how good conductive is.

In real world fixtures and bulbs, I have seen as high as 60+% radiative, and as low as 10% under typical operating conditions, with the trend to higher with better high heat handling on LEDs but also large surface areas for interior ambient lighting.

snakebite · Feb 24, 2017

the best i have tested was flat black.
helps with convection too as the rougher surface has more surface area too.like adding more fins but not that dramatic.
the best heatsink for its size in my altilon desklamp mods is painted flat black and has angled,long,widely spaced fins.
its a fanless sink from a gpu.built for a htpc where you dont want fan noise.

ssanasisredna said:
Practically most basic paints are almost as good as specialized coatings with emissivity of 0.93 and higher common. There are many charts on the web with emissivity numbers.

DIWDiver, how important radiation is all comes down to:

- Area of the emissive surface
- How hot the emissive surface is w.r.t. the surrounding (Note, when outside, up is often very cold)
- ... and those two above in combination with how good conductive is.

In real world fixtures and bulbs, I have seen as high as 60+% radiative, and as low as 10% under typical operating conditions, with the trend to higher with better high heat handling on LEDs but also large surface areas for interior ambient lighting.

degarb · Feb 25, 2017

DIWdiver said:
Well, there's black, and then there's black. What you want is black in the IR region where heat would be radiated. Anodization doesn't have an inherent color; the color is added with dyes. You want a dye that's black in the IR region. That's not necessarily the same as black in the visible region. I know that an iodine solution can be nearly black to the eye, but nearly transparent to IR. So you can't tell by looking at a heatsink if it has a good surface finish.

The oxide has poor conductivity, both electrically and thermally. But it's so thin and easily damaged that it is not generally considered a reliable electrical insulator. You'd have to anodize and handle the part very carefully, or use a very thick and tough anodizing, to make it useful for electrical insulation. That would increase costs and thermal resistance. In extreme cases, you want to remove the oxide from the thermal path from your source into the heatsink to minimize thermal resistance.

I read somewhere that the amount of heat radiated from a heatsink is pretty small compared to the amount that is transferred by convection. I don't know how true that is, so test results like this are great. It appears from your tests that radiation can account for at least 1/6th of the total dissipation in a passive system.

Another thing to consider in radiation is visibility. With two parallel fins with small separation, much of the energy radiated from one fin will impinge upon the other, and get re-absorbed. In this case much of what's 'visible' to one fin is the other fin. If the two fins are separated, either by distance or by angle, much more 'cold background' is visible to each fin. This improves heat dissipation, not because radiation is increased, but because re-absorption is decreased.

This is beautiful information. You covered so many questions, normally overlooked, a tear trickled from the corner of my eye.

I have one headlamp I tried to solder a Jameco AL avid heatsink (for the bezel, being hand shaped, with fingers to hold the aluminum reflector and lens) to a cpu style head sink (attached to bolt for pivot) , burning off the anodized finish, and failing to get a good connection, before thermal epoxy as before. I have been tempted to spray it with flat high temp, but figure I will choose the wrong can of spray paint. Your post only lack a good, attainable product recommendation. Also will need to think and Google more about your shade of black.

best coatings for heatsink to radiate heat

mds82

Enlightened

Steve K

Flashlight Enthusiast

mds82

Enlightened

monkeyboy

Flashlight Enthusiast

DIWdiver

Flashlight Enthusiast

Pinarello

Newly Enlightened

ssanasisredna

Enlightened

snakebite

Flashlight Enthusiast

degarb

Flashlight Enthusiast

Similar threads

Latest posts