Effect of colour in SSD heatsinks

Radiation cooling is tiny for low temp differences, so the finish doesn't matter much.

Plastic packaged ICs usually have a tiny hot spot, visible on a thermal imager. The real virtue of a heat sink is that it spreads the heat laterally. A piece of flat aluminum woud work about as well.

The 1 mm "conductive adhesive" will dominate theta.

A surface which radiates well will also absorb well. If there are other components nearby with higher surface temperatures, a 'radiating' finish could be a disadvantage because it will absorb more energy than it radiates.

Radiation cooling goes as the 4th power of the temp difference, which is microscopic at human-bearable temperatures. And the cooling goes as everything the heat sink sees in all directions.

I have a program that calculates radiation cooling. In cases like this, it's negligable.

Don't the SSDs throttle on temperature too?

The gold sinks would glamourize a PC but wouldn't have much effect otherwise.

I've heat sunk FPGAs and such, mostly to stabilize prop delay. The lateral heat spreading effect dominates. Fins don't make much difference.

Maybe I should have mentioned that these are for M.2 SSDs - those tiny

I'm talking about the simplest and cheapest type of heatsink for these devices. There are more elaborate - and more expensive - heatsinks for these SSDs but they are overkill for the vast majority of personal computers. Even the simplest ones are not strictly necessary for most use cases. Airflow is by natural convection incidentally augmented by the fans on the CPU, graphics card and case.

Perhaps this is comparable to a TO-220 device dissipating half a watt idle, occasionally going up to a watt or so. These SSDs dissipate something like a watt when idle and up to 4-5W when actively writing, but they have a much larger surface area than a TO-220.

To reiterate, the question is whether the colour of the heatsink matters. I'm inclined to think it will make little difference as most of the heat dissipation will be by convection rather than by radiation.

Yes, I'm thinking of replacing those pads with the much thinner ones intended for mounting TO-247 devices. Larger uncut sheets will be ideal but I don't know if I can get them here.

There's that factor too. So a shiny golden heatsink may actually be better than a black one.

They do. But SSDs in most personal computers used for non-demanding tasks won't reach such temps.

Yeah, they do look nice. PC enthusiasts these days, especially the younger ones, often use cases with transparent covers and lots of coloured LEDs inside - CPU fans, half a dozen case fans, even the memory sticks and the case itself have running and programmable lights on them.

We're engineers. All our PCs and phones are grey wrinkle finish.

Black will be always be the best choice even though as others have written the differences will be tiny.

Folk in really hot desert countries wear black instead of white as although black absorbs better it also radiates better - the wisdom of the ages?

This thread is about *personal* computers, mostly standalone units. Most reasonably new ones have at least one M.2 NVMe SSD these days. Many cases intended for personal use no longer have traditional drive bays and may come with some provision for fitting just one, maybe two, 2.5 and 3.5-inch drives somewhere on the floor or vertically the wall.

That would be the correct approach *if* this were about designing a new computer system from the ground up but it isn't. This is about adding an additional safety margin to a well established norm that already works fine in most cases without the added heatsink.

As above.

Probably anodized. I haven't seen one up close yet but it's highly unlikely that the colours are painted on.

The M.2 SSDs are easily accessed in most cases, more so than HDDs in traditional drive bays, especially when only one M.2 drive is installed. It's usually located in a clear space between the CPU and the graphics card port. A second drive may be obscured by a graphics card if one is installed.

I've noticed that in pictures and movies. It must be based on the assumption that people will spend more time in the shade than in the sun.

Best to use silicone grease or epoxy. Theta goes as thickness and grease or epoxy squash down to microinches.

Our Amazon has tons of Chinese thermal pad stuff in various thicknesses. Most claim about 7 w/mK and I've tested some that was actually about that.

Mica is good too. Specific theta is poor but it's very thin. And it's super cheap.

Here's my Insult program:

Sort of like "Go Faster Stripes" on cars 60-70 years ago. Those that didn't have the mechanical skill to work on their cars did fancy paint jobs.

It depends a lot on how hot things get. Metallic surfaces make lousy radiators and so are very bad for heatsinks if deltaT > 30C or so.

Forced air ventilation it won't make that much difference. Unless the SSDs are being hammered it seems unlikely that they need heatsinking. Worth monitoring them rather than waste your money on toy heatsinks.

I once had a 12v PSU with a 7812 regulator go into thermal foldback because the case was nice mirror finish shiny aluminium metal. A coat of black paint fixed it.

(any colour of non-metallic paint is black in thermal band IR)

Only if there is something red hot nearby. A decent rule of thumb is that with an ambient of 300K and radiative cooling scaling as T^4 the radiative component only really becomes important when deltaT > 30K.

IOW Things that get too hot to touch it becomes a factor to consider.

Right, most anything organic has high emissivity. Kapton or electrical tape, most any paint, even white white-out glop.

I had a brass-based incendescent lamp. The brass part looked room temp in a good Flir imager, but would burn your finger bad if you touched it. Brass and copper are mirrors at 10u.

Interestingly, water or ice or snow are black at thermal wavelengths.

That is more likely to be because white is so difficult to keep looking clean in such a water starved environment. The difference between visibly white and black materials in the thermal band IR is negligible. White has the advantage of reflecting heat away.

There is a good reason why tubes in solar hot water systems are black!

The tricky one is shiny polished metal which is a terrible radiator and so can get very hot in the sun. There are also a handful of designer metamaterials now which can lose heat by radiation in sunlight (ie become colder than their surroundings by losing heat skywards).

Telescope domes back before there was air conditioning were painted the whitest white that could possibly be manufactured and at great expense. A highly loaded MgO pigment white that was painful to look at. It is still completely black in the thermal IR band and so loses heat by radiation but reflects most of the daytime sunlight.

It turned out that this super white was actually too good and after sunset it allowed a supercool layer of air to form on the dome surface and spill down through the dome slit creating turbulence aka dome seeing. Aircon takes up the slack inside the dome during the day.

Modern observatories now use a mixed white pigment combined with aluminium dust to create a surface that is effectively neutral for radiation losses after dark whilst still being very effective in daytime at reflecting sunlight away. Dome seeing is no longer a problem.

Being involved in satelite coatings, the wave length dependance comes into play. In the desert you would choose a coat that reflects visible light and most infra red that comes from the sun, at the same time radiate well (infrared) at 37 C. There is more to this as selecting black or white. Islam women that wear black all the time are rarely allowed to leave the house, so there it may be applicable. Wearing black in a blazing sun is a sure looser.

Groetjes Albert