I've been reading up on thermal material and potting compounds, and silicon carbide powder mixtures are mentioned. There's currently an eBay listing for 20 lbs. of Fujimi "GC 1000" silicon carbide powder (http://www.ebay.com/itm/360805239021), and I clicked buy before thinking too much about it. I don't know what I'm ever going to do with 20 lbs. of this stuff, but anyway...
I'm thinking of mixing some of this with Fujik to glue down MCPCBs to pills in a thermally-conductive and reversible way, should there be a need to replace components in the future. I may also use it in potting compound mixtures in the distant future.
Anybody that has experience with this stuff... Would there be any issues with this product and the planned uses?
I’ve only ever come across silicon carbide as an abrasive used for in stones for sharpening, you could use it in a paste to polish metals, depending on grit size, if the 1000 grit ws correct then it’s quite a fine particle size.
It may have other applications but I’m not sure of any.
Reading the Wiki page on it reveals some surprising properties besides abrasion.
Start small, things of no consequence, and test. Reading the above linked wiki, it seems like it will be current insulating not conductive, yet thermally conductive, so suitable for the job.
20lbs? If it works start selling film canisters full, or some such similar.
And a third if I recall was using a theoretically better ‘silicone’ replacement too. Cant think of that thread… but it was recent
I think you can make a paste and apply it to leather to make a Barber’s strop or to polish blades or metals, maybe making a tumbler to polish metal parts, not sure if a paste of 1000 grit would polish glass in a nice way.
If it’s like wet/dry paper grits, I think it’s the same, then 1000 grit would produce a good frosted glass effect, bench water stones go to 12000 and beyond.
It would make a good abrasive paste.
Thanks for the input. I've been following several of the threads on BLF, which is where I got the idea of mixing in SiC. I don't have any tools to do actual testing, but I can report back with anecdotal results. I'll also look for some GE Silicone II the next time I go to the home improvement store.
Prior to this, the only thing I had ever heard of SiC was from Top Gear:
You will not be able to measure a difference in light output or anything else doing a comparison between plain cheap Radio Shack white silicone-based thermal paste and the most space-age whiz-bang nano-whatever thermal paste.
Back when CPUs had tiny exposed dies of around 1cm x 1cm the fancy pastes could make a measurable difference in core temp/max stable speed, but with more recent parts that have a big ol' integral heatspreader on top, the pastes make much less difference. Similarly, in a flashlight, if you had to use the paste directly between the tiny LED thermal pad and the MCPCB it would be something to worry about, but given the huge (in comparison) area of the MCPCB-to-pill contact, it really doesn't matter.
Measurable does not mean significant, it just means the numbers that pop up on the measuring equipment are different from one test to the next. If it's not placed into a real world context of how those tiny but measurable temperature differences affect the LIGHT OUTPUT it's not much more than a potentially interesting intellectual exercise.
I'm not saying that 'oh well a 5% difference in light output isn't really noticeable so you shouldn't bother', I'm saying 'How much difference in light output is there with a 0.5* difference in temperature?' A flashlight is a thing that exists to shoot light out the shiny end, not a thing that exists to keep the emitter at the lowest possible temperature. If all you're worried about is how cool it runs, remove the batteries.
I use #320 grit silicone carbide abrasive powder all the time. I know about the dielectric and excellent thermal properties of it but never really care. Even the silver nano particles in Arctic Silver only add tiny improvement over the pure epoxy.
Silicone carbide thermal conductivity is approximately 120W/M.K at room temperature and silver is 400+ W/M.K
AS5 thermal conductivity advertised as high as 7 W/M.K, IMHO SiC would add pure epoxy thermal conductivity far below 7 W/M.K
I use #320 SiC powder as separator between heat sink and electrically active MCPCB. Mixed in the thermal epoxy. But remember to lap the components with finer grit (#1000 or above). This way I can spread the epoxy as thin as possible without short circuit the bonded components. Dow Corning offer silica balls filler option in some of their thermal adhesives for exactly the same reason.
This abrasive powder is so sharp, once applied don’t move the components to be bonded to often while the epoxy is still wet. Otherwise it will grind the soft aluminum/copper and the chips could potentially short circuit.
In this chart you can compare the efficacy of materials as thermal compounds, including mayonnaise, lipstick, toothpaste, and the finest purpose-specific thermal compounds. The difference is not remarkable.
