Oh, but that depends upon what got stuck to where... there are places where the cure is worse than the disease. Hint: don't CA in the nude (I know a guy that found that out the hard way. The staff at the hospital emergency room have quite a tale to tell their grandkids). And from personal experience... don't solder in the nude.
I have a couple projects I would like to try out the epoxy carbide thermal glue on. So what grit would would be the best / least expensive to use? I found this on ebay http://www.ebay.com/itm/8-ounce-1500-Silicon-Carbide-Rock-Glass-Polish-SC-15008-/320663383903?pt=LH_DefaultDomain_0&hash=item4aa906ef5f#ht_2835wt_767
Thanks
You want to maximize the packing density of the silicon carbide solids. This is best done with a mix of different grain sizes. You want large grains that maximize the uninterrupted heat flow path through the carbide and finer grains to fill in the spaces between the large grains. Packing densities of over 90% solids/10% binder are very difficult to obtain and work with. 85% solids is usually a putty like mixture. 80% is usually pourable.
The optimum ratio of particle sizes would depend upon the size and shape of the particles. Determining that ratio is a very difficult science/art. For potting, I might try 80% 120 grit and the rest 1500 grit... Or just go with all 1500 grit and see how it works out.
Thanks texaspyro, I think i'll try try the straight 1500 grit is there any easy way to do some rough testing to see how it performs?
I'm basically looking to use it as a filler and backer, like filling some pill's about half full to aid in heat transfer between the led and pill.
A RELATED QUESTION about potting:
Does anyone have an opinion about hot glue as potting material? Would the initial heat (=hot temperature of the glue) destroy the electronic IC parts of the PCB?
My lumintop worm has a sandwich driver in an aluminum pill, and i feel tempted to fill in the sandwich with hot glue :D
I have actually wondered the same thing in the past.
i've got a dead lumintop worm driver. i will practice the hot glueing technique on its sandwich. HDS uses polymer as well (hot glue is a polymer!) for the potting and as we know hot glue gets cold after a few seconds already and it would be a GREAT IDEA to cool the driver in a refrigerator before dripping the glue on it.
my personal assumption is that the (refrigerator cooled) IC parts dont get damaged by the short-time heat of the glue. It's an assumption only and we would have to conduct experiments to find out how well IC parts can endure short-time heat of hot glue.
Also, if the drivers can stand up to a bit of water you can use water to cool it down. Make sure to keep us updated. I have been thinking of using it instead of aluminum foil for wrapping p60s.
the same topic was discussed on an electronics forum and it seems safe to do so.
i will test it on the dead driver (sandwich) of the lumintop worm!!, and then post pictures of it.
never touch a running system .. that's why i wont desolder the cables from my running lights (e.g. Tank E09) only for the purpose of potting their sandwich'ed drivers. (i cant solder fine cables anyway.)
Help me out here, why do y'all use potting compounds at all? I mean it's not like they are very effective at heat transfer. Yes, better than air, but a very small aluminum heatsink is still over 100 times more effective than any potting compound around. Yes, No? At least everything I read suggests never use potting compound unless you absolutely need to for vibration resistance and or moisture barrier.
I don't know, so I'm asking...
Why use it?
my thoughts are:
Oh, I tend to forget, some people actually use their lights. Walk, jog, camp, ride or work with them. LOL, I’m out of touch with reality, (truth in that)… I always think of them as shelf queens…
The temperatures that hot-glue guns run at will in NO WAY damage any electronics on a printed circuit board. The vast majority of circuit boards are reflow soldered which means that in their production they are exposed to temperatures exceeding 260C (500F) with no problems. Now, potting in hot glue will do nothing for thermal transfer at all - and I mean nothing. It's absolutely horrible at transferring heat, and even worse, when the electronics get hot, it will melt and 'run'. What it will do, however, it to insulate the electronics (quite well, in fact) from moisture.
PPtk
Has anyone made any potting compounds using Silicon carbide as texaspyro discusses in this thread. I have no doubt his recommendation it good. Wondering how it worked for those that tried and if they have any recommendations on how to mix and use.
Hey 4Wheelr, you did see the part about the silicone carbide epoxy being a permanent mix, right? Not much way to remove it once it’s in there.
Yeah, I mention that in the other thread. Thank you for making sure though bro.
I'm following this closely. ImA, are you planning to mix the silicon carbide with the GE Silicone II caulk (non-corrosive silicone caulk) you mentioned in your other thread? If so, how much silicon carbide is mixed in?
-Garry
Yes, but I trying to find a Alkoxy cure silicone, if possible. The GE Silicone II is natural cure, but it could slightly corrode copper while actively curing. I don't know how much silicon carbide to mix in . I guess as much as I can, but sill be able to flow the potting mix into the pill cavity.
I’m just throwing this idea in here. Anyone, just anyone, actually ever thought of simply using mortar?
I mean, portland cement + fine sand. Really cheap. Literally dirt cheap.
Circuit can be coated with lacquer spray or something for protection during drying. If this can work, you possibly get potting that does around 1.6W / m.K. Seriously good potting performance.
Any thoughts?
What ever happened to using JB weld?