No insulator. Could glue a chip to it and have some advantages with that. Power transistor cases, much better than plastic and if you can get the silicon to stick to it you have a really good package.
In my opinion, ceramic is a bit too hard to machine for use as heat sinks per se. This stuff about near has to be molded to its shape the first time. With holes.
I wonder how brittle it really is. (was mentioned "drop it") Maybe I'll get one and find out. I do have a cement floor.
If I remember I might, got other things to do right now.
I sure hope that the semi contact area is ground flat. Moulded ceramic is usually rough, and roughness wrecks thermal conductivity. But I guess that's not so important for a small heat sink.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
If it's HIP like I suspect, the surface will be not only mirror finish, but so smooth that you have little hope of getting a strong epoxy bond, if you wished to glue to it.
John Larkin wrote in news: snipped-for-privacy@4ax.com:
The flatness is what counts. Coplanarity is what is important when mating the heat source to the sink.
No, surface texture at that level has ZERO effect for or against heat flow. The interface is where most is lost, and that is about the interface medium. Air = no good. So we put fillers in. sil-pad tabs or liquid 'paste'. The surface quality of molded ceramic is fine grained, and presents exactly ZERO difference.
If it were actually 'rough' you would have a valid point. But the profile of the texture of a molded ceramic face is far too fine to have any effect against thermal flow.
IF it was ground with 36 grit. MAYBE. But the surface quality of ceramic has zero effect. Far more important is a tight, flat interface. Whereas the 36 grit ROUGH surface WOULD impede thermal flow. But not really a lot. You said it "wrecks it" but it doesn't really. Non-coplanarity does.
You could find out with your IR imager. The difference is practically nil until the roughbess gets pretty far up there.
We had a guy who insisted on soldering the big FETs down before mounting the heat sink. If they were individual sinks sure, but a gang of 12 FETs across the edge of a board there to mate with a heat sink poses problems with his method. We were frying FETs within seconds of use. I showed him that we needed to mount the heat sink and FET packages onto the PCB and THEN solder the FET legs.
"Tim Williams" wrote in news:qeg3uj$8qr$ snipped-for-privacy@dont-email.me:
His claim is false. A ROUGH surface degrades an interface. It would have to be very rough though. The smoothness a cast or moulded ceramic face has is far too fine grained a surface profile for it to degrade thermal flow. And it will also allow adhesive attachment a bit more.
And the other problem mirror finish has (for epoxy attachment) is that if your epoxy detaches, it is as if it is not even attached at all. Smoke ensues.
Both roughness and flatness matter for good heat conduction. Thermal grease is not actually a very good heat conductor, and it needs to be thin to minimize theta. If the contacting surfaces aren't flat and smooth, that hurts. We machine aluminum extrusions flat for high power stuff, because they tend to be wiggly as-extruded. Our ceramic AlN insulators are lapped to basically optical limits. Transistor packages tend to be very good.
Thermal silicone grease has ceramic filler particles, which may help conduction but increases gap. They compress to below 100 microinches, which is good. I tried some diamond filled grease, but the particles were so big that they increased the gap, with no net benefit.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
John Larkin wrote in news: snipped-for-privacy@4ax.com:
Yes, but what you fail to note is the actual degree of roughness. I used to polish stainless steel plates so I know what surface quality is. I also know about the interface mediums used between heat sources and heat sinks.
Flatness matters, but once one is below a certain degree of surface smoothness, there is no difference in conduction.
Yes, but it comes down to mils, not microinches. The 'matte' finish on a perfectly flat A1N insulator has been perfectly fine for decades.
Oh boy!
Yes. Most credible companies mill mating surfaces for the heat source to mate to the heat sink.
By you guys or the maker? I'd bet immeasurable difference between an off the shelf A1N insulator and an "optically lapped" version. It is serious overkill with a nil net gain result.
Not really. A copper slab plated. That 'slab media' is extruded and shear cut when they are building those FETs and transistors. It is very flat, but not in any way an "optical quality lapped surface". You have taken your obsession a bit too far and no longer have a picture of what is happening.
SOME thermal grease has ceramic filler particles. Some have Silver. Some have other media. Some have NO granular media at all.
As long as electrical conduction is not an issue, a silver filled thermal paste or epoxy is best.
Why do I say this? Because THAT is what the entire microchip industry uses for their die attachment.
Their stuff is NOT polished to a mirror finish either.
There is a point at which any 'returns' or gains are minimal or even immeasurable. You are worrying about a molded ceramic insulator surface quality that is far and well below the profile level which causes a conduction differential.
You have problems looking at the bigger picture stuff, and apparently at the micro scale as well.
Zero gap is zero gap. Flat, co-planar mating is what is required MOST. A mirror finished surface intrfaces no better than a surface with a 15 micron pebble matte finish. You are well below the point where it matters or even shows a difference.
Also, not that the tabs have a hole that is offset from the conduction surface. That means that the clamping force on it causes it to want to tip a bit away from co-planar mating. There are a lot of mounting clamps out there now, which clamp against the main body of the part. That too is important.
John Larkin wrote in news: snipped-for-privacy@4ax.com:
We used those too. They rely on the spring characteristics of the clamp material. I think they could have made them slightly thicker, and produce about 15 or 20 lbs more force aginst the tab.
Nice design where the plate offset matches the leg heights. Easy to build and then solder.
Ours was perpendicular to the PCB and was only 2kW but that was continuous. We ended up using a sink with no fins. It was a rectangular tube and allowed better air management and worked fine for the task.
John Larkin wrote in news: snipped-for-privacy@4ax.com:
You are incorrect. *I* said flatness is important. YOU ranted on about the surface condition.
Flat is better. Polished? Does not matter.
The flatness matters. The roughness will only matter AFTER it rises above a certain level on the profilometer.
You jacking off at the mouth about a flat face with a matte finish comprised of 15 micron surface undulations causing less cooling efficiency is a joke. Period.
The flatness is ALL that matters once the surface quality is past a certain point. That point is way up in the "36 grit" scratch profile range, not anywhere near the surface profile of a matte finish ceramic face with 15 micron bumps.
Rick C wrote in news: snipped-for-privacy@googlegroups.com:
He is experiencing science observer overkill.
I know for a fact that the difference is so small as to not even be measurable. The term nil works here.
One can take ten of the FETs and do a 'thumbprint' job with them on a known flat surface (an inspection block), and you will find that even they are not all as flat as you might like to think they are. We even "honed" some to prove it. Aside from the copper oxidation problem with the now exposed bare copper, they actually conduct heat away better. But the difference is so slight as to not matter at all.
Does the thermal resistance of a power resistor coating matter all that much (within reason)? At least for a resistor that isn't intended to be connected to a heat sink. Power resistors run at much higher temperatures than the normal semiconductor parts and the area of the case is larger than the element. Thermal conductivity helps moving heat. It doesn't help dissipate it.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.