cool AlN things

Those do look pretty cool. Where do you get them?

Cheers

Phil Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

hobbs at electrooptical dot net

Thats a humongus pdf file, I thought it was a full catalog.

Novel idea, Alumina in SMT resistor packages.

Cheers

Pyrolytic graphite tape is neat stuff too.

Maybe just a typo on your part, but it's AlN not Al2O3. ~10x the thermal conductivity.

George H.

It's electrically conductive, isn't it?

We need diamond, preferably isotopically pure diamond.

Isotopically enriched is good enough - 50% higher thermal conductivity with 15 times less C13 than natural carbon, And it is easier to isotopically en rich carbon than it is to enrich uranium. C12 and C13 are more different th an U235 and U238 - as in 8.33% opposed to 1.28%.

I'm not sure that you actually need it. Natural diamond is a great deal mor e thermally conductive than anything else - excluding super-fluid liquid he lium and other stuff that it's hard to accommodate in regular manufacturing - and the extra 50% that you can get by making it more nearly isotopically pure is rather gilding the lily.

Pyrolytic graphite can be purchased with insulation on one or both sides. The insulated adhesive tape is meant for chilling tiny SMDs and circuit boards where direct soldering to metal bulk isn't practical. Naked pyrolytic graphite adheres very well with epoxy.

The stuff is expensive as hell. It's practical for custom jobs where weight and size matter more than another $50.

Oh, and don't test it by holding a thin strip at one end with your fingers and placing the other end of the strip over a flame. It works better than you'd think.

Given that one wants thermal conductivity and electrical insulation, the graphite just makes things worse. The adhesive is doing all the work.

It may be useful as a lateral heat spreader, if the adhesive is on one side. But copper can do that, too.

AlN and BeO are good thermal conductors and electrical insulators. They are available metalized, for soldering. The IMS things are cute, stocked parts.

Diamond is much better. It's surprising that nobody has come up with a process to make affordable bulk diamond.

Just what I needed. They replied about samples quickly too. Thanks for posting.

ChesterW

The thermal conductance/capacitance ratio looks pretty constant for the various parts, around 1.5 pF per w/K. That depends mostly on the material, AlN. BeO is a little better, but it's toxic.

Diamond is really good.

Really fast circuits need high power dissipation in tiny parts. So sometimes they need good cooling with low capacitance.

Don't usually have to worry about the pF. Spend most of my time nowadays on the other end of the spectrum dealing with 1/f noise and counting statistics.

ChesterW

Somebody probably has, but DeBeers has enough money to hire the *good* hit men.

Matt Roberds

They're not that *good*. DeBeers couldn't stop them from making gem quality diamonds. DeBeers doesn't use hit men, rather much worse (marketeers). ;-)

Exactly. So pick one. Both will work well. Graphite will be lighter and more conductive.

Sometimes you are a trip...

It's already affordable where you really need it. People wouldn't be making diamond heat sinks if there wasn't any market at all, but it will be a while before it shows up in consumer products.