0805 thermals

If there's nothing flammable under it, more than 'almost'. Molten glass doesn't indicate resistor value change, according to my measurements. It was someone else's design.

The smaller EPC BGA GaN fets do NOT like being globbed with hard epoxy.

Potting tends to crush things. And it makes probing and rework a tad more difficult. And it's messy.

How much did silica improve thermal conductivity? We tried a diamond filled epoxy and weren't impressed.

I did manage to melt one once, glass enamel dripping off.

Thermal conduction is like magnetics--the theta or reluctance goes like

Cheers

Phil Hobbs

And if the fillers increase the gap, they make things worse.

snipped-for-privacy@highlandsniptechnology.com wrote in news: snipped-for-privacy@4ax.com:

Good thing it wasn't the hypersonic response of the females. That's almost as bad for the gear as an EMP. :-)

That is a really nice layout, John. Is that lead free solder, because it sure looks too good not to be 63/37.

Anyway... are those purple stains on top of the caps from an over- exhuberant QA pen marking?

Looks like maybe four FETs on the bottom side along with two really big pot core coils.

All those EL caps non-polarized?

And no need for "LIAR 1" to be on the silk screen (har har).

John Larkin <jlarkin@highland_atwork_technology.com> wrote in news: snipped-for-privacy@4ax.com:

This was for HV potting not thermal interface optimization. So the gaps are huge regardless of the medium between the components and any sinking elements.

We made HV supplies that *required potting*. As such, we also knew that most potting media has very poor thermal characteristics. Especially since potting is meant to replace the air around HV components with a more reliable insulative medium than air. As such, cross sectional thicknesses are usually pretty high, and thermals move pretty slowly.

So we tried the introduction of the silica strands and it improved over the base potting compound and it was a large amount but not anywhere near 85%.

It did not improve much, but it was enough that we decided to use it for that particular job.

Polarized but used bipolar to small voltages.

More pics: