Unless the hermetically sealed volume is filled with transformer oil, sulph ur hexafluoride or the like. You can add a bellows to the hermetically seal ed volume to cope with thermal expansion and the like."Potting" implies a r igid and self-supporting filling, which isn't - strictly speaking - necessa ry.
Sadly it's now your name. You've earned it, AlwaysWrong.
No kidding? Just from the vapor pressures?
Hmm that is what I was told. At the FEL we used lead solder in the not so high vacuum areas.
But I'll admit when I went looking for the vapor pressure of lead it didn't look that bad, ~10^-8 torr at 600 K.
Maybe it's a problem during the bake-out phase?
George H.
Thank you all for the responses, by the way.
The main concern was out-gassing from the component. Initially I thought the generated heat would be a problem as well, but the current flowing through an 18-kOhm resistor will only be a few milliamps.
So from what I've gathered, any surface mount resistor should do?
I have some samples of a 1T ohm 0603 resistor. It would arc at atmospheric, or at low pressure, before it would overheat.
I couldn't measure the chip temp, but I calculated that if the cooling was all radiation, Tj must have been around 130C.
It's a jfet amp, buffering the signal from an electrostatic pickup plate. At that Tj, the input bias current was huge, so they added a 1 Meg resistor to ground. The Johnson noise dominated.
I wanted to try to detect a single molecule in orbit, which I figured was just barely possible. Agilent acquired the operation and shut it down; pity. They have also shut down the Varian NMR operation; bigger pity.
Ha, I just did this calc, (over at sci.repair) with emmisivity = 1. you get 1kW/m^2 from 400K to 300K. (100mW/ cm^2)
George H.
I sure hope that doesn't run in the family.
On Fri, 22 May 2015 08:40:31 -0400, Spehro Pefhany Gave us:
Then why do you claim they outgas?
My calc was very rough. There must be some rediative and conductive interaction with the ceramic block. But the t^4 thing reduces the calculated temp error some.
Anyhow, it was a really silly design. I did a really cool amp design (he said modestly) but never got to try it out. I was planning to conduction cool into the machine structure.
FTMS is a neat technique, but it starts with a megabuck of helium-hungry magnet, so it is a good candidate to replace with something cheaper.
People seem to be capturing and re-liquifying boiloff helium these days.
3 mA into 18K is 160 milliwatts, potential thermal hazard in vacuum. It might outgass if it gets hot (think smoke!)
On Fri, 22 May 2015 09:24:47 -0700, John Larkin Gave us:
You have obviously never done a conduction cooled design.
You remain 100% fact-free.
Do not know about space applications , but an least one ICBM has high voltage circuits that are not potted.
Dan
Potting is a mess. It can stress and crush parts. It makes testing and analysis a nightmare.
And it's heavy. Heavy is not good for spacecraft.
An ICBM is a space application. Probably the primary one in terms of dollars spent.
No surprise there - to anybody except AlwaysWrong.
Who knows what they put into thick film ink? A bit of detergent to make the ink wet the substrate could leave volatile-ish residues after firing.
On Fri, 22 May 2015 13:44:51 -0700, John Larkin Gave us:
Only the type an idiot like you would chose..
Only the type an idiot like you would choose.
No, it makes every circuit uniform in characteristic operation. Open air makes for an "each circuit assembly is different" situation.
No s*it.
If it is HV, it MUST be potted for space applications, and the weight is already factored in when they come looking for one, and they are ALL custom. Even NOAA buys them, and they factor in the weight as well.
You ain't real bright.
On Fri, 22 May 2015 13:44:51 -0700, John Larkin Gave us:
Only the type an idiot like you would chose..
Only the type an idiot like you would choose.
No, it makes every circuit uniform in characteristic operation. Open air makes for an "each circuit assembly is different" situation.
No s*it.
If it is HV, it MUST be potted for space applications, and the weight is already factored in when they come looking for one, and they are ALL custom. Even NOAA buys them, and they factor in the weight as well.
You ain't real bright.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required