Surface Mount Resistor in Vacuum Chamber

Any surfmount resistor will be OK in a vacuum. The real issue is cooling. Convection is out, so you have to use conduction or radiation.

High voltages can be a problem, too.

George Harold may kick in here, but if you're checking, the problem doesn't occur at vacuum but at some magic pressure that's significantly below atmospheric. Apparently the dielectric strength of both atmospheric pressure air and hard vacuum is better than very rarefied air.

I dunno more than that, though.

There's the Paschen's Law thing, which says that the lowest breakdown voltage happens at some non-zero pressure. But if the thing is powered up as you pump down, you may as well assume the worst case.

(There's also the fun Farnsworth Multipactor effect with RF. That has killed some satellites.)

I have zero experience here, but isn't out gassing an issue?

On Thu, 21 May 2015 15:14:12 -0700 (PDT), snipped-for-privacy@gmail.com Gave us:

Pretty much all of them.

Sure. UHV projects are often sensitive to outgassing. Some need to be baked and pumped down for days to get crud out of the system. I'd expect that a little solder flux could be bad news. PC boards will outgass.

Here's a UHV "pc board."

It was designed by some chemists, who didn't think about cooling the opamp.

On Thu, 21 May 2015 17:39:37 -0500, Tim Wescott Gave us:

Both the ceramic (silica) substrate, and the soldered termination soak the heat, not air.

So, I have no idea what Larkin is OFF about again. I just know that he IS OFF.

Soak the heat? What does that mean?

Say something specific and we can discuss it.

Right. Voltage breakdown occurs when the mean free path is long enough for a particle to gain enough energy to ionize the gas when the ion hits a mol ecule. If the mean free path is longer than the distance between electrode s, then an ion will on average not hit a molecule . So you do not have bre akdown at really high vacuum.

So the worse pressure for breakdown is usually about 4 mm of Hg.

Dan

On Thu, 21 May 2015 16:42:03 -0700, John Larkin Gave us:

Not here on earth under normal atmospheric pressures.

Utter bullshit.

ANY VOCs are bad news, including simple "coffee breath" after the cleaning segment of the assembly process.

So do parts if they take on water. A ceramic dipped cap will grab water molecules on the first few microns of its surfaces unless cleaned properly. That rules out your poor use of the right machine.

snipped CRAP examples.

Again Larkin spouts nonsense.

The board's HV section need only by without a solder mask. i.e. a bare board. G-10 is usually a bit better than FR4.

The solder nodes on HV need to be ball shaped, so that PCB assembly in his example is NOT part of the HV section. If it were, it would also need to be potted.

After a NON-residual cleaning (not in Larkins sauce bath), a simple, HEATED vacuum session for about 40 minutes at 60C will suffice to remove any water in the PCB. Any proper HV circuit must have the HV section fully potted. His petty 1kV 'circuit' is the limit. Any voltages above that require potting to avoid failure modes.

There is only ONE material on the acceptance list by NASA for space HV applications, and that is CONAP, and it does NOT outgas, nor detach from the circuit board or component elements. Even 1kV boards fail often enough to need potting.

That CRAP he posted a picture of was done by an idiot "engineer", NOT an HV engineer.

The front end (drive section) does not need to be potted and can have a solder mask applied. The transformer takes care of the isolation, and even that gets vacuum potted in most cases.

If properly vacuum oven heated after cleaning, any HV section's PCB will be free of water, and it does NOT take "pumping down for days".

Only after it comes out of one of Larkin's 'special vapor phase sauce pits'.

And the idiot spams his pathetic self with nearly every post too.

More like LOWland techNOTogy.

On Thu, 21 May 2015 20:58:32 -0400, DecadentLinuxUserNumeroUno Gave us:

Epoxy dipped, that is.

No argument there. That design was absurd. I said so.

It was designed by chemists. The company was IonSpec, acquired by Varian, and then by Agilent. Agilent killed the product line. FTMS.

Potting outgasses too much for UHV.

As I said, it was done by a couple of chemists. The opamp is a jfet type operating at unity gain. I figure they gave up about 30 dB of s/n.

That board did operate in the UHV section of an FTMS system.

Serious UHV systems can need to be baked and pumped for days. FR4 and most epoxies are lethal at UHV.

Since this thread is about using a surface mount resistor in a vacuum chamb er, we aren't talking about normal atmospheric pressure.

It's been true of every vacuum system I ever worked on, and I've never actu ally needed anything better than a chemical vacuum - the electron gun ares in the electron microscopes I worked sometimes did need a physical vacuum, and tended to have a local ion pump to get it when required, but that was i solated from the parts I worked on and stayed somebody else's problem.

BET theory points out that the first layer of water molecules on a surface is usually a lot more strongly bound to that surface than any subsequent la yer of water molecules - which only have water to bond to. Getting rid of t he micron-thick layer is trivial, but the last nanometer or so tends to be more strongly bound.

Without the vacuum oven it will.

On Thu, 21 May 2015 19:29:48 -0700, John Larkin Gave us:

What part of CONAP does NOT outgas do you not understand, idiot?

It is approved for space applications by NASA, and they do not allow gassing. It is the ONLY approved potting material for HV space applications, and its volt per mil resistance IS acceptable for UHV. You couldn't be more stupid if you were gassed.

It is what you deserve though. You act like you already were.

On Thu, 21 May 2015 19:29:48 -0700, John Larkin Gave us:

Fully isolated from the HV return and ground, idiot. It was AT the HV level and therefore needed NO protection in the locality it resided in.

I'd bet that you also have no clue why a 4800 turn transformer secondary has to be segmented to keep from failing.

On Thu, 21 May 2015 19:29:48 -0700, John Larkin Gave us:

Make up your mind, dumbfuck.

ELECTRONICS GROUP. UHV MEANS ULTRA HIGH VOLTAGE.

You are truly stupid.

On Thu, 21 May 2015 19:48:44 -0700 (PDT), Bill Sloman Gave us:

You're an idiot too. The voltage drop across an SMD is NOT going to arc over in ANY vacuum. If it is high enough to do so, the resistance media will have already failed short long before it ever got placed into a chamber.

On Thu, 21 May 2015 20:01:22 -0700 (PDT), Bill Sloman Gave us:

UIHV means ultra high voltage. IN an ultra low pressure scenario an SMD is not going to arc over. The actual point is ABOVE that level. And even then, the SMD is safe, because IF it had that much voltage across it, it would have failed before ever entering the chamber.

HV resistors are LONG between the nodes.

You all fail.

That's plain wrong. The arc-over voltage at the Paschen minimum is a couple of hundred volts, and your "any vacuum" is a broad enough pressure range to include the Paschen minimum.

I don't know of any "resistance media" that would fail at that kind of voltage.