PCB under vacuum

t)

ducer.

so

get

page,

According to the kit instructions you put the SENSOR in the vacuum, not the PCB. So one can assume it is not vacuum rated... Electrolytic capacitors perhaps?

John :-#)#

As long as I'm here with experienced people, is there another vacuum pressure kit that is cheaper, or works in a vacuum? Mikek

Yes, I hoping for a picture tomorrow, but, I doubt it has electrolytics, but, I kinda hope it does!

cer.

Electron microscopes put printed circuit boards under vacuum all the time.

The Cambridge Instruments machines used off-the shelf Pirani gauges all the time.

You do have to worry about heat dissipation - anything that relies on conve ction cooling has a problem. The Pirani gauge actually measures pressure by monitoring the convection cooling of the sensor wire.

and it's not much use for high-vacuum measurements - below 10^-4 torr.

Your son - or his lecturer - should ring up MKS and ask them about the prob lem.

Having the sensor on a separate board probably does suggest that MKS didn't expect to see the drive electronics put in the vacuum chamber.

There are all sorts of things you can do to get regular electronics to work under vacuum - at one point we clamped thick chunk of copper braid onto th e top a device that was running hot, and clamped the other end of the braid onto the metal wall of the vacuum chamber, and that provided enough coolin g to let the system work continuously.

One half-wit we ran into used to run his experiment for 20 minutes, then le ave it to cool off for a few hours - graduate students sometimes have to re sort to this kind of expedient, but this guy had got his Ph.D. and had the resources to do better. I don't know if it got him fired, but he left the job shortly after we'd commented on his approach.

That's interesting, now I have to ask, Does the gas that is in the vacuum alter the convection cooling characteristic?

ie, if you pulled a vacuum on a chamber of air vs a chamber of xenon

would the gauge read the same for the same vacuum?

I'll need to ask him, he mentioned -7 tor, so I don't know what vacuum they are working at.

Yes, and looking at they're mounting, it does shew external mounting of the sensor to a hole in the chamber. (That won't work for their experiment.)

Thanks for the info, Mikek

OH, found my answer in the wiki! Mikek

I don't see it on the Pirani page, but do you think this is a pulsed sensor? It looks like it measures lower than what is listed for a standard pirani sensor on the wiki.

Mikek

sducer.

e,

me.

the time.

onvection cooling has a problem. The Pirani gauge actually measures pressur e by monitoring the convection cooling of the sensor wire.

From the wikipedia page

No. See the image.

You said he was monitoring a gas pressure let into a second chamber

problem.

dn't expect to see the drive electronics put in the vacuum chamber.

t.)

Then they need advice. As I mentioned, there are ways of tackling an over-h eating problem.

work under vacuum - at one point we clamped thick chunk of copper braid ont o the top a device that was running hot, and clamped the other end of the b raid onto the metal wall of the vacuum chamber, and that provided enough co oling to let the system work continuously.

n leave it to cool off for a few hours - graduate students sometimes have t o resort to this kind of expedient, but this guy had got his Ph.D. and had the resources to do better. I don't know if it got him fired, but he left the job shortly after we'd commented on his approach.

Yes, but still under vacuum, he mentioned the vacuum was 1/600 of the outer chamber.

I'm just trying to understand a little of what he is doing, in 4 or 5 months he will have a chemical engineering degree. Right now, I think he's just free labor. But he seems to be enjoying this, but I hope he gets exposed what other options in his field.

t)

ansducer.

so

get

page,

ow.

time.

ll the time.

convection cooling has a problem. The Pirani gauge actually measures press ure by monitoring the convection cooling of the sensor wire.

-Vakuummeter.jpg

Still at less than atmospheric pressure. It's a "controlled atmosphere" rat her than a vacuum. I used a vacuum line for my Ph.D. work, and the main poi nt about being able to pump it down was to get rid of any oxygen.

Once I'd got it clean, it served to let me play with nitric oxide, bromine and nitrosyl bromide. Some of the bromine got out once, and made a cleaner rather unhappy for about half an hour.

In another incident, air got into to my stock of nitric oxide which turned into nitric oxide quite quickly, so I had to make up - and purify - a new b atch.

For equipment to be used in a satellite, the material outgassing is a significant issue. The largest "component",, the PCB board itself, can be a problem, check the PCB material used.

10^-7 torr is a couple of orders of magnitude harder vacuum than the sensor can reliably detect so all you can be certain of is 120C for 24 hours and the Noble gas kit used to get a couple of days at 200C with the flight tube peaking at 300C! How brutal the treatment is depends on local conditions but this was typical in the UK.

No. It is a measure of rate of heat loss from the filament and so depends critically on the speed of sound. Traces of Helium will carry away a lot more heat than something heavy like Xenon. Brief intro:

The usual trick is a means to clamp the thing that runs warm to a heatsink attached to the metal walls of the vacuum chamber.

Beware that for hard vacuum kit they may be baking the thing quite hard at some point and have forgotten to mention this minor detail.

Yes, I have already sent him an email asking for clarification of the pressure he is working at. Mikek

MKS makes a range of vacuum sensors, Pirani and other types. We use one of their modular type that attach to a standard fitting on the outside of the chamber. It wasn't much more expensive than the full kit you linked to. Wide-range Pirani sensors are prone to damage from pump oils and other contaminants in the vacuum system, we already had to change out our sensor element once.

Jon

The Pirani scheme basically has a light bulb sort of filament that is run on a constant current. In air, it stays cold, in vacuum it self heats quite a bit. The resistance changes with temperature, so that is measured to read vacuum. So, the electronics needs to regulate some supply voltage down to a very low voltage across the sensor, and therefore dissipates power. That's why it is a huge mistake to have the electronics in vacuum. I know our MKS Pirani gauge runs QUITE hot in operation. I can't tell how much of that heat is in the sensor and how much in the electronics.

Jon

No, but it is a modest effect for most gases. The manufacturer can supply correction tables for any gas you need.

That must be 10 ^-7 Torr, which is quite hard vacuum. Wide-range Pirani gauges can read down to 10^-5 Torr, you need ion gauges of a variety of types to read below that level. Mostly, there are cold-cathode and hot- cathode types. The Piranis are generally used to determine when it is safe to turn on the high-vacuum pumps and more sensitive gauges.

Jon

cer.

Yeah get it out of the vacuum! Any idea of the voltages involved? There's a sweet spot in pressure where it's easier to get 'HV' breakdown of the gas. The other thing you can do is turn off the voltage till the electronics is in high vacuum... (Is it going to high vacuum? Some thing more than a mechanical pump.) If the part is always sitting around 7 torr then you are stuck with taking it out

George H.

In the FWIW dept, I did get a little more info from my son. The outer chamber does have up to -7 torr and yes, it uses a different type sensor. On the internal chamber, they run -2 Torr after the gas is allowed in. Their plan to reposition the pcb by cutting a slot in an access cap and run a ribbon cable through the slot to the sensor. Then use Torr seal to plug the hole.

Can they make a good seal on a ribbon cable through a slot?

Will the ribbon cable leak?

Is there any difficulty sealing a -7 Torr slot passing a ribbon cable? He mentioned maybe a 1/2mm slot width.

Just want to put some ideas in his head to think about if anyone has experience with this.

Thanks, Mikek