Is nesting vacuum chambers a practical approach for creating a medium vacuum?

[Smacal1072]

Hi all,

I don't presume to know anything about vacuum engineering, but I'm considering trying to build a small homemade medium vacuum, and I was wondering: Is it practical to nest vacuum chambers inside each other?

I've been surfing the web and can't seem to find any info about it - but it sounds feasible.

In case I wasn't detailed enough: Say you had a vacuum chamber of maybe 3kPa. Couldn't you encase this chamber with another of equal quality, and simply evacuate them in stages? I'm sure I'm missing something, or they would be doing/have tried this already.

 

[MATLABdude]

Hi all,

I don't presume to know anything about vacuum engineering, but I'm considering trying to build a small homemade medium vacuum, and I was wondering: Is it practical to nest vacuum chambers inside each other?

I've been surfing the web and can't seem to find any info about it - but it sounds feasible.

In case I wasn't detailed enough: Say you had a vacuum chamber of maybe 3kPa. Couldn't you encase this chamber with another of equal quality, and simply evacuate them in stages? I'm sure I'm missing something, or they would be doing/have tried this already.

It might be possible, but it wouldn't be done because it offers no real advantage (over, say, a 'simple' vacuum chamber that's as large as your outer vacuum chamber). It would take longer to pump your chamber down (due to the inner chamber(s) acting as a baffle--applies only when you're in the molecular flow regime, a few hundred mTorr, atmospheric is 760 Torr). Additionally, the increased surface area would allow more contaminants (primarily water, and any heavier / stickier hydrocarbons) to remain in your system.

However, what is frequently done is to use a load-lock (basically an airlock for your samples). You keep your main (large) vacuum chamber at very high vacuum (10^-6 Torr or lower), and then load in samples through your much smaller load lock, which is repeatedly brought up to atmospheric to load in samples. In this way, you don't have to spend as much time pumping down your main chamber, and you don't have to worry as much about contaminants getting into your main chamber (the nice thing about the really persistent contaminants is that they tend to stick on walls--i.e. in your load lock).

EDIT: Result of a Google search on vacuum loadlock:
http://www.mdc-vacuum.com/searchs/doc/LoadLock-Intro.htm

 

[mgb_phys]

Another trick is to have two concentric O-rings and pump out the gap between them.
The outer O-ring takes the mechanical load between atmosphere and vacuum and the inner ones stops anything that leaks through the outer O-ring getting into the chamber.
It's also a good way of telling if a leak is a real leak through an O-ring or some outgassing/contamination issue.

 

[Smacal1072]

Thks, that definitely sheds some light on the issue - I'll probably go with a conventional chamber, and the two concentric O-rings sounds like a good idea...

My reason for suggesting it was the buffer zones provided by nesting the chambers might make whole apparatus safer by distributing the pressure stress between two chambers, instead of having it concentrated on just one (I'm a little terrified of vacuum chambers to be honest).

 

[mgb_phys]

It's not really worth it from a stress point of view. As you get to higher and higher vacuum it becomes pregressively harder to get those last molecules out but even at the rough vacuum you will have almost all of the atmospheric pressure.

Mechanically holding 1 atmosphere isn't difficult (it's only 14psi) and so there isn't much energy stored in a vacuum chamber.
Gas bottles on the other hand, 6ft tall, 150lbs of steel and pressurised to 3000psi - those scare me ****less!

 

[MATLABdude]

It's not really worth it from a stress point of view. As you get to higher and higher vacuum it becomes pregressively harder to get those last molecules out but even at the rough vacuum you will have almost all of the atmospheric pressure.

Mechanically holding 1 atmosphere isn't difficult (it's only 14psi) and so there isn't much energy stored in a vacuum chamber.
Gas bottles on the other hand, 6ft tall, 150lbs of steel and pressurised to 3000psi - those scare me ****less!

Are you more likely to be killed or maimed in a catastrophic vacuum chamber failure, or a catastrophic gas cylinder failure?

My money would be on the gas cylinder (especially if you get suffocated / blown-up by the former contents), or the cylinder rockets towards you at whatever velocity once the head is sheared off (link goes to a Mythbuster video):

 

[Bob S]

I have seen nested vacuum chambers when we wanted a vacuum chamber with the walls at 2 kelvin, so we had on the outside an insulating vacuum with lots of superinsulation. For vacuum chambers at room temperature, 10^-9 torr is easy, as long as you have found all the real leaks.

 

[mgb_phys]

Are you more likely to be killed or maimed in a catastrophic vacuum chamber failure, or a catastrophic gas cylinder failure?

I've never heard of a catastrophic vacuum chamber failure accident.
But every chemistry lab will gleefully show you the hole in the wall where a cylinder got dropped.
Personally I just put a hole in the ceiling after failing to vent a line before taking a spanner to a regulator

 

[Borek]

It was always my idea to use this approach (of one pressure vessel inside the other) not for vacuum, but for deep sea vehicules.