Building a vacuum chamber

In summary, the conversation discusses the construction of a vacuum chamber with manipulation capabilities. The structural specifications are debated, with a proposed design of a 1x1x1 foot cube made of 0.25" thick carbon steel with windows and gloves. The necessary vacuum level for the chamber is also discussed, with higher levels requiring more complex and expensive materials. The potential danger of working with reactive elements, specifically cesium, is also brought up. Finally, there is a calculation of the deflection in the center of the chamber and its potential impact on the seals.
  • #1
Upsilon
32
0
I have been interested in constructing a vacuum chamber for quite some time, but I am not completely certain on what the structural specifications need to be so that it can hold up to the immense external pressure.

My main motivation for building a chamber like this is in the hope that I will be able to manipulate the inside of the chamber while it is in vacuum state. I am imagining a 1x1x1 foot cube made of 0.25" thick carbon steel. I will also cut out a 4x4" hole for a window on two of the sides, either out of Plexiglas or Lexan (thickness not yet determined). Also, on the left and right sides of the cube, there will be holes where layers of gloves will fit inside. These gloves will also have to withstand the atmospheric forces, so the outer glove will be made of some kind of heavy-duty welding glove bolted to the wall of the chamber. The inner glove will be some sort of airtight PVC glove that is sealed to the wall. This will provide a strong, relatively flexible yet airtight place for my hand to manipulate the internal of the chamber.

Assuming 16 psi as the atmospheric pressure near sea level where I live, each side of the chamber will need to withstand 2304 pounds of force for extended periods of time. The windows must hold 256 pounds each, and according to some rough calculations the gloves will need to hold around 700 pounds each.

I have all of the necessary materials to cut, weld, seal, bolt, etc, but I need to know if this project is plausible with the given parameters. I have never before seen a vacuum chamber with manipulation capabilities, so I really don't know what exactly I'm dealing with here. Does anyone know a bit more about this and could help me out?
 
Engineering news on Phys.org
  • #2
First of all, you never specified what vacuum level you want. Do you want just a low vacuum, or do you require high vacuum, ultra-high vacuum, etc.? Each of those require not only different material and different types of seals, but also different vacuum hygiene and pumping. Ultra-high vacuum (10-9 Torr or better) typically requires conflat-type seal using copper gaskets. Plexiglass will not do. You also have to be aware of materials with high outgassing rate.

The complexity, and the cost, go up considerably the higher the vacuum level you require.

Zz.
 
  • #3
ZapperZ said:
First of all, you never specified what vacuum level you want. Do you want just a low vacuum, or do you require high vacuum, ultra-high vacuum, etc.? Each of those require not only different material and different types of seals, but also different vacuum hygiene and pumping. Ultra-high vacuum (10-9 Torr or better) typically requires conflat-type seal using copper gaskets. Plexiglass will not do. You also have to be aware of materials with high outgassing rate.

The complexity, and the cost, go up considerably the higher the vacuum level you require.

Zz.

I suppose it depends on how powerful of a vacuum pump I can get within my price range. More importantly, my plans for this chamber will include isolation of highly reactive elements, so let's just say that I need a vacuum that will prevent cesium from spontaneously combusting.
 
  • #4
Upsilon said:
I suppose it depends on how powerful of a vacuum pump I can get within my price range. More importantly, my plans for this chamber will include isolation of highly reactive elements, so let's just say that I need a vacuum that will prevent cesium from spontaneously combusting.

I hope you know what you are doing. If you are using pure Cs out of a boule, then I hope you know that you have a very dangerous substance that can explode quite easily upon contact with enough oxygen/water vapor. Your "plexiglass" window won't stand a chance.

I don't know if you are doing this on your own, or part of your job. If it is the latter, then your own workplace should provide you with the necessary safety training and equipment. If it is the former, than I don't want to have any part of this.

Zz.
 
  • #5
ZapperZ said:
I hope you know what you are doing. If you are using pure Cs out of a boule, then I hope you know that you have a very dangerous substance that can explode quite easily upon contact with enough oxygen/water vapor. Your "plexiglass" window won't stand a chance.

I don't know if you are doing this on your own, or part of your job. If it is the latter, then your own workplace should provide you with the necessary safety training and equipment. If it is the former, than I don't want to have any part of this.

Zz.

I admit this is something I'm doing at home, but I realize the dangers. I'm not some kind of kitchen chemist though who is doing this on the counter and pours everything down the drain when finished. Even so, the suggestion of isolating cesium would only happen in the distant future, if at all. I only want to use it as a reference in case I want to do something like that in the future. I wouldn't ever possibly need a vacuum more potent than required to store cesium safely, so it serves as a good reference point to how pure of a vacuum I need.
 
  • #6
Since ZZ hasn't locked the thread as being a "dangerous activity", I guess it's harmless to start on some structural analysis.

Assume each side of your cube is a square plate supported at the edges, the deflection in the center will be about
$$.0442 Pa^4 / Et^3$$
where a is the length of the side, P the pressure, E young's modulus, and t the thickness.

In SI units (because I'm not an American!) take a = 0.3m, P = 10^5 Pa, E = 210 x 10^9 Pa, t = 0.0063m.
That gives a deflection of about 0.7mm (0.028 in), which doesn't sound much but it might be enough to crack open the seals between your glass windows and the steel plate.

http://www.roymech.co.uk/Useful_Tables/Mechanics/Plates.html

0.7mm is an under estimate (and possibly a seriously wrong underestimate) because it ignores the compressive stress in the plate, caused by the pressure on the four sides of the cube that are supporting it. But it should be enough to warn you that you have to design something like this properly, even it's going to be used for "non-hazardous" experiments.

(I don't know of a "formula" that take account of the compressive stress in the plate, and I'm not going to make a finite element model and run it for you just to see how wrong the 0.7mm is!)
 
  • #7
AlephZero said:
Since ZZ hasn't locked the thread as being a "dangerous activity", I guess it's harmless to start on some structural analysis.

Assume each side of your cube is a square plate supported at the edges, the deflection in the center will be about
$$.0442 Pa^4 / Et^3$$
where a is the length of the side, P the pressure, E young's modulus, and t the thickness.

In SI units (because I'm not an American!) take a = 0.3m, P = 10^5 Pa, E = 210 x 10^9 Pa, t = 0.0063m.
That gives a deflection of about 0.7mm (0.028 in), which doesn't sound much but it might be enough to crack open the seals between your glass windows and the steel plate.

http://www.roymech.co.uk/Useful_Tables/Mechanics/Plates.html

0.7mm is an under estimate (and possibly a seriously wrong underestimate) because it ignores the compressive stress in the plate, caused by the pressure on the four sides of the cube that are supporting it. But it should be enough to warn you that you have to design something like this properly, even it's going to be used for "non-hazardous" experiments.

(I don't know of a "formula" that take account of the compressive stress in the plate, and I'm not going to make a finite element model and run it for you just to see how wrong the 0.7mm is!)

I'm not particularly concerned about the steel itself bending significantly/snapping. I am mostly concerned with the welds and the gloves. I'm certainly not a professional welder so I am a bit skeptical on how they'll hold up. I could add some rebars on the inside edges if need be. As for the gloves, I'm fairly clueless.
I'm not completely sure if I accurately described the setup, so please let me know if you need more details. However, it shouldn't be too hard to calculate the maximum force for the gloves, as tensile strength of the material will be pretty much the only factor.
 
  • #8
The typical vacuum chamber is either a short cylinder, or a sphere ... thus eliminating a lot of problems.

In your case a short cylinder with a thick plate welded to the bottom would be better than a cube.

If your goal is to handle reactive materials you don't need a vacuum chamber: instead you only need to pump out the reactive atmosphere and replace it with argon. Then the pressures don't matter. You should over-pressurize it a little to help avoid leaks.

Unfortunately your gloves will always be a source of leakage; the danger is related to how much leakage.
 
  • #9
Upsilon said:
I'm not particularly concerned about the steel itself bending significantly/snapping.

I agree that "snapping the steel itself" is nothing to worry about given your proposed dimensions, but you should be concerned about bending, if you want to design windows into the device.

However, it shouldn't be too hard to calculate the maximum force for the gloves, as tensile strength of the material will be pretty much the only factor.

The total force on the glove is easy to calculate. It is just the area of the hole in the wall, times the pressure difference. The hard part is finding the stress in different parts of the glove. That will depend very much on the shape of the glove, and not much on the tensile strength of the glove material. You could make a start by modeling parts of the glove as hemisphseres and cylinders. But what happens where the "fingers" join to the "hand" would be hard to model.

But don't feel too despondent about all this negative stuff. The last thread I can remember on PF about designing a vacuum chamber was from somebody who wanted to make it mainly out of plywood :eek:

Given your apparent level of engineering knowledge, your first attempt will most likely fail, so be prepared for that "learning experience".
 
  • #10
AlephZero said:
But don't feel too despondent about all this negative stuff. The last thread I can remember on PF about designing a vacuum chamber was from somebody who wanted to make it mainly out of plywood :eek:

Plywood can actually work surprisingly well, provided you only need a very rough vacuum. A friend of mine made a plywood vacuum chamber (with one plexiglass side) to test barometric sensors, and it did quite well down to a vacuum of a few torr. It isn't perfect, but it works better than you would initially think (provided you seal the joints well).
 
  • #12
How much water vapor are you willing to present to the cesium?

I would think that you want hermetically sealed gloves (to seal off your hands) which work down to the target pressure level.

See http://wiki.answers.com/Q/What_happens_when_caesium_is_added_to_water
Your entire project seems to be quite hazardous, and your precautions don't seem to be taking into account the issues surrounding your plan.

Why are you fooling with pure cesium in the first place?
 
  • #13
UltrafastPED said:
How much water vapor are you willing to present to the cesium?

I would think that you want hermetically sealed gloves (to seal off your hands) which work down to the target pressure level.

See http://wiki.answers.com/Q/What_happens_when_caesium_is_added_to_water
Your entire project seems to be quite hazardous, and your precautions don't seem to be taking into account the issues surrounding your plan.

Why are you fooling with pure cesium in the first place?

As I mentioned earlier, my mention of cesium was mostly designed as a reference point to build this chamber. I may or may not try it eventually, but that would not be any time soon. I would not ever, for any reason, need a vacuum more pure than required to store cesium safely.
 
  • #14
As has been noted, you might be better off working with an inert gas atmosphere than a vacuum. If the main need is to inert the work space, pulling a vacuum is not the optimal solution.
As is, your gloves will need to hold 700lbs of pressure, which imho is the weakest part of your design. Why you assume welding gloves would resist blowing out better is not clear to me. They are not designed to cope with
15 psi internal pressure differential.
The astronauts all note that their gloves, pressurized to only 5 psi, make working a bear, as the hand gets exhausted very quickly. Your setup is 3x more demanding, at normal atmospheric pressure. You will find working in those gloves clumsy in the extreme.
 
  • #15
Well, I did some research and I found that there is a wonderful compound called propylene carbonate, whose properties allow simulated electrolysis of a molten salt. The salt, when dissolved in the propylene carbonate and electrolyzed, will break the salt up into its components. It is also non-reactive so the metal will sink, depending on its density.

Now I still want to make a vacuum chamber the same way, but I think I will also use it as an inert gas chamber. I'll just suck out all of the air and then let in a bunch of gas from an external tank. So the gloves will still need to withstand a vacuum, but I won't necessarily be working with them in vacuum state.
 
  • #16
There is no need to pump down your chamber, with all the headaches that carries with it.
Simply flush it with inert gas, nitrogen is dirt cheap, argon much more pricey but much less reactive and still cheap. Inert gas setups are a lot easier to deal with in all respects. You can set up a flow to filter out fumes or moisture easily enough. If you electrolyze cesium salts, you'll need that.
 
  • #17
etudiant said:
There is no need to pump down your chamber, with all the headaches that carries with it.
Simply flush it with inert gas, nitrogen is dirt cheap, argon much more pricey but much less reactive and still cheap. Inert gas setups are a lot easier to deal with in all respects. You can set up a flow to filter out fumes or moisture easily enough. If you electrolyze cesium salts, you'll need that.

I still will want to use the chamber for other vacuum purposes, and I don't particularly mind pumping out the air before filling it with a gas. On that note, I can't seem to find any actual argon anywhere except for those tiny overpriced canisters used for wine preservation. Does anyone know a good source?
 
  • #18
You need an industrial gas supply house for large Argon or Nitrogen cylinders. Try a welding supply shop to start.
I still think that the gloves will be the Achilles heel of your setup in a vacuum. Of course, you'll be able to see how it goes as you draw down the pressure. Nevertheless, I would be very leery of this, simply because if the gloves let go while your hands are inside them, you will seriously injure your hands. Smacking them together with 700 pounds of force will not be good for your fingers. Please be very careful.
 
  • #19
Alright, now to get back on track. Should I add rebars to the 4 vertical edges as added support for the welds? Also, what material should I use for the windows? I'm thinking of Lexan (polycarbonate). As for the gloves, does anyone have any relative idea on how they will hold up? Do I need something more rugged?,
 
  • #20
Alright, so I have more information regarding the vacuum pump. I have one that is around 20 years old; it's a Robinair 15101-b and is advertised to pump at 5 CFM, though I couldn't find a micron rating. I also have a newer, smaller one that pumps at 3 CFM and is rated to 25 microns. Which should I use?
 
  • #21
Last edited:

What is a vacuum chamber?

A vacuum chamber is a sealed container that has had all air and other gases removed from it, creating a low-pressure environment. It is commonly used in scientific, industrial, and medical applications for experiments, testing, and manufacturing processes.

Why is it important to build a vacuum chamber?

Building a vacuum chamber allows scientists to create and control a low-pressure environment for experiments and processes that require it. It can also help prevent contamination from air and other gases, which is crucial for certain scientific experiments and manufacturing processes.

What materials are typically used to build a vacuum chamber?

Vacuum chambers are typically made from materials that can withstand high temperatures and pressures, as well as being able to maintain an airtight seal. Common materials include stainless steel, aluminum, glass, and acrylic. The choice of material depends on the specific needs and requirements of the experiment or process.

How is a vacuum created in a vacuum chamber?

A vacuum is created in a vacuum chamber by removing all the air and other gases from the chamber. This can be done using a pump, such as a rotary vane pump or a turbo pump, which sucks out the air and creates a low-pressure environment. The level of vacuum can be controlled by adjusting the pump and monitoring the pressure using a pressure gauge.

What safety precautions should be taken when building and using a vacuum chamber?

When building and using a vacuum chamber, it is important to follow all safety precautions and guidelines. This may include wearing protective gear, such as gloves and eye protection, and being aware of the potential hazards of working with high pressures and temperatures. It is also important to regularly inspect and maintain the chamber to ensure it is in good working condition and prevent any accidents or malfunctions.

Similar threads

  • Mechanical Engineering
Replies
1
Views
1K
  • Mechanical Engineering
Replies
28
Views
7K
  • Mechanical Engineering
Replies
5
Views
1K
Replies
3
Views
3K
  • Mechanical Engineering
Replies
6
Views
2K
  • Classical Physics
Replies
4
Views
1K
  • General Engineering
3
Replies
99
Views
7K
  • Mechanical Engineering
Replies
12
Views
8K
Replies
4
Views
5K
Replies
1
Views
6K
Back
Top