Vacuum leak question

Every fusor and fusion system seems to need a vacuum. This area is for detailed discussion of vacuum systems, materials, gauging, etc. related to fusor or fusion research.
Post Reply
Corby Dawson
Posts: 41
Joined: Mon Nov 23, 2015 9:31 pm
Real name: Corby Dawson

Vacuum leak question

Post by Corby Dawson »

I am putting together a small system to do some testing.
It consists of a 20 liter/second ion pump and various conflat fittings.
The ion pump is connected to a conflat manual valve, currently closed, and is running at about 1X10-7 Torr.
Connected to the valve is a flange with 3 radially mounted VCR ports, two are blanked off and the third is connected via a VCR to NPT adaptor and an NPT union to a thermocouple guage.
The flange then connects to a 3" conflat straight section which connects to the center of a conflat "T". One end of the "T" has a flange with a swagelok bellows valve that connects to my turbo pumping syatem. The other end of the "T" is blanked off.
After running the turbo until the TC guage reads zero I close the Swagelok to evaluate how well the plumbing is holding the vacuum.
Unfortunately there seems to be a leak somewhere!
Here is the time versus pressure reading once the Swagelok valve is closed.
Minutes pressure (milliTorr)
0 0
1 4
2 5
3 8
4 10
5 12
10 23
90 190

All the 2 3/4" conflat connections are using viton gaskets.
I tried spraying dustoff into the fittings to see if the TC guage reading would change but no luck!
Any suggestions on how to isolate the culprit?

Thanks,

Corby Dawson
David Kunkle
Posts: 284
Joined: Thu Apr 17, 2014 12:43 pm
Real name: David Kunkle

Re: Vacuum leak question

Post by David Kunkle »

When I built my chamber and had leakage issues, I had the best luck finding leaks by pressurizing my chamber. Cobble together whatever fittings you need to attach your system to a compressor- even a bicycle pump would probably work. Since you already have an NPT fitting, hardware store should do the trick. I had mine as high as 40 psi, then put soapy water on all the joints and watch them carefully. Can take a while for the smaller leaks to blow a bubble.
If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: Vacuum leak question

Post by Jerry Biehler »

Did you bake the chamber out? That could just be water vapor. And if you are pumping the system down through a little bellows valve you will be pumping a very long time before you get much H2O out.
Corby Dawson
Posts: 41
Joined: Mon Nov 23, 2015 9:31 pm
Real name: Corby Dawson

Re: Vacuum leak question

Post by Corby Dawson »

There is no "chamber" just the volume of the conflat fittings, Since the "leak" rate is somewhat linear over time I assumed a true leak versus outgassing. Would water vapor in such a tiny volume give the rate shown? I.E. 190mTorr in 90 minutes.
Corby Dawson
Posts: 41
Joined: Mon Nov 23, 2015 9:31 pm
Real name: Corby Dawson

Re: Vacuum leak question

Post by Corby Dawson »

OK, when I got home 12 hours after closing the Swagelok valve the pressure had risen to 1 Torr!
I decided to take the 2 NPT joints apart and redo them with the addition of Loctite 290.
Could not find any locally but placed an order which should come Friday.
If still leaking do you think the ion pump can handle that leak rate? If so I could then use the dustoff and observe the ion pump current to find the leak.
Has anyone tried using a homemade Viton gasket in place of a VCR metal gasket? I'm not sure of one of the VCR joints and trying the Viton would help eliminate that.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Vacuum leak question

Post by Dennis P Brown »

Are you just testing for leaks for fusor system? If so, one hardly needs 10^-7 torr so just use the turbo. As for NPT joints for a high vac system, not a good idea. If for a fusor, a different matter but the ion pump isn't what you want for the system testing because getting really low requires a super clean system and baking is the fastest way to get there.

Ion pumps are not meant for anything but HV work with no real leaks or back pressure.

Your numbers look a lot like simple out gassing. Any system that has been open to the atmosphere (read a lot of water vapor) can easily raise your system to 1 torr in that time period. Maybe a leak but also, maybe out gassing (mostly.)

The simplest method is: run the turbo and get to some base pressure where a high vac gauge works; then use a can of "Dust Off" and spray all areas. Your high vac gauge will jump at a leak. If you can't pull such a small volume down with a turbo to below 10^-5 torr (or takes a long time), you have rather serious leaks and it isn't out gassing alone.

The FAQ's cover a lot of good info and you need to search and read posts related to leaks/out gassing; really.
Corby Dawson
Posts: 41
Joined: Mon Nov 23, 2015 9:31 pm
Real name: Corby Dawson

Re: Vacuum leak question

Post by Corby Dawson »

Thanks for the advice everyone!
Dennis, it's not for a Fusor, I need to keep a small 2" X 3" glass "bulb" under a vacuum and then flow a very low pressure Hydrogen through the bulb while ionizing it with an RF discharge.
This is for a Hydrogen Maser I am repairing.
I did search leaks and outgassing and you were right, lots of good stuff there.

Jerry, you were correct! I cracked open the isolation valve on the ion pump and while observing the current I heated up all the conflat stuff with a heat gun until it was too hot to touch. The current rapidly rose to around 50ma and then started dropping. after an hour the current was down to 1ma or 3X10-6 Torr.

I repeated the bakeout and the current only rose to 1.8ma this time and is dropping down past 1ma as I write this!

Spraying dustoff around all the joints fails to make any change in the ion pump current!
Hooray.
prestonbarrows
Posts: 211
Joined: Sun Jun 24, 2012 1:27 am
Real name:

Re: Vacuum leak question

Post by prestonbarrows »

I think most of this has already been covered, if only tangentially, but to sum up a bit.

A good way to differentiate a 'real' leak versus outgassing is to blank off the chamber and record the pressure over time. Outgassing is dependant on pressure so will result in a non-linear, concave-down, pressure rise in time. A true leak to atmosphere will result in a choked flow scenario where the mass flow rate is essentially independent of the pressure resulting in a linear change in pressure over time.

Repeating this test a few times in a row while keeping the chamber sealed is also illustrative. If there is an real leak to atmosphere, the rate of rise will remain constant over each cycle. If due to outgassing and conditioning, the rate of rise will be reduced on each cycle.

Baking out will help with outgassing alot. The majority of outgassing on a recently opened chamber will be water. Water takes a lot of energy to remove from the chamber surface so pumping it is energy limited unless you raise the temperature quite high. Simple heat guns will work in a pinch, but a heat tape wrapped in some fiberglass or even aluminum foil insulation will be much more efficient. Take care not to get hot enough to damage any elastomer gaskets you might be using. On a well sealed, but wet, system you should see the pressure skyrocket then crash when you bake out the chamber while pumping on it. Expect the final cold pressure to be a few orders of magnitude lower than where you get with the heaters on.

Again, if you cycle the bakeout procedure multiple times, you should see a reduction in the rise in pressure and final base pressure over each cycle.

If you have a real leak to atmosphere, large leaks can be found by spraying methanol over any seals and looking for a spike in pressure; smaller leaks can be found by spraying non-nitrogen gas such as helium or 'canned air' and monitoring a gas-dependent gauge such as a cold cathode, Baird-Alpert, ionization, or any other gauge else with a gas correction factor. An RGA is the best if you can get your hands on one.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Vacuum leak question

Post by Dennis P Brown »

Very good post but I suggest avoiding methanol- that stuff is rather toxic for people and bad for the environment; rubbing alcohol or even high proof alcohol can be used (yes, they have water.) But dust off works as well for me and has the advantage that electrical connections can be sprayed safely.
Corby Dawson
Posts: 41
Joined: Mon Nov 23, 2015 9:31 pm
Real name: Corby Dawson

Re: Vacuum leak question

Post by Corby Dawson »

Thought I'd update!
After a few bakeouts I'm stable now at around 4X10-7 Torr!
Hooray and thanks for the inputs.

Corby
prestonbarrows
Posts: 211
Joined: Sun Jun 24, 2012 1:27 am
Real name:

Re: Vacuum leak question

Post by prestonbarrows »

Dennis P Brown wrote:Very good post but I suggest avoiding methanol- that stuff is rather toxic for people and bad for the environment; rubbing alcohol or even high proof alcohol can be used (yes, they have water.) But dust off works as well for me and has the advantage that electrical connections can be sprayed safely.
Yes, this is a good point for the uninitiated.

Methanol is fairly toxic, you should not make a habit of lathering it over your skin. However, it is quite safe with modest procedures centered around wearing gloves and avoiding contact.

Ethanol is much better from a functional/safety standpoint. unfortunately, due to silly laws in the US, pure ethanol is difficult/impossible to come by. The average DIYer will find it impossible. At best, the average citizen will be able to purchase ethanol spiked with methanol to the point that it is poisonous, aka denatured alcohol.

Of course, it is well within the means of anyone technically minded enough to construct a fusor to construct an apparatus which produces laboratory grade pure ethanol. But that is a topic for another corner of the web.
Dan Knapp
Posts: 402
Joined: Wed Aug 06, 2008 8:34 am
Real name: Dan Knapp

Re: Vacuum leak question

Post by Dan Knapp »

You can buy 95% (190 proof) ethanol in many liquor stores as "Everclear," although you're paying the tax to be able to drink it. The other 5% is water. The 95% concentration is the highest you can get by simple distillation, because the 95% ethanol/5% water azeotrope boils lower than pure ethanol. The 95% ethanol is fine for cleaning purposes. To get absolute (100%) ethanol, you add some benzene and distill off the benzene-water-ethanol azeotrope to get rid of the water. The remaining ethanol contains traces of benzene, which is why one never should drink absolute ethanol unless you are certain it has been further processed to remove traces of benzene. This is probably more than you wanted to know; but the point was that producing pure ethanol by the amateur is not as simple as it might appear, and that 95% ethanol is readily available (to those of legal age) if you're willing to pay the taxes.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum leak question

Post by Richard Hull »

Another way to get to absolute ethanol is to bake some Copper Sulfate crystals in an oven to chase off the water. (they go from a beautiful blue to snow white). Immediately transfer the baked material to a tightly stoppered small bottle of everclear. The copper sulfate will turn blue rapidly. Leave it overnight. Decant the now water reduced everclear to another tighly stoppered small bottle. Rebake the copper sulfate until snow white again and drop it into the new bottle of higher ethanol and if that sees the copper sulfate turn blue or bluish then repeat until the baked copper sulfate remains snow white in the bottom of the bottle of ethanol. The white copper sulfate in the bottom remains as a great indicator of the state of your absolute ethanol into the future. Absolute ethanol sucks water from the atmosphere. It is up to you to keep it absolute by quickly opening and closing the storage bottle and seeing that the copper sulfate remains white over time.

Note: the above is an old microscopist trick and works great. For our purposes, straight everclear would do fine by itself.

I have had to go to Maryland or North Carolina to buy everclear as the ABC stores here stopped handling it after some 3 or 4 darling little collegiate "cup cakes" at UVA killed themselves drinking purple passions some years ago at some frat party. Fortunately, I make regular trips to those states yearly on other business.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Post Reply

Return to “Vacuum Technology (& FAQs)”