Major vacuum problems

[Ryan Catalano]

When I first began running my system, I noticed that it only got down to around 70 microns max after a 2-5 minute pump down and refused to go any lower. The tolerable fore pressure for my diff pump, as stated in the pump manual, was 500 microns so I decided to attempt a test. I barely got the system below 1 micron. I also noted from a previous test with just the roughing pump that when valved off, the chamber pressure rises about 20 microns a second. I've been reading on the forums that even something like 1 micron every 10-15 minutes is a concern. I tried removing the NPT fittings pictured below but still the system bottom out to the exact same pressure with diff running as measured by my cold cathode gauge (a gauge which was tested against a fully calibrated BA hot cathode gauge by a grad student and myself on a turbo pumped system at Princeton Plasma Physics lab). I also tried valving it off again during this same test and noticed almost no change in the pressure rise. I also attempted to tighten all high vacuum connections along with applying grease to the foreline hose connections and still no change.
(I'm using a Varian HSA diff pump and a Precision D-75 mechanical pump (6 microns deadheaded); my gauges include a Inficon pilot plus pirani and a Edwards AIM-S cold cathode gauge; I do also vent the system to atmosphere after every test.)
I'm confident that there must be some kind of massive leak in the system. Is this most likely what it is? If so, where could it be coming from? Also if I can't repair the leaks, do I still have an optimal pressure to back fill with deuterium for fusion?

[Jack Puntawong]

Ryan,

from my experience, the leak is most likely coming from the area between the 2.75" conflat cross and your pirani gauge. I never had any problem with NW flange. However, the leaks might also be coming from the conflat flange if the bolt is not tighten enough. Apparently, you have to be really careful to tighten up the bolt in a patern bit by bit, not just in one go. I once had this problem so I change all of my copper gasket to viton and the problem is solved. Also, on the side note, you don't really need the cold cathode gauge. I have one my self and the range that it is operating is far beyond what most fusor need.(Unless you have a specific experiment in mind).

The leak of 20 micron/second is pretty massive. You will be wasting lots of deuterium in the process and there will be a lot of air contamination which would reduce the fusion count too. Your system is pretty small so the fix should be easy.

I would recommend you to close off the right angle valve, take off NW tee, replace it with the connector from your roughing pump. That way if there is still a leak, you'll know that the leaks comes from the side with the conflat and the gauge.

Good luck!

Kunakorn (Jack) Puntawong

[Jerry Biehler]

Get some acetone in a squirt bottle and dribble some on the joints and watch the pressure gauge for a change.

[John Futter]

Ryan
I personally detest teflon tape for sealing vacuum joints.
Many times I have done this only for it to leak.
For threaded joints or any suspect joint I use Loctite 290 it wicks in to the smallest high vacuum leal.
This stuff has little or no vapour pressure once set and is good down to 10 to the minus 8 millibar /torr.
I'm also not a fan of those refrigeration compression joints you are using but a little loctite on them prior to assembly should allow them to work.

Word of caution Acetone dissolves the loctite very quickly so use ethyl alcohol to assist is finding those leaks

[Dennis P Brown]

A leak that causes a two stage oil pump to stall at 70 microns in such a small system as well as exhibiting a leak rate of 20 microns a second will be tough to detect with any "vacuum" gauge and a liquid since the rise in pressure is so fast anyway. Have you considered pressurizing the system and listening for escaping air? Also, have you tired removing components (if possible) and checking the remaining system? That has worked for me.

Check all the KF fitting surfaces for scratches using a magnifying glass. Any scratch can cause a serious vacuum leak. Redressing the a surface or replacement is the only real fix if the KF surface is scratched. Vacuum grease is not the solution.

By the way, Teflon tape isn't used for sealing a connector - ever; it allows the threads to be broken apart more easily and prevents galling - if the joint needs to be sealed, than it isn't something that belongs in your system. Get a proper vacuum connector system for the gauge to connect to your system - consider making your own (if you have access to the equipment and have the skill.)

Also, I have never had much success with those brass compression fittings and copper tubing (they fail so easily with over tightening and/or re-tightening); rather, I use all stainless steel "Swagelok" fittings and SS tubing for my system - expensive but these work extremely well for high vac. Besides, you have SS connector fittings and vacuum chamber so way add such cheap (read poor performing) components like copper/brass? Besides, the cost of SS fittings is small compared to those vacuum components. With SS fittings/tubing I get to the low 10-6 torr easily and leak tests with "Duster" spray shows no leakage either.

One last point - I've seen bad leaks with older cold cathode detector heads (yours looks new, however) - consider removing that and retesting the system. Those aren't very difficult hard to reseal if that is the leak or one of them.