TEFLEX seals for ISO-K

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
CaveJohnson
Posts: 7
Joined: Mon Oct 29, 2018 4:37 pm
Real name:

TEFLEX seals for ISO-K

Post by CaveJohnson »

Hi !

I have three viton seals in my chamber at the moment, the rest is entirely metal sealed. It gets down to 7e-8 but id like to go a bit further just for the sake of the number.
In my modified TW290/20/20 Leybold used TEFLEX seals for the intermediate stages and they seem tempting for parts of my build. They are Teflon O-Rigs with an inner core made from fkm.

Has anyone tried them for uhv applications? Im curious if they will actually give an advantage over normal fkm seals that is worth the 10-15€ the each cost.

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

Re: TEFLEX seals for ISO-K

Post by Jerry Biehler »

Teflon is no good for atmosphere to uhh seals, it is permeable. Just stick with nitrile or viton. You can get metal seals for ISO flanges: https://www.pfeiffer-vacuum.com/en/prod ... PdoId=6609

Probably pretty expensive but and unless you are baking the crap out of everything and being very careful you are not going to get much lower.

Be happy with what you have. There is no need to go any lower for most scientific applications.
Michael Bretti
Posts: 175
Joined: Tue Aug 01, 2017 12:58 pm
Real name: Michael Bretti

Re: TEFLEX seals for ISO-K

Post by Michael Bretti »

Georg,

If you want to get lower with elastomer seals, then the seal needs to be chilled below ambient temperature. Viton, buna-N, butyl, and even teflon have been shown to be good to down to 10^-9 torr if chilled to about 6C or less. In order to accomplish this when between flanges in a system, both of the flanges must be chilled down to this point so that the overall temperature of the o-ring reaches this temperature. The ultimate limitation of ultimate pressure for a elastomer seal is determined by permeation of gases through the bulk material, which is directly related to temperature. Utilizing extremely heavy compression, in addition to chilled seal flanges with a properly baked and outgassed viton seal will get you down to 10^-9 torr. However, this is really absolute best case scenario, and requires a lot of prep work. At regular ambient conditions, you are already hitting the limit for permeation through viton, which means your system is impressively well conditioned. Depending on how large the seal is, for very large o-rings you can also achieve lower than typical pressures utilizing concentric o-rings with a channel in between them pumped down to 10 Torr. This limits the gases permeating through the primary seal and helps reduce overall permeation into the chamber. No reason why you can't push your numbers further, just takes some extra work to move past the limit you are hitting now. Best of luck pushing those numbers further into the UHV region!
CaveJohnson
Posts: 7
Joined: Mon Oct 29, 2018 4:37 pm
Real name:

Re: TEFLEX seals for ISO-K

Post by CaveJohnson »

Mhh,
thank you for these answers, it brought some stuff to my mind. I just assumed since they originally used these seals on the pump they have to be good for the application or they would not go for the expense. But i did not consider they dont seal against atmosphere but only between intermediate stages. Most likely they are used because the cartridge pump is usually used for mass spectroscopy and its not so much about sealing than about contamination and outgassing of the material itself. Here i guess teflon is very suitable.

I will try to ask leybold for the all metal seals. They have been very supportive so far, maybe i can get them for a reasonable price. If not i guess ill have to spend another day on the lathe ^^
Its just a little sad to have all metal construction and then ruin the advantage by having rubber at the pump flange, but the reason is that the tw290 is my best and largest pump, so i dont want it to be more or less permanently stuck to something by a cf flange.
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: TEFLEX seals for ISO-K

Post by John Futter »

Georg
you are doing about as well as a turbo can do in the real world
I'm not talking aboout the un-real ultimate vacuum they report on the datasheet
we use viton at work on ion pumped systems that go down to the low 10^-9 millibar and we do not cool the flanges so all orings are at lab temp of 21 degrees C.
Michael Bretti
Posts: 175
Joined: Tue Aug 01, 2017 12:58 pm
Real name: Michael Bretti

Re: TEFLEX seals for ISO-K

Post by Michael Bretti »

For reference, the numbers I found for chilled flanges in vacuum texts are in comparison for a system that can be pumped down to only 10^-8 normally. With chilling, the ultimate pressure dropped to 10^-9. I believe this was for diffusion pumped systems. While chilling would not see very much gains in a UHV system where small or few o-rings are used, for larger o-rings where permeation loads greatly increase with bulk material, the effects would become more noticeable. At that point however it may be more economical to go with the differentially pumped guard o-ring topology to reduce permeation for large diameters (if you can get a custom adapter made). I agree with John Futter, that you can certainly get lower with viton without chilling, however it depends on your setup. Ion pumping continuously for long periods of time would definitely get you there, assuming everything else is at optimal running conditions. As long as the o-ring sealing diameters are not too large, it should be fine. Our facility has also used viton with ion pumps down to 10^-9 mbar for very small seals as well, without any chilling. This of course depends on the feasibility of your setup and whether you can add an ion pump to your chamber, as well as the size of your seals. Fortunately there are a bunch of ion pumps floating around ebay for pretty cheap. They also would need a controller, but a controller could be built in a pinch if really needed. If you are absolutely stuck with large o-rings and cannot use an ion pump, chilled flanges can help drop your numbers some still, but again the effort to do this would probably be very large for the actual gain you are getting.
CaveJohnson
Posts: 7
Joined: Mon Oct 29, 2018 4:37 pm
Real name:

Re: TEFLEX seals for ISO-K

Post by CaveJohnson »

Thank you for your Answers!

Chilled flanges would be an idea for a short try after bakeout, though in this case i think condensation will become a very annoying problem. Since my flanges of interest are the pump flanges it will get the pump wet and i need to cool down the entire pump as well. Probably i will not go for that yet. I have another Chamber with an old tpu240 completely cf sealed, maybe i should just do my low pressure attempts wit this one. Though the bearing is problematic as its not the original one and im not sure for how long i will be able to run the pump in one go without having to relubricate the bearing. We will see. In this case i might as well add another cf100 and add the cryopump, then lower pressures should be possible without much baking and cooling.

One idea i just had is to plate the o-rings in metal, maybe that could work as well to at least reduce permeability. first applying a physical vapor layer of copper and then electroplating it until a certain thickness is reached that will not crumble off. Im not sure if it can work since the metal will not be pressed against the sealing surfaces with much pressure but who knows...
Post Reply

Return to “Vacuum Technology (& FAQs)”