A collaberation for gas handling gear

Stuff you made and how to make cool stuff

A collaberation for gas handling gear

Postby Doug Coulter » Sat Aug 13, 2011 2:13 pm

Finally, the weekend and I can do fun stuff! In this case, we've been trying to perfect our ability to deal with deuterium and hydrogen at very high (5 9's) purity, and not be stuck with just the one tank we get from the supplier. In the case of the very pure H we recently got, they didn't want to even sell us the tank, so we want to repackage it and get out of the lease as well.

Bill Fain found all this stuff - some very nice aluminum tanks originally for medical oxygen or firemen breathing apparatus, and some correct valves for hydrogen. The tanks were pretty "used" and nasty looking, while the valves were threaded for a larger tank, while the oxygen tanks have a thread "all their own". Obviously a little machining and fab has to be done to get to the goal.

Since Joe Jarski does precise machine work better than I, and this is a place where that really matters, Joe did the valves.
HValves.jpg
Valves before and after Joe's work

Luckily, these had enough extra meat on the stems to allow this. Now they fit perfectly. "Like buttah". He won't brag, so I'll do it for him.

Then I took a bunch of those ratty tanks, and cleaned them up on my lathe:
Tanks.jpg
Tanks, outsides cleaned up.


I now have to make a little punch/die to make the teflon gaskets - many were lost before the tanks got to me, and we found (the hard way) that viton isn't good enough -- slow permeation by the hydrogen. I will also be detail-cleaning the insides of these tanks, and then baking under vacuum before doing the gas transfers, to keep at least 4-5 of the 6 nines purity of the gas we have.
It's not easy to get these outgassed -- takes a long time through the valve. About a week of heating (not too hot!) and pumping each is required. Since we're working with hydrogen, I may put a few pellets of lithium into each to act as getters for residual air and water as well.

We like working with the small tanks for a variety of reasons. For one thing, in case of accident, you don't lose too much, or create the possibility of an enormous explosion. It also makes it much easier to keep track of how much you're using (or wasting). Further, those interested in getting some now have a way (or will shortly) though these can't ship legally at this point. They are mostly out of hydro for one thing, not rated for flammable gas and so forth. But better than nothing, and we've already jumped through the paperwork hoops to get the gases in the first place.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: A collaberation for gas handling gear

Postby chrismb » Sat Aug 13, 2011 4:09 pm

I've always had bad luck with sealing up gas lines gas-tight, so anything I will say here is just casual comments.. (Funnily, I tend to read folks having problems with vacuum leaks but never gas leaks, yet for me it is the opposite. I've never had a prob with vac so far, hope the luck holds up!, but I hate gas fittings' work.)

I got fed up with little leaks on my system to the point that I pressurised with helium then put the whole lot into a bath of water! Bottle, regulator, lines, the lot. Waited to spot the bubbles. Y'know, I reckon that is actually the most sensitive way of doing it - you can spot a sub-mm^3 bubble forming over a period of several minutes. Can you detect a ~0.0001sccm leak by any other means?

Anyhow, as I say probably best not to follow my practice as I always suffer leaks eventually.

One comment though - you really gonna use teflon gaskets and hope that H permeation is lower than viton?!

Can you rather use indium wire in the threads, or some other clever material?
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: A collaberation for gas handling gear

Postby Doug Coulter » Sat Aug 13, 2011 8:03 pm

They come with teflon gaskets, and those don't leak H (or D). Since I know that works, and the viton I tried to replace a missing one with doesn't....I just figured on going with what's know to work here. The cold flow properties of teflon are "just so" for a gasket in this device. A viton sealed one went from 400 psi to 125 just sitting for a couple of months, after having apparently passed the bubble test. These tanks were originally meant for 2500 psi, and the gasket had to be strong. I'm using them at 400 or so, which is what I feel safe with for the adapter that lets me take the air out and transfer gas into the tank after that.

If there's any sort of labyrinthine seal, you can leak a lot of gas before a bubble appears on the surface...and then you're never sure if it was just air trapped in there during assembly.

For vacuum systems of course the best way is to just use a mass spec. That gets you to really tiny leaks and usually finds them real quick. It's for that reason that the new beam device is going on a pump station that has one at first (my little upstairs fusor comes off, as I'm stealing a few parts from it too, like an 8 pin feedthrough), though I don't expect any special troubles with it.
You can get a very mass spec like effect with the good old mark 1 eyball and some gas that lights up a different color when leaked in too. I have a teachers hand spectroscope that helps with that one as well. I'm lucky (and poorer, it was expensive) to ahve a big tank of Ne, so that's what I use, or He if I have mostly air/water in there -- very noticeable color change in a discharge.

John Strong used to put a discharge tube in his system foreline to look at for this reason -- always enough pressure there to get a glow.

Copper flare fittings work great, hardly ever leak if prepared right. And of course you can always solder copper (and most other metals). Flex them too much, they'll start leaking though. I know of no type of plastic for flex pipe that really works well -- pretty much tried them all. Either H leaks out or air leaks in, seems like every time. I've had just about zero problem with plain old soldered copper here. It's always the threaded fittings that have issues. With a flare fitting and annealed copper, the copper itself makes a real good gasket and conforms to the mating surfaces nicely.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: A collaberation for gas handling gear

Postby Starfire » Sun Aug 14, 2011 6:38 am

Doug - I take it that you did not remove much material from the cylinders as it would reduce their max pressure - what amount and what reduction of working pressure do you expect?

They look like ' test gas ' cylinders
Starfire
 
Posts: 143
Joined: Thu Aug 05, 2010 4:26 pm
Location: North Ireland

Re: A collaberation for gas handling gear

Postby Doug Coulter » Sun Aug 14, 2011 7:24 pm

They were something over 1/4" thick originally, or .250"+. I took off about .050" (they weren't perfectly round) to clean them up. They started out rated at 2500 fill pressure. I'll be using them not over 400psi. That's partly because that's as far as I trust my other plumbing to do the transfer, but it's a happy number in general. As the big tanks get drawn down, I won't be able to even hit that -- my big D tank is about 550-600 psi now. Not sure what you mean by "test gas" though of course that's what they are now! Gas to test fusors and accelerators with! I've gotten the valves into a few of them, letting the thread seal goop get hard, might try to test-fill one tomorrow (for a bubble leak test) or the next day. Patience seems a good idea where glue type things are involved. I'll see how bad they outgas once the goop is good and solid. I really cleaned up the insides today. Water+ detergent, 4 rinses, rinse with ethyl alk, then put in front of a heater with compressed air blowing into them for a few hours.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: A collaberation for gas handling gear

Postby Doug Coulter » Tue Aug 16, 2011 12:59 pm

REady for action real soon on this. Here's the setup. I used a stainless steel cross, a ball valve in a vacuum line (hooks to the upstairs small fusor system) and some CGA fittings to make a gas transfer setup. The drill is, you open the ball valve, and the valve on the recipient tank, and pull vacuum on the whole thing. Once satisfied you're done outgassing (perhaps heating the new tank to make that go faster), you close the ball valve and open the supply tank valve -- slooowwwlly. To reduce the inrush (and let me control the pressure to within what the 400psi gage will take), I've drilled and tapped the CGA fitting on the supply side 1/4-20 and put a bolt in there, tight. That's still enough leak to flow a lot of gas, plenty fast enough. Right now, after 20 min, the vacuum system is reporting 1.4e-4 mbar. This system has a base of 4e-7, so there's a couple decades to go before gas transfer would be good -- if I want to keep the purity, which of course I do. Like in stock trading -- hurry up and wait is still the fastest way even if it's sometimes excruciating to do.

GasSetup.jpg
Gas setup and upstairs fusor


I might put a space heater or IR heating lamp on that small tank to speed things up a little...next time I get a sunny day, that is.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA


Return to Machining and Fabrication

Who is online

Users browsing this forum: No registered users and 5 guests

cron