RGA spectrum indicates backstreaming of oil?

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
RobertMendelsohn
Posts: 85
Joined: Mon Jul 20, 2020 4:58 pm
Real name: Robert Mendelsohn

RGA spectrum indicates backstreaming of oil?

Post by RobertMendelsohn »

Hello,

I am trying to debug my vacuum chamber (and diffusion pump) with a residual gas analyzer (see attached spectrum). The largest peak is at 20amu, which is typically associated with neon, doubly ionized argon, and Fomblin diffusion pump oil - I strongly suspect the latter since there are not the other peaks associated with neon or argon. The other peaks present possibly indicate oil from the mechanical pump - if you have an alternative explanation, I'd love to hear it. I am going to install a zeolite trap between the mechanical pump and diffusion pump. There's still a lot of water in the chamber, but I have no gear for bakeout. It looks like there is not a leak, since the N2/O2 ratios are not close to that of air, and leak checking with canned 'air' and isopropyl didn't budge the RGA.

I'm only getting ~4*10^-5 torr with a 2.5" air cooled diffusion pump, which seems odd. With just the mechanical pump it gets to 1 micron. There is no line of site between the diffusion pump and chamber, so I am going to try cooling it with dry ice and acetone, as well as thermally insulating the base of the diffusion pump, and see if that does anything. The longer I run it, the more the pressure starts to rise (albeit very slowly).

Debugging this chamber has been frustrating! If you have any advice or ideas about why the chamber is not pumping down, why I'm getting this spectrum, or any useful diagnostics let me know!

And if anyone has any questions about RGAs, feel free to message me. I got mine for around $1000 with $150 shipping, and the seller has more if you are interested.

Sincerely,

Robert
Attachments
IMG_3491.jpg
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: RGA spectrum indicates backstreaming of oil?

Post by Dennis P Brown »

My system, exposed to air for a long time and once the micro leak was sealed, could only reach the exact same level with a similar diffusion pump. The behavior - after pulling vacuum and over night sitting under vacuum (pump off), improved (i.e. slowed) the rise in the system with the gate valve closed but not the ultimate pressure. Out gassing, especially water on surfaces, would very much look like what you are seeing.

A bake out is what you need to prove/disprove but that you don't have. You could order a heat tape or maybe just use a hair dryer to heat all the surfaces a bit just to see if some improvement is achieved.
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: RGA spectrum indicates backstreaming of oil?

Post by richnormand »

I agree with Dennis that a good bakeout is needed before going forward.
Do you have a cold trap for the diff pump?

To blame the diff pump oil it would be useful to do a 100amu scan as many of the tale-tell signs will be in the 50 to 90amu range.
Diff pump oil will have a primary at 45 followed with 59 43 41...
Polyphenylethers has a primary at 446 but cracking peaks at 77 and 50 and a small at 39. A cluster at around 60 to 65 too for some.
I don't really know the pattern for Fomblin but I am sure its related or available.

All the present peaks (oxigen, nitrogen,water) are usual including the carbon monoxide with primary at 28/42to49 due to the filaments from the gauges.
Peak at 20???? Neon or some combo?

Can your RGA go to 100?

Hope that helps.
Cheers and keep fighting!
RobertMendelsohn
Posts: 85
Joined: Mon Jul 20, 2020 4:58 pm
Real name: Robert Mendelsohn

Re: RGA spectrum indicates backstreaming of oil?

Post by RobertMendelsohn »

Thank you all.

Trapping the diffusion pump doesn't seem to be doing anything.

The peak at 20 is what's so odd. When running really high resolution scans, it seems like it could be 19.5, and there is a peak for 38/39. I see some notes that this could be common fragments of hydrocarbons or fluorinated chemicals. How could the vast majority of remaining atoms at 4*10^-5 torr be oil fragments??

I ran the scan again with more AMU (up to 100) and it revealed no more peaks.
Hitting it with the blow drier did increase the pressure by 20%, so it looks like a bakeout is necessary - although it did not affect the 20amu peak at all.

If anyone figures out what the issue is, your steak at HEAS is on me!
RobertMendelsohn
Posts: 85
Joined: Mon Jul 20, 2020 4:58 pm
Real name: Robert Mendelsohn

Re: RGA spectrum indicates backstreaming of oil?

Post by RobertMendelsohn »

An update: after some very careful measurements of the partial pressures, readings from other gauges and MFCs and known pumping speeds, I've identified a couple of issues.
One of them is that the pressure reading drops when the copper refrigeration coil that attaches to the chamber (1/8 swagelok) is removed - I suspect this as a source of contamination.
Repeated attachment and detachment from the chamber confirmed that this affected the pressure. Since the pressure went DOWN when this tubing was removed, is it possible that the copper refrigeration coil has some coating? It is generic and from Home Depot.

I'm pumping it down for a couple of hours (pressure slowly going down) and I'm going to try heating it up while flowing inert gas. Any other ideas still welcome! Vacuum debugging is no joke, even with fancy gauges, connections etc... thanks for the help gentlemen.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: RGA spectrum indicates backstreaming of oil?

Post by Dennis P Brown »

As someone kindly reminded me, using "Dust Off" is very useful in locating very small leaks - worked on my system. I assume said tubing is inside the vacuum system. Handling tubing and the very fact it is in open air before it is placed under vacuum means the tubing will out gas a great deal. Cleaning it with solvents and maybe heating in an oven, then only handling with clean gloves will reduce out gassing from its surface when under vacuum.
RobertMendelsohn
Posts: 85
Joined: Mon Jul 20, 2020 4:58 pm
Real name: Robert Mendelsohn

Re: RGA spectrum indicates backstreaming of oil?

Post by RobertMendelsohn »

An update: the issue was... the RGA itself. I started noticing that small fragments <30amu were being recorded properly, but the large peaks weren't. After speaking with the seller and the manufacturer, the seller agreed to replace it. Weird!! Thanks for your help everyone!
Post Reply

Return to “Vacuum Technology (& FAQs)”