Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

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
ian_krase
Posts: 534
Joined: Mon Nov 28, 2016 7:48 am
Real name: Ian Krase

Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by ian_krase » Thu Mar 02, 2017 6:39 am

In the months I've been working with vacuum I've gone though a respectable amount of Dust-Off when leak hunting.

While a Pirani gauge or other gas dependent gauge works pretty well for finding the leaks, what's really sensitive is a glow discharge in your chamber. Dust-Off (in my case, difluoromethane) sucked through a leak will cause an instantaneous and substantial change in the color and geometry of a glow discharge, even when a gauge barely registers.

However, I'm a little worried I might be decomposing the fluorocarbon into Nasty Stuff (such as HF!), and then condensing it in my pump oil or something. This is a bit alarming given that vacuum chambers tend to contain teflon, viton, fluoro greases, etc, and my gaskets are pretty exposed. A 1 liter chamber of 500 mtorr of difluoromethane is about 1 mg which... isn't *nothing*.

Is this a real concern?

John Futter
Posts: 1428
Joined: Thu Apr 22, 2004 2:29 am
Real name:
Contact:

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by John Futter » Thu Mar 02, 2017 7:59 am

I would be more worried by your home made oil baffle on your backing pump
pleural Pneumonia caused by oil vapour ingestion is a real threat
1mg of HF in the air is going to help your teeth no end with no permanent damage to your skeleton.

I work with HF regularly and bone density x-rays every 3 years show no abnormalities and I Know I ingest much more than a mg or two every time I play with it

ie water bath @ 80 degreess C and 50% HF used to etch various samples day after day. ( one way to remove Si oxides from wafers before experiments)

ian_krase
Posts: 534
Joined: Mon Nov 28, 2016 7:48 am
Real name: Ian Krase

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by ian_krase » Thu Mar 02, 2017 8:12 am

Yeah. Thus far I haven't had a reason either to run the pump for really long periods or with much gas load beyond the 5 second pumpdown of my tiny chamber. Plus it's directly opposite a window.


I'm not so much worried about small quantities of non-inert fluorostuff in the air as I am about turning my pump oil into a corrosive mess (either to pumps or to hands).

Jerry Biehler
Posts: 864
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by Jerry Biehler » Thu Mar 02, 2017 9:10 am

If you are that worried about it stop using it. Just get a small bottle of helium for balloons or find the small disposable cylinders of argon that are used for welding. Both will make a noticeable change on your gauge readout with a leak.

User avatar
Dennis P Brown
Posts: 1704
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by Dennis P Brown » Thu Mar 02, 2017 1:42 pm

There is something fundamentally wrong about a simple vacuum system that requires that level of leak detection - that is, repeated checks that find new leak points. If one can get an ionization gauge working then your leaks aren't too serious for a fusor.

But back to the issue - any small system has a very finite number of leakage points. Either one has bad surfaces or isn't using standard connectors or components (then there is your problem, not leak detection.) Otherwise, leaks, when found are fixed and the problem is over.

Of course, virtual leaks are of a totally different class and external detection methods are useless.

Real HF is a very toxic substance and one I avoid (when I can ... .) However, the spray cans with fluoride has those molecules extremely tightly bounded and little is converted in your vacuum system because little ever gets into the vacuum system via any leak that still allows an ionization gauge to operate (and if one has no hot filimant no fluorine can be created.) Also, 99.9999% of what you spray is just going into the atmosphere and is nontoxic and harmless.

Jeroen Vriesman
Posts: 277
Joined: Mon Feb 07, 2011 9:08 pm
Real name: Jeroen Vriesman
Location: Netherlands
Contact:

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by Jeroen Vriesman » Thu Mar 02, 2017 2:03 pm

Some time ago I tried leak detection with the glow colour change method.

I tried with argon and helium, but even letting in some helium or argon on purpose didn't change the colour.

I didn't try difluoromethane, what is it that makes this gas so effective? Are there any other gases we should or should not try?

ian_krase
Posts: 534
Joined: Mon Nov 28, 2016 7:48 am
Real name: Ian Krase

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by ian_krase » Thu Mar 02, 2017 4:58 pm

To be clear, I only have a forepump and a pirani. We're talking 5-500 microns, mostly closer to 5-25 lately.

Most of leaks have been bell jar gaskets or extremely disappointing attempts at feedthroughs, along with the occasional insufficiently compressed high durometer O ring.



I'm going to have a real test gas after Easter.

User avatar
Dennis P Brown
Posts: 1704
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Plasma + Fluorocarbon -> Leak-finding and... bad stuff?

Post by Dennis P Brown » Thu Mar 02, 2017 6:34 pm

Using liquid ethanol is by far the easiest method to find leaks at pressures in that range. Dust off is better used with an ionization gauge. Besides, the alcohol often helps to seal the system if trace (only!) amounts of vacuum grease are present.

Bottom line - when using "home-made" and rough vacuum connectors, vacuum leakage is a common problem that generally is hard to over come. One is far away better off finding deals on ebay for proper connectors and feed-thru's (time wise.)

Post Reply