Progress Update

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
User avatar
Adam Ingle
Posts: 27
Joined: Mon Oct 11, 2010 10:04 pm
Real name: Adam Ingle
Location: Springfield, MO

Progress Update

Post by Adam Ingle » Wed Jan 19, 2011 5:13 am

Thought it's been a while since I've checked in with the group, I've been quite busy on my end piecing together all of the parts I've amassed in my scrounging.

Tonight I got enough of the pieces to put my chamber together and run the whole vacuum system for the first time. The pumps have been tested by themselves, but not together and not with the chamber in place. The chamber itself is not fully complete, on the right hand side in the picture were the convectron gauge is connected there will be an additional NW25 tee. This will permit the gauge to be mounted vertically for better accuracy and it will also provide a connection point for the deuterium feed system. Additionally, at the top of the chamber there is currently a blank flange until I receive the 30kv hv feedthrough I ordered from MDC.

My vacuum equipment:
Welch DirecTorr 7 8920
Pfeiffer Balzers TPU-170 and TCP-300 controller
Granville-Phillips 275 Convectron Gauge
4.5 OD cross with copper gaskets etc.

Here's how the first run went:
Started the backing pump with the throttle/valve fully open. Ran it for about 25 min. The convectron gauge bottomed out at about 25 mTorr.

I started the turbo pump. In less than 5 seconds the pressure dropped to 8 mTorr.

I ran the turbo pump for an additional 15 min and adjusted the throttle/valve completely closed. The gauge remained at 8 mTorr.

After shutting down the turbo pump and waiting for it to fully spin down, I then shut off the backing pump.

With the throttle/valve closed, the pressure rose 1 mTorr every 3 to 5 min. My throttle/valve is used and I suspect probably out gasses the most out of the whole chamber when isolated.

At about 14 mTorr, I slowly opened the valve/throttle and the pressure began to rise beyond 300 mTorr in about 5 min. and continued to climb.

I did have some questions, however. While running my backing pump, according to my gauge, the pressure would seem to bottom out, then jump down. Fluctuate, then jump down again. I would have more expected it to be a more linear decrease than the many rapid changes.

Additionally, when I kicked on my turbo pump, the gauge dropped from the 25 mTorr to 8 mTorr in less than 5 seconds. But then it stopped. Is this my gauge?

I'm sure I have quite a bit of out-gassing from chamber contamination. I'm working on getting that taken care of. Additionally, I could also see a possibility of a leak, but a hard stop at 8 mTorr seems a little strange. I would have thought my turbo pump would have easily pulled down 1e-5 before stalling due to a leak. Perhaps not.

Thanks again for the help and advice!
Attachments
DSC02063 (Small).JPG
DSC02063 (Small).JPG (46.67 KiB) Viewed 2101 times
DSC02062 (Small).JPG
DSC02062 (Small).JPG (46.49 KiB) Viewed 2101 times
DSC02060 (Small).JPG
DSC02060 (Small).JPG (50.84 KiB) Viewed 2101 times
DSC02059 (Small).JPG
DSC02059 (Small).JPG (48.45 KiB) Viewed 2101 times

User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 11:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: Progress Update

Post by Carl Willis » Wed Jan 19, 2011 5:35 am

Hi Adam,

Very nice gear you have there.

If you start the turbo and the gauge indication stops falling at 8 mtorr, but the mechanical pump generates 25 mtorr by itself, I suggest you need to calibrate the gauge readout. I have never used a Convectron, but the process for zeroing out the controller is probably straightforward. You should do this when the gauge is attached to a known good vacuum system capable of well below the minimum measurable pressure.

Minor observation: always firmly secure chambers, pumps, etc. that are attached to other components with a bellows flex-hose...that's potentially an accident waiting to happen.

I can't really conclude that your vacuum is being degraded by serious leaks based on the brief measurements with a probably-unreliable gauge. I am sure there will be copious outgassing for several hours unless you actually took apart and properly washed the turbo, the CF hardware, and the flex hose (notorious for being dirt sponges) before assembly. Outgassing can be implicated as a cause of poor vacuum (and simultaneously accelerated) by playing a mild torch flame or heat gun on the components of the system while observing the gauge. Pressure might reasonably surge an order or two in magnitude if the parts are of average off-the-shelf cleanliness. Pumping for a few hours and adding some mild heat should put a major dent in such a problem.

Nice photos, and nice system.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277

Chris Trent
Posts: 291
Joined: Tue Jan 17, 2006 9:24 pm
Real name:

Re: Progress Update

Post by Chris Trent » Wed Jan 19, 2011 6:59 am

Looks good, Great work.

I would let the vacuum system run for a few hours and then see what your gauge reads. A leak is always possible, however in a brand new system contamination could be serious problem. If it keeps improving during a long pump down then that's the likely culprit.

Like Carl mentioned, a miscalibrated gauge could be the issue as well.

-Chris

User avatar
Steven Sesselmann
Posts: 2111
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: Progress Update

Post by Steven Sesselmann » Wed Jan 19, 2011 8:20 am

Adam,

Nice minimalistic setup.

Even if you have read it before, take a look at the FAQ posts in the vacuum forum, Richard covers most of the basic vacuum issues that you and everyone else struggle with.

Creating a good vacuum was a much bigger problem for me, than I expected, when I first started out, I thought it would be simple to suck the air out of a chamber....but I was wrong


Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

User avatar
Adam Ingle
Posts: 27
Joined: Mon Oct 11, 2010 10:04 pm
Real name: Adam Ingle
Location: Springfield, MO

Re: Progress Update

Post by Adam Ingle » Wed Jan 19, 2011 9:37 pm

Thank you guys for the added insight. I'm amazed that the further I go in my experience, I continue to re-read the FAQs and keep grasping more of not only the content but the concept behind it all.

I'll be re-calibrating my gauges tonight and running the pump longer to out-gas more of the contaminants.

Also, looking again at my pictures, I noticed that I have my turbo pump lower than the backing pump. I can only assume that would not be good for oil vapor. I'll be correcting that too.

User avatar
Adam Ingle
Posts: 27
Joined: Mon Oct 11, 2010 10:04 pm
Real name: Adam Ingle
Location: Springfield, MO

Re: Progress Update

Post by Adam Ingle » Wed Jan 19, 2011 9:42 pm

Carl,

Thanks for the info. I'll be setting some heat to the system tonight in addition to a longer running time and see how things shake-out.

I'm probably just mis-reading the statement about the observation you made. Could you tell me what you mean? I defiantly want to avoid any caveats or pitfalls.

Thanks again!

- Adam

User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 11:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: Progress Update

Post by Carl Willis » Wed Jan 19, 2011 10:35 pm

Hi Adam,

I just meant to say that the presence or absence of significant leaks isn't revealed by your testing so far because of what is probably a gauge readout error. In fact, you probably have no major leaks. On the other hand, outgassing always attends the inaugural operation of a vacuum system and can take many hours of pumping to fade into the background.

To speed up the process, wash parts and bake them right before use. CF parts I typically wash with ordinary dish soap and water, sometimes preceded or followed by alcohol or acetone if there is "gunk," then bake with a high flame on my kitchen's gas rangetop. Stainless parts suffer some minor discoloration from the flame but I could care less about cosmetics. Viton gaskets I put in an aluminum pie pan in the toaster oven at maybe 350 F for an hour or so. I avoid using alcohol or other organic solvents on Viton.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277

User avatar
Adam Ingle
Posts: 27
Joined: Mon Oct 11, 2010 10:04 pm
Real name: Adam Ingle
Location: Springfield, MO

Re: Progress Update

Post by Adam Ingle » Wed Jan 19, 2011 10:50 pm

Hi, Carl!

I should have been more specific in my reply, sorry about that.

You had said, "Minor observation: always firmly secure chambers, pumps, etc. that are attached to other components with a bellows flex-hose...that's potentially an accident waiting to happen."

I was just curious about that.

To discoloration of the SS; I made the mistake of acting before researching and decided to try a bake-out in my oven. Needless to say the oil and other "gunk" in the oven baked themselves to my parts. Inside and out. So I may have gotten rid of the water vapor, but I gained likely more in terms of out-gassing with the added "glaze" my parts received. Looked bronze by the time I removed them. Good scrubbing, chemical bath and another good scrubbing got them cleaned. I have read many more better suggestions, such as yours that I hope to be able to try if time permits tonight.

Good notes about the Viton gaskets. I'll assuredly need to be mindful of them as well.

Thanks again and have a good one!

- Adam

DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 5:13 am
Real name:

Re: Progress Update

Post by DaveC » Thu Jan 20, 2011 8:45 pm

Hi Adam - Nice gear.

To just add a note to Carl's observation about calibration:

The usual method for calibration of these "higher pressure" gages, is to connect them, exactly as you have done, and then bring up the mechanical and then the turbo pump. After the turbo gets to full speed, (and I assume from your description that it did), the presssure in the system will be way below the minimum pressure the Convectron gage can read.

Depending on which gage you have, and the usual calibration and cleanliness factors, a Convectron will read down to about 1 milliTorr more or less ( in other words, 1 micron Hg). Since they have a heated element inside, and operate in the mechanical pump vacuum range, it's not unusual for them to have oil contamination.

I have sucessfully cleaned them with a fill of methanol or acetone, followed by a carefully done dry out. Heavily gunked gages, could need several rinses. A little heat gun on the outside, will speed up the dry out process. But keep the gage in your gloved hand so it doesn't get too warm. It's got a plasticsocket on the end.

As Carl described, you need to find the calibration adjustment on the gage controller/readout... and bring the reading to zero, after the system has pumped with the turbo on, for a while. I'd guess the gage is reading about 8 mTorr high....which isn't bad, anyway.

You won't be able to read high vacuum conditions until you get an ion gage of some sort either the cold cathode, or heated filament type. Then you should see pressures down to 10 -7 Torr, fairly promptly, and probably down into the -8's after a few hours or overnight pumping.

The bake- out Carl mentions is very important.... for all "new" or used gear. You'll save hours of pumpdown time, with some simple cleaning, as he has described.

Dave Cooper

User avatar
Rich Feldman
Posts: 1242
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Progress Update

Post by Rich Feldman » Thu Jan 20, 2011 8:54 pm

Dave Cooper wrote:
>
> I have sucessfully cleaned them with a fill of methanol or acetone, followed by a carefully done dry out. Heavily gunked gages, could need several rinses.
>
> Dave Cooper

The Convectron manual talks about the solvent soak, with a warning.
Do not shake the gauge tube when it's -partly- filled with solvent;
the sloshing can immediately damage the gold-plated tungsten filament.
All models are wrong; some models are useful. -- George Box

Post Reply