Vacuum 'Accuracy' Check

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
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Vacuum 'Accuracy' Check

Post by Dennis P Brown »

The issue of accuracy for any vacuum gauge is something we depend on for the manufacture to get right. With older/surplus this isn't exactly something one can depend upon (especially the detector units). However, we do have the fusor - that is, most average sized fusors tend to run in the 5 -15 micron range. This can provide a sanity check for a gauge that is significantly off.

Another method is use two gauges but if both are surplus/unknowns one really doesn't then know.

So I decided to check my thermocouple (TC) gauge against an older cold cathode gauge (Varian). They differed a great deal - so that leads to the issue of which is then right? Luckily, I can borrowed a more modern cold cathode unit (an Edwards.) Using this I made comparisons among all three simultaneously (used a multi-port adapter.)

I used the fusor's plasma for a general cross check (I believe my fusor operates from 5 to 7 microns at 32 - 25 kV). The newer cold cathode gauge appeared to work/read values that agreed fairly well with the fusor operation range. The old Varian cold ion gauge nearly matched the Edwards gauge where the fusor operated (5 - 10 microns.) The TC gauge I used had issues (much too high.) So I adjusted its scale (it has a calibration screw.) So now my TC gauge agreed with the other two in the range the fusor operates.

Bottom-line: in the range I am most interested all gauges and the fusor operation voltages agreed to within a few microns. So I am satisfied my old TC gauge is now ok for the range I need to operate the fusor: 5 - 12 microns.

As a sanity check, I bottomed out my chamber (not super clean) with the turbo in 'standby mode' and the new Edwards gauge read mid 10^-5 torr while the older Varian said mid 10^-6 torr. No way my system could reach that low so I now know I can't trust the Varian below mid 10^-4 torr (that gauge has no easy access calibration and frankly, I don't care about that range. I now know what it really should be reading.)

While not critical to check a given vacuum gauges real values - because the fusor and its plasma determines the desire operation range, its nice to know that my TC gauge is reading fairly accurately in the range my fusor needs to operate.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum 'Accuracy' Check

Post by Richard Hull »

It is a pity that the TC gauge in the 15 micron and below range is crushed. The best tube that I have related to this issue is the 1518 tube, (The DV-4M is similar), which does have a slightly better low end spread. Naturally, I do not rely on this for accurate deuterium operating pressure readings, but use my 0.1 torr heated baratron gauge which is forever gas independent. I have used it to calibrate my TC gauge.

The pity is, we operate in a tiny range of pressures 1-30 microns. A range where a crossover point exists for the common, inexpensive, gas sensitive and specific gauges do not like to function with any sort of expanded scale.

The most linear and gas independent gauge in this realm is always going to be the capacitive gauge which also remains the most expensive of all gauges.

Leaving out this capacitive gauge, and If I was only limited to one gauge I would choose the reliable TC gauge using the 1518 tube and calibrate it using the table of voltage vs. pressure for several tubes in the "make your own TC gauge" article on the Bell Jar site. I think I gave that URL in the vacuum FAQs.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Vacuum 'Accuracy' Check

Post by Dennis P Brown »

Agree with you. My Cap gauge just is an utter pain so I gave up on it. It just would not remain calibrated - guess a bad unit. Hence, my dependence on TC's - my cold cathode unit just barely reads in the 10^-3 torr range. I guess I should have tried calibrating the Cap gauge again to see if it drifts as bad as I thought it did. Mine isn't heated (the unit) but the room is a fairly constant temp when I'm using it (even now, I am heating the lab up.) If I wasn't baking out the fusor, I'd just bite the bullet and install that gauge to check it*. Maybe tomorrow if I have time - if the weather holds today, and part of tomorrow, I need to install new strut assemblies on the front of my car - ugh. Not fun work unlike the fusor. ;)

* I could added it below the turbo/main gate valve w/o the install affecting the bake out. But then I'm dependent on the mechanical pump for calibration range and other gauges are a bit up stream from them (pass the turbo/gate valve.) Might be worth while to try once the lab warms up enough.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Vacuum 'Accuracy' Check

Post by Dennis P Brown »

Since I was working on the system accessories, I finally decided to check the Baratron (a capacitor type unit). First step was simply to check the connections to see if there was a problem. Accessing the readout unit and connection cable is a bit difficult but turns out necessary. A loose connection. So I reinstalled the vacuum detector sensor into the fusor and checked it against my various other gauges. This Baratron covers 10 torr to 1 micron. In the upper range it tracked exactly with a known gauge (10 torr down to 0.2 torr) and when I got below 100 microns it tracked very close (+/- 5 microns) with my TC gauge. Below 20 microns it and the TC agreed exactly down to 5 microns (below which the TC doesn't really read.)

However, with the turbo running (and throttled) and a cold cathode gauge installed (which has a tiny analog range between 10^-3 and 10^-2 torr) I was able to track it from 5 microns down to its bottom reading (1 micron) and it appears to be a bit off (never went to zero with the turbo pulling low 10^-4 torr.) Re-zeroing the Baratron sensor unit fixed that error (hopefully, temperature and other issues don't cause the readings to drift with time.)

While I still need to do some chamber plasma tests to confirm the operating range that will be performed next week since I have to make some mod's to my lower section of the cathode assembly for the fusor. Always something.

Just surprised I overlooked that issue awhile back - pays to check connections even if they are difficult to reach ;)

An aside - my primary ion gauge was fried due to a plasma flash over awhile ago. Since I have a cold cathode running, decided to check my spare ion gauge unit for function and accuracy. It apparently is working (and agree's with the cold cathode) so I'll continue to use it. It is old and a bit flaky so not convinced the unit will hold up.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum 'Accuracy' Check

Post by Richard Hull »

I have baratrons here with ratings of 100 torr, 50 torr, 10 torr, 1 torr and my favorite is the 0.1 torr that is on my fusor. According the literature on the 0.1 torr, it demands a heater and must be on for 4 hours prior to use or calibration to zero. All the rest are usable within the normal room temperature range. The assumption is that if you need readings below 1 micron, the 0.1 torr for which you spend the big bucks is expected to be accurate in the 4th decade down in the millivolt range. Thus, the heater is needed as the flexure of the capacitive diaphragm below 1 micron is hyper small and critical.

As long as your baratron is thermally stable near room temperature, zeroing at 10e-4 torr should be adequate to secure reasonable accuracy in the 4th decade down. TC gauges do crunch down below 10 microns and relying to +/- 5 microns in that area can be a crap shoot unless critically set and adjusted using a baratron reference over that tiny range, which might make the TC gauge so, adjusted at the low end, be a bit off at 100 microns.

The proper conditioning and working of a TC gauge 30 to 0 microns would demand an electronic thermocouple amplified millivolt solution at just the low end using interpolated tables generated using voltage readings over this low range based on the use of a baratron. Naturally, the TC gauge and its wiring become critical here as it is a thermally critical device. A boot over the TC gauge to insulate it and hair fine TC pin return wires would assist in keeping the thermocouple from non-vacuum related ambient, heat sinking, thermal conductivities. All of this is doable but is pushing the stock TC gauge into territory it was never designed for. It is to be remembered that for use in a fusor with no baratron or other sub-micron gauging system, that you would have to calibrate it with the TC tube in a deuterium atmosphere.
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Post Reply

Return to “Vacuum Technology (& FAQs)”