Is baking worth it?

For Short Term Learning Discussions ONLY. This area is for CURSORY questions and connecting with other users ONLY. ALL technical contributions need to be made in the appropriate forums and NOT HERE. All posts are temporary and will be deleted within weeks or months. You should have already search the extensive FAQs in each of the forums before posting here as your question may already be answered.
Lukas Springer
Posts: 15
Joined: Sat Mar 03, 2018 9:37 pm
Real name: Lukas Springer
Contact:

Is baking worth it?

Post by Lukas Springer » Sun Mar 11, 2018 12:24 pm

Hi!

I am currently able to reach 2e-6 mbar with my vacuum setup and a small oil diffusion pump.
I know to get further down I have to bake out the chamber, but a proper heating mantle can be very expensive.
Since I don't need that high of a vacuum to do fusion I was wondering if baking would bring any advantage to my setup, such as less water in the chamber or if it is not necessary.

-Lukas

User avatar
Richard Hull
Site Admin
Posts: 10986
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Is baking worth it?

Post by Richard Hull » Sun Mar 11, 2018 12:40 pm

Baking is not necessary. I never baked. If you have a nice beefy supply, you can do a glow cleaning at the higher pressure of 20 microns with about 4kv and 30ma. This will effectively bombard the chamber walls and heat the chamber at 120 watts. This is best done with the diff pump not started and running just the mechanical pump. I tend to open all my valving and blast the chamber as the mechanical pump works to evacuate the entire system.

This can take 5-10 minutes on a cold start after a long down time...(weeks, months). If there is a heavy water and contaminant load, I might reach 15 microns and never have the mechanical pump go low enough to kill glow mode. I just blast away for at least 10 more minutes and then turn on the diff pump. As the diff pump heats, the pressure rises as the boiler nears the boiling point of water and then drops off shortly after that as it starts to pump and glow mode ends as the system goes sub-micron.

You have to watch your current and the grid as you do not want to melt it due to bombardment/glow mode current.

All of this assumes you have a fully metered system and a view port.

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
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Bob Reite
Posts: 281
Joined: Mon Aug 26, 2013 1:03 am
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: Is baking worth it?

Post by Bob Reite » Mon Mar 12, 2018 2:35 am

I "bake" after a long time to atmosphere by running the system on air with about 10 microns pressure and about 30 KV at 8 MA, My grid can take that much input without hassle. After running like that for a couple of hours the chamber is too hot to touch. I then shut down the HV supply and open the valve to the turbo all the way and let the system run like that overnight. After doing that I'm down to 1.5 X 10-6 torr, more than good enough for fusion work.
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.

Andrew Seltzman
Posts: 698
Joined: Mon Feb 02, 2004 1:02 am
Real name: Andrew Seltzman
Contact:

Re: Is baking worth it?

Post by Andrew Seltzman » Mon Mar 12, 2018 3:28 am

I bake with resistive heaters on the shell to about 70-100C; it definitely desorbs water from the system. I recommend this method since it dosen't sputter grid material onto the HV feedthrough like running HV on the grid for the sole purpose of heating the chamber. Towards the end of the bakeout I also run HV in a ~10mTorr deuterium plasma at ~17ma and a few kV to bombard the inner walls with plasma. These cleaning methods noticeably improve neutron production rates.
Andrew Seltzman
www.rtftechnologies.org

prestonbarrows
Posts: 211
Joined: Sun Jun 24, 2012 5:27 am
Real name:

Re: Is baking worth it?

Post by prestonbarrows » Sat Mar 17, 2018 3:55 am

Baking essentially reduces your H2O partial pressure.
In a typical system, water is the most significant contributor to pressure (assuming there are no leaks).

Getting down to ~1E-6 Torr with a diff pump is just on the boarder of needing to care at all about water adsorption, with like a few hours timeline.
Most likely, you could leave your vacuum system running passivly at room temperature for a couple weeks or you could do a proper bakeout.

The difficulty with a 'proper' bakeout is that the ENTIRE vacuum system must be raised to well past red hot. Any little bit that is left cold eliminates the benefit of the whole thing. To do this requires metal-to-metal conflat seals everywhere, significant thermal insulation(every damn bit covered), and a large enough power supply (thousands of Watts of more).

Are you sure everything in the chain can withstand this?
Are you sure it is worth just passively waiting an extra few days?

Timbits-to-donuts your diff pump does not have metal seals and this whole inquiry is just silly.

Andrew Seltzman
Posts: 698
Joined: Mon Feb 02, 2004 1:02 am
Real name: Andrew Seltzman
Contact:

Re: Is baking worth it?

Post by Andrew Seltzman » Sat Mar 17, 2018 1:43 pm

For a "proper bakeout" (limited to 450C by the conflat o-rings), these high temperatures are only used when desorbing hydrogen from the stainless parts. This in only required for systems pushing the 1e-13torr range, for baking out H2O a little over 100C is a reasonable temperature.

You don't have to bake out the entire system to see a noticeable improvement, unbaked parts will still desorb H2O from the surface, but the total amount of desorption in the system will be reduced, so you will achieve a higher vacuum for a given pumping rate.

Baking without metal seals is still worth it, though H2O does permeate viton pretty readily, so you will have to rebake before every run. This is the procedure I use (~40min bake before running, with 160W of tape heaters on my 6" dia fusor shell), and it does yield noticeable improvements in both vacuum and neutron rates.

Here is how I arranged my heaters
viewtopic.php?f=6&t=10294&hilit=mark+3&start=50

you can find them on ebay by searching for:
Polyimide heater 12v
Andrew Seltzman
www.rtftechnologies.org

Michael Bretti
Posts: 154
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Is baking worth it?

Post by Michael Bretti » Sat Mar 17, 2018 2:55 pm

For water vapor in a system, very high temperature is not critical for desorption to occur to see an effect, especially when only working in the high-vacuum regime up to 10^-7 torr. Higher vacuums, and hydrogen becomes a more dominant factor, as mentioned above. Any thermal energy will help the process, although there is a certain threshold energy needed to break the bonds sufficiently. For systems that are limited by other o-rings and sensitive electronics, 150C is a generally recommended max number for indefinite bakeout times. Higher temperatures will allow for faster bake-out times in general, but the system can be baked out at 100-150C with good results. Depending on what vacuum level you want to achieve and how clean it is initially, this could take longer. Viton o-rings can also be pre-baked in a proper oven in air at this temperature, but is needed over a period of several days straight to see an effect, and should be installed immediately after baking. Vacuum baking will reduce these times to a few hours. While even gradients of temperature are better, it is not necessary to see a benefit at non-ultra-high vacuum levels. If using heat tapes, you can, and probably should, wrap more layers of insulation and tin foil to help increase efficiency of the process and retain heat, as well as spread it more evenly.

I have posted this in another forum topic, but I will post it for reference here as an example since it is directly relevant. For my system, I needed to know the ultimate pressure during pumpdown. This was calculated for three cases: unbaked and pumped for 1hr, unbaked and pumped for >24 hrs, and baked and pumped for > 24 hrs, utilizing calculations for water vapor as the system gas. These numbers were then used to calculate the maximum gas handling loads for air, argon, and deuterium for my system, at a wide range of vacuum levels. As you can see from the numbers, baking does produce a noticeable effect, and under certain run parameters, is necessary for me to achieve certain gas loads at certain vacuum levels. Long-term pumping is also noticeably beneficial as well, though baking is better with pumping.


However, total surface area vs. pumping speed needs to be considered. There is a critical pumping factor of 0.01 L/sec/cm^2 that provides a practical limit, where the calculated factor should be higher than this number. Lower than this, and the desorption rate rapidly falls off, and enters a realm where cryopumping is needed for water vapor to be effectively pumped at a reasonable time period. This number has also been calculated for my system, and even worst-case scenario it is above the critical pumping factor for my chamber, so desorption vs. pumping time and speed is within a reasonable limit for my system.

Another thing to note is the effect on permeability of o-rings due to temperature. Baking will increase the permeability of an o-ring while heat is applied, but will desorb the adsobed water as well. If the surface of the o-ring is chilled, say to 0C between the two adjacent plates, the permeability greatly reduces, and can allow for o-rings to be operated an order magnitude lower pressure than generally rated permeability limits of 10^-8 torr under normal temperatures.

Alternatively, though it is not discussed or mentioned much on these forums, UV desorption can be very beneficial and attain faster results. A power level of 0.4mw / cm^2 of of UV over the internal surface area will reduce the pumpdown time by a factor of 2 compared with if no UV is applied. This can be achieved with hot cathode mercury discharge tubes used in ozone production. The two peaks associated with this type of light source are 254nm (90%) and 185nm (10%). The 185nm wavelength is the wavelength that is primarily effective for the removal and desorption of the water vapor. This can be run directly in the chamber, or through an optically acceptable viewport, though directly in the chamber is always preferable. However, with this method, direct light is not necessary on all surfaces due to reflections, and will still be effective.

Lukas Springer
Posts: 15
Joined: Sat Mar 03, 2018 9:37 pm
Real name: Lukas Springer
Contact:

Re: Is baking worth it?

Post by Lukas Springer » Sat Mar 17, 2018 4:33 pm

Thanks for all the information!

I'll see how hot I can get the chamber with plasma once I've connected my new power supply (-30kV 100mA).
When my neutron detection setup is ready I will make a couple of runs with baking / no baking / with UV / no UV etc and report back.
Since my Boron corona tube seems to act weird I went ahead and ordered a ZnS(AG)/B2O3 scintillator, but shipping from the ukraine takes some time.

Michael Bretti
Posts: 154
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Is baking worth it?

Post by Michael Bretti » Sat Mar 17, 2018 7:33 pm

A quick additional note on the effects of UV desorption. Not all wavelengths will work or be equally effective for water, only certain ones. I'm not sure what wavelength a fusor emits depending on the gas run. UV in general however should help break down other volatiles that may be present in the system.

Running the fusor for glow discharge cleaning, as mentioned often here, will also help in general. For pump down and vacuum conditioning, you don't need to run the fusor while baking. Baking again can be accomplished pretty cheap and very effectively with heat tape and insulation. You can also use the insulation on your chamber to help heating efficiency if heating through just fusor operation alone, though for very long hours of conditioning heat tape is probably preferable and safer.

If you have viton or glass/ceramic feedthroughs, you will also want to monitor the temperature. Especially for glass/ceramic feedthroughs or viewports you will want to increase temperature slowly not to thermally stress the bonds between the metal and glass/ceramic seals. If you want to push into the 10^-7 torr range, pumping and baking times start to get significantly longer unless using much higher temperatures. Gas loading however starts to limit you if using o-rings without special modifications. You can however calculate your theoretical ultimate pressure to get an estimate if the level you want to achieve is even feasible, though it is a very long and involved process. Good luck with your vacuum conditioning!

John Futter
Posts: 1321
Joined: Thu Apr 22, 2004 2:29 am
Real name:
Contact:

Re: Is baking worth it?

Post by John Futter » Sun Mar 18, 2018 12:59 am

Michael et al
when you get energy above about 3.5 eV bonds are broken there are a few which require more than this ie N2
but 185nm is 6.4eV and the oxygen in water is converted to O3 ozone so the water molecule is dissassociated into component atoms that are energetic

now when you establish your glow plasma you will have atoms energised up to your acceleration voltage so all molecules will be dissassociated.
a 5-10kv discharge will excite some atoms to that voltage and they will bombard other molecules and break them down
so lets look at the enegy

10 kv glow clean 10,000eV
UV clean 3-7eV
bakeout using a heater tape 0.0001eV

which do you think will do the fastest cleaning job???

But all are valid

Post Reply