Question About Design Trade-offs for Backing Pumps

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.
Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Tue Jan 09, 2018 8:36 pm

I would just like to ask a quick question to the community to see what some of the experienced member's recommendations would be for my current dilemma. I have recently purchased a Robinair 15800 2 Stage 8CFM pump to use as a backing pump to my system. My diffusion pump is an Edwards EO4, with a backing requirement of at least 4CFM. However, I somehow failed to realize that the Robinair pump has a factory rated final vacuum of 35 microns. I don't know why I thought it was lower, but unfortunately I missed this point. Going through all of the forum posts related to backing pumps for diffusion pumps, this ultimate rated pressure makes me a bit uneasy now. I know it is acceptable, and can be used, but is certainly not ideal, especially if I want to achieve lower ultimate vacuum in my system. The pump and oil that I ordered are both brand new, however I know that it is still not guaranteed that I will reach the rated vacuum, and I have to assume that my system will perform worse than expected upon initial startup. I also should assume that performance could degrade over time as well, which could raise the minimum ultimate pressure. My roughing line is all KF25 hardware, and is extremely short, connecting to the diff pump backing port through a 90 degree isolation valve, KF25 tee, very short bellows, and to a KF25 90 degree foreline trap. You can see the layout at the bottom of this thread that I just recently posted in: viewtopic.php?f=18&t=12011.

Here is my dilemma: would you recommend to use the 8CFM pump capable of 35 micron, or go with a 6CFM pump capable of 15 micron? Is this pump currently sufficient, or is it safer to go with a lower micron capable pump, though at slightly reduced CFM, which also leads to the question if the extra 2CFM is enough overhead, or should I spend a lot more extra to go with an 8CFM pump capable of 15 micron? A large issue is cost here. I can return the pump I bought (still currently in its shipping package unopened) and proceed to purchasing an upgraded pump, though at a loss. However I am on an extremely tight budget, which I have been carefully planning out and researching parts to maximize my cost savings for months. Getting the new pump will undoubtedly require a sacrifice of high vacuum components that will make me most likely fall short of my goal for this year with this system, which is ultimately to pump down past the 10^-6 torr range, and go even lower with the aid of an ion pump on my chamber, and very well baked out, prepped, and outgassed for future experiments. However, if this current pump will greatly struggle to back the system to achieve down into the 10^-6 range, then it wouldn't really be worth using it anyway. The plus side to upgrading the pump is it gives me more overhead for ultimate backing pressure, and may just as well be better in the long run. I am more interested in ion beam and beam on target systems than a traditional fusor, so I will need the higher vacuum for my projects.

How big of a change would one expect in performance between 35 microns and say 15 microns at the inlet, assuming all other variables equal and that the system is reasonably well sealed and prepped?

Any thoughts or recommendations would be tremendously appreciated.

User avatar
Richard Hull
Site Admin
Posts: 10728
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Question About Design Trade-offs for Backing Pumps

Post by Richard Hull » Tue Jan 09, 2018 9:24 pm

if you can reach 35 microns, fine. However the rule is go as low as you can go on the foreline pump. A diff pump goes lower if you can go below 10 microns on the foreline at the inlet. As we have in the FAQs Find out what your pump can do with a known good TC gauge at the inlet. That is what the diff pump will see. (assuming you don't have 10 feet of pipe between the fore pump and the diff pump.)

I would return the pump. I am rather stuned that an 8CFM, 2 stage, Robinair won't go below 35 microns. I worry that first line U.S. companies go out of business and sell their name to the ChiComs, allowing a famous name to become crap. Worse, still, is a first line U.S. company stop manufacturing in their plants and import a ChiCom product with their name on it.

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.

Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Tue Jan 09, 2018 9:38 pm

Richard Hull,

Thank you very much for your fast reply. Note that I can't actually test the pump since that would require me to open the box, fill it with oil, i.e. essentially use it, in which the seller would not accept returns. The 35 microns is just the rating from the datasheet, and it always seems that vacuum systems generally do worse than expected due to a whole host of variables in the system. I was kind of leaning towards returning it now diving more into the subject, but I didn't want to do so if it would work fine. However, it appears that I may need to get a different pump.

That leaves me with a couple of replacement options. I can go with something like a Robinair Cool Vac 15600, which is 2 stage rated down to 15 microns @ 6CFM, which would not be that much more than the price I got my current pump. Or, I could go with something like a Yellow Jacket 93580, which is also rated to 15 microns or lower @ 8CFM, though at a significantly higher cost. While I can get used pumps off ebay that are supposedly much better ultimate vacuum rating (more scientific grade backing pumps), I'd rather not take any risks with potentially non-working used pumps since I gotta get all this right from the get-go otherwise I'll be out of funds until next year. I won't be getting to beam or plasma systems, but I want to spend this year prepping my vacuum system and taking measurements, as well as completing my calculations, simulations, and designs for the things I would like to test with it. Unfortunately, the higher the cost, the less I will be able to do with my current system. I still need at least one high vacuum gauge, and while I would like the ion pump, I may have to forego that depending on the cost of the roughing pump for now.

If you or anyone else has any recommendations or experience with either the above mentioned pumps I am looking at or other alternatives in the similar ebay-price range, I would be very grateful.

Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Wed Jan 10, 2018 12:23 am

Richard Hull,

I also see going through previous forum discussions that you have mentioned the yellow jacket brand pump several times that you seem very happy with its performance. Would it be safe to assume that a yellow jacket 6CFM pump would be a good replacement based on your experience with it? It costs more than the Robinair, but not as much as the Yellow Jacket 8CFM that I mentioned looking at before, so it is probably a good trade-off between quality and price for what I need. I also like that it includes a built in gauge, which could be very useful for quick troubleshooting too.

User avatar
Richard Hull
Site Admin
Posts: 10728
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Question About Design Trade-offs for Backing Pumps

Post by Richard Hull » Wed Jan 10, 2018 4:49 am

The 5 CFM Yellow jacket that I had was 15 years old and was 100% U.S. made. In spite of it having been used in HVAC work for God knows how long, it still pulled to 10 microns in the bell jar of fusor II. I can't speak for current product. The gauge is worthless and is in inches of vacuum. Only a good TC gauge should ever be used to test a pump.

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.

Jerry Biehler
Posts: 814
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: Question About Design Trade-offs for Backing Pumps

Post by Jerry Biehler » Wed Jan 10, 2018 10:50 am

Anything under 50 mtorr will be fine while running. Your chamber is so small you could get away with a pretty tiny roughing pump. You will need a good gauge to watch it, I am sticking convenctrons on the new control I am building for my big system.

Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Wed Jan 10, 2018 2:09 pm

Thank you both for your replies. I have decided to return the pump, and just sent it out this morning. I figure that even though I may be technically ok with it, better safe than sorry and give myself a little extra room to work with on the micron range for the roughing side. A pump rated to 35 microns will probably struggle more to stay under 50 than something rated for 15. Apparently the pump I got is also in the Robinair economy line, and is made in China, so it may not be the best quality. The yellow jacket pumps still seem to be made in the US, and unlike the Robinairs, they do offer a complete list of replacement parts for everything in the pump. Again, since it is technically rated to 15 microns it gives me some extra theoretical overhead to work with hopefully.

I also think that for this chamber, the 8CFM rating may be unnecessarily large. The diff pump I have requires a min of 4CFM to achieve its maximum throughput, 600L/s. However, the pipeline conductance and effective pumping speed of the system is so grossly under that, that I won't need even close to the full throughput. For my system, 600L/s or 100L/s, really does not change the final numbers too much anyway, so I could go with a smaller CFM pump.

I may end up going with the 6CFM yellow jacket. Its probably a good upgrade from the robinair, and may allow me to get my chamber ion pumped as well if I can negotiate prices on ebay low enough. At least the gauge, even though very inaccurate, would be a good rough indicator - if it is not completely bottomed out, then I know I got a bad leak somewhere in the line. I also like that it has a gas ballast, which the current pump I got does not include. I will be using a new thermocouple gauge in the line as well, which has been set into a KF25 adapter and helium leak checked by LDS Vacuum, so I'm not worried about leaks or if the gauge working.

I also just found out last night to a bit of good surprise that the foreline trap that I purchased is actually a molecular seive style trap. I originally thought it was a metal filter one, but it came completely intact with the basket filled with adsorbent material. I will probably need to refill it since I don't know how old it is, but I can get new zeolite for these traps from KJL for $25.00, which is cheaper than the metal screen stuff, and has the benefit of reducing oil backstreaming while also absorbing water, protecting both the diff pump and the roughing pump oil from contamination.

User avatar
Dennis P Brown
Posts: 1431
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Question About Design Trade-offs for Backing Pumps

Post by Dennis P Brown » Wed Jan 10, 2018 5:00 pm

If you are hoping to use a diffusion pump without a real cold trap your system might not get below 10^-6 torr. Worse, the oil vapor back flow from the DP will contaminate the system even if you gate the DP off after reaching low enough for the ion pump to start. You might need a dry ice/alcohol or similar temp trap on the diffusion pump but liquid nitrogen is better; however, if you have access to liquid nitrogen, a cryo-sorption pump is far better (using an absorber like a molecular sieve) than a mechanical pump and is ideal for for ultra clean systems that use an ion pump. While a turbo would also solve the issue, I guess that cost is prohibitive.

Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Wed Jan 10, 2018 5:35 pm

Dennis P Brown,

Thank you for your input regarding the diffusion pump. From what I have seen though, it appears that DP oils seem to be rated for much higher than just 10^-6 untrapped. I plan on using DC-705 or equivalent oils. From what I have read, it is suitable for systems in the range of 10^-9 to 10^-10 untrapped. The sources also say that traps or refrigeration are not necessary for some systems with this oil due to its vapor pressure and very low backstreaming rates:

https://www.tungsten.com/products/diffu ... fXEALw_wcB

https://inlandvacuum.com/portfolio/dow- ... ump-fluid/

Maybe I am missing something crucial, but I was originally going off the assumption that given if my system is built to close to ideal high vacuum sealing, minus the ultimate limit set by any viton (at 10^-8), I shouldn't really have much trouble getting over 10^-6 given I properly bake out, prep, and pump my system sufficiently well. If I can hit 10^-6 at this point I would be extremely happy. Long term my goal is higher, but for now 10^-6 is my first short term goal.

What I want for the ion pump is not to run anything when it is being ion pumped, but just sealing off the main chamber and keeping it pumped down at high vacuum levels so I can condition it and keep it clean and baked out for very long periods of time, so I don't need to run the diff and backing pumps all the time. So effectively the ion pump would be on only when I am not actively using the system for experiments. Ion pumps can be started as low as 10^-4, though really not ideal and is quite a struggle, so if I can hit 10^-5 or even 10^-6 at the end of an experiment I can switch it on, seal it off, turn off the diff pump, and keep my system clean and ready until next time.

Michael Bretti
Posts: 115
Joined: Tue Aug 01, 2017 4:58 pm
Real name: Michael Bretti

Re: Question About Design Trade-offs for Backing Pumps

Post by Michael Bretti » Wed Jan 10, 2018 7:05 pm

A bit of a correction to the above statement. While I generally will not be running with the ion pump except for keeping the system under vacuum when everything else is off and already pumped down, I would like to try and run it if I am doing tests with either electron guns or miniature thrusters.

Thinking more about it now though, since I won't really have the funds to get these systems built until next year (and after I run simulations and CAD everything out first), I might just put the ion pump on hold for now and focus on getting both my roughing side and high vacuum side rock solid and well running before adding more capabilities. I do have a spare 2.75" conflat with some small ceramic insulated feed-throughs, so what might be more economical short term instead is to use that to do either some plasma cleaning or build a simple titanium sublimation pump with that and a spare nipple section I have and play around with that for now.

Post Reply