Vacuum pump

[JacksonCzarniak09]

Hello,

Would I get away with a high quality vacuum pump, or incorporating a secondary pump necessary? Thanks

[Richard Hull]

A good two stage pump is all you need for a forepump and a demo fusor. For fusion you will need a diffusion or turbo pump as a followup to the mechanical pump. Read the FAQs in the vacuum forum.

Richard Hull

[JoeBallantyne]

Sorry, Richard, but per your own experience, which you have discussed previously on this board, one does NOT require a 2 pump setup in order to do fusion.

A single very good mechanical pump will suffice.

I have done fusion many times using only a Welch 1402 mechanical pump that pulls down to ~1 micron.

I wish we would STOP telling people that they HAVE to have a two pump setup in order to do fusion.

IT IS NOT REQUIRED TO HAVE A DIFFUSION OR A TURBO PUMP TO DO FUSION WITH A STANDARD GLOW DISCHARGE FUSOR.

Period.

Must one have a VERY good mechanical pump? Yes.

They do exist. I have 4 of them. (Welch 1402s that pull down to 2 microns or less.)

And I haven't rebuilt any of them. All of them were previously used.

Joe.

(Actually I have yet to acquire a 1402 that HASN'T pulled down to less than 2 microns. And I am not trying to do anything special when I purchase them.)

The factory spec on a new Welch 1402, is that it will pull down to 0.1 microns. So pulling down to 1, is 10x worse than that spec.

New ones do pull down to 0.1 microns or even less. I have a new style Welch 1402 that was little used, and I have bottomed out my MKS 901p vacuum gauge at 1e-5 torr with that pump. (1e-5 torr is 0.01 microns)

[Dennis P Brown]

The vast majority of used vacuum pumps struggle to get below a few microns; yes, mine gets below 2 microns but that is not representative of what many used pumps can do that I've seen newbie's operate here.

Also, having a clean system is not what most beginners for a variety of reasons can generally achieve due to having a poor pump, questionable oil, dirty system, used vac connectors and even a marginable neutron detector and low end power supply.

There are certainly advantages to having a high vac system. Two such reasons are, first and ironically, they enable the user to detect/find minor leaks much easier; and second, the fusor will become more free from water vapor with a good high vac pump down before operation. This may also reduce the amount of deuterium needed to flush/start the fusor for stable operation. Hence, for the added cost of a high vac the fusor performance can improved and operation could be more stable.

Yes, with a very good two stage pump, and a user able to maintain a clean system (high end components and well sealed/rarely opened), then a operator does not require a turbo or DP. But for most novices these high vac systems can certainly improve their odds of getting a stable plasma (for many a major issue) and better neutron counts.

Considering the cost of a good two stage pump, vacuum gauges, the fusor chamber, connectors, power supply, and especially the neutron detection system, a used turbo or a very low cost DP is not a significant extra purchase; especially since it will improve overall fusor performance.

[Richard Hull]

I never did fusion without a second auxiliary. I used a micro maze which pulled the fusor III submicron. I still wasted almost half a $300 cylinder of deuterium gas trying to drag out the remaining garbage to reach a pure Deuterium gas in the fusor. The amazing micromaze is no longer sold which can even take a 5 micron bottomed mechanical pump sub micron. Got Deuterium to waste? Recommend a newbie with a filthy, leaky system to use just a plus ultra mechanical pump.

I learned quickly the value of a second real pump. A second pump got me to the mega mark. Fusor III was abandoned quickly and fusor IV was done correctly, right from the start using a Diff pump that I bought for $25 at one of the HEAS events in 2002.

Is a turbo pump better? Not really, based on my experience. It buys you expensive repairs and tangles your vacuum system with electronics you may or may not understand or be able to fix. By comparison Diff pumps, operated correctly are cheaper and easily fixed with the only electrical component a mere 120 volt heater coil.

I am now on a turbo. Why? I got a turnkey system, pump, controller and matching cables for $150, again at HEAS in 2010. I never used it until 2017 as the diff pump was doing great. Once installed it was fabulous! However, in 2020, the controller died from internal fusor arcing. Fortunately, as a retired electronics engineer, I fixed it myself for the price of 3 TTL IC chips (about $3.00). Plus, a week's worth of my time was involved tracking down the issue.

For the newbie with zero skills in vacuum, electronics, and fusor operation, a fusor is, as we have seen, often a bridge too far. A frustrating first effort that sees them leave and sell their stuff in the trading post or in a second pass with money and gained experience go through to a win.

Money, of course talks and best intentions often walk.

Dennis is correct, newbs are often set adrift in a world they have no concept related to best practices. What does the average newbie bring to the effort? That is the question.

We live and we learn with each iteration.

Richard Hull

[JoeBallantyne]

I'm not going to let this one go without futher comment.

Its funny how one notices what one wants from others posts. When Richard posted on my fusor build thread about my using just a mechanical pump to do fusion, what I noticed was that he used a mechanical pump. I didn't notice or pay attention to the fact that he was also using a secondary micromaze device with that pump. Sorry for that oversight Richard, I guess I focused on what I wanted to see.

Be that as it may, there have been a few folks, myself included, who at some point did fusion with just a mechanical pump.

See the following posts:

viewtopic.php?p=82065#p82065

viewtopic.php?p=94053#p94053

viewtopic.php?p=98063#p98063

While I agree with Dennis that a significant number of new folks have dirty, leaky systems, I think there is also a VERY inexpensive fix for that. Its called, push the new folks to build fusors using KF/QF/NW parts. Preferably KF50 sized. They are inexpensive, easy to assemble, easy to clean to a basic first order level (rubber gloves, hot water, palmolive dish soap, and a blue backed scrub sponge (NOT the green scrub sponge which will scratch the stainless)), and as long as the flange sealing faces are in decent shape, hold an excellent vacuum. I built my first fusor with a mix of used and new KF parts, and it doesn't leak. (Ok, it may possibly have a VERY small leak rate, but that rate appears to normally be less than the outgassing rate.)

I've been lurking on fusor.net since about 2008, and have been periodically looking at fusor related equipment on ebay for the duration. When I first started looking, diffusion pumps were plentiful and very cheap ($100 or much less). That is no longer the case. Last week I checked, and most currently available diffusion pumps are $400+ and are rather large. Turbo pumps that are claimed to be working generally start about $800 minimum, and that is usually without a controller or cable.

Probably the cheapest place one could get a diffusion or turbo pump is from a fellow fusioneer, or at Richard's HEAS. But still, once you have the pump, you immediately are faced with interfacing it with your system. Which for a duffusion pump in many cases means you may need some kind of custom adapter - or at least an ASA adapter. More cash outlay and time as you try to figure out what you actually need, and wait for it to show up on ebay. You also need a gate valve or equivalent so you can throttle the now very high vacuum pulled by the pump on your system. Another several hundred to a thousand dollar expense + time trying to find one that will connect to your pump. You also need to plumb your mechanical pump to the diffusion pump/turbo, for which you need more adapters and bellows. Further cash outlay and delay waiting to find the right parts. Then if you got a diffusion pump, you need to buy VERY expensive diffusion pump oil, that goes for ~$600 a gallon. Fortunately if you get a smaller pump you likely need only a pint or so. Realistically for most folks, when all is said and done, adding a second hand high vacuum pump is going to cost on the order of $1000 minimum, likely significantly more, and likely will take a month or two just to get all the parts. If not longer.

Furthermore, the pump might have issues, especially if it is a turbo, which then need to be identified and resolved. More time, expense and frustration.

I really think that we do a disfavor to new members by pushing them all to try to do 2 pump setups before attempting fusion. It would be better IMO if we pushed them to make sure to buy a mechanical pump with the performance needed to do fusion without a second very high vacuum pump. Then they have the ability to do fusion first, and upgrade their pumping system later if they choose to.

So, since I am not shutting up about this, I've decided to put up as well, and checked on Amazon to see if there are any reasonably priced new mechanical pumps that at least claim on paper to pull a vacuum that would work for doing fusion in a small fusor. Which I arbitrarily decided was being able to pull a vacuum of 3 microns or less.

It turns out that there are a number of inexpensive two stage direct drive rotary vane mechanical vacuum pumps for sale on Amazon that on paper meet that requirement. So, yesterday I ordered a couple of them. One claims to pull down to 2.5 microns with a pumping speed of 5CFM ($120), the other down to 1.5 microns at 9.6CFM ($235). When they show up (one today, and one tomorrow), I plan on connecting them directly to an MKS901p to see how they do. Then I plan on connecting each of them to my fusor, to see if I can do fusion using those pumps.

Hopefully one or the other of these pumps will work reasonably well, at which point there will be a documented way for folks to get a single new pump for ~$250 or less that is good enough to do fusion.

I will post on the process I use to bring the pumps up, what kind of vacuum they initially pull, as well as how my fusor works with each of them. It will likely take a couple weeks or so to complete the whole process.

Joe.

[Dennis P Brown]

Impressive goal and those results will certainly be worthwhile. Such a test will be interesting - do provide all equipment used with this pump - chamber, fittings, gauges, deuterium system, power supply (and its voltage/current), ultimate vacuum and leak rate/out gassing over a day. Neutron detection system and values obtained. This way one can compare the results to required secondary systems used for a given result.

In any case, a good mechanical pump is ALWAYS the absolutely best first purchase and the critical component. Add an accurate micron range gauge and one has the primary stuff to seriously consider a fusor. A HV pump can be ignored if one has skill, money and even then, some good skill (unless a complete turn key $$$$ system.)

[JoeBallantyne]

So these are the two pumps I ordered.

https://www.amazon.com/dp/B01N0SYCL4

https://www.amazon.com/dp/B09KBJS36B

One thing you will immediately notice, is that it is difficult for folks writing the marketing descriptions of the pumps to do math properly, as BOTH of the pumps have a factor of 10 or more difference in the claimed ultimate vacuum rating in pascals vs microns. This is also an issue with the listings for many of the other inexpensive dual stage pumps you can find on Amazon. The vacuum rating in Pascals is generally 10x better than the vacuum rating in microns. Which is just stupid, because 1Pa is 7.5 microns.

So a pump rating of 0.3Pa is 2.5 microns, not 25 microns, and a pump rating of 0.2Pa is 1.5 microns, not 15 microns. I'm not quite sure why so many of the pumps get the conversion wrong. They list their ultimate vacuum as some fraction of 1Pa, but more than 10 microns. Makes no sense at all.

This is the current marketing verbage for the Specstar pump on Amazon:

"STRONG PERFORMANCE - The dual-stage air vacuum pump delivers high speed and efficient performance; With 9.6 CFM free air displacement and factory rated to 0.2 Pa (15 Micron), it ensures a faster evacuation and cleaner process that will meet your requirements; High-quality filter elements used to effectively reduce oil mist and improve the filtration efficiency"

And the similar verbage for the Kozyvacu:

"Specification - Flow Rate: 5.0 CFM, Ultimate Vacuum: 50 Micron, 1/2HP AC Air Tool Rotary Vane, Inlet Port: 1/4 Flare and 1/2 male, Voltage /Frequency: 110V/60Hz,Oil Capacity: 380 ml"

BUT when you zoom in on the Kozyvacu pump label it shows a rating of 0.3Pa on the faceplate of the pump, AND the math on the faceplate appears to have been done correctly, because it also shows 2.5 microns.

One of the reviews of the Kozy TA500 pump stated that they used it with some scientific equipment, and were able to pull a 4 micron vacuum.

So, I decided to trust the ultimate vacuum specs stated in Pascals, not the micron numbers.

This means that the Kozy should pull ~2.5 microns, and the Specstar should pull 1.5 microns. Factors of 10 be damned.

Joe.

[JoeBallantyne]

So both pumps showed up on my doorstep exactly on schedule per the Amazon delivery date stated when I ordered them. The Kozy showed up next day, and the Specstar took 2 days. Both were very well packaged in sturdy cardboard boxes with custom fitted solid white styrofoam inside the box and around the pump. Neither pump was damaged in transit. The packaging for both pumps was excellent.

Both pumps came with sufficient vacuum oil for 1 fill.

I decided to use the oil that came with the pumps initially, to see if they can meet their specs with the oil they ship with. After they have been tested with the in box oil, I plan on flushing them out, and replacing the oil with Inland 19 Ultra which I use in my Welch 1402 pumps to see if they do better with that oil. I fully expect that they will per my experience here:

viewtopic.php?p=98621

I am currently waiting for appropriate hardware so that I can get a KF16 connector connected to both of these pumps. These are sold as HVAC pumps, not scientific pumps, and therefore ship with HVAC connectors, but I am going to use them with KF, so need to make the conversion. This is completely normal, once you get a pump, then you have to get the proper hardware to connect it to your system, and that can take as long or longer than getting the pump in the first place.

So, once the hardware I need shows up in the next couple of days, I should be able to report on the vacuum they pull.

As far as build quality of the pumps themselves, well, when brand new pumps are being sold for so little money, you really do get what you pay for. You get a lot more for another hundred bucks: the Specstar seems like a much higher quality pump overall than the Kozy. If it pulls a good vacuum, and works for fusion I think I would recommend getting it over the Kozy, unless saving the extra cash outlay is the first priority. I was really quite impressed with the Specstar - it seems like a really nice pump, and a steal for $230. But of course, what really matters is what kind of vacuum it can pull as that is where the rubber meets the road. We shall see.

Joe.

[JoeBallantyne]

So during the last couple of days I was able to test the Kozy and Specstar pumps vacuum performance. All of the pressure measurements were made with an MKS901p vacuum gauge.

The Kozy as shipped using the in box oil, consistently pulls between 6 and 6.5 microns while warming up, and stays at 6.5 microns when up to temperature. After pumping for an hour, the Kozy was pulling 6.5 microns and the pump head was at 115F and the motor was at 98F.

The Specstar as shipped using in box oil, pulls between 5.5 and 14.5 microns while warming up, and stays at about 14.5 microns when up to temperature. After pumping for an hour the Specstar was pulling 14.3 microns with the pump head at 117F and the motor at 101F. The Specstar in box oil vapor pressure is much more temperature dependent than the Kozy oil. The ultimate vacuum almost triples from 5.5 microns at a pump head temperature of 68F to the 14.3 microns at 117F.

I flushed out the oil in both pumps with Inland 19 ULTRA, and then filled them both with that oil, which is the same oil I use in my Welch 1402 pumps, and both pumps worked better with the Inland 19 ULTRA oil.

The Specstar pulled down to 4.3 microns after 5 minutes at a pump head temp of 87F. After pumping for an hour, its pressure had risen to 6.6 microns at a pump head temperature of 115F. So the better oil, vastly reduced the pump head temperature dependence of the ultimate vacuum, but it was still worse by a little over 2 microns after an hour. This is likely because there was still some residual original oil mixed in with the new oil, but it could also be because pump performance gets worse due to tolerances going up when the pump is hot.

With the Inland 19 ULTRA oil, the Kozy pulled down to 3.8 microns in about 5 minutes with a pump head temperature of 90F. The pressure remained remarkably steady as the pump warmed up, the minimum pressure was 3.3 microns at the 10 minute mark at a pump temp of 99, and after half an hour the pressure was 3.8 microns with a pump head temp of 114F.

Bottom line, with good oil, the Specstar pulls about 6.5 microns when hot, and the Kozy pulls 3.8 microns when hot. The Specstar was $230, and the Kozy was $120, so the Kozy is about a (6.5/3.8)*(230/120) = 3.28x better over all value than the Specstar. It pulls almost 2x lower than the Specstar for almost 1/2 the price.

I am going to try both of these pumps with my fusor, and see if I can successfully do fusion with them. I expect that I will be able to, although it will likely be easier to do so with the Kozy.

Based on these results, I would recommend new folks who want to do fusion get the Kozy, and not the Specstar, because of the better price, and the better ultimate vacuum. (Even though when running the Kozy sounds like it is going to die, and the Specstar sounds much better.) Both of these pumps performed within their marketing specs in microns, but neither one met their performance specs stated in Pascals. (Remember that the specs stated in Pascals are ~10x better than the specs stated in microns - because evidently doing unit conversions is difficult...) The Kozy came close to its Pascal specs (0.3Pa) with good oil, but not with the in box oil.

It's pretty amazing that for $120 you can get a new vacuum pump with a 1 year warranty that pulls 3.8 microns. Granted you will have to spend ~$30 on adapters, and ~$25 on a liter of very good vacuum oil - like the Inland 19 ULTRA, but that still means that all in all done, you can have a brand spanking new pumping setup that should allow you to do fusion for ~$175 plus tax.

LASCO 17-6783 1/4-Inch Female Flare by 1/4-Inch Male Pipe Thread Brass Adapter
https://www.amazon.com/dp/B008E5CUFG

Female NPT/FNPT to KF16 Flange Fitting (ISO-KF KF16 QF16 NW16) (1/4" FNPT to KF16)
https://www.amazon.com/dp/B06ZZRQ2ZP

Kozyvacu Dual-Stage HVAC Vacuum Pump, 5.0 CFM
https://www.amazon.com/dp/B01N0SYCL4

Inland® 19 ULTRA Rough Pump Oil (1 liter bottle)
https://www.sisweb.com/vacuum/sis/inland19-ultra.htm

Joe.

[Richard Hull]

Great comparison report and well done, indeed.
Newbs take note! Joe has done some hard work with you guys in mind.

Richard Hull

[JoeBallantyne]

So, I hooked the Kozy up to my fusor, using a ~36 inch long KF16 bellows connected to a normally blanked off KF16 test port that I have on my pump station, and successfully pulled the fusor down to 5.4 microns after about 20 minutes of pumping. Then I did two 10 minute fusion runs.

It took a little getting used to controlling the pressure with the new pump, because the Kozy pumping rate does not compare to the Welch 1402, as the inner diameter of the 1/4 inch male flare intake connector on the Kozy is about 1/8th of an inch, maybe 3/16ths max, whereas on the Welch 1402, the KF25 intake connector inside diameter is more like 5/8 to 3/4 of an inch. So the conductance of Kozy intake is MUCH less than that of the Welch. Normally with the Welch, if I opened the KF40 butterfly valve on the end of the fusor just a tad it would immediately pull the pressure down, and drop the cathode current. With the Kozy, there were a few times at first, when opening the butterfly valve all the way was not sufficient to drop the pressure, and the current would run away, and the power supply would shut down.

Initially I had set the Spellman DXM50N1200 supply to -24kV, but after having the current run away on me a couple of times, I decided to drop the cathode voltage down to -20kV to see if that would give me better control of the pressure with the less powerful Kozy. It did end up helping. At -20kV, it was easier to control the pressure and keep the current from running away.

The first 10 minute run was at 20kV, a pressure range from 12-14 microns, and a current range from 0.01mA up to about 6mA. I kept the current pretty low most of the time, at 1mA or 0.5mA. The neutron count for that 10 minute run was 522.

I did a second run also at 20kV, at a slightly higher pressure and current range, the pressure was usually between 13 and 17 microns, and the current was usually between 1 and 8 mA. Typically around 1.5 to 2mA. The neutron count for that 10 minute run was 3842.

I then shut the fusor down, and did a background count run for 10 minutes, the total neutron count was 65. For an average of 6.5 cpm.

So, it is possible to do fusion using only the Kozy pump that pulls a 3.8 micron vacuum. The fusor I used is the same one I describe in my neutron club entry request here:

viewtopic.php?p=98063#p98063

The changes since then, were an upgrade of the power supply to a 50kV 24mA Spellman, from a 70kV 8.5mA supply, lead shielding added to the fusor, and oil added to the custom built feedthrough connector. Although I still haven't been able to push higher than 36kV due to what I believe is internal arcing in the fusor.

Joe.

[JoeBallantyne]

An update on the Kozy pump. I did another run today with the MKS901p directly connected to the pump intake, and it actually pulled down to a consistent 3.0 microns. It ran for 30+ minutes, and was stable at 3.0 microns for most of that time. Which is better than the 3.8 microns I got on the first day I ran it with good oil. So the Kozy meets my arbitrary performance requirement for a single mechanical pump to be able to do fusion of pulling a 3 micron vacuum. It also does work for fusion as shown by yesterday's runs.

Because I do very much like the build quality of the Specstar pump, I plan on doing one more oil flush/replacement with another fill of Inland 19 ULTRA, to see if that will cure the 2 micron rise in pressure (from 4 to 6 microns) as the pump gets hot. I am hoping that it will, and that the Specstar can pull a ~4 micron vacuum consistently. That should work for fusion just fine, even though its a bit above my arbitrary spec of 3. In fact, I bet the Specstar will work for fusion as is, with its 6 micron vacuum, but I would like to see if I can cure the pump head temperature dependence of its ultimate vacuum. It would be nice to know for sure it is due to remaining contamination from the original in box oil.

Joe.

[JoeBallantyne]

So, I ran the Specstar pump a second time today as well, and unlike the Kozy which did better than it did the first time around, the Specstar started out a little better - pulled to like 3.9 microns at the start, but after pumping for an hour plus, it actually did worse, and was only pulling down to 7 microns. As planned, after the pump head and oil were good and hot (117F), I drained the first load of Inland 19 ULTRA oil, and flushed it with a bit more 19 ULTRA down the throat of the intake while running, with the exhaust mostly plugged and the pump tilted towards the drain at about 20 degrees or so.

After flushing, I shut the pump down, replaced the plug, filled it with a fresh batch of Inland 19 ULTRA, and fired it up. It was already still pretty warm, as it was hot when I drained it, and off the bat, it was pulling about 5 microns. Left it running with the MKS901p on the intake for an hour plus, and it was pulling about 6 microns again. Much like it did after the first 19 ULTRA oil change. The Inland 19 ULTRA oil I drained from it today looked pristine - clear as day. There was no visible evidence of any contamination. I suspect if the drained oil were loaded into a better pump, it would pull down to 1 micron or less no problem.

So, it looks like the Specstar is just designed and built so that as it gets hot, the ultimate vacuum it is capable of pulling gets WORSE. Regardless of the oil. When cold it will pull to about 4 microns, when hot, will only get down to 6 or so. That is a shame, because it appears to be a pretty solidly built pump. Too bad it doesn't have the chops of the smaller, lighter, cheaper, and noisier Kozy as far as ultimate vacuum capability.

I will still test it out with the fusor, sometime in the next couple of days to see how it does, but I was really hoping it would pull a consistent 4 micron vacuum or better. Alas, it was not to be. 6 is certainly not bad, but 3 is much better. Per its stated 0.2Pa ultimate vacuum specs, it ought to be pulling 1.5 microns. It may be called the Specstar, but specs you don't meet don't matter, and you can't be a star if you don't meet your specs.

Joe.

[JoeBallantyne]

So I did a neutron run on the fusor using the Specstar pump today, and it worked just fine. It seemed a little easier to control the pressure with the Specstar than with the Kozy, possibly because the Specstar uses a 3/8 flare fitting instead of the Kozy's 1/4 flare, so its intake conductance is higher. (Likely more than 2x that of the Kozy.)

The Specstar did take significantly less time to pump the fusor down, I started it up cold, and was done with the whole fusion run within 30 minutes or so, so it likely never even got completely up to temperature - which is good because its ultimate pressure gets worse as it heats up. Pump head temperature now is 100F and that is about 5 minutes or so after shutting everything down. It took less than 10 minutes for the Specstar to pump the fusor down to 5.8 microns. I used the same 36 inch long KF16 bellows to connect the Specstar that I used for the Kozy.

I set the voltage on the Spellman DXM50N1200 to -20kV and did a 10 minute run with the pressure between 13 and 19 microns. Pressure was usually between 13 and 17 microns. The current ranged between 0mA and 6mA, and was usually between 1.5 and 4 mA. The reason why it was at zero for a bit, was that I left my Ludlum 2200 powered on batteries for a day or two (forgot to turn it off after the Kozy neutron runs), and ran the batteries down, and had to connect a power cord to fire up my neutron counter. So I dropped the pressure in the fusor while doing so, and that meant the current was at zero for a bit after I started the 10 minute count until I could adjust the pressure back up to where the plasma lit off, and the cathode current rose.

The total neutron count for the 10 minute run was 3859. I did not do another background neutron count run, as it should give comparable results to the one I did for the Kozy tests.

So, one can use either the Kozy or the Specstar to make neutrons with a fusor. The Kozy has a lower ultimate vacuum, but the Specstar has a faster pump down due to its larger CFM rating. Its also more pleasant being in the room with the running Specstar than the running Kozy, although neither one approaches the MUCH quieter and pleasant lub lub lub lub and ticking of a belt driven Welch 1402.

I am going to stick with using my Welch 1402, but for new folks who want to do fusion on a budget, and with minimal hassle, either the Kozy or the Specstar will work for a one vacuum pump fusor setup, as long as you use excellent vacuum oil in the pumps and build a decent chamber that doesn't leak and is clean. (ie: PLEASE use KF/QF/NW parts - or conflat if you really want to go that route - I would NOT use conflat for a v1 low cost fusor. KF is great. Try not to use any connection or connector that is threaded on your vacuum system. Threads generally only cause problems, because they are HARD to get sealed leak free.) You will also need a good neutron dectection system.

Joe.

[Dennis P Brown]

I'm impressed by this work and you proved beyond any doubt, what-so-ever, that a marginal system can not only do fusion, but rather decent levels of fusion. This certainly means that not only can someone do measurable fusion w/o a secondary High Vac pump but can do it with a very inexpensive pump. My doubts were totally misplaced*.


* Post Edit: meaning, as in, I was wrong

[JoeBallantyne]

Thanks Dennis!

I think it would be good if we could have a FAQ post in the vacuum section about this, so that folks can see in that section that inexpensive single pump systems can do fusion.

It can really save a huge amount of effort and troubles if the second high vacuum pump is dispensed with.

I would be willing to write a FAQ post.

Joe.

[Liam David]

I think this is FAQ-worthy material. This is yet another way the barrier-to-entry for building a fusor has been reduced. Good work!

[Richard Hull]

I agree! Write it up and submit it with the the normal beginning FAQ -
Might I suggest FAQ - Single vacuum pump fusion
Post it in both the Vacuum forum FAQs and the Construction forum FAQs to catch hapless lost newbies attention....You hope.
You might mention in the FAQ that single pump fusion can waste a bit of flowing deuterium in an attempt to reach a pure deuterium pressure in the fusor. This is common with inexperienced first pass fusion efforts, especially with inadequate valving.

Richard Hull

[lanewaddell]

This is wonderful information, thank you so much!

Quick question, this pump is limited by the 1/4" NPT fitting on the top of the pump. Does that small constraint set an upper limit on what volume it is able to pump down? I am looking at an ISO-80 cross chamber right now. If it's a matter of waiting a little longer due to low volumetric flow rate that is fine, but I wanted to make sure it could pump down that size void.

What size did you test it on out of curiosity?

Thanks!

[Richard Hull]

I assume you are asking someone else.

You are correct pump down time will take longer due to a pump constriction or a small bore long, pump hose. more time in pump down means hotter oil and more vapor backstreaming and higher chamber pressure over time. More D2 loss if fusion is to be done fighting to get to a more pure D2 into the chamber with no secondary pump. It can be done, but at the cost of useless pumped out D2.

Richard Hull

[Frank Sanns]

I think it was Wilfred Heil that pointed this out many years ago. Those of us that were doing fusion were all using two pumps. He was one of the first to do it with a small chamber and a two stage pump.

It can be done but I personally have never seen a 3 micron pull down from a mechanical pump on a chamber. I have a larger chamber and have had a few high performance commercial grade pumps on there an could never get down low enough to get over 20 kv on my chamber. That is probably up around 12 microns. It is a Looooonnnnng way down to 3 microns from there.

[JoeBallantyne]

I pulled the low cost v2 fusor chamber I am currently working on down to 1.3 microns with my Welch 1402. This was the very first time it was pumped down. Took a couple of hours or so, but that Welch 1402 pump's ultimate vacuum is about 1 micron, so getting that close to the pump ultimate vacuum generally takes some time.

I pull my v1 fusor down to less than 3 microns on a regular basis using that same Welch 1402 rotary vane mechanical pump.

I can pull the v2 fusor down to less than 6 microns using the Kozyvacu TA500 discussed in this thread in about 20 minutes or so, as long as the chamber is leak free. With my leaky v1 sparkplug based cathode it goes down to about 8 microns in 20 minutes.

Lane, the Kozyvacu TA500 should be able to pull your ISO80 reducing cross down to 5-6 microns in 30 minutes or so no problem. That cross doesn't have a lot of volume. Maybe a liter or so. Frank has a relatively LARGE vacuum chamber.

Joe.

[lanewaddell]

Joe that is great practical information.

Thank you!

[Carl West]

This is great information that significantly lowers the barrier to entry for newbies.