AC Window Unit Compressors

[Monroe Lee King Jr]

I just pulled a compressor out of a window unit AC. It pulled 29" with no effort I think perhaps if I get a second unit to run in tandem it might back up a diffusion pump. Tomorrow I'll find another unit in my junk pile and give that a try. I also found a Black Widow nestled in right next to the pump! So watch out lol. Unfortunately that photo came out too fuzzy (wonder why?) I wonder if some better oil would help this pump? I've got a really good pump I can use but I want to try and see how cheap I can do this with just junk I have around here. See if I can get to star mode for less than $100. I'll probably build several fusors this is just the first one for fun!

[Ross Moffett]

I believe that the working fluid is used as the lubricant in these pumps. Extended operation without some constant lubrication will burn it up, most likely. Old pumps using R40 (I think?) were lubricated by the refrigerant, modern ones using the non-CFC refrigerants have a squirt of oil in there as well.

[Monroe Lee King Jr]

Yeah I can leak a little oil in there with a needle valve from a cup (modify this air line oiler I don't use). I think I can clean this pump out real good and use some quality oil I have here for my good pump. These compressors have oil mixed in with the working fluid and a little settles in the pump and in the filter dryer. I might not be able to use these if it wont work I'll let you know. I've seen a lot of this and that said both ways. I know in the old days like in 1992 when I started reading the Bell Jar we could still find the old pumps that would work. I've never tried these newer ones we will see I suppose.

Monroe

Thinking about it I guess I should make a catch container for the exhaust to reclaim the oil too I'll do that if I can make a 2 hour run test. I guess I might should do that with just this one pump and see if it burns out or not before I go scrounge up another one.

[Monroe Lee King Jr]

This pump will work for sure if the AC units cant cut it. Anybody recognize this pump?

[Richard Hull]

Air conditioner pumps? Again, no good for any fusion work. Gotta' have a real pump.

The second pix is better but what kinda' shape is it in. You will have no idea without a decent gauge. As mentioned in my mechanical pump testout FAQ, 50 microns barely cuts it for diff pump use and 30 microns is merely OK. 15 microns or less is good.

Richard Hull

[Monroe Lee King Jr]

Back to back compressors. I need to adapt the outlet to inlet tubing. I'll see what we can pull with this set up. Everyone will know exactly then what to expect. Just for fun!

[Mike Veldman]

Way back in the 60s I played with cascaded compressors as vacuum pumps. I was a kid with limited funds and there were some articles in Scientific American amateur scientist column which suggested they'd work. BZZT! They did make some vacuum, at the cost of refrigerant oil back-streaming all over the inside of the glass chamber under vacuum. I messed with jars stuffed with dish towels as filters, tubes filled with desiccant and finally refrigeration line filters, which did the best but still didn't stop the oil contamination of the chamber. The chamber was a long piece of glass tube I was making a CO2 laser with. Once I put the same amount of work as I'd put into the compressors into an old worn out service pump, my vacuum was clean and workable. Don't waste your time trying compressors as vacuum sources, use the two to make a two stage cooled cloud chamber.

mike

[Richard Hull]

Mike's advice is sage. Tandem A/C pumps will never function to any usable level for fusion. They do pollute the fusion chamber very well and are efficient at distributing all manner of filth into the fusion gas environment. Without a calibrated, scientific vacuum gauge you'll never realize just how terrible they are and why you can't seem to fuse at all.

Richard Hull

[Monroe Lee King Jr]

Oh well this is indeed NOT for fusion! This is just to see what I can do with the junk I have laying around! I'll get a turbo pump if I'm serious about fusion! This is just for fun on no money! Just seeing how I can do for nothing first. I am going to make a diffusion pump too! Out of exhaust parts! I think I'll nickel plate them though. This is just exactly for a broke 13 year old to follow if he want's to play with vacuum. I'll present the results! Good bad or indifferent! It's not like I'm worried about if this works or not. This is a $100 fusor I've decided I wont spend more than that and junk I have around here and see what I come up with.

This is just the way I do things. Do what you can for next to nothing first and make dang sure you want to spend the hard cash for the good stuff. Then you appreciate having the good stuff. If you cant do it with scraps in a cave! Then what good are ya!

Monroe

[Carl Willis]

The tradeoffs between time and money, between learning the hard way and getting to a destination promptly, are all personal matters. That said, the economic argument for home-rolled diff pumps and refrigerator compressor vacuum systems has scant credibility. Today's conversation sounds just like Mike Kan's posts from years ago, or the many posts from people whose most up-to-date reference on vacuum technology was evidently an Amateur Scientist column from 1960. Needless to say, the world is a different place now.

$100 for a small, scientific-grade vacuum pump (with free or included shipping) is a common, almost daily, occurrence on eBay.

$100 for a refrigeration-service vacuum pump new in the box from China is an everyday occurrence.

Under $100 for a diffusion pump is an everyday occurrence.

People who are willing to wait and watch for a few weeks are rewarded with even better deals.

-Carl

[Rich Feldman]

Monroe,

I'm eager to see vacuum measurement results from one of your little chiller compressors,
and from two in tandem to see if that makes a difference.

Am guessing that in many designs, the vacuum is limited by the compression ratio and the need for compressed gas to push open a reed valve.

Here are some pictures from a trephination experiment I did in early 2005.

There might be more info about the valves on websites like "sam's laser faqs" or belljar.net.

[Monroe Lee King Jr]

Ahhh yes indeed but in my book there's just no better way than to make things yourself to get a root understanding of the nature of things. It allows incite into ways of doing things others wont try because they don't understand the true nature of it. For instance the scientist vs the engineer I'm a bit of both + mechanic. I make my own chemicals, telescopes, microscopes and tools. The rocket guy's feel the same way about me, but I've gone higher and farther than most wanna be rocketeers and spent far less money doing it.
I know for instance if you want to get to space cheap you fly a balloon to high altitude and launch from there. For some reason the rocket guy's just don't like it? Makes absolutely no sense to me to launch a small rocket from the ground!
Makes no sense to me to spend a dime on a fusor demo either! I know for a fact I can get plasma with my set-up here for nothing! Can I pull a medium vacuum with this set up maybe not! But maybe?
I know refrigerator/freezer piston compressors wont work! But A/C window unit rotary pumps with the proper oil might! I can easily set up an automatic oiler to keep the pumps alive and put in a gas ballast valve between the 2 stages. 2 things the average experimenter leaves out! I don't see experiments with these in this configuration with the proper oil!
I also know there are lots of kids out there that could get these pumps that have no chance to get a real pump. I was over 20 years old before I could afford a real pump! I know I can make a demo fusor with this setup and so can others less fortunate! I also remember I was at my peek of learning capacity when I was 13 years old! I was lucky a neighbor noticed I was sharp and helped me attend collage when I was 13 because I was into physics so deeply at that age. I want others to have a chance that's all. Being in collage at 13 is a bit weird for a 13 year old but cool indeed. Perhaps some other kid will do this and get noticed?
I'll get around to the man sized fusor soon enough. This one's for the very young to try with pocket change.

Monroe

[Monroe Lee King Jr]

@Rich
I see what you have there. That is a piston pump correct? How did your experiment go with that? The rotary style pumps don't have any valves These A/C window units are rotary compressors and have a much higher chance of working at lower pressures.

Monroe

By the way it's super cool to see the guts of a pump like that isn't it!

The limit is how close a tolerance you can hold, the smaller the better and the vapor pressure of your oil. There are other factors but these are the main 2 the true limit is in the oil if you can build the machine (pump) good enough.

Oh and by the way so far I don't know how low I can pump with these compressors. I can pull 29 inches with a single pump but I found out I can't measure any lower than that with my current gauge (I guess it has a small leak) I discovered this last night as I was cleaning these with mineral oil (in preparation for higher grade pump oil) I pinched my line and I could see bubbles coming from the gauge side if I held the pinch just right. If I pinched it off all the way no bubbles! So yeah my gauge leaks a little.
I don't have a swaging tool to mate up the lines for soldering just yet. I can make one on my lathe as I don't need many sizes. You could make a crude one easy enough from a flairing tool or buy a simple one for $20 but I have a lathe so I'll just make one. I'll post some photo's of the swaging and soldering process soon.
I'm going to use a saddle valve ($5 each) for the oil leak and the gas bypass valve.

[Mike Veldman]

Thanks for the personal history lesson it helps. Perhaps you should have mentioned this was a "play time" project and saved yourself our attempts to save you time and trouble. I played some in the 70s running refrigeration compressors as engines. Both using actual steam and freon 504 as the working medium, heated by solar collectors and parabolic dish boilers. Some success in both cases, even made electricity from a sealed unit motor run as a generator. The scroll compressor is much easier to use in this application. So, now I'm kinda interested in what you turn up, as I'm sure are others, since we all seem to have natural child like curiosity.

mike

[Richard Hull]

On fusor III, I was still in "make-it-myself" mode, trying to hack stuff. Good vacuum stuff is hard to hack. I came to the table with a few years of low level vacuum experience, but really dug into the texts related to conductance and took the "Bell Jar" subscription for its entire run.

Our local scrap yard had a ton of large SS 316 steam fittings. Among them I found 2 - 8" diameter blank offs that used 8 - 3/4" bolts! Cost for the pair?.... $40.00. Not to be stopped by a web of solid metal, I struggled to use my milling machine and rotary table to hog out the 1" thick circle required to turn those monsters into rings to allow me to TIG weld my two Braun 6" hemispheres to make my SS fusor III chamber. Over a 2 week period and about 100 hours work, I fashioned fusor III. (A week's vacation from work - Full time and attention) The machining, drilling, and welding, it was a real chore. Fusor III was my first fusing fusor. It suffered from a plethora of leaks and other cheesy, but cleaver, bothers. I sealed its flat mating surfaces with pure indium wire and this was good enough to let me do limited fusion. Even the view port was home made. I was proud and would be the first amateur to do fusion in a Farnsworth fusor. On its best day, Fusor III could kick out about 200,000 n/s but was more often in the 100k n/s range.

I limped along until 2003 when I had enough hardware for fusor IV. This time I was going to do it right. A pair of 8" professional conflat rings were picked up in 2001 off e-bay for $45.00 each. Another pair of Braun hemispheres were purchased in 2002 for ~$100.00! I purchased mostly swagelock gas fittings and valves all for about $100.00. The final item, a good 2.75" conflat window with a 1" clear aperture cost me $20.00 at an HEAS fleamarket in 2003. I was set for only a little over $300.00 to build a proper, professional fusor! No tedious machining this time except for the boring of the port holes in the hemispheres and a bit of welding....A snap!

On the vacuum end, I had determined to use a diffusion pump with my mechanical pump this time. (Fusor III was 100% mechanical only using a professional 5CFM Precision lab pump, but did utilize the good offices of a Lesker micro maze. Still, I wasted a lot of D2 in getting fusor III to perform with this less than adequate vacuum system.) I found a nice little air cooled diff pump at HEAS 2001 for $50.00, so, I was all set.

At first pump down of the new Fusor IV chamber, I was stunned to hit 10e-5 torr in mere minutes! WOW! Having the right stuff made all the difference. I learned my lesson. Whenever I think of machining what I need out of some kludged metal item, I think again and either wait to seek out what I need made for the purpose over time on e-bay, hamfests or fleamarkets or let go of the cash, if desperate. I do less machining and get better results. Fusor IV went on to exceed 1.5 million n/s a few years ago and can regularly hit nearly a million n/s.

You can't beat having the right stuff based on years of vacuum technology advancement in the art. Understanding vacuum technology takes a lot of self-directed reading as well. One learns that 1/2" vacuum lines will never pump well at the scientific level and that if you have to rely on gobs of vacuum greases to seal your vacuum, you are doing something really wrong.

If one is willing to burn off a lot of time, spend little money and have a gross underperformer as a result then have at it. The ultimate success of a fusion project proceeds as the inverse square of the number of home spun kludges in the system.

Richard Hull

[Monroe Lee King Jr]

We shall see indeed. Hey so far I ordered 2 saddle valves and some oil so I've got $20 invested and about 2 1/2 hours. Most of my time right now is researching so the little time I've spent playing with these compressors is a good break. By the way the pumps pulled much faster after sitting overnight with the mineral oil in them. Now I need to make the Pirani Gauge so I can see what I'm really pulling. Making this gauge http://www.drkfs.net/pirani.htm I'll have to order a 1A meter from fleabay for $5

Monroe

[Ross Moffett]

Harbor freight muktimeters make great cheap panel meters.

[Rich Feldman]

Hi Monroe.
Re. the Pirani gauge, there are many ways to make 'em. Perhaps even easier than the one in your DrKFS link is to open a miniature incandescent lamp, or take a glow plug, and expose the filament to your vacuum under test. Reminds me of your proposal to turn an ordinary vacuum tube into an ionization-type vacuum gauge.
Here's one thread (talking about both) that came up right away in fusor.net archives.
viewtopic.php?f=10&t=3544&p=22917

Are you trying to copy the DrKFS design so closely that you can use his calibration table, with no other vacuum reference?
Then (with all due respect) I wouldn't expect absolute accuracy closer than a half-octave, maybe a full octave.
Did DrKFS post any data about design repeatability, or sensitivity to details like wire diameter?
IMHO, factor of two errors in pressure measurement (if stable) would be pretty minor as you characterize and operate a fusor.

As for displaying pressure values, from nonlinear voltage or current measurements.
Let's push Nick Peskosky to realize the open-source digital solution he mentioned in his post today ("06SEP13").
I once took an analog meter & made a nonlinear scale card, not unlike DrKFS's solution, and reported it here: viewtopic.php?f=10&t=4047&p=21455

[Monroe Lee King Jr]

Yeah your right! I used one I got at Autozone on a little bench supply I made out of a PC Power Supply for fun. Has a 24v variable 1.5 amp and the rest you see there. Even has a standby and active mode.

[Monroe Lee King Jr]

@Rich thanks for the links! Yeah, I was kinda thinking about using his calibration. However I'm still looking at other designs. I like the glow plug idea even better using the platinum wire and already made for $7. Still looking! Keep the info coming!

Monroe

[Jerry Biehler]

Hi Monroe.
Re. the Pirani gauge, there are many ways to make 'em. Perhaps even easier than the one in your DrKFS link is to open a miniature incandescent lamp, or take a glow plug, and expose the filament to your vacuum under test. Reminds me of your proposal to turn an ordinary vacuum tube into an ionization-type vacuum gauge.
Here's one thread (talking about both) that came up right away in fusor.net archives.
viewtopic.php?f=10&t=3544&p=22917

You can use a thermistor to measure vacuum. There are plans out there. Here is one design of many. Of course you need something to calibrate it.

http://www3.telus.net/schmaus2/vacf/thermis.html

[Monroe Lee King Jr]

Looks like all these charts look pretty much the same too! Thanks! Yeah I can work something out. I have the 2 pumps running together now and I received the vacuum oil today and I primed them with it. Should get the saddle valves Monday or Tues. So now I need the gauge to test them With both pumps running it does increase the cfm for sure because the gauge I have now drops like a rock to 29 inches (where my gauge leak stops me from getting 30) I want the gas ballast valve in there before I go for a long run test.
I found a board to mount them on still need to make up a good catch cup for expelled oil and the oil drip for the intake.

[Richard Hull]

Naturally, the 29-30 inch readings are absolutely valuless in the technical sense as you could still have up to 1000 times too much remaining air in the lines to get to star mode. Such a gauge really isn't worth having when qualifying any sort of scientific, technical forepump as it tells you nothing about the bottom end, only that you have some sort of vacuum of totally unknown level.

Richard Hull

[Monroe Lee King Jr]

For sure the bourdon gauge is just a mechanic's tool to determine rough vacuum. It does mean the pumps are working and I can can get a rough idea of cfm (how fast the gauge drops) it is handy for beginning to look at vacuum. I now have to construct a valid gauge to determine how deep the rabbit hole goes. That's next on the list.

Monroe

[Monroe Lee King Jr]

Ok here are the saddle valves in place. The thing to remember about these valves is they pierce the tube and are now needle valves with the tube as the seat so you have to barely pierce the tube and when you close the valve Do Not apply much pressure to close it. Keep an eye on the handle position when closed because over time if you close to hard it will move and at some point you will run out of needle for a proper seat! So the one on the top is the Gas Ballast valve between the 2 pumps and the one on the suction just above the dryer is for the oil drip cup to keep the pumps lubricated. The pumps are filled with oil so as long as you have a little oil in your exhaust you are getting lubrication. It also helps the pump seal and pump lower as the oil is the seal in the pumps.