Refrigerator compressors in series as vacuum pumps

[Brian_Gage]

Fridge compressors are fairly easy to obtain used, but as I recall they have welded 1/4" input and output pipes. Years ago, I saw one of these hooked up and used as a simple vacuum pump, but eventually it started cooking and venting burned oil. Can larger fittings be bored into these standard compressors to improve the through-flow? What is done with regards to oil to keep one operational. With two of these compressors mounted side-by-side in series with larger plumbing, would they need an additional diffusion pump to achieve suitable vacuum for a demo fusor? Would they be adequate for a simple neutron producing fusor?

[Richard Hull]

The answer is a great big old... NO!

Odd how they call them refrigerator COMPRESSORS and not refrigerator vacuum pumps.... They were designed to compress only from the instant the first engineer worked on their design.

By their very nature, both a vacuum pump and a compressor have a high pressure side and a low pressure side...Sure.... But, the valves and all the internals are designed to do the job the the device is named for. In short, they have a designed-in, "preferred direction" of gas flow. Thus, reversing them from their designed role is not a winner idea.

No one here has done what you suggest, though a few newbies have asked in the past. All were disuaded from the effort due to the complications involved in trying to reverse engineer a bus into being a car.

The medium or "technical" vacuum "fore pump" is but a tiny part of the fusor system. Buy, steal or borrow a real vacuum pump. Then, worry about the diffusion or seconary high vacuum pump you will also need. Then, worry about the conductance of the vacuum plumbing and obtaining a vacuum metering system. Then, worry about getting a true fusion capable power supply. Then, worry about the vacuum chamber, the HV feed-thru insulator, the welding, and the gas ports. Then, worry about getting the deuterium gas or making a heavy water electrolysis system manifold and dryer, (your choice). Then, worry about the electronic detection gear needed to prove to yourself, and others, you actually are doing fusion.

I wouldn't invest a lot of time fumbling about re-engineering one of the easier to obtain items in the chain or you might wear yourself out on a failed mission at this low level and decide it is all too hard, abandoning the effort altogether.

Richard Hull

[Carl Willis]

To add a little to Richard's response, there is some distant precedent for the hobbyist re-purposing of refrigerator compressors as vacuum pumps: F. B. Lee's 1960s-vintage Van de Graaff accelerator article in Scientific American. His project used two different types of refrigerator compressors, operated in series, to back a mercury diffusion pump. Mercury diffusion pumps are practically extinct today, but their uniquely-high foreline pressure tolerance did make them usable with refrigerator compressors.

In four decades, much has changed, and there are much better options. Cheap Chinese exports now include refrigeration servicing pumps (for evacuating the coolant loop, not for compressing the coolant) that can perform as roughing pumps for fusors, getting into the millitorr range starting at about $150. The Internet-driven market for used and refurbished goods did not exist in F. B. Lee's time, but now it's a reliable way to find a suitable scientific vacuum pump for $150-$300. I have little doubt that in today's world, this would have been his choice without batting an eye. These options will serve adequately for "demo"-class fusors and will back oil diffusion pumps and turbomolecular pumps for neutron-generating fusors.

-Carl

[Brian_Gage]

Thanks Richard. I'd just read a recommended article, Vacuum Pump Basics, which suggested using two air conditioner pumps in series. Scrub that. At this stage, I'm just reading and considering options, and I am looking at vacuum pumps first, though today I did make a couple of inquiries about obtaining microwave ovens from recycling (for transformmers and capacitors), but again, just asking, not building, not even planning to build ... yet.
Watched your Youtube interview. Great stuff, even just for listening. Some of the Youtube demos of fusors really need help to get intelligent narrative or better audio into their videos. I've heard a couple without any description, just music or vacuum pump noise.

Thanks for the quick reply.

[DaveC]

There's a bit more to this, than a simple no ain't gonna work. All vacuum pumps are compressors. This is the basic message in the Welch Instruction manual that (used to?) comes with their pumps.

To put matters into perspective, and shown where the limitations are, consider that a vacuum pump, rated for 10-4 Torr blanked off ultimate pressure, must exhaust the gas at slightly above atmospheric pressure. This gives a compression ratio of about 7.6 million ( 760 Torr x1/10-4 Torr) .

Refrigeration compressors operate in a range of a few hundred PSI at the high pressure side, and a few tens of PSI on the intake side. This gives an effective compression ratio of 10 to perhaps 20. This is not actually the pump's limit, but the limits of the working fluid in refrigeration systems... one of the various Freon compounds.

I did exactly what Brian has asked about, years ago, going even to the step of sawing the welded case open, with my Dremel MotoTool.... (took about 20 cutoff disks as I recall. It was a piston type refrigerator compressor from a small water cooler.

I replumbed the intake so that it went directly to the pump rather than into the oil sump, changed out the oil, built a transparent plexiglas cover and silicone rubber sealed the whole thing back together, after measuring the innards, cleaning the exhaust leaf valve and checking the head gasket for leaks. I still have it and it still works... but it is not a high vacuum pump.

The compression ratio by volumetric measurements of piston displacement, and the residual dead space ( can't be pumped) in the valve body... indicated a maximum compression ratio of about 30.

In use, the pump ran noiselessly, and produced a measured intake pressure of just below 20 Torr. Two pumps in series, as you suggested, would improve things to give an overall compression ratio of 400, but this is far away from the 7.6 million of the typical belt drive or direct drive pump, (when new!).

You can , of course, do the math and see that ignoring all the leaks and stackup of inefficiencies, it would take about 6 stages of pumps - definitely not practical at all.

So what Richard and Carl have advised is sound and practical. Maybe this little bit of anecdotal support, helps to illustrate why.

Dave Cooper

[Brian_Gage]

Thanks for the addendum Dave. I won't be attempting improvisations on the high vacuum end of anything. Was just asking, as mentioned in an earlier post, because a recommended page on "Vacuum Pump Basics" suggested two air conditioner pumps in series. I immediately started thinking the existing gas lines would be too small in diameter, so asked the questions.
Carl's suggestion about the Chinese pumps sounds good and the prices low. If I can cultivate a sighted partner locally, then I'll eventually be searching for a real good primary system with appropriate quality plumbing.