Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Cryocooler Stirling 1-stage build, some questions

  1. Apr 11, 2016 #1
    Hello everyone!

    Just the forum I was hoping to find! There are really not many ppl to ask about this subject.

    Had nothing to do about a month ago so I thought that it would be quite fun and interesting to build a cryocooler, it will be a long time project though since I lack many things right now.

    Since it will be a fun thing I had no plan to use it to anything special only a goal to get it as cold as it would get whatever that now might be since I doubt it would be even remotely possible to get close to any temperatures to fex make liquid air, which would be really fun.

    Ok, before anyone say it so yes I have gone through around 40+ patents during the month of how they are constructed and what regenerator material that could be used and stuff and looked on many pages on the internet.

    I have found much info in these patents but I lack the answers to some questions that I hope someone here might be able to answer to me:

    1. The compression in a stirling cryocooler don't seem to be much over 1.5 if I have got things right which as I see is hardly any pressure at all. Wouldn't a higher pressure make a bigger volume of chilled helium and thus bigger active work or it is just not needed due to the high pressures in these?

    2. The regenerator materials that is seemed to be used (that I can get my hands on that is) is lead, stainless steel and even some plastics. However these materials are more to be used in the area of 80K and lower which I unfortunately don't think I will even get close to. What materials would work good about over 80K? I guess stainless would work quite long but I have not found any sharts to over 80K for materials in general, but if there are better ones over 80K then stainless Im all ears. I have not found any info on this mainly because cryocoolers are not normally used in the upper Kelvin segment.

    3. Normal say small football sized cryocoolers, fex those you see on Ebay sometimes and YouTube, seem to draw around 100W or so. For a compressor with no physical resistance that only have to compress something 1.5 times seem to be dam high, or have I miss something here? Sure its around 300 psi of helium in them but still they still compress from 1 Atm to 1.5 sort of speak since its all in the same pressurized atmosphere.

    4. Im quite sure the length of the regenerator have to be a length that accommodate the volume in the bottom at the cold finger and have read about strokes at around 20 - 25 mm here, which are smaller then I thought. But the lengths of the regenerator I have seen on patents have a much much bigger volume then this then what is really needed for that or the compressors 1.5 pressure increase, so what is going on here?

    A longer regenerator then necessary can´t be of any advantage since all the helium will not go thought it all but only the first portion and not go back to be compressed so there is obviously something I miss here.

    5. Im sure that the regenerator material (or regenerator it self) should come as close to the bottom of the cold head as possible to get as much of the helium volume into it?

    Thanks guys!

    Regards William
    Last edited: Apr 11, 2016
  2. jcsd
  3. Apr 12, 2016 #2


    User Avatar
    Science Advisor
    Gold Member

  4. Apr 12, 2016 #3


    User Avatar
    Science Advisor
    Gold Member

    Andy Ross Book

    I've worked on Stirling Engines .
    Last edited: Apr 12, 2016
  5. Apr 12, 2016 #4
    Thanks for your answer!

    Oh, then you might have some detailed information about cryocoolers?

    Thanks for the links as well! Have already seen all of them though and have not found any info on my questions there. But I will search a little more with Reversed Stirling Cycle but I'm not after to just get some frost or so I'm aiming far below that but to get even close to this there are things that I need to look up and wonder about but I have not found any answers on those yet.

    I'm not at all after to build something to get little frost slightly below freezing I'm going full bore with gas bearings and such to even have a remotely chance for this.
  6. Apr 13, 2016 #5


    User Avatar
    Science Advisor
    Gold Member

    Draw out a tentative design and give us the reasons for some of your design decisions and then we can have a proper discussion .
  7. Apr 14, 2016 #6
    I have a design in my head how I like it to be in the end and that is pretty much the same principal and build of a typical cryocooler design like the beta design in this patent. I thought that it would will be better to use a detailed picture then to write pretty much the same my self:


    If I would have write up any basic design it would look quite similar to the one above, I have however thoughts on using magnets instead of the springs just to see how that would be but I will se how I do with that since I have not got to that test phase yet.

    Gasbearings do I see as must for this design as to instead use any Teflon, graphite or similar would not hold in the length and thus planned to use stainless. That would however not be allowed to hit the walls due to wear that could prevent function and thus gasbearings.
  8. Apr 16, 2016 #7


    User Avatar
    Science Advisor
    Gold Member


    A 1.5:1 pressure ratio applied to a 300 psi initial charge pressure gives you 450 psi which is about as high as it is practical to use .

    Think of the average force on a piston during compression phase and the amount of power needed to move that piston many times a second .

    Note :

    (a) The actual calculations of pressure over one cycle for a real cryocooler are quite complicated .
    (b) Pressure ratio always needs strict definition . Some systems work between the charging pressure and a higher pressure and some systems work between high and low pressures with the charging pressure as the mean value .
    (c) Piston may see pressure on one face or it may see differential pressure on two faces depending on design details .


    You may have considerable difficulties in converting that patent drawing into a practical design .

    Personally I would go for something based on the simple and elegant Phillips - Stirling concept with mechanical drive .

    We can discuss any proposed design a bit more though if you want to .
  9. Apr 19, 2016 #8
    This thread came at a convenient time! I was in the same boat as you, I was interested in making liquid air for a while and tinkered with JT throttles a little bit. I finally decided I was going to pick a design and try and recreate it! I chose this after a bit of research:


    A simplified version mind you, If this thing runs for more than 6-10 hours in its life I will be surprised. Weight was no problem (it's not going to space). Ease of machining was key. I'm going to externally drive it. I will be using lead shot in my regenerator. I will initially test run pressurized with air. Then hopefully if it works switch and try helium. It's not finished yet, I have been puttering away at this in my spare time.



    For scale, The regenerator is 5.1875" overall length.

    You might be interested in researching a "hilsch tube", It doesn't get that cold but is extremely easy to make.

    I learned a lot from this website : https://translate.google.ca/transla...s.de/Hotcold/Kryo/main.htm&edit-text=&act=url

    It was German, I had to google translate it. They have reached impressive temperatures with simple instruments!

    Good luck!
    And keep us updated!

  10. Apr 22, 2016 #9
    And even from 300 to 450, and not from a 0 to 450 psi which would be something else, is it still 150 PSI which require some strength to do, that is true.

    Lesser sized piston means lesser physical pressure back and thus smaller work for the motor, but there are of course other reasons for the size of the piston that that make it easier for the motor I guess.

    (a) I happened just recently got a glimpse of some calculations (don´t think It was all) for a cryocooler and I can tell you that that was way more advanced math then I wanted. It was way more then I ever want to deal with and in that case I pretty much rather go after the trial and error principle, of course with some directions so I´m not fumbling around totally blind.
    (b) That is true.
    (c) Yea that what I thought about in the above patent drawing that there is equal 300PSI pressure on both sides on the piston and thus it only need to create a pressure of 150 PSI and also easier if only a small distance. But Im quite sure there are other reasons for the piston/cylinder size apart from this thing. But ok then I know the basic reason for that.

    Im not going to rip the measurement straight of it or so but the principle only because this type have fast cooling and is small. But yes I have had the Phillips - Stirling variant in my head as well. There is this video on youtube, (doesn´t work to post youtube links?) just search for Stirling Cycle Part 2 (Stirling Cryogenics) there, where they use one of these types and they wnt down to temp really quick and created liquid air quite fast and a lot of it as well and that was a classic Philip - Stirling.

    Have not found the patent on that exact variant though which bugger me, but that would be alternative two to make if I will not have time to build something like the one in the patent or any other problems arise. Some of the reasons I would like to build the one in the patent is because they have a very long life since there are pretty much no friction between the piston, regenerator and the cylinder wall due to the gas bearings but also since it could be made very compact.

    If it takes to long to make this I would probably go over to the mechanical Philip - Stirling concept and just have the other one done the day it will be done. But even that one require some advanced math which I don't like to deal with.

    Thanks for the reply!



    What a coincidence then and good work! Have not got this far by my self yet, lot of things to do, and I did not want to deal with any high pressures or such otherwise so would it be quite easy to get really low in temperature.

    Good you mentioned the size there since it looked much longer, where did get the calculations and such for for that? And those, what it seem, O-rings would not hold for liquid air temperatures and any oil would freeze as well, but graphite dust could work for a time as example.

    I don't want to deal with high pressures so hilsch-tube and such are of and I´m after building something that can be cold enough to make liquid air not only make something that gets cold, I can make a refrigeration system to do that. But to make liquid air something else is needed and there are many different designs for this and what I choosed is probably one of the more advanced ones but fun to make if I will get the time for it and the things I needed as well that is.

    Yes I have seen that patent as well and something like that could be the second design I would make if the other will take to long or I can´t get some things made or whatever the reason. I don´t really have access to the same machinery you seem to do which make it take so much longer for me, but keep up the good work and keep us updated you to!

  11. Apr 25, 2016 #10
    Just an update then, I got it together enough to try. I was unsuccessful. The temperature differential I achieved was a measly 10 C. I tried charge pressures ranging from 0-200psi. And rotation speeds from 1-2000rpm. Everything was sealed up nicely, If I left my charge valve open I could see a pressure spike while rotating.
    If I reversed rotation I got a higher temperature on the cold head.
    I'm not sure what to try next.... I think perhaps I have too much parasitic loss. Maybe the wall thicknesses on the regenerator and the cold head leg were too thick and are conducting too much heat to the cold head.

    I tried lead shot and screens in the regenerator.
    *just a note on the orings, I figured they would be fine to seal, at least temporary for proof of concept, especially since they are on the "hot side" of everything.
    To get my sizes I printed the image out and scaled everything off of that. I have cad data that I have sketched out. It, isn't neat or tidy though! Just a quick sketch!

    Any Ideas?

  12. Apr 25, 2016 #11
    Just a little more info, The piston diameter is 1.200" regenerator is .500" OD And the stroke is .35"
  13. May 1, 2016 #12
    Im not an expert in this by far but as far as I have understood it all so to get down to low temperatures you among many other things need to cool the compressed hot gas (air in your test?) as much as possible and also prevent any losses as far as possible. If you want to get down to cryo temperatures then it is an absolute must.

    Don't know if this next thing makes much difference in your design but to fully utilize the regenerator the material you use need to be the one that was planned to be used and thus not fex the material of a thick walled tube which draw out the heat from the generator material and will not release it back as fast or at all as the regenerator material was design to do.

    The use of o-rings, I don't know if this might make a huge difference in your case but this could perhaps be one of the reasons since these make a lot of friction and thus heat. I would personally instead have tested to make anything that move with fex graphite mixed epoxy instead with closely made clearance to the cylinder. But the cylinder then need to be very smooth like the inside of a glasstube or the piston clearance would very fast be bigger and quickly loose its sealing ability.

    Btw, where have you got the calculations from to make the sizes of the regenerator and piston?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted