Ebay Stirling Cryocoolers

[CM800]

Hi All,

I found a seller on ebay selling large quantities of these Stirling Cryocoolers:

I am considering getting one of them, the price looks pretty good (Is it?)
My main concern is the chances of them being broken / gas leaked out.

http://www.ebay.co.uk/itm/RICOR-Mini-CryoCooler-Stirling-CryoStat-Liquid-Nitro-IR-Detector-FLIR-526-/231840615874?hash=item35fac70dc2:g:NFIAAOSwQYZWueiZ

[PointyOintment]

It's split-cycle (two pistons in separate modules) and the connecting tube in the representative photo looks like they might have been bent somewhat carelessly, or yours might get bent carelessly in shipping. You might end up having to replace the tube and refill the helium. That might be more difficult than a regular plumbing job because indium seals are often used for helium, and it might be necessary to compress the helium (depending on operating pressure and the pressure of your helium tank), which is probably difficult.

I've been looking into building a cryocooler recently, though of the pulse tube type, which has only a single piston. I still don't know how I'll compress the helium if necessary, and purify it if necessary (because Balloon Time helium tanks now contain up to 20% air, and industrial helium tanks seem to be hard to get). I've found some piston seal technologies that might work, and if not, I'll just put a sealed chamber of helium around the back of the piston so leakage doesn't matter other than for efficiency.

[Fraser]

I have a little experience with Sterling coolers in thermal cameras.

1. The fill of Helium is at around 200psi
2. The Helium fill leaks out over time, so reducing pressure and the ability to reach normal operating temperature (-196C)
3. You cannot tell a good sterling cooler from looking at it.
4. Untested Sterling coolers should be considered faulty and priced accordingly. A broken Stirling Cooler is an expensive paperweight.
5. A Sterling cooler that is described as working can mean many things. The motor runs, the cold end gets frost on it or more helpfully, the cold end reaches -xxxC. A Stirling cooler with low gas fill will still cool, just to not the normal operating temperature.
6. I have yet to come up with a safe way to refill a Sterling cooler. The fill port does not appear to contain a valve. It is a conical seat into which a threaded plug screws and indium is used as she sealing gasket. This needs a clever adapter to allow turning of the threaded plug whilst under 200psi of Helium pressure. I have all the required gauges and pipe work though.
7. Ballon Helium from BOC in the UK is pretty good but not ultra pure, as normally required for a Sterling Cooler. Any water content will freeze, causing piston issues. Ultra pure Helium may be ordered from BOC but costs around £150 in total and you are paying rental on the cylinder. The Cylinder is at around 250psi so a regulator can provide the required 200psi fill pressure for the cooler.
8. Air is being added to Helium tanks for ballon filling to protect the worlds reserves of this gas. A 20% air to 80% Helium mix is fine for balloons but will ruin the helium for Sterling Cooler use in my cameras.
9. Suicide by Helium gas hood is a well known method for those wishing to end their lives. For this reason, some suppliers are very suspicious about why you want pure helium as opposed to balloon helium. The 20% air in balloon Helium tends to make it non lethal.

I have two Sterling coolers that need a Helium refill. The OEM isn't interested and boding it and it would be very expensive anyway. If you buy an old Sterling Cooler, expect it to have lost most or all of its Helium fill. Helium molecules even get through the aluminium casing material of the cooler !

Fraser

[Fraser]

You have not specified what you intend to use a Sterling Cooler for. Remember that applications that require the lowest possible temperature have to use a vacuum Dewar around the Cold Finger and whatever it is cooling. Cooling capability is severely limited. In thermal cameras, the sensor die is physically attached to the cold finger and enclosed in the vacuum Dewar, with a Germanium window in the end.

Fraser

[mackrc1]

Price is definitely good. Looks like they are coming out of old RF gear. I'd be surprised if they didn't work currently however after shipping I'm not so sure. That tube does look rather flimsy. I've never seen one with the cold finger separate before.

[CM800]

You have not specified what you intend to use a Sterling Cooler for. Remember that applications that require the lowest possible temperature have to use a vacuum Dewar around the Cold Finger and whatever it is cooling. Cooling capability is severely limited. In thermal cameras, the sensor die is physically attached to the cold finger and enclosed in the vacuum Dewar, with a Germanium window in the end.

Fraser

I really want to set one up to make Liquid Nitrogen or Oxygen. Also want to mess around with some superconductors.

Thanks for the information given. Do you have any ideas on how I can actually get a Stirling cryocooler or anything along these lines. How much does this kind of thing actually run new?

[CatalinaWOW]

There are published accounts of successfully using these for generating LN2, but I have avoided the temptation to follow this path.  Cryocoolers are notorious for short life.  Life in 1000-2000 hour range is common, while even the best only do an order of magnitude or so better.  Leakage of the operating fluid is one failure mode, but seal failure on the regenerator is another common one.  Making seals is something of a black art.  I was once involved with a vendor of cryocoolers who had temporarily lost the recipe for his version.  Replacement of the seals is basically not possible.

Before I bought I would think long and hard about the fact that these coolers are on the surplus market for a reason. 

[jwm_]

Hi, I have bought three cryocoolers from ebay, they are pretty darn neat. You can be the envy of every volt-nut if you rig of a josephen-junction voltage reference.

just some notes, there is a non trivial chance the cooler will be broken no matter what they say, some are _very_ fragile once removed from whatever they were in and the helium can leak out of an invisible leak. if you can't take the financial hit from a broken one, then don't get it. My record is 2 for 3 working, but one of the working ones has just a beatuful IR sensor on it i can't bring myself to remove. I want to make a fantastic FLIR out of it if i can figure out how to talk to it.

Most of the ones on ebay come from missles or missle tracking systems for cryocooled IR (heat seeking) sensors, usually salvaged from somewhere in the middle east, expect them to be beat all to hell and you may be put on a watch list or something, never had issues though.

two of mine are straight DC, one is brushless with hall feedback wires, don't know the general distribution.

If you don't have a way to dry the air you will have an instant ice maker. those flower preserving crystals in a beaker with a tube running through it worked fine for me to dry the air enough to get ln2/lox out of them.

[HAL-42b]

You can always recharge with Hydrogen. That is if you like explosions 

[jwm_]

You can always recharge with Hydrogen. That is if you like explosions 

pressurized hydrogen and metals don't play nice together.
https://en.wikipedia.org/wiki/Hydrogen_embrittlement
I don't expect it to last long with hydrogen independent of explosion risk.

[HAL-42b]

You can always recharge with Hydrogen. That is if you like explosions 

pressurized hydrogen and metals don't play nice together.
https://en.wikipedia.org/wiki/Hydrogen_embrittlement
I don't expect it to last long with hydrogen independent of explosion risk.

To my knowledge rarely a problem unless high temperatures are involved. Stainless steel can retain much of its strength while exposed to liquid Hydrogen.

[jwm_]

You can always recharge with Hydrogen. That is if you like explosions 

pressurized hydrogen and metals don't play nice together.
https://en.wikipedia.org/wiki/Hydrogen_embrittlement
I don't expect it to last long with hydrogen independent of explosion risk.

To my knowledge rarely a problem unless high temperatures are involved. Stainless steel can retain much of its strength while exposed to liquid Hydrogen.

True, but depending on the cryocoolers situation, the hot side can get very hot. though, I use water cooling for mine so it can be kept below  100C so perhaps it isn't an issue.

I have been wondering if i can repair my broken cryocooler by repressurizing it somehow, but the break is a tiny crack in the line between the cold head and pump and there isn't an obvious fill port or way to seal the crack... if anyone has any ideas, it would be great to bring my third one back to life.

[sarepairman2]

<100 days of continuous operation only..?

not great for a thermal telescope.

[ejeffrey]

1000-2000 hours is definitely not a normal lifetime for cryocoolers, but if you are buying 10 year old obsolete device off ebay, you don't know how it was handled, and you can't contact the manufacturer for maintenance parts, then I could believe that.

[PointyOintment]

Ben Krasnow's Stirling cryocooler came from a cell tower, where it was used to cool superconducting RF components. I doubt they replace those multiple times a year.

[coppice]

Ben Krasnow's Stirling cryocooler came from a cell tower, where it was used to cool superconducting RF components. I doubt they replace those multiple times a year.

Those cellular coolers are rated for 5 years continuous operation, like most other parts of a basestation, and need to achieve 77K.

[Wolfram]

The STI Superfilter cryocoolers, as used by Ben Krasnow, are a bit easier to work with than the split cycle ones. They consist of a single unit, which is less fragile, and the helium fill port is a pinched off metal tube like the ones used on Helium Neon lasers, protected with epoxy. They also seem to be pretty reliable: http://link.springer.com/chapter/10.1007/0-306-47919-2_11

[Stray Electron]

  I used to work on military FLIRS. The Sterling cycle coolers that we used cooled the IR sensor to below 8d K but only had 1 W cooling.  The smaller one only had 1/4 W cooling. At that temperature, any kind of lubricant would freeze solid so the Sterling cycle pumps were unlubricated. Therefore the operational life was very short; tens or at most hundred hours or so.  The cold finger and the IR sensor were contained in a well insulated dewer flask and we also had to limit the current to the motor after starting to reduce the heat generated to absolute minimum. The entire system was contained in a well cooled pod.  The point is, that even if these still work, their cooling power and life span is extremely limited and they require exceptional conditions to work at all.  Oh, and the incoming optics was heavily filtered to only pass the wavelength of interest and to block all other wavelengths since they would add to the heat load on the cooler and in many cases could actually burn out the VERY sensitive IR sensor.

    BTW the motor on all the ones that I've seen require 115 VAC 400 Hz 3 phase power and as I said there was a dropping resister assembly that reduced the voltage and current to the motor after it started and was up to speed.

    One more thought, I've played around with modifying old digital cameras (Olympus E-1 and others) for IR work and I've looked into cooling the image sensor to increase their sensitivity, but what I found was the material that is used in most of them (ALL that I looked at) does NOT get more sensitive as they're cooled.  You can't just cool any sensor and make it more sensitive. You have to you very special material that is chosen for the temperature that you intend it to operate at.  I strongly suspect that that would also be the case for electronic devices.  Cooling them a few tens of degrees might improve their performance but cooling to cryogenic temperatures is probably a waste of time and very likely would damage them. 

   FWIW.

[CM800]

So the general consensus is that It's not worth buying one unless you have the cash to waste on the chance of a broken one... fair 'nuff.

What about the ones on MRI scanners? I believe they use larger versions of these too. I might contact one of my buddies and see if he can ask one of his that work in the industry.

[Photon939]

So the general consensus is that It's not worth buying one unless you have the cash to waste on the chance of a broken one... fair 'nuff.

What about the ones on MRI scanners? I believe they use larger versions of these too. I might contact one of my buddies and see if he can ask one of his that work in the industry.

Those small ones have very tiny cooling capacities, useless for doing gas liquefaction. The heat leakage from your collection dewar is likely larger than the cooling capacity.

If you can get your hands on an MRI cooler you would probably be in business but may be a bit too large for experimenting. (large power requirements or possibly 3 phase) I don't have experience with those but I do have one of the STI cryocoolers. See if you can find one of those.

[CM800]

So the general consensus is that It's not worth buying one unless you have the cash to waste on the chance of a broken one... fair 'nuff.

What about the ones on MRI scanners? I believe they use larger versions of these too. I might contact one of my buddies and see if he can ask one of his that work in the industry.

Those small ones have very tiny cooling capacities, useless for doing gas liquefaction. The heat leakage from your collection dewar is likely larger than the cooling capacity.

If you can get your hands on an MRI cooler you would probably be in business but may be a bit too large for experimenting. (large power requirements or possibly 3 phase) I don't have experience with those but I do have one of the STI cryocoolers. See if you can find one of those.

Easier said then done finding one of the STI versions in the UK / EU

[PointyOintment]

  I used to work on military FLIRS. The Sterling cycle coolers that we used cooled the IR sensor to below 8d K but only had 1 W cooling.  The smaller one only had 1/4 W cooling. At that temperature, any kind of lubricant would freeze solid so the Sterling cycle pumps were unlubricated. Therefore the operational life was very short; tens or at most hundred hours or so.  The cold finger and the IR sensor were contained in a well insulated dewer flask and we also had to limit the current to the motor after starting to reduce the heat generated to absolute minimum. The entire system was contained in a well cooled pod.  The point is, that even if these still work, their cooling power and life span is extremely limited and they require exceptional conditions to work at all.

Some of the reasons I prefer the pulse tube type for DIY. There's only one piston, and it's at one of the room-temperature ends (the cold part being in the middle, with heat being rejected halfway toward each end).

You can't just cool any sensor and make it more sensitive. You have to you very special material that is chosen for the temperature that you intend it to operate at.  I strongly suspect that that would also be the case for electronic devices.  Cooling them a few tens of degrees might improve their performance but cooling to cryogenic temperatures is probably a waste of time and very likely would damage them.

I remember an experiment a few years ago where some overclockers (I think) put a Raspberry Pi in liquid nitrogen. I don't remember if it was more overclockable than usual, but I'm pretty sure it was at least stable at regular speed. Similar experiments have been done with AVRs as well.

[edpalmer42]

I only heard about cryocoolers a few months ago.  If you look on Youtube, you'll find videos from people who got the cryocoolers from cell towers.  They are a one-piece unit.  Basically, add some copper onto the cold finger to increase the surface area, plug it in, and produce somewhere in the range of a liter per day of LN2.  Is that just too optimistic?



Did he win the lottery?

[wkb]

Did he win the lottery?

Well, maybe not the lottery but well underway for a Darwin Award of sorts.  Playing with cryogenic liquid in a total mess like that without any sensible protection is downright stupid.

[CM800]

Should I just keep looking on ebay, or is there anyone in the UK that has been successful elsewhere? I could budget up to £450 on this if I can get something good.

[CatalinaWOW]

It varies by region, for but LN2 is pretty cheap to buy.  Cheaper than gasoline.  Unless you have continuous use it buying LN2 is a far more economical way to go than to set up to produce it.

That said, the nerd coolness factor of making your own is over the top.

[CM800]

It varies by region, for but LN2 is pretty cheap to buy.  Cheaper than gasoline.  Unless you have continuous use it buying LN2 is a far more economical way to go than to set up to produce it.

That said, the nerd coolness factor of making your own is over the top.

I agree on the latter!

The main reason is that I only really want to use it periodically. I'm mostly interested in actually making the unit, mostly just messing around.