Cooling Rotary Stage with Conduction | Solutions

[nyxynyx]

I have a circular stage (made of stainless steel) that rotates continuously. How can i cool it by conduction? I thought of using water cooled copper braids that contact the sides of the stage, but it'll cause a lot of wear since the stage will be rotating a lot. Any suggestions on how i can go about cooling it?

 

[brewnog]

How much heat do you need to reject? Could you spray coolant onto the rotor? Could you chill the surroundings? Why is it getting hot?

 

[nyxynyx]

The stage will be holding a sample that gets heated to around 500 deg C.

I cannot spray any coolant because the stage will be in a ultra high vacuum chamber(sorry forgot to mention), which means I cannot generate too much loose particles in the process, and I can only cool the stage via conduction or radiation since there's almost no air in the vacuum chamber.

I cannot chill the surroundings (no air), but I can cool the stage itself. The sample does not need to be cooled, just the stage it is mounted on (and everything in contact with the stage should not get too hot)

 

[nyxynyx]

Here's a view of the stage.

 

[Redbelly98]

Can you thermally insulate the stage from the sample better? Perhaps by minimizing the contact area between the stage and sample holder? Or use some kind of insulating sheet in between?

 

[mgb_phys]

It is very difficult to cool the stage purely by radiation, even with multiple overlapping black painted vanes ( it's a similair problem cooling a filter wheel in a cryostat ).

I would put the rotation stage outside and use a rotary feedthrough.

 

[nyxynyx]

I'm planning to use a ceramic disc to thermally insulate the sample from the rest of the instrument, but I think I will still require a cooling block to remove the heat. So I will have to cool the cooling block eventually.

There is a shaft (stainless steel) that connects everything together, which I think will still allow heat to be transferred from the sample to the other components. I don't think I can minimise contact areas further.
Haha I guess cooling radiation's probably out of the question. I have to stick to conduction methods.

I cannot place the other components outside the vacuum chamber, everything has to be inside. This stage will be connected to another stage, and will have 3 axis of rotation in total.

 

[mgb_phys]

You haven't said how much power - is this a microgram of sample heated to 500deg or a 100kg part being e-beam welded?

If you need to remove a large amount of heat the only way I can think of is a recirculating fluid - there are gas tight rotating fluid feedthroughs. Basically a concentric pipe with o'rings and magnetic fluid seals.
You can run either gas (eg helium) or oil through them - Helium is easier to clean up when it leaks!

edit - oops just read the 3deg of movement, that's tricky!
Do you need continual rotation in all axis or can you have a cable wrap that goes through 360 deg and then back again?

 

[nyxynyx]

I'm not sure of the exact power. It's a 2in x 2in (x1-5mm)sample that will have sublimed metal ( using a electron beam evaporator, metals are like gold, titanium) depositing on it.

I need only continuous 360 deg rotation on one of the axis. This continuous rotation will be provided by the stage whose image I've posted a few posts before. I can have cables going from outside the chamber to the stage. The other axes of rotation are 360 degrees but do not have to be continuous.

 

[brewnog]

Well you could have a swivel hose attached to a cooling gallery inlet at the centre of the table.

 

[Redbelly98]

Is the sample just resting on the stage, or is it rigidly attached/bolted? If it's just resting there, you could make a simple 3-point contact so that very little heat gets transferred to the stage.

edit:
What material is the ceramic disk? Stainless steel would be a better insulator than aluminum oxide, for example.

 

[nyxynyx]

The swivel is a nice idea! I'll try that.

The sample is rigidly attached to the stage via 3 screws. I think raising the sample such that there are only 3 contact points, as opposed to the entire sample touching the stage will slow down the heat transfer. But the heat will eventually accumulate and gets hot?

The ceramic disk is made of ceramic. I'm not sure what kind of ceramic is suitable, have to look more into that.

 

[Redbelly98]

If you do nothing to cool it, and wait long enough, then yes eventually the stage would get hot -- but the heat load and cooling requirement will be much lower with just 3 contact points. Um, the screws will transfer heat as well. How long will one of your coating runs last?

For even less heat transfer, the insulating disc could have a thin-walled section as shown in the attached figure. (Sample is on top of the disc, rotary stage is below it.)

Macor is a machinable ceramic with a low thermal conductivity (1.4 W/m-K), and would be a good choice for the insulator. For comparison, aluminum oxide has a thermal conductivity of 25-30 W/m-K, about twice as conductive as stainless steel. But stainless could be machined to a very thin wall, which would have a low conductance, and gets used as an "insulator" for that reason.
http://www.accuratus.com/macorprps1.html

edit:
Whatever you end up doing, you might use a thermocouple to monitor the stage's temperature during a run.

 

[FredGarvin]

You could also look into some peltier cooling if your power requirements aren't too high.

 

[Danger]

How about a hollow metal drum sealed onto tubular axles? Mount it so it impinges upon the rim of the stage like a pressure roller, maybe greasing it up with heat-sink compound, then pump coolant through it from the axles.
I don't think that it would work because the contact area is so small, but I'm throwing it out there as an 'just in case'.