Extreme cooling loop

Hello, this is my first, and likely one of my only threads.

I come to you from a laser forum, looking for a method for closed-loop cooling, to super cool a laser diode.

The basics.

I want a copper block, with two tunnels drilled out that will connect to a continuous loop of super-cooling liquid. In this copper block, the laser will be mounted. though perhaps I will simply line this block with indium foil, and mount the laser to it. Either way, that's neither here nor there.

I originally planned on using LN2, though I read the pressure to liquefy the LN2 as it cycles through, would be immense.

Also, condensation here is not an issue, I can have water drip off this thing all day long, I could care less. The laser itself will be protected, so if the cooler itself gets wet, no biggie.

That's really all of it. If you have any ideas, let me know. I'd like to get <100C if possible. I've seen -191C from a LN2 setup, but it was from a gentleman pouring the LN2 onto the laser basically. I'd like a more elegant design if possible.

Something like a multi stage cascade coupled to the block that your laser is mounted to would probably work best, and they can easily get below -100, plus you can switch it on and off as desired as it is electrically powered.

Just an FYI, I don't build computers or know anything about cooling a CPU. So the less jargon, the better.

Though before someone jumps on me to use the search function, I will do some googling and such.

Thanks for somewhere to start.

Ah, so layered TECs. That's something I can work with.

Is there any ratio to how many layers, or output wattage that equates to certain temps?

You need to be able to remove the heat from the TEC's which takes a lot of energy and getting to a temp of -100 with them is going to be a major undertaking if at all possible, a cascade is the only option I can see to get lower than -78or if you wanted a temp closer to -78 you could use dry ice

Yea, continuous operation sounds like a phase system to me as well. You will probably want something custom built for you, where are you located? There may be someone local.

Something I forgot to mention, laser diodes heat up quite a bit. So there will be a lot of output that a cascade won't be able to cope with. At least, not while maintaining the temperatures I need.

Do any of you know of any decent "elegant" methods of using LN2? I'm not looking for some large bulky apparatus that requires constant refilling, or a lot of area to cool, thus requiring a lot of liquid.

Location? Currently stationed at Offutt AFB out here in lovely sunny Nebraska.

How much heat output are you talking about?


Edited for mobile grammar...

Just your ordinary output heat from any self-respecting electronic.

With more current, the diode will continuously heat up, until it eventually overheats itself. Typically if overdriven without a heatsink, they can get hot enough to make it terribly uncomfortable to hold one.

However, the diode will be sealed in a large Cu heatsink.

As far as how hot, or how fast, all depends on input current, threshold current, and actual laser output. These diodes are not terribly efficient. Figure 1.5W max heat output.

A cascade can hold something putting out over +100c heat at below -100, how large is the laser/housing you want to cool?

Was going to say... Some cascades are tuned to 200w+ at -100c. 1.5w is nothing for the smallest of cascades. I mean we use them for cpus which some fan output over 200w and they are kept low.

I also remember seeing a LN2 maker that you can run, though I believe it will be cost prohibitive, especially versus a small cascade.

Also, since 1.5w is so little, you can likely just use ln2 straight up on the copper block that surrounds the diode. Is it important to keep the laser at a very stable temperature?

A picture of the device in question and the copper block to cool it would be helpful.

Cascade cooling would easily do that.
There are some custom builders and some that build for scientific and industry that post over at the xs phase change cooling area. I would post there for more info. http://www.xtremesystems.org/forums/forumdisplay.php?80-Vapor-Phase-Change-Cooling
With that small of a heat load, you might be able to use liquid nitrogen on a drip and just let it evaporate. Like we do with our liquid nitrogen. I suppose it depends on your run time.

Give Ron at Under the Ice a call.

http://www.under-the-ice.com/

A stable/constant temperature is not required. I simply need low.

Basically, this laser runs 638nm (red-orange) at room temp. At -191C it's been clocked at 608nm (very bright pure orange).

The laser diode is only 5.6mm in diameter. The module it will be in will be 12mm in diameter. Though the Cu block I plan to put it in, likely 4x4x2". Though I am considering Al, and if I go that route, 5x5x2".

Though that block will be inside a second metal closure, with desiccant (moisture absorbers like what you find in beef jerky) inside the second enclosure. This will be completely air-tight to prevent moisture condensing on the laser optics. This outer enclosure is to be cooled. I could always put this second enclosure in a pan/tub of sorts, and pour LN2 over it. Though I feel it would evaporate too quickly.

Have you thought about an R134a refrigeration system? 1.5W is almost nothing compared to any appliance or overclocked computer.

how many amps and what voltage is this diode using... unless your running 300w+ of power though it a cascade will be able to cool it just fine. it doesnt matter how hot you think its going to get a cascade designed to pull 200w+ while maintaining -100 degrees is going to be able to handle pulling that amount of heat from what ever the cold bit is attached to.

im sure coming from the laser forums you know how to calculate wattage... volts x amps = watts so as long as the interface you design to attach the cascade system to can transfer the heat efficiently enough it is going to cool it just fine, if not you could submerge the whole thing in ln2 and it isnt going to make a difference.

edit: sorry im on my htpc and missed the part where you said 1.5w part... even a small cascade could handle that all day and not break a sweat, they are designed to keep 200w cpu's at subzero temps a 1.5w diode is a cake walk, and about the only realistic way to keep it that cold for long periods of time without refilling ln2/dice even though with either of those a little would go a long way with such little wattage. ln2 would easily bring it sub 200f probably even sub 300f right around 200c. dry ice is gonna bring it to about 85c give or take either of those are fine if you are wanting to run it for short amounts of time and dont mind feeding it.

Bloompyle said:

I could always put this second enclosure in a pan/tub of sorts, and pour LN2 over it. Though I feel it would evaporate too quickly.

Click to expand...

What I am saying is... it wont really... if you cool this Cu or Al medium to LN2 temps, and set it in an LN2 bath, I highly doubt the meager heat being applied it would flinch.

How long are you trying to run these lasers? Is pouring some ln2 viable?

I added 1/4" copper tubing wrapped around the base of the block. I attached it to a 230L pressurized dewer to force feed LN2 to the base of my block taking me colder than before.

Click to expand...

I wonder if something similar to what Fugger did here would work.

http://www.xtremesystems.org/forums...cascade-2014&p=5225742&viewfull=1#post5225742

MetalRacer said:

I wonder if something similar to what Fugger did here would work.

http://www.xtremesystems.org/forums...cascade-2014&p=5225742&viewfull=1#post5225742

Click to expand...

I was just about to say I saw on Facebook that Fugger has a new system to run a cascade with a nozzle to add LN2 as necessary. Great minds think alike.

So a cascade could get the OP ~-100-110°C 24/7, with LN2 on demand to go colder.