High Power Cooling / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

High Power Cooling

previous 1 2

Move Thread

LAN_403

Steve Conner

Wed Jan 24 2007, 03:42PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I just took self-contained to mean "Doesn't need water and drain connections", since the no-brainer way of keeping something around 25'C is just to run water from the tap through a waterblock to the drain, and twiddle the flowrate to get the desired temperature. At least it is here, where the domestic tap water is only about 18'C even in summer.

But now I guess you're building a high power diode pumped laser, and you want it to look neat and finished like a commercial laser system? Peltiers and radiators would make a much more compact, neat, and pro looking system than random buckets and chiller units.

"Neat Looking" wasn't part of the spec though, so we just tried to think of the cheapest and nastiest solutions that didn't actually need water lines

PS: I know fine well about laser diode temperature tuning, since I've been designing commercial laser and TEC drivers for the last 6 years or so. But I was under the impression that you just wanted to get some diode bars working long enough to burn holes in stuff with them, and you didn't know that they don't need to be exactly 25'C to work. I guess I underestimated your laser skillz.

ConKbot of Doom

Sat Feb 10 2007, 07:57AM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

I'm not even sure if a beer chiller/refrigerator would be up to snuff for this, not for extended periods of time.

If there is ~1kw of heat going into the fridge, its going to have a ton more energy being put into it then normally. and running the compressor all the time when it wasn't designed to will overload it

According to google calc, 1 kwh = 3412.14163 btu

5000BTU window air conditioner
and a 5K BTU window air conditioner is only $99 at home depot, even cheaper if you get one used. Aim the cold air exhaust duct from that into a radiator for the watercooling loop, and recirculate that around to the intake of the evaporator, and there you have your cooling.

...

Sat Feb 10 2007, 08:30AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

If I were to use phasechange cooling, I would need a source of 3gpm of water at 10c, and be able to keep it at 10c with a 300w load. That shouldn't be all that hard decent sized fridge, but you do bring up a valid point that the compressors are not designed to run continously.

But for now the decision has been made for pielter cooling (and the parts bought), so hopefully that will work. The requirnements actually look a lot looser now, as it looks like the rod is only doped for ~60w of pump, so I only need to run each diode for 30w of lasing (~80w a piece) so I am getting closer to 200-250w on the pielters, which should allow another few "C of leyway. For the current renders see my projects thread.

Or if you are too lazy to let that page load... (that clear plate in the middle is actually made of copper, just set to be clear for clarity)

ConKbot of Doom

Sat Feb 10 2007, 09:16AM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

... wrote ...

If I were to use phasechange cooling, I would need a source of 3gpm of water at 10c, and be able to keep it at 10c with a 300w load. That shouldn't be all that hard decent sized fridge, but you do bring up a valid point that the compressors are not designed to run continously.

But for now the decision has been made for pielter cooling (and the parts bought), so hopefully that will work. The requirnements actually look a lot looser now, as it looks like the rod is only doped for ~60w of pump, so I only need to run each diode for 30w of lasing (~80w a piece) so I am getting closer to 200-250w on the pielters, which should allow another few "C of leyway. For the current renders see my projects thread.

Or if you are too lazy to let that page load... (that clear plate in the middle is actually made of copper, just set to be clear for clarity)

Hmmm well then perhaps a large buffer would be suited, as using a chest-style freezer full of water, close to freezing, or saltwater close to freezing. Add a heat exchanger of some sort so you aren't piping the excessively cold water to your diode blocks. I did all the math earlier before I figured out an air conditioner could handle all the load, and 200 gallons of -17C saltwater (assuming still close to 1cal/CC, not sure how much the salt affects the specific heat) would take something like 46 hours to heat up from -17 to 35F with 1kw of heat input. Not sure if I did all my math correctly there. but if I did, it should be somewhat workable. Not sure if 200 gallons in a freezer is even reasonable, but even if its half that, I think 23 hours shouldn't be too bad of a limit...

Though I was talking about using the air conditioner to cool the water which cools the peltiers, not replacing the peltiers as with a waterchiller. Rube Goldberg cooling anyone?

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob