NUBM44 Dry Ice Run

[waffailtrager101]

Thats a great link you found Andrew124C41, somehow in my hours of searching for a good tutorial, I never came across this one.

Good find :beer:

 

[Andrew124C41]

Bottom line is there's no free lunch when it come to removing waste heat. It all comes at a huge cost, whether you use forced air circulated through heat fins or use a TEC and remove heat from the back side.

Yep, evil has a way of being persistant!

 

[Andrew124C41]

Hey, let's face it, everyone should be able to mix up a batch of liquid nitrogen. I used to get it the easy way from a company...for the office....to burn off small lesions...warts...and other growths. Being able to produce it yourself...,.now that is something!

 

[paul1598419]

Unless you have some of those components already, making liquid N2 is going to cost more than $500 to do in sufficient quantities.

 

[logsquared]

So, I got to thinking that running back and forth to the store to pick up dry ice to cool the hot NUBM44 was getting a bit old. Paying high school kids to wave a hand fan at the diode proved not only inefficient but expensive.

While what I am about to propose is also inefficient, I wonder if it would work. One would have one set of batteries for the diode and another for cooling. Perhaps putting the 44 into an aluminum radial finned heatsink with one of these in between the sink and the module.

Otherwise, we are going to have to use mind over matter and just keep thinking cool!

Peltier Cooling

How about using ice? Make a cylindrical cavity with oring seal to house an ice cylinder in the host. Then make an ice tray from silicone or something to freeze ice "cartridges". I would guess you could get a 1/2 hour or more continuous duty from 1/2 pound of ice .

 

[Light superglue]

Yes, with its high enthalpy of fusion ice is the best way to absorb a load of calories.

Andrew, do you want the cooling for an outdoor or indoor device?

I have once built a setup for keeping BDR-209 at 2°C, look at pic. in post 7 here:
https://laserpointerforums.com/f38/2-w-single-mode-combining-4-lds-theory-100356.html

The right shaped Al piece can be found in this kind of mini mug coolers:

https://www.ebay.com/itm/NEW-DRINK-...ash=item4d566a4003:g:x1wAAOSw2gxY0rzC&vxp=mtr

Sure my setup was for a test at home but one could use cold coke cans as cartriges when outdoors or even make a "torch" with a sealed version of this...

 

[RedCowboy]

I have cooled my large heat sink units down by rolling them in my freezer against the frost build up, problem is if I cool it too far the lens fogs up, yours in the pic looks fogged up.

Maybe use a TEC pad between the lasers holder and the cold aluminum so you can regulate the temp.

 

[paul1598419]

I agree that the best way to regulate a diode's temperature is with a TEC on a square or flat module with a feedback loop. That way you can set and keep the temperature at a desired level without causing fogging of your lenses. I have been planning to build one of these in the near future to keep the diodes I test at a single temperature regardless of the waste heat generated by the diode itself.

 

[logsquared]

TEC's are only about 10% efficient. So plan on 130 plus watts just to run the TEC element. Not to mention the giant finned heatsink needed. Also need power to run a fan on the hotside plus the 20 watts to run the diode. Doesn't seem practical for a handheld device with self contained power source.

A small resistor glued to the lens should keep the fog away if its a problem. Or design the thermal resistance of the diode mount to keep the module temp where you want it (above dew point)

 

[paul1598419]

That's true for the 44, but mine is as a test module for far less powerful diodes. It is to maintain the 20* C or so, to test wavelengths of new diodes that have been coming out just recently. If I can maintain the temperature, the shift in WL will only be caused by current increases.

 

[deadshadow]

you guys are right about the lens foggin,i got the same problem sometimes on my tec cooled 638nm build when the temperature gets close to 0c(5 or less)

 

[Light superglue]

Ha ha, do not you find it a little strange to discuss lens fogging in a topic where Dry Ice and liq. N2 coolling were suggested?

On that pic. I indeed put much more ice than it was needed for BDR-209 into the cup. But one could also put less and keep water in the cup at 10 or 20°C adding more ice ocasionally.

And once you change BDR for NUBM44 and turn it on - how many seconds will it take for all fog to evaporate from that lens?

One day I may attach a NUBM08 block with T sensor to this cup and see how much time will it take for all ice to melt...

 

[Light superglue]

Hi logsquared,

You have written that TECs are only 10% efficient.

I just have bought the so called "Beverage cooler&warmer" from ali and checked it. It cools 300g of wáter put inside of the can at rate of half ºC pro minute from rt to about 2-3ºC (that takes 1h but at rt cooling starts at rate of 1/2 degree pro min).
Knowing that 1 calorie (needed to cool 1g of wáter at 1ºC) = 4.2 Ws, it makes 10.5W of heat extraction while energy consumption (as was given in specs) is 12Vx3A = 36W.
So the TEC in it must be 29% efficient (if we only care about wasted energy).

I took the thing apart - the TEC inside was not bigger than ones purchased from ali before.

And if something is cooled from temperaures higher than rt to rt, the efficiency should be even higher, should not it?

 

[Mattronium]

Hi logsquared,

You have written that TECs are only 10% efficient.

I just have bought the so called "Beverage cooler&warmer" from ali and checked it. It cools 300g of wáter put inside of the can at rate of half ºC pro minute from rt to about 2-3ºC (that takes 1h but at rt cooling starts at rate of 1/2 degree pro min).
Knowing that 1 calorie (needed to cool 1g of wáter at 1ºC) = 4.2 Ws, it makes 10.5W of heat extraction while energy consumption (as was given in specs) is 12Vx3A = 36W.
So the TEC in it must be 29% efficient (if we only care about wasted energy).

I took the thing apart - the TEC inside was not bigger than ones purchased from ali before.

And if something is cooled from temperaures higher than rt to rt, the efficiency should be even higher, should not it?

TEC efficiency depends on how hard the TEC is run and how much heat is moved. I found a useful peltier calculator that I've been using to test some values for a multi-stage peltier cooled 638nm labby that I am working on.(TEC Calculator)

 

[Light superglue]

If anyone is still interested in the topics around crazy laser ice cooling subject I dismantled this beverage cooler/heater, there was a TEC with no marks but current was 2.6A at 12V (will post pictures later when my handy finds connection), so 31W were consumed and the TEC should be one of those 12703 ones (3A rated).
I took regulating PCB out, put TEC back in and just soldered to work at full power all time. Added temperature reader and in 10-15 min the empty Al can came down to -6°C.

With 300g of water cooling speed of 0.5°/min (from my calculations above meaning 10.5W heat extraction or 34% TEC efficiency) was only at start when T difference between TEC sides was still very low but 1h after when water became 1.6°C cold the cooling rate fell to 0.2°/min, so the efficiency down to 14%.
And it took almost one more hour of work until the water started to freeze into ice.

I wonder what would be a better way to increase speed of cooling here - replace this cheap 3A TEC by a 6-8A rated TEC or make a multi-stage TEC from two 3A ones. If efficiency depends highly from T gradient (and heat backflow because of it) could it be a better way to increase efficiency of the whole setup, could not it?