Super "chilled" ethanol Cooling system for NUBM44. Pic Heavy

[Down with Umbrella]

Preface

This experiment/build has evolved and evolved over time in many different directions. First the idea of the host, I welded it at work on lunchbreak, then I moved onto the batteries and driver. The next and final step was cooling this system. I have no formal experience here but I am always willing to experiment and learn :na: In this thread I will discuss:
-the build of the host
-My choice for ethanol
-The coolent system a trial run for a wavelength Shift in a future build.
I did this with limited means so it's more of a proof of concept theory not a true experiment like I wanted to do. I would need:
-an Ophir head to measure wattage and plot values over time.
-A spectrometer to measure WL shift.
-the proper thermometer to measure sub zero temperature!



Here is where I will start!
While NOT technically considered "super cooling" as that would imply I altered the nucleus of a Crystaline structure of a liquid to prevent the transition into a solid. I did cool down liquid ethanol to approximately -60C(-75F) via MULTIPLE dry ice baths in a 2000mL flask. (I say approx. because my thermocoupling died. And a regular thermometer bottoms out at -20c) the ambient temperature was 3C. I picked up 10 LBS at the super market and started dumping ice in the flask. 100 grams at a time. Dry icy expands very rapidly especially in an ethanol bath. 100grams at a time was a good way of maintaining pressure and temperature. I pre-cooled the ethanol before hooking up the laser's coil because the dry ice to tepid ethanol is quite violent. I prefer pellet type dry ice but what I got were solid blocks I had to smash with a hammer. :cryyy:



The only thermometer I have tapped out. And broke as well...
INSULATED GLOVES ARE A REQUIREMENT WHEN HANDING COOLED ETHANOL


Host construction
Without the coolent system it's actually portable!
I welded schd 10 stainless pipe rings 1.5"rings and 2" rings to sandwich 6 1/8th inch SS tubes.
Well, it was kinda like building a ship in a bottle until I cut one of the tubes. There is a small stainless wire the slides up and down. As long as there is tension on the tubing the "Pin" will always stay in place. (See photo below regarding this pin) the benefit here is the host can be worked on and I can swap diode modules.


One purpose of this experiment was to cool a powerful laser head that has a copper core, aluminum body and copper tubing coiled around the body with a thermal paste to increase conductivity. The module itself was custom made by Ehgimus! Thank you friend :beer:


Battery tube- 3/4in copper tube with male threads for a brass cap. This holds the negative spring and at the top is the positive battery contact. Very low resistance here.


Working at the bar literally? Am I the only one who has a work space like this?
I have the HeNes in the back for encouragement.


The internals
Two switches a momentary and a toggle on/off. The on/off has a protective covering when I actually start lasing.
Buck driver set to 5amps
G2 lense
2x 18650 LG he2 batteries. 26650 would have been ideal but I didn't actually have two on hand.
Looking back I would have built this with bigger batteries...no brainier.


The bypass.


With the host complete its time to address cooling this beast.
Where was I?

Cooling
What's your first choice if you wanted to cyro cool a laser. Liquid nitrogen.
I have access to liquid nitrogen and argon at work but what I don't have is a vacuum flask to transport this. Anddd I'm not about to put this in a pop bottle so I had to explore cheap options that were available to me.
Dry ice and ethanol.
Dry ice was used because its temperature where it sublimates is -78.5C
Ethanol was used because of its extremely low freezing point. -114C.
I estimate the efficiency to be close to 80% regarding the solid CO2 to the ethanol which maybe a little generous.
Therefore there is no way the alcohol could freeze and it would maintain its viscosity.
Pumping the ethanol through the coil was tricky because most mechanical pumps would freeze up so I sealed the flask and allowed the sublimating solid CO2 to build pressure forcing the cold ethanol through the stem into the coil. I control the pressure from a valve on the flask which controlled the flow rate. safety glasses were a requirement because the flask could have blown up!! This is a noncyclical system. The spent ethanol was discharged into the reservoir then back into the main flask(here vv)with 100g of additional dry ice.

(Interesting note: ethanol's PH is just a hair over 7. [7.33 to be exact.] After all the dry ice the carbonated ethanol measured a ph of 5.5. from slightly basic to acidic!)

To be honest I wasn't sure what would happen. I'm Familar with how steam expands and contracts but never low low temps.
The coil
In this photo there are two different colored arrows. The black and magenta arrows show solid arctic alumina epoxy. Those two points are fixed. The blue arrows denote "loose" coiling; this allows for expansion and contraction when the laser module cools/warms arctic silver 5 grease was used here to maintain conduction.
The circle shows where I actually epoxied an AL heat sink for the driver! I'd say it's working!
Important note: moisture here on the driver was a concern. I used canned air to blow away any accumulation.
The brass tubing connectors were ditched. Hose and hose clamps worked perfect because there was no real pressure...



Up and running- OD 5+ glasses used here. I set the valve to a trickle and made various notes and photos.
Hmm something I did not account for, condensation on the optics. I didn't notice any difference in beam profile but in the future it might need to be considered. The ambient room temperature was a chilly 3C so that might have helped.




Let's get to some beam shots.




Conclusion.
Well folks it's been a fun build. 6 weeks id say working on it on and off.
Was I successful? I'd say the system worked, there were some things that I would have measured if I could have but I lack the tools. For now I will be happy with this..
I have read blue doesn't shift like the red 638s and that will be what I will try next . What did the color look like? Kinda hard to judge because you shouldn't be looking at 7 possible watts! However with the diode being pushed it's likely that it only shifted back to 445 and that's what it appeared to be.
Maybe a mit 500 will be a better choice and if you toast it It's not a pricy.

If you have read this far down appreciate your time. Comments, concerns, constructive criticism always welcome.
Jefferson

 

[Alaskan]

Nice work, I really like the way you put things together, looks very classy! Well worthy of the +rep I just give, outstanding. Wish you had a meter to measure the wavelength running at low to full power output.

 

[Benm]

I'd say keep researching it further

Dry ice and ethanol mixtures are a well documented lab method of cooling this rapidly. Since you don't get the leidenfrost effect it is actually superior in cooling speed compared to liquid nitrogren an such, and used for biological samples needing very rapid cooling.

As for the effects on this temperature on lasers, i'd be interested to see results in wavelength and power variation for a given current.

For the method used i'm not really sure you would need to make a cooler like that. Perhaps it would be easier to submerge the laser diode and optics in the cold ethanol since it's not that conductive to electricity.

And as a word of warning to others that want to try this: dry iced cooled ethanol is FAR more dangerous to work with than liquid nitrogen. It doesn't bubble away when you get it on you, but at that temperature is a very viscous liquid (somewhat like motor oil at room temp) that sticks to virtually anything. This is a great property if you're looking for rapid cooling, but also very dangerous. While you can just pour some liquid nitrogen on your hand and bush it off, this stuff sticks like napalm and will cause severe freeze burns that will not heal quickly because the remaining ethanol will kill live cells ones the injury warms back up to body temperature.

 

[Thejoker301]

Very nice to see how you put it all toghether. Creative idea

Like said above would really like to know the wavelength of this beast :beer:

 

[crazyspaz]

Ooh, the mod master speaks again I agree, throwing a p73 would give much more exciting results. I wonder how low you could shift the wavelength?

 

[Down with Umbrella]

Thanks guys! Benm, I updated the OP. That was something I should have included in the post. +rep When I can. I had Tillman leather gloves with a thick layer of insulation.
My skin never came in contact with the ethanol so that was good.

Yup, I pretty much have to go red hopefully orange I should say. If I did that I could use my radiant X4 and the wavelength would be noticeable in a side by side comparison.

 

[RedCowboy]

Very cool, literally.

 

[CDBEAM777]

WOW...now this is a cool thread !!! Remember..." Ice is Nice "....another famous Laser Scientist/Actor once said !! Thanx for the education/teaching ! See the attached pic...."you've earned it"
Beam Out...:eg::eg:

 

[Down with Umbrella]

Hahaha that's awesome cdbeamz. I'm saving that.
What is the poll? Oclaro 700 or a mit P73. Setting the diode at a more conservative amperage would be more app to shift I would think. Also if I'm not mistaken these diodes are open can. I wonder if the cooling would damage the diode even if the module is sealed. I've kill one in the past.

 

[Benm]

I don't think submersion in ethanol would kill a diode, open can or otherwise, as long as you allow it to change temperature gradually. Dropping one at room temperature into the cold ethanol could cause cracks and such because of the very sudden 100K temperature jump. This could be a problem for the window of a canned diode, but also for the die on either type.

One problem with open can diodes could be the difference between refractive index of air versus ethanol. Depending on how they did the reflective coating for the cavity it may (reversibly) ruin it. But as long as the reflective coatings are on the insides that wouldn't be much of a problem.

Aligning a collimating lens in alcohol would be different from in air, which would be a problem if you want to produce a beam from the unit. If you are just looking at wavelength you could omit that and look at the output straight from the diode though a glass dewar vessel or even flask.

I would recommend the dewar though, since water from the air will constantly freeze onto a single walled vessel containing cold ethanol.

 

[CDBEAM777]

Hahaha that's awesome cdbeamz. I'm saving that.
What is the poll? Oclaro 700 or a mit P73. Setting the diode at a more conservative amperage would be more app to shift I would think. Also if I'm not mistaken these diodes are open can. I wonder if the cooling would damage the diode even if the module is sealed. I've kill one in the past.

Well....Glad u like my warped humor !!! You are seriously on the right track !!!!

See this video....done by "Planters" Mr. Planters is just amazing !!! He is my hero !! He has some serious toys !!!

http://youtu.be/z8EuZwgdqQ4

OK....I vote for the Oclarro 700 !!! SOoooo...
Stoke that HH with some Dry Ice and Juice....and " Let'er Rip" !!!:san::san::san:

Beam out

 

[RedCowboy]

Hahaha that's awesome cdbeamz. I'm saving that.
What is the poll? Oclaro 700 or a mit P73. Setting the diode at a more conservative amperage would be more app to shift I would think. Also if I'm not mistaken these diodes are open can. I wonder if the cooling would damage the diode even if the module is sealed. I've kill one in the past.

I'm starting to prefer open can or removing the can, I just popped another window on a 7875 switching out lenses testing them at 65 feet in cool humid air.

I'm starting to see a pattern with cool humid air and a hot diode windows.

 

[Down with Umbrella]

Aligning a collimating lens in alcohol would be different from in air, which would be a problem if you want to produce a beam from the unit. If you are just looking at wavelength you could omit that and look at the output straight from the diode though a glass dewar vessel or even flask.

I would recommend the dewar though, since water from the air will constantly freeze onto a single walled vessel containing cold ethanol.

I've seen some submersion videos where they dip the diode in and observer the wavelength shift. I would like to see the collmulated beam.

Here's my other question mark. It you submerge a module with say a 3element lens into a clear beaker of ethanol it's possible a gas/liquid mixture would build between the lens and diode window because light passes through each differently this would change the beam profile. I would think.

If you sealed the lense threads could a pressure differential damage the diode? Not sure. Red cowboy might be on the right lines here. Even humidity.
Here is a photo of a 405 passing through alcohol with uv dye, mineral oil and water.

I don't have a dewer flask but I want one . They are pricy on eBay.

Open can diode just appear to be so much more fragile XD if that spring were to ever twist loose it could contact the wires and there's always the possibility of brass shavings.
With that said I have read many threads where the window breaks and you have to decann the diode. Never done it myself.

 

[CDBEAM777]

Doctor U....MMMMmmmm ???? I just think submersion...of the module into a liquid bath...is a bad idea !!! Maybe I am wrong ?? Dunno....but....your goal....is to lower the temp of the LD....in the most efficient manner. I speculate that submersion would be an optical nightmare.

Cool the diode via convection !! Deep cool the base pate....which....via convection chills the LD mount...which...in turn ... cools the LD....the standard " Cold finger " approach.

To provide maximum efficiency...the concept of entombing the entire assemble...in an enclosed insulated chamber is likely a good idea....as in it contains the lowered temp atmosphere that is so difficult to achieve. And... a slight heating of the output window to prevent condensation/frost build-up is likely a desirable feature.

Now....hmmm...running a DI/Ethnol mix....thru a CU heat sink block...may be another avenue ?? And have the LD mount mounted to the CU block via an Iridium foil interface. I am sure we can study the computer/overclocker group's adventures to apply to the super-cooling of just the LD + Mount. They have been experimenting with super cooling for quite some time !!...and they share their experiments !!!

Just some thoughts !! And...how we apply these things to a HH unit is yet another hurdle !! I am sure others will have a more educated insight !!

Beam Out

 

[Benm]

Here's my other question mark. It you submerge a module with say a 3element lens into a clear beaker of ethanol it's possible a gas/liquid mixture would build between the lens and diode window because light passes through each differently this would change the beam profile. I would think.

If you submerged the entire module it would behave differently than in air. It's just the huge difference in breaking index of ethanol and air that causes this. If you submerged it in toluene the lens would hardly have any effect at all since it's breaking index is very close to that of glass.

I would not attempt to submerge it counting on no ethanol getting between the diode and lens either. What you could do is submerge the entire assemby into ethanol and room temperature and adjust the optics at that point, hopfully you are able to.

There still is quite a difference between ethanols optical density at -80 versus room temp, but perhaps this can be adjusted for once the whole system is cold. This lens adjustment would have to be done at low temperature using tweezers to turn the lens, but that should be doable. One thing here would be to use a single element lens, not a 3 element, since it could take a long time for ethanol to get between lens elements.

 

[Trendkilla254]

I'm excited to see 638nm in that host.