"Super"cooling a laser?

[brtaman]

Hello everyone,

I am new to this forum (been reading for quite a while). I have a few burner diodes salvaged from dvd burners, a DX true 50 and a 803t blu ray diode on the way. I run the diodes through the DDL driver, that was graciously posted on here by deadal.

My questions, what would the effect of super-cooling a diode, not true super cooling...only in the -50-60C range...have on the laser diodes function?

I have a phase-change cooling system left over from my computer overclocking days (still hold the WR for a CPU ), and have been trying to find a good use for it. The system is run by a 1/2 hp Danfoss compressor, it uses r-402a gas (good temp and capacity). It can handle a 150w at around -47 C, so I believe the <1w load from the laser shouldnt be a problem . However on the other hand I am thinking whether or not the laser will even be able to lase at such a cold temperature (as they usually need a couple of seconds of heating up before they run at their full potential).

What are your thoughts on the subject? At the moment I have a ton of work to do, but I will be trying out the experiment as soon as I have time!

I attached an old picture of the system, to give you better idea of the type of cooling system I am referring to (basically a superpowered-concentrated fridge)


Thanks
brtaman

 

[GooeyGus]

Well, I think the colder the diode, the lower the wavelength will skew from the rated output. I really have no idea how much it would change. Maybe 5nm? Maybe more, I'm really just guessing here.

Also, we all know the death of all semiconductors come from heat. So in theory with a cooler such as this, if you can keep the diode cold you could pump a whole lot of power through it. But I also wonder about the stress of the die itself going from very cold to hot rapidly.

I'm very interested in how this turns out. How much would one be able to pick up a system like that for?

 

[brtaman]

Hey gooeygus,

Yeah I cant wait until I have some free time to perform the experiment.

Hmm, about the cost I am not too sure it really depends, (this one was custom built for my needs at the time so was more expensive). But there are many commercially available ones, that would be more than adequate to cool down such a small load.

(its been a while, but of the top of my head)

Check out:

asestek Vapochill cooling systems (they were pretty cheap a couple of years ago, should be even less expensive now)
Prometia systems (they were considered better but still commercial by the "underground" OC scene hehe)

Also I recomend you check out some overclocking forums, there may some systems for sale for pretty cheap:

xtremesystems.org (great site)
there are many others but it has been a couple years since I got out of overclocking (too many busted cpus, motherboard et. ..too expensive...now I found lasers :-X...another way to "burn" money)

I would suggest that you perhaps wait to see if the LD's will even function at there temperatures before investing in something like this system, cause beleive me there isnt exactly a market for them once you dont need them anymore )

Thanks
brtaman

 

[brtaman]

I have done some preliminary testing with the cooling system using a stock aizix line module (better to determine color change?) and I have come to two conclusions. The beam seems to be putting out a much higher amount of energy, for a 5mw laser the ammount of light it put out was quite shocking (though I have no meter to confirm any of this ...), but it could just be my eyes playing a trick on me

The next observation was that the wavelenght had changed a noticable bit, I could actually discern a difference with my eyes as it was shifting towards a more yellow/orange hue, though the system was not powered up completely during these tests...I have taken a couple of pics (sorry for the horrendous quality, but I was a bit excited , i have experience photographing lasers, and the camera is ancient from the 2mp times :I)

 

[brtaman]

Picture #2 (couldnt figure out how to attach more images to a single post)

these were all done at 0 exposure

 

[brtaman]

Last pic...(got more dont want to spam )

Next up is the 16x burner with ddl driver...

One thing to note the biggest problem was the condensation forming crystals over the lens which required me to wipe the lens with a q-tip every couple of minutes, diminishing out put quite a bit.

Thanks
brtaman

 

[flogged]

What sort of the diode are you using? This would be useless to do on a DPSS system - the pump diode needs to lase close to the pump frequency (usually close 808nm).

For 'direct injection' diodes, the cooler the diode the higher the frequency.

For red diodes (635-670nm) and 808nm (pump) you see a .15-.3nm (per degree C) decrease in frequency with decreasing temperature. So no more than 3nm for every 10(C) of cooling. Closer to 1-2nm per 10C with red diodes. You may be able to cool a 635nm diode to the point where it would lase under 630nm - orange. I did take one of my 635nm pointers out on a very cold night (-20 C), and after letting it cool the output looks more orange the usual.

Apparently violet (405nm) laser diodes are less effected my temperature - perhaps .04nm/C.

There's a limit to this, based on the physics of the LD. So you can't just keep cooling it and increasing the lasing frequency. Below a certain temperature the glues holding the LD together may also come apart

See SAM's laser FAQ for more information:
http://www.repairfaq.org/sam/laserdio.htm#diobcc6

 

[amkdeath]

Hello everyone,

I am new to this forum (been reading for quite a while). I have a few burner diodes salvaged from dvd burners, a DX true 50 and a 803t blu ray diode on the way. I run the diodes through the DDL driver, that was graciously posted on here by deadal.

My questions, what would the effect of super-cooling a diode, not true super cooling...only in the -50-60C range...have on the laser diodes function?

I have a phase-change cooling system left over from my computer overclocking days (still hold the WR for a CPU ), and have been trying to find a good use for it. The system is run by a 1/2 hp Danfoss compressor, it uses r-402a gas (good temp and capacity). It can handle a 150w at around -47 C, so I believe the <1w load from the laser shouldnt be a problem . However on the other hand I am thinking whether or not the laser will even be able to lase at such a cold temperature (as they usually need a couple of seconds of heating up before they run at their full potential).

What are your thoughts on the subject? At the moment I have a ton of work to do, but I will be trying out the experiment as soon as I have time!

I attached an old picture of the system, to give you better idea of the type of cooling system I am referring to (basically a superpowered-concentrated fridge)


Thanks
brtaman

YOU HOLD TEH WORLD RECORD?!?!?!?


OMG


Im a bit of an overclocker myself, but my biggest achievement has only been a measly 2.53---> 4.5 GHz overclock (CPU)... :-[ I take it the world record was like a 6 GHz Overclock or something? I always thought it was archived with liquid nitrogen...


anyhow I R TEH BIG FAN!!!!


</threadjack>

amk

 

[brtaman]

Thanks for the info flogged, I was sure that the diode had turned orange but could not confirm, it was the base 5mw within the aizix module (just with focused line), I am going to try with the burner diode, but am waiting to recieve/order glasses because 200ish mW pumped into a small room with white walls is not something I wish on my eyes.

With your calculations, and the ~-50C the cooling system pumps out I should get -10nm from the diode (going by the 2nm change per degree and from 0C, if ambient at 20C is meant, then in theory -14nm in the wavelenght? Cant wait to see what comes out of this.

BTW flogged where do you live, -20 outside? even at night brrrrr


Hehe amk, yes I hold a world record (a measly one though ) for a specific CPU core not the general OC WR, I got a 3000+ athlon 64 (754 pin), from 2.0-3.3, not that hard though, I was kind of waiting for someone to challange me before pushing it (then i killed the cpu...), but everyone was busy with the 939 pin version at the time, so it has stayed and will stay i guess , but its still cool to say that I have a WR in something hehehe

My OC name was ZIZI, here is a link to my lowly achievement. the one posted is only 3.2 IIRC (no one challanged nad i lost the link to the CPU-Z confirmation for the 3.34 one), these old athlons were pretty stuburn to push compared to Intels of then and now though, but great fun and had amazing memory latency.

http://www.xtremesystems.org/forums/showthread.php?t=59753 its the single core 3000+ Venice 754 (they werent too easy to overclock).

Good to see a fellow overclocker on these board (nice overclock though which core) (though its not too surprising, tech related things seem to interest all oc'ers ), though my OCing days are pretty much over as I cant find the time, and those tests and processes took up ALOT of my time.


Thanks
brtaman

 

[Jimmymcjimthejim]

Those are some interesting results...
I never knew that cooling a laser diode will increase the frequency of the light (in nanometers).

 

[flogged]

uh, actually cooling the diode will decrease the frequency of light in nanometers.

 

[Jimmymcjimthejim]

If cooling a 650nm diode makes it go to 640nm, then that would be increasing the frequency.
Less wavelength = Higher frequency
Correct me if I'm wrong.

 

[roSSco]

The wavelength is related to the frequency by the formula: wavelength = wave speed / frequency. Wavelength is therefore inversely proportional to frequency. Higher frequencies have shorter wavelengths. Lower frequencies have longer wavelengths, assuming the speed of the wave is the same.

 

[Hemlock_Mike]

R402 ?? I haven't heard of that in years !!! Used in climate test chambers cascaded with R12 or 20 -- WOW that's $$$ stuff.
Remember that cooling the die super cold still leaves only hair fine wires to carry the current.
You are on a new edge here and everyone is watching ;D
Carry on -- Mike

 

[brtaman]

Thanks for the comments everyone!

Yeah hemlock mike r402a is the bees knees

On a side not I just had an epiphany on how to use even more overclocking equipment that is gathering dust.

I will be cooling the diode from both sides:

On one side the phase change system

on the other side a 127w peltier (which will soon be going to igor) cooled by one of my old watercooling systems

I made quick schematic on how it will look like when finished it should lower the temp by around an extra 10 degree, too bad my old overclocking thermometer was sold..., I will really be interested in the temp, maybe I will have to get a new one... ;D

I cant wait to see what happens

Thanks
brtaman

 

[thejunkmonger]

ha time to break out my peltier that has been gathering dust