Building a orange (613nm) Laser from scratch with off-the-shelf parts (and for cheap!)

[icecruncher]

I am hoping to archive that. I did every „good practice“. The casing should hold the vacuum just fine. I did go to grate efforts to make every seal a metal/metal or metal/Glas seal. Everything should be very tight.

The diode is still in it’s can. There is only one Bar pressure difference, so no problems with the can window.
I will do experiments on longevity (keeping vacuum) and best diode. If I do have these Informations, I am thinking about a groupbuy for the 120k Tec. Price should be around 100€. Power usage would be around 80w. Maybe I could even offer CNC milled casings.
nice thing is: we could make this fully tunable from 613 to 640nm (heating and cooling).

I will keep all of you updated.

This is very cool

FIRST, I'm not knocking this. It is very well done and it screams awesome. Only thing I would be concerned with is total cost. By the time you buy the diode and the TEC and milled casings and pump and other bits, it could push the price toward where another option might be more cost effective or consistent.

I say this, because I've caught myself doing things like this and ending up spending more than if I went ahead with the original in the first place (even though I learned a lot)

Still interested though!

 

[farbe2]

I understand your point. But this project is special. First: Its not as pricy, I am still at around 150€ price point for everything. The Tec: currently 40€, better one 100€
Material for the mill: 40€
Laserdiode with colimator: 30€, better one 130€
Indium foil: 10€
So the deeper module would be around 300€, the. Current one is around 120€.
I don’t count the Time invested in engineering and milling the parts, that’s essentially free for me. I think worst case would be 1000€ for every part milled by a company. This would make the price to 1300€ for the expensive variant. I don’t see where I could get a laser that’s tuneable between 640 and 613nm with 100mW for this money.
Nearest would be 607nm which costs around 2500€.

 

[LSRFAQ]

Two stage Peltier ordered. Milling machine I have. Vacuum sealing tools I have. Current Feedthroughs I have. Trick will be to keep the heat conducting pastes off the optics, and indium, I have as well. Have to see if my Cryocoolers work.

Steve

 

[icecruncher]

I understand your point. But this project is special. First: Its not as pricy, I am still at around 150€ price point for everything. The Tec: currently 40€, better one 100€
Material for the mill: 40€
Laserdiode with colimator: 30€, better one 130€
Indium foil: 10€
So the deeper module would be around 300€, the. Current one is around 120€.
I don’t count the Time invested in engineering and milling the parts, that’s essentially free for me. I think worst case would be 1000€ for every part milled by a company. This would make the price to 1300€ for the expensive variant. I don’t see where I could get a laser that’s tuneable between 640 and 613nm with 100mW for this money.
Nearest would be 607nm which costs around 2500€.

You are absolutely right and If price isn't too high I'm interested also. You can't get it new - especially a tunable one. But, I bought my last couple of yellow HeNe lasers for about $100 each with driver and my orange was about $500, only they are less then 5mw.

I think its a great project, it is special and I love it. Just don't go selling your house to finish it - end up homeless in a cardboard box on the corner with an awesome tunable laser - looking for a place to plug it in.

I just know that personally, I've dumped more money than I want to admit in certain projects. I then realized, I could have saved by buying the item completed or a new version (upgrades on computers for example). You might kick yourself if a $200 300mw 612 diode comes out! (I can dream)

 

[kecked]

Very cool. Hahaha. Pun intended.

is there any risk to the diode running that cold? I have no idea but wonder if the thermal gradient over time might crack it. Diodes cheap so who cares. Just curious to see how it handles it over time. Also does it cleanly modulate at the cold temp?

Did I read you have a vacuum? Could you not use dry nitrogen or argon at room pressure? Same seals and all but less stress on window. Window might bow a bit and mess with beam at vacuum.

whatever. Fantastic job. Better than this arm chair quarterback could do.

 

[farbe2]

I think just a argon fill will still be to good of a thermal conductor.
Regarding the bowing window: I dont think this will do anything. The free area is quite small (5mm bore with a 10mm window) and the window is still thick with 2mm. One could also use a thicker window to have even less bow.
I am more afraid of the colimation. The thermal expansion could shift everything out of place, i need to do more research.

I also dont know how the diode will handle the temperature swing. But i dont think this will be a problem. Blue diode dies run at up to 130°C (osram) without a problem. So every power cycle gets you 100k temp swing.
I also do 100k swing but in the other direction, so i dont think thermal expansion will crack anything.


Update: i got the better vacuum pump and could verify that its already better.

Next step:
- Improving the setup
- measuring promising diodes for temperatur shift
- buying better tec (ouch)
- buying 633nm diode (ouch)

Are some people interested in buying Tecs and/or diodes? We might get a better price.
Two stage chinese low quality tecs dont cut it here. We need the best we can get. 5 Stages with high temperature differential and efficiency.
I do have a quote from a good supplier that has two suitable modules:
4-stage with 70W and 5-stages with 52W
These modules are both capable of -101°C without heat load. So i expect to get down to -80, maybe even a little lower.
The 5-Stage module is obviously better because of the lower input power for the same pump power. However its more expensive.
Quote:
160$/pc and 960$/10pcs for the 4stage one. So if we have 7pc ordered we can already get 138$/pc.
190$/pc and 1400$/10pcs for the 5stages one. So 8pc ordered would give us 175$/pc.

I would already take two. Maybe we can reach 7/8 or more.
I would prefer the 5-stage one as its suitable for a pointer. 52W is almost easy to dissipate.
Also the 5 stage one would be easy to improve further, just add a hefty tec to the bottom an go to -120°C. Just the power will be astronomical. I calculated almost 300W for the 5+1 stack.

Is anybody onboard?

For the diode: does anybody have a 633nm diode or knows a cheap (single pcs) source?
I would like to confirm its nm/K drift and measure it at low temperatures.

 

[farbe2]

Hello Guys, Progress!

I will just leave this here.

 

[farbe2]

I hope this does not count as spam.

A quick update. I got it down another 1,3nm!
Just mind that this is even the running wavelength at 60mW output, not a "let it cool down and then turn it on" wavelength.
Thats with a 700mW 638nm diode. So even way more power available (at a higher wavelength of curse).


I have the money shot for all of you waiting. 616nm compared to 660nm. Thats no joke orange, no way anybody would mistake that for red. Picture looks quite accurate on my display.


Please don't mind this janky setup. A lot of stuff going on that needs to be tested.
Actually I do have another update. The seal is quit good. Should hold vacuum for a long time. At least the case lid seal. For the pump down tube, I am still investigating...

 

[CurtisOliver]

I hope this does not count as spam.

A quick update. I got it down another 1,3nm!
Just mind that this is even the running wavelength at 60mW output, not a "let it cool down and then turn it on" wavelength.
Thats with a 700mW 638nm diode. So even way more power available (at a higher wavelength of curse).

View attachment 73241
I have the money shot for all of you waiting. 616nm compared to 660nm. Thats no joke orange, no way anybody would mistake that for red. Picture looks quite accurate on my display.
View attachment 73242View attachment 73243

Please don't mind this janky setup. A lot of stuff going on that needs to be tested.
Actually I do have another update. The seal is quit good. Should hold vacuum for a long time. At least the case lid seal. For the pump down tube, I am still investigating...

Very good job. Looks great. You got down pretty low in the end.

 

[RA_pierce]

Great work.
I have 3 633 nm 100 mW diodes. One is new (I'm keeping this one) and two are used surplus. I have tested only one of these, so I can't be sure about the functionality of the other.
The one that I did test works just fine. I verified the wavelength with a home-made spectrometer (calibrated with 3 DPSS wavelengths).
The price I paid for these used diodes was high, but substantially less than what I could get for a new one at the time. Given the high cost and uncertainty, I'm hesitant to offer a used diode but you can PM me if you're interested.

 

[kecked]

Nice! Now I both yellow And orange fever.

 

[Gabe]

Very cool project! Wish I could have more to add than that, you seem to have figured out all the tricks with stacked peltiers that I know of. I’ll stay tuned!

 

[CurtisOliver]

I’ll stay tuned!

Nice to see you Gabe. Pun intended there?

 

[Gabe]

Nice to see you Gabe. Pun intended there?

Hi there, yes! absolutely very much intended! i am indeed quite clever, yes very much so

 

[farbe2]

Hello guys,

I am currently running tightness test for the module.
I will report in 3 weeks (I am in vacation) if the module is vacuum tight. If so I will start ordering the new tecs.

best regards

 

[farbe2]

Update:

My tests are very promising. Seems like everything works out. I am currently redesign the case to only include hermetic metal to metal seals wich should be air tight to such a high degree that the module will work for at least a decade.

I am in the process of ordering tecs.
One member did already commit to a fully build module.

Does somebody want a module/Tec? I can't give a accurate estimate for price.
I dont have enough informations for my assembly time involved (evacuating and making the seals).
I would also make a groupbuy for these TECs, one should be around 160USD excluding shipping.
If you want a high performance TEC or would want me to build a module for you, let me know.

For Module Performance i expect (Wavelength on highest tec power):
150mW @ 617nm with the cheap 185mW sharp diode (30USD)
90mW @ 612nm with expensive 633nm sharp diode (190USD)
500-700mW @ 620nm with 700mW sharp diode (30USD)

The power is an estimate as i dont know how hart i should drive the diodes if cooled.
The wavelength would be tunable up to the nominal wavelength of the diode, maybe even higher as we can also heat.
The wavelength will also drop if we use lower power.

I would use the 633nm Single mode diode as standard because the wavelength is more important (to me). But this can be changed to anything.

The temperature gets measured with a expensive platinum RTD sensor, driver (Tec+LD) will be supplied (of the shelf one).
So without case (and work), we are already looking at around 400USD (with the expensive diode).
We still need to add the collimating optics, driver, output window, temp sensor and cnc milled casing (5 parts + one 3D print).
The case will be the most expensive parts and i dont have informations on it either (i need to make the redesign first).
So not really cheap but also not expensive for this special wavelength.