The Tunablue - wavelength tunable blue laser

[Atomicrox]

[Warning: pic heavy!]

Here's my latest build. It is, by far, the coolest laser in my collection.

There have been a few similar builds before, mine isn't fancy but I hope you'll enjoy it!

I really like lasers in boxes as long as they are still portable. That's the case here, it can still be carried around and has the obvious advantages of boxes: stands on it's own with no chance of rolling, easy to build, has space for a modulating pot.

Specs:
Diode: NDB7675
Driver: BlackBuck 8M
Current: 0-2.8A controlled by an external pot while the laser is on
Lens: 3 element
Approximate wavelength: 462-476nm
Power:0-1.5W (it peaks around 2A and then folds back to 1.1W at 2.8A and keeps dropping, can't leave it on at that current very long)

This isn't a very efficient diode but it's higher wavelength than my other NDB7675 even at lower currents.

The tape is used to pull the batteries from the holder.

I measured wavelength using my ghetto spectrometer (a cheap spectroscope glued to a webcam). The second, weaker line is a 473nm DPSS for comparison. My spectrometer is obvisouly not precise, but at max current the laser is clearly above 473nm. Here are the readings at multiple current settings:

Note that the pictures below serve to show the wavelength variation, but the actual color looks nothing like the pictures. At low power the beam looks royal blue with maybe a very slight touch of violet, then it goes to a pure, cobalt blue. At max current it looks light bluish-green, not cyan. ZRaffleticket's picture here is actually closer to the beam color. The dot color remains blue throughout, but gets lighter as it goes.

Here's some comparison shots at multiple currents:

 

[Atomicrox]

Some pictures without the lens. The multiple modes change a lot while the current is altered:

Some pictures with a diffraction grating. These were meant to show the first order dot moving while wavelength changes but bloom and camera movement ruined the effect. The first one is before threshold, you can still see the LED emission.

 

[Atomicrox]

Last but not least.. a few beamshots! No use for laser threads without them.

This one is a comparison shot with an NDB7875. The difference is larger IRL.

 

[paul1598419]

I do agree that my 7675 is a much different color than my 7875 even at the minimum current necessary to get the 7675 to lase. If you are getting fold over at 2.8 amps, why push it that hard? I will measure mine with my spectrometer at 2.4 amps. I think it will be pretty close to as far as this diode will shift. It is an interesting experiment you've done with this diode and the Blackbuck driver. How did you set the calibration on your spectrometer? I have heard it is a two point system where you take the pixel number at one WL then the pixel number at another. If so, that is a linear relationship which is not accurate for any diffraction grating. Just trying to understand how that thing works. Thanks for the results you've come up with.

 

[BobMc]

Very nice read Atomicrox. Nice build with the 7675. A nice little "lab laser" build. Interesting spectrometer graph readings, like the color shift. Nice compairsion with the 7875. beams shots are always a plus. :gj:

 

[badboybilly]

Awesome buddy, these diodes are just awesome!

 

[Benm]

What is the mechanism behind the wavelength change?

Is it purely thermal shift due to the diode die heating up at higher current?

 

[paul1598419]

I'm sure that is a large part of it. As people tried to get more power out of these diodes, it was found that they do shift considerably at higher currents. Apparently, the diode folds over at 2.8 amps and people have driven them into fold over now to see how much of a shift they can get. Not too healthy for the diode, but they can be had on eBay for $28.00 now. I have two myself and a driver set at 2.4 amps for my build. Mine is going into a large copper heat sink, so heat may be less of a problem, but that remains to be seen.

 

[Atomicrox]

Thanks, guys!

I do agree that my 7675 is a much different color than my 7875 even at the minimum current necessary to get the 7675 to lase. If you are getting fold over at 2.8 amps, why push it that hard? I will measure mine with my spectrometer at 2.4 amps. I think it will be pretty close to as far as this diode will shift. It is an interesting experiment you've done with this diode and the Blackbuck driver. How did you set the calibration on your spectrometer? I have heard it is a two point system where you take the pixel number at one WL then the pixel number at another. If so, that is a linear relationship which is not accurate for any diffraction grating. Just trying to understand how that thing works. Thanks for the results you've come up with.

You're correct, it's two point calibrated and that's bad. I used the same software they were using with that <$100 ebay spectrometer that popped up a few months ago. I used DPSS 473nm and 532nm for calibration, so it should be at least in the ballpark.

Even if the spectro readings are off I'm positive it's above 473nm and the most visible change unfortunately comes after it starts folding back...
I'm pushing it that hard because I have another diode at a more reasonble 1.8A, because I got this diode dirty cheap in an auction and, most important, because the color is AMAZING

I look forward to seeing your spectrometer readings!

What is the mechanism behind the wavelength change?

Is it purely thermal shift due to the diode die heating up at higher current?

Looks like it's mostly current related, but thermal shift seems to have some effect, after it heats up it gets a few nm higher.

 

[lasersbee]

Nice experimental build...

Jerry

 

[Alien Laser]

Awesome gives me an idea to build something like it

 

[ZRaffleticket]

Actually guys, the 7675's really shift in wavelength when put all they way near the brink of death.

RHD had one that was binned at 473 @2.4A, and went to 478 @ 2.8A. For an extra 400mA, 5nm is a pretty freaking huge jump.

http://laserpointerforums.com/f40/blues-449-460-478-495-a-93399.html

 

[badboybilly]

Actually guys, the 7675's really shift in wavelength when put all they way near the brink of death.

RHD had one that was binned at 473 @2.4A, and went to 478 @ 2.8A. For an extra 400mA, 5nm is a pretty freaking huge jump.

http://laserpointerforums.com/f40/blues-449-460-478-495-a-93399.html

Exactly I'm glad your on board and agree Zach!

There are threads that are years old and peaple still don't believe it,
Buy the way have you seen my adjustable 7675 I just sold!

 

[ZRaffleticket]

Yes I have! Makes me wish the maglite I have has a knob instead of a regular old button. Having something tunable has to be pretty sweet.

 

[badboybilly]

Yes I have! Makes me wish the maglite I have has a knob instead of a regular old button. Having something tunable has to be pretty sweet.

Yep it's sweeter than sweet, I miss it terribly, I'm sure Andrew is going to post some delicious pics soon maybe a better one for my avatar!
The best thing is you can change the power out for different situations, there is no need to have the pot at full 2.7a all the time, I would only put it full to see that buitifull 475 colour now and then, that is why I don't understand people saying 2.7-2.8 is too high, I think my diode-build was a very efficient one considering it hits 2.3w at 2.7a and doesn't start reducing it would probably go higher if I gave it 3a , it's also heatsinked like a sandwich very well,

Do you want to buy some maglight housings new in few different colours!

 

[Benm]

Actually guys, the 7675's really shift in wavelength when put all they way near the brink of death.

RHD had one that was binned at 473 @2.4A, and went to 478 @ 2.8A. For an extra 400mA, 5nm is a pretty freaking huge jump.

http://laserpointerforums.com/f40/blues-449-460-478-495-a-93399.html

It's a huge wavelength jump for a relatively low (percentage) change in current.

I wonder that the mechanism really is though.

If it is purely thermal you could just heat up the laser diode (or supply insufficient heatsinking) and get it that hot at a much lower current (say 1 amp).

Typical wavelength variation for temperature is something in the order of 0.2 nm/K, so a 5 nm shift would equate to a change in temperature of 25 degrees C (or K), which seems unlikely when going from 2.4 to 2.8 amps in the same heatsink.

Another option is that you are actually damaging the laser diode causing the wavelength shift, which would become permanent if it didn't die first. This is somewhat similar to how a LED can have it's output shifted from an overcurrent event and remain that way even when operation at normal current is resumed.