The Tunablue - wavelength tunable blue laser

[Atomicrox]

It's not permanent, the wavelength goes back to ~462nm as soon as current drops.

Either way I'll report back here if it dies or looses too much power.

 

[badboybilly]

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.

Honestly was expecting u to jump in her again regarding this diode and spoil everyone's excitement again to do what a diode is supposed to do for a mere $50 or less, Not $100 like a year ago when you said this is not possible!

 

[Alien Laser]

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

well yes its shifting in wavelength the diode is about to see its own light threw the tunnel hehe :crackup:-:na:

 

[Alaskan]

Thank you for sharing the project with us, nice. Wish I had the equipment to measure something like that! One day!

 

[Atomicrox]

Thanks, but actually the equipment is just a common spectroscope like Portable Spectroscope Gemstone Identification Small Diffraction Gem Gemological hot glued to a webcam. All you need is patience to glue them well aligned and a few DPSS lasers for calibration. As Paul pointed out, it's not very accurate, but it's surely better than eyeballing and can easily compare close wavelengths.

 

[RedCowboy]

Electrons are driven into a higher orbit then collapse back down and in so emit that energy as a photon of light, the wavelength of light shooting out of the P/N junction depends on the substrate, so when overdriving, if it's not damaging the substrate it is stressing it, but simply energizing the laser diode stress it, will it wear out faster by overdriving it?

It is likely, by how much we don't know, and if the cost is acceptable then enjoy it, but it's likely to accelerate degradation by some amount, it could be very little or a substantial amount, we don't know but I know overdriving LED diodes in flashlights doesn't take all that much out of the lifespan, as long as you don't go too far, I have heard of some green shift when they are pushed really hard, but I digress.

If the output falls at the tail end of the color shift then that's about the limit, if output does not fall and only color shift is observed then that's not so bad, most these diodes at least the more recent ones will dim all the way down to almost nothing when pushed to 7 amps but recover as soon as the current is reduced, that's probably not good for them, but again if the output is not falling hard as the current is increased then the stress may not be all that severe.

This is my guess based on playing with a lot of decommissioned diodes, a theory that would show up in the raw output is that the dominant center wavelength is full and the adjacent bars are getting more of the load, if you look at how the raw output fills in on a wall as current is increased you will see it starts in the center and spreads out, so maybe that's a part of it? Probably not as the adjacent bars are higher frequency/darker blue.

 

[Atomicrox]

I don't think the bars are different wavelengths, at least not visibly so. That's an artifact of the camera on those pictures.

Edit: OTOH the power starts mostly on the middle bars and spreads from there.

 

[Benm]

My interest is not in what the diode costs, just in what mechanism could result in such a big shift in wavelength apart from thermal changes that are not big enough to explain this observation at all.

One thing i wonder about is if actually a near-death phenomenon for these diodes or not. How long can it survive running at 2.8 amps with good cooling - minutes, hours, days or even months?

 

[paul1598419]

One thing I did notice about this diode is it outputs a broad band of light over about 20nm before it reaches the current needed to get it to lase. It almost looks like a fluorescent ion at that point. As you increase current it then starts to lase and the band turns into a peak. I've seen other laser diodes do this, but don't know if it has anything to do with the WL shift.

 

[Atomicrox]

That is true, I think all diodes do it but it's more noticeable in some due to the range of colors covered by the LED emission (i.e. 20nm on the green-blue edge covers a lot of colors, while 20nm on red covers just red). Also that emission isn't coherent, there are no speckles.

 

[paul1598419]

No, many diodes I have measured don't start out with a broad band of wavelengths. The 473nm ones didn't and some of the 505nm didn't, but one did. Some start out lasing right away. It can't be predicted by modes either as I have seen it with single modes and multi modes. When I see it, I always think it kind of strange. It's always as the current is just enough to see light. But, many diodes just lase as soon as you get any light out of them at all. The 520nm and 510nm I have seen don't do it.

 

[Atomicrox]

Will try it with my 480nm and 505nm when I get around to it.

 

[Benm]

One thing I did notice about this diode is it outputs a broad band of light over about 20nm before it reaches the current needed to get it to lase. It almost looks like a fluorescent ion at that point. As you increase current it then starts to lase and the band turns into a peak. I've seen other laser diodes do this, but don't know if it has anything to do with the WL shift.

This broad output spectrum when driven below lasing threshold is common.

Red diodes certainly do it, but then again it all looks like red light even if it is all the way from 700 to 630 nm. A small change in wavelength is much more visible in the blue-green spectrum though, so you may notice it more there.

This doesn't explain how these lasers change wavelength when running -far- above threshold though. Thermal effects also do not seem that plausible, so it's quite mysterious.

 

[paul1598419]

I don't look at the laser's colors. I measure them with my spectrometer. It beats the hell out of subjective impressions as the spectrum is there to see. I don't know why some laser diodes lase only and others seem to start out with a band of wavelengths before they lase. I can only tell you that some do and others don't. And none of it is subjective.

 

[Benm]

If you look at it with a spectrometer you can indeed be sure.

I reckon most people don't have one though. Many laser diodes will produce a wider spectrum below treshold current compared to when actually lasing. How wide the spectrum is and how fast it closes when you get near lasing depends on the diode.

 

[paul1598419]

Well, since you have made the statement that ALL laser diodes do this, I'll be watching very carefully to see if I can indeed find a definitive answer to this question. I know I have seen diodes that, at the moment any light is available, lase and don't have a broad spectrum of light. I will be extra vigilant from now on.