an amusing (but useless) trick

[Things]

I would say some of the 445nm energy is pumping something else beside the die, but overpowers it once you go above threshold. I originally thought it may have been fluorescence, but it seems to stay collimated.

[Stoney3K]

I would say some of the 445nm energy is pumping something else beside the die, but overpowers it once you go above threshold. I originally thought it may have been fluorescence, but it seems to stay collimated.

If you have an LPM, you might be able to figure out how much power is going into the green (532nm?) and violet (405nm?) lines. Filter the green out with a dichro, and hit the LPM with it.

Some pics with a bit of smoke might also be totally worth it, if it's even half a mW of direct green!

I don't know what material the 445nm diode is exactly, but the violet and green may just correspond to different bandgaps in the P-N junction of the diode. The 445nm is most sensitive to stimulated emission and is dominant in lasing (just like the 488nm in Ar ion tubes).

[Things]

I'd say thats the case, infact it's probably why green diodes are suddenly taking off after these 445nm diodes. They realized how close they were to getting green, so they are just refining the process or something.

I have a thermal LPM, so I doubt it'd register such low powers, though if anyone has an optical LPM, I'd love to see what it's doing

[Stoney3K]

I'd say thats the case, infact it's probably why green diodes are suddenly taking off after these 445nm diodes. They realized how close they were to getting green, so they are just refining the process or something.

Well, that would probably mean the end of DPSS, which has its drawbacks as well.

If they manage to squeeze 100-150mW of diode green out, the triad is pretty much complete and RGB would reduce to three diodes. Some manufacturer might even be crazy enough to put them all together on one die, although alignment of that will be a pain.

The higher end market can probably transition to OPSL. If only we could get some decent beam specced red...

[Meatball]

An RGB diode would be impossible with one die, such a setup would consist of 3 separate dies, which would lase separately. GaN can only be push so far. Collimating such a diode would be a nightmare.

Its possible that variations in the length of the die would allow other nodes to exist simultaneous inside the medium, but they would be fighting each other for energy.

GaN (Gallium Nitride) diodes such as this have only been pushed up to ~524nm. That's the absolute longest wavelength I seen achieved as reported my the media. Some shorter wavelength greens at ~515nm have been manufactured with the capabilities of a stable 500mw. But that's still pushing the capabilities of the material. Those diodes are currently available for sample testing I believe, since they are "green enough" for some projector applications. 532nm is only a wavelength achieved through the DPSS units we see all the time.

[-bart-]

515nm will even give you a greater chromatic reach than 532 would.

[catalanjo]

GaN (Gallium Nitride) Some shorter wavelength greens at ~515nm have been manufactured with the capabilities of a stable 500mw.

So who is going to start the GB then ?

I want one pls

Cheers

[Stoney3K]

An RGB diode would be impossible with one die, such a setup would consist of 3 separate dies, which would lase separately. GaN can only be push so far. Collimating such a diode would be a nightmare.

Would such an RGB diode (on multiple dies) have better beam alignment when collimated than the separate-laser setup we've been using so far?

I can imagine, if you manage to align three dies up to the micrometer and then collimate them, the beams will be slightly misaligned due to two colours always being off-axis. It's possible the alignment is still a lot better than the mm-accurate alignment we're using today. Especially angular matching of the beams and keeping them parallel are things you can be less concerned about.

A single die would also be quite impractical, as it only produces white light in the best of cases and there's no way to control R/G/B current independently, knocking us back to Ar/Kr ion tech and PCAOMs. Three (power matched) dies in one package would be a more practical option.

Some shorter wavelength greens at ~515nm have been manufactured with the capabilities of a stable 500mw. But that's still pushing the capabilities of the material. Those diodes are currently available for sample testing I believe, since they are "green enough" for some projector applications.

I'd like to see one of those in action (and, am actually pretty curious to the wavelength of this diode's green) and I'm also getting a bit curious as to other wavelengths we could pull off with GaN. Anything in between 405 and 520nm? There are some really nice blue and cyan hues yet to be explored in there....

TBH, I'd rather have a 250mW 515nm diode that isn't prone to mode hopping or jellybeaning when modulated, and is a lot easier to mount/cool, than a cumbersome DPSS which needs to have its temperature stabilized to be of any effective use.

532nm is only a wavelength achieved through the DPSS units we see all the time.

Don't forget the good old YAG's.

[Meatball]

Would such an RGB diode (on multiple dies) have better beam alignment when collimated than the separate-laser setup we've been using so far?

No. Focusing each of the different wavelengths would be impossible without separating the beams, and then recombining them. Plus, the source of the white light could not be considered as a point source anymore.

I can imagine, if you manage to align three dies up to the micrometer and then collimate them, the beams will be slightly misaligned due to two colours always being off-axis. It's possible the alignment is still a lot better than the mm-accurate alignment we're using today. Especially angular matching of the beams and keeping them parallel are things you can be less concerned about.

What?

A single die would also be quite impractical, as it only produces white light in the best of cases and there's no way to control R/G/B current independently, knocking us back to Ar/Kr ion tech and PCAOMs. Three (power matched) dies in one package would be a more practical option.

The KES-440A diode from the PS3 gaming console was the first 405nm diode available to hobbyists. Inside the can was 3 dies, one for the IR, one for the red, and one for the BR. The diode simply had more pins, and was common cathode. The reason it works in a PS3 is because the PS3 sled, like any other sled contains active optics which can move accordingly.

I'd like to see one of those in action (and, am actually pretty curious to the wavelength of this diode's green) and I'm also getting a bit curious as to other wavelengths we could pull off with GaN. Anything in between 405 and 520nm? There are some really nice blue and cyan hues yet to be explored in there....

TBH, I'd rather have a 250mW 515nm diode that isn't prone to mode hopping or jellybeaning when modulated, and is a lot easier to mount/cool, than a cumbersome DPSS which needs to have its temperature stabilized to be of any effective use.

Its the same 515nm that Ar+ put out. Its a very strong line in the multi-line tubes. There's only one 515nm wavelength, right?

Don't forget the good old YAG's.

Point proven. YAGs are still PSS when you get technical about it. I forget that not every medium is pumped with an actual 'diode'.

[Stoney3K]

No. Focusing each of the different wavelengths would be impossible without separating the beams, and then recombining them. Plus, the source of the white light could not be considered as a point source anymore.

Suppose you have a can with three dies in it, for R, G and B. The easiest option is to put one of the colours in the center of the can, on the collimating lens axis, creating a perfect match (and thus, parallel beam) for that specific color. The other colours are slightly off axis, and therefore, are effectively (slightly) out of focus with the collimation lens. This means the two other colours would be more divergent when compared to the colour in the center.

Another option is to place the three dies in a 120 degree pattern equidistant from the center of the diode, so all three emitters are at the same distance to the collimating lens. This would probably bring the overall divergence down, but would create a similar effect to chromatic aberration found in photographic lenses. The colours would be more separate near the far end of the beam, and the beam divergence would improve if the colours space closer together.