Review: 510nm Direct Green Diodes / Build Photos (DGH-N1, DGH-N2)

I'm going to jump right in here and tackle some brief initial comments, and Aryntha will jump in for the second and more substantial portion of this review. About a month ago I was able to start testing some direct green diodes. I spent some time testing the diodes, testing drivers, and experimenting with M9x0.5 lenses. Eventually I transferred the diodes into portables, and one of these then headed out to Aryntha.



First, some important notes:

• We're going to talk about as much as we think is appropriate to talk about. If there's something we don't address, it's probably for a reason, and it's probably for a reason that isn't selfish or silly. One thing I'm not going to talk about, is where these diodes came from. It's not necessarily a secret, but it's also not my place to say. I'm more than happy to acknowledge the source of these diodes if that source so desires, but otherwise, I'm staying mum on the subject. That said, I think a lot of people on the forum already seem to know where these diodes originated from. Our reason for not bringing it up here directly, is out of courtesy - to avoid an inbox flood that may be unwanted.
  
  
• What I will say about the origin of these diodes (to save everyone a lot of wasted time) is that they're not harvests. These were legitimate, and moderately pricey, purchases of bare diodes, and, as far as I know, they aren't available again at this very moment anyway. Though I'll mention that their cost was not nearly in the range of what sample diodes like the exotic 488 and 473s cost. Think of this sort of like those 445s that started popping up a year or so before the KasEO diodes. So again, these aren't harvests from a retail consumer product, and probably won't be found in any such product any time soon. 510nm is not the ideal colour for projection purposes, so this is kind of like a "mistaken hop" on the pathway to the more desirable ~520 range.
  
  
• The other note, is that this really *shouldn't* cause any silly drama like we saw with 445s. These are still costly diodes, and they're relatively low power (much lower than even very inexpensive DPSS 532 modules). So, this isn't an earth-shattering moment in the community. I sort of think of it as somewhat on par with getting a new DPSS wavelength from CNI

After testing, I ultimately decided to give the diodes a more permanent home in a portable unit. Out of this came the DGH-N1 and the DGH-N2. Not real product names, just descriptors used to identify the first and the second (to my knowledge) direct-green diode portables. One (the 1st) is with me, the second is now with Aryntha. They're essentially the same, save for the fact that the unit Aryntha now has was built using a direct-press housings from Lazeerer (btw, big thanks for your quick turnaround, I know you lost a few hours of sleep on them!).

The two portables are also engraved, with their rough build dates, and a small claim to fame as the "World's 1st" and "World's 2nd" Direct Green Handhelds. To my knowledge, this is true, but of course it very well may not be - I'm sure there are some lab scientists chuckling at the title somewhere



My diode review / build notes:

• Being 3.8mm in form factor, I initially turned to Pontiac's fantastic copper-modules and had one pressed into one his 12mm-compatible housings. Pontiac's modules are fantastic. I've used them for single-mode blues as well, and they very well may be at the top of my list of diode housing solutions.
  
  
• The diodes require an incredibly high voltage supply, nearing 8V at spec-sheet current levels. That's no easy task. I elected to go with 3x cell hosts and a buck driver.
  
  
• Regarding spec-sheet current levels, these diodes don't require much. Both builds are running at 191mA to output roughly ~45mW. However, to put that into perspective, at a Vf of 8, we're looking at about 1500mW of input for 45mW of output, which is a diode efficiency of slightly under 3%.
  
  
• In context, my single-mode blue has a diode efficiency of somewhere around 15%, and my single-mode red (the HL63133DG), before I sold it, would hit almost 20%. Or if we're comparing apples to apple (colour wise), a 532nm FireDragon's module gets somewhere on average of around 7% IR diode input to 532nm output. In other words, these aren't high efficiency diodes just yet.
  
  
• I love the colour. I'll let Aryntha address that in more detail, but for my part, I'll just say that it's a beautiful "forest green". In fact, at 510 (or whatever these end up being), they actually seem like a more "authentic green" than 532nm DPSS. Of course, this is at odds with the reality that 510nm is not "ideal" in terms of colour mixing, and isn't where they want these diodes to ultimately land colour-wise.

To my knowledge the engraving is accurate, but who knows

Re: placeholder

And so begins my portion of the review, of the 510nm (50mW) Direct Green Diode and the World's 2nd Direct Green Handheld, built by the one and only RHD.

Overview:

So, in a way, this is kind of the moment we've all been waiting for. At this point, we finally really do have direct red (of course), rather recently (only 5-6 years ago), 405nm-violet diodes thanks to blu-ray, and then of course direct blue (not too long ago DPSS was the only way we got blue - and 445 Diodes have been driving down the price and demand of DPSS 473 Blue) and now direct green. This means - at least a bit down the road- a whole lot in terms of efficiency, space, and reliability for picoprojectors, as well as (potentially) laser show scanners, as well, once output increases. Things have come a long way since I got my first 670nm diode laser in 1992, and most of these developments in diodes have happened in the last 5 of 6 years.

The first thing everyone's going to ask is, at 510nm, "how green is it". The answer is: Very much green, and NOT CYAN. The emission is about 22nm below 532nm green, and 22nm above 488nm cyan. So it really sits equidistant between those two frequencies, and I think what has to be understood is, 488nm is starting to get into the realm of "subjectively blue-green", whereas even 473nm has been described as "turquoise" and not true blue. So, really, 510nm is definitely a true green, making 532nm look closer to a 543nm HeNe (because it is closer!) in the "yellowish-green" category, when compared side by side. (An odd corollary, but it makes sense: Viewing 532nm after viewing 561nm makes 532nm look positively blue-green!)

Subjectively, it looks very close to 515nm ArIon. I did a visual comparison and it was very hard to tell the two apart; indeed the 515nm ArIon looking ever so slightly greener, but, not in any way that any camera could represent this difference.

There's an interesting thing to be said here about relative color differences. Moving on momentarily to another laser I've been working with, my REO Tunable HeNe, the phenomenon of relative color perception displays itself very evidently. On that laser, when going from "633->612->604->594->543", 594nm looks absolutely yellow. Then, when walking back from "543->594->604->612->633", 594nm looks unmistakeably orange. (And 612nm often looks red - I often think I got to 633 but haven't yet.) So, side by side is going to show a marked difference in wavelength, often, but on its own, 510nm looks "more like 532 than you'd think."


ON TO THE REVIEW AND PICS


The Host



I don't want to replicate anything that RHD already did above, so, as you can see, my host/build is almost exactly the same as his (Sans the mirror polishing.) It uses 3x3.7v 10440s, for the 11-12v needed for the driver to supply the ~8Vf to the diode.



The business end of the host. RHD provided a great custom focusing ring and a glass lens. Note, the only 'detail' RHD and I had with this host, is that "flat top" 10440s don't quite work right. They don't make full contact, as it's designed to be a 3xAAA flashlight host. When I first got it, I got quite nervous as it wasn't turning on... A few solder beads on top of the 10440s brought it to life. Because of the diode's voltage requirements, I figured I'd note this as 3xAAA hosts may be common for these diodes in the future.


The Beam



This is still 50mW of green, so it's as visible as you'd expect 50mW of green to be. Looking "down-beam" in an ambiently lit room, the beam is most certainly visible with room dust.




Sorry for the condition of the outdoor shot -- it's COLD around here lately. I usually would have shown the laser hitting the mountain behind these trees (Which it does -- easily -- probably due to its insanely low divergence, which I'll detail further on down in this review.) It's interesting to note, that possibly due to the night-adapted vs day-adapted visibility centers noted in some of my earlier reviews, this beam may actually be MORE visible than an equivalent 532nm beam at night. In fact, (as I'll note below) - it may be the most "visible at night" laser we have so far. To me, at least, it certainly seemed to be "apparently brighter" at night in the sky than 532nm.

Beam Profile



This is a real single-mode green diode laser, so its output when unfocused looks like what you'd expect to see from a single-mode diode; an oblong ('diode stripe') sort of shape, similar to diodes we've seen before. No secondary stripes or modes like the C'io 445s. The output and falloff is very clean, nothing unexpected from a diode here - except the fact that it's green. Seeing this, even after a dozen times or so, is still a cool sight to behold.



The beam spot, focused down, at 40 feet distance. I made this image grayscale to show the shape of the dot better. The lines you're seeing appear to come from the optics, not the diode (as they do move with focus) - and at this point, at 40ft, the "fast axis" (if it can be called that) is still under 1CM. No really, it's under 1cm at 40ft. I still can't believe how non-divergent this thing is. To the eye, the beam appears to be an "oblong gaussian TEMoo" - not really a circle, but sort of an oval with even falloff. It was hard to get a good image of it, but it really is nothing like the 445s we're used to, and is a lot closer to (though less divergent than!) red diodes we've seen.

Color Comparisons



Here we see a comparison with 532, at about 20 feet. Notice how the Direct Green is certainly "more into the cyan range" than the 532nm DPSS, but it is still unmistakeably green. It was a lot closer to 532nm than RHD and I expected, but still is a cool new color to have in a handheld! Note, as usual with dot photographs, the 'wings' and 'lines' visible here are a LOT less visible to the human eye than they are to the camera -- meaning the dot doesn't really look this messy to the eye.



Viewing "down-beam", you can see the "yellow shift" of the 532nm DPSS and the "cyan shift" of the 510nm DG. This is pretty representative as to how it looks in actuality.



Another color comparison, this time with 561nm DPSS, 532nm DPSS, 510nm DG, and 473nm DPSS. It's really interesting how absolutely blue 473nm looks in comparison to 510nm DG.



Another color comparison between 561, 532, 510 and 473.



And for good measure, one more. It's also important to note that 510nm does seem to at times appear a bit more "intense' than 532 of the same output.. The outputs here were as equivalent as I could get them: 561nm@40mW, 532nm@50mW, 510nm@54mW, and 473nm@50mw. I firmly believe that this is because 510nm is right at the center of 'scotopic', or night-adjusted vision rather than photopic or day-adjusted vision (555nm).

To bring up a graph from a past review:



As you can see, with 510nm we have a laser that's just a few nm away from dark adjusted or "scotopic" vision. So I think with this new wavelength, we've got something new in terms of nighttime beam visibility. Even at 54mW, it really is an amazing thing to see, even on a clear night.

Measurements - Power



The diode is rated at 50mW when driven at 200mA. I'm not sure if I got a more efficient one or not, but our builds are "conservatively" driven at 191mA, as we don't want to push such a rare diode. Even so, at 191mA, I'm getting a rock solid 54mW out. There is no change. There is no warmup or die-off or variation. 54mW, all the time. This is a VERY stable diode.



And the Radiant+TEC/thermopile concurs exactly with the Kenometer+Ophir.

Measurements - Divergence

What I have to say about this is, the divergence on this diode is INSANELY low. I may have to double check my measurements but they really are borne out visually. At 28ft away, the dot really was not much larger than it was at 7 ft away. At 40ft, the 'oblong pattern' started to show itself, but it was still very tight and just unbelievably coherent. These are the measurements I came up with:







Conclusion

It would appear that at this point, "direct diode R,G,B,V" is a reality. We're not sure how availability of these diodes in a hobbyist-usable package (3.8mm) will play out, but if I'd hazard a guess I'd say (and hope) that the availability, so long as something else doesnt happen, may follow that of the recent 450nm Single Mode 3.8mm diodes. My rationale for thinking this is that ultimately, they have the same marketing/distribution/demand/purpose profile as these Direct Green 3.8mm diodes would. (Picoprojectors, etc.)

I know a lot of folks even on this forum won't be very interested in them, as a large contingency of folks are only worried about how much something will burn, and whether it can be pushed to 2 watts or not. So to those folks, this may look like "Oh, another weak green, who cares." However, in reality, and to those who do follow the actual technology behind this stuff, the implication of direct green availability (at least once it hits mainstream) is huge. Smaller and more efficient scanner setups. Longer duty cycles, lower power requirements, less heat. And the fact that these diodes are starting out with great beam characteristics makes them all that much more appealing.

Not to mention the likelihood that we'll see completely direct-driven picoprojectors that are even smaller than the ShowWX and the like, that we've seen in the past, relatively soon. Like 405 and 445, we've hit another milestone here with Direct Green - it's no longer "three-to-five-years-away", it's happening right now. There's a bit of work to be done until it gets well into the 520+ range where it becomes the best for color mixing, but this is certainly an interesting step along the way.

So um can you tell us where you got the diodes from?

Sweet. I may need to get me one of these.

Great job guys.:beer:

Real nice work there guys. Nice colo(u)r! Do you have a power curve yet?
That might tell some more about the diode but we can't expect too much yet from a 3.8 mm.
HMike

Sweet lasers, guys. Great job on the review.

aryntha - I bet you had fun with that RGBE camera. That is the ultimate beamshot machine.

Love that color.
So are they really expensive?

Amazing work! I'm beyond excited for what lies ahead :drool:

Wow. That's an amazing sight to behold. :angel:

Green diodes are actually here... I can't wait to get my hands on one.

Awsome green diode,love the host you two made :beer:very soon they will have achieved 1 watt or better.
We might get lucky next year and see bigger power improvements:eg:We have to:takeit: learn to crawl before we can walk.

i donnt this this is near to the same as "a new dpss from CNI" this a diode for lasers sake!! this is huge to me i find this totally incredible a new DIODE WL. thats awesome. thats huge!! who cares about "MORE POWER" new WL are awesome. :drool:

heh wow that was a really poorly put together paragraph... anyways seriously this is really intriguing to me. i really like the color. i dont know allot about the science of laser diodes...but that 3% efficiency is terrible right? 532 dpss modules can get up to 30% right? logic in my mind says they wouldnt end up using these diodes in a projector as they are to inefficient to be practical. would that be a correct statement?

anyways :drool:

Holy hell at the divergence.....? how is that even possible for a dioide? I'm guessing it's a side-effect of the unique nature of the green diode. With a 20x beam expander that thing can pretty much go forever.

joeyss said:

Holy hell at the divergence.....? how is that even possible for a dioide? I'm guessing it's a side-effect of the unique nature of the green diode. With a 20x beam expander that thing can pretty much go forever.

Click to expand...

I *did* measure twice on the divergence. It blew me away, but it's so low that I want to measure *again*, and will do that when I can find more than 40ft of space to reliably calculate from.

But, yeah, really, even at 40ft, it was "like that". I had a "this can't be right" moment and did double check.

great job =+rep to Matt when system allows.
+rep tp rhd now--

Awesome! Awesome! Awesome!

I knew we were getting close to these diodes but I had no clue they were actually in-hand!

Oh man I can't wait to get one :eg:

This made my day. Thanks for the great pictures guys! :beer: