I’ve decided I will go ahead and try to make the 3x3 array and put it in an emisar D1S. I already bought some supplies to do it and I’ve been wanting to do something with a D1S. It should pull around 24A from a high drain 18650 and do about 5500 lumens OTF and 83Kcd.
The performance might increase a bit when in array form, which it seems is the intended purpose of these LEDs. The amount of blue light escaping out the sides of a single die is significant, and I figure most of this light will become recycled and used to pump phosphor in neighboring dies in the 3x3 array. This would increase the output and luminance.
This is very odd. According to Cree’s PCT tool, this emitter is performing like an r2 or r3 bin. Lumen values are about 20% lower than s2 specs between one and two amps.
Maybe EasyB is onto something with the light recycling in an array, but I’d be quite surprised if they changed binning procedure and didn’t annotate that.
I also noticed that it was below the cree spec for this bin. At 1A I measured 320 lumens so about 15% lower than the spec’ed 365 lumens.
The measurements I did to investigate the XPG3 (The XP-G3 and the mystery of the disappearing luminance) are relevant here. The XPG3 has the similar flip chip design. I measured the candela above the die just sliced and after I scraped all the phosphor from the area to the sides of the die. As in this measurement the candela above the die is proportional to the total lumen output. There is 15% decrease in output when I scrape all the phosphor off, allowing the blue light to escape out the sides. That situation is similar to with the XD16 by itself or surrounded by other XD16s, so I might expect a 15% increase in output when n in array form.
Those DIY board is very similar to my “twig-light”.
Based on my research with Jensen567, I predict the performance would likely to get worse in tight packed array.
E21A designed with sidewall reflector to minimize photon cross talks, big problem in general luminaire when one need to put arrays in very tight space. Even with those side wall reflector, I couldn’t close the gap narrower than 0,3mm. This phenomenon was urban legend to me until all those measurements revealed the ugly truth.
This XD16 doesn’t have any barrier (hence the blue side spill), the cross talks could be even more severe. I guess you’ll get increased luminance at the expense of much lower current capability. This should not be a problem if throw is what you’re after (hopefully it can beat the black flat). As long as the luminance gain vs. max. current net to useful numbers.
Let’s hope we have something better than XPL HI. I’m really curious about it too. Looking forward to see the result.
I’m not familiar with photon crosstalk. I looked at some of your tests. Do you just mean the edges are heated more and burn when two dies are next to each other? It seems to me that problem could be more to do with the construction of the E21A and it’s problem with phosphor cooling?
I don’t expect the XD16 to beat the black flat or XPL HI in luminance. I would expect up to a 15% increase from what I measured above, which doesn’t put it that high.
under the same setup the gapless design maxed in 4,8A (Jensen tested up to 5,5A). Spaced by 0,2mm made no difference. But using the alpha board spaced at 0,4mm it peaked very close to 12A.
250% max current increase.
No it’s not about phosphor cooling. Yes, the phosphor in E21A is indeed hotter than average but it still productive up to 3,2A/die. While in gapless quadtrix (4xarrays) max current was limited to only 1,2A/die.
This very well could be important for the XD16, but the extent of the effect is completely dependent on the chip design and phosphor heating. For example the edges of the E21A phosphor are probably more susceptible to burning because they are farther from the contacts leading to less cooling. I’ll keep it in mind during testing.
Interesting results but sadly about what I expected, in fact a bit better I would say. I only expected around 800 lumens max based on the info I had.
Although you also have a DTP mcpcb, which most could not do or use with flashlights. So expect performance to be worse in actual use with a normal mcpcb.
Sadly like many have said, not very interesting for flashlight use. Although I am surprised that it still suffers the tint shift issues, wasn’t that one of the big marketing points that it fixed the tint shift issues?
A single XD16 is not interesting, but a 2x2 or 3x3 array can have respectable performance (good output and no donut hole) with the possibility of being 3V. If my speculation is correct about the output going up by 15% upon making into an array then it can compete and even beat the XHP70.2 or XHP50.2. Of course the lack of proper MCPCBs would still limit the performance.
:innocent: