I did not even try that because on every led other than Cree leds I failed. I will certainly try that too!
So how do you like the resulting tint?
Thanks a lot for that picture djozz!
Thanks to Web Plot Digitizer I was able to calculate the die size of 4.69 mm².
already thought that someone would like that pic for that reason. And how does it compare to the XM-L/XP-L die size? (it is bigger, right?)
Bigger. This explains why it’s floodier.
XM-L2 3.92
XP-L HD 4.16
LH351D 4.66
XHP35 5.32
Not as good for throwers as I hoped for, but not bad either. I calculate 110 cd/mm² at 6A, 130 cd/mm² at 9.25A.
So throw alone mediocre when compared to top dogs, but when you take into account other features it becomes OK. Especially with linear or FET drivers. Not sure if it makes sense in throwers, though in more general-purpose lights it’s not bad.
EDIT: Fixed calculations. And conclusions.
Where does Luxeon V fall into that?
DB Custom, if I do a build like this again, I’m definitely going to orient the stars as you describe. I don’t think I’m going to step backwards to revise this one, so I’ll just have to make do.
My latest thinking is to pour a layer of epoxy to just below the mcpcbs, let it harden and then dremel out groves for the conductors, fit them in, and then epoxy them in place. I’m going to be out of town for a bit though, for my oldest nephew’s high school graduation, so there is a good chance that my plans change before I have time to work on this build more.
I think I mentioned in my earlier post about this mod that I used lead free solder between the sinkpads and the copper disc with the idea that it shouldn’t melt when I reflow the emitters with Sn63Pb37 solder paste.
That combo plate photo has accelerated my appetite. I better call in my previously planned carne asada plate order, and go pick it up.
I think my years of going for the hottest peppers are behind me. I was pretty fond of chili pequins and tepins. My father in law turned me on to them. He’d eat them almost like popcorn. You’d think nothing could kill a guy like that. You’d be wrong. Stubbornness and too-tight socks did him in. Never got a chance to back the BLF X6 SE I’d given him. He’d complained of some problems with it, so I cleaned it up for him, and had a new S2+ for him, some new batteries and a charger, to go along with it.
Ghost pepper has a good flavor. Even if you’re not adapted to the level of capsaicin, it could still be used sparingly to reach a comfortable level of spiciness.
Habanero has a good flavor too. Some of my favorite sauces and salsas are habanero-based.
As for jalapeno, I usually prefer them smoked chipotle style, though the usual sour variety can also be nice sometimes.
If you just want to make food really spicy though, you could spritz it (lightly, from a distance) with pepper spray… but it doesn’t taste very good.
For some reason, I can’t handle even medium concentrations of capsaicin.
The only time I liked using capsaicin was as a painkiller when I was still a black belt two years ago in Karate Kyokushin. Amazing pain killer compared to even pain patches.
Koef3 calculated that by adding side emitting surfaces,which is not the correct way to calculate LES.
Chip size is 4mm^2 according to Lumileds.
Thanks for the correction.
Added a blue banggod lighted tail switch to my Eagtac D25C and a red Led4power lighted tail switch to my recently arrived SWM V11R. The D25c I had to lathe up a shim for it to fit nicely.
Ok, I was lured into hot dedoming the LH351D. I have hot-dedomed many Cree leds succesfully and on some Nichia’s I managed to break off the silicon just above the phosfor (so a succesful result but not a full dedome). I tried many other led brands and always failed (minus some mediocre but ok functioning dedomes of the SST-40) I’d say my skills are adequate for the old Cree leds but I’m sure I do not have a perfect technique.
The dedome was a fail, the phosfor was clearly better attached to the silicon as it was to the die, as with Nichia leds my assumption is that the phosfor is embedded in the silicon. Wildcat’s technique must be more sophisticated than mine.
So I’m leftover with another high power royal blue led, in 3D with the stereomicroscope you can see that the design is flip-chip, weird connection pattern btw 
Can relate.
Re-made the switch on the triple XP-L HD carbon fibre with magnets.
Made a CRX Brass 405nm Laser - 120mAh Lipo - Clicky Side Sw - Rechargeable.
Although difficult to implement I quite like this way of incorporating a magnetic tail.
Having become very enthousiastic about the tint, I prepared another sliced Samsung LH351D 4000K 90+CRI today, this time for my daily EDC (S2+shorty). And after messing up one more led by slicing through the phosfor, I reached perfection with the last one: sliced to the exact right height just above the phosfor with the help of a steel washer of the right thickness , and sanded with 1000/3000/5000/7000 grit (the 1000 grit with the washer still around the led for guidance). The tint is great, on a Aspire 18350 I get 940lm at start, 860lm at 30 seconds, and it is quite a bit throwier than the triple that I EDCed for a while.
Looks good :+1:
Nice. But performance is quite a bit lower that I’d expect. What is the current draw?
Now that is a pretty sliced emitter !
Wow, really neat 
The long answer 
The current draw on a fresh battery is 6.3A.
Using TA’s test of a 5000K 70CRI LH351D, at 6.3A the led is at 90% of its maximum output, which is in its quite inefficient region, I usually aim at 80% in my builds, I should remove the spring bypass at the driver side.
Also, assuming 18% light loss from reflector and lens, the led lumen of my led (at 30 seconds) would be 1050 lumen, which is 50% of the 5000K 70CRI version that TA measured at 6.3A, so I paid a dear price for tint and CRI. (my lumen may be slightly lower than TA’s lumen though).
The Aspire cell at 6.3A is at about 3.7V, so the power draw is 23.3W. That makes the system efficiency at maximum output 37 lm/W. Yeah, pretty bad, it is the efficiency of my first XR-E DX-cheapo’s six years ago
Edit: checked under 350mA (it was 286mA), system efficiency is quite a bit better: 66 lm/W. That is still about half of what the most efficient flashlights (with moderately driven cool whitehigh binned XM-L2 leds) do. Btw, I stacked an extra 7135 chip on top of the first one on the BLF-A6 driver, so the flashlight should have ok efficiency up to 700mA/200lm.