Nope, at 6V they’re running 2S2P with the dies, so you’re looking at 3.16A per die I believe. 1.58A each would be if the dies were all in parallel (4p). Sounds about right?
ECHO
I forgot about the 6V thingy. 
It IS 1.58A per die, can’t divide 19 x 12 any other way.
Yeah, but you can’t add currents together in series wired leds.
Just like cells in series don’t share current between them like parallel cells would.
2s1p supplying 5A is 5A per cell.
1s2p supplying 5A is only 2.5A per cell
Same thing with the leds.
the math is easier with watts (at least for me)
19A x 6v = 114W
114W / 3 emitters = 38W per emitter
38W / 4 die = 9.5W per die
9.5W / 3v = 3.16A per die at 3v
So they aren’t double as efficient as separate XML2’s, they’re right in the same ballpark.
Same thing with the leds when talking electricity, be it current or voltage or whatever.
But output, lumens, total is divided equally by the number of dies. The very reason they’re wired 2S2P is that they are 3V dies to begin with. XM-L2.
My point is that given the amount of current flowing…19A…seeing 13,903 lumens… each die is kicking butt and taking names. Sure, they’re holding hands to share the 6V burden, but effectively doing the work nonetheless.
So, if I took 12 10mm SinkPADS and put 12 XM-L2’s on em and wired em up the same way as these, would I get the same end result?
If you put 6V 3 amps into a 2S/2P led, then each of the parallel sets will get 3 amps?, but that would mean each of the series dies gets 1.5 amps? So overall the four dies are getting 6 amps?
Diagram please, my little half dead brain is aching.
Yes. 3amps at 6volts power into the driver LED= 1.5amps at 3v to each of the dies = 6amps at 3v total power to the LED.
18watts in = 18watts out, just in a different form
Only one way to find out 
But is it really surprising for a highly binned cool white XML2 die to output around 1000lumen at 3A? I thought that was about expected?
No, the parallel sets will share the current that you supply. Exactly how cells in parallel will share a current load.
So each parallel set of 2S leds will draw 1.5A @ 6V. Half of the total current going into the emitter.
And yes in turn each die in those sets will have 1.5A passing through it (since series currents don’t add) and a voltage across each die of 3v.
One cell will drive one LED in direct drive. 2 cells will direct drive 2 LED’s in series, the 2 LED’s are sharing the current, so there’s still one LED per cell…effectively.
The XHP (Which of course stands for Extreme High Performance for some reason, which we are presently showing the performance is the same that we’re used to) uses this doubling of the emitters to be 6V or 12V capable. Only 2 dies would not allow 12V, so the 4 die pattern allows configuration to run it either way, as 4 3V emitters for 12V (series) or 2 3V emitters in Parallel- twice- for 6V.
It’s just juggling. Better heat management because the individual die is in the lower current range where it’s more efficient. Same reason we see high performance from a triple XP-L.
It's high performance for the lighting industry. It eliminates as much as 75% of costs in mcpcb's, optics, and wiring connections per unit of light.
We are not the intended recipients of this Extreme High Performance.
Shit almost 14000 lumens……i don’t want to want it 
I wonder if you could sell those as a weapon to various LEO’s, this thing must be like a soundless handheld flashgranate if pointed at you…….
i can’t even imagine a 14000 lumens strobe
Mouse over for a comparison at the same camera settings of a pretty hot Fenix TK75
That tree is going to catch on fire one of these days….
I still have a few of the LD –1 drivers and this FET might be able to handle driving a triple or quad board unlike the stock drivers. Probably need to stack sense resistors to up the current beyond the stock 5A but might be another driver option for the 35 mm triple. Do you have a pic of this FET? None shown in the link.
Holy smokes!!
So that is ~14,000 lumens w/ the TK75 as a host?!
I am a THROW-A-HOLIC,But that is unbelievably BRIGHT!
@Dale the mouseover don’t work for me.
@Rufusbduck, that is a great idea i would love a triple or quad without any PWM.
I hope led4power makes a LD-2 with one or he starts selling the LD-1 so we could upgrade it with this.
Please if you decide to make it do a write up on how to do it, if it is anymore to it than reflowing the Toshiba MOSFET and some more sense resistors to it and even then
cajampa Did u take the “Cursor” outside of the picture?
The expander board(which I posted over in the Oshpark projects thread)is a daughter board for the FET and extra sense resistors needed to run either at more than stock 5A(extra sense resistors) or multiple parallel Leds( lower Vf means FET needs more heat sinking)or both but this FET might not need the daughter board and stacking a 1206 or two might be fun. 8) Rest assured if it happens there will be pics. In any case it would be with a different led.