Not de-domed; stock XP-G2 reflowed onto a copper DTP board. When powered on the first time, there was an initial bright flash, then nothing. Afterward, I noted a significant discoloration/distortion under the dome at the bond wires.
Strange, I’ve found the XP-G2 takes more current than that to fry it. I’m guessing a bad emitter.
I have one more question:
How about a courui with 3 18650 cells wired in a parallel + a dedomed xp-g2 with a direct drive, no driver at all, just thick wires and good heatsinking or maybe even active cooling. Would the current be within some normal range ( like 5-6 A) and yet not too high to damage the LED or would it burn the emitter? The batteries will probably be these protected trustfires http://www.dx.com/p/trustfire-protected-18650-3-7v-3000mah-rechargeable-li-ion-batteries-pair-120476#.Vh_rwVOqqko
I know this question is too hypothetical to be answered exactly, and I will probably end up frying an emitter or two during the build, but anyone with some experience will be helpful! Is it worth trying or not? Thanks!
with those trustfire batteries? I think you’ll be lucky to hit 5 amps, even with 3P.
On a related note, why buy cheap cells when you can get samsung for cheaper?
I was afraid of too much current running through the LED. Those samsung cells look definitely better, but will the emitter handle so much current?
Another possible solution could be using 2 high end cells in a parallel instead of 3 lower end cells. Is it achievable or should I focus on some other solutions and approaches ?
You’ll be fine with 3-4P in direct drive to an XP-G2. I doubt you’d even see 5A to a newer bin XP-G2, the Vf is just too high to hit on 3.7V (4.2v charged) batteries. Run 4 or more in parallel that charge to 4.35v and you might have something to be worried about.
If that is true, what is the best way to deliver maximum amps to a XP-G2 led in a Courui? Wiring those 3 cells into a series and use a buck driver or use really good cells? I thought that if a DD driver delivers 5amps to the emitter, then omitting the driver should save some current and voltage ( that turn into heat in the driver because they are not 100% efficient) and increase the power output. Are my calculations wrong because I did not consider the high Vf or is there something else worth mentioning? thanks everyone, I really appreciate every comment!
I got one test subject flashlight with XP-G2 with FET driver pulling 4,3A (ncr18650PF) and I see no problem for everyday 30 minutes(constantly) use, night after night for more than 2 months. It is hot if hand held(but you can still hold it).
So if properly heat sinked I don't see any problem. Similar like OC cpu :)