Hhmm. You sure you got a good reflow? The main thing I don't like about the MCPCB is no escape for extra solder, so I was as careful as I could be to not use too much. The reflow seemed like it came out perfect - one of the best I've done, but I've had a few stinkers in my time 
.
Tom, since your ledboard was not soldered down either, I will do a re-test in my usual way without soldering the board. After the test I pulled the led off the board and back on, that may have improved the reflow a bit. Also, I will use larger stiffer rings under the screws so that the pressure is more to the center of the board.
Maybe tonight but it can be later as well.
Do you guys quickly pop down the emitter during reflow to eject the excess solder?
Look here at the 7:40 to 8:20 mark.
Actually the whole video is really good.
No time to check the vid but sometimes yes, usually no. Not long ago I had a reflow with too much solder - pressed down and it just popped right back up - just couldn't get it to sit right. I need to check the vid though to see. Suppose if you hold it down and let it cool, should get it down. Gotta be so careful though to not move the LED off of proper pad alignment - not sure I could do that.
If you “pop” the LED down hard enough it will spit the excess solder out. I generally do this on all the LED’s I reflow. It was a very good idea as it is very hard to get enough but not too much solder otherwise.
Thank you all for your tests and posts! The XHP70.2 sounds finally like the LED which I have been waiting for so many years 
Yep, hold it down and blow on it…that’s my weird method…doesn’t take but a few seconds to cool and stay down.
Yeah, no, I don’t build flashlights for “regular use” or walking dogs or stuff like that.
If I need a flashlight for work or EDC I buy a factory assembled light with enough lumens and throw.
What I build is stuff that pushes limits, usually not very practical for carrying around often.
If I can upgrade my XHP70 at 12A to an XHP70.2 at 15-20A I certainly will 
I hate DD though so I would need to find a CC driver that can do that much current.
Imho most factory assembled light sucks… And that counts even for some “premium” China brand manufacturers.
So I re-did the test with extra emphasis on getting the mounting right. I did see some improvement in the numbers: slightly higher voltages and the maximum is now 6860 lm at 12.5A, but from there it goes slowly but steady down again.
After a complete cooldown I will try a 15A and a 20A burst but I’m afraid that I will not get to the previously reported results.
I’m not sure what to do with this, I may have a particularly low performing led, or perhaps one way or another my test method is not suitable for such high powers. I’m reluctant to post emitter test results that several others have found to be not correct.
djozz, it is troubling when different measurements don’t agree. Back when the XHP70 came out and you measured it there were also discrepancies between your measurements and measurements done in a flashlight at turn-on. Then you took some measurements at turn-on and that cleared up the discrepancy, right?
I’m interested to see your results.
More goodies:
Little pricey but if you wanna play...
Hoping for 3 of these to go in my MT03.
Just to satisfy my curiosity I put the 70.2 P2-1A JAX X6 (18-20amps) using a L6 OP reflector against a L6 70 P2-1C (12-13amps) OP reflector and on the meter and visually (side by side) the 70.2 P2-1A “Out Performed” (stunningly) the old XHP70 P2-1C, as apples to apples as I can get! 
Carry on…… :+1:
Bare led results for going directly from 0 to 15A and 19A (the PS did not go higher) starting with the led at room temperature:
15A: start 8015, 30seconds 7400
19A: start 7780, 30seconds 6070
Also not really different from my first test, my XHP70.2 led in my test situation does not like going over 15A, warmed up 12.5A
Hmm, hard to say what the source of discrepancy is. One thing is that Kawiboy and Tom measured the current with a clamp meter. Depending on the resistance of the tailswitch, the current when the light is in normal operation could easily be 1.5-2A less than with the clamp meter. So if they did the lumen and lux measurements with tailcap in place that could bring the measurements a little closer.
I put the 70.2 P2-3B in the SR90, I think I finally found the emitter that suits this Bowl type of a reflector well enough to finally leave it be! 
Now only battery pack will have to be rebuilt and upgraded, one day, some day? :person_facepalming:
That's true for me. I could measure lumens with the tail jumpered with same wire used with the meter.
I’m not too concerned about lumen differences, ordinary calibration differences could explain that already, what is the real worry is that Tom and Kawi found a very significant increase in output between 15 and 20 amps, and I see a clear drop.
Across 4 measurements by KB, post #108 here, it certainly does show consistent rise in output from 15A to 20A. I should get some more data points in this range as well. I only tested at 11A and 18A here at post #94.
I will just give up for the moment, a proper re-test would require a new $16 dollar led that takes weeks to arrive, soldering a fresh ledboard onto a 30mm copper disc and mount that on my test-rig.
In the meantime I guess that over many people combined enough knowledge is gathered on this led that a test has little to add anymore.