Could it just be variances in Vf between emitters? Or would that not account for that much difference?
No, it's not Vf spread. On mine it did 2-2.6A in stock form, with just the stacked '000' resistors removed and all 3 resistors in place, current dropped to like .45A.
Instant 5A+ on the emitter and probably rising… I switched mode right away… Low seemed to still work nicely… 
Awesome! So they are not R0 then. try adding R12 resistors like I did to ‘tweak’ the current where you want it on each channel. Don’t forget to remove that dead short 
I'm really confused by how this thing works, usually after playing for a while things start to make sense but this one just keeps getting more confusing.
Mine, with a jumper, did the same as above, way too much current for direct drive from 4 cells. BUT, as it sits now, with 3 '000's instead of the 2-1RO/1-000 setup, it runs at 3.40A and does not increase when I jumper across the 000s. See? Totally makes no sense.
yes, that was what I though too. But I did not dare to say it in case my noobish skills was not as good as I hoped…
So, right afterwards, I used my noobish skills and made some artwork…
Wires should make some resistance, on the end of them sits this , 0,5R resistor. Its the only ones I have…
That bumped the amps up to about 2,4A, (from 2,2). Added another one, in the same professional way… And then saw about 2,47A… Not much increase there
I used the wires since the resistors was considerably smaller then the ones on the board. For testing, I felt more comfortable doing it the safe way, up in the air…
Its clearly possible to increase the current.
Do you have a link to the most appropriate resistors. Adding less of them would be better… Its not a big deal if I end up going slightly beyond 3A.
The R000 resistors have some resistance… probably around a milliohm or two or three. Instead of shorting them with copper wire, try some nichrome.
I ordered the complete line of sub-1 Ohm resistors from Fasttech. I added R12 resistors to each channel, enough to get about 2.8A. Each resistor added about 250mA. I used three resistors on two channels, four resistors on the other one. It worked for me, no idea how yours will behave though. Comfy’s seems strange too.
All of these drivers are strange. It seems there are inconsistency in what people get too.
I ordered some 0,12R resistors and some others.
When I get access to my light again, I need to look at the circuit closely. I doubt the MOSFET RDSon is part of the current feedback resistance to the 7136. It is too variable from FET to FET. SO, the variance is in the traces and resistors used. Mine seems to have the most sensible resistor values (1R, 1R, 1R, 0.03R for 0.0275 Ohms total). Working on the 7136 Vfb of 50mV, that’s ~1.8A. After adding three R12 resistors, it’s 0.0163 Ohms, for 3.08A. This seems add up for me. Trace resistance will account for the slight reduction to 2.8A. I also confirmed, that the FET on-resistance cannot be part of the feedback resistance. It would not work at all.
So, the main issue for Comfy and RaceR86? What is going on with those resistors??? :~
I don't really care except curiosity for curiosity's sake, as mine will be used in... well, an unconventional manner (see here). 5 XMLs in parallel, using only one channel (sort of, lol) of the original driver. It may set itself on fire, I've not yet tested it with all five LEDs connected at the same time yet. If it does catch fire no biggie, it's all leftover spare parts I probably only would have kept set aside for testing anyway.
:party:
Five XM-L in parallel, trying for 15A? On one channel? You need two things handy before you try that; video camera and fire extinguisher
I’m serious on the first one, not completely serious on the second. It might melt, but probably not catch fire.
I see you stacked two FETs, which is better than using just one.
I dunno how much current you will get, but it should be substantial. J)
Edit: Wait, you only got 3.4A??? Maybe replace the three R0 with a piece of 14ga wire or a massive blob of solder. Regulated 15A of current to five XM-Ls from four 18650s is something silly only I would do (and did). You are on copper so it’s all good J)
Does anyone know what pins the PWM is coming out of on the MCU? Would it be same three pins as the MCU on the “good” SRK driver?
I bought one of these for Father’s Day. I started looking at it last night and noticed one led was dead and another was very bright. Opened up the light and there is no FET on L3. L1 had extra resistors (3.25 amp at the LED). I want to have this light working in time for Father’s Day. So I intent to connect to a 7135’s board for L3. Just need to know where to pick up the PWM.
The FET is completely missing?? Nice quality control.
I do not recall which pin, will take a look if no one else knows.
I’ve been looking at close up pics here on BLF. Wires are in the way of all of them. I tried to take pics, but my crappy camera doesn’t macro well at all. I will try to back light the driver tonight when I get home.
It looks like Pin 7 of the MCU is the PWM output. It goes to an R680 resistor before going to the 7136 chips. I suggest you connect after the resistor.
In this pic from comfy, it’s identified as R6:
You should connect to the top side of that resistor over to the Vcc on your slave driver.
Correct... pins 5 & 7 are NOT additional PWMs as in the original SRK's MCU.
Thank you relic38. I’ve been looking at the driver, but haven’t been able to figure it out with my limited electronics knowledge. I was just about the search for the data sheets on the FET amps so see the most likely pwm feed pin and then try to trace back from there. I was expecting a resistor in the PWM path (like the 10K’s in the “good” driver) and was going to connect after the resistor. I appreciate you advice on that. I feel better about doing it that way.
Probably going to just use 6 7135’s (for about 2 amps) without a circuit board. I’ll report back on the results. Hopefully tonight. Thank you again! 
Thank you again relic and comfy. Got it working. Sorry for the dark picture. What you see is a solid copper wire on it’s side. It is soldered to several ground points to increase thermal transfer to the driver board. Kapton tape is under one one spot that is not a ground connection. The six 7135’s are on their side and soldered to the missing FET’s biggest pad (one that feeds L3-) with copper braid. The PWM feed is connected to the top pins by the red wired. That skinny red wire was one of the original LED wires.
Now I just need to sort out what is going one in the resistor bay for L1. It’s currently driving 3.25 amps. Too high for a light for my Father.
Wallbuys is sending me a new driver. They sounded like they were more than willing to send a whole new light, but I told them a driver is all I need. They were prompt and professional. I’m really starting to like that outfit. I don’t care what anyone else says.
Try taking off a couple of the 1R0 resistors (the pile of resistors at the top, in your image), nut sure it will do much though.
Taking off the R030 will probably drop it too much.
If you have access to resistors, try replacing the R030 with an R10, see if that helps.