I’ve been looking for that ‘ideal’ driver myself. So far, I think the one you mentioned is the best bet. Although, the way this light is configured now with the stock driver, it is quite nice. Right around 5000lm OTF, manageable heat (driver probably gets toasty though), and enough NW flood for a football field. I would like to have a 5A turbo mode for showing off, but it would not be practical for regular use in this host.
now thats what im talking about!!!
Nice. 
Next step: Make a 9x MTG2 flashlight with 21,000 lumens OTF! :bigsmile:
and liquid cooled
That would be a 300W flashlight if all emitters were driven at 5A. Yowza!
Insert 10 26650 cells, and three canisters of liquid nitrogen for a one hour burn J)
Nice build, nice beamshosts.
Does anyone know if the contact board on this driver can be removed, or does it have circuitry on it?
Never mind. I heard from Hank, and the board has a circuit, and can’t be removed.
I wonder how much of this board you could chop down without harming anything and still be able to attach a contact board to it?
Maybe you should mod a Trustfire x100 (better heatsink vs LEDS and 4x26650) to a mere 14,000 lumen on turbo light and sell them. 
Oh yeah…and I’m in on the pre-order, how much are you charging??
Ouchy, you probably can get rid of most dof the outer gold ring on the contact board. I would guess that would shave 2mm off the diameter. What host do you want to put that in?
B42, I didn’t think about using an x100. Have to look into that. I think the group buy will have to wait a bit. Once I get the bill of materials and build instructions I’ll send em over to you
I think I’m going to have to tuck my driver right inside the head. I’ll probably stuff a 20mm copper mast right down the pill almost to the battery board.
3 X MT-G2 + 3 X 26650
So far I’m looking at these two drivers. I need advice.
http://intl-outdoor.com/boost-7led-3-modes-23-cells-input-circuit-board-p-545.html
http://intl-outdoor.com/xml-multicell-circuit-board-35a-ouput-55126v-p-543.html
I need some input. Pros and cons or alternatives. This will be my first multi emitter light and most exciting project yet.
The first driver, the 'boost 7*LED', might run them but at what current I don't know. Those listed output numbers are assuming 3v LEDs, three 6v LEDs in theory should look the same to the driver as six 3v ones. But it doesn't always come out that way.
The second driver isn't a boost driver so would only work with LEDs in parallel, and the 3.5A output would be divided between them. Very low.
Another option is the '9A' monster with LEDs in parallel, that would give 3A to each one. And buck drivers are almost always more efficient than boost drivers, so it would take less input current to get similar output.
I’m kinda unknowledgable when it comes to “parallel” setups.
Ouchy, here’s a schematic of a few ways that drivers can be wired to multiple emitters.
The ‘Emitters in Parallel Home-Run’ is how the LusteFire 3XU2-X3 is wired, before and after my mod.
The ‘Emitters in Series’ option is how the BTU Shocker is wired.
The ‘Drivers in Parallel’ option is how my Copper DIY light is wired. Most SRK drivers, and texaspyro’s custom SRK driver work like this too, even though it’s all on one board.
Hope this helps. 
Edit: updated picture to differentiate parallel wiring methods.
If I’m understanding this diagram…from driver + the red wire would jump from + emitter to + emitter to + emitter and then fom the driver - the black wire would jump from - emitter to - emitter to - emitter in the same sequence as the + wire with both wires ending at the third LED ?
Yup, that’s what it is. You can do it with one wire from the driver (parallel daisy chain, as drawn above) or three pairs, one pair per emitter (parallel home-run).
I prefer the home-run method, since the current is slightly more balanced for each emitter.
Updated drawing. Electrically from a schematic perspective, daisy-chain and home-run are equivalent. In reality we are dealing with significant current, so the wiring needs to account for that.
In parallel all the LED - pads connect to the common - on the driver, and all the + LED pads connect to the driver's common +. The voltage output from the driver is the same as it would be with only one LED, but the current is divided equally among them.
In series the driver has to put out a voltage equal to the combined voltage of all the LEDs in the string. But since the series LEDs are all in one continuous circuit, if the driver puts out 3 amps it will give 3 amps to the first LED, and 3 amps to the second, and 3 amps to the third.
If you calculate the total wattage output for both designs they will be the same. Volts times amps equals watts.
So assume each LED has a forward voltage of 6 volts. And you want 3 amps per LED.
In the series circuit, you add the forward voltages of all LEDs: 6v + 6v + 6v = 18v. And the driver's 3 amp output gives 3 amps to all LEDs. 18v x 3a = 54 watts.
In parallel, the voltage is only 6v. But to get 3 amps per LED the driver has to output 9 amps. 6v x 9a = 54 watts.
The catch is, a boost driver is incredibly inefficient. Lots of waste heat and shorter runtimes and requires a much much higher current from the cells. The numbers for a 3x MTG2 at 3 amps from 3 cells work out to give the advantage to the buck driver and the LEDs wired in parallel.
ew! ick!
Yup, same is true (and even worse) for daisy-chain. The 9a has to flow from one place.
As I hinted at earlier, schematic diagrams generally ignore wire loss and do not document wire length. You need to deal with that when you wire it.
Keep the ‘home’ node as close as possible to the source (i.e. driver pads). Alternatively, use larger wire for the ‘common’ wire that comfychair pointed out is carrying 9A.
Also, try and keep each home run equal in length to balance the emitter current as much as possible.