Man. These diagrams would be a lot easier to understand if the wiring was color coded.
You recall this ,right? 
It could be added. I’m too lazy though :weary:
LEDs in parallel, single-channel driver (and color coded!)
Bold is mine…
I totally got what you meant, but I want to add that good circuit design coupled with quality components that are properly spec’d can yield outstanding efficiency: TaskLED’s HyperBoost/HBFlex is over 88% efficient even as input amperage approaches 5A. AFAIK the powerful boost drivers from Chinese vendors that are intended for driving multiple XM-Ls aren’t that efficient.
Just wanted to say that so people know it is out there 
Carry on…
Okay. Comfychairs photo makes total sense to me. Five wires sprouting from a common bulb splitting the current five ways into five separate emitters. (why does it split the 9A five ways and not send 9A up each wire?) I could never solder three wires to one small,pad without destroying something. I would have to send a short stalk from the pad and connect the three wires like a shamrock…the holy trinity. Am I getting this right? This way seems simple enough for me.
The Home run looks like a mess. All I see are a handful of possible shorts. Isn’t that asking for trouble, splicing wires all over the place. The diagram itself probably isn’t the best learning tool. What do you use to insulate all those splices against shorts?
The daisy chain seems simple enough. The diagram shows small spliced in wires, but wouldn’t you just hop from pad to pad?
I think all these diagrams need to be redone photographicly. Even the series diagram is confusing.
Photos using clearly marked emitter pads with real red and black wiring laid out in front of you. Relic? How to? I think it’s sorely needed.
If the driver is set to output a fixed current, that's all it's going to put out no matter what's connected on the other end of the wires. Multiple LEDs in parallel divide the available amperage among themselves. Nine LEDs in parallel would use 1 amp each, total 9 amps. Two LEDs in parallel would use 4.5 amps each, total 9 amps.
Thanks for the photos comfy. Easy to understand.
Q: if a 9A driver is driving three LEDs in parallel at 3A each, what kind of current will you be getting at the tailcap?
It depends.
That IOS 9A driver, running an actual measured 9A to a single MTG2, I got 6.57 amps at the tailcap with 3 18650s (and 4.85 amps with just 2 cells, which is close to the same current it did with no driver at all running from 2 cells, just direct drive). I didn't measure the cell voltage under that load however, so I can't calculate the input wattage to compare with the output wattage to get the driver efficiency numbers. Since the voltages are different between the input & output sides the numbers aren't directly comparable unless you figure out the power (watts) on both sides.
I assume you mean single/multiple parallel cell xm-l/xp-g, since that’ll change drastically with mt-g2’s much higher vin?
if so around 9a at start and dropping as tge leds forward voltage goes above cell voltage.
Damn. My MT-G2 triple plans have been squashed. The Roche M170 reflector fits beautifully into the Convoy L2 host, but the emitter is too big and the holes can’t be expanded. It’s a shame too. What a beautiful reflector. It’s a really hefty chunk of aluminum, almost a heat sink unto itself. Deep with a mirror finish. Rats!
Comfychair’s picture is a Home-run setup just like the schematic shows, believe it or not.
The problem with the schematic is the wiring lengths and nodes cannot be taken literally, and I assumed that was known. That happens when you draw and read schematics for a living. I didn’t mean to confuse things with the schematic, maybe I should just delete it.
Anyway, if you are not comfortable putting all three wires to the same pad on the driver like Comfy did (ideal parallel setup), you can use a larger wire to extend the pad, then solder the wires to the end of it. It’s basically the same thing.
Bummer about the reflector, would have been a nice setup.
Awsome job on this dude hope to see this light, ( and your many others ) at the BBQ/Beamshot meet later in the summer.
Nah. The diagrams make you think about what your seeing and make your mind envision little electrical arrows circling around in some sort of order.
After the reflector fiasco, I had to head over to the bar accross the street for a while. Unfortunately, diet coke doesn’t do much to alleviate the letdown, so I’m back to square one. I’ve still got a nice reflector and two Convoy L2 hosts, so I need a new direction to go with this thing. I’ve been building that MT-G2 triple in my mind for over two weeks.
So the best driver for a 3xMT-G2 is so far the modified stock driver in the Lustefire it seems. :)
What kind of batteries are you using?
Just curious, in general, how do you like the Lustfire as a host compared to the BTU?
Last but not least, if you had put some NW emitters in your AK90 and cranked it up a little more. Which beam do you think you would have preffered and why? (modded AK90 VS MT-G2 Lustefire)
I guess your referring to his camera! 
Nice photos.
you win
i ordered a camera tonight
13 megapixel
Yay! :party: Once you have learned to use it you can buy any light you want! ;)
On a sidenote. You don't need megapixels, you need a camera that can do manual settings. And preferably a tripod as well. 0:)
No, TWO tripods... trying to hold the light steady when you're doing a ~2 second exposure is no fun!
-----
What about the IOS 4.5A (formerly 3.5A) driver? Isn't it resistor-moddable?
Tom E says it’s direct drive on high with three emitters. I have one of them on the way as well but not for an MT-G2.