Looks plausible. Something close to whatever is in there, in parallel, should cut the resistance in half and bump that up. Need a teardown or info for further guidance.
1 Thank
The dream would be have a similar headlamp in dual channel and adjustable CCT, powered by AA.
Please tell me you’re talking Eneloops…
1 Thank
Maybe this driver will be implemented in D2 
1 Thank
Yeah, Eneloop and 14500 support
I somehow suspect that making it both boost and dual-channel would make it hard to shave 2mm off the diameter. The D2’s tailcap fits in the D3AA’s head. Maybe for a single-channel boosted D2 though.
1 Thank
It would be nice too! I keep finger crossed for implementation of this great driver into more lights.
1 Thank
The only disavantage
One downside of these L-shaped dual LED lights in 14500 format is that tend to be a bit long and heavy.
I don’t really like have contact with metal, would like a bamboo one made!
I would like that too, but I think the D3AA driver might be too big for the D2 body tube. And the driver has a 9V output, so a single 3V 519a would not work, but a 12V B35AR, XHP35 or GT-FC40 could work. Hank would have to use a different mcpcb for the 12V LEDs.
When the D2 first came out, I asked about a single-channel boosted D2 and was told that there were no plans. I wonder if the Freeman driver may alter those plans. However, I still think that that depends a bit more on whether someone wants to (or even can) tweak the component layout for a smaller board. And while “can” is likely, “will” is a lot more variable.
I seem to recall reluctance to make a 21700 version of the D4V2 though… 
By about 2mm by my guesstimation. Looking at one side of the driver, there’s some hope that a layout change could alter that, but I think that that sort of thing is best left to the person who designed the D3AA driver in the first place. When it comes to this sort of thing, I think argumentum ab auctoritate is not a logical fallacy.
Going from 9V to 6V output is (most likely) a relatively simple affair. Assuming that it remains 18W, it’d increase the output to 3A/emitter, and the increased efficiency might allow that without thermally overwhelming the D2.
Changing it to 12V is likely no harder, but more dubious. The only emitter coming to mind that needs 12V that doesn’t have 6V variants/capabilities is the FC40, and while I can see a 5050 MCPCB for the D2, I have a hard time imagining a 7070 even for a domeless emitter. The ID on the D2’s bezels is only 8mm, so it’d practically be a mule. And I can’t help but think Jackson will have 2*SFT40 D2’s on his site at some point.
The B35AM would be an interesting option though. Barely larger than 3535, so optics will be less of an issue. Not fond of high amperage, so also a non-issue. The only problem is the footprint needing a special MCPCB, which is fairly trivial considering that it’s close enough to 3535 to fit inside a D2. Maybe pop the optics on a drill press with a reamer. It’d be great to have a tiny E21 DT8 
2 Thanks
The only thing that put me off Hank flashlights was the outdated drivers. Am I correct in assuming that we can hope for an updated boost driver from @thefreeman on all other lights?
1 Thank
Is there enough space for dual channel driver? really want 2 white leds and 1 red.
I cant comment on Hank’s future plans.
With dual fuel support It should be possible but not something as efficient/powerfull as the D3AA driver.
A lower power boost converter + 2×3V linear channels should fit.
Something like 0.8A (global) max on NiMH and 3+3~4A linear channels on Li-ion, probably limited to 4~5A global.
2 Thanks
1 Thank
yes, but not 14500.
hopefully this will happen in the near future, if there is a market for that of course.
I also have a particular way i want it to function. 1H from off always brings up red instead of moon mode. 2C, 2H, 1H works like normal. 3C changes to white. When in white 3C switches smooth/discrete ramping, can’t go in to red, unless enabled with 9H.
I guess red replaces moon mode and functions as separate channel with regular controls.
Unless you have a better idea how it should work.
1 Thank
Just received a Red D3AA w DD 4500K 519a
It is smaller than I expected, and the output range is very impressive, even on Eneloop.
The most important things I want to change is in the firmware:
Eliminate the Blinks at 10 lumens and at top of ramp, and delete 5C momentary mode
I do not know how to create a custom hex and request someone to please help. I do know how to reflash. Thanks in advance.
Everything else is really nice! The Oring grooves have been enlarged, are anodised, and fitted with a thicker O ring than the first batch.
Really impressed how low it goes, 0.01 lumens with NoPWM and no flicker.
love the ultra low Flicker Index of 0.0040, and the really low modulation frequency, and the very flat lux flux on the right panel
5 Thanks
There is no firmware mid ramp blink, it’s hardware related and happens at the transition (or ”gear change” between the low and high range (or ”gear”).
It’s due the inacuracy of the MCU’s integrated DAC on the low end (offset), which create a mismatch between the top of the low range and bottom of the high range, resulting in a jump or a zigzag when it’s respectively higher or lower than it should be.
It’s amplified by the large op-amp compensation required with this boost IC which makes this transition more visible.
So this varies between lights, ones with low DAC offset will have a smooth transition, the ones with higher offset will have a more visible bump.
I could have used a lower HDR ratio, which would have lowered the effect of the DAC offset and lowered the required compensation value, making the transition less visible, but it would have been at the cost of a higher moonlight.
Probably a ratio of ~1:100 would have been better compromise than the 1:135 ratio used, but this is hardware set, it can’t be changed now.
Another option would have been to use 3 sense resistors like on the new FF Lume drivers, as the ratio for each gear is smaller the effect of the offset is smaller. But in the D3AA case it would still be a bit visible due to the boost IC used, and there would be 2 transitions instead of one, not sure it would have been better in this case. (Also it takes more space).
9 Thanks
thank you… that is too bad
fwiw, the moonlight is WAAAAY lower than necessary… otoh, the blink or dip in the ramp near 10 lumens is REALLY aggravating!
I would gladly take a higher moonlight if it would get rid of the uneven ramp.
oh well… I might just use stepped mode instead… (I like 12 steps)
next light comes tomorrow, a Skillhunt EC200s mini w 519a 4500K and dual channel Red
1 Thank
What emitters do you have in yours? Mine has 519a’s and other than the blink at the top of the ramp, which I believe is an Anduril thing, I’ve not seen this behaviour.
I love how low the moonlight can go, although I’ve realised that I can have moonlight set higher than I thought without it affecting night vision.