Wow, WELCOME all new BLFers!
Will update the lists later!
interesting so just more power to the LED of the TN42 got the numbers up much higher, what is it driven at?
dont know , didnt do the mod
its putting out 2500 lumens by my lightbox though, and 770 k lux, give or take
im thinking its pushing 3 amps
Hi there,
Great project!
I would like to get one as well.
Thanks.
Regards Frank
The Miller requested that I give a little detail on the driver that we are putting together for this light. The XHP35 emitter, together with the 4S cell arrangement, presents some opportunities and challenges. On the one hand DD drivers are out - too easy to kill the emitter. 7135s are out - voltages are too high. On the other hand the 1-4 V voltage differential between the cells and the emitter (and the relatively low current) makes a buck driver ideal and relatively easy to implement.
The boards for the prototype shipped this week, should have them in hand sometime next week. The driver is built around the LM3409 buck controller chip. This chip ticked a bunch of boxes. It has also been previously recommended by RMM, based on his experience with buck drivers, and TA and Flintrock applied it in the Texas Buck driver.
Highlights:
- It is currently using a tiny85 MCU, retaining compatibility with popular existing firmwares. We may migrate to a tiny841 or better before production starts.
- Except for the very low modes, the buck controller will be in full analog current control mode, no PWM. Not that relevant for a thrower, but for very low modes the MCU would set the buck circuit in a low mode and then PWM it to get even lower output. The buck circuit can tolerate 20 kHz or even higher for the PWM frequency.
- Expected efficiency of the driver is 92 %. Together with the real current control it will provide for a pretty efficient flashlight.
- We have a 46 mm diameter driver. This gives a lot a flexibility for component selection. To simplify sourcing, and limit cost, the most popular and common formats were selected for parts.
- The nominal design is for 2.5 A, but all components are generously sized and it would not take much to get 5 A for other applications.
- This is an ‘e-switch’ driver, to be controlled by a small tactile switch only. The expected parasitic drain is a little high, around 0.14 mA. This is mostly due to the LM3409 chip and can be managed better when/if we upgrade to an MCU with more IO pins.
The board is now shared on OSH, for anyone that wants to play along at home 
. It will, however, change before this is all done:
Please add me to the list.
Thanks.
Very nice, DEL!
EDIT: What does a 46mm board fit into?
SRKs David
Oh yeah! SRKs! 
EDIT: Hmm… at 2.5A it isn’t interesting for a SRK. But, if it were built to 5A output, that would make a VERY interesting SRK!
Well maybe someboy wants to mod a Boss1
that would be awesome
I was thinking of a possible mod for a dedomed XPL or XP-G2 in the GT, bucking down the 16V to 4V, 5A+. Should gain throw I'm think'n. Maybe 1.5+ mcd. Hopefully can be done with just some part changes on the driver.
But first is first of course -- let's do the XHP35 HI and be the best reflector thrower --- in the world? 
Double-up (half-down?) the sense resistor and this guy will do 5 A. Will run a little hot of course. Efficiency will be somewhat lower at 4 V output, nature of the (buck) beast.
EDIT: Running 1S, like in a standard SRK, is not really an option.
Well, I was assuming the work required to change SRK battery tube to a 4S set-up. That’s why I said that this driver wouldn’t be very interesting in SRK. The same performance can be gained from a regular 1S driver and 4P cells. But, it might be worth the battery tube 4S mod if the driver were bumped up to 5A.
If it could handle 5A you could run 4x LED’s in series @ a regulated 5A each (the same as to 20A with 4p LED’s for those not aware) and have a pretty nice amount of output.
For example the Q8 would make about the same power it is now but would have a fully regulated non-PWM driver.
Yeah, or if you got some very low Vf LEDs maybe even squeeze a fifth one in there. You’d lose a little of the lower voltage end of your cells, making for an automatic LVP. But, you’d gain lumens on the full charge and possibly a bit more driver efficiency. Even better, get a 10x SRK clone and do 5S2P with low Vf emitters and get a little more.
Nice looking driver, those graphics are awesome.
That sounds like plan…… A great plan in fact. :+1:
“let’s do the XHP35 HI and be the best reflector thrower —- in the world? ”
I like it………… 
Really have no idea how much space the Boss1 driver cavity has
But it has a nice big reflector and if this driver could be placed on top of original driver a mod to make it 4S with 26650 should be feasible placing a XHP35HI in it
Not 1s, but could it drive a 3v led with 3s batteries?
Pretty sure the BOSS1 is standard SRK size, but can't find the diameter in my notes:
I hit over one mile on mine, dedomed XPL (696 kcd), but later with focusing hit ~720 kcd.
Hey, what about 2S? Could this driver do that into a ~6V emitter? Or is it good for 4S only? I’m thinking about a SRK with 18350’s stacked inside, feeding some XHP-50’s or something like that. Then, there’s no need to re-arrange the battery tube.