Yes, you are correct. It doesn’t go as low as many would like. The lowest level is maybe 40lm? That is a very crude guess only from memory and not by direct comparison. But it’s definaltely not a moon mode and more like a moderate low mode.
Hahaha that ramp is funny, not a wide range but funny yeah.
It seems to go a little lower this way.
Thanks for the answer LightRider!! Well…that is not what I expected when I saw stepless dimming, but ok, we’ll survive this UI 
It will always be a nice BUDGET light as long as it works well in the modes it has, once that modding it is not an easy option, as you and TomE said!
Eheheh, at least it is funny :partying_face:
Maybe - I’m just saying - maybe the manufacturers would be open to use one of the mentioned firmwares, Narsil, NarsilM, others, if we “advise” them to do that!
Do you think this was a good option, or was it desirable? If these firmwares are open source, it would be interesting to get to these manufacturers and tell them about it.
I don’t have much experience contacting manufacturers. Sofirn and Amutorch are just 2 of them to whom I provided some feedback on the flashlights I bought (through AliExpress).
- Sofirn - represented on BLF by Tracy Wan - incorporated some feedbacks given here about their AA lights (were from H to ML, now are L to H, and strobes were “dismissed”).
- Amutorch contact, despite didn’t make the changes (yet?) took the given suggestions to their responsibles.
This Jaxmnve manufacturer already uses a “ramping UI” (+ clicks). Maybe they are open to make some changes on their drivers and UIs and make a more “nice” light.
I’m just saying! Sorry if I completely ignore how things can happen on this manufacturing world, as I have no experience on that :person_facepalming:
As a user, I would just like to have the best flashlights possible, keeping the budget side on it! :money_mouth_face:
I wonder how would they react to the legal burden of GPL…
BLF firmware, whilst wonderful, is not the be-all and end-all.
Neither are simple FET drivers, necessarily.
And is predicated on relatively expensive MCUs.
And is an acquired taste, maybe more than many will ever want, or use fully.
Plenty of good stuff being done commercially with much more humble stuff, E.g. I have a huge 4D cell “10W Cree LED” torch that cost £12 (reduced) from Lidl/Aldi last year, with simple high, low, then for more control a great ramping UI, up and down, flicker at each limit, hidden strobe, perfect.
All good, genuine Cree XML LED in good neutral white tint, driven well, good reflector, the metalwork alone weighs over 700g, when loaded with included x4 Ds (a few quid themselves), 1.3 kilos. Fit, finish, o-rings, anodisation quality, all great.
Bought it having a project for the chassis in mind, but so good I never bothered. Perfect for walking the dog, or investigating “bumps in the night”.
it’s very good but price is………
The driver is pretty simple. However it looks like there are some things hurting the efficiency. Idk?
The driver uses two large limiting resistors that have a total resistance of 1.32 Ω.
It is interesting that the driver uses the same fet for reverse polarity protecton and for the output mosfet. So the current travels through an fet twice before it reaches the led. I believe this is less efficient than using a diode for the MCU only like the BLF drivers do?
However the reverse polarity protects both the driver and her charging circuit. So after the rpp mosfet the batt- is isolated from the rest of the driver. The only other part connected to the host is the switch.
All of the charging circuit is contained on the vertical switch pcb apart from the rpp fet. I am unsure if the charging uv has rpp or not. I’m assuming not.
The charger has a program resistor of 1.5k Ω. This should give a charge current of .66 amps. This is assuming that the charging ic closely resembles the MCP73831/2 as I couldn’t find a data sheet for the HXN-MH. Someone somewhere mentioned a connection but I’m not sure. The pin layout seems a little different so there might not be any relation st all. Idk?
Nice Info! Thanks!!
I’ve this clone m3 modded with rechargeable Narsil. The charging circuits is separate from driver.
edit: sorry, still figuring out how to posting pics.
bwings that looks awesome, great job on this mod
+1 here!!! I’m sure the light now worths, at least, the double !! :sunglasses:
Wish I had the skills for it :person_facepalming:
Just WOW bwings !!! Work of art right there 
Bwings, that looks great! You beat me to it. 
I’m stuck on finishing the firmware but I should have done today or tomorrow. Was the pill hard to pound out on your sample? What is the chip marked A6SZ?
Thanks for all the likes and comments.
Yeah, the pill is press fit and hard to pound, was tried to heat before, I used a 16mm brass tube and press from battery side. The tiny A6SZ is AO3416 Sot23 Mosfet.
Is that for reverse polarity Protection?
Little late, but yes - Wow!! Impressive mod there bwings. Don't think my 60 yr old eyes and hands can do that level work, least not in a timely manner...
Forgot what I did to loosen the pill in my M3 mod (yes 60 yr brain on many beers), but recall it being a little stuck. Worked on a bunch of similar press fit pills though, bout always can find a way though.
Ohh - sure looks the same as the OTR M3 - same driver setup and brass pill piece, switch board, etc.
The only difference I see is the reflector isn’t screwed into the pill like the m3 but outherwise it’s the same.
Tom, you willing to take a look at the narsilm page? I’m in the middle of trying to figure something out. Thanks.
It’s FET PWM at pin6, since I couldn’t fit LFPAK56/SO8 (not enough space for its size).
This light has definitely given me the biggest mechanical challenge of any mod thus far. It’s fun though
I’ve also had a hard time choosing the right emitter. It’s dd but dd is only accessible via double click. Still, if I want to use 219b, 219c I will need to reduce the 2x click turbo as well. Maybe an xpg3 would be ok and xpl2 would probably do well though only for short bursts. But putting a direct drive driver in a light like this shows it’s desire to make an impression. Any of the old gen leds will work well but will fade more with the battery on regulated modes.
So, I think I would like to stick to the plan and make an impression. I have an xpl2 here with 80+ cri and if the beam is not satisfactory I can add a shuttle filter.
Ahhh! What am I thinking! A useful edc light like this without 90+ cri! I can’t let that happen! Maybe tomorrow I will reconfigure my hex file to accommodate a 219c 4000k 9050. Xpg3 90+ would be an option as well but only if I could find a tir lens to fit this thing.
What do you think? Am I making the right choice? 219c?