.
I see your pic with the turbine engine, Sweet :+1:
Thx CNC!
This is the only reason why I like flashlights, working in dark and visibility is essential. Rainbow tints give me a headache and give the wrong perception of a hydraulic leak (yellow-green corona looks the same as hydro fluid on a titanium material)
Recently received S21B, and when I shake it I can clearly hear rattling sound, when I shake head only - sound persists. And when I touch glass by finger I can definitely fell glass and reflector moving (and making that sound). I try to screw down bezel - but it’s screwed down to the max, no space left. And for s21b I can’t screw down the pill (because there is no pill). Any ideas what should I do?
Check the o-ring between reflector and lens.
Replacing it with a thicker one would help too.
Hello!
Just a quick question about early versions of the 12-mode bastard biscotti drivers. I had a ∅17mm driver ordered from FastTech on July 9th sitting in a drawer which I just used for a “dormant” build aaand… holy sh1t! The thing seemed to work right at first, but started misbehaving nearly from the start doing all sorts of glitches: incorrectly changing modes or not in any way I could discern, random strobes, and right now it just doesn't powers on. Already dismantled the S2+ flashlight in which it is installed for inspection a couple times, everything else inside is in full working order. Is the software so terribly bugged? 
That’s a review of a specific bin by Rngwn. I hope Simon got the same
I’d like to comment on the “12-mode driver” (Biscotti clone), based on my experience with a few units that use it:
Convoy Z1 (SST40, 12-mode driver)
Convoy C8+ (Osram NM1, 12-mode drive)
Convoy S21B (SST40, 12-mode driver)
Convoy L21A (Osram PM1, 12-mode driver)
I notice these inconsistencies:
- when in battery check mode, there are some slightly different behavior: on some lights, the main light will stay lit (like Moon mode level), while it blinks the battery level (1-5 blinks), but on some other lights, the main light is off while blinking the battery level.
- when in default group 1 (5-modes), the 1st level and 2nd level brightness appear to be quite similar in visual brightness (although this could be due to the LED and current applied to the LED)
I have purchased LEDs from the factory. Each of the 4000K and 5000K color temperatures will be produced at 500PCS. I specifically require XY to be below the black body locus.
I was engaged in the packaging of LEDs in 2008 and deployed phosphors. As long as a very small amount of 660 phosphor is added to the phosphors, high color rendering LEDs can be produced.
It may be because the O-ring is too thin or the gasket is too thin. I have some 1.3mm thickness O-rings, if you need them, please send your address to me via private message.
The earlier version only had the bug that the group was not saved, which has now been fixed. I feel that your drive problem is caused by poor contact. But if you need it, I can send you a new driver.
Thanks for your link
Simon, how many amps in sft?
Simon, you really get nice LEDs recently. 
I’d imagine the same as the SST variant. Peak output around 8A likely. Maybe 15-20% less output
My estimate is that the SFT70 is a great high output long thrower led. A bit smaller die and bit less output and much cheaper than the SBT90.2 and with comparable throw.
But 6V or 12V which greatly limits the driver choices.
Still, a good option. Especially for someone like Simon who doesn’t worry about Andruil compatibility.
With the XHP35 HI version of the K1 there is certainly be a way to use Anduril on a boost driver (granted it may be designed specifically for that task). Wouldn’t whatever method is used for the firmware coding to run on it be open-source as well?