My personal opinion on the number of modes: Five modes is simply too much to step through.
Four is much better, and can cover the 0-100% quite well.
Three is good for usability, but can't cover well enough.
=> based on these findings I started to design the TheStar FW, which has the usability of 3-mode FW, but it offers 5 modes (+ hidden blinkies, but those are not the point here).
Here's how it's done:
- LowLow mode is hidden (can be easily found when needed, but it's not bothering in the main mode cycle) - Low & Medium are the first two modes
- High is not accessible as a standard mode, just as Turbo rampdown level (-> not in the main mode cycle, but still usable)
- Turbo is the third normal mode
For LowLow, cycle through the modes twice, and you'll find the LowLow after Turbo, like this: L -> M -> T -> L -> M -> T -> LL For High, set the light on Turbo and simply wait 90s (or whatever the turbo timeout is programmed to be) For returning to Turbo after timeout, simply half-click and you are there.
The more I use it in this 3-mode configuration, the more I like it. :)
Edit: And if you like H->L mode order better, just short the star #3 => T -> M -> L -> T -> M -> L -> LL
That’s sounds great. I am just going to go with that. I’ll be chatting with them tomorrow to confirm that I can get this firmware on them, but I don’t see any reason why they won’t. They will probably steal it though.
That's very much possible, but would that be a bad thing? Better flashlights for everyone!
I was anyways planning to release TheStar as "free for all BLF users for non-commercial use", which pretty much means that it will be freely available => will be stolen and used in commercial things too.
If it's NANJG105C compatible (same circuitry except 12x7135) AND that stacking is reliable (a photo would be nice) AND shipping isn't a killer, I'm in for 50. Firmware is irrelevant as I'll reprogram them anyway.
Sounds good, I’ll ask them for a picture and see if it’s nanjg105c compatible. I told them I would have a bunch of questions for them tomorrow. Well you know how it is. Tomorrow for them is early AM for me. So very late night today actually.
If you really want to sell a boat load of these, have them add a 1uF capacitor and load off-time memory FW. That will allow folks that can't flash and or solder SMD's to truly upgrade their lights with both more power and much improved UI.
Funny, I actually meant TheStar, did not realise that JohnnyC's firmware had that similar name (but that's a nice UI too btw). I wouldn't mind the extra cap/off-time memory either :-)
Why not just have them build 1000 drivers using the BLF-17DD design files? Once you get up to that many 7135s the runtime before it falls out of regulation is so short there's not much benefit there to offset the higher parts count needed and the resulting bulky double-sided driver.
The current (probably final) revision of the 17DD works with only 6 components, 7 if you add an off-time capacitor. Gate & pulldown resistors have been eliminated since we changed the MCU's power supply circuit (it uses the same parts, just with the capacitor moved from after the D1 diode to before the diode).
Because it’s not regulated. Contrary to popular believe, not being regulated is actually a disadvantage. Once you install a BLF17DD you can’t just hand someone a light and assume they won’t acquire an IMR cell or 4.35v cell/charger down the road. You can blame whatever happens after that on them (whether it’s minor heat issues or poofing the LED), but it’s really on you (me/18sixfifty/whoever). If I build a light with 7135’s I can be pretty confident about what’s going to happen with all the batteries we know about right now. (Grrr)
But even when connected to a triple in parallel and the hottest single cell currently in existence, you won't get much more than 8.5-9A. The many-stacked-7135 drivers aren't really 'regulated' in the true sense, they just have a maximum limit. Their output starts falling (no longer 'regulated') after just a handful of minutes.
That’d drive up the costs because the company has a stockpile of NANJG drivers and all they need to do is flash the MCU and piggyback 4 more AMC7135 chips.
On the flip side, I don’t think it’s possible to blow an LED with direct drive to a 1x li-ion cell, whether it be IMR or charged to 4.35V. Plus, we won’t be randomly handing out flashlights with these drivers anyways.
