Looking good! Well done. 
why do you think a bleeder could help?
Theory: It sounds like whatever current your tailcap ia drawing, it’s more than the voltage divider resistors on the tailcap allow through, so the extra current is being pulled through the mcu itself, and it’s enough power to keep the mcu awake, and not thinking the switch was ever pushed. A bleeder should divert enough power past the mcu so that it functions normally.
Hi all!
I just received the new Astrolux S41, and for what i believe i have come across in the long thread dedicated to this light, someone mentioned the tailcap PCB is ready for lighting but not populated.
If someone managed to get it working, i would be happy to see a component list and little how-to… :innocent:
Thanks
Quick report - got a tailcap working perfectly in a Convoy S2+.
Notes:
- You want to order 0.8mm boards, not the standard 1.6mm. The normal ones are too thick and the tailcap assembly doesn’t sit right (too far from metal switch, too thick to tighten down the retaining ring more than a half turn).
- I used a 47 ohm bleeder on the driver and 400 ohms on the tailcap. You definitely want a 1:6 ratio or lower (going off what someone earlier in the thread pointed out - thank you!). The 450 ohm total is necessary to see the low-efficiency red LEDs through the rubber. This means the battery will drain in about two weeks but I’m ok with this, can always loosen the tailcap to lock out.
- On the MTN-17ddm driver, you want to put the bleeder before the diode (NOT across the capacitor). I actually ended up using a 1/8 through-hole resistor and soldering it between the spring and the ground ring since my SMD resistors were not quite long enough for that and the 1/8w doesn’t interfere with anything. Modes work fine for me.
I’ll post some pictures when I have some time to get the camera out. Thank you for all the hard work PD86 and everyone else!
Hi! I want to do this lighted tailcap in an X6. I found that I need the v5.1 version for 14mm boot, and I ordered high amp tailcap board too because it will be a triple XP-L setup. But what size leds I need under the boot to fit nicely? 0805 or 0603? Everything else is clear for me. Thank you!
I have made them with both sizes of LED’s. I think technically it is designed for 0805’s. If nobody answers with authority I can measure the pads tonight. Might check the info in the thread, should be in the section that introduces the board.
All boards are designed for 0805 except the crazy looking one
Thank you! Now I will order some 0805 leds for it.
A verson with 1206 LED pads was requested so here it is:
Yesterday I made a lighted tailcap on a Convoy S2+ shorty that has a triple Nichia 219B red emitters, I used red leds for the tail of course. It proved useful today already when supervising a photograpy school project in a dark room (a blinded class room), I could find the red flashlight back immediately all the time :-) .
What is new (to me at least) is that I did not use a transparant tailcap but a red one, it is fairly translucent for red light. Here it is by daylight:
parts list:
-Qlite driver from intl-outdoor link
-0805-size bleeder resistor of 470 Ohm on the Qlite (without the resistor everything goes wrong with the UI)
-PD68's dumb 16mm tail washer circuit board with 3 pairs of 2, makes 6 led pads link
-3x 5.7 kOhm 0805-size resistors before the 3 strings of 2 leds
-6x 0603 size red leds from Fasttech link
-16mm red tailcap from kaidomain link
With these resistor values and leds, the Qlite's behaviour is entirely normal. The lighted tail draws 1.02 mA, so will empty the Sanyo 18500 cell in between two and three months. The light is clearly visible in the dark at a distance without being too present, a good locator brightness. The entire tailcap is very evenly illuminated, thanks to the red tailcap that diffuses the light more than the transparant one.
Crazy idea, would it be possible to add a photoelectric sensor in the tail, something like this ?
Run time could possibly be extended as then we’d only need to run the tail LEDs when ambient light drops below a certain level. Just a thought :).
That would be interesting, certainly. Other than the problem of available space, the cost is prohibitive. We would have to buy 3,000 of them.
Well that one is available from Mouser in qty 1 @ €1.7, but there could be better/simpler options for the same task.
Exactly - I’m almost certain that’s not the perfect sensor to use, but the idea may have merit for someone more clever on this than me.
Looks awesome djozz! What firmware are you using on the driver and where does the bleeder go? Did you put the bleeder on C1?
I first tried it without bleeder and that messed up the user interface bigtime. I had soldered the driver in thoroughly to heatsink the 7135 chips (they have to burn up almost 2V overhead voltage in red led flashlights), so the bleeder on the led side of the driver was no option anymore. I had soldered the 4th star of the (stock firmware) Qlite to ground (mode group 4) and that trace goes quite far to the middle, with scratching off some solder mask I could solder a 0805 resistor from that trace to the central batt+ pad
I’ve ran into this project in a thread here, and it looks like it could be adapted for tailcap. In a nutshell, guy is running Luxeon Z from CR2032 with more than a year of runtime (26.3 μA draw), for the same reasons as us. I will try to make more sense of it, all of the source is available.
That’s an interesting idea. How would a low sensor reading trigger LED activation?
I’m wondering because an attiny13 could do it, but… the ADC parts of the chip use more power than simply leaving the LED on all the time at a low level. I guess it could keep everything off most of the time though, waking up once per second to take a reading and then go back to sleep.
I wonder if it’s possible to get the light sensor and tail LEDs in different frequencies so the LEDs won’t make it think it’s in a bright place.
Where did ya get that triple nichia 219b red emitter ? 
Havent even found triple red ….