D.I.Y. Illuminated Tailcap - gChart Editions

Ah, I see.

That is really kind of the only thing I ended up liking about my Nitecore EC23. Because I had to always lock it out (or else it comes on in my pocket, virtually guaranteed), when I went to use it I'd get the battery voltage blinked out the second I tightened the tailcap. I'd love to have my lights tell me their battery status the second I turn them on, in a non-disruptive way. The best way, to me, seems to be to use switch lights. Since all my lights that I care about will start in low or moonlight, I wouldn't mind checking a light by briefly turning it on. For bonus points, the dumb-switch lights would still have absolutely 0 standby drain still, and I wouldn't have to worry about bleeder resistors on the drivers.

But yes, I can see what you mean from the design standpoint. There'd have to be another voltage drop in series with the main LED/driver, and that just wouldn't work. Alternately, there'd have to be a second current path, which leads to the FW3A all over again. Obviously this works fine with E-switch stuff because there's current already there, and some manufacturers already do something like it.

And no, I can't PM you, you're up to far too many awesome things. I'd have to go sell plasma or something. It seems like half the new interesting things I stumble across here have you involved :P

Still, the standby drain you mentioned is low enough, even at 0.35mA that's 1800 hours to drain my 650mAh Olight 16340s. None of my small lights have space for this in the switch anyway, so it would be going on a 3000/3500mAh 18650, which is around a year to drain. And that was the overly high standby drain you mentioned, with the wrong resistors.

Ugh, you monster. I'll have to remember to PM you in a few months. I just took two cats to the vet, and bought a new chair.

I have build colorful tailcaps, but they know only 3 stages
LVP -> OFF
LOW -> red
Normal -> all LEDs lit, red off

a battery indicator on light os can be realized with Aux LED board

I have made a solo board that has the 3 stages and an external switch board with different LED channels same as my Aux boards
its untested so far but should work in most e-switch lights
the base board is 14mm diameter and only 3mm tall

750 Ohms did it !
Memory is back and Biscotti is acting normally :slight_smile:
Thanks !

Hi everyone, great to see the development work here on taillight switches! Here's one I made about a year ago but haven't had time to write much about it. It's nothing special, just another LED taillight board. I designed it a while back with my GFS16 tailswitch system, built a few, but haven't completed writing all the features of it yet in firmware.

Here's how it looks like. I'll try to find a photo of it when I'm not traveling. On board there's a micro, as well as a battery measurement system.

Some other interesting features including a microphone (initially I designed this to be a 'clap' sensor, so you can change modes on the effects of the tailcap, as well as four RGB LEDs). So for example, you could have the micro do fun things like make the LEDs breathe, do a rainbow fading pattern, or just show the battery voltage in terms of colour. Or you could use it for other things as well, such as dancing LEDs to external music. Though in reality the most practical is to have the micro do some clever pattern on the LEDs then go to sleep to save battery power.

Very cool, LO. And welcome back!

I was wondering what that component was until I read your description (the microphone); interesting idea indeed.

I’m just now learning about these small RGB LEDs. Before them, all I had seen are the 5050 type. I may have to make a version of my “smart” (MCU-driven) tailcaps with some of those. Where did you source yours from?

What did you do for battery management? I was just using the internal 2.5V VREF in the Attiny416 for mine.

Thank you for your comments, and fantastic job with your taillight board! It seems much more practical than this one!

I'm currently using this LED: https://www.digikey.com/product-detail/en/broadcom-limited/HSMF-C114/516-1795-1-ND/1966501. No particular reason for this choice, the only reason is because I used it before in previous projects and I have a bunch of them lying around. I'm sure you can find many alternatives.

For battery voltage measurement, I simply use a regular LDO to generate a reference rail (and power the micro). The regular ADC on the micro (I use a ATtiny841, again arbitrary choice) seems good enough. By controlling each group-anode and each group-cathode, the micro has full control of each of the 12 LEDs using 7 pins.

Using microphone to make flashlight responds to audio signal is brilliant. :heart_eyes:

Finally got a chance to put more smart tailcaps together. I got request for a dual-color one… I think I like it!

Note: the tailcap does periodically re-check voltage on it’s own, but I have it set for hourly checks. So for testing purposes, I just disconnected/reconnect between voltage changes.

Smart tail cap by a smart guy. :+1:

:smiley: I just wish I had more time. I have so many ideas and designs rolling around in my head… not to mention all of the partial builds and custom drivers awaiting tweaked firmware. Slowly chipping away one 2AM night at a time.

While browsing for some ICs, I ran across these HM1160 chips which are bare-bones lithium battery life indicator ICs. They’re stupid simple and make it possible to create the battery monitoring tailcaps without a fancy teensy-tiny MCU and flashing firmware.

First off, the circuit boards using Oshpark’s new After Dark service. These are 15.8mm: small enough to fit a Tool AA (or SF14, etc) but should also fit Convoy S2/M1/C8/etc.

Here’s it in operation:

Public safety reminder… use protection. :person_facepalming:

I wanted to see if this chip could get away without a protection diode/FET. NOPE. I don’t want the voltage drop of a diode, so a V1.1 will be coming soon using a small FET for protection.

For the test shown above, I used a 680 Ohm resistor in series to mimic the bleeder resistor (which will have a marginal amount of voltage drop across it). With that in place, LEDs extinguished when falling below these approximate voltage levels:

  • 3.85 V
  • 3.63 V
  • 3.46 V
  • 2.98 V

Those aren’t where I’d necessarily pick them to be, be really it’s not too bad considering how convenient (and cheap!) these little HM1160 buggers are. To be continued… :smiley:

How does current consumption compare to MCU based tail cap?

I haven’t measured yet, but it should be minimal. The HM1160 datasheet lists “Quiescent operating current” as 8 uA.

With how I’ve got the MCU’s programmed, they have a very low current consumption as well (using 32768 Hz clock with only periodic ADC activity). Datasheet says active power consumption using the 32768 Hz clock is 10 uA for the clock itself (2.6 uA in idle). I think I’ve only got the ADC turning on for a moment of every hour, so that shouldn’t draw much current.

Good. :slight_smile: 4 years on 10440. Does it include LED current? I guess no?

Nope. That is dependent on what you use for the resistor. In the one I just put together for my Tool AA Ti, I used a 20,000 Ohm resistor. I haven’t measured actual current draw yet, but it should be roughly 0.08 mA (80 uA) on a full battery. Calculation goes a bit like this:

Battery voltage: 4.2 V
LED forward voltage: 2.6 V
Resistor used: 20,000 Ohm
(4.2 - 2.6) / 20000 = 0.00008 A = 0.08 mA = 80 uA

Battery drain (worst case scenario, current will actual drop as the battery voltage drops):
1000 mAh battery ÷ 24 hours/day ÷ 30 days/month ÷ 0.08 mA = 17 months standby time

It’s bright enough to be mildly visible in a decently lit indoor room. Illumination not visible outside during daylight. Relatively bright in a dark room.

Thats interesting gchart. Very simple, just wish it had voltage adjustment.
Couldn’t you add a small resistor in series to vcc to create a slight v-drop in the 10 to 20 mv range to lower the led voltage cut off points.
I haven’t a clue how you could increase them.
.
Are these pcb’s shared on Oshpark?

A series resistor can/does create a slight voltage drop, but probably not a enough to make it worthwhile to chase. Honestly, I think it’s just convenient & cheap enough to not worry about the small details. They’re directionally correct and informative.

Oshpark links:

  • Version 1.0, lacks reverse polarity protection but works fine otherwise (this is what is in my Tool AA)
  • Version 1.1, adds a small PFET for reverse polarity protection. (waiting these to arrive, should be here in a week or so)
  • Version 1.2, opened up the center hole a bit for better fitment (not yet tested)

I did notice that while installing v1.0 in my Tool that the center hole is slightly narrower than expected and I had to spend a minute lightly sanding the switch to allow the PCB to slide on it. The dimension I used was the same as with other boards, so perhaps it’s just board-to-board variation. Or perhaps it’s a process difference between the regular service and the After Dark service. If my v1.1 arrive with slightly narrow cutouts, I’ll likely open that hole up and call it v1.2.

Thanks for the links gchart. I got too many projects on going right now, but I’ll give it a try in a week or so. :+1:
Sending you a PM.

Sounds good. FYI, I just went ahead and made a v1.2 with a slightly larger center hole. That way if anybody cares to order some, that’s already taken care of just in case.