GFS16 - Battery Indicator Tailcap Light & 1mR FET Tailswitch! (Rev B)

I really like the idea and the realization. Is it possibble to use 3x2 leds in 3 different colors for battery indicator? So that way maybe the switch can be green from 4,2V to 3,7V and orange from 3,7V to 3V and red under 3V?

With the current PCB design with no modifications, nope, but otherwise, yes very possible with different logic :).

Just asked because mostly I don’t discharge my batteries all way down and I will be happy with an earlier reminder.

If you read the original post, it's very easy to change the resistor divider values to suit your threshold voltages, and to also take into account any effect the series driver may have on the tailcap circuit. :)

edited with new info

Hi loneoceans, I’m planning on changing some values to the following…….

Vbatt x R4 / R3 + R4 =Vref

  • When battery level drops below 3.30V, Light up Red.————————- 3.3V x 360K/116K+360K = 2.5V

Vbatt x R6 / R5 + R6 =Vref

  • When battery level drops below ~3.00V, Turn off all LEDs.————— 3.0V x 360K/ 72K+360K =2.5V

Does this look right or should I decrease R4 to keep R3 at a lower value, sorry I looked up the datasheet and i couldn’t figure it out, what’s the rule when setting R2 value in a voltage divider circuit pertaining to the MAX9052BEUA+ datasheet? Also how do I factor in the voltage drop across the bleeder resistor to have compensated values for R3&R5….

Vi= Vt x Ri/ RB+R4+R3

So to set a trigger point of 3.3V, bleeder resistance= 100R, R4=360K, R3(compensated for bleeder)=115K. drop across R4= 2.5V

If any one need help setting different trigger point or led brightness, just PM me.

note a 100ohms bleeder is required… any higher value messes up the ref voltage for the compactor even if your calculations are correct, the video loneoceans posted implies that he’s hooking up a bench power supply without any resistance and I’m sure his driver must have some sort of resistance from positive to ground other wise it’ll short circuit, also not shown is an internal hysteresis of about 0.02V where both channels are trigger.

Also note that the 100R bleeder will be in parallel once the flashlight is turned on… consuming 37mah average, a minimum of 1/5 watt power rating required, 1/2 watt recommend for reliability to cope with the heat generated from the led, easily available in 0603 package.

I would add a 2.8V LDO to the tail cap, so you get constant brightness of the LEDs

R4/R6 should be large to have a low current over the voltage divider, so do not lower them
their current draw is even present when all LEDs are shut down

The reason the calculated brightness varies from calculated resistors id that the bleeder consumes voltage,
the drop differs with the current drawn from the LEDs, so constant current is at least easier to calculate

for example if your 6 LEDs draw together 0.3mA then on a regular 100 Ohms bleeder the voltage drops 30mV
Each LED draws 50uA
LEDs are powered over a 2.8V LDO so we do not have to mess with different currents on different battery voltages
LV 3.2V
SD 2.9V

on this example the current drawn of the 3 regular LEDs is removed when you hit 3.2V
so the current drops from 0.3 to 0.15mA also the drop on the bleeder jumps from 30mV to 15mV
so the comparator notices when 3 LEDs shut down a voltage rise of 15mV
The question is how big is the hysteresis of the comparator is this basically dictates how big the bleeder can be

With such a low hysteresis not really big LED currents or big bleeder of 1kOhm are possible without flickering

I looked in the datasheet of the comparator and adding 2 resistors per channel from Output adds hysteresis,
this would mean a bigger bleeder possible

I also noticed the comparator shows 130uA supply current which is more than many low brightness tail caps consume, so storing the light without checking tail color is not a good idea

Using an LDO would pretty much remove the need of a comparator with internal reference

Thanks Lexel, makes a lot of sense… I finally got it to work exactly as I want it, changed LEDs current & R3/R5 - R4/R6 values works good with no hysteresis noted, ill upload a video in a little bit. I have the compactor set for 3.7V LV and 3.5V SD.

old values showing hysteresis

new values, works better

Video shows off the switch design nicely. Good job

credit goes to loneoceans, its his design, hope he shows up again with his driver release!

I have made a new enhanced version

2x3 colors above first shut down
1 single LED for low battery signal

28uA 2.8V LDO up to 5.5V
dual pad for a 10V LDO for 2S lights
capacitors for LDO

2.8uA comparator without reference voltage, significantly reduced standby drain
the reference voltage comes from the 2.8V regulator and a voltage divider

added for each comparator a hysteresis voltage divider, so a larger bleeder like 470 or 1000 Ohms should be possible

17/20mm version

20mm with trimmers for LED brightness

Wow… awesome job Lexel! Did you just design this in the past 24 HR :+1: !? Well when is this gonna be available to buy?

Lexel is from a different planet somewhere distant in the universe, the life forms from over there work at a much faster clock rate than we humans.

Apologies if I missed it, but where did you get the stainless steel switch?

If your referring to the switch on page 1 of this thread…, they come with the flashlight, I think they are standard on the red and clear anodized s2+.

If, like me, you have one of the S2+ lights with a black metal switch you can buy the stainless metal switches from Banggood and the light will shine through the clear gasket (barely). So my Tan/Sand colored S2+ has a stainless metal switch I replaced and then added a lighted switch to.

Yeah that is the switch, do you have a link on Banggood?

https://www.banggood.com/Convoy-Metal-Clicky-Switch-DIY-For-S2-RedGreenBlueSilverSand-p-1161437.html?rmmds=search&cur_warehouse=CN

It only works on the ones already threaded for a metal switch though.

Thanks!