D.I.Y. Illuminated tailcap

Lower, got it. I wish I understood why I need a lower one but honestly I don’t. That’s okay though, I’m learning and I really appreciate you holding my hand through this :slight_smile:

I’ll report back later this evening after some experimentation.

I’m no electrical engineer, but the way I think of it is that to make the driver work normally, we have to make it oblivious that the illuminated tailcap exists. So we want all of the power the tailcap needs to bypass the driver and just come from the bleeder. If the driver is still messing up, it must mean that not enough power is coming from the bleeder, and it must be pulling some through the driver, keeping the mcu awake or keeping the OTC charged instead of draining like it should. So we need to lower the resistance to make more go through the bleeder.

I hate to muddy up this thread with a post that doesn’t contribute in any way… but I just gotta say, holy wow have I learned a lot in the last two pages!

Ordering parts Trying to order parts as we speak, I wanna play too! Thank you guys!

Okay, I got home shortly ago and make some quick progress. I dropped in several different sizes of bleeder resistor, just working my way down through the assortment I own. I got to 220 ohms but then my driver started working properly again! Tailcap draw is way up, 944 µA now, and the tailcap brightness is much higher too, so I may crack the tail back open and increase those resistors a bit. If I understand things that shouldn’t affect the driver though, so hopefully I won’t have to touch it again!

We’re getting there! Thanks a lot PD and others!

That is kind of high. And current going through the driver may be keeping the OTC from draining or could be keeping the SRAM from decaying to its powered-off state. Either of these can break the interface on an offtime-based driver.

I’m not entirely sure how guppydrv (guppy2drv) works, but DrJones’ page says it uses off-time memory. So, it probably needs a bleeder to let power through without keeping the rest of the driver lit up.

It seems you’ve already figured this out though, and are in the process of fine-tuning it. I can at least say that the lower the tailcap power, the less the offtime-disrupting effect. Also, the less-resistive the bleeder, the lower the offtime-disrupting effect. However, less-resistive bleeders also make the normal running modes less efficient, IIRC. So, it should probably be the highest value which still works.

Hehh. I would avoid getting too excited. Very little of that has been fully thought out. :-/

The 84mA number was a mistake - turned out to be 84uA, much lower. The bleeder isn’t hurting the efficiency in regular modes very much. For example:
4.2v / 200ohms = 0.021amps

So in the case of a 200 ohm resistor: with the light turned on we’d be wasting 21mA constantly through the resistor if the battery maintained 4.2v (which it doesn’t). That’s less than 1/10w, which is probably quite a bit for a little 0805 resistor… but definitely not too much I think.

Your observations seem accurate. This is a pretty weird thing to be doing to the overall circuit; I think that your assessment is close enough to the truth to run with.

I’ve had that issue, with pot boards, and once you dial in the leds so the modes/memory works right, you’ll run into issues with the light after it heats up, even just warms up. I was using guppy but decided to build a few blf a6(+7135) drivers for a change, and the same happens. Also attempted lighted tailcaps the same way you did with a 3.04A qlite, same outcome as you experienced, I pulled the lighted parts and gave up on that because it was just to try it. Ive put the bleeder on the +spring to ground, and on top of the OTC cap, from otc to gnd, I have no idea what could be going on here.

Could the bleeder resistor and the led resistor in series acting as a voltage divider be something we aren’t taking into account?

When the light is off we are putting +470/560/200 ohms worth of positive power into the host, when we turn the light on we overpower that with - power making the host more neutral than the bleeder resistors bleed off, killing the tail leds and not losing any power at all from the main led(s).

Maybe we need a diode in series with the bleeder? Or is there already one lined up with c1? Or maybe there should be one in line with the OTC cap to keep power from backfeeding when the light is off? It seems like power is backfeeding into the mcu or a component causing it to get heated and mess with the modes.

I measured the voltage at the tail leds set up with blf a6/mtn17ddM/470ohm resistor/.67mA draw and it was only 2.1 V, same light before without any lighted tailcap parts it was 4.2v & .06mA draw.

I really don’t know, I’m just spitballing what I’ve done hoping someone with actual knowledge can say “hey dummy do this…” I’ll try about anything.

I got the alu X5 in today (what a great little light! The bling of the CU/SS set was nice, but IMHO this little black X5 is what this GB was all about) and since I received the 6-led dumb board as well this week, I could not resist instantly modding it. Pulled out the blue led tail assembly and build a new one from the 6-led ring, a new small Omten switch and a BLF-A6 switch board that had come spare. The resistor before the single blue led in the stock light was 2K Ohm, so I did not dare to go different in fear of messing up the driver UI: 3x 6.8K Ohm parallel for the new board, 6 orange leds. It all worked very well and I also reflowed another led: XP-L Hi 3000K 80CRI for added cuteness. :-)

Happy camper. Thanks PD68 for another winning design!

Looks very nice djozz.

I’m interested in installing a Rev5.1b in the S2+ and Solarforce tailcaps. Will that 19mm PCB fit in either?

But the extra 21mA is pretty huge in some modes.

On an unmodded Kronos X5, I measured 14mA on moon. On an unmodded BLF A6, I measured 3mA. It’s a pretty big jump from 3mA to 14mA. FWIW, neither measurement included the tailcap.

Anyway, it matters to me when the moon runtime drops from ~42 days to ~9 days.

But then, on the lights which use moon a lot, I don’t install bleeders. :stuck_out_tongue_winking_eye:

Very nice.

Also, good to know the board actually fits those small boots. Might be a good idea to flatten the +/- solder blobs though, to help with waterproofing. (more even pressure)

I wouldn’t worry about side effects of using higher-value resistors. It’s not like the driver was actually calibrated for this in the first place… I didn’t even get the opportunity to calibrate it this time, though I did try regardless. The hardware got changed like four times after sending the code they used, and each change invalidated the calibration. In the end, it was way out of spec. You might actually be able to improve it a bit with higher-value resistors.

I did, after taking the picture I filed them flat :slight_smile:
As for the higher resistor, I like the current brightness, but will measure the current current to see how long the battery will last.

I measured the current through the tail when switched-off on an almost (4.14V) dark purple Efest battery, it was 0.855mA. So about a month to drain the battery if this current will be constant, which it will not be, so perhaps 1.5 months? Good enough for all the uses I have for this flashlight, keeping it as an emergency back-up light in a glove box will not be one of them.

Whenever I change my mind and after 30 years without it after all decide to obtain my driver’s licence and then buy that matt black DB9 in which glove-box this gem of a flashlight will have its worthy residence, I can always unscrew the tail a quarter turn for lock-out of the tail-leds.

With that high of draw on a Rev5, it must be pretty bright, eh? Or maybe the LEDs are a little inefficient, it doesn’t look blindingly bright in your pictures.

emarkd, did you ever get yours dialed-in?

It appears about as bright as that one blue led of the stock tail. In other words, in the dark it is a good bright indication light visible from quite a distance but not disturbingly bright (the orange colour helps with that), it can be much dimmer and still be useful.

Here it is next to my keychain fob, that has three big bright trits (way brighter than the small ones in my Ti Reylight):

Hi guys!

@djozz

very nice X5 you got there!

@wight

like you, I want to upgrade some Solarforce tailcaps. I can’t help you with the S2+, maybe someone else can chime in here. I opened up the three Solarforce tailcaps I’ve got and found that they differ. My L2T is one of the first batches, it has the glossy type ano, smaller tailcap boot and smaller PCB in there. It’s about 19mm, maybe the 19mm LTC-board would need a little filing. My L2N and L2M have similar tailcap internals, same boot size, about 21mm PCB. Notice, all these lights have the forward clicky installed.

Some pics:





Obviously I converted the L2M to reverse clicky

I’m very sad since I found out the L2N is discontinued. Well, at least I got one. But it’s such a nice light to play Solarforce lego with… :-((
Pick one up if you find one for a acceptable price…
Also, the spare switches are gone from the Solarforce site.

@PD68

What are the specs I should watch out for, when searching for some nice LEDs for LTC purposes? 0603? current rating?

The L2n is discontinued?? That was my favorite one… guess I won’t be selling the one I have left.

I have had LTC in multiple solarforce tailcaps, but none with the Rev5/6 ring design yet. Previously it was hard to get the forward-clickies to fit on the board, but that shouldn’t be an issue with the ring design as long as the ring fits over the post on the forward clicky

I have successfully used a Rev4 board in an S2+, and at least one other person (emarkd) has used the Rev5.1 board, I’m guessing the smaller size.

There are a couple boards that require 0603 size, but most use 0805 size pads. You can use 0603 on 0805 pads though, which is what Djozz has done.

Yes, L2N is discontinued, I asked them… Couldn’t resist and picked up a spare L2N from KD, I hope it’s genuine.

Thanks for answering my question about the LEDs. What kind of current rating would you chose? I’m currently packing some stuff into my mouser cart and can’t decide what LEDs to get. For example, blue Osram 0603 LEDs, which one should I order, 2mA, 10mA, 20mA rated? Or even higher?

I buy my LEDs from Lighthouse (linked in OP). For example the blue 0805 is rated for 20ma and luminosity of 120mcd. The red is rated 20ma and 180mcd. I think what you want is the highest mcd possible.

maybe the lower power rated will be more efficient at the lower power we give them (less than 1ma)?

Lighthouse also has theirs rated as “LED Brightness Class: Super/Ultra/Extreme”, but I don’t know if that’s actually some kind of standard or if it’s just marketing hype.

Thanks PD!

That’s an interesting point. I will be digging through some datasheets and eventually someday I’ll complete that mouser order… :smiley: