Musings on the FET+1 and lighted tailcap.


I’ll be following along here, though I can’t contribute much. I have zero training in this type of thing, just a little first-hand experience and reading here on BLF. I will say I think you have the tail LED polarity reversed in your diagram, though that probably doesn’t make any difference.

I haven’t had much of an issue with usability on my drivers, except when trying to use TK’s “medium press” with a lighted tailcap. That said, I’d be happy to start using a “better” method as long as it doesn’t increase part count or driver cost dramatically. I just need to be told what to do, and shown that it will make a noticeable improvement.

Mild derision first:

  • Love the copyright statement on a circuit diagram for an existing circuit.
  • The C1 overvoltage problem is not just “plausible” - it’s fully explored. See comfychair’s thread. EDIT: make sure you have JS enabled to get images…
  • Feel free to try and fit in the PFET onto the board. (actually I might be setting myself up to have the joke be on me here, see below)

I like this suggestion.

In my opinion one of the most difficult things about making improvements to what we have now is that we really don’t understand this. Nobody here knows exactly how this is happening and as you can see the effect is pretty strong!

RE: a few other things

  • IMO there is no reason why C1 can’t “decouple” a 2-component circuit (MCU & diode) just like it could “decouple” 1 component (MCU). That said, if you read comfychair’s thread you’ll see that while I suggested the solution it wasn’t my favorite idea at the time. It’s still not. That’s why I recently shared a fresh layout: [WIP] 17mm DD+single-7135 driver / single sided / Dual-PWM - #326 by wight
  • I think the point I made above about not understanding how the OTC gets drained in the first place applies to this suggestion of yours. “Alternatively consider relocating the OTC to connect between R2 and ground, and perhaps adjust R1 and R2 upwards. This would free up another MCU pin for future use, and could give both on-time and off time measurements with the single component. Off time determined by R2, on time determined largely by R1 . Moving the OTC here might also smooth any noise at this point, perhaps improving the stability of battery voltage measurement.” Whatever interaction is pulling down the OTC now would still be happening with that setup… plus the fact that the OTC would start charging as soon as power was re-applied to the driver.
  • IMO we are guaranteed to have some poor interactions with this many bad/weird things being done. I’m always game to try and minimize/eliminate poor interactions and bad behaviors where possible.

:) I’m glad you started a conversation addressing (a) the OTC and (b) the tail LEDs. :)

Finally… re your “wacky idea”: … hmm. It’s certainly something we could play with, but it sounds like yet another calibration to me. We’d need to have an established value / table in the MCU since checking at startup could give us the value from a hot LED.

I was forgetting that in this case the labelled “GND” is not actually a constant Neg- connection, but just the battery tube of the flashlight (where the tail LED connects to Batt+). So you do have it right in your diagram. I’m a bit rusty on these diagram things since I never use them.

What would be some good resistor values to try for R1 and R2, assuming about 0.08ma needed for the tail led at 4.2v?

Oh, I like where this discussion is going. I have been going back and studying basic electronics, have not had to since university in about 1985. I like the PFET idea, but I am not sure there will be enough room on a 17mm driver. Please continue to explore away. An yes, the lighted tailcap turns the mundane into a challenge for each different driver!

You lost me at “First Draft”…

But I can hand solder, order parts, and I’m more than willing to try suggestions for improvements if you can lay it out in laymans (better yet, full Rtard) terms.

It makes sense. So it really wouldn’t make a difference if we changed the voltage divider values vs just using a higher resistance bleeder?

The bleeder is easier for me because most of my drivers have a spot for it already, and I’ve already stocked up on 19.1k and 4.7k for R1 and R2

It very well could be 22k on the X6 driver, I know it was on the A6. But 22k has been shown to cause more OTC issues than using 19.1k. Maybe you can figure that issue out too. (specifically the OTC timing inconsistencies that happen when the light gets very hot)

Great here!! You certainly got the current driver design doc'd correctly, but I probably can't contribute much more.

But (a Big Butt), I feel the 13A is dead (D-E-D dead), and all our attention/brains (ok - your brains) should be paid to the 25 and 85, preferably 85, and help to solve the problems we (well seems like mostly me) is having with the 85.

I posted a ways back that the BLF X5 and X6 (v2) are using a 12 uF cap, not a 10 uF cap. Did anyone else confirm this? It's easy to do, and if so, it confirms why I needed to go to 20 uF (2 10 uF caps) to get the dang board to work reliably with a 25 or 85. From my inside info sources (ok - Dale and TK), it was mentioned Manker was having the identical problems with the new 25 based FET+1 driver as I've had, and no one knew what they did to fix it -- now we know --- they changed the cap out from 10 uF to 12 uF.

I admit I can't follow all this, but I think y'all (southern talk from a northerner ) are headed in the right direction, but keep in mind the target platform - 17mm round drivers, and at some point the theoretical has to become practical.

I’m all for progress, but I have yet to use anything but a 13. Partially because I’m lazy, but mostly because I haven’t had any reason to. It’s just preference thing. The only FW I want but don’t have is a simple dual-switch variety, and it isn’t yet available on the 25 or 85 either.

Long live the 13!

Unless of course you are dealing with 6 different led’s of 3 different colors set by 2-3 different resistors to balance and a master potentiometer to control brightness.

I do not understand what the “dead easy to work out what is inside a tailcap” was in response to.

Yeah, I’m still just missing what your post is in response to. As far as I can see your PS/hint has nothing to do with the quoted text. Were you responding to some out-of-band inquiry? Or maybe your PS is intended as a continuation of Post #9 and you just quoted the wrong thing?

Just seems like I’m out of sync with the discussion. I still don’t get why we’re measuring the contents of an illuminated tailcap. PD did not ask about it as far as I can see. All I’m looking for is a handle on the conversation.

EDIT: Maybe I’m being too pedantic and should just watch from a distance for a while?

Suit yourself RE: TTFN / Your thread / etc.

If my rather lengthy reply made it look like I wasn’t interested in the plethora of subjects your massive OP brought up then I don’t know what I did wrong. I even put in two smiley faces just for you!

I can’t tell if you are being facetious with “A bit of a stunner, that.”. Remember, there is no inflection in forum text. If you are honestly surprised it doesn’t hurt to say so specifically. Right now I’m reading some spite in your posts, but I can’t tell if it’s all in my head or what. If you mean for the spite/vitriol to leak out then by all means be specific about that too… if only to avoid hearing questions like this from me.

Personally I was quite surprised when I realized how limited the understanding of drivers and general electrical stuff was in a flashlight forum. One might think that this place would be full of EE’s. It’s not, and those EE’s that are here don’t want to fool with it or are outside their specialty.

Dude…. Everyone knows it’s magic.

I read this as just being straight up rude. It is your thread, but it’s everyone’s forum.

Unless I misunderstood how comments like this are helpful in some way.

Man PD, you are keeping us in the dark ages . I much prefer e-switch lights - navigating more that 3 modes on a clicky is a PIA, and no matter what many say, just can't get any of my clicky lights with Bistro or the BLF A6 firmware to consistently handle the long hold correctly - every one I have varies with temperature, whether it's a BLF A6 or the BLF X5/X6 v2's. e-switch timing is precise and reliable every time on a 85, and most side switch's, even on $10 budget lights, are much easier to operate than a clicky. Just about any feature in the $40-$120 brand name lights can be implemented.

Do you mean dual e-switch's or e-switch and clicky switch combo? Probably should take this to pm's or another thread, but I have e-switch and clicky support in one firmware version w/limitations. If you want 2 e-switch's, should be easy to add to what I have. Pm me or refer me to another thread - seem to recall you raising this issue, but all depends on exactly what features you want from the dual switch's, and if I can get my hands on a light/host that has that capability. I'm using my Narsil firmware now with a LED ON/OFF control on pin #3 for a locator light and low battery indicator, but could also configure it for a 2nd switch easy enough -- devil is in the details of the required functionality. I'm very interested in a dual e-switch but don't have one, or know of one worth buying that I know is modd-able, budget preferably. I'd love to be able to add my own e-switch's but lack the knowledge or maybe skill set/tools to accomplish that. I've seen lots of mods posted with adding switch's but not enough details, or lacking a lathe or mill for me to reproduce it. Ideally 2 e-switch's is a great setup, if you want one as ON/OFF, or one dedicate to blinkies/bike modes, etc. - lots of possibilities.

Beware the wrath of wizards, I believe they say?

But — in every field I’ve ever fiddled in — amateurs and hobbyists make Rube Goldberg contraptions, trying for a desired result.
There’s often a useful lesson in observing the result (because a contraption says what’s wanted clearer than any wishlist can)

Once the idea is clear, the expert might well start from scratch rather than fixing the contraption.

Sometimes, though, the contraption produces worthwhile surprises you wouldn’t imagine otherwise.

‘Do not meddle in the affairs of Wizards — It makes them soggy and hard to light.’

It seems that the kettle, upon being called black, has responded in turn. Unfortunately the pot wasn’t prepared for this, and has boiled over.