Ah you want super capacitor solar charger…doh
There is a guy that built a single cell sun jar, he was experimenting with a new chip solar charger that doubles the charge rate and controls the charging more efficiently
Oh cool…V3.0 released!
Organizing the projects into a table might be good. Those that haven’t been tested as much should be clearly marked. Making them easy to short through may also cause some people to jump in who haven’t really followed a project.
Op is organized this way only because this is how it got done. Not against tables other than I prefer it this way. I like to read and to me a table isn’t a paragraph but others might see things differently. I agree that other sources can be cheaper but they are usually temporary whereas digikey and mouser are stable. If RMM where to set up DIY kits I would certainly link to them if he wanted them linked here. I’m happy to entertain suggestions. Long term I think it should be set up for open access/editing but I’ve no idea how this is done and wanted both to see whether the idea would take root and to give it direction first.
I don’t think a resistor would work as voltage drop across it would fluctuate with mcu current which varies greatly due to the pwm signal output. The voltage divider used now is mostly independent of the rest of the circuit(except with the Zener mod) so current through it drops evenly with cell depletion. I could have this all wrong though. Maybe different values on the voltage divider with independent access to raw input voltage and separate voltage control for mcu would do it. Right now the voltage divider is in series with the Zener mod so the current through it is pretty much fixed, disabling voltage sensing.
You're on the right track, but you forgot that the zener diode will prevent the MCU from ever seeing anything above 4.3v (at least the one most of us are using.) The fix (at least in theory) is to bypass the zener diode/capacitor for the voltage readout circuit. Matt has a few untested boards on Oshpark that implement this idea. In theory, it should work, but I am no EE and am just working with the limited information and knowledge I've got.
Above is basically what Matt has done with the modified BLF boards.
Basically just cutting the trace that runs after the diode and zener mode that feeds the voltage sense resistors and then running the batt+ line directly to them. The zener mod makes it so that they never see over 4.3v, rendering the voltage readings useless.
Here’s the math I used to find the resistance values:
(2.9 v per cell, voltage taken directly from bat + w/ no diode)
(5.8 * 4700 * 255) / (19100 +4700)*(1.1) = Approx. 265.
6,951,300 / 26,180 = Approx. 265. Too high! (255 max range)
Now we look at replacing the 4700 resistor with a 19100 resistor:
(5.8 * 4700 * 255) / (19100 + 19100)*(1.1) = Approx. 165.
6,951,300 / 42,020 = Approx. 165. Bingo!
So, with two 19.1K resistors in the circuit it should work, I'll let ya know when I find out.
Also, regarding kits: I have one available for the BLF17DD / BLF20DD and will be expanding it soon to offer unflashed MCUs, the smaller FETs, kits without FETs, etc. to meet everyone's needs at a fair price. I have a large run of Oshpark boards coming so that I will be able to offer them at the same price as Oshpark; I don't want to charge more for something you could get somewhere else with free shipping. The current kit gives you a preflashed MCU and all of the components, including the FET, gold plated spring, copper braid, 22 AWG wires, etc. you need to make one run.
Looks great, ordered! I have one [simple] question I’m embarrassed to ask; what way does D1 go?
There is a line on the diode that corresponds to a line in the silkscreen.
Added your kits Rich.
I'm actually not sure yet, but it's slightly larger than the Quark CR2 which was only 14mm. I'm going to guess between 15mm and 16mm. I just haven't had time to take it apart yet sorry. I considered power cycle, but I'm not sure how easy that will be to use, but I guess it depends on the UI.
Don't be embarrassed....it's missing. #@$%! There is always something lol.
I'll add the missing mark, but for your records (as you've already ordered), D1 should be pointing to the right if you use this image as reference. So if the mark was there, it'd be on the right hand most pad for D1.
- Matt
ah yes, you are right, I forgot about that 
Anyway, it look like things are progressing, I just have to wait a bit more for 2 cell, Attiny13 with V sensing and ~4A 
And once again, thanks to all involved in making progress in this thread 
I only wish If I had more time to devote to all this interesting projects…
I may regret this but here goes...
I don't like DD drivers. NFI why, but I don't like the idea of the LED dimming as the light dims. So I prefer constant current drivers. Of course a constant current driver will probably never be as bright as a DD driver, but that is a small trade off to make in my opinion. There are of course no CC drivers that can even come close to the power these DD drivers can deliver.
Why don't we work on something like this: http://www.instructables.com/id/Circuits-for-using-High-Power-LED-s/?ALLSTEPS#step8
Use the FETs we have included in the BLF15DD and BLF17DD drivers. Maximum drive current can be set on a case by case basis (R3). All we really need to do is find a suitable NPN transistor and we're away.
I know all it's really doing is burning off as heat what could otherwise be light so I'm not sure how useful it would be. That being said 7135 boards are stupidly popular but are effectively the same thing, why not? With the BLF15DD FET, and a suitably small transistor we could be getting a constant 4A out of 10/12mm drivers without resorting to chip stacking.
Thoughts?
You need to verify which end of the diode is which based on the datasheet for the exact diode you use. Markings aren't consistent between different manufacturers or diode types.
For instance the ones I use and in the Digikey lists only has a '41' marking on it, no line anywhere. The end with the '4' goes toward B+.
What if you set the ATtiny max PWM instead of 255 (full on) to say (215) 85%~ on…this way it always has a PWM regulated output?
Well, PWMing a MOSFET will still dim at a given PWM level as the input voltage drops...but what if you started at around 85%, then used the voltage sense bank to tell the MCU to automatically adjust the PWM tables (feedback loop) in all of the modes to compensate as the voltage drops? This is actually probably far too complex to be reasonable because every emitter, host, batteries, etc. would need their own set of tables. Too much work for me, but it's an interesting idea.
It still won't be regulated, it will be a percentage of the direct drive current and will still vary depending on battery type & input voltage. 85% of a 20R is different from 85% of a 28A, as is 85% of 4.1v and 85% of 3.6v. Using the PWM to limit the output gives you 100% current for 85% of the time, and zero current for 15% of the time.
Yeah my link is by far the simplest way to do it.
I agree with you about DD but it is a solution that is very space conservative. Your buck driver is great but has a lower current limit and takes more space. Personally I think 4A is plenty in a host that has limited space. It likely also has limited heat sinking as well and a larger host with better heat sinking probably also has room for a larger, more powerful buck driver. How about matching needs with capabilities? What about designing a slightly larger, higher output, buck driver?
I'm not sure there's enough of an advantage there to justify a more complex board, just for a regulated output alone. My primary interest is having an easily flashable controller, anything else is secondary. It'd only be worth it if you go all the way and do something that's non-PWM and still easy to reprogram which I'm not sure is possible. Since I can't do much of anything with the code myself I can only work with hardware that has a large enough pool of smart people coming up with interesting firmware. I can hack away at existing code but anything from scratch is more than my ol' brain can handle.
I agree Matt, regulation at a bit more moderate output would be a key factor for a very nice work light. Many times there are occasions where a constant output is needed for nearly an entire work shift, and there’s not much that will supply that with reasonable output.
I’m having fun building the drag racers, and I carry extra cells for my Ti light in case full output is needed, but that’s a band-aid to the problem instead of a solution.
I’ve even considered a small wired host, with a belt mounted battery pack. Heat, of course, is a constant when it comes to alleviating the issues. Would be nice to get a new generation of emitters with much more efficiency. 
Not over my need for the Ultimate drag racer just yet, but learning to appreciate the value of a more steady longer lasting output….
Dunno, you'd have to work out all the variables, like how long it runs vs. how long the charge time is, and so on. But I don't think capacitors would be a good choice for that, they're best at really awesomely huge loads for very short periods of time (like, a starting battery). You'd probably be better off with a small lead acid batt, like a 7Ah or smaller, from a UPS. SLA life sucks though, they just wear out too quick. A bank of moderately affordable LiIons (new old stock laptop batteries are great for that kind of thing) would have better density and longer life too.
Yeah, it only needs to be able to do a relatively small charge rate, just enough to counter any parasitic drain from the car and self-discharge in the caps themselves. It doesn't need to recharge from a heavily discharged state or anything like that.
As far as I understand it before really getting started, a switching controller will make the panel design easier. It would technically work with nothing between the cells & caps except a diode, but the panel layout has to be such that the max no-load voltage is within the safe range for the capacitor array, and that's not always possible. A higher voltage panel would also allow higher charge rate, since the panel gets disconnected from the battery when the voltage reaches the cutoff point.
With DIY you don’t need a reason or excuse or permission so I’m game if you are. Let’s build it. If it works and does something other boards can’t then it was worth it.