Except that the X5 doesn’t have a boost driver, so we still need that, too.
I have one in 2AA, but have been unable to purchase a single AA tube for it. I would love to find a single AA tube, used with a 14500 it would be a great little light. They are very well made, I like it better than the Solarforce L2r.
The other issue is that there do not seem to be any really good driver out there. I have a couple of older AA lights that could use a bump, but really need a driver with the new LED.
Matt
You said it. I would join the band wagon for a 2AA light, but an Attiny13A driver for a 1AA light would blow my mind.
There are more new members every day and some have EE knowledge so keep asking and sooner or later it will happen. Remember it took several years to get a programmable 17mm buck driver. If someone were paid to do it then it would get done sooner but as it is you have to wait for someone with a hobby interest to take it up and run with it. Testing, boards, and parts costs money so setting up a way for people to pitch in could accelerate the process but you first need to find your EE. Then there’s always the DIY method and you might find lots of help where you might not find volunteers.
One of those (kinda) new members here, although I’m not an EE. Sorry. Just wondering, how did the other drivers get designed? Obviously some smart folks have put in the effort here before, folks like Wight or Alex Wells, maybe. Those are names I’ve seen but know nothing about. Are they not up to this job? Do they just not want it? Have they left the community?
For us new guys, what’s the story there?
Ah heck, I just can't resist VoB :)
- 14mm diameter
- ATtiny85V microcontroller
- 0.7 - 3.0V input voltage
- 0.4 - 420 mA LED drive current (XP-G2; 1x Eneloop)
- Full current-control through all drive currents
- Off-time capacitor mode switching
- End-user programmable via custom firmware:
- 1-4 output levels
- Brightness of each output level fully adjustable
- Toggle between classic mode memory, hybrid or no mode memory
- 2xAA/1xCR123 not yet tested but should work with higher LED currents possible/likely
- Li-ion likely to run direct-drive (at least until the sense resistor burns out :P)
Nice. Where do I order from?
!!!
Welcome to BLF :-) That is one heck of a first post calipsoii ! 
That sounds and looks like a great driver, both the hardware and the software! So what's the story? You designed it and kept it a secret till now? 
edit: I should check CPF a bit more often ;-) http://www.candlepowerforums.com/vb/showthread.php?373167-Build-Log-The-1-5V-Project
That is a great build log, I'm enjoying reading it :-)
Hey calipsoii, thanks for sharing the teaser with us, now how about some details and a sales thread? 
Some things I’d like to know:
1) With your circuit design, is the 420mA output the most that a single Eneloop can do, or is it set at that level for safety, or efficiency, or parts are cheap, or what?
2) You said maybe 2xAA or 1xCR123A might do higher current. Any way we could get testing done on that? If you don’t have the time or the means to test, could you send a sample to one of the guys here that do testing, like djozz?
3) Will you be selling this driver? And, if so, when, where, and how much?
4) Is the design open? Will you be sharing your board layout? Could someone here legally make PCB changes in order to improve the design, make different sized boards, or change components?
Wait, is that 2 different drivers? or one 2-sided driver without a spring contact?
Two-sided: with current control on all levels, an Attiny85 and a torroid there's not an option for a 14mm one-sided board.
I just read some parts of the build thread. Thanks to djozz for the link! I’m really interested in this driver!
Great looking driver you created there calipsoii. Checked out your build log on CPF. You really do some first class work.
I don't have time to look at the data sheet for the Linear's LTC352 today. Did you consider the PAM2803 when choosing a boost converter? Supposed to be designed for a 1 amp load.
PAM2803 is too large for a 14mm board. The boost converter stats aren’t important anyways since it’s only jumpstarting the ATtiny which then takes over switching. The electrical path components are all rated to 3A (well, the sense resistor is only rated to 0.25W) so this circuit has a higher ripple ceiling than one using the PAM2803.
I scavenged an AA boost driver from a cheap zoomie and it uses a ME2180 DC-DC regulated output voltage chip that might be of interest.
A picture of the driver can be found in my build post.
The datasheet can be found here.
Here is the website for the company that makes this chip.
http://microne.com.cn
Many of their step-up DC-DC converters look suitable for an LED driver because they only need two external parts: an inductor and a capacitor. Since these are voltage regulators and not current regulators like the PAM2803, they are better suited to power the ATtiny.
I have a hard time finding a source for the boost driver that uses the ME2180. Has anyone else come across a similar driver?
1) The 4R7 inductor is probably determining peak current. A 2R2 (used on most boost boards) would likely give higher current at the cost of higher ripple. ~350-400mA was always the target: decent brightness with a decent runtime. If I need brighter I'll reach for a li-ion light. :)
2) Did a couple tests! This is through 15" long 22awg alligator leads connected to the battery with neodymium magnets so total circuit resistance is extremely high. Board was floating in air with no heatsinking and overheated before I could finish the 2x Eneloop run.
3) and 4) are undecided :)
They’ve all left for reasons of their own. There’s a tendency to keep asking more and more of those few who can to develop drivers and such for the rest of us. If I had to guess I would put it down to burn out and suggest we give thanks for what they have left us and move on.
Some of the early driver designs centered around finding ways to use the attiny13A micro controller piggybacked onto an existing board since there was an already growing group of software savvy engineers and hobbyists that were fooling with it and it was in common use on certain linear drivers. Some designs began with minor modifications to allow two cell (6V) input to an ostensibly one cell driver to allow it to control a 6V led, specifically the MTG. Figuring out how to piggyback the attiny13A onto an FET based driver opened up another avenue of designs and during all of this more and more drivers and other items were being designed using Eagle CAD software and shared on the forum. Some boards were designed to use pic platform mcu’s but the vast majority was based on the 13A. It is now being supplanted by it’s more powerful cousins in the newest work and smaller cousin the mmu on the smaller board designs but for now remains the most commonly used mcu for DIY. Hope that answered your question to some extent.
Wow! That 2x Eneloop line! Let me know if 3) and/or 4) become decided.