[Oshpark] HQ 15mm/17mm programmable boost driver with ATtiny13A

guys, thats me: :laughing:

.

@fixed it
That sounds promising.
I admit I understood only every second word of what you said, but i’ll dig into it.
I’m no firmware guy but I usually can adapt what others come up with.

.

@ImA4Wheelr
Here in EU these Coilcraft inductors even increased in price about 70% in the last 12 months. They were expensive back then, they need to be weighed in gold by now.

Any other inductor with the µH-specs and size should do, I tried several (look at my 3rd pic in post#4).
The other (non-Coilcraft 2.2µH) inductors have mostly higher DC-resistance and thus are less efficient, like, losing some 5-10% output in 1AA and (some) runtime in 2AA.
All inductors I tested (same board, swapped inductors, same cell, measured output current; very humble and unprecise testing scenario) did work on the Nanjg110. Only difference I experienced was efficiency.

Datasheet does not say above 4.7µH, it says minimum 2.2µH, 4.7µH recommended. All commercial driver of this kind I recollect have 2.2µH inductors. I tried 4.7µH and the only difference was that they were less efficient at similar size (as they have by design a higher resistance than the 2.2µH). 4.7µH might/will have advantage in other terms (saturation?) but i wouldn’t know anything about it. I’ll stick with 2.2µH. You can take the inductor from the FT-driver for starters.

y makes for µ on a keyboard without a µ.

.

@mattlward
Allow for 12-15 hours and you get another 17mm driver board to choose from.

Hey HQ, thanks for the update… I will wait a day or 2. So, the build is a parts steal right? Any idea what kind of output might be there?

Good point.
I added the SMD part sizes in the HQB15 v3 post. All HQB15 (v1/v2/v3) have the same parts.
In the HQB17 v1 post the footprints are stated. Notable difference is the larger inductor.

.

Yep, oncoming HQB17B is specifically meant as the simplest possible part swapping build (keeping fingers crossed that C2 will be sufficient).
Output should be in the 750mA range with 2AA. Very well within the spec of the PAM2803. Output depends on R1 (Sense Resistor) and the FT driver has R1 = 0.120 Ohm (“R120”), could change with any production batch, as always.

Maybe I should have summarized:

- The code I posted is to test the behavior of the FET with a partially charged gate. Quick & dirty way to see low output. Don’t leave it running too bright too long.

- Plan B: It can be integrated into other firmware for low modes, with some difficulty.

- Plan A: The hardware can be changed a little to get similar low modes with slightly modified existing attiny13a firmware. The resistor value can be approximately worked out from behavior of the code I posted.

  • Plan C: Use lower PWM frequency. Also complex but definitely possible and with no hardware changes. Has risk of visible PWM eventually. Might eat up valuable firmware space.

The major unknown is how much the results of A & B will depend on part variations. I’ve not done enough testing to have any clue. It could be that each driver must be flashed with different settings to get the same output. I think that would be annoying but still better than no low modes at all.

Yeah, some clarity about output expectations for 1xAA or 2xAA would be nice. Also, I don’t know anything at all about driver development. Does this need a custom firmware written for it, or do (any/all) the already available Tiny13 firmwares work? I’ve been trying to follow along, but I’m way over my head.

That is a very interesting approach, fixed it, thanks for posting! Most spend their time trying to bypass triode, not linger in it. Like you said, it'd be interesting to test what kind of dissipation occurs if you drive it that way and whether the trade-off in efficiency is worth it.

This isn’t about high power. Output remains the same and is limited by the power handling ability of the boost ic to ~1A Max or lower by choice of sense resistor. What’s new is the addition of the Attiny 13A. This opens all the AA and 2xAA cheapo and not so cheapo lights up to mods without annoying strobe modes or with modes of choice. Non muggle modes for muggle lights.

Thanks for the responses HQ. I’ll raid my parts bins tonight and then order what I need.

HQB17B v1

17.6mm boost driver board
designed to source as many parts as possible from the FastTech boost driver

.

EDIT: More capacity really is needed: see post #30
Either you need to add a capacitor at C3 or swap for a larger C2 (like 22µF)

.

Oshpark Link

.

Parts size
(more on parts and values see partlist overview in post#2)

Switch: PAM2803, SOT-23-6
L1: max 4.3x4.3mm
D1: DO214AA
C1: 0805
C2: 0805
C3: 0603 (optional)
R1: 0805
R2: 0603
R3: 0603
MCU: ATTiny13A-SSU, 8S1
OTC: 0603 (optional)
FET: N-Channel, SOT-23-3

.

.

The highlighted parts are those from the FT-driver, just add an ATtiny, C3 and - if needed - an OTC.

Thanks! That’s a very helpful highlight.

It’s time for some field testing.
I just built a HQB15 v2 and installed it in a Sipik SK68 clone.

First memorable result: A swap of the FT driver will need additional capacitance (higher C2 or additional C3).
I did at first a straight transfer of only the FT parts and got only initial flickering. Like: strobe for half a second. Same after every mode change. I changed between different driver frequencies, still just flickering.
The FT driver had 10µF as C2 (10.7µF according to DMM, I tested it after harvesting the part). Added a 10µF 0603 (tested 7.8µF) as C3 and - voila - the light stays bright in every mode.

R1 is 0.120 Ohm (R120), no ‘overclocking’ for now.

Flashed is the already mentioned adapted MiniDrv with 3 modes (PWM 1,8,255), with on-time memory (no OTC).

I started with 4.7kHz (9.6MHz, fuse 6a (divider:8), fastPWM) and now flashed to 38kHz (9.6MHz, fuse 7a (divider:1), fastPWM). Both work, 4.7 is whining considerably and has higher low, 38kHz is silent and slightly better at PWM1.

Let’s see how it handles some heavy use. Pics later.

Edit:
Of course I swapped for a BAT60A before installing into the SK68, as the original diode was only attached by wire.

Good to know, thanks. Still waiting on the boards.

The HQB15 v2 is still up and running in the SK68. First results after some days of heavy use

- the driver does not press-fit well, I had to back it up with wire or output was not stable

- a notch really helps to pry out a press-fit driver…

- re-prgramming while installed did work, just take care that the LED-wires are soldered away from the MCU

  • the modes {1,8,255} are far too close together (at least with 1AA), I’ll look into lowering moon next

But, well, in the end this is my first 1AA with own programm.

.

SK68 pill with HQB15 v2 (0.8mm board), original SK68 driver in the background:

.

.

The built driver so far: one HQB15 v2 (0.8mm), two HQB15 v1 (1.6mm).

Well done HQ. Its good to see this up and running in a light.

Little status update

I successfully built one HQB15 v3 and one HQB17B v1, which both are running fine in lights. Still with simple firmware, I tweaked the MiniDrv firmware with several mode groups which took me a lot of time.

To be honest I at least postponed the idea for LVP. None of my AA light has it and I can live with a dead or degraded eneloop once in a while. I just won’t find the time to dig deeper here. I will finish the planned 17mm LVP boards (C/D) so if anyone else wants to progress LVP he has something to go on. Just ask if you need a certain layout.

I started working on HQB15M and this one looks really promising in Eagle so far. Made me chuckle more than once. I redesigned my Attiny MMU breakout board and that took me a while, but now it all might get into shape.

Using the Attiny13a-MMU would even allow for sub-15mm variants. Just sayin’ :wink:

How far sub 15mm? I am still interested in trying these boards. What would be the best rev to go with at this point?

14mm was easily sketched
13mm already looks doable.
Less will be tight.

Mind it’s the 3x3mm MMU package. :wink:

All the smaller AAA boost drivers I’ve seen with modes use stacked boards to fit an mcu. Somehow they look bigger on your table than mine. :wink:

With 12mm you serve quite a lot of AAA lights :slight_smile:

I’m hoping at some point these boards will be sold ready to go, Mtn Electronics is the obvious seller, but a source in Europe or Asia is easier for us overseas.

To that effect it seems GB’s with custom drivers can make that happen. The drivers end up sold separately afterwards.