Ouchy - have you done full continuity checks? Be sure where you should have connections, then check for shorts? I do that with every board I build and have found problems - even with a couple of these boards, but all were obvious shorts that were easily fixed with the iron. Yours has nothing obvious from the pic, but agree with DEL - can't see the labeling on the zener, so who knows - but sounds like you got that right too (with the "W7" label, the W should be towards the C1 cap).
This is a view of mine:
Was the MCU pre-programmed with Bistro from MtnE? Richard and Lisa don't make mistakes often but it has happened - maybe it's not programmed?
Thanks for the pic Tom. They sure come in handy for reference.
Yes, I have the zener mounted with the line towards the large capacitor.
The problem with me checking with a DMM is that I really don’t have any idea how these components relate to each other, so I don’t know what’s right or wrong.
What would happen if the MCU had no program flashed in it?
I removed the MCU and replaced it with another one and it works perfectly.
I guess the flashing didn’t take on the first one. I have ten of these, so I hope it was just the one MCU that didn’t hold the Bistro program. I really hate whining to RMM about one defective MCU, but he doesn’t have the pre flashed attiny25s for sale yet.
Our AVRDude command files always verify, byte by byte. Maybe you missed the verify, or just got unlucky with a bad MCU. Not sure if I ever got a bad one, can't recall, certainly not recently and all 25's and 85's have been good so far. I just got one of these boards in a Skilhunt H15, headlamp style light. It can't be held after 30 secs with an EFEST IMR 14500, it gets that hot. But real nice with the locator LED in the switch I got working, and full ramping I used a XM-L2 T5 5D - real nice tint. I replaced the LED in the switch with a pink one, but they always come out looking purple for some reason, but nice! Smallest light I've built so far w/Narsil on the 85.
I use a 350 mA 7135, not the 380 - I can get a really low moon mode, using a PWM value of 2 (programmable from 1 to 7).
I didn’t flash it myself. That’s way above my pay scale. It came with a parts kit. I will try installing the old one on another board just to make sure it wasn’t a bridging problem or something, but I doubt it. I’m just glad I got the problem solved. Just when when I got confident in my reflowing skills, I was starting to doubt my abilities.
Ouchy, I’ve gotten a few drivers from Richard with blank MCU’s as well, I think he might have had a bit of an issue at one point and missed it. I did the same thing, built a light and troubleshot the heck out of it, finally reflashed it and it worked perfectly. All along it had no firmware.
Could be a firmware issue, or a ground issue. Those MCU pins don't look fully reflowed to me (lots of gold still showing, especially on the ground pin).
I did reheat the Board and wiggled and pressed down on the MCU to make sure it was properly soldered, but who knows. I haven’t given up on it yet. I plan to reflow it onto the next board I build before I come to any conclusions. At least now I know what part of the board was causing me problem and it will be easier for me to do a complete trouble shoot.
Richard. When are you going to be stocking the attiny 25 preflashed with bistro, and those 40K-60K Pulldown Resistors?
lol, yeah well, when I forget I forget all the way! Did all kinds of proofing looking for why that first light wouldn’t work, totally clueless the MCU was empty. It was really pretty funny when I figured it out, finally.
Bought one of Sulman's Titanium 18350s the other day just to upgrade the light engine for a client and when I removed the old LE, it appears that the guy who modded it after Sulman originally built it did just that. After thinking through it for a minute I could tell exactly what firmware that he had loaded to the MCU by a small personal code that he wrote directly to the PCB. Also had the LED bin's written on the mcpcb. Great idea. Especially if you're building this stuf in bulk and storing it for a later time.
The last few month I just resorted to sticking, unwaveringly, to guppydrv2. Hard to not know what's on an MCU when you use the same firmware for every build lol
I do that so I don’t forget what bin I used in the build. Easy to write it on the mcpcb, and the date too if it’ll fit, then nothing has to be remembered.
Ok so refresh my memory, (again) Richard… I have one driver left, an FET+1 with ATTiny13A and the lage center via for the spring bypass. It’s got a Zener on it and the high value gate resistor… it won’t flash, the USB ASP turns off when I clip it onto the MCU. What’s up with that?
Yeah, second I hit "send" on that post I thought to myself, "I bet he bought it from Dale" hahaha He never said where he got it, just that it was modded. Had a triple XPG2 in it (was marked "R5" on the mcpcb, indicating the bin) and I know that Sulman doesn't use XPG2's, or doesn't currently anyway :)
Well, I’m pretty much the US warranty dept for Sulman, so there’s quite a few I’ve had my hands on out there in the wild. Virtually all of em that I’ve had to crack open were the typical USER ID10T error code. Seems like people just like opening up lights, whether they have a clue about how a light is put together or not. Dome’s broken off emitters, glow in the dark stickers melted onto the domes, you name it… even a primary CR123 cell installed and the guy couldn’t get it to work. :person_facepalming:
I'm pretty sure that I told you about the client who I installed, in a sinner, one of Serge's Lux-RC 371d LEs...? He wanted to see it and ended up ripping the contact poles off when removing the optic. Serge's going to get me a replacement LE when he gets back from Vacation...In September. lol
User error...well that's nothing new. I have had several lights that "didn't work" that were obviously opened up and either broken or put back together incorrectly.
Did all kinds of proofing looking for why that first light wouldn’t work, totally clueless the MCU was empty. It was really pretty funny when I figured it out, finally. 
Seems like people just like opening up lights, whether they have a clue about how a light is put together or not. Dome’s broken off emitters, glow in the dark stickers melted onto the domes, you name it… even a primary CR123 cell installed and the guy couldn’t get it to work. :person_facepalming: