You’ve already shown that the LEDs are hooked up correctly by doing the test comfychair suggested.
The capacitor you are referring to is a “decoupling capacitor”, it is not for offtime. If you get battery voltage (or battery voltage minus ~0.2v) at the MCU’s Vcc/Vdd/whatever pin then you do not have a problem with the protection diode.
Post large, detailed photographs of everything involved. The LED MCPCB, the wiring from the MCPCB to the driver, both sides of the driver, pictures specifically showing your soldering job on the swapped MCU, etc.
Strange; that I forgot to flash the driver 2 times in a row sounds unlikely to me. Hm.
But indeed, usually the ATtinies survive a lot: Someone accidentally applies 2*16340 to H17F, the LED died, the driver survived.
Oh, just reread #13, your pin numbering is wrong: 1 is lower left, 4 (GND) is lower right, 5 is upper right, 8 (Vcc) is upper left. Pin 3 is the switch.
A few more tests…
Measure voltage between pin 8 (Vcc, upper left) and 4 (GND).
Use a small jumper wire or fine tweezers to directly connect pin 8 (Vcc) to pin 7 or 6 or 5 or 2.
Maybe it’s the switch?
Test continuity between pin 3 (switch) and 4 (GND) when pressing the switch.
Use a jumper wire or fine tweezers to connect pin 3 to pin 4.
Oh yes. I wasn’t thinking when numbering the pins.
3.75
Jumping 2,4,6,7 to pin8 turns on the corresponding color. Fully bright.
There is continuity between pins 3 and 4 only when the switch is pressed.
Jumping pin 3 to pin 4 does nothing.
This circuit is very simple so this is confusing to me. I can’t see what else could be wrong. I’m going to retest everything I can, review my methods and see if I’m missing something.
PPDB22 - as far as determining for certain what’s wrong… I would say that we’re back at the point you brought up earlier. That is to say, can we determine whether the MCU is blank? We can. (We just can’t determine whether the MCU is programmed incorrectly.)
Once that’s out of the way, adjust the command for the ATtiny85 and for dumping rather than flashing. I believe that it would look something like this: The guide actually covers this and provides a command, so all that’s required is swapping “T13” for “ATTINY85”….
Ok white. I think you are right. It’s time for me to learn to flash. Pffff… It’s something I really want to learn to do(especially because I have a handful of attiny drivers with flashy modes) but I’ve been a bit intimidated.
One of my challenges is that my days resemble the move “ground hogs day”. Don’t know if you’ve seen it, but the actor tom hanks wakes up and it is Groundhog Day everyday. He experiences thing over and over but every morning when he wakes up he has accomplished nothing. In the movie, it is everyone else that forgets(never new) all that has been accomplished. Which is what it is kind of like for me. Every night my brain resets and I forget most of what I’ve learned. So I read things over and over again until I am able to learn fast enough to cover it all in one day. Once I’ve learned something it seems to stick around but the process of learning is difficult.
I’ve worked out a little system now where I stash the things I learn throughout the day. Learning to flash will put my system to the test;)
Dr Jones
You are MORE than generous! I fear it’s me who has messed up at your expense. I will continue trying to figure out the problem here and if in the meantime you are willing to send another, pm me with shipping and parts expenses. I thank you for your help!
What caused the original dead short, just the nose of the battery pushing the button too deep and then contacting the ground part of the retainer? And then it was totally unrelated that there was also some other problem with both MCUs?
It’s highly unlikely those two (three?) are unrelated, but not impossible. ‘Highly unlikely’ things happen all the time.
What does your meter show, checking ohms from the battery spring to the ground ring? It should be some very high number but not a complete ‘O/L’. Like 30k-50k or something.
Those last tests showed that everything else but the MCU is working; further tests on anything else are probably wasted. I can think of two possible reasons: a) You killed them. But that’s quite improbable; the first with 10Wh from the cell blowing out in 30s (that would be 1200W, hm, probably not, instead the cell probably killed itself. There must be still some energy left in there), that might have done something, but the second? b) I actually forgot to flash BOTH drivers. Not very likely either, as they weren’t made on the same day. However one of those reasons it must be, and since everything else survived the ‘incident’, a) is even more unlikely than b). Quite embarassing.
I’ll additionally send you a (flashed!) MCU to swap in, so you’ll have 2 working drivers then. And a H17F+ (you’ll be the first to have one).
Ya. Idk? There really isn’t much to the circuit component wise. Which says a lot about your abilities to create such firmware that is stable with so little components. Like you said, every possibility seems unlikely. This was the first time I transfered an mcu, but I watched a number of videos beforehand and it when quite smoothly. I did not have to use much heat. I soaked it with flux, wicked the solder, then jently popped up one side at a time by dragging my iron from one pin to the next. To reinstall, I aligned and clipped the mcu, then one pin at a time I touched the iron tip to the board, then to the pin and then pulled away the iron. After, I checked for continuity between each pin and the end of its lead to ensure they had latched to the board. I also checked for shorts. So, I think I did ok. That said, things can go wrong even when doing good. I have been biting my nails waiting for the H17F+! I don’t know what to say! You are being very kind to me. Thank you! I’d like to say more… I’ll start preparing a review right now:)
Comfychair, I measured infinate resistance between spring and gnd. I measured 4.67k from Vcc to gnd. But I don’t trust my measurements on this one. It doesn’t seem right to me? That was without power applied.
No, you are waiting for the H17F (still takes a while; I’m waiting for PCBs); you’ll get an additional H17F+, it’s not even announced (it has a better FET for more current on turbo).
I have the programmer wired up with drivers installed. I am following steps by comfy chair to creat a hex file from luxdrv .c file version 0.30b. My first attempt failed with one error, but I think I know what was wrong and will try again tomorrow. The error was, “Variable ‘modes’ must be const…” I think I just need to add “const” to the corresponding command line.
I’m getting there. I will flash a driver or two and then check those attiny85’s just to see. I don’t know what I will find or what to look for but we will see…
When doing new stuff like that I have to see it work first. So, to start, build a .c file with no changes, flash it, and see the driver work like it’s supposed to. Even if that’s not the actual firmware you want to use. Then you can go back and make changes. Since you already saw it work once, and if you’re smart only making one change at a time, if you break it you’ll know exactly where it stopped working and then have an easier time figuring out how to fix it.
If you can’t build the .c file as-is, it’s almost always something goofy with your settings in Atmel Studio. Like the ‘release’ vs. ‘debug’ profiles create hex files of different sizes, and sometimes the ‘debug’ version comes out too big for the chosen MCU and AS chokes on it. Change the little drop-down box to ‘release’ and try again, it usually goes through fine with no other changes.
I have not continued my driver flashing lessons. I have been dealing with taxes(we sold a rental house), my phone(ive been out of a phone now for two weeks), and my wife and daughter and I are going to be able to move out on our own, our offer on a humble house has been accepted! This is a big deal for us; we’ve lived with my parents since I first became ill about 6 years ago. I think I am excited, but it is going to take a lot of my energy so I will have to cut back on my flashlight building for a while. However, when we are settled there is a large area in the new house that I will be able to convert into my own flashlight building workshop! No more laying sheets down in my mothers bedroom just to use my dremel for a bit;) Dust will fly, wires will be snipped, and solder fumes will freely swirl through the air like a dancing eel! And when I am done, I will walk away from my mess without a care and I will feel like a man again!!! Maybe I am a bit excited
So… I received a new driver as well as another mcu. Thanks dr jones!!! I’ve been able to play with each and I have found that both the new driver and the old driver with the new mcu installed work perfectly. So the problems I was having we’re definatly a result of two defective mcu’s. I still can’t see how dr jones could have forgotten to flash two mcu’s before sending them. There must have been something I did or maybe they were damaged in transit? I was wondering if the magnetism of the automatic letter sorters could have erased the mcu’s? Idk! But I now have a working Roche f6 with rgbw driver and one chip added to each color channel and two chips added to the white channel. I love it! It’s been worth the efforts for sure. Now I’m working on cracking the Easter egg code. Thanks dr jones!
Thanks comfychair! I finally got around to trying again. Your suggestion worked and I now have a couple drivers programmed. Idk why I was so intimidated by the process, it was easier than expected.
