Oshpark Projects

Matt’s first FET board, the one with no ground pads on the spring side.

Can you post a photo of both sides and I can tell you if it's a hookup issue or if there is (I hope not) a potential trace issue?

Mine acts like the early attempts when the code didn't work right without an added capacitor. Just weird, unpredictable mode changes. Sometimes it even does a strobe mode that's not there. Sometimes it does weird patterns of blinks like morse code. Checking the board with the meter, the cap is connected to where it should be.

Check the Vcc voltage of the MCU to make sure the diode is not connected in reverse.

If the diode were backwards it wouldn't work at all.

The diode seems to be compatible with what comes on the 105C, I get 9 blinks in luxdrv's battmon mode from a cell that measures 3.87v, close enough for me.

Now that I think back, on the big 46mm version, mode changing went goofy as well when I was testing without the big 470uF cap. I'm using this one, should be correct right? Does it just need one rated higher than the 105C's original?

Comfy, I think that capacitor is correct, I think I've seen mentioned that the stock cap is also 10uF. This one should even be better since it's an X7R from a name brand... dunno what's going on there. You could try and stack a second cap on top of that one to see if you get any better results.

Pics of your build please?

Diode backward?

What OSHPark board are you building the 15,17, or SRK?

Comfy…is that the 17DD initial release the one w/o the ground ring on the battery side…might not be getting a good ground (floating ground)

Looks to me like he's got a wire soldered directly to the FET pin for ground, is that correct?

Ah…duh…bench test

Why did you go to a 10μF from a 470μF? Wouldn’t it be advisable to go higher rather than MUCH MUCH lower?

I have a separate ground wire stuck right onto the FET's ground pin, testing right now with a battery box with alligator clip leads but I'll be using a separate wire soldered to the inside of the pill, and rely on just a press fit to hold it in place, it's plenty tight like that if you don't completely file off the flashing.

I think I have a giant bag of 33uF 'A' case tantalum caps around here somewhere...

Because he's using this one on a 17mm board where there isn't room for a huge electrolytic cap. We know that it can function from the stock 105C cap., so it should be able to function here as well.

The 470uF/16v on the SRK driver is only for noise reduction, it shouldn't be needed for the MCU to function (but for some reason, it looks like it is letting the MCU work right, because the little 10uF/16v doesn't seem to be enough).

Ah…and just did a SMD 0805 search on digikey…largest SMD is 47μF

Well…that cap could be storing enough energy to keep the MCU alive at 470μF, opening up that huge MOSFET more than likely causes large voltage sag and drops the power below the MCU voltage range causing it to go all squiffy

But why does it work when you stick the same FET onto a 105C, using all the 105C's original parts? Is it possible the original cap isn't 10uF? There's nothing different about the new PCB, I've checked everything until I'm crosseyed.

Also there's nothing higher than 10uF in 0805 & X7R. 15 & 22uF 0805s only exist in JB, X5R, & X6S.

Is there any difference in the firmware between working and non working drivers?

Take a break, attack it from a different direction in a bit man