Mitochondrion 3.4 progress

36 jobs out of 46 done.

Most of this weekend was spent checking over things to be made next weekend, seeing if bits fit and if things will be connected up properly, when they are connected up.. Yup, there were bugs. There are now less bugs. Is it right? We will find out. And assorted other little jobs, where many little jobs add up to lots of time:

Checking for fit, before making a real circuit board, not a paper one. Shit, the pins on the 40X1 (the square chip) are tiny:

If you look carefully, you can spot the BMSAF.

(Okay, so the hardest job is number 38, requiring extreme soldering skills. But hey, I like to live life on the edge, especially when soldering.)
(And once the hardware is done, I need to rewrite the software, coz it currently runs some shonky test code I threw together without really thinking, and it does approximately a hundredth of what I’d like it to do, while taking up a third of the memory space. Obviously, this isn’t good enough. But hey, I’ve only got 4k of program memory to fill, how long can it take to write 4k of code…?)

10 thoughts on “Mitochondrion 3.4 progress

    1. Err… this is one of the dodgier parts of the design. They’re always connected to the main bus, for resistance to mechanical shock, so it’s just a case of plugging a external DC source in, detecting the voltage jump when this happens, turning everything else off, then waiting for the thermistor on one battery pack to show the temperature rise when batteries are full. And then just turning things back on to reduce the charging current to zero-ish.

      Yeah, simple, skody, not ideal. But I’m not sure how to fit anything better into the space I have. For a start, NiMH packs in parallel is tends to munt one pack. I’d love a disconnect on full charge, but no idea how. WTH, it’ll do for this version.

      (Next version may use lithiums. Next version may catch fire and die.)

      1. Also, you’ve just made me work out a much better way to catch the voltage jump on charge start. Cheers.

        Life sucks without external switches.

  1. Things

    1. When did it go from 49 to 46 and what broke to make this happen?

    2. Is ‘check it works’ one of these jobs?

    3. The BMSAF looks like it’s been cut out with scissors. But not while listening to Spencer Gordon.

    4. Sellotape? Oh yes, I see why.

    5. And as an added bonus, if it flies apart and lands in water, your batteries will float and therefore be salvageable.

    6. Check out the title of this image:

    1. Re: Things

      1) There may have been job rationalisation when one stone killed two fatted calves. And there may have been times when I said to myself – “nah, that’s a stupid idea”.

      2) The checking is pre-hoc. That way, I’ll finish making it, plug it in, and it’ll just work. Also, pigs will fly.

      3) It was birdies going tweet, I seem to remember. Or possibly the Diplo EM.

      4) Sellotape coz then you can adjust diameter in 25 micron increments for a tight fit. Things wot rattle are bad, for reasons you’ve already seen.

      5) Given the construction, it should float for almost a minute. I’m not keen to try this, no matter how pretty it might look.

      6) Ffs! Tagging individual river fish with RFIDs? How about just not fucking with the river? If you’ve got to do that degree of active management to keep the river alive, then it’s a pretty screwed up river.

        1. Re: Things

          Yeah, coz I am so hardcore, so long as by ‘suspended over a flooded torrent’ you mean leaning near the kitchen sink.

          Also, stainless steel counter-tops => anti-static workstations!

Leave a Reply

Your e-mail address will not be published. Required fields are marked *