#239

Поділитися
Вставка
  • Опубліковано 2 лис 2024

КОМЕНТАРІ • 33

  • @pldaniels
    @pldaniels  5 років тому +7

    A bit of programming for a change, though not at my normal programming desktop, because I wanted to be next to the meter as I did the testing.
    Anyone who wants to complain about the fact that I bought a "cheap sh*t meter", send me a better one and I'll use it if it covers my requirements (including having the data protocol documented).
    Meter has a single use environment (board repair) where the cheapness is not an issue. Aside from this one quirk everything else is covered well (diode mode, resistance, volts and frequency) at a cost that was hard to pass up; difficult to justify spending 5x the price for no real use gains.

    • @pldaniels
      @pldaniels  5 років тому +1

      That's the idea, yes. I've got my "higher spec" hand held meters for when I need precision work; for board repair it's a low demand criterion and this did a good job fitting the need.
      I bought the Protek 806 many years ago, with calibration, back for when I was doing micro power circuit design and needed 0.1uA resolution; it however wasn't really a good meter for board repairs.

    • @ewastesydney7470
      @ewastesydney7470 5 років тому

      What may work for you may not work for others, but ..... who cares for others 😁
      Great job to see how you can reprogram to fit your requirements

  • @davidlegault9745
    @davidlegault9745 5 років тому +3

    If you are happy with it and it does what you want it to do - win - win. Regards, Dave

  • @Motorman2112
    @Motorman2112 5 років тому +1

    Thanks for showing this. You're doing this and I'm sitting here with a bunch of post it notes with variable names and "true" or "false" on either side. I'm flipping them over as I read through my code trying to debug a state machine.

    • @pldaniels
      @pldaniels  5 років тому +1

      oh that can drive you insane when you get through so far and still manage to mis-flip things. I know people love debuggers for a lot of stuff but often doing it like you are is the best way. Good luck with the state machine.

  • @chucktodd7329
    @chucktodd7329 5 років тому +2

    The problem of key reception is that you need a delay before the first command is sent. When you manually change the range your software immediately sends the change range command, while the meter is in a debounce timeout from the manual key. After you detect the wrong mode, delay for an initial debounce period, then send the first mode change, debounce delay, send second mode change.

  • @wallflips
    @wallflips 5 років тому +1

    Man you should do more coding videos, really inspirational work there!

  • @rars0n
    @rars0n 5 років тому +5

    As soon as you fix this, Louis will complain about something else.
    Wait, who am I kidding? Louis never updates his software.

    • @pldaniels
      @pldaniels  5 років тому +4

      At least this software is a different variant to the one he's using, same core engine though but this at least is going to be specific for my use here. I do find it so useful being able to code up software to my requirements.
      Getting Louis to update software is quite a process.

  • @deanrubine2955
    @deanrubine2955 5 років тому +1

    Writing C with vi; take me back to 1979.

  • @QsTechService1
    @QsTechService1 5 років тому +1

    Awesome coding 👍 Lot more faster response Looks like you get out of the meter 😃

  • @EldaLuna
    @EldaLuna 5 років тому

    that's kinda neat. didn't even know could reprogram those kinda things.. specially the cheap crappy ones that mentioned in the pinned comment. i guess there's always away to do something to stuff i guess with enough force or know how or it has a thing for it lmao.

    • @pldaniels
      @pldaniels  5 років тому

      In this case, I'm not reprogramming the meter itself but rather I'm using the software I wrote for displaying/collecting the data through the serial interface to act as a manager for the meter, to look out for the condition that annoys me and apply the fix required.
      If I don't run the software on the PC with the data link the meter will behave as per normal with the quirk that I dislike.

    • @EldaLuna
      @EldaLuna 5 років тому

      oh i see now. so just a passive fix rather a permanent one. well that sucks but i guess its something for now aha.

  • @TheDefpom
    @TheDefpom 5 років тому +1

    Is there a command to turn off the beeper? Might be an option

    • @pldaniels
      @pldaniels  5 років тому

      Not that I can find; would be nice if there was. I suppose I could try cycle through a range of byte codes... and hope I don't cause a mess to the calibrations

    • @robc8468
      @robc8468 4 роки тому

      @@pldaniels why not unsolder the beeper and put an on off switch in series with the beeper?

    • @pldaniels
      @pldaniels  4 роки тому +1

      @@robc8468 would likely just be as much drama remembering to turn it on/off. I've come to get used to it, especially since now the software I've written forces it out of auto-range mode when volts are selected.

  • @Dratchev241
    @Dratchev241 5 років тому +1

    Don't delay, program with paul daniels today!
    is the VC8145 firmware locked? ie can't update it at all? just wondering cause if user can't do a firm upgrade that could be a bugger if a firmware flaw was found and the user was sol being unable to upgrade.

    • @pldaniels
      @pldaniels  5 років тому

      Looks like the HT48R30 they include is actually able to be reprogrammed, but not sure if anyone has a sane tool to do it other than from Holtek themselves. In some ways, it almost might be easier to produce an daughterboard with a different MCU (ATMega likely would suffice with a lot of performance/capacity to spare). Likely just not worth it, easier to simply work around the behaviour as I've done or get a different meter.

  • @TheDefpom
    @TheDefpom 5 років тому

    @25:00 the unreliability is probably due to collisions in the serial communications.

    • @pldaniels
      @pldaniels  5 років тому

      by the looks of it, I think it was more to do with potentially causing undefined behaviour due to not reading back the response from the meter after I sent the range-change request.
      If you check the code where I do the send_cmd() -> byte_read() -> send_cmd() -> byte_read() sequence to shift in to the new state, you'll see there's no delays there, it's as fast as the meter will receive reply.
      @15:44 you can see I've highlighted the two 0xA1 bytes which were in the meter output buffer waiting to be picked up because at that stage in the coding, I wasn't listening for the reply back. By the time I get to about @19:56 I've put in the new function to retrieve the response byte and from there it works perfectly.

  • @cls9474
    @cls9474 5 років тому +1

    Hi Paul, if you make yourself a member of the dialout group you don't have to launch your program with sudo...

    • @pldaniels
      @pldaniels  5 років тому

      Thanks, I forget that I can do that. Laziness.

    • @cls9474
      @cls9474 5 років тому +1

      You're more than welcome :)

  • @janzimny594
    @janzimny594 5 років тому

    Good for ya! Glad your happy with it! could you find some firmware to upgrade my noggan? Ha ha good vid again Paul!

    • @pldaniels
      @pldaniels  5 років тому

      Would be nice to get some firmware upgrades for the head... and body :D

    • @janzimny594
      @janzimny594 5 років тому +2

      Paul Daniels If you do come to Canada, make sure to let me know !

  • @sidp1494
    @sidp1494 5 років тому +1

    I SUBBED TO PAUL DANIELS