techtravels.org

java client works

Well I’m producing correct .ADF’s with the new java client.

The terminal mode is also finished and working.

I have some various odds and ends to wrap up with it, namely allowing the user to save his preferences, actually putting the preferences to use(I’m currently ignoring them) and using the correct filenames, etc. Bunches of little stuff really.

Besides those, the other major thing that needs implemented, written, etc, is proper error handling.  I’m not reporting or taking many error conditions into consideration.  I don’t save “bad data”, but I also don’t retry, etc, to hope to fix the problem.

I *am* currently having a heat problem where my SX28 is heating up way too much. I tried posting on the parallax forums, but that was unfruitful. I think I must be asserting something, and that same lead is being connected to ground, but heck if I can figure out what it is.

I’m going to try disconnecting the floppy and see if it continues to heat up. and then I’m going to check all the pins into the floppy and see if any go straight to +5v or GND.

keith

Amateur Electronics Design Engineer and Hacker

2 comments

  • performed a little maintenance tonight.

    in the process broke a few things but I know what I’m doing! 🙂 or so says the mad scientist!

    I’m trying to get the whole thing wrapped up and automated, but that isn’t happening yet. My transfer checksum routines aren’t working, because I’ve added some stuff at the beginning that can throw off the starting point.

    I instituted using a header 0xff 0xff, awhile back, and this helps me find the start position, and the offset from there.

    Some things are getting screwed up, so I’m going to go back and figure out what I’m doing wrong there.

    Everything is coming together nicely though, overall, and I’m getting closer to having a full working piece of software.

  • I’ve seen several protoboards where each and every pin of the microcontroller connects to nothing but its own 10 Ohm resistor.
    (Except the power and GND pins, which are also connected to the capacitor that runs directly between power and GND pins.)

    Then when the microcontroller gets too hot, it’s easy to put the 2 probes of a voltmeter across any one resistor and measure how much current is going in or out the associated pin: I = V/R = (voltage measured) / 10 Ohms.

    Oh, except the resistors for the power and GND pins were 1 Ohm, just to keep you on your toes when doing the mental math:
    For the power pins, I = (voltage measured) / 1 Ohm.

    You can also probe both sides of the resistor with the 2 probes of a dial-trace o’scope.
    With the o’scope, you could distinguish between
    * high current cause by low resistance: microcontroller tries to pull the pin to 0 V or +5V, and the output immediately snaps to a new constant voltage, but that constant voltage is further away from 0 V or +5V than we expected.
    * high current cause by high capacitance: microcontroller tries to pull the pin to 0V or +5V, and the output very gradually drifts all the way to 0 V or +5V, after a much longer time than we expected.

    — DavidCary