Loading ...
Sorry, an error occurred while loading the content.

RE: Re: [cosmacelf] (Attempting to) diagnose/fix memory problems on a Membership Card

Expand Messages
  • joshbensadon
    Hi, Hmm, sounds a little intermittent. If you RESET, then go into LOAD mode and load a bunch of 00 s (say 10 bytes of 00 s). Can you return to RESET, flip up
    Message 1 of 4 , Sep 23, 2013
    • 0 Attachment

      Hi,


      Hmm, sounds a little intermittent.


      If you RESET, then go into LOAD mode and load a bunch of 00's (say 10 bytes of 00's).

      Can you return to RESET, flip up Memory protect and go into LOAD mode again and see the 00's?


      Try this again with FF's, AA's, 55's, and ultimately with a single bit for each byte (ie, 00, 01, 02, 04, 08, 10, 20, 40, 80).


      Does it write and read the ram correctly?


      PS. The bottom board looks a little messy, are there any solder shorts? I can't tell from the photo.


      Josh




      --- In cosmacelf@yahoogroups.com, <cosmacelf@yahoogroups.com> wrote:


      On Sep 23, 2013, at 8:45 PM, boba_teff@... wrote:

       

      I got one of Lee Hart's Revision F Membership cards (All of it, with front cover, etc.), back in June/July and assembled it then. I did some testing then, and found that there were some, best described as, anomalies in operation.



      To elaborate, I found that there were some semi-subsequent errors when toggling in programs, where what I had input was not what was taken down to memory. Additionally, I found that, with write enabled, my instructions would sometimes be partially correct, say if I entered 0B I might get 5B, the previous instruction having been from a test write of FF. I found that when going through all of the memory with FF or 00, there were times when the newly written instruction was not such.

      This problem seems to extend into programs, not just Load Mode. Lee Hart's program, which is designed to take whatever is on IN4 and put it on OUT4, seems to function some of the time, then cease to functio n.

      I took some pictures of the board itsself. All of the instructions I've tested seem to work properly otherwise. I didn't really get around to trying to debug it until now due to the lack of a project requiring a device like the 1802MC.

      Here's all of the pictures I took, uploaded to an imgur album: http://imgur.com/a/bVTwh#0

      If I need to, I'll make a video showing this abnormal operation. Any speculation as to what's causing this? Misproperly placed resistor, bad chip, joint, switch, etc?


    • Lee Hart
      ... It sounds like you have an intermittent connection somewhere, Bob. If you can t find it, send it to me and I ll have a go at it. With CMOS, an open circuit
      Message 2 of 4 , Sep 24, 2013
      • 0 Attachment
        boba_teff@... wrote:
        > I got one of Lee Hart's Revision F Membership cards (All of it, with
        > front cover, etc.), back in June/July and assembled it then. I did some
        > testing then, and found that there were some, best described as,
        > anomalies in operation.

        It sounds like you have an intermittent connection somewhere, Bob. If
        you can't find it, send it to me and I'll have a go at it.

        With CMOS, an open circuit lets an input "float". It may float high, or
        low. In one state it may work, and in the other, it doesn't work. There
        may be some high resistance (megohms) across the open circuit, so that
        the circuit still works, but is unreliable. Handling the board while it
        is in operation, or even breathing on it to condense some moisture on it
        may be all it takes to go between working and non-working.

        Open circuits can be caused by forgetting to solder a connection, or a
        bad solder joint. You may also find an IC pin that folded under the
        chip, and so didn't enter its socket or hole on the board.

        In your photo, I see some little splatters of solder flux. I just fixed
        an intermittent that was caused by this. A tiny drop of solder flux had
        landed on one of the connector pins. When the connectors were plugged
        together, the flux just happened to wind up right between the contacts.
        It caused a weak, intermittent connection. Cleaning the flux off with
        isopropyl alcohol fixed it.

        Your soldering looks good, though it's hard to be sure from the photo.
        There might be solder shorts at R11 pin 1, U9 pin 2. The top terminal of
        S8, and the bottom two of S7 might not be soldered. There appears to be
        a blob of solder between R13 7-8 that might be causing trouble. A number
        of pads have a lot of solder -- this can cause shorts. Try holding the
        board upside down, put the tip of your soldering iron on them, and let
        the excess solder run off onto the iron.

        Hope this helps!

        --
        ICEs have the same problem as lightbulbs. Why innovate and make
        better ones when the current ones burn out often enough to keep
        you in business? -- Hunter Cressall
        --
        Lee A. Hart, http://www.sunrise-ev.com/LeesEVs.htm
      Your message has been successfully submitted and would be delivered to recipients shortly.