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

I think I found my so-called bad chip problem.

Expand Messages
  • wfahle
    It looks like R15 is shorted to ground between it and C6+ (the non-hairpin side. With a ground lead on R50 hairpin, the ohmmeeter reads 0 ohms at the pin just
    Message 1 of 9 , Nov 3, 2012
    • 0 Attachment
      It looks like R15 is shorted to ground between it and C6+ (the non-hairpin side. With a ground lead on R50 hairpin, the ohmmeeter reads 0 ohms at the pin just below the TX jumpers (R15). Relevant pictures here, ignore the green wire that ties in the new op-amp chip (the old was destroyed in the process of diagnosis). The other side of R15 reads 10 ohms (9.9 actually indicated), and the other side of C6 goes to infinity quickly after it charges, so the short has to be in the board itself, I think. I don't see any bridges. I'll pull the two components and see what happens.

      http://imgur.com/NIOL1
      http://imgur.com/rUzMv
    • wfahle
      Fixed it. I ran the probe around the edge of R15 before I pulled it, and a shiny bit of copper was uncovered shorting the center of the pad to ground. I
      Message 2 of 9 , Nov 3, 2012
      • 0 Attachment
        Fixed it. I ran the probe around the edge of R15 before I pulled it, and a shiny bit of copper was uncovered shorting the center of the pad to ground. I scratched through it and now it indicates an open circuit. So, I will patch it all back together and see what happens.

        --- In softrock40@yahoogroups.com, "wfahle" <billfahle@...> wrote:
        >
        > It looks like R15 is shorted to ground between it and C6+ (the non-hairpin side. With a ground lead on R50 hairpin, the ohmmeeter reads 0 ohms at the pin just below the TX jumpers (R15). Relevant pictures here, ignore the green wire that ties in the new op-amp chip (the old was destroyed in the process of diagnosis). The other side of R15 reads 10 ohms (9.9 actually indicated), and the other side of C6 goes to infinity quickly after it charges, so the short has to be in the board itself, I think. I don't see any bridges. I'll pull the two components and see what happens.
        >
        > http://imgur.com/NIOL1
        > http://imgur.com/rUzMv
        >
      • wfahle
        And, reassembling, the voltages now check out (2.5 everywhere but pin 8 and 4, which are 5v and 0v, and bridged with a resistor also gives the right voltages
        Message 3 of 9 , Nov 3, 2012
        • 0 Attachment
          And, reassembling, the voltages now check out (2.5 everywhere but pin 8 and 4, which are 5v and 0v, and bridged with a resistor also gives the right voltages of 2.75 and 2.25 at 1 and 7). If you examine the pin in question in this picture (the after) you can now see a tiny trace of copper where the board defect was. I broke that trace with a voltmeter probe, and now it all works. Scratch one chip and one capacitor in the process. Hopefully there was a spare.

          http://imgur.com/oY6qI

          --- In softrock40@yahoogroups.com, "wfahle" <billfahle@...> wrote:
          >
          > Fixed it. I ran the probe around the edge of R15 before I pulled it, and a shiny bit of copper was uncovered shorting the center of the pad to ground. I scratched through it and now it indicates an open circuit. So, I will patch it all back together and see what happens.
          >
          > --- In softrock40@yahoogroups.com, "wfahle" <billfahle@> wrote:
          > >
          > > It looks like R15 is shorted to ground between it and C6+ (the non-hairpin side. With a ground lead on R50 hairpin, the ohmmeeter reads 0 ohms at the pin just below the TX jumpers (R15). Relevant pictures here, ignore the green wire that ties in the new op-amp chip (the old was destroyed in the process of diagnosis). The other side of R15 reads 10 ohms (9.9 actually indicated), and the other side of C6 goes to infinity quickly after it charges, so the short has to be in the board itself, I think. I don't see any bridges. I'll pull the two components and see what happens.
          > >
          > > http://imgur.com/NIOL1
          > > http://imgur.com/rUzMv
          > >
          >
        • KF4BQ
          ... Bill, Don t take this bad: there appears to be some pieces of wire/solder near many of the solder connections in your pictures. Many use a Qtip with
          Message 4 of 9 , Nov 3, 2012
          • 0 Attachment
            --- In softrock40@yahoogroups.com, "wfahle" <billfahle@...> wrote:
            >
            > And, reassembling, the voltages now check out (2.5 everywhere but pin 8 and 4, which are 5v and 0v, and bridged with a resistor also gives the right voltages of 2.75 and 2.25 at 1 and 7). If you examine the pin in question in this picture (the after) you can now see a tiny trace of copper where the board defect was. I broke that trace with a voltmeter probe, and now it all works. Scratch one chip and one capacitor in the process. Hopefully there was a spare.
            >
            > http://imgur.com/oY6qI

            Bill,

            Don't take this bad: there appears to be some pieces of wire/solder near many of the solder connections in your pictures. Many use a Qtip with alcohol to clean up flux and junk. To weak for me; I use Acetone with a Qtip and you can really see any troubles after that. On some heavy flux areas I even use a small brush (trimmed down) to scrub the area. It might save you some future rework.

            73, Mike Collins KF4BQ
          • Alan
            ... Subject: [softrock40] Re: I think I found my so-called bad chip problem. ... Well done. You should have followed Robby s method, and told us the voltages
            Message 5 of 9 , Nov 3, 2012
            • 0 Attachment
              ----- Original Message -----
              Subject: [softrock40] Re: I think I found my so-called bad chip problem.


              > And, reassembling, the voltages now check out (2.5 everywhere but pin 8 and 4, which are 5v and 0v,

              Well done.

              You should have followed Robby's method, and told us the voltages before you did anything more. Zero volts on pin 2 would have
              indicated where the fault was.
              Zero volts on pin 2 is the reason for 5V on pin 1.

              73 Alan G4ZFQ
            • wfahle
              No offense taken; it s my first PCB. I did try rubbing alchohol but it made more of a mess. The whole board is a mess, but especially this part after all the
              Message 6 of 9 , Nov 3, 2012
              • 0 Attachment
                No offense taken; it's my first PCB. I did try rubbing alchohol but it made more of a mess. The whole board is a mess, but especially this part after all the diagnosis I went through to find the fault, including destroying a chip and soldering green wire onto it, etc. It will be a miracle if it works.

                --- In softrock40@yahoogroups.com, "KF4BQ" <mikecol@...> wrote:
                >
                > --- In softrock40@yahoogroups.com, "wfahle" <billfahle@> wrote:
                > >
                > > And, reassembling, the voltages now check out (2.5 everywhere but pin 8 and 4, which are 5v and 0v, and bridged with a resistor also gives the right voltages of 2.75 and 2.25 at 1 and 7). If you examine the pin in question in this picture (the after) you can now see a tiny trace of copper where the board defect was. I broke that trace with a voltmeter probe, and now it all works. Scratch one chip and one capacitor in the process. Hopefully there was a spare.
                > >
                > > http://imgur.com/oY6qI
                >
                > Bill,
                >
                > Don't take this bad: there appears to be some pieces of wire/solder near many of the solder connections in your pictures. Many use a Qtip with alcohol to clean up flux and junk. To weak for me; I use Acetone with a Qtip and you can really see any troubles after that. On some heavy flux areas I even use a small brush (trimmed down) to scrub the area. It might save you some future rework.
                >
                > 73, Mike Collins KF4BQ
                >
              • John
                Use a toothbrush and Isopropyl alcohol and the pcb will look like it just left the factory. John G3UGY From: wfahle Sent: Saturday, November 03, 2012 9:50 PM
                Message 7 of 9 , Nov 4, 2012
                • 0 Attachment
                  Use a toothbrush and Isopropyl alcohol and the pcb will look like it just left the factory.
                   
                  John G3UGY
                   

                  From: wfahle
                  Sent: Saturday, November 03, 2012 9:50 PM
                  Subject: [softrock40] Re: I think I found my so-called bad chip problem.

                   

                  No offense taken; it's my first PCB. I did try rubbing alchohol but it made more of a mess. The whole board is a mess, but especially this part after all the diagnosis I went through to find the fault, including destroying a chip and soldering green wire onto it, etc. It will be a miracle if it works.

                • William Fahle
                  Yes, I wish I had been more patient in trying things before starting to unsolder. In particular the solution came from continuity testing with the ohms turned
                  Message 8 of 9 , Nov 5, 2012
                  • 0 Attachment
                    Yes, I wish I had been more patient in trying things before starting
                    to unsolder. In particular the solution came from continuity testing
                    with the ohms turned up to 200k. This allowed me to see that certain
                    circuits were not open, but were in fact 21k, which led to the
                    diagnosis. In any case, I learned a lot doing this, so no regrets.

                    So far it passes every test; I now discover I need a second soundcard
                    to actually use the transmit. Also my primary soundcard is mono in, so
                    it is useless as a receiver card, so while I get semi-readable signals
                    from a backyard dipole, I'm not 100% sure what the receiver
                    performance will be yet, as there are too many variables. A quickly
                    thrown-up antenna, a bad soundcard, a new softrock build with problems
                    along the way (and no scope to check the output), default calibration,
                    all conspire to give me a big screen full of white noise with the
                    occasional discernable cw signal, and muddy signals that I can only
                    assume are ssb. I'm going to get a USB soundcard for the receiver side
                    and use the built-in for transmit and then report back. The built-in
                    definitely puts out stereo, so that shouldn't be an issue for
                    transmit.
                  • pond.james_pond
                    You re not alone. Took me several days of frustrated reworking before finding the etch failure. I felt less charitable than yourself at the time.
                    Message 9 of 9 , Nov 5, 2012
                    • 0 Attachment
                      You're not alone. Took me several days of frustrated reworking before finding the etch failure. I felt less charitable than yourself at the time.

                      --- In softrock40@yahoogroups.com, William Fahle <billfahle@...> wrote:
                      >
                      > Yes, I wish I had been more patient in trying things before starting
                      > to unsolder. In particular the solution came from continuity testing
                      > with the ohms turned up to 200k. This allowed me to see that certain
                      > circuits were not open, but were in fact 21k, which led to the
                      > diagnosis. In any case, I learned a lot doing this, so no regrets.
                      >
                      > So far it passes every test; I now discover I need a second soundcard
                      > to actually use the transmit. Also my primary soundcard is mono in, so
                      > it is useless as a receiver card, so while I get semi-readable signals
                      > from a backyard dipole, I'm not 100% sure what the receiver
                      > performance will be yet, as there are too many variables. A quickly
                      > thrown-up antenna, a bad soundcard, a new softrock build with problems
                      > along the way (and no scope to check the output), default calibration,
                      > all conspire to give me a big screen full of white noise with the
                      > occasional discernable cw signal, and muddy signals that I can only
                      > assume are ssb. I'm going to get a USB soundcard for the receiver side
                      > and use the built-in for transmit and then report back. The built-in
                      > definitely puts out stereo, so that shouldn't be an issue for
                      > transmit.
                      >
                    Your message has been successfully submitted and would be delivered to recipients shortly.