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

RXTX Ensemble, low output on TX. Repair of someone elses build.

Expand Messages
  • Chris Wilson
    07/02/2013 02:34 I have acquired an RXTX Ensemble a friend built, and although the receiver seems to work perfectly it has negligible TX output. He says he
    Message 1 of 14 , Feb 6, 2013
    View Source
    • 0 Attachment
      07/02/2013 02:34

      I have acquired an RXTX Ensemble a friend built, and although the
      receiver seems to work perfectly it has negligible TX output. He says
      he damaged the 2N2222 in the driver / output stage, and fitted one
      from his little box of assorted common transistors he had off Ebay. It
      *WAS* a 2N2222, but of unknown make and origin. Apart from that he
      can't recall anything of note. The voltage checks in the PA
      instructions check out OK, but on my scope the output into a dummy
      load, measured at the Ensemble antenna socket, at the board
      connectors, is very low, and it won't drive a watt meter to show
      anything at all. He's not very patient and has given up on it. The
      standard of soldering and component placement is excellent. I doubt
      there's that much wrong with it. I have double checked the values of
      the 2.2 ohm and up red / brown resistors in the PA stage, the ones
      very easily mixed up, but he has them correct. I need some help as how
      best to test it though, as the Softrocks I have built myself worked
      mainly OK, or if not, showed a localised issue by testing them at each
      stage of the build. I am not too good at diagnosing a fully built one,
      though. He claims all test were passed save for output level.

      Using USB on 20 meters (the lowest frequency band it's built for),
      using HDSDR, the app I am most familiar with, it needs full power
      output and full microphone gain, on the sliders, plus talking loudly
      right into the mike, to show much change on the scope. I can probably
      get a photo of the scope's screen to show the rather odd waveform. We
      are talking millivolts here, not volts, although I do believe it just
      once flickered into a stronger output, and I only just believe I saw
      it. My scope probes are not high frequency ones, although my scope is
      good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
      probe, or an RF Voltmeter (at least not one that works, my ham fest'
      bargain one showed the vagueness of the expression "needing slight attention"
      and is probably beyond economic repair ;))

      Any tips to narrow the issue much appreciated, I would like to fix it.
      Thanks.


      --
      Best Regards,
      Chris Wilson. 2E0ILY
      mailto: chris@...
    • John Greusel
      How did he damage that transistor and how did he know? Grounded the case against the board? I would inspect the transformer before the PA very carefully for
      Message 2 of 14 , Feb 6, 2013
      View Source
      • 0 Attachment
        How did he damage that transistor and how did he know?
        Grounded the case against the board? I would inspect the transformer before the PA very carefully for good and correct connections.

        John
        KC9OJV


         



        From: Chris Wilson <chris@...>
        To: softrock40@yahoogroups.com
        Sent: Wednesday, February 6, 2013 9:00 PM
        Subject: [softrock40] RXTX Ensemble, low output on TX. Repair of someone elses build.



          07/02/2013 02:34

        I have acquired an RXTX Ensemble a friend built, and although the
        receiver seems to work perfectly it has negligible TX output. He says
        he damaged the 2N2222 in the driver / output stage, and fitted one
        from his little box of assorted common transistors he had off Ebay. It
        *WAS* a 2N2222, but of unknown make and origin. Apart from that he
        can't recall anything of note. The voltage checks in the PA
        instructions check out OK, but on my scope the output into a dummy
        load, measured at the Ensemble antenna socket, at the board
        connectors, is very low, and it won't drive a watt meter to show
        anything at all. He's not very patient and has given up on it. The
        standard of soldering and component placement is excellent. I doubt
        there's that much wrong with it. I have double checked the values of
        the 2.2 ohm and up red / brown resistors in the PA stage, the ones
        very easily mixed up, but he has them correct. I need some help as how
        best to test it though, as the Softrocks I have built myself worked
        mainly OK, or if not, showed a localised issue by testing them at each
        stage of the build. I am not too good at diagnosing a fully built one,
        though. He claims all test were passed save for output level.

        Using USB on 20 meters (the lowest frequency band it's built for),
        using HDSDR, the app I am most familiar with, it needs full power
        output and full microphone gain, on the sliders, plus talking loudly
        right into the mike, to show much change on the scope. I can probably
        get a photo of the scope's screen to show the rather odd waveform. We
        are talking millivolts here, not volts, although I do believe it just
        once flickered into a stronger output, and I only just believe I saw
        it. My scope probes are not high frequency ones, although my scope is
        good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
        probe, or an RF Voltmeter (at least not one that works, my ham fest'
        bargain one showed the vagueness of the expression "needing slight attention"
        and is probably beyond economic repair ;))

        Any tips to narrow the issue much appreciated, I would like to fix it.
        Thanks.


        --
              Best Regards,
                          Chris Wilson.  2E0ILY
        mailto: chris@...



        ------------------------------------

        Yahoo! Groups Links

        <*> To visit your group on the web, go to:
            http://groups.yahoo.com/group/softrock40/

        <*> Your email settings:
            Individual Email | Traditional

        <*> To change settings online go to:
            http://groups.yahoo.com/group/softrock40/join
            (Yahoo! ID required)

        <*> To change settings via email:
            softrock40-digest@yahoogroups.com
            softrock40-fullfeatured@yahoogroups.com

        <*> To unsubscribe from this group, send an email to:
            softrock40-unsubscribe@yahoogroups.com

        <*> Your use of Yahoo! Groups is subject to:
            http://docs.yahoo.com/info/terms/



      • warrenallgyer
        Chris: You need a solid tone source for your testing. If you want to use HDSDR it is ok. Use it in CW mode, put it in TRANSMIT and hold down the RETURN key to
        Message 3 of 14 , Feb 6, 2013
        View Source
        • 0 Attachment
          Chris:

          You need a solid tone source for your testing. If you want to use HDSDR it is ok. Use it in CW mode, put it in TRANSMIT and hold down the RETURN key to generate a signal. I prefer to use Rocky for testing as you can select the "Tone" mode to make a stable signal... but HDSDR will work just fine. I think SSB modes are pretty inconvenient for this work.

          There are four points to check and you can quickly isolate the problem.

          Important! Use the proper ground for your probe! For these four tests you should ground your scope probe at the hairpin of R50 as a ground. For testing the output you must use the shield of the output connector.

          1) Collector of Q6 - just touch your scope probe to the case of the 2N2222 - you should see 6-10 Volts P-P. If you don't then the driver stage is bad.

          2) The high side of C24 in the output filter - you should see 20 volts or so P-P. If you have it at the driver and not here then the PA stage is bad..... likely a bad transformer soldering job but not necessarily.

          3) The high side of C25 - should still be in the neighborhood of 20 Volts P-P. If it was good to this point then look at soldering on L2.

          4) The high side of C26 - still around 20 Volts and, if not, then soldering on L3.

          If you pass all of these but still do not have a watt or better output then check the soldering and continuity of T6.

          Good luck! You have great tools and you can do this. Someone here will help if this does not locate the problem.

          Warren Allgyer - W8TOD

          --- In softrock40@yahoogroups.com, Chris Wilson wrote:
          >
          >
          >
          > 07/02/2013 02:34
          >
          > I have acquired an RXTX Ensemble a friend built, and although the
          > receiver seems to work perfectly it has negligible TX output. He says
          > he damaged the 2N2222 in the driver / output stage, and fitted one
          > from his little box of assorted common transistors he had off Ebay. It
          > *WAS* a 2N2222, but of unknown make and origin. Apart from that he
          > can't recall anything of note. The voltage checks in the PA
          > instructions check out OK, but on my scope the output into a dummy
          > load, measured at the Ensemble antenna socket, at the board
          > connectors, is very low, and it won't drive a watt meter to show
          > anything at all. He's not very patient and has given up on it. The
          > standard of soldering and component placement is excellent. I doubt
          > there's that much wrong with it. I have double checked the values of
          > the 2.2 ohm and up red / brown resistors in the PA stage, the ones
          > very easily mixed up, but he has them correct. I need some help as how
          > best to test it though, as the Softrocks I have built myself worked
          > mainly OK, or if not, showed a localised issue by testing them at each
          > stage of the build. I am not too good at diagnosing a fully built one,
          > though. He claims all test were passed save for output level.
          >
          > Using USB on 20 meters (the lowest frequency band it's built for),
          > using HDSDR, the app I am most familiar with, it needs full power
          > output and full microphone gain, on the sliders, plus talking loudly
          > right into the mike, to show much change on the scope. I can probably
          > get a photo of the scope's screen to show the rather odd waveform. We
          > are talking millivolts here, not volts, although I do believe it just
          > once flickered into a stronger output, and I only just believe I saw
          > it. My scope probes are not high frequency ones, although my scope is
          > good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
          > probe, or an RF Voltmeter (at least not one that works, my ham fest'
          > bargain one showed the vagueness of the expression "needing slight attention"
          > and is probably beyond economic repair ;))
          >
          > Any tips to narrow the issue much appreciated, I would like to fix it.
          > Thanks.
          >
          >
          > --
          > Best Regards,
          > Chris Wilson. 2E0ILY
          > mailto: chris@...
          >
        • Alan
          ... Subject: [softrock40] RXTX Ensemble, low output on TX. Repair of someone elses build. ... Chris, If you have a 500MHz scope then get some decent probes! Or
          Message 4 of 14 , Feb 6, 2013
          View Source
          • 0 Attachment
            ----- Original Message -----
            Subject: [softrock40] RXTX Ensemble, low output on TX. Repair of someone elses build.


            >
            > The voltage checks in the PA
            > instructions check out OK, but on my scope the output into a dummy
            > load, measured at the Ensemble antenna socket, at the board
            > connectors, is very low, and it won't drive a watt meter to show
            > anything at all. He's not very patient and has given up on it. The
            > standard of soldering and component placement is excellent. I doubt
            > there's that much wrong with it. I have double checked the values of
            > the 2.2 ohm and up red / brown resistors in the PA stage, the ones
            > very easily mixed up, but he has them correct. I need some help as how
            > best to test it though, as the Softrocks I have built myself worked
            > mainly OK, or if not, showed a localised issue by testing them at each
            > stage of the build. I am not too good at diagnosing a fully built one,
            > though. He claims all test were passed save for output level.
            >
            > Using USB on 20 meters (the lowest frequency band it's built for),
            > using HDSDR, the app I am most familiar with, it needs full power
            > output and full microphone gain, on the sliders, plus talking loudly
            > right into the mike,

            Chris,

            If you have a 500MHz scope then get some decent probes!
            Or at least make a diode probe. It is easier if you can get an idea of where the signal stops.
            Although putting your probes across a dummy load with a known 14MHz power fed into it should give a rough idea how it responds to
            that frequency. Unless you are confident the waveforms you see are true then just signal levels are all you need at the moment.
            As John says the transformers are often a problem.
            Warren gives figures for max output, here https://sites.google.com/site/g4zfqradio/levels-measured-in-a-softrock-tx are mine for 1/2
            watt starting at audio levels.

            As you have the makings of a good signal tracer make use of it. It is possible to diagnose without instruments but nothing like as
            easy.

            73 Alan G4ZFQ

            to show much change on the scope. I can probably
            > get a photo of the scope's screen to show the rather odd waveform. We
            > are talking millivolts here, not volts, although I do believe it just
            > once flickered into a stronger output, and I only just believe I saw
            > it. My scope probes are not high frequency ones, although my scope is
            > good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
            > probe, or an RF Voltmeter (at least not one that works, my ham fest'
            > bargain one showed the vagueness of the expression "needing slight attention"
            > and is probably beyond economic repair ;))
            >
          • Chris Wilson
            ... 07/02/2013 11:20 Thanks Alan, I have found a decent Marconi RF probe like the ones in RF voltmeters that a pal has no further use of, so I am buying it off
            Message 5 of 14 , Feb 7, 2013
            View Source
            • 0 Attachment
              > Chris,

              > If you have a 500MHz scope then get some decent probes!
              > Or at least make a diode probe. It is easier if you can get an idea of where the signal stops.
              > Although putting your probes across a dummy load with a known 14MHz
              > power fed into it should give a rough idea how it responds to
              > that frequency. Unless you are confident the waveforms you see are
              > true then just signal levels are all you need at the moment.
              > As John says the transformers are often a problem.
              > Warren gives figures for max output, here
              > https://sites.google.com/site/g4zfqradio/levels-measured-in-a-softrock-tx are mine for 1/2
              > watt starting at audio levels.

              > As you have the makings of a good signal tracer make use of it. It
              > is possible to diagnose without instruments but nothing like as
              > easy.

              > 73 Alan G4ZFQ


              07/02/2013 11:20

              Thanks Alan, I have found a decent Marconi RF probe like the ones in
              RF voltmeters that a pal has no further use of, so I am buying it off
              him. So I'll have a BNC connected RF probe very soon! Most of the 350
              to 500 MHz `scope probes seem to be X10 probes, will that be an
              issue? I have checked my probes against my TS-590 on 5 Watts, direct
              and via a 30dB 100 Watt capable attenuator. They read fine at 14 MHz,
              so the weird traces are down to the Softrock. Armed with your info,
              and the info Warren and John very kindly posted, I am hoping I can
              now get to work on this in a more effective manner. Thanks everyone!

              Oh, and John, not sure why he changed the 2N2222, I'll try and grab
              him tonight and see what the story was.

              --
              Best Regards,
              Chris Wilson. 2E0ILY
            • R. R. (Robby) Robson
              Warren: Great stuff. I have taken the liberty of referencing this thread - and especially your input herein - in the PA Stage documentation, RF Output Test
              Message 6 of 14 , Feb 7, 2013
              View Source
              • 0 Attachment
                Warren:

                Great stuff.  I have taken the liberty of referencing this thread - and especially your input herein - in the PA Stage documentation, RF Output Test section, to wit:
                "Also, if you have very low or no voltage output from this stage, check out the testing guidance from Warren Allgyer, W8TOD, in message 72129 on the Softrock40 reflector in Yahoo Groups.  He has some good test points and measurements to help in troubleshooting the low output power situation."

                On Wed, Feb 6, 2013 at 9:31 PM, warrenallgyer <allgyer@...> wrote:
                 

                Chris:

                You need a solid tone source for your testing. If you want to use HDSDR it is ok. Use it in CW mode, put it in TRANSMIT and hold down the RETURN key to generate a signal. I prefer to use Rocky for testing as you can select the "Tone" mode to make a stable signal... but HDSDR will work just fine. I think SSB modes are pretty inconvenient for this work.

                There are four points to check and you can quickly isolate the problem.

                Important! Use the proper ground for your probe! For these four tests you should ground your scope probe at the hairpin of R50 as a ground. For testing the output you must use the shield of the output connector.

                1) Collector of Q6 - just touch your scope probe to the case of the 2N2222 - you should see 6-10 Volts P-P. If you don't then the driver stage is bad.

                2) The high side of C24 in the output filter - you should see 20 volts or so P-P. If you have it at the driver and not here then the PA stage is bad..... likely a bad transformer soldering job but not necessarily.

                3) The high side of C25 - should still be in the neighborhood of 20 Volts P-P. If it was good to this point then look at soldering on L2.

                4) The high side of C26 - still around 20 Volts and, if not, then soldering on L3.

                If you pass all of these but still do not have a watt or better output then check the soldering and continuity of T6.

                Good luck! You have great tools and you can do this. Someone here will help if this does not locate the problem.

                Warren Allgyer - W8TOD



                --- In softrock40@yahoogroups.com, Chris Wilson wrote:
                >
                >
                >
                > 07/02/2013 02:34
                >
                > I have acquired an RXTX Ensemble a friend built, and although the
                > receiver seems to work perfectly it has negligible TX output. He says
                > he damaged the 2N2222 in the driver / output stage, and fitted one
                > from his little box of assorted common transistors he had off Ebay. It
                > *WAS* a 2N2222, but of unknown make and origin. Apart from that he
                > can't recall anything of note. The voltage checks in the PA
                > instructions check out OK, but on my scope the output into a dummy
                > load, measured at the Ensemble antenna socket, at the board
                > connectors, is very low, and it won't drive a watt meter to show
                > anything at all. He's not very patient and has given up on it. The
                > standard of soldering and component placement is excellent. I doubt
                > there's that much wrong with it. I have double checked the values of
                > the 2.2 ohm and up red / brown resistors in the PA stage, the ones
                > very easily mixed up, but he has them correct. I need some help as how
                > best to test it though, as the Softrocks I have built myself worked
                > mainly OK, or if not, showed a localised issue by testing them at each
                > stage of the build. I am not too good at diagnosing a fully built one,
                > though. He claims all test were passed save for output level.
                >
                > Using USB on 20 meters (the lowest frequency band it's built for),
                > using HDSDR, the app I am most familiar with, it needs full power
                > output and full microphone gain, on the sliders, plus talking loudly
                > right into the mike, to show much change on the scope. I can probably
                > get a photo of the scope's screen to show the rather odd waveform. We
                > are talking millivolts here, not volts, although I do believe it just
                > once flickered into a stronger output, and I only just believe I saw
                > it. My scope probes are not high frequency ones, although my scope is
                > good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
                > probe, or an RF Voltmeter (at least not one that works, my ham fest'
                > bargain one showed the vagueness of the expression "needing slight attention"
                > and is probably beyond economic repair ;))
                >
                > Any tips to narrow the issue much appreciated, I would like to fix it.
                > Thanks.
                >
                >
                > --
                > Best Regards,
                > Chris Wilson. 2E0ILY
                > mailto: chris@...
                >




                --
                Cheers,
                Robby
                 
                Richard R. (Robby) Robson
                LTC USA (RET)
              • John Williams
                Warren, Do you have a similar, well documented, way to test the receive section? I have a 6.3ng that has a i/q balance issue. The bugger is being able to probe
                Message 7 of 14 , Feb 7, 2013
                View Source
                • 0 Attachment
                  Warren,

                  Do you have a similar, well documented, way to test the receive section? I have a 6.3ng that has a i/q balance issue. The bugger is being able to probe those darn 805 SMDs. So tiny!

                  John - ke5ssh
                  On 2/6/2013 9:31 PM, warrenallgyer wrote:
                   

                  Chris:

                  You need a solid tone source for your testing. If you want to use HDSDR it is ok. Use it in CW mode, put it in TRANSMIT and hold down the RETURN key to generate a signal. I prefer to use Rocky for testing as you can select the "Tone" mode to make a stable signal... but HDSDR will work just fine. I think SSB modes are pretty inconvenient for this work.

                  There are four points to check and you can quickly isolate the problem.

                  Important! Use the proper ground for your probe! For these four tests you should ground your scope probe at the hairpin of R50 as a ground. For testing the output you must use the shield of the output connector.

                  1) Collector of Q6 - just touch your scope probe to the case of the 2N2222 - you should see 6-10 Volts P-P. If you don't then the driver stage is bad.

                  2) The high side of C24 in the output filter - you should see 20 volts or so P-P. If you have it at the driver and not here then the PA stage is bad..... likely a bad transformer soldering job but not necessarily.

                  3) The high side of C25 - should still be in the neighborhood of 20 Volts P-P. If it was good to this point then look at soldering on L2.

                  4) The high side of C26 - still around 20 Volts and, if not, then soldering on L3.

                  If you pass all of these but still do not have a watt or better output then check the soldering and continuity of T6.

                  Good luck! You have great tools and you can do this. Someone here will help if this does not locate the problem.

                  Warren Allgyer - W8TOD

                  --- In softrock40@yahoogroups.com, Chris Wilson wrote:
                  >
                  >
                  >
                  > 07/02/2013 02:34
                  >
                  > I have acquired an RXTX Ensemble a friend built, and although the
                  > receiver seems to work perfectly it has negligible TX output. He says
                  > he damaged the 2N2222 in the driver / output stage, and fitted one
                  > from his little box of assorted common transistors he had off Ebay. It
                  > *WAS* a 2N2222, but of unknown make and origin. Apart from that he
                  > can't recall anything of note. The voltage checks in the PA
                  > instructions check out OK, but on my scope the output into a dummy
                  > load, measured at the Ensemble antenna socket, at the board
                  > connectors, is very low, and it won't drive a watt meter to show
                  > anything at all. He's not very patient and has given up on it. The
                  > standard of soldering and component placement is excellent. I doubt
                  > there's that much wrong with it. I have double checked the values of
                  > the 2.2 ohm and up red / brown resistors in the PA stage, the ones
                  > very easily mixed up, but he has them correct. I need some help as how
                  > best to test it though, as the Softrocks I have built myself worked
                  > mainly OK, or if not, showed a localised issue by testing them at each
                  > stage of the build. I am not too good at diagnosing a fully built one,
                  > though. He claims all test were passed save for output level.
                  >
                  > Using USB on 20 meters (the lowest frequency band it's built for),
                  > using HDSDR, the app I am most familiar with, it needs full power
                  > output and full microphone gain, on the sliders, plus talking loudly
                  > right into the mike, to show much change on the scope. I can probably
                  > get a photo of the scope's screen to show the rather odd waveform. We
                  > are talking millivolts here, not volts, although I do believe it just
                  > once flickered into a stronger output, and I only just believe I saw
                  > it. My scope probes are not high frequency ones, although my scope is
                  > good, an analogue Tek 7854 that goes to 500 MHz. I don't have an RF
                  > probe, or an RF Voltmeter (at least not one that works, my ham fest'
                  > bargain one showed the vagueness of the expression "needing slight attention"
                  > and is probably beyond economic repair ;))
                  >
                  > Any tips to narrow the issue much appreciated, I would like to fix it.
                  > Thanks.
                  >
                  >
                  > --
                  > Best Regards,
                  > Chris Wilson. 2E0ILY
                  > mailto: chris@...
                  >


                  -- 
                  
                  John Williams
                  
                  KE5SSH - ham since 2007
                  WQKA523 - GMRS for family use on the farm
                  
                • Chris Wilson
                  ... 07/02/2013 15:28 Some great diagnostic info there! I have found a winding of the T4 transformer wasn t soldered properly. The PA FET s were not both
                  Message 8 of 14 , Feb 7, 2013
                  View Source
                  • 0 Attachment
                    >
                    > Chris:

                    > You need a solid tone source for your testing. If you want to use
                    > HDSDR it is ok. Use it in CW mode, put it in TRANSMIT and hold down
                    > the RETURN key to generate a signal. I prefer to use Rocky for
                    > testing as you can select the "Tone" mode to make a stable signal...
                    > but HDSDR will work just fine. I think SSB modes are pretty inconvenient for this work.

                    > There are four points to check and you can quickly isolate the problem.

                    > Important! Use the proper ground for your probe! For these four
                    > tests you should ground your scope probe at the hairpin of R50 as a
                    > ground. For testing the output you must use the shield of the output connector.

                    > 1) Collector of Q6 - just touch your scope probe to the case of the
                    > 2N2222 - you should see 6-10 Volts P-P. If you don't then the driver stage is bad.

                    > 2) The high side of C24 in the output filter - you should see 20
                    > volts or so P-P. If you have it at the driver and not here then the
                    > PA stage is bad..... likely a bad transformer soldering job but not necessarily.

                    > 3) The high side of C25 - should still be in the neighborhood of 20
                    > Volts P-P. If it was good to this point then look at soldering on L2.

                    > 4) The high side of C26 - still around 20 Volts and, if not, then soldering on L3.

                    > If you pass all of these but still do not have a watt or better
                    > output then check the soldering and continuity of T6.

                    > Good luck! You have great tools and you can do this. Someone here
                    > will help if this does not locate the problem.

                    > Warren Allgyer - W8TOD




                    07/02/2013 15:28

                    Some great diagnostic info there!

                    I have found a winding of the T4 transformer wasn't soldered properly.
                    The PA FET's were not both working, only one was. I now get a trace on
                    pressing the TX button in HDSDR. It is low level though, I'll link to
                    a trace photo at the end. It seems no drive level is being applied, as
                    there is 12.9 V DC on the case of Q6 (collector) no measurable RF on
                    the scope.

                    I need to see why no drive is being applied, either in CW mode with
                    the TX button pressed, and the ENTER key pressed, or in SSB mode with TX button pressed and
                    shouting into the microphone. The HDSR display responds to the Enter
                    key, or speaking into the mike though...

                    Thanks, will attempt driver circuit diagnosis next, any tips
                    appreciated! As usual, a great reflector, an much appreciating the
                    knowledge on here. A big thanks to Ed, too, for his help this AM.

                    http://www.chriswilson.tv/testing.zip

                    --
                    Best Regards,
                    Chris Wilson. 2E0ILY
                  • Alan
                    ... Subject: Re: [softrock40] Re: RXTX Ensemble, low output on TX. Repair of someone elses build. ... Chris, The easy way with HDSDR is to use AM or FM. Both
                    Message 9 of 14 , Feb 7, 2013
                    View Source
                    • 0 Attachment
                      ----- Original Message -----
                      Subject: Re: [softrock40] Re: RXTX Ensemble, low output on TX. Repair of someone elses build.


                      >
                      > I need to see why no drive is being applied, either in CW mode with
                      > the TX button pressed, and the ENTER key pressed, or in SSB mode with TX button pressed and
                      > shouting into the microphone. The HDSR display responds to the Enter
                      > key, or speaking into the mike though...
                      >

                      Chris,

                      The easy way with HDSDR is to use AM or FM. Both will output a constant carrier rather than keeping return pressed or shouting.

                      > Thanks, will attempt driver circuit diagnosis next, any tips
                      > appreciated!

                      Check the audio drive as on my web page.
                      See if there is any sign of output from the mixer.

                      73 Alan G4ZFQ
                    • Chris Wilson
                      ... 07/02/2013 20:41 I think I may have narrowed down another fault. I did as you suggested and used iqout to to input a signal of known level. I have taken
                      Message 10 of 14 , Feb 7, 2013
                      View Source
                      • 0 Attachment
                        > Chris,

                        > The easy way with HDSDR is to use AM or FM. Both will output a
                        > constant carrier rather than keeping return pressed or shouting.

                        >> Thanks, will attempt driver circuit diagnosis next, any tips
                        >> appreciated!

                        > Check the audio drive as on my web page.
                        > See if there is any sign of output from the mixer.

                        > 73 Alan G4ZFQ



                        07/02/2013 20:41

                        I think I may have narrowed down another fault. I did as you suggested
                        and used iqout to to input a signal of known level. I have taken
                        screen shots. the higher voltage trace is at the hairpins of R30 and
                        R31 (both give the same trace). At the T2 end of R30 and R31 I see the
                        low level trace. At the collector of Q6 I see the "messy" trace. Am I
                        right in suspecting T6? Any clue as to what to look for, or should I
                        just pull it and rewind it? R30 and R31 are correct values (49.9 Ohms)


                        http://www.chriswilson.tv/testing2.zip Scope shows volts /div and
                        time /div settings.

                        Thanks Alan. I am suspecting the signal is getting messed between the
                        TX Mixer and the driver transistor?

                        --
                        Best Regards,
                        Chris Wilson. 2E0ILY
                      • Chris Wilson
                        ... 07/02/2013 20:51 Oops, typo, I am suspecting T2 *NOT* T6!! -- Best Regards, Chris Wilson. 2E0ILY
                        Message 11 of 14 , Feb 7, 2013
                        View Source
                        • 0 Attachment
                          > 07/02/2013 20:41

                          > I think I may have narrowed down another fault. I did as you suggested
                          > and used iqout to to input a signal of known level. I have taken
                          > screen shots. the higher voltage trace is at the hairpins of R30 and
                          > R31 (both give the same trace). At the T2 end of R30 and R31 I see the
                          > low level trace. At the collector of Q6 I see the "messy" trace. Am I
                          > right in suspecting T6? Any clue as to what to look for, or should I
                          > just pull it and rewind it? R30 and R31 are correct values (49.9 Ohms)


                          > http://www.chriswilson.tv/testing2.zip Scope shows volts /div and
                          > time /div settings.

                          > Thanks Alan. I am suspecting the signal is getting messed between the
                          > TX Mixer and the driver transistor?



                          07/02/2013 20:51

                          Oops, typo, I am suspecting T2 *NOT* T6!!

                          --
                          Best Regards,
                          Chris Wilson. 2E0ILY
                        • warrenallgyer
                          Hi John These steps are based on the Softrock RXTX V6.3 - Xtall. If this is not what you have the same steps will work but you will need to translate the
                          Message 12 of 14 , Feb 7, 2013
                          View Source
                          • 0 Attachment
                            Hi John

                            These steps are based on the Softrock RXTX V6.3 - Xtall. If this is not what you have the same steps will work but you will need to translate the component numbers to equivalents in your version of Softrocks.

                            For testing the receive section I use my MFJ antenna analyzer to put a strong signal directly into the antenna input of the receiver. The trick is to get a signal strong enough that it makes measurable levels out of the receiver QSD mixer.

                            Set the MFJ or other signal generator 5-10 KHz above the crystal frequency of the 6.3. This will result in a 5-10 KHz audio signal out of the QSD mixer.

                            1) Measure the level of the OP Amp outputs at the hairpins of R55 and R56. The actual level should be between 1-5 Volts P-P depending upon the input level from the signal generator. The important thing is the levels at the two points should be roughly equal. If you can measure both at the same time with 2 probes you can check that one side should be 90 degrees out of phase with the other..... but the most important thing is roughly equal levels.

                            2) If Step 1 is ok then suspect a bad audio cable, a sound card setting for mono, or a sound card fault. If not then proceed backward in the chain with Step 3.

                            3)Measure the level of the 5-10 KHz audio signal at the hairpins of R49 and R50. The actual level will depend upon the signal generator level but it should be somewhere more than 50 millivolts P-P. The key thing is the levels should be approximately equal. Again, if you can measure two at the same time then look for them to be 90 degrees out of phase with each other. If the levels look good proceed backward in the chain with Step 4.

                            4) Measure the RF signal at the hairpins of R46 and R47. You should see more than 50 millivolts P-P, the same levels as measured in Step 3 but at RF frequency, not audio, and the two levels should be roughly equal. If one level is significantly higher than the other then suspect the connections or winding of transformer T100 on the RX Bandpass Module.

                            Again.... all Softrocks receivers have the same signal flow and this procedure will work on any of them...you just need to find the equivalent component designations in your particular model. Measurement at the hairpins is convenient and does not require probing the SMDs.

                            Good luck and come back if this does not help you.

                            Warren Allgyer - W8TOD



                            --- In softrock40@yahoogroups.com, John Williams wrote:
                            >
                            > Warren,
                            >
                            > Do you have a similar, well documented, way to test the receive section?
                            > I have a 6.3ng that has a i/q balance issue. The bugger is being able to
                            > probe those darn 805 SMDs. So tiny!
                            >
                            > John - ke5ssh
                            > On 2/6/2013 9:31 PM, warrenallgyer wrote:
                            > >
                          • warrenallgyer
                            Hi Chris You are right in suspecting T2. I am guessing one of the two primaries is reversed. There should not be so much voltage drop across R30/31. I would
                            Message 13 of 14 , Feb 7, 2013
                            View Source
                            • 0 Attachment
                              Hi Chris

                              You are right in suspecting T2. I am guessing one of the two primaries is reversed. There should not be so much voltage drop across R30/31.

                              I would pull it, rewind, and reinstall.

                              Hint: two good ways to clear the PCB holes for the re-installation of T2: First and fastest is a "solder sucker". Heat the board from one side and use the sucker to remove the solder from the hole from the opposite side.

                              If you don't have a sucker then you can use a piece of the #24-26 enamel wire used for the coils. Heat the hole from one side and push the wire through. Keep the heat on and keep pushing until it has dragged the solder out the other side. The solder will not stick to the enamel and you will have a clear hole for the new wires.

                              Warren Allgyer - W8TOD
                              --- In softrock40@yahoogroups.com, Chris Wilson wrote:
                              >
                              >
                              >
                              > > 07/02/2013 20:41
                              >
                              > > I think I may have narrowed down another fault. I did as you suggested
                              > > and used iqout to to input a signal of known level. I have taken
                              > > screen shots. the higher voltage trace is at the hairpins of R30 and
                              > > R31 (both give the same trace). At the T2 end of R30 and R31 I see the
                              > > low level trace. At the collector of Q6 I see the "messy" trace. Am I
                              > > right in suspecting T6? Any clue as to what to look for, or should I
                              > > just pull it and rewind it? R30 and R31 are correct values (49.9 Ohms)
                              >
                              >
                              > > http://www.chriswilson.tv/testing2.zip Scope shows volts /div and
                              > > time /div settings.
                              >
                              > > Thanks Alan. I am suspecting the signal is getting messed between the
                              > > TX Mixer and the driver transistor?
                              >
                              >
                              >
                              > 07/02/2013 20:51
                              >
                              > Oops, typo, I am suspecting T2 *NOT* T6!!
                              >
                              > --
                              > Best Regards,
                              > Chris Wilson. 2E0ILY
                              >
                            • Chris Wilson
                              ... 07/02/2013 22:27 Thanks Warren, I think we have another fix here :) Hopefully that s all that s wrong with it. Thank you for the ideas for removing the
                              Message 14 of 14 , Feb 7, 2013
                              View Source
                              • 0 Attachment
                                >

                                > Hi Chris

                                > You are right in suspecting T2. I am guessing one of the two
                                > primaries is reversed. There should not be so much voltage drop across R30/31.

                                > I would pull it, rewind, and reinstall.

                                > Hint: two good ways to clear the PCB holes for the re-installation
                                > of T2: First and fastest is a "solder sucker". Heat the board from
                                > one side and use the sucker to remove the solder from the hole from the opposite side.

                                > If you don't have a sucker then you can use a piece of the #24-26
                                > enamel wire used for the coils. Heat the hole from one side and push
                                > the wire through. Keep the heat on and keep pushing until it has
                                > dragged the solder out the other side. The solder will not stick to
                                > the enamel and you will have a clear hole for the new wires.

                                > Warren Allgyer - W8TOD



                                07/02/2013 22:27

                                Thanks Warren, I think we have another fix here :)
                                Hopefully that's all that's wrong with it. Thank you for the ideas for
                                removing the transformer, I spoilt myself last Christmas and invested
                                (and that is definitely the right description) in a proper Pace
                                de-soldering station. I saw an acquaintance using one on a multi layer
                                PCB and was hooked. I expressed my sentiments and the old whatsit went
                                out of his shack and produced another, at a price too tempting to
                                refuse. Best thing I bought all year, awesome it of kit! I love
                                quality gear, and so long as you use these things, what the hell, they
                                have good residuals, and make life a bit less of a pain :)

                                I got the T2 off intact, lead wise, and as far as I could tell the
                                leads were crossed. I could perhaps of refitted it correctly, but the
                                leads were a bit short, so I rewound it entirely.

                                I have scope shots of it now, hopefully this is how it should be, will
                                test with a microphone next, and see about checking for imaging.

                                Thanks to all who have assisted, not as painful a fix as I feared. If
                                it *IS* indeed all OK now....

                                http://www.chriswilson.tv/testing3.zip Note the changing
                                volts/division settings.

                                --
                                Best Regards,
                                Chris Wilson. 2E0ILY
                              Your message has been successfully submitted and would be delivered to recipients shortly.