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

Re: Tx Problem

Expand Messages
  • Dave
    ... Also, the software doesn t simply put out continuous Tx I and Q signals in CW mode. Keying waveform shaping is done on the I and Q, while PTT remains
    Message 1 of 8 , Mar 2, 2007
    • 0 Attachment
      --- In softrock40@yahoogroups.com, "Dave" <dmertz@...> wrote:
      > the software needs to know that it is
      > to output the Tx I and Q signals appropriate for your operating mode
      > and frequency.

      Also, the software doesn't simply put out continuous Tx I and Q
      signals in CW mode. Keying waveform shaping is done on the I and Q,
      while PTT remains active until keying stops for long enough and the
      system returns to Rx.

      Dave, WA3OFF
    • Tony Parks
      Hi Colin, With an SDR type transmitter as in the RXTX all modes of tranmission, including CW, require quadrature audio signals from the soundcard line-out.
      Message 2 of 8 , Mar 2, 2007
      • 0 Attachment
        Hi Colin,
         
        With an SDR type transmitter as in the RXTX all modes of tranmission, including CW, require quadrature audio signals from the soundcard line-out.  For CW two fixed frequency keyed quadrature sine waves are required where the frequency of the sine waves is equal to the frequency offset from the center frequency and the phase between the two sinewaves is + or - 90 degrees depending on if the RF output is to be above or below the center frequency.
         
        Thus the software needs to respond to key inputs to generate the quadrature sinewaves and a serial port input is a high speed way of connecting the key to the PC. 
         
        73,
        Tony KB9YIG
         
        ----- Original Message -----
        From: bty230314
        Sent: Friday, March 02, 2007 10:40 AM
        Subject: [softrock40] Tx Problem

        First let me add my thanks to Tony and others who have supported him
        in the SoftRock project. I know its been said before, but the
        program has allowed me an insight into SDR technology, including the
        satisfaction of having built working Rxs.

        Now to the problem, --where my limited knowledge is letting me down.

        I have built the Rev6 Tx/Rx 40/80 M option, and the Rx is playing
        fine with Rocky and WinRad software, I have a good quiet baseline
        and can pick out CW and SSB signals easily. Not so good with M0KGK
        software, as the baseline is very noisy, and the signals quite
        difficult to pick out. However with care, I can pick up CW sigs. I
        suspect a set up problem.

        I have set up the Tx by adjusting R29 to give approx 50mV at R30 and
        R31. I expected that if I apply 12V to PTR input, the Tx should be
        transmitting into my dummy load, but I see no rf sig at the ant on
        my scope. I thought this would simulate a CW sig., which is my
        intended mode of operation.

        Probably I have the theory wrong. As my next question would have
        been, Why not have the CW keyer at the PTT input, rather than the
        COMM port ???

        I also have a hardware problem as COMM1 port does not function. Is
        there a way to change the set up to use COMM2 port for the key???

        Any comments would be appreciated

        Colin M3WCK

      • cesco12342000
        ... I wonder if this is a good concept. There are numerous problems with the low-latency required for this type of keying. This low-latency stuff is
        Message 3 of 8 , Mar 2, 2007
        • 0 Attachment
          > Keying waveform shaping is done on the I and Q,
          > while PTT remains active until keying stops

          I wonder if this is a good concept. There are numerous problems with
          the low-latency required for this type of keying. This low-latency
          stuff is responsible for quite some hiccups on the software side.

          A better thing would be to FORGET the shaping and key the tx
          directly, I and Q signals always on. The rx soundcard could be active
          while transmitting, allowing to listen in between the dits. I think
          this would work much better than to route those signals trough a PC
          with a lame non-realtime OS.
        • Lyle Koehler, K0LR
          My computer is a couple of years old and has an AMD Athlon 3200 processor. The sound card I use for I/Q input and output is an Audigy 2 NX USB card, and I run
          Message 4 of 8 , Mar 2, 2007
          • 0 Attachment
            My computer is a couple of years old and has an AMD Athlon 3200
            processor. The sound card I use for I/Q input and output is an Audigy
            2 NX USB card, and I run everything at 96 kHz, supposedly 24 bits. If
            anything, I would expect the USB connection on my card to impose an
            additional speed penalty on the system. To test the system, I loaded
            up the memory in a K1EL K10+ keyer with a test message and played it
            back into the RXTXv6.1 at various speeds. The speed is "software"
            selected by entering it with the paddles. Although I didn't check it
            with a scope, the keying sounded fine in another receiver at 35 wpm
            and even at 50 wpm (to the extent that I can copy at that speed!).

            The whole point of software defined radio is to perform as many of the
            radio's analog functions as possible in software, to minimize the
            complexity of the hardware. "Soft" keying is one of those functions,
            and I think it would be difficult to come up with a simple hardware
            implementation that works as well as the CW shaping in KGKSDR.

            Lyle, K0LR

            --- In softrock40@yahoogroups.com, "cesco12342000" <cesco1@...> wrote:
            >
            >
            > A better thing would be to FORGET the shaping and key the tx
            > directly, I and Q signals always on.
          • Ford Peterson
            Colin wrote: ...snip... I have set up the Tx by adjusting R29 to give approx 50mV at R30 and R31. I expected that if I apply 12V to PTR input, the Tx should be
            Message 5 of 8 , Mar 2, 2007
            • 0 Attachment
              Colin wrote:

              ...snip...


              I have set up the Tx by adjusting R29 to give approx 50mV at R30 and R31. I
              expected that if I apply 12V to PTR input, the Tx should be transmitting
              into my dummy load, but I see no rf sig at the ant on my scope. I thought
              this would simulate a CW sig., which is my intended mode of operation.

              Probably I have the theory wrong. As my next question would have been, Why
              not have the CW keyer at the PTT input, rather than the COMM port ???

              I also have a hardware problem as COMM1 port does not function. Is there a
              way to change the set up to use COMM2 port for the key???

              Any comments would be appreciated

              Colin M3WCK

              ************************

              Colin, and others...

              I cannot help with the software issues as I have an issue of my own. I
              dislike involving the comm port in the first place as the PC has always had
              problematic serial port issues. Now with the ubiquitous USB ports, using
              serial port ops is a virtual no-no. I did make an observation regarding the
              R30/R31 measurements and setting R29 that was interesting to me.

              When I set R29 according to the instructions, it took some fussing with the
              pot to get them just right. What I did notice is that I could hear the
              opposite side band.

              For example, set to USB and listen to the transmitter using another
              receiver. If you tune down to zero beat, and go past zero to the opposite
              sideband, I could hear it just fine. Setting R29 according to the manual
              does not produce as good a null as when you tune for minimum distortion of
              the sidebands. In fact, it was very easy to set R29 to almost null out the
              opposite sideband. What was interesting is that the setting of R29 using my
              method did not result in the same setting as per the instructions.

              I can only speculate as to why this is so. Perhaps the two transistors are
              not well matched? Differences in the balance of the transformers? I don't
              know for certain. I set mine for minimum spurious response instead of using
              the volt meter on R30/R31.

              Ford-N0FP
              ford@...
            • bty230314
              ... on ... mode ... Is
              Message 6 of 8 , Mar 4, 2007
              • 0 Attachment
                --- In softrock40@yahoogroups.com, "Dave" <dmertz@...> wrote:
                >
                > --- In softrock40@yahoogroups.com, "bty230314" <cweaving@> wrote:
                > > I expected that if I apply 12V to PTR input, the Tx should be
                > > transmitting into my dummy load, but I see no rf sig at the ant
                on
                > > my scope.
                >
                > You also need the Tx I and Q signals supplied from the sound card
                > to get an RF output. Just applying 12V to the PTT input puts the
                > rig in Tx mode with no drive.
                >
                > > Why not have the CW keyer at the PTT input, rather than the
                > > COMM port ???
                >
                > Because, as mentioned above, the software needs to know that it is
                > to output the Tx I and Q signals appropriate for your operating
                mode
                > and frequency.
                >
                > > I also have a hardware problem as COMM1 port does not function.
                Is
                > > there a way to change the set up to use COMM2 port for the key???
                >
                > Although I have not tried it, I believe KGKSDR allows setting the
                > port to COM2 on the Transceiver Options->Hardware tab (in the Ports
                > block). Winrad and the current version of Rocky do not support Tx
                > operation.
                >
                > Dave, WA3OFF
                >
              Your message has been successfully submitted and would be delivered to recipients shortly.