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

Getting more than 1W out

Expand Messages
  • sparxx@btinternet.com
    I am making the odd QSO now with my ensemble on 40m having sorted out the software issues I was having. I think the output power could do with boosting though
    Message 1 of 10 , Jul 13, 2014
    • 0 Attachment

      I am making the odd QSO now with my ensemble on 40m having sorted out the software issues I was having. I think the output power could do with boosting though as some people can't quite get all my callsign when I reply to their calls.


      Is there a simplish PA stage mod to get more power? If the PA stage had a separate supply, higher than 12V say, would that work? I am not a PA designer and have just started to look at the schematic.


      I know that there are external PA's with rx/tx switching if there is no internal mod possible.


      Thanks


      Lucien 

      G4SYI

    • warrenallgyer
      Lucien You can get more power out of the RXTX by running it at 13.8 volts or even higher. The electrolytics in the power supply are rated at 16V so that is
      Message 2 of 10 , Jul 13, 2014
      • 0 Attachment
        Lucien

        You can get more power out of the RXTX by running it at 13.8 volts or even higher. The electrolytics in the power supply are rated at 16V so that is probably a practical limit. My tests indicate the signal is cleaner when the rig is run at a higher voltage as well.

        But you need to be realistic about your expectations. If you are able to double the power out that will result in one half S unit increase in received signal on the other end and will not likely contribute much to intelligibility. Raising the power to 10 watts would ad a full S unit plus. Again, not likely a significant improvment.

        The single factor that will most improve SSB for the RXTX is the addition of a speech processor. Unlike standard SSB transmitters, the simple RXTX has no speech processing or ALC so the average voice power is very low and the voice peaks easily overmodulate the transmitter and cause splatter.

        I recommend you try to use Voice Shaper (google it), a free, inline speech processor. Or you can purchase an external processor. Either one will tremendously improve your SSB penetrating power and raise your average power without overloading or raising the peaks.

        I have posted a video demonstration which clearly shows how Voice Shaper improves RXTX SSB here:
        Optimizing Softrock RXTX SSB Performance

         

        Warren Allgyer
        9V1TD
      • ed.murphy31
        Lucien, Have a look at my modifications described in the files section for a total transformation of the softrock. Files GM3SBC QRO Modifications. Regards
        Message 3 of 10 , Jul 14, 2014
        • 0 Attachment
          Lucien,

          Have a look at my modifications described in the files section for a  total transformation of the softrock.

          Files > GM3SBC > QRO Modifications.

          Regards
          Ed Murphy GM3SBC

        • Chris Wilson
          Re: [softrock40] Re: Getting more than 1W out This mod of Ed s works very well, and is well documented, so long as you don t over modulate the SR it sounds
          Message 4 of 10 , Jul 14, 2014
          • 0 Attachment
            Re: [softrock40] Re: Getting more than 1W out This mod of Ed's works very well, and is well documented, so long as you don't over modulate the SR it sounds fine on air with good panadptor display reports.  I was surprised it never really caught on, it's a cheap way to some more power!




            Best Regards,
                               Chris Wilson.    2E0ILY (UK)


             

              
            Lucien,

            Have a look at my modifications described in the files section for a  total transformation of the softrock.

            Files > GM3SBC > QRO Modifications.

            Regards
            Ed Murphy GM3SBC

          • sparxx@btinternet.com
            Thanks for that, I didn t know this mod existed until now. I am getting excited about doing it. Cheers Lucien G4SYI
            Message 5 of 10 , Jul 14, 2014
            • 0 Attachment
              Thanks for that, I didn't know this mod existed until now. I am getting excited about doing it.

              Cheers

              Lucien
              G4SYI
            • ed.murphy31
              Lucien, I have worked the world since I incorporated this modification. BTW the Flex 2.4.4 software has all the facilities you need regarding getting punch
              Message 6 of 10 , Jul 14, 2014
              • 0 Attachment
                Lucien,

                I have worked the world since I incorporated  this modification.

                BTW the Flex 2.4.4 software  has all the facilities you need regarding getting punch on speech. Please read the Flex manual. You do not need any other software for your audio.

                If you go through this simple modification carefully you shall be rewarded with a cracking little rig. I have had so many reports complimenting me on the quality and strength of the signal. In the first instance you do not need to have the band switching of the filters. Just fit the appropriate values for one band and check it out for performance, then you will be keen to do the remaining . I can send you a band changing switch should you require it.


                Best Regards

                Ed Murphy GM3SBC.


              • warrenallgyer
                Ed I have admired your ingenuity and workmanship in adding this QRO mod to the Softrocks. I would like to make you a friendly bet however. I am betting that
                Message 7 of 10 , Jul 14, 2014
                • 0 Attachment
                  Ed

                  I have admired your ingenuity and workmanship in adding this QRO mod to the Softrocks. I would like to make you a friendly bet however. I am betting that you will get better signal reports running it at one watt and using the Flex speech processor than you will running it at full power with the speech processor off.

                  The standard RXTX has plenty of power to work the world. I do it every day on 40 meter WSPR and, on 20 meters I routinely do it with 10 dB less than one watt. With a reasonable antenna that is enough power to "get the signal around the world". The challenge then is to make it intelligible. WSPR does this with digital ears taking advantage of forward error correction.

                  In SSB we have only analog ears and the analog equivalent of FEC which is speech processing.

                  I am in no way criticizing those who want more power, especially done is such a professional and ingenious way as you have. My personal objective in using these radios was to leave the hardware as simple as possible and to enhance the performance on the software side.

                  Lucien, operating with HDSDR and no speech processing, is leaving probably 6-10 dB of performance on the table and looking to make up for it by complicating his hardware. To each his own but this would not consistent with what I personally want to achieve with these radios.

                  I am sure your signal sounds great by combining a little more power with speech processing. That is the best of both worlds. I just don't want all to believe that they have to go to the hardware in order to achieve startling SSB performance improvement from the Softrocks.

                  Warren Allgyer
                  9V1TD
                • sparxx@btinternet.com
                  I have had a look at what I would need to do to modify my 80/40m ensemble to the 10W mod state on 20/30/40m. I don t hear much on 80m so sacrificing it for a
                  Message 8 of 10 , Jul 20, 2014
                  • 0 Attachment
                    I have had a look at what I would need to do to modify my 80/40m ensemble to the 10W mod state on 20/30/40m. I don't hear much on 80m so sacrificing it for a higher band is what I want to do. 

                    My unmodified rig also does 160m and 60m with almost no harmonics as far as I can tell with an external rx. 

                    The mod seems do-able even with my soldering skills. There are some potential issues for me though:

                    There is 'gunk' (probably hot melt glue) on most of the toroids to stick them to the pcb. The gunk covers some of the windings and tracks also. I think I may be able to get it off with rubbing alcohol/soldering iron heat. I don't want to pull tracks off the board or break toroids.

                    I can't get Farnell parts as I don't have a trade account with them. So that switch and fan would need to be designed out or maybe I could get them off Ed. I thought of using a bigger fan on the top of the enclosure but it would look a bit ugly. I could start with no switching as mentioned already.

                    I use data modes mostly rather than speech so the extra oomph from pumping up the power should help.

                    I wonder what the limiting factor(s) would be to simply turning up the input voltage?
                    The 5V regulator can go to 30V, the various electrolytics to 16V. If those caps were changed to 20V or higher (if they still fit) and if the 5V regulator can take the higher dissipation then would that be an acceptable easy mod?

                    Another idea I had is to (somehow) increase the peak voltage output from the soundcard.

                    Just some throw away thoughts for discussion.

                    Thanks,

                    Lucien 
                    G4SYI
                  • Nick Norman
                    ... Gentle use of a hot air gun ? ... You don t need one (I don t think) just register and away you go, or use CPC (who are not farnell - honest :-) ) I also
                    Message 9 of 10 , Jul 20, 2014
                    • 0 Attachment
                      On 20/07/14 13:20, sparxx@... [softrock40] wrote:
                      > I have had a look at what I would need to do to modify my 80/40m ensemble to the 10W mod state on 20/30/40m. I don't hear much on 80m so sacrificing it for a higher band is what I want to do.
                      >
                      > My unmodified rig also does 160m and 60m with almost no harmonics as far as I can tell with an external rx.
                      >
                      > The mod seems do-able even with my soldering skills. There are some potential issues for me though:
                      >
                      >
                      >
                      > There is 'gunk' (probably hot melt glue) on most of the toroids to stick them to the pcb. The gunk covers some of the windings and tracks also. I think I may be able to get it off with rubbing alcohol/soldering iron heat. I don't want to pull tracks off the board or break toroids.

                      Gentle use of a hot air 'gun'?
                      >
                      > I can't get Farnell parts as I don't have a trade account with them. So that switch and fan would need to be designed out or maybe I could get them off Ed. I thought of using a bigger fan on the top of the enclosure but it would look a bit ugly. I could start with no switching as mentioned already.

                      You don't need one (I don't think) just register and away you go, or use
                      CPC (who are not farnell - honest :-) )

                      I also use RS occasionally again just register your credit card is just
                      as good as anyone else's

                      Or try Rapid-online



                      > I use data modes mostly rather than speech so the extra oomph from pumping up the power should help.

                      > I wonder what the limiting factor(s) would be to simply turning up the input voltage?
                      > The 5V regulator can go to 30V, the various electrolytics to 16V. If those caps were changed to 20V or higher (if they still fit) and if the 5V regulator can take the higher dissipation then would that be an acceptable easy mod

                      Not tried but confess I'd thought about it
                      >
                      > Another idea I had is to (somehow) increase the peak voltage output from the soundcard.

                      Watch out or various distortions, especially if you've not increased the
                      voltages - remember you can not increase the voltage to the
                      logic/analogue stuff just the PA, so likely not much use.

                      > Just some throw away thoughts for discussion.
                      >
                      >
                      > Thanks,
                      >
                      >
                      > Lucien
                      > G4SYI
                      >
                      Just my 2ds worth

                      Nick

                      MoHUG

                      --
                      "I'd like to know if I could compare you to a summer's
                      day. Because, well, June 12th was quite nice, and..."
                      (Wyrd Sisters)
                      15:15:01 up 1 day, 19:58, 5 users, load average: 0.02, 0.04, 0.05



                      --
                      In the fetid fleapit of Rincewind's brain the projectionist
                      of memory put on reel two. Recollection began to flicker.
                      (The
                      Last Continent)
                      15:55:01 up 1 day, 20:38, 5 users, load average: 0.01, 0.22, 0.30
                    • warrenallgyer
                      Lucien My unmodified rig also does 160m and 60m with almost no harmonics as far as I can tell with an external rx. I recommend you check this statement
                      Message 10 of 10 , Jul 20, 2014
                      • 0 Attachment
                        Lucien

                        "My unmodified rig also does 160m and 60m with almost no harmonics as far as I can tell with an external rx. "

                        I recommend you check this statement again. The standard 80/40 build of RXTX suppresses the 80 meter second harmonic less than 20 dB. That is why there has been a recommendation to use an external low pass filter and the components to build one are supplied with the kit. The filter is absolutely required in order to meet the 40 dB harmonic suppression requirement.

                        Operating the 80/40 RXTX on 160 meters will have even less second harmonic suppression and will also have a additional problem. The inter-stage transformer and coile values are such that distortion levels are likely to be well outside the required specifications if used on 160.

                        Measurement of harmonics and distortion to the levels that are legally required is going to take more than just "ear-balling" it. Unless you have the necessary instruments to actually measure distortion and harmonics, you should probably hold to the recommended bands and practices in the instructions.

                        Warren Allgyer
                        9V1TD
                      Your message has been successfully submitted and would be delivered to recipients shortly.