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

Re: [powersdr-iq] Re: Image

Expand Messages
  • Pete Ferrand
    Hi Andrew: Thanks for the reply. Yes, that s exactly what I m seeing. And yes, I did see your previous postings on this but it didn t sink in. I don t have a
    Message 1 of 17 , Jun 1, 2010
    • 0 Attachment
      Hi Andrew:

      Thanks for the reply. Yes, that's exactly what I'm seeing. And yes, I did see your previous postings on this but it didn't sink in. I don't have a mobo and won't until they're available again I guess.

      The image is, in fact, symmetrical around the hump.

      Anyway on the powersdr-iq screen I have the bottom of the range is -90 db and a strong signal is -40...if I run the sig gen there's a real "tipping point" at -10 when the image suddenly vanishes. Wow, that's amazing.

      Of course this procedure has to be repeated whenever I switch bands, etc.

      Thanks for discovering this and setting up the test gear! Should be a solvable problem.

      -Pete






      -----Original Message-----
      >From: irbsurfing <irbsurfing@...>
      >Sent: Jun 1, 2010 5:21 PM
      >To: powersdr-iq@yahoogroups.com
      >Subject: [powersdr-iq] Re: Image
      >
      >Hello Pete,
      >Just seen your posting and I think you may be seeing similar things to me. The images you are seeing I think are consistant if you take into account the 9kHz offset that that is applied to the tuned frequency.
      >Basically if can see the usual noise hump in the middle of the panadapter the images should be symetrically placed either side of it.
      >Have you tried increasing the signal level as I find that the image drops to nothing with an input of about -40 to -45dBm?
      >Andrew
      >
      >--- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@...> wrote:
      >>
      >>
      >> I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
      >>
      >> The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
      >>
      >> With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
      >>
      >> With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
      >>
      >> I am running Windows 7 32 bit with a Delta-44.
      >>
      >> Thank you for any help!
      >>
      >> -Pete
      >> WB2QLL
      >> Somers, WI
      >>
      >
      >
      >
      >
      >------------------------------------
      >
      >Yahoo! Groups Links
      >
      >
      >
    • Graeme
      I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to
      Message 2 of 17 , Jun 12, 2010
      • 0 Attachment
        I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.

        73 Graeme ZL2APV

        --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@...> wrote:
        >
        >
        > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
        >
        > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
        >
        > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
        >
        > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
        >
        > I am running Windows 7 32 bit with a Delta-44.
        >
        > Thank you for any help!
        >
        > -Pete
        > WB2QLL
        > Somers, WI
        >
      • g8voip
        Hi Graeme, Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds. As far as I know it is done
        Message 3 of 17 , Jun 12, 2010
        • 0 Attachment
          Hi Graeme,

          Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.

          As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.

          If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!

          I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.

          For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.

          On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.

          Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.

          Once that process has been 'forced' all appears well.

          There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.

          73, Bob G8VOI


          --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@...> wrote:
          >
          > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
          >
          > 73 Graeme ZL2APV
          >
          > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
          > >
          > >
          > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
          > >
          > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
          > >
          > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
          > >
          > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
          > >
          > > I am running Windows 7 32 bit with a Delta-44.
          > >
          > > Thank you for any help!
          > >
          > > -Pete
          > > WB2QLL
          > > Somers, WI
          > >
          >
        • sv1eia
          Hi Bob, I m using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I ve placed that usbtoi2c board that I have
          Message 4 of 17 , Jun 12, 2010
          • 0 Attachment
            Hi Bob,

            I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).

            In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.

            In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.

            Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.

            73,
            Christos SV1EIA



            --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@...> wrote:
            >
            > Hi Graeme,
            >
            > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
            >
            > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
            >
            > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
            >
            > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
            >
            > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
            >
            > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
            >
            > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
            >
            > Once that process has been 'forced' all appears well.
            >
            > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
            >
            > 73, Bob G8VOI
            >
            >
            > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
            > >
            > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
            > >
            > > 73 Graeme ZL2APV
            > >
            > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
            > > >
            > > >
            > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
            > > >
            > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
            > > >
            > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
            > > >
            > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
            > > >
            > > > I am running Windows 7 32 bit with a Delta-44.
            > > >
            > > > Thank you for any help!
            > > >
            > > > -Pete
            > > > WB2QLL
            > > > Somers, WI
            > > >
            > >
            >
          • g8voip
            Hi Christos, It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but
            Message 5 of 17 , Jun 13, 2010
            • 0 Attachment
              Hi Christos,

              It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!

              I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.

              I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!

              It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.

              v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.

              I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!

              He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.

              Thanks again,

              Bob G8VOI


              --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@...> wrote:
              >
              > Hi Bob,
              >
              > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
              >
              > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
              >
              > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
              >
              > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
              >
              > 73,
              > Christos SV1EIA
              >
              >
              >
              > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
              > >
              > > Hi Graeme,
              > >
              > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
              > >
              > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
              > >
              > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
              > >
              > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
              > >
              > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
              > >
              > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
              > >
              > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
              > >
              > > Once that process has been 'forced' all appears well.
              > >
              > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
              > >
              > > 73, Bob G8VOI
              > >
              > >
              > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
              > > >
              > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
              > > >
              > > > 73 Graeme ZL2APV
              > > >
              > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
              > > > >
              > > > >
              > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
              > > > >
              > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
              > > > >
              > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
              > > > >
              > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
              > > > >
              > > > > I am running Windows 7 32 bit with a Delta-44.
              > > > >
              > > > > Thank you for any help!
              > > > >
              > > > > -Pete
              > > > > WB2QLL
              > > > > Somers, WI
              > > > >
              > > >
              > >
              >
            • sv1eia
              Hi Bob, Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex. If I remember correctly, that
              Message 6 of 17 , Jun 13, 2010
              • 0 Attachment
                Hi Bob,

                Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.

                If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15

                Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)

                73,
                Christos SV1EIA




                --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@...> wrote:
                >
                > Hi Christos,
                >
                > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                >
                > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                >
                > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                >
                > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                >
                > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                >
                > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                >
                > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                >
                > Thanks again,
                >
                > Bob G8VOI
                >
                >
                > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                > >
                > > Hi Bob,
                > >
                > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                > >
                > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                > >
                > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                > >
                > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                > >
                > > 73,
                > > Christos SV1EIA
                > >
                > >
                > >
                > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                > > >
                > > > Hi Graeme,
                > > >
                > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                > > >
                > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                > > >
                > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                > > >
                > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                > > >
                > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                > > >
                > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                > > >
                > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                > > >
                > > > Once that process has been 'forced' all appears well.
                > > >
                > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                > > >
                > > > 73, Bob G8VOI
                > > >
                > > >
                > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                > > > >
                > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                > > > >
                > > > > 73 Graeme ZL2APV
                > > > >
                > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                > > > > >
                > > > > >
                > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                > > > > >
                > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                > > > > >
                > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                > > > > >
                > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                > > > > >
                > > > > > I am running Windows 7 32 bit with a Delta-44.
                > > > > >
                > > > > > Thank you for any help!
                > > > > >
                > > > > > -Pete
                > > > > > WB2QLL
                > > > > > Somers, WI
                > > > > >
                > > > >
                > > >
                > >
                >
              • sv1eia
                ooops ... I meant 1.12.23 , you are right Bob, even I get confused here :-))
                Message 7 of 17 , Jun 13, 2010
                • 0 Attachment
                  ooops

                  > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15

                  I meant 1.12.23 , you are right Bob, even I get confused here :-))



                  --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@...> wrote:
                  >
                  > Hi Bob,
                  >
                  > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                  >
                  > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                  >
                  > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                  >
                  > 73,
                  > Christos SV1EIA
                  >
                  >
                  >
                  >
                  > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                  > >
                  > > Hi Christos,
                  > >
                  > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                  > >
                  > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                  > >
                  > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                  > >
                  > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                  > >
                  > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                  > >
                  > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                  > >
                  > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                  > >
                  > > Thanks again,
                  > >
                  > > Bob G8VOI
                  > >
                  > >
                  > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                  > > >
                  > > > Hi Bob,
                  > > >
                  > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                  > > >
                  > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                  > > >
                  > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                  > > >
                  > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                  > > >
                  > > > 73,
                  > > > Christos SV1EIA
                  > > >
                  > > >
                  > > >
                  > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                  > > > >
                  > > > > Hi Graeme,
                  > > > >
                  > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                  > > > >
                  > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                  > > > >
                  > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                  > > > >
                  > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                  > > > >
                  > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                  > > > >
                  > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                  > > > >
                  > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                  > > > >
                  > > > > Once that process has been 'forced' all appears well.
                  > > > >
                  > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                  > > > >
                  > > > > 73, Bob G8VOI
                  > > > >
                  > > > >
                  > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                  > > > > >
                  > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                  > > > > >
                  > > > > > 73 Graeme ZL2APV
                  > > > > >
                  > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                  > > > > > >
                  > > > > > >
                  > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                  > > > > > >
                  > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                  > > > > > >
                  > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                  > > > > > >
                  > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                  > > > > > >
                  > > > > > > I am running Windows 7 32 bit with a Delta-44.
                  > > > > > >
                  > > > > > > Thank you for any help!
                  > > > > > >
                  > > > > > > -Pete
                  > > > > > > WB2QLL
                  > > > > > > Somers, WI
                  > > > > > >
                  > > > > >
                  > > > >
                  > > >
                  > >
                  >
                • Rob
                  Hi Bob & Christos, With the arival of the sporadic-E season in Europe, some of the signals are large enough to make the image rejection work for me. I also
                  Message 8 of 17 , Jun 13, 2010
                  • 0 Attachment
                    Hi Bob & Christos,

                    With the arival of the sporadic-E season in Europe, some of the signals are large enough to make the image rejection work for me. I also found a local CB operator makes it work... the image rejection works on signals containing swearing <LOL>

                    I find V1.12 rejects the image faster than the beta version. Changing the EIA "amount" down to lower levels doesn't make it "kick-in" with low signals any better.

                    As Bob says, sometimes switching it off then on again makes it work.

                    Over the next month, I hope to construct a preamp for 10m. It will use a dual gate MOSFET, so plenty of gain. It may overload the mixer, so it will have to be switchable with a relay.

                    We will be watching the Flex website for hints on progress of the new version.

                    -Rob


                    > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                    >
                    > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                    >
                    > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                  • tom
                    Christos, I m sure it s been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                    Message 9 of 17 , Jun 13, 2010
                    • 0 Attachment
                      Christos,
                       
                      I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version?  I wish the Yahoo search function worked properly
                       
                      Thanks,
                      Tom
                       
                      ----- Original Message -----
                      From: sv1eia
                      Sent: Sunday, June 13, 2010 6:56 AM
                      Subject: [powersdr-iq] Re: Image

                       

                      ooops

                      > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15

                      I meant 1.12.23 , you are right Bob, even I get confused here :-))

                      --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@...> wrote:
                      >
                      > Hi Bob,
                      >
                      > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                      >
                      > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                      >
                      > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                      >
                      > 73,
                      > Christos SV1EIA
                      >
                      >
                      >
                      >
                      > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                      > >
                      > > Hi Christos,
                      > >
                      > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                      > >
                      > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                      > >
                      > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                      > >
                      > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                      > >
                      > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                      > >
                      > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                      > >
                      > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                      > >
                      > > Thanks again,
                      > >
                      > > Bob G8VOI
                      > >
                      > >
                      > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                      > > >
                      > > > Hi Bob,
                      > > >
                      > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                      > > >
                      > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                      > > >
                      > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                      > > >
                      > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                      > > >
                      > > > 73,
                      > > > Christos SV1EIA
                      > > >
                      > > >
                      > > >
                      > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                      > > > >
                      > > > > Hi Graeme,
                      > > > >
                      > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                      > > > >
                      > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                      > > > >
                      > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                      > > > >
                      > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                      > > > >
                      > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                      > > > >
                      > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                      > > > >
                      > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                      > > > >
                      > > > > Once that process has been 'forced' all appears well.
                      > > > >
                      > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                      > > > >
                      > > > > 73, Bob G8VOI
                      > > > >
                      > > > >
                      > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                      > > > > >
                      > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                      > > > > >
                      > > > > > 73 Graeme ZL2APV
                      > > > > >
                      > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                      > > > > > >
                      > > > > > >
                      > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                      > > > > > >
                      > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                      > > > > > >
                      > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                      > > > > > >
                      > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                      > > > > > >
                      > > > > > > I am running Windows 7 32 bit with a Delta-44.
                      > > > > > >
                      > > > > > > Thank you for any help!
                      > > > > > >
                      > > > > > > -Pete
                      > > > > > > WB2QLL
                      > > > > > > Somers, WI
                      > > > > > >
                      > > > > >
                      > > > >
                      > > >
                      > >
                      >

                    • sv1eia
                      Hi Tom, - http://code.google.com/p/powersdr-iq/downloads/list Also if you see on the first page of this group - http://groups.yahoo.com/group/powersdr-iq/ it
                      Message 10 of 17 , Jun 13, 2010
                      • 0 Attachment
                        Hi Tom,

                        -> http://code.google.com/p/powersdr-iq/downloads/list

                        Also if you see on the first page of this group
                        -> http://groups.yahoo.com/group/powersdr-iq/
                        it shows

                        'Announcements
                        Oct 15, 2008 : Downloads and SVN'

                        this link which it points to the same software repository

                        73,
                        Christos SV1EIA



                        --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@...> wrote:
                        >
                        > Christos,
                        >
                        > I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                        >
                        > Thanks,
                        > Tom
                        >
                        > ----- Original Message -----
                        > From: sv1eia
                        > To: powersdr-iq@yahoogroups.com
                        > Sent: Sunday, June 13, 2010 6:56 AM
                        > Subject: [powersdr-iq] Re: Image
                        >
                        >
                        >
                        > ooops
                        >
                        > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                        >
                        > I meant 1.12.23 , you are right Bob, even I get confused here :-))
                        >
                        > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                        > >
                        > > Hi Bob,
                        > >
                        > > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                        > >
                        > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                        > >
                        > > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                        > >
                        > > 73,
                        > > Christos SV1EIA
                        > >
                        > >
                        > >
                        > >
                        > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                        > > >
                        > > > Hi Christos,
                        > > >
                        > > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                        > > >
                        > > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                        > > >
                        > > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                        > > >
                        > > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                        > > >
                        > > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                        > > >
                        > > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                        > > >
                        > > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                        > > >
                        > > > Thanks again,
                        > > >
                        > > > Bob G8VOI
                        > > >
                        > > >
                        > > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                        > > > >
                        > > > > Hi Bob,
                        > > > >
                        > > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                        > > > >
                        > > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                        > > > >
                        > > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                        > > > >
                        > > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                        > > > >
                        > > > > 73,
                        > > > > Christos SV1EIA
                        > > > >
                        > > > >
                        > > > >
                        > > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                        > > > > >
                        > > > > > Hi Graeme,
                        > > > > >
                        > > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                        > > > > >
                        > > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                        > > > > >
                        > > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                        > > > > >
                        > > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                        > > > > >
                        > > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                        > > > > >
                        > > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                        > > > > >
                        > > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                        > > > > >
                        > > > > > Once that process has been 'forced' all appears well.
                        > > > > >
                        > > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                        > > > > >
                        > > > > > 73, Bob G8VOI
                        > > > > >
                        > > > > >
                        > > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                        > > > > > >
                        > > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                        > > > > > >
                        > > > > > > 73 Graeme ZL2APV
                        > > > > > >
                        > > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                        > > > > > > >
                        > > > > > > >
                        > > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                        > > > > > > >
                        > > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                        > > > > > > >
                        > > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                        > > > > > > >
                        > > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                        > > > > > > >
                        > > > > > > > I am running Windows 7 32 bit with a Delta-44.
                        > > > > > > >
                        > > > > > > > Thank you for any help!
                        > > > > > > >
                        > > > > > > > -Pete
                        > > > > > > > WB2QLL
                        > > > > > > > Somers, WI
                        > > > > > > >
                        > > > > > >
                        > > > > >
                        > > > >
                        > > >
                        > >
                        >
                      • tom
                        Thanks Christos....A question...I am presently using V1.12.20 which works great on my PC at about %18-20% cpu load. When I run 1.19.3.15 the usage only goes
                        Message 11 of 17 , Jun 13, 2010
                        • 0 Attachment
                          Thanks Christos....A question...I am presently using V1.12.20 which works great on my PC at about %18-20%  cpu load.  When I run 1.19.3.15 the usage only goes up to maybe 20-25% but the audio starts to break up. I would prefer to run V1.19.3.15. Does this sound normal to you?   Using a P4-T533 2.7 GHZ single core w/Win XP SP3 with only PSDR running.
                           
                          Tom
                          ----- Original Message -----
                          From: sv1eia
                          Sent: Sunday, June 13, 2010 9:02 AM
                          Subject: [powersdr-iq] Re: Image

                           

                          Hi Tom,

                          -> http://code.google.com/p/powersdr-iq/downloads/list

                          Also if you see on the first page of this group
                          -> http://groups.yahoo.com/group/powersdr-iq/
                          it shows

                          'Announcements
                          Oct 15, 2008 : Downloads and SVN'

                          this link which it points to the same software repository

                          73,
                          Christos SV1EIA

                          --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@...> wrote:
                          >
                          > Christos,
                          >
                          > I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                          >
                          > Thanks,
                          > Tom
                          >
                          > ----- Original Message -----
                          > From: sv1eia
                          > To: powersdr-iq@yahoogroups.com
                          > Sent: Sunday, June 13, 2010 6:56 AM
                          > Subject: [powersdr-iq] Re: Image
                          >
                          >
                          >
                          > ooops
                          >
                          > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                          >
                          > I meant 1.12.23 , you are right Bob, even I get confused here :-))
                          >
                          > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                          > >
                          > > Hi Bob,
                          > >
                          > > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                          > >
                          > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                          > >
                          > > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                          > >
                          > > 73,
                          > > Christos SV1EIA
                          > >
                          > >
                          > >
                          > >
                          > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                          > > >
                          > > > Hi Christos,
                          > > >
                          > > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                          > > >
                          > > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                          > > >
                          > > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                          > > >
                          > > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                          > > >
                          > > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                          > > >
                          > > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                          > > >
                          > > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                          > > >
                          > > > Thanks again,
                          > > >
                          > > > Bob G8VOI
                          > > >
                          > > >
                          > > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                          > > > >
                          > > > > Hi Bob,
                          > > > >
                          > > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                          > > > >
                          > > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                          > > > >
                          > > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                          > > > >
                          > > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                          > > > >
                          > > > > 73,
                          > > > > Christos SV1EIA
                          > > > >
                          > > > >
                          > > > >
                          > > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                          > > > > >
                          > > > > > Hi Graeme,
                          > > > > >
                          > > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                          > > > > >
                          > > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                          > > > > >
                          > > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                          > > > > >
                          > > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                          > > > > >
                          > > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                          > > > > >
                          > > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                          > > > > >
                          > > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                          > > > > >
                          > > > > > Once that process has been 'forced' all appears well.
                          > > > > >
                          > > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                          > > > > >
                          > > > > > 73, Bob G8VOI
                          > > > > >
                          > > > > >
                          > > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                          > > > > > >
                          > > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                          > > > > > >
                          > > > > > > 73 Graeme ZL2APV
                          > > > > > >
                          > > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                          > > > > > > >
                          > > > > > > >
                          > > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                          > > > > > > >
                          > > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                          > > > > > > >
                          > > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                          > > > > > > >
                          > > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                          > > > > > > >
                          > > > > > > > I am running Windows 7 32 bit with a Delta-44.
                          > > > > > > >
                          > > > > > > > Thank you for any help!
                          > > > > > > >
                          > > > > > > > -Pete
                          > > > > > > > WB2QLL
                          > > > > > > > Somers, WI
                          > > > > > > >
                          > > > > > >
                          > > > > >
                          > > > >
                          > > >
                          > >
                          >

                        • tom
                          Christos, sorry, let me be more specific. I adjusted all the buffer sizes between 1.12.20 and 1.19.315 and the audio dropouts seem to have stopped in
                          Message 12 of 17 , Jun 13, 2010
                          • 0 Attachment
                            Christos,
                             
                            sorry, let me be more specific. I adjusted all the buffer sizes between 1.12.20 and 1.19.315 and the audio dropouts seem to have stopped in 1.19.3.15. However, in CW mode,  everything being equal in both revs setups, using 1.12.20 the dots and dashes are exactly the same length when I send a string of dots or dashes. In V1.19.3.15, the lengths of the dots or dashes vary randomly as I send a string of either. Any solution for this?  I would prefer to use 1.19 rather than 1.12 as I think the display (skin) is much nicer.
                             
                            I think Bob mentioned at some point that you came out with a rev of 1.19 that solved this problem.
                             
                            Tom 
                            ----- Original Message -----
                            From: tom
                            Sent: Sunday, June 13, 2010 9:33 AM
                            Subject: Re: [powersdr-iq] Re: Image

                            Thanks Christos....A question...I am presently using V1.12.20 which works great on my PC at about %18-20%  cpu load.  When I run 1.19.3.15 the usage only goes up to maybe 20-25% but the audio starts to break up. I would prefer to run V1.19.3.15. Does this sound normal to you?   Using a P4-T533 2.7 GHZ single core w/Win XP SP3 with only PSDR running.
                             
                            Tom
                            ----- Original Message -----
                            From: sv1eia
                            Sent: Sunday, June 13, 2010 9:02 AM
                            Subject: [powersdr-iq] Re: Image

                             

                            Hi Tom,

                            -> http://code.google.com/p/powersdr-iq/downloads/list

                            Also if you see on the first page of this group
                            -> http://groups.yahoo.com/group/powersdr-iq/
                            it shows

                            'Announcements
                            Oct 15, 2008 : Downloads and SVN'

                            this link which it points to the same software repository

                            73,
                            Christos SV1EIA

                            --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@...> wrote:
                            >
                            > Christos,
                            >
                            > I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                            >
                            > Thanks,
                            > Tom
                            >
                            > ----- Original Message -----
                            > From: sv1eia
                            > To: powersdr-iq@yahoogroups.com
                            > Sent: Sunday, June 13, 2010 6:56 AM
                            > Subject: [powersdr-iq] Re: Image
                            >
                            >
                            >
                            > ooops
                            >
                            > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                            >
                            > I meant 1.12.23 , you are right Bob, even I get confused here :-))
                            >
                            > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                            > >
                            > > Hi Bob,
                            > >
                            > > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                            > >
                            > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                            > >
                            > > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                            > >
                            > > 73,
                            > > Christos SV1EIA
                            > >
                            > >
                            > >
                            > >
                            > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                            > > >
                            > > > Hi Christos,
                            > > >
                            > > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                            > > >
                            > > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                            > > >
                            > > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                            > > >
                            > > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                            > > >
                            > > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                            > > >
                            > > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                            > > >
                            > > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                            > > >
                            > > > Thanks again,
                            > > >
                            > > > Bob G8VOI
                            > > >
                            > > >
                            > > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                            > > > >
                            > > > > Hi Bob,
                            > > > >
                            > > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                            > > > >
                            > > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                            > > > >
                            > > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                            > > > >
                            > > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                            > > > >
                            > > > > 73,
                            > > > > Christos SV1EIA
                            > > > >
                            > > > >
                            > > > >
                            > > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                            > > > > >
                            > > > > > Hi Graeme,
                            > > > > >
                            > > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                            > > > > >
                            > > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                            > > > > >
                            > > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                            > > > > >
                            > > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                            > > > > >
                            > > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                            > > > > >
                            > > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                            > > > > >
                            > > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                            > > > > >
                            > > > > > Once that process has been 'forced' all appears well.
                            > > > > >
                            > > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                            > > > > >
                            > > > > > 73, Bob G8VOI
                            > > > > >
                            > > > > >
                            > > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                            > > > > > >
                            > > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                            > > > > > >
                            > > > > > > 73 Graeme ZL2APV
                            > > > > > >
                            > > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                            > > > > > > >
                            > > > > > > >
                            > > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                            > > > > > > >
                            > > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                            > > > > > > >
                            > > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                            > > > > > > >
                            > > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                            > > > > > > >
                            > > > > > > > I am running Windows 7 32 bit with a Delta-44.
                            > > > > > > >
                            > > > > > > > Thank you for any help!
                            > > > > > > >
                            > > > > > > > -Pete
                            > > > > > > > WB2QLL
                            > > > > > > > Somers, WI
                            > > > > > > >
                            > > > > > >
                            > > > > >
                            > > > >
                            > > >
                            > >
                            >

                          • sv1eia
                            Tom, Have a look at the links section, it has some information for CW that could be of use for you - http://groups.yahoo.com/group/powersdr-iq/links 73,
                            Message 13 of 17 , Jun 13, 2010
                            • 0 Attachment
                              Tom,

                              Have a look at the links section, it has some information for CW that could be of use for you
                              -> http://groups.yahoo.com/group/powersdr-iq/links

                              73,
                              Christos SV1EIA



                              --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@...> wrote:
                              >
                              > Christos,
                              >
                              > sorry, let me be more specific. I adjusted all the buffer sizes between 1.12.20 and 1.19.315 and the audio dropouts seem to have stopped in 1.19.3.15. However, in CW mode, everything being equal in both revs setups, using 1.12.20 the dots and dashes are exactly the same length when I send a string of dots or dashes. In V1.19.3.15, the lengths of the dots or dashes vary randomly as I send a string of either. Any solution for this? I would prefer to use 1.19 rather than 1.12 as I think the display (skin) is much nicer.
                              >
                              > I think Bob mentioned at some point that you came out with a rev of 1.19 that solved this problem.
                              >
                              > Tom
                              > ----- Original Message -----
                              > From: tom
                              > To: powersdr-iq@yahoogroups.com
                              > Sent: Sunday, June 13, 2010 9:33 AM
                              > Subject: Re: [powersdr-iq] Re: Image
                              >
                              >
                              > Thanks Christos....A question...I am presently using V1.12.20 which works great on my PC at about %18-20% cpu load. When I run 1.19.3.15 the usage only goes up to maybe 20-25% but the audio starts to break up. I would prefer to run V1.19.3.15. Does this sound normal to you? Using a P4-T533 2.7 GHZ single core w/Win XP SP3 with only PSDR running.
                              >
                              > Tom
                              > ----- Original Message -----
                              > From: sv1eia
                              > To: powersdr-iq@yahoogroups.com
                              > Sent: Sunday, June 13, 2010 9:02 AM
                              > Subject: [powersdr-iq] Re: Image
                              >
                              >
                              >
                              > Hi Tom,
                              >
                              > -> http://code.google.com/p/powersdr-iq/downloads/list
                              >
                              > Also if you see on the first page of this group
                              > -> http://groups.yahoo.com/group/powersdr-iq/
                              > it shows
                              >
                              > 'Announcements
                              > Oct 15, 2008 : Downloads and SVN'
                              >
                              > this link which it points to the same software repository
                              >
                              > 73,
                              > Christos SV1EIA
                              >
                              > --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@> wrote:
                              > >
                              > > Christos,
                              > >
                              > > I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                              > >
                              > > Thanks,
                              > > Tom
                              > >
                              > > ----- Original Message -----
                              > > From: sv1eia
                              > > To: powersdr-iq@yahoogroups.com
                              > > Sent: Sunday, June 13, 2010 6:56 AM
                              > > Subject: [powersdr-iq] Re: Image
                              > >
                              > >
                              > >
                              > > ooops
                              > >
                              > > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                              > >
                              > > I meant 1.12.23 , you are right Bob, even I get confused here :-))
                              > >
                              > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                              > > >
                              > > > Hi Bob,
                              > > >
                              > > > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                              > > >
                              > > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                              > > >
                              > > > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                              > > >
                              > > > 73,
                              > > > Christos SV1EIA
                              > > >
                              > > >
                              > > >
                              > > >
                              > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                              > > > >
                              > > > > Hi Christos,
                              > > > >
                              > > > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                              > > > >
                              > > > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                              > > > >
                              > > > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                              > > > >
                              > > > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                              > > > >
                              > > > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                              > > > >
                              > > > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                              > > > >
                              > > > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                              > > > >
                              > > > > Thanks again,
                              > > > >
                              > > > > Bob G8VOI
                              > > > >
                              > > > >
                              > > > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                              > > > > >
                              > > > > > Hi Bob,
                              > > > > >
                              > > > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                              > > > > >
                              > > > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                              > > > > >
                              > > > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                              > > > > >
                              > > > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                              > > > > >
                              > > > > > 73,
                              > > > > > Christos SV1EIA
                              > > > > >
                              > > > > >
                              > > > > >
                              > > > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                              > > > > > >
                              > > > > > > Hi Graeme,
                              > > > > > >
                              > > > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                              > > > > > >
                              > > > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                              > > > > > >
                              > > > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                              > > > > > >
                              > > > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                              > > > > > >
                              > > > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                              > > > > > >
                              > > > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                              > > > > > >
                              > > > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                              > > > > > >
                              > > > > > > Once that process has been 'forced' all appears well.
                              > > > > > >
                              > > > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                              > > > > > >
                              > > > > > > 73, Bob G8VOI
                              > > > > > >
                              > > > > > >
                              > > > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                              > > > > > > >
                              > > > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                              > > > > > > >
                              > > > > > > > 73 Graeme ZL2APV
                              > > > > > > >
                              > > > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                              > > > > > > > >
                              > > > > > > > >
                              > > > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                              > > > > > > > >
                              > > > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                              > > > > > > > >
                              > > > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                              > > > > > > > >
                              > > > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                              > > > > > > > >
                              > > > > > > > > I am running Windows 7 32 bit with a Delta-44.
                              > > > > > > > >
                              > > > > > > > > Thank you for any help!
                              > > > > > > > >
                              > > > > > > > > -Pete
                              > > > > > > > > WB2QLL
                              > > > > > > > > Somers, WI
                              > > > > > > > >
                              > > > > > > >
                              > > > > > >
                              > > > > >
                              > > > >
                              > > >
                              > >
                              >
                            • tom
                              Thanks Christos. Lots of useful information there... Tom ... From: sv1eia To: powersdr-iq@yahoogroups.com Sent: Sunday, June 13, 2010 12:20 PM Subject:
                              Message 14 of 17 , Jun 13, 2010
                              • 0 Attachment
                                Thanks Christos. Lots of useful information there...
                                 
                                Tom
                                ----- Original Message -----
                                From: sv1eia
                                Sent: Sunday, June 13, 2010 12:20 PM
                                Subject: [powersdr-iq] Re: Image

                                 

                                Tom,

                                Have a look at the links section, it has some information for CW that could be of use for you
                                -> http://groups.yahoo.com/group/powersdr-iq/links

                                73,
                                Christos SV1EIA

                                --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@...> wrote:
                                >
                                > Christos,
                                >
                                > sorry, let me be more specific. I adjusted all the buffer sizes between 1.12.20 and 1.19.315 and the audio dropouts seem to have stopped in 1.19.3.15. However, in CW mode, everything being equal in both revs setups, using 1.12.20 the dots and dashes are exactly the same length when I send a string of dots or dashes. In V1.19.3.15, the lengths of the dots or dashes vary randomly as I send a string of either. Any solution for this? I would prefer to use 1.19 rather than 1.12 as I think the display (skin) is much nicer.
                                >
                                > I think Bob mentioned at some point that you came out with a rev of 1.19 that solved this problem.
                                >
                                > Tom
                                > ----- Original Message -----
                                > From: tom
                                > To: powersdr-iq@yahoogroups.com
                                > Sent: Sunday, June 13, 2010 9:33 AM
                                > Subject: Re: [powersdr-iq] Re: Image
                                >
                                >
                                > Thanks Christos....A question...I am presently using V1.12.20 which works great on my PC at about %18-20% cpu load. When I run 1.19.3.15 the usage only goes up to maybe 20-25% but the audio starts to break up. I would prefer to run V1.19.3.15. Does this sound normal to you? Using a P4-T533 2.7 GHZ single core w/Win XP SP3 with only PSDR running.
                                >
                                > Tom
                                > ----- Original Message -----
                                > From: sv1eia
                                > To: powersdr-iq@yahoogroups.com
                                > Sent: Sunday, June 13, 2010 9:02 AM
                                > Subject: [powersdr-iq] Re: Image
                                >
                                >
                                >
                                > Hi Tom,
                                >
                                > -> http://code.google.com/p/powersdr-iq/downloads/list
                                >
                                > Also if you see on the first page of this group
                                > -> http://groups.yahoo.com/group/powersdr-iq/
                                > it shows
                                >
                                > 'Announcements
                                > Oct 15, 2008 : Downloads and SVN'
                                >
                                > this link which it points to the same software repository
                                >
                                > 73,
                                > Christos SV1EIA
                                >
                                > --- In powersdr-iq@yahoogroups.com, "tom" <loeblt@> wrote:
                                > >
                                > > Christos,
                                > >
                                > > I'm sure it's been posted in the past, but where can I download v1.19.3.14 and your latest version? I wish the Yahoo search function worked properly
                                > >
                                > > Thanks,
                                > > Tom
                                > >
                                > > ----- Original Message -----
                                > > From: sv1eia
                                > > To: powersdr-iq@yahoogroups.com
                                > > Sent: Sunday, June 13, 2010 6:56 AM
                                > > Subject: [powersdr-iq] Re: Image
                                > >
                                > >
                                > >
                                > > ooops
                                > >
                                > > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                                > >
                                > > I meant 1.12.23 , you are right Bob, even I get confused here :-))
                                > >
                                > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                                > > >
                                > > > Hi Bob,
                                > > >
                                > > > Yes, I know, I owe some enhancements but I keep pushing everything after a not-so-quicksand version gets out from Flex.
                                > > >
                                > > > If I remember correctly, that issue you mention of disabling/enabling EIA to start working I got it fixed some time now in 1.19.3.15
                                > > >
                                > > > Anyway, if all go well, I believe your patience will be rewarded with a quantum leap for homebrew SDR in a couple of months, speaking for both S/W and H/W.. :-)
                                > > >
                                > > > 73,
                                > > > Christos SV1EIA
                                > > >
                                > > >
                                > > >
                                > > >
                                > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                                > > > >
                                > > > > Hi Christos,
                                > > > >
                                > > > > It can get a little confusing at times with your two PSDR versions. Hopefully I try to remember to comment on which one I am talking about, but maybe sometimes I forget!
                                > > > >
                                > > > > I agree v1.12.20 can take a short time to fully reduce the image but depends on the EIA setting used. I settled for a compromise of 0.5. I have seen it sometimes not work, but found if you just turn EIA off and on again, then works fine.
                                > > > >
                                > > > > I have been using v1.19.3.14 now since you released your customised version with my v6.3 RXTX and running WSPR for many long hours. Generally it is working very well, but of course there are some problems in the main Flex code, so no point in complaining to you about them!
                                > > > >
                                > > > > It will be superb when it matures and as far as I am concerned its my 'main' SDR program already for transceive operation with all my SoftRock's.
                                > > > >
                                > > > > v1.19.3.14 is running on my other PC at the moment with Tony's VHF Ensemble RX on 2m monitoring for sporadic Es and stations in the PW Wireless QRP contest today.
                                > > > >
                                > > > > I saw your exchanges with Andrew G4XZL who is a very good friend of mine, we have know each other for over 30 years!
                                > > > >
                                > > > > He is out on a local hill top today taking part in the 2m QRP contest and what he is working towards is a complete 2m SDR station using a v6.3 RXTX, modified v4.3 mobo and transverter. We both are mainly VHF enthusiasts, so a lot in common and I am watching his progress with great interest.
                                > > > >
                                > > > > Thanks again,
                                > > > >
                                > > > > Bob G8VOI
                                > > > >
                                > > > >
                                > > > > --- In powersdr-iq@yahoogroups.com, "sv1eia" <sv1eia@> wrote:
                                > > > > >
                                > > > > > Hi Bob,
                                > > > > >
                                > > > > > I'm using the 1.12.20 here and used the 1.19.3.x only when I initially developed it (I almost forgot where I've placed that usbtoi2c board that I have :-) ).
                                > > > > >
                                > > > > > In 1.12.20 you do need a couple of minutes in order to get some good rejection. That time though is pertinent also to the coefficient, larger works faster and with large signals, smaller works slower but with low signals.
                                > > > > >
                                > > > > > In the old version, I didnt notice that effect, yet again there are some reports building up in the new one (1.19.x) that it needs a 'boost' signal in order to kick in. The algo is not significantly different so I do not expect that drastic behaviour. On the other hand I did not payed too much attention to it since its newer and I thought that it would behave better than what I did in 1.12.20.
                                > > > > >
                                > > > > > Anyway, lets wait until this 1.19 testbed became a valid supported v2.0 being available also for SDR1K, and then I hope these early version isues will go away. If not then I'll see what it needs to be done for our homebrew staff and I'll fix it using the algo of the 1.12.20.
                                > > > > >
                                > > > > > 73,
                                > > > > > Christos SV1EIA
                                > > > > >
                                > > > > >
                                > > > > >
                                > > > > > --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@> wrote:
                                > > > > > >
                                > > > > > > Hi Graeme,
                                > > > > > >
                                > > > > > > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                                > > > > > >
                                > > > > > > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                                > > > > > >
                                > > > > > > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                                > > > > > >
                                > > > > > > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                                > > > > > >
                                > > > > > > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                                > > > > > >
                                > > > > > > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                                > > > > > >
                                > > > > > > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                                > > > > > >
                                > > > > > > Once that process has been 'forced' all appears well.
                                > > > > > >
                                > > > > > > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                                > > > > > >
                                > > > > > > 73, Bob G8VOI
                                > > > > > >
                                > > > > > >
                                > > > > > > --- In powersdr-iq@yahoogroups.com, "Graeme" <gvj@> wrote:
                                > > > > > > >
                                > > > > > > > I had to re-flash my firmware and set up my PowerSDR-IQ again and of course the image supression was changed. I thought of your message as I got my image to slowly go down. I wonder if you realise that it takes a fair while (30 mins or more) but if you sit the rx on a reasonable signal (-70 dBm or more) and just wait the image sinks slowly into the noise. Mine went from about -15 dBr wrt carrier down to more than -60 dBr wrt carrier. Can't tell exactly how far as my noise floor with my EDIROL FA-66 is about -135 dBm absolute.
                                > > > > > > >
                                > > > > > > > 73 Graeme ZL2APV
                                > > > > > > >
                                > > > > > > > --- In powersdr-iq@yahoogroups.com, Pete Ferrand <petef@> wrote:
                                > > > > > > > >
                                > > > > > > > >
                                > > > > > > > > I am wondering what to do about a lack of image suppression for the RxTx 6.3 with I2C I just built. There is an image attenuated by 10db only. This is the same on 80-17, haven't built the other bands.
                                > > > > > > > >
                                > > > > > > > > The odd thing is that with the radio and signal set to 7.100 kHz, the image is at 7082. With the radio left at 7100, a signal at 7105 has an image at 7077. A signal at 7110 has an image at 7072.
                                > > > > > > > >
                                > > > > > > > > With Rocky I get suppression into the grass. I'm using powersdr-iq 1.19.3.15. Whether I use an isolation transformer or not doesn't matter; I know Christos says not to use one.
                                > > > > > > > >
                                > > > > > > > > With PowerSDR 1.18.6 and this same computer setup I can adjust image suppression into the grass on my SDR-1000.
                                > > > > > > > >
                                > > > > > > > > I am running Windows 7 32 bit with a Delta-44.
                                > > > > > > > >
                                > > > > > > > > Thank you for any help!
                                > > > > > > > >
                                > > > > > > > > -Pete
                                > > > > > > > > WB2QLL
                                > > > > > > > > Somers, WI
                                > > > > > > > >
                                > > > > > > >
                                > > > > > >
                                > > > > >
                                > > > >
                                > > >
                                > >
                                >

                              • Graeme
                                Hi Bob, Nice to hear from you. Yes I agree with all you say but I think I remember this from the first time I installed 1.19.3.15 also but must confess I am a
                                Message 15 of 17 , Jun 13, 2010
                                • 0 Attachment
                                  Hi Bob,

                                  Nice to hear from you. Yes I agree with all you say but I think I remember this from the first time I installed 1.19.3.15 also but must confess I am a little hazy on this now. I am pretty sure the same thing happened and I had to wait a long time -lots of minutes rather than seconds - for the image to disappear and then after that it was instant. Silly thing is I can't verify this right now as I am taking the plunge and installing my softrock in case and it is dismantled. What I am sure of is that for the first time I set it up in PSDR-IQ this time round it took many minutes for the image to fade away. I was doing some screenshots for Alex 9V1AL to show him what the noise floor of my FA-66 looked like with a sig gen only connected and each sucessive screenshot showed less and less image until eventually it disappeared into the noise altogether. I wish I could switch it on now to verify that the image is still nulled but that will have to wait. I am also wondering if this effect is a result of my phase being a long way off 90 deg. I say this because I am using long audio leads and transformers for isolation. I just know Christos will be shaking his head if he reads this haha. When the softrock is packed in its new case the FA-66 will be mounted on top and the transformers will not be used along with the long leads. Then I can test this theory. In the early days I set up the transformers as a security blanket for my expensive sound card but have got over that now and am happy to make direct connections with the advantage of better phase response.

                                  I should add that I have been using 1.19.3.15 from the time that Christos set it up for CW using USB2IC and it has gone brilliantly. By far the best software for an operator although latency is a factor. SDR-Shell is the best I have tried in that area and uses dttsp rather than intel's fea but the interface is a programmers tool rather than an operators tool. In many hours of operating including contests I have had only 1 inexplicable crash so flex has been very stable for me.

                                  73 Graeme ZL2APV

                                  --- In powersdr-iq@yahoogroups.com, "g8voip" <g8voi.reeves59@...> wrote:
                                  >
                                  > Hi Graeme,
                                  >
                                  > Usually when the automatic image rejection is working correctly it reduces the image to the noise floor within seconds.
                                  >
                                  > As far as I know it is done on the 'fly' and there is no progressive building up or storage of gain / phase correction curves as in Rocky.
                                  >
                                  > If you stop and think about it, if you had to wait 30 minutes, it would be pretty useless!
                                  >
                                  > I have been following the ongoing thread, and can confirm I have seen the same effect as others have reported on many occasions recently.
                                  >
                                  > For the past couple of months I have been carrying out several beta tests for Tony KB9YIG for his new Ensemble designs.
                                  >
                                  > On many occasions whilst carrying out image rejection or MDS measurement with a signal generator attached, I have seen a large image present, which has not been automatically removed.
                                  >
                                  > Simply increasing the signal input level to 50dB above the noise floor instantly kicks the auto image rejection into action and removes the image to the noise floor.
                                  >
                                  > Once that process has been 'forced' all appears well.
                                  >
                                  > There is almost certainly some strange behaviour, and the auto image rejection either does not work as expected, or can get into a 'confused' state. The problem possibly might be less apparent on 'real' band signals as the basic SDR hardware typically has 20 - 30dB or more image rejection to start off with.
                                  >
                                  > 73, Bob G8VOI
                                  [snip]
                                Your message has been successfully submitted and would be delivered to recipients shortly.