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

Re: Wierd I/Q Balance

Expand Messages
  • alanzfq
    ... Paul, I do not have an answer. How does it work with other software? I think I have seen more than 10° in Rocky, does Skimmer have a smaller range?
    Message 1 of 11 , Mar 22, 2013
      --- In softrock40@yahoogroups.com, "pmackanos" <k2db@...> wrote:
      >
      > I have a Softrock Ensemble RXTX that I built for 40,30,20 meters. It appears to work fine receiving except:
      > When I put it on CW skimmer on 40 meters (I haven't checked on 30 or 20 yet).
      > Any signals that appear to be above an SNR of 30 to 32 db, it spots images, the exact frequency differential on the other side of the LO.
      > This has been a major problem during a few of the CW contests, as the big gun signals get spotted on two frequencies. (and they have my e-mail address :-) )

      Paul,

      I do not have an answer.
      How does it work with other software?
      I think I have seen more than 10° in Rocky, does Skimmer have a smaller range?
      Winrad, a long time ago was modified because it had not enough range.

      As far as the Softrock goes I tried once to see how badly a transformer could be wound. I even found it balanced with one turn difference in the secondaries.
      The main thing that affects balance seems to be the antenna. The TX rejection seems better and is not affected by the antenna.
      Try the other bands.
      Can you try a different antenna or use a preamp to add a little isolation? Or was your sig gen test without the antenna?

      There is not much in the RX to affect phase balance. Try rewinding the transformer maybe try replacing the capacitor that isolates the centre tap first.

      John, The image is exactly opposite the centre from the true signal and is actually on the opposite sideband.
      In PSDR TX it is in the opposite sideband position due to the frequency change in PSDR.

      73 Alan G4ZFQ


      > I have been working lately with Pete, N4ZR, and on this and other items with Tim, KQ8M trying to find the problem and solution, and we are at a stumbling point.
      > When I use a signal generator and run it up and down the band, using the Automatic I/Q Balance function in CW Skimmers, all appears to be ok, except that:
      > The phase goes from about 12 degrees to 16 degrees across 7.000 to 7.096
      > The gain stays along about .980 + or - across the same range.
      > Pete informed me that in a chat with Alex, CW Skimmer cannot auto correct above a phase of 10 degrees, and my plot line is totally above 10 degrees.
      > My question is this, what can it be?
      > Is it a possibility that it is the winding of the coils? Something else? or just inherent in the fact that is is not just a stand alone receiver?
      > Phase angle stays about the same with different sound cards (above 10 degrees), changing the slot it is installed in, and changing out all of the usb and audio lines?
      > I hope someone can help, my solution right now was to put the LO at 6.999 and just scan from 7.000 to 7.048, but I loose 48 k of skimming.
      > I am hoping one of you "GURU'S" can help me out here,
      > I do not want to put a skimmer on the air that gives out image spots !!!!
      > Thanks,
      > Paul K2DB
      >
    • pmackanos
      ... Yes, direct into the RXTX, disconnected antenna. Paul K2DB ... I will try that when I get a chance, Thanks for the feedback !!!
      Message 2 of 11 , Mar 22, 2013
        > The main thing that affects balance seems to be the antenna. Or was your sig gen test without the antenna?

        Yes, direct into the RXTX, disconnected antenna.

        Paul K2DB
        >
        > There is not much in the RX to affect phase balance. Try rewinding the transformer maybe try replacing the capacitor that isolates the centre tap first.

        I will try that when I get a chance,

        Thanks for the feedback !!!
      • vbifyz
        High and variable phase imbalance can be a symptom of a soundcard that needs 1 sample delay. There are soundcards out there that have fractional sample delay
        Message 3 of 11 , Mar 22, 2013
          High and variable phase imbalance can be a symptom of a soundcard that needs 1 sample delay.
          There are soundcards out there that have fractional sample delay between channels (IDT/Sigmatel, used in many laptops), again resulting in variable phase imbalance that can not be compensated by any existing software.

          73, Mike

          --- In softrock40@yahoogroups.com, "pmackanos" <k2db@...> wrote:
          >
          >
          > > The main thing that affects balance seems to be the antenna. Or was your sig gen test without the antenna?
          >
          > Yes, direct into the RXTX, disconnected antenna.
          >
          > Paul K2DB
          > >
          > > There is not much in the RX to affect phase balance. Try rewinding the transformer maybe try replacing the capacitor that isolates the centre tap first.
          >
          > I will try that when I get a chance,
          >
          > Thanks for the feedback !!!
          >
        • Alan
          ... Subject: [softrock40] Re: Wierd I/Q Balance ... Yes, I should have thought of that. Try the sample delay settings. 73 Alan G4ZFQ
          Message 4 of 11 , Mar 22, 2013
            ----- Original Message -----
            Subject: [softrock40] Re: Wierd I/Q Balance


            > High and variable phase imbalance can be a symptom of a soundcard that needs 1 sample delay.
            > There are soundcards out there that have fractional sample delay between channels (IDT/Sigmatel, used in many laptops), again
            > resulting in variable phase imbalance that can not be compensated by any existing software.

            Yes, I should have thought of that.
            Try the sample delay settings.

            73 Alan G4ZFQ
          Your message has been successfully submitted and would be delivered to recipients shortly.