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

DDS-->PLO Attenuation

Expand Messages
  • Sam Wetterlin
    Back in Message 3674, Mike noted that he got lower spurs by attenuating the DDS signal by 10-14 db as it runs to the PLO. This is on top of the 6db attenuation
    Message 1 of 2 , Nov 25, 2008
      Back in Message 3674, Mike noted that he got lower spurs by
      attenuating the DDS signal by 10-14 db as it runs to the PLO. This is
      on top of the 6db attenuation he got by having R12 in place on the PLO
      reference input.

      To achieve a similar effect, and get 50 ohm output impedance for the
      DDS, I left that R12 (and the associated C27) in place on the PLO
      board and modified the DDS squarer output. I changed R14 from 33 ohms
      to 150 ohms, and then attached a 10nf capacitor and 75 ohm resistor in
      series with each other running from DDS J4 (i.e. the squarer output)
      to ground. There is plenty of space on the PCB for the added
      components. This gives 16 db of attenuation relative to the original
      schematic (i.e. R12 deleted) and 10 db of attenuation relative to the
      original with R12 in place, and so is equivalent to what Mike did by
      adding his 10 db attenuator. As a side benefit, both source and
      destination are now close to 50 ohms, so any small reflections off the
      PLO will be completely absorbed by the source.

      I don't have the full MSA up and running, but I checked PLO3 directly
      for spurs and phase noise, before and after this modification. A
      significant spur at -25 KHz disappeared after the change, and one near
      +25 KHz was somewhat reduced. Phase noise out to 75 KHz on each side
      appeared unchanged.

      Given my limited test and Mike's more extensive tests, this appears to
      be a good move to make. This modification applies directly to both
      PLO1 and PLO3. PLO2 is fed from the master oscillator rather than a
      DDS, but conceivably some amount of attenuation in that connection
      would also be beneficial.

      Sam W.
    • william sprowls
      Hi All,  Mike and I are working his spur problem.  Until we come to a resolution on the originating problem, I highly suggest you do not change your hardware
      Message 2 of 2 , Nov 25, 2008
        Hi All,
         Mike and I are working his spur problem.  Until we come to a resolution on the originating problem, I highly suggest you do not change your hardware design, just because it eliminates someone else's symptoms.  If you are having the same spur problems, we sure want to hear about it.
          If this really is a problem with all PLO's, then Sam's design may be viable.  As of now, there is only a single instance.
        Scotty

        --- On Tue, 11/25/08, Sam Wetterlin <swetterlin@...> wrote:
        From: Sam Wetterlin <swetterlin@...>
        Subject: [spectrumanalyzer] DDS-->PLO Attenuation
        To: spectrumanalyzer@yahoogroups.com
        Date: Tuesday, November 25, 2008, 1:39 PM

        Back in Message 3674, Mike noted that he got lower spurs by
        attenuating the DDS signal by 10-14 db as it runs to the PLO. This is
        on top of the 6db attenuation he got by having R12 in place on the PLO
        reference input.

        To achieve a similar effect, and get 50 ohm output impedance for the
        DDS, I left that R12 (and the associated C27) in place on the PLO
        board and modified the DDS squarer output. I changed R14 from 33 ohms
        to 150 ohms, and then attached a 10nf capacitor and 75 ohm resistor in
        series with each other running from DDS J4 (i.e. the squarer output)
        to ground. There is plenty of space on the PCB for the added
        components. This gives 16 db of attenuation relative to the original
        schematic (i.e. R12 deleted) and 10 db of attenuation relative to the
        original with R12 in place, and so is equivalent to what Mike did by
        adding his 10 db attenuator. As a side benefit, both source and
        destination are now close to 50 ohms, so any small reflections off the
        PLO will be completely absorbed by the source.

        I don't have the full MSA up and running, but I checked PLO3 directly
        for spurs and phase noise, before and after this modification. A
        significant spur at -25 KHz disappeared after the change, and one near
        +25 KHz was somewhat reduced. Phase noise out to 75 KHz on each side
        appeared unchanged.

        Given my limited test and Mike's more extensive tests, this appears to
        be a good move to make. This modification applies directly to both
        PLO1 and PLO3. PLO2 is fed from the master oscillator rather than a
        DDS, but conceivably some amount of attenuation in that connection
        would also be beneficial.

        Sam W.


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