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

old generation DDS + PLL?

Expand Messages
  • Art
    Greetings, Many who have the DDS-60 don t use them much, and even fewer than that feed them into a transmitter due to the multitude of close in spurs. Even
    Message 1 of 21 , Jan 31, 2007
    • 0 Attachment
      Greetings,

      Many who have the DDS-60 don't use them much, and even fewer than
      that feed them into a transmitter due to the multitude of close in
      spurs. Even those that use them for receivers note that the spur
      content is high enough so that the MDS is impacted on receive. I have
      never built my DDS-60 in fact, it's still in the sealed bag!

      We can clean up the spur ridden old technology DDS units such as the
      DDS-60 by using it to feed a PLL (as the reference frequency) and
      using the output of the PLL to feed into our radios. While PLL's are
      known for having phase noise, it can be minimized if the PLL doesn't
      have to lock on a new frequency quickly-because we can use a lower
      loop gain. The point being that we can make a PLL that is more
      suitable for this purpose by careful design rather than using a
      general purpose design.

      Many of us want higher frequencies to run quadrature based detectors
      as well, so the PLL becomes a spectrally clean means to double or
      quadruple or multiply the output frequency by a very large margin.

      It seems the DDS-60 and similar DDS units can be operated in
      conjunction with PLL technology to make them useful again rather than
      shunned as a little too dirty for prime time use::>

      There are thousands of DDS-60 units out there, and I wonder if a
      companion PLL can be built to accommodate them and the many other DDS
      units that have less than stellar performance?

      Additionally, I can't see any practical reason to upgrade to newer
      DDS units that use much more power due to higher clocking rates
      (plus, they need multipliers on their clock inputs just to have the
      clock operate at 500 MHz). Does anyone else here see the
      practicality of the old technology DDS + PLL combination, or is it just me??

      Regards,

      Art
    • Terry
      Art: The PLL approach may be a good alternative, some have done it in the past (such as a QEX article, I think). It will also help if the output freq needs to
      Message 2 of 21 , Jan 31, 2007
      • 0 Attachment
        Art:

        The PLL approach may be a good alternative, some have done it in the
        past (such as a QEX article, I think). It will also help if the
        output freq needs to be very high.

        I'm not sure I agree with the comment about the newer DDS chips,
        however. The middle-generation (98532, 9854, etc) may draw a lot of
        power. But the AD9951, 9952, 9954 all draw less power than the
        9851. The 9851 can draw over 600mW at 5V, which is required for
        180MHz. The 9952 draws only 170mW, granted at 1.8V

        Plus, the AD9951-4 use 14-bit DACs, unlike the 10-bit DAC of the 9851
        in the DDS-60. They create far fewer spurs.

        The DDS-60 uses an AD9851, which has a PLL, just like the newer DDS
        chips. In fact, the DDS-60 comes with only a 30MHz oscillator, I
        believe, which requires a x6 multiplier to reach 180MHz. The AD9951
        family can use a multiplier of only 4x, if a higher input oscillator
        frequency is used, I think that may lead to lower jitter given the
        same DDS output frequency.

        Phil's comment earlier today about femtoclocks may be what is needed
        to reduce jitter from the DDS. Plus, these chips seem to create a
        higher clock freq (approx 125-250MHz on one). It will not fix spurs,
        but the better, 14-bit DACs go a long way to fix that. I have been
        pretty impressed with the new AD9951 series DDS devices. I can't
        wait to see when Analog Devices comes up with 16-bit DAC DDS units.

        Adding a PLL may be counter-productive, since unless the loop
        filtering is very carefully designed, more short and long-
        term "jitter" may result. Nevertheless, it should be interesting to
        experiment with, especially with all the 9850 and 9851 DDS chips out
        there. I agree 100% with you that if you plan to use a 9850 or 9851
        DDS, the PLL idea could be cleaner if properly designed.
        Terry


        --- In softrock40@yahoogroups.com, Art <KY1K@...> wrote:
        >
        > Greetings,
        >
        > Many who have the DDS-60 don't use them much, and even fewer than
        > that feed them into a transmitter due to the multitude of close in
        > spurs. Even those that use them for receivers note that the spur
        > content is high enough so that the MDS is impacted on receive. I
        have
        > never built my DDS-60 in fact, it's still in the sealed bag!
        >
        > We can clean up the spur ridden old technology DDS units such as
        the
        > DDS-60 by using it to feed a PLL (as the reference frequency) and
        > using the output of the PLL to feed into our radios. While PLL's
        are
        > known for having phase noise, it can be minimized if the PLL
        doesn't
        > have to lock on a new frequency quickly-because we can use a lower
        > loop gain. The point being that we can make a PLL that is more
        > suitable for this purpose by careful design rather than using a
        > general purpose design.
        >
        > Many of us want higher frequencies to run quadrature based
        detectors
        > as well, so the PLL becomes a spectrally clean means to double or
        > quadruple or multiply the output frequency by a very large margin.
        >
        > It seems the DDS-60 and similar DDS units can be operated in
        > conjunction with PLL technology to make them useful again rather
        than
        > shunned as a little too dirty for prime time use::>
        >
        > There are thousands of DDS-60 units out there, and I wonder if a
        > companion PLL can be built to accommodate them and the many other
        DDS
        > units that have less than stellar performance?
        >
        > Additionally, I can't see any practical reason to upgrade to newer
        > DDS units that use much more power due to higher clocking rates
        > (plus, they need multipliers on their clock inputs just to have the
        > clock operate at 500 MHz). Does anyone else here see the
        > practicality of the old technology DDS + PLL combination, or is it
        just me??
        >
        > Regards,
        >
        > Art
        >
      • Dan Andersson
        On Wednesday 31 January 2007 23:30:17 Art wrote: Greetings, Many who have the DDS-60 don t use them much, and even fewer than that feed them into a
        Message 3 of 21 , Jan 31, 2007
        • 0 Attachment
          On Wednesday 31 January 2007 23:30:17 Art wrote:
          > Greetings,
          >
          > Many who have the DDS-60 don't use them much, and even fewer than
          > that feed them into a transmitter due to the multitude of close in
          > spurs. Even those that use them for receivers note that the spur
          > content is high enough so that the MDS is impacted on receive. I have
          > never built my DDS-60 in fact, it's still in the sealed bag!
          >
          > We can clean up the spur ridden old technology DDS units such as the
          > DDS-60 by using it to feed a PLL (as the reference frequency) and
          > using the output of the PLL to feed into our radios. While PLL's are
          > known for having phase noise, it can be minimized if the PLL doesn't
          > have to lock on a new frequency quickly-because we can use a lower
          > loop gain. The point being that we can make a PLL that is more
          > suitable for this purpose by careful design rather than using a
          > general purpose design.
          >
          > Many of us want higher frequencies to run quadrature based detectors
          > as well, so the PLL becomes a spectrally clean means to double or
          > quadruple or multiply the output frequency by a very large margin.
          >
          > It seems the DDS-60 and similar DDS units can be operated in
          > conjunction with PLL technology to make them useful again rather than
          > shunned as a little too dirty for prime time use::>
          >
          > There are thousands of DDS-60 units out there, and I wonder if a
          > companion PLL can be built to accommodate them and the many other DDS
          > units that have less than stellar performance?
          >
          > Additionally, I can't see any practical reason to upgrade to newer
          > DDS units that use much more power due to higher clocking rates
          > (plus, they need multipliers on their clock inputs just to have the
          > clock operate at 500 MHz). Does anyone else here see the
          > practicality of the old technology DDS + PLL combination, or is it just
          > me??
          >
          > Regards,
          >
          > Art


          Art,

          You are partly correct. There are no reason to dump the older generation DDS
          designs a they can be re-vitalised as PLL references. This works OK as long
          as you don't expect to sweep over the frequency range or change frequency to
          often.

          However, When using these DDS chips with PLL's with a high N rate, you benefit
          from the 14 bit resolution but one of the more important reasons is the
          increased frequency agility of the new breed of DDS chips.

          I use the older AD9850 and the AD9851 in combination with the PLL chip
          ADF4007. I'm using an extra external divider to be able to run the vco well
          up in the microwave bands. And as you say, "slow" frequency changes are not a
          problem.

          OK - so your DDS don't reach 440MHz because it's an AD9850? Wrong! it does.

          Look in the DDS information from Analog about aliased images.

          If you intend to use your DDS as a reference to drive a spur insensitive
          application like a PLL, you can use the DDS aliased images even if you might
          add 6 or 12 dB (or so... ) gain after the DDS output. Also, the DDS output
          filter needs changes as well. There area number of blind spots in the
          frequency range but that can be controlled by using a suitable system clock
          for the DDS. And if you intend to run low frequencies like 440MHz or so,
          there are cheap PLLs available like the ADF4007 from Analog Devices.

          So don't bin your old DDS-60 yet!

          //Dan


          --
          Dan Andersson, M0DFI
          dan@...
        • Kees & Sandy
          Using the AD9851 .....is there anything to be gained by better grounding schemes ? analog and digital ground isolation (seems like that was key according to
          Message 4 of 21 , Jan 31, 2007
          • 0 Attachment

            Using the AD9851 .....is there anything to be gained by better grounding schemes ? analog and digital ground isolation (seems like that was key according to AD), etc to improve (but not eliminate) the spur situation ? Just because the DDS-60 has a certain magnitude of certain spurs, doesn't mean all designs using the AD98xx do.

            73 Kees K5BCQ

              

            -- Dan Andersson <dan@...> wrote:
            On Wednesday 31 January 2007 23:30:17 Art wrote:

            > Greetings,
            >
            > Many who have the DDS-60 don't use them much, and even fewer than
            > that feed them into a transmitter due to the multitude of close in
            > spurs. Even those that use them for receivers note that the spur
            > content is high enough so that the MDS is impacted on receive. I have
            > never built my DDS-60 in fact, it's still in the sealed bag!
            >
            > We can clean up the spur ridden old technology DDS units such as the
            > DDS-60 by using it to feed a PLL (as the reference frequency) and
            > using the output of the PLL to feed into our radios. While PLL's are
            > known for having phase noise, it can be minimized if the PLL doesn't
            > have to lock on a new frequency quickly-because we can use a lower
            > loop gain. The point being that we can make a PLL that is more
            > suitable for this purpose by careful design rather than using a
            > general purpose design.
            >
            > Many of us want higher frequencies to run quadrature based detectors
            > as well, so the PLL becomes a spectrally clean means to double or
            > quadruple or multiply the output frequency by a very large margin.
            >
            > It seems the DDS-60 and similar DDS units can be operated in
            > conjunction with PLL technology to make them useful again rather than
            > shunned as a little too dirty for prime time use::>
            >
            > There are thousands of DDS-60 units out there, and I wonder if a
            > companion PLL can be built to accommodate them and the many other DDS
            > units that have less than stellar performance?
            >
            > Additionally, I can't see any practical reason to upgrade to newer
            > DDS units that use much more power due to higher clocking rates
            > (plus, they need multipliers on their clock inputs just to have the
            > clock operate at 500 MHz).  Does anyone else here see the
            > practicality of the old technology DDS + PLL combination, or is it just
            > me??
            >
            > Regards,
            >
            > Art


            Art,

            You are partly correct. There are no reason to dump the older generation DDS 
            designs a they can be re-vitalised as PLL references. This works OK as long 
            as you don't expect to sweep over the frequency range or change frequency to 
            often. 

            However, When using these DDS chips with PLL's with a high N rate, you benefit 
            from the 14 bit resolution but one of the more important reasons is the 
            increased frequency agility of the new breed of DDS chips.

            I use the older AD9850 and the AD9851 in combination with the PLL chip 
            ADF4007. I'm using an extra external divider to be able to run the vco well 
            up in the microwave bands. And as you say, "slow" frequency changes are not a 
            problem.

            OK - so your DDS don't reach 440MHz because it's an AD9850? Wrong! it does.

            Look in the DDS information from Analog about aliased images.

            If you intend to use your DDS as a reference to drive a spur insensitive 
            application like a PLL, you can use the DDS aliased images even if you might 
            add 6 or 12 dB (or so... ) gain after the DDS output. Also, the DDS output 
            filter needs changes as well. There area number of blind spots in the 
            frequency range but that can be controlled by using a suitable system clock 
            for the DDS. And if you intend to run low frequencies like 440MHz or so, 
            there are cheap PLLs available like the ADF4007 from Analog Devices.   

            So don't bin your old DDS-60 yet!

            //Dan


            -- 
            Dan Andersson, M0DFI
            dan@... 




             
            Yahoo! Groups Links

            <*> To visit your group on the web, go to:
                http://groups.yahoo.com/group/softrock40/

            <*> Your email settings:
                Individual Email | Traditional

            <*> To change settings online go to:
                http://groups.yahoo.com/group/softrock40/join
                (Yahoo! ID required)

            <*> To change settings via email:
                mailto:softrock40-digest@yahoogroups.com 
                mailto:softrock40-fullfeatured@yahoogroups.com

            <*> To unsubscribe from this group, send an email to:
                softrock40-unsubscribe@yahoogroups.com

            <*> Your use of Yahoo! Groups is subject to:
                http://docs.yahoo.com/info/terms/
             

          • Graham Haddock
            Art: The DDS-60 was intended to be an educational tool, not a universal high performance VFO. I think it does that beautifully. Learn about how to program,
            Message 5 of 21 , Jan 31, 2007
            • 0 Attachment
              Art:

              The DDS-60 was intended to be an educational tool, not
              a universal high performance VFO. I think it does that beautifully.
              Learn about how to program, control and manage a DDS, with a simple
              PIC controller. It's performance is good enough to be a useful
              signal generator, sweeper, drive an antenna analyzer, etc.

              I think your time looking for a higher performance universal
              VFO will lead you in the direction of the higher performance
              DDS, like the 9951 series, rather than coming up with the
              ultimate PLL. Single band or limited bands, yes you can
              come up with a PLL, but all band HF coverage will take
              multiple PLLs. If you start with a 9851, it might take more
              than one stage of PLL filtering to get to world class receiver
              LO performance.

              The 9951 series has a lot more capability, but needs more
              control lines to manage it. I have one running from a 125 MHz
              chip oscillator. No multiplier needed. I don't perceive any
              jitter or spur problems that will limit my attaching it to a
              transmitter. It is not quite as good as the Drake R8, so
              still not in high performance receiver territory, but still
              some 20+ dB better than the 9850/9851.

              --- Graham

              Art wrote:
              >
              > Greetings,
              >
              > Many who have the DDS-60 don't use them much, and even fewer than
              > that feed them into a transmitter due to the multitude of close in
              > spurs. Even those that use them for receivers note that the spur
              > content is high enough so that the MDS is impacted on receive. I have
              > never built my DDS-60 in fact, it's still in the sealed bag!
              >
              > We can clean up the spur ridden old technology DDS units such as the
              > DDS-60 by using it to feed a PLL (as the reference frequency) and
              > using the output of the PLL to feed into our radios. While PLL's are
              > known for having phase noise, it can be minimized if the PLL doesn't
              > have to lock on a new frequency quickly-because we can use a lower
              > loop gain. The point being that we can make a PLL that is more
              > suitable for this purpose by careful design rather than using a
              > general purpose design.
              >
              > Many of us want higher frequencies to run quadrature based detectors
              > as well, so the PLL becomes a spectrally clean means to double or
              > quadruple or multiply the output frequency by a very large margin.
              >
              > It seems the DDS-60 and similar DDS units can be operated in
              > conjunction with PLL technology to make them useful again rather than
              > shunned as a little too dirty for prime time use::>
              >
              > There are thousands of DDS-60 units out there, and I wonder if a
              > companion PLL can be built to accommodate them and the many other DDS
              > units that have less than stellar performance?
              >
              > Additionally, I can't see any practical reason to upgrade to newer
              > DDS units that use much more power due to higher clocking rates
              > (plus, they need multipliers on their clock inputs just to have the
              > clock operate at 500 MHz). Does anyone else here see the
              > practicality of the old technology DDS + PLL combination, or is it
              > just me??
              >
              > Regards,
              >
              > Art
              >
              >
            • Leon Heller
              ... From: Art To: Sent: Wednesday, January 31, 2007 11:30 PM Subject: [softrock40] old generation DDS + PLL?
              Message 6 of 21 , Jan 31, 2007
              • 0 Attachment
                ----- Original Message -----
                From: "Art" <KY1K@...>
                To: <softrock40@yahoogroups.com>
                Sent: Wednesday, January 31, 2007 11:30 PM
                Subject: [softrock40] old generation DDS + PLL?


                > Greetings,
                >
                > Many who have the DDS-60 don't use them much, and even fewer than
                > that feed them into a transmitter due to the multitude of close in
                > spurs. Even those that use them for receivers note that the spur
                > content is high enough so that the MDS is impacted on receive. I have
                > never built my DDS-60 in fact, it's still in the sealed bag!
                >
                > We can clean up the spur ridden old technology DDS units such as the
                > DDS-60 by using it to feed a PLL (as the reference frequency) and
                > using the output of the PLL to feed into our radios. While PLL's are
                > known for having phase noise, it can be minimized if the PLL doesn't
                > have to lock on a new frequency quickly-because we can use a lower
                > loop gain. The point being that we can make a PLL that is more
                > suitable for this purpose by careful design rather than using a
                > general purpose design.
                >
                > Many of us want higher frequencies to run quadrature based detectors
                > as well, so the PLL becomes a spectrally clean means to double or
                > quadruple or multiply the output frequency by a very large margin.
                >
                > It seems the DDS-60 and similar DDS units can be operated in
                > conjunction with PLL technology to make them useful again rather than
                > shunned as a little too dirty for prime time use::>
                >
                > There are thousands of DDS-60 units out there, and I wonder if a
                > companion PLL can be built to accommodate them and the many other DDS
                > units that have less than stellar performance?
                >
                > Additionally, I can't see any practical reason to upgrade to newer
                > DDS units that use much more power due to higher clocking rates
                > (plus, they need multipliers on their clock inputs just to have the
                > clock operate at 500 MHz). Does anyone else here see the
                > practicality of the old technology DDS + PLL combination, or is it just
                > me??

                It's a good solution (I think my FT-817ND uses it), and the same controller
                could be used for both the DDS and the PLL. The Nat Semi PLL chips are quite
                cheap and easy to get hold of, I started designing a PCB for a PLL using one
                but never got round to finishing it. I could resurrect the project if there
                is any interest.

                73, Leon
                --
                Leon Heller
                Amateur radio call-sign G1HSM
                Yaesu FT-817ND transceiver
                Suzuki SV1000S motorcycle
                leon355@...
                http://www.geocities.com/leon_heller
              • Art
                ... I m not sure there is enough interest Leon, the reception here has been somewhat (barely) luke warm. But, I know they made a million of those DDS-60 s and
                Message 7 of 21 , Jan 31, 2007
                • 0 Attachment
                  >
                  >
                  >It's a good solution (I think my FT-817ND uses it), and the same controller
                  >could be used for both the DDS and the PLL. The Nat Semi PLL chips are quite
                  >cheap and easy to get hold of, I started designing a PCB for a PLL using one
                  >but never got round to finishing it. I could resurrect the project if there
                  >is any interest.


                  I'm not sure there is enough interest Leon, the reception here has
                  been somewhat (barely) luke warm. But, I know they made a million of
                  those DDS-60's and few are happy with them. If someone made a
                  companion board to go with the DDS-60 (same size, maybe stackable),
                  it would seem to me that many would buy it.

                  I've been looking at the PLL's commonly on the market today-many (the
                  majority) are called clock multiplier PLL's, with a built in rc
                  oscillator that typically operates over a very very wide frequency
                  range. These are fairly common among clock multiplier PLL's, and this
                  type of vco needs to be avoided like the plague::>

                  Anything with an outboard VCO should perform well, as long as the vco
                  itself is clean and the loop gain is kept low. The Analog Devices
                  AD4007 PLL as mentioned by Dan Anderson is very easy to use, with
                  fixed 8, 16, 32 or 64 divide ratios that can be set with 2 inputs.

                  I might clock the DDS with a 20 Mhz crystal, set the divide by at 32
                  and ask the DDS to produce 100 KHz to 5 MHz-giving me a clean 3.2 MHz
                  to 160 MHz capability. I suspect several vco ranges would be needed,
                  so some inductors and caps could be switched in/out with dip switches
                  on the PCB. This would give me a relatively low power DDS with at
                  least 4 samples for each output cycle, so the output signal would be
                  fairly pure.

                  I didn't happen to see the article in QEX about this, anyone know if
                  it is published on the web?

                  Regards,

                  Art
                • Terry
                  From Terry, WB4JFI: I have recently moved, and am still trying to find all my magazines. However, I did find a two-part article in the March/April 2001 and
                  Message 8 of 21 , Feb 1, 2007
                  • 0 Attachment
                    From Terry, WB4JFI:

                    I have recently moved, and am still trying to find all my magazines.
                    However, I did find a two-part article in the March/April 2001 and
                    May/June 2001 QEX, titled "Beyond Fractional-N" parts 1 & 2. It is
                    more of a microwave DDSdriven PLL LO. Id does use an AD9850. The LO
                    is supposed to be used in an upconverting HF rig with a 75MHz IF, so
                    there may be some interest for us. The freq range is 770-1050MHz,
                    then uses a divide-by-ten to improve phase noise.

                    That article isn't the one I remembered, however. I will look
                    through magazines as I get a chance. It may have been in a QST
                    instea of QEX, it seems like it was about three years ago.

                    Sorry I couldn't find it.
                    Terry


                    --- In softrock40@yahoogroups.com, Art <KY1K@...> wrote:
                    >
                    >
                    > >
                    > >
                    > >It's a good solution (I think my FT-817ND uses it), and the same
                    controller
                    > >could be used for both the DDS and the PLL. The Nat Semi PLL chips
                    are quite
                    > >cheap and easy to get hold of, I started designing a PCB for a PLL
                    using one
                    > >but never got round to finishing it. I could resurrect the project
                    if there
                    > >is any interest.
                    >
                    >
                    > I'm not sure there is enough interest Leon, the reception here has
                    > been somewhat (barely) luke warm. But, I know they made a million
                    of
                    > those DDS-60's and few are happy with them. If someone made a
                    > companion board to go with the DDS-60 (same size, maybe stackable),
                    > it would seem to me that many would buy it.
                    >
                    > I've been looking at the PLL's commonly on the market today-many
                    (the
                    > majority) are called clock multiplier PLL's, with a built in rc
                    > oscillator that typically operates over a very very wide frequency
                    > range. These are fairly common among clock multiplier PLL's, and
                    this
                    > type of vco needs to be avoided like the plague::>
                    >
                    > Anything with an outboard VCO should perform well, as long as the
                    vco
                    > itself is clean and the loop gain is kept low. The Analog Devices
                    > AD4007 PLL as mentioned by Dan Anderson is very easy to use, with
                    > fixed 8, 16, 32 or 64 divide ratios that can be set with 2 inputs.
                    >
                    > I might clock the DDS with a 20 Mhz crystal, set the divide by at
                    32
                    > and ask the DDS to produce 100 KHz to 5 MHz-giving me a clean 3.2
                    MHz
                    > to 160 MHz capability. I suspect several vco ranges would be
                    needed,
                    > so some inductors and caps could be switched in/out with dip
                    switches
                    > on the PCB. This would give me a relatively low power DDS with at
                    > least 4 samples for each output cycle, so the output signal would
                    be
                    > fairly pure.
                    >
                    > I didn't happen to see the article in QEX about this, anyone know
                    if
                    > it is published on the web?
                    >
                    > Regards,
                    >
                    > Art
                    >
                  • Len Warner
                    ... Those seem to be the articles available here:- http://www.doug-smith.net/drentea1.pdf http://www.doug-smith.net/drentea2.pdf Regards, LenW
                    Message 9 of 21 , Feb 1, 2007
                    • 0 Attachment
                      At 7:26 am ((PST)) Thu Feb 1, 2007, Terry wrote:
                      >However, I did find a two-part article in the March/April 2001 and
                      >May/June 2001 QEX, titled "Beyond Fractional-N" parts 1 & 2.[snip]
                      >there may be some interest for us.

                      Those seem to be the articles available here:-
                      http://www.doug-smith.net/drentea1.pdf
                      http://www.doug-smith.net/drentea2.pdf



                      Regards, LenW
                    • Len Warner
                      ... Could this be the missing article? http://www.thegleam.com/ke5fx/synth/article.doc A Versatile Hybrid Synthesizer for UHF/Microwave Projects John Miles
                      Message 10 of 21 , Feb 2, 2007
                      • 0 Attachment
                        At 7:26 am ((PST)) Thu Feb 1, 2007, Terry wrote:
                        >I have recently moved, and am still trying to find all my magazines.
                        >...find...article in...2001 QEX, "Beyond Fractional-N" parts 1 & 2.
                        >
                        >That article isn't the one I remembered, however. [snip]
                        >it seems like it was about three years ago.
                        >
                        >Sorry I couldn't find it.

                        Could this be the missing article?

                        http://www.thegleam.com/ke5fx/synth/article.doc

                        A Versatile Hybrid Synthesizer for UHF/Microwave Projects
                        John Miles KE5FX
                        Richard Hosking VK6BRO

                        | Authors' draft of article published in
                        | March/April 2004 QEX/Communications Quarterly
                        |
                        | Please refer to <http://www.qsl.net/ke5fx/synth.html> [**]
                        | for the latest updates and errata"

                        [** Redirects to http://www.thegleam.com/ke5fx/synth/synth.html ]

                        The design is based on an AD9852.


                        Regards, LenW
                      • Len Warner
                        ... Could this be the missing article? http://www.thegleam.com/ke5fx/synth/article.doc A Versatile Hybrid Synthesizer for UHF/Microwave Projects John Miles
                        Message 11 of 21 , Feb 2, 2007
                        • 0 Attachment
                          At 7:26 am ((PST)) Thu Feb 1, 2007, Terry wrote:
                          >I have recently moved, and am still trying to find all my magazines.
                          >...find...article in...2001 QEX, "Beyond Fractional-N" parts 1 & 2.
                          >That article isn't the one I remembered [snip]

                          Could this be the missing article?
                          http://www.thegleam.com/ke5fx/synth/article.doc
                          A Versatile Hybrid Synthesizer for UHF/Microwave Projects
                          John Miles KE5FX & Richard Hosking VK6BRO
                          [ Authors' draft of article in March/April 2004 QEX
                          See also http://www.thegleam.com/ke5fx/synth/synth.html ]

                          You may also find these Fractional-N Synthesizer
                          discussions are useful background.

                          http://www.skyworksinc.com/products_display_item.asp?did=4722
                          Dual Fractional-N Synthesizers/PLLs 103037C.pdf

                          http://www.skyworksinc.com/products_display_item.asp?did=4725
                          Basics of Dual Fractional-N Synthesizers 101463B.pdf


                          Regards, LenW
                        • Art
                          ... No, that s not it. The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title is A PLL Spur Eliminator for DDS VFOs . I looked on the web using
                          Message 12 of 21 , Feb 2, 2007
                          • 0 Attachment
                            >
                            > >Sorry I couldn't find it.
                            >
                            >Could this be the missing article?
                            >
                            >http://www.thegleam.com/ke5fx/synth/article.doc


                            No, that's not it.

                            The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title is
                            'A PLL Spur Eliminator for DDS VFOs'.

                            I looked on the web using google, got no hits.

                            Art
                          • FRANCIS CARCIA
                            I think the article was by Drenta not John Miles. John did a very nice job with his set up and looks pretty clean. Len Warner wrote: ...
                            Message 13 of 21 , Feb 2, 2007
                            • 0 Attachment
                              I think the article was by Drenta not John Miles.
                              John did a very nice job with his set up and looks pretty clean.

                              Len Warner <novost@...> wrote:
                              At 7:26 am ((PST)) Thu Feb 1, 2007, Terry wrote:
                              >I have recently moved, and am still trying to find all my magazines.
                              >...find...article in...2001 QEX, "Beyond Fractional-N" parts 1 & 2.
                              >That article isn't the one I remembered [snip]

                              Could this be the missing article?
                              http://www.thegleam .com/ke5fx/ synth/article. doc
                              A Versatile Hybrid Synthesizer for UHF/Microwave Projects
                              John Miles KE5FX & Richard Hosking VK6BRO
                              [ Authors' draft of article in March/April 2004 QEX
                              See also http://www.thegleam .com/ke5fx/ synth/synth. html ]

                              You may also find these Fractional-N Synthesizer
                              discussions are useful background.

                              http://www.skyworks inc.com/products _display_ item.asp? did=4722
                              Dual Fractional-N Synthesizers/ PLLs 103037C.pdf

                              http://www.skyworks inc.com/products _display_ item.asp? did=4725
                              Basics of Dual Fractional-N Synthesizers 101463B.pdf

                              Regards, LenW


                            • Terry
                              This article is further back than I rembered Art, but I believe you are correct. That sounds like the one I was remembering. I cannot find my QEX mags from
                              Message 14 of 21 , Feb 2, 2007
                              • 0 Attachment
                                This article is further back than I rembered Art, but I believe you
                                are correct. That sounds like the one I was remembering. I cannot
                                find my QEX mags from that time yet.
                                Terry


                                --- In softrock40@yahoogroups.com, Art <KY1K@...> wrote:
                                >
                                >
                                > >
                                > > >Sorry I couldn't find it.
                                > >
                                > >Could this be the missing article?
                                > >
                                > >http://www.thegleam.com/ke5fx/synth/article.doc
                                >
                                >
                                > No, that's not it.
                                >
                                > The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title
                                is
                                > 'A PLL Spur Eliminator for DDS VFOs'.
                                >
                                > I looked on the web using google, got no hits.
                                >
                                > Art
                                >
                              • gandalfg8@aol.com
                                In a message dated 02/02/2007 13:10:17 GMT Standard Time, KY1K@verizon.net writes: No, that s not it. The missing article is in Jul 2000 QEX, page 40, by
                                Message 15 of 21 , Feb 2, 2007
                                • 0 Attachment
                                  In a message dated 02/02/2007 13:10:17 GMT Standard Time, KY1K@... writes:
                                  No, that's not it.

                                  The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title is
                                  'A PLL Spur Eliminator for DDS VFOs'.

                                  I looked on the web using google, got no hits.

                                  Art
                                  ------------------------------------------------------------
                                  Hi Art
                                   
                                  I've just extracted a copy of the article and will email it to you direct.
                                  I haven't posted it in the files section as I'm not sure about copyright but please feel free to pass it on if anyone else wants a copy.
                                   
                                  regards
                                   
                                  Nigel
                                  GM8PZR
                                • Art
                                  Well, it might be a little dated, most of the synthesized chips made back then aren t made today. But, I m sure it talks about vco design to keep the phase
                                  Message 16 of 21 , Feb 2, 2007
                                  • 0 Attachment
                                    Well, it might be a little dated, most of the synthesized chips made
                                    back then aren't made today.

                                    But, I'm sure it talks about vco design to keep the phase noise down-

                                    I'd gladly sacrifice a slower lock up time if I could get low phase noise.

                                    I think the PLL filter will make/break this project....so it's
                                    important to do the synthesizer right.

                                    I also read a technical bulletin last year about modifying the method
                                    of supplying correction voltages to the vco-by doing so, there was
                                    some amplitude variations of the output, but much less phase noise. I
                                    can't remember or locate that document now-unfortuinately::>

                                    Regards,

                                    Art


                                    At 10:41 AM 2/2/2007, you wrote:
                                    >This article is further back than I rembered Art, but I believe you
                                    >are correct. That sounds like the one I was remembering. I cannot
                                    >find my QEX mags from that time yet.
                                    >Terry
                                    >
                                    >
                                    >--- In softrock40@yahoogroups.com, Art <KY1K@...> wrote:
                                    > >
                                    > >
                                    > > >
                                    > > > >Sorry I couldn't find it.
                                    > > >
                                    > > >Could this be the missing article?
                                    > > >
                                    > > >http://www.thegleam.com/ke5fx/synth/article.doc
                                    > >
                                    > >
                                    > > No, that's not it.
                                    > >
                                    > > The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title
                                    >is
                                    > > 'A PLL Spur Eliminator for DDS VFOs'.
                                    > >
                                    > > I looked on the web using google, got no hits.
                                    > >
                                    > > Art
                                    > >
                                    >
                                    >
                                    >
                                    >
                                    >
                                    >Yahoo! Groups Links
                                    >
                                    >
                                    >
                                    >
                                    >
                                    >--
                                    >No virus found in this incoming message.
                                    >Checked by AVG Free Edition.
                                    >Version: 7.5.432 / Virus Database: 268.17.14/658 - Release Date:
                                    >1/29/2007 2:49 PM
                                  • p_a_0jrv
                                    if possible i like also to have a copy from that article jan pa0jrv ... direct. ... copyright but
                                    Message 17 of 21 , Feb 2, 2007
                                    • 0 Attachment
                                      if possible i like also to have a copy from that article
                                      jan pa0jrv
                                      --- In softrock40@yahoogroups.com, gandalfg8@... wrote:
                                      >
                                      >
                                      > In a message dated 02/02/2007 13:10:17 GMT Standard Time, KY1K@...
                                      > writes:
                                      >
                                      > No, that's not it.
                                      >
                                      > The missing article is in Jul 2000 QEX, page 40, by WA6NUT. Title is
                                      > 'A PLL Spur Eliminator for DDS VFOs'.
                                      >
                                      > I looked on the web using google, got no hits.
                                      >
                                      > Art
                                      >
                                      >
                                      >
                                      > ------------------------------------------------------------
                                      > Hi Art
                                      >
                                      > I've just extracted a copy of the article and will email it to you
                                      direct.
                                      > I haven't posted it in the files section as I'm not sure about
                                      copyright but
                                      > please feel free to pass it on if anyone else wants a copy.
                                      >
                                      > regards
                                      >
                                      > Nigel
                                      > GM8PZR
                                      >
                                    • jabauzit
                                      Most of the references posted previously are old, not that old but DDS technology evolves quite rapidly. At the time these projects were quite performant and
                                      Message 18 of 21 , Feb 2, 2007
                                      • 0 Attachment
                                        Most of the references posted previously are old, not that old but DDS
                                        technology evolves quite rapidly. At the time these projects were
                                        quite performant and the only ones doable, but rather complicated and
                                        expensive.
                                        Today using a DDS-60 in such projects would still be complicated and
                                        expensive. Using an AD-99XX would be a lot simpler and cheaper.
                                        There are some projects which are simpler but would be limited to one
                                        band. For example:
                                        http://www.gw0etf.btinternet.co.uk/Pages/vfo.htm

                                        The article mentions a page with tools to calculate PLL parameters,
                                        the address is no longer valid but can be found at:
                                        http://home.mchsi.com/~kd9jq/hamradio/kd9jq.html

                                        The one-on-one Tracking PLL from EMRFD is also a simple way to do it.

                                        Jean-Claude PJ2BVU
                                      • Art
                                        Thanks Jean-Claude, I know the newer DDS units work better and don t need PLL s to clean them up. But, tons of us had DDS-60 boards and suffer from less than
                                        Message 19 of 21 , Feb 2, 2007
                                        • 0 Attachment
                                          Thanks Jean-Claude,

                                          I know the newer DDS units work better and don't need PLL's to clean
                                          them up. But, tons of us had DDS-60 boards and suffer from less than
                                          stellar performance. It seems that a basic PLL with a fixed
                                          multiplier used in conjunction with a DDS-60 would work well and
                                          would be the answer for those who might want higher frequencies too.

                                          Building multiplier stage(s) with a crystal oscillator input just to
                                          get the clock up to 500 MHz is not my idea of fun either, and that's
                                          exactly what you have to do to get DSS frequencies over 100 MHz. And,
                                          with clock frequencies that high, the PLL chip nearly melts, again,
                                          not my idea of fun.

                                          But, the PLL happily runs at higher frequencies, and doesn't draw
                                          much extra power in doing so....it just seems the PLL/DDS offers
                                          higher frequencies without needing an atomic power plant to run the
                                          DDS::> While it's true the VCO would not operate over wide ranges, HF
                                          could be covered with 3 VCO's and phase noise can be minimized by
                                          careful design of the VCO.

                                          Whether DDS/PLL is the right answer for many of us remains to be seen.

                                          I don't know about the one-on-one Tracking PLL from EMRFD, can you
                                          give any more information about it???

                                          Regards,

                                          Art



                                          At 12:01 AM 2/3/2007, you wrote:
                                          >Most of the references posted previously are old, not that old but DDS
                                          >technology evolves quite rapidly. At the time these projects were
                                          >quite performant and the only ones doable, but rather complicated and
                                          >expensive.
                                          >Today using a DDS-60 in such projects would still be complicated and
                                          >expensive. Using an AD-99XX would be a lot simpler and cheaper.
                                          >There are some projects which are simpler but would be limited to one
                                          >band. For example:
                                          >http://www.gw0etf.btinternet.co.uk/Pages/vfo.htm
                                          >
                                          >The article mentions a page with tools to calculate PLL parameters,
                                          >the address is no longer valid but can be found at:
                                          >http://home.mchsi.com/~kd9jq/hamradio/kd9jq.html
                                          >
                                          >The one-on-one Tracking PLL from EMRFD is also a simple way to do it.
                                          >
                                          >Jean-Claude PJ2BVU
                                          >
                                          >
                                          >
                                          >
                                          >Yahoo! Groups Links
                                          >
                                          >
                                          >
                                          >
                                          >
                                          >--
                                          >No virus found in this incoming message.
                                          >Checked by AVG Free Edition.
                                          >Version: 7.5.432 / Virus Database: 268.17.14/658 - Release Date:
                                          >1/29/2007 2:49 PM
                                        • Leon
                                          ... From: Art To: Sent: Saturday, February 03, 2007 6:12 AM Subject: Re: [softrock40] Re: old generation DDS
                                          Message 20 of 21 , Feb 3, 2007
                                          • 0 Attachment
                                            ----- Original Message -----
                                            From: "Art" <KY1K@...>
                                            To: <softrock40@yahoogroups.com>
                                            Sent: Saturday, February 03, 2007 6:12 AM
                                            Subject: Re: [softrock40] Re: old generation DDS + PLL?


                                            > Thanks Jean-Claude,
                                            >
                                            > I know the newer DDS units work better and don't need PLL's to clean
                                            > them up. But, tons of us had DDS-60 boards and suffer from less than
                                            > stellar performance. It seems that a basic PLL with a fixed
                                            > multiplier used in conjunction with a DDS-60 would work well and
                                            > would be the answer for those who might want higher frequencies too.
                                            >
                                            > Building multiplier stage(s) with a crystal oscillator input just to
                                            > get the clock up to 500 MHz is not my idea of fun either, and that's
                                            > exactly what you have to do to get DSS frequencies over 100 MHz. And,
                                            > with clock frequencies that high, the PLL chip nearly melts, again,
                                            > not my idea of fun.
                                            >
                                            > But, the PLL happily runs at higher frequencies, and doesn't draw
                                            > much extra power in doing so....it just seems the PLL/DDS offers
                                            > higher frequencies without needing an atomic power plant to run the
                                            > DDS::> While it's true the VCO would not operate over wide ranges, HF
                                            > could be covered with 3 VCO's and phase noise can be minimized by
                                            > careful design of the VCO.
                                            >
                                            > Whether DDS/PLL is the right answer for many of us remains to be seen.
                                            >
                                            > I don't know about the one-on-one Tracking PLL from EMRFD, can you
                                            > give any more information about it???

                                            I've just ordered a couple of sample ADF4007 PLL chips.

                                            73, Leon
                                            --
                                            Leon Heller
                                            Amateur radio call-sign G1HSM
                                            Yaesu FT-817ND transceiver
                                            Suzuki SV1000S motorcycle
                                            leon355@...
                                            http://www.geocities.com/leon_heller
                                          • jabauzit
                                            ... Yes the clock is problematic. ... The link for the HW-9 VFO is one and uses several VCOs and Xtal oscillators. The one described in EMRFD has only one VCO
                                            Message 21 of 21 , Feb 3, 2007
                                            • 0 Attachment
                                              --- In softrock40@yahoogroups.com, Art <KY1K@...> wrote:
                                              >
                                              > Thanks Jean-Claude,
                                              >
                                              .....
                                              >
                                              > Building multiplier stage(s) with a crystal oscillator input just to
                                              > get the clock up to 500 MHz is not my idea of fun either,

                                              Yes the clock is problematic.

                                              >
                                              > I don't know about the one-on-one Tracking PLL from EMRFD, can you
                                              > give any more information about it???
                                              >

                                              The link for the HW-9 VFO is one and uses several VCOs and Xtal
                                              oscillators.
                                              The one described in EMRFD has only one VCO for the 14MHz band, which
                                              is mixed with a 12.5MHz Xtal oscillator, the resulting 1.5MHz signal
                                              is filtered by a 1.7MHz BPF and goes to one of the input of the phase
                                              comparator - 74LS74. The reference oscillator is a 1.5MHz analog VFO -
                                              would be replaced by the DDS - and goes to the other input of the
                                              phase comparator. The output goes to the OP amp controlling the VCO.
                                              There is no divider, the reference oscillator is variable and the VCO
                                              tracks the reference oscillator Hz for Hz. That should clean most of
                                              the spurs, only what is in the bandwidth of the loop filter will show
                                              up: 10kHz for the EMRFD one and 5kHz for the HW-9.

                                              Jean-Claude PJ2BVU
                                            Your message has been successfully submitted and would be delivered to recipients shortly.