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

Re: Ensemble RXTX for 40/30/20 debate

Expand Messages
  • warrenallgyer
    As I hit SEND one other possibility occurred to me. It could be that I had an intermittent cable connection to the resistive load. That has happened before and
    Message 1 of 25 , Jan 5, 2013
    • 0 Attachment
      As I hit SEND one other possibility occurred to me. It could be that I had an intermittent cable connection to the resistive load. That has happened before and it would explain higher than normal voltages measured.

      At this point, I am not curious enough to disassemble it again to confirm..... it is working too well now and I am, in fact playing with David's new 96 KHz software for the Peaberry.

      For my data I have made a note of the cable possibility and I am treating those measurements as only relative indications of the power among the bands.

      Warren A.

      --- In softrock40@yahoogroups.com, "warrenallgyer" wrote:
      >
      >
      > Alan
      >
      > You are not going to upset me at all. I scratched my head when I recorded the results and I double checked them just because they did not make sense to me either.
      >
      > My only possible explanation: When I am measuring the output directly from T4 it is into a solid resistive load. Once this is connected to the LPF I have no idea the actual load. I do know, although I did not note the values, the total current draw changed dramatically when I switched over to the filter.
      >
      > I think this has been one of the bugaboos of measurements taken at the T4 secondary with the filter in place. Clearly the old filter was rolling off significantly at 14 MHz and the actual impedance reflected back to the finals was likely not resistive and not 50 ohms. So the voltage measurements taken at that point are not necessarily comparable as you move down the skirt of the filter..... at least I think so!
      >
      > The reason I terminated and measured at T4 in the first place was to find out what actually was coming out of the finals.
      >
      > But that is what my gut is telling me and someone smarter than me will have to explain it.
      >
      > Or someone else can remove their filter and make confirming or better measurements. I am pretty confident of my numbers.... not confident at all of the reason.
      >
      > The third harmonic number at -6 is wrong however.... I was recording numbers in dBm and then calculating the attenuation in my head..... and the calculator did not kick in on that one.
      >
      > Alan, Alan, Alan (sigh)...... why would you EVER think you would upset me? Hahaha!
      >
      > Best regards,
      >
      > Warren Allgyer
      >
      >
      >
      > --- In softrock40@yahoogroups.com, "Alan" wrote:
      > >
      > >
      > > ----- Original Message -----
      > > Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate
      > >
      > >
      > > Warren,
      > >
      > > I'm going to try to upset you now:)
      > > No filter, 3.6 watts into 50 ohms on 40m? Just maybe.
      > > Put this through a filter and into 50 ohms, 5.7 watts? How come? I'd say impossible.
      > > And the 40m 3rd harmonic, -13dB no filter, -6dB with filter.
      > >
      > > Sorry to say it but these sort of figures make me doubtful of the rest.
      > > Are they typos or do you need to test again?
      > >
      > > Maybe it is due to errors in your temporary setup. The figures from your Peaberry mentioned in your subsequent post seem good.
      > >
      > > 73 Alan G4ZFQ
      > >
      > >
      > > >
      > > > I have concluded a set of tests this morning that gives me more confidence in the 30 meter harmonic performance using the new
      > > > filter values. I am seeing just 40-41 dB of second harmonic suppression using the new values and testing them two ways: one with
      > > > the toroids wound full coverage and one with them wound with approximately a 90 degree gap.
      > > >
      > > > My tests indicate the greatly improved performance on 20 meters makes this worth the change.
      > > >
      > > > Now the details of the latest set of my tests for those who are interested.
      > > >
      > > > I removed the output filter completely and put a temporary connector driven by the secondary of T4. In this test I was interested
      > > > to measure what is actually coming out of T4 using a stable resistive load. Here is what I found:
      > > >
      > > > T4 connected directly to a 50 ohm load
      > > > 7.03 MHz
      > > > 7.03 MHz Pout 3.6 W
      > > > 2nd Harmonic -32 dB
      > > > 3rd Harmonic -13 dB
      > > > 10.13 MHz
      > > > 10.13 MHz Pout 3.9 W
      > > > 2nd Harmonic -30 dB
      > > > 3rd Harmonic -18 dB
      > > > 14.03 MHz
      > > > 14.03 MHz Pout 1.4 W
      > > > 2nd Harmonic -28 dB
      > > > 3rd Harmonic -22 dB
      > > >
      > > > The significant fall-off in power output on 20 Meters needs further investigation. It is masked when the filter is in place and T4
      > > > is probed, I expect due to a high impedance presented by the filter at that point. Nevertheless there is enough power available if
      > > > we can get it through a properly constructed filter.
      > > >
      > > > Next I connected this output to an external filter using the original values and tested it with full coverage turns on the toroid
      > > > and 270 degree coverage. Results are below:
      > > >
      > > > With Original Filter, full coverage
      > > > 7.03 MHz
      > > > 7.03 MHz Pout 5.7 W (!)
      > > > 2nd Harmonic -33
      > > > 3rd Harmonic -6
      > > > 10.13 MHz
      > > > 10.13 MHz Pout 6.0 W
      > > > 2nd Harmonic -53 dB
      > > > 3rd Harmonic -
      > > > 14.03 MHz
      > > > 14.03 MHz Pout 0.8 W
      > > > 2nd Harmonic -59 dB
      > > > 3rd Harmonic -
      > > >
      > > >
      > > > With Original Filter, 270 degree coverage
      > > > 7.03 MHz
      > > > 7.03 MHz Pout 3.3 W
      > > > 2nd Harmonic -33 dB
      > > > 3rd Harmonic -39 dB
      > > > 10.13 MHz
      > > > 10.13 MHz Pout 2.6 W
      > > > 2nd Harmonic -47 dB
      > > > 3rd Harmonic > 60 dB
      > > > 14.03 MHz
      > > > 14.03 MHz Pout 0.6 W
      > > > 2nd Harmonic > 60 dB
      > > > 3rd Harmonic > 60 dB
      > > >
      > > > Some notes here: Obviously winding the toroids as David Turnbull suggests, full, evenly spaced turns makes this filter perform
      > > > better for 20 meters. However, the combination of lower 20 meter drive and the characteristics of this filter really do
      > > > significantly degrade 20 meter performance in both cases.
      > > >
      > > > Finally I tested the new filter values on an externally connected version, again with full coverage and with 270 degree coverage
      > > > on the toroids. Those results:
      > > >
      > > > With New Filter, full coverage
      > > > 7.03 MHz
      > > > 7.03 MHz Pout 4.5 W
      > > > 2nd Harmonic -32 dB
      > > > 3rd Harmonic -34 dB
      > > > 10.13 MHz
      > > > 10.13 MHz Pout 5.1 W
      > > > 2nd Harmonic -41 dB
      > > > 3rd Harmonic > 60 dB
      > > > 14.03 MHz
      > > > 14.03 MHz Pout 1.6 W
      > > > 2nd Harmonic > 60 dB
      > > > 3rd Harmonic > 60 dB
      > > >
      > > > With New Filter, 270 degreel coverage
      > > >
      > > > 7.03 MHz
      > > > 7.03 MHz Pout 6.1 W
      > > > 2nd Harmonic -30 dB
      > > > 3rd Harmonic -32 dB
      > > > 10.13 MHz
      > > > 10.13 MHz Pout 5.4 W
      > > > 2nd Harmonic -40 dB
      > > > 3rd Harmonic > 60 dB
      > > > 14.03 MHz
      > > > 14.03 MHz Pout 1.8 W
      > > > 2nd Harmonic > 60 dB
      > > > 3rd Harmonic > 60 dB
      > > >
      > > > Here I found significant improvement in the 20 meter output; 1.6-1.8 watts depending how you wound the toroids. The 30 meter
      > > > harmonic performance just squeezed in under the wire and this is a small concern but I am back to thinking, overall, this is a
      > > > good move.
      > > >
      > > > Hope this is helpful and I stress this is one test. Alan is working on something similar and it will be interesting to see if he
      > > > concurs.
      > > >
      > > > Best regards,
      > > >
      > > > Warren Allgyer - W8TOD
      > > >
      > > > --- In softrock40@yahoogroups.com, "R. R. (Robby) Robson" wrote:
      > > >>
      > > >> The changes were made to the RF I/O stage (
      > > >> http://wb5rvz.org/ensemble_rxtx/05_rf) and involved changed values for C24,
      > > >> C25, C26, L2, and L3. The new values are shown in the published
      > > >> documentation (along with a note describing the change and referencing the
      > > >> thread of which you spoke).
      > > >>
      > > >
      > >
      >
    • Parks, Tony
      Good Morning All, For now all RXTX Ensemble kits shipped will include the capacitors such that the previous or new LP filter for 40m/30m/20m may be built. I
      Message 2 of 25 , Jan 5, 2013
      • 0 Attachment
        Good Morning All,

        For now all RXTX Ensemble kits shipped will include the capacitors such that the previous or new LP filter for 40m/30m/20m may be built.

        I very much appreciate the efforts of Warren, Alan and others to get things right!

        73,
        Tony KB9YIG

        On Fri, Jan 4, 2013 at 11:55 PM, warrenallgyer <allgyer@...> wrote:
         

        I hope this is my final post on the subject. The production filter is now installed in my on-air Peaberry (identical RF section to the RXTX) with the following results:

        Power Supply is 13.8 VDC

        7.03 Vrms Out 13.1
        Watts Out 3.4
        2nd harmonic -30 dB
        IMD3 ref Carrier -28 dB

        10.13 Vrms Out 11.3
        Watts Out 2.6
        2nd harmonic -40 dB
        IMD3 ref Carrier -29 dB

        14.03 Vrms Out 9.2
        Watts Out 1.7
        2nd harmonic -58 dB
        IMD3 ref Carrier -25 dB

        I am going to call this one "finished" so I can try to get to installing David's new 96 KHz software before the weekend vanishes.

        Warren Allgyer - W8TOD



        --- In softrock40@yahoogroups.com, "warrenallgyer" wrote:
        >
        > All
        >
        > I have concluded a set of tests this morning that gives me more confidence in the 30 meter harmonic performance using the new filter values. I am seeing just 40-41 dB of second harmonic suppression using the new values and testing them two ways: one with the toroids wound full coverage and one with them wound with approximately a 90 degree gap.
        >
        > My tests indicate the greatly improved performance on 20 meters makes this worth the change.
        >
        > Now the details of the latest set of my tests for those who are interested.
        >
        > I removed the output filter completely and put a temporary connector driven by the secondary of T4. In this test I was interested to measure what is actually coming out of T4 using a stable resistive load. Here is what I found:
        >
        > T4 connected directly to a 50 ohm load
        > 7.03 MHz
        > 7.03 MHz Pout 3.6 W
        > 2nd Harmonic -32 dB
        > 3rd Harmonic -13 dB
        > 10.13 MHz
        > 10.13 MHz Pout 3.9 W
        > 2nd Harmonic -30 dB
        > 3rd Harmonic -18 dB
        > 14.03 MHz
        > 14.03 MHz Pout 1.4 W
        > 2nd Harmonic -28 dB
        > 3rd Harmonic -22 dB
        >
        > The significant fall-off in power output on 20 Meters needs further investigation. It is masked when the filter is in place and T4 is probed, I expect due to a high impedance presented by the filter at that point. Nevertheless there is enough power available if we can get it through a properly constructed filter.
        >
        > Next I connected this output to an external filter using the original values and tested it with full coverage turns on the toroid and 270 degree coverage. Results are below:
        >
        > With Original Filter, full coverage
        > 7.03 MHz
        > 7.03 MHz Pout 5.7 W (!)
        > 2nd Harmonic -33
        > 3rd Harmonic -6
        > 10.13 MHz
        > 10.13 MHz Pout 6.0 W
        > 2nd Harmonic -53 dB
        > 3rd Harmonic -
        > 14.03 MHz
        > 14.03 MHz Pout 0.8 W
        > 2nd Harmonic -59 dB
        > 3rd Harmonic -
        >
        >
        > With Original Filter, 270 degree coverage
        > 7.03 MHz
        > 7.03 MHz Pout 3.3 W
        > 2nd Harmonic -33 dB
        > 3rd Harmonic -39 dB
        > 10.13 MHz
        > 10.13 MHz Pout 2.6 W
        > 2nd Harmonic -47 dB
        > 3rd Harmonic > 60 dB
        > 14.03 MHz
        > 14.03 MHz Pout 0.6 W
        > 2nd Harmonic > 60 dB
        > 3rd Harmonic > 60 dB
        >
        > Some notes here: Obviously winding the toroids as David Turnbull suggests, full, evenly spaced turns makes this filter perform better for 20 meters. However, the combination of lower 20 meter drive and the characteristics of this filter really do significantly degrade 20 meter performance in both cases.
        >
        > Finally I tested the new filter values on an externally connected version, again with full coverage and with 270 degree coverage on the toroids. Those results:
        >
        > With New Filter, full coverage
        > 7.03 MHz
        > 7.03 MHz Pout 4.5 W
        > 2nd Harmonic -32 dB
        > 3rd Harmonic -34 dB
        > 10.13 MHz
        > 10.13 MHz Pout 5.1 W
        > 2nd Harmonic -41 dB
        > 3rd Harmonic > 60 dB
        > 14.03 MHz
        > 14.03 MHz Pout 1.6 W
        > 2nd Harmonic > 60 dB
        > 3rd Harmonic > 60 dB
        >
        > With New Filter, 270 degreel coverage
        >
        > 7.03 MHz
        > 7.03 MHz Pout 6.1 W
        > 2nd Harmonic -30 dB
        > 3rd Harmonic -32 dB
        > 10.13 MHz
        > 10.13 MHz Pout 5.4 W
        > 2nd Harmonic -40 dB
        > 3rd Harmonic > 60 dB
        > 14.03 MHz
        > 14.03 MHz Pout 1.8 W
        > 2nd Harmonic > 60 dB
        > 3rd Harmonic > 60 dB
        >
        > Here I found significant improvement in the 20 meter output; 1.6-1.8 watts depending how you wound the toroids. The 30 meter harmonic performance just squeezed in under the wire and this is a small concern but I am back to thinking, overall, this is a good move.
        >
        > Hope this is helpful and I stress this is one test. Alan is working on something similar and it will be interesting to see if he concurs.
        >
        > Best regards,
        >
        > Warren Allgyer - W8TOD
        >
        > --- In softrock40@yahoogroups.com, "R. R. (Robby) Robson" wrote:
        > >
        > > The changes were made to the RF I/O stage (
        > > http://wb5rvz.org/ensemble_rxtx/05_rf) and involved changed values for C24,
        > > C25, C26, L2, and L3. The new values are shown in the published
        > > documentation (along with a note describing the change and referencing the
        > > thread of which you spoke).
        > >
        >


      • Chris Wilson
        ... 05/01/2013 10:48 I see your dilemma, I feel this is going to run and run. The availability of choice to newbies like myself is sometimes a cause of
        Message 3 of 25 , Jan 5, 2013
        • 0 Attachment
          >
          > Good Morning All,

          > For now all RXTX Ensemble kits shipped will include the capacitors
          > such that the previous or new LP filter for 40m/30m/20m may be built.

          > I very much appreciate the efforts of Warren, Alan and others to get things right!

          > 73,
          > Tony KB9YIG




          05/01/2013 10:48

          I see your dilemma, I feel this is going to run and run. The
          availability of choice to newbies like myself is sometimes a cause of
          indecision and very protracted discussion on reflectors ;) I hope some
          agreement and a recommended build format without the need for great
          knowledge is reached. But of course, those who wish to learn more can share
          in the experience of the ongoing development in this great series of kits.

          All the best for 2013, the Softrocks have been my most used amateur
          devices, despite me having far more costly gear. I can no longer
          operate without a panadptor display, twiddling a dial listening for
          new signals is simply not an option, even if the receiver is a tad
          better. And to be honest, the times my TS-590 resolves a signal the
          Softrock will not, on the same fairly decent antenna, is very rare.

          --
          Best Regards,
          Chris Wilson. 2E0ILY
        • richard.lawn@ymail.com
          I m frankly a bit frustrated with this discussion which I ve tried to follow. I ve not finished my kit yet and want to do it right. Could someone please post a
          Message 4 of 25 , Jan 5, 2013
          • 0 Attachment
            I'm frankly a bit frustrated with this discussion which I've tried to follow. I've not finished my kit yet and want to do it right. Could someone please post a simple statement about what changes should be made to the build. I've been directed to the WB5 website but its not clear to me where this errata can be found.
            Tnx
            Rick
            W2JAZ

            --- In softrock40@yahoogroups.com, Chris Wilson wrote:
            >
            >
            >
            > >
            > > Good Morning All,
            >
            > > For now all RXTX Ensemble kits shipped will include the capacitors
            > > such that the previous or new LP filter for 40m/30m/20m may be built.
            >
            > > I very much appreciate the efforts of Warren, Alan and others to get things right!
            >
            > > 73,
            > > Tony KB9YIG
            >
            >
            >
            >
            > 05/01/2013 10:48
            >
            > I see your dilemma, I feel this is going to run and run. The
            > availability of choice to newbies like myself is sometimes a cause of
            > indecision and very protracted discussion on reflectors ;) I hope some
            > agreement and a recommended build format without the need for great
            > knowledge is reached. But of course, those who wish to learn more can share
            > in the experience of the ongoing development in this great series of kits.
            >
            > All the best for 2013, the Softrocks have been my most used amateur
            > devices, despite me having far more costly gear. I can no longer
            > operate without a panadptor display, twiddling a dial listening for
            > new signals is simply not an option, even if the receiver is a tad
            > better. And to be honest, the times my TS-590 resolves a signal the
            > Softrock will not, on the same fairly decent antenna, is very rare.
            >
            > --
            > Best Regards,
            > Chris Wilson. 2E0ILY
            >
          • Alan
            ... Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate ... Rick, Yes, difficult to be sure when Robby has so much
            Message 5 of 25 , Jan 5, 2013
            • 0 Attachment
              ----- Original Message -----

              Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate


              >
              >
              > I'm frankly a bit frustrated with this discussion which I've tried to follow. I've not finished my kit yet and want to do it
              > right. Could someone please post a simple statement about what changes should be made to the build. I've been directed to the WB5
              > website but its not clear to me where this errata can be found.



              Rick,

              Yes, difficult to be sure when Robby has so much http://www.wb5rvz.org/ensemble_rxtx/index?projectId=14 and set the band option. The
              revision is on RF/IO.
              If you have got that far I'd say just remove two turns from L2 and L3 and it should be correct. But to do the complete change the
              capacitors need to be changed.

              It seems that some before have been frustrated so much that they gave up, so you are lucky!

              73 Alan G4ZFQ
            • allenbh57
              Hi Tony, Warren and Alan, Very interesting discussion on the filters for 20-30-40. Warren, I liked your thought that the Z of the filter may not be the same in
              Message 6 of 25 , Jan 5, 2013
              • 0 Attachment
                Hi Tony, Warren and Alan,

                Very interesting discussion on the filters for 20-30-40. Warren, I liked your thought that the Z of the filter may not be the same in your test setup and the board, therefore the differences. Just for fun and related information (isn't information being fun one reason we do this? Hi Hi), would a VNA (don't have one available) or other suitable piece of equipment measure the impedance of the filters as a black-box? For that matter, thinking off the top, could a directional coupler between the amp and the BPF be used to show what is happening here? Except that wold probably introduce another impedance into the mix.

                This takes me back to the discussion between Bill and Mike on Slodersmoke, trying to figure out the input Z of an amplifier!

                Tony - good decision.

                73,
                Bill AJ4TX

                --- In softrock40@yahoogroups.com, "Parks, Tony" wrote:
                >
                > Good Morning All,
                >
                > For now all RXTX Ensemble kits shipped will include the capacitors such
                > that the previous or new LP filter for 40m/30m/20m may be built.
                >
                > I very much appreciate the efforts of Warren, Alan and others to get things
                > right!
                >
                > 73,
                > Tony KB9YIG
                >
                > On Fri, Jan 4, 2013 at 11:55 PM, warrenallgyer wrote:
                >
                > > **
                > >
                > >
                > > I hope this is my final post on the subject. The production filter is now
                > > installed in my on-air Peaberry (identical RF section to the RXTX) with the
                > > following results:
                > >
                > > Power Supply is 13.8 VDC
                > >
                > > 7.03 Vrms Out 13.1
                > > Watts Out 3.4
                > > 2nd harmonic -30 dB
                > > IMD3 ref Carrier -28 dB
                > >
                > > 10.13 Vrms Out 11.3
                > > Watts Out 2.6
                > > 2nd harmonic -40 dB
                > > IMD3 ref Carrier -29 dB
                > >
                > > 14.03 Vrms Out 9.2
                > > Watts Out 1.7
                > > 2nd harmonic -58 dB
                > > IMD3 ref Carrier -25 dB
                > >
                > > I am going to call this one "finished" so I can try to get to installing
                > > David's new 96 KHz software before the weekend vanishes.
                > >
                > > Warren Allgyer - W8TOD
                > >
                > >
                > > --- In softrock40@yahoogroups.com, "warrenallgyer" wrote:
                > > >
                > > > All
                > > >
                > > > I have concluded a set of tests this morning that gives me more
                > > confidence in the 30 meter harmonic performance using the new filter
                > > values. I am seeing just 40-41 dB of second harmonic suppression using the
                > > new values and testing them two ways: one with the toroids wound full
                > > coverage and one with them wound with approximately a 90 degree gap.
                > > >
                > > > My tests indicate the greatly improved performance on 20 meters makes
                > > this worth the change.
                > > >
                > > > Now the details of the latest set of my tests for those who are
                > > interested.
                > > >
                > > > I removed the output filter completely and put a temporary connector
                > > driven by the secondary of T4. In this test I was interested to measure
                > > what is actually coming out of T4 using a stable resistive load. Here is
                > > what I found:
                > > >
                > > > T4 connected directly to a 50 ohm load
                > > > 7.03 MHz
                > > > 7.03 MHz Pout 3.6 W
                > > > 2nd Harmonic -32 dB
                > > > 3rd Harmonic -13 dB
                > > > 10.13 MHz
                > > > 10.13 MHz Pout 3.9 W
                > > > 2nd Harmonic -30 dB
                > > > 3rd Harmonic -18 dB
                > > > 14.03 MHz
                > > > 14.03 MHz Pout 1.4 W
                > > > 2nd Harmonic -28 dB
                > > > 3rd Harmonic -22 dB
                > > >
                > > > The significant fall-off in power output on 20 Meters needs further
                > > investigation. It is masked when the filter is in place and T4 is probed, I
                > > expect due to a high impedance presented by the filter at that point.
                > > Nevertheless there is enough power available if we can get it through a
                > > properly constructed filter.
                > > >
                > > > Next I connected this output to an external filter using the original
                > > values and tested it with full coverage turns on the toroid and 270 degree
                > > coverage. Results are below:
                > > >
                > > > With Original Filter, full coverage
                > > > 7.03 MHz
                > > > 7.03 MHz Pout 5.7 W (!)
                > > > 2nd Harmonic -33
                > > > 3rd Harmonic -6
                > > > 10.13 MHz
                > > > 10.13 MHz Pout 6.0 W
                > > > 2nd Harmonic -53 dB
                > > > 3rd Harmonic -
                > > > 14.03 MHz
                > > > 14.03 MHz Pout 0.8 W
                > > > 2nd Harmonic -59 dB
                > > > 3rd Harmonic -
                > > >
                > > >
                > > > With Original Filter, 270 degree coverage
                > > > 7.03 MHz
                > > > 7.03 MHz Pout 3.3 W
                > > > 2nd Harmonic -33 dB
                > > > 3rd Harmonic -39 dB
                > > > 10.13 MHz
                > > > 10.13 MHz Pout 2.6 W
                > > > 2nd Harmonic -47 dB
                > > > 3rd Harmonic > 60 dB
                > > > 14.03 MHz
                > > > 14.03 MHz Pout 0.6 W
                > > > 2nd Harmonic > 60 dB
                > > > 3rd Harmonic > 60 dB
                > > >
                > > > Some notes here: Obviously winding the toroids as David Turnbull
                > > suggests, full, evenly spaced turns makes this filter perform better for 20
                > > meters. However, the combination of lower 20 meter drive and the
                > > characteristics of this filter really do significantly degrade 20 meter
                > > performance in both cases.
                > > >
                > > > Finally I tested the new filter values on an externally connected
                > > version, again with full coverage and with 270 degree coverage on the
                > > toroids. Those results:
                > > >
                > > > With New Filter, full coverage
                > > > 7.03 MHz
                > > > 7.03 MHz Pout 4.5 W
                > > > 2nd Harmonic -32 dB
                > > > 3rd Harmonic -34 dB
                > > > 10.13 MHz
                > > > 10.13 MHz Pout 5.1 W
                > > > 2nd Harmonic -41 dB
                > > > 3rd Harmonic > 60 dB
                > > > 14.03 MHz
                > > > 14.03 MHz Pout 1.6 W
                > > > 2nd Harmonic > 60 dB
                > > > 3rd Harmonic > 60 dB
                > > >
                > > > With New Filter, 270 degreel coverage
                > > >
                > > > 7.03 MHz
                > > > 7.03 MHz Pout 6.1 W
                > > > 2nd Harmonic -30 dB
                > > > 3rd Harmonic -32 dB
                > > > 10.13 MHz
                > > > 10.13 MHz Pout 5.4 W
                > > > 2nd Harmonic -40 dB
                > > > 3rd Harmonic > 60 dB
                > > > 14.03 MHz
                > > > 14.03 MHz Pout 1.8 W
                > > > 2nd Harmonic > 60 dB
                > > > 3rd Harmonic > 60 dB
                > > >
                > > > Here I found significant improvement in the 20 meter output; 1.6-1.8
                > > watts depending how you wound the toroids. The 30 meter harmonic
                > > performance just squeezed in under the wire and this is a small concern but
                > > I am back to thinking, overall, this is a good move.
                > > >
                > > > Hope this is helpful and I stress this is one test. Alan is working on
                > > something similar and it will be interesting to see if he concurs.
                > > >
                > > > Best regards,
                > > >
                > > > Warren Allgyer - W8TOD
                > > >
                > > > --- In softrock40@yahoogroups.com, "R. R. (Robby) Robson" wrote:
                > > > >
                > > > > The changes were made to the RF I/O stage (
                > > > > http://wb5rvz.org/ensemble_rxtx/05_rf) and involved changed values
                > > for C24,
                > > > > C25, C26, L2, and L3. The new values are shown in the published
                > > > > documentation (along with a note describing the change and referencing
                > > the
                > > > > thread of which you spoke).
                > > > >
                > > >
                > >
                > >
                > >
                >
              • Orin
                ... liked your thought that the Z of the filter may not be the same in your test setup and the board, therefore the differences. Just for fun and related
                Message 7 of 25 , Jan 5, 2013
                • 0 Attachment
                  --- In softrock40@yahoogroups.com, "allenbh57" wrote:
                  >
                  > Hi Tony, Warren and Alan,
                  >
                  > Very interesting discussion on the filters for 20-30-40. Warren, I liked your thought that the Z of the filter may not be the same in your test setup and the board, therefore the differences. Just for fun and related information (isn't information being fun one reason we do this? Hi Hi), would a VNA (don't have one available) or other suitable piece of equipment measure the impedance of the filters as a black-box?


                  Yes, a VNA works fine for measuring the Z of the filter.  You get something like this:

                  VNWA scan of 15/12/10m filter 

                  though that's for the 15-12-10m build and I didn't have the actual impedance displayed.   An antenna analyzer such as the MFJ 259B works too if you terminate the filter output in 50 ohms; it will display impedance, though the 259B doesn't give the sign of the reactive component. 


                  > For that matter, thinking off the top, could a directional coupler between the amp and the BPF be used to show what is happening here? Except that wold probably introduce another impedance into the mix.
                  >

                  About to wander slightly off topic here on what they don't tell you about filters; at least what they didn't tell me!

                  That could work depending on what you have to look at the output from the coupler.  What you should see with these filters is no (ideally) reverse power in the passband and for it to increase rapidly after the cutoff frequency.  Yes, I don't think the amateur radio exams I've taken mention it, but this kind of filter doesn't do anything with input at frequencies higher than the cutoff frequency, they merely bounce it back to the source.

                  Even though I might have said filter attenuation in the past, using the term 'attenuate' can be misleading.  Unlike an attenuator where power that isn't passed through is turned into heat, these filters bounce the harmonics back to the PA!  (Rather unfortunate when you select the wrong band/low pass filter on you amplifier.  Say you leave you 40m LPF in and transmit on 20m, almost all your power bounces back.  Hopefully the PA survives this abuse!)  In the passband, there will be some slight attenuation due to losses in the coils and capacitors, but it's usually only a few tenths of a dB, depending on the Q of the coils and capacitors.

                  Orin.

                • warrenallgyer
                  The first clue I had on the varying impedance issue was when I built the external filter breadboard to test the new values in the first place. I terminated on
                  Message 8 of 25 , Jan 5, 2013
                  • 0 Attachment
                    The first clue I had on the varying impedance issue was when I built the external filter breadboard to test the new values in the first place. I terminated on side of the test filter and then fed the MFJ directly into the other side, measuring the voltage across the termination on the scope. It was a sort of manual sweep generator and I could very quickly see when the output level started to drop off as I swept up through 11 MHz or so. In the process I noted that the VSWR indication on the old MFJ was very low, nearly 1:1 at 7 MHz and it started to rise rapidly in correspondence with the output level falling off at higher frequencies.

                    Similarly I have noted in the past when tuning up my antenna tuner, using a scope on the input to the tuner as a power indicator, that the indicated voltage at the scope rises rapidly as the tuner VSWR goes up.

                    I like Orin's characterization of "attenuation" in these filters. The power in those harmonics and the portion of the fundamental that gets "scraped off" by the filter is substantial and it is not heating the filter. It is being reflected back and dissipated in the finals and one indication of this appears to be a rise in the measured voltage at the filter input.

                    I really do apologize for the tedium this subject has brought to some. Allen has it exactly right from my point of view in that this is a learning opportunity and my excitement at that gets a little overactive at times. I have learned a ton from what you guys have written in these exchanges and I hope those I have bored to tears can just make allowances and skip over the topic.

                    Thanks again for a great discussion!

                    Warren Allgyer - W8TOD

                    --- In softrock40@yahoogroups.com, "Orin" wrote:
                    >
                    >
                    > --- In softrock40@yahoogroups.com, "allenbh57" wrote:
                    > >
                    > > Hi Tony, Warren and Alan,
                    > >
                    > > Very interesting discussion on the filters for 20-30-40. Warren, I
                    > liked your thought that the Z of the filter may not be the same in your
                    > test setup and the board, therefore the differences. Just for fun and
                    > related information (isn't information being fun one reason we do this?
                    > Hi Hi), would a VNA (don't have one available) or other suitable piece
                    > of equipment measure the impedance of the filters as a black-box?
                    >
                    >
                    > Yes, a VNA works fine for measuring the Z of the filter. You get
                    > something like this:
                    >
                    > VNWA scan of 15/12/10m filter
                    >
                    JzHAUEj7Cg?feat=directlink>
                    >
                    > though that's for the 15-12-10m build and I didn't have the actual
                    > impedance displayed. An antenna analyzer such as the MFJ 259B works
                    > too if you terminate the filter output in 50 ohms; it will display
                    > impedance, though the 259B doesn't give the sign of the reactive
                    > component.
                    >
                    >
                    > > For that matter, thinking off the top, could a directional coupler
                    > between the amp and the BPF be used to show what is happening here?
                    > Except that wold probably introduce another impedance into the mix.
                    > >
                    >
                    > About to wander slightly off topic here on what they don't tell you
                    > about filters; at least what they didn't tell me!
                    >
                    > That could work depending on what you have to look at the output from
                    > the coupler. What you should see with these filters is no (ideally)
                    > reverse power in the passband and for it to increase rapidly after the
                    > cutoff frequency. Yes, I don't think the amateur radio exams I've taken
                    > mention it, but this kind of filter doesn't do anything with input at
                    > frequencies higher than the cutoff frequency, they merely bounce it back
                    > to the source.
                    >
                    > Even though I might have said filter attenuation in the past, using the
                    > term 'attenuate' can be misleading. Unlike an attenuator where power
                    > that isn't passed through is turned into heat, these filters bounce the
                    > harmonics back to the PA! (Rather unfortunate when you select the wrong
                    > band/low pass filter on you amplifier. Say you leave you 40m LPF in and
                    > transmit on 20m, almost all your power bounces back. Hopefully the PA
                    > survives this abuse!) In the passband, there will be some slight
                    > attenuation due to losses in the coils and capacitors, but it's usually
                    > only a few tenths of a dB, depending on the Q of the coils and
                    > capacitors.
                    >
                    > Orin.
                    >
                  • Anthony Casorso
                    The AADE filter designer program that I use allows you to plot the filter input and output impedance vs frequency. I suspect that ELSIE does too. If you take a
                    Message 9 of 25 , Jan 5, 2013
                    • 0 Attachment
                      The AADE filter designer program that I use allows you to plot the filter input and output impedance vs frequency. I suspect that ELSIE does too. If you take a typical filter design, say a Chebyshev 5th order bandpass, and look at the impedance variation with frequency you will see that it is quite variable. A 50 ohm filter is almost never exactly 50 ohms and you can expect excursions from 10 to 150 ohms, depending on the filter specifics. This is just the nature of the beast. 

                      Tony





                    • Dave Miller
                      Awesome discussion. As a newbie I am entitled to a couple dumb questions. ;-) I have a Peaberry 40,30,20. I have been struggling with testing out the TX. From
                      Message 10 of 25 , Jan 9, 2013
                      • 0 Attachment
                        Awesome discussion.
                        As a newbie I am entitled to a couple dumb questions. ;-)
                        I have a Peaberry 40,30,20.
                        I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and 3rd harmonics on 40 m were hardly attenuated. Building and testing filter filter got waylaid by getting software working on TX. :-(
                        My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                        I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW. Finally resorted to SSB and the whistle test. If I could only whistle.

                        Briefly tried Rocky but seemed not to have much joy.

                        I finally got PowerSDR-IQ running on RX. Have not got to TX yet but have not tried.

                        So the question is. Can someone outline in some detail what software and method used to test TX on Ensemble RXTX (will work fine on Peaberry :-) ) ?
                        I have the gear (dummy load, sampling Tap and Spec Analyzer) and knowledge to capture the data just too dumb to coax a continuous carrier out of the beast to capture.

                        I have all the parts for the External Filter and will build it soon. I will sweep it when complete. I will post results. I got a VNWA3 for Christmas so I can post pretty pictures. ;-).

                        Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                        How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing requirements.

                        Thanks all. Just amazed at the on air performance of the receiver. Like other have said after SDR it's hard to go back to no panadpter.
                        Dave
                        VE7PKE



                        --- In softrock40@yahoogroups.com, "warrenallgyer" wrote:
                        >
                        > The first clue I had on the varying impedance issue was when I built the external filter breadboard to test the new values in the first place. I terminated on side of the test filter and then fed the MFJ directly into the other side, measuring the voltage across the termination on the scope. It was a sort of manual sweep generator and I could very quickly see when the output level started to drop off as I swept up through 11 MHz or so. In the process I noted that the VSWR indication on the old MFJ was very low, nearly 1:1 at 7 MHz and it started to rise rapidly in correspondence with the output level falling off at higher frequencies.
                        >
                        > Similarly I have noted in the past when tuning up my antenna tuner, using a scope on the input to the tuner as a power indicator, that the indicated voltage at the scope rises rapidly as the tuner VSWR goes up.
                        >
                        > I like Orin's characterization of "attenuation" in these filters. The power in those harmonics and the portion of the fundamental that gets "scraped off" by the filter is substantial and it is not heating the filter. It is being reflected back and dissipated in the finals and one indication of this appears to be a rise in the measured voltage at the filter input.
                        >
                        > I really do apologize for the tedium this subject has brought to some. Allen has it exactly right from my point of view in that this is a learning opportunity and my excitement at that gets a little overactive at times. I have learned a ton from what you guys have written in these exchanges and I hope those I have bored to tears can just make allowances and skip over the topic.
                        >
                        > Thanks again for a great discussion!
                        >
                        > Warren Allgyer - W8TOD
                        >
                        > --- In softrock40@yahoogroups.com, "Orin" wrote:
                        > >
                        > >
                        > > --- In softrock40@yahoogroups.com, "allenbh57" wrote:
                        > > >
                        > > > Hi Tony, Warren and Alan,
                        > > >
                        > > > Very interesting discussion on the filters for 20-30-40. Warren, I
                        > > liked your thought that the Z of the filter may not be the same in your
                        > > test setup and the board, therefore the differences. Just for fun and
                        > > related information (isn't information being fun one reason we do this?
                        > > Hi Hi), would a VNA (don't have one available) or other suitable piece
                        > > of equipment measure the impedance of the filters as a black-box?
                        > >
                        > >
                        > > Yes, a VNA works fine for measuring the Z of the filter. You get
                        > > something like this:
                        > >
                        > > VNWA scan of 15/12/10m filter
                        > >
                        > JzHAUEj7Cg?feat=directlink>
                        > >
                        > > though that's for the 15-12-10m build and I didn't have the actual
                        > > impedance displayed. An antenna analyzer such as the MFJ 259B works
                        > > too if you terminate the filter output in 50 ohms; it will display
                        > > impedance, though the 259B doesn't give the sign of the reactive
                        > > component.
                        > >
                        > >
                        > > > For that matter, thinking off the top, could a directional coupler
                        > > between the amp and the BPF be used to show what is happening here?
                        > > Except that wold probably introduce another impedance into the mix.
                        > > >
                        > >
                        > > About to wander slightly off topic here on what they don't tell you
                        > > about filters; at least what they didn't tell me!
                        > >
                        > > That could work depending on what you have to look at the output from
                        > > the coupler. What you should see with these filters is no (ideally)
                        > > reverse power in the passband and for it to increase rapidly after the
                        > > cutoff frequency. Yes, I don't think the amateur radio exams I've taken
                        > > mention it, but this kind of filter doesn't do anything with input at
                        > > frequencies higher than the cutoff frequency, they merely bounce it back
                        > > to the source.
                        > >
                        > > Even though I might have said filter attenuation in the past, using the
                        > > term 'attenuate' can be misleading. Unlike an attenuator where power
                        > > that isn't passed through is turned into heat, these filters bounce the
                        > > harmonics back to the PA! (Rather unfortunate when you select the wrong
                        > > band/low pass filter on you amplifier. Say you leave you 40m LPF in and
                        > > transmit on 20m, almost all your power bounces back. Hopefully the PA
                        > > survives this abuse!) In the passband, there will be some slight
                        > > attenuation due to losses in the coils and capacitors, but it's usually
                        > > only a few tenths of a dB, depending on the Q of the coils and
                        > > capacitors.
                        > >
                        > > Orin.
                        > >
                        >
                      • Alan
                        ... Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate ... Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just
                        Message 11 of 25 , Jan 9, 2013
                        • 0 Attachment
                          ----- Original Message -----
                          Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate


                          > I have a Peaberry 40,30,20.
                          > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and
                          > 3rd harmonics on 40 m were hardly attenuated.

                          Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just have not built it.
                          What values did you use in the integral LPF?

                          Building and testing filter filter got waylaid by getting software working on TX. :-(
                          > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                          > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW.

                          You will get a constant carrier on AM or FM.

                          >I got a VNWA3 for Christmas so I can post pretty pictures. ;-).

                          Make sure we can understand what they tell us:) Gerhard posted some and I'm puzzled!

                          >
                          > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                          > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the
                          > bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing
                          > requirements.

                          Probably much the same. Just look at the CPU usage, if it goes over 50% it could be borderline.
                          Probably the latest PDSR-IQ is the best bet if you go that way.

                          73 Alan G4ZFQ
                        • Dave Miller
                          Alan, thanks for reply. See below. First off I see that as I failed to refresh my cache a significant amount of discussion has occurred on the topic of what
                          Message 12 of 25 , Jan 9, 2013
                          • 0 Attachment
                            Alan, thanks for reply. See below.
                            First off I see that as I failed to refresh my cache a significant amount of discussion has occurred on the topic of what component vales to use for LPF. I was going to use what are in Dave T's build instructions. I printed them off last week. Given all the debate where is the most up to date and definitive turn and cap values?



                            --- In softrock40@yahoogroups.com, "Alan" wrote:
                            >
                            >
                            > ----- Original Message -----
                            > Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate
                            >
                            >
                            > > I have a Peaberry 40,30,20.
                            > > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and
                            > > 3rd harmonics on 40 m were hardly attenuated.
                            >
                            > Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just have not built it.
                            > What values did you use in the integral LPF?

                            Whatever were in the AE9RB online directions? What should I use?
                            >
                            > Building and testing filter filter got waylaid by getting software working on TX. :-(
                            > > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                            > > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW.
                            >
                            > You will get a constant carrier on AM or FM.
                            I shall try that. Still want to get CW working. Have been talking to LC via email and understand HDSDR state on this.
                            >
                            > >I got a VNWA3 for Christmas so I can post pretty pictures. ;-).
                            >
                            > Make sure we can understand what they tell us:) Gerhard posted some and I'm puzzled!
                            I shall. You can display so many traces . ;-) Its very confusing at first. starting to getba handle on it. I will try to post a couple of versions. The easy to understand and the full meal deal. For this just a simple thru measurement showing attenuation verses frequency is what counts and is easy to do on an external LPF. Orrin points out doing it properly on an internal gets more difficult.
                            >
                            > >
                            > > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                            > > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the
                            > > bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing
                            > > requirements.
                            >
                            > Probably much the same. Just look at the CPU usage, if it goes over 50% it could be borderline.
                            > Probably the latest PDSR-IQ is the best bet if you go that way.
                            Ok Alan, thanks. Lots to learn. I appreciate all the assistance the group provides.
                            I will have to keep up to date here.

                            Thanks
                            Dave
                            VE7PKE

                            >
                            > 73 Alan G4ZFQ
                            >
                          • Dave Miller
                            Alan, Just has a look at the wonderful pictures. The crib notes are just look at the S21 traces. S21 is the amount of the signal passing thru the filter. So
                            Message 13 of 25 , Jan 9, 2013
                            • 0 Attachment
                              Alan,
                              Just has a look at the wonderful pictures.
                              The crib notes are just look at the S21 traces.
                              S21 is the amount of the signal passing thru the filter. So the hump should be where we want the signal to pass.
                              The memory traces are other captures. Not clear on what they are.
                              S11 is the reflection as measured from the input.

                              I hope this helps.

                              Many thanks for the OP for posting those pictures.

                              Thanks
                              Dave
                              VE7PKE

                              The memory traces are previous traces and without a detailed explanation rather difficult to follow

                              --- In softrock40@yahoogroups.com, "Alan" wrote:
                              >
                              >
                              > ----- Original Message -----
                              > Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate
                              >
                              >
                              > > I have a Peaberry 40,30,20.
                              > > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and
                              > > 3rd harmonics on 40 m were hardly attenuated.
                              >
                              > Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just have not built it.
                              > What values did you use in the integral LPF?
                              >
                              > Building and testing filter filter got waylaid by getting software working on TX. :-(
                              > > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                              > > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW.
                              >
                              > You will get a constant carrier on AM or FM.
                              >
                              > >I got a VNWA3 for Christmas so I can post pretty pictures. ;-).
                              >
                              > Make sure we can understand what they tell us:) Gerhard posted some and I'm puzzled!
                              >
                              > >
                              > > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                              > > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the
                              > > bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing
                              > > requirements.
                              >
                              > Probably much the same. Just look at the CPU usage, if it goes over 50% it could be borderline.
                              > Probably the latest PDSR-IQ is the best bet if you go that way.
                              >
                              > 73 Alan G4ZFQ
                              >
                            • Alan
                              ... Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate ... Tony has altered the design slightly. It was found the HF cutoff was too low. See Robby s
                              Message 14 of 25 , Jan 9, 2013
                              • 0 Attachment
                                ----- Original Message -----
                                Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate


                                > Given all the debate where is the most up to date and definitive turn and cap values?
                                >

                                Tony has altered the design slightly. It was found the HF cutoff was too low. See Robby's pages
                                http://www.wb5rvz.org/ensemble_rxtx/ and set the band option. The revision is on RF/IO.

                                If already built I reckon two turns maximum removed from L2 and L3 should make the LPF conform to the original design if problems
                                are seen with the 20m output.

                                73 Alan G4ZFQ
                              • John Greusel
                                Two turn is exactly what I did and then I measured the inductance and it was spot on to the original specs. Whistling yields almost 2.5 watts on 20 meters so
                                Message 15 of 25 , Jan 9, 2013
                                • 0 Attachment
                                  Two turn is exactly what I did and then I measured the inductance and it was spot on to the original specs.
                                  Whistling yields almost 2.5 watts on 20 meters so I've had to throttle back the output now.

                                  John
                                  KC9OJV
                                   



                                  From: Alan <alan4alan@...>
                                  To: softrock40@yahoogroups.com
                                  Sent: Wednesday, January 9, 2013 10:13 AM
                                  Subject: Re: [softrock40] Re: Ensemble RXTX for 40/30/20 debate

                                   

                                  ----- Original Message -----
                                  Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate

                                  > Given all the debate where is the most up to date and definitive turn and cap values?
                                  >

                                  Tony has altered the design slightly. It was found the HF cutoff was too low. See Robby's pages
                                  http://www.wb5rvz.org/ensemble_rxtx/ and set the band option. The revision is on RF/IO.

                                  If already built I reckon two turns maximum removed from L2 and L3 should make the LPF conform to the original design if problems
                                  are seen with the 20m output.

                                  73 Alan G4ZFQ



                                • Orin
                                  If you are talking about my VNWA scan that I posted a few days back in this thread, the memory traces are the theoretical filter responses as calculated by
                                  Message 16 of 25 , Jan 9, 2013
                                  • 0 Attachment
                                    If you are talking about my VNWA scan that I posted a few days back in this thread, the memory traces are the theoretical filter responses as calculated by "Elsie".

                                    Orin.

                                    --- In softrock40@yahoogroups.com, "Dave Miller" wrote:
                                    >
                                    > Alan,
                                    > Just has a look at the wonderful pictures.
                                    > The crib notes are just look at the S21 traces.
                                    > S21 is the amount of the signal passing thru the filter. So the hump should be where we want the signal to pass.
                                    > The memory traces are other captures. Not clear on what they are.
                                    > S11 is the reflection as measured from the input.
                                    >
                                    > I hope this helps.
                                    >
                                    > Many thanks for the OP for posting those pictures.
                                    >
                                    > Thanks
                                    > Dave
                                    > VE7PKE
                                    >
                                    > The memory traces are previous traces and without a detailed explanation rather difficult to follow
                                    >
                                    > --- In softrock40@yahoogroups.com, "Alan" wrote:
                                    > >
                                    > >
                                    > > ----- Original Message -----
                                    > > Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate
                                    > >
                                    > >
                                    > > > I have a Peaberry 40,30,20.
                                    > > > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and
                                    > > > 3rd harmonics on 40 m were hardly attenuated.
                                    > >
                                    > > Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just have not built it.
                                    > > What values did you use in the integral LPF?
                                    > >
                                    > > Building and testing filter filter got waylaid by getting software working on TX. :-(
                                    > > > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                                    > > > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW.
                                    > >
                                    > > You will get a constant carrier on AM or FM.
                                    > >
                                    > > >I got a VNWA3 for Christmas so I can post pretty pictures. ;-).
                                    > >
                                    > > Make sure we can understand what they tell us:) Gerhard posted some and I'm puzzled!
                                    > >
                                    > > >
                                    > > > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                                    > > > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the
                                    > > > bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing
                                    > > > requirements.
                                    > >
                                    > > Probably much the same. Just look at the CPU usage, if it goes over 50% it could be borderline.
                                    > > Probably the latest PDSR-IQ is the best bet if you go that way.
                                    > >
                                    > > 73 Alan G4ZFQ
                                    > >
                                    >
                                  • Dave Miller
                                    Orin, I was actually referring to Gerhards new pictures. I got around to measuring the inductors for my external 40 m low pass on my 40,30,20 unit. Once I
                                    Message 17 of 25 , Jan 9, 2013
                                    • 0 Attachment
                                      Orin,
                                      I was actually referring to Gerhards new pictures.
                                      I got around to measuring the inductors for my external 40 m low pass on my 40,30,20 unit. Once I removed the magic 2 turns they are now 1.3uH @10MHz according to my VNWA. Caps measured as marked so I figure I had things calibrated properly.

                                      Slowly figuring this all out. I will have to pull the two inductors in on board Bandpass and measure them.

                                      Thanks
                                      Dave

                                      --- In softrock40@yahoogroups.com, "Orin" wrote:
                                      >
                                      > If you are talking about my VNWA scan that I posted a few days back in this thread, the memory traces are the theoretical filter responses as calculated by "Elsie".
                                      >
                                      > Orin.
                                      >
                                      > --- In softrock40@yahoogroups.com, "Dave Miller" wrote:
                                      > >
                                      > > Alan,
                                      > > Just has a look at the wonderful pictures.
                                      > > The crib notes are just look at the S21 traces.
                                      > > S21 is the amount of the signal passing thru the filter. So the hump should be where we want the signal to pass.
                                      > > The memory traces are other captures. Not clear on what they are.
                                      > > S11 is the reflection as measured from the input.
                                      > >
                                      > > I hope this helps.
                                      > >
                                      > > Many thanks for the OP for posting those pictures.
                                      > >
                                      > > Thanks
                                      > > Dave
                                      > > VE7PKE
                                      > >
                                      > > The memory traces are previous traces and without a detailed explanation rather difficult to follow
                                      > >
                                      > > --- In softrock40@yahoogroups.com, "Alan" wrote:
                                      > > >
                                      > > >
                                      > > > ----- Original Message -----
                                      > > > Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate
                                      > > >
                                      > > >
                                      > > > > I have a Peaberry 40,30,20.
                                      > > > > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and
                                      > > > > 3rd harmonics on 40 m were hardly attenuated.
                                      > > >
                                      > > > Assuming the Peaberry is like the Softrock it will need an extra LPF, Oh I see you just have not built it.
                                      > > > What values did you use in the integral LPF?
                                      > > >
                                      > > > Building and testing filter filter got waylaid by getting software working on TX. :-(
                                      > > > > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                                      > > > > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW.
                                      > > >
                                      > > > You will get a constant carrier on AM or FM.
                                      > > >
                                      > > > >I got a VNWA3 for Christmas so I can post pretty pictures. ;-).
                                      > > >
                                      > > > Make sure we can understand what they tell us:) Gerhard posted some and I'm puzzled!
                                      > > >
                                      > > > >
                                      > > > > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                                      > > > > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the
                                      > > > > bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing
                                      > > > > requirements.
                                      > > >
                                      > > > Probably much the same. Just look at the CPU usage, if it goes over 50% it could be borderline.
                                      > > > Probably the latest PDSR-IQ is the best bet if you go that way.
                                      > > >
                                      > > > 73 Alan G4ZFQ
                                      > > >
                                      > >
                                      >
                                    • joelweder
                                      I used to practice trying to whistle different tones when I worked repairing Motorola rigs years ago. But if you can t whistle, PowerSDR has a nice tone
                                      Message 18 of 25 , Mar 3 9:31 AM
                                      • 0 Attachment
                                        I used to practice trying to whistle different tones when I worked repairing Motorola rigs years ago. But if you can't whistle, PowerSDR has a nice tone generator built right in!

                                        There may be some easier way to do this, but I used the Two Tone test feature in Setup. I'm using PowerSDR-IQ 2.4.4. Click on Setup, then go to the Tests tab. By default the tones are set to 700 and 1900 Hz. I set them both to the same value eg 1000 Hz.

                                        The MOX button will put the radio into transmit but minimal power is transmitted when you're on a Sideband mode, so that doesn't work unless you can figure out a way to whistle into your mic! (or get an audio tone generator)

                                        Joel VE6EI

                                        --- In softrock40@yahoogroups.com, "Dave Miller" <ve7pke@...> wrote:
                                        >
                                        > Awesome discussion.
                                        > As a newbie I am entitled to a couple dumb questions. ;-)
                                        > I have a Peaberry 40,30,20.
                                        > I have been struggling with testing out the TX. From what I have seen the filters do deserve a good look. In my testing 2nd and 3rd harmonics on 40 m were hardly attenuated. Building and testing filter filter got waylaid by getting software working on TX. :-(
                                        > My biggest problem is getting the thing to transmit ia fashion suitable for testing.
                                        > I was using HDSDR but have concluded its not really ready for transmit. I was trying to test on CW. Finally resorted to SSB and the whistle test. If I could only whistle.
                                        >
                                        > Briefly tried Rocky but seemed not to have much joy.
                                        >
                                        > I finally got PowerSDR-IQ running on RX. Have not got to TX yet but have not tried.
                                        >
                                        > So the question is. Can someone outline in some detail what software and method used to test TX on Ensemble RXTX (will work fine on Peaberry :-) ) ?
                                        > I have the gear (dummy load, sampling Tap and Spec Analyzer) and knowledge to capture the data just too dumb to coax a continuous carrier out of the beast to capture.
                                        >
                                        > I have all the parts for the External Filter and will build it soon. I will sweep it when complete. I will post results. I got a VNWA3 for Christmas so I can post pretty pictures. ;-).
                                        >
                                        > Last question. Yes I know I may be exceeding number of allowable questions for the day. ;-)
                                        > How much computing horsepower is required to run PowerSDR? I am using the version from the links on SdRBuzz? For testing on the bench I would like to use my old EEPC netbook. Will it work? HDSDR seemed to work on it. Ate they similar in computing requirements.
                                        >
                                        > Thanks all. Just amazed at the on air performance of the receiver. Like other have said after SDR it's hard to go back to no panadpter.
                                        > Dave
                                        > VE7PKE
                                        >
                                        >
                                        >
                                        > --- In softrock40@yahoogroups.com, "warrenallgyer" wrote:
                                        > >
                                        > > The first clue I had on the varying impedance issue was when I built the external filter breadboard to test the new values in the first place. I terminated on side of the test filter and then fed the MFJ directly into the other side, measuring the voltage across the termination on the scope. It was a sort of manual sweep generator and I could very quickly see when the output level started to drop off as I swept up through 11 MHz or so. In the process I noted that the VSWR indication on the old MFJ was very low, nearly 1:1 at 7 MHz and it started to rise rapidly in correspondence with the output level falling off at higher frequencies.
                                        > >
                                        > > Similarly I have noted in the past when tuning up my antenna tuner, using a scope on the input to the tuner as a power indicator, that the indicated voltage at the scope rises rapidly as the tuner VSWR goes up.
                                        > >
                                        > > I like Orin's characterization of "attenuation" in these filters. The power in those harmonics and the portion of the fundamental that gets "scraped off" by the filter is substantial and it is not heating the filter. It is being reflected back and dissipated in the finals and one indication of this appears to be a rise in the measured voltage at the filter input.
                                        > >
                                        > > I really do apologize for the tedium this subject has brought to some. Allen has it exactly right from my point of view in that this is a learning opportunity and my excitement at that gets a little overactive at times. I have learned a ton from what you guys have written in these exchanges and I hope those I have bored to tears can just make allowances and skip over the topic.
                                        > >
                                        > > Thanks again for a great discussion!
                                        > >
                                        > > Warren Allgyer - W8TOD
                                        > >
                                        > > --- In softrock40@yahoogroups.com, "Orin" wrote:
                                        > > >
                                        > > >
                                        > > > --- In softrock40@yahoogroups.com, "allenbh57" wrote:
                                        > > > >
                                        > > > > Hi Tony, Warren and Alan,
                                        > > > >
                                        > > > > Very interesting discussion on the filters for 20-30-40. Warren, I
                                        > > > liked your thought that the Z of the filter may not be the same in your
                                        > > > test setup and the board, therefore the differences. Just for fun and
                                        > > > related information (isn't information being fun one reason we do this?
                                        > > > Hi Hi), would a VNA (don't have one available) or other suitable piece
                                        > > > of equipment measure the impedance of the filters as a black-box?
                                        > > >
                                        > > >
                                        > > > Yes, a VNA works fine for measuring the Z of the filter. You get
                                        > > > something like this:
                                        > > >
                                        > > > VNWA scan of 15/12/10m filter
                                        > > >
                                        > > JzHAUEj7Cg?feat=directlink>
                                        > > >
                                        > > > though that's for the 15-12-10m build and I didn't have the actual
                                        > > > impedance displayed. An antenna analyzer such as the MFJ 259B works
                                        > > > too if you terminate the filter output in 50 ohms; it will display
                                        > > > impedance, though the 259B doesn't give the sign of the reactive
                                        > > > component.
                                        > > >
                                        > > >
                                        > > > > For that matter, thinking off the top, could a directional coupler
                                        > > > between the amp and the BPF be used to show what is happening here?
                                        > > > Except that wold probably introduce another impedance into the mix.
                                        > > > >
                                        > > >
                                        > > > About to wander slightly off topic here on what they don't tell you
                                        > > > about filters; at least what they didn't tell me!
                                        > > >
                                        > > > That could work depending on what you have to look at the output from
                                        > > > the coupler. What you should see with these filters is no (ideally)
                                        > > > reverse power in the passband and for it to increase rapidly after the
                                        > > > cutoff frequency. Yes, I don't think the amateur radio exams I've taken
                                        > > > mention it, but this kind of filter doesn't do anything with input at
                                        > > > frequencies higher than the cutoff frequency, they merely bounce it back
                                        > > > to the source.
                                        > > >
                                        > > > Even though I might have said filter attenuation in the past, using the
                                        > > > term 'attenuate' can be misleading. Unlike an attenuator where power
                                        > > > that isn't passed through is turned into heat, these filters bounce the
                                        > > > harmonics back to the PA! (Rather unfortunate when you select the wrong
                                        > > > band/low pass filter on you amplifier. Say you leave you 40m LPF in and
                                        > > > transmit on 20m, almost all your power bounces back. Hopefully the PA
                                        > > > survives this abuse!) In the passband, there will be some slight
                                        > > > attenuation due to losses in the coils and capacitors, but it's usually
                                        > > > only a few tenths of a dB, depending on the Q of the coils and
                                        > > > capacitors.
                                        > > >
                                        > > > Orin.
                                        > > >
                                        > >
                                        >
                                      • Alan
                                        ... Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate ... If you want a carrier why not use AM or FM with the mic gain right down? Works in HDSDR or
                                        Message 19 of 25 , Mar 3 9:59 PM
                                        • 0 Attachment
                                          ----- Original Message -----
                                          Subject: [softrock40] Re: Ensemble RXTX for 40/30/20 debate


                                          >I used to practice trying to whistle different tones when I worked repairing Motorola rigs years ago. But if you can't whistle,
                                          >PowerSDR has a nice tone generator built right in!
                                          >
                                          > There may be some easier way to do this, but I used the Two Tone test feature in Setup. I'm using PowerSDR-IQ 2.4.4. Click on
                                          > Setup, then go to the Tests tab. By default the tones are set to 700 and 1900 Hz. I set them both to the same value eg 1000 Hz.
                                          >
                                          > The MOX button will put the radio into transmit but minimal power is transmitted when you're on a Sideband mode, so that doesn't
                                          > work unless you can figure out a way to whistle into your mic! (or get an audio tone generator)
                                          >

                                          If you want a carrier why not use AM or FM with the mic gain right down?
                                          Works in HDSDR or PSDR.

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