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

Re: [softrock40] Midldle Noise Hump(s)

Expand Messages
  • Alan
    ... From: Bill Dumke ... I have mainly used the PC power for my Softrock Lite RXs with no problems around the centre, apart from the spike which varies
    Message 1 of 12 , Aug 3, 2007
    • 0 Attachment
      ----- Original Message -----
      From: "Bill Dumke"
      >
      > 4. I also don't believe the AC power supply was contributing anything to
      > the middle hump(s) since the SoftRock 40 did not show any noise at all
      > at the center frequency.

      I have mainly used the PC power for my Softrock Lite RXs with no problems
      around the centre, apart from the spike which varies depending on the
      soundcard I use.

      I have seen your problem with the RX/TX 40/80 however. After ensuring there
      were no ground loops I rerouted the audio cables and the hump was gone. I
      think the RX/TX has a different grounding than the Lites.

      The soundcard input is very sensitive to interference. Try first with just
      the D44 input connected. Make sure the lead is good quality. Try different
      cables. Alter the position of the breakout box and Softrock relative to the
      monitor and PC. Try different settings of jumpers in the D44 box.

      You may eliminate the PSU by using a battery, I think a 9 volt PP3 would be
      OK for an RX only test, certainly it works on a Softrock Lite.
      To make certain check the 5 volt line.

      73 Alan G4ZFQ
    • Bill Dumke
      Alan, Thanks for the comments. I was about to try the battery, but now the hump is gone on the RXTX v6.2 40/30 using the D44 Interface with the 1:1 RF
      Message 2 of 12 , Aug 3, 2007
      • 0 Attachment
        Alan,

        Thanks for the comments.  I was about to try the battery, but now the hump is gone on the RXTX v6.2 40/30 using the D44 Interface with the 1:1 RF transformer, still using the AC power supply.  I am not doing anything else different, so it must be the routing of the cables.  All I did was disconnect the old SoftRock40 and replaced it with the RXTX v6.2 on top of the wood desk.  Well, I don't understand it, but at least there is hope now!

        I posted tonight's photos in the photo section under WA9PWR.  They end with the suffix 8/3 meaning today's date of 8/3/07.  You can see the hump totally gone with the 1:1 RF transformer in line. Signals tonight are strong on 40 meters as well.  The RXTX sure sounds good on SSB.  Used a Russian Shortwave broadcast station to calibrate the frequency of the receiver.  Wish I could transmit SSB with Rocky.  Hopefully, some day I will be able to do that.

        Anyway, with this good news, I am going on to calibrating the receiver IQ in Rocky and then going on to testing the transmitter.  (I use a DDS-60 (amplifier removed from the DDS-60 circuit) with a step attenuator in sweep mode to do the Rocky IQ calibration.)  I will have to do it now for both the CW and SSB crystals, independently, somehow.

        BTW, for the 1:1 RF transformer I am using about 24 turns of #26 AWG Beldsol wire wound bifilar on what I believe is a T50-6 yellow core toroid.  How good it will be in transmit, I don't know.  Also don't know how much loss or bandwidth I have either.  Just sort of guessed at it.   

        Bill wa9pwr

        Alan wrote:
        ----- Original Message ----- 
        From: "Bill Dumke"
          
        4. I also don't believe the AC power supply was contributing anything to
        the middle hump(s) since the SoftRock 40 did not show any noise at all
        at the center frequency.
            
        I have mainly used the PC power for my Softrock Lite RXs with no problems 
        around the centre, apart from the spike which varies depending on the 
        soundcard I use.
        
        I have seen your problem with the RX/TX 40/80 however. After ensuring there 
        were no ground loops I rerouted the audio cables and the hump was gone. I 
        think the RX/TX has a different grounding  than the Lites.
        
        The soundcard input is very sensitive to interference. Try first with just 
        the D44 input connected. Make sure the lead is good quality. Try different 
        cables. Alter the position of the breakout box and Softrock relative to the 
        monitor and PC. Try different settings of jumpers in the D44 box.
        
        You may eliminate the PSU by using a battery, I think a 9 volt PP3 would be 
        OK for an RX only test, certainly it works on a Softrock Lite.
        To make certain check the 5 volt line.
        
        73 Alan G4ZFQ 
        
        
        
         
        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/
         
        
        
        
          
      • Chuck Carpenter
        Had double humps here at first too. They went away when I changed the length of the cable from the audio out to the sound card from 6 feet to 18 inches.
        Message 3 of 12 , Aug 4, 2007
        • 0 Attachment
          Had double humps here at first too. They went away when I changed the
          length of the cable from the audio out to the sound card from 6 feet to 18
          inches. Only have small middle spike now. Probably won't get any better
          as I'm not using an optimum sound card.



          Chuck Carpenter, W5USJ, Point, -|- Rains Co., TX -|- EM22cv -|- 72/73
          52 years -|- 19 as K2OFN and 33 as W5USJ -|- Most fun = QRP since 1984.
          Website = http://www.w5usj.com hosted by Hamnutz.com
        • g3vnc
          Hi Bill Try a ferrite core for antenna isolation. I ve used 2+2T bifilar wound on a FX1115 ferrite bead successfully on receive. I ve also used binocular
          Message 4 of 12 , Aug 4, 2007
          • 0 Attachment
            Hi Bill

            Try a ferrite core for antenna isolation. I've used 2+2T bifilar
            wound on a FX1115 ferrite bead successfully on receive. I've also
            used 'binocular' ferrite cores with 2 or 3 turns on each winding.
            These transformers are inherently broadband.

            Interested to hear you use a DDS60 for the Rocky IQ calibration.
            I've been using an old Heathkit RF-1U signal generator manually
            sweeping the band; much better than off the air calibration.

            Nick G3VNC
          • Alan
            ... From: g3vnc ... Or if you have an ordinary transceiver and a dummy load. Set it to AM/CW/FM low power and sweep it over the range. 73 Alan G4ZFQ
            Message 5 of 12 , Aug 4, 2007
            • 0 Attachment
              ----- Original Message -----
              From: "g3vnc"
              > I've been using an old Heathkit RF-1U signal generator manually
              > sweeping the band; much better than off the air calibration.

              Or if you have an "ordinary" transceiver and a dummy load. Set it to
              AM/CW/FM low power and sweep it over the range.

              73 Alan G4ZFQ
            • Bill Dumke
              The following test was done with the Delta 44 sound card feeding an Audigy 2 sound card for the speaker, as were the other tests in the past. I just found that
              Message 6 of 12 , Aug 4, 2007
              • 0 Attachment
                The following test was done with the Delta 44 sound card feeding an
                Audigy 2 sound card for the speaker, as were the other tests in the past.

                I just found that switching the crystal jumper back and forth will
                sometimes trigger the noise hump(s) to appear. It doesn't matter which
                of the two crystals are being selected. It can appear with either one
                selected. But when it does appear, turning the power supply off and
                then on gets rid of it. (When the power is turned off a single smaller
                noise hump also appears, but this is because of the lack of band noise
                which covers that up in normal operation. This shown in the last photo
                in WA9PWR dated 8/4)

                I always wondered why sometimes the noise hump(s) was/were there and
                sometimes not. I never saw any significant variation in the amplitude
                of the hump(s) moving the various cables and wires around that are
                attached to the RXTX. (BTW The only cable attached to the computer D44
                interface card is the line in cable.)

                When the noise hump does appear it almost sounds like audio feedback,
                listening to it with the CW filter in rocky set to 600 Hz BW with the
                hump(s) centered in the passband. (When the power to the RXTX is turned
                off, the single smaller noise hump that always appears also sounds like
                audio feedback.)

                Anybody experience anything similar when switching crystals? Anybody
                have any ideas?

                Bill wa9pwr

                Bill Dumke wrote:
                > Alan,
                >
                > Thanks for the comments. I was about to try the battery, but now the hump is
                > gone on the RXTX v6.2 40/30 using the D44 Interface with the 1:1 RF transformer,
                > still using the AC power supply. I am not doing anything else different, so it
                > must be the routing of the cables. All I did was disconnect the old SoftRock40
                > and replaced it with the RXTX v6.2 on top of the wood desk. Well, I don't
                > understand it, but at least there is hope now!
                >
                > I posted tonight's photos in the photo section under WA9PWR. They end with the
                > suffix 8/3 meaning today's date of 8/3/07. You can see the hump totally gone
                > with the 1:1 RF transformer in line. Signals tonight are strong on 40 meters as
                > well. The RXTX sure sounds good on SSB. Used a Russian Shortwave broadcast
                > station to calibrate the frequency of the receiver. Wish I could transmit SSB
                > with Rocky. Hopefully, some day I will be able to do that.
                >
                > Anyway, with this good news, I am going on to calibrating the receiver IQ in
                > Rocky and then going on to testing the transmitter. (I use a DDS-60 (amplifier
                > removed from the DDS-60 circuit) with a step attenuator in sweep mode to do the
                > Rocky IQ calibration.) I will have to do it now for both the CW and SSB
                > crystals, independently, somehow.
                >
                > BTW, for the 1:1 RF transformer I am using about 24 turns of #26 AWG Beldsol
                > wire wound bifilar on what I believe is a T50-6 yellow core toroid. How good it
                > will be in transmit, I don't know. Also don't know how much loss or bandwidth I
                > have either. Just sort of guessed at it.
                >
                > Bill wa9pwr
                >
                > Alan wrote:
                > > ----- Original Message -----
                > > From: "Bill Dumke"
                > >
                > >> 4. I also don't believe the AC power supply was contributing anything to
                > >> the middle hump(s) since the SoftRock 40 did not show any noise at all
                > >> at the center frequency.
                > >>
                > >
                > > I have mainly used the PC power for my Softrock Lite RXs with no problems
                > > around the centre, apart from the spike which varies depending on the
                > > soundcard I use.
                > >
                > > I have seen your problem with the RX/TX 40/80 however. After ensuring there
                > > were no ground loops I rerouted the audio cables and the hump was gone. I
                > > think the RX/TX has a different grounding than the Lites.
                > >
                > > The soundcard input is very sensitive to interference. Try first with just
                > > the D44 input connected. Make sure the lead is good quality. Try different
                > > cables. Alter the position of the breakout box and Softrock relative to the
                > > monitor and PC. Try different settings of jumpers in the D44 box.
                > >
                > > You may eliminate the PSU by using a battery, I think a 9 volt PP3 would be
                > > OK for an RX only test, certainly it works on a Softrock Lite.
                > > To make certain check the 5 volt line.
                > >
                > > 73 Alan G4ZFQ
                > >
                > >
                > >
                > >
                > > Yahoo! Groups Links
                > >
                > >
                > >
                > >
                > >
                > >
                > >
                >
                >
              • Len Warner
                ... mini Ring Core Calculator 1.2 (minirk12) will tell you something about this. ARRL HANDBOOK 2001 [6.84] says the reactance of the windings should be at
                Message 7 of 12 , Aug 4, 2007
                • 0 Attachment
                  At 7:37 pm ((PDT)) Fri Aug 3, 2007, in Digest 1867 Bill Dumke wrote:

                  >BTW, for the 1:1 RF transformer I am using about 24 turns of #26
                  >AWG Beldsol wire wound bifilar on what I believe is a T50-6
                  >yellow core toroid. How good it will be in transmit, I don't
                  >know. Also don't know how much loss or bandwidth I have either.
                  >Just sort of guessed at it.

                  mini Ring Core Calculator 1.2 (minirk12) will tell you something
                  about this. ARRL HANDBOOK 2001 [6.84] says "the reactance
                  of the windings should be at least four times the impedance that
                  the winding is designed to look into at the lowest design
                  frequency." (minirk12 Help agrees.)

                  The RXTX v6.2 40/30 is a 50 ohm system, so at 7.5MHz
                  the reactance of the transformer winding should be 200R.

                  Using minirk12, this requires 33 turns. It will also calculate
                  core loss for you - it's very small (even with your 24 turn
                  winding - but about doubled).

                  It isn't clear to me whether minirk is using peak or RMS
                  voltage, but 20Vp-p at 1W is quoted in RXTX Builder's
                  Notes and entering that only gives about a 0.02W loss,
                  a temperature rise of 2 degC, and flux 26% of maximum.

                  If minirk assumes RMS volts, at 7.071Vrms the losses
                  are only 1/8 as big, which must be negligible.


                  Regards, LenW
                  --
                  A: Because it destroys the flow of the conversation
                  Q: Why is top-posting bad?
                  (and because it's usually followed by atrocious quoting ;-)
                • Bill Meahan K8QN
                  ... [snip] ... [snip] I ve just started building my RXTX so I can t speak from direct experience, but I note there is nothing grounding the crystal cans. Many
                  Message 8 of 12 , Aug 4, 2007
                  • 0 Attachment
                    --- In softrock40@yahoogroups.com, Bill Dumke <billd@...> wrote:
                    >

                    [snip]

                    > I just found that switching the crystal jumper back and forth will
                    > sometimes trigger the noise hump(s) to appear. It doesn't matter which
                    > of the two crystals are being selected. It can appear with either one
                    > selected. But when it does appear, turning the power supply off and
                    > then on gets rid of it. (When the power is turned off a single smaller
                    > noise hump also appears, but this is because of the lack of band noise
                    > which covers that up in normal operation. This shown in the last photo

                    [snip]


                    I've just started building my RXTX so I can't speak from direct
                    experience, but I note there is nothing grounding the crystal cans.
                    Many times in the past, failure to adequately ground crystal cans,
                    either for IF filters or frequency-determining elements, has led to
                    all sorts of oddball problems. The NorCal 38 Special I built was a
                    good example of this. Even in commercial equipment, it is not unusual
                    to see some sort of extra grounding applied to crystal cans.

                    Perhaps a piece of 22-ga. bare tinned wire soldered to the cans and to
                    the groundplane at the base of the cans would help?

                    This is just a guess, but it does have some rational basis behind it. :-)

                    73 de K8QN
                  • Bill Dumke
                    Bill. The crystal cans are grounded with a small wire from the PC board to each can as per the instructions. But thanks for the idea. Bill wa9pwr ... [snip] I
                    Message 9 of 12 , Aug 4, 2007
                    • 0 Attachment
                      Bill.

                      The crystal cans are grounded with a small wire from the PC board to each can as per the instructions.  But thanks for the idea.

                      Bill wa9pwr

                      Bill Meahan K8QN wrote:
                      --- In softrock40@yahoogroups.com, Bill Dumke <billd@...> wrote:
                        
                      [snip]
                      
                        
                      I just found that switching the crystal jumper back and forth will 
                      sometimes trigger the noise hump(s) to appear.  It doesn't matter which 
                      of the two crystals are being selected.  It can appear with either one 
                      selected.  But when it does appear, turning the power supply off and 
                      then on gets rid of it.  (When the power is turned off a single smaller 
                      noise hump also appears, but this is because of the lack of band noise 
                      which covers that up in normal operation.  This shown in the last photo 
                          
                      [snip]
                      
                      
                      I've just started building my RXTX so I can't speak from direct
                      experience, but I note there is nothing grounding the crystal cans.
                      Many times in the past, failure to adequately ground crystal cans,
                      either for IF filters or frequency-determining elements, has led to
                      all sorts of oddball problems.  The NorCal 38 Special I built was a
                      good example of this. Even in commercial equipment, it is not unusual
                      to see some sort of extra grounding applied to crystal cans.
                      
                      Perhaps a piece of 22-ga. bare tinned wire soldered to the cans and to
                      the groundplane at the base of the cans would help?
                      
                      This is just a guess, but it does have some rational basis behind it. :-)
                      
                      73 de K8QN
                      
                      
                      
                       
                      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/
                       
                      
                      
                      
                        
                    • Len Warner
                      ... If I were staring from scratch I would probably use a ferrite too but Bill has already chosen a T50-6 and it s quite adequate. ... But this is for an RXTX:
                      Message 10 of 12 , Aug 5, 2007
                      • 0 Attachment
                        At 4:07 am ((PDT)) Sat Aug 4, 2007, in Digest 1868 g3vnc wrote:

                        >Try a ferrite core for antenna isolation.

                        If I were staring from scratch I would probably
                        use a ferrite too but Bill has already chosen
                        a T50-6 and it's quite adequate.

                        >I've used 2+2T bifilar wound on a FX1115 ferrite bead
                        >successfully on receive.

                        But this is for an RXTX: there is already isolation in
                        the SR Lite RX (there could be on the RXTX too**).

                        Is the FX1115 capable of 1W? (Obviously the
                        binocular BN-43-2402 is.)

                        At 1W, the T50-6 still has plenty of headroom left.

                        [** were it not for the DC drive requirement of the
                        T/R switch BS170 fets. It would be good if
                        someone worked out an alternative isolated drive,
                        stacked optocouplers might be one not very
                        economical way.]

                        >These transformers are inherently broadband.

                        The higher permeabilty of ferrite over iron dust makes it
                        easier to extend the lower end of the frequency range
                        - but this is not a broadband requirement, it is for a
                        7.5MHz - 10MHz rig, close frequencies which lie right
                        in the middle of the application range of type 6 material.

                        All Bill needs is more turns.

                        With his chosen wire gauge, 33 bifilar turns won't quite fit
                        in a single layer but he should be able to manage a slightly
                        "banked" winding where the turns overlap on the inside
                        in order to avoid any need to wind a second layer.


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