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

Re: ATS3B passband

Expand Messages
  • Vojtech
    Hi Fred. FYI, this is what I measured 3 years ago when I replaced the shunt capacitors in the ATS-3b with varactors and tuned the voltage:
    Message 1 of 4 , Jan 9, 2011
    • 0 Attachment
      Hi Fred.

      FYI, this is what I measured 3 years ago when I replaced the shunt capacitors in the ATS-3b with varactors and tuned the voltage:
      http://www.kufr.cz/~ok1iak/temp/ATS3b-filter/

      Interesting thing, the output response after the audio peak filter changes slightly with load (with earphone impedance). I suppose there is some feedback between the two op amps in the same IC package.

      I would recommend not to tinker with the filter. If you really need to tune the filter, you would have to experiment with the input / output impedance (the output impedance is mostly defined by the loading resistor, if I remember correctly) and with the input / output / shunt capacitors. Ideally you would desolder the crystals and replace them with closely matched ones. IMHO Just blindly modifying the capacitors will likely destroy the board before reaching improvement though.

      I am not sure whether the filter input is correctly matched by the input LC section. I tried to experiment with that, but it did not improve really. On the other side I did not invest too much time and I was worried to destroy the board. I believe the double peak indicates too light loading of the filter. If the filter is loaded too heavily, the peak will be round and attenuated.

      73, Vojtech OK1IAK, AB2ZA

      --- In AT_Sprint@yahoogroups.com, "Fred Maas" <just.one.hill@...> wrote:
      >
      > Attached below is the spectrogram of my ATS3B passband. There is something of a dip in the middle of the center, and I can hear that when using the radio. The step tuning sometimes lands me right in that dip, with the next step off on the skirt.
      >
      > Years back on this reflector, I read that there was a mod that fixes it. I e-mailed the author, that was some time ago, and no response.
      >
      > Does anyone know the mod that fixes this???
      >
      > 72, Fred - KT5X / W5YA
      >
    • Fred Maas
      On the other side I did not invest too much time and I was worried to destroy the board. I believe the double peak indicates too light loading of the filter.
      Message 2 of 4 , Jan 10, 2011
      • 0 Attachment
        On the other side I did not invest too much time and I was worried to destroy the board. I believe the double peak indicates too light loading of the filter. If the filter is loaded too heavily, the peak will be round and attenuated.

        73, Vojtech OK1IAK, AB2ZA

         
        Wow.  That is exactly the information I was looking for.  You did good work, and thank you for sharing.  That dip that I hear (and see with Spectrogram) in the center of the passband is sure real.
         
        Now ... pardon my ignorance; how does one "load" the filter??? 
         
        73,  Fred KT5X / W5YA
      • Vojtech
        Hi Fred. I measured the spectrogram by feeding the receiver with noise generator. You load the filter the same way you are supposed to load 50 Ohm transceiver
        Message 3 of 4 , Jan 10, 2011
        • 0 Attachment
          Hi Fred.

          I measured the spectrogram by feeding the receiver with noise generator.

          You load the filter the same way you are supposed to load 50 Ohm transceiver output with 50 Ohm antenna.

          http://www.stanford.edu/class/ee133/datasheets/SA612A.pdf

          Reading the datasheet I believe each input pin and each output pin has impedance of 1.5kOhm. I wonder what the working impedance of the crystal ladder filter shall be to show a nice single peak. It really depends on parameters of the given crystals. I guess the impedance will be around 600 Ohm. Therefore the impedance needs to be transformed from 1k5 to 600 Ohm before crystal filter and from 600 Ohm to 1k5 after the filter.

          If I remember correctly, the input transformation is done with a LC network similarly to an LC antenna tuner. The output is loaded with resistor. Two transformers would work, but they are bulky. Elecraft KX1 has the input transformation done with a single transistor emitor follower stage.

          It is interesting to compare the filter circuits of the Elecraft K1, KX1, SW40+ and ATS. They are all 2xNE612 type of superhetodyne with narrow band ladder filter.

          73, Vojtech

          --- In AT_Sprint@yahoogroups.com, "Fred Maas" <just.one.hill@...> wrote:
          >
          > On the other side I did not invest too much time and I was worried to destroy the board. I believe the double peak indicates too light loading of the filter. If the filter is loaded too heavily, the peak will be round and attenuated.
          >
          > 73, Vojtech OK1IAK, AB2ZA
          >
          >
          >
          > Wow. That is exactly the information I was looking for. You did good work, and thank you for sharing. That dip that I hear (and see with Spectrogram) in the center of the passband is sure real.
          >
          > Now ... pardon my ignorance; how does one "load" the filter???
          >
          > 73, Fred KT5X / W5YA
          >
        Your message has been successfully submitted and would be delivered to recipients shortly.