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

72448Re: [softrock40] Anomalous R 32 and R 33 voltage versus frequency values

Expand Messages
  • Paul Marbourg
    Feb 25, 2013
    • 0 Attachment
      Thanks for the reply, Alan. Do you think this is a part specific problem that may improve by swapping in another FST3253? Or, is this a general device/layout problem that is inescapable? 
      I built a NorCal 2030 that used discrete architecture for each 90 degrees of wave sampling and combined wave phase externally for IQ generation. It is a sweet implementation straight from the inventor's bench. 

      Perhaps this or even the double balanced QSD mixer approach would have yielded truer quadrature phase and amplitude relationships over the softrock ensemble's frequency domain. 

      73 & thanks again, Alan
      Paul WN7T
      Sent from my iPhone

      On Feb 25, 2013, at 12:00 AM, Alan <alan4alan@...> wrote:

       


      ----- Original Message -----
      Subject: [softrock40] Anomalous R 32 and R 33 voltage versus frequency values

      > Hello group. After assembling my ensemble 2 receiver, I tested the voltages on resistors R32 and R33 versus frequency. The R32
      > voltage values changed considerably, while the voltage vs frequency for R33 remained very close to the nominal 2.50 volts. The V
      > vs Freq remained within 0.3 V for R32 & R33 up until filter bands 2&3, when they began to diverge. There was a big broad dip
      > between 8 MHz & 16 MHz bottoming at about 1.83 V @ 13.2 MHz for R32. R33 @ this freq was 2.42 V on a Fluke 289 VOM.
      >
      > At 16Hz, R32 = 1.86 V, dropping from 2.18 V as the Freq passed into filter band 3.
      >

      Paul,

      In fact this is not unusual. It is said to be due to a slight imbalance of stray capacitance in the mixer.

      > All other voltages are normal. The receiver works very well on the lower bands, but degrades with respect image suppression on 20
      > meters and above.
      >
      > I reflowed T3's solder points to no avail. All pin/pad voltages are nominal. Soldier joints look good. Clock performance okay on
      > my Tek 454 scope.
      >

      Uncorrected image rejection may reduce at higher frequencies due to them being more affected by stray capacitance and tolerances.

      73 Alan G4ZFQ

    • Show all 4 messages in this topic