This page details some experiments building a SoftRock 6 or 40 and 20 meters. My goal is to use a DDS 60
in place of the crystal controlled oscillator. My DDS 60 can drive about 700 mVpp into a 50 ohm load at 56 MHz
(4x LO needed for quadrature generation). My first attempt at this was to drive an LT1719 comparator with the DDS and then
take the resulting square wave from the comparator and feed it to the SoftRock's quadrature generation chain This approach did not work well for 20 meters, it appears the LT 1719 runs out of gas somewhere around 45 MHz. For 40 meters it worked find, but the goal was to get 40 to 20 meter coverage so will have to find a faster comparator or a better sine wave to square wave converter to pursue this approach.
The QSD circuit also responds to odd harmonics of the clocking frequency.
So the thought was to try a clock rate that would result in sampling on the 3rd harmonic.ie: 4/3 the frequency of interest.
This results in an LO frequency of 18.8 Mhz to receive 14.1 MHz and 9.466 MHz to receive 7.1 MHz - well within the range of
my DDS 60 + LT1719 setup. Some loss compared to baseband sampling is expected in using harmonic sampling,
as only 1/2 the energy is being captured when sampling using the 3rd harmonic compared to the 1st harmonic.
Below are shown some panadapter screen shots from my setup. The first setup I tried was baseband sampling for 40 meters - a 28.4 MHz clock signal applied to the SoftRock. With this setup a -14 dbm signal applied to the antenna input was the maximum signal receivable without distortion, and it produced a 4.14Vpp output IQ signal. MDS for this configuration was measured at -111 dbm. This measurement is comparable with crystal controlled Soft Rock 6 I've previously measured.
Sampling the same signal at the third harmonic (a clock signal of 9.466 MHz) resulted in a maximum signal receivable w/o distortion of -4 dbm producing producing an output signal of 4.54 Vpp at the IQ output. MDS of this configuration was measured as -106 dbm. This is the second panadapter screen shot below.
The final panadapter screen shot is of a 20 meter 3rd harmonic configuration (18.8 MHz clocking). This configuration resulted in a maximum receivable w/o distortion signal level of -10 dbm producing an
output of 4.3Vpp. The MDS of this configuration was measured as -109 dbm. The
20 meter panadapter screen shot below shows more spurs than the 40 meter setup. I suspect the 18.8 Mhz clocking may have some small oscillation in it as the spurs seems to wander between two states at a 1-2 Hz frequency. The MDS was done at a region that was relatively spur free (12 khz or so above the central frequency). I'll have to try and eliminate this oscillation and suspect it may help a little on the MDS measurement.
One other point of interest is that for the 3rd harmonic configurations, the spectrum is inverted. Swapping I/Q will reverse the spectrum to get a correctly oriented spectrum.
Comments to: Bill Tracey (kd5tfd@ewjt.com)
Last Updated 23 April 2006 Copyright © Bill Tracey 2006