by 5B4AIY » 10 Oct 2017 05:56
Hi, Jef,
OK, let's see if we can explain this. The output from the DDS on pin 21, Iout, is a current that is converted to a voltage by the input resistor, R15/200 ohms, of the low pass filter. The complementary output from pin 20, Ioutb, is similarly converted to a voltage by R12/100 Ohms. Both the normal and inverted DDS outputs are summed by R13 and R14 and should be exactly zero, decoupled by C9. This is the reference comparison voltage for the comparator and is coupled to pin 15, the negative comparator input. The normal output from the DDS is low-pass filtered and the sinusoidal signal is coupled to the positive input of the comparator on pin 16. The normal and inverted comparator outputs are coupled to the dual-D cross-connected bi-stable to provide the I and Q local oscillator signals at half the frequency of the synthesiser's output.
In a perfect arrangement the polyphase networks would have exactly 90 degrees of phase shift, but in the practical world there are always minor errors. R42 alters the switching threshold of the comparator by a small amount, and this has the effect of slightly advancing or delaying the phase of the I and Q signals, allowing you to obtain a better unwanted sideband null.
To adjust, inject a S9, 50uV CW signal into the transceiver on your favourite band. Tune the signal in using either the LSB or the USB mode. Switch to the opposite mode and the signal should be greatly attenuated. Carefully adjust R42 for the greatest null in the audio output. There should be a distinct reduction in the audio signal. If you do not get any significant change to the audio output when you adjust R42, then carefully examine the connections involving R55, R42, R40, R14, and R13. In my case, a poor soldered joint was causing the problem.
Adrian, 5B4AIY