I've worked on this a couple of times over the last few days and I've figured out the right way to control the tuning of the front end section. With the right test setup I can now tune the front end and see a test signal appear at the IF output, at the expected 10.3 MHz frequency. Switching on and off the output of the signal generator or varying the amplitude confirms the same happens at the IF.
The unit uses a first mixer, fed with the input and PLL signals. This selects the input frequency that is tuned to the first intermediate frequency passband.
The PLL can be tuned from 512 MHz to around 1 GHz, I can successfully get a PLL lock and by measuring at the divider can confirm the frequency is as commanded.
The first mixer output appears to be bandpass filtered, centred around 520Mhz, if my figuring is correct - I can't check this precisely, it's just outside of the range of the other spectrum analyser I have access to. This is then mixed with the 512MHz local oscillator to produce the 2nd intermediate frequency around 10 MHz.
Fortunately I can determine the local oscillator frequency from the divider in the PLL loop for this clock.
My test code now includes control via serial of the sweep start and end frequencies and I'm getting near to adding connections to other controls like the attenuation/gain selection.
It looks like I can now test the synthesiser board with the detector board, capture the output signal and digitise then start plotting some spectra instead of viewing the IF on my scope FFT plot.
If anyone is interesting in resurrecting the analyser section of the Seaward Sceptre, let me know, I have gathered lots of information.