Having changed the design to stay in-line with my modular approach, I have split out the first mixer, post mixer amplifier and roofing filter in to a separate module.
This approach will allow easier modification as the radio evolves, for example if I change the roofing filter all I have to do is fabricate another daughter board to accommodate the replacement and as the filter matching networks are on the daughter board this means I don’t have to touch the mixer mother board.
My initial calculations for the mixer IF port diplexer were wrong as my first tests noted a severe loss in gain, the mixer output was in the region of -6dB and the output from the diplexer was in the region of -22dB – something was clearly wrong. Using the tracking generator on the analyser I swept the diplexer and noted a sharp peak around 5MHz, way below my 10.7MHz IF so I re-calculated the LC networks and re-wound the T50-6 toroid inductors and the peak was closer to 10.7MHz and relatively flat everywhere else – and only seeing a 2dB loss.
The final output from the crystal filter was in the region of +3dBm, more than enough to drive the first IF amplifier stage (work in progress after re-design) so I can tick this one off the list and mark as complete – pictures and videos below.