The problem with doing it in the sequence suggested by Whopity is that the diversion is started after a 180 turn on instruments and after a delay whilst the radnav fix is sorted out - all potentially disorientating. Although possibly more realistic I think it's a step too far under test conditions.
In future I think I'll move the radnav fix to the diversion leg but keep everything else unchanged (roughly as per Homeguard and Beagle's approach).
This was my second renewal test and it all seems OTT for renewing an active PPL FE. It's bl**dy expensive (FIE+aircraft+lost time+CAA fee) and all the points could be adequately covered by discussion or seminar.
HFD