In the RAAF, the main reason for an SRA was to approach a different airfield from where the radar was situated, e.g.radar at Pearce, fly the approach to Gin Gin 15 miles north with the runway at 90 degrees to the radar. The controllers did a reasonable job, but they were dealing with students. And unlike the GCA where the radar sweeps painted the aircraft twice a second, they only saw the blip every 6 or 7 seconds, so the heading changes needed were considerably larger than on a GCA.
Chief Galah said, many years ago:
Sort of training for both sides of the fence. One of these "practise diversions" undignifyingly ended up with a wheels up on Rwy 34 one day.
Reputed to be the only survivable wheels up by a MIR3.
I'm unsure if a SRA was involved with that, but the "practise diversions" seem to disappear from the repetoire after that, funnily enough.
That wheels-up came from a short hop from RAAF Laverton to Melbourne, not a diversion. The weather was a bit ordinary and he wanted to stay VFR under the cloud, and he only had UHF radio, so I think he was talking to the tower via Mel Approach, as tower only had VHF at the time. At 250t plus it doesn't take long to cover the distance, and it was a straight-in approach, whereas the knuckleheads were only used to an initial/pitch/oval circuit to land. So, because there was no "downwind leg" he didn't do his "downwind checks" and screeched to a halt on his big drop tanks on the wet runway, the tanks being the only reason he survived - otherwise, the nose-high approach attitude would mean that when the tail hit the ground, the cockpit would slam onto the runway and drive the pilot's @ss into his helmet.
This was the cause of 2 things:
Changing the name to "Pre-landing checks"; and
The joke "How does a fighter pilot know he has done a wheels-up landing? He needs full afterburner to taxi."