The basic problems with the radar were:
1) Made by GEC not Ferranti - bit like hiring Laurel and Hardy when you could have hired Einstein and Edison.
2) Folded Cassegrain antenna. Better in theory but not in practice. Big sidelobes which would kill you given half a chance by generating submarine targets on low level oversea intercepts - I got that twice, once at night. The good news was that the Soviets had nicked that design for our primary opposition, the Flanker. So, with a single seat and half the yearly flying hours, they were far more screwed than we were! ;)
3) The computer speed. I had done some display programming on early computers at school, so I could see where all the errors were occurring. Essentially, the radar resolution coupled with the processing cycle times meant the computer could not work out reliably between a target that had moved or a potential false echo. Add that to the antenna being more prone to generating false echoes in the first place, and one had plots dipping in and out of display, velocity vectors jumping all other the place, and any target turns showing up way late. I flew a bit with an ex-Gannet radar operator as my nav, and he knew all the tricks with a pulse set. It was quite an education in radar. After he'd said his incantations, we could see the targets in pulse, even look down on fighter size targets, at 30 miles over sea. However, the computer display would be a mess of plots jumping around, and appearing and disappearing. The junior navs (not many, thankfully) sometimes couldn't find anything in any mode.
Obviously for targets that were bigger, closer, look up, and not turning much, it performed better. The best approach was to literally sit on one's hands for the first 30 seconds and let the thing settle down. Next, adopt the standard Lightning intercept profile - Stick him on the nose, watch him drift, call him a 140 anyway, accelerate to hold him at 20 degrees off the nose. Then once you'd got inside 20 miles you would get something usable if you had made the right sacrifices to the Gods that morning. If you hadn't, you relied on the one guy in the formation who did have a half-decent air picture (it would be no better) to guide you all in, then once inside 20 miles you'd be in the right bit of sky to do a rapid scan, sort, lock, shoot when the plots did start appearing. Done well, most of you could get head sector missiles off, but you'd be working like a one armed paper hanger with an oven glove on.
Because of the false target problem, the radar was in practice quite vulnerable to clever jamming. It would work OK against straight power jamming, but gate-stealers and the like would fool it every time. Hopefully the Skyflash would go into its magic mode and relieve you of the problem, but there were a number of manual overrides, front and back, which might enable you to keep the radar pointing at the real target. One had to be ready to use them at a moment's notice.
I fired one skyflash for real, against a simulated cruise missile right on the deck, so a very tricky target. The radar lock was all over the place, but I had the wizard nav behind me, who reckoned it had the right target, so for a fraction of a second when the HUD picture looked right, I squeezed the trigger and prayed. Direct Hit! Deep Joy!
Once we got the Stage One radar with the faster computer, it got a lot better. I remember an early intercept with it, when we picked up a box four of mudmovers on the deck, overland, at 55 miles, steady plots. We couldn't believe it, thought it must be a lie, but it wasn't and we simu-shot all of them in the face and went home for tea and medals. Tick VG. Of course, sometimes it was a lie still, but it was a lot better than Z list.
As to the radar tech side of things, I know the QWIs and Nav Radar Leader would often be off talking to the radar fitters, but I was not involved in that. Essentially, I think every little tweek had to be made to maximise transmitter and scanner performance, in order to make the raw signals as clear as possible. There was a lot of swapping of modules between aircraft in order to get the best combination of kit that worked well together. This occurred with all the backseat stuff, including INs and computers, especially on detachment when spares were short. I know the techies would sometimes be up all night trying to get a "matching set" in as many aircraft as possible. This was not unpleasant work in Cyprus or Gib, but we also went to Stornoway and Noggieland
I should perhaps add that we had a very good atmosphere on both my squadrons between aircrew and engineers. I think most had been picked for some level of experience or skill because it was a new jet, and we all realized that everyone was working their best to get something out of the aeroplane. The mixed parties on detachment were very pleasant, and the Bosses were keen on annual opportunities for the SNCOs to give the aircrew a poke in the chest.