Thanks to all for helpful suggestions, apologies for the delay in reply, I've been busy at work so have only had a couple of evenings to fiddle with things and to listen to the Bronze.
I've been using two alignment protractors, a combined single point/two point protactor from turntable basics and the card protractor that comes with the HiFi News test record. The turntable basics is a mirror for setting azimuth. This has three separate protactors on it, I've used the single and two point points. Using the single point protactors I have tried both eyballing the spindle and a thread to align the protractor. I end up with the same cartridge alignment - as per photo above whichever method I use. I suppose it's possible all these protractors are faulty but it seems unlikely. I spent an evening playing with VTA and came to conclusion that this is not the issue - I tried =/- .5 cm, the most the Michell VTA adapter for the RB300 can manage on the Orbe, in roughly 1 mm steps.
I started to worry that I was hearing things that weren't there or that the test records I was using were at fault so I set up the following three systems:
1. Michell Orbe SE - RB300 - Cadenza Bronze
2. Thorens TD160- SME 3009 Series 2 detachable head shell version - Stanton 881S. The Stanton has been safely in the drawer for the best part of the last 40 years since I started my MC adventure. I started with a lowly Ortofon MC10 Super mk 1 before moving on to an MC20II, MC30II, Transfiguration Spirit III then Rohmann. The last died of old age, the cantilever sheared at the diamond mounting point.
3. A homebrew turntable using the platter and bearing rescued from a defunct 401. (Sacrilege I know but the rest of the useable bits went to a good home and are now part of some other proud owner's system). The spindle runs on a PTFE bottom bearing. This cured the rumble which was dire before - it still had the original brass domed bearing, except the dome had worn away to a flat when I got it. The motor is an Audio Origami thing based on a 3.5" floppy disc driving a belt round the platter rim. (Wow is surprising not a major problem). The arm is also homebrew, a unipivot using a tapered carbon fibre tube sourced from an F1 supplier whose name I forget. The bearing cup is TiAl6V4. The cartridge is my venerable Transfiguration Spirit III.
In case anyone's interested, the amplification is a Sugden Masterclass phono stage into a Meridian G91A, thence to Sugden Symetra power amps and finally QUAD ESL 63s. All are secondhand purchases acrued over many years.
The test records used were the HiFi News test record - a new copy bought specifically to help set up the Bronze and the Decca Pavarotti Rossini William Tell (2 copies). A few tests were also conducted with the DG Guilini Verdi Requiem however this simply mirrored the results of the HiFi News and Rossini discs so I didn't spend a lot of time listening to it.
All 3 systems produced an enjoyable sound in their different ways.
The Cadenza clearly extracts more information from the groove while within its operating tracking window. Everything is sweet and clearly defined with lots of air and space around individual voices. I'm sure I'm hearing things in the mix I didn't hear before. Using the HiFi News test record, it tracks the 1st torture track on side 1 securely, with a lot of fiddling it just about, more or less manages the 2nd track and gives up on the 3rd. So better than in the Wilson-Benesch but not great.
It'll come as no surprise that the moving magnet Stanton is not as sweet sounding nor has as much space and air around the voices as the Cadenza but is still an enjoyable listen. It's also no surprise that it sails through torture tracks 1, 2 and 3 and will even have a stab at the last torture track. It tracks nearly every climax in the Rossini. This is really why the Stanton was included, the mistracking I'm hearing with the Bronze is not groove damage or a faulty pressing.
The biggest surprise of all was the 3rd system. I think we'd all agree that this has no right to be even mentioned in the company of the Orbe/Cadenza Bronze and yet it held it's own with ease, I found myself listening to a whole side of the Rossini, not just a couple of climaxes to test the tracking ability, then turning the record over to continue enjoying the performance. Although it didn't have the absolute clarity of the Bronze system, it was not far off. The beguiling Transfiguration openess and sweetness were all there, and it tracked better than the Cadenza/RB300. It managed torture track 2 and was prepared to have a go at track 3. It tracked all but the most dense climaxes of the Rossini. Having started the tests with the Cadenza, with the Transfiguration I found myself, as climaxes approached, preparing myself mentally for mistracking that never came.
Where does this leave me? Well, I had a brief moment where I said to myself, "If I'd realized the Transfiguraation was this good in this arm I could have saved myself a fortune". Then I thought, but a gimball bearing RB300 is so much easier to use and the Bronze does have a magical sound (until it doesn't). So perhaps I'll cut out the opera for a bit (or stick to theTransfiguration system for that) and enjoy the deftness of touch of the Bronze with instrumental music in the hope that after the Bronze has run in for a bit the problems iron themselves out.
Once again many thanks to those who have suggested possible cures and for letting me know that the Bronze can be made to track.