Hi Howard,
I did indeed bring my Whitworth, BSF, UNF, UNC taps and dies with me!
I've done a bit of a study of the available information about the rear bearings – pictures I've found, Tony Griffiths' lathes.co.uk accounts etc.
Text copied from lathes.co.uk in Bold
On the Britannia 2 and 3 the design of the headstock was initially pretty simple, a front bearing, and a centre spike and socket controlling both end-float and axial accuracy at the rear end of the headstock mandrel.
Both models could also be ordered, for £1 : 10s : 0d extra, with its headstock spindle and bearings in hardened steel.
At half the price of the basic lathe the hardening, however it was achieved, was very costly!
Back-gear was also available on the No. 3 lathe but, as this involved an entirely different headstock with two spindle bearings, and the end thrust taken by a swan-neck casting bolted to the end of the bed (the arrangement can just be made out in the picture above), the cost almost doubled to £4 : 0s : 0d. When fitted with backgear the lathe could also be specified with a heavier pattern of bed, based on that used for the No. 13 lathe, where the section that bowed downwards as it approached the headstock held a proper detachable bridge – and all for an extra charge of just £1.
This arrangement was further developed for the 13, in that the rear of the headstock casting incorporated the end-float control screw (as it had now become) into a development of the rear back-gear carrier, rendering the interim bracket bolted to the end of the lathe bed redundant. Tony’s account suggests that the 13 was already available when backgear was made optional on the 3. This makes it hard to understand why they didn’t just use the 13 headstock casting?
I have the 1899 catalogue, which seems to list pretty much all the lathes, albeit not always in the form that one sees nowadays. E.g. the table between lathe and legs is not often show, but seems often to be visible today?
It would be very good to find out just when the different models became available!
From this and the pictures I've accumulated it looks to me as though a rear bearing was incorporated on the 13, and in my case it looks to have a means of moving it axially. But it looks to me that end-float was still controlled by a 'spike in the adjustable piece which passes through the modified rear-end of the headstock casting.
I don't have anything very obvious with which to try and turn the cap on the rear bearing, but I'm guessing it is over-tight. While I've got the backgear spinning perpetual-motion style I'm still well short of that on the mandrel, even when the backgear and the tumbler gears a in neutral. I guess I need a bit of suitably thickish plate to drill holes in!
Here is how mine was:
You can see what looks like a brass nut with nasty little radial holes for me to try and turn it with? This, and the integrated bracket holding the 'push from behind' object are characteristic of the 13, and the later No.3's
Here, on the other hand, courtesy of Tony Griffiths, is an earlier arrangement, on a No.3. You can see the cone and the simplified rear end of the headstock casting (which clearly, like mine, has a thread in it!).
Also, in this case there needs to be an adapter to permit a MT1 centre to be fitted (and something to drive the work). On my 13 I think the mandrel has a MT1 bore – it certainly has a big enough hole in it!
And here is the later arrangement, again depending on a ball as yu have described, but now differently supported on a No.14 (this also courtesy of the lathes.co.uk site. Actually this is more similar to mine than I initially realised. There is the brass – or maybe not brass – nut apparently holding the rear bearing in place, while the bridge-and-two-studs scheme does the axial pushing?
Best regards, Tony