I seem to remember a very simple design appearing in one of the model engineering mags some time ago based on using a bar for the bed. Might have been MEW. Probably square bar rotated by 45 degrees.
There are some interesting ideas from older lathes. For instance one plain lathe uses a totally different saddle for screw cutting. That is then moved along the bed to where it's needed just like the usual slide arrangement.
Some old cheap lathes used nothing more than a pair of round bars for the upper slides rather than dovetails. Some also used Myford type bed rails on the upper slides.
My father reckoned the best way to get a super precision small lathe was to make one myself. I'd been thinking in terms of using flat ground stock sections. His attitude was why – ordinary steel sections only vary by thou's. Makes sense really as things are going to have to be fitted and dowelled / double roll pinned.
He mentioned that an interesting lathe with an unusual design of bed was produced by one manufacturer just after WWII. A slant bed. They can have larger swings and be very rigid compared with ordinary beds. The saddle hangs down from the back so the bearing surfaces can be very large. Trouble is that the cross slide needs to run horizontal. I'd guess he was thinking in terms of a very thick fairly wide steel section. I feel it would be too difficult to make but he might have been suggesting a rather unusual spindle to bed arrangement – bed flat and spindle moved back rather than on centre. Not sure. He'd usually only gave me clues especially as he got older.
One thing that is a fact is that plain bearings are best. It struck me that oilite's could be used to save keeping on having to make them. Maybe even running on needle roller bearing inners pressed onto the spindle. it wouldn't be too much of a problem to make an attachment that could be fitted to the bed to finally size both the bearings in the head stock and the tail stock by hand. It should be just a case of removing a few thou of material. From what I can gather this was done by people a long time ago to save the pain of having to accurately scrape them. The manufacturers have used similar ideas even jigs to position and size them often rotating so that both bearing can be at the same height and alignment.
Thrust – a race of some sort will be best for that but for insanely high speeds I suspect Pultra used hardened steel running against phos bronze as I feel that the thrust bearing they used on some would have a hard life at 10,000rpm.
I think about the slant bed and wonder how every now and again. I wonder if what he was thinking about would be better described as a slant head. In my minds eye it makes sense.
John
–