Is .0005 runout in 10" acceptable?

[EdK]

It seems to me you need to have one known reference to go by and if you don't know if the lathe bed is untwisted and you don't know if the head axis is aiming at your groin instead of where it should be aiming then I don't see how that method can possible work.

Ed

[Sunset Machine]

Hi Ed,

Sorry to take so long getting back, been busy. My yard was a mess; trees fallen, grass high, sprinkler system leaking, weeds everywhere. Plus! I'm finally making a permanent RPC and now have it mounted to a wall. Still need to do some wiring to get that 3-phase power out to where it's needed.

About Rollie's... You're correct, but maybe confused about that reference - it's all in the spindle, those two points (spindle bearings) which make a straight line. Everything has to relate to that. On a worn lathe you are making a new reference: extending an imaginary line out from the spindle and averaging out the errors. Like waves on a lake, you are shooting a line above the lows and below the highs. Unlike a lake, a lathe has two planes to deal with, but the idea is the same. It's kinda like sighting in a rifle; the target is the tailstock.

People tend to focus on the condition of the bed, but it's not that important (well, if you are buying a lathe it is - "look at that worn out bed! The price is too high..."). Do some trig and you'll see. If the bed is worn evenly front and back (like a good design would), that dip in the middle will have very little, if any, significance. A guy might even improve accuracy by artificially inducing wear (emery cloth). Errors come from the tool bit moving in and out, not so much up and down (in and out coming from the differing bed wear front to back). The up and down movement affects small diameter work far more than large ones, but in that case the length of the work is usually short and the error close to unmeasurable on the finished part. Long, skinny work offers it's own set of problems with deflection and might require a bit of hand work anyway.

A new lathe with a properly aligned headstock is almost a piece of cake to "level" with Rollie's, but the worn ones take some imagination. In either case, you need to followup by making chips and testing the full work envelope out to its limits and up close. You'd do this even using a real level, which, btw, won't correct a misaligned headstock either.

Oh heck, just bolt the sucker down and run it! :)