Myford S7 Spindle Nose / Chuck fitting.

re bogstandards post august 2011 on myford spindle threads i have recently purchased an er collet chuck which is rather tight fit on the lathKke spindle thread i would like to know where i could borrow or buy the tap to try to rectify this

sidetrack

Surely the fit should a slight interference fit to maintain the chucks accuracy, especially when work is removed with the chuck for secondary work and then returned to the lathe.

---

That's my understanding

I was always impressed by the accuracy of the nose on the wee Unimat SL, and how well original parts fitted onto it

The flat part parallel with the thread and the enlarged rear flat part are the all important register which is responsible for radial and axial concentricity

The thread is only responsible for tensioning everything up

Steve,

I agree with your understanding of the design; and I would venture to suggest that "a 0.0005" taper towards the threaded portion" was Myford's way of ensuring that the register would be effective.

... It would, of course, be a disaster if the taper ran the other way.

Without access to the design drawings; we can only guess, but I would expect to see that taper expressed as something like +0.0000/-0.0005 at the minor diameter.

With a suitably bored backplate [!] that would allow the chuck to screw on freely, with reducing radial clearance, to the point where the rear-face register takes effect.

Thus making the best of a theoretically imperfect design.

MichaelG.

Michael Williams has the correct understanding here.

Martin.

To my mind, 3-jaw chucks tend by their nature to hold stock slightly eccentrically, often by an amount which varies with the diameter of the stock, and which might well swamp the effect of a very small amount of looseness in the register diameter. Slight eccentricity in the body of an independent chuck is immaterial, because the work itself has to be centered up.

However, an ER collet chuck needs to run true. Geoff, you might try a bit of emery paper (followed by a good clean-up) on the crests of the thread inside yours, in case the crests are poorly finished. And are you sure that it's the threads that are tight, rather than the register diameter?

Andy

I agree Michael Williams is correct. I always undestood and practiced that screw threads are not an accurate form of location or register but only a means of fastening parts together.

Tony

Tell me something then.

I have an ML7 clunker here with worn spindle nose, it's that worn you can see it.

The thread looks like a pipe thread and the register is about 3 thou under and also visibly tapered.

When the chuck is screwed on it flops about until it reaches the end of the register but when checked it has less than a thou run out.

A brand new backplate has the same runout.

An earlier lathe I have has just a thread, no register at all but still has a decent and REPEATABLE run out.

John S.

In the first case perhaps the register diameter is not undersize right up to the shoulder. In the second case without a register, it's obviously pulling up on the thread, but I would still suggest it isn't the best way of locating a chuck to ensure concentricity?

Tony

The thread does play a part in centralising the chuck as JS points out or if it's that poorly made then the thread is only pulling the chuck up tight on one side. I would imagine that after a while, some deformation of the materials involved will make a larger area of thread become in contact but that too only reinforces the idea that the thread is playing a part in positioning the chuck. A parallel register can only limit the amount of eccentricity, maybe to extremely fine and very acceptable tolerances but there must be a clearance in order to be able to get the thing on.

OK, just as a test been out in the shop, put a bit of 3 thou shim [ only small bit I could find ] between the chuck and the back of the register on one side, screwed the chuck on, went on OK so proving the register must be well undersize but now got that much of a wobble on it's a joke to try and measure it.

Remove the shim and back to acceptable limits.

[EDIT] Unfortunately this machine has no motor or countershaft fitted, as I say it's a clunker and used for trying thing to fit but if I get time next week I'll get power to it and turn the register off.

Spindle is worn out anyway so it's not sacrilage.

Edited By John Stevenson on 27/12/2012 14:04:51

Once the rear face of the backplate comes into contact with the spindle, it establishes the final angles between the thread faces in the backplate and the thread faces on the spindle and as the chuck is drawn on further, the two will settle into a centralised position where the radial forces are equal in all directions. Hopefully that position should lie where the backplate shoulder isn't in contact with any part of parallel register. In practice, it probably will be in contact with one side of the register but that means the register is then providing a sideways force to limit the eccentricity. Come on guys, this is basic mechanics.

It's the same principle as a centre in a tailstock except the conical surface is a spiral rather than a straight cone.

Edited By Chris Trice on 27/12/2012 14:10:15