A couple of screwcutting MATH questions!!!!

[Martyn GroganParticipant @martyngrogan32924]

Can anyone tell me how to convert the error in Martin cleeve’s book from 1 in… To mm

I think I just divide 1 by the error. But just checking.
E.g.
1÷465=0.00215??

Also is there a way of measuring leadscrew error and don’t laugh if this is wrong but I was thinking if we could. Could this be offset by the change gear error.
I’m gathering a worn LS would give -error???? If so we could avoid more – error???
Hope I’m not making a fool of myself with this but thought I’d ask?
Thanks for your time

[Martyn GroganParticipant @martyngrogan32924]

Help!

[Brian WoodParticipant @brianwood45127]

Hello Martyn,

I can't follow you at all, perhaps if you provide the page number and paragraph to Cleeve's error we could use that to understand what you are concerned about.

You can measure error in ;leadscrews but in my opinion they will be significantly more accurate than will have any effect on the sort of work a Myford ML4 lathe can ever turn out and most unlikely to be 'tuned out' by change wheel re-arranging

On my ML7R for example the leadscrew error is 0.0018 inches over the whole 25 inch length, the test section was deliberately chosen on the most heavily used end of the leadscrew at the chuck end. The error was measured against an industry quality DRO.

Regards Brian

[JasonBModerator @jasonb]

1 in xxx is a ratio not an imperial measure

So for example an error of 1 in 465 would in imperial be 1inch in every 465 inches or in metric 1mm in 465 mm

If you want that as an actual length then as you have done divide 1 by 465 which then gives an error of 0.00215" per inch or 0.00215mm per mm

 

Or if you want the pitch error then multiply pitch x (1/465)

Edited By JasonB on 20/04/2018 09:00:14

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Martyn Grogan on 20/04/2018 07:56:39:

I'm gathering a worn LS would give -error???? If so we could avoid more – error???

.

Continuing Brian's on theme …

It is most unlikely that wear of a lead-screw would introduce a consistent negative error; generally it just increases the backlash in the mechanism. Measured very carefully, you might find that it also introduces a localised pitch variation, but that is not something that could be compensated-for in the way you are suggesting.

It is possible to adjust the effective pitch of a lead-screw … but the methods are devious and, not really applicable to correcting for wear.

MichaelG.

.

P.S [for general interest] Here's how pitch correction is usually done: http://lathes.co.uk/bryantsymons/

That lathe is 'an object of desire' but my old toolmaker's microscope uses a similar arrangement.

Edited By Michael Gilligan on 20/04/2018 09:36:11

[SillyOldDufferModerator @sillyoldduffer]

Wouldn't work in this example but the idea of compensating for errors is respectable. Capacitance in a stable electronic oscillator is sometimes provided by using two capacitors in parallel, one having a positive temperature error, the other negative. The two errors tend to cancel each other out. Not a new idea – Harrison's gridiron pendulum is a mechanical example from 1726.

Dave

[HopperParticipant @hopper]

If you're worried about the worn leadscrew on an ancient ML4 etc. you might remedy it with the old dodge of cut the leadscrew where the threaded section joins the plain section just near the headstock. Then turn the threaded section around (end for end) and rejoin the threaded section to the plain section again. This puts the worn section of leadscrew at the tailstock end where its never used and the unworn section at the headstock end, where it is most used. Drilling the ends of the pieces of leadscrew and inserting suitable dowels and pins ensures concentricity is maintained.

But really, for most home hobby work, it's not worth worrying about. Threads are made to a clearance fit so there is room for small errors. And until you get very proficient at screwcutting, you'll probably be using a die, or at least a thread file, to get the final finish on your threads anyway, so minor pitch errors will be moot.

[John HaineParticipant @johnhaine32865]

Martyn, if you could be more specific with your question we could perhaps comment.

[Author]

Posts