Lathe cross slide slack

[LittleWatchShop]

I have a cross slide that I have cleaned up pretty well.  Operation is smooth.  However, there is "slack" in all of the controls.  Perhaps several hundred microns.  Looking at the design, there does not appear to me that there was any design feature to null this out.  I have other precision microscope slides that have a feature to eliminate any slack.  My guess is that the design is such that you only expect calibrated movement when going in just one direction.

Any thoughts on this?

Am I the only one who has noticed?

BTW, at first when I got it working, I did not appreciate its purpose.  Now I recognize one very important purpose is truing a metal rod that is out of round.  This seems very hard by hand with a graver, but quite easy using the cross slide.

[nickelsilver]

That's normal, even a new slide might have 2-3 hundredths of play in the screw. Most small slides have an adjustment for the "bearing", usually a cylindrical portion of the screw near the crank which is captured in the slide mount with a nut, this can be brought until there is near zero play and the screw is still free. Often some of the play that's perceived as between the nut and screw is actually there.

 

More modern slides, like later Levin and larger ones will have preloaded ball bearings here, smooth action with no play.

[LittleWatchShop]

Thanks @nickelsilver.  I looked at my slide carefully and there is no way to null out the slack.  I am showing a picture of a part from a microscope slide (part of an electronic probe station where microns count!) showing how the slack is eliminated.

[nickelsilver]

Correct, there is no provision to take up play between the screw and nut. So you always approach from the same direction, taking up the slack before. With a good slide, with hundredths of mm of slack, no problem working to microns.

[LittleWatchShop]

12 minutes ago, nickelsilver said:

Correct, there is no provision to take up play between the screw and nut. So you always approach from the same direction, taking up the slack before. With a good slide, with hundredths of mm of slack, no problem working to microns.

Thanks a bunch...this make me feel better!!

[Michael1962]

The correct term for what you are describing is called backlash.

When I did my apprenticeship, I was fortunate to work on some of the more high end lathes in the workshop. All of them had backlash. Some more than others.

Mill had backlash. Horizontal borer had backlash.

I am not aware of watchmakers lathes and the cross slides as to whether you could move the measurement scale on the slide after you have taken up the backlash. We could do that on the larger machines.

I am not aware of newer machines these days, but in the eighties, if you wanted to get a machine that had backlash adjustment capability, break out your cheque book.

[jdm]

17 hours ago, nickelsilver said:

Correct, there is no provision to take up play between the screw and nut. So you always approach from the same direction, taking up the slack before. With a good slide, with hundredths of mm of slack, no problem working to microns.

For the OP or anyone that has never done or is just at the beginning with lathe or milling work. 

The approach above is the way to follow if you have only the wheel dials to rely on, possibly of the non-zearoable type to add pain. But as soon you back off "the controls", voluntarily or not, then returning precisely to the same point just "taking up the slack" does not work anymore. In other words, using dials on the wheels you can't measure distance in both directions, and kind of fastidiously in one direction only.

The solution is simply to use an instrument (dial indicator) which tells where your tool holder or workpiece is, on whatever axis you apply it. It does't know or care about backlash, nut play, double nuts, ball bearings, or anything else.  It saves you from remembering how much a division on the wheel is, if it doubles for diameter or not, or having to convert from metric to inches. You can get one including mag base and arm for about USD20. 

Truth is however,  real watchmaking work on a watchmaker's lathe do not use dials, indicators and not even slides or wheels mostly. Just gravers, handfeeling, loupe and micrometer. But as my mentor says, better start learning on the large and get to the smaller gradually. 

[LittleWatchShop]

10 hours ago, jdm said:

For the OP or anyone that has never done or is just at the beginning with lathe or milling work. 

The approach above is the way to follow if you have only the wheel dials to rely on, possibly of the non-zearoable type to add pain. But as soon you back off "the controls", voluntarily or not, then returning precisely to the same point just "taking up the slack" does not work anymore. In other words, using dials on the wheels you can't measure distance in both directions, and kind of fastidiously in one direction only.

The solution is simply to use an instrument (dial indicator) which tells where your tool holder or workpiece is, on whatever axis you apply it. It does't know or care about backlash, nut play, double nuts, ball bearings, or anything else.  It saves you from remembering how much a division on the wheel is, if it doubles for diameter or not, or having to convert from metric to inches. You can get one including mag base and arm for about USD20. 

Truth is however,  real watchmaking work on a watchmaker's lathe do not use dials, indicators and not even slides or wheels mostly. Just gravers, handfeeling, loupe and micrometer. But as my mentor says, better start learning on the large and get to the smaller gradually. 

I ordered this setup in addition to a runout indicator gauge.  Just for kicks.