No announcement yet.

Leadscrew Imprinting on Lathe.

  • Filter
  • Time
  • Show
Clear All
new posts

  • Leadscrew Imprinting on Lathe.

    In another thread ( someone mentioned their lathe imprints the leadscrew into the workpiece. I commented about this in that thread but don't wish to hijack it so I'll ask here...

    The issues is that a workpiece ends-up having an irregular surface which is the imprint of the leadscrew -obviously caused by a variety of anomolies in the carriage drivetrain. In my case, the lathe is very old and worn. It's been corrected as best I can and I don't plan to spend more time on it.

    The questions are, how common is this and is it to be expected to some extent?

    I'm shopping around for a new lathe and wish to know if certain types/designs are more prone to this than others. At the moment, I have an Atlas TH42 -which I've long outgrown. Hoping that my next lathe will be the last for a very long time.



  • #2
    By imprinting you are telling us you can see the "flaws" in the finish that follow the lead screw thread?

    If that's the case, look at simple things like tighten the gibs, make sure there is adequate lubrication on the bed/carriage. You also said it's been used quite well (worn), I would look at the half nuts or lead screw to see if new ones would be worth the trouble to install. Not too expensive or difficult to replace. And that should improve the "flaws" imprinted on the finished piece.

    A larger lathe is gonna be expensive, so unless you have a couple thousand to spend, it might be easier to rebuild the one you have.

    I haven't always been a nurse.........


    • #3
      Your going to get that imperfection with any lathe whether it's a lead screw that moves the carriage or a rack and pinion. It is the loading and unloading of the cutter as it progresses along the work. The heavier and deeper the cut the less it will show. When you take light finish cuts all kinds of issues show up.

      To correct that use a radius on the cutter tip or finish over size and file to size or use a shear cutting tool.
      It's only ink and paper


      • #4
        I had the same problem with my lathe.
        My problem was was that cuts little ridges in the work piece.
        Sad to say it was just cus my lathe is of bad design, but well it is fixed now.
        I doubt that this is your problem but here is what I did.
        My lathe is a china POS called a HQ500 check www. I think this type of lathe is called a smithy in the US.
        The lathe has a Auto feed for the carriage and for the cross slide.
        The way it does the cross slide auto feed is threw having a wide grove on the leadscrew that runs in a bevel gear that transfers the power to the cross slide.
        The problem is the sliding through the bevel gear caused drag on the carriage and made it slow down everytime the half nut gets to the grove in the lead screw.
        I got sick of it and took the bevel gears out and problem solved, no more auto cross slide but no more marks in the work piece either.
        It now cuts a perfectly smooth surface and feeds a heck of alot easier.

        If you are using violence and it does not work, You are not using enough or it is upside down.
        You can always just EDM it...


        • #5
          What does these flaws look like?

          Are you sure its not gear noise? Patterns can show up from gears, more so if they're not a well fit mesh.


          • #6
            On better lathes the leadscrew does not do the driving for sliding cuts. Only the cross slide screw comes into play for surfacing work

            The leadscrew is only engaged for screwcutting.

            A decent quality machine ( albeit it may have a few years under its belt) will probably be available for less than a far eastern import.


            • #7
              Here's an example from my lathe. I'm pretty sure this is what he's referring to:

              Finish with small DOC and slow feed rate on power-feed

              Here I took the same tiny DOC but fed by hand using the carriage handwheel and cut the first half of the part again for a side-by side.

              The little ripples generally represent about 0.0002 to 0.0005 of waviness in the surface of the part. This generally has meant that I have to finish close fitting shanks and bores by hand so that while the surface finish may suffer at least the dimension is stable over the whole part.

              Deep cuts don't outwardly show it but if you hit it with sandpaper it will show up. I know the leadscrew is somehow involved because if I put the gears in neutral and turn the screw with a cordless drill it will show the exact same thing just not with the regular timing of being linked to the spindle. Grizzly replaced the screw for me and while it got better it is still there.

              Most folks showing their work after a power-feed finish pass don't end up with a raccoon's tail so I don't think it's just a fact of life on all lathes.


              • #8
                I take it this lathe doesn't have a feed screws...


                • #9
                  I was under the impression this was caused by an alignment problem with the leadscrew/halfnuts.

                  Pop a dial indicator on the leadscrew and detect if closing the halfnuts causes the leadscrew to flex a bit.

                  If so, the leadscrew is becoming a spring attempting to push your cutter one way or another as it goes along.


                  • #10
                    If you care to look at post #12 of the following thread ( it outlines the steps I've taken to correct/mitigate it.

                    On my lathe (a very basic lathe indeed) there's only one leadscrew which drives the carriage and the crossfeed. I gather nicer, more recent lathes have two leadscrews; one to drive the carriage and one for the crossfeed. The photos posted by 'photomankc' above are similar to what I experienced but my symptoms were much worse. The salient issue is that the "imprint" showed exactly the same pattern as the thread spacing of the leadscrew.

                    In any event, it seems that anything that interrupts smooth travel of the carriage will result in "flaws' in the workpiece -and that some lathes are less prone to this than others. -And that pretty much satisfied my curiosity.

                    -Much appreciate everyone's time and help...


                    PS: I'd like to create another thread (later today perhaps) to ask questions about different gearing mechanisms on lathes. After a year of learning the ins/outs of my Atlas TH 42, I'm honing in on an upgrade...


                    • #11
                      leadscrew imprinting

                      If I understand the situation correctly, it is a bit misleading to say it imprints the shape of the leadscrew on the work. Apparently, the problem stems from having a leadscrew which is rotating off center or has a bend in it so that it effectively is off center. The carriage, which is loose or insufficiently rigid, then deflects in and out or up and down like a jalopy changing the depth of cut producing a slight pattern on the work which happens to have the same pitch as the lead screw. In and out motion of the tool will have a greater affect than up and down (unless, perhaps, up motion puts the tool above centerline so it digs in, lower the tool slightly). This would still happen if you wiped all the threads off the lead screw, other than that you would no longer have the longitudinal motion. This surface finish irregularity is a function of the pitch of the lead screw, and probably sinusoidal in shape, but does not mimic the thread form on the lead screw; if your lead screw threads were V shaped, acme, Whitworth, buttress, or sinusoidal it would not make a significant difference.
                      The depth of the surface finish would be expected to vary along the length of the work (position of the carriage). If the lead screw is bent, rather than off axis, this will contribute to it being more severe in the center of the travel but this competes with another effect which is that whether the leadscrew is bent or off axis, it is more flexible in the center of its length and thus should have less effect there. If the looseness of your carriage varies along the length (i.e. ways are worn or not uniform as shipped) this could also affect the depth of the surface finish irregularity along the length.

                      If your leadscrew was perfectly straight and concentric and the bearings were concentric, this would not happen. If your half nuts were free to float up and down and in and out, this would not happen. If your gibs were perfectly tight this would not happen, though it may not take much force to produce a very small surface irregularity. If your power feed used a separate shaft and the lead screw was disengaged during feeding, then this would not happen unless the power feed shaft was off center or bent by more than the play in the follower in which case it would happen; indeed the power feed shaft would make the same pattern reminiscent of a lead screw (possibly with a different pitch, depending on power feed gearing) even though the power feed shaft has no threads. Again showing that it is a surface finish error with a pitch that is equal to the lead screw pitch or related to the power feed gear ratios and rack pitch and a magnitude which is partially a function of the ecentricity vs longitudinal position of the leadscrew or power feed shaft and not an actual imprint of the lead screw shape.

                      One thing that can make the effect non-sinusoidal is if you are taking a finish pass where the depth of cut is less than the deflection due to the leadscrew eccentricity. This will give a clipped sinusoidal pattern which is more noticeable. From this it follows that, for aesthetics, you would need to avoid "cutting air" type finishing passes and go a little deeper on your final finishing passes. If you need to erase the memory of the raw stock shape (i.e. tool deflection which varies depending on how much work was removed distorting the final shape), try taking multiple finish passes at, for example 3,2, and 1 mils from final diameter (assuming <1 mil surface finish waviness) instead of taking one real finishing pass followed by cutting air. Then you will probably end up with a more pleasing sinusoidal surface finish rather than a clipped sinusoidal surface finish. Which surface finish is better depends on the application. If you have a sliding o-ring seal, the sinusoidal is better. If you have a sliding bushing, the larger and flatter bearing surface, interrupted by "oil groves" would be superior. If it is a rotating bearing seat, the clipped sinusoidal is probably better. If you are concerned about appearance, the sinusoidal is likely better.

                      There also could be a slight effect due to effective pitch errors along the length of the lead screw or a similar effect from angular velocity effects of the play in the power feed shaft follower vs the eccentricity of the power feed shaft causing the longitudinal travel of the cutter to be non-linear changing the width of each cut and therefore affecting the workpiece and tool deflection. A related effect would be if any of the change gears or gearbox gears were badly off center; in particular the gear on the leadscrew would have an effect that matched the pitch of the lead screw. These effects I would expect to be somewhat smaller but unlike the others can't be eliminated by perfectly tight gibs as the motion error happens in the direction the carriage is free to travel.

                      Make sure the carriage is riding on the ways and not on the lead screw. If the lead screw mounts are displaced up, down, in, or out such that they lift, pull, or push the carriage off the ways it will not only interfere with tightening the gibs but the carriage will be riding on the lead screw (and its ecentricity) and following that instead of the (hopefully) flatter or smoother and much more rigid ways. Chatter will be a lot worse, too. Lead screws tend to make lousy ways. In some cases, engaging the half nuts too far could cause this, especially on lathes where the half nuts are indeed half nuts and not two half nuts. Conversely, even if this isn't the problem, more loosely engaging the half nuts may reduce the effect by substituting some longitudinal position non-linearity for carriage deflection.

                      Obvviously, you want to attack the core problems as much as possible. But workarounds may still prove useful for getting work done in the short term and for dealing with the flaws you can't fix.

                      One of the things about machine tools is that any imperfection in their manufacture or due to wear or improper adjustment or lack of rigidity will tend to show up in the workpiece in some way. Lack of flatness, lack of parallelism, lack of squareness, lack of concentricity, periodic errors, lack of rigidity, lack of balance, lack of alignment, play, and operator pressure on the controls, all show up somewhere to some degree. When machine tools are built to, or have worn to, typical woodworker tolerances you can expect some problems.


                      • #12
                        Sorry for starting any confusion, I never said it was imprinting 'threads' into my finish in my other post OP's referring to. I said it was imprinting a pattern that matched the threads, I can see that being confusing though. On my machine I have clamped the gibs on the carriage as tight as they will allow me to. It reduced the effect but did not eliminate it. So far I can't ascertain what is causing the deflection though I am considering taking the carriage apart to look at it some more again. Light engagement on the leadscrew does also help but the only way to obtain a smooth shaft or bore is to hand feed to finish. The depth of the features I get is very modest but really sucks on something like a shank on a custom tool-holder meant to be held in a collet. You want every sq cm of contact you can get, not raccoon tails.

                        I do take your point on perhaps the carriage being forced up off the ways a bit when engaged. I'll have to look at that. It only really chaps my behind on boring. HSS boring at 150 to 300RPM and hand turning the carriage wheel just sucks. I can see my daughter growing while I wait. With carbide tools it's less of a hardship to finish off the final couple of passes by hand where the surface really matters.
                        Last edited by photomankc; 06-16-2011, 06:11 PM.


                        • #13
                          You need to put some indicators on your leadscrew and figure it out if's bent (indicator moves as the leadscrew rotates) or misaligned (leadscrew at different positions along it's length) or if it gets bent when you close the halfnuts (location of halfnuts not correct relative to position of leadscrew).

                          Except for the leadscrew being bent (which is hopefully the least likely) these issues should be easy to resolve.


                          • #14
                            The lead screw shouldn't cause a problem unless the wobble is obvious. If running true position the carriage about midway, loosen the half nut mount bolts, close them, retighten the mount bolts and see if that had an effect.