View Full Version : Lathe Turning unwanted taper

09-20-2007, 10:05 PM
Hey All,

I got a lathe from a friend who used it and since the move it is turning an quite unwanted taper. It is a 14x40 lathe. I moved into my shop and since then without using the tailstock - just the chuck to hold a piece of work it is turning about a .020 taper in 2". Any suggestions on what to try to fix this - or what could be causing it.


09-20-2007, 10:46 PM
So it was turning true before you moved it???

.020" in two inches is pretty far out. How much lathe experience do you have?

J Tiers
09-20-2007, 10:53 PM
Sounds very much like it is taking up the lost motion, or "backlash".

The 0.020 is really too much to explain by most any other means, unless the lathe now obviously looks like a pretzel.

To cure the backlash problem, be sure you always move the tool to cutting position by a motion which takes at least one turn of the positioning screw (crosslide or compound) in the direction that moves the point of the tool towards the work.

Another possible is if the machine has a taper attachment, and it is not secured right for non-taper turning. it could be set for a taper, or it could be loose and allowing the crosslide to move.

09-20-2007, 11:01 PM
Push off from slop in the crossfeed would not produce a taper. It would produce a slight taper or rounded starting cut and then start cutting a straight cut.

He needs to tell more about the lathe and his experience as a machinist.

This is another thread with little info expecting a good answer. Any answer given is all assumption and guess work. Don't get sucked into it. Require more information.

09-20-2007, 11:15 PM
If it is a run of the mill Chinese lathe there is the possibility that the headstock has been knocked out of alignment in the move. Try running an indicator across the face of the head stock spindle and see what the result is.

09-20-2007, 11:36 PM
Running the indicator across the face of the chuck using the crossfeed would be a better way.

09-20-2007, 11:40 PM
The DRO on the machine doesn't move one bit when feeding toward the chuck - which is the bigger end. I've done plenty of machining before - have an A.S. in machining from the local JC and am an ME. It does have a taper turning attachment and I've unlocked it so it just rides along. I've thought of temporarily combating the problem by setting the turning taper. I never thought of everything getting knocked out of alignment. It was moved very carefully and rolled to the current position. I'm not sure if it was producing a taper before, but knowing my friend - it wasn't. He's very meticulous. I think I'm giong to take a dial indicator across the face and then re-level the machine.


09-21-2007, 01:05 AM
He might be turning 1/2" hanging out 2"+ with no center taking .050" a pass.

He just dont know what he is doing at all.

Dont blame the tool, the problem is in the mirror 99% of the time. Iv run junk and made working parts.

I am from the school that never uses a taper attachment, its a way to screw thing up by a mile, fake all tapers and use a file. Iv run lathes where some idiot used the taper attachment and left the lathe all messed up with .200" backlash.

Im sure once you get a 14X40 lathe figured out you might see <.001" per inch or less taper.(on a good size part)

Almost all lathes cut some taper. I used to run a german Mouser lathe that would do less then .002" in 6 feet in a 6" diameter, I had to dial it in on roughting passes. you just need to learn how metal bends away from the tool and make up for it.

Just slow down and learn to use the machine and you will do fine.

09-21-2007, 02:35 AM
Ok Ed, run an indicator mounted on the cross slide across the face of the chuck and see if it stays on "0" or runs out and report back.

Your Old Dog
09-21-2007, 05:30 AM
Make sure you arent' trying to turn a coat hanger wire. Turning thick stock like maybe 1" thick and only 2 or 3" inches long can be done without a tailstock. Suggest lighter cuts, shorter stock sticking out of chuck or use of a tailstock. Good luck, have fun!

Ed Tipton
09-21-2007, 08:15 AM
If the workpiece is long enough, try using a slip bushing on the outboard side of the headstock. it's possible that the workpiece is simply being pushed over due to tool pressure. Without the tailstock there to hold the piece on the correct axis, the piece will move, resulting in a taper. Simply relying upon the chuck to hold the piece true is a recipe for failure. With the workpiece securely held in place, and running true, .020 would certainly be excessive in a two inch turning. Something else is happening.

09-21-2007, 08:21 AM
I'm turning graphite - 3" diameter and 3" long. It is pretty soft stuff so I'm not thinking it deflects. I'm turning them in .010 passes until I get to my final cut, which is .005".


J Tiers
09-21-2007, 08:24 AM
Push off from slop in the crossfeed would not produce a taper. It would produce a slight taper or rounded starting cut and then start cutting a straight cut.

Not an EVEN taper, but it WOULD be different at start and end....... taper shape depends on how stiff the crossfeed is....and the DOC.

Bigger at chuck end disposes of that anyhow, and also knocks out the idea of workpiece flex.

If DRO is not showing any issues then the cutter is moving straight. It appears that the OP knows something, and won't be in trouble from a real basic lack of understanding.

IIRC, a number of chinese lathes are not scraped-in and instead have screw adjustments for the headstock alinement. Some just have a bunch of beer can shims here and there.

Either way, the headstock alinement sounds like it has been disturbed, and may need to be re-adjusted.

09-21-2007, 08:32 AM
How about a gap bed lathe with the gap shifted out of alignment? When the carrige hits the gap and moves in or out? I have to align my gap with a dial indicator after removing the gap. JIM

09-21-2007, 06:04 PM
How was the lathe set up after it was moved? Was any attention paid to being sure it was aligned? The usual (but not the only) way to check a lathe is to put a sensitive level across the ways at the headstock end and the tailstock end and be sure the lathe bed isn't twisted.

If the lathe is one of those on which the headstock alignment can be adjusted relative to the lathe bed...that's another possible source of error.

09-21-2007, 06:48 PM
If a 14X40 lathe needs the headstock fixed and has that much error it has had a bad crash or two.

chuck up a big peice of hydraulic induction hardened shaft, dial it true then loosen the headstock bolts and tap it with a dead shot hammer and do it like you are dialing in a mill vice. if you get it within <.0005 in 12 inches you are good to go.

Its not rocket science.

09-21-2007, 09:04 PM
.020 in 2 inches is .2 in 20 inches- easily seen at the tailstock, and easily seen along the bed if it's twisted that much. First sight across the bed from a distance, say 6 feet, and align both way tops with your vision. If they look to be parallel, they won't be out by much, probably not enough to cause that much taper. Then mount a long bar, round and clean but not needed to be exactly straignt. Turn the chuck by hand and watch the outboard end wobble. Stop at a position where the wobble is either at the very top of the rotation, or the very bottom. Now sight from the top and compare the relative position of the bar with the center of the tailstock.

If the bed looked ok visually, the headstock is probably out of alignment. If the bed didn't look ok, then at least part of the solution would be obvious.

I dion't know if that is a gap bed lathe, but if it is, I can sure see a problem developing there during a move, especially if there's crud around the gaps, and maybe the gap is somewhat loose. I can see crud dropping into the cracks as the lathe is flexed during a move, and then the alignment would be affected afterwards. If there is this possibllity, maybe it would be a good idea to remove the gap and clean well, being careful to reposition the gap afterwards. I have not yet read anything which talks about the benefits of having a gap, except the one condition where a larger than normal workpiece must be turned. Everything else is a negative about the gap being there- and in fact if it looks like the gap hasn't been disturbed, it would probably be best to not remove it. Put that operation farther down on the list of things to look for as the source of the problem.

That's my take on it- look for the obvious first, like visually noticeable misalignment of the ways, then the alignment of the headstock axis with the ways, as described above.

There's still the possibility of foreign material between the carriage and the ways. One way to check for this is to remove the way wipers and note what visible gap might exist between the ways and the carriage. Run the carriage back and forth and see if that gap changes- if it does, the carriage could be going out of parallel with the ways. Because it's carrying the cutting tool, that could translate into a deviation from a straight cut. There are several errors that could occur if there isn't a smooth and consistent contact here.

And then there is bed wear- did the machine get a ton of use with the carriage moving over only a very small range- seems like for that much error a problem like that would be very obvious visually, so maybe placing a straightedge along the sides of the ways would show this up quickly. If you can't see any hollowing out, then even if there is some wear it wouldn't cause that much taper, and you could probably eliminate that from the list of possible problems.

09-21-2007, 11:33 PM
The DRO on the machine doesn't move one bit when feeding toward the chuck - which is the bigger end.
That is because the source of the error is independent of movement measured by the DRO.

J Tiers
09-22-2007, 12:27 AM
Hmmm--- a lot of the asian lathes are (or were) gap-bed. And they are known to fit as tight as a pencil in a water barrel.

That could cause a problem, but on the other hand, that would be obvious enough that the OP, who is clearly not a dummy, should have spotted or considered it.

That 0.020 per inch is 0.240 per foot....... if the lathe has a 4 foot bed, it would be twisted 1 inch if all this is legit, and a twist is the problem. Now THAT is a scary number, one which very obviously CANNOT be the problem. I think the bed would crack if you tried it.

However, it is far easier to believe that the headstock adjustment is off-position enough to put it out 0.020 per inch. We don't know that it was EVER RIGHT, after all, it could have been held by paint originally, with the adjusters loose. This IS a chinese lathe, remember.

09-22-2007, 02:02 PM
heh, held by paint and shimmed by grit, play taken up by cosmoline or whatever its substitute is, bushings lubed by humidity- scary stuff:)

09-22-2007, 06:45 PM
SO,, edward what have you figured out to be your problem??? JIM

09-24-2007, 09:57 PM
I've printed out the suggestions and put them on the lathe. I've not made it further as my real job has been having me work more than I'd like. I'm planning on getting it all sorted out this week as I've got some spare time.


I put my dial indicator to the face of the chuck and it was a little off. The chuck is 8" across and has a .030" difference across the face. So indeed they are misaligned. Now for the fun part :) Getting it all back to alignment :) I'll do some looking around but it seems as if I've got a few bolts to unbolt and then a couple screws to go in/out to align it.


Spin Doctor
09-25-2007, 01:17 PM
Often one will see on lathes the headstock has a machined vee to mate with the vee on the bed. If the head is off and the vees are mating the only way to fix that is to remachine or scrape. Rather than that type of set-up myself I would prefer to see a set of push/pull blocks with solid shims in the .125" range between the block and headstock. This is a standard practice on industrial machinery in which the hork head/s may be mounted either on a slide or even lathes and grinders. Makes aligning things a lot easier. Although it doesn't hurt to set up indicators to make sure your moves are what you think they are. Shims should be shims, but it does not work that way sometimes.

09-28-2007, 11:49 PM
Hey guys,

Tonight I undid the bolts that hold down the headstock and adjusted the two push/pull screws. I've got the chuck to read .003 difference in 8" diameter. Much better than the .030 in 8" that I was seeing. I can't get it any further because the holes that the bolts go in have nearly nil clearance. Would it be okay to make them slightly more oversized (1/32 larger than the bolt) so that I can dial it in to nearly perfect?


09-29-2007, 02:16 AM
Yes, drill out the bolt holes to give you the needed room, or file them in the proper direction.
If you've gone that far, then you should be aware that the axis of the headstock and tailstock should be the same. Can the headstock be moved forwards/backwards with the adjusting screws? If it can, then it may go out of line with the tailstock. The tailstock may need to be adjusted forwards or backwards, tilted left to right, moved up or down, or rotated somewhat on its base.

Once you get the headstock aligned to get less than one brazillianth of an inch taper in 8 feet ( :) ) don't stop there. Lightly pinching a flat disc between centers gives a pretty good indication of the relative positions of the center points. Run the tailstock ram nearly fully in, check with the disc, then run it nearly fully out and check again. I wouldn't be surprised if you find that the point rises or falls as the ram is extended, and if it goes north/south also. If its axis is actually parallel to the spindle axis, it could still be high or low, or forward or backward. When you can pinch a disc between centers and it doesn't tilt, and can do this over the extension range of the tailstock ram, then it's good to go. It's still possible for the tailstock to be out as it's moved away from the headstock, so it might be prudent to do a test bar between centers. You might have to re-adjust the forward/backwards in order to eliminate a taper when turning between centers.

Obviously, you'll want to have the ways clean and the play in the ram lightly snugged out throughout the test and adjust procedure.

09-29-2007, 09:18 AM
Before drilling out any bolt holes, remove the chuck, check for any burrs, swarf, or any reason for a poor chuck mount. Then run a dial test indicator on spindle registor to see if there is any runout in spindle. Just a thought. JIM

J Tiers
09-29-2007, 09:45 AM
I suggest using the DTI on the internal taper instead for spindle checking.....

The "register" (on threaded spindles only, and probably not applicable here) is not always finished to the same standard as the ID taper, and in any case is often not contacted by chuck mounts.

The ID taper is always contacted by centers and collet closers. And it is used when you really want to be sure the work is on center. Chucks may be a few thou out (aside from "split chucks". AKA collets). So it makes sense to check on the ID taper for concentricity.