Why so many ways to align a tailstock?

Got a project coming up involving turning between centers and want to be sure my tailstock is dead nuts on. Poking around Youtube I see all kinds of ways to do it, including:


Turning a couple collars and measuring them.

Making an adapter to indicate the O/D of the tailstock quill.

Turing a piece that same O/D as the quill and indicating.

Using a test bar between centers.


Assuming my bed isn't badly worn if I just stick a test indicator in the spindle and indicate the inside of the tailstock quill to zero runout won't that align it?

Why all the drama?

You don't need to care unless this is a new (straight, unworn bed) machine.

Take your stock, throw it in the machine, take a light cut, measure along the length, adjust if needed, take another light cut. Repeat if needed.

Simply measuring runout between headstock and tailstock does nothing unless the bed is dead straight AND the headstock is axially aligned to the bed. If they're not, any adjustment you make to the tailstock will be incorrect.

With a worn bed you'd still likely to have some barrel or other error along the length, so you have to decide what's the best compromise. Frankly, I've taken cylinders, gotten them as close as possible to size, then used careful sanding to bring in high spots (with care not to get caught in the machine!).

With the part between centers, you can place an indicatior at the center line and zero-zero on both ends. You need to find the mean of run-out to do this.

You can turn the part just enough to turn halfway around clean up, and that will be the part ends cleaner, thn lne up your tail adjustment. You can touch your tool bit to a shim against the part with looking at your cross feed dial for a match.

Likely these methods will get you to .001 or so.

Like Milland implied the part is king, not the machine if the machine has some wear.

For fine grinding between centers with no stock to take, I would first have to correct the part centers before grinding the part.

Probably lots of different ways for two reasons: because there are lots of people giving advice that don't really understand the geometry, and because there is more than one way to skin a cat.

On a lathe in very good condition you can sweep the tailstock barrel/spindle with an indicator (tailstock clamped down tight, spindle clamp on) on a mag base stuck to the chuck face to check for height alignment but you need to be aware of indicator sag due to gravity. If your setup isn't very stiff it can be significant. Then the T/S spindle can be checked for pointing up, down or sideways with carriage travel.

On a heavily worn lathe you don't want to use carriage travel because it won't be true to the tailstock ways. In that case you can extend the tailstock spindle and zero the indicator on the extended end of the tailstock spindle (again, T/S clamped down tight, T/S spindle locked) and then unclamp the tailstock and slide it toward the headstock, clamp it down and sweep again, note the difference in indicator reading, half of this is how much your tailstock is pointing in whichever direction is indicated across the distance between the points that the indicator was touching.

A quick easy way to test for sag amount is to chuck up a bar (preferably a piece of tubing or pipe) the same size (or close) as your T/S spindle and mount the indicator to the chuck face with your mag base. Register the indicator on the tube, then set and zero it at the top of rotation. Rotate the chuck to 90°, 180° and 270° and note the readings. If they deviate from zero, use the readings to compensate the readings you get when you sweep the tailstock spindle.

Note that even this method relies on the tailstock spindle O.D. being concentric with the I.D. taper - not always the case - so using a test bar would be better.

Neither of my lathes have a taper attachment so I do have to offset the TS from time to time. I don't think you can beat a test bar in my opinion. Turn two collars and adjust TS until they match. Put the test bar in your tool chest and drag it out when you need it. If you are still turning tapers after adjusting the TS with the test bar then something else is wonky.

On a related note, I was doing a few hours' work at a guy's shop the other day (about all the time I've got until I am done for the day - with my back issues) and my first time using his lathe I started drilling a hole - it didn't feel right straight from the start. I stuck an indicator on the chuck and swept the T/S - indicator was zeroed at the top and got ~+.050" at the bottom! Holy shit. Worst tailstock wear I've ever seen... .025" low and pointed down in the front by .010" over 6" length.

Thanks for the reply's, made me think some more.

I levelled and aligned the headstock to the ways a while back but got some small deviations both vertically and horizontally when I ran an indicator down a 12" test bar. I aligned the headstock so it zeroed out at the end.

Emco S11 - spindle alignment. - YouTube

Emco S11 - alignment 2. - YouTube

If I just align the t.s. with an indicator near the spindle (or out per Kretz), when I move the t.s. out 8" to make my part it could be off, no?

What if I turn a piece (the same length as the part I'm making) between centers and mic both ends? If they mic the same over that distance then my part shud be dead on, right?

But you can't turn a sample part for every length you might turn.
and the test bar may not be the same length you wish to turn.

often the bed washout may be at worse 12 " or so out and then at the far rear, the bed may be near perfect.

The tail nose fail/drop may be constant checked on a plate but exaggerated by the washout of the bed in places.

You may only be able to have perfect centers on a perfect lathe.

or you have a different extension of the quill.

Do the best that you can to align the tailstock, but don't be disappointed if you don't get a perfectly straight cut on a long piece. For the most part, a thou taper is typically good, and may or may not hold steady at that value. There will always be residual spring of the workpiece, plus, if the part is flimsy, and you need to use a steady rest, you can easily correct or force a bend into the part, which will give you a taper. Each part is more of a science experiment than many realize, and a well experienced machinist has learned all the tweaks as if by rote. CNC is only wonderful in the way that the 'experiment' is carried out in a very repeatable fashion at near optimum cutting speeds with tools in near mint condition (which stay that way because they are run near optimum speed with near zero dwell), so the tip suffers a very low rate of damage.

I seldom have to turn very long straight cylinders (thankfully) and my lathes are in good condition. But different material hardnesses affect the springback, so what is a good setting for a long piece of aluminum tube won't be the same as for turning a large bar of 4140. You'll have to learn to master the skim cut with a very sharp tool, or incremental cut adjustments along the way as you try to compensate for the spring.

In the long run, you will be filing and polishing the part to some degree, so equip yourself to do so. Air powered belt sanders can be had fairly cheap, and those are good for sizing and removing toolmarks. But, handling the sander is yet ANOTHER art to learn, depending on the grit, how fresh the belt is, how fast the part is revolving, etc, and how fast your lead is across the part, near the ends and near the shoulders. I do like finishing with about a 180 grit sanding belt, which when fresh, takes off a thou or two rather quickly, but then after that, it kind of settles into a more stable finisher. It may leave what you consider a suitable surface as is, or you can further shine it up with finer grits carefully held in a rig or by hand. But once you've sanded a uniform finish onto a fresh turned part, then it becomes a lot of work to remove all the material for a further sizable reduction. A full thou then seems like a mountain to remove by hand polishing.

Wear a dusk mask while sanding, or ventilate to the outdoors. We all know how important having a good set of lungs has become.

HuFlungDung said:

We all know how important having a good set of lungs has become.

Click to expand...

Ask Dolly Parton how important a good pair of lungs are!

(What a wonderful woman: Dolly Parton helped fund Moderna's Covid-19 vaccine research - CNN)

One other thing to consider when turning a long shaft is heat. Not everybody wants to use flood coolant so they may just apply lube with a brush or hand sprayer. As your part heats up you are going to turn a taper. I learned this years ago turning cast iron dry. Don't confuse this with the TS being off.

I use a precision cylinder mounted in head and center. Then indicate up and down length.
Second, I indicate the tailstock barrel. On my lathe I've been told that there is no adjustment.
But I don't follow that advice.

When you type that you indicate the inside of the tailstock quill.
You might get it right on. But then extend the ram and you are off because (looking down) the direction moved would nudge the center position a little.
Reason is the ram may not parallel. If there is no adjustment for that then there should be.

Terry --

Depending on what you mean by "inside of tailstock quill" you would either be 1) twisting the tailstock to be parallel to one side of the quill taper, or 2) setting the quill housing bore parallel to the ways.

The first is, perhaps self-evidently, silly. The second is only part of what you want. A theoretically perfect job of tailstock positioning would put the centerline of the tailstock quill COAXIAL with the headstock spindle axis. But in the real world of machinery -- often worn out after years or decades of use and abuse -- the most important aspect of tailstock alignment is to have the tip of the tailstock center in the vertical plane containing the spindle axis, at nearly the height of that axis. Certainly having the tailstock quill axis parallel with the spindle is nice, but having the workpiece center located correctly is more important.

John

John Garner said:

Terry --

Depending on what you mean by "inside of tailstock quill" you would either be 1) twisting the tailstock to be parallel to one side of the quill taper, or 2) setting the quill housing bore parallel to the ways.

The first is, perhaps self-evidently, silly. The second is only part of what you want. A theoretically perfect job of tailstock positioning would put the centerline of the tailstock quill COAXIAL with the headstock spindle axis. But in the real world of machinery -- often worn out after years or decades of use and abuse -- the most important aspect of tailstock alignment is to have the tip of the tailstock center in the vertical plane containing the spindle axis, at nearly the height of that axis. Certainly having the tailstock quill axis parallel with the spindle is nice, but having the workpiece center located correctly is more important.

John

Click to expand...

Ahh, I see what you mean.

Then my thinking should be right.

I rarely do any turning between centers, the reason I'm doing it for this job is I need to be very accurate. If I make a test bar the exact length as my project piece, lock the TS quill and don't move it, turn two collars between centers and adjust as necessary, my part should be good.

Thanks gents....

I think your first sentence has a lot packed in it. I was going to say much the same.

The idea of a perfect lathe with ways, head stock, and tail stock all completely parallel and the head and tail stocks the exact same distance above the ways is easy to imagine but difficult to achieve. And with four of the six degrees of freedom at issue here, there are many ways that a tail stock can go wrong.

Yes, I suspect that many do not understand the geometry very well. That is one reason for the many different methods. And yes, there are different ways that will work. Throw in the fact that all lathes are made of rubber and you have a real mess. A person must understand the geometry AND must be able to tell just what the problem really is before attempting to correct it. Otherwise he is shooting in the dark.

eKretz said:

Probably lots of different ways for two reasons: because there are lots of people giving advice that don't really understand the geometry, and because there is more than one way to skin a cat.

On a lathe in very good condition you can sweep the tailstock barrel/spindle with an indicator (tailstock clamped down tight, spindle clamp on) on a mag base stuck to the chuck face to check for height alignment but you need to be aware of indicator sag due to gravity. If your setup isn't very stiff it can be significant. Then the T/S spindle can be checked for pointing up, down or sideways with carriage travel.

On a heavily worn lathe you don't want to use carriage travel because it won't be true to the tailstock ways. In that case you can extend the tailstock spindle and zero the indicator on the extended end of the tailstock spindle (again, T/S clamped down tight, T/S spindle locked) and then unclamp the tailstock and slide it toward the headstock, clamp it down and sweep again, note the difference in indicator reading, half of this is how much your tailstock is pointing in whichever direction is indicated across the distance between the points that the indicator was touching.

A quick easy way to test for sag amount is to chuck up a bar (preferably a piece of tubing or pipe) the same size (or close) as your T/S spindle and mount the indicator to the chuck face with your mag base. Register the indicator on the tube, then set and zero it at the top of rotation. Rotate the chuck to 90°, 180° and 270° and note the readings. If they deviate from zero, use the readings to compensate the readings you get when you sweep the tailstock spindle.

Note that even this method relies on the tailstock spindle O.D. being concentric with the I.D. taper - not always the case - so using a test bar would be better.

Click to expand...

A tailstock setting is not a once and done thing. If you don't like the taper, adjust it. Don't fuss every time with indicating the tailstock barrel, or two collar tests, or other foolishness. Make a pass on your workpiece, measure, and adjust as necessary.

A useful technique is to indicate the back side of the workpiece. You should see the full taper of the workpiece. If not, part of the taper is probably spring in the workpiece.

An OD grinder is a machine made to far closer tolerances than a lathe. But I plan on adjusting the table so it grinds straight every time I change the tailstock position.

The point of checking the tailstock spindle in such a way is not to set the taper. It is to take corrective action if necessary. A hole can't be drilled or reamed using the tailstock without some pretty severe scuffing going on if the axis of the headstock spindle centerline and the axis of the tailstock spindle centerline aren't at least fairly close to inline. It is not meant to be done more than once, or at least not repeated until considerable wear occurs.

The axes being out of line with each other causes more problems than just that - those two being out of whack means a center won't seat correctly in a center hole also - potentially causing rigidity issues, runout or workpiece deflection. And also means that as mentioned earlier, differing tailstock spindle extensions produce differing amounts of taper when turning between centers.

If you are just worried about one turning job then make a couple spot cuts to check for taper and kick the center over to wherever you need it to get a straight cut.

when it really matters you want to look at the center hole. for smaller work i prefer the round variety. sometimes i make the hole with a countersink in the toolholder and single pointing would be best. sometimes i "cheat" on a small lathe and push the center drill to the side for a final touch up of the hole. then its about the surface finish of the hole but i dont mess with that, i feel a center hole grinder would be needed. i blow it out every time before installing.

This is a very good thread. The point made by a few of you that it's more "about the workpiece" rather than about the machine. Is an important consideration to not loose sight of. Something I have been guilty of in the past.