My lathe is turning a taper

[T Bredehoft]

center drilled both ends, hold it with dead center and live center

That certainly will get you a lot closer than you are now. Go for it.

 

[Winegrower]

I just went through this with the Takisawa. I found that aligning centers in headstock and tailstock was just not accurate enough. I made a test bar, and that got me down to a few 001’s. I happened, by accident, to obtain a 15” precision test bar, cut a headstock center in place to get as exactly centered as possible, and using that I could see that vertically the tailstock/headstock were right on. With the test bar, I could just run an indicator along the (stationary) bar and precisely move the tailstock down to less than a half thou.

I would have fussed for a long time without the test bar.

 

[RJSakowski]

.0075 would be hard to determine by aligning centers by eye. For aligning the tailstock, I use a modification of the RDM method.

If you mount a round bar between centers and measure runout with a test indicator or dial indicator, you can precisely align the tailstock. You measure the maximum and minimum reading at the headstock and take the average. Then, without changing the indicator, compound, or cross feed, you move the carriage to the tailstock and do the same at the tailstock. If the two averages are equal, the tailstock is aligned. If not, the difference is half the taper.

The only requirement is that the bar be truly round and of equal diameter at either end. The diameters can be different but that adds a correction for the difference in diameters to the mix. Note that this will also take into account and correct for any misalignment of the ways.

 

[RJSakowski]

One issue that I have with the above method is that the alignment is only valid at the test position for the tailstock and the particular extension of the tailstock quill. If the bed is twisted so that the lathe turns a taper on an unsupported bar, moving the tailstock will change the alignment. Similarly, if the tailstock quill axis is not parallel with the spindle axis, moving the quill in or out will change the alignment.

Mr. Pete did a You Tube video where he made a test tool for the Morse taper in the tailstock where the end of the adapter was turned to a cylinder. Using a suitable sleeve for the headstock and a test indicator, the indicator was zeroed on the surface of the cylindrical portion. Then the adapter was moved to the tailstock, the carriage moved, and the measurement repeated. Any difference would be half the amount of taper at tha position for the tailstock and quill extension.

This is essentially the same as above described method, the difference being that the procedure no longer is limited by a fixed length test bar and the tailstock and quill can be set at, or approximately at, the position used in turning the work. A potential source of error would be if the cylinder were not perfectly concentric with the Morse taper but this can easily be checked. If there was an eccentricity, the process could still be used by placing a mark on the adapter and measuring at the mark in either position. The tool can easily be made by drilling a socket for press fit cylindrical plug in aN MT center or starting with an MT to JT adapter and turning a cylinder to true concentricity.

 

[MrWhoopee]

I took off the chuck, put a dead center in and live center on the tailstock, aligned the tips by eye



the tailstock side is smaller, and gets larger as I get closer to the headstock.

Also, could a live center cause it by not being ridgid enough? I got a cheap live center from eBay for $35

I'll align the headstock to the tailstock again later, maybe it's shifted from the last time I did it.

The taper doesn't matter much to the project I'm doing, but it's annoying when the taper can be noticeable visually.

A cheap live center will give you chatter, might give taper that is larger at the tailstock end, but not smaller.
Aligning centers point to point by eye is not nearly close enough and does not give any indication of alignment when the tailstock is further down the ways.

I used the two collar method, here's shaft that I used. I made if from what I had lying around.



I got lucky. To my amazement, my Heavy 10L was within .0003 over 14 inches as received. I put my tools away and went out to buy a lottery ticket.

Don't fool around with this, get it right. It may not be critical now, but it will be.

 

[Cadillac]

One thing to think about is even if you get the tailstock aligned it’s gonna be challenging to cut a 1/2or 5/8 rod 14” long between centers without deflection of the part. Using a followrest would help a lot.

 

[Winegrower]

I would think the first step in all of this would be to ensure the headstock alignment with the ways is good, or else all efforts will fail. I used the longest straightest bar I could find and ran it through the spindle with a 4 jaw chuck, centered it at the left end of the spindle and zeroed runout at the chuck. Then, fortunately, running an indicator in the toolpost along the length of the rod (not supported by the tailstock, of course), I could see that the headstock was aligned parallel to the ways to few tenths.

It seems to me if you have leveled the lathe to eliminate twist, aligned the headstock with the ways, adjusted the tailstock offset and height appropriately, and are still cutting tapers, a new lathe is the next step.

 

[projectnut]

Here are a couple YouTube videos by Tubalcain (mrpete222). The first is on how to make a test bar and the second is the completion of the bar and how to use it.

 

[WarrenP]

Aligning the tailstock center with the spindle center will only show if it is close at that point. Plus its pretty hard to "see" a few thousands off by eye. You also need to align the them further out from the spindle to see if it is still lined up further out. A test bar seems to be the best way to find out.

 

[Pcmaker]

I got it to within .0005 within 15" or so using regular 1 3/16" 1018 cold rolled, 20" in length. I faced and center drilled both ends, held it between 2 dead centers.

I indicated along 15" of it, occasionally turning the round bar to make sure it's somewhat concentric. I ran the indicator back and forth, moving the tailstock until I had less than .0005 of runout. It may not be a perfect set up to align the lathe, but I bet it's a heck of a lot better than what it was before.

It's always confusing when you're moving the tailstock to the direction you THINK would lessen the difference between both ends, but it goes the opposite way... until you get to the point where the needle goes the way you want it to go. I've never figured out how it does this. It also happens when I'm squaring up the mill vise.