chasing accuracy and trouble shooting unwanted taper on manual lathe

stoneaxe · Jan 11, 2020

I am making a toolholder for a mill, on the lathe. ER 25.

The 5" long part has a .875" shank with a 1.65" dia. x 2" long end for the ER collet pocket.

The end of the shank is gripped in a four jaw, has -.0001" TIR at the chuck.

The other end, that will hold the endmill, is held in a live center in tailstock.

This 1.65" diameter is turned round to - .0001" TIR. , but showing a .0003" taper over 1" of travel on the 2" long cylinder.

The large end of the taper is the tailstock end.

The live center is so-so (crap), having .0005" runout on the taper.

Is the center the problem, or is the tailstock out of alignment?

I am inclined to believe it is the tailstock, because if it were the center, that end should be smaller diameter, yes?

Milland · Jan 11, 2020

Could also be bed twist or general Z way wear, but if it's uniform you should be able to "tune it" by offsetting the tailstock.

Are you sure you don't want to try a new live center? If yours is that far out I'd want to set it aside, especially when you trying to make tool holders.

eKretz · Jan 11, 2020

This isn't even a problem. Just adjust the tailstock center. Taper direction isn't going to tell you much of anything, as the tailstock can move. Most live center issues manifest as runout or roundness error, not taper. I would ditch that center if its .0005" out and just turn in the chuck. If that won't fly due to taper, try a dead center and lower RPM with some good center lubricant. Make a plug if you need to.

john.k · Jan 11, 2020

Tailstock height not being centre height is likely your problem.....this cant be adjusted with a screw.......and cant be measured with a wobbly dial indicator on a stand either.......spindle axis not parallel to bed is also a possible.

stoneaxe · Jan 11, 2020

eKretz said:
This isn't even a problem. Just adjust the tailstock center. Taper direction isn't going to tell you much of anything, as the tailstock can move. Most live center issues manifest as runout or roundness error, not taper. I would ditch that center if its .0005" out and just turn in the chuck. If that won't fly due to taper, try a dead center and lower RPM with some good center lubricant. Make a plug if you need to.

That is a good plan. I will get the tailstock on line and use a dead center for now, till I get a better live center.

eKretz · Jan 11, 2020

I doubt that it has anything to do with center height. By my calculations, a .0003" change in diameter with the tool on center would require around an .016" change in relative height from start of cut to finish.

If the diameter is regular in it's taper (⅓ of the way measures smaller by .0001", ⅔ of the way measures smaller by .0002" and the end measures .0003" smaller) tailstock offset will make it all good, as Milland mentioned.

john.k · Jan 12, 2020

I agree with 90% here....the taper can be eliminated by adjustment of the tailstock.Called cut and try....just be sure you dont overcut ,and need the putting back on tool.

steve-l · Jan 12, 2020

A cut taper on a lathe can be caused by a myriad of problems. These problems, if present, require different solutions. So it is very important to know what the culprit is first before doing anything. You can of course, just adjust the tail stock to compensate or you can be professional and isolate the real issue with a spindle test bar...........your call.

eKretz · Jan 12, 2020

The "real issue" may simply be a tailstock offset. Finish the job, then worry about starting the formal investigation. We're talking about a .0003" taper here, give me a break with the "professional." A lot of places would barely even notice that. If it wasn't for the fact that he's trying to make a toolholder and minimize TIR I'd have told him to ignore it.

Tyrone Shoelaces · Jan 12, 2020

eKretz said:
The "real issue" may simply be a tailstock offset. Finish the job, then worry about starting the formal investigation. We're talking about a .0003" taper here, give me a break with the "professional." A lot of places would barely even notice that. If it wasn't for the fact that he's trying to make a toolholder and minimize TIR I'd have told him to ignore it.

Most blokes I worked with would cure that taper with a length of emery cloth. Just saying.

Regards Tyrone.

fciron · Jan 12, 2020

As a user of old, clapped out manual lathes, I regularly mess with the levelling to remove taper. Bed wear near the chuck means I can move the "level" part of the bed to where it's most needed.

My South Bend is particularly handy for this: There is a leveling screw. In the tail stock foot that makes leveling simple. I imagine I'll be messing with the offset more since my new-to-me Sheldon lacks this feature and I can't see the dial indicator from under the cabinet.

Also, you can induce error with the chuck/center combination if you remove and replace the part. Ideally, you'd chuck the part once, center drill, then bring in the center and do everything without removing the part from the chuck. The final part should then be true to the lathe axis regardless of slight misalignments. (This won't prevent cutting a taper if your saddle travel its not parallel to the headstock axis, so all previous tail stock alignment still applies. I should probably stop talking before I get in over my head.)

stoneaxe · Jan 12, 2020

Some basics- the lathe is a pretty decent Taiwan 1340, in excellent condition. Not great, but not a total POS. About like a Sharp 1340.

The ways are not worn, and I leveled it a few years ago.....! Better do that again!

I am relying on a straight cylinder cut, as the part is coming off the center and will be held in the chuck for boring the collet recess, and I need a straight part to indicate alignment.

Pulling the center, and putting the free end on a center rest seems iffy, as there is little being held in the chuck (7/8" deep" x 7/8" dia, with aluminum pads, on the external drawbar threads, in a four jaw) and the drill-bore is substantial- the part could be pushed back into the chuck. It could be done, for sure, with care- is this a better option?

I suspect it is what many have mentioned, a slight tailstock misalignment- so slight it may have passed factory inspection.

I am striving to make parts as accurate as possible, machining has a steep learning curve and being rigorous about errors and sloppy work is the only way I know to improve. Every time I think I am getting better, I learn how to measure more accurately..... :scratchchin:

Thank you all for helping me- this is a wonderful forum!

gbent · Jan 12, 2020

Tyrone Shoelaces said:
Most blokes I worked with would cure that taper with a length of emery cloth. Just saying.

Regards Tyrone.

What Tyrone said. Its a lathe, not an od grinder. You can chase your tailstock offset from now to next Sunday, but the next time the tailstock is in a slightly different position, or the spindle has a different amount of extension, or you rotate the center before you seat it the taper will be different.

eKretz · Jan 12, 2020

When you are turning free in the chuck you should have a stop behind your part to prevent it from being pushed back into the chuck. You can run it in a steady just fine as long as you have it aligned well. Sweep the tailstock with an indicator mounted to your workpiece to get ballpark very quickly. Bore a test cut or skin the O.D. and correct from there.

You can get the tailstock set to cut dead nuts this time, but as mentioned it may well be a bit off for next time. If you have even a little more torque on one of the offset screws than the other the tailstock will gradually work its way off center as you clamp and unclamp the tailstock also. I like to adjust the tailstock and try to leave the offset adjusting screws with only something like the weight of my hand on the wrench worth of torque, they stay put pretty well like this. Chasing it is fine for this one-off job if you want to as you are making a tool. But...

If it were me doing this job as a one-off I'd probably do what Tyrone suggested and just straighten it out with a fine file and some emery then scotch-brite and a little oil. Probably about a minute's worth of work all told. You could make a lap and lap it but that's a LOT more work. I like a file rather than just emery, it keeps things a little straighter. Use the emery under the file too, not with your fingers.

stoneaxe · Jan 12, 2020

Having set the job up in the four jaw and a center, I was surprised to see .005 runout in the free end, after I pulled the center out. Stuck the dead center in there, and the runout was a .001 or so.

That was when I got disgusted with pushing a rope, pulled the part out and stuck it the mill, where it is going to get turned, drilled, and bored. That is where it has to be concentric, and that is where it is going to get made! (still have to put it on the lathe to turn the threads, though).

eKretz · Jan 12, 2020

Yeah you need to true it before you put the center in... set it up in the chuck only gripping a short bit or using a copper wire around the gripped area and tap the tail end true. Dial in the chuck and check the tail end again, tap true if necessary. Back and forth until you're dead set. Then put the center in. Accurate work requires you to work accurately...

stoneaxe · Jan 12, 2020

eKretz said:
Yeah you need to true it before you put the center in... set it up in the chuck only gripping a short bit or using a copper wire around the gripped area and tap the tail end true. Dial in the chuck and check the tail end again, tap true if necessary. Back and forth until you're dead set. Then put the center in. Accurate work requires you to work accurately...

^^Yep- a really good machinist suggested that to me some time ago, and I forgot.

Plan B- in the Mill- so far so good- put the toolholder in the mill, (it already had the taper, keyway, drawbar threads and dog slots in it.) Then drilled, bored to diameter and bored the ER taper. I did this by pulling the compound off the lathe and bolting it to the mill table, with the mill head in horizontal position.

Used the sine of 8 degrees, to get the angle right. (hopefully) One DI to measure distance traveled (1"), and another to pickup angular displacement (the sine). The ER collet dropped in with no shakiness...

Problem is, with a hand feed, the bore is a bit rough. So I took a 1 1/4" bar stock, put it in an endmill holder on the mill, and used the same setup to make a male lap. Blued out great, we will see- I will try some of the non-embedding lapping compound on a slow rpm on the mill. Wish I had a real 8 degree plug gauge!

jim rozen · Jan 13, 2020

stoneaxe said:
Having set the job up in the four jaw and a center, I was surprised to see .005 runout in the free end, after I pulled the center out.

Old time machinist would say of course you see that. You should be turning the part between centers. The practice of holding a part in both
a chuck and a center is best avoided, you wind up with all sorts of issues like this.

fciron · Jan 13, 2020

Glad you got it all worked out.

For future work, like you might want to examine your order of operations to avoid doing the most critical operations while gripping a short, threaded section in the chuck.

chasing accuracy and trouble shooting unwanted taper on manual lathe

Stainless

Diamond

Diamond; Mod Squad

Diamond

Stainless

Diamond; Mod Squad

Diamond

Titanium

Diamond; Mod Squad

Diamond

Stainless

Stainless

Diamond

Diamond; Mod Squad

Stainless

Diamond; Mod Squad

Stainless

Diamond

Stainless

Similar threads