Making a lathe test bar

[Tozguy]

I took a 3/4” diameter 24” long rod, nothing special, held it in a 4 jaw chuck with minimal stickout, adjusted for minimum runout and center drilled. Did the same to the other end, optional but seemed appropriate. Then I pressed a couple of about 2” diameter, 1” thick aluminum donuts on either end, about 2” in. Then center one end in the 4 jaw, other in tailstock center. Make a couple of light passes to clean up, and now you have a pretty good test bar where you can make a very light cut on one donut and without changing the tool position make the same cut on the other donut. If the two diameters match, your lathe is perfectly aligned. My guess is it won’t be, but you can get the error to a few tenths with some fiddling around.

Any feedback? Would appreciate any critique. It was cheap, didn’t take much time to make, but plenty of time messing with tailstock alignment, and let me see up close some of the surprising error sources when you are pursuing tenths.

Not sure about holding one end in a four jaw even if run out is zeroed. Isn't the tailstock alignment test supposed to be run between centres?

 

[Tozguy]

Considering the variables we face when turning (quill extension, position of TS on the ways, flex in the holding systems, etc.) is it logical to adjust the TS to the last tenth using a delicate test procedure that does not duplicate the conditions of the work to be done?

In a job where tailstock alignment is critical what tests or measurements can we make on the actual work to assure that the TS is set right for the job at hand?

 

[rwm]

If you drill a center on one end but it is off center a little it doesn't matter. You can use a 3 jaw. Also tailstock alignment doesn't matter. After center drilling you place the bar so the center is at the tailstock. When you turn the first collar you work at the TAILSTOCK end. That will make a collar necessarily concentric to the center. Then you flip the bar and repeat.
Robert

 

[Stonebriar]

oops misread

 

[Winegrower]

In a job where tailstock alignment is critical what tests or measurements can we make on the actual work to assure that the TS is set right for the job at hand?

I would say that if your turnings are the same diameter at both ends, to whatever level of accuracy you need, then your lathe is adequately set up.

If not, then pursue to the extent you find necessary and fun. I did, got the taper to around I recall about 8 tenths over 24", but I posted on this and could look it up.

More than good enough for a few washers and bushings.

 

[Tozguy]

Winegrower, just wondering what mechanism your lathe (Takisawa ?) has for adjusting the tailstock alignment. My lathe has a very crude mechanism that would take a miracle to adjust in tenths increments.

 

[Suwannee Tim]

If the stock is 4 jaw chucked close to the chuck jaws, and indicated closely, the center drill will crowd to center and make a closely concentric center, if a recentering tool; is used to further true it up, it should be quite concentric. How is this bar to be used?
Also if the bar is held with the far end projecting some distance and held in a steady rest, the result will be the same.

I intend to use the bar for aligning the lathe. I already chucked the bar with 30 inches sticking out and no support on the tailstock end, run an indicator along the bar, 0.030 over 30 inches, clearly unacceptable, that is the headstock misalignment with the ways. I probably need to correct that before I do anything else. Once I have that corrected and have concentric holes in the ends of the bar I would hold the bar in a 4 jaw, carefully indicated but not very tight in the chuck and the other end on a center in the tailstock to align it. I would run an indicator up and down the bar to set the alignment of the tailstock.

I don't know that I agree that a center drill will find the center of a piece naturally, sounds too good to be true and we know how that works. I see the center drill touching off center and being grabbed by the work piece and the end of the center drill orbiting around the geometric center due to flex in the headstock, the chuck and whatever you call it thing that extends from the headstock, not a spindle cause it doesn't spin, quill I guess it is called. I do agree that an off center hole could be trued with a tool off the tool post and that should be true. I also think after such truing I could put a live center in the headstock and extend the quill out as far as it would go to make it as limber as possible and engaging it in the hole, running the lathe and put a dial indicator on the live center to detect and read the runout of the hole.

 

[rwm]

Whoa! Be careful here. If you hang a 30" bar out of the chuck it will flex under gravity. A perfectly aligned headstock will read incorrectly with a dial indicator in this setting.
Robert

 

[Suwannee Tim]

Whoa! Be careful here. If you hang a 30" bar out of the chuck it will flex under gravity. A perfectly aligned headstock will read incorrectly with a dial indicator in this setting.
Robert

I did think about that, maybe today I will calculate the deflection. The misalignment so severe errors caused by deflection would not be an issue but as I get these alignment issues worked out it will become a consideration. Thank you for pointing it out.

 

[Tozguy]

A lathe inspection record from Grizzly shows that the headstock alignment test uses a 12'' test bar. The spec for vertical alignment is 0-.001'' at the end with actual results probably close to .0005''. Horizontal alignment spec is a bit tighter at 0-.0008'' with actual results comfortably inside the tolerance.
The test bar fits in the spindle taper and is probably made of a very rigid material but there would still be some deflection. There always is.

A chuck is an unreliable means of holding a test bar because of the lack of precision in most chucks. Results will likely reflect misalignment of the chuck more than anything else. A bar can be carefully zeroed (almost) in a four jaw chuck but there is no garantee that the axis of the chuck is parallel to the axis of rotation of the spindle.

For tailstock alignment, to get a center in the headstock accurate enough for testing, it should be turned fresh every time it is installed.

To get an accurate center hole in the bar, after first drilling a center hole as accurately as possible using the 4 jaw and center drill in the tail stock, mount a center reamer in the tool holder on the carriage and ream the hole. Results will depend entirely on how well you have zeroed the 4 jaw.

Aligning parts on a lathe does not have to involve calculating deflection or measuring runout of centers or center holes. If you use the right approach and tools, your centers and holes will be the best you can make them whatever that is.