A head scratcher here …

[Christianstark]

Perfect timing for this thread. It's been quite a while since I checked level. I'd been planning to do it for a while and, boy, was it out of level. I got it level with the machinist level front to back and checked the headstock using the bar that David listed. It's been quite some time since I aligned the headstock. I had it perfect with the bar over the entire rod with an Interapid. I got it all tightened and verified alignment again. I faced a 3" aluminum cutoff and had a difference of .001" from center to edge. I did not expect that.

Did you continue to the other edge to see if you got back the .001 and are in fact machining a dish?

 

[parshal]

You mean put it in reverse and cut the other side?

 

[Christianstark]

You mean put it in reverse and cut the other side?

No, just indicate it from front to center to back. If front to center changes .001, it should change back to match the front measurement. The only reason I might suggest this is I have seen some recent postings on the castings on some mills not being dead square. Granted the thread I am recalling was for a milling machine, and I’m sur I am wrong, but if you indicated front to middle to back, and the indicator does not return to the same indicated value, then the cross slide/ casting may have an issue.

 

[RJSakowski]

No, just indicate it from front to center to back. If front to center changes .001, it should change back to match the front measurement. The only reason I might suggest this is I have seen some recent postings on the castings on some mills not being dead square. Granted the thread I am recalling was for a milling machine, and I’m sur I am wrong, but if you indicated front to middle to back, and the indicator does not return to the same indicated value, then the cross slide/ casting may have an issue.

The headstock out of alignment will also cause that problem. I mount a boss on my faceplate and rotate the faceplate so the boss in near the front of the lathe. I retract the cross slide and zero a dial indicator on the boss. Then I rotate the faceplate so the boss is at the rear of the lathe and advance the cross slide until the indicator is in contact with the boss. If the indicator reads zero, the headstock is aligned to the cross slide. If the reading is less, the headstock is rotated counterclockwise to the cross slide when viewed from above. If the reading increases, the headstock is rotated clockwise to the cross slide. If the cross slide is perpendicular to the ways, the former case will result in turning a taper which is larger on the tailstock end while the latter case will result in a turning a taper which is smaller at the tailstock end.

On my lathe, I have no means of correcting a dished or domed facing cut aside from rotating the headstock. With that alignment made, I can turn my attention to correcting any taper due to bed twist.

 

[Christianstark]

The headstock out of alignment will also cause that problem. I mount a boss on my faceplate and rotate the faceplate so the boss in near the front of the lathe. I retract the cross slide and zero a dial indicator on the boss. Then I rotate the faceplate so the boss is at the rear of the lathe and advance the cross slide until the indicator is in contact with the boss. If the indicator reads zero, the headstock is aligned to the cross slide. If the reading is less, the headstock is rotated counterclockwise to the cross slide when viewed from above. If the reading increases, the headstock is rotated clockwise to the cross slide. If the cross slide is perpendicular to the ways, the former case will result in turning a taper which is larger on the tailstock end while the latter case will result in a turning a taper which is smaller at the tailstock end.

On my lathe, I have no means of correcting a dished or domed facing cut aside from rotating the headstock. With that alignment made, I can turn my attention to correcting any taper due to bed twist.

Agreed. But a cross slide out of square would double the error from front to back. A dished or domed face would read the same measurement recorded at the front, and the back.

 

[Parlo]

The headstock out of alignment will also cause that problem. I mount a boss on my faceplate and rotate the faceplate so the boss in near the front of the lathe. I retract the cross slide and zero a dial indicator on the boss. Then I rotate the faceplate so the boss is at the rear of the lathe and advance the cross slide until the indicator is in contact with the boss. If the indicator reads zero, the headstock is aligned to the cross slide. If the reading is less, the headstock is rotated counterclockwise to the cross slide when viewed from above. If the reading increases, the headstock is rotated clockwise to the cross slide. If the cross slide is perpendicular to the ways, the former case will result in turning a taper which is larger on the tailstock end while the latter case will result in a turning a taper which is smaller at the tailstock end.

On my lathe, I have no means of correcting a dished or domed facing cut aside from rotating the headstock. With that alignment made, I can turn my attention to correcting any taper due to bed twist.

Are you setting the indicator on the face of the boss? Would setting the indicator directly on the faceplate give the same result?
I guess it's a choice between aligning the headstock parallel to the spindle axis or square to the cross slide.

 

[ddickey]

I feel as confused as the original poster.
Doesn't a headstock misalignment and a twisted bed produce the same error? A tapered cut?
How do you distinguish one from the other?

For a new lathe, place a precision level on the cross slide and adjust the stand feet or lathe feet so that there is no movement of the bubble.
Then you take a test cut 5"-6" and measure both ends of the bar. Type and size of material for the bar, tool to use, spring passes & DOC's are very important.
After you get zero taper you're done.

 

[ddickey]

I invested in one of these - makes aligning the headstock very easy. Once the headstock is aligned you can adjust out any twist in the lathe bed.

Lathe Alignment Test Bar Shank Size 1MT, 2MT, 3MT, 4MT, 5MT, 6MT Hawk | eBay

2MT 283mm (11-1/8”) 0.0002”. 3MT 335mm (13-3/16”) 0.0002”. 4MT 386mm (15-1/4”) 0.0002”. 6MT 507mm (19-7/8") 0.0002". 1MT 151mm (5-15/16”) 0.0002”. 5MT 400mm (15-3/4") 0.0002”. Hawk Werkzeuge comprises of a dedicated and skilled team of engineers with an aim of providing you tools of highest...

www.ebay.com

I've bought two of these both out of spec. In fact the last one I bought had .003" TIR. However, I was refunded the total amount I paid without needing to send back to India (2nd one).

 

[RJSakowski]

Are you setting the indicator on the face of the boss? Would setting the indicator directly on the faceplate give the same result?
I guess it's a choice between aligning the headstock parallel to the spindle axis or square to the cross slide.

You can run on the face of the faceplate. I chose to add a boss because my faceplate has radial slots and concentric grooves and the indicator would drop into the slot and/or groove as I was changing positions. The boss is domed slightly so there is no abrupt jump when making contact again. I made it from a hex bolt and it is easily removed when I need to use the faceplate. A Sharpie cross on the face of the boss ensures that I measure on exactly the same point front and back.

I chose to align the headstock square to the headstock because I had no other easy way to make that alignment. Once the headstock was aligned, I could then correct any taper turning issues by controlling bed twist.

 

[parshal]

I took the chuck off and ensured there was no swarf and lightly stoned it and the lathe's boss, reassembled and made another cut. It's close to .001" across a 3" aluminum cutoff. The dish/dome measures the same all the way across. I moved my Interapid a bit above center to avoid the very center of the cut and, of course, it read just a little less. As much time as I spent yesterday getting the headstock aligned with the test bar I decided not to go after it given I'm in an uninsulated barn at it's near 100 outside and in.

I cut a two bar test over 12-15" and it cut perfectly so I'm leaving it alone.