Very confused after aligning the tailstock.

[zippyslug31]

I just completed the alignment of my tailstock on a PM-1130v: I have it running within a half thou over 8 1/2" inches which seems plenty good enough to me. But the direction of adjustment on the tailstock makes no sense to me. Prior to the TS aligning, I leveled the ways of the lathe and have it also running very well.

I've include a kindergarten drawing of my lathe (oh the colors!) to help with the discussion -



I turned a test bar (orange) between dead centers and took a reading - the TS end of the bar was smaller by 7 thou. To me this means that the tool (yellow) is closer to the work at the tail end (green) and farther away at the headstock end (red). I had an indicator (brown) attached to the backsplash and indicating against the TS so I adjusted it about 3.5 thou closer to the indicator (upward in my picture).

I did another pass, mic'd the bar and this time I was nearly 12 thou out?!? Trusting the indicator, I readjusted away from the indicator (downward in my picture) which, to me, would bring the test bar even closer to the toolpost. I took another pass on the test bar and mic'd again - this time I was very close.

Assuming I have the ways (blue) level, how can this be? Either I'm completing misunderstanding how this works or there's a deeper problem.

After all this fun, I remounted the 3-jaw and turned a part using the tailstock... I am indeed cutting pretty straight. This is great, but I'm super confused...

 

[pontiac428]

This test makes the assumption that the cutting point is on center. Bed twist where the tool (for example) drops below center as it moves toward the tail will show a larger test cut than the same cut at the head stock. Run your test indicator on both the top and the front vertices of your part with the indicator base mounted on the top slide, so you can see how the tool tip moves in relation to long travel.

Once you've proven the bed level and the tool tip tracks vertically, then you can do the tailstock offset.

OTOH, .0005/foot is not too shabby.

 

[Janderso]

OTOH, .0005/foot is not too shabby

#1

 

[PhilJustLearning]

Hi zippyslug31,
I'm new on HM, but what you said seems right to me and why did it increase (I'm also wondering). I was thinking when you moved the tailstock, did you put a twist in it by mistake so that the indictor may have moved the 3.5 thou in the correct direction but the tip of the tailstock live center went the other way making the test bar worst? The green block doesn't want any twist. I like the drawing BTW, very helpful!
Phil

 

[Huub Buis]

To adjust the tail stock, you have to loosen the tailstock lock & quill lock and lock them again after adjusting. It is possible that the tail stock has rotated a bit on the lathe bed, you did(not) lock the quill, there is wear on the quill, dirt on the ways, dirt in the test bar center boring, etc.

Put an indicator on the test bar and measure a few times how repeatable you can reposition the tail stock.

 

[Justin.P]

I wouldn't trust indicating off of the backsplash.

 

[gard]

Put an indicator on the test bar and measure a few times how repeatable you can reposition the tail stock.

This is what I was thinking also, put the indicator tip on the test bar or the tailstock quill. Nice drawing, I think you were correct about which way it should be moving.
I will admit there have been times I have misinterpreted which way the point was moving vs which way the dial on the indicator was turning. Usually I move the indicator tip with a shim or small screwdriver to try to verify it. Let us know what you find out.

 

[WarrenP]

It sounds like your doing the measurement from the opposite side as your cutting tool. That would mean it would need adjusted towards the opposite side. If it measures smaller on the opposite side as the cutting tool then it would be further from the cutter but closer to the indicator.

 

[zippyslug31]

I was out in the shop last night experimenting with several points brought up here; thanks for the replies as they were helpful.

A couple things I learned:
- I agree that it was a mistake of me to leave the TS locked during my adjustment... if anything I introduced bed twist, which could be part of the explanation.
- I also found that the TS does indeed have a small amount of vertical variance so I'm investigating how to adjust the attitude (z-axis, if you will) of the TS; the only adjusting bolts I've found are for lateral movement.
- The TS adjustment screws move the TS in the opposite direction that I originally assumed. This is partly due to my inexperience, and partly an odd design IMO. This makes my task a real brain twister since I have to (first) consider which way I want to move the TS, (second) think about which way the indicator should react, and (third) turn the adjustment screws in the correct direction. I guess I already realized this during my first session, but last night I confused myself a few more times! (would be slightly easier if I moved the indicator to the other side of the TS... I think)

Addressing a couple of comments/questions that were made:
- The lathe is less than 6 months old, with hobby-only use so there should be no wear yet.
- While I agree that a backsplash isn't super rigid, it should be good enough for this test. During an actual turning operation I watched the indicator and the needle bounce within a 1 thou variance. With the lathe turned off and me making adjustments to the TS, this mounting point was stable enough to show me relative adjustments.

After my screwing around with my setup last night, I lost my .5 thou setting, but I'm currently adjusted to "pretty good". Not a problem though since doing this setup/adjustment more than once is just as educational as actually turning real parts.

 

[tjb]

It seems to me it's a near guarantee that you will get less than accurate results by indicating off of the backsplash. Not only is it not 'super rigid', it's in no way designed to be rigid at all, much less attached at a level of measurable precision. You could probably grab and move the top-right corner of the backsplash and see significant movement on your indicator. I would be inclined at the very least to come up with a method of determining how accurate such a reading would be. I'm assuming you're using a magnetic holder, so if you could rig up an indicator that would measure from the backsplash to the back rail of the bed, you could take several readings along its length between the headstock and the tailstock. The bed's not going anywhere, so your readings would give you an idea of how much distortion is presented by using the backsplash.

Regards