Lathe leveling - do you do it with the lathe fully dressed or just the headstock and ways?

[ddickey]

I've spent hours trying to do this correctly. I used to use two levels. A cheap one and a nice one made in France I think. I could never get the cheap one calibrated so now I only use the one, I was able to cal it. Put it on the cross slide like Mike and Rich said.

 

[ddickey]

Yes. It seems even it I get it off the peg at 180, then at 0 the bubble is pegged.

Interesting. Did you post what brand you have? Do you have a picture?

 

[WobblyHand]

I've spent hours trying to do this correctly. I used to use two levels. A cheap one and a nice one made in France I think. I could never get the cheap one calibrated so now I only use the one, I was able to cal it. Put it on the cross slide like Mike and Rich said.

Will be putting the level on the cross slide. As long as I don't move it from the cross slide, I can measure the twist of the ways.
However, it bugs me that it won't calibrate. The darn thing is supposed to be self proving, but just can't quite figure it out.

 

[WobblyHand]

Interesting. Did you post what brand you have? Do you have a picture?

Usual suspects - no name. I can post a picture. Here it is. Came in a padded wooden box.

Adjustment is with a special wrench. You remove the black plug under the red screw. There are two nuts that control the level of the level.

 

[rabler]

I’m guessing your calibration is so far out that your initial level line is not within the range of the level.
My suggestion is to to adjust the level so as near as you can tell the bubble moves to full deflection at about the same rate in either direction when flipped end for end on the surface plate. Once you’ve done that, find a new level line on the surface plate and try again. May take several iterations.

I wouldn’t trust wooden floors to be stable.

 

[Firstram]

The darn thing is supposed to be self proving, but just can't quite figure it out.

Could be the spot you chose isn't level enough to reuse after you tweak the 180°? Try rotating to a new level spot and adjust again. Might take a few times if it was off enough to begin with!

 

[WobblyHand]

Unless of course you inadvertently flip the level end for end between measurements.

No, not doing that. It's obvious what the orientation is, see the picture in previous post.

 

[WobblyHand]

I’m guessing your calibration is so far out that your initial level spot is not within the range of the level.
My suggestion is to to adjust the level so as near as you can tell the bubble moves to full deflection at about the same rate in either direction when flipped end for end on the surface plate. Once you’ve done that, find a new level line on the surface plate and try again.

I wouldn’t trust wooden floors to be stable.

Bingo! This is what I suspect. From my calculations, I need the initial level axis to be level within 0.002" over 12" or I am out of range. So far, that seems to be pretty tough to find.

Wooden floors are not terribly stable, at least in the 1875 part of this house. If I get close to the end of the level I can see the level move 1 division. I have tried to stand perpendicular to the level long axis to minimize that disturbance.

 

[WobblyHand]

Could be the spot you chose isn't level enough to reuse after you tweak the 180°? Try rotating to a new level spot and adjust again. Might take a few times if it was off enough to begin with!

Feels like shooting in the dark, but that's all I can think of.

 

[mikey]

Yes, going through beginners hades. Everything that could mess me up is happening. At this point, I'm stuck on level calibration. Seem to be getting nowhere. The document you attached assumes that the surface plate is flat and level, or at least level to the measurement range of the level. At the moment, I can't guarantee that, not with 0.02 mm/M. That would require the slab be level to within 9 divisions * 0.02 mm/M, or 0.002" over 1ft. The counter is level to a mm over a meter, I'd guess.

Have a small 9 x 12 surface plate, and a 6" level. Right now the surface plate is sitting on the kitchen counter, because it's a bit warmer than down in the basement. I know about letting things settle thermally and mechanically. The counter isn't particularly stable, but 71F in the kitchen is a lot nicer for me than 61F in the basement on a similarly unstable surface. I use the cross level to indicate roughly where the level plane might be, then rotate the level 90 degrees to search for a candidate area.

What confuses the daylights out of me is the following situation. Say we pick 0 degree rotation and it reads high on the left side. Right side of bubble is on left side. Then rotate the level by 180 degrees. Now the bubble is high on the right side. Does one adjust the bubble towards the center? It doesn't seem to help the way I'd expect.

I use a toolmaker's flat (a 9X12 Starrett) that sits on a 3/4" thick plywood base that has three threaded legs that run in threaded inserts. The tripod arrangement allows me to level the support/bearing surface under the surface plate, just like the article recommends. If you cannot adjust the surface your plate is sitting on then calibrating the level is going to be difficult.

This is the fastest and most accurate way to calibrate your level. I place the level on the surface plate and use 1-2-3 blocks to locate it so I can accurately flip it. If I recall correctly, I've only needed to do this once and my level has remained calibrated since. I also used the same set up to calibrate my Starrett 98-6.