Master precision Machinist level

Dude, document this in your fettling thread.

Seriously, it will be cathartic.
I will...when I've either:

Performed a successful installation of the new angular contact bearings in the new headstock (yeah, I think I buggered up my original pair of new angular contact bearings too; I have another new pair of those now)

Or

Messed it all up again and need to confess my further clumsiness and folly for the sake of my immortal engineering soul (such as it is).

I reckon it will be the former as I know the utterly fatuous mistake I made and have identified some other risks too.

The correct and non-risky procedure will take me longer but eh...I'm only 52; got loads of time left. :D
 
My tools are reasonably accurate, but not "tenths per foot" accurate.

I'm not building rockets, and getting within 0.001" is fine for the tractor parts and other crap I make. My mill is 25 years old. And I do this for fun.

I'd rather spend my play money on a new tool or material than making sure my mill is at NASA spec.

That said, if you want your stuff to be completely dialed in, and it makes you happy, go for it. It's not going to hurt, even if it doesn't add any value.
Nope not making parts for nasa. Hahahhaha
 
Fellow beginner here. :)

Eh, I've yet to even make any chips on my little 7x14 (had a little accident with a hydraulic press when installing new angular contact bearings in the headstock*, so have had to order a new headstock casting; arriving today hopefully, Royal Mail willing).

When I read about levelling a lathe (reducing/removing twist in the bed really; no need for a lathe as a whole to be level), I figured I'd need a machinist's level and bought a cheap Dasqua 6" (150mm) level. However, I then read about Rollie's Dad's technique and turns out I could have done without.

£100 ($126) is about two-thirds of the cost of a decent clone Aloris style quick change toolpost with some toolholders, or most of the cost of a nice Mituyoyo dial test indicator, or a decent shop vac or will go towards a decent bench grinder...

So maybe don't be like me! :D






*yes I know, why not use the thing with the original deep-groove bearings and learn and have some fun? Eh I had it in bits anyway for cleaning and fettling and I have a terror of not being able to discriminate between my lack of skill and the failings of kit when starting a new hobby. I am aware this is not necessarily a balanced, constructive and healthy perspective. ;)
Well the point is to learn from other experiences so I am learning. I will not buy that one, and put the money for the dial test indicator.
 
I used my lathes for almost fifty years before buying a precision machinists level. You can align a lathe to cut without a taper without a level. However. There are points to consider. A lathe will cut a taper for several reasons. The lathe bed can be twisted, the bed can be worn from use, or the headstock can be misadjusted. Or it can be a combination of all three.

By using the two collar method or Rollie's Dad's method (RDM method), it is possible to adjust a lathe to cut without a taper. But two wrongs don't necessarily make a right Case in point; if the headstock is misadjusted, the spindle axis is not parallel to the ways and you cut a taper. Also though, the cross slide isn't perpendicular to the spindle axis so when facing, you make a domed or dished cut. Twisting the bed will correct the taper issue but it won't correct the facing issue.

This indeed was the issue that I had with my Grizzly lathe. Using the two collar test and shims, I was able to cut without a taper but I was still cutting a dome when facing. So the problem at hand was how to parse out two related problems. My answer was to get back to basics. First, level the lathe to remove any bed twist. Then adjust the headstock to cut a flat face. Then use the RDM method and shims to correct any remaining taper issues.

This was the reason that I bought a precision level.

On another note, My Tormach CNC mill has had some issues with the table not being perpendicular so that the table rises or lowers relative to the head as it is traversed in either x or y. Again, there are multiple reasons as to why this condition occurs. The table can be twisted, the table surface could be non parallel with the ways (factory defect), or the table can be lifting due to a combination of loose gibs and cantilevered weight.

Tormach's recommendation to correct this was to level the machine with a precision level and then shim the base to remove any residual issues. When I bought the mill, I had set it up with a carpenter's level. When I bought my precision level, I was amazed at how far out of level the machine actually was.

Here is an more detailed account of my lath experiences. https://www.hobby-machinist.com/threads/on-the-level-further-adventures-with-the-g0602.72338/
 
1) the .0002 per foot levels are very VERY hard to calibrate. They do NOT come calibrated from the factory. Mine has a messed up adjuster screw, masked by blue locktite. When the locktite was removed, there was a galled and frozen adjuster thread. When I fixed it and set it up on my surface plate, I found the adjuster screw thread to be too coarse to get it even close to being calibrated. One day I might make a differential screw adjuster for it, but I have 4,273 more important projects to do.
I use one of those that's never been calibrated, I just never rotate it! Worked fine for taking the twist out.
 
Regarding using a plumb bob for leveling a lathe, consider this. The sensitivity of a plumb bob in detecting a plumb condition is governed by to factors. first the length of the plumb bob and second, the user's visual acuity. I would expect that a person's uncertainty in viewing and out of plumb condition would be on the order of a 1/16th inch. In order to achieve the sensitivity of a Starrett 98 level (.005"/ft), the length of the plumb bob would have to be about 12 ft. To equal the sensitivity of a Starrett 199 level, it would be 120 ft long. This is an impractical situation in most shops.

I happen to like plumb bobs and have several which I use on a fairly regular basis. But in the case of leveling a lathe, a plumb bob isn't the tool to use. A plumb bob will precess, tracing out an oval path and takes forever to come to a dead stop. If you think that a precision level is fiddlesome, wait until you try a plumb bob.
 
Well the point is to learn from other experiences so I am learning. I will not buy that one, and put the money for the dial test indicator.
Yep, having a reliable DTI is important.

I recently used my DTI to check the spindle on my old Record Power drill press and thankfully the runout on that is entirely acceptable (just under a couple of thou on the spindle and just over 3 thou, 4 inches down on a piece of half inch drill rod; the previous owner was a professional machinist and up until the last couple of years before he died, he looked after his kit well).

Mitutoyo or Starrett tend to be the most recommended (I have a vintage Mitutoyo tenths indicator that seems to be still going strong) but the one I've used so far is a cheap Dasqua 0.01mm fella (half thou practically speaking), so no real need to go overboard. ;)
 
Yep, having a reliable DTI is important.

I recently used my DTI to check the spindle on my old Record Power drill press and thankfully the runout on that is entirely acceptable (just under a couple of thou on the spindle and just over 3 thou, 4 inches down on a piece of half inch drill rod; the previous owner was a professional machinist and up until the last couple of years before he died, he looked after his kit well).

Mitutoyo or Starrett tend to be the most recommended (I have a vintage Mitutoyo tenths indicator that seems to be still going strong) but the one I've used so far is a cheap Dasqua 0.01mm fella (half thou practically speaking), so no real need to go overboard. ;)
Thanks a lot for your thoughts
 
Back
Top