Hi Ian,
When the runout of a straight, constant diameter bar is nil at all points along that bar then the bar is perfectly aligned with the spindle's axis of rotation. This will perform the same as a test bar which is fitted in the lathe's spindle taper. The part of the bar which the indicator will touch has the same relation to the spindle in either case; it's centered on the spindle's rotation axis - just held differently.
It does take some fiddling to get the arbitrary bar aligned but it's much cheaper than a test bar and a Thomsen bar has equivalent or better specs for roundness, diameter, straightness and surface finish. Since most owners align a lathe infrequently, often only once, the time vs cost tradeoff may be reasonable.
In either case, if the spindle axis is not aligned with the ways then as an indicator mounted on the carriage and zeroed at the headstock end of the bar is traversed along the bar it will read the change in distance between the ways and the spindle axis, i.e. the alignment error. Depending on whether the indicator finger touches the top or front of the bar it will read the vertical or horizontal error. Note that if you stop at any point and rotate the spindle there should be no runout if the real test bar or the arbitrary bar held in the chuck is perfectly straight, constant diameter, and perfectly aligned with the spindle axis. In the real world, where things are seldom perfect, one can use RDM to reduce the error due to imperfections, but limited by the resolution and accuracy of the readings, of course.
John
---In 7x12minilathe@yahoogroups.com, <ian_new@yahoo.com> wrote :
Hi John,
OK - I'll bite:
The point of using the test bar is to check the alignment of the system.
The process you describe in your post outlines a possible way to get a test bar aligned in a 'less than perfect' three or four jaw chuck. Doesn't this method assume the axis is already in perfect alignment?
All the best,
Ian
It is fair to point out that unless you're concerned with alignment of the spindle's MT taper to the spindle axis, an accurately ground bar (e.g. 3/4" Thomsen case hardened linear shaft, $12 at ENCO) can provide the same measurements as a more expensive test bar.
Or if you have an accurately ground bar you can make acceptably accurate measurements for alignment of a 7x12 with that. If you're willing to work at it you can lap a straight 1" round steel bar (preferably a tube), use a 4 jaw and set the runout near the chuck to nil, then set the runout at the far end to 4/10 or less by tapping it in the right direction as you incrementally tighten the jaws. It takes a few minutes to do this but at that point you have the test bar aligned exactly on the spindle axis. This should perform just like a commercial test bar. Of course, you should use RDM to correct for any minor runout remaining, just like when using a real test bar. Or, you could chuck the lapped bar in a 3 jaw, accept a thou or so runout, push the high side on the far end as you tighten the chuck incrementally and get runout to a thou or so, then use RDM - my guess is there will be little difference in results.
More thoughts on 7x12 alignment here: MiniLathe Alignment There's more to it than a test bar, a precision level may also be helpful, upping the cost for alignment tools.
John
