If a smaller chuck is being used, probably on a smaller lathe, then the size will
need to be scaled down appropriately. If the projection is then only 50mm the above
values would be 0.0025mm and 0.001mm but this is getting into values that would be
more difficult to detect, even with a quality micrometer.
There are other possible methods, one consists of taking a bar, say 300mm long by
25mm diameter, and mounting this in the four jaw, setting it to run true at the jaws,
testing this with a dial test indicator. Without moving anything, other than running
the saddle to the outer end of the bar, rotate the lathe BY HAND and take note of
the dial indicator readings and determine the mean value, that is midway between
the maximum and minimum readings (it is unlikely that the outer end will be running
By adjusting the lathe's mounting, the aim is to get the mean reading at the outer
end the same as that at the chucks jaws. Do take note though, that having made adjustments
it will be necessary to check the dial test indicator reading at both the chuck and
the outer end of the bar. An advantage of this system is that no machining takes
place, and if 300mm, then the figures above would be 0.006mm and 0.015mm for the
same degree of error. These theoretically being more easy to detect.
However, for the method to be adequate in its own right, a more sensitive indicator
would be required than those normally available in the home workshop. If not, then
it is just a good method of getting close to the required values, but a machined
test piece still being necessary. Do take note that the test must be carried out
with indicator horizontal, within a degree or two.
The Tailstock effect
Having now ensured that the lathe is set up to turn parallel within close limits,
the lathe owner may feel that it would be the end of the subject, this is not so.
It is possible that the tailstock could be off centre resulting in work being tapered
when the tailstock centre is used to support the workpiece, whether this be with
a workpiece held in the headstock chuck or mounted on the headstock centre.
Before turning a test piece to check the situation, both centres must be checked
for concentricity. Place each in turn in the headstock, after having scrupulously
cleaned centres and internal bore. Turn the lathe by hand and check with a dial test
indicator that the centre runs true, Photograph 3. If, as is most likely, both run
true (at least better than 0.005mm total indicator reading) all's well.
Should a soft centre be at fault it can be turned in situ in the headstock, if a
hardened centre is off centre it needs grinding, not a job for the majority of home
workshops and it is probably better to purchase a new one. If both are out of true
you should suspect the lathe mandrel's bore, hopefully it is just a small dent on
the entry to the socket that can be removed with a scraper, if soft.