2899-Testing Lathe Accuracy
2899-Testing Lathe Accuracy
2899-Testing Lathe Accuracy
LATHE ACCURACY
T
HB most important test of a
lathe lies, no doubt, in the
accuracy of the work and
the manner in which it is produced
-a skilled operator being able to
overcome considerable basic in-
accuracy. Even if the work is of
a relatively simple character, ample
scope for error exists-provided
there is sufficient variation to test
all aspects of alignment.
This is to say, a lathe may be
accurate for one type of work but not
for another, and experience of it
can indicate where to expect errors,
though the reason why may not
always be immediately apparent. How-
ever, there are various simple tests
which are largely a substitute for
work experience and which can
be useful for discovering errors, for
truing the machine (if possible), or on
occasion for setting up.
FI XED CENTRE /
13 DECEMBER 1956 831
A simple but important test, A, is
the meeting of headstock and tail-
stock centres. When the fixed centre
has noticeably dropped, wear of the
underside of the tailstock and possibly
of the bed itself is indicated. The
effect on between-centre turning may
be small or non-existent, but care
will be necessary when using centre
drills or boring cutters from the
tailstock-a degree of lift then
being necessary for the tools to centre.
The same effect also obtains when
supporting chucked work from the
tailstock. This test should be made
with the barrel both close in and well
extended. A sideways error of the
same kind can often be corrected by
adjusting the tailstock.
A more severe test of the same sort,
B, can be made with an indicator,
which can be a dial type or one of the
small inexpensive varieties. The
indicator is mounted in a chuck or on
a driving plate with its plunger bearing
on the fixed centre. Then the machine
spindle is turned, when a steady
reading shows perfect alignment, using
a small mirror to see the instrument
upside down and from behind.
Where there are variations in read-
ings, as is virtually always the case, it
can be seen in which direction (vertically
or sideways) they occur, and the tail-
stock adjusted to correct sideways
error.
A live centre test for running truth,
B, can be made of the one normally
in the spindle and of others of the
same taper with the indicator mounted
on the slide-rest. This can sometimes
reveal that the centres themselves
have not been ground true; and in use
it may be advisable to keep them to J
one position-spindle or tailstock-
marking for fitting in a certain
manner.
GENERAL ALIGNMENT
A test for general alignment of
headstock and tailstock for between-
centre turning can be made employing
a mandrel, C. Any suitable piece of
rod can be used, carefully centred,
reduced in its length, and with the
ends turned the same size.
A tool mounted on the slide-rest
can be brought close to one diameter,
leaving a small gap, t
checked on the other
piece of white paper on the bed
providing a light background against
which to see the gaps. Finally, the
tool may be set to touch the diameters
lightly when traversing the saddle.
This aids reasonable setting of a
lathe before work begins, and as an
alternative to a tool an indicator
can be used.
A faceplate may be checked with a
rule, D, and when mounted on the
spindle and rotated is tested for face
wobble. If chuck work is machined
true, a test can be made on the face-
plate of cross-slide alignment-in two
stages, E and F, using a tool or an
indicator.
Testing along the near side along
line X-Xl, no error may be shown if
the faceplate was machined on the
lathe, for alignment corresponds to
the cross-slide. Testing on the far
side, however, on line X2-X3, any
error is doubled and can be easily
seen.
Topslide setting can be tested as
G, an indicator on a mandrel and a
rounded rod on the slide-rest. With
the slide out of alignment, movement
is along such as X4-X5, and checking
with saddle traverse, variations are
shown, whereas with a true setting
readings will be all the same.
rzl
MODEL ENGINBBR
,