f the punch has a sixty-degree taper, a fair
center can be formed in this way, but it is not a method to be
recommended, especially when accurate work is required. Sometimes
centers are made with punches that are too blunt, producing a shallow
center, such as the one shown in the upper left-hand view, Fig. 27. In
this case all the bearing is on the point of the lathe center, which is
the worst possible place for it. Another way is to simply drill a
straight hole as in the upper view to the right; this is also bad
practice in more than one respect. The lower view to the right shows a
form of center which is often found in the ends of lathe arbors, the
mouth of the center being rounded, at _r_, and the arbor end recessed as
shown. The rounded corner prevents the point of the lathe center from
catching when it is moved rapidly towards work which is not being held
quite centrally (as shown by the illustration), and the end is recessed
to protect the center against bruises. Stock that is bent should always
be straightened before the centers are drilled and reamed. If the work
is first centered and then straightened the bearing on the lathe center
would be as shown in Fig. 29. The center will then wear unevenly with
the result that the surfaces last turned will not be concentric with
those which were finished first.

[Illustration: Fig. 30. Tool Steel should be centered Concentric, in
order to remove the Decarbonized Outer Surface]

=Precaution When Centering Tool Steel.=--Ordinarily centers are so
located that the stock runs approximately true before being turned, but
when centering tool steel to be used in making tools, such as reamers,
mills, etc., which need to be hardened, particular care should be taken
to have the rough surface run fairly true. This is not merely to insure
that the piece will "true-up," as there is a more important
consideration, the disregard of which often affects the quality of the
finished tool. As is well known, the degree of hardness of a piece of
tool steel that has been heated and then suddenly cooled depends upon
the amount of carbon that it contains, steel that is high in carbon
becoming much harder than that which contains less carbon. Furthermore,
the amount of carbon found at the surface, and to some little depth
below the surface of a bar of steel, is less than the carbon content in
the rest of the bar. This is illustrated diagrammatically in Fig. 30 by
the shaded area in the view to the