In article ,
"Charles Friedman" wrote:

The size of the piece would be a cylinder of 2.5 inches diameter and a length
of 5 inches. The working end of the punch would have a rounded flat bottom
(radius of 3/16 inch) and could also be spherical (diameter of 2.5 inches) or
bullet shape or a taper.

My initial gut reaction is that hard maple would do better than oak. My
second reaction is that actual numbers are available in reference books.
Parts of one of the best are available on-line:

http://www.fpl.fs.fed.us/documnts/fp...tr113/ch04.pdf

Skip down to table 4.3b for Lb. and inch units - 4.3a is for
metric-thinkers.

That sort-of confirms my gut, but my gut is wrong depending on species
of oak (the red that's common here is weaker in compression parallel to
grain than sugar maple (which is aka hard maple). But if you can find
some black locust you're in who-hoo land!

In any case, if your figure of 3500PSI is correct, many woods will do
(and probably more than once) IF the load is primarily taken parallel to
the grain (ie, on end-grain). Live oak appears to be the best bet for
side-grain loading (in US woods). There must be some side loading in
punching a 6 inch disk with a 2.5-3 inch punch, to form the sides. The
die might benefit from hoops or pipe clamps to restrain it from
splitting.

Lingnumvitae (steel might be cheaper) and Ipe look worth investigating
for imported woods. Ipe is becoming common as deck material.

--
Cats, coffee, chocolate...vices to live by