I am cold rolling bi-metal steel coils (Copper Alloy on steelback). My
challenge is correlating the finished steel hardness with the percent
reduction. I need this to choose the proper steel size to begin with
to ultimately achieve the proper final hardness and size.
I'm an engineer and not an experienced metalurgist. I assume there
curves or formulas available to determine the steel hardness vs %
reduction? Does anybody know where I can find these and what other
factors I should be considering in this process?
Bascially your final as-rolled hardness is a fuction of the orginal
input hardness transformed by the metal flow model base on the amount
of strain (%reduction).
Some stress verses strain curves are published ASM International in a
book (check the web site). Unfortunately these will be next to
useless fo you. Standard hardness verses % reduction curves (really
plots) from steel companies (if you can find them!!) are only
applicable to their machinary due to redundant deformation.
What you really need is a metal flow model. Generally for steels
hardness follows the tensile strength (other will say yield strength
but they don't realize that it only applies to the annealed state -
they are ignorant of the process of work hardnening).
The metal flow model for you case must use tensile strength (not
hardness - hardness is not a ratio scale - has no zero!) can convert to
hardness (using tables from per ASTM A360 or SAE J417)
For % reduction e (measured in true strain only!!!)
OCRT=3DICRT* exp (e) for en
ICRT =3D Input Cold Rolled Tensile Strength (from input hardness)
OCRT =3D Output Cold Rolled Tensile strength (can convert to hardness)
e =3D natural or true strain from rolling =3D ln (input thickness/output
thickness) + Redundent strain
n =3D necking strain - strain at tensile load.
A =3D intercept factor (specific to steel) - in ksi or N/mm=B2 units
B =3D post necking strain hardening factor (specific to steel)
ksi/strain or N/mm=B2/strain
Actually it would be easier IMHO to plot these surves from your product
using the flow model above rather than search for data.
Ed Vojcak PE