The materials presented here are NOT FD+E official documents, and are only meant as working documents, with the usual working document errors. The material should not be used for any other purposes except research.
>. Ductile materials 1020 (or less carbon) will exhibit elongation
> after yield but the cyclic curve will have a smooth transition and
> immediately harden.
>
A: Yes, low carbon steels will usually exhibit upper-lower yield
behavior followed by strain hardening to the ultimate. (assuming no
prior cold work due to forming or rolling or something).
The cyclic curve will be smooth yes, except where it leaves the elastic region
due to the addition of elastic + plastic terms in any Ramberg-Osgood type of equation.
For low carbon steels it will typically "leave" the elastic region
at a stress lower than monotonic yield, then cross over the monotonic
and run above the monotonic for many low carbon steels.
Examples:
Keshevan's thesis monotonic data:
http://fde.uwaterloo.ca/Fde/Materials/Steel/Lowcarbon/sae1015KeshavanTens.html
I.Smith (mono. + cyclic):
http://fde.uwaterloo.ca/Fde/Materials/Steel/Lowcarbon/iSmith1120-mono+Cycic-SS.png
in folder: http://fde.uwaterloo.ca/Fde/Materials/Steel/Lowcarbon/lowcarbon.html
A.Conle (mono. + cyclic):
http://fde.uwaterloo.ca/Fde/Materials/Steel/BoronSteel/strainLifeCyclicMonoBoron.png
in folder: http://fde.uwaterloo.ca/Fde/Materials/Steel/BoronSteel/boronSteel.html
A.vander Zanden (mono. + cyclic):
http://fde.uwaterloo.ca/Fde/Materials/Steel/Sae1500/Vanderzanden/vanderzanden_Base_Tens+Cyclic.png
in folder: http://fde.uwaterloo.ca/Fde/Materials/Steel/Sae1500/sae1500.html>
In the database, when monotonic data is available, it is in a separate file.
The cyclic stress-strain raw data points are defined by column 1 and 3 in all the
fatigue files such as
e.g.: A36 steel
http://fde.uwaterloo.ca/Fde/Materials/Steel/Lowcarbon/mergedA36.html
When you hit the "Send for Processing" button the returned graph will show the
cyclic, and the parameters are K' and n' in the table. -or you can take the
"fitted" curve points (Columns 1 and 3) from the bottom of the returned web page.
> FEA models should reference cyclic monotonic stress strain curve
> data which is different than a true stress curve.
>
A: FEA models that wish to simulate monotonic deformation behavior
should use the monotonic true stress-strain curve. At the bottom of
the monotonic web page(linked above) is an Abaqus compatible text file
Abaqus type models that wish to simulate deformation events that
occur during or after plastic deformation fatigue cycles should use
the Cyclic stress-strain curves from the fatigue plots. (K', n' equation etc).
The equation has no limit however, and one must decide at what strain to
end it.
> A single overload instance on 1020 or lower carbon the true stress
> curve should be referenced which includes the elongation after yield.
>
A: Both monotonic and the cyclic have a "true stress-strain" curve version
so you have to pick which of the two you want to use. For Abaqus
models one should always use the true stress-strain curve. Although I
believe that Abaqus has a eng --> true converter option, but this should
be used with caution. The upper/lower yield point for example would give a
very hard time to Abaqus models.
Personally I think one should always plot the data to see if it makes
sense before running the Abaqus models.
> Materials that harden from draw processes have great tensile
> properties but FEA models should use the cyclic stress strain curve because
> the gains will be lost after cyclic loading.
>
A: If you are modeling an event after cyclic loading (with plasticity) has been
applied to the object, yes one should probably use the cyclic true-stress-strain curve.
If one is modeling a single high deformation event after cyclic loading
that does NOT have any plastic deformation, one should probably use the
cold worked monotonic curve. (note that the latter can be estimated from the
non-cold worked monotonic curve.)