Correlation Curves PP Abaqus

[avieira]

Hi all, Jorgen.

I'm trying to correlate the physical tensile test with the virtual testing, the elastic part is going well, but the plastic does not. I calibrated the material using the software MCalibration, in attached I put the images of the calibrated material and the result of comparison between the virtual and physical curve.
The Abaqus material I'm using is the Two-Layer.

For the stress data I'm dividing the reaction force from the model by cross-section area of my speciment.

The material is a PP and I'm using the abaqus.

I put the viscous data, the plastic data too, what I need more to correlate this material, in the plastic behavior?

Thank you.

[Jorgen]

I am not quite sure I understand your question. One of your figures show a stress-strain curve for PP (?), and also a model prediction. What material model is that (Two-Layer plasticity?). The data in that figure looks good.

I am not sure what the other figure is showing. :(

-Jorgen

[avieira]

Hi Jorgen,

Yes, the figure is the stress-strain curve, is the comparison of the physical test, in red, with my virtual test, in blue. The Two-Layer is a abaqus material, which I can enter the viscous part.
My queation is, how can I correlate the plastic curve?

Thanks.

[Jorgen]

I am still confused about what the difference is between the two figures (except that one looks good and one looks bad :(

Perhaps the Two-Layer model is not good enough...

-Jorgen

[avieira]

The good picture is only to show the calibration on MCalibration Software, which is really good, another figure is to show my correlation between the tensile physical test and my virtual test. I made a virtual test and extract the curve virtual strain-stress, and compared with the physical test, but the plastic behavior of my virtual test is different from the physical test, I'd like to know, how can I calibrate my material for give me a good correlation between the tensile physical test and virtual test.
Below the data material that MCalibration informed me, I need to put some more data?

Thanks.

*Material, name=MCalibration material
** Calibrated with MCalibration
** Units: [length]=millimeter, [force] = Newton, [time] = seconds, [temperature] = Kelvin
** Calibration file name: simulation.mcal_com_tinta.mcal
** Two-Layer Metal plasticity with rate-dependence
*Elastic
2200.82, 0.414733
*Plastic
**(you can specify an aribtrary number of points here)
10.4343, 0.0
12.9201, 0.0112683
15.5475, 0.0521779
13.5884, 0.385452
*Viscous, law=strain
7555.12, 3.38049, -6.41553e-05, 0.463099
*Density
1e-09

[Jorgen]

I still don't quite follow :(
Can you explain more about what you mean by "physical test" and "virtual test"? What are the tests? How did you get the experimental data?

-Jorgen

[avieira]

Hi Jorgen,

I'm Trying to correlate the physical tensile test with the virtual tensile test, using the two layer abaqus material.

I get the stress x strain curve from physical tensile test and calibrate on the MCalibration Software, but when I put the calibrate material from MCalibration on my abaqus model and perfomance the virtual tensile test, the result curve (stress x Strain) from the virtual test is diferent from physical test, I'd like to know, how can I fix this?

I put in attached the material´s physical curve, and a virtual and physical tests pictures.

Thanks

[Jorgen]

I think I finally got it :)

That is indeed odd. I have one idea: it appears that the specimen in the physical test may have necked during the test. That is, the stress and strain fields in the physical test may not have been homogeneous.

Note that MCalibration is using a "single element" for the stress-strain calculations which will not allow for necking.

What do you think? Could that be part of the reason?

-Jorgen

[tdalrymple]

Avieira,

You have shown a simple tension pull test. What other test data do you have for this material? If you want to calibrate the Two-Layer viscoplastic model you will need some test data that tells you about the time-dependent nature of the material.

- Tod