NX Nastran composite tensile/compression properties

[bulldogbaja]

I am using Nastran NX 8.0. I'm using the MAT8 card to input 2D properties for a unidirectional CF. The material properties I have specify different tensile and compression modulus and Poisson's ratio. My load case is complex so bending is impossible to predict.
Is there a way to input different tensile and compression properties? If not, what is the preferred method of selecting which material property values to use?

 

[BlasMolero]

Hello!,
In order to enter MAT8 NX NASTRAN (the solver) card data in NX 8.0 AdvSim (the pre/post) you need to select ORTHOTROPIC material properties, if not by default you are defining an isotropic material. Then in the interface you will see request for:
• Young’s Modulus (E1)
• Young’s Modulus (E2)
• Young’s Modulus (E3)

Also, in the Strength tab, Orthotropic properties, you can enter Allowable stress and strain for laminate failure analysis, with Stress/Strain Limits at:
• Tension (ST1, ST2, ST3).
• Compression (SC1, SC2, SC3)
• Shear (S12, S13, S23)

Also you can define the Tsai-Wu Interaction Coefficient (F12).

Best regards,
Blas.

~~~~~~~~~~~~~~~~~~~~~~
Blas Molero Hidalgo
Ingeniero Industrial
Director

IBERISA
48011 BILBAO (SPAIN)
WEB:

http://www.iberisa.com

Blog de FEMAP & NX Nastran:

http://iberisa.wordpress.com/

 

[rothers]

But that doesn't allow the input of different tensile & compressive Young's modulus does it ? Maybe a function could be used but do they allow negative stress values, i.e a stress/stain curve covering the full compressive to tensile range.

 

[hezhj2000]

Maybe I am wrong, but I don't see this is a linear problem. For example, for a single beam model, if I apply 10 unit force in tension for subcase 1, and apply 10 unit force in compression for subcase 2. By definition of linear problem, the results of subcases 1 and 2 should be able to sum together to get the deformation under the summation force of these subcases. Due to the difference in tensile and compression modulus, the summed deformation will not be zero, but the summed force is zero.

So I think you probably need to use nonlinear solution sequences for this problem.

 

[rothers]

Correct it's a non-linear material problem

 

[hezhj2000]

I know for MAT1, you can use MATS1 and TABLES1 to define strain-stress curve. I am using MSC Nastran, and I saw MATS8. But it says it is for Sol 400 only. I am not sure if NX has something similar.

 

[jotunn]

bulldogbaja,

To the best of my knowledge, no, you cannot specify both a tensile and a compressive modulus. Doing so would essentially create two different ABD matrices (laminate theory) and thus two different solutions.

For what it's worth, my general approach is to average the tensile and compressive moduli in each direction...

 

[bulldogbaja]

Thanks for all of your help- I ran my model twice, once with the compression properties and once with the tension properties. I saw a 1.5% increase in ply stress and a 4.7% increase in deflection using the tensile properties. I will therefore just use the tensile properties, since for my application this is good enough and I need to be on the safe side.