Stiffener matrix A,B,D

[Ing.marco.soave]

Good morning everybody...

 
I have a serious problem to understand the way used by Hypersizer in order to calculate the stiffener matrix A,B,D through the coefficient A11,A12...B11,B12...D66. I know that Hypersizer creates a PSHELL property with MAT2 materials into the text file with extension .PM1. My problem is to compare these values with values provided by nastran output .f06 and with a mathcad sheet in which analythical formula given by RObert Jhones in  "Mechanics of composite Materials".  I have a composit with 2 layer (a and b) so in this particular case the coefficient of matrix are:

A11=Q11a(z1-z0)+Q11b(z2-z1)

D11=1/2(Q11a(z1^2-z0^2)+Q11b(z2^2-z1^2))

B11=1/3(Q11a(z1^3-z0^3)+Q11b(z2^3-z1^3))

where z0 is 1/2 of the total thickness and z1,z2 are the distance between the centroid and the middle plane of the whole plate for the layer a and the layer b respectively.


Thanks a lot for each reply!!!!

Marco

 

[Phil]

Hi Marco,

Are you saying that you have a two ply laminate and you are trying to compare HyperSizer's ABD results to the classical lamination theory ABD results? 

For composite laminates, HyperSizer implements classical lamination theory exactly therefore it should exactly match ABD results as implemented in Robert Jones book (with which we are very familiar).  This has been verified many times.

Are you saying in your email that you have found this to not be true? 

Perhaps you can give us an example of what you are looking for.

One other thing.  If your component is an unstiffened laminate or sandwich, then by default, HyperSizer will not export a PSHELL and MAT2 properties into the .PM1 file.  Rather it will export PCOMP and MAT8 material cards.

Thanks,
Phil

[Ing.marco.soave]

I have used a grid stiffner family concept so Hypersizer export in .PM1 file a PSHELL with MAT2.

By using the Jhones formula all the coefficients of stiffnes matrix are correct minus D66 and B66, so i don't understand the way used by Hypersizer in order to calculate the coefficients.

If you want try i give u the geometrical and the material data.

I want obtain an equivalent isotropic laminate from a grid stiffner plate:

0 web thickness 4mm
90 web thickness 4mm
top face thickness 2mm
0 and 90 web height 20 mm
0 and 90 web spacing 120 mm

The equivalent 2 layers (a,b) plate has:

E_a=72400 MPa
E_b=2413   MPa
nu_a= 0.3
nu_b=0.0
rho_a= 2.8E-6
rho_b=1.836E-7


PM1 file from Hypersizer is:

PSHELL*                1              10             1.0               9
*                   12.0                                 .8140377725E-02*
*            .393701E-01    -.826772E+00               7
MAT2*                 10  .207380947E+06  .477358722E+05  .000000000E+00*       
*         .207380947E+06  .000000000E+00  .557168721E+05  .000000000E+00*       
*         .179173311E-04  .179173311E-04  .000000000E+00  .211111111E+02
MAT2*                  9  .750193050E+07  .159119574E+05  .000000000E+00*       
*         .750193050E+07  .000000000E+00  .218078291E+05  .000000000E+00*       
*         .247100420E-07  .247100420E-07  .000000000E+00  .211111111E+02
MAT2*                  7  .530914537E+06  .000000000E+00  .000000000E+00*       
*         .530914537E+06  .000000000E+00  .294952396E+03  .000000000E+00*       
*        -.180000000E-06 -.180000000E-06  .000000000E+00  .211111111E+02

The wrong value are the red values.
B66 has to be equal to zero while D66 has to be 0.1856E+05 instead .218078291E+05.


Thanks a lot.

Marco

[Stressed]

This process should be documented by Collier. We have spent much time trying to understand the assumptions used. Collier will state this is standard text book stuff, but when a stiffener is included on a panel, the assumptions used to calcualte the stiffness properties need to be clarly defined by either TAGS with help notes or links to a detailed stiffened panel example showing the process and assumptions.