Cancel
Showing results for 
Search instead for 
Did you mean: 

Help with modelling rigid elements on a sandwich structure plate

Experimenter
Experimenter

Hi all,

 

First time posting here, but I have a Femap problem that is most likely very easy to clear up. I am modelling a composite sandwich structure in Femap for a chassis plate design with multiple elements attached to the plate. When I have meshed my plate with laminate properties, and I attempt to create RBE2 mass elements and tie them to my plate using rigid elements, I cannot run any correct analysis. I am curious to what basic things I may be overlooking as I am extremely new to the world of FEA and am running this analysis for a project, and being taught Femap in my spare time. I've tried to use RBE3 as well.

Please check out the neutral Femap file that I have attached and look at the analysis sets to get a picture of the problem. I've been running normal modes analysis to try and see where my errors are. I think that it may have something to do with rigid elements in general, but like I said I am very very new to this and doing this all in my spare time. Any thoughts are appreciated! Thank you, Ethan

file : https://drive.google.com/open?id=1-P7BS4a_m1Oh3nZhMdiPg3W5a5OiTXv4

 

 

4 REPLIES

Re: Help with modelling rigid elements on a sandwich structure plate

Siemens Phenom Siemens Phenom
Siemens Phenom

I'm not sure what you mean by 'correct analysis'.  I was able to run this model without errors when using RBE3's to attach the Mass elements to the mesh.

 

However, I'm curious about your stiffness values for the material HC (honeycomb core material?).  These are very low values and can definitely contribute to suspect results.

 

Finally, when running a model with laminates, I recommend that you also enable the PARAM,SRCOMPS card as shown below to output Ply Strength Ratios.

 

2018-01-10_20-26-24.jpg

Best Regards,
Chip Fricke
Principal Applications Engineer - Femap Product Development

Re: Help with modelling rigid elements on a sandwich structure plate

Experimenter
Experimenter
Could you show me the result plot that you got in the analysis? The result
I got for the Laminate Failure Index was extremely low, around 0.000143,
which would imply that I had an extremely strong laminate that was nowhere
near failure.

Re: Help with modelling rigid elements on a sandwich structure plate

Siemens Phenom Siemens Phenom
Siemens Phenom

Have you looked at the stiffness of your laminate as shown below?  The bending stiffness is high, and with the small loads you have on this plate and with a almost zero stiffness of the honeycomb, you're getting higher strains on the XY plane vs bending.

 

I recommend that you run a test coupon model of this with loads in both axial directions and a simple fixed-fixed beam with a center load to see the behavior.

 

Laminate Equivalent Properties
 
11 Plies - Total Thickness = 0.037
 
In-Plane Properties
Ex = 3.1202E+10  Ey = 3.1202E+10  Gxy = 1.3577E+10
NUxy = 0.524589  NUyx = 0.524589
Alphax = 0.  Alphay = 0.  Alphaxy = 0.
 
Bending/Flexural Properties
Exb = 5.5038E+10  Eyb = 4.8873E+10  Gxyb = 2.5432E+10
NUxyb = 0.586824  NUyxb = 0.521089
Alphaxb = 0.  Alphayb = 0.  Alphaxyb = 0.

A Matrix
  1.60150E+9      8.44254E+8      -6.6241E+7
  8.44254E+8      1.60150E+9      -6.6241E+7
  -6.6241E+7      -6.6241E+7      5.05926E+8

B Matrix
  -1.1642E-8      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0

D Matrix
  3.37445E+5      1.77176E+5      -1.7413E+4
  1.77176E+5      2.99957E+5      -1.7413E+4
  -1.7413E+4      -1.7413E+4      1.08580E+5

A-Inv Matrix
  0.00000E+0      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0

B-Inv Matrix
  0.00000E+0      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0
  0.00000E+0      0.00000E+0      0.00000E+0

D-Inv Matrix
  4.30442E-6      -2.5259E-6      2.85213E-7
  -2.5259E-6      4.84742E-6      3.72294E-7
  2.85213E-7      3.72294E-7      9.31524E-6

 

Using RBE3s (the proper way to load a model using Mass Elements, I get a max Failure Index of 1.78e-5 as shown below.

 

2018-01-11_21-18-37.jpg

 

Best Regards,
Chip Fricke
Principal Applications Engineer - Femap Product Development
Highlighted

Re: Help with modelling rigid elements on a sandwich structure plate

Solution Partner Phenom Solution Partner Phenom
Solution Partner Phenom

Hello!,

As an add-on to great Chip comments if you run a normal modes/eigenvalue analysis using NX NASTRAN (SOL103) and plot the deformed mode shapes you can see the weird behaviour of the composite: you have a problem with material properties values for axial & bending stiffness.

COMPOSITE_weird-eigenvalue-results1.png

In fact, revising the material properties of HC material (core Honeycomb) I see that a value of EX = 1 Pa = 1e-6 MPa is used, this could cause to have a very low stiffness for shear & bending in the core:

COMPOSITE_weird-eigenvalue-results2.png

To learn more in composites take a look here:
http://www.iberisa.com/soporte/femap/composites/nafems_benchmark_composite_test_r0031_3.htm

To learn more about RBE3 vs. RBE2 take a look here:
https://iberisa.wordpress.com/2015/10/13/rbe2-vs-rbe3-on-femap-with-nx-nastran/

Best regards,
Blas.

Blas Molero Hidalgo, Ingeniero Industrial, Director
IBERISA • 48004 BILBAO (SPAIN)
WEB: http://www.iberisa.com
Blog Femap-NX Nastran: http://iberisa.wordpress.com/