isponmo
Aerospace
- Jul 20, 2012
- 39
Dear Eng-tips community,
I am running a FE analysis in Hyperworks and I have some issues with one of the model checks. In particular, when I run a free-free check analysis with GROUNDCHECK card, the resultant strain energies are above the predefined threshold, leading to a fail flag:
The results of rigid body checks of the whole stiffness matrix:
Print results in all six directions against their limits
Direction Strain Energy Limit PASS/FAIL
--------- ------------- ------------- ---------
1 -1.569670E-04 1.987832E-01 PASS
2 6.098720E-04 1.987832E-01 PASS
3 -9.923881E-04 1.987832E-01 PASS
4 1.829827E-02 2.285839E-02 PASS
5 3.144717E-02 2.285839E-02 FAIL
6 1.773684E-02 2.285839E-02 PASS
Some possible reasons may lead to the failure:
1. CELAS elements with one end grounded.
2. Mis-aligned CELAS elements.
Possible Solution: Convert mis-aligned CELAS to CBUSH.
The results of rigid body checks of the SPC constrained stiffness matrix:
Print results in all six directions against their limits
Direction Strain Energy Limit PASS/FAIL
--------- ------------- ------------- ---------
1 -1.569670E-04 1.987832E-01 PASS
2 6.098720E-04 1.987832E-01 PASS
3 -9.923881E-04 1.987832E-01 PASS
4 1.829827E-02 2.285839E-02 PASS
5 3.144717E-02 2.285839E-02 FAIL
6 1.773684E-02 2.285839E-02 PASS
Some possible reasons may lead to the failure:
1. Single point constraints which prevent rigid-body motion.
My FE model is a sandwich panel in which the core is modelled as solid elements and the facesheets as shells of composite material. There are some rbe2, rbe3, point masses and CBUSH elements attached to it, but no CELAS at all. There are also several parts attached to each other by means of the CONTACT card, using a glue algorithm. In order to find out where the problem is, I run several instances removing one feature at a time, and I found out that there are no problems with the rbe2, rbe3, point masses, and bushes. However, by removing the CONTACT card I get rid of the FAIL flag.
The questions are: how can the contact algorithm be causing the problem? how could I fix it? I need to glue several components to the sandwich panel, so I need to use the CONTACT card since the meshes are very different and making coincident meshes would be almost impossible. Would it be acceptable to use different thresholds for the strain energies?
Thank you very much for your help.
Best regards,
Israel.
I am running a FE analysis in Hyperworks and I have some issues with one of the model checks. In particular, when I run a free-free check analysis with GROUNDCHECK card, the resultant strain energies are above the predefined threshold, leading to a fail flag:
The results of rigid body checks of the whole stiffness matrix:
Print results in all six directions against their limits
Direction Strain Energy Limit PASS/FAIL
--------- ------------- ------------- ---------
1 -1.569670E-04 1.987832E-01 PASS
2 6.098720E-04 1.987832E-01 PASS
3 -9.923881E-04 1.987832E-01 PASS
4 1.829827E-02 2.285839E-02 PASS
5 3.144717E-02 2.285839E-02 FAIL
6 1.773684E-02 2.285839E-02 PASS
Some possible reasons may lead to the failure:
1. CELAS elements with one end grounded.
2. Mis-aligned CELAS elements.
Possible Solution: Convert mis-aligned CELAS to CBUSH.
The results of rigid body checks of the SPC constrained stiffness matrix:
Print results in all six directions against their limits
Direction Strain Energy Limit PASS/FAIL
--------- ------------- ------------- ---------
1 -1.569670E-04 1.987832E-01 PASS
2 6.098720E-04 1.987832E-01 PASS
3 -9.923881E-04 1.987832E-01 PASS
4 1.829827E-02 2.285839E-02 PASS
5 3.144717E-02 2.285839E-02 FAIL
6 1.773684E-02 2.285839E-02 PASS
Some possible reasons may lead to the failure:
1. Single point constraints which prevent rigid-body motion.
My FE model is a sandwich panel in which the core is modelled as solid elements and the facesheets as shells of composite material. There are some rbe2, rbe3, point masses and CBUSH elements attached to it, but no CELAS at all. There are also several parts attached to each other by means of the CONTACT card, using a glue algorithm. In order to find out where the problem is, I run several instances removing one feature at a time, and I found out that there are no problems with the rbe2, rbe3, point masses, and bushes. However, by removing the CONTACT card I get rid of the FAIL flag.
The questions are: how can the contact algorithm be causing the problem? how could I fix it? I need to glue several components to the sandwich panel, so I need to use the CONTACT card since the meshes are very different and making coincident meshes would be almost impossible. Would it be acceptable to use different thresholds for the strain energies?
Thank you very much for your help.
Best regards,
Israel.
