Learning Forum

[Matthew Pirkle]

I am trying to obtain the nodal reaction force on a composite laminate due to an applied displacement over a number of load steps. The composite laminate has 24 laminas. The goal is to determine the reaction force of the constrained edge of the composite laminate at the onset of damage for a given load step and substep due to an applied displacement along the free edge of the laminate. I am simulating a simple tension test of a composite laminate, where SHELL281 elements are used for each lamina.

The current approach is to list the all the nodes, and then find the node numbers at the constrained edge (x = 127 mm). Once these particular node numbers are found, the nodal loads for Fx are listed (List > Results > Nodal Loads > Fx). All of those nodal loads are then summed to obtain the nodal reaction force for each lamina (0, +-45, 90 degrees) at a particular load step and substep. Once all of the nodal reaction forces for all nodes at x = 127 mm for the 0, +-45, 90 degree laminas at a given load step and substep have been found, they are multiplied by 6 (4 different fiber orientations*6 laminas for each fiber orientation = 24 laminas) and summed to obtain the reaction force of the laminate at the constrained edge (x = 127 mm). The applied displacement is at x = 0 mm. The nodal reaction force values returned by ANSYS are negative, as one would expect. Is this approach correct to obtain the nodal reaction force of the laminate?

 

The concern is that once the nodal reaction forces are summed for all nodes at the constrained edge for all 24 laminas, the total nodal reaction force exceeds the strength of the laminate. Is it possible that PDA does not consider total failure of the laminate? I am using continuum damage mechanics (CDM) with progressive damage analysis (PDA).

 

Also, the node numbers at the constrained edge (x = 127 mm) are the same for each of the laminas (0, +-45, and 90 degree fiber orientations). Does the nodal reaction force for a given lamina account for all the other laminas with the same fiber orientation? For example, does the nodal reaction force for the 0 degree lamina include the reaction force for all of the 0 degree laminas? If there are 6 laminas that have a 0 degree fiber orientation, is the nodal reaction force for a selected 0 degree lamina (ex. RSYS 11 for LAYER 1) a summation of all the 0 degree laminas in the composite?

 

Thank you so much!

[Chandra Sekaran]

Also, the node numbers at the constrained edge (x = 127 mm) are the same for each of the laminas (0, +-45, and 90 degree fiber orientations). Does the nodal reaction force for a given lamina account for all the other laminas with the same fiber orientation? For example, does the nodal reaction force for the 0 degree lamina include the reaction force for all of the 0 degree laminas? If there are 6 laminas that have a 0 degree fiber orientation, is the nodal reaction force for a selected 0 degree lamina (ex. RSYS 11 for LAYER 1) a summation of all the 0 degree laminas in the composite?

The solver calculates the total nodal forces. There is no differentiation for different lamina. You can use the command PRRFOR to print the reaction forces at the constrained edge. The PRRFOR will also give you the sum of the reaction forces at the end of the reaction listing. I do not see a need to multiply the nodal reaction forces (or nodal forces) by 6. 

If you have other constraints then you can select just the nodes on the edge (NSEL command)  and then list the reaction forces (PRRF command).

 

[Matthew Pirkle]

[Matthew Pirkle]

 

 

Chandra,

 

I understand! Multiplying by 6 made the nodal reaction force too high, which didn’t make physical sense.

 

Just to reiterate and confirm with you, what I have been doing to obtain the total nodal reaction force of the laminate at the constrained edge (x = 127 mm) at the time of damage initiation for a given damage mode, say MT in the +45 lamina, is to find the load step and substep when damage first appears in the +45 lamina. For that particular load step and substep, I then sum the Fx nodal reaction forces at x = 127 mm for the 0 degree fibers (Layer 1, RSYS 11), sum the Fx nodal reaction forces at x = 127 mm for the 45 degree fibers (Layer 4, RSYS 14), sum the Fx nodal reaction forces at x = 127 mm for the -45 degree fibers (Layer 5, RSYS 15), and sum the Fx nodal reaction forces at x = 127 mm for the 90 degree fibers (Layer 10, RSYS 20). Finally, I add those 4 summations together to obtain the total nodal reaction in the laminate at that load step and substep. So let’s say for the 0 degree layer summation I get -4000 kN, for the +45 layer summation I get -3000 kN, for the -45 layer summation I get -3000 kN, and for the 90 degree layer summation I get -.00001 kN, my answer for the total nodal reaction force in the x direction at that particular load step and substep is -10,000.00001 kN.

 

Is the above approach correct to obtain the nodal reaction force of the laminate at the constrained edge?

 

Thank you so much for your help!

 

 

[Chandra Sekaran]

The nodal force (NFORCE or FSUM or PRRS) is for the entire element i.e. all the layers. There is no separate force stored for each layer.

[Matthew Pirkle]

Chandra,

 

So there is no need to sum the 0, +45, -45, and 90 layer Fx nodal reaction forces at x = 127 mm for a given load step. If I want the nodal reaction force for the laminate in the x-direction (the direction the tensile load is applied) I simply choose the load step and substep I want, choose the 0 degree layer (fibers oriented in the x-direction) and use NFORCE and sum the nodes that correspond to x = 127 mm.

 

Thank you!

[Matthew Pirkle]

Chandra,

 

Does the above sound correct?

 

Thank you for your time!