Learning Forum

[gb17]

When performing modal analysis, the mode shapes and their corresponding frequencies are the eigenvectors and eigenvalues obtained when solving the governing equations. I am solving similar designs in Mechanical APDL, where everything is exactly the same except for slight differences in the design geometry and the meshes, yet the mode shapes are sometimes being returned in reverse (i.e. they are backwards) when compared to the other solutions. I can easily fix them by multiplying the eigenvector and eigenvalue by -1, but ideally the eigenvectors would always be in the same direction. Is there a way to fix this?

[Aniket]

Can you please explain with images what do you mean by reversed mode shapes?

-Aniket

How to access Ansys Online Help Document

How to show full resolution image

Guidelines on the Student Community

How to use Google to search within Ansys Student Community

 

[gb17]

I do not have an image of the mode shapes readily available. The x, y, and z displacements of every node are opposite of what they are supposed to be, based on the other tested designs. For example, mode one of most designs might have a displacement close to 1 for the first node, while another design has a displacement close to -1 for the same node.  When this occurs, it is true for every displacement of every node. Its like every displacement has been multiplied by -1. The relative magnitudes and shapes of the mode are the same, just reversed of each other.

[gb17]

Here are some images to show what I am talking about. The first image shows the mode 1 node x-displacements of the first 12 nodes (columns) for 20 different designs (rows). There are a lot more rows and columns to my data but this shows what is going on. Mode 1 does not have any mode switching, which is why all of the displacements shown here are positive (i.e. they are all in the same direction).

The second image shows the mode 2 node x-displacements of the first 12 nodes (columns) for 20 different designs (rows). Mode 2 shows the problem I am having with the mode shapes being reversed. About half of the designs have positive displacements (red) and about half of the designs have negative displacements (blue) for the same set of nodes. They are of the same magnitudes but opposite in sign. I have to manually flip about half of the designs so that they are all oriented in the same direction. I have the same issue with modes 3, 4, and 5. These are the displacements from the modal analysis in ANSYS, so something weird is happening in ANSYS where it randomly flips the mode displacements.

[Jacques]

Hi gb, i have exactly the same problem, did you ever sort it out?nFailing that, how does one multiple the eigen vector and values by -1? I really cant see an option for this anywhere and am getting very fustratednCheersn