Modal mass missing in higher mode shapes

[MRCH94D9W]

Hi

I'm experiencing some incorrect modal masses calculated by Robot 2020.

To investigate I have made a 10m simply supported IPE100 beam, and checked the mode shapes.

As per analytical calculation, all "sinus mode shapes" should have the modal mass of: 1/2*8.34kg/m*10m = 41.7kg

This fits quite well with mode shapes 1-3. However, when looking at mode shape 4, the modal mass drops to 26kg and even further beyond that.

I have traced it to what I think is a discrepancy in the normalized eigenvectors.

Looking at mode shape 1 the normalized eigenvector has a value from 0 to 1, mode 2-3 has a value from -1 to 1, but mode 4 has values from -0.8 to 0.8.

So, why this incorrect normalization, and is that what causes the missing modal mass?

I'm using Lumped mass matrix with rotations. I have tried discretizing the beam by 10-100 nodes, but it doesn't affect the modal mass to the extend I need.

The eigenfrequencies seem to be correct though.

I have attached the model as a zip file, as the forum doesn't allow me to attach it as *.rtd file.

Kind regards

Martin

[Simau]

Hi @MRCH94D9W 

Could you do the same analysis for the example attached?

[MRCH94D9W]

Same thing happens 😕

[Simau]

And what about this example?

Look at mode 1-6 and mode 7-20

Add another beam and look what happens

[MRCH94D9W]

By introducing the extra beam you have changed the static system.

Only the odd modeshapes are comparable, i.e. 1 = 1, 2=3, 3=5, 4=7 etc. and the modal mass should be divided by 2 to be comparable.

We still see the same issue

• Shape 1 (1) mZ = 81kg --> 40.5 kg
• Shape 2 (3), mZ = 90kg --> 45 kg
• Shape 3 (5), mZ = 81 kg --> 40.5 kg
• Shape 4 (7), mZ = 53 kg --> 26.5 kg

[Simau]

I must look more closely at this subject, but not for now

Maybe someone else has an idea

[MRCH94D9W]

Can someone from Autodesk please address this issue?

Would be prudent to identify whether this is a modelling issue or if it is an error in the software.