Learning Forum

[AR]

Hello,

I am performing a stress combination between a static and a harmonic analysis. However, the maximum values ​​are smaller than the maximum values ​​of the static-only case.

In the structural static analysis I have a maximum stress value of 528 MPa. In the harmonic analysis I have a very small stress value, around 0.1 Pa. However, the solution combination gives me a maximum of 390 MPa.
The combination I am performing is an SRSS with factors of 1 for the static analysis and 1 for the harmonic analysis.

Do these results make sense? How can this be explained?

 

[dlooman]

Keep in mind when dealing with harmonic results that there are two, the real and imaginary.  It doesn't matter in this case, but there is a apples and oranges relationship between a single static result and a complex harmonic result.  Obviously the SRSS of a large number with a small number should return the large number.  When you maximum stress, do you mean maximum equivalent stress?  That's a derived quantity and it's possible that by removing the signs of the component stresses by SRSS, the equivalent stress got smaller.  For example, a 2d state of stress where sx is positive and sy is an equal and opposite negative value will produce an equivalent stress of around sx + sy, but if sx and sy are both positive the equivalent stress would be closer to sqrt(sx^2+sy^2).  

[AR]

Hello Dlooman,

 

Thank you for your answer!

I understand what you mean about the real and imaginary parts, but if I extract the stress at a specific frequency and phase (136.21 Hz and 81º, for example), which I have identified as the moment when the maximum equivalent von Mises stress occurs, I wouldn’t have the issue of the result having both real and imaginary parts, right? If that’s not what you mean when you say it’s the same to compare static stress with harmonic stress, what are you referring to?

On the other hand, I think I understand your example about the sigma components and the importance of their signs in the von Mises equivalent stress. Therefore, I understand that the Solution Combination applies the Linear or SRSS method to the normal and shear stress components for each node, and then with the result of that, it calculates the equivalent von Mises stress (if I request it). Is that correct?

On the other hand, I have performed another combination of the results, this time using the linear method. The final result is much closer to what I would expect, that is, to the static analysis in terms of maximum values of von mises stress. As can be seen in the image below, the stress distribution below 180 MPa is very similar in both solutions, only the high stress values change. Does this mean it is more appropriate to perform the combination using the linear method for static and harmonic?

[dlooman]

Yes, I think linear is a better choice.  I agree with everything you say above, but I wasn't aware that a Solution Combination would allow you to specify a phase angle for the harmonic results.  If it does, then linear is especially good.  SRSS is necessary when combining results that have lost their sign such as Response Spectrum and Random Vibration results, but you don't have that.

[AR]

Ok,

 

Thank you for your response!