News: HyperSizer.com has a Community Board and Customer Support System. Submit a ticket at http://hypersizer.com/ticket

Author Topic: Panel Buckling - SS v FF  (Read 25066 times)

garyjh

  • Client
  • **
  • Posts: 41
    •  
Panel Buckling - SS v FF
« on: July 05, 2018, 08:42:15 AM »
Been doing a lot of FEA and non-FEA project panel buckling analysis on monolithic composite panels, with all simply supported BC’s. Results acceptable and compare well with in house tools.

When changing Project Sizing- Buckling Tab Boundary Condition all four BC’s from SIMPLE to FIXED, the results are not acceptable and are nowhere near results from in house tools.

The example attached is from an FEA project, with all panels, loading and material properties the same, the only change is from simply supported to fully fixed. For SS, various MS’s but when FF all MS’s are between -0.97 and -0.99. This has been seen on other projects and by other users within my organisation.

Please can you provide some help or advice as to why this is happening and how best to perform fully fixed panel buckling, with confidence in the results. Thanks.
« Last Edit: July 05, 2018, 08:57:10 AM by garyjh »

Stephen

  • Administrator
  • *****
  • Posts: 70
    •  
Re: Panel Buckling - SS v FF
« Reply #1 on: July 05, 2018, 08:56:24 AM »
Hi Gary,

Can you check for me whether you have "exhaustive search" active for this project? It is a setting in the backdoor options form for the project. Depending on your version of HyperSizer (we have occasionally changed the descriptions for some of these settings), it should look like the attached image.

There is a known issue when running the energy solution panel buckling method with fixed boundary conditions while exhaustive search is active that can cause margins returned to be very conservative. The best advice to avoid this issue for now is to simply disable the exhaustive search function if you need to use fixed boundary conditions.

Exhaustive search was added to ensure that the energy solution will find the lowest-energy mode even for atypical component dimensions (and therefore m,n pairs). For "standard" geometry (panels without very high aspect ratios), this panel buckling method should return the lowest energy mode without exhaustive search active.

Exhaustive search is most often required for full cylinders, where there may be a large number of half-waves along the length, but only a few circumferentially. This is not a typical situation for rectangular panels,.

Let me know if you need any clarification!

garyjh

  • Client
  • **
  • Posts: 41
    •  
Re: Panel Buckling - SS v FF
« Reply #2 on: July 05, 2018, 09:02:26 AM »
Using v7.3.37. Does not look like that at all - see attachment.

Stephen

  • Administrator
  • *****
  • Posts: 70
    •  
Re: Panel Buckling - SS v FF
« Reply #3 on: July 05, 2018, 09:12:35 AM »
That first option ("Shear Buckling Force All Modes") is the equivalent option in 7.3.37. We have since renamed it in later versions.

Is it actually set to "No" on the projects in which you have noticed this behavior?

garyjh

  • Client
  • **
  • Posts: 41
    •  
Re: Panel Buckling - SS v FF
« Reply #4 on: July 05, 2018, 09:22:53 AM »
Yes. It was set to No as shown in the picture. Changed it to yes and made no difference. Gone back to No to double check and no difference.

However, found the following (see attached) in different backdoor data tab, Buckling Coefficients. Which is like your original reply. Changed to NO and looks OK now.

Thanks for quick reply and help.
« Last Edit: July 05, 2018, 09:34:46 AM by garyjh »