News: Need training? HyperSizer Training Videos are available now! Learn more here: https://hypersizer.com/trainingevents/e-learning/

Author Topic: Design-to-Loads and Object Loads  (Read 33195 times)

Ruben

  • *
  • Posts: 40
    •  
Design-to-Loads and Object Loads
« on: December 12, 2013, 11:54:08 AM »
Hello,

In my FEM I have modelled the skin nodes at the OML with no offsets. In HyperSizer 6.4, I have chosen the Unstiffened family, single stack concept. In the Options tab I have set the Reference plane to 'Top' to indicate that the skin is at the OML. This will therefore introduce additional moment terms into the analysis, however I am not clear as to how these are calculated. Also, in the example that I am looking at, the moment terms in the Object Loads tab are different again, and these higher loads appear to be used in the sizing calculations.

My question is therefore, how are the moment terms calculated in the Design-to-Loads tab and the Object Loads tab.

Regards,
Ruben

James

  • Administrator
  • *****
  • Posts: 286
    •  
Re: Design-to-Loads and Object Loads
« Reply #1 on: December 16, 2013, 06:04:04 PM »
Hello Ruben,

For solid laminates the design-to moments and object moments are described below.

Moments on the Design-to loads tab & FBD tab are at the HyperSizer reference plane.
Moments on the Object Loads tab are at the Midplane of the laminate

The design-to moments (on the FBD tab and the design-to loads tab) are derived from the FEA forces + virtual moment. The virtual moment is calculated as the moments required to transform the loads from the FEA reference plane (detected on import) to the HyperSizer reference plane (set on the options tab). When the PCOMPs are imported, the Zoff (offset) is read in from the .bdf. Then if the reference plane is not modified in HyperSizer, the virtual moments should = 0, and the FEA moments are displayed on the design-to loads tab. If you modify the reference plane, HyperSizer must add additional moments to transform the Nx, Ny and Nxy to the new reference plane. This additional moment is called a "virtual moment" and is calculated by HyperSizer and superimposed with the FEA moments. 

The moments on the object loads tab, are the moments correspond to the loads being applied at the midplane of the laminate. So if you have the Nastran Zoff = 1/2 laminate thickness, then the moments on the object loads tab should = FEA moments.

For more information, see: http://hypersizer.com/download.php?type=analysis&file=HyperSizer_Stiffness_Method_-_Thermoelastic_In-Plane_Stiffness_Formulation.HME.pdf
Section 2.1.2, Load Eccentricity.

Ruben

  • *
  • Posts: 40
    •  
Re: Design-to-Loads and Object Loads
« Reply #2 on: January 07, 2014, 09:55:54 AM »
Hi James,

The original bdf has a PSHELL property with a 3mm thickness, with no offset. In HyperSizer I have set the Reference plane to Top, such that the nodes represent the OML. The optimised thickness is 24mm.

I can calculate the forces and moments on the element from the f06 file and match the numbers in the FBD tab, such that:

Nx, Ny, Nxy = 111.8, 431.0, 378.4 N/mm
Mx, My, Mxy = 9.9, 14.9, 5.6 Nmm/mm

Applying the 12mm offset I can calculate the Object Loads, which have the same in plane loads, with updated moments, such that:

Mx, My, Mxy = 9.9-111.8*12, 14.9-431.0*12, 5.6-378.4*12 Nmm/mm
            = -1331.7, -5157.1, -4535.2 Nmm/mm

However, I'm not sure how the Design-to-Loads have been calculated, as these are not the same as the FBD tab, and their calculation is not mentioned in the pdf that you reference.

FYI, in the Design-to-Loads tab, the following moments are shown:

Mx, My, Mxy = 1042.0, 4012.3, 3521.3 Nmm/mm

Regards,
Ruben

James

  • Administrator
  • *****
  • Posts: 286
    •  
Re: Design-to-Loads and Object Loads
« Reply #3 on: January 07, 2014, 10:52:32 AM »
The design-to loads tab reports the statistical processing of the FEA loads extracted from the FEA results. These are the raw loads, before transforming them to a new reference plane. The "strength" loads are computed based on the selected FEA Load Extraction method (2-sigma, element based, etc.). The "buckling" loads are the average compression loads in the component. The controlling case should match the loads displayed on the FBD tab.

Do you have the "Zero-out FEA Computed Moments" option active on the FBD tab? This will add additional moments to the FEA loads that will enforce zero X and Y curvature.



Ruben

  • *
  • Posts: 40
    •  
Re: Design-to-Loads and Object Loads
« Reply #4 on: January 08, 2014, 04:16:59 AM »
The moments on the "strength" and "buckling" loads appear to take into account the optimised offset in some way, such that the moments on the controlling case do not match the loads displayed on the FBD tab. I do not have the Zero-out FEA Computed Moments selected.
I've taken some screen grabs of the various tabs, which you can see in the attached spreadsheet. (HS 6.4.62)

Regards,
Ruben

James

  • Administrator
  • *****
  • Posts: 286
    •  
Re: Design-to-Loads and Object Loads
« Reply #5 on: January 13, 2014, 12:04:05 PM »
The FBD tab and the design-to loads tab should stay in-sync. Something that may cause these two forms to report different moments is if you change the reference plane and don't immediately reanalyze. Can you try to reanalyze the component? Do the FBD tab and Design-to loads tab still show different moments? If so, can you send the database?

-James

Ruben

  • *
  • Posts: 40
    •  
Re: Design-to-Loads and Object Loads
« Reply #6 on: January 14, 2014, 07:00:49 AM »
James,

Rather than email you the database and results, I've attached a text file of an example panel and tell you how I analysed the data.

Perform a NASTRAN analysis on the file.
Create a new HyperSizer 6.4 database.
Preferences / Analysis Defaults: Limit 0.667, Ultimate 1.0, FEA Stat Method = 2 sigma.
Change Units to SI, so Weight in kg, Length in mm, Force in N.
Import FEM.
In the Sizing Form, create a new group in the Unstiffened Family.
Add the Component to the Group Membership.
In the Concepts tab, deselect Honeycomb and Foam.
In the Variables tab, for the Top Face, set Min=2, Max=25, Perm=47.
Select the "Al 2024" material.
Perform the analysis.
The component result should be 16.5.

The FBD data is for BUCKLING and is equal to the Buckling Load Set Controlling Unfactored on the Design-to Loads tab. Object Loads are equal to the STRENGTH case.

Now change the Reference Plane in the Options tab to 'Top' and Analyze.

The FBD momemt terms are at the same as before, so not transformed to the reference plane.
The moment terms on the Design-to Loads tab for the Buckling Load Set Controlling Unfactored are different.
The moments on the Object Loads are also different as these are now at the centre of the Reference Plane, and these are correct.

Regards,
Ruben