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

Author Topic: Why use DSP?  (Read 23114 times)

David Hughes

  • Client
  • **
  • Posts: 3
    •  
Why use DSP?
« on: February 20, 2020, 10:22:25 AM »
Hi,

What are the benefits of sizing with Discretely Stiffened Panels compared to sizing the Stiffeners and Skins separately?

If the FEA and HS geometry match, would it be best to set the backdoor setting "use summed panel loads for object analysis" to "No" to avoid unusual object loads calculated when combining segment load components from the tension/compression loops?

I believe one advantage of DSPs is that more global failure modes can be considered, including flexural-torsional buckling, plastic bending and crippling (inc. effective skin). Is this right? Are there any others that require the summed segment loads?

Kind Regards,
David.

Brent

  • Administrator
  • *****
  • Posts: 26
    •  
Re: Why use DSP?
« Reply #1 on: March 11, 2020, 10:56:09 AM »
Hi David,

Benefits of sizing with discretely stiffened panels(DSP) compared to stiffeners and skins separately:
  • Like you mentioned, DSP can account for failure modes which are dependent on the geometry and loads of neighboring structure. With this information, the effective area for skins and stringers are known and can be used for various analysis criteria. In addition to the failure modes you mentioned, DSP supports compression post buckling.
  • Reduce the number of sizing and FEA iterations. In the stiffener direction, DSP assumes strain and curvature compatibility between the stiffener and adjacent skins. Therefore, when sizing, this approach analytically accounts for local load redistribution.
  • The Grid Stiffened plugin package, provided in the install, supports DSP modeling. This plugin package has text book implementation of methods in addition to capability for tuning conservatism. Some of these features are missing from the standalone skin and stiffener criteria.

If the FEA and HyperSizer geometry match and the panel segment definition is typical, the summed panel load analysis should produce similar results to the object based analysis. But, as in the case you mentioned, when there are peaking loads or load reversals in the uniaxial direction of a DSP, then the segment approach could inaccurately capture these load states when using the 0-Sigma load extraction technique. From our experience, this is a corner case that can be addressed by using object loads or a different load extraction technique.

Important Points:
  • DSP still runs tension and compression sizing loops so it does't average tension and compression element loads. The tension/compression sizing loops have independent segment loads and object loads. 
  • Only the loads in the direction of the stiffener are summed for panel loads. All other directions of loads are treated as object based.

Please reach out if more explanation is needed. Provided below is a link to the Grid Stiffened Plugin package documentation in our help system.

Link to documentation: http://hypersizer.com/help_7.3/#Udef/Plugin-HS_plugins_grid_stiffened.php%3FTocPath%3DAnalysis%2520Plugins%7CPublished%2520Plugin%2520Packages%7C_____4

-Brent