Hello, I am new to the world of FEA - my apologies if this is somewhat trivial.
I am modelling an aircraft wing. I would like to model only the ‘wingbox’, which is the load-bearing structure highlighted in green in the attached image. However, I currently need to also model the non-load-bearing leading and trailing edge sections. This is because the aerodynamic loads are applied as pressures on each element of the wing, through the use of *DLOAD. Hence, I cannot currently remove the non-green areas from my model, as I will then lose the aerodynamic forces which act on those elements and my overall aerodynamic force will be less than reality.
My questions then are:
Is it possible to keep the non-green areas in the model for the purpose of applying the pressure loads, but then NOT have them included in the FEA calculation?
Literature points to the use of so-called ‘rigid elements’, to apply the equivalent loads of the leading and trailing-edge areas onto the 2 beams in the wing, along the length of the wing. Is this the correct method and if so, how would this be done in CCX? - I have read the documentation but cannot seem to wrap my head around this. Any help or examples would be appreciated.
hi, I don’t have an answer for you on this. but it’s a useful feature idea. i’ve talked to the developer of Mecway about adding something similar. Right now, what I do is break the surface mesh up so that nodes are exactly where my loads are.
You can make some regions of the model rigid but this may of course change the global response. Instead, it might be better to use submodeling or substructures to speed up the analysis. But it also depends on your model - if it’s something similar to the OP’s case or a different kind of structure.
I would calculate the loads in discrete points spanwise and chordwise. With the loads known, apply them to the nearest hard points of the wing such as skin to spar, skin to stringer, skin to rib to stringer, and so on. The best way of doing this would be to use interpolation elements such as DCOUP3D, or COUPLING with DISTRIBUTING variant. This way, concentrated stress due to load application would be mostly avoided as well as artificially rigid areas.
Usually, the leading and trailing edges are of no interest to the wing’s general structure response, but it is not a big problem if you want to include them as well. Keep in mind that their sizing load is the direct pressure of the airflow (bird strike as well but that is another story ) for the LE and control surface loads for the TE.
I would personally use submodelling as @Calc_em suggested. Separate your load cases of interest and store the results from the main box to use them in TE and LE calculations.
Thanks for the answer, one more question though for clarification. I specifically am trying to run buckling analysis on a wing, but the LE and TE are buckling before the wing box. I need the TE and LE modeled because they apply some twist to the wing box from the airflow pressure, but I’m not actually sizing them. Would the methods you provided be applicable for this case?
this is a standard method in aerostructures analysis, in Nastran RBE3 elements are used (not really rigid but interpolation elements).
In CCX you have distributing coupling. Typically the “rigid” element is attached to rib-spar corners and the reference point placed in the wing elastic axis.
As in this paper (picture extracted from it): https://www.dglr.de/publikationen/2016/420186.pdf
More details about methodology on this book (however based in Nastran code, but you can find equivalent tools in CCX): FEA Academy - The Book
) for the LE and control surface loads for the TE.
