i already know, pre-tension of force with solid element has been implemented in PrePoMax. however, it seem required high computer resource due to large number of element and nodes. will be accumulate with counting number of the bolts.
the strength and behavior of bolt with pre-tension only depend on faying surface and frictional resistance (slip critical joint type). so it could be simplified by beam element and nut/head surface coupling.
above picture was taken form official examples pret3.inp
some interested benchmark of both model approach has been done by another here. i did not further review personally, but i think it will be shown as is.
may it would be useful feature when PrePoMax can generates automatically of beam element, dcoup3d, additional nodes and equations.
Right, 1D modeling of bolts is very useful. But maybe it would be better to implement the components of such a model (distributing coupling and beam elements) first, e.g. to make it possible to create 1D bolt models without pre-tension.
Also, there’s a quite important limitation of this approach in CalculiX. Here’s a quote from the documentation regarding the *PRE-TENSION SECTION keyword:
For beam elements a linear multiple point constraint is created between the nodes belonging to the beam element. The beam element itself is deleted,i.e. it will not show up in the frd-file. Therefore, no other boundary conditions or loads can be applied to such elements. Their only reason of existence is to create an easy means in which the user can define a pretension.
i was tried in the past, but failed due to rotational movement in which dcoup3d element not allowing. beam element (bolt shank) must transmit bending by couple of shear forces between head and nut positions.
it may something i missed or another approach is required.
In general I would love to see full 1D modeling in ProPoMax, I used that in Abaqus more frequently than I expected. You can save a ton of time using beams when they are a good match to the structure.
At Boeing we did use beam elements for fasteners in medium complex Abaqus models and beams and shells in everything highly complicated. As far as I know a properly applied fastener really shouldn’t need to be modeled with friction unless you are looking at highly detailed failure cases in which case you need the whole fastener for bearing, bending etc. In structure it is really the preload and stiffness that is of interest for how the load moves.
Speaking of 1D elements you simply can’t do complex structures like full aircraft structures in 3D elements, no matter how much computer power we have. Particularly with skin and rib structures like aircraft even doing the stringers and beams with surfaces gets intensive. All the Boeing 787-8 global loads come from Nastran/Patran models made using 1D and 2D elements (and something like 1200 cases when left 10 years ago) only details are evaluated using 3D elements.
i’m thinking again about visualization. if it’s a problem, pre-tension force maybe changed to initial conidiation type stress by equivalency.
right, this is classical approach to apply pre-stressed of element. also detailed similar examples from @mkraska available at GitHub.
this feature of bolt pre-tension forces using beam element has a purpose in significantly reduced computational times.
however, all of method still doubting me for some reasons.
pre-tension section (beam&solid), how about partially applied force. mixed between friction/slip and bearing type?
thermal load, how about the model subjected to fire in second steps?
because of it, i have build a models by following some guidelines using another approach. the force applied at bolt and nut surface in opposite directions.
