This connection is a part of a bigger assembly made of welded steel profiles.
The two beams in the picture are supposed to be bolted together using two steel plates on the sides and 16 bolts.
I already have the whole thing simulated in PrePoMax but now i wonder what the best way would be to find the force that needs to be held by the bolts.
My first thought was to create a surface contact on the contact faces and then check the surface contact force with a history output.
Is there a better way?
Or use *SECTION PRINT if it’s modeled as a single compound part. Section print is not supported in PrePoMax yet but can be added using the Keyword Editor: Section print in history outputs
Probably fastening analysis is one of the greatest discussions in the FEM community. You will find lots of ways of doing this, each one with its advantages and disadvantages. Here are some options:
Model the splices and bolts as 3D geometry
Model the connection through interpolation elements + beams
Same as above, but using springs instead of beams
Estimating the flexibility of 5 rows of fasteners using hand calculations
And so on…
Using the last option you would find that each fastener takes different loads and the last ones on each side will take the highest load transfer. As @FEAnalyst said, you can use the section force to do this. Be aware that highly loaded joints in bending, torsion, or shear may not be suitable for this uniaxially loaded approach.
probably, using shrinkage effect such as thermal load or initial condition type stress/displacement can eliminate this condition. It will be use real beam element with section forces capable instead of imaginary/fictitious preload element.
So if I use a tied contact and then get the total forces on the surface and then use those forces for the manual calculation of a bolted connection, that would be accurate?
It might be best to use this approach and then compare it with the simplest one of those more advanced (e.g. model the plates and bolts as springs). Otherwise, it’s hard to say how inaccurate the simplified approach will be.
external link at CalculiX forum provide is only for example to use *Contact Print with CF,CFN,CFS options. In test case attachment using Tied Contact type, to get accurate results of bolt response needed Penalty contact instead.
As pointed out by @ANYS, in this case, it is a double-shear problem, but I showed the plot for single-shear. The right one should be 5 fasteners instead of 10 being analyzed. Again, the first and last ones (from each side) must have the highest load transfer.
I still need to think a little about them having different magnitudes in the worksheet. My first thought is that they should be equal (symmetric load transfer about the 3rd fastener) .
.