There are several advanced approaches, but they need so much effort that in most cases solids are preferred. Especially since in other software 1D bolt models are helpful to easily define many bolts (with the help of some semi-automated tools or scripts) while in CalculiX the effort would be unreasonably large for more bolts. Unless, someone picks reliable enough approach and automates it too in some way (maybe with OS commands and scripts parsing text files).
right still prefer common approach as available in example, for pretension bolt simplification with imaginary beam element also. It has been requested available in GUI since keyword editors is not convenient to use.
Bolt pre-tension force with beam element - Feature Requests - PrePoMax
Those “disappearing” beams can be helpful if only preload is of interest, but in most cases, it is followed by some operational loading, and the bolt must still be present in the model.
IMO, even highly simplified, coarsely meshed, and not so ideally connected (via tie constraints or simple contact) solid bolts will still be superior in the majority of the cases, but it’s always good to have alternatives, even if they are very hard to define and highly limited. After all, some degree of automation is always an option, as I’ve mentioned above.
in building and bridge regulations did not allow this mixed type of bolts (friction and bearing) to act together meaning does not exist. Specific case such as in piping may exist due to small amount of clamping required to fit the surface.
it’s common in many FEA to provide several bolt modeling approaches i.e spring, beam and solid element. One modeling approach can not fit to all condition in facts, this related to computational resource and reliable times by trial and error in designing.
Ok, let’s stop at that since we have that large thread about bolt modeling approaches, and we’ll soon have another on the CalculiX forum. So let’s just sum up that the OP should try a solid (or at least beam) representation of the bolt to account for the preload.
since this look as detailed analysis shown by solid part with fine mesh it could be suited with truly solid element. However, the treads made me a bit confusion when spring is mixed with preloads. To get clear OP need to notice are bolt interaction (friction and bearing) at next step analysis will be considered or not. Even possible, it can be complex when initial preloaded and further spring response is actually needed in analysis.
Thank you very much to all for your answer and effort 
Since I need a reliable model and a stable approach I prefer to switch to solid bolt; using the tutorial as you provided.
Thank you again
Marco F.