I have to model a 4-point bending test that I conducted in the lab on composite decking floor systems. The setup consisted of steel decking with a Fibre Cement Board fixed on the top and a CaSi board fixed on the bottom ( steel decking is sandwiched between the two boards). I am having massive trouble with convergence due to the model having very large contact areas (I have tried both contact modelling approaches, general contact and contact pairs) and how to model my fixities that I used in my experiment (Self-drilling screws) to fix the boards to the decking (what connector section type to use). Attached below is a photo of the experimental setup and the inp file of the model that I have so far. Any any help would be very much appreciated.
Dear Jaleel,
Contact is the one of the most problems in modeling. try to use the surface to surface contact (Standard) in four point bending.
Dear Salar'
I have most certainly learned this the hard way, I have been using surface to surface contact pairs most of the time or general contact. I get too many discontinued iterations caused by openings and overclosures during the analysis. I need to start using advanced techniques to make the problem converge.
Im also struggling with the modelling of the fixities that I used in my experiment, as to how to model the self-drilling screws that I used in my setup.
Dear Jaleel,
You need to specify the user task: the type of expected load should determine the requirements for the contact form. For example, if you expect loads of a transport nature (loader), it is worth using a spherical "probe", if static (snow, on the roof or furniture in the apartment) - the nature of the contact flat panel/flat panel.
the I beams shown in the photos will be fixed of reality by bolts or will be welded , so use the friction between the two surface as tie , it will be converge , Also recheck the fixation of the deck itself ( must be constraint only for one direction )
Plenty of useful suggestions already given by experts above.
As you said you have Convergence problem, it might be due to the fracture energy value as well. You may check (change) this value and find it useful.
Ahmed Hassan Nada
The beams seen in the photo is only used to apply the load from the actuator, and are not involved with the analysis. Its the screws that I used to attach my boards to the decking system that I am referring to. I have been trying different combinations of connectors types but get the same error message. "there are unconnected regions in the model' .I used attachment lines to model the nodes that need to be constrained between surfaces, then assigned connector sections to it. Im not using fasteners as I need 2 of the translational dof to be unconstrained.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Inorganic Materials
- Metals
- Alloys
- Steel