The following query relates to the appropriate use/setting of the torsional stiffness of two-way slab elements specifically in RAM Concept. Note that the default settings for a clean RAM file are as follows:
i) Slab torsional constant, kMrs = 1.0
ii) CS Torsion Design = None
There are many forum posts on ‘Eng Tips’ discussing what is the most appropriate stiffness modifier that should be assigned to the torsional stiffness of slab elements. The forums discuss FEA modelling in a general sense, not specifically for RAM Concept, however, the common theme is that the torsion stiffness should be reduced to say 30% (kMrs=0.3). Some threads go on to mention the need to include the torsional moments specifically in RAM Concept in the design strips (CS Torsion Design = Wood-Armer), otherwise they will be ignored.
1) In a post on Eng-Tips by Seth Guthrie, he notes that in RAM Concept specifically, when calculating deflections using the Load History method, the slab torsional constant should remain unchanged (kMrs=1.0), noting that ‘cracking of the slab is part of the design’. Please confirm that RAM Concept does indeed consider the reduction in torsional stiffness during the analysis and that the modifier kMrs is recommended to be taken as 1.0 for two-way slabs when using load history calculations? Therefore, would I be correct in saying that using a load history calculation with a forced reduction in torsional stiffness should be over-estimating deflection results?
Link to Eng Tips post: http://www.eng-tips.com/viewthread.cfm?qid=321443
2) Based on the above, what is considered reasonable for a strength (limit state) analysis? Is a separate ‘strength’ model with a reduced torsional constant (say, kMrs=0.3) more appropriate to better model the slabs behaviour as cracking occurs approaching the ultimate loading condition?
3) The default settings in RAM Concept appear to be contradictory? That is, 100% of the torsional stiffness is allowed to occur in the slab, whilst the design spans are set by default to not convert these twisting moments into additional moments in the primary directions. Is there a specific reason for doing this, as it would lead to an un-conservative design for strength? Please comment.
Thanks