Offset to varying RC deck using FEM

[mohamed.zakir9GETY]

Hi,

I'm currently assessing the bridge approach/backspan. We are planning to replace the centre suspended span and check the adequacy of the backspans to the updated loading condition. I have created the initial model and applied the section properties (please see attached screenshot). The backspan is varying RC deck and in ASBD I have assigned Grillage to the suspended span and FEM to the backspan.

I see there is an option to offset the FEM-Beam member. However, is there an option in ASBD to offset the FEM-shell element to allign with the carriageway level? Your assistance will be of great assistance.

Many thanks,

[dave_geeves]

Hi,

Unfortunately there is no option to to have shell finite element offsets so the node points will always be at the mid height level of the elements.  My recommendation is to model the backspan as a grillage of beam members so that member offsets can be applied.  I would also define the longitudinal and transverse members of the back-span using design beams.  Reinforcement moments, using ether the Wood Armer method for designd, or Denton Burgoyn method for Assessment can be obtained from a grillage as well as an FE slab so this may be useful for the verification process.  If you use design beams for the longitudinal and transverse members then don't forget to halve the  beam torsion constant in both directions, otherwise you will be doubling up on the torsion stiffness of the slab.

I hope this helps and if so, please mark my reply as a solution so that others may benefit. Thanks.

Kind regards.

Dave Geeves.

[mohamed.zakir9GETY]

Hi Dave,

Thanks for your reply. I have proceeded with the model using the FEM. I might need to do a lot of alterations to change the back-span to grillage and connect the piers and abutments ribs to the deck slab. 

Say, if I don't change the offset and undertake the assessment as it is, will it affect on the results? 

Many thanks,
Zakir

[dave_geeves · Accepted solution]

Hi,

It is quite difficult to answer this definitively as it will depend on other modelling aspects. In general it will have some affect on the results  but this may be quite small, especially if the bearings on the bank seat are able to slide in lateral movement so that no membrane forces will be generated. (ie the back-span is essentially a propped cantilever)  Any membrane forces will act at the element mid level, which is not of course at the centroid of the actual slab so this may underestimate the actual moments in the real structure. If you were to make the assumption that it will not make a significant difference, I would check out this assumption by creating a simple line beam model (using a refined model option but with just one beam) with the beam members connected so that the centroids are at the same level as the detailed model and the support conditions are also the same.  This will highlight the difference that this assumption may incur.

I hope this is reasonably clear.

Kind regards

Dave Geeves