Composite girder Deflection

[david_jefferson]

Hi 

I would be grateful if someone could help me with Autodesk SBD,

It is important for me to obtain an accurate estimate of girder deflection allowing for construction staging and creep and shrinkage effects.

I created a model to test the predicted deflection against hand calculations and other software.

The problem is I'm getting significantly differing results from hand calculations other software.

The main difference appears to relate to long term deflection under dead load and superimposed dead load.

SBD calculates a long-term deflection "dead stress increment" of 78mm.

The only dead loads are:

Self weight of girder = 13.75 KN/m

Weight of topping slab = 9.0 KN/m

SDL = 6.8 KNm

The If I add deflection "at erection" to deflection from "dead stress increment" i get a total deflection of -10mm ie 10mm down (sag).

For the same beam under the same design actions (including creep and shrinkage) MIDAS is predicting a long-term upward deflection of 50-60mm 

Any help would be appreciated.  

[dave_geeves]

Hi,

Thanks for including your data file, which I successfully downloaded and opened.

I don't have Midas so I decided to make a comparison by building a stage construction structural model of a single beam using a long term elastic modulus and applying the various UDLs and Moment loads for each stage.  I used the UDL's that you supplied in your post and extracted the prestress moments, the prestress losses (before and after composite) and the differential shrinkage moments, from the beam analysis stress summary table.

The moments were applied as nodal moments (one positive the other negative) at each end of the beam 

The results of the analysis showed the following breakdown.

showing a total value of -11mm.

I know that this method is a little crude but it should provide a reasonable estimate of the build up of deflection for checking purposes.

I find it strange that Midas is predicting such a large upward deflection at mid span as the sums of the various moments here are definitely positive so that the curvature of the beam at this location will be of a sagging sense.  I am wondering if differential shrinkage and SDL have been included in the Midas analysis, as you don't mention them in your post?

Sorry I can't be of more help at this point, but I hope this helps a bit.

Kind regards

Dave Geeves