STAAD Pro Questions - Plate Shear Force and Stress Output

[BloEngineer]

Hello All,

First time posting on Eng-Tips, but I've been using the forum as a resource for many years. I am working on a fairly complicated finite element model using quadratic and triangular plate elements. I am at a point where I am comfortable with the model results given the geometry and loading and am now evaluating the results. I have somewhat "subdivided" my structures into separate walls and slabs and am designing the structural elements according to worst-cast reactions. I have a two-part question, the first being technical and the second being more output based:

1) Since most of my structural elements are two-way spanning, I am grabbing the worst-case Mx and My values and using them to design the structures outside of STAAD. Where I am getting hung up is designing for shear. I've done research on this area, but am still unclear about how to utilize results. The way I have typically derived shear force is to identify max SQY and SQX values x plate width/length (in direction of shear) x thickness of plate. I understand the theory of FEM in regards to force matrices, but am not a master by any means. I wanted to do a "dummy" experiment, to see how the method I described works if you vary slab thickness of the particular member I'm designing. My initial thought was that the shear force computed should be consistent for each slab thickness. In my dummy experiment, I varied the slab thickness as follows: 3', 2', 1'. What I found was that the shear force was highest for the 3' slab (19.2 kip), similar for the 2' slab (18.1 kip), and much lower for the 1' slab (12.8 kip). In my opinion, I think that the method I'm using is correct. My theory on why the shear force drops is that the model I have is three-dimensional and I am dropping the stiffness of a certain member, thus now distributing forces to the stiffer elements around it. Can someone confirm that my approach to deriving shear forces is correct? Is there a more straight-forward method for grabbing shear forces from STAAD? I've read recommendations for "Print Element Forces" and "Results Along Line", but these methods don't seem as intuitive.

2) Please see the attached, which shows the SQY plate stress distribution in one of the walls I'm designing. The way I isolated the wall was to do "View Selected" in order to get a graphical output of the results. For this load case, I found that the max SQY was 154 psi. But, the table in the upper left-hand corner is telling me that the max shear stresses are much higher. What I believe is happening is that even though I isolated this wall, STAAD is showing the max stresses elsewhere in the model. Is there any way to correct this?

Thanks much!
BLoEngineer

 

[WARose]

The slab shear stress should increase as the thickness decreases. You can find the max. SQX/SQY by clicking on the summary tab in the plate results section. Sometimes those results can get deceiving because of stress concentration factors (i.e. from supports, changes in geometry, etc.). Usually I've backed out of an allowable (ult.) shear stress based on the allowable in ACI 318 (i.e. phi*2*(f'c^0.5)). But you have to factor that allowable down a bit (or factor up your stresses) to compensate for the fact that the full depth of your slab is not effective for shear resistance (i.e. that depth is "d").

Also with regard to your moment design.....do not forget to include the Mxy (i.e. the twisting moment) in your computations. Most of the time I just combine the absolute values of Mx & Mxy and My & Mxy to get my design values.

 

[BloEngineer]

Thanks WARose.

The shear stress I'm seeing does increase as the thickness decreases. This makes sense because you end up multiplying the shear stress by the slab thickness to get shear force. My point was that I saw a sharp decrease in shear force as the thickness decreased, signifying to me that the stiffness reduction in the slab was also dropping the shear.

I have read a couple forums that recommend adding the twisting moment component. Why is this? If I were doing a hand calculation on a two-way slab, I would not add any type of twisting component. I'm interested in hearing the reasoning for adding this additional contribution in.

 

[WARose]

It's just equilibrium (just like in a beam). The layout/loading of these slabs sometimes dictates that the force will have to be transferred by twisting.

 

[Rainbowtrout]

The addition of Mxy is from Wood-Armer formula. I believe it has been discussed on this forum before. This design method is based on Johansen's yield criterion. The yield line is the line about which plastic rotation occurs and across which the rebars in both directions are yielding across. Using a lower bound approach, the Wood-Armer formula is used to account for the twisting moment. It is particularly important for the corner regions of slabs where the twists are high. See Reinforced Concrete Slab by Robert Park et al.

You can find other useful material online.

There is usually not much you can do with STAAD's graphic interface. I am afraid you cannot do much with the second problem

 

[BloEngineer]

Thanks for your input Rainbowtrout. I did include Mxy in my design after doing significant research, along with your recommendations, and determined it was prudent to include it.

On the 2nd item regarding STAAD's visual output, I did find a Bentley forum that explains how to view stress contours for a selected view in a larger model. What you do is select all the plates in a given element that you want to isolate (a wall, for example), right click and select "New View" and select "Display View in Active Window". The stress contours will adjust to the selected plates. To get back to the global model, go to the View Tab at the top and select "Whole Structure".