Modelling Girder Openings for Deflections in SAP2000 2

[efFeb]

Hi!
I have several plate girders with long (3') openings. I have reinforced these for strength, but am also concerned about how this will affect deflections. At first I thought to use section wizard to design the section with the opening, keeping other dimensions constant, but I worry that over the 3' opening length, it would be unconservative to consider the added benefit I get to the I value from the parallel axis theorem based on the overall section depth. Has anyone run into a similar issue before or have any thoughts on this?
Thanks so much!

 

[STrctPono]

I wouldn't think that deflection would be an issue if you have openings in your web. The way in which you strengthen it to allow a proper load path for the compression strut and tension tie is more important. Are you adding both horizontal and vertical stiffeners?

If you did want to check deflection, model a simple beam element and apply the appropriate section properties to the beam at the locations in which they correspond. I think the affects on the deflection will be smaller than you think. You can neglect the added benefits of the horizontal stiffeners when solving for the section modulus. You will need to utilize the parallel axis theorem anyway if you have a composite deck or your openings are not centered on the non-composite section's centroid.

 

[KootK]

1) How deep is your girder?

2) What is your girder span?

3) How tall are these 3' long openings?

The only potential for issues that I can see is if the proportions lead to a situation where your girder is acting like a vierendeel truss across the opening. If that's the case, you could work out an effective shear stiffness for that and apply it over the length of the beam where the opening is at plus a foot or so each side. Unless you go balls to the wall plate FEM, whatever you do will be a rough approximation. in concrete speak, this will very much be a disturbed region where Bernoulli beam simplifications will be only marginally applicable.

 

[STrctPono]

Your inquiry is an example of a time that I would actually model the opening using a full 3D FEA analysis. Not to check the deflection but to check to make sure your strengthening scheme of the opening in the web was sufficient and you aren't getting too high of stress concentrations. In this particular case, you will have to be familiar with stress analysis. I would model the adjacent beam sections using line elements and the section in question using shell elements. The node where the line element connects to the plate element node will need to be master-slaved (rigid linked) to the entire vertical face of the plate girder web and flange.

Something like the image below.

If you have a good FEA that you are familiar with, you should be able to model this in approximately an hour. Interpretation and manipulation of the results to take longer, however.

As always, you will want to substantiate your results.

 

[AllenAdams]

AISC Design Guide 2, Steel and Composite Beams with Web Openings, describes a few possible approaches for calculating deflection. See Chapter 6. At the end of Section 6.5 it states that for large openings the deflection due to the opening is approximately equal to the deflection due to shear, given there as delta_s = wL²/(8A_yG) for a uniformly load beam. For small openings the increase in deflection is less than 4%. As a side note, AISC Design Guide 31 Castellated and Cellular Beam Design indicates that for those beams the deflection increases by about 10% (and remember, this is for a beam that has more holes than web).

If you prefer British Standards approach there is an SCI Advisory Desk article, AD 183: Deflection of composite beams with large web openings, that has some nice tables indicating the percent increase in deflection due to openings of various Opening Depth / Beam Depth ratios and Opening Width / Opening Depth ratios. That is a quick and easy approach, valid regardless of the code you are designing to.

A Eurocode approach is given in SCI Publication P355, Design of Composite Beams with Large Web Openings. See Section 6.1 Calculation of Additional Deflections.

The impact on the overall deflections is likely to be quite small.

Also, based on the design methodologies of AISC Design Guide 2 and SCI P355, you may not need to add any stiffeners. Don't just assume that you need to, take the time to go through one of these and see if stiffeners are required. They usually aren't.