FIELD FIX: Glitch in manufacturing Plate Girder

[SmithJ]

I'm working on a project where the wrong size plate was used for the web during the manufacture of some plate girder columns. These columns have already been fabricated and shipped to the jobsite.

Apart from all the acrimony and finger pointing that is sure to occur; I want to have understanding of the different options available to increase the shear carrying capcity of a plate girder.

In a case such as this, is it okay to include the flanges in computing the shear strength of the member? If so, how would the contribution of the flanges be calculated?

Any other ideas, suggestions or comments would be more than welcome.

Thanks.
JS.

 

[DRC1]

The shear capacity is (VQ)/(It) and for W shapes V/dwtw is a good approximation. One option is to compute the shear capacity based on the above eqation. The other is to allow a slight increase if the capacity is close if you are using allowable stress. AISC tells you that the basis for fv=.14Fy is old and very conservative. Otherwise you will probably need to put shear plates on the web. Is the axial capacity of the column sufficent?

 

[Lutfi]

First let me start by saying that I sympathize with you. It is not fun when such things happen! I also would like to applaud your action by wanting to find a solution and not dwelling on the acrimony of whose fault was it. We need to do what is right and keep public safety the number one concern.

I am assuming that you are using AISC specification for the girder design. It has been my experience that steel members are hardly governed by shear stress! Is the girder overstressed in shear with the thinner web plate? Are the flanges and web continuously welded or they have intermittent welds?

I recall in my early days in college, my steel professor did the shear calculations including the flange thicknesses as part of the shear carrying element for hot rolled sections. In other situations, he deducted the flange thicknesses for same hot rolled sections. I think it is a judgment call. You can argue for each method when using hot rolled sections.

What is the fv compared to Fv for the thinner web plate?


Regards,
Lutfi

 

[Lutfi]

This link is a great one. It has welded plate girder chapter from Ohio State University.

http://hcgl1.eng.ohio-state.edu/~ceg532/chap9/chap9.htm

Regards,
Lutfi

 

[Lutfi]

The plate girder applet is great. The link is in the bottom of the page on the link above. It is easy to use with good documentation. Verify results!

Regards,
Lutfi

 

[JStephen]

I think more details would be needed for specific answers.

Your allowable bending moment will be reduced slightly as well.

In some cases, you could add stiffeners to the web to increase shear capacity- depends on what governed the design in the first place.

The design codes are always written in terms of minimum specified yield strength. Actual yield strength, as shown on mill test reports, will run several KSI higher. If it can be approved by all concerned (and that's a big IF), you could redo the design using actual yield or tensile strengths.

Who designed the beam? Get them involved if they're not. There may be some inherent overdesign included. They may can remove some approximations and reduce the required strength.

Depending on the configuration, you could possibly weld a strip of plate on the web, near one of the flanges. Fair bit of work, but better than scrapping the beam.

 

[IFRs]

You may want to revisit web crippling also.

 

[SmithJ]

Lufti, Thanks for the link. It seems to be filled with a wealth of great info. I'm sure it will come in handy.

A little more background: due to space constraints, we were forced to use a shallower column than we normal and as a result, the thinner web plate does result in overstressing the member in shear.

The fv compared to the Fv is at a ratio of 1.6 overstress. Axial capacity is okay and even the reduced allowable bending moment is sufficient.

I did use AISC Allowable stress to design the beam based on Section F4 of the 9th Edition. However, looking at the commentary C-F4, it describes the shear yield stress as

Fy/(Sqroot 3).

My understanding is that this is the inelastic shear yield of the web plate. When can I use C-F4 to compute shear capacity?

Thanks for all your responses so far.

JS.

 

[shepherd]

Could you take some round bar and bend it in the configuration of a warren truss web and weld at panel points to the flanges similar to an open web joist and neglect the web plate? Are web stiffeners in the way?