Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Lateral Torsional Buckling

[Subject Prev][Subject Next][Thread Prev][Thread Next]

Padmanabhan Rajendran wrote:

The allowable flexural stress of a rolled W-Section can be assumed to be 0.66Fy if the laterally unsupported length of the compression flange is less than a certail span, Lc, which depends on the section properties.

A 70' span, steel beam/wood deck, bridge has come for my review. The bridge is 14' wide. Two steel beams (W36X160), 9' apart support 6X6 wood members, laid continuously across the top flanges of the steel beams. A continuous angle member (L5X3) is bolted to the underside of each of the 6X6 wood beams. The outstanding leg of the angle (pointing downwards) is bolted to the W36 at 30" spacing. In addition, There are transverse diaphragms spaced at 17'.

The design engineer considered that the above arrangement provides effective restraint against lateral torsional buckling and applied an allowable stress of 0.66Fy. I contend that, even if the deck effectively restrains the lateral moveemnt of the top flange of W36, a lateral brace should prevent rotation of the cross section. On this basis, the laterally unbraced span should be 17' which is the spacing of the diaphragms.

I recall reading somewhere that in order to qualify for an effective restraint against lateral torsional buckling, the brace should restrain, at least, the upper third dedpth of the W-section.

May I have some opinions?





In answer to your question, no, I do not consider that the beams are adequately braced by the deck.  The best way to fasten wood decking that I have seen or used involves clips - a plate with a notch into its center, which notch fits half the beam flange - which are nailed to the side of the wood members.  The clips are alternated side to side of the beam, and thus off both lateral and torsional bracing.

I have seen stream crossing bridges built this way, hit by large drift during a flood, resulting in all the beams curved about 6 inches off of line in a 20 foot span.  Yet the bridge performed well for many years afterward because the beams were torsionally braced.

I don't entirely understand how the angle you describe is bolted to the beam, but in any event it does not seem to me that it gives you much torsional restraint.

Another response mentioned that the stress should be 0.55 Fy.  This is the value listed by AASHTO for highway bridges, not 0.66 Fy.  AASHTO has equations similar to AISC for determining the allowables for a given unbraced length.  You should use these rather than AISC.  You should also apply the impact factor and distribute the wheel loads to the stringers per the Code.

Another point is that you've got about 2.5 feet of cantilever each side. If a vehicle can get a wheel out there (it can), then you will probably have uplift on the deck connections.  This is to be avoided if at all possible.

It sounds like this is a bridge yet to be built.  Given what you have told us, I would recommend it be redesigned by a bridge engineer.  While most structural engineers can read the AASHTO Code and apply it intelligently, there are a myriad of detailing and construction issues (especially regarding the foundations and the railings) that are not codified, but are nonetheless important.


Mike Hemstad

who used to design a lot of bridges

Minneapolis, Minnesota