Laterally unsupported length and LTB of bridge floor beam

[tmalik3156]

Good day all.

I am analyzing a bascule bridge span constructed in 1929. As shown in the floor plan, the vertical lines indicate floor beams (4-feet deep). The horizontal lines are stringers (20- inch deep). Stringers are riveted to the floor beam as shown in the photo.
To reduce the weight of the bascule span, steel grating is used instead of a concrete deck. That means, the top flange of the floor beam is not continuously restrained against LTB. My question is:
1. Are those stringers able to provide lateral restraint to the floor beam at their connection points?
2. Stating otherwise, in performing Moment Capacity with LTB calculation, should the unsupported length be taken as L (full length of the floor beam) ? Or should it be S (the stringer spacing)? Thank you.

 

[BridgeSmith]

In the photo, I can't see the riveted connection between the floor beam and the stringers, but if there is indeed a connection, the stringers provide bracing for the compression flange of the floor beam, so the unbraced length of the floor beam for LTB would be the stringer spacing.

Theoretically, this would be predicated on the stringers themselves being restrained against movement longitudinal to the bridge superstructure, which I'm sure they are somewhere along their length, adequately enough to prevent LTB of the floor beams, anyway. It doesn't take much; I believe the lateral restraint capacity I've seen thrown around would be 2% of the floor beam vertical load.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmalik3156]

@ BridgeSmith, thank you for the reply. I meant to say that the angle that connects the stringer to the floor beam is riveted. The small beams above the stringer are grating beam; they carry the steel grating on top. Grating beams are welded to the stringers.
I have labeled the photo below.

 

[BridgeSmith]

Ok, in typical bridge lingo, the 2 large beams would be girders and the small beams supported by the floor beam would be the stringers. Anyway, if there's no connection between the stringers and the floor beam to brace the top flange of the floor beam, it would be normally be unbraced between the 2 main girders.

I say normally, because there is some research indicating that at the instant the the floor beam typically needs braced (when a truck is passing over it) the friction between the bottom flange of the stringer and the top flange of the floor beam, because of the weight of the truck on the stringer, is sufficient to brace the flange against LTB. That makes sense intuitively, since the required restraint force is only 2% of floor beam load, and even the low-end of steel-on-steel friction coefficient is around 0.2 (20%).


Rod Smith, P.E., The artist formerly known as HotRod10

 

[bugbus]

Are those stringers able to provide lateral restraint to the floor beam at their connection points?

Potentially, yes, but because the adjacent floor beams are also prone to LTB, there is a possible failure mode where all the floor beams buckle concurrently and offer little or no restraint to one another.

It probably depends on the nature of the loading. E.g. for a predominantly uniformly distributed load, the adjacent floor beams will experience a similar load as the critical floor beam and may offer little or no effective restraint; or if the loading is more localised, then you can probably rely on the restraint provided by the adjacent beams. I think you would have to crunch the numbers to satisfy yourself.

 

[BridgeSmith]

Potentially, yes, but because the adjacent floor beams are also prone to LTB, there is a possible failure mode where all the floor beams buckle concurrently and offer little or no restraint to one another.

The stringers the OP was referring to are actually the main girders. There is no connection between the floor beams and the stringers they support.

At some point, most likely the stringers are braced against longitudinal movement, but there remains the question of whether the floor beam is braced against LTB by the friction between the top flange of the floor beam and the bottom flanges of the stringers above.

If LTB causes an unacceptable reduction in the floor beam capacity, there's always the option retrofit a connection between the flanges.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmalik3156]

@ gusmurr, thank you for your reply.

@ BridgeSmith

Actually, this bridge does not have girders. It's a lift bridge with trusses at two longitudinal ends.

I understand that the photo can be confusing. You are thinking that the top gray members are stringers (but they are small beams carrying the steel grating). Stringers and floor beams are exactly as I labeled in the photo. These large transverse floor beams rest on longitudinal truss bottom chords - which are not visible in the photo. May be that is the reason of confusion.

Anyway, I looked online for the 2% restraint. I did not find any North American source, but the British take 2.5%, and that's a small load indeed. So I concluded that the large transverse floor beams are adequately restrained by the stringers through the riveted angle connection shown in the photo.

Here is is a cross section showing floor beam and truss. The stringers are perpendicular to the screen so only their ends are visible.
Hope this dispels any confusion.

 

[BridgeSmith]

I was indeed misunderstanding the configuration. My apologies for not paying closer attention to the labeling in the photo.

I would say it's very clear that the floor beams are braced against LTB by the stringers. Even if the stringers are not restrained from longitudinal movement, I'm confident you could easily prove there's adequate moment restraint in the riveted connection and the stringers to produce stability against global LTB failure.

Intermediate restraint of the stringers themselves (by the grating beams) would be questionable, but as long as their unbraced length from floor beam to floor beam is not long enough for them to be subject to LTB, then you're ok there, too.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmalik3156]

Thank you BridgeSmith.
Stringers are 6.35 m long (20' - 10") from FB to FB. But each stringer carries as many as 12 grating beams on top; grating beams are welded to stringers. So based on your earlier comments, I believe the stringer themselves are also laterally restrained.

 

[BridgeSmith]

As long as the welds stay intact, you should be able to assume the compression flanges of the stringers are braced. We've found connections like that to be susceptible to weld breakage, and it's been a subject of much debate as to whether the flanges can be considered braced if the welds are broken.

Rod Smith, P.E., The artist formerly known as HotRod10