Can anyone explain why these steel highway girders have this uniform deformed-like appearance?

[Mylzie]

I drive under this on my way to work every day and have always been curious about it. My boss told me, at the time this bridge was built, these were the deepest wide-flange girders in the world (or maybe it was just the U.S.?). At first, I thought it was just the bridge above me casting a wavy shadow, but all 4 girders appear like this.

The lighter portions look like a uniform \/\/\/\/\/\/\/\/\/\/\/\/ strain. My guess is the darker and lighter portions of the flange are deflecting outward opposite of one another. Is this a defect from the fabrication process? Flange local-buckling E: web buckling? Just spit balling at this point.

Located in NH by the way.

 

[JoelTXCive]

Are you referring to the slightly darker regions of paint? That is odd... my first instinct was "that's the pattern of how the painter sprayed the beam,", but I don't think that is the case.

The pattern would look more like web shear buckling or web crippling to me. Are there cross braces located on the other side of the dark regions? Maybe the marks are a result of welding during initial fabrication?

 

[Mylzie]

@Joel welding marks is actually a pretty good guess that's probably what it is. There is cross bracing but it doesn't align with the pattern.

Also just realized I said flange buckling when I meant web.... sigh, I'm taking my PE exam in a couple months, too.

 

[Tomfh]

It looks like it could be web buckling. The way it’s inclined towards the support matches that. Maybe it’s just from fabrication stresses. It’s a big beam.

 

[SandwichEngine]

I don't think it's web buckling. I think that's normal for deep sections with thin webs. PEMBs look like this all the time straight off the assembly line.

 

[phamENG]

I'm thinking distortion from welding up the plate girder. I seriously doubt that's a standard, milled beam shape...looks easily as deep as a sedan is tall.

 

[Tomfh]

Yeah, I remember seeing a funny photo of a plate girder under fabrication tagged “F***ED”, after it had warped beyond the point it could be recovered.

 

[Mylzie]

@phamENG Yeah it's kinda shocking how large they are. Pic doesn't do it justice from the angle but the spans are insane, longest i have or problably ever will see for a non-truss-like steel member. I don't blame them but I always think its funny how most people probably don't even blink an eye towards them...

 

[phamENG]

We have some long ones, too.

 

[Compositepro]

It is most probably due to the hot rolling process. The thin web cools much faster than the thick flanges, so after forming, the thick flanges cool and shrink much more than the web they are attached to. The result is longitudinal compression of the web and visible waviness.

 

[phamENG]

No chance that's hot rolled. A standard traffic light is about 42" tall. The tallest 'standard' hot rolled, wide flange shape is 44" tall. That's a plate girder all day along.

 

[271828]

The next time you get a chance, look on the inside face of the web and see if there are transverse stiffeners like in the following.

https://www.bgstructuralengineering.com/BGSCM13/BGSCM008/Design/BGSCM0080603.htm

 

[Compositepro]

Welding distortion would look the same.

 

[Tomfh]

What’s the address?

 

[BridgeSmith]

I'd guess web shear buckling, and the shear capacity is based on post-buckling tension field action. I'd bet money there's vertical stiffeners on the backside of that web right where the diagonal waves meet the top and bottom flanges.

Do the diagonal waves angle the other direction in other parts of the span.

That's a plate girder, for sure, but not an extraordinarily large one. Looks to be about 8' deep. I designed a bridge with a 9.5' deep girder a few years ago. Missouri has done some plate girders approaching 12' deep.

 

[Compositepro]

If you look closely, I think you will find that those are not diagonal waves. There are two sets of longitudinal waves. The waves at the bottom of the web a off-set half a wavelength from those at the top of the web. The wavelength appears to be the same as the height of the web. (One wavelength top to bottom of the web).

 

[BridgeSmith]

You might be right, Compositepro. It's hard to tell, but I still lean towards diagonal buckling of each panel (between vertical stiffeners) caused by tension between the top right and bottom left of each panel, and compression between the top left and bottom right of each panel. If you're right, though, what would cause that?

 

[wrantler]

Does the pattern mirror at midspan?

 

[hokie66]

I concur that the diagonals represent the web under compression. Think of a truss analogy. If that is correct, the diagonals should be in the opposite direction at the other end.

 

[271828]

This is probably not due to shear buckling. The out of plane displacements from shear buckling are not very visible until you get near the ultimate strength. Also, in this case, they seem to be no worse near the support where the shear is higher.

There are probably transverse stiffeners on the other side. No web is perfectly flat. The stiffeners are making the out of plane deflections be in the pattern that is shown.