Span configuration over stream with many bends

[OSUCivlEng]

I am trying to layout a series of parallel bridges over a creek that is more crooked than a politician. The creek meanders from northwest to southeast with multiple 180 degree U shaped bends, while the road is running east to west.

Given the span lengths, 175' to 300' depending on the bridge, I am looking at 3 and 4 span continuous plate girders for the superstructure. The bends make it a challenge to achieve balanced span lengths (all equal spans or end spans ~75% of center span). Also, I will have to justify my design to other engineers who are not bridge or structural engineers, and one person I have already talked to doesn't understand why you can't just use a smorgasbord of different span lengths for a continuous girder.

I have been able to fit more conventional span configurations such as 185'-185'-185'. I suppose you could try and be non-conventional and use various non-symmetrical span lengths, but I think the girder design would be very inefficient and probably expensive. Has anyone ever encountered a similar situation?

 

[BridgeSmith]

Symmetrical spans are nice, but not necessary, especially for bridges as large as what you're designing. A 300' bridge will typically have at least 4 (usually 5) plate girder pieces, so you can economize by using different flange thickness for each section as necessary.

Although generally not as efficient, you can typically push the span ratio for a 3 span to around 1:2.5:1 before getting any significant uplift. We've gone as high as 1:3:1 using fully integral abutments on steel H-piles encased in a drilled shaft (with the top 15' of the pile left uncased to accommodate thermal movements. The girders end up somewhat larger, but it's a small percentage increase in overall cost of the bridge, and usually presents a more economical solution than more exotic options.

 

[OSUCivlEng]

BridgeSmith what about when the end spans are longer than the center span? I don't know if I have ever seen that before. I have done 1:2.4:1 with integral abutments before. A now retired coworker did 1:2.8:1 with no problems.

Or what about when none of the spans are the same length?

 

[BridgeSmith]

End spans longer than the center span is probably going to be less efficient, but doable. Depending on the depth available for the superstructure, a 2 span bridge, even an unsymmetrical 2 span, may be the more economical option.

Generally, we've found a depth/span ratio around .04 to result in the most economical girders, with ratios down to .035 being only slightly more. Even a ratio of .032 doesn't add any significant cost to the overall cost, especially for composite superstructures, which is pretty much always preferred, and usually becomes necessary for shallower depth/span ratios, in order to meet the fatigue stress limits.