Camber Diagram for Steel Girders but not for Concrete Girders

Question

Why is it that, at least in Ontario Canada, camber diagrams are required for detailed design drawings (Not shop drawings) for bridges with steel girders, but are not required for CPCI Girders?

My understanding is that the camber drawing is required for the steel manufacture to put in a vertical curved to the beam so that when the dead load of the superstructure is added on top of the girders the whole thing bends to meet the road profile.

For concrete girder this is not done, though a similar process seems to be followed. The screed elevation is adjusted to account for deflection of the concrete, including long term deflection from shrinkage and creep.

I would have thought a screed elevation table would be required for both steel and concrete girder designs, so why have to two different approaches for dealing with deflection to match the road profile?

Answer 1

CPCI girders are prestressed precast concrete girders, right? (Every region seems to have their own acronym for these things so it's not the name I'm familiar with but the mechanics should be the same.)

The short version of the answer is that the prestressing will usually give you the appropriate amount of precamber even if the girder is cast in a form with a flat bottom.

Roughly speaking, the amount of prestressing is chosen to give a bending moment that counters the permanent load and a small percentage of the variable load (by placing most of the prestressing in the bottom of the beam). This gives the girder an upwards precamber straight out of the form, but once it's installed and the bridge completed, the total deflection will be around zero.

Obviously, it will not always be possible to make it end up perfectly horizontal. And the resulting deflection is nearly impossible to calculate with millimeter accuracy due to uncertainties in creep both before and after installation and the elasticity modulus of the concrete when the prestressing is applied. But in most cases it will even out nicely enough for a precamber diagram not to be worth the trouble.

Answer 2

It might have something to do with the issue of quality assurance.

Steel beams are highly uniform, both in dimensions and material properties. Therefore, if you calculate the deflection of a steel structure and get 2 mm at midspan under dead load, you'll likely get something very close to 2 mm in the real world.

Meanwhile, concrete elements are far less uniform. Their real-world cross-sections tend to deviate further from what was designed than steel beams', and the material properties are obviously less accurately determined. So if you likewise calculate a 2 mm deflection at midspan, you'll probably get something... around 2 mm, more or less. Which is perfectly adequate for a serviceability assessment, but not if you're concerned about the road profile. So in the case of concrete beams, you'll probably need to adjust the screed elevation anyways.