Web crippling ASD Specs

DPAJR

I've always read that section to mean perpendicular to the web. Also, if this moving load is for a a monorail, bridge girder, or crane girder; you need to check for fatigue. Did you increase your loading for impact? If this is a high use beam you need to be carefull about your stiffener detailing. On some high use beams with fatigue considerations, the stiffeners are not supposed to extend to the bottom flange. If I remember correctly it can cause a stress riser of some sort and crack the beam at that location.

Hope this helps more than hurts.

Hi Socks 15,

Thanks for the quick response. Your answer is helpful.

Yes it is a bridge. It is existing and crosses a small creek. It has a clear span of 26.5 feet and the beams are two HP12x50's The land on the other side will be subdivided into 6 lots. The HS20 loads will probably only be seen during construction of new homes and possibly delivery of appliances etc. I did use the impact factor of 1.30 even though traffic on this bridge will probably never exceed 10 mph because it dead ends into a road about 20 feet beyond the span.

I have seen the perpendicular type of stiffeners used in shipbuilding but. I have done a couple of privately owned bridges before but they were from scratch and the beams were sized to avoid web crippling.

Thanks
dpajr

DPAJR – for bridge design you should be looking at the AASHTO specifications….see 10.48.8.1. It gives stiffened and un-stiffened shear capacities for beam webs. If you exceed the unstiffened shear capacity of the web, transverse web stiffeners will be required.

Here’s a drawing from Michigan DOT of a transverse stiffener detail to give you an idea of what they typically look like (usually use them in pairs). Go to Chapter 8: structural steel, guide 8.06.02A.


It will be difficult to install these on an existing bridge over water. You may want to think about some other alternatives.

MichST

So the Standard Specification doesn't base the need for a bearing stiffener on web crippling, but on web shear stress. I never realized that. Do you know why there is such an apparent discrepancy between AISC and AASHTO?

DPAJR
I didn't realize that this was a traffic bridge. When I said a "bridge girder" I meant for a bridge crane. I'm afraid that I'm going to have to agree with MichSt and say you need to use AASHTO for this. Granted, it sounds like you could get away with using AISC, but it's still not the proper design code. Along with the HS-20 there are other considerations on bridge design such as longitudinal forces that can result from vehicles braking. At least that's something I remember from my bridge course, but I dont' remember much more than that. Also, I know from experience that concrete trucks tend to carry more concrete than what they should and weigh more. Hopefully the AASHTO code will have safety factors to account for these things.

I’m a bit confused now. In you’re initial question you are asking about concentrated moving loads. I assumed you where finding your web was overstressed in shear due to the HS load, in which case you would need the intermediate transverse web stiffeners. I wouldn’t be surprised either if shear was an issue with bridge with how short the span is.

But, now it sounds like you’re concerned with shear at the end of the beam. To my knowledge the AASHTO standard specs don’t have a specific web crippling criteria like AISC. AASHTO requires that if the shear at the end of a beam is over 75% of the allowable shear stress you provide bearing stiffeners. See sections 10.33.2 and 10.34.6.1. I think the difference is just AASHTO using old tried and true conservative requirements here.

That said, if you are considering using web stiffeners or bearing stiffeners you may want to think about drilling and bolting angles to the web instead of welding.

DPAJR

As noted above, you need not meet AISC criteria for a bridge. However, if you were to check K1-4 for web crippling under a wheel load, and it failed, then you would need a bearing stiffener everywhere.

I don't have dimensions handy for an HP12x50, but if I use those for an HP12x53, and conservatively assume N=0, then 1.3x16k/0.435"x(0"+2x1.12")=21.3ksi < 0.75Fy.