Web stiffeners

[rabedrak]

I am trying to make a strong beam that is low in height. Two W12x72's side by side, stitch welded together, then attach 3/4" plate top and bottom. I want to use only fillet welds along the sides of the plates to attach them to keep cost down. I am looking at flexure, stress in the weld at the interface between plate and beam, and web shear capacity.
There are two loading conditions, point load in center, and symmetric loads separated by 10 ft.
I have analyzed the flexure and the capacity of the fillet welds. I am stuck understanding how to analyze and design the web stiffeners. I know the web is not sufficient by itself. I would like to just have transverse stiffeners on the outsides instead of both sides of each web. Can anyone point me in the right direction?

 

[KootK]

Try the plate girder provisions of the AISC specification. Given what you're trying to accomplish, you might be better of with a section built up from plate. That way, you can design your webs to work without the stiffeners. How will load be applied to the beam? Your setup may not be great for distributing load between beams.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[CBSE]

It seams to me once the plate is loaded, it may try to "bow" out your top flanges of your beams and induce some really large prying stresses at your weld interface.

 

[dhengr]

Rabedrak:
I’ll second what KootK suggests. I’d fabricate a box beam/girder with a single top flg. of 42-50ksi mat’l., two thicker webs of 36-42ksi mat’l., and a single bot. flg. of 50-65ksi mat’l. Fillet welds, both sides on webs to top flg., and single fillet welds from the outside only at the webs and bot. flg. You start the fab. with the top flg. upside down on a stiff fab. table and apply the webs. You can apply cross stiffeners, were you want/as needed, clipped corners and welded to the top flg. and webs with fillets. You want to be able to roll the member from side to side for welding. Then, roll it over onto the bot. flg. to complete the bot. flg. welds. With those magnitudes of loads, you have to pay attention to top flg./web bearing stresses and may nibble bevels 12-16" long in the top of the webs at the load application regions to increase the weld metal area for the concentrated cross web loading. You must distrib. those line loads over some length of the webs. Then, you have essentially the same bearing type problems at the reaction points on the bot. flg. You have to pay attention to deflection under loading, and camber due to the fab./welding process.

 

[rabedrak]

Thank you all for the feedback. My dilemma is partially constrained by the fact that I am working with another engineer that feels a mandate to use up these beams because we have them leftover from another project and they represent a portion of the assembly that is available at basically scrap value, which reduces his cost for the project. Ultimately, with the help of KootK's comments about the box girder, I have recommended that he turn the W12's 90 degrees, using the flanges as the webs. This gives a substantial increase in web strength at the expense of about 10% of the bending strength which was not the limiting factor anyway. The dimensions are nearly the same. The limit in strength is going to be the weld interface as noted by CBSE once we get above a certain load, but the weld we can get is sufficient for our current needs.

 

[Shu Jiang PEng SE]

The moment is 3110k-ft from diagram 1 and the shear is 1992/4.5=442.67kips. I calculated the section properties of the built-up member as mentioned above with plate dimension 26x0.75 (to accommodate the option of rotating W12 90 degree):

Ix 2651 in4 Zx 470 in3 Sx 384 in3 Q 127 in3

Rotate W12 90 degree option

Iy 1720 in4 Zy 314 in3 Sy 255 in3 Q 108 in3


Per the information, I doubt the beam strength, deflection and the weld.

 

[cliff234]

How are you going to connect this double beam at whatever is supporting it? There are several complications when you try connecting double wide flange members such as this.

 

[rabedrak]

The beam will not be attached at the ends. They will be supported by platforms at each end. Their purpose is to hold large engines in a staging area. We will transfer them from railcars to a special hydraulic trailer, then move them to the staging area, slip the beams beneath the engines, then lower the trailer to let the beams support the load. The only concern is to make sure we have enough strength to hold the loads safely. Deflection is not a great issue as long as it does not hinder the range of the trailer to lower them and then move out from under them. Engines weigh 500 kips, and we are using 3 beams per engine, so the load per beam would be ideally around 167 kips. I found that, for the dual point load, weld strength controls, followed by web shear. For the single point load, the web shear controls, followed by flexure. I am working on the web shear issue using Chapters F and G of the AISC.
The diagram I clipped just shows the capacity of the beam in flexure with no regard to other factors. I included it mostly to show the load configurations that we are considering. The dual point load is what we expect for the engines, and the single point load is a worst case scenario that we evaluate in order to put a load rating on the assembly for general use without regard to engineering analysis. This company is a crane and rigging company and we would need to put a rating on the part just for general use because it could eventually be used for other jobs with unforseen configurations.