built up section - sction modulus

[morganjoe]

I have a situation where I am checking the strength of a built up steel section composed of a plate tension flange, a plate compression flange and a square tube steel. The compression flange length is limited so the tension flange is always longer. This causes the centroid to be shifted toward the tension side. Therefore you get two different section modulus for the section. I always use the compression side section modulus which is always the smallest. The question I have is that if I use a thicker plate on the compression side I can actually get the centroid to be closer to the compression side. In that case do I still use the compression side section modulus (which is now larger), or do I use the smaller of the two which would then be the tension side section modulus.

Thanks in advance for any help

 

[Lion06]

You can absolutely get the centroid closer to the top flange by making the plate thicker, but how much thicker depends on the cross section.
I would use Stop for the compression and Sbottom for the tension. It is possible (although unlikely given a square tube section), that the allowable bending stress is different for compression and tension (since you have an unsymmetrical section). Are you using the 9th edition ASD manual? It sounds like it since you are concerned about elastic section modulus.

 

[MikeHalloran]

"Therefore you get two different section modulus for the section. "

Uh, no. If the tubes and plates are welded to each other, you compute a composite section modulus for each cross section. I.e., you would have some modulus where top and bottom plates are present, and a different modulus where only the bottom plate is present, and a different modulus where no plates are present. But at no point do you have two different moduli at the same cross section. Okay, strictly speaking you do, one for X-X and one for Y-Y, but in one plane of bending, there's only one modulus.

At least, that's how I think it works.

I think there's a page in Machinery's Handbook that covers how to compute a composite section without making your head hurt.


Mike Halloran
Pembroke Pines, FL, USA

 

[JAE]

Mike, you are mistaken here.

If you have an unsymmetrical section (one with a top plate smaller than a bottom plate, then you have ONE moment of inertia of the entire section, but you have two section moduli - one for the top and one for the bottom since

Sx = I / y

and y is different for the top than the bottom.

(I'm only talking bending about one axis here).

Also - morganjoe - your statement: "I always use the compression side section modulus which is always the smallest." isn't the proper way to go about designing a section (it may be conservative but not proper).

Depending on the bending (positive or negative) you would use the S(top) or S(bott) depending on whether you are determining the max. compressive or tensile stress in the section. For most beams, and for steel, compression usually controls the design.

If you use a large enough plate on the compression side, and get the neutral axis to move closer to the compression side, then you might have a smaller S(comp) but whether you use S(comp) or S(tens) depends on the unbraced length of the compression side and the magnitude of the moment on the section. There is no standard answer in that case.

 

[MikeHalloran]

Well, oops all over me. That's what I get for working from dim memory. I apologize to all.



Mike Halloran
Pembroke Pines, FL, USA