Latticed Channel Design - Slender Members (Webs)

[Howlyn2]

Hello All,

I have recently run across several older water tank structures where we have been asked to review their structural capacity. Unfortunately, original design drawings are hard to come by and structural information is taken from field measurements. Quite often the legs are composed of built-up latticed channels with a considerable gap between channels. Currently we are reviewing (2) C15x?? B2B channels (Flange & Web 0.25" Thick / Flange = 3.5" Long / Gap = 9" with flanges turned outward / laced every 7.5" with PL2.5" x 0.4375" / Lu = ~40'). These members make up the water tower's legs. I have done some research & Bethlehem Steel's 1907 Design Manual contains Latticed Channel Safe Loading Tables. Also, since labor was cheap in the era I believe the structure was constructed (early ~1900s) and as the major steel producers show in their manual it was typically less expensive to roll wide flanges than to build up large sections by riveting together other smaller sections.

Based on the limiting width-thickness ratios for compression elements (Case 5) the web is slender. How should one go about calculating these member's allowable compression capacity today without using the older Design Manual's Safe Loading Tables? The member is built-up so I believe I should follow E6 --> Built-Up Members BUT the member also contains a slender element that kicks me to E7 --> Members with Slender Elements. I am thinking that I should check the member locally based on E7 and double the value based on the Consistent Deformation Method and compare these values to the global built-up member capacity where the lesser value controls.

The same concept applies to the moment capacity of the legs. Is the moment capacity simply M = S *Fy where S is the global section modulus, or should I be looking locally at each channel in flexure, determining it's compactness criteria and subsequent flexural capacity, and doubling due to adjacent member?

Thank You.

 

[KootK]

I believe that a common approach would be:

1) adjust your available yield stresses fictitiously until it would satisfy slenderness limits.

2) use #1 and and an assumption of composite action to develop aggregate member capacities.

 

[Howlyn2]

Hello KootK. Thank you for your response. I’m afraid I’m not following your explantation. Is there perhaps a specific code check you can reference or a design example available I can cross reference? I have to imagine there is a procedure for these types of analyses but I’ve yet to find it. Thanks again.

 

[r13]

Basically you have a built-up columns, with back to back channels as chords, and the straps as web members. Note that the straps are spaced to prevent local buckling of the channel flanges. You still have to check the built-up column for global buckling. I don't understand where the moment is coming from, at that time, most lattice structures were designed for axial loads only.

 

[KootK]

I’m afraid I’m not following your explantation.

I'm afraid that I wouldn't be able to locate a reference for you without going on an egg hunt myself. I'm pretty sure that that both AISC and the cold formed steel standard have methods like this, at least for flexure. I'll try describing it from a different angle and we'll see if that helps.

1) Our normal approach is to assume that we'll want to mobilize all of the material stress capacity of our member prior to local buckling (Fy etc). To facilitate this, we ensure that we meet the b/t limits that guarantee this for us. Note that those limits themselves are based on Fy. Then, we dive into the more macroscopic code provisions, no longer worrying about local buckling.

2) An alternate approach is to work with the b/t that you have available to you (renovation) but then use the b/t limit equations to work out the max material stress that you can take the design too before local buckling occurs. Then use this modified stress in place of Fy etc in the macroscopic design equations. This second approach is what I advocated in my previous post. It's quite simple to execute and allows you to avail yourself of most of the available capacity of your members.

From a design philosophy perspective, it is quite a nice thing to know that you can plastify your members fully before local buckling occurs. And that aspect of robustness is compromised with #2 above. We generally accept that as a reasonable sacrifice, particularly in renovation work.