Cruciform Column - AISC 1

[slickdeals]

Folks,

I am analyzing an existing building for additional loads. The framing consists of cruciform columns.

The cruciform column consists of a W36 member with 2 WT 18 welded to the webs forming a + shape. The lateral resistance is gained by knee braces framing into the flanges in each direction at various floors. The knee braces frame at 5' below the beam into the column and 5' into the beam beyond the column.

Can the column be analyzed for combined forces using the equations in Chapter H (H1-1a, b)?

Can I calculate the axial force capacity using E3 and E4?

Can I conservatively assume bending capacity in each direction only provided by an analogous wide flange bending in the strong axis neglecting the other member?

Are there any recommended methods for analyzing cruciform columns such as the one above?

 

[BAretired]

Torsional flexural buckling has to be considered and that seems to be covered in E4 so I think your approach is the correct one.

BA

 

[csd72]

I would do it like that although it will not be exact. As long as you dont design it right down to the bone you will be fine.

 

[ToadJones]

I'd be inclined to use H2-1

 

[slickdeals]

The WT is attached to the web of the W36 at mid height. As a result, it provides no restraint to LTB? Am I right?

I will calculate the flexural capacity based on the unbraced length = floor height?

 

[slickdeals]

In calculating Cw (warping constant), is the warping constant additive? For example, can I calculate the Cw for individual I shapes that form the cruciform and add them together?

I am calculating torsional constant J approximately as the summation of individual rectangular pieces.

 

[ishvaaag]

I have highlighted in p.5 of the pdf printout the formula I used for J of the mathcad 2000 worksheet dedicated to these columns some ten years ago. I classiffied it in the "near OK" status, because some refinements would likely be required to properly portrait the LRFD 1999 setup. It considers (as well) flexural-torsional buckling, optimized the shape and provided the "adequate" stiffness reduction to be considered at service level and factored level when using such optimized dimensions at a time where the current stiffness reduction statement was not extant. In a later post, the optimizing worksheet and the corresponding to just check.

 

[ishvaaag]

I have not the worksheet for just checking the section but the one that optimizes the shape I have. Maybe I didn't produce the check one for this type since I was doing some kind of serial work in columns, some of rare sections, and I omitted the check worksheet. By just setting the appropriate constraints to dimensions can approximate the check behaviour. Remember, a Mathcad 2000 worksheet, might not work properly with later versions.

 

[slickdeals]

Thanks Ishvaag.
In your calculations, the Cw is computed as 2 times the Cw for an I shape.
Are there hand computation methods for computing the Cw for different open shapes?

 

[ishvaaag]

The cozier path will be to use some section designer that gives Cw. Otherwise one may program oneself following the procedures in the books, too long to locate and scan for here to post, then to program proficiently.

I have joined in a zip a pair of tables for the axial (factored level) capacities of crosses of IPE and IPN european shapes for A42 steel, and some formulations for J and Cw. Hope it helps to someone. I have more tables for the axial capacity from the Ensidesa manual for such European profiles in other steels.

 

[BAretired]

This link may be of some help:

http://www.cisc-icca.ca/files/technical/techdocs/updates/torsionprop.pdf

BA