Buckling Resistance: Rigid lean-on bracing.

[wadavis]

See the attached screenshot to save us a thousand words.

I'm analysing a slender system (guy-wire tower) that is making use of rigid lean-on bracing system to resist buckling. The column to column lateral connections (the 4'x4'x3/4" plates in the screenshot) are stiff enough that I want to include their moment resisting connection in the analysis. I'm assuming a pinned-pinned analysis will be overly conservative.

Can you point me toward some reference material for this buckling bracing?

Most of what I'm digging up is for columns with discrete bracing to a rigid fixture, with the bracing stiffness based on an assumed deflection value. I'm sure I can work back and rework the analysis for an assumed rotation in the bracing which is fixed to the stiffest member, but I'd rather not reinvent the wheel.

Thanks, please ask about anything that is unclear.
Wadavis

Structural E.I.T., Alberta

 

[KootK]

1) At one extreme, one could consider the plates to transmit longitudinal shear with near perfect rigidity. In this case, you've got yourself a composite member and could base your compression strength checks on that assumption. Basically, your plates would be the webs of what is effectively a vierendeel truss. I'm skeptical of this approach.

2) At the other extreme, one could consider the plates to be simply lateral spacers. In this case, the buckling capacity of the assembly would be something closer to the sum of the buckling capacity of the individual members within the assembly. Or, more simply, just the buckling capacity of the big pipe in the middle. This probably defeats the purpose of the spacers.

3) As an intermediate solution, you could proceed along the lines of #1 while making a conservative approximation of the flexibility introduced by the spacer plates acting as vierendeel truss webs.

While I don't know for certain, I'd be surprised if you didn't find something relevant in one of these comprehensive stability references:

1) Ziemian: Guide to Stability Design Criteria for Metal Structures
2) Bazant: Stability of Structures - Elastic, Inelastic, Fracture, and Damage Theories

Hopefully someone familiar with this kind of construction will know of something more industry specific.

Some questions for you:

1) What do things look like at the guy wire connections?
2) Do the members ever come together at a point such that the assembly could be considered a simple truss?
3) There are at least three different sized pipes in the assembly and not all are located to maximize structural efficiency? Which are intended to be structural members. Just the central pipe and the tubes at the four corners?

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.

 

[KootK]

I'd go with option #3 above. Create a model that reflects a reasonable degree of composite flexural action. Apply some dummy loads to the model and use the deflections to work out an effective EI. Then you're off to the races with the ability to treat the assembly as a single compression member.

I see this scenario more as a composite compression member rather than true lean-on bracing. The lean-on concept would basically just be #2 above which wouldn't make for very efficient use of the spacer plates.

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.

 

[wadavis]

Kootk,
Thanks for the reply. Until yesterday I've never heard of the name Vierendeel or seen a truss designed that way. Thanks for the literature references, I've found a lot of material for effective column lengths in braced continuous frames, but nothing yet that tackles sidesway.

A response to the proposed methods:
1) I suspect the structure will behave very similar to this, no rotation at the joints but sidesway, but without knowing how similar... well the error is in the wrong direction.
2) This how previous designs were completed, the plates do nothing mid-span and are considered pinned at the guy wire connection plates. K=1, simple buckling lengths and combined loads.
3) Going to attempt this by applying a dummy lateral displacement to match a buckling mode with the factored compressive load for a dummy second order moment. Assuming zero web shear/moment transfer I'll get a rotation and displacement at the web. I'll then apply the rotation and displacement from the dummy displacement to the web and compare the moment generated to the dummy second order moment. If the moment generated is greater that the second order moment, that buckling mode will not occur. I can run down the most likely buckling modes to calculate the effective length this way. It wont give me the additional support it provides to the final buckling mode, but it will tell us which mode governs. (This all a rework of the direct method of analysis for the stability of structures and members from CSA S16 9.2.6.2 or AISC Sesign Examples Version 14.1 Example A-6.2)

Are you suggesting applying a dummy lateral load to a static FEA model to calculate the EI for the buckling analysis? Hmmm. That would give the behavior of the whole system, then I could repeat the process with a dummy load on the individual members to get an effective length from the joint stiffness to review local buckling.

To answer the questions:
1) Guy wire connections to every third plate. Translation is restrained, rotation is not.
2) Not so lucky, the members never connect to give truss behaviors.
3) The piping arrangement was something thrown together by drafting just to give something to draw on. It is likely that the three inner members will more toward the outside for a greater moment of inertia, or they will stay where they are for greater plate stiffness. They are my tool to maybe affect the buckling mode.



Structural E.I.T., Alberta

 

[KootK]

I think that you're making things more complicated than they need to be with the dummy displacement business. If you're going to account for composite action, then I see the whole assembly as just an axially loaded column between supports (guy wires). You just need to get an effective EI for the columns and then proceed more or less as you normally would with column design. Modeling the assembly and applying a dummy load is simply about back calculating that effective EI.

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.