Effective Buckling Length for Top and Bottom chord 3

[Mohamed Maher]

Dear all

Greetings

In the attached file you can find 3D arch with span 40 m. I'd like to ask about Top chord and bottom chord buckling length.. I believe it should be the geometry length (For the in & out buckling) ..means the distance between the diagonals. anyone has another opinion(taking into consideration that there is no horizontal bracing)

or I should deal with it like lacing columns with total buckling length 40 m and get the inertial for the three pipes.

Thanks
Maher

 

[YuleMsee]

Yes, its the distance btw the between diagonals, get the compression force from the calculation for the element with the highest load and design a section.
If you take the entire 40m, there might not be a section big enough for the compression

 

[mchen96]

I would argue both are important (if it acts as an arch and not as a curved beam).

The distance between diagonals will control local buckling of a section within the arch.

The 40 meter arch will control global buckling of the whole arch.

 

[JoshPlumSE]

I think this is a good example of where the Direct Analysis Method (per AISC) would probably provide better design results than just guessing an unbraced length per the effective length method.

Add an out of plane notional load (or model in an initial imperfection) then let the 2nd order analysis determine if buckling is going to be an issue. Though you will have to use a reduced flexural stiffness to account for material inelasticity.

 

[Agent666]

A second vote for both the global and local buckling being important. Another vote for use of the DAM also.

As an alternative you can use a rational buckling analysis to determine the capacity in accordance with your code of choice. Most analysis software have options for undertaking a buckling analyses these days. See this post for the general procedure that is involved (specific to NZ code here but other codes have similar approaches for correctly accounting for the imperfections by application of the code column buckling curves, there is also a few more linked threads where similar concepts were discussed more in relation to AISC).

Whatever you do, don't guess the effective length, you'll feel like you have an answer. But it could be grossly conservative or unconservative and you simply won't know without doing a more rigorous analysis method. It doesn't take much more time to do these type of analysis, but since they are not mainstream you really need to understand how to do it and how to interpret the results.

I'd also highly recommend running through the self learning Stability Fun modules with Mastan2 (free software) to grasp the fundamentals with buckling analyses and the DAM.

 

[Mohamed Maher]

Thank you all for your replay that was useful