Effective Length of Unsupported Compression Chords 1

[CANPRO]

I am looking for some references to determine the effective length to be used in compression chords which are laterally unsupported. Application will be similar to a pedestrian bridge where the top chord only has the web members to provide lateral support.

I've been on

http://www.cidect.org

which has a lot of good resources...still reading through them now. All of the references I have found so far give a general description on how to solve the problem and some have worked examples...they all seem to reference "Mouty, J., 1981: Effective lengths of lattice girder members. CIDECT Monograph No. 4, CIDECT. " which I have not been able to find. Apparently there are 60+ design charts for determining the effective length of unsupported compression chords.

If anyone could point me in the right direction for the above reference or any other relevant reference, it would be very much appreciated.

 

[DamsInc]

Are they HSS members?

 

[CANPRO]

Sorry, should have been more specific. All members of the truss will be HSS (rectangular or square). I don't have all of the specifics of the problem right now, I have just been asked to look into the effective length of the unbraced compression chord.

 

[DamsInc]

Looking at this book in front of me:

http://www.cisc-icca.ca/solutions-c...-section-connections-and-trusses#.VLfKnHse2zQ

It references the same CIDECT Monograph, as well as more recent design guides. The simplified rules are stated as follows:

HSS Chord members: K=0.9
HSS Web members: K=0.75

These values are valid only for HSS members connected around their full perimeter without cropping of flattening. Chord members must be parallel. The smaller dimension of the web member must be at least 0.25*(chord width). Web wall thickness must not be bigger than chord wall thickness.

For a pedestrian bridge like you describe, there is a longer section with a more detailed calculations. I recommend picking up the book, or those CIDECT guides.

 

[Guest262653]

Hi. Maybe the following references can help:

http://publications.cce.iastate.edu...l Buckling Analysis of a Steel Pony Truss.pdf

http://saice.org.za/downloads/journal/vol47-3-2005/vol47_n3_b.pdf

Additionally, a FEM model that replicates the stiffness of the transversal U frames (either a full model or a beam on elastic springs model) may be used effectively to determine the top chord buckling load.

 

[CANPRO]

DamsInc, I have that book as well. Very useful. Unfortunately they don't seem to get too in depth with unsupported compression chords. The rules you stated (K=0.9 for chords) is for when the top chord is braced (with purlins, joists, whatever may be framing into the truss). The simplified rules are shown in section 2.3.1, my problem is more like the problem shown in section 2.3.3 (two pages over).

avscorreia, those look like great references...looks like that will get me to where I need to be, will review later. We are eventually going to get into FEM...I would like to tackle this with a hand calc/spreadsheet as well. Thank you very much.

 

[Guest262653]

You're welcome. Glad I could help!

 

[DamsInc]

Yes, I figured that Section 2.3.3 of that book would describe your laterally unsupported condition (i.e. pony truss). Hadn't gone through it myself though.

 

[Guest262653]

By the way, the following books deal with this question extensively:
- Theory of elastic stability, Timoshenko - Section 2.6.
- Guide to Stability Design Criteria for Metal Structures - 6th Ed, Ziemian - Chapter 15

 

[JoshPlumSE]

A couple of ideas:
1) Run an eigen buckling analysis to determine what the buckled shape of the chords really is then use a length based on that value.

2) Use your best guess during the analysis. But, use the Direct Analysis method from AISC and make sure to use some initial imperfections or notional loads that will displace the chords in the direction of buckling. That way, the P-Delta analysis will capture the appropriate level of moment amplification from 2nd order effects.