Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Buckling out of plan 8

Status
Not open for further replies.
Mohamed Maher

Mohamed Maher

Structural
Dec 31, 2017
132
Dear ALL,

GREETINGS

I'd like to ask about the attached file truss.. I have 2D truss not supported in out of plan .. how much I should take the buckling length out of plan(For the Top and Bottom chord) if all the truss members are welded to each others and also welded to columns.

Thanks
Maher
 
 https://files.engineering.com/getfile.aspx?folder=036a5f69-2a56-4e3c-b057-005f3cbe08be&file=Truss.pdf
Replies continue below

Recommended for you

Sort by date Sort by votes
Probably the full span of the truss.

But the way you treat the calculation of buckling limit depends on the approach of your steel code.

There is obviously some torsional fixity at the columns which will help minimize your out of plane buckling of the chord.




Check out Eng-Tips Forum's Policies here:
faq731-376
 
Upvote 0 Downvote
@JAE

I use British code is that possible to consider ends of truss fix fix and as a result buckling length equal 0.7 total length ?
 
Upvote 0 Downvote
For curved members you should really be going down the route of using a rational buckling analysis to determine the critical buckling load under axial loading. This type of analysis leads you to a member capacity, in which is embodied an effective length if you desire to back calculate it.

Not sure how BS treats the use of a rational buckling analysis to determine the design buckling capacity, but this thread goes through the process according to several other standards (NZ and AISC) and may be of some help in showing you the general steps involved.
 
Upvote 0 Downvote
I would use either the full radiused span length or use an elastic critical buckling analysis.

I imagine it'll be somewhat less than 1.0*L due to the torsional restraints of the column - exactly how much less it'll be will come from your model.
 
Upvote 0 Downvote
I would also go with the full radiused span of the compression chord.

OP said:
consider ends of truss fix fix and as a result buckling length equal 0.7 total length ?
Click to expand...

Definitely not. The columns do add some restraint but I'd not call that rigid restraint. Consider the possible sources of flexibility:

1) give in the connections between the truss and the columns.

2) slack and give in the column splice connections.

3) twist in the columns taken over their length.

4) slack and give in the column base connections.

5) rotational flexibility of the foundation itself.

Looking at what you've got here, it seems to me that:

1) It's an arch but a pretty shallow one.

2) This appears to be an aesthetic element not likely to carry any serious load.

If those statements are correct, I'd keep it simple and a little conservative. Hence my recommendation at the top.
 
Upvote 0 Downvote
Hi
I did a few projects several years ago that included something similar. It was pedestrian bridges with two trusses (RHS-members) and the walkway was HEA-beams with steel plates.

Step 1: Rotational stiffness of the HEA-beams end. That gives the rotational stiffness where the HEA-beam meets the bottom chord of side truss.

Step 2: Translational stiffness for the top chord (out of plane) based on the side truss and the support is the rotational stiffness from step 1.

Step 3: Buckling length for the top chord. The top chord is a member supported by the translational springs from step 2.

What I worked with at the time was simple trusses, not arces. But the idea should work with the appropriate modification.

The buckling length typacally became twice the spacing between the connections to diagonals. I later tested the analysis by doing it in 3d software and the result was pretty much the same. But the work needed was far less [smile].

Good Luck

Thomas

 
Upvote 0 Downvote
To be honest, if you already have an analysis model in software capable of doing an elastic critical buckling analysis, then you're 95% of the way through solving for the design capacity. There's really no need to guess what length applies, this is the real beauty of this type of analysis. It's so simple, yet never seems to be used in practice to justify complex effective length problems.

If you need more information as a background to the use of critical buckling analysis to determine design capacities check out Mastan2 stability fun modules Link. Many people incorrectly think that the softwares buckling analysis result is the design capacity, this is not correct, you still need to apply the code column curve to the result to correctly account for initial imperfections. This is usually outlined in the particular code you might be working to.
 
Upvote 0 Downvote
Thank you all for your valuable feedback ..this truss carried arround 170 kN ultimate load ..it supports tensile fabric ..the load not small ..and the program don't have buckling analysis
 
Upvote 0 Downvote
The Mastan2 program that Agent666 mentioned is free. It's just a little tedious for real world design problems with rampant load cases etc.
 
Upvote 0 Downvote
Hi
I did a very simple check in 3d. The buckling length was similar to the length of the upper chord. It can actually be longer because the inflection points seemed to be outside the columns.

Thomas
 
Upvote 0 Downvote
Again thank you all for your support

I'll take is as top chord length to get rid of any problems.. but I have question the forces in the middle of truss is small relative to its edges is it possible to change edges only with bigger cross section (same diameter but more in thickness)which will be safe under the top chord buckling length... and the middle smaller cross section is safe also under the total buckling length but lower force.. or I should keep all the top chord with same cross section for better behavior of buckling

Thanks

 
Upvote 0 Downvote
Its not a terribly long truss, so its likely more economical/practical to use a single length section for the top/bottom chords. Also keeps your buckling calc simpler!
 
Upvote 0 Downvote
FYI, if you changed the section and/or had other axial forces adding into the chord forces at intermediate nodes from roof bracing (analogous to having an axial load applied at some point along a column member for a variation in axial load between restrained ends) then going down the elastic critical buckling analysis route is the only true way to accurately estimate the axial design capacity.
 
Upvote 0 Downvote
I have another truss with 15 meter span.
The load is distribute over all the top chord length.. what i mean that internal axial comp. forces are small in the middle and increase more at edges.

I'll try to learn more about bucking analysis .

Thanks
 
Upvote 0 Downvote
Again this is something you can only really determine from a buckling analysis. Rule of thumbs like 0.7L will only get you so far.

A bit more advice I'm not sure if its just a one-off portal, but the connections for what you've shown can be quite expensive. Fabricators are generally not a fan of those types of arrangements.

I'd probably detail those as a gap - note that this will result in additional moments due to joint eccentricities. Also from experience check your connections early on - the new Eurocode rules are quite onerous and can warrant additional thickness in the chord members.
 
Upvote 0 Downvote
100% agree with UKbridge, the problem with application of simplified rules of thumb in buckling problems is you'll never know if you are conservative or not unless you confirm it by via a rational buckling analysis. So unless your structure (and axial load distribution) is simple enough to apply one of the pre-made code effective length factors or apply one of the alignment charts based on end restraint stiffness, then you are on shaky ground. Your calculated capacity could be out by 50-100% and you don't know it. You wouldn't deliberately under or over design something to that magnitude and accept it, but that's exactly what you might be doing by guessing at the buckling behaviour.

If you want to learn more about buckling analyses, I can highly recommend working your way through the mastan2 modules I referenced earlier. It's self directed learning but it'll make a lasting impression figuring it all out on your own.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

asalisbury
  • Question
Lateral Torsional Buckling of Thin Beam
Replies
15
Views
5K
BAretired
BAretired
bugbus
  • Question
Buckling capacity of twisted plate strut 3
Replies
12
Views
8K
bugbus
bugbus
StrEng007
  • Question
Wood Header/Beam out of plane (F2) connection
Replies
11
Views
2K
phamENG
phamENG
palk7 EIT
  • Locked
  • Question
Out of plane moment (lateral force)
Replies
11
Views
2K
TRAK.Structural
TRAK.Structural
SeizeTheMoment
  • Question
Bent/local buckling of joist chord analysis
Replies
3
Views
2K
ANE91
ANE91

Part and Inventory Search

Sponsor

Back
Top Bottom