Analysis model

[atee001]

I have an analysis model composed of bracings,beams and columns.

When I have verical loads only applied to my structure, I find the axial forces in bracing (in some of them) greater than the columns connected to it.

What could be the reason?

 

[Ron]

Could be one of several reasons. If you have an animation feature in your software, animate the model and see if it reacts as you expect.

Next look at nodal restraints. You might have some restraint conditions that are not correct.

What type of software are you using?

 

[JoshPlumSE]

Is this a Chevron braced frame? If so, then the natural load path will go through the braced frames rather than through beam flexure. That's why the seismic codes tell you to analyze / design the beams as if the braced frames were not present.

If this isn't a chevron, then is it possible that the axial stiffness of the brace (AE/L) is greater than the axial stiffness of the column?

 

[atee001]

No axial stiffness of the brace is not greater than axial stiffness of the column.Both have same area but the brace is longer than the column.

 

[csd72]

atee01,

we need more info, a sketch maybe?

 

[271828]

It sounds like you have a full building model, or at least a model with multiple frames and columns.

Pull out one braced frame with only its nodal loads. You should be able to manually compute the brace and column loads or at least estimate them. Also, at least you'll have an easier model to let you zero in on the problem.

This kind of exercise should be one of the *first* steps when the program seems to have done something weird.

 

[sshields]

I know that when I use RAM Structural System, we design the gravity framing first, using gravity loads only. Then enable the RAM Frame module. What I'm trying to say is that by the time I'm on my lateral analysis, the gravity design is pretty close to complete, save for any iteration coming from RAM Frame. But generally if you are just analyzing vertical loads, I'd not include braces for the intial sizing of your members. Certain analysis softwares are quirky and things have to be done in a certain order or fashion to get expected results. Depending what you're using I'd follow JoshPlum's advice.

Scott Shields
Ghafari Associates, LLC

 

[atee001]

Hi--here is a picture--can anyone advise now?

I have some hinges also at member ends as you may notice

 

[kslee1000]

Check sshield's response, it make sense to me - take out the diagonals for gravity load analysis, and make sure your model is stable under such load.

 

[271828]

Again, yank out one of the braced frames into a model by itself and see what happens.

 

[atee001]

I removed the bracings and checked that the frame remains stable.

The compressive force in the column (the column which has less axial force in the presence of bracing) without bracing was 7.89 KN

When, I put all the bracings the compressive force in that columns was 3.5 KN and in the brace was 8.5 KN?

Any thoughts?

 

[csd72]

It makes sense if you lateral loads on these braced frames are high compared to the vertical loads.

 

[271828]

Does the bldg drift in any direction due to gravity loads only?

 

[haynewp]

That was my first thought about drift too. But what is all of that you are representing in blue?

 

[kslee1000]

Ok, does the program allows "offset" commend?
If it does, "offset" the diagonal brace a certain distance from the upper brace-column joint.
If not, do offset manually.
The program couldn't recognize the secondary nature of the brace, it sees it as the stiffer main force resisting member in a truss, make sense? The offset should help to identify the main member - the column.

 

[steve1]

It appears from your graphic that the two outside bays that are normal to your end bay are not braced symetrically. In addition the end bay opposite your problem is not braced at all. If all that is true then you may be getting some global torsion about your center of stiffness. This would result in an axial force in the top horizontal member parrallel to your brace. This axial load would then be resisted by the brace which would be stiffer than the column.

Check your deflected shape for evidence of torsion and look for an axial force in the horizontal member framing into your brace. If this is the cause then your model is fine and you are getting realistic results.

 

[atee001]

Thanks Steve.Will check on that.Yes, there exist lateral drift.Enclosed is the displacement (translation)at the nodes.What does presence of drift(ux in the attached file) indicative of?

The blue colour are the hinges to have free rotation at those ends of the member

 

[kslee1000]

I think Steve got it. The graph did shown two locations of excessive rotation for no obvious reasons, but caused by geometric arrangement of the diagonal braces. It is also evident that without the braces, the model is stable under gravity load alone.

 

[atee001]

Steve,did you mean the horizontal members highlighted in the figure in the attached file?

 

[atee001]

I am also attaching the results of deformation for the selected beams above.

But,Steve as i mentioned , the brace is not stiffer than the column-both have same cross sectional area but length of brace is more than the column length

kslee2000, the graph attached was for translations-not rotations