Follow-up on App. 6

[Lion06]

I figured I would start a new thread so that JAE's wouldn't get hijacked.

While I agree that the tension force in the flange keeps it from twisting, I believe that is only true for a section that is braced against twist at its ends. This is, after all, the basis of all the AISC equations. If that condition isn't met then all the equations go out the window.

Additionally, if it were true that bracing the compression flange against lateral tranlation prevents LTB, then 1.) how do you explain the sidesway web buckling phenomenom?
and 2.) Why does App.6 explicitly state that the section must be braced against lateral translation of the compression flange AND twist of the section? These are two distinct requirements.

 

[dik]

I think it would depend on the magnitude of the tension force. A tension force increases the flexural stiffness of a member; this would have the effect of increasing the lateral stiffness.

Dik

 

[kslee1000]

StrEIT:

I think you knew the foundamental on structural deformation, but let's just re-examine and review the geometries of member subjects to simple concentric axial & gravity loads.

For a member under axial compression, it has to shorten. When the material reaches buckling stress, it tends to bend side way, which is free to translate (buckling up or down is prevented by web).

On the other hand, a member subject to axial tension, it lengthens and stiffens (in all directions but axial). We all know the fact that a stiffened/tensioned string would not deform side ways without external forces to force the issue. Thus, the member remains straight in its horizontal plane until something else happens.

Now project these two members on a common plane, the distances between the two lines (representing the members) form the twist planes in between the members along their length. This (twist) is the phenomenon the lateral brace is to prevent.

Now let's look the effects of eccentric grivity load to the beam, assume the beam is free of lateral support but its ends, and loaded on the top flange.

The eccentric load tends to deflect the flange and rotate about flange-web joint, however, owing to rigid nature of the joint (flange-web maintains 90 degrees), bending is created in the web plate. The stress tends to travel down with diminishing intensity, with the distance of travel depending on the rigidity of the web. If the web is relatively rigid, I can see there is potential of buckling near the flange. Thus, brace the compression flange, with the brace at least half way down to the beam depth, makes sense.

For both cases, the distance between braces plays vital role.

 

[kslee1000]

Correction:

"I think you knew the foundamental on structural deformation, but let's just re-examine and review the geometries of member subjects to simple concentric [red]axial &[/red] gravity loads.

Please ignore the text in red.

 

[Lion06]

kslee-

I'm not sure how your post relates to my question. If bracing against lateral translation of the compression flange is the same as bracing against twist of the section, then why does AISC explicitly state that the equations are based on the sections being braced against twist at the ends?

 

[kslee1000]

StrEIT:

Sorry the above does not help. No comment on AISC.

 

[BAretired]

According to Article 9.2.4 and 9.2.5 of the CSA Commentary (condensed):

The possibility of twisting of a member at brace points should be investigated and if necessary, braced to prevent twisting.

The top (tension) flange of a cantilever, if not braced, can deflect more laterally than the bottom flange. Bracing of both flanges should be considered.

In cantilever-suspended span construction with open web steel joists, it is essential to analyze potential lateral displacements at the tops of supporting columns because the beam web is also in vertical compression.

An inflection point cannot be considered a brace point.

Simply supported beams in single curvature typically require only lateral bracing at the compression flange.



BA

 

[haynewp]

Additionally, if it were true that bracing the compression flange against lateral tranlation prevents LTB, then 1.) how do you explain the sidesway web buckling phenomenom?

I thought sidesway buckling is more a local consideration from point load bearing, and a function of the web slenderness. The web is in compression and the tension flange may not have the stiffness needed to restrain web sidesway.

2.) Why does App.6 explicitly state that the section must be braced against lateral translation of the compression flange AND twist of the section? These are two distinct requirements.

Where does it say that?

If bracing against lateral translation of the compression flange is the same as bracing against twist of the section, then why does AISC explicitly state that the equations are based on the sections being braced against twist at the ends?

Sounds like preventing lateral displacement of the compression flange at the ends would qualify as preventing twist.

 

[haynewp]

If bracing against lateral translation of the compression flange is the same as bracing against twist of the section, then why does AISC explicitly state that the equations are based on the sections being braced against twist at the ends?

It may be that they defined it that way since the moment is often zero at the ends, so it wouldn't be correct to provide a brace only at one flange at the ends of the beam to prevent twist.