SCBF compression only??

[aaronmcdevitt]

Hi, I am a month and a half out of school, and am modeling a 3 story, 340,000 sqft building in high seismic zone. In grad school, my professor skipped SCBFs altogether, saying in essence that they suck and to always use BRBFs instead. The owner and another PE are planning on using - as a single frame - 2 prefab SCB bays connected by a single BRB at the middle story. This is in addition to rocking STMFs at other locations (stiffness compatible). The current concentric braces are angled only one direction, so that they would all be in compression (or all tension) in a single frame line.

My question is does the AISC allow this, and how is it justified? I would think both braces would buckle, causing a story mechanism.

Aaron

 

[SteelPE]

I thought the AISC seismic provisions required you to have a maximum of 70% of members in either direction be tension bracing (section 13.2c in AISC 341-05). Now, if you comply with this provision then having compression bracing is not a problem. You can achieve this by flipping the direction of the bracing so that a maximum of 70% are in tension at any one time.

 

[aaronmcdevitt]

I am looking through AISC documents, in particular the Seismic Specifications. It says:

"Along any line of braces, braces shall be deployed in alternate directions such that,
for either direction of force parallel to the braces, at least 30% but no more than 70%
of the total horizontal force along that line is resisted by braces in tension, unless the
available strength of each brace in compression is larger than the required strength
resulting from the application of the appropriate load combinations stipulated by the
applicable building code including the amplified seismic load."​

Does "required strength" really mean the elastic stress that would go to the member, if it remained linear elastic? If so, this is still considering the loads with R factor, so wouldn't the braces still buckle? In addition, I can't seem to find a definition of "amplified seismic load" in AISC.

 

[aaronmcdevitt]

Nevermind the definition - it is in there, but for some reason is not italicized. On another note, is it possible to edit these posts so I don't end up posting over and over and over again?

 

[JAE]

Amplified seismic load is that load that has been factored up by the overstrength factor ([Ω][sub]o[/sub])

 

[aaronmcdevitt]

So, correct me if I am wrong, but what I am seeing is that assuming I=1, R=6, and Ωo=3, I may put all braces in one direction, so long as they are designed to take a load reduced by 6x and then increased by 3x. So I am seeing all braces buckling at 1/2 the actual expected seismic forces. I never studied dynamic buckling, but it doesn't sound good to me for an entire lateral frame to buckle at half load.

 

[SteelPE]

From what I remember, R is basically equated to how well the system can dissipate the seismic loads. The higher the R value the better the system performs. Most building I design use R=3. In my area, I don't see any advantage to using a higher R value.

In your case you are selecting a system with R=6..... which is a system that is expected to dissipate a fair amount of energy during the seismic event. However, you are using a system that doesn't quite comply with the provisions (by having all tension only bracing in the system). For this you pay a penalty by increasing the loads by omega in certain critical areas.

and no, you cannot edit your posts once they are in.... I imagine that might cause confusion when researching older posts or when coming in late to a post.

 

[JoshPlumSE]

Aaronmcdevitt -

Just to clarify, the base seismic value (before it is reduced by R) doesn't really represent your expected seismic force. It represents the expected seismic force if the system were to deform ELASTICALLY.

If you detail out the system per SCBF then you should get a good amount of inelastic behavior which will greatly reduce the forces your structure sees. Hence the reason why you are allowed to use the lower forces you get by dividing by R.... But, the only way you are allowed to get away with those lower forces is because your structure is detailed to provide the non-linearity. Part of that detailing involves designing critical elements to amplified (omega) forces.

Few people appreciate this right out of school. Many a young engineer has seen that his wind shears are 10% higher than his seismic shears and decided that wind controlled.... But, he (or she) needs to remember that the demand on his connections will still be higher for the higher seismic forces

 

[aaronmcdevitt]

Josh,
Let's say an elastic structure would see the equivalent of 10000 kips lateral. It is designed for 1667 kips, so it first experiences inelastic behavior at 1667 kips (ideally). Then a certain part is designed for extensive inelastic deformation, protecting the elastic members, which I assume should theoretically remain elastic up to and surpassing Cd*delta. This would mean the protected members need to support 8333 kips, assuming Cd=5.

I suppose one could design some members other than the braces to yield before the braces buckle, thus protecting the braces. Still, if the braces were designed for 10,000*Omega/R, they would buckle at 5,000 kips. In addition, it was my assumption that AISC required the braces to deform inelastically before other members. Where am I making a mistake?