AISC 341 Connection Design with Redundacny Factor

[T_Bat]

Hello all,

I'm designing a structure is SDC D with ordinary moment frames. To design the moment connections I'm using E1.6b.(b) which allows the use of the limiting earthqauke force (forces calculated with R=1). Since I'm essentially designing for elastic behavior (R=1) do I need to include the redundancy factor (rho)?

 

[T_Bat]

Seems like per ASCE 12.3.4.1 part 6 this (NOT using rho with R=1) would be consistent with the spirit of the code. Just wonderied if anyone had strong feelings.

 

[JoshPlumSE]

Use the Rho for the demand forces of members and columns (calculated using your real R value). But, use R=1, rho=1 for the design of the moment connections.

 

[JAE]

In looking at ASCE 7-10, section 12.3.4 Redundancy, the rho factor has several exceptions in 12.3.4.1. Item 6 in that list suggests that if you use an overstrength factor (Omega)is used in connections - which is an amped up seismic demand used for connections. This suggests that an R=1.0 condition - elastic response - would possibly negate that.

But since you are SDC D, the next section 12.3.4.2 requires two other conditions to be satisfied as well.

If you satisfy all that then obviously you can use Rho = 1.0.
But if you don't satisfy the two conditions in 12.3.4.2, then I still think that the Rho = 1.3 value is an attempt to add safety to structures with limited quantities of brace systems along a line - i.e. redundancy; which still seems to me to be independent of what R value you use.





Check out Eng-Tips Forum's Policies here:
faq731-376

 

[T_Bat]

Thanks for confirming. I've alsways done it this way but for some reason started questioning this job.

 

[T_Bat]

JAE - I am using rho for all member designs. But for SFRS connections I am designing with R=1 and rho=1.

 

[JAE]

I'm not sure I see that as correct -

Rho is dealing with overall structural redundancy which I think is dealing with the statistical chance of some unique, weird or other strange possible flaw, damage, poor workmanship, or insufficiency in a single brace along a line...this is independent of elastic behavior, R values, etc.
(where loss of a single brace in a line would be significant to structural stability)

The use of R=1 does not deal with that issue.

The use of various R values in your design takes you from a purely elastic condition (R=1) to a post elastic condition (R = something) which is reflecting the ability of your structure to survive in an inelastic condition.

Rho doesn't deal with that and even with R=1 you still would have a redundancy issue overall in your structure - assuming you fall under a Rho = 1.3 condition.

Using Rho = 1.0 regardless is saying - "I have no redundancy issue if I design to remain elastic".

You could still have a flaw, damage, or possible higher event than predicted and this would result in a connection that would "break first" over the member limit if you use R = 1 and Rho = 1 for everything.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[JAE]

Perhaps there is commentary on this somewhere (none found in ASCE 7) on the issue so I'm open to authority saying otherwise.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[jittles]

I agree with JAE's take on it - the rho factor is applied to the demand regardless of the R factor. In this case, I myself would use rho = 1.3.

 

[JoshPlumSE]

There are two different levels of force we're talking about. Design of the lateral force resisting system (beams and columns and such) which would use an R = 3.5 (for OMF) and a rho of 1.3.

Then there is the design of the OMF connection which is governed by the requirements of AISC 341 where the connection gets designed for an elevated force level based on the probable plastic moment of that connection, including expected yield (as opposed to min yield), and even strain hardening. This should be a much larger force than the rest of the lateral force resisting system is designed for. Essentially, this is a codified version of saying "the maximum moment that can be developed at the connection". It is conceptually the maximum you can get and conceptually there is no need to amplify this by rho. You certainly would never do this with an IMF or SMF where you're using AISC 358 connections.

Because this connection is an OMF instead of a SMF, we're only expecting minimal ductility, and you get let off the hook with a few things. The important one (for this discussion) being that the probably plastic moment need not be taken as greater than the demand moment you'd get if you used R = 1.0. This should still be 3.5 times the force you'd get if rho = 1 and R = 3.5 and 2.7 times the force you get using rho = 1.3 and R =3.5. So, there is a lot of extra safety built into this value even though we're only expecting minimal ductility out of the system.

 

[JAE]

This should be a much larger force than the rest of the lateral force resisting system is designed for.

Yes but.... the OP stated that they selectively used R=1.0 for their whole system.

For me - I would have a hard time designing the connections for R=1 and Rho=1 AND THEN the framing for R=1 and Rho=1 in an SDC D situation.
I would want my connections to not be the brittle link in the behavior, which is what I would fear with that approach.


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[T_Bat]

JAE - Sorry if I was unclear. I'm talking about using R=1 for connection design only. The overall system was designed with OMF criteria.

 

[JAE]

That makes a big difference I think.

Check out Eng-Tips Forum's Policies here:
faq731-376