Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: HSS and FEMA 350

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Wow. you must have been typing all night! (-:

You have hit on a key "secret" in EBF design -- reducing the section of the
beam in the link is a very favorable way to simplify the design and
construction. The requirements outside the link can require expensive beam
reinforcement. It is most often much more economical to reduce the section
within the link instead. So I suggest you not try to eliminate that in your

All else looks logical to me -- you are creating the ductile fuse in the
link and designing the remainder of the system for the forces from your
analysis but also strong enough to remain nominally elastic as the link
undergoes any deformation during an earthquake.

It seems to be not well known, but EBF can also be used in the R=3 concept
when R=3 is permitted in the applicable building code. Essentially, the EBF
is designed as a nominally elastic system for the higher forces associated
with R=3 (versus the high-seismic alternative with R=8 or 7). Said another
way, treat the R=3 seismic forces just like you would wind forces. This
force penalty is most often significantly outweighed by the elimination of
the high-seismic detailing requirements, particularly stiffeners through the
link and the reinforcement issues I mentioned above. If you are interested
in this, the following paper is available:

    Stan Lindsey and Arvind Goverdhan, "Eccentric Braced
    Steel Frames for Wind and Low-to-Moderate Seismic Loads",
    Proceedings of the 1989 AISC National Engineering Conference.

Note that the authors refer in the lingo of the day to "Low-to-Moderate
Seismic Loads". In today's lingo, they would be talking about R=3 systems,
which are designed for nominally elastic behavior.

Hope this helps.


-----Original Message-----
From: Keith Fix [mailto:kefix(--nospam--at)]
Sent: Thursday, February 20, 2003 8:37 AM
To: seaint(--nospam--at)
Subject: RE: HSS and FEMA 350

Charlie, while I have your attention:

My latest foray into structural engineering has required me to design
eccentrically braced frames per the AISC Seismic Provisions.  The building
a single-story fire station, Sesimic Zone D.  One of six braced frames was
required to be an eccentrically braced due to architectural constraints, so
went ahead to make all the rest eccentric to reduce the R-value (not
insulation).  Since I hadn't run thru the new code yet, it took several
iterations to discove what I needed to do to get a code-compliant design. 
Hopefully I got it right.  Here's my final design procedure (after several
false starts):

0. Assumptions.  Although other approaches might work, certain constraints
limited my design approach.  First, I needed to keep the brace size to a
minimum.  Also, I needed to keep the beam flange size small.  Columns and
braces were both HSS.  Brace connections are modeled as pins.  In my case,
frames were inverted v-braces.  [Link to column connections seem daunting.]

1. Select initial sizes.  The sizes should not be conservatively large, as
will cause beams, braces, and columns to be unnecessarily large to create a
plastic hinge in the beam.  Also, I had to dog-bone the beam to form the
otherwise, I couldn't get the process to work.  More about that later.

2. Run analysis.  Include all code-required load combinations.

3. Check link rotation angle.  I used a method recommended in the military's
809-4.  Link rotation angle appears to be the most critical piece of
information for eccentric braced frame design, except possibly eVp/Mp of the
link.  If the link rotation angle is too large for the ratio eVp/Mp,
frame (read:link) stiffness and re-run analysis.  Eventually, the rotation
angle meets code requirements, and design continues.

4. Check the link.  The link needs to be initially selected as a dog-boned
section of the W-section beam, otherwise, the forces the link generates will
exceed the code-defined capacity of the rest of the beam, and the plastic
cannot form.

5. Check the brace.  Forces for the brace must be calculated by multiplying
brace force by the ratio of the the link plastic hinge forces over the
generated in the analysis.  Which forces?  I chose Brace Earthquake Load x
Capacity / Link Earthquake Load.  If somebody has another opinion, I'd like
hear it.  If the brace was too small, I picked a larger one, and re-ran the
analysis.  Analysis forces don't usually change much based on the brace

6. Check beam.  This is where things can get sticky.  I went over and over
again trying to get the frame to work w/o dog-boning the beam for the link,
success.  After dog-boning the beam, things worked better, except that axial
load sometimes becomes a problem if most of the frame force comes from only
side of the frame.  In such a case, the link axial load tends to be half
of the beam load, so that if the link beam is nearly at capacity, the beam
likely to be over capacity.  In such a case, the beam (and link) must be
re-selected, based on the forces available, and the analysis must be re-run.

With a new link size, link rotation angles, forces, and forces required to
the plastic hinge all change, so member checks must start over.  I got
and only had to do this once.

7. Check column.  This part is easy.

8. Design brace connections.  The brace to column connection is
straightforward.  I welded the gusset to the base plat and column face, and
tube brace to the gusset.  The brace to beam/link connection is a little
difficult.  Where the gusset connects to the beam, it cannot infringe on the
length of the link, so odds are the gusset weld to the beam will have an
eccentric load at an angle.  I've used the instaneous center of rotation
in the AISC Manual to handle that.

There are some other design checks, but the rest are fairly
Am I on the right track?

-Keith Fix
-Little Rock, AR

--- "Carter, Charlie" <carter(--nospam--at)> wrote:
> >I presume this discussion is limited
> >to moment-resisting frames?  For example,
> >are eccentrically-braced, non-moment-resisting
> >frames impacted?
> Yes, FEMA 350 is limited in scope to moment frames, so CBF, EBF, etc. are
> not affected. One caveat: if you put the EBF link directly to the column,
> opposed to internal between braces, that moment connection to the column
> be affected.
> Charlie

Do you Yahoo!?
Yahoo! Tax Center - forms, calculators, tips, more

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ******** 

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: 
******* ****** ****** ****** ******* ****** ****** ********