Need a book? Engineering books recommendations...

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

Re: FEMA Bolted Connection

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Dear SEAINT friends: here is another question i had outside SEAINT pages and thought it is an interesting item to be posted here with my 2.5 cents response. Again the personal info is redacted. it would be interesting to hear from you on this.
Here is my reply , the question is below.

Dear xxxx: Thank you for the e-mail and kind words. Things are fine
around here although I got pretty busy with the Mac Arthur Maze fire.
Hope all is well with you  and the family.
As for your item on the bolted connections and the use of over-sized
holes in the column splices, I don't know if your subscribe to the
SEAINT posting page from SEAOSC?  If not,  I posted a response to a
similar question there just this past weekend. i have entered the
question and my reply below FYI.  It would be great to hear from you
what do you think?   Am I off?

By the way I was pretty interested and campaigned for getting this
slotted hole and specially over sized hole allowance into the AISC
seismic provisions for bolted connections. Of course the short slotted
and over-sized holes should be in the gusset only and not in the member
not to further reduce the net area of the member.

I agree with you completely that OS holes help in erection and
fabrication specially for large connections such as those large and
complex multi plane gussets  you sometimes  have in your tall buildings.
As for the column splices, everything that I know tells me that having
slippage in the column splices will be beneficial. Having cyclic
slippage of 3/8" or so that happens in over sized holes  instead of the
3/16 or so of  cyclic slippage  expected in the standard hole splices
will only help the situation.  Of course one should check the additional
drift , which simple calculation indicates to be in the order of 0.001.
Specially important is the fact that unlike the bolted moment
connections that the bolts are subjected to combination of gravity and
seismic and can slip under gravity negative moment,  the column splices
cannot slip under  gravity load since the column sections are in contact
bearing at all times under gravity load and the bolts can only slip due
to uplift of the column due to lateral loads.

Professor Ray Clough of UC Berkeley, in 1970's did shaking table tests
of steel structures with uplifting base of columns and showed that it is
very beneficial if one lets the columns uplift.
Later during 1980's , we did tests of uplifting column base plates in
braced frames and showed that it is very beneficial to allow column
bases to uplift a controlled amount. We also developed an innovative
"uplifting base plates" to do just that. All of this is in my upcoming
Steel TIPS report:  " Seismic Behavior and Design of  Steel Base Plates
in Steel Braced Frames", By Abolhassan Astaneh-Asl, to be released in
July 2007. The concept of uplifting base plate was used in seismic
retrofit of the Golden Gate bridge and the Carquinez bridge in the Bay
The beneficial effect of column uplift was best manifested itself during
the tests of steel shear walls we did a few years ago for  GSA/MKA where
the boundary columns and steel shear walls had bolted double lap plate
splices at floor mid-heights. Because of the splice slippage, not only
there was some damping but more importantly, the slippage of the bolted
splice added to the ductility of the system and prevented large
compressive strain built-up in the columns delaying column buckling.

To summarize it, I think we should use bolted splices in columns more
often and allow short slotted and even over sized holes in the splice
plates (up to 1/4' gap in the hole).  As for the drift, as I mentioned
above, my simple calculation indicates that such an additional slippage
might add 0.001 to the drift.
Hope this helps and sorry for the length of this e-mail. As a famous
person has said-don't remember the name now- I could not find the time
to make it shorter.
With your permission after redaction and removal of personal info, I
will post this on the SEAINT.
Best wishes.

xxxx, yyyyy wrote:

Dear Hassan, I hope everything is well with you and family.

Fema 350 allows for their pre-qualified Bolted Flange plate (BFP) connection to use Oversize hole for the plate while considering full bearing value of the bolt shear capacity. AISC Seismic provision also now recommends using OVS holes for Bracing connections based on the same Fema 350 prvisions. I like the concept very much since it facilitaes a site bolted connection for high demand connections while maintaining a manageable number of bolts. I would like to get your opinion on that. Also can this be extended to column splices?



* *

7 Message:0007 , June 3, 2007 from SEAINT 7
From: Abolhassan Astaneh <astaneh(--nospam--at)>
To: seaint(--nospam--at)
Subject: Braced Frame Connection, Slip-Critical or Bearing Bolts?

Dear SEAINT Friends: I have received the e-mail below from a SEAINT
list member privately and thought the question is very important and
decided to reply to it on these pages to be able to hear from other
SEAINT members as well. To preserve his/her privacy I have redacted the
e-mail and my response.

Here is the question followed by my 2.5 cents response. Please let me
know what you think either on the list or my e-mail:

Dear Dr.Astaneh,

I have been reading your posts on the SEAINT list and find them very
informative and in great detail. I am a structural engineer working in
xxxxx and a member of the list for more than xxx years. Although I use
the posts almost daily, I have never put up a question in public
domain. I have a question regarding steel bracing connections and would
greatly appreciate your thoughts on it.


We are designing a 1x story steel concentric braced frame structure in
xxxx. The codes used are UBC-94 & AISC ASD-89. Seismic zone is 2A. The
proposed connections are slip critical bolted connections using A-325
Assuming item 2 of section 2212.6.3.1 of UBC controls, the connections
need to be designed for 3 (Rw/8) times the force in the brace due to
prescribed seismic forces. Since we are designing the connections at
strength level using ASD, 1.7 * allowable stresses can be used.

The strength of the bolts in bearing type connections (Type N) is about
30% more than the bolts in slip critical connections. I am assuming the
connections will slip at strength level forces and the bolts will be in
bearing condition. This will increase the drift of the structure but
that is not critical.

The question is:

When designing the connections for strength level seismic forces, can
we use the strength of the bolts in bearing condition rather than
slip-critical condition?

xxxx yyyyy
Senior Engineer
zzzzzzz Consulting Engineers P.C.
(Address and Tel.)
A. Astaneh:
The current AISC Seismic Provisions (2005)  has a section (7.2) on the
use of bolted connections in Seismic Load Resisting System (SLRS) and a
commentary in the back of the document on this subject.  If I understood it
correctly, it says that the bolts in the connections of diagonal
members should be designed using bearing (and not slip critical
strength) but the bolts should be tightened and the faying surface
should have a Class A surface to deliver a coefficient of friction of
0.35. Also, it allows to use Short Slotted Holes perpendicular to
the direction of the load. Oversize holes (with hole diameter being maximum of 3/16" greater than the bolt diameter) are also allowed. For standard holes, nominal
bearing strength is limited to 2.4dtFu. There is a good commentary on
this in the back of the Provisions that clarifies these and more.

However, all of the provisions and commentary in AISC Seismic
Provisions (2005 and earlier) focuses on seismic issues only. I wish,
AISC had brought in some concerns and had some recommendation regarding
wind behavior of seismically designed braced frames into the design of
Seismic Lateral Load Resisting System. This is specially important in
medium and low seismicity areas such as yours. But this is the way
things are for now as long as separate code committees look into the
wind and seismic and other loads acting on a single structures.  What
happens is that, at least in some cases, when seismic load is governing
and you just follow the seismic design procedures and design your
bracing connection for bearing capacity, when the service wind load is
acting on your braced frame, the wind can cause  slippage of the
connection bolts in shear, creating not a safety problem but quite
annoying noises to the otherwise satisfied occupants! Don't even
mention what the owner would think of you as structural engineer!   I
have head that this actually occasionally happens although not very

In any event, I use the following approach in design and teaching and
recommend it in my consulting to others. For whatever it is worth, here
it is:


(Adapted in content and not verbatim from "Behavior and Design of Steel
and Composite Structures-Vol. I, Stael Structures", by and copyright
2007 Abolhassan Astaneh-Asl, Ph.D., P.E., all rights reserved. To be
released Jan 1, '08)

1. Design the connection as bearing connections for the governing
seismic or wind axial force in the brace. If seismic is governing, use
additional provisions in Section 7.2 of the AISC Seismic Provisions
(2005). If wind governs, use the AISC specification (2005).

2. Regardless of which load , wind or seismic, governs, use tightened
bolts and at least a Class A faying surface with 0.35 coefficient of

3. Check the design of connection for slippage under service
(un-factored) wind load and ensure that the shear strength based on the
slip-critical values is greater than 1.25x(design service wind load).
This serviceability requirement ensures that the structure will not
slip back and forth when there is high wind.

4. There is no need to check , or even avoid the slippage during
moderate or strong earthquakes since such slippage during earthquakes
that are relatively rare event (compared to service wind) can act as
"poor  person's friction device" and dissipate  quite a lot of energy
and cut out the path of transmission of the ground acceleration to the
the masses of the upper floors  (to some extent) resulting in reduced
inertia forces. Both effects are beneficial. Besides, to me it looks
like the last thing one should worry about during a moderate or major
earthquake is the noise created by the slippage of otherwise perfectly
safe bolts!  You may have other things to worry about including falling
objects such as that nice framed painting above your bed!

5. As for the drift, the research and shaking table tests of frames
with bolted connections , for example, (Nader and Astaneh-Asl, 1990)
have clearly established that during dynamic loading having slippage in
the bolt holes dos not necessarily increase the drift! That is true for
static loading and service wind, but, the drift due to seismic loads
are created by dynamic inertia forces, which depend on stiffness,
damping, character of the ground motion and mass. By changing the
stiffness , which is reduced due to slippage, and damping which is
increased during slippage, there is no reason to state that drift will
increase if bolts slip. To be safe a rule of thumb can be that you can
consider the drift might increase about 10-15% due to slippage. This is
based on a number cases we have looked at. If the case is important of
course you can always do time history-analysis using your ETABS or SAP.
  But, just taking the actual slippage in the bolt hole and converting
it via geometry to establish drift will not be quite correct. It might
be more reasonable to assume the connection will not slip. Do your
typical elastic analysis, get the elastic drifts, convert them to
inelastic drift using Cd, then increase it by 10-15% at the most.
Hope this helps and best wishes as always.
Abolhassan Astaneh-Asl, Ph.D., P.E.
Professor, UC-Berkeley
(Contact info 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: ******* ****** ****** ****** ******* ****** ****** ********