Need a book? Engineering books recommendations...

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

Re: Lumber Moment Frame Connection

[Subject Prev][Subject Next][Thread Prev][Thread Next]

I won't do a nice long dissertation since Ben already had a fairly decent

My short response is that I would suggest to keep in mind that it is not
just a matter of doing calculations per a material specification (i.e. NDS
or AISC spec) to show that there is "positive margins of safety using
rational engineering analysis" when it comes to dealing with connections
resisting seismic loads.  One can easily design a nice moment connection
for a steel structure that will adequately resist some prescribed
moment/forces.  The problem is such specifications don't really have a
nice easy way to design/analyze the non-linear performance of such a
connection and whether or not it will truly respond ductily.  In practical
terms, it is a matter of being about to determine how well such a
connection will handle the loads in a ductile fashion (i.e. what R value)
such that we can then more accurately determine what psuedo-static lateral
loads to use.  So, even in structural steel, it is NOT just a matter of
using some rational anaylsis to design/detail a moment connection per the
AISC must also detail that joint to have ductility and this is
basically done by either testing a joint or using pre-approved joints that
have shown how ductile there are or are not.

Thus, for a wood moment connection, while one can certainly do a rational
analysis per the NDS (and AISC spec) to show it can resist some load and
show it to some degree that a code official would have little arguement,
one cannot, however, easily show how that joint will perform in the
non-linear range and how much ductility it will or will not have at least
just by some rational analysis/engineering calculation.  One would also
have to do some testing (since there are no pre-approved joints in wood).
And in this regard, I would argue that you would not likely come up with
any real joint configurations that would show much ductility with
testing...but if you could, then a code official would not have as much of
a leg to stand on.

Now, I will note that even if you do come up with testing showing some
sort of ductile performance of some joint or system, you still will have
problems.  This is due to the somewhat political nature of the R values.
Ben did a nice little explanation of the theory of how an R value
should/would be determined, but the truth is that the current process for
determining R values is not too terribly based upon rational engineering
thought, but more upon political "back room" negotiating.  So, even with
some testing showing a ductile performance of some wood moment connection,
one must still convince a code official of some reasonable R value to use.

Now, a lot of the above "falls away" to some degree when you talk about
loads other than seismic.


Adrian, MI

On Thu, 1 Mar 2007, Jordan Truesdell, PE wrote:

> If the calculations were performed per the NDS and AISC spec and all show
> positive margins of safety using rational engineering analysis what
> grounds does the plan checker have - other than their gut feel - to deny
> the permit? What magic is occurring in the moment connection which does
> not result in tension, compression, and shear forces of a known quantity?
> Given the dimensional change and deterioration of lumber in the natural
> environment, why is it that bending moment, shear, and defined
> fastener-steel-wood interactions  are allowed for nearly every other
> case?  Is the lumber somehow aware that there is a moment connection at
> the free end?  Does a post embedded 3' into concrete or soil figure
> there's no use in failing because there's a special formula in chapter 18
> that allows such a use?
> From experience we've all seen "moment connections" in poorly built
> lumber/timber that just don't have much rigidity left, but is it a good
> idea to avoid such designs because of anecdotal evidence of poor
> craftsmanship? I know of gambrel lumber roofs built with plywood gusset
> plates which have withstood the test of time, yet they function thanks to
> two moment connections using just plywood and nails.
> While it's always fun to argue whether or not we can reasonably predict
> the forces imposed on structures from earthquakes, we nevertheless have
> all settled on a set of values. We also have a reasonable idea of where
> how a structure will respond, and can use conservative values where there
> is less certainty. Currently, there is no data in the IBC2003 which
> addresses the lateral shear performance of log homes. Should we outlaw
> their construction without steel moment frames, or perhaps require
> sections of SST panels with log siding interspersed among the 20-30"
> diameter logs? I'm not aware of any exhaustive research which would
> characterize such a system to the point that an effective R value has
> been determined, yet they appear to get built.
> I don't necessarily mean to take the discussion in this direction, but
> have we, as engineers, fully given up engineering as a science in favor
> of checking off boxes and filling numbers into equations in a book? I
> suppose that depends on whether the plan checker asked for calculations
> or simply dismissed the plan as "not in the code". I believe it was the
> latter, in which case we're just checking boxes.
>  Jordan
> Benjamin Maxwell wrote:
>       Jordan,
> I thought I'd respond to your post to keep the conversation
> (gentleman's debate?) going...
> (Yes, it's a slow day at work, but I'm not moving to Palm
> Springs...yet)
> I don't think anyone on the List argued that a wood moment frame
> connection would not physically work (in terms of preventing
> collapse), just that it would not work within the defined
> limitations required by our building codes.
> We all would probably agree that for a small canopy, a metal
> strap connection between the beams and posts would probably hold
> the structure together for most earthquakes.  The question is that
> would the system do so with a degree of reliability that
> would assure a plan checker to allow its use.  I think this is the
> aspect of the question most engineers on the List responded to (me
> included).
> Without testing the proposed system, one could not decisively
> conclude that there is adequate ductility capacity and hysteretic
> energy dissipation to limit the performance of the structure to
> acceptable levels (stresses, strains, displacements, etc.).  Even
> with testing, there is still a considerable amount of uncertainty
> (acknowledged or ignored) in how a system will perform under actual
> earthquake loads (Need I mention steel moment frames?).
> The wood - bolted steel plate moment connection in question could
> undergo brittle fracture of the steel strap at re-entrant corner in
> the first cycle of ground motion.  Or, the wood could shrink and
> crack over the course of time, substantially reducing the capacity
> of the connection as intended.  I won't bring up cross grain
> bending.  The point is that you can't assume that a given
> connection possesses ductility and will exhibit robust hysteretic
> energy dissipation.
> With that established, I hold that the building official in this
> case was performing his job in asking the engineer of record to
> substantiate his design, since the connection is not a recognized
> lateral load-resisting system.
> -Ben
> ******* ****** ******* ******** ******* ******* ******* *** * 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: 
******* ****** ****** ****** ******* ****** ****** ********