Need a book? Engineering books recommendations...

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

Re: Rigid vs Flexible residential diaphragm discussion

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Thor, I would do it like you have.  I think your analysis is rational,
follows the intent of the code, and is therefore correct.  The UBC says to
"consider the rigidity of the diaphragm."  What it means is that one should
consider the stiffness of the diaphragm IN RELATION TO the stiffness of the
supporting shear walls.

You have done this, and your conclusion is that the two corridor walls will
share the tributary loads to the two walls nearly equally, due to the fact
that the diaphragm will span across the corridor with very little deflection
at midspan.  The diap behaves more as a continuous beam on springs (in this
case), than as a series of simple-span beams.

This is why, I believe, if the code were to ALLOW tributary forces to be
used in all cases (of wood diaps in residential buildings), as has been
proposed by Bill Allen, there should still be the OPTION to "consider the
rigidity".

Mark Swingle, SE
Oakland, CA

These views are my own....

-----------------------------------------

Thor Tandy wrote:

I have been following this thread with much interest and now I have a
question.

With the flexible diaphragm approach how does the UBC handle the situation
where you may a number of walls relatively close to each other.  E.g. where
you may have a corridor or two that are bounded by walls.  I had two cases
recently (I was doing the concept review) where we had a church with a large
open area and corridors at the perimeters.  Also there were stairs that had
several walls associated with the assembly.  I.e. a whole bunch of walls at
the perimeters and corners.  From a flexibility point of view tributary
areas to those walls results in very low forces while a rigid  diaphragm put
significant load on the walls.  We sidestepped the issue by considering very
close walls as single block entities and sharing the resultant in proportion
to the wall lengths.

How would you do it?

Thor Tandy