Need a book? Engineering books recommendations...

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

Re: Diaphragm Chords

[Subject Prev][Subject Next][Thread Prev][Thread Next]
"Any comments?", you ask. Sure. Thanks for raising the questions.
Reply comments are added within your text:

At 04:31 PM 12/2/98 -0600, you wrote:
>Here is a question that I hope might generate some discussion........
>Have any of you ever wondered if a diaphragm can work without having a
>specific edge member, etc. assumed to resist the calculated chord force.

        Yes, I sure have. I was intrigued in the early 1980's by the
question of chords and their location with respect to the diaphragm edge and
when splices don't happen (like in your example below), and especially for
plywood diaphragms and tilt-up concrete or masonry-walled buildings.
Explored the principles of mechanics of this from many directions until
satisfied, and came to many conclusions, some quite startling and at odds
with "conventional wisdom". 

>Can a portion of the diaphragm along the edge actually work as the
>"chord"? It obviously has to be fastened together to resist those

        Every portion of the diaphragm will actually work the way it wants
to. The assumed need for a "specific edge member" is a convenient and simple
mental conceptualization that isn't unreasonable, or necessarily a gross
misrepresentation, but it is still a human-made assumption. 
        To give an example from a different branch of engineering, it was at
one time believed that if an aircraft was able to fly, it had to be in
conformance with the specific laws of aerodynamics as were assumed true at
the time. Then someone calculated how well a bumblebee complied with those
well-accepted laws. It came out that a bumblebee was so far out of proper
conformity that there was no chance it could fly.  

>This is similar to a large steel plate or concrete wall working as a
>shear panel  with no edge members. The bending stresses are simply
>resisted by the deep rectangular section with a  triangular stress
>distribution which is maximum at the edge (assuming deep member theory
>does not control).

        It seems that if there is bending, and there is only a deep
rectangular section available, it will have to resist in some manner. But
you appear to be imposing some assumptions on the thing, as to its stress
distributions within. I think for the inquiry at hand, it is better not to
assume that any conclusory, human-invented theories "control", and instead
discover from observation and open-ended tools of investigation what the
thing apparently does.   

>I could see this not working for fluted metal deck in one direction
>where the tensile chord forces would tend to open up the flutes.

        I see the same thing. We are looking to see what it wants to do,
rather than telling it what it has to do.

>To talk in practical terms, let's say we have a building with steel
>joists, metal deck and load bearing masonry walls. The joists span to
>the long walls (span parallel to the short walls).
>The bond beams and the reinforcing are cut at each masonry control joint
>(so that it cannot act as a chord). How does the building work to
>transfer the lateral forces. I believe the deck works as the chord
>parallel to the long wall. 

        Well, if the middle of the diaphragm length drifts sideways compared
to the ends, it would take on a shape we associate with flexural strain. As
long as there is some elastic or plastic modulus, strain means stress.
Flexural stress. The deck would appear to work to furnish its own chord
internally, to the extent that its elements and connections are willing to
resist strain and keep on participating.
        How the example building works to transfer the lateral forces is a
larger question however. Suppose additionally that the deck itself were cut
at the masonry control joints (or at tilt-up panel edges) so that NOTHING
above the floor slab ran through longitudinally across those transverse
joints. Except there is a shear transfer link from one segment of diaphragm
to the next, across the otherwise complete transverse diaphragm cut I have
added. Thus, each segment of building can transfer crosswise shear to its
neighbor, but nothing longitudinal connects the building segments in either
tension or compression above the slab and foundation. OK?

        Now, how does the the building work to transfer the lateral forces?   

        If this setup is checked out with an open mind and unashamed use of
old time free body diagrams and principles of static equilibrium, some
rather startling realizations can be found. These extend to applications
well beyond the situation you asked about.

        Suppose however, your question were supplemented with another:
"Regardless of how a diaphragm behaves with respect to any chord it might
have, how should an engineer behave in designing a diaphragm?"

        Galileo encountered something like this with his behavior after he
observed things about the solar system more authoritatively than anyone but
Copernicus had done before. His findings were most enlightening, and also
most unwelcome. He soon became a defendant in court, in Pope vs. Galileo.
Galileo's position was correct, of course, but the Pope won the case.

        Thus another problem beyond knowing how a natural phenomenon works
is the permissibility of making use of the knowledge, if it differs from
accepted "understandings" others have become habituated to.

        Charles O. Greenlaw, SE   Sacramento CA