Need a book? Engineering books recommendations...

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

RE: Rigid vs. Flexible Diaphragm

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

I will grant you that shake table models may not exactly match field conditions though I am sure CUREE would have made attempts to match. Maybe somebody directly associated with CUREE could respond to this. However, in spite of the exact match, I would think the test results would point us in the direction we should take in our design.

I would take exception to your statement that RDA is a justification used by most engineers who don?t design wood framing for a living. Opinions on the use of RDA or FDA may vary but I don?t think it is fair to brand engineers that use RDA in that category.

To the many engineers that claim that they have been doing wood design (or any design) in a certain fashion for so many years ? my question is ?so what? ?. People in the steel industry that had been designing pre-Northridge moment connection for years before Northridge hit and were forced to change. I am sure the Japanese engineers were cock-sure of the validity of their design until KOBE hit. Why, the floor rigidity was not even an issue until after Northridge.

Deflection due to nail slip is generally the largest contribution to the wood shear walls and wood diaphragms. Because the floor diaphragms are usually glued to the floor joists, the nail slip contribution is marginalized in the floor diaphragm making it much more rigid. In spite of the fact the deflection equations are based on experiments done using monotonic forces (I believe backbone curves are now available for shear walls), I think you would agree that diaphragms would act as a rigid member ? especially in a single family homes where close-spaced shear walls further reduce the diaphragm aspect ratio.

One of your main objection to the use of RDA was the potential hardship and liability to the engineer hired later on to make changes because of lack of records. This condition (lack of records) would apply whether the design was based on RDA or FDA. In either case, the new engineer that performs the change takes on the responsibility. I do not understand the logic to flouting the code now (and to expose yourself to possible lawsuit) so that somebody down the line several years from now does not have the liability. The engineer down the line takes on the liability regardless of whether the original design was based on RDA or FDA and whether or not he has access to original drawings/calcs.

I agree with you that quality control in the field is a big issue. It is a sorry state in which the owner usually wants the cheapest engineer and the cheapest contractor and will not hire an engineer to observe the construction. However, this condition would apply regardless of the design basis ? RDA, FDA, or 20% increase. I think, as engineers, we should strive to provide clear and adequate details so that the contractor does not have a need to ?field design the building? for us. But if the contractor does not know how to read the drawings, then ? well, you throw up your hands and look towards the sky.

Regarding the openings in the walls ? my intention was not to promote perforated shear wall but rather to accept the fact that the plywood around the opening will contribute to the overall wall rigidity. If one neglects this contribution, the argument for RDA becomes even stronger.

You correctly stated that deflection is an issue engineers should watch out for. The FDA method does not require the designer to worry about the wall deflection. On the other hand, the RDA method forces the designer to review the wall deflection as the wall rigidity is based on the wall deflection.

The use of proprietary walls, in my opinion, compounds the weakness of FDA. The proprietary walls will not have the problems about nail spacing, nailing, etc, you mentioned plywood shear walls would have. Since the proprietary walls will be stiffer than the plywood shear walls and since the deformation compatibility must be maintained, it would sure look like the proprietary walls would see more load, especially if the proprietary walls and the plywood shear walls are aligned (not recommended, though) or are close by. FDA will not account for this.

As Bill said, the 20% increase method unnecessarily punishes the weak wall. Based on RDA, the forces would be less in the weak wall. I would agree with him that maximum load on the weak walls should be based on the FDA. I strongly recommend the use of envelope method and, as you mentioned, keep on eye on the wall deflection. With computers these days, this information is at the tip of your finger.

Whether or not we agree or disagree on the analysis method, it is a fact that earthquake forces in the building will behave the way it wants to. As practitioner, we should try to understand the behavior. We can either try to change the behavior by providing adequate structural fuses or accept the behavior and design accordingly. Based on the arguments I made in my earlier e-mail, my vote is for RDA.

Gautam Manandhar, SE

From: "Dennis Wish" <dennis.wish(--nospam--at)>
Reply-To: <seaint(--nospam--at)>
To: <seaint(--nospam--at)>
Subject: RE: Rigid vs. Flexible Diaphragm
Date: Sun, 18 Apr 2004 20:08:11 -0700

Let me expand on some of the issues you raised. CUREE's shake table
models are fairly pristine - very boxy, and well constructed with
plywood shearwalls and diaphragms. But these are hardly examples of real
construction methods that we find in the field.

My greatest objection to Rigid Diaphragm Analysis (RDA) that I believe I
mentioned in my original e-mail, is the lack of practicality for use on
single or multiple family residential structures that "will" require
modification at some point in the future. Commercial buildings maintain
a preponderance of shear distribution to exterior load bearing walls and
designated interior walls that are not expected to change over time.
However, residential construction is dynamic - it changes on the whim of
the owner and future owners. In most cases this results in modifications
to walls (interior and exterior) that have been designed to resist
shear. Without the ability to reproduce the load path it becomes a
hardship to the future owner and a potential liability to the engineer
or architect hired to make the modifications. With the exception of
large tract developments, it is rare for a homeowner to produce the
original design package AND in most jurisdictions; it is not required to
list the engineer of record on the building permit application. In
addition, most engineers do not maintain their records past a certain
number of years - due to changes in recording media, lack of space to
store paper drawings and a number of other reasons - including the
potential for liability if the drawings are scrutinized by an expert
witness in the event the owner decides to pursue the engineer for errors
and omissions.

" 1. The material for the walls and the diaphragms are the same (plywood
however, the plywood in the floor diaphragm is usually thicker) ;"

I won't debate the rigidity of the diaphragm - inasmuch as the unblocked
diaphragm is only an arbitrary approximation of the calculated blocked
diaphragm. I would even grant you that the diaphragm is more rigid than
the walls based on the mass and aspect ratio - but whether we call the
old method of design Flexible or just Proportional distribution of
shear, we have yet to see a home that has been designed and attended to
in the construction phase by the architect, engineer or an inspector
representing the engineer to insure decent construction quality. Rather,
we see mistakes, we see plumbing chases placed in the middle of a
shearwall - windows added that were not on the original drawings but
never questioned in the field, shearwalls that stop at the ceiling and
are not connected to the roof, nails overdriven in 3/8" plywood or OSB
because the developer does not want to spend the money for 5-ply panels,
nails driven too close to the edge of the stud, over-drilled anchor
bolts, basically shoddy construction that is not corrected in the field.
Most of all, we see homes allowed to be designed by prescriptive methods
that even CUREE warned a year or so after they started testing, would be
the most critical and damaging of the building constructed.

Still, we are not talking about loss of life - but cost to replace
property which was added to the 1994 or 1997 Uniform Building Code in
the term "Major Structural Damage".  This was due to the Insurance
company leaving California and states in hurricane regions that were at
high risk.

" 2. The deflection equations for the shear walls and diaphragms are
(the floor diaphragm is generally glued to joist and therefore will
experience less deflection);"

Again, no argument but engineers are at fault for most of the damage
that occurred due to excessive deflection of shearwalls in past
earthquakes. This is a harsh statement, but to be truthful, I've been
designing wood structures for nearly twenty years and the issue before
the 1994 UBC was to disregard the wall drift, but to heavily load the
wall as long as you maintained a 3.5:1 aspect ratio. There is nothing
wrong with a narrow wall - even proprietary walls have aspect ratio's
that don't meet the 2:1 requirement by code and these are allowed
because they are tested for deflection. If we only calculated deflection
on shearwalls rather than adopted a limit of 3.5:1 in previous codes, we
would have seen that highly loaded narrow walls over-deflect and by a
great margin.

" 3. The shear walls generally have lot of window/door openings compared
the floor diaphragm, which usually does not have opening except for the
stair well"

Again this is true, but most of us who design wood structures don't
design perforated walls in high risk area's. Therefore, regardless of
the number of openings, the engineer designs with the walls that they
have and if these are insufficient, we can rely upon stiffer proprietary
walls such as the Hardy panels, or ShearMax, or Strongwalls. There are
choices out there and the openings can simply be ignored as long as the
continuity is maintained in accumulating enough diaphragms to
accommodate the shear demand.

" Based on the observation above, it appears to me that in typical
residential structures and in long narrow commercial structures, the
rigid diaphragm behavior concept is the way to go rather than the
flexible behavior concept."

This is the justification by most engineers who don't design wood
framing for a living. I am sorry if it sounds critical, I don't mean to
as your assumptions are valid, but impractical.

Look, we might be engineers but we serve the public - people like you
and I. Would you buy a car that would be next to impossible to repair?
Would you buy a home and want to add an addition or make changes to it
only to find out that the cost will be outrageous because the designer
needs to tear the home apart in order to reconstruct the lateral load
path? I don't think so - I think that we want to be assured of
performance, but most people are ignorant of the quality of the home
they buy and the potential out-of-pocket cost that they will be
responsible for should the home be damaged. This is because the industry
has no disclosure and owners don't understand "capitalism" of home
construction. They have a false understanding of what a building permit
means. To them it means quality that has been inspected and assured by
the building department. To us it means nothing more than minimum
compliance to the building code and having to design to the standards
established by a free-market to profit as much as possible by using as
little materials as possible to maximize the profit. To do this means
that we produce a home that has the potential for greater damage than if
we had a free hand to add an additional 2% in the cost of framing to
improve the performance of the home. We can do this - and do, for homes
that are custom built rather than tracts.

Whether you are an engineer, building department official, and educator
- you can't simply disregard the human element in housing design. To do
so is unethical and highly unrealistic. We are responsible to design the
best home we can within the guidelines of our clients. If a developer,
our hands may be tied, but as a professional community, we should have
an outlet to disclose the differences in construction between those
designed to maximize profit and those designed to improve performance
for an owner built home.

The bottom line is that RDA is practical when walls are not intended to
change in the life of the building - it is not practical in the dynamics
of designing and redesigning homes for the public. Stick to RDA in
Steel, Concrete and Masonry, but realize that there are too many
variables in Wood design that make it difficult to build a decent home
by Flexible analysis let alone complicate the issues by including RDA
into the equation.

We have to start thinking like practitioners rather than theorists who
believe we should design wood structures according to text book theory -
it doesn't fit in the real world.

Dennis S. Wish, PE

Dennis S. Wish, PE

California Professional Engineer

Structural Engineering Consultant


Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (
Version: 6.0.572 / Virus Database: 362 - Release Date: 1/27/2004

******* ****** ******* ******** ******* ******* ******* ***
*   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:
******* ****** ****** ****** ******* ****** ****** ********

From must-see cities to the best beaches, plan a getaway with the Spring
Travel Guide!

******* ****** ******* ******** ******* ******* ******* ***
*   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: ******* ****** ****** ****** ******* ****** ****** ********