From: "Martin W. Johnson" <MWJ(--nospam--at)eqe.com>
Date: Fri, 16 Mar 2001 09:03:15 -0800
I suggest you design the beam and supporting columns for omega (2.8) times the
forces from the wall, times 1.4 if your forces have already been reduced to ASD
levels.. Remember that the shear strength of the wall that you are using is an
ASD value. If you look at the plywood shear wall test data that was developed
at UCI using shear panels with height/length of about 1, the actual ultimate
strength tended to be about 2.5 to 3 times the tabulated allowable shear values.
The code provision is intended to ensure that the supporting beams, trusses,
columns, etc., do not fail due to downward forces caused by this overstrength.
Although the code applied force will be large, the code allows you to use
special inrease factors when you calculate the strengths of the supporting
The SDM example suggest that the hold-downs for the shear wall might also need
to be designed using this increased force. I personally do not agree with this.
While the irregularity condition can be detrimental to the elements below the
wall, the shear wall itself is not different in essence that any other shear
wall in the building, so should we single out this one wall to require stronger
hold-downs? The concern is for the downward force, not the upward one.
Presumably, if the ultimate tested sthrength of a shear panel can be as much as
2.5 to 3 times greater than the ASD shear, while the maximum strength of the
hold downs may only be about 1.7 times greater than the permitted design
strength, then the hold-downs that we use in walls may often be a "weak link"
that develop yielding. If we start increasing the strength of the uplift
connection, then the possible downward force at the toe of the wall is able to
correspondingly increase, and it is the compression force that the code is
Incidentally, the UCI test data also includes some data for shear panels with
height/length of 2. The ultimate strength of these panels tends to be about 30
percent less than for the more square panels. So why don't our codes make any
distinction between allowable shear values for shear walls based on h/L? It
seems that the factor of safety is varying.