On behalf of Norm Scheel, I am submitting the following letter he sent
to SEAOCC. He is responding to the "Perspective" section of the February
2000 Newsletter. Should anyone want to email Norm directly on this
subject, his address is Norm(--nospam--at)NSSE.com.
Terri Schneider PE
February 16, 2000
P.O. Box 2590
Fair Oaks, CA 95628-9590
To Whom It May Concern:
This letter is in response to the article in the "Perspective" section
of the February 2000 Newsletter, entitled "Hold-Down Eccentricity and
Capacity of the Vertical Wood Member".
While the exact test set-up is vague, I suspect it was a post pulled on
one end and resisted by a holdown device on the other end.
I have the following disagreements with the opinions expressed in this
1. I do not believe the holdown post is clearly a "tension member"
as defined by the 1991 NDS Section 3.9.1 and Figure 3G. This section
clearly depicts a member of a given length being pulled with a force on
each end. A holdown post has a connection at some location along its
length, which resists many smaller shear forces applied to it by the
sheathing nailing. I do not believe the allowable tension stresses are
intended for a member that only has the maximum tension stress at a
unique location along its length. This would combine the probabilities
that all the strength reducing characteristics allowed in the grade
would occur at the exact location where the tension force is highest (at
the top bolt in the holdown bracket). The published allowables are
based on in-grade testing which established values by pulling actual
pieces on each end (tension uniform over a length of +/- 16'). These
tests established the statistical distribution function of the pieces in
each grade. The 95% exclusion value was then adjusted by the
appropriate safety factors. These allowable tension values would be
completely different (higher) if they were based on this scenario where
the piece is not pulled on each end.
2. Secondly, it should be clarified that the tension in the post is
not the uplift on the holdown. The tension at the "critical section" of
the post could be far less than the force in the holdown bracket
depending on the vertical location on the posts.
For example, on an 8' wall, the top bolt in an HD10A is at least
18" up the wall. The maximum tension force in the holdown post would be
81% of the tension in the holdown bolt, and less if it was installed a
little higher as is the usual practice.
3. It appears the authors have treated the stiffness contribution
of the sheathing as inconsequential. I believe that the sheathing
contribution is very substantial. My opinion against theirs. Wood
shear wall systems are one of the most complex structural components we
design. The holdown and the post are only a small part of this
component. Isolating these and neglecting all the other corrective
influences is bogus.
4. The contribution of the holdown bracket fixity is not discussed.
The holdown bolt does provide lateral restraint to the holdown bracket,
thus reducing the moment in the post. I do not agree with the statement
in the fifth paragraph "the issues is one of simple statics". I assume
that if the authors actually did design shear walls, they would be
forced to specify a holdown manufactured by a company who has a
concentric holdown device.
Based on the authors' suggestion that the eccentricity be assumed to be
from the centerline of the posts to the centerline of the anchor rod, I
have prepared tables showing results for typical holdowns and posts.
Please note that on a 4x4 No. 1 DF post the HD2A will carry 2644# and a
HD10A will carry 2105#. The tabulated value for a HD10A is 9540#.
Something seems wrong.
It seems presumptuous of the authors to assume that the practicing
structural engineers who use these devices or their manufacturers have
never noticed the simple, obvious concept they are presenting. Could it
be that we are all that stupid or that we have considered all the other
factors that make the shear wall "system" work? I know of no reports
from past earthquakes pointing to holdown posts fractures at the top
bolt due to "combined axial and tension interaction". To me, that is a
good indication that this is not the weak link and that these
calculations should be tossed in with the calculations that prove a
bumble bee can't fly.