The proposed structural system as you describe it has a lot of problems, in
my opinion. Several of the recent threads on this list regarding Chapter 16
of the 97 UBC are manifest in this building. There is little consensus on
many of these issues, as you may know.
I recommend strengthening the first floor (plywood?) shear walls, and the
diaphragm above, to resist the full base shear. Then you could use a higher
R of 5.5 (if plywood) for the WHOLE building, reducing the base shear. Then
the cantilevered columns can be deemed to be "not required by design to be
part of the lateral-force-resisting system," and would simply be designed
for deformation compatibility per 97 UBC 1633.2.4.
This approach would allow you to simplify the columns' detailing so that
they do not resist lateral forces, hopefully offsetting some of the
increased costs of the paragraph above. Or, you could still design the
columns with lateral resistance, although the code will not allow you to
"count" it. What I mean is, this approach will likely increase rho, and MAY
result in rho becoming greater than 1.0, which would be unfortunate.
My guess is that there are other architectural constraints of which I am
unaware, that may make this approach unfeasible. You may have to lengthen
the shear walls, for instance.
Otherwise, I believe you are stuck with the soft story that you describe,
unless more radical design changes are undertaken.
The code DOES allow a soft story. However, the penalty is severe. A soft
story is a.k.a. a vertical irregularity of Type 1 (Table 16-L), and you must
therefore perform a dynamic analysis per 1629.8.4.
I don't envy you trying to do a rational dynamic analysis/design of a system
for which we cannot agree upon the "correct" STATIC approach.
Mark Swingle, SE
Ken Tarlow wrote this.....
Thanx responding so quickly. In my example we have a frame (cantivered
columns) on the lower floor and plywood shearwalls on the upper floors.
When considering regidies the plywood walls may deflect 1/8" while the frame
will deflect .8 inches under similar loading. The lower floor is therefore
more flexible. Rigid Frames are a more flexible system than a shear wall
system. We therfore have a fleable system supporting a rigid system (soft
I would like to find away around this penalty on cntilevered columns, and if
I could am overlooking something I would like to believe I could use section
Please read C105.3.1 page 127 of the 1996 Blue Book
I regard it as apenalty because you penalize the whole first floor in my
case. Ther are two other shear walls on the lowest level but they are all
being penalized. In reading the blue book the lowering of the R value was
done because the factor of safety for cantilevered columns was reduced.
Perhaps a more appriate response was to change the K value ofthe column from
2 to 4. Drift governs anyway, these colums never really got stressed up.
But ifis true I canuse two R values in one direction. The lower value must
be used in the lower levels, How do I get around the problem of sitting a
rigid structure on top of a flexable structure. It seems to me the code
doesn't allow that.