From: John Rose <jrose36(--nospam--at)EARTHLINK.NET>
Date: Wed, 02 Jun 1999 19:03:57 -0700
Most walls have been tested with 1:1 aspect ratio, but these are made up with
two panels having 2:1 aspect ratio. Shear wall allowable design loads in UBC,
and deflection formula, works ok for walls having aspect ratio of 2:1 or less.
For higher aspect ratios, there seems to be agreement that a modification factor
(aspect ratio factor) should be applied to shear wall values in code. We (APA)
are studying various methods of developing such factors.
With regards to wall deflection, try using the UBC shear wall deflection formula
and see what happens to wall deflection when the aspect ratio changes. Start
with a given shear force (such as 720 lb) and an 8 ft high wall, and common
sheathing (3/8" STR I wood structural panels) and nailing (8d common spaced 4"
oc) and hold-down slip of 1/8", then calculate deflection for various aspect
ratios: 2:1(4 ft length), 3:1(2.67 ft length), 4:1 (2 ft length), and 6:1 (1.33
ft length). You will see that the wall deflection increases pretty fast as the
aspect ratio increases, such that about 3.5:1 aspect ratio, 0.5% drift is
exceeded. It appears that application of a shear wall aspect ratio factor could
compensate for deflection by reducing allowable shear load in these situations,
so that deflection would be within the range for "normal" walls.
John Rose/APA, Tacoma
Jeff Smith wrote:
> Mr. Rose:
> Thanks for the reply. What shear wall heights and aspect ratios were
> |~| The cyclic load shear wall test project sponsored by APA at
> |~|UC-Irvine in 1995 showed that UBC deflection formula predicts deflection
> |~|accurately for shear loads up to 40% over UBC allowable values,
> |~|which is still
> |~|less than yield limit state. Two other factors need to be considered: the
> |~|hold-down will now be required to resist the total force (based
> |~|on 900 plf for
> |~|example); and check bearing stress of end posts on plates. The
> |~|end posts may
> |~|have to be increased in size or number so that bearing does not
> |~|control. (I
> |~|recommend using 1977 NDS bearing stress f-c perp. = 385 psi for
> |~|Douglas fir.
> Someone should invent "Supersill"(TM), a sill plate with fc perp=fc par of
> the hd post.
> Jeff Smith