From: Charles Greenlaw <cgreenlaw(--nospam--at)speedlink.com>
Date: Fri, 03 Mar 2000 10:26:35 -0800
I disagree that the UBC "partially enclosed" definition applies to this
situation. That definition applies to openings in windward projected areas,
and their ratio to openings on all other projected areas. "Openings" is
defined as pertaining to the exterior WALL boundary of the structure. The
presence of a roof is implicit in the whole wind code for these definitions.
Here, you have no wall openings, but no roof. The UBC wind section does not
explicitly address such a situation, except at its dimensional extremes: The
low-rise extreme is parapets, where the opposite side parapet is a large
multiple of the parapet height away, and no shielding credit is given. The
high-rise extreme is a chimney, where the distance between windward and
leeward "walls" is short compared to height, if not to width. The base shear
for chimneys is little higher than for a single parapet wall.
Your situation is in between, as are rectangular tanks, but the code does
not indicate whether such tanks have lids or not, or whether it matters.
I conclude the code, which was intended to simplify ANSI A58.1, didn't
undertake to address your situation. The Wind Commentary to the UBC, by SEA
of Washington, 1991 ed, didn't reach your situation either, but did explain
"partially enclosed", and the ballooning effect that results.
My estimate is that there is substantial shielding of the leeward wall for
the dimensional ratios you have.
It would be interesting to arrange for a windy day and compare base shears
for two cardboard-box scale models, one open-topped and the other with a lid
on it. The two could be held by a common string looped to windward around
pulleys, so a tug of war results. Then the open-topped one could have its
rear panel cut loose at its vertical edges, and the pull it takes to
restrain the top compared with the pull to hold an isolated, parapet-like
panel of the same dimension.
Charles O. Greenlaw SE Sacramento CA
>James Lane wrote:
>Have an existing 20'x20'x16'height box with the top open. The box sets on
>the top of a roof.
>Question: Does the second (leeward wall) pick up full wind pressure as the
>first wall? Any good references for how much the second wall will pick up.
>In the past I have always been conservative and used full pressures on both.
>What if the bax were smaller the first wall would shield the second wall and
>you would think pressures would be less.
>Sounds like you have a partially enclosed structure. The near wall must be
>designed for both internal pressures pulling in and windward pressures pushing
>in. The far wall must be designed for both internal pressures pushing out and
>leeward pressures pulling out. In both cases, the effects are additive. Of
>course, for the structure as a whole, the interior pressures are equal and
>opposite. Depending on which code you are designing for, see either 1997 UBC
>Table 16-H, Item 2, Wall elements or ASCE 7-95, Tables 6-1 and 6-4.4
>Rick Drake, SE
>Fluor Daniel, Aliso Viejo, CA