Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: C and C wind loading (Roof Trusses are not C&C!)

[Subject Prev][Subject Next][Thread Prev][Thread Next]
{Not sure this got through on first attempt: didn't show on archives}

Andrew Kester wrote:

I have had to send back pre-eng wood truss calcs because they were based on
MWFRS. You better have a good argument prepared if you want to convince me
your average wood trussed roof (2ft o.c.) should be treated as MWFRS.

<end quote>

Andrew why make such demand? It seems unjustified and potentially erroneous
and unconservative: a truss is neither cladding or immediate supporting
component and the C&C pressures do not give the correct distribution of
pressures on the building surfaces. The MWFRS pressures provide the
distribution of pressures as airflows over the various building shapes. The
C&C pressures provide increased values of pressure which are highly
localized: they do not occur over large areas.

I am not overly familiar with US methods of house construction. But I
believe you fully sheath the roof with plywood or similar for bracing, then
install battens to provide venting between sheathing and external timber
shingles placed over. So the roof trusses are not immediate supports to the
cladding.

The C&C pressures are for highly localized peaks in pressure, they do not
occur over large areas. That is they do not occur over the entire span of
the truss at one time. Figure 6-3 in ASCE7-05 clearly shows the zones where
the pressures occur. What ASCE7-05 and its commentary doesn't make clear is
that the spikes in pressure only occur over a small area within these zones:
to AS1170.2 the areas are (a x a ) and (a/2 x a/2).

{Variable 'a' is defined differently in our respective codes: a =
min(0.2b,0.2d,h) where b=across wind building dimension, d=along wind
building dimension, h=average height,  In similar manner ASCE7-05 the local
pressure extent a= max(min(0.4b,0.4d),min(0.1b,0.1d,0.4h), 3ft)} 

ASCE7-05 has simplified to a pressure within a zone: but the pressure does
not occur across the entire zone at the same time. The pressures are small
turbulent peaks: micro-tornadoes if you like. These small peaks can fully
encompass a shingle or roof tile and remove them if only designed for the
MWFRS pressures. The pressures may also have an effect on thin materials
near the buildings surface, such as the plywood sheathing used for bracing.
But these peaks in pressure do not occur across the entire surface of the
roof: the peaks occur near the roof edges: ridge, eaves.

To help further clarify: wind pressure tests are done on small solid models
in a wind tunnel or monitored buildings. If the building surface was large
enough to be clear of the edge turbulence, then the pressure in the airflow
over the surface would be variable but not highly turbulent but not quite
laminar either. The peak values measured in this airflow are the external
pressure coefficients given in the codes. The Australian code considers it
unreasonable to apply these peak values over large areas for the design of
the primary structure or MWFRS, and so we have our area reduction factor
k[a]: ASCE7-05 Figure 6-6 provides some reduction for certain situations. 

The surface of the building however will experience these peak values, and
it may cause localized failure of the building fabric, and so to AS1170.2
k[a]=1 for cladding and components. Using ASCE7-05 you use MWFRS pressure
coefficients for the primary structure and C&C pressure coefficients for the
interior zones Figure 6-3, or Figures 6-11A -D to check the C&C. 

Near the edges of the building, wall corners, roof ridge and eaves, verge
and hips, there is high turbulence, small pockets of air break from the main
air mass, and small vortices form, all of which produce very localized high
pressures. The distance 'a' marks the extent of the localized pressure zone,
where this turbulence occurs. To the Australian code we apply the local
pressure factor k[l] to magnify the pressure coefficients, and apply this
localized peak pressure to areas (a x a) or (a/2 x a/2) the rest of the
element experiences the unfactored (k[l]=1) pressure, and is only applicable
to C&C {eg. Steel sheeting, concrete tiles, battens, girts and purlins.}.
The smaller areas (a/2 x a/2) experience the highest load increase (k[l]=2)
and the larger areas a smaller load increase (k[l]=1.5). Using ASCE7-05
simply look at the additional tables for C&C pressures in the edge zones.

Since ASCE7-05 places a limit of 3ft on the dimension 'a', anything smaller
than say 3ft x 3ft, or encompassed by (a x a) and within the edge zones,
such as an air-conditioning unit or solar hot water system, would be better
designed for the C&C pressures. But the MWFRS  pressures should still be
assessed to check the effects of the pressure distribution which may result.

The roof trusses maybe closer together than the minimum dimension for 'a',
but it is unlikely that the entire span of the truss is encompassed by an (a
x a) high pressure zone. If have a small gable roof then the MWFRS pressure
on the leeward roof plane is typically less than that on the windward: there
is a uniform pressure on each roof plane. For large roofs the MWFRS pressure
coefficients are stepped and generally reduce in value from the windward
edge to the leeward edge: both transversely and longitudinally. This stepped
reduction in loading can provide benefit to large buildings, because a large
portion of the building may only be experiencing a very low pressure: to
AS1170.2 buildings with plan dimensions greater than 3h see the benefit to
ASCE7-05 it is 2h. The MWFRS pressures are not low values considered
unconservative for small structures: they are relatively high and
conservative values. The MWFRS pressures are the appropriate pressures to
use for roof truss design. To further clarify ASCE7-05 Figure 6-11B (C&C)
shows that the wind pressures on either side of the gable roof are the same,
this does not represent the airflow over the building, it is simply the
requirement for checking localized effects on either roof plane: it is a
simplification.

To AS1170.2 there is no simplification. Instead we would use the equivalent
of Figure 6-6 (ASCE7-05) and apply the localized pressure factor k[l] to
both roof planes: since the wind can change direction, the maximum net
pressure coefficient gives the design pressure for cladding and purlins. For
example for a 10 degree roof with pressure coefficients of (-0.7,-0.3) we do
not design the purlins on the leeward roof plane for a lower load. The
ASCE7-05 C&C tables have already made the decision for the users.

The MWFRS pressures are the correct pressures to design the roof truss and
its tie-downs. The roof truss will not experience the MWFRS peak pressure
over its entire surface, and when the peak value does occur it will be for a
short period of time. The MWFRS pressures are therefore conservative. The
C&C pressures however may have an effect on the top chord of the roof truss:
since battens and bracing need to be fastened to the chord. The top chord
needs adequate depth and breadth to accommodate the required fasteners.

The size of fastener and depth of penetration into the timber is determined
from the C&C pressures applied to the cladding surface and transferred to
the supporting battens. The C&C load doesn't have to be traced back any
further than the adequacy of the batten support to accommodate the support
reaction. If the shingles and battens are designed to resist the localized
peak pressures, then they should not expose the surface of bracing to
pressures/suctions which may detach the bracing sheathing from the frame.

All structural elements should be designed for the MWFRS pressures, those
elements on the surface directly in contact with the airflow, and those
supporting such also need to be checked for the C&C pressures.

Of course this is my view from the perspective of another code, and only a
brief look at ASCE7-05. But I suggest that using the C&C pressures for the
truss design is erroneous, and the difference between MWFRS and C&C
pressures should be looked at more closely.

Probably not convincing enough. But it's a start.


Regards

Conrad Harrison
B.Tech (mfg & mech), MIIE, gradTIEAust
mailto:sch.tectonic(--nospam--at)bigpond.com
Adelaide
South Australia




******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at: http://www.seaint.org/list_FAQ.asp
* 
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*
*   http://www.seaint.org/sealist1.asp
*
*   Questions to seaint-ad(--nospam--at)seaint.org. Remember, any email you 
*   send to the list is public domain and may be re-posted 
*   without your permission. Make sure you visit our web 
*   site at: http://www.seaint.org 
******* ****** ****** ****** ******* ****** ****** ********