Need a book? Engineering books recommendations...

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

RE: component and cladding vs MWFRS : A curious dependence of ASCE7-05 on AS1170.2 ?

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Here's a curiosity: ASCE7-05 commentary indicates that the pressure
coefficients for open buildings were taken from AS1170.2 free roof pressure
coefficients. Free roof has zero walls and exposed frame and roof cladding
subject to frictional drag coefficients.

Suggests didn't do new empirical research. AS1170.2 doesn't have MWFRS and
C&C pressures. Instead we have local pressure factors kl, which magnify
pressures on small areas. For an area 'a x a' typical value is kl=1.5, and
for area 'a/2 x a/2' typical value is kl=2. These areas are located in
turbulent zones, outside defined turbulent zones kl=1. These local pressure
factors only apply for assessment of cladding and immediate supporting

For wind direction theta=0 (transverse) the turbulent zones are the sharp
edges at wall corners, anywhere on windward wall, and sharp edges at eaves
and ridge. For direction theta=90 (longitudinal) the turbulent zones are
sharp edges at wall corners, anywhere on windward wall, and sharp edge at
roof ends (verge of gable end). Since wind can change direction: for a
symmetrical building shape, theta=0 gives similar distribution to theta=180,
and theta=90 gives similar distribution for theta=270. We only get the one
set of pressure coefficients: say the equivalent of ASCE7-05 Fig6-6, and
Figs6-18A to Fig6-18D.

To AS1170.2 the designer has to work out which direction is the most
critical case for application of the local pressure factor kl, to the
coefficients ASCE7-05 classes as MWFRS. As such ASCE7-05 Figs6-19A to 6-19C
are some what misleading. Assuming they have been derived applying kl to the
MFWRS coefficients adopted from AS1170.2. (eg. no research to determine

To AS1170.2 for enclosed buildings kl is applied to external pressure
coefficients only, for free roofs kl is applied to net pressure coefficient.
Partially enclosed buildings are enclosed buildings with dominant openings:
getting Cpi is more complicated than ASCE7-05: it depends on the external
pressure coefficient on the surface where the opening is located and that in
turn depends on the relative direction of the wind.

AS1170.2 does not have a gust factor for static analysis. We have ka and kc.
The area reduction factor ka accounts for the fact that those so called
MWFRS pressures are extreme values and do not occur simultaneously over the
entire surface area: so for large areas ka provides reduction. The authors
of the timber framing code (AS1684.1) introduced kc, and it was adopted into
the wind loading code in 2002. For light timber framed construction the area
reduction factor is typically ka=1 : no reduction allowed. What they argued
was that the pressure coefficients are extremes and they do not peak on all
surfaces of the building at the same time, therefore for certain loading
distributions kc<1 and provides further reduction. But ka.kc not permitted
to be less than 0.8. Roughly if ka applies then kc doesn't, and vice versa.

The point is the MWFRS pressure coefficients in ASCE7-05 are in the main
equal to those in AS1170.2. With ASCE7-05 gust factor having default value
of 0.85, whilst AS1170.2 has area reduction factor with maximum reduction
0.8. To AS1170.2 these so called MWFRS pressures are considered conservative
for design of all structural elements except for: cladding and its immediate
supporting structure (eg. girts/purlins/battens). Unconservative is adoption
of an internal pressure coefficient of GCpi= +/- 0.18, when have no
guarantee doors and windows will be closed.

Since the extreme value for local pressure factor is kl=2, the typical
solution for roof purlin design is to half the spacing of purlins at the
ridge, and at the eaves, and in the gable end bays. For small buildings, it
is not worth the effort setting out different spacings, and equal spacing is
adopted throughout based on the extreme pressure zone. We do not however
declare this to be the prescription for all buildings.

The misleading part of the Figs for C&C pressures for open buildings in
ASCE7-05 is they imply a distribution of wind loading on the building: that
is pressures occurring simultaneously over the surface of the building. This
is not what AS1170.2 says. The extreme pressures are randomly distributed in
time. An extreme spike in pressure may occur at the ridge, but not
necessarily at the same time on both sides of the ridge. A spike may occur
at ridge and eaves at the same time, but low probability of doing so. And
the extremes do not occur along the full length of the purlin in the ridge
zone, only at certain points. Further more the extremes at eaves and ridge
are for transverse wind flow, whilst spikes in pressure at the gable end
roof edge occurs for longitudinal wind flow: thus not at the same time. How
do you apply those C&C pressures from ASCE7-05?

AS1170.2 approach is that the fabric of the building should be resistant to
the direct action of peaks in wind action. That is high pressures on small
areas which can lead to a breach of the building fabric and generate
dominant openings and high internal pressures. The local pressure factor kl
is there to provide added resistance to the building fabric. For example do
not want to loose the ridge cap, and then have the wind working away at the
free edge of the roof cladding, and tearing a dominant opening in the roof,
creating an internal suction of say Cpi=-0.9, to be combined with windward
wall Cpe=+0.8. We don't want small concrete roof tiles torn free which then
put a hole through a window, generating Cpi=+0.8. We want to maintain the
integrity of the building fabric, so that assumptions about the primary
structural elements remains valid.

So when the older engineers have an intuitive feel that the traditional
loading of 15psf was adequate for most buildings, they are potentially
correct. And when the manufactured timber truss manufacturers oppose
applying ASCE7-05 C&C pressures over entire truss span they also potentially

>From a perspective of AS1170.2. I would say for ASCE7-05 that gust factors
needs to be G=1 to be conservative for static analysis of small buildings,
not G=0.85. Secondly the ASCE7-05 commentary advises that internal pressures
may be higher than those put in the code: so a more rigorous assessment of
GCpi and adopting values greater than GCpi=0.18, would improve tie-down
requirements and strengthen/stiffen cladding and battens.

Thirdly if the C&C pressures are applied as spikes in loading in the zones
considered but not along entire length then the influence on over all
structure is diminished. For example a purlin in the ridge zone does not
experience the pressure from zone 1 along its entire length at the same
time, nor those peaks from zone 2 and zone 3. Therefore consider pattern
loading of the purlin, with peaks on individual spans, the rest loaded with
the so called MWFRS pressures. If only considering as single span then the
peaks control, but if consider as continuous span, then can gain some
benefit on long structures: note the longitudinal distribution of loading
Fig6-6. Similarly, if must apply to the truss, the peaks most certainly do
not appear on both roof planes at the same time. So once again consider
pattern loading combinations in the peak zones: with MWFRS elsewhere.

Conservative prescriptions are good, but not surprising housing industry has
some opposition if C&C pressures are being used across the board.

Also when I refer to high wind load tie-down, I typically mean a requirement
for tie-down rods from the roof to the footing: or more substantial
structure. In Australia we don't have any framing brackets as substantial as
those by Simpson strong tie: and if we did I believe carpenters would oppose
installing all the fasteners required. In any case steel straps for rafters,
and wall studs is a typical wind loading requirement: unless have heavy
concrete roof tiles in wind class N1 (qzu=0.69kPa), which permits
traditional skew/toe nailing. From which perspective the toe-nailing of
rafters, for light timber framing, in IBC:2003 for Vb=100mph seems highly
unconservative. Further to which the reference pressure qz, would be a
better constraint than regional velocity. There is more than one way to get
a given value of qz: vary Vb, or kz, or kzt. Building designed for given qz
should have same resistance irrespective of how qz arrived at.

As for minimum load, AS1170.2 has a minimum design velocity Vzu=30 m/s,
which would give qzu=0.54kPa, that is for ultimate strength limit state.
Wind class (AS4055) N1 has Vzu=34m/s (qzu=0.69kPa), this wind class does not
affect size of members for light timber framing, where wind class N1 and N2
(qzu=0.96kPa) span tables are combined. N1 and N2 only affect lateral
bracing and tie-down. Wind class N3 (qzu=1.5kPa) affects member sizes.

Basically I am saying there appears to be a misunderstanding in the
interpretation of the ASCE7-05 C&C pressure coefficient diagrams. Plus the
cause of failures seem to be unwarrantedly associated with misuse of MWFRS
and C&C pressures, rather than the most likely cause of failure being under
estimating the value of GCpi: which affects the result of both MWFRS and C&C
pressure states.

Conrad Harrison
B.Tech (mfg & mech), MIIE, gradTIEAust
South Australia

******* ****** ******* ******** ******* ******* ******* ***
*   Read list FAQ at:
*   This email was sent to you via Structural Engineers 
*   Association of Southern California (SEAOSC) server. To 
*   subscribe (no fee) or UnSubscribe, please go to:
*   Questions to seaint-ad(--nospam--at) 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: 
******* ****** ****** ****** ******* ****** ****** ********