Need a book? Engineering books recommendations...

• To: <seaint(--nospam--at)seaint.org>
• Date: Fri, 9 May 2008 15:25:46 +0930

```As Scott suggests it is the size of the thing, that determines the choice,
but not entirely.

AS1170.2:2002, there is no explicit calling up of MWFRS and C&C. There is
just a reference to cladding and immediate supports.

The code has:
k[a] = area reduction factor
k[l] = local pressure factor

and newly introduced in 2002 is:

[] = subscript

The external pressure coefficients listed are the peak values from tests,
and it is considered unreasonable to apply the peak value over large areas,
the area reduction factor (k[a]) accounts for this by taking the tributary
area into consideration. For small areas there is no reduction in value and
the pressure coefficients remain unchanged.

The introduction of the load combination factor, is largely because the
authors of our timber framing code AS1684, thought it unreasonable to apply
the peak value pressure to all surfaces of a building at the same time. So
for certain load combinations the pressure coefficients can be reduced
further. However there is a constraint on the net reduction that can be
achieved from k[a] and k[c] in combination.

The local pressure factor is to allow for turbulence and small vortexes,
which can produce high pressures over small areas. In AS1170, these areas
are either (a x a) or (a/2 x a/2). It is to these areas that we apply the
local pressure factor k[l]. For (a x a) the value is typically 1.5, and for
(a/2 x a/2) it is typically 2.

These turbulent areas tend to be near the edges of buildings, but there is
also a localised high pressure zone anywhere on the face of a wall. There is
some confusion about the application of k[l]. It is meant to be applied over
an area (a x a) or (a/2 x a/2), thus whilst it does occur in the zones shown
in Figure 6-3 ASCE7-05, to AS1170.2 it is only applied to the localised
area.

Thus if designing the roof purlin, for in the localised pressure zone, a
distance 'a' from the eaves, then there is a localised spike in pressure
somewhere along the length of the purlin. The spike in pressure could be in
the middle of the span, or at the end of the span. The area could encompass
one purlin only, or several. The span of the purlin is subject to the main
pressure plus a localised peak. Rather than deal with the stepped loading,
manufacturers catalogues and texts have recommendations for averaging the
peak, and using a uniformly distributed load only.  (So if purlin is 3a
long, then the localised pressure factor could be on a length, either in the
middle 1/3rd or at the ends.)

For metal cladding, the permitted end spans are typically shorter than
internal spans, and so purlins are typically closer together anyway. In
extreme conditions however the solution is to simply half the spacing of the
purlins near the ridge and eaves, and in the end bays of the building.
Similarly the spacing of wall girts have half spacing in the end bays. If
using cold-formed c-sections or Lapped Z-sections, then a solution is to
increase the steel gauge thickness in the localised pressure zones.

These localised spikes in pressure could fail the cladding and immediate
supports, and thus turn the enclosed building into an open building.

To the Australian code, the internal pressure coefficient is equal to the
external pressure coefficient that would be generated on the building
surface if it was there. So for a wind ward wall with Cpe=+0.7, if a
dominant opening is placed in the wall then the internal pressure
coefficient would be Cpi=+0.7. The commentary permits allowing for holes in
several surfaces at the same time, assuming flow in equals flow out. And so
relative areas of dominant openings can be used to adjust the internal
pressure coefficients. For common situations Cpi can be simply obtained from
a table similar to ASCE7-05. Also most of what ASCE7-05 calls open buildings
we refer to as free roofs. Only net pressure coefficients are given for free
roofs. Some pressure coefficients are given for roofs attached to sides of
buildings: like carports and verandas. For most situations however it tends
to be a judgement call, taking a conservative estimate of pressure
coefficients from the available information.

The authors of our timber framing code (AS1684) for example choose windward
wall pressure of +0.7 below and external pressure of -0.9 above, to give a
net pressure coefficient of 1.6 upwards on the overhang of a house rafter.
Which many believe is excessive compared to the pressure coefficients for
attached canopies. Further AS1684 only uses this pressure for the rafter
span tables, it ignores it when considering uplift and main connections for
which it uses a Cpn= -1.1 (noncyclonic). The timber framing code also
arbitrarily defines the local pressure extent as 1.2 metres to simplify
calculations for the span tables, and adopts a net pressure coefficient of 2
for the zone, and is applied to the whole span of roof battens, or the whole
tributary area of batten fasteners (cladding fasteners are usually specified
by manufacturers and determined from tests.) The localised pressure factors
are not used for design of rafters, wall studs, hanging-beams, and the
likes.

>From a brief reading of the commentary to ASCE7-05 I don't believe it is
that much different than AS1170.2, just some minor differences in the
parameters used, affecting what is detailed and what is averaged. From the
commentary the C&C pressure coefficients seemed to have been averaged for
the zones, rather than give the peak for the (a x a) areas.

Here for standard verandas and carports, which are mainly steel cladding on
a frame, no one appears to have considered the C&C pressures for the frame:
by a back channel at the house wall, and a fascia beam on the other end of
the cladding span. If the local pressure factor was considered then other
factors would be considered in more detail to drop the pressures to which
k[l] is applied to.

Verandas and carports are also secondary structures: so could boost pressure
coefficients to those used for C&C, and adjust life expectancy and drop
basic wind speed (equivalent to changing importance factor: which we no
longer have.). In other words can get the same net pressure to keep the
market happy, but with a different justification. So the design-codes can
change but the product-specifications remain unchanged, just greater
restriction on where suitable to use.

So it is not entirely the size that determines whether MWFRS or C&C
pressures should be used, it is the actual function as indicated by the
abbreviations.

So for a small awning, it may be entirely within the local pressure zone,
and not worth the effort doing calculations for both MWFRS and C&C: and
simply opt for the C&C pressures. If there are rafters with purlins, then a
more economical structure may result if both MWFRS and C&C are appropriately
considered. Noting that the C&C pressures are not transferred all the way to
the foundations/footings.  The primary purpose is to stop ripping a hole in
the building fabric.

As for a sail-shade awning: isn't that a cable-net and tension-membrane and
complex to analyse? Starting with form finding of the initial shape, and
pre-tensions in the cables and membrane. And whilst not yet a mandatory
design case, drunken youths climbing into the canopy and seeing if they can
stretch it to the ground seems a common problem. Due to someone cracking
their skull open and dying, many of our local sail-shades have been removed.

This may be of interest for analysis:

http://www.arcaro.org/tension/

Regards
B.Tech (mfg & mech), MIIE, gradTIEAust
mailto:sch.tectonic(--nospam--at)bigpond.com
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