Need a book? Engineering books recommendations...

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

Overview ASCE7-05 and IBC:2003 Wind loading.

[Subject Prev][Subject Next][Thread Prev][Thread Next]
A quick overview on ASCE7-05 and IBC:2003 Wind loading.

clause 6.1.4.1 (ASCE7-05)
minimum pressure 10 psf (0.48 kPa)

>From quick scan of code:

Max. GCp = -4.1
Max GCpi = +0.55

Net ignoring modification factors which may have to be applied separately
and differently to internal and external coefficients.

Cpn = 4.1+0.55 = 4.65

qz = 0.613 Kz*Kzt*kd V^2*I
kz=1, kzt=1, kd =1, I = 1 
V= 37 m/s (85 mph)

qz = 0.613 * 1 * 1 * 1 * 37^2 * 1
qz= 839.2 Pa
qz = 0.839 kPa

p = qz*Cpn = 0.839*4.65 = 3.9 kPa (81.5 psf)

Unless dealing with large structures expect net pressure on an element of
the structure to be significantly less than this.

For any project qz will be pushed upwards and closer to 1 kPa (20 psf) and
higher than unity for tall buildings, whilst Cpn will typically be pushed
down and closer to 1. By definition kz=1 for open terrain and 10m (33ft)
above ground level. For suburban terrain and mostly 1 or 2 storey
residential it will get lower than unity.

So if adopt qz=1 kPa and Cpn =1 for design purposes, get p=1 kPa (20 psf).
For suburban buildings typically less than 10m, expect qz to move away from
1 and get lower, whilst otherwise expect Cpn to be higher than 1. The rise
of one variable and the fall of the other should have a net effect of
keeping p= 1kPa (20 psf).

Alternatively adopt net presssure coefficient of 1 and calculate qz, which
effectively reverts back to more traditional codes which simply calculated
qz and applied to building. It is not complicated to calculate qz for each
project, most 'k' values can be set to unity, and only kz needs to be looked
up to account for the height of the building. For a 10m high building in
exposure 'B' it should be less than unity.

Anycase from above qz=0.84 kPa, therefore p=qz.Cpn=0.84*1=0.84 kPa (17.5
psf) as a typical design pressure.

Table 6-3 doesn't have 10m (33ft), but for 30ft exposure B get kz=0.7, and
thus pressure of p=0.7kPa (14.6 psf) {assuming had qz=1*kz}, which is close
enough to the 15 psf, some have reported as being the value in more
traditional codes. And if Cpn=1, and Cpi=0.55, then need to check that Cpe<=
0.45: which typically it won't be, hence my original earlier post about
adopting Cpn=1.2, requiring a check that Cpe<= 0.65. 

{May have to adopt a higher value for use with ASCE7-05, since to AS1170.2
most users are using an internal pressure coefficient which is constantly
being argued as too low: typically somewhere between 0.2 and 0.4. To
AS1170.2 however an opening on a windward face would generate Cpi=+0.7,
whilst a side wall opening a maximum of Cpi=-0.65, and a skylight opening
Cpi=-0.9: assessing internal pressure coefficients to AS1170.2 can get
complicated balancing the flow in and the flow out from various openings.
Hence adoption of a simple value which doesn't quite make the building
sealed, but not open either. There are a multitude of factors which can push
the load up or down: what matters is the net load the building is able to
resist, not the details of how the applied load was obtained.}

The code may push in a lot more detail and involve more running around from
page to page, but what I have presented is the general gist of the code. And
both competent designers and certifiers should be capable of assessing
qualitatively that the code is not applying greater loading to a specific
building: and do so without need for presenting detailed working to the
formula in the code for each and every project, and most certainly not
presenting excessive detail for simple structures. As for the gust factor
most of the time it appears already combined with Cp to give GCp in the
tables.

The code may have been placed into law, but it is none the less still only a
guideline. Leave the excessive detail for self-study between projects and
should you ever experience a failure and need to demonstrate faulty
construction rather than faulty design. But for day to day design, keep it
simple.

However, it is the designers task to simplify the code, and find practical
means of reaching appropriate design-solutions. It is the certifiers task to
have adequate qualitative understanding of the code to check compliance of a
proposal without conducting extensive calculations or demanding submission
of such.

Last year list members provided links to sites on professional ethics, some
of the legal case descriptions clearly indicated it is not necessary to
carry out extensive calculations: the requirement is that the proposal
and/or finished buildings have adequate resistance and such can be
demonstrated to still be the case if more extensive assessment is conducted.
A competent engineer would know so, without having to do the more extensive
assessment: they would know the relationship between the simple and the
extensive and know which direction the result is going to be pushed. More to
the point they would carry out the simple assessment to provide a safety
line, before wading into the detail of the more extensive calculation
method. They also would not wade into such detail unless there was some net
benefit to be gained.

So if 25 psf gets the builder complaining about expense of construction,
then wade into the code to find a means of pushing the load closer to the
minimum of 10 psf.

And to clarify the word "simple" in the simple method, refers to restricted
and limited in scope, not to ease of use. The simple method in IBC:2003,
should be found to be aligned with ASCE7. I would say using pressure
coefficients (GCp) and calculating qz, is faster and easier than looking up
pressures.

So for say a roof angle of 20 degrees, then IBC table 1609.6.2.1(2) is
simply varying the value of qz with variation in wind speed, for constant
pressure coefficients GCp. So for example the table of lambda values is
simply an adjustment of the kz values, with origin of exposure 'B' rather
than origin exposure 'C'. So set kz=0.7, and calculate qz for each wind
speed in Table 1609.6.2.1(2), with all other values defaulting to 1. Then
for say Zone E estimate GCp for each windspeed, and should get same value
for each. (eg. GCp=1.06 to 1.07 from figure 6-10 ASCE Zone 2E has a  value
1.07 hence some rounding error in the pressures put in the IBC table). (I
have a trivial Excel scratchpad I used to check my guess of the
relationship: if any one wants.)

There appears to be a slight problem in that IBC seems to be ignoring the
influence of the internal pressure coefficient: but then have that
cumbersome MWFRS and C&C prescription. When considering forces on over all
building the internal pressures tend to cancel out: internal pushes down on
floor with equal and opposite force to which lifts up on roof. However when
considering an individual element such as a rafter, that is not the case and
both external and internal pressures need to be taken into consideration. Or
put another way: tying the building down doesn't tie the roof down, and
stopping the building from racking over doesn't stop the wall stud from
being bent and snapped.

So may be those who contributed to the simplified method, would like to
explain more clearly the relationship between the simplified method and the
analytical method of ASCE7, so that users of IBC/UBC can see the benefit of
ASCE7 and ease into its use.

For I don't see how IBC Table 1609.6.2.1(2) can be considered easier or
better to use than Figure 6-10 ASCE7. Using ASCE7 draw a diagram and apply
appropriate pressure coefficients, if lucky then 80% of the time the
coefficients will be the same across projects. Then calculate qz with kzt=1,
kd=1, and I=1, and Vb as required, and look up kz instead of lambda. The
rest of ASCE7 analytical method is complication which disappears for simple
structures for which the simple method is applicable. But don't use the
simple method use the analytical method and refer to the simple method for
guidance on the complications which are redundant.

The benefit is that there is more than one way to get a given value of qz,
and as I have explained before the WFCM guides for 90, 100, 110, 120 and 130
mph, have been unwarrantedly restricted in application by referencing
exposure 'B' and the basic wind speed: since the only difference between
these guides should be the value of qz used: the pressure coefficients
should be the same.

Maybe the AWC and AF&PA would consider introducing the concept of an
equivalent design wind speed Vz. In Australia we calculate Vz first then qz,
so we use multipliers 'M' which are effectively the square root of the 'k'
values used by ASCE7. But rather than mess around with the 'k' values,
calculate qz as usual, then transform to an equivalent Vz.

For Example
[Australia AS1170.2
Vz = Mzcat * Ms * Mt * Md * VR  [m/s]
qz = 0.5*1.2*Vz^2 / 1000 [kPa]
]

Using ASCE7
qz = 0.613 Kz*Kzt*kd V^2*I [metric]
qz = 0.00256*kz*kzt*kd*V^2*I [imperial]

qz = 0.00256 * Vz^2 
therefore
Vz = sqrt( qz / 0.00256)

or
Vz = sqrt(qz / 0.613) [metric]


Therefore with default values:
Exposure B (Vb=85mph, h=30 ft, kz = 0.70) get Vz = 71.1 mph [32m/s]
Exposure B (Vb=90mph, h=30 ft, kz = 0.70) get Vz = 75.3 mph [34m/s]
Exposure B (Vb=100mph, h=30 ft, kz = 0.70) get Vz = 83.7 mph [37m/s]
Exposure B (Vb=110mph, h=30 ft, kz = 0.70) get Vz = 92.0 mph [41m/s]
Exposure B (Vb=120mph, h=30 ft, kz = 0.70) get Vz = 100.4 mph [45m/s]
Exposure B (Vb=130mph, h=30 ft, kz = 0.70) get Vz = 108.8 mph [47m/s]

whilst
Exposure C (Vb=85mph, h=30 ft, kz = 0.98) get Vz = 84.1 mph

and so the WFCM guide for 110 mph exposure B should provide adequate design
solution for 85 mph exposure C, for that matter within round off error WFCM
guide for 100 mph Exposure B should be adequate. What could be easier than
that? Assess a site/building and adopt a pre-engineered solution, and let
the builder get on with their job, instead of causing unnecessary delay. So
if seismic is the critical issue, select suitable WFCM guide for wind
loading as evidence of suitability for wind load then check seismic by
calculation.

As for wind loads lying between those of the guides, does the engineer
really provide any economic benefit when construction costing is based on
crude measures such as $/sq.ft or $/ton.

In Australia we have a wind classification system (AS4055) derived from wind
loading code AS1170.2, our timber framing code (AS1684) makes use of AS4055:
there are 6 wind classes N1 to N6 for non-cyclonic and C1 to C4 for
cyclonic. The wind speeds for the cyclonic are the same as for wind classes
N4 to N6. For timber sizes, span tables for wind classes N1(Vzp=28m/s,
Vzu=34m/s) and N2(Vzp=33m/s, Vzu=40m/s) are combined, only tie-down and
racking/bracing are separated. {Vzp=permissible/allowable stress design,
Vzu=limit state ultimate strength design}

For low wind loads the long term gravity loading and material creep are the
critical design issue. There is thus a critical point at which wind load
kicks in and becomes an issue. Wind class N1 typically has the traditional
connections, used when winding was not considered.

Further the pressure coefficients (GCp) are not that accurately known for
real buildings and the complex roof shapes adopted, and the calculation of
qz is based on a series or guesses or judgments about the environment of the
building. Further more cannot choose any size of section want, have to use
what is commercially available. In general most builders appear happy to use
timber span tables for wind class N3(Vzu=50m/s), but otherwise like to
reduce tie-down and racking requirements and so otherwise get an engineer to
assess wind load direct to AS1170.2 and specify bracing and tie-down
requirements, rather than select to suit the wind class.

The wind classes could be considered inappropriately sized, for most N2 and
N3 sites are at the lower end of the range of speeds for the class. Even so,
the wind classes do provide benefits to manufacturers.


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 
******* ****** ****** ****** ******* ****** ****** ********