Andrew,
Point taken. From my perspective the critical issues are proper determination of:
1) qz the reference pressure
2) internal pressure coefficient Cpi.
Ignoring the gust factor: assume equals 1 for simplicity. Then net pressure coefficient Cpn=Cpe-Cpi. Reference to Fig6-6 for Cp.
Indicates if get a hole in the roof of a house then the internal pressure coefficient could rise to -0.9, greater in magnitude than the -0.55 for partially enclosed, and G.Cpi=0, for open. Combined with windward wall Cp=+0.8, then have high probability of loosing wall if the building was designed for G.Cpi=-0.18.
In similar manner if the windward wall pressure removes doors or windows, then internal pressure coefficient will rise to +0.8 rather than the +0.55 or +0.18, and certainly not 0 for open. Adopting the appropriate value of Cpi for a given state of the building at the time of a given wind condition is the critical issue. For serviceability level winds speeds (or loads if thats where adjustment is made), then assuming the building is sealed may be appropriate. But as the wind speed increases, the chances of maintaining the seal during a hurricane or tornado diminishes: higher internal pressure coefficients should be adopted. Though assuming building is open at serviceability loads is not a big issue. (Here we define the serviceability conditions: not the code.)
Also the hurricane is likely to clear trees and other small obstructions which influence surface roughness, so adopting exposure C to push qz upwards may be more realistic. That is for serviceability adopt exposure B and for strength adopt exposure C. The issue is that to limit state design, there is no doubt the building does not survive the ultimate strength loading in a state which could be considered an habitable dwelling. The building only survives serviceability loads and remains habitable and serviceable. Using limit state design we can push ultimate loads upwards without necessarily increasing the traditional size of anything, we otherwise refine our definitions of serviceability states (more than 1 and each has different expectations of performance).
I agree strengthening the connections is not a big deal: costs are relative: builders who complain tend to be the ones who don't cost accurately and want to do the least amount of work.
My issue is that ASCE7-05 seems highly prescriptive, and using the C&C pressures is not conservative if other parameters are inappropriate.
Just curious but it seems each and every year, that a township in the USA is flattened by a tornado. Are those stick dwellings engineered to IBC/ASCE7, or engineered to the WFCM or seIected form IRC prescriptions? Presciptions which many list members seem to have suggested don't comply with the engineering codes.
It seems to me that the use of IRC should be dependent on the calculation of qz, and GCpi, if these are of appropriate magnitude then the IRC presciption can be used: if not then need to engineer or need additional prescriptions available. For those who are always looking up pressures (p), qz varies between sites and pressure coefficients (Cp) are constant for a given building shape. So at the simplest only need to calculate qz for a building/site combination to select a prescriptive solution. In such manner the failure rate of conventional construction can be reduced by increasing the restriction on its use, and adding more pre-engineered solutions.
My point is make sure qz, and Cpi are appropriate before arguing whether a failure is due to inappropriate use of MWFRS pressures and C&C pressures. For qz and Cpi are more likely to be the crtical issue responsible for the failure.
Regards
Conrad
