Wind Speeds and Drifts
Hemal Modi
Structural engineer who works hard to translate your concept into something beyond a simple brick-and-mortar reality
Wind Speed
As per IS 875 (Part 3):2015, important structures are designed for 100-yr MRI as compared to 50-yr MRI for normal structures with speeds that are about 8 percent higher based on the k1 that is similar in approach to ASCE 7-05. The k1 factor will effectively increase speeds by roughly 30 percent for 500-yr MRI & approximately 40 percent for 1000-yr MRI. However, the recently introduced cyclonic region factor k4 is apparently too conservative as it increases wind speeds by additional 15 percent for industrial and 30 percent for post disaster emergency structures.
It is worth noting that ASCE 7-05 "service level" 3-sec peak gust wind speeds at 33 feet (10 m) above ground in Exposure C conditions are used for design with a 50-year mean recurrence interval (MRI), so the annual probability of exceedance = 0.02 for general buildings in occupancy category II. The code used importance factor (I) to extend the MRI on certain buildings (i.e., using I=1.15 for occupancy category III & IV to adjust the design-level wind velocity to MRI of 100 years or I=0.87 (0.77 for Vasd > 100) for occupancy category I that adjusts the design-level wind velocity to MRI of 25 years).
ASCE 7-05 Commentary C6.5.5 Importance Factor (p. 286) discusses this further, and Table C6-7 (p. 318) gives conversion factors for different MRIs, as shown in the snapshot above. The non-hurricane wind speed multiplier was based on Fc = 0.36 + 0.1 In (12T), with T is the MRI in years.
However, the wind speed maps are now “ultimate” and were revised to more closely align with MRIs on seismic maps for consistency in load combinations, etc. since ASCE 7-10. The load factors used in the load combinations have been revised accordingly from 1 ASD / 1.6 LRFD for ASCE 7-05 to 0.6 ASD / 1 LRFD ASCE 7-10 and wind speeds were expected to be roughly increased by sqrt(1.6) or roughly 26 percent higher to maintain similar levels of wind pressures. And wind speed maps are now provided for different Risk Categories, with Category I buildings now at a 300-year MRI (15% probability of exceedance in 50 years), Category II at a 700-year MRI (7% probability of exceedance in 50 years) , Category III at a 1,700-year MRI (3% probability of exceedance in 50 years) and Category IV buildings recently updated from 1,700-year MRI in ASCE 7-10 to 3,000-year MRI in ASCE 7-16. The importance factor (I) has been eliminated because the maps themselves now consider the extended MRI for higher Risk Categories. The wind maps for various MRI were based on the 50-yr MRI wind maps using the multiplier Fra = 0.45 + 0.085 ln(12T), where T is the MRI in years and refined further based on the additional weather data with special emphasis on thunderstorm and non-thunderstorm data for non-hurricane level winds and apparently further refinements are expected for tornadoes in ASCE 7-22.
It is worth noting that the ratio of wind speeds for post disaster structures (Risk category III or IV) is 6 to 11 percent higher than normal structures designed under risk category II in ASCE 7 and achieved by k1 factor in IS 875 as shown in the first image.
The definition of a hurricane-prone region is the region where the basic wind speed for Risk Category II is greater than 115 mph or 51.4 m/s at ultimate per ASCE 7-10 and 90 mph or 40.2 m/s at service level per ASCE 7-05.
Wind Drift Limit
The ASCE 7-16 standard still does not suggest an allowable drift limit for wind design as it does with a seismic design but, according to the non-mandatory Appendix CC (Serviceability Considerations) of ASCE 7-16, common usage for building design is on the order of 1/600 to 1/400 of the building or story height without more details. Typical wind drift limits in common usage vary from H/100 to H/600 for total building drift and h/200 to h/600 for interstory drift, depending on building type and the type of cladding or partition materials used. The most widely used values are H (or h)/400 to H (or h)/500 (since 1988). An absolute limit on interstory drift is sometimes imposed by designers in light of evidence that damage to nonstructural partitions, cladding, and glazing may occur if the interstory drift exceeds about 0.4 inches (10 mm). Also note that the Appendix C Commentary presents maps for return periods of 10, 25, 50, and 100 years (Figs. CC.2-1 through CC.2-4 of ASCE 7-16). However, the decision of which map to use is not explicitly stated and is left to the discretion of the design engineer. MRI of 10 and 50 years is recommended but, under certain circumstances, the design engineer can use a higher MRI wind speed in consultation with the client. The height of the structure, type of cladding materials, and type of cladding detailing are among the most important reasons that may encourage using wind speed maps with high return periods.
ASCE 7-05 Section CC.1.2 discusses drift due to wind effects and states “Use of the factored wind load in checking serviceability is excessively conservative.” For wind serviceability checks, which again are short-term loading, the Commentary recommends a load combination of D + 0.5L + 0.7W, and this has an annual probability of exceedance of 0.05. Interestingly, this would be close to a 10-year MRI and not 20-yr MRI as stated. The ASCE 7-05 service level 50-year design wind speed 90 mph, has a corresponding 10-year wind of 76 mph (=90 mph * 0.84 factor per Table C6-7). The velocity pressure is a function of wind speed squared (Eq. 6-15), the 10-year velocity pressure is almost 0.7 times the 50-year velocity pressure (76^2/90^2=0.713). The AISC design guide 3 recommends 75 percent of 50-year wind pressure as a reasonable (plus or minus 5 percent) approximation of the ten-year wind pressures compared to 70 percent recommended by ASCE 7-05 and AISC guide also recommends 0.006 (1/167) for ultimate behavior for drywall on studs.
ASCE 7-10 Commentary CC.1.2 says that the 700-year or 1,700-year MRI winds are “excessively conservative” for serviceability considerations, and it recommends a serviceability load combination of D + 0.5L + Wa. For the majority of the country, Figure CC-1 shows the 10-year MRI wind speed as 76 mph, Figure CC-3 shows the 50-year MRI wind speed as 90 mph…both of which are the same as ASCE 7-05. But is left to engineering judgement or performance requirements.
When using software to check H/400 (0.0025) interstory drift at a 10-year MRI, the ASCE 7-05 design-level wind speeds (50-yr) can be used for analysis and design, but the allowable drift could be checked to meet H/280 (=H/(400*0.7)). This allows you to check design at the 50-year winds and serviceability at the 10-year winds without having to add more Load Cases. Similarly, to check H/400 (0.0025) interstory drift ratio for 10-yr serviceability (76 mph wind), the ASCE 7-10 700-yr ultimate (115 mph wind) drifts could be taken up to H/175 or 0.0057 (H/(400 x 76^2/115^2)). Or you may find it easier to simply create a new service Wind Load Case with the 10-year wind speed specified.
For steel structures, AISC 360-05 introduced some checks on stability, including a Tau-Beta factor which reduces column stiffness to account for mill tolerances and residual stresses. However, drifts are typically checked using the unreduced stiffness with Tau-Beta turned off, while member strength checks are to be done with Tau-Beta turned on.
NOTE: The motivation for this article is the Webinar organized by CTBUH-India today (19th June 2020) titled "CTBUH : Revisiting India's Wind Speed" by Dr. Suresh Kumar of RWDI.
Structural design engineer at DSG Formerly Known as MZ&P W.L.L.
1 年that was very useful indeed. Thank you
Structural engineer who works hard to translate your concept into something beyond a simple brick-and-mortar reality
3 年Please watch the video recording of the RWDI webinar for additional details. https://www.youtube.com/watch?v=NFydFenFUwQ
Architect and Engineering consultant at Sarnath Consultants
3 年Very useful Information. Good effort Thanks.
Head - Structural Engineering at PNC Architects (Sobha Group, Dubai)
4 年Problem is when on any subject, if we are not thorough and doesn't have enough research recommendations to back up the problem, we end up taking onerous points from different sources and compiling into one document that becomes very conservative.
Technical Director, VHPT system Structural engineer
4 年(1) Agree, k4 factor is highly conservative. (2) Though IS 16700 now requires serviceability checks with 20 year return period wind after amendment 1, it is advisable to have separate wind speeds for SLS and ULS without importance factor (k4) like in ASCE 7-16. Including k4 factor even with 20 year return period wind is highly conservative for serviceability checks. (3) If wind loads are factored, similar property (stiffness) modifiers can be used for wind and seismic loads for ULS. (4) Considering along and across wind components simultaneously as per IS 875-3 is also highly conservative. This requirement is from australian/newzealand wind code of 1989 which removed in recent version of these codes. These codes along with other codes allows reduction upto 35% in along wind loads when they are combined with across wind loads.