Academia.eduAcademia.edu

Figure 3 – uploaded by Aly Mousaad Aly

See full PDFdownloadDownload figure
Table 3 Response time comparisons As described in Table 3 the response time for Test 7a (7.5% door opening, located at the center of the windward wall) was 0.09s for the three wind AoAs considered. Test 8a (9% window opening located close to the right side of the windward wall) was 0.08s at 45° and 90° wind AoA while 0.07s for 75° wind AoA. This reveals that the response time of transient internal pressure overshooting is comparatively faster for a larger dominant opening as the opening area (A) governs the resonance frequency (Eq. (3)) for a given internal volume (V). Comparing the response time with and without internal volume correction, Test 8b for the building with no volume correction exhibited 4 times faster response than that of Test 8a (i.e., 0.03s vs 0.08s) irrespective of their similar opening size. This underlines the necessity for volume correction for transient response experiments in a BLWT with velocity ratios other than unity. The distribution of the peak internal pressure after sudden breach is shown in Fig. 12 for tests 7a and 8a. The data points starting from the time the door or window opened were taken by computing the peak values. The peak internal pressure is fairly uniform across the building cavity for all tests undertaken.

Table 3 Response time comparisons As described in Table 3 the response time for Test 7a (7.5% door opening, located at the center of the windward wall) was 0.09s for the three wind AoAs considered. Test 8a (9% window opening located close to the right side of the windward wall) was 0.08s at 45° and 90° wind AoA while 0.07s for 75° wind AoA. This reveals that the response time of transient internal pressure overshooting is comparatively faster for a larger dominant opening as the opening area (A) governs the resonance frequency (Eq. (3)) for a given internal volume (V). Comparing the response time with and without internal volume correction, Test 8b for the building with no volume correction exhibited 4 times faster response than that of Test 8a (i.e., 0.03s vs 0.08s) irrespective of their similar opening size. This underlines the necessity for volume correction for transient response experiments in a BLWT with velocity ratios other than unity. The distribution of the peak internal pressure after sudden breach is shown in Fig. 12 for tests 7a and 8a. The data points starting from the time the door or window opened were taken by computing the peak values. The peak internal pressure is fairly uniform across the building cavity for all tests undertaken.

Related Figures (16)

conservative in current codes (Sharma and Richards 2003, 2005). The present study focuses on the characterization of internal pressure due to sudden door or window breaching, effects of volume corrections, varying dominant opening areas and their location with respect to the incoming (upstream) wind direction. It also compares aerodynamic data obtained from experiments carrie¢ out at small-scale (boundary-layer wind tunnel) and at large-scale Wall of Wind (WoW). The WoW is a 6-fan large-scale testing facility primarily used to study wind and wind-driven rair effects on low-rise structures (Chowdhury et al. 2009, Bitsuamlak et al. 2009, Aly et al. 2012). Recently, a new 12-fan version of the WoW is constructed (Aly et al. 2011). For comparisor purposes, the WoW results presented in Tecle et al. (2012) are provided in the current study. The study also examines the peak internal pressure loading and compares with the wind load provisions of the ASCE 7-2010.
conservative in current codes (Sharma and Richards 2003, 2005). The present study focuses on the characterization of internal pressure due to sudden door or window breaching, effects of volume corrections, varying dominant opening areas and their location with respect to the incoming (upstream) wind direction. It also compares aerodynamic data obtained from experiments carrie¢ out at small-scale (boundary-layer wind tunnel) and at large-scale Wall of Wind (WoW). The WoW is a 6-fan large-scale testing facility primarily used to study wind and wind-driven rair effects on low-rise structures (Chowdhury et al. 2009, Bitsuamlak et al. 2009, Aly et al. 2012). Recently, a new 12-fan version of the WoW is constructed (Aly et al. 2011). For comparisor purposes, the WoW results presented in Tecle et al. (2012) are provided in the current study. The study also examines the peak internal pressure loading and compares with the wind load provisions of the ASCE 7-2010.
VV LINLw | AO Ulibv MMALOLYY Mi Gilby 42 LO Uli oVMitiviliv GQItnGA VE UWS UV ITTIIIICe Ad ieee =) te iO Uw characteristic length of the air slug, P, is the ambient pressure of air, P.(t) is the external pressure driving the inertial force, C,; and C,. are internal and external pressure coefficients respectively, V, is the effective volume of the cavity, L is the characteristic geometric length scale; @ is eave height wind speed, and subscripts m and f represent model and large-scale, respectively. For correct internal volume scaling and the appropriate measurement of the internal pressure fluctuations, the nominal volume obtained through length scaling needs to be magnified as given by Eq. (5). This could be done by providing an additional volume chamber underneath the wind tunnel turntable. In the present study, the model was prepared at a length scale of 1:9 and the test was conducted at a velocity scale of 1:4. Thus, the volume needed to be amplified by a factor of 16 as shown in Fig. 1(c). An airtight volume chamber box was attached to the base of the model building underneath the wind tunnel turntable, following the recommendation of Sharma et al. (2010). Sudden door or window opening test setup: The mechanism implemented to create door/window sudden failure during the wind tunnel testing utilized a digital servo motor system shown in Fig. 4. In order to obtain consistent aerodynamic data, it was necessary to operate the door without interfering with the building’s pressure taps. The best suited option for this application was the use of a remote controlled device. A radio control system normally used for model-aircraft was used as part of the electromechanical system and a radio transmitter was used to send out a signal of instructions (open/close the door), which were collected and interpreted by a radio receiver. The
VV LINLw | AO Ulibv MMALOLYY Mi Gilby 42 LO Uli oVMitiviliv GQItnGA VE UWS UV ITTIIIICe Ad ieee =) te iO Uw characteristic length of the air slug, P, is the ambient pressure of air, P.(t) is the external pressure driving the inertial force, C,; and C,. are internal and external pressure coefficients respectively, V, is the effective volume of the cavity, L is the characteristic geometric length scale; @ is eave height wind speed, and subscripts m and f represent model and large-scale, respectively. For correct internal volume scaling and the appropriate measurement of the internal pressure fluctuations, the nominal volume obtained through length scaling needs to be magnified as given by Eq. (5). This could be done by providing an additional volume chamber underneath the wind tunnel turntable. In the present study, the model was prepared at a length scale of 1:9 and the test was conducted at a velocity scale of 1:4. Thus, the volume needed to be amplified by a factor of 16 as shown in Fig. 1(c). An airtight volume chamber box was attached to the base of the model building underneath the wind tunnel turntable, following the recommendation of Sharma et al. (2010). Sudden door or window opening test setup: The mechanism implemented to create door/window sudden failure during the wind tunnel testing utilized a digital servo motor system shown in Fig. 4. In order to obtain consistent aerodynamic data, it was necessary to operate the door without interfering with the building’s pressure taps. The best suited option for this application was the use of a remote controlled device. A radio control system normally used for model-aircraft was used as part of the electromechanical system and a radio transmitter was used to send out a signal of instructions (open/close the door), which were collected and interpreted by a radio receiver. The
Fig. 3 BLWT flow characteristics (a) mean wind velocity profile, (b) turbulence intensity profile, (c) spectra and (d) integral length scale profile: fis the frequency, I, is the turbulence intensity, L, is the integral length scale, S, is the power spectra, o is the root mean square value of the along-wind velocity component (U) receiver then translated these instructions to a servo motor, which carried out the instructions. The servo motor provided high operational speed necessary to simulate rapid failure of the door/window and substantial torque to hold the door/window closed against the wind-tunnel flow prior to opening. The completed servo motor time response is 0.1 s per 85° door turning.
Fig. 3 BLWT flow characteristics (a) mean wind velocity profile, (b) turbulence intensity profile, (c) spectra and (d) integral length scale profile: fis the frequency, I, is the turbulence intensity, L, is the integral length scale, S, is the power spectra, o is the root mean square value of the along-wind velocity component (U) receiver then translated these instructions to a servo motor, which carried out the instructions. The servo motor provided high operational speed necessary to simulate rapid failure of the door/window and substantial torque to hold the door/window closed against the wind-tunnel flow prior to opening. The completed servo motor time response is 0.1 s per 85° door turning.
Table 1 Dominant openings and background leakage distribution in small-scale dimensions ee Se Table 2 describes the various test scenarios performed to investigate internal and external pressures. Test cases | to 6 investigate the effects of the dominant opening sizes, while test cases 7 and 8 focused on transient internal pressure response during sudden openings of a door and a window, respectively. For test cases 7 and 8, the ceiling partition was removed so that the whole building acted as a single room (i.e., attic and living rooms were combined). Test cases 8a and 8b represent the same opening with and without volume correction, respectively, to assess the necessity of internal volume correction for transient response wind tunnel studies.
Table 1 Dominant openings and background leakage distribution in small-scale dimensions ee Se Table 2 describes the various test scenarios performed to investigate internal and external pressures. Test cases | to 6 investigate the effects of the dominant opening sizes, while test cases 7 and 8 focused on transient internal pressure response during sudden openings of a door and a window, respectively. For test cases 7 and 8, the ceiling partition was removed so that the whole building acted as a single room (i.e., attic and living rooms were combined). Test cases 8a and 8b represent the same opening with and without volume correction, respectively, to assess the necessity of internal volume correction for transient response wind tunnel studies.
CU IO BP - 7 ers ne all of the wind AoA examined, the internal rms values are lower than the external rms values measured at the periphery of the respective dominant opening. This is believed to be due to the uniform nominal background leakage and the presence of the vents, which caused damping and hence reduction of the intensity of the internal pressure fluctuations. The trend of the peak C,; and Cre, however, illustrates good correlation between the internal pressure and the external pressure at the dominant opening. For example, the rms of the internal and external pressure coefficients for the two biggest dominant openings (i.e., D}=7.5% and W=9%), as shown in Fig. 6, did not indicate enough internal pressure excitation. In addition, spectral analysis of the internal pressure time history did not show any significant peak due to Helmholtz resonance (see Fig. 7).
CU IO BP - 7 ers ne all of the wind AoA examined, the internal rms values are lower than the external rms values measured at the periphery of the respective dominant opening. This is believed to be due to the uniform nominal background leakage and the presence of the vents, which caused damping and hence reduction of the intensity of the internal pressure fluctuations. The trend of the peak C,; and Cre, however, illustrates good correlation between the internal pressure and the external pressure at the dominant opening. For example, the rms of the internal and external pressure coefficients for the two biggest dominant openings (i.e., D}=7.5% and W=9%), as shown in Fig. 6, did not indicate enough internal pressure excitation. In addition, spectral analysis of the internal pressure time history did not show any significant peak due to Helmholtz resonance (see Fig. 7).
Table 2 Summary of test cases for low-rise building with gable and hip roof
Table 2 Summary of test cases for low-rise building with gable and hip roof
similar internal pressure variation regardless of the opening size (3.75% vs 9%). The distance between the upstream wall corner and the dominant opening has a significant impact on the r.m.s. values. For shorter distance between the windward wall corner and window opening, the r.m.s. fluctuation is considerably higher than the case when this distance is larger. For example, the r.m.s. for the 9% window (W2) at 30° wind AoA is 0.36; however, at 150° AoA the r.m.s. value is 0.1. A similar trend was observed for 20° and 160° AoAs. Fig. 8(c) shows the internal pressure coefficients due to window openings reaching a peak value at 70-75° wind AoA, whereas in the case of door openings, the maximum peak value occurs at 100° wind AoA. For wind AoA above 90°, the location of the window opening is far from the left corner (where upstream flow occurs) as compared to that of the door opening, which is located at the center of the wall. Consequently, the magnitude of the internal pressure due to both window openings is lower than that due to the door openings regardless of its smaller size. It can be
similar internal pressure variation regardless of the opening size (3.75% vs 9%). The distance between the upstream wall corner and the dominant opening has a significant impact on the r.m.s. values. For shorter distance between the windward wall corner and window opening, the r.m.s. fluctuation is considerably higher than the case when this distance is larger. For example, the r.m.s. for the 9% window (W2) at 30° wind AoA is 0.36; however, at 150° AoA the r.m.s. value is 0.1. A similar trend was observed for 20° and 160° AoAs. Fig. 8(c) shows the internal pressure coefficients due to window openings reaching a peak value at 70-75° wind AoA, whereas in the case of door openings, the maximum peak value occurs at 100° wind AoA. For wind AoA above 90°, the location of the window opening is far from the left corner (where upstream flow occurs) as compared to that of the door opening, which is located at the center of the wall. Consequently, the magnitude of the internal pressure due to both window openings is lower than that due to the door openings regardless of its smaller size. It can be
Fig. 6 Relationship of (a) rms and (b) maximum external pressure at dominant opening periphery vs internal pressure concluded that the dominant openings located outside of the center region of the wall exhibit larger internal pressure for an obtuse wind AoA. For example, for wind AoA between 0° and 70°, the window W; with 4% opening generates larger internal pressure than that due to doors D; (7.5%) and D> (5%). On the other hand, for wind AoA between 100° and 180°, the door openings D, and D, generate higher internal pressure compared to windows W; and W>. This attests to the significance of the dominant opening location.
Fig. 6 Relationship of (a) rms and (b) maximum external pressure at dominant opening periphery vs internal pressure concluded that the dominant openings located outside of the center region of the wall exhibit larger internal pressure for an obtuse wind AoA. For example, for wind AoA between 0° and 70°, the window W; with 4% opening generates larger internal pressure than that due to doors D; (7.5%) and D> (5%). On the other hand, for wind AoA between 100° and 180°, the door openings D, and D, generate higher internal pressure compared to windows W; and W>. This attests to the significance of the dominant opening location.
Fig. 8 Living room internal pressure (mean, rms, max and min) distribution due to various dominant Oneningec Fig. 7 Internal pressure spectra for sudden opening, (there is no significant Helmholtz resonance)
Fig. 8 Living room internal pressure (mean, rms, max and min) distribution due to various dominant Oneningec Fig. 7 Internal pressure spectra for sudden opening, (there is no significant Helmholtz resonance)
Fig.9 Comparison of internal pressure coefficient inside attic for an opened and closed vent (a) mean and (b) maximum This indicates that vent operation could be useful not only for ventilation optimization, but also for prevention of wind driven rain intrusion, and regulation of the internal pressure.
Fig.9 Comparison of internal pressure coefficient inside attic for an opened and closed vent (a) mean and (b) maximum This indicates that vent operation could be useful not only for ventilation optimization, but also for prevention of wind driven rain intrusion, and regulation of the internal pressure.
Fig. 10 Comparison of internal pressure inside living room for an opened and closed vent: (a) mean and (b) maximum
Fig. 10 Comparison of internal pressure inside living room for an opened and closed vent: (a) mean and (b) maximum
Fig. 11 Internal pressure coefficient time history for sudden breach of 7.5% door (a) and 9% window (b), and the corresponding external pressure coefficient on the periphery of 7.5% door (c), and 9% window (d) The experimental study on the transient response of wind induced internal pressure to sudden breaching was carried out with the 7.5% (Test 7a) and 9% (Test 8a) dominant door opening located at the center and window opening located off-center with volume correction, respectively. Additional tests were also performed (i.e., Test 8b) without internal volume correction to examine the sensitivity of the transient response to the internal volume correction. All these tests were conducted for 45°, 75° and 90° wind AoA.
Fig. 11 Internal pressure coefficient time history for sudden breach of 7.5% door (a) and 9% window (b), and the corresponding external pressure coefficient on the periphery of 7.5% door (c), and 9% window (d) The experimental study on the transient response of wind induced internal pressure to sudden breaching was carried out with the 7.5% (Test 7a) and 9% (Test 8a) dominant door opening located at the center and window opening located off-center with volume correction, respectively. Additional tests were also performed (i.e., Test 8b) without internal volume correction to examine the sensitivity of the transient response to the internal volume correction. All these tests were conducted for 45°, 75° and 90° wind AoA.
Fig. 12 Uniformity of internal pressure distribution at various wind AoA (45°, 75° and 90°)
Fig. 12 Uniformity of internal pressure distribution at various wind AoA (45°, 75° and 90°)
Amanuel S. Tecle, Girma T. Bitsuamlak and Aly Mousaad ALY
Amanuel S. Tecle, Girma T. Bitsuamlak and Aly Mousaad ALY
Related topics:
Wall of WindWind loadsInternal PressureHelmholtz ResonanceLow-Rise Buildings
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved