Academia.eduAcademia.edu

Figure 4 – uploaded by valter lori

See full PDFdownloadDownload figure
Technical features of tested materials. Table 4

Table 4 Technical features of tested materials. Table 4

Related Figures (21)

Maximum equivalent sound absorption areas for a room volume V = 200 m?. Table 1
Maximum equivalent sound absorption areas for a room volume V = 200 m?. Table 1
Fig. 1. Schematic plot of the floor plan of the different rooms in the ILT. Same scale for all labs.
Fig. 1. Schematic plot of the floor plan of the different rooms in the ILT. Same scale for all labs.
Geometric properties and mean reverberation time of the rooms. Table 2 3.2. Samples
Geometric properties and mean reverberation time of the rooms. Table 2 3.2. Samples
Fig. 2. Equivalent sound absorption area of damped and non-damped empty room: yellow Ajdampea; fed A1-354; dark blue, Amaxzoo; light blue Amax.
Fig. 2. Equivalent sound absorption area of damped and non-damped empty room: yellow Ajdampea; fed A1-354; dark blue, Amaxzoo; light blue Amax.
Description of the damping treatments applied by the different labs. Table 3
Description of the damping treatments applied by the different labs. Table 3
Fig. 3. Photographs of the three samples under test in three different laboratories participating in the ILT: a) 20 cm sample; b) 10 cm sample; c) 2 cm sample
Fig. 3. Photographs of the three samples under test in three different laboratories participating in the ILT: a) 20 cm sample; b) 10 cm sample; c) 2 cm sample
Fig. 4. Sound absorption coefficient, measured at normal incidence in an impedance tube, averaged over the four laboratories, in red the average. a) 20 cm mineral fibres; 10 cm mineral fibres; c) 2 cm polyester fibres.
Fig. 4. Sound absorption coefficient, measured at normal incidence in an impedance tube, averaged over the four laboratories, in red the average. a) 20 cm mineral fibres; 10 cm mineral fibres; c) 2 cm polyester fibres.
Fig. 5. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to ISO 354, using T20 values, in the nine selected Labs. Panels b), d), and f) give the resulting a, calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2sp.
Fig. 5. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to ISO 354, using T20 values, in the nine selected Labs. Panels b), d), and f) give the resulting a, calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2sp.
Fig. 7. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to Draft:2019, using T30 values, in the eight selected Labs that could be considered damped. Panels b), d), and f) give the resulting a calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2s,p.
Fig. 7. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to Draft:2019, using T30 values, in the eight selected Labs that could be considered damped. Panels b), d), and f) give the resulting a calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2s,p.
Fig. 9. Plot of y coefficients calculated according to ISO 354 draft, obtained as ratios between measured absorption coefficients for each laboratory and: a) the average of the’ damped laboratories; b) a theoretical value calculated assuming a finite size sample having the same flow resistivity as the sample under investigation. Eight laboratorie were considered, excluding Lab 1 due to its odd behaviour. At this point, in order to proceed with the application of the y correction it was necessary to set the required equivalent absorp- tion area (Areq) values to be used for the calculation. A quite obvi- ous option was represented by the average values coming from the ILT itself (after removing Lab 1 which proved to be a clear outlier, and Lab 9 whose room was not damped because of timing factors). This consideration takes into account the well-known fact that, due to the typology of the sample test in acoustic measurements, a ref- erence value of the measurand does not exist; therefore an esti- mated value has to be used [44,43]. Of course, the best evaluation of the measurand is the average value coming from an ILT (Fig. 8). In accordance with the prescriptions of Draft:2019, the reference absorber value was determined with the damped reverberation rooms (i.e. without considering Lab 9). The use of the average values (best-estimate values) from the ILT itself had the obvious consequence that the laboratories could not comply with STEP 1 for what concerns the use of the reference absorber (see paragraph 2.2 above), i.e. to modify the room in order to mea- sure the minimum value of the equivalent absorption area of the reference absorber (A;eq). Another unavoidable consequence result- ing from the adoption of a reference value calculated as stated above is that there will always be laboratories having an equiva- lent absorption area lower than A;eg, and thus a y value greater than one. Conversely, Draft:2019 inherently implies that laborato- ries complying with the standard should have y values below one. Consequently, a viable option may be to use a reference value cal- culated as a function of the flow resistivity of the selected sample. According to Mandel’s h test, Lab 1 shows 4 stragglers (at 160 Hz, 315 Hz, 2000 Hz and 2500 Hz) and 8 outliers (at 200 Hz
Fig. 9. Plot of y coefficients calculated according to ISO 354 draft, obtained as ratios between measured absorption coefficients for each laboratory and: a) the average of the’ damped laboratories; b) a theoretical value calculated assuming a finite size sample having the same flow resistivity as the sample under investigation. Eight laboratorie were considered, excluding Lab 1 due to its odd behaviour. At this point, in order to proceed with the application of the y correction it was necessary to set the required equivalent absorp- tion area (Areq) values to be used for the calculation. A quite obvi- ous option was represented by the average values coming from the ILT itself (after removing Lab 1 which proved to be a clear outlier, and Lab 9 whose room was not damped because of timing factors). This consideration takes into account the well-known fact that, due to the typology of the sample test in acoustic measurements, a ref- erence value of the measurand does not exist; therefore an esti- mated value has to be used [44,43]. Of course, the best evaluation of the measurand is the average value coming from an ILT (Fig. 8). In accordance with the prescriptions of Draft:2019, the reference absorber value was determined with the damped reverberation rooms (i.e. without considering Lab 9). The use of the average values (best-estimate values) from the ILT itself had the obvious consequence that the laboratories could not comply with STEP 1 for what concerns the use of the reference absorber (see paragraph 2.2 above), i.e. to modify the room in order to mea- sure the minimum value of the equivalent absorption area of the reference absorber (A;eq). Another unavoidable consequence result- ing from the adoption of a reference value calculated as stated above is that there will always be laboratories having an equiva- lent absorption area lower than A;eg, and thus a y value greater than one. Conversely, Draft:2019 inherently implies that laborato- ries complying with the standard should have y values below one. Consequently, a viable option may be to use a reference value cal- culated as a function of the flow resistivity of the selected sample. According to Mandel’s h test, Lab 1 shows 4 stragglers (at 160 Hz, 315 Hz, 2000 Hz and 2500 Hz) and 8 outliers (at 200 Hz
Fig. 8. Comparison of absorption coefficients to be used as “reference” values for the calculation of y. Measured diffuse-field values are based on the ILT average resulting from the 7 laboratories according to Draft:2019 method. Predicted diffuse field values for finite dimension samples are based on the use of TMM according to Allard and Atalla [46] formulation. Measured and predicted normal incidence values are also given for comparison.
Fig. 8. Comparison of absorption coefficients to be used as “reference” values for the calculation of y. Measured diffuse-field values are based on the ILT average resulting from the 7 laboratories according to Draft:2019 method. Predicted diffuse field values for finite dimension samples are based on the use of TMM according to Allard and Atalla [46] formulation. Measured and predicted normal incidence values are also given for comparison.
Fig. 10. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurement: according to ISO 354 draft, using T30 values, in the eight selected Labs, including y correction based on ILT average. Panels b), d), and f) give the resulting ay calculatec according to ISO 11654. Error bars in panels a), c), and e) correspond to 2s,p. ‘true’ diffuse field absorption coefficient obtained by accounting for the size effect both in the absorbed and in the incident power, as a function of the normal impedance calculated with the Trans- fer Matrix Method (TMM). ‘True’ is used in [46] in the sense that its value is always less than one; the results of this formulation, analytically calculated with the Bonfiglio et al. formulation [47] In fact, Fig. 7 shows that, with reference to the normal incidence absorption values, there is an almost perfect agreement between values calculated using the Delany-Bazley-Miki [45] model and those measured. Thus, it seems reasonable to start from flow resistivity to obtain the diffuse field absorption coefficient assum- ing a finite dimension sample. In [46] Allard and Atalla introduce a
Fig. 10. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurement: according to ISO 354 draft, using T30 values, in the eight selected Labs, including y correction based on ILT average. Panels b), d), and f) give the resulting ay calculatec according to ISO 11654. Error bars in panels a), c), and e) correspond to 2s,p. ‘true’ diffuse field absorption coefficient obtained by accounting for the size effect both in the absorbed and in the incident power, as a function of the normal impedance calculated with the Trans- fer Matrix Method (TMM). ‘True’ is used in [46] in the sense that its value is always less than one; the results of this formulation, analytically calculated with the Bonfiglio et al. formulation [47] In fact, Fig. 7 shows that, with reference to the normal incidence absorption values, there is an almost perfect agreement between values calculated using the Delany-Bazley-Miki [45] model and those measured. Thus, it seems reasonable to start from flow resistivity to obtain the diffuse field absorption coefficient assum- ing a finite dimension sample. In [46] Allard and Atalla introduce a
Fig. 11. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to ISO 354 draft, using T30 values, in the eight selected Labs that could be considered damped, including y correction based on predicted finite dimension diffuse field value. Panels b), d), and f) give the resulting o,, calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2sp.
Fig. 11. Plot of absorption coefficients for: a,b) mineral wool fibres, 20 cm thick; c,d) mineral wool fibres, 10 cm thick; e,f) polyester fibres, 2 cm thick. All measurements according to ISO 354 draft, using T30 values, in the eight selected Labs that could be considered damped, including y correction based on predicted finite dimension diffuse field value. Panels b), d), and f) give the resulting o,, calculated according to ISO 11654. Error bars in panels a), c), and e) correspond to 2sp.
Fig. 12. Summary of results for 20 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (in black and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measurement configuration d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. When the 10 cm sample was used, again the high frequency fluctuations were greatly attenuated and those in the low frequency range significantly reduced. However, inter-laboratory differences were consistent with those appearing when y j.r was used (Fig. 11c,d). Considering ow, results varied between 0.95 and 1.0, with the lowest values appearing in some of the Labs where y produced the lowest values in the medium frequency range. Particularly, it is worth to notice that Lab 3, which always showed the highest values, with the y correction, returned a lower As expected, the use of the y factor calculated from the ILT aver- age works well when applied to the 20 cm sample (Fig. 10a,b). In fact, with the exception of a few intra-laboratory fluctuations, mostly located at the lowest frequency bands, most of the varia- tions observed before (Fig. 8) completely disappeared. However, the supposed usefulness of the y factor should appear when applied to different samples. Application to the 10 cm sample (Fig. 10c,d) showed a significant reduction of the dispersion in the high frequency range (where absorption coefficients are very similar to those of the thicker sample), while some odd behaviours were observed in the low frequency range, although to a very lim- ited extent. In terms of oy all the labs returned the same value, apparently confirming the usefulness of the y correction. However, some very odd behaviours appeared with the polyester fiber mat (Fig. 10e,f), because the Labs that originally returned the highest values of absorption coefficients were strongly affected by the application of the y correction which (being lower than one) caused a change in the shape of the absorption curve and a reduc- tion of ow from 0.45 to 0.40. Conversely, as an obvious conse- quence of having chosen the reference value as the average
Fig. 12. Summary of results for 20 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (in black and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measurement configuration d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. When the 10 cm sample was used, again the high frequency fluctuations were greatly attenuated and those in the low frequency range significantly reduced. However, inter-laboratory differences were consistent with those appearing when y j.r was used (Fig. 11c,d). Considering ow, results varied between 0.95 and 1.0, with the lowest values appearing in some of the Labs where y produced the lowest values in the medium frequency range. Particularly, it is worth to notice that Lab 3, which always showed the highest values, with the y correction, returned a lower As expected, the use of the y factor calculated from the ILT aver- age works well when applied to the 20 cm sample (Fig. 10a,b). In fact, with the exception of a few intra-laboratory fluctuations, mostly located at the lowest frequency bands, most of the varia- tions observed before (Fig. 8) completely disappeared. However, the supposed usefulness of the y factor should appear when applied to different samples. Application to the 10 cm sample (Fig. 10c,d) showed a significant reduction of the dispersion in the high frequency range (where absorption coefficients are very similar to those of the thicker sample), while some odd behaviours were observed in the low frequency range, although to a very lim- ited extent. In terms of oy all the labs returned the same value, apparently confirming the usefulness of the y correction. However, some very odd behaviours appeared with the polyester fiber mat (Fig. 10e,f), because the Labs that originally returned the highest values of absorption coefficients were strongly affected by the application of the y correction which (being lower than one) caused a change in the shape of the absorption curve and a reduc- tion of ow from 0.45 to 0.40. Conversely, as an obvious conse- quence of having chosen the reference value as the average
Fig. 13. Summary of results for the 10 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (it black) and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measuremen configuration; d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. In order to provide a quantitative estimate of the effect of the application of different standards (and their variations), sound absorption values (frequency based and single-number) and uncer- tainties were compared and discussed. For comparison purposes, With reference to the 20 cm sample (Fig. 12 and Table A.1) it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in a slight increase in absorption coefficients and, more importantly, in a substantial increase in the repeatability standard deviation balanced by a reduction of the reproducibility standard deviation that, at fre- quencies above 400 Hz, was at its lowest without the application of the y correction. The use of Draft measurement configuration without y correction returned slightly lower sound absorption coefficients in the medium frequency range and a mixed behaviour in the lowest bands, where the “damped” configuration returned the highest absorption coefficients, while the non-damped rooms returned values comparable to those obtained with ISO 354/T20. In terms of repeatability standard deviation the adoption of the Draft configuration yielded a general improvement, while in terms of reproducibility standard deviation values, results were substan- tially comparable to those obtained with ISO 354/T20 above 400 Hz, but in the lowest bands the use of damping proved effective in reducing the differences. Finally, the application of y correction had obvious effects on the absorption coefficients, which coincided with the “reference” values adopted in the two cases; it also had a small positive effect on the repeatability stan- dard deviations (particularly when using y:ym), and, as expected, it caused a dramatic reduction of reproducibility standard deviations.
Fig. 13. Summary of results for the 10 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (it black) and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measuremen configuration; d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. In order to provide a quantitative estimate of the effect of the application of different standards (and their variations), sound absorption values (frequency based and single-number) and uncer- tainties were compared and discussed. For comparison purposes, With reference to the 20 cm sample (Fig. 12 and Table A.1) it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in a slight increase in absorption coefficients and, more importantly, in a substantial increase in the repeatability standard deviation balanced by a reduction of the reproducibility standard deviation that, at fre- quencies above 400 Hz, was at its lowest without the application of the y correction. The use of Draft measurement configuration without y correction returned slightly lower sound absorption coefficients in the medium frequency range and a mixed behaviour in the lowest bands, where the “damped” configuration returned the highest absorption coefficients, while the non-damped rooms returned values comparable to those obtained with ISO 354/T20. In terms of repeatability standard deviation the adoption of the Draft configuration yielded a general improvement, while in terms of reproducibility standard deviation values, results were substan- tially comparable to those obtained with ISO 354/T20 above 400 Hz, but in the lowest bands the use of damping proved effective in reducing the differences. Finally, the application of y correction had obvious effects on the absorption coefficients, which coincided with the “reference” values adopted in the two cases; it also had a small positive effect on the repeatability stan- dard deviations (particularly when using y:ym), and, as expected, it caused a dramatic reduction of reproducibility standard deviations.
Fig. 14. Summary of results for 2 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (in black and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measurement configuration d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. With reference to the 10 cm sample (Fig. 13 and Table A.2) it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in a notable decrease in absorption coefficients below 500 Hz (which is some cases also affected the single number value). In terms of repeatability stan- dard deviation, the use of T30 returned slightly better values in the low frequencies, while they were substantially the same in the highest bands. Finally, the use of T30 instead of T20 resulted in a very small increase of reproducibility standard deviation, being a little bit more evident in the highest frequency bands. Use of Draft measurement configuration without y correction returned slightly higher sound absorption coefficients in the med- ium frequency range and a mixed behaviour in the lowest bands, where the “damped” configuration returned again the highest absorption coefficients, while the non-damped rooms returned values comparable to ISO 354/T20. In terms of repeatability stan- dard deviations, the adoption of Draft configuration yielded again a general improvement, while in terms of reproducibility standard deviations, results were substantially comparable to ISO 354/T20 above 400 Hz, but in the lowest bands the use of damping proved effective in reducing the differences. Finally, the application of y correction caused a notable decrease of absorption coefficients when Yrmm Was used, with a reduction of o&w in half of the Labs. The effect on repeatability standard deviation was barely notice- able, with more evident improvement only at the lowest frequen- cies when Yrmm was used. With reference to reproducibility standard deviation, the comparison offered the first important check of the usefulness of y correction, showing that a significant With reference to the 2 cm sample (Fig. 14 and Table A.3), it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in barely noticeable differ- ences in terms of sound absorption (although aw changed in some of the Labs), while, in general, using T30 resulted in a smaller uncertainty. The use of Draft measurement configuration without y correction returned negligible variations in absorption coeffi- cients, with just one Lab recording a lower ow, with substantially similar results between damped and non-damped configurations. Repeatability standard deviation also decreased with the only exception of the 125 Hz and 160 Hz bands where an odd peak was observed. The same odd behaviour at the same frequencies appeared in the reproducibility standard deviation while in the other bands (except for 1250 Hz, where another odd peak is evi- dent) no noticeable variation could be detected in comparison with results derived from the current ISO 354 configuration (particu- larly if T30 was used). Starting from 2500 Hz, an improvement of inter-laboratory behaviour is noticeable, being the absorption coef- ficient values higher than and closer to 1, where the y correction had proved to be effective. Finally, as anticipated, the application of y correction caused some changes that need to be carefully dis- cussed. First, in terms of absorption coefficient, the use of Yur resulted in some “inversions” in the assignment of the mw values because some of the Labs that previously yielded a 0.45 value
Fig. 14. Summary of results for 2 cm sample: a) ILT average (8 Labs) of sound absorption as a function of frequency and measurement configuration; b) ILT average (in black and individual Labs ow resulting from different measurement configurations; c) repeatability standard deviation as a function of frequency and measurement configuration d) 95% probability (reproducibility standard deviation with coverage factor k = 2) as a function of frequency and measurement configuration. With reference to the 10 cm sample (Fig. 13 and Table A.2) it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in a notable decrease in absorption coefficients below 500 Hz (which is some cases also affected the single number value). In terms of repeatability stan- dard deviation, the use of T30 returned slightly better values in the low frequencies, while they were substantially the same in the highest bands. Finally, the use of T30 instead of T20 resulted in a very small increase of reproducibility standard deviation, being a little bit more evident in the highest frequency bands. Use of Draft measurement configuration without y correction returned slightly higher sound absorption coefficients in the med- ium frequency range and a mixed behaviour in the lowest bands, where the “damped” configuration returned again the highest absorption coefficients, while the non-damped rooms returned values comparable to ISO 354/T20. In terms of repeatability stan- dard deviations, the adoption of Draft configuration yielded again a general improvement, while in terms of reproducibility standard deviations, results were substantially comparable to ISO 354/T20 above 400 Hz, but in the lowest bands the use of damping proved effective in reducing the differences. Finally, the application of y correction caused a notable decrease of absorption coefficients when Yrmm Was used, with a reduction of o&w in half of the Labs. The effect on repeatability standard deviation was barely notice- able, with more evident improvement only at the lowest frequen- cies when Yrmm was used. With reference to reproducibility standard deviation, the comparison offered the first important check of the usefulness of y correction, showing that a significant With reference to the 2 cm sample (Fig. 14 and Table A.3), it can be observed that the use of T30 instead of T20 under current ISO 354 measurement conditions resulted in barely noticeable differ- ences in terms of sound absorption (although aw changed in some of the Labs), while, in general, using T30 resulted in a smaller uncertainty. The use of Draft measurement configuration without y correction returned negligible variations in absorption coeffi- cients, with just one Lab recording a lower ow, with substantially similar results between damped and non-damped configurations. Repeatability standard deviation also decreased with the only exception of the 125 Hz and 160 Hz bands where an odd peak was observed. The same odd behaviour at the same frequencies appeared in the reproducibility standard deviation while in the other bands (except for 1250 Hz, where another odd peak is evi- dent) no noticeable variation could be detected in comparison with results derived from the current ISO 354 configuration (particu- larly if T30 was used). Starting from 2500 Hz, an improvement of inter-laboratory behaviour is noticeable, being the absorption coef- ficient values higher than and closer to 1, where the y correction had proved to be effective. Finally, as anticipated, the application of y correction caused some changes that need to be carefully dis- cussed. First, in terms of absorption coefficient, the use of Yur resulted in some “inversions” in the assignment of the mw values because some of the Labs that previously yielded a 0.45 value
Fig. 15. Summary of results as a function of frequency (for all repetitions) and as a function of different calculation methods in Lab 3: a) 20 cm sample; b) 10 cm sample; c) 2 cm sample.
Fig. 15. Summary of results as a function of frequency (for all repetitions) and as a function of different calculation methods in Lab 3: a) 20 cm sample; b) 10 cm sample; c) 2 cm sample.
Summary of results for 20 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sr - referred to ISO 354:2003 in force expressed a: a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM. Table A.1
Summary of results for 20 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sr - referred to ISO 354:2003 in force expressed a: a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM. Table A.1
Summary of results for 10 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sp - referred to ISO 354:2003 in force expressed a: a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM.
Summary of results for 10 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sp - referred to ISO 354:2003 in force expressed a: a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM.
Summary of results for 2 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sp - referred to ISO 354:2003 in force expressed as a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM.
Summary of results for 2 cm sample - average (avg), repeatability standard deviation s, and reproducibility standard deviation sp - referred to ISO 354:2003 in force expressed as a function of T20 reverberation times, and to Draft:2019 including the effect of room damping and the use of different y corrections, ILT and TMM.
Related topics:
Mechanical EngineeringEnvironmental ScienceAcousticsArchitectureClassical PhysicsSound absorptionApplied AcousticsMeasurement Uncertainty
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved