Home

DETERMINATION OF SOUND POWER LEVELS USING

1. 33 gt owl ho Direct NCINOG none aaa daq 33 9 2 DEO TIS sd e i e il aa iel la 37 3 4 Validity of Sound Intensity MeasurementData 42 od T Pressure lillenisiti NGSX nc ve 42 7 2 D ei eri a na yal e 44 3 4 3 Pressure Residual Intensity ndex 45 522 Dvmxm ieocapab l ly aa ne alis akal oa dese 49 De MEL O inci m a Ya delindi 50 S A utfoluetione a a a al a 50 3 5 2 Sound Pressure Calibration 51 3 5 3 Verifying Sound Intensity and Particle Velocity z 3 5 4 Measuring the Pressure Residual Intensity Index 54 DETERMINATION OF SOUND POVVER 57 4 1 Introduction to Sound PovveT 57 4 2 Comparison of different Sound Power Techniques 59 4 2 InttoduCtfOH ei ti el ea lt Bh a eta ett ades 59 od PICS qune qutt ROY 59 aki e ee ade aad d bi 60 4 3 Determination of Sound Power using Sound ntensity 61 4 3 1 Introduction and Definition l 4 3 2 Influence of Background Nolse 64 4 4 Determination of Sound Power according to ISO 9614 2 66 66 4 4 2 Measurement 5 LACE sc eee meli le
2. Intensity t gt Residual Intensity VV LLL Signals are not in phase Pressure Hesidual Intensity Intensity is detected Figure 3 14 Pressure Residual Intensity Index 5 Sound intensity measurement instruments are classified in the IEC 1043 standard according to the measurement accuracy achieved The accuracy specifications are based on frequency response filtering PRI index and environmental sensitivities There are two grades of accuracy designated as class 1 and class 2 The class 1 where the processor and calculating accuracy is better than 0 2 dB and the class 2 where the processor and calculating accuracy is better than 0 3 dB The same requirements apply to both classes the differences are only in the tolerances and in the pressure residual indices A class 2 requirements are stringent than those for 47 class 1 There is an additional class designated as 2X which has the same requirements and processor accuracy as class 2 except that real time performance is not required 8 The minimum PRI index requirements for class instruments with 12 mm nominal microphone separation is indicated in figure 3 15 e IEC 1043 1993 Electroacoustics Instruments for Measurement of Intensity with Pairs of Pressure Sensing Microphones Gives minimum requirements for PRI of probe signal conditioning and processor Also gives guidance on how to properly measure PRI of a system Minimum Pressure
3. JRipple ni Area 1 Area FA Practical filter and Practical filter and definition of LE definition of 3 dB Bandwidth i Noise Bandwidth I E i eres EE hof Frequency Frequency Figure 2 15 Bandpass Filter and Bandwidth 5 In acoustics there is a long tradition for using what is called constant percentage bandwidth CPB filters and they are best presented on a logarithmic frequency axis Constant percentage bandwidth means that each filter has a bandwidth which is of a certain percentage of the centre frequency 27 B 1 3 Octave 1 3 Octave f V2 x f 125 x f B 0 23 xf 23 Frequency Hz Figure 2 16 Third octave Filter 5 The third octave bandwidth filters are typically shown on a logarithmic display because they appears very compressed together at low frequency while they get wider and wider at high frequencies if shown on a linear display Third octave filters are characterized by the upper frequency limit is the third root of 2 multiplied by the low frequency limit and that turns out to be 1 26 which means that the upper frequency limit is 26 higher than the low frequency limit See figure 2 16 and table 2 2 Table 2 2 Third octave Passband 5 Third octave Frequency Hz Passband Hz 757 O 7 8910 11200 17 8 22 4K For third octave frequency spectra at I Hz the bandwidth is approximately 14 of an Hz At 10 Hz the bandwidth is 2 6 Hz and at 10 kHz t
4. Phase match and accuracy of intensity probe and analyzer together is indicated with the pressure residual intensity index PRI index A higher PRI index means that the instrument is better phase matched The PRI indexes is measured on the intensity analyzer by mounting the intensity probe into a sound intensity calibrator and expose both microphones to the same pink noise sound This method 46 corresponds to either having the intensity probe in a free field with a 90 angle of incidence or having it in a diffuse field situation When both microphones are exposed to identical pink noise sound the estimated intensity 1s expected to be equal to zero see figure 3 14 But real measurement probe and analyzers have phase mismatch and therefore some intensity is detected This detected intensity is called the residual intensity of the analyzer and is a false intensity produced by the phase difference between measurement channels If the entire measurement chain was perfectly phase match the residual intensity would be zero W m or 00 dB Since phase is not an integral part of a pressure measurement the pressure remains the same at both microphones The level difference between the indicated sound pressure level L and the indicated residual intensity level L in one octave or one third octave bands is then called the Pressure Residual Intensity index and is expressed as L L dB 5 19 Real probe with phase Ay f mismatch
5. The L is the energetic average of the noise L dul yib al R l l dnt py t i EPR l vil Figure 2 9 Equivalent Level Leq and measuring Leg with linear averaging 5 Exponential averaging is a running averaging which means it puts high emphasis upon what happens right now and gradually forget about what happened in the past see figure 2 10 While with a linear averaging time all data is equally weighted With an exponential averaging with fast time constant there will be continues random variations in the measurement data Figure 2 10 Leg for Transient Noise 5 22 2 2 3 Measuring Sound in Practice Measuring sound in practice is a question of microphone position The measurement position should in general be far enough away from reflecting surfaces and where it does not disturb the sound field The sound level meter should be mounted on a tripod and the operator should be standing both behind and aside the sound level meter A field check of the sound level meter with an acoustical calibrator or a pistonphone should be done before and after each measurement Measurement for standard procedures are defined in the ISO standard 1996 International Standards Organization Requirements for sound level meters are defined in the IEC standards International Electrotechnical Commission where they are divided into class 1 Precision and class 2 General purpose category For mos
6. 19 sound pressure levels Because the same sound is then only radiating into much smaller angle than if it was in a free field where it would be radiating in all agr ire Directivity Factor Index dB Free field directions On a flat plane At a junction of two planes Ata junction of three planes Figure 2 5 The Directivity Index sl 2 2 Measuring Sound 2 2 1 Basic Sound Level Parameters Sound level meters are integrated measurement equipment with standardized time weightings and also standardized measurement parameters They consist of a microphone a pre amplifier and detectors The detectors for the most basic sound level quantities are the RMS detector and the Peak detector RMS stands for Root Mean Square see figure 2 6 This is the most important detector that there exists because this one will indicate the amount of energy or power there is in the sound signal RMS 7 tat Root Mean Square Peak Crest factor AMS Figure 2 6 Basic Sound Level Parameters 5 20 Time constants or averaging times used with a RMS detector are typically fast slow or impulse Fast is a time constant of 125 ms and slow 1s a time constant of I s A fast detector will respond much faster to variations in the signal than the slow detector But after some time they would have reached the same level the RMS level A short time constant makes it easier to follow and track changes in the signal and is typical
7. G 2 SKA 46 Q9 gt 100057 00000 0900 ge ii M e o P 49 9 9 a 000009 00000000 i 202 b 0 2000 00000000009 00000 oo SL 20 209 200 Ze 2o N 99 fame AT Coo Sum Vertical plane O N c 9922 9 UU SH kHz VW 0 S 222 7 N gt o 2 K 7 o E SAZ 4 EN lf y 9630 Hz 4 NOVI 4 120 E et 9927 ML a 0997 NS N09 amp 98008209300998 990 N 9 900 99997 zi 10 820927 090907 7 4009900000965 7 40000000090 ogg El 2240005 Bo 90 S 99 3 e 9 dz N 4 EE x ELE eS 0 902 4 o 900000000099 es 5000000600099 7 00000000000 9 7490 2 2099 x CF wat X M 40 WT 000 AKNE S EAEN i 09977 o E 977 w 70 e 7 0087 b 009 a 208 g20928 9997 Figure 5 11 Typical directivity characteristics of Type 4224 7 The level control on the front panel of Type 4224 which enables the voltage across the loudspeaker to be controlled in steps of 10 dB from 40 dB to 0 dB was selected to 0 dB and the other level control were selected to 5 dB The level controls were in these same positions as indicated in figure 5 12 during the time when the loudspeake
8. k bk 10k 20 Frequency Hz Figure 2 18 The Auditory Field 5 29 Threshold of pain are typically up around 130 dB in the full frequency range But the limit of damage risk 1s at lower sound pressure levels Listening to a sound pressure level of approximately 100 dB in the frequency range of 2 3 kHz may damage the hearing system The frequency range for music is wider and has a more dynamic range in sound pressure levels than for speech More detailed information about the sensitivity of the human hearing system is defined by the Equal Loudness Contours for Pure Tones also called Phon curves see figure 2 19 For example listening to a 1 kHz pure tone with a sound pressure level of 40 dB will generate a loudness impression of 40 Phon But in order to get a loudness impression of 40 Phon at 20 Hz the sound pressure level have to be increased to 90 dB The Phon curve becomes more flat as sound pressure level increases Phon curves 120 sound 100 pressure B level L F dB re 20 uPa C weighting B weighting A weighting 20 Hz Figure 2 19 Equal Loudness Contours for Pure Tones 5 The 40 70 and 100 Phon curves are the background for the acoustic weightings called A weighting B weighting and C weighting The A weighting is an approximation to the 40 Phon curve by normalizing it to 0 dB at 1 kHz and then inverting it see figure 2 20 The idea with the different weighting curves was to use the A we
9. 360 or 27 Thus the angular frequency is 27f The wave number k is an alternative spatial descriptor 2m 21 A C C The wave number is used to non dimensionalize size and distance in acoustics 5 k 8 For characteristic dimension or distance d kd gt gt 1 the item is acoustically large or the distance is acoustically far kd lt lt 1 the item is acoustically small or the distance 1s acoustically close The characteristic dimension or distance kd is of course dependent upon the frequency see figure 2 12 The position might be acoustically in the near field at low frequencies but acoustically in the far field at high frequencies Likewise an item might be acoustically small at low frequencies but acoustically large at high frequencies 1 8 1 4 1 2 2k 5k 10k 20K 50k 100k Hz 27k d 71 2 Figure 2 12 Example of characteristic dimension d free field correction 5 24 2 3 2 The Microphone There exists many types of microphones and they can be divided into cheep microphones used for example in mobile telephones studio microphones for recording purposes and measuring microphones The measuring microphone consists of housing an insulator a diaphragm and a back plate behind the diaphragm see figure 2 13 Measuring microphones are pressure sensing condenser microphones and use a constant electrical charge for converting the diaphragm displacement into an analog electrical signal Measuring microphones detect what th
10. Residual Intensity Index IEC 1043 Class 1 12 mm spacer 25 50 100 200 400 800 1600 3150 6300 Figure 3 15 PRI index 6 10 and the IEC 1043 standard 5 The error due to phase mismatch can be both positive and negative If for example the sound fields phase change between the microphones is 1 and instrument is only matched with 1 accuracy then the estimation of phase change would be either 2 or 0 If the PRI index of the instrument and the P I index of the sound field is exactly the same then there is an error due to phase mismatch which can be 3 dB or dB see figure 3 16 This is because the error introduced due to phase mismatch is either adding or subtracting to the estimation If the intensity of the sound field 15 a positive value and the residual intensity of the instrument is a negative value then they subtract But if the intensity of the sound field is a positive value and the residual intensity of the instrument also is a positive value then they add together This is in difference to sound pressure measurement where background noise only adds to the measurement results and thus giving a too high estimate Error due to phase mismatch x reme For Fy 6 10 dB the error is lt 0 5 dB For Fy l 7 dB the error is lt 1 dB Figure 3 16 Error due to Phase Mismatch of the Instrument 5 48 An example of PI index PRI index Lac
11. cC wor WO O 0 50 63 80 100 125 160 200 250 1000 1250 1600 2000 2500 3150 4000 5000 6300 Frequency Hz Figure 5 16 Comparison of the Total Sound Power Levels of each measurement Likewise for the check of the validity on measurement data similar graphs were created The comparison of the validity checks for adequacy of the measurement equipment of each measurement is indicated in the graph in figure 5 17 e B amp K 4296 o B amp K 4296 44224 e B amp k 4224 e B amp K 4224 4296 Check for validity Ld Fpin gt 0 dB o o o LO LO o N LO LO I N lt 00 o N o LO e N N Frequency Hz Figure 5 17 Comparison of validity check s La Fpm gt 0 of each measurement 84 The comparison of the validity checks for the limit on negative partial power of each measurement is indicated in the graph in figure 5 18 B amp K 4296 B amp K 4296 4224 e B amp K 4224 o B amp K 4224 4296 Check for validity Fpln abs Fpin lt 3 dB R 5 0 O o0 o 50 63 80 100 125 160 200 250 315 400 0 800 1000 1250 1600 2000 2500 3150 4000 5000 6300 Frequency Hz Figure 5 18 Compariso
12. 22 96 22 45 22 21 22 07 21 52 21 71 22 55 21 22 21 73 23 42 21 94 21 71 21 72 21 34 20 87 20 33 20 33 20 33 20 33 43 2 78 891 6 42 25 15 18 06 19 43 15 96 15 96 15 45 15 21 15 07 14 52 14 71 15 55 14 22 14 73 16 42 14 94 14 71 14 72 14 34 13 87 13 33 13 33 13 33 13 33 Dynamic Capab 1 dB 3 43 22 5 91 2 42 2215 15 06 16 43 12 96 12 96 12 46 12 21 12 07 11 52 11 71 12 55 11 22 11 73 13 42 11 94 11 71 11 72 11 34 10 87 10 33 10 33 10 33 10 33 480 5 62 836 461 817 6 50 8 58 934 7 72 706 7 55 734 7 06 802 7 19 6 94 655 642 634 606 6 11 599 594 581 533 458 357 88 98 108 118 128 138 148 158 158 158 158 158 158 158 158 158 158 158 158 158 158 158 25 31 40 50 63 80 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000 6300 8000 10000 A L Figure 5 15 Example of the application displaying PI index and more 5 3 2 Results For a comparison of the total sound power levels from each of the four measurements a graph were created by Microsoft Excel with the measurement data generated by the software application created in this thesis The comparison is indicated in the graph in figure 5 16 83 B amp K 4296 109 1 dB B amp K 4296 4224 108 8 dB e B amp K 4224 107 7 dB e B amp K 4224 4296 107 2 dB o e o o co O 85 80 75 Sound Power Levels A vveighted dB re 1 pW O O O o o
13. 3 3 5 4 45 5 55 s Figure D 15 Multispectrum no 13 Figure D 16 Multispectrum no 14 dib LER GS LA bi S C GS eee yon SR SE RA 0 05 1 1 5 2 25 3 3 5 4 45 5 55 0 05 1 1 5 2 25 3 3 5 4 45 5 55 s Figure D 17 Multispectrum no 15 Figure D 18 Multispectrum no 16
14. 3 8 There might even be an incorrect direction of the estimated intensity Approximation Error L 1 101094 a Bias Error dB 1 dB limit 1 20k Frequency Hz Figure 3 8 Limitations at BIE n order to estimate the sound field with IdB accuracy at for example 500 Hz where the phase to be detected 1s 6 28 the system must have a phase match which is at least better than 1 26 approximately five times better At low frequency it is also a phase issue see figure 3 9 Here it turns out that the phase change over the spacer due to the sound field must be at least five times bigger than the phase mismatch between the two microphones in order for the sound intensity to be estimated within an accuracy of 1 dB The low frequency limitation of an analyzer is determined by the phase matching between the entire measurements chains of the system The low frequency limit for sound intensity estimation depends on the PI index of the sound field and the Dynamic Capability phase match and spacer of the analyzer The frequency 40 limit can be lowered by decreasing the PI index by measuring closer to the sound source or by adding absorption material into the room Dynamic Capability can be improved by increasing the size of the spacer and by improving phase matching e The Sound Intensity is Ores Puls Pla Changa Dev Seaver Banan proportional to the phase change over the spacer Li A 1
15. 46 22 53 22 25 21 7 22 85 22 30 21 63 20 93 19 96 19 96 19 96 19 96 Dynamic Capab 748 227 85 33 12 93 24 91 16 75 21 35 22 55 2240 18 84 17 26 16 58 17 24 12 53 17 91 17 88 16 46 15 53 15 25 14 77 15 85 15 30 14 63 13 93 12 05 12 85 12 56 12 56 Dynamic Capab 104 2 333 953 21 91 13 75 18 35 19 55 19 40 16 94 14 26 13 55 14 24 1463 14 81 1458 13 46 12 53 12 25 11 77 12 85 12 30 11 53 10 53 956 90 56 90 56 8 06 Pl index 3 51 651 11 62 5 43 11 35 10 43 12 81 1281 800 818 Sri 38 532 934 926 950 954 5 31 9286 902 954 58 73 7 58 7 53 2 556 5 00 4 17 95 s 105 115 125 135 145 155 155 155 155 155 165 165 1565 1565 155 155 158 158 158 15 5 dB 34 25 31 40 S0 63 80 100 125 160 200 250 315 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 5000 10000 A L Figure C 9 PI index Lac IEC minimum Values and StatusCodes 104 MS Noise 0046 5TP BK2260 Bz 7205 2 1 0 Total Extraneous Noise mi x Status Code x x x x x Extraneouz Noise 000 O O 1 35 0 56 0 20 33 1 18 1 02 0 00 0 00 O04 DOD O59 D OF62 O23 DOS 00 37 0 12 0 89 0 36 0 01 0 13 m h 0 00 O00 0 00 25 31 40 50 63 100 122 160 200 250 545 400 500 630 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A L Figure C 10 Spectra of Extraneous Noise Limit on Negative Partial Power 3 dB and StatusCodes 105 Appendix D Reverberation Time 120 Measurement Results Date of measurement April 2009 A
16. 93 47 91 69 2485 72 25 73 47 63 82 107 67 110 31 dB 130 120 110 100 90 20 70 60 50 30 20 20 31 40 S0 3 80 100 125 160 200 250 515 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 5000 10000 A L Figure B 3 A weighted Total Sound Power Spectrum with StatusCodes MS Noise 0045 5 2260 B27205 2 1 0 Total PI index nmi xil96 C T Status Code xDE xDE x x x PRI index 5 57 2878 15 51 12 42 32 15 2506 26 43 2256 2256 22 45 22 21 2207 21 52 21 71 22 55 21 22 21 73 23 42 2154 21 71 21 72 21 34 20 57 20 33 20 33 20 33 20 33 Dynamic Capab 748 43 278 821 542 2515 18 06 19 43 15 96 15 96 15 45 15 21 15 07 14 52 14 71 15 55 14 22 14 72 18 42 14 94 14 71 14 72 14 34 13 87 13 33 13 33 13 33 13 33 Dynamic 1046 2 43 22 5 81 2 42 22 15 15 06 16 42 12 06 12 96 12 45 12 21 12 07 11 52 11 71 12 55 11 22 11 73 13 42 11 94 11 71 11 72 11 24 10 37 10 33 10 33 10 33 10 23 Pl index 480 5 52 8 36 451 3 17 650 848 024 7 72 706 7 55 7 34 70G 802 7 19 694 655 8 42 6 4 60G 6 11 5 99 5 04 5 81 5 2 458 2357 989 98 105 118 125 135 145 155 155 155 155 158 165 165 168 1556 1556 155 155 158 158 15 8 dB 34 25 31 40 50 63 100 125 160 200 250 315 400 500 630 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A i Figure B 4 PI index Lac IEC minimum Values and StatusCodes 91 MS Noise
17. analyzers sensitivity to sound intensity and particle velocity is then verified against the known corresponding levels indicated in the supplied calibration chart The acoustical medium between the two chambers puts up a time delay which corresponds to a 50 mm spacer By setting up the analyzer for a microphone spacing of 50 mm it is possible to verify that the sound intensity and the particle velocity is calculated correctly in the analyzer 54 In figure 3 23 a simplified block diagram is indicating how sound intensity is estimated The particle velocity signal is given by Pa Pa PAP u 4 26 Figure 3 23 Block Diagram of Sound intensity Estimation 4 3 5 4 Measuring The Pressure Residual Intensity Index The purpose of measuring the pressure residual intensity index 1s to minimize the phase mismatch of the sound intensity analyzer The phase mismatch is largest at low frequencies and if the phase matching between the two measurement channels is poor the pressure residual intensity index will be low and the sound intensity at these frequencies will not be estimated accurately The upper chamber of the intensity coupler is used with a broad band sound source for measuring the pressure residual intensity index of the analyzer or in other words to determine the phase mismatch between the two measurement channels see figure 3 24 The broad band sound source produces pink noise creating a diffuse so
18. because essentially sound intensity estimation is really a phase measurement of the sound signal see figure 3 6 The limitations of a sound intensity analyzer are that the phase matching must be good in order to calculate low frequencies accurately If microphones are too far apart then high frequency resolution degrades and if they are too close together the signal to noise ratio degrades Lower Frequency Limit Upper Frequency Limit Phase change Phase change b x phase mismatch lt 60 1 dB accuracy 1 dB accuracy Frequency range for 1 dB Accuracy in a Free Field with 12 mm Spacer Phase Mismatch in Probe and Analyzer 12 mm _ _ 000 E et ee vin Frequency 20 40 80 125 5k T k Hz Figure 3 6 Frequency Range Depends on Phase 5 38 The finite different approximation for the gradient of the instantaneous sound pressure will generate a larger error at high frequencies see figure 3 7 Finite Difference Approximation Errors Thaoretical approximation errors at high oqlu HERB T MEM A EE l Low Frequency ip r r DER Accuracy within 1dB spacer Limit 50mm upto 1 25 kHz m ups 12mm upto 5 kHz rak 5 1 withiout cormechon for resonance s gi high frequency Figure 3 7 Theoretical Limitations at High Frequencies 5 It is obvious that if the separation between the two microphones were equal to half a wavelength then the two micr
19. doubled under free field condition the same amount of sound power or sound energy must penetrate a surface which is four times larger When the surface to penetrate is four times larger the sound pressure level is decreasing by 6 dB This is why sound pressure level measurements in free field condition highly depend upon the distance from the sound source Point source r L 2r L Figure 2 2 Three different types of sound sources 5 15 A line source can be a highway or a motorway If the distance to a line source is doubled the sound pressure level is decreasing by 3 dB The radiated noise from a line source has the shape of a half cylinder A doubled radius is increasing the cylinders surface by a factor of two thus decreasing the sound pressure level by 3 dB A plane source can be a tube with a piston at the end of it or a stethoscope If there is no absorption in the tube the sound pressure level will be the same at all positions in tube regardless of the length of the tube When combining sound pressure levels from multiple sound sources their sound pressure levels must be first converted back into sound pressure and then squared before they are added together and then converted into decibels as shown earlier in formula 1 This calculation is called power addition or RMS addition There exist also tables and charts for addition and subtraction of dB values If there are two uncorrelated sound sources and the sound pre
20. each of them with different acoustic directional characteristics were to be positioned next to each other The sound power levels was then to be determined as the sound source under test was running alone as well as when the other sound source was running parallel with the sound source under test Parallel to the thesis work computer software for measurement data analysis was to be created 10 1 3 Restrictions The determination of sound power levels using sound intensity is carried out according only to the standard ISO 9614 2 in a reverberant environment on two sound sources which have different directional characteristics Measurements are carried out as they run alone as well as when they run together Only the sound intensity scanning method ISO 9614 2 is used in the tests and is therefore studied more deeply in the theory part of this thesis This version of the computer software application created for analysis of measurement data stored in the analyzer by the sound intensity software BZ 7205 is only displaying the data and it does not perform any calculations by itself 11 2 BASIC KNOWLEDGE 2 1 Basic Concept of Sound 2 1 1 Sound and Noise Sound is something that surrounds us each and every day The most important thing about sound is that we use it for communication Communicating can mean many things it could be a truck driver just listen to the sound from the engine and the engine will communicate to him when
21. is moved forwards and backwards over the surface as if it was being painted The accuracy of the Sweeping introduces a random error and the accuracy of the surface area introduces a BIAS error on the measurement results ISO 9614 Part 2 amp 3 sweep ISO 9614 Part 1 point Figure 4 2 Scanning and Point Measurement Methods 5 The definition based upon intensity 1s that sound power is the surface integral of the sound intensity over a controlled surface or measurement surface which is completely inclosing the sound source of interest see figure 4 3 Figure 4 3 Definition of Sound Power from Sound Intensity 5 The sound power level Ly W from sound intensity measurement on a surface area S is calculated as S L L 10log E dB 5 30 0 Where L is the estimated intensity level dB S 15 the surface area m and So the reference surface area of 1 m2 63 Any measurement surface can be used for example hemisphere or a box surface or a conformal surface very close to the sound source see figure 4 4 Conformal Surface Hemisphere Figure 4 4 Measurement surfaces 5 The point method ISO 9614 1 gives precision engineering or survey grade of accuracy The disadvantage of this method is that it 1s difficult to follow There are four types of field indicators and precision sound power determination may require hundreds of measurement points to be measured and that would normally take muc
22. source and this gives a high amount of signal to noise ratio and therefore suitable where there is background noise Manual scanning will introduce a random error due to lack of precision in terms of distance to the sound source and accuracy of the scanning path A longer measurement time will minimize this random error A BIAS error is introduced due to the accuracy of the size of the surface because either the surface is too large or too small An error in the size of the surface by a factor of 2 will introduce an error of 3 dB If a surface of a sound source is close to some boundaries with reflecting surfaces there is no reason to measure this surface Intensity 1s zero on a hard surface because all the energy hitting that surface is reflected back from the surface and gives no contribution For a sound source standing in a corner and a box surface is defined around it it makes sense to only have three segments But according to the standard it must be divide into at least four segments Then one of the segments must be subdivided into two segments to meet the standard 4 4 3 Partial Power Repeatability Check Partial power repeatability check is a criterion that indicates the repeatability of two independent scans This criteria means that the difference between the sound powers of the two scans must be smaller than a certain number The partial power repeatability check detects error due to source or environmental variability s
23. source and a correction barometer see figure 3 19 The correction barometer determines the correction terms to the calibration levels of sound pressure and particle velocity due to changes in atmospheric pressure The sound intensity calibration level is not influenced by changes in atmospheric pressure A calibration chart is also needed that states the levels that should be detected during calibration 4 Figure 3 19 B amp K Sound Intensity Calibrator Type 3541 Components in Their Case An intensity coupler consists of housing and two chambers connected by a coupling element There is also a removable dummy microphone for closing unused chamber openings as indicated in figure 3 20 Pistonphone or sound source Upper chamber Dummy microphone Microphone A removable Coupling element Microphone B Lower chamber 871887 1e Figure 3 20 Simplified Cross Section of a Sound Intensity Coupler 4 51 3 5 2 Sound Pressure Calibration The sensitivity of each microphone is supplied by the manufacturer and has to be keyed in into the analyzer setup The gain adjustment factor for each microphone channel is obtained by a sound pressure calibration The coupler and the pistonphone are used for sound pressure level calibration of each measurement channel The pistonphone is mounted on the coupler and the microphones are positioned in the upper chamber of the coupler see figure 3 21 The pistonphone produ
24. 00 10000 A L Figure C 6 PI index Lac IEC minimum Values and StatusCodes 101 m 4 I 5 Moise 0047 5TP 2260 627205 2 10 Total Extraneous Noise Exrarecur Noire VK HZ O SS EE Status Code xDE xDE xDE F ORE ORE DE Extraneous Moise 998 1401 7 65 0 09 oO 02 DE DE xDE xDE DRE 0 08 0 12 O80 2 342 13 06 10 90 9 42 2 57 1 D8 0 69 0 89 1354 O46 60 0 32 0 50 146 1 36 3 38 916 16 50 15 09 25 531 40 50 63 100 125 160 200 250 315 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 8000 10000 A L Figure C 7 Spectra of Extraneous Noise Limit on Negative Partial Power 3 dB and StatusCodes 102 IMS Noise 0046 5 BK 2260 BZ7205 2 1 0 Total Sound Power Status Code x xO xO x x Power re 12 127 18 74 21 35 30 43 51 08 53 17 66 55 75 81 82 17 88 47 9404 96 15 98 84 OF 4 97 51 57 54 88 34 58 45 96 63 87 35 587 31 87 86 85 42 582 57 88 56 87 17 85 22 78 47 108 75 112 38 130 25 31 40 50 63 100 125 160 200 250 315 400 500 630 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A L Figure C 8 A weighted Total Sound Power Spectrum with StatusCodes M5 Noise 0046 5 2260 Bz 7205 2 1 D Total PT index LY 103 Status Lade x xO xO x x PRI index 927 13 33 15 93 31 91 23 75 28 25 28 55 2040 25 04 2426 23 88 2424 2453 2451 24 88 23
25. 0045 5 BK2260 BZ7205 2 1 0 Total Extraneous Noise F x m E r Z T Status Code KDE xDE xL LI x x Extraneous Moise 5404 470 184 122 0 11 O08 oFf4 063 0 77 b 016 O87 14 O25 z of0 t 0 04 004 0 11 O63 019 063 019 004 003 D dB io bea 25 31 40 350 63 80 100 122 160 200 250 315 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 5000 10000 A L Figure B 5 Spectra of Extraneous Noise Limit on Negative Partial Power 3 dB and StatusCodes 98 Appendix C Measurements Results for B amp K 4224 B amp K OmniPower Job name no 1 Job name no 2 Date of measurement Ambient temperatur Ambient pressure Sound intensity analyzer Analyzing software Sound source no 1 Sound source no 2 Amplifier 0047 S TP Sound Source Type 4224 0046 S TP Sound Source Type 4296 April 2009 20 C 1011 mbar B amp K Modular Precision Sound Analyzer Type 2260 Sound Intensity Software BZ7205 version 2 1 0 Br el amp Kjar Sound Source Type 4224 Level controls at 0 db and 5 dB see figure 5 12 B amp K OmniPower Sound Source Type 4296 Pink noise generatorer Minirator at 4dBu B amp K Power Amplifier Type 2716 C Level controls of A and B channel at max 99 Eme Status Code xDE xDE xDE R DRE DRE DE DE DE xDE xDE DRE Power re 1 1 7 2 0 YUB 22 46 48 02 60 55 FOS T8 74 83 23 85 60 77 35 8337 87 35 54 26 85 80 387 25 98 26 95 06 06
26. 40 88 4 80 48 95 19 93 25 01 51 83 94 74214 653 65 58 33 1 7 24 108 14 MS Noise 0047 5TP BE2260 BZ7205 210 Total Sound Power dB 130 120 110 100 au 20 70 60 50 30 20 25 31 40 50 63 100 125 160 200 250 312 400 500 630 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A L Figure C 5 A weighted Total Sound Power Spectrum with StatusCodes M5 Noise 0047 5 BK2260 2 7205 2 1 D Total Pl index Status Code xDE xDE xDE FR DRE DRE DE DE DE xDE xDE DRE PRI index 927 13 33 15 93 31 81 23 75 25 35 28 55 2040 25 04 2426 23 88 2424 2453 2451 24 88 23 46 22 53 22 25 21 7 22 85 2230 21 63 20 93 19 96 19 96 19 96 19 96 Dynamic 748 227 6 33 12 93 24 91 16 75 21 35 22 55 2240 18 84 17 26 16 88 17 24 12 53 17 91 17 88 16 46 15 53 15 25 14 77 15 85 15 30 14 63 13 93 12 05 12 56 12 56 12 56 Dynamic 104 2 333 953 21 91 13 75 18 25 19 55 19 40 16 94 14 26 13 55 14 24 1463 14 81 1458 13 46 12 53 12 25 11 77 12 85 12 30 11 53 10 93 9556 90 56 90 56 8 06 Bl irdex 14 81 128 22 15 70 7 74 10 21 8 02 28 75 11 65 13 03 2519 2204 19 47 12 76 11 59 10 36 8 75 11 29 919 235 58 751 Ger 936 9054 1124 17 52 2407 22 58 295 s 105 115 125 135 145 158 158 158 158 158 156 165 1565 158 155 158 158 158 158 158 dB 34 32 25 31 40 S0 63 50 100 125 160 200 250 515 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 50
27. 4296 is an omni directional sound source and is designed to simulate a pulsating sphere when it is radiating sound see figure 5 5 The sound source consists of a cluster of twelve loudspeakers radiating from the dodecahedral enclosure All loudspeakers are connected in a network to ensure in phase operation and that the impedance matches the Br el amp Kjar Power Amplifier Type 2716 C Figure 5 5 Briiel amp Kjzr OmniPower Sound Source Type 4296 with accessories When the sound source Type 4296 is connected via a bridging cable to both output channels of the amplifier Type 2716 C and the pink noise generator of B amp K 2260 Investigator is used the sound source can deliver a sound power level Ly of 122 dB see figure 5 6 3 nil ater H H Figure T dh Mu power levels L for sitos 3 77 The Power Amplifier Type 2716 C see figure 5 7 has two channels which may be used independently or jointly Signals enter electronically balanced inputs via XLR and jack connectors Total output power is 300 W and matches the requirements for driving the OmniPower Sound Source Type 4296 The amplifier uses passive cooling during operation and therefore do not need a cooling fan which makes it quiet during operation 1 Figure 5 7 Br el amp Kjzr Power Amplifier Type 2716 C During the determination of sound power the analog audio generator Minirator MRI was used to generate a pink noise signa
28. 8 20 24 76 34 74 51 74 60 63 69 36 76 19 63 00 88 79 04 56 95 95 98 42 98 12 88 21 97 94 08 64 98 93 97 19 98 25 OF BO 98 15 85 56 92 73 88 780 87 18 85 38 70 46 108 07 113 25 dB 130 120 110 100 25 31 40 50 63 80 100 125 160 200 250 315 400 500 630 500 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A L Figure A 3 A weighted Total Sound Power Spectrum with StatusCodes M5 Noise 0043 5 BK2260 BZ 7205 2 1 D Total Pl index Status Code x x x x x PRI index 8 00 035 1048 15 78 13 02 15 35 18 68 19 84 25 05 34 80 25 40 24 38 2400 23 28 22 42 27 57 23 23 21 07 2241 21 14 20 50 20 71 20 50 19 94 19 94 19 94 19 94 Dynamic Capab 798 1 99 235 348 8 78 662 8 35 11 68 12 84 18 55 27 89 19 49 17 38 17 09 16 28 15 42 20 57 16 23 14 07 15 41 14 14 13 90 13 71 13 50 12 34 12 34 17 94 12 04 Dynamic 1 dB 1 1 01 65 48 5 78 3 62 5 35 8 68 984 15 65 24 50 15 40 14 38 14 09 12 28 12 42 17 67 13 23 11 97 12 41 11 14 10 90 10 71 10 50 994 90 54 0 04 2 04 Pl index 3 87 414 7 39 354 7 20 558 7 78 7 70 664 7 34 7 66 805 6 61 695 MA 643 623 6 22 5 80 6537 6532 5 53 491 477 455 403 338 88 035 108 118 125 138 148 158 153 153 153 158 158 158 158 158 158 158 158 158 158 158 dB 34 32 25 31 40 0 63 80 100 125 160 200 250 515 400 500 630 600 1000 1250 1600 2000 2500 3150 4000 5000 6300 5000 10000 A L Figure A 4 PI index Lato Lac IEC minimum Values and
29. Fp of the sound field is compared with the dynamic capability Lg of the instrument for each segments group as L gt Fa dB 9 37 The dynamic capability index 4 1s obtained by measuring the pressure residual intensity index jo of the sound intensity instrument and then subtracting the BIAS error factor K of either 7 or 10 dB depending upon the grade of accuracy indicated in table 4 2 The defaults L4 data in the instrument are the minimum requirements for a class 1 sound intensity instrument 8 70 Table 4 2 BIAS error factor K 9 Bias a factor Grade of accuracy grade 3 1 Defined in ISO 12001 4 4 5 Limit on Negative Partial Power The negative partial power indicator F detects possible error due to extraneous sound sources or nearby reflecting surfaces and indicates if there 1s a substantial flux of sound power into the source volume through parts of the measurement surface The amount of extraneous noise is checked by comparing the number of positive and negative measurements for each frequency band and how much contribution the negative intensity gives Negative partial power indicator F 1s evaluated for all frequency bands of measurement in each segment group as F 20 dB 9 38 Where P is the partial sound power of segment i and is given by the formula 34 The limit on negative partial power is F 3 4 9 39 4 4 6 Flowchart and Corrective Actions Flowchart o
30. In a free field where there is a purely active sound field the Fk is equal to zero dB and the sound pressure and the corresponding particle velocity 1s in phase with each other when pointing the intensity probe towards the sound source In a purely diffuse sound field the F is very large and the pressure and the particle velocity are uncorrelated When pressure and particle velocity at the same point are uncorrelated then the intensity is equal to zero An extreme type of sound field where intensity is behaving quite different from pressure than in any other sound field is the standing wave A standing wave is characterized by that the pressure and the particle velocity is 90 out of phase with each other and their product becomes zero In a standing wave the intensity level is equal to zero because it is the mean level between the maximum sound pressure and the minimum sound pressure 3 4 2 BT product The is used for evaluating both random and systematic BIAS errors For random error on a sound pressure level measurement data there 1s a confidence level or a standard deviation in terms of accuracy of the correct level of 68 given in the chart in figure 3 13 as a BT product and extended as a function of Fp For BT product B is the bandwidth of the filter and T is the measurement time If for example the BT product is equal to 600 for sound pressure level measurement then the accuracy in terms of standard deviation i
31. StatusCodes MS Noise 0043 5TP BK2260 Bz 7205 2 1 0 Total Extraneous Noise 4 H l L xi S iz Status Code x x xD D D X X D Extraneouz Moise 0 01 m h 0 00 1 20 O 00 O00 h D DD D DD D DD 00i DOZ O25 m h 0 00 O00 DOD 0 00 h D DD dde 0 00 l O 0 00 00 DD dB 16 15 14 13 12 11 10 9 5 D 6 5 4 m 25 31 40 50 63 S 100 125 160 200 250 315 400 500 630 6500 1000 1250 1600 2000 2500 3150 4000 5000 5500 5000 10000 A L Figure A 5 Spectra of Extraneous Noise Limit on Negative Partial Power 3 dB and StatusCodes 94 Appendix B Measurements Results for B amp K 4224 Job name 0045 STP Date of measurement March 2009 Ambient temperatur 20 C Ambient pressure 1007 mbar Sound intensity analyzer B amp K Modular Precision Sound Analyzer Type 2260 Analyzing software Sound Intensity Software BZ7205 version 2 1 0 Sound source no Br el amp Kjar Sound Source Type 4224 Level controls at 0 db and 5 dB see figure 5 12 Sound source no 2 B amp K OmniPower Sound Source Type 4296 switched off but in position Figure B 2 Measurements 95 P BEZ760 EZ 7205 2 10 Total Sound Power MS Noise 0045 5T E Status Code xDE xDE xD D X X D Power re 15 12 8 18 2416 4612 60 14 69 58 783 84 83 94 84 14 66 78 9001 41 71 15 12 95 69 97 61 98 64 95 46 96 58 98 50 99 36 98 45
32. VAASAN AMMATTIKORKEAKOULU UNIVERSITY OF APPLIED SCIENCES DETERMINATION OF SOUND POWER LEVELS USING SOUND INTENSITY Michael S derback 2009 PREFACE This master s thesis is a part of my studies for a master s degree in construction engineering at Vaasan Ammattikorkeakoulu University of Applied Sciences The thesis served to provide myself a general introduction to acoustics and an in depth study of sound intensity for sound power determination As an additional result of the thesis a computer software application was created for measurement data analysis Michael S derback Miku gt M Z Ca Pa Vasa June 10 2009 VAASAN AMMATTIKORKEAKOULU UNIVERSITY OF APPLIED SCIENCES Rakentamisen koulutusohyelma ylempi AMK tutkinto ABSTRACT Author Michael S derback Title Determination of Sound Power Levels using Sound Intensity Year 2009 Language English Pages 89 4 Appendices Supervisor Tapani Hahtokari This thesis is a study in the determination of sound power levels of noise sources by using the sound intensity method The sound intensity method is suitable when the measurements are carried out in field conditions where the acoustical environment varies The sound intensity method for sound power applications is regarded as less sensitive to background noise than sound pressure measurements and therefore requires less special test facilities This thesis also deals with acoustics in general basic concepts of so
33. a vector quantity and the unit is W m Sound intensity is scaled into sound intensity level in dB when divided by the reference intensity 1 pVV m 1 107 The real part of sound intensity or the propagating part of the sound field is called active intensity The imaginary part of sound intensity or the non propagating part of the sound field is called reactive intensity 5 Speed of sound c Density of air Power WW W Intensity 1 Jisim W m Pressure p Pa N m Figure 3 1 Definition of Sound Intensity 5 The relationship between the three basic parameters of sound under free field condition is indicated in figure 3 1 where an omni directional sound source is emitting a certain amount of power W The intensity on the surface at the distance r is the radiated sound power W divided by the surface area 42 Or the intensity 7 on the surface at the distance r is the sound pressure p squared and divided by the impedance pc of the air The impedance of the air in which the sound wave propagates is the speed of sound c multiplied by the density of air p 32 The human hearing system is able to detect from which direction the sound is arriving because the sound is arriving at the two ears at a slightly different time Likewise an intensity probe consists of two microphones and thus enabling detection of both direction and level of sound An intensity probe has a cosine like directivity pattern of
34. ag and IEC minimum values in the same graph produced by the software created in this thesis work is indicated in figure 3 17 BESS B T205 Fo 0 T kal Fiili M 1133 H DM 1275 290 2055 204p 1154 1420 12008 2424 2481 124081 2410 1146 1143 1125 1177 22484 2220 2181 13 03 R ER ee BAAI 327 i 040 144 1825 20238 2255 2240 11 94 17 70 MT 1724 ee Mr 11 53 1424 147 PE ME 1400 1191 1198 12396 11304 dud Tebr 133 840 ELIMIN E M E M PMI Nr 1148 1424 162 Mr 1253 12256 1177 di 1230 1842 R OM Mr M 351 88 187 40 1135 Da FPE N28 BD EG 77 813 934 GAB 000 DON AH ucl Gi BT 765 75 BRIT 535 500 417 EA SE OE LE GA DA M idi di PEE CLE SA DA MA HB di SA BA HA MA H Ii Figure 3 17 Example of P index dp10 Lac IEC minimum values 49 3 4 4 Dynamic Capability The PRI index of the instrument minus a safety margin must be higher than the PI index of the sound field For the estimation of the sound intensity level to be within an accuracy of 0 5 dB there must be a safety margin or a difference of 10 dB or higher between the PRI index of the instrument and the PI index of the sound field Likewise to have an accuracy of dB the safety margin must be not less than 7 dB This safety margin is called K factor and described as BIAS error factor in the standard ISO 9614 The dynamic capability index is the difference between sound pressure level and sound intensity level w
35. al sound power level Note that all three accuracy criteria failed for L total but none for A total Thus the accuracy for the A total of B amp K 4224 is within an engineering grade of accuracy 0 5dB The difference between the two determinations of sound power levels A total 107 7 dB and 107 2 dB for B amp K 4224 is equal to what is indicated by the engineering grade of accuracy 0 5dB Common for both sound sources is that the A totals 1s decreasing in the presence of background noise It seems like that there 1s slightly more background noise going in to the measurement surface than going out thus the negative and the positive intensity 1s not completely canceling each other 86 The conclusion is that the sound intensity method for determination of sound power levels of noise sources can be used with a high level of accuracy even in a reverberant chamber with the presence of high background noise But background noise is degrading the level of accuracy in the determination of sound power levels for a sound source which 1s not fully omni directional 87 7 PROPOSALS FOR FUTURE WORK AND IMPROVEMENT Proposals for future work is to study and use the sound intensity technique for applications in building acoustics for example measurement of noise reduction index of building elements and noise source mapping of sound leakage detection in buildings and of building elements A proposal for future work is also to develop comput
36. alibrated Another advantage is that the sound pressure and the particle velocity are calculated at the same time and at the same location Mean Pressure Pra 1 p Pa Also referred to as Sound Intensity siz Pa Pa lpa p dt the Direct Method Figure 3 5 Estimation of Sound Intensity using Constant Percentage Bandwidth 5 When estimating sound intensity using a constant percentage bandwidth analyzer see figure 3 5 the microphone signals from the pre amplifiers are first converted from analogue to digital signals Then the output from the third octave filters are after summation squared and averaged thus giving the mean pressure 37 Sound intensity is calculated by taking the sum and the difference of the output from the third octave filters This difference is then integrated with respect to time The difference is a quantity which 1s proportional to the particle velocity and the sum is a quantity which is proportional to the pressure midway between the two microphones Then the difference and the sum are multiplied and averaged Finally the scaling factor of 1 2p4r generates the result where p is the density of the air and Ar is the separation between the two microphones This is called the direct method because both intensity and mean pressure can be calculated directly according to the formulas 3 3 2 Limitations The frequency range of a sound intensity analyzer depends upon phase matching
37. an obas na 68 4 4 3 Partial Power Repeatability Check 68 4 4 4 Adequacy of the Measurement Equipment 69 4 4 5 Limit on Negative Partial Povver 70 4 4 6 Flowchart and Corrective Acttons 70 MEASUREMENT JO Sut West Seti DESC 1 Oi uus a Gl Gan e a aiio 73 5 2 Description for Measurement and Test Equipment 74 5 2 1 Sound Intensity 74 5 2 2 Sound Sources and Accessorf1es 76 2 0 Resule arid nab na yl de aaa yl e ri 30 5 3 1 Application for Measurement Data Analysis 80 200 Pi ci o io b 82 6 COLOLU Um imi kille ll S EA ED EU 85 7 PROPOSALS FOR FUTURE WORK AND IMPROVEMENT 87 6 TLFELELEobvscso e 88 Appendix A Measurements Results for B amp K OmniPovver 90 Appendix B Measurements Results for B amp K 4224 94 Appendix C Measurements Results for B amp K 4224 B amp K OmniPower 98 Appendix D Reverberation Time T20 Measurement Results 105 1 INTRODUCTION 1 1 Technobothnia Research Centre The Technobothnia research centre is a joint project involving the ministry of education the city of Vaasa the University of Vaasa the Vaasa University of Applied Sciences and
38. at higher frequencies At a frequency where the diameter of the microphone is equal to the wavelength of the sound field overestimation might be as much as 10 dB 5 By drilling holes in the back plate the frequency characteristics can be changed at higher frequencies Microphones have different sensitivities depending upon the angle of incidence and depending upon of what type of sound field they are used in 1 Free field microphone 2 Pressure field microphone 3 Diffuse field or random incidence microphone Microphones with large diaphragm have high sensitivity and microphones with small diaphragm have lower sensitivity The smaller microphones are for measuring high sound pressure levels and large microphones are for measuring low sound pressure levels On the other hand the larger the microphone is the more it disturbs the sound field Large microphones are therefore used only up to a limited frequency range small microphones can be used up to very high frequency ranges 5 26 2 3 3 Frequency Analysis With frequency analysis of sound means the decomposing of the sound signal into its various frequencies as indicated in figure 2 14 Frequency Frequency Figure 2 14 Waveforms and frequencies 5 Typically the bandwidth of the filter 1s defined as being the upper frequency limit minus the lower frequency limit see figure 2 15 Bandwidth f f Centre Frequency f Ideal filter Frequency
39. called a reactive sound field and constitutes the imaginary part of a complex sound field 18 An example is a standing wave where the net flow of acoustic energy is zero see figure 2 4 Standing Wave 1 0 Pressure p OOOO 2 dd S206 58 Figure 2 4 Standing wave 11 There exists also a test facility where there are two rooms next to each other and they can both be reverberation chambers Or one of them can be an anechoic chamber the other one can be a reverberation chamber These are typically used for measurement of transmission lost of panels Pressure fields are small enclosures where the sound pressure is the same in terms of magnitude and phase everywhere in that sound field and they are in size small compared to the wavelength Typically used in small couplers With the directivity index means that the sound pressure level will increase if the sound source is positioned close to some boundaries compared to the sound pressure level of the same sound source in a free field situation see figure 2 5 The directivity index can be identified by positioning the sound source next to a wall where the index will be a factor of two corresponding to 3 dB In a junction of two planes there will be an increase of 6 dB and in a junction of three planes there will be an increase of 9 dB This means that the best place for a loudspeaker is in a corner because then it will produce the highest amount of
40. ces the same sound pressure with respect to both amplitude and phase at each microphone Each microphone channel is then gain calibrated against this known sound pressure level The pistonphone produces a pure tone typically at 250 Hz In the calibration setup of a sound intensity analyzer the user typically has to key in the calibration level the ambient temperature the ambient pressure and also the correction terms from a correction barometer The pressure from the pistonphone is depending upon the ambient pressure For the calibration level the sound pressure level given on the calibration chart of the calibrator is used Figure 3 21 Sound pressure calibration using a pistonphone and an intensity coupler The error in estimated intensity Lg gain due to error in the gain adjustment factor is i kas T S Qn E1010 E gain be 2 Where CA and Cp 15 the error in gain adjustment factors dB If only channel B is calibrated and channel A has an error in sound pressure level of for example 1 dB Then the error introduced on sound intensity measurement due to the error in the gain adjustment factor for channel A 1s 0 5 dB Generally this is not a critical error the phase match of the analyzer is much more critical 52 3 5 3 Verifying Sound Intensity and Particle Velocity Calibration of sound intensity determined from a two microphone pressure measurement requires knowledge of sensitivity and gain adjustment phase match and ef
41. d power level L is the logarithmic measure of the sound power passing through segment i of the measurement surface and is given by Ly 10log P P dB 9 35 Two different types of field indicators are calculated to indicate the quality of the sound power determination and there are three different types of criteria that have to be fulfilled in order to meet the standard If a criterion fails there are corresponding actions given in the standard on how to increase the grade of accuracy in the determination The sound power is calculated by multiplying the average intensity with the surface area of each segment and finally adding together the results of all segments Select Source Installation amp Operating Conditions select Initial Measurement Surface amp Make Initial Measurement Calculate Sound Power amp Indicators Make new measurement Accuracy Critera Take specified Satisfied action Final Result Figure 4 6 ISO 9614 2 Procedure 5 68 4 4 2 Measurement Surface A measurement surface that is completely inclosing the source under test is defined and for the engineering grade it must be divided into at least four segments The surface can be a conformal surface a box surface or a hemisphere Hemisphere is not very practical with the scanning technique but it may be possible with a robot system Conformal surface means that the measurement surface will be very close into the sound
42. d sources see figure 5 1 and 5 2 Wires were used to divide each of the five sides of the box into segments in three rows and three columns The box was positioned around the sound source under test with the horizontal acoustic center of the sound source in the horizontal center of the box The horizontal point midway between the acoustic centers of the sound sources were in the horizontal midpoint of the chamber Figure 5 Box around sound source no 1 Figure 5 2 Box around sound source no 2 The endpoint of each scan line was marked on the steel frame and on the wires in order to keep an exact and repeatable path of scanning Each segment was scanned twice as required in the standard ISO 9614 2 All calculations monitoring of field indicators and criterions were done by the software in the sound intensity analyzer The purpose of the post processing of measurement data was only to display the results The reverberation time was also measured 74 5 2 Description for Measurement and Test Equipment 5 2 1 Sound Intensity Analyzer Br el amp Kjar Modular Precision Sound Analyzer Type 2260 Investigator supports one and two channel applications such as building acoustics sound intensity measurements and FFT analysis with pure tone detection It is a programmable platform for a dual channel real time hand held analyzer With one of its installed software applications active it takes input from the acoustic front end and sends the
43. e A phase mismatch of the s 5 analyzinq system introduces an error to the calculated sound intensity The phase mismatch error is most severe at low frequencies Figure 3 9 Limitations at Low Frequencies 5 For example if a sine signal at 100 Hz which has a wavelength of 3 6 m is estimated with an intensity probe where the microphone separation is cm then the estimation should be a 1 difference of the sound field 360cm 360 Because there is 1 degree difference between the microphones if they are 1 cm apart Then of course the matching of the two channels must be better than 1 and it must be five times better for not having a significant error If the microphones were matched to 1 and the estimated sound field signal is 1 then the result of the estimation would be 2 or 0 depending upon if the phase difference of this instrument is subtracting or adding to the phase difference of the sound field If the estimation is 2 rather than 1 then it is an over estimation of 3 dB and if the estimation is 0 rather than 1 then it will be an under estimation which is infinitely high A change of the spacer size shifts the frequency range For example a 12mm spacers upper frequency limit is 5 kHz and with a phase match of the system of 0 17 the low frequency limit is 40 Hz Likewise a phase match of the system of 0 057 which is two times better then the frequency range is two times l
44. e human ear detects namely the pressure variations in the sound field The diaphragm and the back plate are isolated from each other There are a number of holes in the back plate and depending upon how many holes there are the microphone is to be used in a free field or in a diffuse field The number of holes is determining how much damping the microphones has A measuring microphone has to have a number of special features such as wide frequency range flat frequency response wide dynamic range low distortion robust with long term stability and simple design Diaphragm Backplate Housing y Insulator Figure 2 13 Cross sectional view of a classic microphone type 2 25 A typical sensitivity of a microphone is 50 mV Pa So when Pa is applied to the microphone it corresponds to 96 dB 5 Measuring microphones are not symmetrical in all directions and therefore not fully omni directional Also the mounting of the measuring microphones on a sound level meter will introduce a change in directional characteristics at higher frequencies When a microphone is inserted into a sound field it will due to its size disturb the sound field Typically there will be an increase of sound pressure in front of the microphone when the microphone is inserted into a free field and there is a plane sound field arriving to the microphone diaphragm Thus there will be an overestimation of the sound pressure in the sound field especially
45. e quiet machinery and equipment sound power is the best and only quantity that really describes the noise of a sound source and enables noise emission comparison 4 2 Comparison of different Sound Power Techniques 4 2 1 Introduction The sound pressure and the sound intensity based methods for determination of sound power levels are standardized in the ISO committee The idea with standards in acoustics is to make sure that everybody use the same method thus simplifying comparison of results but the standards does not guarantee that everything is done correctly it is just to ensure that everybody do the measurement in the same way Standards for Sound Power determination comes in three grades of accuracy survey grade least accurate engineering grade medium accuracy and precision grade most accurate The different techniques of sound power determination will yield the same result because the sound power is a property of the sound source which is independent of where the sound source is placed Common for all techniques is the assumption that the sound source is stationary and that octave and third octave sound power levels are calculate from the measurement and then the overall sound power level is Just synthesized from octave or third octave data 4 2 2 Sound Pressure Sound power determination by sound pressure measurements is based upon the ISO 3740 series of standards and they typically require that there is a known acoustic e
46. e would give a positive contribution to the measurement and the sound power would be over estimated Sound intensity measurement in general 1s not less sensitive to background noise than sound pressure measurement it is sound intensity when used for sound power applications When making noise source mapping with sound intensity background noise is an issue Any individual intensity measurement is always sensitive to background noise and maybe even more sensitive than sound pressure measurement might be For example when measuring sound pressure level in front of a sound source and unfortunately the background noise 1s just as strong as the sound source under test then there will be an over estimation of 3 dB in the sound pressure measurement But if that was an intensity probe and 66 sound that propagates from the sound source and from the background noise had equal strength and exactly opposite directions then there will be introduced an infinite amount of error in the sound intensity estimation Because then theoretically the estimated sound intensity would be zero In practice this is not possible but in theory this is possible and then the error is infinitely high Due to the influence of reflections on sound intensity estimation it is extremely important that the operator never stands behind the intensity probe while doing measurements The operator has to be standing to one side of the intensity probe and keep it perpend
47. er applications for analysis in building acoustics This includes both the traditional method as well as the sound intensity method for measurement of sound reduction index Future work could also be to create software for noise source mapping and mapping of sound leakage detection in building acoustics Proposals for improvement of the software introduced in this thesis is to add calculating functions and functions that enables comparison of different measurements 88 8 REFERENCES 10 Br el amp Kjer February 2007 Product Data Sheet Power Amplifier 100W Stereo Type 2716 C Rosendahls Bogtrykkeri N rum Denmark Br el amp Kjer July 1996 Microphone Handbook Volume 1 Theory Naerum Denmark Briel amp Kjer June 2000 Product Data Sheet OmniPower M Sound Source Type 4296 Nerum Denmark Briel amp Kj r March 1996 Product Data Sheet Sound Intensity Calibrator Type 3541 N rum Denmark Br el amp Kj r Lecture notes Retrieved from the course material presented by Svend Gade at the training course in Advanced Acoustics Br el amp Kj r University N rum Denmark 17 18 November 2008 Br el amp Kj r Revision August June 1998 User Manual Sound Intensity Software BZ 7205 Naerum Denmark Br el amp Kj r Revision June 1988 Instruction Manual Sound Source Type 4224 From Serial No 973197 Naerum Denmark CEI IEC 1043 1993 Electroacoustics Instruments for the measuremen
48. er free field condition when sound is propagating along the probe axis in one direction only sound intensity level and sound pressure level is the same In three different other situations the sound intensity level is smaller than the sound pressure level 1 Sound propagates at an angle towards the probe axis 2 There are two or more sources in a free field 3 The sound field is diffuse meaning there is reflections When the orientation of the probe is pointing in the direction of the sound propagation the phase change Ag across the spacer 15 Ar 360 A where Ar is the spacing between the microphones and 4 is the wavelength But if AY 5 16 the orientation of the probe is not in the direction of sound propagation see figure 3 12 then the detected phase change Ag across the spacer is modified as _ Ar cos 0 360 A where is the angle of propagation As a result the detected intensity will be Ap 5 17 lower and the effective spacer distance reduced as Ar Ar cos mm 5 18 Effective Space Distance Ar cos 6 Sound i propagating Phase change at an angle AM across spacer tothe probe fi _ Arcos 360 A Sound propagating along the probe axis Figure 3 12 Sound propagating towards the probe at an angle or along probe axis 5 44 The is of course frequency dependant some frequencies have a more free field character than other frequencies
49. erial inside the control surface might introduce an under estimation of the sound power The sound intensity method is best suited for research and development and engineering testing because it sets higher demands to the operator s skills The method is used for engineering measurements in development of new product in situ and survey measurements and for measurements where it is not convenient or possible to put the device under test in an acoustic test facility Sound intensity rather than sound pressure 1s used because it is a vector quantity that measures the energy flow thus giving directionality information With directionality information it is possible to determine if a surface area is radiating sound or absorbing sound This information can not be found with a sound pressure measurement Sound pressure measurement in most cases must be carried out in the far field Sound intensity measurement can be used both in the near field and in far field without having any near field artifacts 4 3 Determination of Sound Power using Sound Intensity 4 3 1 Introduction and Definition When using sound intensity method for sound power determination a measurement surface 1s defined around the sound source under test Then either by point or scanning measurements on the surface the results are averaged together see figure 4 2 The scanning method is easier to use than the point 62 method With the scanning method the intensity probe
50. f the procedure for achieving the desired grade of accuracy in the sound power determination is given in figure 4 7 Actions to be taken to increase the grade of accuracy of sound power determination are indicated in table 5 In addition to actions given in table 4 3 the dynamic capability Ly of the sound intensity instrument can be increased by modifying the microphone separation A or reducing the phase mismatch by means of calibration 71 Define initial measurement surface and segments Next measurement Measurement of L and Lzy on the measurement surface Evaluate indicators Fe and FJ Action a Engineering Grade or bor f Si Action a QO or f zi s l g DA Action a or b Ly l Lm 2 gt 8 Final result Action c or d Action e Figure 4 7 Scheme of the procedure for achieving the desired grade of accuracy 9 72 Table 4 3 Actions to be taken to increase the grade of accuracy of determination 9 Action code see figure B 1 Fu gt by Halve the average distance of the measurement surface from source to not less than a minimum average value of 100 mm and double the scan line and density Fu gt 3 dB Shield the measurement surface from strong extraneous noise sources by means of a screen Reduce the adverse influence of the reverberant sound field by introducing additional absorption into the test space at locations remote from the source Halve the average distance of t
51. fective acoustical separation of sound pressure transducers used with the analyzer Knowledge is also required about the density of the fluid medium which is influenced by the temperature ambient pressure and composition of the medium The estimated sound intensity is then obtained according to the direct method as A 1 5 22 I 2 pA r Pp py Pa pp dt dB 22 Where P4 is pressure in channel A and Pp is pressure in channel B Ar is the effective acoustical separation of the two microphones p is the density of the medium The density varies with temperature ambient pressure and composition of the medium normally air The intensity can be estimated by assuming a reference density pre for the acoustic medium then compensation is made for the true intensity Z If density o is assumed to estimate intensity and the true density isp then the true intensity 115 obtained from p L L 10108 dB 23 ref Intensity can also be estimated by assuming a reference ambient pressure Po and a reference ambient temperature for the acoustic medium then compensation is made for the true intensity I If Po and 15 assumed to estimate intensity Lef and if the true ambient pressure is po and the true ambient temperature is 7 then the true intensity 7 15 obtained from T L 71010810 o_ 101og m dB 5 24 ref t turns out that the density is proportional to the ambient pressure and i
52. h longer time than the precision pressure based method The scanning version ISO 9614 2 gives engineering or survey grade of accuracy It is easier to meet the standard because there are only two types of field indicators and three criteria Experience 1s required to acquire good scanning technique and it does not give precision grade but engineering grade is satisfactory for 90 of sound power tests 5 Usually takes longer time than pressure based methods but scanning is faster than point measurement The scanning version ISO 9614 3 gives precision engineering or survey grade of accuracy But again experience is required to acquire good scanning technique This 1s a new standard so there is luck of practical experiences and feedback All these three methods can be used in situ in the presence of background noise and there are now restrictions on the volume of the sound source The character of the noise should be steady broad band or narrow band or of discrete frequency Sound power levels obtainable are band limited third octave 50 6300 Hz A weighted and in third octave or octave bands Grade of accuracy is determined 64 from field indicators Optional information is available of positive and or negative partial power concentration The reason why the point method ISO 9614 1 gives precision grade and the scanning version ISO 9614 2 does not is a question about repeatability When using the point method there w
53. he bandwidth is 2600 Hz The bandwidth becomes broader and broader at higher frequencies but the relative bandwidth is the same that does not change It is a fix percentage It is 28 standardized so that three third octave bands are covering one octave band This is in contrast to vibration measurement where a Fast Fourier Transform FFT analyzer is to prefer The FFT analyzer uses what is called narrow band analysis and a linear frequency axis It has a constant bandwidth which means that a FFT frequency line has the same resolution at any centre frequency see figure 2 17 y m ctave 1 3 Octave Figure 2 17 The Spectogram FFT 5 2 3 4 Perception of Sound The audible range of the human hearing system is called audio sound and has been defined to be typically in the frequency range from 20 Hz to 20 kHz see figure 2 18 Sound below 20 Hz is called infra sound and sound higher than 20 kHz is called ultra sound But the auditory field is a little more complex and complicated than just going from 20 Hz to 20 kHz Because there is a threshold of hearing also called threshold of quiet This means that there need to be a relatively high sound pressure level for example at 20 Hz before the human hearing system can hear it while for example at 4 kHz the sound pressure level does not have to be very high in order for the human hearing system to detect it m g 2 EL 3 a e Threshold in Quiet 500 k
54. he measurement surface from source to not less than a minimum average value of 100 mm and double the scan line density Reduce the adverse influence of the reverberant sound field by introducing additional absorption into the test space at locations remote from the source Identify and suppress causes of temporal variation in field conditions or if tnis fails double the scan line density on the same segment Lu 1 AL ns Double the average distance from the measurement surface to the source SEN keeping the same scan line density F j 1 dB 73 5 MEASUREMENTS 5 Test Setup Description The purpose of the tests was to carry out sound power determination of two individual sound sources each with different acoustic directional characteristics in a reverberant chamber The chamber was of concrete element isolated from the main structure of the building and has two doors The inner dimension of the chamber was 3 16m x 2 60m x 2 42m width x length x height The sound power levels were determined for each sound source using sound intensity according to ISO 9614 2 with and without the presence of background noise generated by the other sound source The idea was to study the influence of background noise on the sound power levels determined for each of the sound sources The sound sources were placed on the floor A steel frame box with dimension 0 6m x 0 6m x 0 6m was used to define a box surface around the soun
55. icular to the measurement surface while measuring 4 4 Determination of Sound Power according to ISO 9614 2 4 4 1 Introduction The procedure of the determination of sound power levels of noise sources using sound intensity according to ISO 9614 2 see figure 4 6 is to first define a surface that completely encloses the device under test and divide it into segments The average sound intensity for each segment is then obtained by performing two individual scans so that the second scan is orthogonal to the first Each scan must last at least 20 seconds Then both local and global criterions are evaluated Finally the total sound power is calculated by adding the results of all the segments The total sound power P generated by a sound source 1s given by P WP W 90 31 ip and Mez P E 9 32 1 Il where WN is the total number of segments 7 of the measurement surface Sound power level L is the logarithmic measure of the sound power generated by a source and is given by 1 10log P P dB 9 33 vvhere Po 1s the reference sound povver 1071 W 67 Partial sound power P is the time averaged rate of flow of sound energy through a segment of a measurement surface given by P 1 8 W 9 34 where is the signed magnitude of the segment average normal sound intensity measured on the segment 7 of the measurement surface Si is area of segment i m Partial soun
56. ighting for low sound pressure levels B weighting for medium level sound pressure levels and C weighting for high sound pressure levels 30 e 4006 Equal LOoudness Contour Yormalized Io 0 dB at kHz 40 dB Equal L udnaes s Contour inverted nd compared i wd M s A weighting A weighting ared to A weight 5 Figure 2 20 The 40 dB Equal Loudness Contours inversed and comp Over the years it has become a tradition to use the A weighting and today most acoustic measurements are made with A weighting In the most recent standards there is only A and C weightings used A weighting typically under estimate the sound pressure levels at low frequencies and also attenuates the results too much in the frequency range of 3 4 kHz where the human hearing system is most sensitive The D weighting is actually the most proper acoustic weighting to use but A weighting is most commonly used in measurement standards see figure 2 21 C weighting 1s typically used for peek measurements where sound pressure levels might be high as in an industrial environment A and C are standardized frequency weighting filters ki edi Frequency 10 20 50 100 200 500 ik 2k 5k 10k 20k Hz Figure 2 21 A B C and D weighting curves 5 31 3 SOUND INTENSITY AND ITS APPLICATIONS 3 1 Introduction to Sound Intensity Sound intensity is the rate of acoustic energy flow per unit area Sound intensity is
57. ill be a high amount of repeatability Because one can put the intensity probe into a position and make a measurement and at some other point in time put the intensity probe into exactly the same position and do the same measurement again While this is not so easy with the scanning method because it is not possible to manually scan exactly the same path twice The question is which method would give the highest amount of accuracy in the estimation The difference between repeatability and reproducebility is that repeatability means that a measurement can be repeated in exactly the same way while reproducebility means that a measurement can be reproduced somewhere else at another location But repeatability does not necessary mean that there is a high amount of accuracy Because an error in a measurement can be repeated over and over again but all the measurements have errors And it turns out that even though a point measurement can be repeated it does not necessary mean that it has the same amount of accuracy as with the scanning method Because a point measurement is an approximation to the surface integral but the scanning method tends to give a better approximation With the scanning method there will be an infinite number of points over the measurement surface but with the point method there will only be a finite number of measurement points 4 3 2 Influence of Background Noise When measuring over a controlled surface where there is no
58. it 1s time to change gear You can also take pleasure in sound for example listen to music or take a walk in the forest and listen to some sound there But what is noise Every time sound is unwanted it is called noise Noise can be harmful and it may damage the hearing It may also not be harmful but maybe just annoying Typically when the neighbour is playing some music it is very annoying but not when you play music yourself Noise does not have to be loud to be annoying 2 1 2 Basic Quantities of Sound The three different quantities describing sound are sound pressure sound intensity and sound power see table 2 1 Sound pressure is a scalar describing the pressure fluctuation at a given position and is measured in Pascal Pa Sound pressure is typically measured at the receiver s position for evaluation of the harmfulness and the annoyance of a noise source Sound intensity 1s a vector quantity that describes the amount and the direction of flow of acoustic energy at a given position The unit for sound intensity is Watt per square meter W m Measurement of sound intensity needs a special probe consisting of two microphones and a sound intensity analyzer Sound intensity describes the path of sound and is used for noise source location and rating of noise sources 12 Sound power can only be calculated or determined either based upon sound intensity measurement or based upon sound pressure measurement The main use of so
59. ithin which measurements to precision engineering and survey grades of ISO 9614 may be made ISO 9614 gives requirements for the K factor according to the grade of measurement accuracy required This K factor is subtracted from the PRI index to give the dynamic capability index of the analyzer fig 3 18 The dynamic capability indicates the dynamic range for intensity measurement and the dynamic range of the analyzer must be bigger than the dynamic range of the sound field Dynamic capability index Ly is expressed as L dB 5 20 Once the dynamic capability is determined this value is compared the measured PI index of the sound field as if the PI index dynamic capability desired accuracy is achieved if the PI index gt dynamic capability measurement does not meet desired accuracy X MEYS M Exp s meth aes HB Pressure Rs Er l vr Intensity 20 index e P Res l 15 index 25 50 100 200 400 B 1600 Hz sm see Figure 3 18 Dynamic Capability Example 5 50 3 5 Calibration 3 5 1 Introduction Calibration of a sound intensity analyzer includes sound pressure calibration of the individual microphone channels verifying the sound intensity and particle velocity levels and measurement of the pressure residual intensity index This requires an intensity coupler a pistonphone a broadband noise
60. k to the computer In this thesis a 5 way was introduced for transferring and displaying the measurement data Namely to create a computer application that reads information from the binary files in which the Sound Intensity Software BZ 7205 of the analyzer has stored the measurement data This solution has some advantages It is fast it removes the need for a hardware lock and once data 1s read into the program it can be analyzed and post processed in an infinite number of ways In order to be able to read data from binary files the position length and format of the stored data must be known For this reason the files generated by the l analyzer was mapped with a hex editor while compared to the same measurement data obtained by a software package supplied by the manufacturer of the analyzer An example of such a hex editor mapping 1s indicated in figure 5 13 0003 5GM 00000000 2260 27205 2 00000010 1 77 5 00000020 00000030 00000040 00000050 00000060 00000070 00000080 00000090 00000080 0000000 000000 0 00000040 00000020 00000020 00000100 00000110 00000120 00000130 00000140 00000150 00000160 Figure 5 13 The file 0003 8GM edited with a hex editor in MS Visual Studio 2008 As a result a functional computer application was created in MicroSoft Visual Studio 2008 using C which reads and displays the relevant measurement data In this version of the software the application b
61. l to the input channel A of the amplifier The level of pink noise was set to the maximum level 4dBu of the generator see figure 5 8 Figure 5 8 Minirator MRI 78 The Br el amp Kjar Sound Source Type 4224 is a loudspeaker with a built in power amplifier and noise generator see figure 5 9 The type 4224 is specially designed for building acoustics measurements Figure 5 9 Br el amp Kjar Sound Source Type 4224 When driven continuously from a mains supply the Type 4224 can typically deliver up to 118 dB sound power level L in the frequency range from100 Hz to 4 kHz In its wide band mode Sound Source Type 4224 produces a pink noise signal from 100 Hz to 4 kHz Sound power spectra with the 4224 operating at full power for Wide Band mode is shown in figure 5 10 7 Note that the diffuser cone was at no time during these measurements attached to the loudspeaker Sound Power 110 dB re 1 pW Wide band 63 125 250 500 1000 2000 4000 Hz Figure 5 10 Sound power spectra with the 4224 operating at full power 7 79 The Br el amp Kjar Sound Source Type 4224 has different acoustic directional characteristics than omni directional sound sources Typical directivity characteristics of the sound source are indicated in figure 5 11 490008000009 490000000099 0 0000000000009 009 00 gt 000 00 400 00 s EY el PCM 700 COMM SL A n 6 20 69 d b y gt
62. lm ll Intensiteettimenetelma sopii hyvin kentt mittauksille erilaisissa akustisissa ymp rist iss Adnitehotasojen m ritt minen niintensiteetti menetelm ll on yleisesti pidetty v hemm n herkk taustamelulle kuin nipainemenetelm ja n in ollen v hemm n akustiikkaan liittyvi vaatimuuksia asetetaan mittausymp rist lle Opinn ytety ss k sitell n my s akustiikkaa yleisell tasolla Intensiteettimenetelm n k yt n tutkimista varten suoritettiin kaiullisessa tilassa nitehotasojen m ritt minen kahdelle erityppisille nil hteelle Taustamelun vaikutus nitehotasoihin mittaustarkkuuteen tutkittiin m ritt m ll nil hteiden nitehotasot kahdessa erilaisessa tilanteessa ensin yksin ja my hemmin toisen nil hteen vaikutuksen alaisena Ty n yhten osana on C ohjelmointikielell MicroSoft Visual Studio 2008 ymp rist n toteutettu tietokonesovellus akustisten mittaustuloksien analysointia varten Mittaustulokset tiedostoissa olivat bin risess muodossa joten niit kartoitettiin ensin hex editorilla Lopuksi annetaan ehdotuksia uusiin tutkimuksiin ja parannuksia tietokone sovellukselle Avainsanat A nipaine nen intensiteetti niteho VASA YRKESHOGSKOLA Rakentamisen koulutusohjelma ylempi AMK tutkinto ABSTRAKT Forfattare Michael S derback Lardomsprovets namn Best mning av ljudeffektnivaer med ljudintensitetsmetoden Ar 2009 Sprak E
63. ly used when analyzing rapidly changing signals see fig 2 7 Figure 2 7 Time Weighting RMS detector 5 Slow time constant will smooth the data much better and is typically used for averaging stationary signals A longer averaging time minimizes the random error in a signal When the reverberation time is measured in a room the time constant in the analyzer must be shorter than the time constant of the room Otherwise it is the reverberation time of the analyzer that is measured not the room Figure 2 8 Peak Hold Peak Detector 5 A Peak detector must have some sort of hold facility for keeping the Peak level until a new Peak value would arise that is higher than the previous one see figure 2 8 The ratio between the Peak and the RMS value called that the crest factor is 21 1 4 for a pure sinusoid signal For a transient signal like a handclap the crest factor is higher 2 2 2 Leg The Leg is the sound level of a steady state noise source that is producing exactly the same amount of sound energy as a true fluctuating sound source over a certain specific measurement time Linear averaging means that all measurement data is equally weighted and it is an arithmetic averaging of the noise signal see figure 2 9 With the linear averaging there will be some starting random errors and with a longer measurement time the results will be smoother and give a better linear average e Integrating Sound Level Meters
64. mbient temperatur 20 C Ambient pressure 1011 mbar Measurement positions 4 Measurements per position 4 Sound level meter B amp K Modular Precision Sound Analyzer Type 2260 Analyzing software B amp K Building Acoustics Software BZ7204 v 2 3 Sound source B amp K OmniPower Sound Source Type 4296 Amplifier B amp K Power Amplifier Type 2716 C a 28 oss 80 187 oa 24 iso oa cen cn ce len cn m CAN SSc Figure D 2 Measurement equipment setup for measurements 106 Pattee 1242 AA N V Note a 3 2 2 MP 255 0 05 il 1 5 2 2 5 3 3 5 4 4 5 5 55 0 05 1 15 2 25 3 35 4 45 5 55 Figure D 3 Multispectrum no 1 Figure D 4 Multispectrum no 2 p w Meh af DO MAPAN ES am LIE EUER GC TEIN PEE A 7 2262 7 poseer at ae EPRS A 557 9 4 201 r SI S R F y A i l L 4 ar ed yet Ann A YALAN 4 7 C NAI N h 2 5 u 0 O05 1 1 5 2 25 3 3 5 4 45 5 55 Figure D 6 Multispectrum no 4 0 Of 1 1 5 2 25 3 3 5 4 4 5 5 55 Figure D 5 Multispectrum no 3 TEN Mesi DESC DESI ees EE RD EET NS OR a nl SE 3 a l ee 722 27 CAP dar ox BY XC E
65. mi 2 iz ff LR CS fot S 22 Si TLEER aly iz iff 2 7 7 5 ET e Vs ara a ala POR X Ein i PA MEAS 1 E NM 0 O05 1 15 2 25 3 35 4 4 5 5 55 Figure D 8 Multispectrum no 6 4 0 NA NA v paria WEN f p ag i C 3 7X E ur i RR f p Lie 4 ATG SH s EET He Sir o n 7 NT SS s n Fost gt 2 0 05 1 1 5 2 2 5 3 3 5 4 4 5 5 55 0 05 1 1 5 2 25 3 3 5 4 45 5 55 Figure D 9 Multispectrum no 7 Figure D 10 Multispectrum no 8 l3 J EN TREERE sede uec ER RUN 000 i are asec 7 l L P AIT n AG na n d Y xn B 7 A we 74 nf 77500 SC x ME YA I x Abia L X ON AS H 40 8 Nw D NN Bea i e d B a I m dil 11 Bl NNNM S dii 11 SEL T E 2 z ZEY KE 12775 58 PN Lo A EN o gt j EAN LL NN ENNE Paka M E SSS am Ana Reet DAC i a 22 g b 4 2 Pee a P AO a ui 05 1 1 5 2 2 5
66. n r see figure 3 4 there 1s 1 p ni 11 t p r 35 Then theoretically from Euler the particle velocity u is obtained by integrating the pressure gradient with respect to time I p dt u Es 11 12 In practice the pressure gradient is approximated as the pressure at one microphone pg minus the pressure at the other microphone pa divided by the separation distance Ar between the two microphones Thus giving the finite difference approximation for the particle velocity u in the direction r at a point midway between the two microphones as 1 u ard p at 11 13 This is a valid approximation as long as Ar 11 Theoretical Finite Difference from Euler Approximation 1 Ea 1 Pe Pa q uz por p AT 1 _ PatPs r Averaqe instantaneous ressure p DP R P Figure 3 3 How Sound Intensity is Estimated 5 The average of the two instantaneous sound pressures in the two microphone positions added together divided by 2 is an approximation to the sound pressure midway between the two microphones See figure 3 3 36 Figure 3 4 Sound intensity in one direction r is estimated with two closely spaced microphones The advantages with estimating the sound intensity with two microphones are that measuring microphones are common transducers in acoustics they can be used in a wide variety of situations and they can be easily c
67. n of validity check s F 3 dB of each measurement 85 6 CONCLUSIONS When the sound sources were running alone by themselves there were no difficulties in meeting the standard ISO 9614 2 in terms of accuracy indicated by the field indicators and criterions for the measurements even though the measurements were carried out in a reverberant chamber When the sound sources were running together the criterions and field indicators for measurements on the omni directional sound source B amp K 4296 were kept on levels according to the engineering grade of accuracy indicated by the standard ISO 9614 2 The difference between the two determinations of sound power levels A total 109 1 dB and 108 8 dB for B amp K 4296 is smaller than what is indicated by the engineering grade of accuracy 0 5dB The background noise influenced the accuracy of the determination of sound power levels for B amp K 4224 This was specially indicated during measurements in segments other than on the front side of the sound source where the repeatability criteria was very difficult to fulfill although scanning was executed in the same way as for all other measurements This 1s an indication that the acoustic character of the sound field within a segment varies much These segments should according to ISO 9614 2 be subdivided into smaller The influence of background noise also caused one frequency band 315Hz to have a negative direction for the tot
68. ngelska Antal sidor 89 4 bilagor Handledare Tapani Hahtokari Detta l rdomsprov r en studie av ljudintensitetsmetoden for best mning av ljudeffektnivaer Metoden r anv ndbar for f ltm tningar i olika akustiska milj er Ljudintensitetsmetoden f r best mning av ljudeffektniv er anses allm nt som mindre k nslig f r bakgrundsljud n ljudtrycksmetoden I l rdomsprovet be handlas ven grundl ggande kunskaper och begrepp inom akustiken F r att unders ka och utv rdera metodens anv ndning 1 praktiken utf rdes best mning av ljudeffektnivaer f r tv olika ljudk llor i ett m trum med lang efterklangstid De tv ljudk llorna hade olika akustiska egenskaper med avsikt p riktningen f r den ljudenergi som utstr las Ljudeffektniv erna best mdes f r ljudk llorna b de utan och med bakgrundsljud genererat av den andra ljudk llan Som en del av l rdomsprovet har ett datorprogram skapats f r analys av akustiska m tdata Datorprogrammet har utvecklats med C programmeringsspr ket i MicroSoft Visual Studio 2008 Datorfiler med m tdata sparade 1 bin r form kartlades med hex editering Till sist ges f rslag om akustiska unders kningar och f rb ttringsf rslag f r ytterligare utveckling av datorprogrammet mnesord Ljudtryck ljudintensitet ljudeffekt TABLE OF CONTENTS PREFACE ABSTRACT TITVISTELMA ABSTRAKT I INTRODUC HOT e x m Oo e add 9 Il Technobothnra Rescarehe entre iori a adadan 9 1 2 T C Pu
69. nverse proportional to the temperature Acoustical separation between the microphones is the distance between the acoustic centers of the microphones and is depending upon frequency Acoustical 53 separation 1s measured in anechoic chambers as a function of frequency At some frequencies the acoustic center is in front of the diaphragm and at some other frequencies it is behind the diaphragm The error due to the effective acoustical separation Ls separation Of the microphones is given as Ar L on LOlog amp Separation 2 10 Ar 5 25 Where Ar is the actual spacing and Ar is the nominal spacing The acoustical separation between the microphones is normally supplied by the manufacturer The analyzer must also be calibrated with respect to the acoustical separation ars ed Figure 3 22 Setup for sound intensity and particle velocity calibration The analyzers sensitivity to sound intensity and particle velocity can be verified by using both chambers of the coupler Each microphone of the intensity probe is positioned in different chamber and the pistonphone is mounted on the top of the coupler as indicated in figure 3 22 With this arrangement the pistonphone generates a phase difference between the sounds pressures in the different chambers but the sound pressure amplitude in the chambers is the same thus simulating a plane sound wave propagating in a free field along the axis of the probe The
70. nvironment The sound pressure method can be roughly divided in to free fields methods in anechoic rooms and diffuse field methods in reverberation rooms 60 With the free field method typically a hemisphere a parallelepiped or a shoe box is used to define the surface around the sound source and then the measurement points are at the exes on the corners of these surfaces While in a reverberation chamber there is typically a reference sound source RSS and the device under test DUT The determination of sound power levels is then by simply making a comparison between the sound pressure measurements of RSS and of DUT Typically a rotating boom is used to get an average of the sound pressure in the entire room Because no room is perfect if 1t was perfect it was good enough just to make measurement at one point But because of imperfections it has to be measured in a number of points and averaged or measured by using a rotating boom The advantages with the sound pressure method for determination of sound power levels is that it gives reliable result relatively simple to follow and gives a wide frequency and dynamic range But the disadvantage is that it requires a qualified acoustic test facility The pressure methods are used for production audits and testing high volume testing determination of low level sound powers and where the user already has an acoustic test facility Pressure method is best suited for non qualified pers
71. o correct for phase mismatch by measuring the phase mismatch of the system and then correct subsequent estimations automatically This will improve the dynamic range of such a sound intensity system 42 3 4 Validity of Sound Intensity Measurement Data 3 4 1 Pressure Intensity Index The quality of sound intensity measurement is typically checked by comparing the pressure intensity index PI index to the Pressure Residual Intensity index PRI index The PI index is a property of the sound field and the PRI index is a property of the measurement system The PI index is expressed as F L L dB 5 15 The PI index F is the level difference between sound pressure and sound intensity indicating if the measurement was made in free field or reverberant field Under free field condition sound pressure level and sound intensity level is due to choice of references exactly the same and is the easiest sound field to handle see figure 3 11 In other types of sound fields there might be more sound sources or there might be a reverberant sound field where there is an increasing level difference between sound pressure level and sound intensity level thus increasing the PI index These sound fields are more complex and complicated to work with If pc 400 Nsm then L L but pc 415 Nsm at 20 C and 1013 mbar Liz 0 16 dB Figure 3 11 The Relationsship Between Pressure and Intensity in a Free Field 5 43 Und
72. onnel In standardized sound power determination using sound pressure the standard require that measurements are carried out in the far field of the sound source and the distance between the measurement points and the sound source is depending upon the size of the object under test 4 2 3 Sound Intensity The sound intensity methods for sound power determination can roughly be divided in to point or scanning measurement and they are standardized in three versions Version 1 ISO 9614 1 is discrete point measurement with precision engineering or survey grade Version 2 ISO 9614 2 is measurement by scanning with engineering or survey grade Version 3 ISO 9614 3 is measurement by scanning with precision engineering or survey grade 61 These methods can be used in situ in almost any acoustic environment which means that a special acoustic test facility is not required They also includes location ranking and segmentation of noise sources which means that it is possible to calculate how much sound power is emitted from various parts of the device under test Sound intensity gives directional information and isolates the object under investigation There is no restriction on the shape of the control surface Sound intensity method has less demand on the background noise and steady background noise is excluded Sound intensity measurements can be carried out in both the near field and the far field of the sound source Absorption mat
73. ophones would measure the same instantaneous pressure but in opposite phase thus generating a sum which is equal to zero Likewise if the separation between the two microphones were exactly one wavelength then the two microphones would measure the same pressure and in phase thus generating a difference equal to zero At high frequency the separation between the two microphones must be smaller than the wavelength If the separation is approximately six lt 60 of 360 6 times smaller than the wavelength or the phase change over the spacer is smaller than 60 then the sound intensity can be estimated within an accuracy of 1 dB Otherwise there will be what 1s called a finite difference approximation error Higher frequencies can be estimated by making the separation between the two microphones smaller but only up to a point where the signal to noise ratio is becoming to poor Signal to noise ratio depends also on the averaging time High frequency limit depends also on microphone dimensions 39 For limitations at high frequencies there is a formula indicating the error in estimating sound intensity level L as 1 sin kA L 10log 10log dB 5 14 I kAr Where is the estimated intensity is the corrected intensity and kAr is the characteristic dimension for the spacer This means that the estimation is correct up to a certain upper frequency limit and then there will be an under estimation see figure
74. oring installed application software 75 The analyzer can be used for general sound intensity measurements and for sound power determination according to ISO 9614 2 when equipped with a sound intensity probe and the sound intensity software BZ 7205 Figure 5 3 Probe with extension stem on handle and probe on handle 6 The intensity probe can be mounted on a handle or betvveen the handle and the probe can be mounted an extension stem see figure 5 3 The cable from the handle is then connected into the input stage socket at the top of the analyzer Alternatively the extension stem with the probe can be mounted directly into the input stage socket see figure 5 4 The display can be turned around 180 degrees thus enabling control of the software keys with the left hand while performing scanning with the right hand The sound power application can be controlled with a one key operation Figure 5 4 B amp K 2260 with extension stem and probe 5 The use of Br el amp Kjer Modular Precision Sound Analyzer Type 2260 Investigator for sound power determination according to ISO 9614 2 is described in detail in the user manual Sound Intensity Software BZ 7205 76 5 2 2 Sound Sources and Accessories Two sound sources with different acoustic directional characteristics were chosen for the tests An omni directional sound source radiates sound evenly in all directions The Br el amp Kjar OmniPower Sound Source Type
75. ough the medium 3 Sound Power is independent of the acoustic environment and is therefore a good parameter for making comparisons of sound sources Pressure vs Power Temperature t C Power P MW Electrical Healer Figure 4 1 Sound Pressure vs Sound Power 5 In figure 4 1 the analogy of sound pressure vs sound power is described with a heater in a room A heater will emit the same amount of heat from any position in 58 any room but the temperature in the room is highly depending upon in what kind of environment the heater is For example depending on if it 1s winter time or summer time the temperature in the room due to the heater would be different even though the heater is producing the same amount of heat all the time Or in other words the power of the heater is known but its influence on the temperature of the air at the position of the thermometer is unknown and depending on the environment It is the same thing about a sound source a stationary sound source is always producing the same amount of sound But the sound perceived in the room 18 depending upon the acoustic properties of the room it might be an anechoic or a reverberation chamber or any other room and it is also depending upon the distance to the sound source But the sound source itself 1s always emitting the same amount of sound It is just the pressure in the room that would be different due to the acoustic pro
76. ously or in buildings with hard walls for example churches sound fields with acoustic characteristics like a diffuse field may be found Anechoic rooms are set up to produce a free field situation and a controlled acoustical environment Anechoic rooms are used for many different purposes like for example sound power determinations or measurement of the directivity pattern of acoustic noise sources In a free sound field the sound waves can propagate freely without being disturbed by any object along their path A sound field at a distance 1 2 m away from the sound source can be regarded being a free field if no other sound source gives a significant contribution to the sound pressure and there is no influence of reflecting surfaces Reverberation chambers are designed to produce a reverberant sound field by making the walls and the boundaries as highly reflective as possible The idea with such a room is to have a controlled acoustical measurement Reverberation chambers are typically used for sound power determination of noise sources A sound field where the particle velocity is in phase with the sound pressure is called an active sound field and constitutes the real part of a complex sound field In such a sound field all acoustic energy is transmitted and none of it is stored An example of this is a plane sound wave propagating in a free field A sound field where the particle velocity is 90 out of phase with the sound pressure is
77. ower 4 down to 20 Hz A change in spacer from 12mm to 6mm would typically just shift the entire frequency range a factor of two up Or a change in spacer from 12mm to 50mm would shift the range a factor of four lower in frequency See figure 3 10 Frequency range for 1 dB Accuracy in a Free Field Total Phase Mismatch in Probe and Analyzer Microphone 0 05 Spacing Ar ee Figure 3 10 Change of Spacer Shifts Frequency Range 5 Microphones used for sound intensity estimation differs from microphones used for ordinary sound level meters The phase match of these microphones can be improved by adding phase correctors There is a ventilation hole for equalizing changes in the ambient pressure which are large pressure variations compared to sound pressure variations A leak of sound pressure outside the microphone to the inside of the microphone is creating a phase change at low frequencies and the phase matching between microphones is a question of being able to control the tolerances in making these equalization holes By adding more equalization holes further attenuation is made of the sound field that is detected through the ventilation holes Thus making these microphones less sensitive to low pressure through the ventilation hole compared to a normal microphone This is improving the low frequency phase matching and it also improves the low frequency performance for sound intensity estimations It 1s of course possible t
78. perties of the room So the sound pressure is highly depending on not only the sound source but also what kind of environment the sound source is placed in This is why Sound Power is the basic quantity for characterizing a sound source If there is a knowledge of the sound power levels L of a sound source and of the properties of the acoustic environment the sound pressure level L due to that sound source at a given distance r in that environment can be predicted by adding some correction terms to the sound power levels as 1 L 10log Q 10log r 11 dB 5 29 Where is the directivity factor In a free field situation the directivity factor is equal to one or 0 dB But if the sound source is close to a wall the directivity factor is two or 3 dB If the sound source is placed between wall and floor the directivity factor is equal to four or 6dB and if it 1s placed in a corner then the factor is eight corresponding to 9 dB The correction term of 11 dB corresponds to 4n This means that with knowledge of the sound power of a sound source it 1s possible to predict what the sound pressure is at any position in space 59 Sound power is used to determine that a sound source complies with noise specifications which means that in order for manufacturers to be able to sell noisy machinery and equipment they have to declare the maximum sound power level emitted by their product For engineers trying to develop mor
79. pressure level which decays to a smaller value at the receiver s position further away The sound pressure is a small pressure variation on top of the atmospheric pressure and is sometimes called the dynamic pressure A typical sound pressure is 1 Pa Atmospheric pressure is typically 100 000 Pa or 1000 mbar and is 13 sometimes called the static pressure The atmospheric pressure is measured with a barometer and the sound pressure is measured with a microphone The highest and the lowest sound pressure that can be perceived by the human hearing system are called threshold of pain and threshold of hearing The ratio between the threshold of pain and the threshold of hearing is seven orders of magnitude Range of Sound Pressure Levels Sound Pressure p Nim Pascal Pa Sound Pressure Level L 140 dB B 60 40 20 reference p 20 uPa 4 0 0 000 01 Absolute Scale Relative Scale Figure 2 1 Range of sound pressure and sound pressure level 5 2 1 4 The Decibel Scale Sound pressure is measured in the unit Pa on an absolute scale which covers seven orders of magnitude see figure 2 1 But in acoustics it is preferable to use a relative scale With a relative scale it is possible to set the zero point at a certain position and it has been chosen to be at the threshold of hearing The threshold of hearing is then used as a reference sound pressure po and has been determined to be the sound p
80. r was switched on for sound power determinations The loudspeaker was not driven at full power because of overload indications Sound Source Bruel 8 Kj r Type 4224 Sewn tw 2517 Spec I yo P 2 Sentra 2 40 45 Figure 5 12 Positions of level controls of Type 4224 during measurements 80 5 3 Results and Analysis 5 3 1 Application for Measurement Data Analysis Measurement data stored in files in the Br el amp Kjar 2260 Investigator analyzer can be transferred and displayed in a number of ways 1 Measurement data can be sent from the analyzer to a printer via the serial interface by using the print option in the analyzer 2 Measurement data can be transferred using an application called 2260 Investigator Link The measurement data is transferred via the serial interface to a file on a computer by using the print option in the analyzer The files can be edited with a simple text editor 3 Measurement data can be transferred using a software package for example Br el amp Kjar Noise Explorer Type 7815 either directly from files in the computer or via the serial interface from files in the analyzer This typically requires a hardware lock attached to the computer 4 The analyzer can also be remotely controlled via the serial interface by sending commands from a computer to the analyzer When the computer sends commands corresponding to a request of data the analyzer sends the requested data bac
81. ressure amplitude equal to 20 uPa for a large population of young persons with a good hearing system The decibel scale is then defined by comparing the sound pressure p to the reference sound pressure Thus the sound pressure level SPL or L is defined as 10 2 Lp 1010 2 dB 1 0 There are several advantages with the decibel scale One advantage is that the human hearing system actually perceives sound or noise in a logarithmic fashion 14 If for example the sound pressure is increased from mPa to 10 mPa That is an absolute increase in sound pressure by 9 mPa This will be perceived by the human hearing system as the same change as if it first listen to a sound pressure of 1 Pa and then increased to 10 Pa That is an absolute increase of 9 Pa but the human hearing system perceive this change to be the same Another reason for using the decibel scale is that this scale essentially 1s only covering two others of magnitudes This means that the numbers are easier to handle The ambient pressure that surrounds us is around 100 000 Pa and that corresponds to 194 dB It is not possible to produce a sound pressure level that is exceeding the ambient pressure This is of course far above the threshold of pain where at the hearing system would be damaged 2 1 5 Sound Sources Sound sources are sometimes referred to as point sources line sources and plane sources see figure 2 2 Every time the distance from a point source is
82. rpose acere aici r al ai apse G lal 9 1 77 EC LO a a Da a ME 10 2 BASIC KNOW LEDGE ai sede a lala l olsada 11 2 1 Basic Concept of SOM iuc ero ea iii 11 Zel Sound nd INOIS Cis a aa ada daaa 11 2 1 2 Basic Quantities of Sound 11 2 1 SOUNGIP LESS Ute 3 12 DWE Ili SCID eR e 13 2 1 5 SOME olitcon esi Ebene ali ella 14 251 6 Sound Eve Sue a stb ni a kyn ee 16 2 2 Nasir SON rez e ie il e alaya in lak e s arene na 19 2 2 1 Basic Sound Level Parameters 19 7 ee e baba 21 2 2 3 Measuring Sound n Practice seien enne 27 2 3 Basic Frequency Analysis of Sound 22 2 3 1 Frequency and VVavelengith 22 2527 he lciopioneeecun nan ml dale yanliz 24 2 5 XL Bej MC Y NE yS Sa a e A a a A e 26 2 Perception Ol SOUNC ar 28 3 SOUND INTENSITY AND ITS APPLICATIONS 31 3 1 Introduction to Sound ntensity 31 3 2 Applications of Sound Intensity Uca aa le alimi 32 3 2 1 Sound Power Determinat on 32 3 2 2 Noise Source Identification 33 52 BUGIS ACOUSUCS riner early Gates bep no 33 3 3 Estimation of Sound ntensity
83. s 0 2 of a dB In order to get the same statistical accuracy in the sound intensity measurement where F is equal to 10 then the BT product most be equal to 10 If the F p 1S equal to zero the same statistical accuracy is obtained as for sound pressure measurement 45 BT product 108 T T Li r m b 1 r b _ pr or sa p 7 Li r b 1 1 r 20 dB PI Index Pu Figure 3 13 Random Error 68 Confidence Level 5 In order to obtain more accurate results a longer averaging time should be used thus reducing the random error Generally at lower frequency where the bandwidth is smaller and the random error is larger a longer averaging time is needed in order to get the same BT product as for a higher frequency where the bandwidth is larger For a third octave sound intensity analyzer the required averaging time is therefore determined by the lowest frequency of interest An example of where the averaging time makes a difference is when measuring the PRI index Here the microphone probe is mounted into a coupler where a sound field with high F is simulated Then the required averaging time in order to measure inside the coupler is much longer than for a measurement under free field condition The F can be used to establish how much averaging time is needed because the higher 2 there is the longer averaging time is needed 3 4 3 Pressure Residual Intensity Index
84. se or out of phase or what phase relationship there is and also a question about wavelengths In the far field there is a more consistency in the sound field and sound pressure levels will not vary too much when the measurement position is moved slightly The far field is also experienced if there is a free field situation If the sound pressure level decreases by 6 dB when the distance to a point sound source is doubled then there is a free field condition see figure 2 3 Far field Free field Reverberant field Distance r Figure 2 3 Sound fields 5 In the far field far away from the sound source there is a 1 1 relationship between pressure intensity and particle velocity they look the same and they are all in phase with each other In the far field there 1s a free field situation 17 If the measurement position is close to some boundaries like floors walls and ceilings there might be reflections of sound as well as direct sound In such a sound field a change in position further away or closer to the sound source may not give a significant change in the sound pressure level because a lot of the sound pressure is caused by reflections This is called a reverberant field or a diffuse sound field A sound field with a diffuse character exists where the sound waves propagate from all directions with the same probability and the same sound pressure level In factories where many sound sources exists simultane
85. sensitivity as indicated in figure 3 2 Figure 3 2 Intensity Probe Directivity Intensity 5 3 2 Applications of Sound Intensity 3 2 1 Sound Power Determination A major application of sound intensity 1s the sound power determination It might be a non standardized measurement for indication of the sound power levels or it might be a measurement according to a standard There exists several different standards for sound power determination and it is generally regarded to be more time consuming when using standards because there is a number of field indicators that have to be monitored Sound intensity measurement is considered to be not sensitive to background noise when used for sound power determination 33 3 2 2 Noise Source Identification A simple noise source location application is moving the probe forwards and backwards around the sound source and look for quick changes in the intensity sign This technique is suitable only when there is one major sound source presence A more advanced application is the noise mapping which will generate a complete noise map of the sound source thus making it easy to pin point exactly where the noise source is Sound intensity enables source ranking by calculation of how much sound energy or sound power is radiated from the various parts of a sound source One of the advantages of using sound intensity 1s that it gives directional information because sound intensity is a vector quanti
86. sound source inside but there is a stationary sound source outside then this background noise would give a contribution of energy flow going into the surface on one region and the same amount of energy would move out of the surface on another region Then the amount of negative and positive intensity would cancel each other and the contribution from background noise would be zero if the measurements are executed carefully enough This is the basis upon why the sound intensity 65 measurement for sound power determination is considered to be less sensitive to background noise than sound pressure measurements Source inside surface Source outside surface Source inside and outside surface Figure 4 5 Internal and External Sources Combined 5 Depending upon if there is a stationary source inside the surface or a stationary source outside the surface or both the intensity vectors may have completely different directions because it is a summation of the internal source and the external source see figure 4 5 But as a result the calculated sound power will only be the sound power contribution from the internal stationary sound source Unless there is some absorption inside the surface because then there will be more energy flow into the surface than out of the surface and that would give a negative contribution and the sound power would be under estimated This is different from sound pressure measurement where background nois
87. ssure at a certain point due to Source 1 is p and the sound pressure at that same point due to Source 2 1s p then the sound pressure level L due to both sources 1s 10 Lp 10log 1 Pi Ps dB 2 po 0 Where po is the reference sound pressure 20 uPa In the case of two equal sources P1 p so that 10 Lp 101og 2 2 Pi 10log ft pi 10log2 20log 3 dB Po Po 3 Thus two sources which by themselves each cause for example L 40 dB at a certain location will cause L 43 dB at that same location when sounded together If two sound sources independently cause sound pressure levels of L 50 dB and L 2 53 dB at a certain point what is the sound pressure level at that point when both sources contribute at the same time 10 pi p L 101og L 20log dB 4 pb Po Lp pP p 10 6 32 107 Pa 5 Lp pal0 22 8 93 10 6 2 2 Pi p Po 16 2 1 6 Sound Fields Sound fields are typically near far free and reverberant field In the near field close to the sound source there can be a large amount of pressure variations from one position to another position Sound pressure level measurements in the near field are therefore forbidden in many applications In the near field the sound field is very complex and complicated which means that the sound pressure distribution and the sound intensity distribution may look completely different It is a question of are they in pha
88. t of sound intensity Measurement with pairs of pressure sensing microphones International Electrotechnical Commission Geneva Switzerland 1993 EN ISO 9614 2 1996 Acoustics Determination of Sound Power Levels of Noise Sources Using Sound Intensity Part 2 Measurement by Scanning International Organization for Standardization 1996 Geneve Switzerland Rossing Thomas D 1990 The Science of Sound Second Edition Addison Wesley Publishing Company 11 12 89 Gade Svend 1982 Sound intensity Part I Theory Br el amp Kjer Technical Review 3 N rum Denmark Technobothnia Research Centre brochure 1998 Second Edition Stencca Bock s Office 90 Appendix A Measurements Results for B amp K OmniPower Job name 0043 S TP Date of measurement February 2009 Ambient temperatur 20 C Ambient pressure 1028 mbar Sound intensity analyzer B amp K Modular Precision Sound Analyzer Type 2260 Analyzing software Sound Intensity Software BZ7205 version 2 1 0 Sound source no Br el amp Kjar Sound Source Type 4224 switched off but in position Sound source no 2 B amp K OmniPower Sound Source Type 4296 Pink noise generatorer Minirator at 4dBu Amplifier B amp K Power Amplifier Type 2716 C Level controls of A and B channel at max Figure A 2 Measurements 91 MS Noise 0043 5 BK2260 BZ7205 2 1 0 Total Sound Power Status Code x xO xO D D X X D Power re 12 120 1
89. t practical purposes type class 1 sound level meter is the most versatile 2 3 Basic Frequency Analysis of Sound 2 3 1 Frequency and Wavelength Different sound sources are producing sound in different frequency ranges Supertanker typically in very low frequencies and a violin in higher frequencies Frequency is measured in Hz indicating the number of oscillations per second for a sound wave see figure 2 11 For example the legs of a standardized tuning fork are vibrating and producing sound with 440 oscillations per seconds Wavelength A m p e a I r LJ r L I n Li n i a b i I i I L I L I F F 1 T 1 J ESE FSS saa ss ee Oe ee ee ee Of MM OE ML GM A a anl Speed of sound in air c 344 at 23 C 20 4 2734 m s Tistemperature in C Figure 2 11 Wavelength and Speed of Sound 5 23 The time period is the time it takes for the sound wave to oscillate from one maximum to another maximum The time period of a 1000 Hz oscillation is 1 ms The distance between two maxima of the sound wave is called one wavelength when looking at oscillations as a function of space At low frequency the wavelength is long and at high frequency the wavelength is short Angular frequency c is used for simplifying mathematical expressions The unit of angular frequency is radians per second This means that one cycle is exactly
90. the Novia University of Applied Sciences All three educational institutions are on the same campus as the Technobothnia research centre Technobothnia s mission statement is to provide a framework for high standard research and education in the field of technology serve as a channel of cooperation between educational institutions enterprises and other research institutes and technology centres offer education as well as research and product development measurement and testing services to the private and the public sector 12 The acoustic section in the Technobothnia research centre s laboratory of building physics provides acoustic measurements and analysis 1 2 The Purpose The purpose of this thesis work was to give myself a general introduction to acoustics and an in deep study of the determination of sound power levels using sound intensity method The determination of sound power levels were to be carried out according to ISO 9614 2 using the Br el amp Kjer Modular Precision Sound Analyzer Type 2260 Investigator The purpose of this thesis work was also to evaluate the fact that sound intensity measurement used for sound power determination is regarded as not sensitive to background noise The influence of background noise from a sound source with known acoustic characteristics on the determination of the sound source under test was to be evaluated by carrying out the test in a reverberant chamber For the tests two sound sources
91. the input channels of the system Mi Calik Job F Res I 1998 Mar 06 10 06 15 00 5 Nu F Eez I 319 3dE E ui ra lzE zk HZ kb AL Figure 3 25 Pressure residual intensity index shown with IEC minimum values 6 56 rr Lu SPL in coupler Intensity level in coupler Residual pressure intensity index Figure 3 26 An example of where sound intensity and sound pressure levels are measured in the coupler while verifying the pressure residual intensity index By subtracting the sound intensity level spectra from the sound pressure level SPL spectra the pressure residual intensity index spectrum is obtained 4 57 4 DETERMINATION OF SOUND POWER 4 1 Introduction to Sound Power Sound power is the rate per unit time at which airborne sound energy is radiated by a source and its unit is Watt W Sound power is a quantity which can not be measured it can only be calculated or determined based upon either sound pressure measurement or sound intensity measurement The sound power levels Ly is defined as W L dB 5 28 0 where the reference sound power Wo is 1 pW The three basic parameters of sound compared 1 Sound Pressure is dependent on the acoustic environment and is a product of the sound source s and the acoustic environment 2 Sound Intensity is dependent on the acoustic impedance of the medium and is a descriptor of the radiation of sound power thr
92. two input signals through filters to an A D converter The application software processes the digital signal and sends output signals to for example the 192 x 128 pixel LCD screen Application software together with appropriate accessories changes the use of the instrument The analyzer can have several software applications installed in it at the same time Once measured data can be stored and transferred onto memory cards for storage or for transfer to computer Data can also be transferred to a computer or printed via the serial interface Measurement data can be post processed with dedicated acoustic software packages Alternatively data can be exported to other standard software packages such as word processors and spreadsheets The analyzer is operated by pushing either hard or soft keys The hard keys are located on the front panel and are all identified by icons and they have fixed application software independent functions The soft keys are to the right of the analyzers display The functions of these context sensitive push keys are identified by the soft key menu on the right hand side of the display The analyzer has a menu based user interface The analyzer has two drives one for an internal disk and one for an external memory card The internal disk is organized in directories and subdirectories Some of these directories are for storing measurement data others for storing measurement set ups while others are for st
93. ty Sound intensity also enables segmentation of a sound source see figure 4 3 3 2 3 Building Acoustics Applications in building acoustics are for example sound reduction index leakage detection and sound absorption detection For measurement of sound reduction index there is typically a transmitter room and a receiver room The traditional way of measuring sound reduction index is with sound pressure level measurements in two rooms When using sound intensity for sound reduction index the sound pressure level is measured in the transmission room where the sound field 1s made as diffuse as possible In the receiving room the sound intensity is used to measure the energy flow at various positions for mapping of the transmission Sound intensity in the transmitter room can be calculated from a formula and is also called the one sided intensity in a diffuse sound field The sound reduction index is then the difference between the calculated one sided intensity and the measured transmitted intensity 3 3 Estimation of Sound Intensity 3 3 1 The Direct Method A sound intensity probe consists of two microphones and based upon the measurements of the sound field of two closely spaced positions in space the 34 sound intensity can be estimated Sound intensity estimation is essentially a phase estimation of the sound field Sound intensity 7 is defined as the time averaged rate of energy flow per unit area A more mathematicall
94. uch as transient noise change in background levels change in acoustic environment or error due to insufficient scanning scan speed or scan path variations 69 The repeatability of the measurement for each segment and for each frequency band is checked as 1 s dB 9 36 VVhere ku and m are partial sound power levels from two orthogonal scans of segment i and the standard deviation s given in table 4 1 Table 4 1 Uncertainty in the determination of sound power levels 9 Octave band centre One third octave band Standard deviations s frequencies centre frequencies Engineering grade 2 Survey grade 3 Hz Hz dB 63 to 125 50 to 160 250 200 to 315 500 to 4 000 400 to 5 000 6 300 A vveighted 1 NOTE The stated uncertainty of the A vveighted estimate does not apply if the total A vveighted power in the one third octave bands outside the range 400 Hz to 5 000 Hz exceeds the total vvithin this range individual band uncertainties then appiy 1 63 Hz to 4 kHz or 50 Hz to 6 3 kHz 2 The true value of the A vveighted sound power level is expected with a certainty of 95 to be in the range of 3dB about the measured value 4 4 4 Adequacy of The Measurement Equipment This criterion is to determine the adequacy of the measurement equipment in relation to the nature of the sound field being measured The criterion detects error due to instrument phase mismatch The surface pressure intensity indicator
95. und and noise are introduced and studied The practical usage of the sound intensity method is introduced by carrying out measurements in a reverberant chamber on two sound sources each with different directivity pattern By this arrangement in the given environment the influence of background noise on the determination of sound power levels of noise sources by using the sound intensity method could be studied and evaluated As a part of this thesis a software application for measurement data analysis has been developed The application was created with the C language in a programming workbench named MicroSoft Visual Studio 2008 In order for the software to be able to read from the binary measurement data files relevant data properties were obtained by mapping these files by hex editing In the end proposals for future investigation of the sound intensity method and for further development of computer software applications for acoustic analysis have been given Keywords Sound pressure sound intensity sound power VAASAN AMMATTIKORKEAKOULU UNIVERSITY OF APPLIED SCIENCES Rakentamisen koulutusohyelma ylempi AMK tutkinto TIIVISTELM Tekij Michael S derback Opinn ytety n nimi A nitehotasofen m ritt minen niintensiteettimenetelm ll Vuosi 2009 Kieli Englantia Sivum r 89 4 liitteet Ohjaaja Tapani Hahtokari Opinn ytety ss tutkitaan 4nitehotasojen m ritt minen nlintensiteetti menete
96. und field with a high PI index in the chamber Due to the high PI index an averaging time not less than two minutes is required B 4 Figure 3 24 Verifying a pressure residual intensity index 55 When a sound wave propagates at an 90 angle towards the probe axis differences in phase responses of the microphones cause a phase difference between the microphone signals and there appears to be an acoustic flow of energy along the probe axis Because each microphones is exposed to equal sound pressure any intensity detected is residual intensity The residual intensity spectrum is not tied to variations in measured sound pressure level The pressure residual intensity index is a constant and is obtained by subtracting the detected residual intensity spectrum from the measured mean sound pressure spectrum see figure 3 26 An example of pressure residual intensity index spectra after phase calibration shown with minimum values required by IEC 1043 as seen on screen of a B amp K 2260 Investigator sound intensity analyzer 1s indicated in figure 3 25 The error in gain adjustment factor Lg sain due to phase mismatch between the two measurement channels of a sound intensity analyzer 1s given as sin d sin Q 4 Where n is the measured phase and gd is the phase mismatch n un dB 5 27 The largest phase mismatch of a sound intensity analyzer is between the microphones followed by the pre amplifiers and
97. und power is for noise rating of machines For comparison of how noisy various machines are the only way to compare them is to determine the sound power The unit for sound power is Watt W and is telling exactly how noisy the machine 1s Table 2 1 Basic quantities of sound Quantity Sound pressure Receiver Measured 1 Evaluation of the harmfulness Pa and annoyance of noise sources Sound intensity Path Measured Location and rating of noise W m sources Source Calculated Noise rating of machines W 2 1 3 Sound Pressure Sound pressure can be generated for example with a tuning fork The vibration of the tuning forks two legs will activate the air molecule to vibrate and this vibration are then transmitted through the air to the human hearing system A handclap produces a disturbance on the air molecules and this disturbance is then travelling with the speed of sound to the receiver s ears Waves can transport energy from one place to another through a medium but the medium itself is not transported A disturbance is passed along from point to point as the wave propagates A sound wave is a change in pressure and velocity Throwing a stone into the water will cause a disturbance in the water This disturbance will then travel away from the disturbance point in the shape of waves The waves in the water are getting smaller and smaller as they travel away from the disturbance point The handclap will produce a certain sound
98. y definition is that the sound intensity vector equals the time averaged product of the instantaneous pressure and the corresponding instantaneous particle velocity at the same position 5 I pr ur 9 Where p t is the instantaneous pressure u t is the particle velocity and the time averaging is indicated with a bar The particle velocity describes the actual motion of air particles as they oscillate around their equilibrium or rest position The higher sound pressure there is the higher the particle velocity becomes In acoustics an equivalent with the force in Newton s 2 law is the pressure gradient There is no force on an air molecule if there is no pressure gradient With the pressure gradient means that there is a difference in pressure to the sides of a particular point If there was the same pressure in both points there will be no pressure gradient And if the pressure gradient is in a direction from low to high pressure then the force would be in the opposite direction and therefore negative This explains the minus sign in formula 10 The sign due to the direction of the pressure gradient 1s given by the definition of the intensity probe orientation A derivate of Newton s 2 law also called Euler s Relation is used to estimate the average particle velocity u by measuring the pressure gradient p 1 a grad p 5 10 p Where a is acceleration and p is the density of the air In one directio
99. y itself do not perform any calculations it only displays the measurement data as it is stored in the analyzer by the sound intensity software BZ 7205 FE MS Noise 0045 5TP BK2260 B27205 210 Ioj gt File dl di di Gl al d 4 3 20013 18 58 23 4 3 2003 18 56 17 4 3 2003 18 51 15 4 3 20013 18 45 31 4 3 2009 18 46 54 4 3 2009 18 48 44 4 3 2003 15 43 43 43 2005 18 41 53 4 3 2003 15 40 34 c Bes Figure 5 14 The user interface of the application 82 The user interface of the application is simple and straight forward see figure 5 14 Measurement data is inserted into the application by a click of a button in the tool bar which will then activate a file dialog where the project is selected The name of each subsurface in the project are displayed in a tree structure By a click on one the nodes in the tree view segments in the selected subsurface are displayed in the list view where various segments are indicated by their corresponding data In the list view a segment can be selected to display its measurement data as a spectrum in a new window The application also enables exporting of measurement data into files that can be edited with a simple text editor An example of a graph produced by the application is shown in figure 5 15 MS Noise 0045 STP BK2260 BZ7205 2 1 0 Total PI index m Status Code xDE xDE xD D x x D PRI index 6 57 9 78 15 91 12 42 32 15 25 06 26 43 22 96

DETERMINATION OF SOUND POWER LEVELS USING

Contents

Download Pdf Manuals

Related Search

Related Contents