Nor-121 Extension Manual
Contents
1. Sweep Du ration It offers the choice between three values named Short Medium and Long Table below contains nformation about duration per bandwidth inverse of sweep speed and the total duration for a sweep covering he frequency range of third octave bands with midband frequencies from 50 Hz to 5 kHz and octave bands with midband frequencies from 63 Hz to 4 KHz Noted that this parameter has no influence on the sweep duration for oc tave band measurement The lengths of the measured impulse responses IR exceed 10 s Block diagram of implementation in Nor121 is presented in figure 6 It is important to notice that de spite the use of spectral division a linear deconvolution has been implemented This is important as it enables removal of distortion components from the measured im pulse response Setting up for Swept Sine measurement Our traditional motto Measurements made easy has been followed during implementation of the Swept Sine method User already experienced with classical meas urement will find it easy to start using tje Swept Sine technique as the structure of user interface and menus is unchanged Practically there is just a need to get used to a reduced set of parameters as Swept Sine technique does some settings automatically Predefined standard setups allow the user to completely set up the instrument for measurement by loading one of the standard setups from setup menu The setup m2. Extension to INSTRUCTION MANUAL SOFTWARE VERSION 4 0 Introducing Swept Sine measurement technique and other new features nor121 SOUNDANALYSER e NN Norsonic nor121 Introducing Swept Sine measurement technique Nor121 Extension to instruction Manual March 2006 Edition Software version 4 0 Introducing SweptSine measurement technique Im121_5Ed1ROEng Norsonic is a registered trademark of Norsonic AS All other brand or product names are trademarks or registered trademarks of their respective companies Every effort has been made to supply complete and accurate in formation However Norsonic AS assumes no responsibility for the use of nor for the consequential damages of the use of this information and or the instrumentation described herein Further more Norsonic AS assumes no responsibility for any infringement of the intellectual property rights of third parties wherever appli cable which would result from such use Norsonic AS reserves the right to amend any of the information given in this manual in order to take account of new develop ments If you wish to communicate with us please feel welcome Our address is Norsonic AS P O Box 24 N 3421 Lierskogen Norway Find us on the web www norsonic com Tel 47 3285 8900 Fax 47 3285 2208 e mail info norsonic no Copyright Norsonic AS 2006 All rights reserved Introduction General Features Quicker instrument switch o
3. Figure 4 Resulting decay curve gets distorted Instruction manual page 17 Nor121 The reason why Swept Sine technique is less sensi tive to time variance is that response to each frequency appears at exactly one instance of time This means that deconvolution only shifts and aligns responses at differ ent frequencies i e there is no summation of several re sponses at same frequency which might be time scaled differently Implementation in Nor121 The combination of a powerful digital signal processing hardware components in the hand held sound analyser Nor121 and carefully crafted algorithms enabled imple mentation of such an advantageous technique as Swept Sine Furthermore every effort has been made to fit the Swept Sine algorithm into existing user interface of Nor121 s building acoustics measurement mode This requires the minimal effort for an user who is already ex perienced with classical measurements to become effec tive with Swept Sine technique As a matter of fact some functionality became even easier to use The technique is implemented in accordance with in ternational standard ISO 18233 Acoustics Application of new measurement methods in building acoustics a Excitation signal A sinusoidal signal is used for the excitation whose fre quency increases exponentially with time This means that the time to sweep through each fractional octave band is constant The energy per fra
4. Pre set pure tone Standards In the Set up Standards menu there is now a wider range of preset Standards for pure tone analysis This includes all the above mentioned Standards plus the special Joint Nordic Method JNM for wind turbines It is also possible to conduct FFT analysis for the use with the DIN 45 681 standard from 2005 FFT data are stored in the instrument and can be later transfered to a PC where the pure tone analysis can be performed using a special macro for Microsoft Excel Another possibility for performing a pure tone analysis is to make recordings of the sound in the form of 16 bit wav files These files can then be transfered to a PC and pure tone analysis can be done using the NorReview post processing software Measurement mode cano Setup tupe Select setup DIN 45 681 Entwurf 06 01 FFT f r DIN 45681 Ausgabe Marz 2995 ISONDIS 1996 2 2005 JNM for wind turbines Joint Nordic Method Version 2 Save current setup as default Save default setup at switchoff Show this menu at startup In the Set up Standards menu the user may choose among preset set ups in accordance with misc Standard proposals Instruction manual pageo Nor121 Building acoustic mode BA TETE 13 1U 2964 95 25 15 15 44 BGN corr Off 1 2 dB ic corr 5 9 dB M orf Spectrum Adaptation Terms The Average Rating Report is now also calculating the Spectrum Adaptation Terms as given in the ISO 717 Stan
5. broacoustics acoustical profiling sonar and many other areas are closely linked to measurement of impulse re sponses Even high end fields like virtual acoustic which is a part of virtual reality strongly rely on convolution with impulse responses with high signal to noise ratio page 14 Impulse response is also indirectly involved in the classical method with noise excitation as the measured response to such excitation can be described as a con volution between the excitation signal and the impulse re sponse of the room However in such case the response to noise excitation is recorded directly and information about the impulse response is normally not known Several new measurement methods may be applied to obtain the impulse response as described in stand ard ISO 18233 One of those methods based on use of maximum length sequences MLS became very popular in the 1980s due to the ability to use computer features available at that time However MLS based techniques have some undesirable properties such as vulnerability to distortion and time variance Those undesired properties motivated further research which resulted in even more advanced method known as Swept Sine The Swept Sine is a measurement method which uses an excitation signal of a sinusoidal shape whose frequen cy increases from low to high frequencies The response to this excitation is recorded by the analyser and the im pulse response between source and rec
6. dB sol 4 30 10 10 0 00 00 0 00 02 850 x 9 68 00 Rep Reverb KEEK If several measurements are to be made it is important to press STOP after saving each measurement This will put the instrument back in READY state before the next measurement Otherwise every measurement will overwrite the previous one Transferring measured IR from the Instrument to a PC The measured impulse response can be transferred to a PC either by moving the PC card or by a cable connetion between the instrument and the PC In either case use the NorXfer software for transferring the data If PC card is used do the following 7 Take the PC card out and put it in the PC running NorXfer software page 32 8 Click on the connect icon and chose Nor121 as an Instrument and Local as a Connection 9 Copy the measurement to My Measurements If cable is used do the following 7 Connect the parallel cable to the instrument and set the instrument in Transfer mode by choosing the NorXfer button from the MEMORY menu 8 Click on the connect icon in the NorXfer and chose Nor121 as an Instrument and LPT as a Connection 9 Copy the measurement to My Measurements Subdirectory Ch1 will contain file ir1 wav and subdirectory ch2 will contain file ir2 wav NorXfer does not show the 24 bit ndf files although it transfers them to My Measure ments User is advised to access these files from Win dows Explorer or some other
7. Details group Figure 9 Reverberation setup with parameters explained page 23 Nor121 Reverberation setup Reverberation setup menu is presented in figure 9 This menu contains three submenus depending on excitation type Each of submenus contains ia details switch When Details is set to On additional parameters are displayed Although intended for advanced users this menu is quite simple to use If parameter Details is set to Off the details will not be displayed see figure 9 Parameters from De tails group are described in the figure The use of other parameters similar to classical measurement Measurement control Rating setup Rating formula Normalized CA Ao gt Measurement type Resolution dB BGN correction Room description Height Width Length 1 68 m 2 88 Im Loo Jm Area 5 1 98 m2 Volume cur 1 88 ms Figure 10 Rating setup part of Measurement control menu page 24 Rating and BGN setup Please note that parameter BGN correction from Rating setup menu displayed in figure 10 is automatically set to On when any of SweptSine related setups is loaded This is because measured levels are automatically cor rected for background noise with Swept Sine technique Even if user changes the value of this parameter to Off the correction will still be applied since it is a part of the method For the same reason there are no accessible param eters from Measurement control Backgrou
8. SMS messages and give the results from both noise and weather measurements How ever it is now possible to additionally read out the actual date and time of the instrument s clock simply by sending the SMS command CLOCK to the Nor121 Instruction manual page 3 Nor121 O dB reference selection The Nor121 has had the possibility to choose different O dB reference values in order to make measurement in dB relative to either sound or vibration units The choices are 2 0e 5 sound 1 0e 6 acceleration and 5 0e 8 sound particle velocity The normal value is the 2 0e 5 i e 20 micro Pascal When another reference is chosen the dB labels are displayed with inverse video in order to avoid the misunderstandings with normal dB values In the new version the selection of the O dB refer ence value is moved from the Range Calib menu to a separate Range 0 dB reference menu in order to avoid possible misunderstanding or wrong selection by a user that is working under pressure Primary ch 1 Secondary ch 2 6 dB The 0 dB reference selection is now moved to a seperate sub menu page 4 Improved modem connections The Nor121 may be used for long term noise monitoring in a system based on modem connection to the NorXfer version 4 2 software Earlier versions of the Nor121 occa sionally had some problems with the output buffer when the modem connection was broken This has now been improv
9. mono samples stored as Lit tle Endian byte order Intel type LSB first Header and trailer usually contain data about format such as sampling rate bits per samples etc The fact that ndf file contains no header and trailer means that it is us ers responsibility to specify file s format while opening a file if 24 bit data are required Making IR Measurements a cookbook 1 Use the Setup menu to load standard setup Impulse Response Measurement SweptSine By default both cahnnels will be enabled If a one channel measure ment is needed the other channel can be turned off using Meas Control menu 2 Press START button to make a measurement 3 After the measurement is finished status becomes ENDED the echogram of measured impulse re sponse will be shown on display Figure 16 Inspect the echogram to confirm that the measured results are correct 4 Copy the measured impulse responses to Average registry by pressing Copy Avrg soft key We will not perform any averaging here but this step is necessary in order to save the impulse responses to memory 5 Save the measurement on the PC card or instruments hard disk using Memory menu 6 Press STOP button to put instrument back to READY state Instruction manual page 31 Nor121 Figure 16 Echogram of an impulse response measured in Nor121 BA 13 1U ME 2666 62 66 16 43 46 Last T Tsa a 1 Lee 6 6 dB N 1 f SkHz 1 3 10
10. robust for environmental influences such as temperature and wind than the MLS technique Hence Norsonic has taken a new step forward by imple menting a new improved feature Both the MLS and Swept Sine measurement tech nique for use on building acoustics application are de scribed in ISO 18233 This new standard currently at the FDIS stage will probably be turned into a legal standard early 2006 which means that sound insulation tests may be performed using this new technique as an alternative to the traditional ISO 140 technique The Swept Sine measurement technique is available for all Nor121 units equipped with version 4 x and the new option 14 SweptSine technique Instruction manual Swept Sine measurement technique Swept Sine measurement technique is an optional ex tension for the Nor121 Furthermore it is an extension of building acoustic measurement mode of Nor121 Swept Sine based measurements on Nor121 can be performed with two aims in mind The first one is to obtain results used in the building acoustics such as airborne sound insulation and reverberation time Application of Swept Sine technique to this type of measurement is de scribed in the first part of this chapter As a contrast to this highly specialized task the second type of measurement that Swept Sine provides is much more general It is dual or single channel measurement of broadband impulse response and is described in second part of this chap
11. s 0 20 0 14 0 14 0 10 0 09 0 08 0 07 0 06 0 05 0 04 0 O O Octave bands RTmin s Ong O O RTmax s 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 RTmax s 4 4 4 4 i O1 O1 O1 O1 O1 O1 O1 Broadband impulse response measurements using Swept Sine Broadband impulse response measurements are used in every field of acoustics hence it is a great advantage to be able to perform them using a sound analyser Espe cially the ability to use two channels and simultaneously perform two measurements completely independent of each other opens a broad range of applications Some of these might be measurements of microphone character istics analog equipment testing or even measurements of head related transfer functions HRTFs Measured impulse response of the room can be used to calculate room acoustic parameters in accordance with 1503382 Implementation Measurement is always performed using a sampling fre quency of 48 kHz Each measured impulse response is stored in two files i e two copies are made each with different resolution One 16 bit wav file and one 24 bit ndf data files are stored and can be transferred to PC for further processing Note that ndf is a Norsonic data file used internally in the instrument The format used for data storage in an ndf file is raw data format without any header or trailer with 24 bit
12. 1 is used to make true audio recordings of the measured sound it is often necessary for them to be replayed as part of either mediation or legal proceedings n these circumstances it is important to be able to adjust he speaker or headset to the correct level in order for a non technical audience to appreciate the original acous tic environment The new version of Nor121 has therefore he possibility to record a reference tone at a pre defined level that can be used as a reference to adjust the head set or speaker level The reference tone can be recorded at the start of the measurement and additionally at any time during that particular measurement The maximum level of the refer ence tone is 1 V rms but may be set to a lower level in the RangeCal Ref Tone menu The length of the reference tone after activation may also be selected In the same menu the type of signal used for the reference tone may page 8 be specified in addition to a single sine wave at a user specified frequency White Pink or Band filtered random noise are optionally available These latter selections will be helpful where standing waves in the replay environment make it difficult to set the correct level These reference noise feature requires that the Option 3A is available Prior to the use of the reference tone the AC output ch 2 in the Set up I O menu must be selected as Signal gen The reference tone is selected via the Act
13. Criterion example 100 Pz pis CN NC 50 NR 48 RC 46H 80 70 RES Rs SS en 3 A NR7 60 REF aca oe NC 69 NRG 50 RE66 NC 50 pya 50 40 Ea NR 30 SES 20 ke m E oe 16 31 5 63 125 250 500 1000 2000 4000 The Noise Criterion evaluation search the highest 1 1 octave bana that fits below the applicable noise criterion lines The evaluation is performed in 1 dB steps 1 3 octave spectrums are re calculated into 1 1 octave spectrums prior to evaluation page 7 Nor121 Marker for external DC failure When the Nor121 instrument is used for long term noise monitoring powered by an external 12V supply with the internal battery as backup it may be useful to know if the external power source has powered down at any time during a running measurement By entering the name Ext DC fail for marker no 9 the Nor121 will automatically activate this marker each time the external DC is absent When the external power for the Nor121 is restored again the marker will be turned off Hence at the end of a long term measurement the operator may view the absent time periods as a marker along the level vs time display or at the PC using the NorReview post processing soft ware This feature will indicate if the instrument was for example powered down and moved to another location during the measurement Reference tone on recordings When the Nor12
14. Hz 315 Hz 800 2 kHz Gain dB Frequency page 11 Nor121 Equalizer can be used with both one third octave and one octave band resolution When used with octave reso lution gain is applied to all three one third octave bands constituting that octave Equalizer gains are saved at instrument switch off Using EQ with the pink noise excitation a cookbook 1 Place the speaker and microphone in the sending room and connect everything as desired 2 Open generator menu 3 Press Gen On Off software key again in order to start the generator 4 The measured levels in each band are shown on dis play both as numerical values and graphically 5 Use inner LEFT and RIGHT arrow keys to move cursor between frequency bands Current frequency can be seen above the graphical display e g f 1 kHz in the figure 6 While standing on the Equalizer parameter use in ner UP or DOWN keys to toggle between EQ On or Off 7 While standing on the Gain at this frequency pa rameter use inner UP or DOWN keys to adjust gain at the frequency of the cursor Gain is adjusted in 1 dB steps 8 Move the cursor to another frequency and if necessary repeat the step 7 page 12 Using EQ with the Swept Sine excitation a cookbook Basically the same as with the pink noise excitation ex cept that the generator can not be started separately and therefore EQ cannot be adjusted while playing excitat
15. akes it possible to measure signals whose level is lower then the level of background noise i e measure signals in negative SNR environments However this benefit necessiate certain measures to be taken User should be aware that by sup pressing noise not only the useful signal becomes avail able but also all other signal components which would usually be buried in noise One such undesired signal component is the consequence of crosstalk in the instru ment itself When measuring high insulation values signal leve recorded in the receiving room may be considerably low er than the levels in sending room The following meas ures might be used to reduce the influence of crosstalk in such a measurement n 1 Turn off HP filters High pass filters in Range menu should be turned off Note that when HP filters are turned off the instrument becomes more sensitive for overload by low frequencies If an overload occurs the measurement should be restarted and not copied to the average register 2 Use a low generator output level and amplify the sig nal in the power amplifier This will reduce the risk for crosstalk from the generator output to the signal out put 3 For level difference values higher than 80 dB the use of microphone front end Nor336 for microphone in the receiving room is recommended This preamplifier provides the gain up to 40 dB The calibration should be performed with both microphone and front end conne
16. ame type of measurement as with setup ISO 140 4 Airborne sound reduction index just with the Swept Sine method Swept Sine Airborne standard level diff sets the instrument to perform the measurement in accordance to International Standard ISO 140 4 This is the same type of measurement as with setup ISO 140 4 Airborne standard level diff just with the Swept Sine method Swept Sine Fa ade loudsp sound red inx sets the instrument to perform the measurement in accordance to Interna tional Standard ISO 140 5 This is the same type of measurement as with setup ISO 140 5 Fa ade loudsp sound red inx just with the Swept Sine method Setup St andards 7 Swept Sine Fa ade loudsp std level diff sets the in strument to perform the measurement in accordance to Measurement mode International Standard 1SO 140 5 This is the same type Setup tupe of measurement as with setup ISO 140 5 Fa ade loudsp Select setup std level diff just with the Swept Sine method 1S0 146 7 Impact standardized Impulse Response Measurement SWEPT SINE Reverberat ion time Sweep SWEPT SINE Airborne normalized level dif SWEPT SINE Airborne sound reduction inx SUEPT SINE Airborne standard Level diff Tip You are not confined to stick with the selected setup It may be modified as much as you like and even stored as one of your own setups for later use As default the setup menu appears at start up If you don t want this option select No i
17. control Reverberat ion setup Fullscale Channel 1 dE Channel 2 Fep Excitation type Details Min dist to noise floor 19 8 Jap RT parameters Primary Secondary l Fra range 53 JE Hz Bandwidth Oct Broadband Impulse ResponselOff A Upper range for measurement TONMO O B One or both channels can be selected for measurement just as with classical measurement C Excitation type can only be SweptSine Only this exitation is available in Swept Sine standard setup D Details parameter dispalys group of parameters marked with on the screenshot E Primary and secondary level range reverberation times Secondary is measurement if it is not possible to measure primary RT F Defines the distance from estimated level of noise floor to the lower end of the level range from which the reverbera tion times are calculated This parameter directly influences the condition for reliability of results display of sign with the value Meaning is the same as in classical measurement G These parameters define frequency resolution and range for the measurement valid parameters for Swept Sine measurements are in range 63 4k Hz for Bandwidth 1 1Oct and 50 5k Hz for Bandwidth 1 3 Oct H In building acoustic measurement this parameter should be kept Off Turning it on activates measurement of broad band impulse response which is described in second part of this chapter Parameters which constitute the
18. cted The care must be taken that the full scale deflection should be diminished for the amount of gain used on the front end For example if the fullscale has value of 90 dB and front end gain is 30 dB the actual fullscale value is 90 30 60 dB Any level higher than 60 dB will produce an overload condition Instruction manual page 29 Nor121 Technical specifications Bandwidth 1 3 Octaves Swee Durati p duration on per bandwidth Acquisition duration Start frequency Stop Swee Durat requency p duration ion per bandwidth Acquisition duration Short 60 9 s 2 9 s one third octave band 68 92 s 38 13 Hz 5956 5 Hz Medium 336 s 16 s one third octave band 338 92 s Start frequency 44 44 Hz Stop frequency 5956 5 Hz Sweep duration Long 672 s Duration per bandwidth 32 s one third octave band Acqui sition duration Start frequency Stop Leng requency h of impulse response 674 92 s 44 59 Hz 5956 5 Hz 10 72 s Bandwidth 1 1 Octaves Swee Durat Acqui p duration ion per bandwidth isition duration Start frequency Stop frequency Length of impulse response page 30 14 7 S 2 1s octave 23 58 33 19 Hz 5956 5 Hz 10 72 s One third octave bands Frequency Hz 50 63 80 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000 Frequency Hz 63 125 250 500 1000 2000 4000 RTmin
19. ctional octave band will therefore be the same and the spectrum of the excitation signal will have a pink character The sweep is started at or below the lowest band edge frequency of the lowest fractional octave band to be measured and continues upwards to at least the upper band edge frequency of the highest fractional octave band to be measured Fade in and fade out are applied at beginning and end of the excitation signal to protect loudspeaker from transients The acquisition time for recording the sweep response has to be longer than the sweep itself to collect the reverberated sound This should be thought of as if the generator delivers silence for a period of time after the sweep is finished Figure 5 a Time frequency representation of SweptSine excitation b T F representation of response to SweptSine page 18 Measurement of IRs deconvolution Normally in building and room acoustic measurements the aimis to obtain results in octave or third octave bands This is achieved by one continuous sweep through all of the bands The energy in eacg band of interest is sepa rated from the energy in other bands by filtering In the Swept Sine implementation filtering is done by means of time and frequency domain windowing For this purpose hybrid windows were developed Those windows had to satisfy a number of constraints Most important they sat isfy the frequency response requirements for a class 1 fractional octave band fil
20. d noise level Note that Swept Sine method compensates for the apparent background noise automatically Therefore do not attempt to make any further compensation for the noise yourself as you are used to do with con ventional measurements What does the measured BGN and SNR mean It is important to note that signal to noise ratio SNR ob tained with Swept Sine measurement is not the same as SNR obtained with classical measurement SNR in clas sical measurement is ratio between excitation signal and background noise in the receiving room at microphone position What is measured with the Swept Sine tech nique is the SNR with processing gain included It is SNR in the measured impulse response which benefits from the noise suppressing features of the technique It is the term which determines the reliability of the results What does the displayed with level values mean User familiar with measurements of reverberation time in classical measurement might note that reverberation time results can be marked with which means that results are unreliable This can be because reverberation time is too short or background noise level is too high Please refer to main user manual for details on this However no such indication of reliability was used with level results in classical measurement A novelty in Swept Sine based measurements is that 2 symbol can be displayed with level results However note that it does no
21. dard These terms are calculated both for airborne and impact sound insulation measurements When an ex tended frequency range is used for the measurement the corresponding terms are calculated as well All the terms are displayed in a sub menu that is ac cessed from the leftmost soft key below the display when the Average Rating Report is displayed in either graphical or numerical format Rep Rating C Corr Topscale t Topscale Zoom Compress The Spectrum Adaptation Terms are calculated and presented by a push on the C 50 3150 leftmost softkey displayed in Ctr 50 3150 the Average Rating Report C 58 5808 gt Ctr 50 5000 C 199 5882 Ctr lt 100 5000 gt Ci Ci 50 2500 page 10 General Equalizer is a new feature introduced with software ver sion 4 0 It is used with signal generator in building acous tics mode to provide the user with even better control over the excitation signal The international standard ISO 140 4 specifies that the excitation signal shall be such that the level difference between adjacent one third octave bands shall not be greater then 6 dB In other words it is required that the frequency characteristic of excitation signal is sufficiently flat Rooms and loudspeakers hinder us from having this requirement always fulfilled hence we need an equalizer to control the difference This provides the possibility to fine tune the spectrum of excitation signal to every send ing
22. eas urement There is no need for compensation as this is performed automatically An extra display has been added to the Level re port This display which is presented in figure 11 con tains measured levels and measured background noise BGN To display the measurement results from this dis play in numeric form as shown in figure 12 simply press the Num key Press Num key again to return to graphi cal mode Pressing Display key instead will take you to the next display in the sequence The whole sequence of available displays is presented in figure 13 eq eq 8GHz 1 3 fvraSLast Topscale t Topscale Pree View Last Instruction manual Figure 11 Levels in receiving room and measured background noise BGN levels are now displayed for LAST register ERMILOADED 9686 13 10 TE 2005 03 03 12 54 51 Fra ZNetw 173 S Hz 63Hz 8 Hz 1 6Hz 125Hz 16 Hz 2 GHz 25 Hz 315Hz 4GGHz S GHz 63GHz 8QGHz 1kHz 1 25kHz 1 6kHz 6 ol wl 8 12 6 4 2 7 4 6 1 0 8 4 4 7 4 1 1 8 2 8 2 1 AvrasLast ChS lt gt ChR lesr O View Last Figure 12 Channel 2 BGN and SNR as numerical table page 25 Nor121 PELLETER FETES POPDCTPENCE RTS LEVELS IN SENDING AND RECEIVING ROOMS LEVEL IN RECEIVING ROOM AND BACKGROUND NOISE LEVEL LLLLELELELEEE POPOPI PEU EE Mon auRo nas Figure 13 Sequence of displays in Level report page 26 Tip Do not correct for the backgroun
23. ed without any pre weighting left or with the new A pre weighting feature right page 5 Nor121 Noise criterions NC NR RC Many applications particularly in the heating amp ventilation industry require the use of noise rating criteria that com pare the frequency spectra of the measured noise levels against reference curves Version 4 0 of Nor121 provides the standardized NC NR and RC noise rating criteria and hence any spectrum measured in the Environmen tal mode may therefore be evaluated using these criteria This new feature is available from a soft key displayed below all the level vs frequency numerical tables with the results displayed in a dialogue box These noise criteria normally require a 1 1 octave fre quency spectrum however if a 1 3 octave spectrum is selected the Nor121 will automatically re calculate it into an appropriate 1 1 octave spectrum prior to noise crite rion evaluation page 6 STOPPED 12 80 RUB 2004 12 16 13 00 10 Ega Enu E oie Global Profile Profile Fra ZNetw 173 1 2SkHz 2kHz 2 5kHz 3 15kHz 4kHZ SkHz 6 3kHz 8kHz 16kHz 12 5kHz 16kHz 20kHz A netw C netw F netw 5 4 1 9 2 3 4 2 Criterion TT The Noise Criterions evaluation are calculated and presented as a table by a push on the Criterion softkey displayed with the numerical frequency table of the selected report 23 8 de 22 48 dB 66 64 dBoRo Instruction manual Noise
24. ed and the stability of such remote connections is much better with the version 4 0 internal software New weather station connection The original version of the Nor121 only connected to the Reinhard weather station type MWS9 The new version of internal Nor121 software now additionally communicates with the new model MWS9 5 and MWS5 Measurement control Weather data Supported weather station Reinhardt MWS9 MWS9 5 Resolution ia minutes Temperature C Pressure kPa bac pa f Humidity Wet ground 8 13 Rain Cmm Peak wind speed m s OFF Average wind speed m s Wind direction Reset hour for rain sensor Meters above sea level ojo gt Pre weighting of spectra The graphical and numerical displays of frequency spec tra within all the reports in the Environmental mode may now be viewed via a selectable A weighting filter This A preweighting feature is turned On Off in the Set up Dis play menu 12 7V 2004 12 16 13 02 31 Es eq f 256Hz 100 1 1 dB ACF Chi Global a TS es our Instruction manual Environmental mode The presence of A pre weighting is clearly indicated in the various displays as all values are followed by a dB A indication instead of the normal dB label 12 80 2664 12 16 13 62 56 Lrea f 25 Hz 1 71 100 de Meme A Done roues All frequency spectrum displays in the Environmental mode may be view
25. ediately The consequence however is that the instrument may not start up again with the same set up that it had just before switch off Measurement mode LUTTE Setup type Standard setup Select t DIN 45 681 Entwurf 06 01 FFT f r DIN 45681 Ausgabe M rz 2005 ISONDIS 1996 2 2005 JNM for wind turbines Joint Nordic Method Version 2 Save current setup as default Save default setup at switchoff Show this menu at startup In the Set up Standards menu the user may save the cur rent set up as a future default set up as well as turn off the automatic save of current set up at instrument switch off General Features Save a new default set up The Nor121 contains a default set up that will always be loaded by turning the instrument on until a new Standard set up is chosen The operator now has a choice to save any current setting of the instrument as the future default set up just by selecting the field named Save current set up as default in the Set up menu and then pressing the Enter key If the instrument is later turned off with the Save default set up at switch off flag set to No see the new feature described above then the instrument will on subsequent switch on start with exact the set up previ ously chosen as default The Save current set up as default feature may also be activated using the remote control command SDS SMS command for date and time The Nor121 will respond to
26. eiver position is extracted by the use of a mathematical technique known as deconvolution Benefits of using Swept Sine The international standard ISO 18233 Acoustics Appli cation of new measurement methods in building acous tics focuses on use of MLS and Swept Sine techniques Both of them provide the possibility to conduct measure ments in high background noise which is the huge ad vantage over classical method MLS technique achieves this by synchronous averaging which is very effective against random noise but unfortunately increases sensi tivity to distortion and time variance This limits the use of MLS Swept Sine is much more resistant to this and it has reduced sensitivity to time variance extended signal to noise ratio SNR and high distortion immunity The transmission of sound within a room as well as the transmission of sound between rooms may normally be regarded as an approximation to a linear and time invariant system However if time variance increases as a result of temperature changes and air movement this assump tion ceases to hold This can result in distorted shape of impulse response and decreased SNR Measurements with sweeps are less vulnerable to those unwanted effects of time variance than measurements with MLS In outdoor measurements and under windy weather conditions where air movement is expressed sweeps are sometimes the only viable option when measuring impulse responses The measured impulse
27. enu may appear at start up by default The menu is also accessible by pressing the Setup key However if it is necessary to perform measurements which are different from the standard setup parameters can be changed from measurement control menu Ac cessed by pressing the Meas Control key The menu is divided in several subparts each of which is accessible via a corresponding soft key Sweep Duration Total Duration s Duration per bandwith s octave Total Duration s Duration per bandwidth s one third octave sne e a o page 20 Instruction manual The instrument is equipped with 7 standard setups for Swept Sine measurements al Impulse Response Measurement Swept Sine is used for broadband impulse response measurements which are closely described in second part of this chapter Reverberation time Sweep is the setup intended for general measurements of reverberation time Both channels are used for measurements in one third octave bands with midband frequencies from 50 Hz to 5 kHz Swept Sine Airborne normalized level dif sets the instrument to perform the measurement in accordance to International Standard ISO 140 4 This is the same type of measurement as with setup ISO 140 4 Airborne normalized level diff just with the Swept Sine method Swept Sine Airborne sound reduction index sets the instrument to perform the measurement in accordance to Interna tional Standard ISO 140 4 This is the s
28. ff Save a new default set up SMS command for date and time 0 dB reference selection Improved modem connections New weather station connection Environmental mode Pre weighting of spectra Noise criterions NC NR RC Marker for external DC failure Reference tone on recordings Pure tone mode Improved possibilities Pre set pure tone Standards Building acoustic mode Spectrum Adaptation Terms Equalizer General Equalizer implementation characteristics Using EQ with the pink noise excitation a cookbook Using EQ with the Swept Sine excitation a cookbook Swept Sine measurement technique ADA BR WWW W O O O O O1 o O 10 10 11 11 11 12 12 13 Contents Building Acoustics measurements using Swept Sine technique Introduction What is Swept Sine technique Benefits of using Swept Sine Principles behind the method Implementation in Nor121 Excitation signal Measurement of IRs deconvolution Setting up for Swept Sine measurement Level Setup Reverberation setup Rating and BGN setup Correction setup Making Measurements with Swept Sine Level in channel 2 and SNR display What does the measured BGN and SNR mean What does the displayed with level values mean Measurement of high sound insulation values Technical specifications Bandwidth 1 3 Octaves Bandwidth 1 1 Octaves Broadband impulse response measurements using Swept Sine Implementation Making IR Measurements a cookbook T
29. file management program Both ir1 wav and ir1 ndf files will be stored in the Ch1 sub directory of the corresponding measurement Files with impulse response measured on channel 2 ir2 wav and ir2 ndf will be stored in subdirectory Ch2 Technical specifications Bandwidth Broadband covering third octave bands centred at 50 Hz to 20 kHz Sweep duration 1 58 Acquisition duration 2738 Start frequency 42 17 HZ Stop frequency 23713 7Hz Impulse response length 2 73 s Sampling frequency 48000 Hz e NN Norsonic P O Box 24 N 3421 Lierskogen Norway Tel 47 32858900 Fax 47 32852208 info norsonic com www norsonic com Norsonic AS supplies a complete range of instrumentation for acoustics from sound calibrators micro phones amp preamplifers via small handheld sound level meters to advanced yet portable real time analysers but also building acoustics analysers and complete community industry and airport noise monitoring systems Contact your local representative or the factory for information on our complete range of instrumentation
30. hat it is necessary to sum up two responses the right one at t4 and the compressed one at to Resulting decay will be distorted and resulting noise floor will be higher as noise will not be averaged out With Swept Sine we do not shoot pulses at all We can think of Swept Sine excitation as of one pulse stretched in time It is therefore no wonder that synonyms like chirp and time stretched pulse also appear in the literature Let us for the moment look at the pulse not in time but in frequency domain It covers the range of frequencies from 50 Hz to 5 kHz If we would use a sine signal at 50 Hz frequency and constantly increase it s frequency up to 5 KHz the spectrum would be the same So this signal will produce the same effect as a pulse when considered over a period of time Because the frequency of the signal changes we would be at one frequency at exactly one instance of time The response recorded with the instru ment is also a sine signal but it is changed by the influ ence of the system under test To transform this signal into impulse response we need to perform mathematical operation known as deconvolution The effect of decon volution is to transfer responses at individual times and frequencies so that they appear at the same time L dB et t t Figure 3 Decay curve gets compressed as a consequence of increase of speed of sound 4 LA dB t A L dB t A L gt dB ES t
31. ion signal Instead the user is advised to do the following 1 Make a test measurement first look at the display that shows the level in sending room 2 Open the Generator menu and adjust EQ as desired 3 Make a new measurement with EQ On As an alternative you may adjust the equilizer by using pink noise before the Swept Sine mode is entered BA 13 10 TE 2995 05 25 67 28 49 95 2 dB 15 6 Lercspo fF spL f ikHz 1 3 or Level E18 8 4B Sync Noise type Fullscale 2 Equal izer Gain at this freqency EC de ESMMAITASTorscale t Topscale tJ 1 User interface Open the Generator menu and two parameters related to EQ are 1 Equalizer values On Off turns generator on off 2 Gain at this frequency values 6 6 dB adjusts the gain at each frequency band individually Building acoustic measurements under difficult situa tions such as high background noise in the receiving room may be greatly improved by use of the Swept Sine measurement technique This technique may also be used for measurement on high performance sound insulations that is impossible to measure with traditional technique The previous Norsonic analyzer Nor840 offered a similar feature with its unique MLS Maximum Length Sequence measurement mode The new Swept Sine technique within the Nor121 analyzer is also based on the impulse response of the measured rooms but the Swept Sine technique is more
32. ivate refer ence tone field in the RangeCal Ref Tone menu and then activated by pushing the Enter key Alternatively pro viding that one of the single markers is preset to the name Ref tone the reference tone may be activated by select ing that marker number key on the Nor121 front panel In the pre stored setups these new reference tone set tings are implemented within the standard setup named Annoyance Recorder Activate reference tone Noise tupe Fra B88 6 C14 4 ge Exc time fds Level Improved possibilities In the MeasSetup menu there is now a choice between the algorithm specified in the DIN 45 681 Entwurf 2001 and the algorithm specified in both the ISO DIS 1996 2 2005 and in the Joint Nordic Method Version 2 For the ISO JNM algorithm the parameters Delta and Re gression Range may also be adjusted The operator may speed up the pure tone analysis by deactivate the search for moving pure tones selecting the lowest level of analy sis Lmin and or turning off any analysis below hearing threshold Measurement control Pure tone Number of FFT averages Detect moving puretones Analyse below Hearing Threshold o Oo ES 5 o a oo Lmin Algorithm type 1S0 7JNM Delta dB Regeression Range fo The MeasControl PureTone menu now contains more features in order to analyse for pure tones in accordance with different Standard proposals Pure tone mode
33. n the lower parameter field The menu can be 5 f acs oun do inx SWEPT SINE Facade loudsp std level dif Save current setup as default Save default setup at switchoff accessed any time by pressing the Setup key Note that Show this menu at startup changing the setup will delete any unsaved measurements However a warning will always be given Figure 7 Setup Standards menu accessed with SETUP key page 21 Nor121 Level Setup Level setup menu is presented in figure 8 It is used to contol the input channels the frequency range the measurement duration and the dispaly of results Most of parameters are identical to those used for traditional measurement so only the parameters introduced with Swept Sine will be described here If parameter Show unreliable values is set to On then all level values will be displayed but marked with 2 if judged unreliable The symbol is used to mark measured values as unreliable if the measured signal to noise ratio is bellow a chosen threshold More about level values and use of symbol can be found in the chapter What does the displayed with level value mean Parameter Required S N ratio sets the threshold for marking the level results with If this value is set too high there will be too many question marks If this value is too low there will be less question marks but some values might deviate from the correct value witho
34. nd level setup accessible by BGN soft key from the measurement con trol menu A message The measured levels are automati cally corrected for the apparent background noise when sweep is excitation type is displayed instead The use of other parameters from Rating setup menu is the same as for classical measurements Please refer to main user manual for details Correction setup The use of correction setup menu is the same as for a classical measurement Please refer to section Applying corrections in main user manual for details Making Measurements with Swept Sine Making measurements using Swept Sine technique is even simpler than with the classical method as it is not necessary to measure background noise levels separate ly AS measurement is usually conducted in accordance with an appropriate standard it is advisable to use stand ard setups and follow the instructions given in section Measuring airborne sound reduction between rooms a cookbook from the main instruction manual The rest of this chapter will focus on things that are specific to Swept Sine Level in channel 2 and SNR display It is possible to measure levels in source and receiving room as well as reverberation times in receiving room However it is neither possible nor needed to measure background noise levels separately as with a classical measurement With Swept Sine technique background noise level is measured as a part of a normal level m
35. not look like this but responses to those pulses would be buried into thousands of responses from other pulses However this representation is convenient to depict the physical process To obtain one impulse response from noise like re sponse signal we use a mathematical method known as Hadamard transformation which does nothing else but shifts all those responses back in time in proper order so that all of them appear at the same time After this they are summed Let us emphasize that this operation involves summing of energy at each frequency obtained at differ ent time instances Note that all frequencies are meas ured in several time instances Figure 2 a Time frequency representation of part of MLS excitation b t f representation of corresponding part of response page 16 The effect of air fluctuation wind and temperature fluctuations is that the speed of sound changes over time This means that response obtained at lower speed will be stretched in time as it takes more time to gather all the energy The reflections arrive later as they travel slower Let us assume that such a change of speed of sound happens between two time instances t and to Figure 3 represents level as function of time for fixed frequency f4 As an increase of speed of sound it is noticeable that re sponse appearing at to is shorter than response appear ing at ty Synchronous averaging shifts both responses to time O and averages them This means t
36. ransferring measured IR from the Instrument to a PC Technical specifications 14 15 16 18 18 19 20 22 24 24 24 25 25 27 27 29 30 30 30 31 31 31 32 32 Instruction manual page v Instruction manual Introduction The measurement capabilities of the Nor121 sound level meter depend largely on the options installed in the in strument This user guide is applicable to software ver sion 4 0 which contains many new features like e Pre weighting of spectra e Reference tone on recordings e 0 dB reference selection e Quicker instrument switch off e Equaliser for signal generator e Swept Sine measurement technique The options may be either ordered and installed in in struments delivered from the factory or retrofitted to ear lier delivered units This manual does not describe how to install the options Contact the factory or your Norsonic representative if you need more information For description of the sound level meter Nor121 in gen eral reference should be made to the main user guide for the instrument page 1 Nor121 page 2 Quicker instrument switch off The Nor121 was originally designed to save the current set up by switch off in order to start up again with exactly the same set up However this means that the turn off procedure lasts for couple of seconds The new version has a choice in the Setup menu that will de activate this feature allowing the instrument to switch off imm
37. reduce influence of transients is obsolete with Swept Sine Principles behind the method To explain principles behind Swept Sine method let us first take a look at classical pulse excitation by start pis tol for example and its response Let us assume that ex citation pistol will produce a pulse whose energy covers the frequency range of at least 50 Hz to 5 kHz which is our range of interest With such pulse all excitation energy appears at all frequencies at the same time Response to such pulse also contains information at all frequencies at the same time as shown in figure 1 So measurement is undertaken for all frequencies at only one instance of time The inherent drawback is that a large amount of en ergy has to be delivered at once This limits the signal to noise ratio To obtain higher signal to noise ratios more excitation energy has to be used Both Swept Sine and MLS tech nique exploit this approach With MLS we simply shoot several pulses and we do it very fast and for several sec onds or even minutes Each of those pulses delivers en ergy at all frequencies and they are very close so signal has spectral density of white noise Each of pulses will have its own response whose tails will overlap to form a noise like response signal To depict this in a way which is analogous to pulse excitation we need to observe two isolated pulses at time instances t and ty as depicted in figure 2 Note that sig nal itself would
38. rent parameters are the same as for the classical measurement as well as the use of LAST and AVERAGE data registers to perform spatial averaging Results can be saved and transferred to a PC using NorXfer software Reports can be generated using the NorBuild software package 1 2 Leeg eq f 4kHz 13 so de L ata Topscale t Topscale To o Figure 15 Question mark with level difference value page 28 Note that the symbols are displayed only after the whole measurement is finished i e when instrument is in ENDED state Level values on the other hand are already displayed as they are measured i e while the instrument is in RUNNING state In order to provide the information about reliability of measured results while the measurement is still running a warning S N ratio requirement is not fulfilled is used It will be displayed during the measurement as soon as a level which will be marked by when the measurement is finished is detected This can help the user to identify the suspicious results and eventually stop the measurement Measurement of high sound insulation values One of the advantages of the Swept Sine technique is the ability to perform measurements of high sound insula tion values and measurements in presence of high back ground noise In both cases the measurement benefits from the noise suppressing capability of Swept Sine algo rithm The noise suppressing capability m
39. response is normally the com bined impulse response of the system consisting of am plifiers transducers applied filters and the enclosure between the transmitting and the receiving points Thanks to known fact that crest factor which is a ratio of peak to RMS voltage of a sine signal is only 3 01 dB Swept Sine excitation can be fed to amplifier and loud speaker with substantially more power than noise signal As more signal power means better SNR sweep meas urements can provide signal to noise ratios in excess of 100 dB at quiet sites This extended measurement range enables a measurement under situations where no result is obtained by the classical method Due to use of signal processing the new method is more immune to noise from other sources such as clicks pops and even footsteps This is very important in build ing acoustics where measurements sometimes need to be conducted on construction site while work is still un derway The use of linear deconvolution allows all distortion ef fects from an overloaded loudspeaker to be deleted from the results Figure 1 a Time frequency representation of pulse excitation b t f representation of response to pulse excitation Instruction manual page 15 Nor121 Swept Sine does not require synchronization between generator and receiver For MLS this was the issue as the errors in synchronization degrade the measurement Pre sending which is necessary with MLS in order to
40. room ensuring the fulfilment of standard requirements and quality of measurement results Equalizer implementation characteristics The equalizer implemented in Nor121 is a 21 band equal izer with one third octave bandwidth The central frequen cies of filters are at one third octave bands mid frequen cies from 50Hz to 5kHz Gain in each band can be adjusted from 6dB to 6dB If positive gains are used the overall level will be cor respondingly lowered in order to prevent speaker and amplifier overload If the user needs more power it can be added by increasing Level parameter from Generator Instruction manual Equalizer menu However it is advisable to use equaliser in order to lower the levels in those bands where it is necessary rather then to boost level in other bands Figure below shows that positive gain at one frequency upper curve will result in negative gain correction of the output The result is that the actual gain cannot exceed 0 dB lower curve in order to protect the loudspeaker and the amplifier Equalizer can be used with both noise excitation and Swept Sine excitation Equalizer gains can be adjusted while the generator is running with pink noise excitation If Swept Sine excitation is used the gains has to be ad justed prior the measurement Please note the equalizer cannot be used with Swept Sine excitation when sweep duration parameter has value Long 5 kHz 50 Hz 125
41. t have the same meaning as with rever beration results Level report contains several displays Individual level values or level difference between chan nel 1 and 2 might be displayed If an individual level value is marked with this means that the real value is less or equal to the measured value In other words the real value might be lower but we are not able to measure it because the noise hinder us Example of individual level value marked with is shown in figure 14 If a level difference value is marked with you need to take a look at reliabilities of individual values of L1 and L2 in order to draw a conclusion BA 13 10 TH 2004 10 19 13 09 59 f 4kHz 1 3 Instruction manual 4444 _ Topscale f Topscale f d o Figure 14 Question mark with individual level value means ess or equal page 27 Nor121 If only L2 is marked with that means that real differ ence is greater or equal to the measured value In other words the real value might be higher but we are not able to measure it again because of noise Example of level difference value marked with is presented in figure 15 If only L1 is marked with that means that the real difference is smaller or equal to measured value If both L1 and L2 are marked with we can not draw any other conclusion but that the result is simply unreli able All other features including rating and calculation of diffe
42. ter As an impulse response is a fundamental quality in system analysis in general its measurement with Nor121 can be performed in broad range of applications even beyond field of acoustics page 13 Nor121 Building Acoustics measurements using Swept Sine technique Introduction What is Swept Sine technique Classical method for the measurement of sound transmis sion phenomena uses random noise or impulse signals as excitation The resulting levels or decay times are de termined directly from the recorded responses In order to measure with greater accuracy and repeat ability new measurement methods have been developed They use deterministic signals as excitation and apply digital signal processing principles to obtain the impulse response of system under test This is a great advantage as all acoustical parameters can be obtained from further processing of the impulse response This makes these methods applicable to all areas of acoustics In building acoustics this method is applicable to sound pressures measured in rooms as well as to the velocities measured on structures In room acoustics im pulse response plays a central role as many acoustical quantities can be derived from it as specified in Inter national standard ISO 3382 Development of transducers such as loudspeakers and microphones involves numer ous measurements of frequency response function which is linked to the impulse response by Fourier transform Vi
43. ters as specified in IEC 61260 At the same time they produce short virtual reverberation time This is very important as virtual reverberation time limits the shortest reverberation time that can be meas ured In Swept Sine case those times are considerably shorter then at the classical method Generator Instruction manual With present implementation it is possible to measure in one third octave bands with midband frequencies from 50 Hz to 5 kHz i e 21 one third octave bands or in octave bands with midband frequencies from 63 Hz to 4kHz i e in 7 octaves The duration of the sweep is directly propor tional to amount of energy used for excitation The longer the sweep time the greater the resistance toward back ground noise present at measurement site For measurement of reverberation time background noise is seldom a problem as the excitation loudspeak er and measuring microphone s are in the same room Therefore a sweep duration of approx 15 s for octave and approx 60 s for third octave bands has been chosen for measurement of reverberation time These durations in normal measurement conditions can provide signal to noise ratio of up to 100 dB which is far more then what the standard requires Figure 6 Block diagram of implementation in Nor121 page 19 Nor121 For measurement of levels the signal to noise ratio in the receiving room might be an issue In such situations the user is provided with parameter named
44. ut any warnings Measurement control Level setu Fullscale Channel 1 129 Jae Channel 2 ga Jas Fra range BG 5k Hz Bandwidth 173 Joct Measurement type Show unreliable values Required S N ratio Jae Sweep Duration Total sweep time is 69 s Figure 8 Level setup is part of measurement control menu accessed by Meas Control key page 22 Parameter Sweep Duration can have three values Short Medium and Long This parameter together with Frequency range and Bandwidth affect the total duration of excitation which is displayed at the bottom of the menu The idea behind the Sweep duration parameter is fol lowing The Short duration is considered as the default value It is expected that this setup will give satisfactory results for most applications The other two longer dura tions are meant to be used only in cases the noise level in receiving room is too hight such that short duration measurement cannot produce reliable results In this case an increase in SNR can be achieved by exciting the room under measurement with more energy which is equivalent to delivering excitation energy in a longer period of time In order to save time it is suggested to first try to measure with Short sweep duration Only if the results of such a measurement are marked as unreli able with the medium or long durations should be chosen Instruction manual Measurement
Download Pdf Manuals
Related Search