Some standard content:
ICS17.140
National Standard of the People's Republic of China
GB/T25079—2010/IS018233.2006 Acoustics
Application of new measurement methods inbuilding and room acoustics-MLS and SS methods(ISO18233:2006,IDT)
2010-09-02Release
General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of ChinaStandardization Administration of the People's Republic of China
2011-04-01Implementation
Normative references
Terms, definitions and abbreviations
Pulse response measurement
Frequency response function measurement
8Precision
9Test report||tt| |Appendix A (Normative Appendix)
Appendix B (Normative Appendix)
References
Maximum Length Sequence Method
Sine Sweep Method
GB/T25079—2010/ISO18233:200610
GB/T25079—2010/ISO18233:2006Application of New Test Methods in Building Acoustics and Room AcousticsThis standard adopts ISO18233:2006 "Acoustic Methods" (English version) in the same way.
This standard has made editorial changes to the international standards adopted in the same way.Appendix A and Appendix B of this standard are normative appendices.This standard was proposed by the Chinese Academy of Sciences.
This standard is under the jurisdiction of the National Technical Committee for Acoustics Standardization (SAC/TC17).The drafting units of this standard are: Institute of Acoustics, Chinese Academy of Sciences, China Academy of Building Research. The main drafters of this standard are: Lv Yadong, Qiu Bo, Miao Zhenwei, Tan Hua, Cheng Mingkun, Yin, Xu Xin. MLS and SS
GB/T25079—2010/ISO18233:2006 Introduction
The random signal analysis methods used to measure sound propagation phenomena have been developed since 1960, but due to the lack of effective computing power at the time, these methods can only be applied to well-equipped laboratories. With the development of digital circuits, powerful general-purpose computers, and the application of digital signal processing components in field acoustic measurements, the application of measurement instruments based on extended digital signal processing has become increasingly mature. At present, special instruments and professional software that can run on general-purpose computers have adopted these measurement methods and have been widely used. Compared with traditional methods, the new methods have many advantages, such as: suppressing background noise and extending the measurement range. However, if certain guidelines are not followed, reliable results may not be obtained. At the same time, compared with traditional methods, the new methods may be more sensitive to time changes and changes in environmental conditions.
This standard is intended to provide requirements and guidelines for new measurement methods for building acoustics and room acoustics, and these requirements and guidelines can also be applied to the measurement equipment that applies these methods. Even people who are experienced with traditional methods and their measurement equipment may not be aware of the difficulties and limitations of some applications of new methods. Therefore, each user should be encouraged to have a deeper understanding of the theoretical basis of the new methods. At the same time, instrument and equipment manufacturers should be encouraged to provide more guidance on the application of equipment and make it a design goal for instruments and equipment to be able to give timely warnings when measurement results are unreliable. This standard provides requirements and guidelines for the application of new methods in the measurement of sound insulation of buildings and building components, reverberation time and related physical quantities. The references provide relevant standards for traditional methods regarding measurement content, number and location of measurement points, and measurement conditions. 1 Scope
GB/T25079—2010/ISO18233:2006 Acoustics
Application of new measurement methods in architectural acoustics and room acoustics
MLS and SS methods
This standard specifies the application guidelines and requirements for new methods for measuring the acoustic characteristics of buildings and building components. It also provides guidelines and requirements for the selection of excitation signals, signal processing and environmental control, as well as requirements for the linearity and time invariance of the measured system. This standard applies to the following measurements, such as airborne sound insulation between rooms and exterior walls, room reverberation time and other indoor acoustic parameters, reverberation room sound absorption, vibration level difference and loss factor. The methods defined in this standard can replace the traditional methods defined in GB/T19889 (all parts), ISO3382 (all parts) and GB/T21228.1.
2 Normative references
The clauses in the following documents become the clauses of this standard through reference in this standard. For any dated referenced document, all subsequent amendments (excluding errata) or revisions are not applicable to this standard. However, parties reaching an agreement based on this standard are encouraged to study whether the latest versions of these documents can be used. For any undated referenced document, the latest version applies to this standard. GB/T3241 Octave and fractional octave filters (eqvIEC61260:1995, GB/T3241-1998) GB/T3785 Electrical and acoustic properties and test methods of sound level meters (IEC61672-1, NEQ) 3 Terms, definitions and abbreviations
3.1 Terms and definitions
The following terms and definitions apply to this standard. 3.1.1
Traditional method classicalmethod
Traditional measurement method for determining sound pressure level or decay rate directly by recording the response of random noise or impulse signals. 3.1.2
New methodnewmethod
Use various deterministic signals to first obtain the impulse response of the system under test, so as to obtain the required sound pressure level and decay rate measurement method. Note: The new method has some other inherent characteristics that the traditional method does not have, such as the ability to avoid noise interference from other sound sources. 3.1.3
Effective signal-to-noise ratioeffectivesignal-to-noise ratiosignal-to-noise ratio
The ratio of the mean square value of the signal part generated by the excitation source and obtained by the new method to the mean square value of the unnecessary part of the signal generated by the same method and non-excitation source, taken by taking the logarithm to the base 10 and multiplying it by 10. Note 1: Effective signal-to-noise ratio is expressed in decibels.
Note 2: In the test steps of the new method based on the traditional method, the effective signal-to-noise ratio is used instead of the usual signal-to-noise ratio. 3.1.4
Peak-to-noise ratio
The ratio of the square of the peak value of the signal part generated by the excitation source and obtained by the new method to the mean square value of the unwanted part of the signal generated by the same method and non-excitation source, taken by the logarithm to the base 10 and multiplied by 10. Note: Effective peak-to-noise ratio is expressed in decibels.
GB/T25079—2010/ISO18233:20063.1.5
Fractional-octaveband The frequency range from the lower limit to the upper limit of the fractional-octave filter specified in GB/T3241, in Hz. Note: Both octave and fractional-octave filters are specific fractional-octave filters. 3.2 Abbreviations
4 Special References
Maximum Length Sequence Method
Sine Swept Frequency Method
4.1 Maximum Length Sequence Method (MLS)
The MLS method in accordance with this standard is defined as \GB/T25079-MLS". 4.2 Sine Swept Frequency Method (SS)
The SS method in accordance with this standard is defined as \GB/T25079-SS". 5 TheorybZxz.net
5.1 Overview
The sound propagation between rooms can usually be regarded as an approximately linear time-invariant system. Therefore, the general theory applicable to this system can be used to establish the sound propagation relationship between excitation and response. The impulse response is the basis of all measurements. This method can be used for both structural vibration velocity measurement and indoor sound pressure level measurement. 5.2 Indoor sound
Parts 3 to 5, 10 and 14 of GB/T19889 specify the measurement methods for airborne sound insulation of building components and rooms. ISO3382 (all parts) specifies the test methods for reverberation time. In order to test these physical quantities, noise excitation should be used to measure the indoor sound pressure level and reverberation time. For the measurement of reverberation time, the noise source should be turned on for a period of time to obtain a steady-state sound pressure, and then the noise source should be turned off to observe the sound decay in the room. In this standard, the moment when the noise source is turned off is set as time zero, t=0. The record of the change of sound pressure level over time generally contains information about the steady-state sound pressure level and reverberation time of the room. Figure 1 is a typical relationship between sound pressure level and time. The steady-state sound pressure level before the noise source is turned off is the information recorded when t<0, and ≥0 contains decay information. The decay information can be further processed to obtain the reverberation time. The traditional method for measuring room air sound defined in GB/T19889 and ISO3382 series of standards stipulates that a random signal is used as the excitation source. Although the room can usually be described as a deterministic system, the statistical distribution of the random excitation signal makes the final result have a certain random variation, and the standard deviation is used to characterize this randomness. Therefore, it is usually necessary to take the average of multiple measurement results to approximate the expected value in a statistical sense. The traditional method is to average the measurement results of the spatial measurement points to obtain the average value of the room. The method described in this standard is intended to obtain measurement values in fractional octaves. The corresponding requirements and guidelines need to be selected. As shown in Reference 6, the expected decay at a specific observation point can be obtained by directly processing the impulse response between the excitation signal source (loudspeaker) and the observation point (microphone) without averaging. As long as the system is linear and time-invariant, the application of this theory to measure decay curves and steady-state sound pressure levels is valid. The theory can be extended and applied to the sound field measurement of the sound source room and the receiving room and the sound transmission measurement from the sound source room to the receiving room.
Theoretically, the response based on noise excitation measured by the traditional method can be described as the convolution of the excitation signal and the room impulse response. However, in the traditional method based on noise excitation, the response can be recorded directly, while the impulse response is generally unknown. According to the new method introduced in this standard, the measurement result can be obtained by processing the impulse response itself. Note: The impulse response is usually the integrated impulse response of the system consisting of the amplifier, the sensor, the filter used and the closed space between the transmitting point and the receiving point.
Lo Steady-state noise level before the excitation signal is turned off; LN
Background noise level;
A time.
Note: At t=0, the excitation signal is turned off.
GB/T25079---2010/IS018233.2006 Figure 1 Typical curve of sound pressure level changing with time There are many methods to obtain the impulse response and the frequency response function obtained by Fourier transforming the impulse response. All of these methods can be adopted if they show that reliable results can be obtained under normal test conditions. The room system is excited by a steady-state white noise signal and lasts long enough to obtain a stable sound field. The sound source is turned off at t=0. The expected sound pressure level L(t) (unit: dB) at any time t≥0 can be expressed as: L(t)=10lg
W. Constant refers to the power per unit bandwidth of the excitation signal; h(t)
is the impulse response;
is an arbitrarily selected reference value for calculating the sound pressure level. Cref
h2(t)dt
The decay curve corresponding to the expected decay based on the traditional method can usually be approximated as a straight line. (1)
Note: Since time t is the lower limit starting point of the integration, equation (1) can be regarded as a reverse integral. After the formula is equivalently transformed, the reverse integration starts from ten to the actual time. In the past, the reverse integration was achieved using the recording rewind simulation technology. Equation (1) does not include the external noise that is usually associated with the measurement process. If the measurement system uses a fractional octave filter, then equation (1) describes the expected decay of the filter band obtained by the traditional method. Equation (1) can be used to calculate the expected sound pressure level at any time after the sound source is turned off. This formula can also be used to calculate the average sound pressure level L before the sound source is turned off. (Unit: dB). Assume that t=0 in equation (1), then: Lo=10lg
h2(t)dt
Figure 2 illustrates how to obtain the functional relationship between sound pressure level and time using the traditional method and the new method. 5.3 Sound propagation between two rooms
（2）
If the noise source is placed in the sound source room and the sound pressure level measurement point is S, the expected sound pressure level Li (unit: dB) can be obtained from the impulse response h1(t) between the excitation point and the measurement point S according to formula (2): [Wn(e)dt
Lr=10lg
（3）
GB/T25079—2010/IS018233:2006L4
Sound pressure level;
-impulse response;
Time.
a) Traditional method
b) New method
Note: In the traditional method, the approximate value of the expected decay curve Lm(t) is the average value of multiple decay curves Lr(t), L(t),, Ln(t) measured based on the noise excitation method. In the new method, the expected decay curve L(n) is obtained by processing the impulse response (t). Figure 2 Illustration of the difference between the traditional method and the new method Similarly, if the sound pressure level is measured at point R in the adjacent receiving room, the expected sound pressure level Lz (unit: dB) can be obtained from the impulse response h2(t) between the excitation point and the measuring point R:
『w。
L2=10lg
[h?(t)dt
The expected sound pressure level difference D (unit: dB) between the source room and the receiving room is calculated by formula (5):hi(t)dt
DLi-L2=10lg
h(t) dt
It can be seen from formula (5) that the expected sound pressure level difference is independent of the excitation signal power W. and the reference value Crat. Note: The new method of this standard can also be applied to the sound insulation measurement of the exterior wall of a building. One of the measuring points in the measurement should be located outdoors in the building. 5.4 Application of frequency response function
（5）
In the theory of signals and linear time-invariant systems, sinusoidal signals play a unique role. If the transient phenomena generated when the signal is turned on and off are ignored, the response of the linear time-invariant system to the sinusoidal signal is still a sinusoidal signal at the same frequency, but the amplitude (gain) and phase will change. The information about the change in amplitude and phase between the input and output signals is expressed as a function of frequency, which is called the frequency response function of the system. Like the impulse response, the frequency response function can give all the response information of any input signal. The frequency response function can be obtained by Fourier transforming the impulse response. Applying Parseval's theorem, equation (2) can be changed to: 00
Wo|h2(t)dt-
Where:
One-corner frequency;
[H()\da
H()——Frequency response function obtained by Fourier transforming the system impulse response h(t): H(w) =F(h(t)) =
Where: j=V-1
h(t)e-iatdt
Note: In equation (6), it is assumed that h(t)=0 when t<0, which is consistent with the physically realizable causal system. (6)
(7)
From equation (6), it can be seen that the calculation of sound pressure level is only related to the modulus of the frequency response function. The measurement of reverberation time is related to the phase and modulus of the frequency response function.
Combining equations (5) and (6), the expected sound pressure level difference D between the source room and the receiving room can be obtained through the frequency response of the room. The expected sound pressure level difference D (unit: dB) with an upper limit frequency of f2=
octave (lower limit frequency of f1:
) can be expressed as: |H()}2da
D=Li-L2=10lg
6 Impulse response measurement
6.1 Overview
JH2()|2da)
A typical room impulse response is an oscillating signal with many cycles. The envelope of the signal is irregular, but usually has a very short pulse time and then decays exponentially. The response of the room to a very short sound pulse can be regarded as the impulse response of the room. However, in most cases the sound source used is not a loudspeaker, so it is difficult to control the spectrum and directivity of the excitation signal. In order to obtain the necessary control over the excitation signal, the impulse response is obtained by digital signal processing in many practical situations. The room is excited for a period of time with a known signal, and the impulse response of the room is calculated from the response of the room to the excitation signal. The excitation signal is distributed over a long period of time to increase the total radiated energy. This processing method can increase the dynamic range obtained and reduce the influence of external noise. References introduce several methods for measuring impulse response, see references [6]~[8] and [13]~[15]. When measuring the impulse response, it is not allowed to change the sound source and microphone position, because this will violate the time invariance requirement required by the system under test. The impulse response of a room is formed by the interaction of sound waves reflected from the floor, ceiling and walls of the room. Between multiple reflections, the air in the room affects the sound propagation. Air movement or changes in the speed of sound (caused by air temperature) may also violate the time-invariance requirement of the system under test. 6.2 Stimulus signal
6.2.1 Overview
In the traditional method, the excitation signal is a random signal or an impulse signal with a bandwidth at least equal to the bandwidth of the measurement channel. The randomness of the noise signal makes the measured sound level randomly distributed and also limits the repeatability of the measurement. The new method uses a deterministic excitation signal that can be accurately reproduced, thereby enhancing the repeatability of the measurement. 5
GB/T25079—2010/ISO18233:20066.2.2 Spectral requirements
6.2.2.1 Overview
The effective frequency response range of the excitation signal should at least cover the actual measured fractional octave. If a broadband measurement covering the entire audio range is performed, the goal is to make the spectral shape of the excitation signal received at the receiving point close to the spectral shape of the surrounding background noise, so that a frequency-independent signal-to-noise ratio can be obtained. Typical background noise sources (from air conditioning systems, traffic, etc.) have a spectrum that increases with decreasing frequency. Therefore, when measuring the impulse response of the room, the low-frequency components of the excitation signal should be emphasized. In most similar cases, a pink noise excitation signal (with equal energy in each fractional octave) is suitable to obtain an adequate signal-to-noise ratio.
In sound insulation measurements, the sound insulation usually increases with increasing frequency, so it is necessary to increase the energy of the high-frequency components of the excitation signal. The most perfect adjustment method is to compensate for the sound power response of the measurement loudspeaker and adapt to the spectral distribution of the background noise. The ideal method that can achieve both functions is to multiply the spectral distribution of the smoothed background noise with the inverse response of the loudspeaker within the pre-specified measurement frequency range as the sounding mode of the appropriate excitation signal spectrum. 6.2.2.2 Repetitive excitation signal
If a repetitive excitation signal is applied, the spectrum of the excitation signal will contain narrowband spectral lines, the distance between adjacent spectral lines △ is the inverse of the signal repetition period TREP:
Af=TREP
（9）
In order to ensure that all normal vibration modes of the room are excited, the repetition period of the signal should not be shorter than the reverberation time of the room under test. This requirement applies to both reverberation time and sound pressure level difference measurements: TREP≥T
.....10)
Note: Each normal vibration mode of the room can be passed through a second-order bandpass function with a specific quality factor (Q factor). A larger quality factor means a narrower frequency response bandwidth and a longer decay time after the excitation signal stops. For a second-order function with a bandwidth of (decay-3dB)B (in Hz), the actual reverberation time is approximately (2.2/B). The repetition time is required to ensure that at least two spectral lines of the excitation signal fall within the bandwidth of any normal vibration mode of the room.
6.2.2.3 Non-repetitive excitation
The non-repetitive excitation signal can be of any suitable length. However, a certain period of silence is required after the excitation signal to ensure accurate recording of the decay response. The decay should be recorded for a period of time at least equal to 1/2 the reverberation time. For a swept signal that sweeps from low frequency to high frequency (see Appendix B), the required length of the silent period is determined by the reverberation time of the upper frequency limit. 6.2.3 Sound pressure level and linearity
The sound power of the excitation signal should be high enough to obtain an effective signal-to-noise ratio that meets the requirements of the standard of the applicable traditional method. In general, the method using a deterministic excitation signal can better suppress external noise than the traditional method. Compared with the traditional method, the signal-to-noise ratio can be improved by 20dB to 30dB or even more. The use of loudspeakers usually introduces nonlinear distortion into the system. Nonlinear distortion does not meet the linearity requirements of the new method. The nonlinear distortion of a loudspeaker increases with increasing sound pressure level. The user should be aware of this problem and experiment with different excitation signal sound pressure levels to obtain the best signal-to-noise ratio. Sometimes the signal-to-noise ratio can be improved by reducing the sound pressure level of the excitation signal. This problem needs special consideration in the application of the MLS method described in Appendix A (see Appendix A). If properly established, the MLS sine-swept method described in Appendix B can effectively eliminate the effects of harmonic distortion on the measurement results. The region where the impulse response decays to the noise floor is usually most susceptible to nonlinear distortion. This makes reverberation time measurements more susceptible to distortion effects than sound pressure level difference measurements. 6.2.4 Directivity
The directivity of the sound source shall comply with the requirements specified for the applicable conventional method. 6.2.5 Number of sound source locations
The number of sound source locations shall comply with the requirements specified for the applicable conventional method. 6
6.3 Response measurement
6.3.1 Measurement sensor
GB/T25079—2010/ISO18233:2006 The measurement sensor (usually the measurement microphone) shall comply with the requirements of the applicable conventional method. 6.3.2 Frequency weighting
The methods introduced in Appendix A and Appendix B describe the measurement of wideband impulse responses. Further processing of the wideband impulse response can obtain a fractional octave weighted impulse response for the required frequency band. Although equations (1) to (5) are general, the impulse responses in these formulas should be fractional octave weighted so that the final result is valid for each fractional octave band.
The fractional octave weighted impulse response can in principle be obtained as the output of a fractional octave filter (fractional octave filter specified in GB/T3241) applied to the wideband impulse response signal. When selecting methods for the required frequency weighting, for the appropriate type of filter specified by the traditional method, measures should be taken to ensure that the tolerances of the frequency weighting meet the requirements of GB/T3241. The sampling frequency should be selected appropriately and measures should be taken to prevent the adverse effects of frequency mixing. For excitation with repetitive signals, the response should be recorded in accordance with the time and frequency resolution set for the excitation signal requirements, and the response to the excitation signal for one or more cycles should be recorded. For non-repetitive signals and sound pressure level measurements, the recorded portion of the impulse response should cover the time from the start of the excitation signal until the response decays by more than 30 dB in each fractional octave. For reverberation time measurements using non-repetitive excitation signals, at least the decay portion required by the applicable traditional method should be recorded.
6.3.3 Sound pressure level linearity and dynamic range
The signal processing should have sufficient resolution and dynamic range to meet the requirements for sound pressure level linearity in GB/T3785. Measuring instruments designed to obtain results using new methods cannot be tested in the same way as ordinary acoustic measuring instruments. Generally, the microphone signal is digitized and the result is obtained by sampling the microphone signal using a digital processing system. Normal tests can usually verify the normal operation of the microphone and digital circuitry, but not the entire calculation process. As long as the equipment can obtain valid results, the accuracy of the digital processing is determined by the design of the equipment and is not affected by equipment aging or changes in operating environmental conditions. It is recommended to test the effectiveness of the design and operation of the system by conducting experiments that can be compared with the results measured by traditional methods. The room used for verification tests should have well-controlled acoustic characteristics and fixed points should be selected as measurement points. It is more convenient to use a time-invariant system with electronic input and output signals. This system can use a digital reverberator without time modulation. The range of tests needs to cover the entire range of possible reverberation times. The performance of the test equipment under low signal-to-noise ratio conditions can be studied by adding broadband random noise to the analog input or output signals. It is recommended to regularly (and if appropriate, periodically) check the microphone, digital circuitry and stimulus signal source. 6.3.4 Crosstalk
The use of deconvolution measurement techniques allows measurements with a large dynamic range, often extending to situations where the signal level is lower than the external noise level. Even signal levels below the inherent noise level of the microphone and the measurement system can be measured. Therefore, the influence of unwanted signal paths, such as electrical crosstalk, should be carefully eliminated. The cables connecting the excitation sound source, such as the speaker, should be kept away from the cables connecting the microphone and shielded. Sometimes even internal crosstalk of the device, which is usually hidden in the self-noise, will show up. Replacing the normal transducer (microphone) with a dummy device with a very low sensitivity to the measured signal can be sufficient to eliminate the influence of crosstalk. If possible, the impulse response is displayed, which can illustrate possible crosstalk problems. Due to the finite speed of sound propagation and the distance between the transmitting and receiving points, the received sound signal, even the direct sound, is usually delayed. The crosstalk signal is an electrical signal and therefore usually has no delay. To eliminate the influence of residual crosstalk signals, any non-acoustic components can be reduced by windowing the initial impulse response signal. 6.3.5 Time integration limit
6.3.5.1 Sound pressure level measurement
Equation (2) shows that the infinitely long integration time is infinite. This is neither possible nor necessary. The length of the recorded impulse response determines the maximum value of the upper limit of the integration. 7
GB/T25079—2010/IS018233:2006 The measured impulse response is always accompanied by unwanted noise signals from external noise sources and equipment self-noise. The nonlinearity and time-varying nature of the system will increase the noise. As the integration time increases, the impact of these unwanted noises in the integration also increases. If integrated from 0 to t, the sound pressure level can be expressed as:
L=10lg
Where:
Background noise signal.
e2(t)d
W|h2(t)dt+
In equation (11), it is assumed that h(t) and e(t) are unrelated, so the cross terms are ignored. If the upper limit of integration is very small, the result of integration will be very small. Figure 3 shows the influence of t on the sound pressure level calculation result. Css
Sound pressure level (dB);
a) Envelope diagram of impulse response, background noise and the composite signal of the two 0.1
b) Relationship diagram of calculated sound pressure level and upper limit of integration t-the ratio of time to reverberation time;
-the ratio of integration limit time to reverberation time; S—impulse response signal:
N—background noise signal;
CsN—signal synthesized by impulse response and background noise. Note 1: In order to study the influence of background noise on the measurement results, the effective signal-to-noise ratio of this example is only 10dB. Time is the ratio of real time to reverberation time. Note 2: The time is the ratio of the relative reverberation time. The sound pressure level is calculated by equation (11), and the figure shows the first part of the integral S, the second part N, and the total integral CsN. 0dB is the correct sound pressure level in the absence of any noise influence and the integration limit is infinite. Note that in this example, the difference between the maximum envelope of the noise and the impulse response is only 10dB. Figure 3 Integration limit
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.