GB/T 43944-2024Measurement method of weighted sound insulation index of ship internal material (English PDF)

Note Note on the printed English title

The English title on the cover of the document is printed as Measure method of weighted sound insulation index of ship internal material. The first word has been set here as Measurement, which is what the Chinese title and the body of the document call for; nothing else in the title has been changed.

3 Terms and definitions

Four terms are defined. The average sound pressure level in a room (3.1) is the ratio of the space and time average of the squared sound pressure to the squared reference sound pressure; a note says that the space average covers the whole test room but excludes the region directly radiated by the source and the region close to the boundaries, for example the wall faces. The sound reduction index (3.2) is the ratio of the sound power incident on the test specimen to the sound power transmitted through it.

The apparent sound reduction index (3.3) is the ratio of the sound power incident on the partition specimen from the source room to the total sound power transmitted into the receiving room, which includes the power transmitted through the specimen and the power transmitted by other paths. The weighted sound reduction index (3.4) is the single number obtained by weighting a set of band sound reduction index data against a set of reference values.

4 Measuring equipment

4.1 The accuracy of the sound pressure level measuring equipment shall meet the class 1 or class 2 accuracy requirements of GB/T 3785.1-2023, and unless the source room and the receiving room both use microphones with the same frequency response in a diffuse field, the measuring equipment shall be corrected for the diffuse field.

4.2 Before use, the whole measuring system including the microphone shall be calibrated with a sound calibrator meeting the class 1 accuracy requirements of GB/T 15173-2010. 4.3 The one third octave band filters shall meet GB/T 3241. 4.4 The reverberation time measuring equipment shall meet ISO 354.

5 Test specimens

5.1.1 The size of the specimen is determined by the size of the test opening in the laboratory facility; for panels it is about 10 square metres and for deck structures between 10 and 20 square metres, the short side of a panel or deck structure being not less than 2.3 m. A note says that if the wavelength of the lowest free bending wave frequency is shorter than half the shortest side of the specimen the specimen can be smaller, but that the smaller the specimen the more sensitive the result is to the boundary fixing conditions and to local variations in the sound field.

5.1.2 The specimen shall be installed so as to reproduce the normal connection and sealing at the boundaries and at the nodes under real conditions. 5.1.3 Where the specimen is installed in an opening between the source room and the receiving room, the ratio of the niche depth on the source room side to that on the receiving room side should be 2 to 1, as shown in Figure 1; the legend of Figure 1 names those two depths, the specimen and the wall in which it is mounted.

5.1.4 Where one face of the specimen absorbs markedly more than the other, the more absorbent face shall face the source room, and diffusers shall then be installed in the source room. 5.1.5 In a laboratory meeting GB/T 19889.1-2005, the sound power transmitted by paths other than the direct one shall be shown to be negligible against the power transmitted through the specimen; to verify this the maximum apparent sound reduction index of the laboratory installation shall be measured with a construction of very high sound insulation fitted in the test opening, following Annex A of GB/T 19889.1-2005.

5.2.1 The test method for doors, windows, glazing and similar components is the same as for panels and deck structures. 5.2.2 Where the specimen is smaller than the test opening, a special filler wall of sufficiently high sound insulation shall be built into the opening and the specimen placed in it, the sound power transmitted indirectly by the filler wall and the other parts being negligible against that transmitted through the specimen. 5.2.3 A specimen that has to open and close shall be installed so that it opens and closes freely as in use, and shall be opened and closed at least five times before the test. 5.2.4 Where a door is set into the test opening, its threshold shall be as close as possible to the laboratory floor. 5.2.5 For glazing, windows and doors the specimen area means the area of the opening in the filler wall in which the specimen is mounted.

5.2.6 For some glazing systems or components the sound insulation depends strongly on the temperature at which it is measured; for such specimens the temperature in the source room and the receiving room is recommended to be held at 20 °C +/-3 °C and the specimen is to be kept at that temperature for 24 h before the test, and it is better still to measure at the temperature for which the specimen was designed. A note adds that because the sound insulation of windows, doors and small external wall components depends on size, a component in a building whose size differs from that of the laboratory specimen will give a noticeably different sound reduction index; where the areas differ by a factor of about two the single number index generally differs by not more than 3 dB, and a larger real area generally gives a lower index, an accurate and reliable result being obtainable only by measuring a specimen of the same size.

5.2.7 Window sashes shall be installed as close as possible to the real application; when a window is fitted into the test opening the niche depths on the two sides should not be equal, unless the particular design of the window makes this impossible, the ratio being 2 to 1 as in Figure 1. The gap between the window and the test opening, about 10 mm to 13 mm at the window edge, shall be filled with a sound absorbing material such as mineral wool and sealed on both sides with an elastic sealant, or sealed as the window manufacturer's instructions require.

5.2.8 Glazing installed in the test opening shall also keep a niche depth ratio of 2 to 1 as in Figure 1, with a 10 mm gap left between the pane and the opening. The specimen shall be fixed with two wooden beads of 25 mm by 25 mm, with about 5 mm of sealing compound between the glass and the beads, the beads overlapping the glass by 12 mm to 15 mm. A first note says that the sound insulation of the glass does not represent that of a window made from it, which is obtained by measuring the whole window; a second note says that mounting and sealing the glass in the test opening, although not the arrangement used in real work, is a practicable, quick and repeatable method.

6 Measurement conditions

6.1 The laboratory test facility shall meet GB/T 19889.1-2005. 6.2.1 The sound produced in the source room shall be steady and shall have a continuous spectrum over the frequency range considered; where a filter is used its bandwidth shall be at least one third octave. Where broadband noise is used its spectrum shall give a suitable signal to noise ratio in the receiving room at high frequencies as well, white noise being recommended. In every case the difference in sound pressure level between adjacent one third octave bands of the source room spectrum shall be not more than 6 dB.

6.2.2 The sound pressure level in the receiving room shall exceed the background noise by more than 15 dB in every band, and where that cannot be met the correction of 8.2 shall be applied; the correction for background noise is actually given in 8.5 of the document, and the cross reference is translated here as printed. 6.2.3 Where several loudspeakers in a source enclosure work at the same time they shall be driven in phase.

6.2.4 Where another method is used to give the loudspeakers the uniform non-directional radiation required by A.1.3, the following apply. Where several sources may be used at the same time, they shall be of the same type, driven at the same level and radiating uncorrelated signals; a continuously moving source may also be used; where a single source is used it shall have at least two positions. Those loudspeaker positions may be in the same room, or the source room and the receiving room may be interchanged and the measurement repeated in the opposite direction, in which case one or more source positions are used in each room; where one face of the specimen absorbs markedly more than the other, the measurement may be made in one direction only. The loudspeaker shall be placed where it gives as diffuse a sound field as possible and at some distance from the specimen, so that the direct sound is not prominent; the sound field in a room depends strongly on the type and position of the source, and the technical requirements for the loudspeaker and the measurement point positions shall meet Annex A, the use of a moving loudspeaker being described in A.2.5.

7 Measurement requirements

7.1.1 The average sound pressure level may be obtained with a single microphone used at different positions, with a fixed array of microphones, with a single microphone moved continuously, or with a rotating microphone. For all source positions the levels measured at the different points shall be averaged on an energy basis, as in 8.1.

7.1.2 At least five microphone positions shall be used in the source room and in the receiving room, distributed according to the space available and spread evenly through the largest permitted measurement volume of each room, the detailed requirements being given in Annex A. The minimum spacings are more than 0.7 m between microphones, more than 0.7 m between a microphone and a room boundary or a diffuser, more than 1.0 m between any microphone and the source, and more than 1.0 m between any microphone and the specimen. Where a single moving microphone is used the sweep radius shall be more than 1 m, the plane swept shall be inclined by more than 10° to any surface of the room, and the sweep period shall be not less than 15 s.

7.1.3 At each microphone position the averaging time shall be at least 6 s for bands with a centre frequency below 400 Hz; for bands of higher centre frequency it may be somewhat shorter but not less than 4 s. Where a moving microphone is used the averaging time is the total over all the sweep points and shall be not less than 30 s.

7.2 When the sound pressure levels are measured with one third octave band filters, at least the eighteen centre frequencies of Table 1 shall be covered. Table 1 is a single list of frequencies in hertz and runs 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1 000, 1 250, 1 600, 2 000, 2 500, 3 150, 4 000 and 5 000. 7.3 The measurement shall be sufficiently repeatable, the repeatability being determined by the method of GB/T 19889.2, and shall be verified whenever the test procedure or the equipment changes.

8 Processing of the measurement data

8.1 Three formulas are given for the average sound pressure level in a room: one for a continuously moving microphone, whose legend names the sound pressure in pascals, the reference sound pressure of 20 micropascals and the integration time in seconds; one for a number of fixed microphone positions, whose legend names the root mean square sound pressures measured at the n positions; and one for the usual practical case in which a number of sound pressure levels have been measured, whose legend names the levels at the n measurement points.

8.2 The sound reduction index is given first as the ratio of the sound power incident on the specimen to the sound power transmitted through it, and then, where the source room and the receiving room are both diffuse fields, as the difference of the average sound pressure levels in the two rooms plus a term in the specimen area and the equivalent absorption area of the receiving room. The legend names the average level in the source room and in the receiving room in decibels, the specimen area, equal to the area of the test opening, in square metres, and the absorption area of the receiving room in square metres. The legend of the first formula gives the two sound powers in decibels, which is how the original is printed. A note says that in deriving the second formula from the first the sound field is assumed to be completely diffuse and the sound from the source is assumed to reach the receiving room only through the specimen.

8.3 Where the sound power transmitted into the receiving room by flanking or other elements is also appreciable, the apparent sound reduction index is given as the ratio of the incident sound power to the sum of the power transmitted through the specimen and the power transmitted by the other paths; the legend again gives the three sound powers in decibels. Where the transmitted power is made up of several parts and both rooms are diffuse fields, the apparent index is estimated by the same level difference formula as in 8.2. Whatever the actual transmission, in the apparent sound reduction index the power transmitted into the receiving room is related only to the power incident on the specimen.

8.4 The absorption area in the correction term is calculated from the reverberation time measured according to ISO 354 by a formula whose legend names the absorption area in square metres, the volume of the receiving room in cubic metres and the reverberation time in seconds, the numerical coefficient being 0.16. Following ISO 354, the starting point of the decay curve evaluation is about 0.1 s after the source is switched off, or a few decibels down the start of the decay; the evaluation range shall be not less than 20 dB and should not exceed 30 dB, or the decay in the observed range departs from a straight line, and the lower end of the range shall be more than 10 dB above the background noise. At least six decay measurements shall be made in each band, with at least one loudspeaker position and three microphone positions each read twice; where a moving microphone meeting 7.1.2 is used, its travel time shall be not less than 30 s.

8.5 Background noise shall be measured to make sure that the result in the receiving room is not affected by intruding sound, such as noise from outside the test rooms, electrical noise in the receiving system, or electrical crosstalk between the source and the receiving system; crosstalk may be checked by replacing the microphone with a dummy microphone or with an equivalent impedance. The background level shall be at least 6 dB and preferably more than 15 dB below the total level of signal and background noise together. Where the difference lies between 6 dB and 15 dB, the level is corrected by a formula whose legend names the corrected signal level, the combined level of signal and background noise and the background noise level, all in decibels. Where the difference in any band is not more than 6 dB, a correction of 1.3 dB, the value for a difference of 6 dB, is applied, and the test report shall then state clearly that the sound reduction index given is a measurement limit.

8.6 Following ISO 717-1:2020, the measured data are compared with the reference data of Table 2 by shifting the reference curve until the sum of the unfavourable deviations, calculated by formula (10), is as close as possible to 32 dB without exceeding it; the corresponding shift is then added to the reference value of 52 dB at 500 Hz to give the weighted sound reduction index. Table 2 has three columns and sixteen rows and gives, for band numbers 1 to 16, the centre frequencies 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1 000, 1 250, 1 600, 2 000, 2 500 and 3 150 Hz and the airborne sound reference values 33, 36, 39, 42, 45, 48, 51, 52, 53, 54, 55, 56, 56, 56, 56 and 56 dB. The legend of formula (10) names the band number running from 1 to 16 and covering the sixteen one third octave bands from 100 Hz to 3 150 Hz, the unfavourable deviation in decibels, the reference value of the ith band, the shift, positive upwards and negative downwards, and the measured value of the ith band in decibels to one decimal place; the legend calls the reference value the value of the ith band in Table 1, whereas the reference values are in Table 2, and it is translated here as printed.

8.7 All the sound reduction index values, to one decimal place, shall be given in a table and as a curve; in the figure of the test report the ordinate is the sound reduction index in decibels and the abscissa is frequency in hertz on a logarithmic scale, with 5 mm for one third octave and 20 mm for 10 dB. Where the assessment needs octave band data, the octave band value may be calculated from the three one third octave values inside the octave by formula (11), whose legend names the octave band sound reduction index and the sound reduction index of the nth one third octave band, both in decibels; the printed formula lost its logarithm operator in the extraction and is therefore not reconstructed here. Where the test procedure is repeated in the same or the opposite measurement direction, the arithmetic mean of all the results for each band may be taken.

9 Test report

9.1 The test report shall contain the number of this document; the name and address of the testing laboratory; the name of the manufacturer and the product model; the name and address of the client; the test date; a sectional drawing of the specimen and a description of how it was installed, including its size, thickness and mass per unit area and the curing time and conditions of each component, together with the name of the party that installed it, the works or the testing body; a detailed description of the test opening; the volumes of the two reverberation rooms; the air temperature and relative humidity of the test rooms; the sound reduction index of the specimen against frequency; and a short description of the measurement procedure and of the equipment.

9.1 also requires the report to state the results obtained under any measurement limit: where the sound pressure level in a band cannot be measured because of background noise, acoustic or electrical, as in 8.5, or where the measurement of the sound reduction index is affected by flanking transmission, the result may be given as a limit in the form of an apparent index greater than or equal to a stated number of decibels, the appropriate maximum apparent index being given in the second case. The report format is given in Annex B. 9.2 The frequency against sound reduction index curve shall be evaluated by the single weighted sound reduction index, and the report shall state that the result was obtained by the laboratory method.

A Annex A (normative) Technical requirements for the sound source and its arrangement

A.1.1 The position and the directivity of the loudspeaker shall put the microphones outside the direct field of the source and shall keep the sound radiated directly by the source from dominating at the surface of the specimen. The requirement on the radiation characteristic depends on the size of the source room; for uniform non-directional radiation the minimum distance from the source to the specimen and to any microphone is calculated by formula (A.1), whose legend names that distance in metres, the volume of the room in cubic metres and the reverberation time in seconds. The distance from the source to the specimen and to any microphone may be taken as twice that minimum; where the source meets the uniform non-directional single source requirement of A.1.3 the spacings of 7.1.2.2 shall be met, and for other types of source the position shall meet A.1.2.

A.1.2 The microphone positions shall lie outside the direct field of the source, which may be verified by recording how the sound pressure level changes as the microphone moves along the straight line from the surface of the source to the chosen microphone position; the test shall be made for all one third octave bands with a centre frequency above 630 Hz. Each fixed microphone position shall lie outside the region where the level falls noticeably as the distance from the source increases, and for a moving microphone the level measured near the source shall not rise noticeably.

A.1.3 For all source positions in the free space of the room the loudspeaker shall be a driver mounted in a closed box, and all the drivers in one source box shall radiate in phase; drivers mounted on the faces of a polyhedron, preferably a dodecahedron, give approximately uniform non-directional radiation. To test the directivity of a source, the sound pressure level is measured at about 1.5 m in a free field with the source driven by a noise signal, in one third octave bands, and the difference is taken between the energy average over 360° and each of the sliding averages over 30°; the sentence that describes this is left incomplete in the original. The directivity index is calculated by formula (A.2), whose legend names the ith directivity index, the energy average over 360° and the ith sliding average over 30°. The source may be assumed to be non-directional if the directivity index lies within +/-2 dB over the bands from 100 Hz to 630 Hz; from 630 Hz to 1 000 Hz the limit runs from +/-2 dB to +/-8 dB, and from 1 000 Hz to 5 000 Hz it is +/-8 dB. Where the test is made in different planes the worst condition shall be included; for a polyhedral source a test in one plane is enough.

A.2.1 The check is made by exciting the particular modes of the source room, so that the chosen source positions give a sound insulation result as close as possible to the average of a large number of evenly distributed positions in the room. When the selected loudspeaker positions are used for real sound insulation measurements, the loudspeaker model and directivity shall be the same as those used in the qualification test, and this applies to all the characteristics of the laboratory, including the microphone positions or microphone travel paths, the diffusers, the absorbing surfaces and even the position of the specimen, particularly when it is mounted in a filler wall in a test opening.

A.2.2 In the selection procedure, two different loudspeaker positions shall be not less than 0.7 m apart and at least two positions shall be not less than 1.4 m apart; the centre of the source shall be not less than 0.7 m from a room boundary, small irregularities of the interior surfaces not counting, and source positions close to a boundary, particularly in a corner, are covered by A.2.3. The loudspeakers shall not be arranged symmetrically about the central axis or central plane of the source room where the boundaries are parallel, and different loudspeaker positions shall not lie in the same plane parallel to a room boundary, the minimum distance between such planes being 0.1 m. Where a non-directional source is not used, the orientation of the loudspeaker shall be recorded and shall be the same at all positions.

A.2.3.1 The number of loudspeaker positions and a set of optimum measurement positions are obtained as follows. The sound pressure level differences are measured at m loudspeaker positions, the number m being calculated by formula (A.3) from the volume of the source room in cubic metres. The positions are chosen according to A.2.2; where the minimum distance between any two measurement positions has to be less than 0.8 m, the positions shall be spread as evenly as possible so that the minimum distance between two positions is as large as possible while the other requirements of A.2.2 are still met, the printed figure of 0.8 m differing from the 0.7 m of A.2.2. For each loudspeaker position the level difference between the source room and the receiving room is measured in each one third octave band between 100 Hz and 315 Hz and the standard deviation of those differences is calculated by formula (A.4), whose legend names the level difference in the ith band at the jth loudspeaker position, the arithmetic mean of the level differences in the ith band and the number of positions under test.

A.2.3.1 continues: the number of loudspeaker positions to be used in practice is calculated by formulas (A.5) to (A.7), which require it to be at least two and relate it to the standard deviations of the level differences and to the specified maximum standard deviation of the mean over the chosen positions. The standard deviations obtained from formula (A.4) shall be lower than the maximum standard deviations listed in Table A.1 for all the one third octave bands concerned; Table A.1 has two columns and six rows and gives, in decibels, 1.4 at 100 Hz, 1.2 at 125 Hz, 1.0 at 160 Hz and 0.8 at 200 Hz, 250 Hz and 315 Hz. Where twice the number of positions to be used in practice exceeds the number of positions under test, that number is increased to twice the practical number, the added points being chosen so that the requirements of formulas (A.5) to (A.7) are met for the enlarged set. For each loudspeaker position the sum of the squared deviations of the six one third octave band level differences from their means is calculated by formula (A.8), whose legend names that sum, the level difference in the ith band at the jth position and the arithmetic mean of the level differences in the ith band; the summation index of the printed formula is the position index j where the band index i is meant, and the formula is not reconstructed here. From all the positions under test, the q positions with the smallest sums are selected.

A.2.3.2 For additional loudspeaker positions that do not satisfy A.2.2, the following apply: an additional position may be used in practice if its sum of squared deviations does not exceed the largest of the q selected positions; q positions are then chosen, with q at least two; for each combination of q points the total of the squared deviations of the six one third octave band level differences from their means is calculated and the combination with the smallest total is selected; and at least two of the selected positions shall be not less than 1.4 m apart. A note says that where loudspeaker positions close to a boundary are chosen, the results measured at those positions can be repeated very accurately.

A.2.4 The qualification test uses a specimen whose sound reduction index is not lower than the values of Table A.2 and whose size meets the requirements of Annex A of GB/T 19889.1-2005 for glazing; Table A.2 has two columns and six rows and gives, in decibels, 27 at 100 Hz, 28 at 125 Hz, 29 at 160 Hz, 30 at 200 Hz, 31 at 250 Hz and 32 at 315 Hz. The specimen used is a single thin sheet made of a sandwich steel panel of steel sheet, resin and steel sheet, fixed to channel section framing with screws and an elastic plastic sealant. A first note says that the sound insulation of this specimen is not affected by resonance over the whole range below 5 000 Hz, so that it also suits the routine repeatability test recommended in Clause 7; a second note says that where a laboratory does not usually test this type of specimen, a representative specimen in common use is taken instead.

A.2.5 Where the sound pressure levels in the source room and the receiving room are measured at the same time, a loudspeaker moving automatically along a path may be used; the path shall be not less than 1.6 m long and the loudspeaker shall be non-directional, and all the positions on the path nearest to the microphones shall be verified by the qualification procedure of A.1.2. The sound reduction index of the specimen shall be measured by the procedure of A.2.4 along a number of paths, including four diagonals crossing the part of the room volume that satisfies A.2.2, and the path with the smallest sum of squared deviations from formula (A.8) shall be used in the actual measurement.

Annex B gives the format of the sound reduction index test report as Table B.1. The form carries the statement that the results follow GB/T 43944-2024, and fields for the manufacturer and the product identification, the client and the laboratory accreditation number, the party that installed the specimen and the test date, a description of the test facility and of the specimen and its installation, the specimen area and its surface density, the laboratory temperature and relative humidity, the atmospheric pressure and the receiving room volume, then a two column list of the average sound reduction index in decibels at sixteen one third octave centre frequencies from 100 Hz to 3 150 Hz, a line for the weighted sound reduction index obtained by the laboratory method, and fields for the report number, the name of the body, the report date and the signature.