Data for sound pressure level prediction in lightweight constructions caused by structure-borne sound sources and their uncertainties

When predicting sound pressure levels induced by structure-borne sound sources and describing the sound propagation path through the building structure as exactly as possible, it is necessary to characterize the vibration behavior of the structure-borne sound sources. In this investigation, the characterization of structure-borne sound sources was performed using the two-stage method (TSM) described in EN 15657. Four different structure-borne sound sources were characterized and subsequently installed in a lightweight test stand. The resulting sound pressure levels in an adjacent receiving room were measured. In the second step, sound pressure levels were predicted according to EN 12354-5 based on the parameters of the structure-borne sound sources. Subsequently, the predicted and the measured sound pressure levels were compared to obtain reliable statements on the achievable accuracy when using source quantities determined by TSM with this prediction method. In addition to the co-submitted article (Vogel et al., 2023), the sound pressure level prediction according to EN 12354-5 in detail is described. Furthermore, all data used are provided.

In addition to the co-submitted article (Vogel et al., 2023), the sound pressure level prediction according to EN 12354-5 in detail is described. Furthermore, all data used are provided.
© 2023 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ) Table   Subject Civil and Structural Engineering, building acoustic Specific subject area Sound pressure level prediction, structure-borne sound sources, lightweight constructions, uncertainties Type of data Tables; images How the data were acquired

Specifications
The data were measured by microphones and acceleration meters. For measuring plate mobilities an electrodynamic shaker was used to excite the reception plates. For the characterization of the structure-borne sound sources, only surface velocities on reception plates were measured. For those measurements a laser Doppler Vibrometer was also used. Data format Analyzed Description of data collection The data file "Data for calculation.xlsx" contains all numerical values necessary for the calculation of the sound pressure level caused by the structure-borne sound sources. The manuscript also contains a detailed sketch of the building elements considered.
The provided values were measured in small frequency bands ( f = 1 Hz) as well as third-octave bands. To determine the source parameters, the raw data were measured exclusively in small frequency bands to calculate the source parameters from these values. Subsequently, the calculated source parameters were converted into third-octave band values (see Table 1 in the article and the data file). The description of the individual data is given at the top of each

Value of the Data
• The calculation and dataset presented in this article allow other researchers, especially acousticians, to conduct further calculations to reduce the uncertainties of the prediction method. For example: using frequency depending on radiation efficiency as well as new information concerning the sound propagation in buildings, and simulation of the investigated setup. • This full dataset of a sound pressure level prediction provides also detailed information about the structure especially the walls in the test stand • This full dataset of a sound pressure level prediction caused by structure-borne sound sources provides detailed information about the characterized values of the structureborne sound sources • This dataset illustrates the difference between predicted and measured uncertainties to specific frequencies as well as to single values representing the whole frequency range (total sum, arithmetic mean, A-weighted sum levels, etc.)

Objective
The supported article [1] presents analysis, discussions, and insights into the data and measurement method of the two-stage method (TSM) while characterizing a shaker, compressor, extractor fan, and ventilation unit (typical structure-borne sound sources). To determine the uncertainties of the predicted sound pressure levels based on these source parameters, subsequently the sound pressure levels were measured in a lightweight test stand by mounting the sound sources on a flanking wall and compared with predicted data. This article presents the full dataset of these sound pressure level predictions due to the four structure-borne sound sources including the measured data of the source characterization with TSM and all necessary data.

Data Description
In [6] the full dataset used for the sound pressure level prediction is provided. The data consist of numerical values and related formulas, which are necessary for the sound pressure level prediction in rooms due to structure-borne sound sources. The data also characterize the building elements of a lightweight test stand and the vibrational behavior of the sources used. Fig. 1 shows the lightweight test stand, sketches and dimensions, where the measurement of the data was done. Table 1 shows the characteristic structure-borne sound source parameters v f , F b , and Y s . Table 2 shows constant parameters and room dimensions. Table 3 shows the receiving mobility Y r and Table 4 the resulting coupling term D C,i for each source. Table 5 shows the adjustment term D as,i and installed structure-borne sound power L Ws,inst,i . Table 6 provides the sound reduction index R i of the walls. Table 7 contains the structural reverberation time T s,i of the walls. shows the equivalent absorption length a i Table 8 . Table 9 contains the directionaveraged junction velocity level difference . Table 10 provides the reverberation time T 60 and equivalent absorption area A of the source and receiving rooms. Table 11 contains the vibration reduction indices K ij ; the flanking sound reduction index R ij and the flanking sound reduction coefficient R ij,ref . Table 12 provides the sound pressure levels L n,s,ij for paths 1 and 2 Fig. 1. Left -lightweight test stand [3] ; middle, top -flanking wall with tiled section in source room; middle, bottom -separating wall in source room; right top -construction of the flanking walls; right bottom -construction of the separating wall.

Table 1
Characteristic structure-borne sound parameters of the sources used, measured with two-stage method according to [2] . and the resulting sum L n,s in the receiving room, predicted and measured values. Table 13 shows the differences between the predicted and measured normalized sound pressure levels L n,s in the receiving room as mean values across all investigated sources. Table 14 shows the list of measurement equipment.    Table 5 Adjustment term D as,i ; installed structure-borne sound power lever L Ws,inst,i on the flanking wall in the source room.

Experimental Design, Materials and Methods
The data article presents the prediction method including all necessary data concerning the structure-borne sound sources (compressor, shaker, ventilation unit, and extractor fan) and the sound pressure level prediction. The structure-borne sound source characterization was done by the two-stage method, according to [2] . Therefore, the sources were mounted on a heavy and a light reception plate (approx. 3 -5 m ²) and were switched on. The induced structure-borne sound power was determined on the plate surfaces using the measured surface velocity. Using the two reception stages heavy and light one can make simplifications regarding the receiver mobility (very high or very low compared to the source mobility) and this yields to installationindependent source parameters. Detailed information about the structure-borne sound source characterization method itself is provided in [2] . In [1] , the characterization of the sources used is described in detail. The determined source parameters free velocity v f , blocked force F b , and source mobility Y S are provided in Table 1 . All measured data used for the investigation of the sound pressure level prediction were measured in a lightweight test stand at Working Group 1.72 Applied Acoustics, PTB Braunschweig.

Data of the Characterized Structure-borne Sound Sources
Using to the measurement method described above Table 1 provides the measured installation independent source parameters.

Lightweight Test Stand at PTB Braunschweig
The lightweight test stand at the PTB in Braunschweig is a wooden plate construction with a length of 7.10 m and a width of 3.25 m. There are two adjacent rooms on each of the two floors with a room height of 2.55 m, so that sound transmission can be reproduced horizontally, vertically, and diagonally with a coupling of structure-borne sound sources to the partition wall or flanking elements.
The perimeter walls are made of 60-mm x 80-mm timber studs spaced 625 mm apart and filled with 80-mm mineral wool. On the outside, these flank walls are covered with 13-mm chipboard, and on the inside with 13-mm chipboard and 12.5-mm plasterboard. In the interior wall area of one of the flanking walls, on which the structure-borne sound sources were mounted for this investigation, there was also a partially tiled section of approx. 0.80 m x 2.00 m.
The substrate of the tiles (plasterboard) was first treated with deep primer before the tile adhesive was applied, so that it does not lose all of its moisture and thus its adhesive strength on the wall. Then the tiles were glued and grouted. The tiles are standard bathroom tiles with the dimensions 20 × 25 [cm] and a weight of approx. 750 g per tile.
Both separating walls, one per floor, consist of a 60-mm x 155-mm wooden framework that is also filled with 80-mm mineral wool. They are covered on both sides with 13-mm chipboard and 12.5-mm plasterboard. On both floors, the separating walls are butt-jointed to the flanking exterior walls (with continuous planking) and arranged offset to each other on each floor, so that all element connections of the lightweight test stand are always designed as T-joints (no cross-joints existing).
The upper ceiling, which closes off the test stand, is constructed in the same way as the surrounding perimeter walls. The bottom floor consists of a reinforced concrete floor slab on which 20-mm polystyrene, 10-mm wood fiber insulation, and 20-mm Fermacell gypsum fiber boards are laid from bottom to top. The separating ceiling is designed as a typical wooden beam ceiling with 180-mm high ceiling beams. On top, it is finished from bottom to top with a 22-mm flat pressed board, 30-mm mineral fill, a 10-mm wood fiber insulation board, and a 20-mm Fermacell gypsum fiberboard. Since the three investigated sound sources as well as the shaker were connected to a flanking exterior wall on the upper floor and the standardized sound pressure level was only investigated in the neighbor receiving room, the ceiling components were neglected in the prediction according to EN 12354-5. The relevant dimensions of the lightweight test stand and the element constructions are shown in Figs. 1 and 2 . Here, the dimensions of the separating and flanking walls differ, because a higher sound reduction index of the separating wall was chosen.

Prediction Method
The normalized sound pressure level L n,s in the receiving room induced by structure-borne sound sources is predicted with a prediction method according to [4 , 5] . The equations and the full data set for the prediction are given in this section.    The sound reduction index of the flanking walls is taken from the measurement of the sound reduction index of the separating wall, the constructions are similar.
Since the existing joints were not sufficiently known, the vibration reduction indices K ij were determined experimentally according to Equation 10 by measuring the velocity level differences D v,ij, and D v,ji ( Table 9 ) for the relevant transmission paths. The equivalent absorption lengths a i and a j ( Table 8 ) were calculated using Equation 11 and the measured structure-borne sound reverberation time T s,i .  Table 11 presents the vibration reduction index, the flanking sound reduction index, and the flanking sound reduction coefficient for both transmission paths, Ff and Fd.

Comparison of Predicted and Measured Sound Pressure Levels
Columns 1 and 2 of Table 12 contain the predicted normalized sound pressure level components of the individual transmission paths L n,s,ij . Columns 3 and 4 contain the energetic sum of columns 1 and 2 as normalized sound pressure level L n,s in the receiving room. Columns 5 and 6 show the measured values of the normalized sound pressure level L n,s in the receiving room. Table 13 shows the differences between the predicted and the measured values of the normalized sound pressure levels. The values represent the energetic mean value across all investigated sources. Since the shaker is an ideal source of structure-borne sound for the characterization and prognosis method (punctiform one-point contact with the receiving structure), it cannot be regarded as a common source of structure-borne sound. Therefore, the deviations are shown with (columns 3 and 4) as well as without the shaker (columns 5 and 6). For the frequency range relevant to building acoustics in Germany (normative requirements of 100 -3150 Hz), the A-weighted total level results in an average deviation of 5.2 dB, and the arithmetic mean value of all 16 single third-octave band differences is 7.0 dB. It must be discussed, which frequency range is valid and if the levels must be A-weighted because of the typical acting of structure-borne sound sources in the low and very low frequency range and because of their tonal behavior, which can be very disturbing.

Measurement Equipment
In Table 14 , the main components of the measurement equipment are listed, which were used for the investigation of the characterization method, the characterization of the sources, and the sound pressure level measurements.

Ethics Statements
No ethical issues are associated with this work.

Declaration of Competing Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability
Data for sound pressure level prediction (Original data) (Mendeley Data).