Concept and development of a novel timber spring for impact sound reduction in timber floor slabs

Abstract

A novel concept for enhancing timber floor slabs with integrated acoustic decoupling is introduced. The approach was evaluated through a series of measurements on both small-scale and large-scale setups. The tests involved a dual-shell timber slab system with discontinuous timber beam elements within the cavity. The timber beam elements are used to integrate acoustic decoupling into the structural system achieved through the use of targeted processing and geometric modification. The results show up to 5 dB improvement in the weighted normalized impact sound pressure level $L_{n, w}$ and the sound reduction index $R$ compared to the unmodified reference system, without the addition of mass. This demonstrates the potential for improving sound insulation in lightweight timber structures, also in the low-frequency range, while maintaining minimal material use, offering significant implications for the design and construction of future building systems.

Keywords

Acoustic decoupling building acoustics performance impact sound reduction low-frequency sound timber floor slabs timber spring

Introduction

Timber is increasingly being used in construction, offering a sustainable alternative to traditional building materials such as reinforced concrete, also with regard to multistorey buildings.^1–4 This growing trend has led to a steady expansion in the use of timber in various building construction types. Timber construction methods have evolved to meet the demands of modern architecture, with developments such as mass timber structures, hollow box systems and ribbed ceiling slabs enabling spans of more than 5 m.⁵ However, it has been observed that the low weight of timber poses unique challenges for acoustics and sound insulation.^6,7

The acoustic quality at low frequencies in multistorey timber buildings is particularly challenging due to the distinct properties of timber. As a porous material with lower density and stiffness compared to other building materials, such as concrete or masonry, timber provides reduced sound attenuation at low frequencies, leading to sound transmission issues and noise disturbances that affect user comfort. The propagated low-frequency sound is often a result of impact noise.^3,6,8

To address this challenge, various design strategies are employed to enhance the acoustic quality of timber buildings. Acoustic improvements are typically achieved by adding mass and incorporating multi-layered surface structures. These may include the use of mineral materials such as gravel, cementitious screed and mineral wool to create weight-bearing surface structures, fillings and suspended ceilings.¹ Figure 1 shows an example of a typical timber slab assembled from cross-laminated timber (CLT) along with acoustic improvements.^3,9 While the CLT plate has a thickness of 140 mm, the total height of the ceiling slab’s cross section increases by 230 mm, resulting in a total height of 370 mm due to these improvements.

Figure 1.

Example for a conventional wooden floor structure consisting of CLT (140 mm), a heavy weight surface structure (gravel, mineral wool, screed) and a suspended ceiling.^3,9

These known approaches of applying additional layers are based on practices in reinforced concrete slabs and do not show the expected efficiency for timber structures, as they can also lead to a deterioration in the acoustic performance of timber slabs in the low-frequency range.¹⁰ The introduction of certain acoustic treatments can sometimes exacerbate sound transmission issues if not carefully designed and implemented.

In addition, acoustic enhancements in timber slabs can substantially affect the overall environmental impact of building systems. Therefore, a significant correlation exists between the acoustic performance of floor structures and the environmental profile of the entire building. As a natural raw material, wood has the unique ability to sequester $C O_{2}$ from the atmosphere during its growth, which contributes to a reduction in overall $C O_{2}$ levels.¹¹ This carbon sequestration is an important factor that is taken into account in the life cycle assessment of wood products. The use of mineral materials to improve acoustic performance increases the global warming potential (GWP) of the structure due to their higher embodied carbon.³ Depending on the specific wood building system used and the acoustic enhancements implemented, different substitution potentials arise. For example, the ecological improvement potential is particularly notable for timber slabs with heavy mineral elements or floating screeds.¹¹ This underscores the need for a balanced approach that integrates both acoustic performance and environmental considerations. Therefore, selecting and dimensioning acoustic treatments should be done with careful consideration of their acoustic benefits and their potential impact on the overall environmental profile.³

This paper investigates the potential for hollow core timber slab structures to be fitted with structurally engineered acoustic performance features following the principle of a leaf spring. The goal is to enhance sound insulation while simultaneously eliminating the need for mineral weights.

State of the art

In practice, the acoustic performance of timber slabs, particularly regarding impact sound transmission and vibration behaviour, is typically addressed only after the structural design phase. As a result, wooden floor slabs are designed with mass-based layers. This not only increases the cross-sectional thickness but also adds significant weight to the structure. However, these measures are suboptimal, as they fail to effectively improve impact sound transmission in the critical low-frequency range below 100 Hz. To achieve targeted improvements in sound transmission, it is essential to analyse the frequency-dependent stiffness properties of the materials used in conjunction with the overall structural dynamics of the system, with a particular focus on different frequency ranges.⁸ This is especially relevant for cross-laminated timber (CLT), considering its layered substructure and the differing properties of timber along and perpendicular to the fibres.^12,13

Solutions for maintaining or improving the impact sound insulation performance of timber slabs by reducing additional mass have been proposed in Kohrmann et al.¹⁴ and in Schoenwald and Vallely.¹⁵ The approach detailed in Kohrmann et al.¹⁴ focuses on reducing vibrations of the bare slab through the installation of tuned mass dampers which are particularly effective in the frequency range below 100 Hz and also have static relevance. These tuned mass dampers improve the vibration behaviour with regard to serviceability and increase the impact sound insulation by damping the sound-emitting ceiling.¹⁴ On the other hand, Schoenwald and Vallely¹⁵ presents methods for minimizing the need for additional mass by applying functional gradings to the ceilings, where the ballast is localized. This approach utilizes CNC (Computer Numerical Control) processes for targeted machining to explore the application of the so-called acoustic black hole principle in this specific context.¹⁵ Both approaches are primarily suitable for solid wood slabs, for example from cross-laminated timber (CLT). Compared to solid CLT, hollow box systems require less material, allow cavity utilization and achieve an overall lower weight.¹⁶

For timber systems with cavity (such as hollow box systems), Málaga-Chuquitaype and Ilkanaev¹⁷ presents an innovative beam configuration with regard to digital manufacturing that enhances the vibration comfort of timber slabs. These narrow, elongated formations within timber beams function as vibration absorbers, resembling cantilever beams. As a result, they effectively dissipate vibration energy from the slab structure while preserving the ecological benefits of timber. However, all these additional measures – whether applied between or on top of the beams and slabs or through the addition of mass – serve only to insulate or dampen the sound or vibration energy already introduced into the building component. Consequently, their effectiveness remains limited, particularly for the critical low-frequency impact sound.

This study pioneers the investigation of integrative acoustic design in the sense of multi-functional structural elements that both connect structurally and decouple acoustically. By minimally invasive CNC processing of wooden elements in the vertical direction (global Z-axis), the dynamic stiffness can be altered in such a way that acoustic performance improvements are achieved, reducing the need for mineral or plastic materials.^8,18 Therefore, this paper explores an entirely new area within the field of building acoustics, addressing a significant gap in knowledge that has, until now, received little research attention.

Experimental investigation

Concept for integrative sound decoupling

The incorporation of acoustic elements is based on the interdisciplinary development of an innovative, flexible, multistorey timber building system as described in Krtschil et al.¹⁶ and Orozco et al.¹⁹ This system is being created through a process that seamlessly incorporates structural, architectural, fabrication, building physics and building services requirements (co-design) without the need for additional layers for each discipline. It merges the advantages of solid timber construction with hollow box systems to create a point-supported slab that is capable of spanning in multiple directions. The design features top and bottom plates with discontinuous timber beams within the cavity, aligned according to principal stress directions in the global XY-plane (Figure 2).¹⁶ The provided cavity between the two slabs can be used for acoustic improvement, for example, by using insulation material.⁸ In this case, the structurally necessary connecting web elements between the plates consequently act as acoustic bridges and are ideally decoupled.⁸ Leveraging digital technologies, new methods are now available to apply and modernize the design concepts of integrative acoustics.

Figure 2.

Cross-section through the centre of the slab in the initial system configuration.

The decoupling is achieved by geometric modifications of the timber web components using CNC processing to cut a defined slot in the component as illustrated in Figure 3. Due to this and a milled recess on top of the web, oscillating arms are generated. The top plate rests on these, ensuring that the vibrations introduced are transmitted into or through the slab structure with reduced amplitude. The resulting targeted and axially differentiated dynamic stiffness creates a spring effect in the vertical direction (along the global Z-axis), similar to the functional principle of leaf springs used in automotive engineering. As a result, the timber components are provided with an acoustic decoupling function in addition to their structural function. The transmission of horizontal shear forces remains largely unaffected. Using computational design and FEM simulation, areas with low out-of-plane shear stress can be identified and therefore be equipped with timber spring elements. This represents a previously unexplored opportunity to integrate acoustic decoupling functionalities into building components, such as slab structures.⁸

Figure 3.

Developed acoustic-functional timber component with geometric modification (timber spring).

The developed acoustic-functional web element builds on the principle of leaf springs. Leaf springs are a conventional type of spring, offering the distinct advantage of serving as both structural and dynamic element. They are mainly used in vehicles for the front and rear axle. Leaf springs can be of various types, with the simplest form being a rectangular cross-section with a constant thickness and functioning as a bending beam.²⁰ This paper focuses on the conceptual investigation and proof-of-concept of this straightforward design, although several variants have been elaborated. Furthermore, since bending vibrations are the primary concern in the examined system, the developed acoustic-functional web element is employed according to the functional principle of a bending resonator in order to reduce sound transmission caused by impact sound. The developed element is therefore presented as a timber spring and will be referred to as such in this paper.

Dimensioning of the developed timber spring

The sound insulation of timber components is determined by various physical aspects. For monolithic planar structures, the sound insulation is primarily dependent on the mass per unit area and on the bending stiffness. The bending stiffness also significantly influences bending wave resonance, as well as the coincidence frequency and eigenfrequencies.²¹

For multi-layered structures, the resonance frequency is also a relevant factor, as the vibration behaviour is described by a mass-spring-mass model. In the case of floor structures, resonances can occur in the substructures such as floating floors or suspended ceilings. The effectiveness of sound insulation in these cases is largely determined by the damping at the resonance frequency. Suitable insulating materials can reduce sound transmission, enabling higher sound insulation with lower mass by decoupling component layers. However, conventional insulating materials face limitations, as current available options do not offer dynamic stiffnesses lower than 6 MN/m³.²¹

The resonant frequency in a mass-spring-mass system is calculated using equation (1). The two outer shells form the surface-related masses $m'_{1}$ and $m'_{2}$ , which are decoupled by a spring with a dynamic stiffness of $s'$ . Typically, the elastic layer consists of pressure-resistant insulation panels, such as impact sound insulation panels.²¹

f_{0} = \frac{1}{2 π} \sqrt{s' \cdot (\frac{1}{m'_{1}} + \frac{1}{m'_{2}})}

(1)

Airborne or impact sound excitation induces vibrations in the mass-spring-mass system, which are particularly pronounced at the resonance frequency $f_{0}$ . To enhance the sound insulation of a slab structure above this frequency, resonance frequencies greater than 100 Hz should be avoided whenever possible.²¹

For the design of the timber spring, 100 Hz is set as the upper limit and serves as the target value for determining the resonant frequency of the overall system. Considering the masses of the upper and lower plate, the required dynamic stiffness is calculated by using equation (1).

Elastic layers, such as impact sound insulation boards, have been designed for flat surfaces. Their spring stiffness, or dynamic stiffness $s'$ , is calculated based on the modulus of elasticity $E_{i n s u l a t i o n}$ , considering the surface area $S$ and the layer thickness $d$ (equation (2)).

s' = \frac{E_{i n s u l a t i o n} \cdot S}{d}

(2)

Neubrand et al.²⁰ describes the calculation of leaf springs. Equation (3) shows the relation between the deflection s and the applied load F at the end of the cantilever beam,²⁰ as illustrated in Figure 4. The deflection $s$ depends on the modulus of elasticity $E_{m a t e r i a l}$ of the respective material, the moment of inertia $I = b \cdot h^{3} / 12$ and the length $l$ . This allows to determine the spring stiffness for one-half of the leaf spring according to equation (4).²⁰

Figure 4.

Functional and calculation principle for rectangular leaf springs.

The required dynamic stiffness of the elastic damping layer is then equated with the force-displacement ratio of a leaf spring, using equations (2) and (4). This creates an interdependence between the length $l$ and the height $h$ of the oscillating cantilever arm (one-half of the leaf spring).

c = \frac{F}{s}

(3)

c_{s} = \frac{E_{m a t e r i a l} \cdot b \cdot h^{3}}{4 \cdot l^{3}}

(4)

In principle, the geometry is parametric and the subtractive CNC milling of the cavities is digitally adapted to the boundary conditions resulting from the respective slab structure. For the measurements, whether in a small-scale or large-scale test setup, the timber springs are dimensioned and fabricated according to the prevailing boundary conditions, as detailed in the following Sections 3.3 and 3.4.

Small-scale prototype

The effectiveness and functionality of the concept is being investigated using a small-scale test setup. The National Metrology Institute of Germany (PTB) in Braunschweig has developed a compact measurement setup (COMET) for determining impact sound reduction, which is standardized by DIN EN ISO 16251-1.²² The main differences to the conventional method in the slab test facility, according to the procedure described in ISO 10140, are that the acceleration or velocity level is measured on the surface of the test specimen and the test area is reduced to 1 m². Therefore, only statements on the frequency-dependent impact sound reduction can be made.²² The COMET was initially developed for measuring floor build-ups on concrete slabs. However, the use of lightweight wooden floor slabs has also become part of the research at PTB. Schmidt and Wittstock,²³ for example, confirmed the comparability of the results of a large ceiling test stand with the COMET measured values for wooden beam-slab structure.²⁴

The assembly of the small-scale test specimen is shown in Figure 5. The bottom plate is inserted first, followed by the web elements and the upper plate. The upper plate has a thickness of 100 mm, while the lower plate is 200 mm thick. All elements are made of softwood cross-laminated timber (CLT, strength class C24²⁵). During the series of measurements, the web elements are replaced by two variants of prototypes of timber springs (Figure 6). To facilitate easy exchanging of the elements between the plates, they are placed without the use of screws or adhesives. As the specimens are tested and compared for their effectiveness with each other in the COMET, screwing them is not necessary. The test setup with unmodified web elements serves as reference system.

Figure 5.

Reference test setup with web elements shown without top plate (left) and reference test setup with measurement equipment (right) on the COMET.

Figure 6.

Variant 1 (left) and 2 (right) of the test specimens installed in the COMET.

Considering 100 Hz as the reference value for the first eigenfrequency, the dimensions of the test specimens for two variants are determined according to the procedure described in Section 3.2. For softwood CLT in strength class C24, the material parameters are $E = 11.000 M P a$ and $ρ = 420 k g / m$ ²⁶. This corresponds to a mass per unit area of 42 kg/m² for the upper plate ( $m'_{1}$ ) and 84 kg/m² for the lower plate ( $m'_{2}$ ). Based on equation (1), the required dynamic stiffness is 11.05 MN/m³. With a beam element width of 80 mm, the dimensions for the timber springs are shown in Figure 7. Variant 2 represents the inverted configuration of Variant 1, designed to assess the influence of the position of the slot on acoustic decoupling.

Figure 7.

Experimental configurations of the small-scale test specimens, Variant 1 (left), Variant 2 (right).

The standard tapping machine is used to excite the structure for impact sound testing. The resulting velocity levels are measured with four acceleration sensors on each of the two panels, as shown in Figure 5. Details regarding the positioning of the measurement equipment are provided in the associated dataset.²⁷ In order to exclude any possible influence and to decouple the slab from the compact test stand construction, the lower plate is placed on elastomer and nonwoven was inserted as an insulating material between the slab and the surrounding construction.

Large-scale prototype

In order to test the findings from the small-scale test on a full-scale slab system and to assess the impact sound behaviour, a wooden slab is installed in a slab test facility with a test area of $S = 4.8 m \cdot 3.8 m = 18.24 m^{2}$ . The slab structure also consists of two plates forming the outer shells: Both the upper and the lower plate are composed of five plies, but with different ply thicknesses. In the upper plate (total thickness of 80 mm), the plies measure 20, 10, 20, 10, and 20 mm. In the lower plate (160 mm total thickness), they are 40, 20, 40, 20, and 40 mm. The cavity between them contains a total of 18 web elements cut from a full five-ply CLT plate, each measuring 1500 mm in length, 200 mm in height and 80 mm in width. All components are made of softwood CLT (strength class C24²⁵).

As mentioned in Section 3.1, the investigated slab system is based on an innovative and flexible timber building system designed for spans of more than 8 m. Due to the different boundary conditions compared to the setup in the COMET, which affect the vibration behaviour, the timber springs have been geometrically adapted according to the procedure described in Section 3.2. The dimensions for the timber springs used in the slab testing rig are presented in Figure 8.

Figure 8.

Experimental configuration of the large-scale test specimen.

The weighted normalized impact sound pressure level $L_{n, w}$ is determined according to the specifications of DIN EN ISO 10140-3.²⁸ Studies on the perception of the impact sound insulation of wooden slabs have shown a low correlation between the subjectively perceived impact sound transmission and the weighted normalized impact sound pressure level $L_{n, w}$ .^6,7,29 For this reason, and because the ISO standards cover a measurement of frequencies down to 50 Hz, the spectrum adaptation term is determined in the extended frequency range, starting at 50 Hz. This enables a more precise assessment of the sound transmission behaviour in the frequency range relevant for wooden slabs. For comparison with the small-scale measurements performed in the COMET, structural velocity measurements were taken at 18 positions on the underside of the lower plate using four roving accelerometers. The accelerometers are distributed across the plate. In addition, the sound reduction index $R$ is measured according to DIN EN ISO 10140-2,³⁰ to characterize the sound insulation properties of the slab system. The precise locations of the sensors and the measurement equipment are documented in the associated dataset.²⁷

Two series of measurements are conducted. In the first series, the dual-shell system with wooden web elements in the cavity is investigated, which serves as a reference for the adapted slab system with timber spring elements, similar to the investigations in the COMET. In the second series, the web elements are replaced by the timber springs. The components are screwed together to enable disassembly. Figures 9 and 10 show the assembly of the two system variants and Figure 11 pictures the measurement equipment. For the impact sound testing, a tapping machine is used for excitation on five different positions on the structure. A loudspeaker is used to generate a diffuse sound field in the source room. The sound pressure levels are then measured in both the source and the receiving room with rotating microphones. The sound reduction index $R$ is calculated on the basis of the difference between the sound pressure levels, taking into account the equivalent absorption area.³⁰

Figure 9.

Assembly of the dual-shell slab system with beam elements in the slab testing rig.

Figure 10.

Assembly of the dual-shell slab system with timber springs in the slab testing rig.

Figure 11.

Measurement equipment in the source (left) and receiving room (right) of the slab testing rig.

Measurement results

The presented acoustic-functional web element uses the principle of leaf springs to improve sound transmission in dual-shell timber slab structures caused by impact sound. The effectiveness is analysed based on measurements conducted in both small-scale and large-scale test setups.

Small-scale testing on the COMET

The acceleration sensors on the two plates measure the structural velocity $ν$ .

The difference in velocity levels $Δ L_{ν}$ is calculated for each of the two variants by subtracting the velocity levels $L_{ν}$ measured at the lower plate from those of the reference element. In Figure 12 the spectra of the velocity level differences $Δ L_{ν}$ are presented.

Figure 12.

Difference in velocity levels $Δ L_{v}$ at the lower plate, determined for Variant 1 and Variant 2 in comparison to the reference element, as measured in the COMET.

The results of the velocity level differences $Δ L_{ν}$ for the variants investigated show considerable deviations. In Figure 12 it is demonstrated that, in the reference system, the structural velocities of the upper and lower plate differ by a maximum of 4 dB in the frequency range up to 63 Hz. Variant 1 provides the greatest level difference between the upper and lower plate. The values are up to 20 dB higher compared to the reference system, with a deterioration of 3 dB at 25 Hz. The velocity level difference of Variant 2 closely aligns with the results of the reference system in the frequency range of 63–80 Hz. Deviations of up to 10 dB become apparent at frequencies between 80 and 500 Hz. In the higher frequency range, the values of the differences approximate those of the reference system again, similar to the values of Variant 1.

The results show a reduction in the transmitted sound energy from the upper to the lower plate in all variants examined. One contributing factor is the web elements themselves. However, the milled slots on the web elements lead to greater level differences, which demonstrates the functionality of the concept in terms of a decoupling function. Variant 1, in which the slot was positioned in the upper area of the web element, results in a lower dynamic stiffness of the spring compared to Variant 2 and demonstrates a significant improvement, particularly in the low-frequency range.

Due to the influences of bearing and excitation, the results in the very low-frequency range around 25 Hz are subject to greater uncertainties and should be interpreted with caution. Nevertheless, the COMET is suitable for carrying out initial feasibility studies of this kind.

Large-scale testing in the slab testing rig

To assess effectiveness of the timber springs on the impact sound pressure level, the findings from the small-scale test setup are evaluated through testing on a large-scale setup. The structure is excited by positioning a tapping machine at five different locations. The sound pressure level in the receiving room is recorded for each position. Additionally, the reverberation time is measured to determine the equivalent sound absorption area of the room.

The measurement results for both series are compared in Figures 13 –15. Figure 13 illustrates the difference in velocity levels $Δ L_{ν}$ measured on the lower plate, highlighting the improvement achieved by using the timber springs. The greatest improvement occurs at 250 Hz, with a reduction of 10 dB. The level difference is not of the same magnitude as in the measurements conducted in the COMET (Figure 12). It has to be noted that the installation in the large-scale test setup required adjustments. For example, the timber springs are screwed to the two plates, which can affect the vibration behaviour.

Figure 13.

Velocity level difference $Δ L_{V}$ between the floor system with timber springs and the reference system.

Figure 14.

Measured normalized impact sound pressure level $L_{n}$ in the slab testing rig.⁸

Figure 15.

Measured sound reduction index $R$ in the slab testing rig.

The curves in Figure 14 also clearly illustrate the improvement in the normalized impact sound pressure level $L_{n}$ due to the geometric modification of the web elements. This aligns with the velocity difference results, as the impact sound pressure level is proportional to the velocity level but is also affected by the radiation efficiency. At 100 Hz, the impact sound level curve of the normalized impact sound pressure level for the slab system with timber springs shows a dip. Additionally, the broad effective range of the timber springs between 80 and 500 Hz is remarkable. However, it does not exhibit typical resonance phenomena.

The analysis of the sound reduction index $R$ shows a notable improvement due to the use of timber springs compared to the reference system. Figure 15 depicts an increase in airborne sound insulation around 80 Hz, with the two curves converging around 1000 Hz. This indicates an enhancement in airborne sound insulation within a frequency range similar to that observed for impact sound (Figure 14).

However, the typical resonance behaviour expected in dual-shell systems cannot be observed. For instance, the curve of the sound reduction index $R$ does not display a clear dip that indicates the resonance frequency or the coincidence frequencies of the plates. According to Santoni et al.,¹² CLT plates may exhibit a coincidence region between the frequencies of the stiffer and weaker directions due to the differing bending stiffness along the two principal axes. The lowest coincidence frequency is reported to correspond to the stiffest principal direction. Using the material parameters of strength class C24²⁶ for timber components ( $E_{0} = 11.000 M P a$ , $E_{90} = 370 M P a$ and $ρ = 420 k g / m^{3}$ ), the calculated coincidence frequency $f_{g}$ equals 148 Hz in the principal direction of the top plate and 810 Hz perpendicular to the core fibre orientation, while for the bottom plate, it is 59 Hz in principal direction and 323 Hz perpendicular to it. In Figure 15, this is observed as a flattening of the curve in the frequency range of 125 to approximately 800 Hz.

The air in the cavity also impacts the sound reduction index. As the use of the timber springs suggests that they also influence the vibration behaviour of the plates, further research or more detailed measurement data is required.

For the rating of the sound insulation of the investigated slab structure, the sound reduction index $R$ and the normalized standard impact sound pressure level $L_{n}$ are weighted and assessed according to DIN EN ISO 717-1³¹ and 717-2,³² within the frequency range of 100–3150 Hz. As the low-frequency range significantly influences subjective perception, especially in wood-based structures,^6,7 and reflects the practical noise conditions caused by walking,²¹ the corresponding spectrum adaptation terms ( $C_{I, 50 - 2500}$ , $C_{I, 100 - 2500}$ ) are also evaluated, as specified by DIN EN ISO 717-2.³² Additionally, the adaptation terms for the extended frequency spectrum $C_{50 - 3150}$ and $C_{t r, 50 - 3150}$ ³¹ related to the weighted sound reduction index $R_{w}$ are also considered.

The results are presented in Table 1. The reduction in the weighted normalized impact sound pressure level $L_{n, w}$ highlights the effectiveness of the decoupling, not only in the low-frequency range. Specifically, the 4 dB reduction achieved by replacing the web elements with the timber springs, without additional improvement measures, is noteworthy. This effect is further supported by the spectrum adaptation term $C_{I, 50 - 2500}$ as it is reduced by 1 dB. Considering the spectrum adaptation term $C_{I, 100 - 2500}$ , which evaluates from 100 Hz, both constructions have a value of −3 dB. The improvement of the airborne sound insulation is also reflected in the weighted value $R_{w}$ , where the use of timber springs increases $R_{w}$ by 5 dB. With the inclusion of the spectrum adaptation terms, the airborne sound insulation performance decreases for both structures. For the reference system, it is reduced by 1 dB for $C_{I, 50 - 2500}$ and by 5 dB for $C_{t r, 50 - 3150}$ , while for the system with timber springs these values are 1–2 dB lower.

Table 1.

Weighted normalized impact sound pressure level $L_{n, w}$ and weighted sound reduction index $R_{w}$ of the investigated timber slab structures including spectrum adaptation terms ( $C_{I, 100 - 2500}$ , $C_{I, 50 - 2500}$ , $C_{50 - 3150}$ , $C_{t r, 50 - 3150}$ ), as measured in the slab testing rig.

Slab structure	$L_{n, w} (C_{I, 100 - 2500}; C_{I, 50 - 2500})$ [dB], Müller and Leistner³³	$R_{w} (C_{50 - 3150}; C_{t r, 50 - 3150})$ [dB]
System with beam elements	79 (−3; −2)	41 (−1; −5)
System with timber springs	75 (−3; −3)	46 (−2; −7)

Discussion

The conducted measurements demonstrate the functionality of acoustic decoupling by means of minimal geometric modification of linear wooden building system components. In the small-scale tests, the difference in structural velocity between the upper and lower plates suggests a significant alteration in the vibration behaviour of the overall system. When this method is applied to a timber building system in a slab testing rig, the improvement in acoustic performance is evident through the evaluation and characterization of the velocity level difference $Δ L_{ν}$ , the normalized impact sound pressure level $L_{n}$ and the sound reduction index $R$ .

However, despite the improvements, the current weighted normalized impact sound pressure level $L_{n, w}$ of 75 dB does not meet the required standards. In Germany, floor structures are required to achieve a maximum $L_{n, w}$ of 53 dB.³⁴ Nevertheless, the $L_{n, w}$ of a 140 mm thick reinforced concrete slab provides a value of 76 dB.³⁵ Notably, the $L_{n, w}$ of a solid CLT slab with a thickness of 140 mm is 87 dB.³⁶ Therefore, bare structural slabs require acoustic improvement measures and providing floor build-ups will also be advantageous in this regard. This also applies for the sound reduction index $R$ . The normative requirements of at least 54 dB³⁴ for separating floors in residential or office buildings is not met. Nevertheless, filling the cavity with insulation materials can further enhance sound insulation.

Floor build-ups and cavity damping were not included in this investigation, as the primary focus was on improving the bearing slab by integrating acoustic decoupling functions. As a result, fewer additional measures are required, such as surface structures or suspended ceilings. This approach not only offers the potential of reducing the need for heavy materials, such as gravel, thereby reducing the overall weight, but also allows for savings in the overall cross-section.

Additionally, the system with timber springs does not exhibit the typical resonance behaviour that is usually observed in dual-shell structures. For instance, the sound reduction index $R$ does not show the expected dips in sound insulation that would indicate resonance or coincidence frequencies. However, a coincidence region can be identified, potentially resulting from the differing bending stiffness of the CLT. Similarly, the improvement in impact sound insulation due to use of the timber springs cannot be explained by the conventional mass-spring-mass law.

This anomaly can be attributed to the distribution of the surface load caused by the movements of the plates. The timber springs are loaded differently over the entire surface. Further investigation using finite element analysis is required to better understand this behaviour.

Conducting a comprehensive parameter analysis can further improve the performance and effectiveness of the developed timber springs. This analysis should include key factors such as material selection, fabrication techniques and the placement and alignment of multiple timber springs and their connections to the plates. In addition, optimizing the design and arrangement of timber springs tuned to different frequencies could significantly improve the overall performance, especially in the very low frequency range.

Studying these parameters will allow for adapted modifications to improve the performance of the timber springs. Further measurements are also necessary to assess the shear characteristics. Moreover, extending the impact sound transmission studies to include spectrum adaptation terms ranging down to 25 Hz provides deeper insights into the acoustic behaviour in the very low frequency range (with appropriate excitation), as proposed by Ljunggren and Simmons.⁷ Besides that, the investigation of the transmission function $D_{T F}$ enables more accurate predictions of sound insulation for various sources of structure-borne noise, such as building services installations.³⁷

To facilitate the practical implementation of the developed timber spring, further adjustments regarding production and installation need to be made in consultation with industry professionals or design and construction firms. The cost-effectiveness of system component configurations and materials should be evaluated to ensure they are both affordable and beneficial.

The geometry of the timber spring is parametric and can be adapted to different configurations. It is necessary to explore optimization through discussions with industry to ensure that the integration of the timber spring does not compromise other essential functions of a slab system, such as load-bearing capacity and fire safety.

From an ecological perspective, it is important to assess the environmental impact of the materials used, including their manufacture and disposal. Additionally, evaluating the potential savings offered by the proposed acoustic-functional element will provide a clearer understanding of its overall benefits.

Conclusion

This paper presents a promising concept for the integration of acoustic decoupling in timber building systems with cavities, introduced as timber spring. The findings from the investigations provide valuable measurement data that enhance the understanding of this innovative approach.

The concept demonstrates the potential to reduce sound transmission from impact sound within a dual-shell. Measurement data revealed significant differences in structural velocity compared to the reference system with unmodified timber web elements, indicating improved acoustic performance. Specifically, the weighted normalized impact sound pressure level $L_{n, w}$ showed an improvement of 4 dB, achieved solely through geometric modifications without additional mass.

The timber spring was tuned to create a resonance at 100 Hz in the slab system, as shown by the normalized impact sound pressure level results. Achieving this outcome requires a fundamental understanding of structural behaviour and force distribution within the slab structure, which underscores the importance of an interdisciplinary approach during the planning and design phase.

This research lays the foundation for future investigations by establishing a fundamental understanding of this concept. It offers novel potential for significant advances in related fields and redefines current practices in timber construction and acoustics.

Footnotes

Acknowledgements

The authors would like to thank the project team of the University of Stuttgart for the interdisciplinary exchange on this topic: Cristóbal Tapia Camú and Simon Aicher of the Materials Testing Institute (MPA), Luis Orozco of the Institute for Computational Design and Construction (ICD) and Anna Krtschil and Jan Knippers of the Institute of Building Structures and Structural Design (ITKE). Acknowledgements are also extended to those involved in the realization of the measurements: Anna Krtschil for structural verification and Jan Mösche and Bernd Kaltbeitzel from the Fraunhofer Institute for Building Physics (IBP) for the assembly of the measurement setups and the subsequent investigation of the measurements conducted. The authors express their gratitude to Bernd Kaltbeitzel, Mark Koehler and Moritz Späh, also from the Fraunhofer Institute for Building Physics (IBP), for the valuable discussion on this topic.

Data availability

The research data is available at . [Müller T. Replication Data for: Concept and development of a novel timber spring for impact sound reduction in timber floor slabs, 2025. DOI:10.18419/DARUS-5110. URL https://doi.org/10.18419/DARUS-5110.]

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The presented work is supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC 2120/1 – 390831618.

ORCID iD

Theresa Müller

References

Kaufmann

Krötsch

Winter

Manual of multi-storey timber construction. 1st ed. Edition Detail. Munich: Detail Business Information GmbH, 2018.

Tupenaite

Kanapeckiene

Naimaviciene

, et al. Timber construction as a solution to climate change: a systematic literature review. Buildings 2023; 13(4): 976.

Müller

Borschewski

Albrecht

, et al. The dilemma of balancing design for impact sound with environmental performance in wood ceiling systems—a building physics perspective. Sustainability 2021; 13(16): 8715.

Svatoš-Ražnjević

Orozco

Menges

Advanced timber construction industry: a review of 350 multi-storey timber projects from 2000–2021. Buildings 2022; 12(4): 404.

Homb

Conta

Timber hollow-box floors: sound insulation measurement results and analysis. Build Acoust 2022; 29(3): 367–385.

Liebl

Späh

Leistner

Acoustics in wooden buildings: evaluation of acoustic quality in wooden buildings: listening tests and questionnaire field study. Borås: SP Technical Research Institute of Sweden, 2014.

Ljunggren

Simmons

Sound insulation in multi-family houses: proposal of single number limits for acoustical protection and comfort. Build Acoust 2023; 30(4): 387–407.

Müller

Leistner

Ecologically motivated approaches for improving low-frequency sound and vibration performance in multistory timber buildings. In: Proceedings of the 10th convention of the European acoustics association forum acusticum 2023, 2023, pp.1501–1506. Turin: European Acoustics Association. DOI: 10.61782fa.2023.0522/.

Holzforschung Austria (ed.) - Austrian Forest Products Research Society Catalogue of wood and wood-based materials, building materials, components and component connections for timber construction covering thermal, acoustic, fire and ecological performance levels. https://www.dataholz.eu/en/index.htm (2024, accessed 28 August 2024)

10.

Caniato

Bettarello

Fausti

, et al. Impact sound of timber floors in sustainable buildings. Build Environ 2017; 120: 110–122.

11.

Müller

Di Bari

Akustisches Verhalten von Holzgeschossdecken ökologisch neu gestalten. In: Fortschritte der Akustik - DAGA2022(ed Deutsche Gesellschaft für Akustik eV), 2022.

12.

Santoni

Schoenwald

Fausti

, et al. Modelling the radiation efficiency of orthotropic cross-laminated timber plates with simply-supported boundaries. Appl Acoust 2019; 143: 112–124.

13.

van Damme

Schoenwald

Zemp

. Modeling the bending vibration of cross-laminated timber beams. Eur J Wood Wood Prod 2017; 75(6): 985–994.

14.

Kohrmann

Völtl

Müller

, et al. Abschlussbericht zum AiF Forschungsvorhaben VibWood: Planungshilfen zur schall- und schwingungstechnischen Beschreibung von Holzdecken und zur Bewertung und Dimensionierung von angepassten Schwingungsschutzsystemen, https://opus4.kobv.de/opus4-rosenheim/frontdoor/index/index/searchtype/authorsearch/author/M.+Kohrmann/docId/523/start/4/rows/10 (2014)

15.

Schoenwald

Vallely

Cross laminated timber elements with functional grading and localised ballast to improve airborne and impact sound insulation. In: Proceedings of the 10th convention of the European acoustics association forum acusticum 2023, 2023, pp.1529–1534. Turin: European Acoustics Association. DOI: 10.61782/fa.2023.0827.

16.

Krtschil

Orozco

Bechert

, et al. Structural development of a novel punctually supported timber building system for multi-storey construction. J Build Eng 2022; 58: 104972.

17.

Málaga-Chuquitaype

Ilkanaev

Novel digitally-manufactured wooden beams for vibration reduction. Structures 2018; 16: 1–9.

18.

Müller

Wagner

Menges

, et al. Component for producing building parts such as walls and ceilings, 2023.

19.

Orozco

Krtschil

Wagner

, et al. Co-design methods for non-standard multi-storey timber buildings. Sustainability 2023; 15(23): 16178.

20.

Neubrand

Ersoy

Eulenbach

. Komponenten der Federung. In Ersoy

Gies

(eds.) Fahrwerkhandbuch: Grundlagen – Fahrdynamik – Fahrverhalten– Komponenten – Elektronische Systeme – Fahrerassistenz – Autonomes Fahren– Perspektiven. Wiesbaden: Springer Fachmedien Wiesbaden, 2017, pp. 439–485.

21.

Hessinger

Rabold

Saß

et al. Schallschutz im Holzbau. In Fouad

(ed.) Bauphysik Kalender 2022. Berlin, Germany: Ernst & Sohn GmbH, 2022, pp. 275–356.

22.

EN ISO 16251-1:2014-08. Acoustics - laboratory measurement of the reduction of transmitted impact noise by floor coverings on a small floor mock-up - part 1: heavyweight compact floor (ISO 16251-1:2014), 2014.

23.

Schmidt

Wittstock

Messung der Trittschallminderung bei Bauprodukten aus Holz. In: Fortschritte der Akustik - DAGA 2012 (ed Deutsche Gesellschaft für Akustik eV), 2012, pp.81–82.

24.

Haakh

JR.

Integrative measures for increasing the sound insulation of floor slabs in timber beam construction. Master Thesis, University of Stuttgart, Stuttgart, Germany, 2024.

25.

ETA-06/0009: Binderholz Brettsperrholz BBS - Solid wood slab element to be used as a structural element in buildings, 2022

26.

EN 338:2016-04. Structural timber - Strength classes, 2016.

27.

Müller

Wagner

Menges

Leistner

. Replication Data for: Concept and development of a novel timber spring for impact sound reduction in timber floor slabs, https://doi.org/10.18419/DARUS-5110 (2025, accessed 12 June 2025)

28.

EN ISO 10140-3:2021-05. Acoustics - laboratory measurement of sound insulation of building elements - part 3: measurement of impact sound insulation (ISO 10140-3:2021), 2021.

29.

Rabold

. Trittschallschutz von Wohnungstrenndecken - Konstruktive Lösungen für den erhöhten Schallschutz nach DIN 4109-5. In Deutsche Gesellschaft für Akustik eV (ed.) Fortschritte der Akustik - DAGA 2023. pp. 282–285.

30.

EN ISO 10140-2:2021-05. Acoustics - laboratory measurement of sound insulation of building elements - part 2: measurement of airborne sound insulation (ISO 10140-2:2021), 2021.

31.

DIN Deutsches Institut für Normung e V. EN ISO 717-1:2020-12. Acoustics - rating of sound insulation in buildings and of building elements - part 1: airborne sound insulation (ISO 717-1:2020), 2020.

32.

EN ISO 717-2:2020-12. Acoustics - rating of sound insulation in buildings and of building elements - part 2: impact sound insulation (ISO 717-2:2020), 2020.

33.

Müller

Leistner

Integrative Ansätze zur Körperschallentkopplung im mehrgeschossigen Holzbau. In: Fortschritte der Akustik - DAGA 2023 (ed Deutsche Gesellschaft für Akustik eV), 2023, pp.316–319.

34.

DIN 4109-1:2018-01. Sound insulation in buildings - part 1: minimum requirements, 2018.

35.

Leistner

Schröder

Richter

Gehgeräusche bei Massiv– und holzbalkendecken. Bauphysik 2003; 25(4): 187–196.

36.

Di Bella

Granzotto

Barbaresi

. Analysis of acoustic behavior of bare clt floors for the evaluation of impact sound insulation improvement. Proc Meet Acoust 2016; 28(1): 015016.

37.

EN ISO 10848-1:2017-10. Acoustics - laboratory and field measurement of flanking transmission for airborne, impact and building service equipment sound between adjoining rooms - part 1: frame document (ISO 10848-1:2017), 2017.