Effects of Coating Ageing on the Acoustic Properties of Norway Spruce (Picea abies)

Abstract

Norway spruce (Picea abies L. Karst) wood is a preferred resonance material for musical instruments, but the surface coatings used to protect it also alter its acoustic behaviour. In this study, the effects of nitrocellulose and polyurethane coatings on spruce lamellas during an ageing period of 300 days were investigated. Gloss, hardness, impact resistance, resonance frequencies, vibration damping (tan δ) and acoustic conversion efficiency (ACE) were measured. Both coatings initially reduced the resonance frequencies and moduli of elasticity (E), while increasing the tan δ and reducing the ACE, with the nitrocellulose having a greater effect. Ageing led to greater hardness, lower tan δ and improved ACE, which can be attributed to the progressive curing of the coatings. The strong correlation between hardness and acoustic parameters suggests that mechanical surface properties may serve as predictors of acoustic effectiveness. Polyurethane maintained acoustic performance better than nitrocellulose, although impact resistance decreased with ageing. These results emphasize the importance of choosing coating systems that balance durability and long-term acoustic requirements in instrument making.

1. Introduction

Spruce wood (Picea abies L. Karst) is one of the most valued resonance materials for the construction of musical instruments, such as violins, guitars and pianos. Its favourable stiffness-to-density ratio, low internal friction and ability to transmit vibrations efficiently make it particularly suitable for soundboards and resonant plates [1,2,3,4,5]. However, spruce wood is also hygroscopic and sensitive to changes in climate, which can affect both dimensional stability and acoustic performance. To protect, increase durability and improve aesthetics, wooden instruments components are traditionally coated with varnishes or polymer-based finishing systems [6].

However, the application of coating does not only serve protective and decorative purposes. They also change the vibroacoustic properties of the underlying wood. Numerous studies have shown that coatings generally lower resonance frequencies due to increased mass, reduce the effective modulus of elasticity and increase damping, resulting in less sharp resonances [2,7,8,9]. The extent of these effects depends on the type of coating, thickness, penetration depth and curing behaviour. For example, alcohol- and oil-based coatings can significantly change the stiffness and internal friction in different fibre directions of spruce plates [1], while polyurethane coatings used on piano soundboards show thickness-dependent influences on sound absorption and radiation properties [10]. Recent findings emphasize that the acoustic properties can be tuned by the finishing system: The number of layers, the type of resin and the curing state can either enhance tonal qualities or dampen sound radiation [8,11].

Despite extensive research into historical and contemporary coatings, there are still considerable gaps. Studies on old instruments are often unable to separate the effects of varnish composition from those of natural ageing, while experimental studies mostly focus on short-term responses immediately after coating [1,12]. Less attention has been paid to the long-term development of acoustic properties during the maturation of the coating. In particular, nitrocellulose and polyurethane coatings, which are widely used due to their mechanical durability and aesthetics, have not been systematically compared in terms of their ageing-related effects on resonance wood acoustics.

In this study, we investigate the influence of nitrocellulose and polyurethane coatings on the acoustic properties of spruce lamellas, both immediately (14 days) after application and during a 300-day ageing period. By analyzing changes in vibration frequencies, damping, elastic moduli and acoustic conversion efficiency (ACE) and their relationship to surface hardness, the study aims to clarify how different coating systems change the vibroacoustic performance of resonance wood over time. Although modern water-based and acrylic coatings are more environmentally friendly, they are rarely used for high-quality resonance instruments due to their higher damping and limited gloss depth. Therefore, solvent-based nitrocellulose and polyurethane coatings were chosen to ensure comparability with traditional finishing systems used in lutherie.

2. Materials and Methods

Norway spruce (Picea abies L. Karst) wood was sourced from a naturally grown forest stand in the alpine region of Gorenjska, Slovenia (46.3059° N 13.9832° E, altitude 1200 m). The average width of the annual rings was (1.6 ± 0.3) mm, with a regular distribution of earlywood and latewood bands. The wood was air-dried and then conditioned in the laboratory (20 °C, 50% RH; RH–relative humidity) for one month before the test specimens were produced.

Coarse lamellae of 5 mm thickness (n = 20) were sawn from the conditioned, radially oriented test specimens on a circular sawing machine. The lamellae were planed on both sides to a thickness of 3 mm and cut to the final dimensions (length 400 mm, width 80 mm and thickness 3 mm) in the longitudinal (L) and transverse (R–radial, T, h–tangential) directions (Figure 1a). The average moisture content of the lamellas before coating was 9.1 ± 0.4%, and it remained within 8.5%–9.5% throughout the entire ageing period under constant climatic conditions.

For each coating system (nitrocellulose and polyurethane), ten (n = 10) spruce lamellas were prepared. Seven were used for acoustic, gloss, and surface hardness measurements. Due to the destructive nature of the method, the remaining three (n = 3) samples were used for surface impact resistance testing.

2.1. Wood Surface Treatment

The test samples were sanded with P180 grit sandpaper before surface coating. The surfaces were then dedusted with compressed air to ensure the best possible penetration and adhesion of the applied coating. Seven of the samples were then coated with a nitrocellulose coating (N, group 1) and a polyurethane coating (P, group 2) using an air spraying technique. Both coating systems were commercial products. For reasons of trademark protection, the trade names of the systems are not mentioned. The nitrocellulose (N) system was based on cellulose nitrate resin in a butyl acetate–ethanol solvent system. The polyurethane (P) system was two-component coating based on polyether-polyol resin and aliphatic isocyanate hardener.

For the N coating, a transparent one-component coating was used. For the first layer, 30% of the weight of thinner was added to the coating; for the second and third layers, only the nitrocellulose coating was used.

The P coating was a high-gloss coating thoroughly mixed with the hardener in a weight ratio of 1:1 according to the instructions for use. To improve penetration and adhesion, 20 percent by weight of a thinner was added to the first coat.

The target application rate for each layer in both groups was between 100 g/m² and 150 g/m², which was achieved with a total of 3 layers (

Table 1. Average application of nitrocellulose (N) and polyurethane (P) coating on 10 spruce lamellas, by layers (1st, 2nd, 3rd) and in total, and the average density (±Standard Error SE = St.dev./(n_s)^−1/2).

Sample	1st Layer (g/m2)	2nd Layer (g/m2)	3rd Layer (g/m2)	Total Coating (g/m2)	Lamella Density (kg/m3)	Dry-Film Thickness (μm)
Control	-	-	-	-	520 (±2.9)	-
Nitrocellulose	14.8	70.0	56.0	146.7 (±1.9)	567 (±3.9)	200.0 (±10.2)
Polyurethane	24.8	38.0	38.0	105.0 (±3.8)	556 (±3.6)	180.0 (±4.8)

). The application was carried out by air spraying with a nozzle diameter of 1.8 mm and a pressure of 3 bar (Figure 1b). Between the individual coating applications, the samples were conditioned for one day at 20 °C and 50% RH. After the first and second coating, the samples were sanded with P320 grit sandpaper. The average dry-film thickness of the coatings after three layers measured with ultrasonic method (PosiTector 200, DeFelsko Corporation, Ogdensburg, NY, USA; EN ISO 2808 (2019)) [13] was (200 ± 10) µm for nitrocellulose and (180 ± 5) µm for polyurethane (Table 1). To achieve the most uniform coating thickness possible, the samples were rotated 90 degrees between each spraying pass. The coating systems were formed according to the instructions of the coating manufacturers.

2.2. Characterisation of the Coated Surfaces

2.2.1. Gloss Measurements

The specular gloss of the coated surfaces was determined 14 days after completion using a gloss meter (AcuGloss TRI, X-Rite Pantone, Grand Rapids, MI, USA) at a measuring angle of 60°, according to the method described in EN ISO 2813 (2015) [14]. Six measurements were taken from each sample.

2.2.2. Determination of Surface Hardness

The pendulum hardness of the coated samples was determined following the EN ISO 1522 (2022) [15] and using the König measuring method (pendulum hardness tester, model 299/300, Erichsen GmbH, Hemer, Germany). The surface hardness was measured by determining the damping time of the pendulum swinging on the coated surface from 6° to 3°, relative to the normal axis, using a digital timer. The pendulum swung with two balls with a hardness of 63 ± 3.3 HRC and a diameter of 5 ± 0.0005 mm (Figure 1c). Surfaces with longer pendulum damping time signalize the harder surface system [16].

The initial coating hardness was measured one day after the last coating application on coated glass and wood samples. Subsequent time-dependent measurements were then only carried out on coated wood samples, as other time-dependent properties were also measured on these samples.

2.2.3. Assessment of Surface Resistance to Impact

The modified approach was based on ISO 4211-4 (1995) [17], which is still the latest active version of this standard for testing the impact resistance of furniture coatings. For this semi-destructive test, the remaining 6 parallel samples, 3 for each coating group, which were not used for testing the acoustic properties, were used. A cylindrical steel weight with a mass of 500 g was dropped from a certain height through a vertically mounted guide onto a steel ball with a diameter of 14 mm, which was positioned on the test surface. Three impacts were performed on the test sample from the following drop heights: 10 mm, 25 mm, 50 mm, 100 mm, 200 mm and 400 mm. After the impacts, the surfaces were carefully examined with a magnifying glass (10× magnification) under direct light. The minimum heights at which the coating film cracked were then determined.

2.3. Determination of the Acoustic Properties

The lamellae were placed on two elastic supports positioned at the first nodal points of the bending vibration (L = 22.4% of the lamellae length or 89.6 mm from the end of the specimen).

2.3.1. Determination of the Fundamental Resonance Frequency

Pulsed mechanical excitation was performed with a rigid steel ball in the geometric axis at the open end of the lamellae to determine the fundamental vibration frequency in bending of the lamellae. A PCB-130D20 condenser microphone (PCB Piezotronics Ltd., Depew, NY, USA) was attached to the rear end of the lamellae. The sound signal was recorded using the NI-9234 DAQ module (National Instruments Ltd., Austin, TX, USA.) with a resolution of 24 bits and a sampling rate of 51 kHz. The measurements were carried out in a semi-anechoic room separated from the main laboratory hall by a sound-absorbing wall (average ambient noise level of 11.5 dB). The signals were processed with LabVIEW 2025 software (National Instruments, Austin, TX, USA) using FFT to determine the fundamental resonance frequency (Figure 2).

2.3.2. Analysis of the Free-Free Vibration Resonance

To investigate the acoustic response, we excited the lamellae without contact using an Anker loudspeaker A3107 (Anker Ltd., Changsha, China), which was placed 1 cm below the lamellae in the center of their length. The excitation frequency of the sound signal corresponded to the fundamental frequency of the lamella and lasted 6 s. At the end of the excitation, the 1st (f₁), 2nd (f₂) and 3rd (f₃) bending vibration frequencies in the triggered vibration mode were determined using a broadband FFT analysis in LabVIEW. The dynamic moduli of elasticity were determined using the Bernoulli solution, which assumes a very high length-to-depth ratio (L/h >> 1) and disregards the shear and elastic support effect (Equation (1)) [18].

$$E_n={\displaystyle \frac{48{\pi }^2{L}^4\rho }{m_n^4{h}^2}}{f}_n^2$$

(1)

Notation:

E_n–dynamic modulus of elasticity in n-th vibration mode (GPa = 1 × 10⁹ Pa);

L–length of the lamella (m);

ρ–density of the lamella (kg/m³);

f_n–bending frequency of the lamella in the n-th vibration mode (n = 1, 2, 3);

h–thickness of the lamella [m];

m_n–parameter to solve the Bernoulli constants, depends on the vibrational mode (Equation (2)),

$$m_n={\displaystyle \frac{2n+1}{2}}\pi$$

(2)

The vibration damping (tan δ) of the lamellae was determined by measuring the logarithmic decrement of the vibration signal at each resonance frequency (f₁–tan δ₁, f₂–tan δ₂, f₃–tan δ₃). The influence of wood structure and density (ρ) on the mechanical stiffness was evaluated by specifying a specific modulus of elasticity (E/ρ). In addition, the acoustic conversion efficiency (ACE_n) was determined for every vibration mode (n = 1, 2, 3) which represents the pure sound radiation and reflects the influence of the microstructure of the material on the sound radiation [19,20].

$${ACE}_n={\displaystyle \frac{\sqrt{\frac{E_n}{\rho }}}{tan{\delta }_n·\rho }}$$

(3)

2.4. Observation of the Surface and Acoustic Properties of the Coated Lamellae over Time

The coated lamellas were stored under controlled laboratory conditions (20 °C, 50% RH) for 300 days to simulate natural ageing without exposure to light or outdoor weathering. The measurements of the surface and acoustic properties of lamellas began 1 day after the application of the 3rd layer. This was the start of the measurement process, and we then measured the surface and acoustic properties of the lamellas after 1, 2, 3, 7, 11, 15 and 43 weeks of ageing.

2.5. Data Processing and Statistical Analysis

We collected data on sample coating and acoustic measurements in MS Excel (Office 365) and NI LabVIEW2025. We performed simple regression models and statistical analysis in IBM SPSS Statistics v29. Statistical analysis was conducted using one-way ANOVA with Tukey’s HSD post hoc test at a 95% confidence level (p < 0.05).

3. Results

3.1. Properties of Coatings Applied to Spruce Lamellas

The three-layer application of nitrocellulose and polyurethane coating provides the necessary surface protection that is required, for example, when using wood for musical instruments [21] (Figure 3). The application rate of the first layer compared to the 2nd and 3rd layers was lower due to the addition of thinner in both coatings used, which reduces the solids content (Table 1). The final application rate was 146.7 g/m² for the nitrocellulose coating, which increased the average lamella density by 47 kg/m³ (+9.0%). The total application rate was slightly lower for the polyurethane coating (105 g/m²) and increased the average lamella density by 36 kg/m³ (+6.9%). An increase in the average lamella density at the time of coating application is to be expected. Studies on violin varnishes show that varnishes generally have a higher density than wood, with values between 1100 kg/m³ and 1400 kg/m³ [21]. The density of polyurethane coatings is usually given as 1200 kg/m³ [7], while nitrocellulose coatings are slightly denser at around 1350 kg/m³ according to earlier studies [2].

3.1.1. Gloss of Finished Surfaces

The highest gloss, corresponding to high gloss appearance, was achieved with a P coating, with values between 94.5 and 99.8 and an average of 97.4. The gloss values for the N coating were significantly lower and ranged between 83.9 and 94.2 with an average value of 89.9, and exhibited a moderately high gloss (Figure 4). Although gloss is an important optical property for aesthetic evaluation, it does not significantly influence the vibrational or acoustic behaviour of the wood surface. However, the gloss level of the surface to be selected depends on the individual’s wishes.

3.1.2. Surface Hardness

The initial hardness of the N coating was low (t = 18.3 ± 0.3 s; ±SE–standard error), while the hardness of the P coating was significantly higher (t = 50.1 ± 1.3 s), similar to previous studies confirming the superior hardness of polyurethane coatings [22,23,24]. The initial hardness of the N coating on the reference glass surface was comparable to that at the wood surface (t = 20.7 + 0.8 s) (Figure 5). However, the hardness of the P coating on the reference glass surface was significantly lower than that of the wood surface (t = 28.4 + 0.5 s). Previous studies have also confirmed that the pendulum hardness of coatings on glass is lower than that of coatings on wood. The results show that the flow of the coating on both substrates is not the same and the impregnation of the substrate is very different [24]. Research has confirmed that film formation also depends on the degree of cross-linking of the coating and the presence of certain wood components [25]. This indicates that the P coating has penetrated well into the wood structure and has built up good adhesion, resulting in a hard and firm surface layer with lower coating thickness.

Over the period studied, there have been characteristic changes in the hardness of the applied coatings. The pendulum damping time and thus the hardness of the N coating increased by 20 s in the first three weeks of ageing. The curing of the N coating slowed down further, whereby we measured 57 s (SE = ±0.8 s) after 300 days. A similar, clearly increasing trend of the increased pendulum damping time was also observed for the P coating by approx. 30 s in the first three weeks and reached 108.4 s (SE = ±1.1 s) after 300 days of curing (Figure 5).

It is important to note that the value measured by the pendulum method is not a general measure of surface hardness, but essentially a value for the reciprocal of the damping capacity or mechanical loss [26]. Therefore, it is meaningless to compare the rigidities of coating films consisting of materials with different viscoelastic properties using this method, as its validity is limited to homogeneous materials. In this context, it can be said that using the König pendulum, we only confirmed the higher hardness of the P coating compared to the N coating in relative terms over the entire test period. Previous studies show that the hardness of coatings can reflect the development of the aggregation structure of the coating during curing, which is related to nuclear-growth (polymerization) and cross-linking of the resins [24,25,26,27,28].

3.1.3. Surface Resistance to Impact

As a result of the impact resistance of the coated surfaces, the minimum drop height at which the coating cracked was significantly different between the two coating systems tested. On surfaces coated with N coating, the film already cracked at a drop height of 10 mm over the entire time range. On surfaces coated with P coating, the first cracks only appeared from a drop height of 400 mm. After one week, the minimum drop height fell to 200 mm and in a further week to 100 mm, which was the case until the end of the test period of 43 weeks (300 days) (Figure 6).

The P coating, which was also found in some earlier studies, is still much more impact resistant than N, and confirms its higher toughness [10]. However, the decreasing performance of the P coating over time could be due to post-cure embrittlement and microstructural changes such as increased cross-linking density and reduced molecular mobility. The ongoing curing of P in this study is consistent with its thermosetting nature and its ability to age physically [28,29]. As the lamellas were kept in a dark room throughout the test period, UV ageing and photodegradation, which have been reported to occur when coatings are exposed to outdoor climatic conditions [24], were not a possible ageing mechanism in this case. It can be clearly concluded that as the hardness of the P coating increases, the impact resistance of the coated surface decreases over time. A similar conclusion could probably be drawn for N coating as well, but we were unable to prove this with the used method impact resistance, as the N coating cracked from the start at the lowest impact height of 10 mm. As the nitrocellulose coating cracked even at the lowest impact height of 10 mm, the adopted ISO 4211-4 [17] method could not detect subtle differences in its performance over time, indicating limited sensitivity for weaker coating systems.

3.2. Impact of Coating on the Initial Acoustic Properties of Lamellas

Since the acoustic vibration characteristics of each lamella were first determined in the uncoated state, these served as baseline control values. The application of N and P coatings significantly altered the initial acoustic response of spruce lamellas (

Table 2. Mean values (±Standard error) of natural frequency (f (s⁻¹)), modulus of elasticity (E (GPa)), damping (tan δ∙10⁻² ( )) and acoustic conversion efficiency (ACE (km/s)) of spruce lamellas in the first three vibration bending modes (n = 1, 2, 3) before and after surface treatment with nitrocellulose and polyurethane coatings.

	f1	f2	f3	E1	E2	E3	tan δ1	tan δ2	tan δ3	ACE1	ACE2	ACE3
Control	117.4	234.1	352.8	18.7	9.6	5.7	1.13	0.95	0.97	1026	974	697
	(0.9)	(1.7)	(2.6)	(0.4)	(0.2)	(0.1)	(0.02)	(0.10)	(0.10)	(20.2)	(28.1)	(43.9)
N	112.9	226.2	332.5	17.2	8.9	5.0	2.08	1.78	1.37	472	417	597
	(2.0)	(3.8)	(10.0)	(0.6)	(0.3)	(0.3)	(0.13)	(0.22)	(0.22)	(26.7)	(50.9)	(35.9)
P	111.7	222.6	327.2	16.6	8.6	4.9	1.95	1.51	1.46	515	477	383
	(3.1)	(6.3)	(12.5)	(0.7)	(0.4)	(0.3)	(0.14)	(0.10)	(0.10)	(41.8)	(38.4)	(41.0)

). Both coatings caused a reduction in natural frequencies (f₁, f₂, f₃) and moduli of elasticity (E₁, E₂, E₃) compared to the control samples, with P showing a slightly stronger effect. These changes are mainly attributed to the additional mass of the coatings, which reduces the stiffness-to-mass ratio of the system and lowers the vibration frequencies, as also observed in previous studies on varnished soundboards of violins [2,25].

A clearer influence was observed in vibration damping (tan δ). Both coatings significantly increased damping compared to the control, with N showing the highest values in all modes. This reflects the higher viscoelastic losses of N, which can reduce the resonance sharpness and influence the sound properties of the wood. Increased damping due to surface treatments has also been reported for acetylated and lacquered woods, where viscoelastic effects of the coating dominate the vibration behaviour [7,20].

The acoustic conversion efficiency (ACE) also decreased significantly after coating, by more than 50% in the first vibration mode. N consistently resulted in the lowest ACE values, while P showed slightly higher efficiency, but still well below that of the uncoated control (Table 2). These results confirm that while surface coatings provide significant protection, they significantly degrade the acoustic performance of spruce wood. For applications such as musical instruments, the choice of coating is therefore crucial as it can influence resonance behaviour and sound radiation [19,21].

3.3. Time-Dependent Changes in the Acoustic Properties of Coated Spruce Lamellas

The ageing process over 300 days showed clear tendencies in the acoustic behaviour of the N- and P-coated spruce lamellae (Figure 7a,b). The dynamic moduli of elasticity in all three vibration modes remained more or less constant over the entire observation period (R² ≤ 0.14). This indicates that the coatings did not significantly change the intrinsic stiffness of the lamellas over time, which is consistent with previous studies reporting that ageing processes mainly affect the viscoelastic and not the elastic components of coated wood [2,21].

In contrast, the vibration damping (tan δ) decreased steadily during ageing for both coatings and for all types of vibrations (Figure 7c,d). This reduction can be attributed to the progressive curing of the coatings, especially in the case of thermoset P, where progressive cross-linking reduces molecular mobility and internal friction [18,20]. A similar, albeit less pronounced, effect was observed with N, suggesting that physical ageing and solvent evaporation also contribute to a stiffer and less dissipative coating structure [7].

As a direct consequence of the reduced damping, the acoustic conversion efficiency (ACE) of both coatings increased steadily over time in all three modes (Figure 7e,f). The improvement in ACE indicates that ageing improves the ability of the coated lamellas to convert vibration energy into acoustic radiation. This trend is particularly relevant for applications in musical instruments where sustained vibrations and efficient sound radiation are required. Although ageing improved the ACE values of coated lamellas, they remained lower than those of the uncoated control, indicating partial but not complete restoration of acoustic efficiency (see Table 2). This confirms that coatings inevitably reduce acoustic efficiency, although the magnitude of this effect decreases as the coating ages.

4. Discussion

The influence of coatings on the acoustic properties of resonance wood has been extensively documented. The observed reduction in resonance frequencies and increase in damping immediately after coating can be attributed to mass loading and viscoelastic losses in the polymer layer [1,2,7,8]. These effects can alter resonance frequencies, suppress vibrational energy and reduce acoustic conversion efficiency (ACE). However, the extent of these changes strongly depends on the coating type, composition, thickness and curing conditions [11,30]. Our results confirm these established trends while providing new insights into the long-term evolution of acoustic performance during coating ageing.

Both N and P coatings reduced the natural frequencies and dynamic moduli of elasticity of spruce lamellas immediately after application. This reduction is primarily due to the additional mass of the coatings, which reduces the stiffness-to-mass ratio of the system. Such effects have also been reported for alcohol- and oil-based varnishes on violin tops, where the resonance frequencies decreased after varnishing [1,8]. In parallel, damping increased significantly in coated samples, especially N coating, which is consistent with the results of Obataya et al. [7] who observed increased viscoelastic losses in coated wood. The combination of lower resonance frequencies and higher vibration damping resulted in significantly lower ACE values, which is consistent with previous reports that coatings tend to suppress resonance clarity and sound radiation [2,31].

Ageing over 300 days led to significant changes in the acoustic response. While the elastic moduli remained stable, the damping in all vibration modes steadily decreased and the ACE values increased. This indicates that the coatings continued to cure, reducing molecular mobility and internal friction, and improving vibration efficiency, as reported in previous studies [28]. These changes indicate a transition from a more dissipative to a more elastic coating structure. Similar time-dependent improvements were observed in lacquer- and resin-coated wood [7,32] and in varnished tonewood subjected to a controlled curing treatment [1]. The P coating system as a thermosetting polymer showed particularly clear signs of progressive crosslinking, which reduced viscoelastic losses, while N showed more moderate improvements, probably related to solvent evaporation and physical ageing. In addition to progressive polymer curing, the observed changes may also involve redistribution of residual stresses and gradual moisture equilibration at the coating–wood interface, both of which can affect energy dissipation and acoustic conversion efficiency.

A remarkable result of this study is the strong correlation between mechanical hardness and acoustic performance. As the coating aged and the surface hardness increased, the attenuation decreased, and the ACE improved (Figure 8). This relationship supports the idea that mechanical surface properties can serve as indirect indicators of acoustic effectiveness, a conclusion that is consistent with recent studies that link coating microstructure, penetration depth and curing conditions to the vibrational properties of spruce wood [1,11].

Finally, the comparison of the two coatings shows their different acoustic effects. N coating consistently resulted in higher attenuation and a lower ACE value, indicating greater suppression of sound radiation. P coating retained better acoustic effectiveness during ageing despite an initial decrease in impact strength due to embrittlement. This is consistent with the observations that P coatings on piano soundboards can strike a balance between mechanical durability and acceptable acoustic performance if the thickness is carefully controlled [30].

Overall, the results emphasize that coatings inevitably impair the original acoustic quality of spruce wood, but that ageing processes can partially restore the sound radiation. For instrument makers and restorers, this means that coated soundboards may initially sound suboptimal, but improve over time. The observed reduction in damping and improvement in ACE suggest that ageing may enhance tonal brightness and sustain in coated resonance wood. Although absolute timbre cannot be inferred directly from these parameters, the findings provide a physical basis for the commonly perceived improvement in the tonal quality of aged instruments. The choice of coating system must therefore harmonize the protective and aesthetic functions with the long-term acoustic requirements of the instrument [11,12]. Future investigations will combine acoustic testing with microscopic and spectroscopic analyses (e.g., SEM, FTIR) to provide deeper insight into coating curing mechanisms and polymer–wood interactions.

5. Conclusions

In this study, the influence of nitrocellulose (N) and polyurethane (P) coatings on the acoustic properties of spruce lamellas during a 300-day ageing period was investigated. The main conclusions are:

• Immediate effects of the coatings—Both coatings decreased resonance frequencies and dynamic moduli due to the added mass, while increasing vibration damping and decreasing acoustic conversion efficiency (ACE). These effects were more pronounced with nitrocellulose than with polyurethane.
• Ageing behaviour—While moduli of elasticity remained stable, vibration damping decreased and ACE increased over time, suggesting that curing of the coating improves acoustic performance with ageing.
• Relationship between hardness and acoustics—A strong correlation was found between surface hardness, vibration damping and ACE, suggesting that mechanical surface properties can serve as reliable indicators of acoustic effectiveness.
• Comparison of coatings—Nitrocellulose suppressed sound radiation more, while polyurethane maintained acoustic effectiveness better, despite some loss of impact resistance with ageing.

Overall, coatings inevitably change the acoustic behaviour of spruce wood, but ageing can partially restore the favourable vibration and radiation properties. For musical instruments, these results show how important it is to select coating systems not only for their protective and aesthetic properties, but also for their long-term acoustic performance. Future studies will focus on developing eco-friendly, waterborne or bio-based coatings that meet both environmental and acoustic performance requirements.