Document Type : Research Paper
Noise pollution in agricultural machinery, particularly grain harvest combines, not only causes quantitative adverse effects but also leads to significant psychoacoustic impacts on operators. Although most previous studies have primarily focused on evaluating sound pressure levels, limited attention has been paid to qualitative psychoacoustic parameters such as loudness, sharpness, and roughness in combine harvesters. Therefore, the present study aimed to perform a psychoacoustic analysis of acoustic signals generated by two grain combine harvester models, namely John Deere and Sampo, under different operating conditions, while simultaneously evaluating both quantitative and qualitative sound characteristics. In this study, acoustic signals were recorded under various operational conditions including idle speed, high engine speed, free movement, and harvesting operation. The recorded signals were then analyzed using signal processing techniques to extract quantitative sound indices, including sound pressure level, as well as psychoacoustic parameters such as loudness, sharpness, and roughness. The results demonstrated that operating conditions, engine speed, and operational mode significantly affected the acoustic characteristics of the combines. The average loudness values of the Sampo and John Deere combines at idle and high engine speeds were 7.1 and 12.95 sone, and 8.6 and 17.57 sone, respectively. Furthermore, the average sound pressure levels for the Sampo and John Deere combines were measured as 54.1 and 59.6 dB, and 54.7 and 65.0 dB, respectively. The highest roughness values for both combines were observed during harvesting operations, whereas the lowest values occurred during free movement conditions. Overall, the John Deere combine exhibited higher sharpness and roughness values compared with the Sampo combine, indicating a greater potential for inducing auditory annoyance and operator discomfort. The findings of this study can be applied to the acoustic design, cabin optimization, and enhancement of ergonomic conditions in locally manufactured combine harvester cabins in comparison with imported models.
Field measurements were conducted on a John Deere 1055 and a Sampo 3065 combine harvester during actual harvesting and road travel (free movement). Sound signals were recorded using calibrated Brüel & Kjær type-1 microphones positioned at multiple locations: inside the cabin, at the operator’s ear (with and without cabin), and at bystander distances of 7.5 m and 20 m. Engine speed (idle/low vs. full/high), gear position (parking, 1–3, and road gears), and operational mode (harvesting vs. free movement) were systematically varied.
Raw signals were pre-processed to remove environmental noise, A-weighted, and filtered. Quantitative SPL (dB(A)) and psychoacoustic parameters were then extracted in MATLAB using standardized Zwicker models (ISO 532-1). Loudness was calculated according to the time-varying loudness model, sharpness via the weighted spectral centroid method, and roughness through modulation analysis in critical bands. Statistical analysis was performed with ANOVA to assess main effects and interactions, followed by Tukey post-hoc tests at 95 % confidence level.
All four operational factors showed highly significant effects (p < 0.01) on every measured parameter. Engine speed, operational mode, and microphone position affected SPL, loudness, sharpness, and roughness; gear ratio significantly influenced SPL and sharpness. Mean SPL ranged from 54.1–59.7 dB(A) for the Sampo and 54.8–65.0 dB(A) for the John Deere. Loudness was substantially higher in the John Deere (8.6–17.6 sone) than in the Sampo (7.1–13.0 sone). Sharpness followed the same trend (John Deere 1.08–1.18 acum; Sampo 1.05–1.09 acum), as did roughness (John Deere 0.07–0.09 asper; Sampo 0.06–0.07 asper).
Harvesting produced the highest values across all metrics due to simultaneous engagement of cutter bar, threshing drum, straw walkers, and unloading systems. High engine speed increased mechanical energy and high-frequency content, elevating loudness, sharpness, and roughness. Cabin presence reduced SPL by 5–12 dB, loudness by 4–8 sone, sharpness by up to 0.15 acum, and roughness by 0.02–0.03 asper, demonstrating effective attenuation of high- and mid-frequency components. Increasing distance from the source caused the expected exponential decay in all parameters. Interaction effects were particularly pronounced: harvesting at high speed without cabin yielded peak annoyance values, while idle/low speed inside a properly equipped cabin produced the lowest.Across almost all conditions, the John Deere exhibited significantly higher SPL (+5–8 dB), loudness (+1.5–4.6 sone), sharpness (+0.03–0.09 acum), and roughness (+0.01–0.02 asper) than the Sampo, indicating less effective noise control and poorer cabin acoustic treatment.
Although both machines operate below the 85 dB(A) threshold for hearing damage risk, the John Deere’s consistently poorer psychoacoustic profile suggests greater potential for long-term annoyance, fatigue, and reduced operator performance. Cabin insulation proved to be the single most effective measure for improving perceived sound quality, followed by operational mode and engine speed management. Gear ratio effects were complex and sometimes nonlinear, highlighting the need for integrated powertrain–acoustic optimization.
These findings demonstrate that compliance with SPL standards alone is insufficient for genuine auditory comfort. Modern combine design must incorporate psychoacoustic targets — particularly reduction of high-frequency and modulated noise sources — to minimize subjective annoyance. The results provide manufacturers with concrete, quantitative evidence for prioritizing cabin sealing, strategic placement of sound-absorbing materials, and active or passive control of dominant noise sources, ultimately contributing to enhanced operator health, sustained productivity, and reduced occupational risk in mechanized agriculture.
This study was carried out with the financial and moral support of the Vice Chancellor for Research of Shahrekord University.
The authors' contributions to this study are as follows:
First Author (Graduate Student): Sample preparation and processing; conducting experiments and data collection; performing calculations and statistical analyses; data analysis and interpretation; and preparation of the original manuscript draft.
Second Author (Thesis Supervisor): Conceptualization and study design; supervision of all research stages; validation and review of results; and manuscript review, revision, and finalization.
Third Author (Thesis Advisor): Contribution to study design; research supervision; and manuscript review and revision.
Fourth Author (Thesis Advisor): Contribution to study design; research supervision; and manuscript review and revision.
During the preparation of this manuscript, the authors used artificial intelligence solely to improve the language and edit the text. After using this tool, the authors carefully reviewed and revised the manuscript and take full responsibility for its content. No artificial intelligence technologies were used in the writing, interpretation, or development of the scientific content of this manuscript.
All data and information supporting the findings of this study are included in this article.
The authors would like to express their sincere gratitude to Shahrekord University for its financial and moral support in conducting this research. They also extend their sincere appreciation to the anonymous reviewers for their valuable constructive and scientific comments, which greatly contributed to improving the quality of this manuscript.
This study received ethical approval from the Ethics Committee of Shahrekord University under the ethics code IR.SKU.REC.1401.030. The authors adhered to the ethical principles governing the conduct and publication of scientific research, and all authors confirm their compliance with these ethical standards.
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
