Experimental and numerical investigations of rain fall induced noise from roofing sheets

Akarsh S, Abhilash P, Abhinav K V, Akshay C C, Sudheesh Kumar C P 1

Department of Mechanical Engineering Government College of Engineering Kannur Kerala, India - 670563

ABSTRACT Rainfall creates noise as it strikes roofing sheets, and causes a lot of disturbances and distactions to the occupants. A wide variety of roofing sheets with di ff erent materials, thicknesses, profiles and other specifications are currently available in the market. Though metallic roofing sheets have high reliability and long life, they have a major drawback of generating large noise during heavy rainfall. Though rainfall induced noise from roofing sheets cannot be completely eliminated, it can be controlled or reduced to a great extent by wise and e ff ective selection and installation of roofing sheet. The main aim of the present work is to study the e ff ects of various parameters on rainfall induced noise from roofing sheets. The parameters considered in the study are the material, profile, thickness and the orientation of the sheets.

1. INTRODUCTION

One of the primary needs of the human beings is nothing but shelter. The concepts of shelter also evolved with human evolution. Humans shifted from caves and huts to roofed structures. Concrete roofing, terracotta, asphalt shingles, slate are the commonly used roofing materials around the world. Among these, metallic roofing sheets are the widely accepted ones owing to their advantages of high durability, long life, and greater protection with good reliability over conventional roofing structures. Their e ff ective life expected is about 40-70 years. They are free from cracks and pest attacks and can withstand rain, winds, etc. But, the main disadvantage is that they generate large noise during rainfall. The world health organization (WHO) recommends maintaining environmental noises below 70 dB over 24 hours to prevent noise induced hearing losses. Hearing sound above 85db for long exposure can cause serious ear problems. The Environmental Protection Agency (EPA), U. S. also specifies limits for speech interference and annoyance at 55 dB for outdoor activities and 45 dB for indoor activities. Noise can cause hypertension, anxiety, pain and injury to ears, cognitive development in children, etc. So it is necessary to maintain specified sound levels in indoor and outdoor regions. Experimental as well as simulation studies were carried out by many researchers on the topic of rainfall induced noise on roofs. A structure was constructed on a concrete floor and various roofing

1 sudheeshkumar3@gmail.com

a slaty. inter.noise 21-24 AUGUST SCOTTISH EVENT CAMPUS O ¥, ? GLASGOW

solutions were tested. A theoretical model was developed to analyse the features of rain noise [1] in the one third octave band frequency range of 0-1600 Hz. A rainfall simulator for simulating rainfall inside a laboratory, meant for the assessment of soil erosion was designed by Mhaske at el. [2] considering various factors such as soil grain. The christiansen coe ffi cient (Cu) of the designed rainfall simulator varies from 81% to 88% with rain droplet size ranging from 1 mm to 5 mm and striking velocities from 5.56 to 9.63 m / s. Simulation studies on di ff erent roofing sheets were carried out by Sreerag et al. [3]. The rainfall is viewed as a source which will apply a pressure on the roof. This pressure is taken as 0.6324 Pa which is equivalent to 90 dB. The room wall is considered to have zero thickness but its influence over sound is considered in the form of impedance i.e.,the product of density of brick and the velocity of sound in brick. The analysis was done in the frequency range of 125-1600 Hz. A method to predict the rain noise of a roof using experimental results conducted in a laboratory was proposed by Yan et al. [5]. McLoughlin et al. [6] showed that the sound intensity level radiated from a corrugated steel roof having trapezoidal profile is proportional to the sixth power of the impact velocity, third power of the drop diameter and depend on water impact rate. From the literature, it can be seen that not many works have been reported on the analysis and control of rainfall induced noise. The present work is to estimate numerically (using COMSOL Multiphysics software) and experimentally the sound pressure levels inside a room due to the rain impact on the roofs. Also, the influence of parameters such as materials, thickness and profiles of roofing sheets are investigated.

2. EXPERIMENTAL INVESTIGATION

The objective is to estimate the sound pressure level (SPL) in dB inside a room subjected to the rainfall on top of its roof. The experimental set up consists of i) a room like structure, ii) a mechanism for roof angle adjustment and iii) a rain simulator. The simulator is used to adjust the intensity of rainfall to light, moderate or heavy. A room with four walls having a size of 2.10 m × 2.10 m × 2.30 m with an opening of 1.90 m height and 0.92 m width, as given in Fig. 1 is built. Walls are constructed using laterite stones. The mild steel frame to carry the roofing sheets consists of two parts; the bottom one for fixing the frame on the room and the top one for fixing the roofing sheet on the frame. A mechanism with two hinges is made for angular adjustment of the roof. Di ff erent types of roofing sheets, as shown in Fig. 2, are used for the experiment. Al-Zn sheet of 0.47 mm and 0.35 mm thicknesses, Al sheet of 0.4 mm and 0.35 mm thicknesses and a galvanized iron roofing sheet of 0.35 mm thick are used. All the sheets are of 2.70 m × 1.1 m size. A water spraying system is fabricated using 1 inch diameter PVC pipes and specially designed nozzles for simulating rain. Two valves are used, one to provide water supply and the other one to control the flow rate. For pumping water, a submersible pump with 1HP motor is used. The measurement of SPL is taken in one third octave band frequency at a height of 1.70 m using RION NL-42 sound level meter. The sound level meter (A-weighted) readings for aluminium roof with an inclination 25 and 0 are shown in Fig. 3. The regulating valve is adjusted to simulate light and moderate rain. By mounting the roofing sheets one by one the SPLs corresponding to di ff erent roof inclinations in the range 0 o -25 o are also taken.

Figure 1: Experimental set up with rain simulator arrangement

ns

Figure 2: Roofing sheets used in the experiment

Table 1: Maximum SPL in dB for di ff erent inclinations.

Angle Al-Zn (.47 mm) Al-Zn (.35 mm) GI (.35) Al (0.4 mm)

0 0 82.4 81.4 85.2 86.6

5 0 81.2 82.4 87.2 86.5

10 0 80.6 82.2 86.4 86.6

15 0 80.2 81.5 85.9 86.4

20 0 78.9 81.3 85.5 86.3

25 0 78.8 81.3 84.4 86.2

Figure 3: SPL inside the room with aluminium roof at an inclination of (a) 25 o and (b) 0 o

2.1. E ff ects of roof angle and sheet thickness Influence of roof angle on SPL is shown in Table 1. Here, roofing sheets of Al-Zn of two di ff erent thicknesses (0.47 mm and 0.35 mm), GI sheet of 0.35 mm thickness and Al sheet of 0.4 mm thickness are used. It can be seen that for all sheets the minimum SPL is obtained when the roof angle is 25 o . This shows that the roof angle has great influence on the SPL generated inside the room. Also, in the case of Al-Zn (47 mm) sheet and Al (0.4 mm) it can clearly be seen that the the SPL gradually decreases as the roof angle is increased from 0 o to 25 o . However,tThese trends are not so clear in the case of Al-Zn (35 mm) and GI (35 mm) sheets. Hence, further investigations are required. SPLs for inclinations beyond 25 o are not anlaysed as real stuctures with roof do not have large slopes. To see the e ff ects of sheet thickness, the SPLs generated due to roofs made of Al-Zn sheets having thickness 35 mm and 47 mm can be compared. It is observed that, increase in sheet thickness reduces SPL, except for 0 o roof angle. This exception could be due to the e ff ects of surroundings while taking measurement, However, to know the exact reason, it needs to be investigated further.

an foam” ie 0s costo toa m™ 74.1

3. NUMERICAL INVESTIGATION

For numerical estimation of the SPL, the software COMSOL Multiphysics is used. As it is di ffi cult to conduct experiments and take measurements of SPL for all types of roofs with di ff erent thicknesses, simulation is more convenient and useful. Here, the acoustic pressure due to rainfall on top of the roof is assumed as 0.2 N / cm 2 which corresponds to a SPL of 80 dB. The SPL generated inside the room for Al-Zn (0.47 mm) and Al (0.4 mm) roofing sheets are shown in Figs. 4 and 5. The obtained results are close to the experimental results, though there are instances where slight di ff erences are observed. These di ff erences could be due to the assumptions made in the theory in comparison to the practical situations.

3.1. E ff ects of roof angles The influence of roof angle has already been discussed in Section 1. To get a more clear picture of the e ff ects, simulation studies are conducted. From Fig. 6, it can be clearly observed that the SPL decreases with increase in the roof angle as observed in the experimental results. The roof material

foam” ie 0 costo ‘or m™ 78.7 «

Figure 4: SPL in a room with Al-Zn (0.47 mm) roof at 0 o roof angle (simulation)

Figure 5: SPL in a room with Al (0.4 mm) roof at 15 o roof angle (simulation)

used in the simulation is Al-Zn sheet with 0.47 mm thickness. Maximum SPL of nearly 82.5 dB is obtained when the roof angle is 0 o , that is when it is laid horizontally on top of the four walls of the room and minimum of nearly 78.5 dB when the inclination is 25 o .

Figure 6: E ff ects of roof angle on SPL in a room with Al-Zn (0.47 mm) roof (simulation)

‘Max 48 method -AlZn(47mm)

Figure 7: E ff ects of sheet thickness on SPL in a room with Al-Zn (0.47 mm) roof (simulation)

3.2. E ff ects of material and thickness Two sheets made of Al-Zn with thicknesses 35 mm and 47 mm are selected to study the influence of roofing sheet thickness on the generated SPL. It can be seen that for all roof angles, except for 0 o , the higher SPL is found in the case of sheet with larger thickness. To investigate the dependency of the sheet material, two di ff erent sheets with same thickness, made of GI (35 mm) and Al-Zn (35 mm) are chosen. The simulation result, as illustrated in Fig. 8, shows that for all frequencies, except for

‘Thickness Effect on noise - Maximum 68 method

Figure 8: E ff ects of sheet thickness on SPL in a room with Al-Zn (0.47 mm) roof (simulation)

2500 Hz, the GI sheet gives larger SPL compared to AL-Zn are chosen.

4. CONCLUSIONS

The experimental as well as the simulation results show almost similar trend. With increase in the sheet thickness, the noise produced during rainfall decreases. Change in the roofing sheet material also shows significant change in the SPLs. Aluminum is found to be the noisiest roofing sheet. It generates SPL of 85 dB even in moderate raining conditions. The SPLs go as high as 90 dB in heavy rains. GI sheet also produces noise above 85 dB in some conditions. Roof inclination also has great influence on the SPL generated during rainfall. Maximum noise is observed at 0 o roof angle and minimum at 25 o . Among the sheets currently available in Indian market Al-Zn is found to be the best in terms of both noise and cost aspects. Though GI sheets are the cheapest, they radiate considerable noise during rainfall. The results obtained in the present work can be of used in the design of quiet rooms.

ACKNOWLEDGEMENTS

We gratefully acknowledge the Centre for Engineering Research and Development (CERD) under APJ Abdul Kalam Technological Universiy, Kerala, India for the fund provided to carry out this work.

REFERENCES

[1] Ballagh, K. O. "Noise of simulated rainfall on roofs." Applied Acoustics 31.4 (1990): 245-264. [2] Mhaske, Sushil N., Khanindra Pathak, and Arnab Basak. "A comprehensive design of rainfall simulator for the assessment of soil erosion in the laboratory." Catena 172 (2019): 408-420.

[3] RK, Sreerag, Vivek Sasidharan, and Sudheesh Kumar CP. "Influence of the Geometry of Roofing Sheets on Rainfall Induced Noise Reduction." (2020). [4] Jansen, H. W. "Rainfall and impact noise measurements on metal roof tiles, TNO Report." Sound and Vibration Division, Stieltjesweg 1 (2004). [5] Yan, Xiang, Shuai Lu, and Junjie Li. "Experimental studies on the rain noise of lightweight roofs: Natural rains vs artificial rains." Applied Acoustics 106 (2016): 63-76. [6] McLoughlin, J., D. J. Saunders, and R. D. Ford. "Noise generated by simulated rainfall on profiled steel roof structures." Applied Acoustics 42.3 (1994): 239-255.