Evaluation of a Modern Catholic Church by Acoustical ... · the space at issue, elementary room acoustic measurements according to ISO 3382 (Acoustic-Measurement of the reverberation

21

2003/3 PAGES 21 � 25 RECEIVED 5. 6. 2003 ACCEPTED 15. 12. 2003

2003 SLOVAK UNIVERSITY OF TECHNOLOGY

M. RYCHTÁRIKOVÁ, M. MUSABUDA, F. OHRABLO

Evaluation of a Modern CatholicChurch by Acoustical SoftwareBased on the Ray-TracingMethod

ABSTRACT KEY WORDS

This article is based on a acoustical analysis of a modern Catholic church in the regionof Antwerp (Belgium). The main issue is the bad intelligibility of speech. This problemintended to be resolved using Acoustical Software based on the ray-tracing method. Inthe space at issue, elementary room acoustic measurements according to ISO 3382(Acoustic-Measurement of the reverberation time of a room with reference to otheracoustical parameters) using Dirac software were made. Three microphone positions(marked in Fig.1) in the middle of each audience area were defined. A computer modeland simulation in the CATTacoustic ver 7.2 was done. In the analysis, results of in situmeasurements and computer simulation were considered and compared as well as a newpurpose for the ceiling was given. This text includes the results of the computersimulation of the church model with the new purpose for the ceiling.

• room acoustics,• ray-tracing method,• reverberation time (RT),• Clarity C80,• Centre of gravity TS.

Monika Rychtáriková (PhD student)Slovak University of Technology, Faculty of Civil Engineering,Department of Building Structures,Radlinského 11, 813 68, Bratislava, Slovak Republic

Mons. Maxwell Musabuda (PhD student)Catholic University of Leuven, Faculty of Theology.St.-Michielsstraat 6, B-3000, Leuven, Belgium

Prof.Ing.Franti�ek Ohrablo, PhD.Slovak University of Technology, Faculty of Civil Engineering,Department of Building Structures,Radlinského 11, 813 68, Bratislava, Slovak Republic

Fig.1. 3D Model and Ground Floor Scheme of St. Paulus Church. (M1-M3 - microphone positions, B - position of sound source)

rychtarikova.qxd 11.8.2004 20:52 StrÆnka 21

22

INTRODUCTION

The fundamental requirements of a church are a good listening envi-ronment for both speech and music and all the spaces within thechurch require quiet surroundings which are conducive to undistur-bed meditation and prayer.Yet, people still have problems with hearing and understanding whatis being said and music that is being played. Good sound in thechurch is still asking for many congregations. To evaluate a churchas an acoustical space, classic room acoustic criteria of the ISO stan-dard are used. Usually, problems arise with determining the recommen-ded values for the space. The modern day worship service involvesthe active participation of the pastor, the performing musical group,and the congregation. Each participant experiences a unique set ofproblems. This case raises philosophical questions concerning people�sattendance at worship. It also leads to one of the greatest problemsfor churches these days, estimating the percentage of people inchurch during ceremonies in orden actual to obtain values of thetotal absorption in a space.

General Problems, Requirements and BasicRecommendation for Church Acoustics

The prime consideration in the acoustical design of a church is thenoise control (NC) level of the church and the proper control ofreverberation. The noise control level of the church describes the background noisein the sanctuary and is too high, especially in the large cathedralsplaced in the centers of big cities. The typical conversation levels ofpeople are around 60dB on the average. To hear speech at acomfortable level and to be able to understand the words beingspoken, the level of speech usually has to be 25dB above any noise.The other concern is that NC is not a constant throughout thesanctuary. The hum of florescent lights or a ceiling fan above theplatform or altar can raise the NC in the front of the church to NC40,while the rest of the church is NC30. [Tab.1]The average reverberation time (RT) in a church is one of the firstclues as to whether there are problems or not. The traditional orLiturgical style of worship can make good use of longer

reverberation times. However, any reverberation time that is longerthat 2.5 seconds is not needed, and it is no longer musical unless itis well diffused. If the reverberation time in the church is longer than2.5 seconds, then it is most likely to have a problem with both musicand speech, and this also degrades the performance of a churchorgan.But not all reverberation is equal. Long reverberation times in thelow/mid range of speech is not good. Longer reverberation timesbelow 500 Hertz will have a profound effect on speech degene-ration. What is interesting is the effect of bass sound in the church.From experience, the ideal reverberation time at 200 Hertz is whenit is within 0.2 of a second, of the measured average time, for anappropriate style of worship. Because of low frequencies, it is goodto check the position of the sound source. The wave length of 200Hz is 1.7m, and for 100 Hz, is 3.4m, so if the bass player is standingcloser to the loudspeaker in order to hear the sound wave, he has towait for the sound to bounce off a wall that is further than the wavelength and perpendicular to the source.

This often means that the sound has traveled to the other side of theroom and come back. Also, if the bass player is standing against awall, he/she still will have problems hearing the bass sound. Highfrequencies (over 2 kHz) give a sound its character and wholeness.There are high frequency sounds which give us our intelligibilityclues and to music the expressions we love and enjoy.Today many newly - built Catholic churches carry considerablesound absorbent treatment for improved hearing conditions, evenwhen organ music renditions are planned during services. Optimalreverberation time for churches is shown in Fig. 2, where bothspeech and music are considered.

2003/3 PAGES 21 � 25

EVALUATION OF A MODERN CATHOLIC CHURCH BY ACOUSTICAL SOFTWARE ...

125 Hz 250 Hz 500 Hz 1000 Hz 2000 Hz 4000 Hz

NC-40 [dB] 56 50 45 41 39 38

NC-35 [dB] 52 45 40 36 34 33

NC-30 [dB] 48 41 35 31 29 28

Tab.1 Noise Control level values [dB]

2,2

0,8300 3000

2800 m3

(St. Paulus Church)

30000

Catholic churchesProtestant churchessynagogues

V [m3]

RT [s]

1,0

1,2

1,4

1,6

2,0

1,8

Fig. 2 Optimal reverberation time at 500 Hz, assuming two-thirdsoccupied, as a function of enclosure volume [2]


St. Paulus Church

St. Paulus Church is a modern Catholic church constructed in thesecond half of the 20th century (1964) in the region of Antwerp(Belgium), by the architect P. van Aerschot. The main shape of thechurch is simple and logical. The application of traditional materialssuch as brick, wood and glass create an intimate atmosphere in itsinterior [Fig. 3]. The Church is located in the quiet of the forest inWestmalle, close to the Abdij, where one of the famous traditionalBelgian Trappist beers is brewed and bottled.The placement of the church in the forest already avoids any pro-blem of background noise. The main construction of the room, desi-gned by keeping to fundamental room acoustic requirements, makesthe space diffuse enough and the application of reflective materialson the inner surfaces makes it reverberant. The floor is made fromstone tiles without the application of carpet on it. The walls are createdfrom unplastered bricks, and the whole ceiling is made of gypsumboard with very low sound absorption. The church use wooden chairswithout upholstered materials for seats. The total volume V= 2870 m3

with a total floor surface area (audience included) is Sfloor = 439 m2.The area of the audience is Saudience = 200 m2. The total surface areaof the ceiling is Sceiling = 630 m2 (the ceiling over the audience is304 m2 ). The ground clearance of the ceiling is relatively low: h1 = 2.8 m over the altar, and the audience is doubled to h2 = 5.5 m.

Measurements and Computer Simulation

Measurements according to ISO 3382 (acoustical measurement ofthe reverberation time of a room with reference to other acousticalparameters) were done, and the results showed, that the

reverberation time, as a first indicator of the room´s acousticalproblems, is more than 4 sec. in the middle frequencies, which is toolong for having good hearing conditions for speech in the room[Fig.4].On the other hand, the decay curve [Fig. 5] in all the measuredpositions is straight without showing any special acousticalproblems of echoes, bad diffusion or other effects in the room.Fig.4 shows a comparison of the measurements and computersimulation. In fact, the longer the RT is, the more difficult it is toestimate its accurate duration. Also, with the increase in RT, thehuman ear gets less sensitive to recognize the difference in the

23

2003/3 PAGES 21 � 25


Fig.3 Fotodocumentation of the Interiour of the Church

Fig.4 Reverberation time RT, Early decay time EDT, according tomeasurements and computer simulation. Difference in resultsobtained from measurement and computer simulation according toposition of microphone

Fig.5 Decay curve at the position M1 (Dirac software measurements)


length of the reverberation. In our case, the incorrect results are onlyin the low frequencies in T 30, which is the typical problem of ray-tracing programs. The differences in the results of the high andmiddle frequencies are negligible.In order to test the range of the validity of the ray-tracing method,other room acoustic criteria were regarded. Fig. 6 shows thedifferences in Central Time Ts, Clarity C80 and Deutlichkeit D50.Ts � the values of the actual ceiling are highest in the low and themiddle frequencies (Ts= 350ms). In the high frequencies is Ts = 250-150ms. In this case, the simulation results are good. The values ofC80 and D50 are more dependent on the position of the receiver, butbecause of the limited extension of this article, the average valuesare given. According to the statistics, the best results obtained fromsimulation are in the middle frequencies, and the curves are flatter.The differences in Rasti values (3-6) are negligible.

Purpose of the New Ceiling

Because of speech domination (in comparison with the playedmusic), our task was to solve this problem without using or placingany electroacoustical instruments in the church. We decided tochange the total absorption of the room to a value that will still keep

some reverberance appropriate for musics performances. As seen inTab.1, we decided to take RT = 1.6s as a reference value of thereverberation time. Thanks to the diffuse sound field [Fig.5], we canuse the Sabin formula, based on statistical calculations, using thetotal room volume and the total absorption present, as the first stepin our purpose. To obtain the value of the total absorption area, thebasic formula of RT was used:

actual total abs.: Aac = (0.163 * 2870)/4.5 = 104 m2

necessary absorp.: An = (0.163 * 2870)/1.6 = 290 m2

For total absorption we need to add: A = An � Aac = 290 � 104 = 186m2. If we have a look into the interior of the church, the onlypossible surface we can change without violation of its innerarchitecture is the ceiling. The surface area of the ceiling over theaudience is S = 304 m2. We need A = 186 m2. To obtain thenecessary absorption coefficient, we used the simple formula: a = A/S = 186 m2/304 m2 = 0.61.Regarding the requirement of new absorption properties of theceiling over audiences (a = 0.61), a new computer simulation using

[ ] [ ]sRT

VAs

AV

RT*163,0

163,0 =→=

24

2003/3 PAGES 21 � 25


Fig.6 Results and differences in other room acoustic criteria: Ts, C80 and D50


the new properties of the ceiling was made. Fig.7 shows thereverberation time and the early decay time as obtained from thecomputer simulation.The exchange of reflective gypsum boards to new absorptiveboards, with properties a = 0.61, will not change the visualimpression of the church�s interior, but it will considerably changeits acoustical conditions. It will surely cause a new subjectivesensation for the audience. The question is wheter they will like thenew sound of the church, but the hearing conditions will certainly beappropriate to the purpose of this church.

CONCLUSION

Our example confirms that the range of the validity of Ray tracingsoftware is mostly limited to low frequencies. Room acoustics,digitally modeled with impulse responses calculated by ray-tracingor image-source methods, are based on geometrical acoustics. Thisis applicable to high frequencies or in simple rooms, without a largeamount of columns and special interior elements in the space, wherediffraction can be ignoreded and so the results obtained from thesimulation are satisfactory. The article is published with in the grant No. 1/8331/01 VEGA

25

2003/3 PAGES 21 � 25


Fig.7 Computer simulation results of T30 and EDT in alternative of new ceiling with (a = 0.61) and their comparison with actual valuesin, the church.

REFERENCES

� ISO 3382 Acoustic measurement of the reverberation time ofa room with reference to other acoustical parameters

� RETTINGER, M. (1968) Acoustics, room design and noisecontrol, New York

� KNUDSEN, V. O. - Harris, C. M. (1978)Acoustical Designingin Architecture, Acoustical Society of America

� FURRER, W. (1961)Raum und Bauakustik Laermabwehr, Basel


Evaluation of a Modern Catholic Church by Acoustical ... · the space at issue, elementary room acoustic measurements according to ISO 3382 (Acoustic-Measurement of the reverberation

Documents