Simulated acoustic field in mudejar-gothic churches

Simula ed Acous ic Field in Mudeja -Go hic Chu ches M. Galindo’, T. Zama ezo’, S. Gi &, J. J. Send a’. and J. Na a o’ ‘Depa men o Applied Physics, ‘Depa men o A chi ec a al Cons uc ions. Se ille Uni e si y. A da Reina Me cedes. 2; 41012~& illa (Spain) Summa y: In his wo k we ha e unde aken an acous ic simula ion o he sound ield in se e al chu ches in he ci y o Se ille. The simula ion has been ca ied ou h ough he RAYNOISE p og am and has been based on he e e be a ion imes measu ed in dmse space?.. We ha e ocused on Sound Ressu e Le els SPL, Cla i y C80 and De ini ion D50 indices da a in ela ion o hei spec al a ia ions and spa ial dis ibu ions and hei compa ison wi h he expe imen al alues. INTRODUCTION Mudeja s yle is a link be ween Islamic a and wes e n Ch is ian a , so i is a uniquely Spanish a is ic mo emen . These chu ches we e p obably buil be ween he hi een h and i een h cen u ies and hey we e buil o b ick. All he chu ches analysed ha e simila a chi ec u al ea u es: h ee na es co e ed wi h wooden amewo ks ollowing in he Moo ish s yle and a aul ed p esby e y. Send a e al (l-3) ha e poin ed ou ha he smalle size o hese chu ches and mainly hei wooden ceilings signi ican ly imp o e hei acous ical beha iou , as a as his can be indica ed by e e be a ion imes, o e he Go hic con empo a ies. The g ea e abso p ion in he ceilings ends o lead o a dec ease o he g ea e e be a ion imes expec ed in hese la ge olumes, abo e all a low equencies. Besides being used o wo ship which implies he use o speech and music, hese buildings a e some imes eno a ed and used as lec u e halls, hea es, o audi o ia. Acco ding o he specialis Angulo (4), his ype o chu ch is called “ he Se ille pa ish ype”. OBJECTIVES Ou main aim in his wo k is o show he ag eemen be ween he expe imen al da a and he simula ed esul s pe o med by he ay acing me hod (RAYNOISE e sion 2.la) in his ype o a chi ec u al enclosu es. We a e specially in e es ed in esea ching he abili y o he so&wa e o p edic he spec al a ia ions and spa ial dis ibu ion o he measu ed acous ic pa ame e s. We mainly ocus on he SPL and on he acous ic indices C80 and D50. The measu emen es s as well as he simula ions ha e been ca ied ou inside he unoccupied chu ches. GENERAL+ ASPECTS Using an i e a i e p ocedu e we ha e adjus ed he abso p ion coe cien s o he di e en wooden ceilings, which a e he mos singula ones, un il we ob ained he simula ed e e be a ion imes in a ange o +lO% o he in si u expe imen al esul s (see able 1). Those abso p ion coe icien alues a e con as ed wi h o he simila da a in he bibliog aphy. The geome ical and acous ical model is buil in his way and he whole simula ion p ocess is execu ed wi h maximum o de o e lec ion R equal o 10. The numbe o ays N is de e mined by he speci ica ions o he pmg am use ’s manual by he o mula 1256,6 1’ N’< S, wi h 1 he mean ee pa h (m) and S, he cha ac e is ic su ace (m’) o he geome ic model. To decide when o abandon a my, in addi ion o he o de o e lec ion, we use a dynamic ang o 60 dB and a ime window, ollowing he manu ac u e ’s sugges ion, &,0294 1 (s). Calcula ions ha e been ca ied ou using he Conical Beam Me hod (CBM) and we ha e conside ed he ai abso p ion and he sound eloci y a he a mosphe ic condi ions o 20°C and 50% ela i e humidi y, simila o ha a he measu emen ime. The backg ound noise in each simula ion p ocess coincides wi h he measu ed noise spec a. Likewise we ha e made calculus modi ying he di usion condi ions o he ceilings and audience zone. This does no ha e a ma ked e ec on any e alua ed pa ame e s in he same way as Lam (5) has al eady poin ed ou . An omnidi ec ional sou ce wi h he known sound powe spec um o he eal one has been used o ob ain simula ed sound p essu e le els. Bo h calcula ion op ions, image and mapping, ha e been se o ob ain he acous ic pa ame e s implemen ed and o compa e wi h he expe imen al da a Recei e su aces and mic ophones ha e been placed a 1,20 m abo e loo le el. TABLE 1: Measu ed e e be a ion ima Tms and S~IIIU~&~ TSim San a Ma ina San icen e San a Ca alina F eq. (W Tmas TS i Tmcas TSh Tmas TSi#ll 250 3,62 3,71 1,71 1,70 I ,50 1.48 500 3396 4.11 1,81 1.79 1,72 I,64 IO00 4,00 4.01 1,82 1,78 1.75 1.68 2GQo 3.57 3,70 1.68 I,70 I,67 1.57 4000 2,62 2,68 I,40 I,45 I ,45 I,36 RESULTS AND CONCLUSIONS Figu e 1 shows a iew o he 3D model c ea ed o San Vicen e chu ch. Figu e 2 shows he measu ed and compu e calcula ed alues o he sound p essu e le els ( ela i e o ha p oduced by he sound sou ce in a ee ield a 1Om) e sus sou ce- ecei e dis ance a 500, and 2000 Hz oc a e FIGURE 1.3D model used o simula ions in San Vicen e Chu ch - - Adjus men 1 Cm) (ml FIGURE 2. LpLp e (dB) e sus dis ance (m) a 500 and 2ooO Hz oc a e bands o San Vic&e chu ch. (m) 1 (ml FIGURE 3. C80 (dB) e sus dis ance (m) a 500 and 2000 Hz O&W bands o San Vic&e chu ch. bands in his chu ch. The same g aphics show he heo e ical cu es ob ained h ough classical di ise ield model, Ba on’s modi ied model (6) and an adjus ed cu e o he expe imen al esul s. As we can see he simula ed le els con inue o dec ease u he we go beyond he c i ical dis ance acco ding o Ba on’s model, bu he measu ed alues di e app eciably om expec ed heo e ical le els al hough he same endency sugges ed by Ba on’s modi ied model o a enua ion by dis ance is obse ed. Simila ly igu es 3 and 4 compa e he measu ed and simula ed alues o C80 and D50 espec i ely as a unc ion o dis ance a he same oc a e bands. In bo h cases we supe imposed (con inuous line) he alues om Ban-on’s modeL Bo h se s o alues show simila spec al and spa ial pa e ns, and hey show hese wo acous ic quali y desc ip o a e a he sensi i e o local a ia ions, he e being no ag eemen wi h he p edic ed heo e ical model. Qui e a sa is ac o y ag eemen be ween measu ed and simula ed esul s has been achie ed by RAYNOISE compu e simula ion, mainly o he ea ly o la e ene gy indices in hese complex ooms. In his espec i is use ul o emembe he unce ain ies ha a ise om he ep oducibili y o measu emen s and absolu e posi ions o mic ophones in a eal chu ch and i s model desc ip ion (7). a 5 10 15 20 25 30 35 0 5 10 I5 2” 25 3” 35 (ml Cm) FIGURE 4. D50 (%) as a imc ioa o dis ance (m) a 500 and 2000 Hz oc a e bands o San Vicea e chu ch We ha e op imised he simula ion p ocedu e aking in o accoun he measu ed e e be a ion imes, so he simula ion p ocess can be used in his way o imp o e any emodeling p ocess om an acous ic poin o iew. In he u u e we a e in e es ed in con as ing hese ac s wi h o he acous ic indices such as STIiRASTI and LEF, also using o he so wa e ools. REFERENCES 1. Send a, J. J., Zama e io, T., “El camp0 sono o en las iglesias Go ico-Mudeja es con cubie a de made a: Aplicaci6n de1 modelo de Ba on,” in P oceedings o Tecni-Anis icu ‘95, XWI Jomadas Nacionales de Anis ica, La CON@ Spain pp. 87-90, 1995. 2. Se&a, J. J, Zama io, T., Na a o, J., “An analy ical model o e alua ing he sound ield in Go hic-Mudeja chu ches,” in P oceedings o Second In e na ional con e ence on Compu a ional Acous ics and i s En i onmen al Applica ions Compac 97, Temi, I aly, pp.l39-148, 1997. 3. Zama e io, T., Galindo, M., Gi 611, S., “P edic ion de Indices ac is icos en iglesias Go ico- Mudeja es,” in P oceeding o XV7 Reunidn Bienal de la Real Sociedad Espa iola de Fisica y Quimica, Las Palmas de G an Cana ia, Spain, pp.25-26, 1997. 4. Angulo, D., “‘A qui ec u a mud&ja se illana de 10s siglos XIII, XVI y XV,” Ayun amien o de Se illa, Se illa , pp. 32, 1983. 5. Lam, Y. W., Jou nal o he Acous ical Socie y o Ame ica 100, pp. 2181-2203 (1996). 6. Ba on, M., Lee, L. J., Jou nal o he Acous icul Socie o Ame ica 84, pp. 618-628 (1988). 7. Pelo son, X., Vian, J. P., Polack, J. D., AppliedAcous ics 37, pp. 175-198 (1992).