Sound power certification of a dodecahedron shaped sound source

Please download to get full document.

View again

of 9
9 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Sound power certification of a dodecahedron shaped sound source
Document Share
Document Tags
Document Transcript
  Sound power certification of a dodecahedron shapedsound source Gilberto Fuchs de Jesus a , Rodolfo Venegas  b , Marco Nabuco c , Paulo Massarani da GROM Acústica & Automa!o, Rio de Janeiro, "#""#$"%,RJ, 'ra(il  b )e*artamento de Acustica, +niersidad P-re( Rosales, 'ro.n Norte "/#, 0antiago, 1hile, c,d Acoustic 2esting 3aborator4, 5nmetro, )6 de 1a7ias, "8 "8#$#"#, RJ, 'ra(il a gilberto@grom.com.br; b rodolfo.venegas@gmail.com c nabuco@inmetro.gov.br; d pmmassarani@inmetro.gov.br, Abstract Reference sound sources must be stable, omnidirectional and shall *resent afre9uenc4 s*ectrum as flat as *ossible, .ithin standardi(ed limits6 2he reference soundsources most used on acoustic measurements are constructed as centrifugal fans thatgenerate aerod4namic broadband noise6 2hese sound sources are er4 stable due tomechanical and electrical inertia .hich guarantee an almost inariable rotation s*eedduring its o*eration6 2he4 can be o*erated for long *eriods of time .ith neglectingmodification on their out*ut sound *o.er6 Neertheless, due to constructie limitations,the sound *o.er leel of aerod4namic sources is limited to around /8 d':A;6 <enthough this sound *o.er leel is enough for man4 laborator4 and field a**lications,some tests re9uire more sound *o.er in=ection in reerberation chambers or on fieldmeasurements6 Performance and endurance tests of electronic deices, .hen submittedto high sound leels, and in situ  sound isolation bet.een *artitions are t.o commone7am*les of acoustic tests in .hich usual reference sound sources do not generateenough acoustic *o.er6 2his *a*er describes a test set$u* as .ell as some results of anelectro$acoustic dodecahedron sha*ed sound source tested in a reerberation chamber62he data can be used as a *ossible standardi(ed *rocedure to ealuate this >ind of sound source6 0ome endurance and re*eatabilit4 inestigation data are also *resented as *art of the alidation tests for a ##? 'ra(ilian made dodecahedron6 1.INTRODUCTION )odecahedral louds*ea>ers hae been .idel4 used as omnidirectional sound sources for along time6 0eeral manufacturers *roduce this >ind of source .hich are used in numerousa**lications6 Neertheless, no reference .as found in the contem*orar4 literature of an4attem*t to erif4 .hether these sources could be used as reference sound sources according tothe international standard 50O @/"@/// BC6 2he abilit4 to generate high sound *o.er leelsof louds*ea>ers is normall4 neglected in face of the lac> of confidence on their stabilit46  2he necessit4 of *o.erful and stable sound sources for 5NM<2RODs Acoustic 2ests3aborator4 :3A<NA; and the increasing 9ualit4 of toda4Ds louds*ea>er ins*ired thedeelo*ment of a ##? 'ra(ilian dodecahedron6 2he source model used for the tests re*ortedherein .as initiall4 designed b4 3A<NADs researchers and their first *rotot4*e unit, testedduring this stud4, .as handcrafted b4 5NM<2RODs car*enters in *l4.ood6 As GROM Acústica & Automa!o decided to turn that *rotot4*e into a *roduct, the srcinalfour inch 'ra(ilian made coa7ial louds*ea>er model .as re*laced b4 a similar and more *o.erful one from the same manufacturer6 2he *entagon sha*ed faces .ere them machinedfrom M)F .ood .ith industrial *recision so the sound source could be assembled in a *roduction scale62.o different units .ere inestigated for this stud46 One is the first handcrafted *rotot4*eo.ned b4 5NM<2RO and the other one .as the first unit sold b4 GROM6 2he second unit belongs to 5NP<, the National 5nstitute of 0*ace Research, that >indl4 authori(ed the authorsto use the data collected during the certification tests *erformed .ith their unit62he sound *o.er measurements made for both dodecahedron units tested sered different *ur*oses6 2he tests made .ith the 5NP< unit .ere used to generate an accredited re*ort for that *articular unit, indicating its sound *o.er s*ectrum for a single in*ut oltage6 2hemeasurements made .ith the 5NM<2RO unit .ere more com*lete and coered a .ider rangeof conditions6 2he different conditions .ere tested to inestigate the dodecahedron as a *ossible reference sound source6 2.MEASUREMENT DESCRITION ! TEST SECI"ICATION2.1 #enera$ 2he electrical characteristics of the louds*ea>er, >no.n from the manufacturer data sheet,.ere indiiduall4 chec>ed for each of the " units at the time of the first *rotot4*econstruction6 2he directionalit4 of that unit .as measured b4 FEbio Arruda B"C bac> in ///62he tests *erformed b4 Arruda did not coer the entire fre9uenc4 range *rescribed b4 50O@/"@ and sho.ed the source directiit4 *resented lobular *atterns aboe >(6 2hea**arentl4 inade9uate directiit4 inde7 : N di ; of the handcrafted *rotot4*e ma4 be caused b4the lac> of *erfect s4mmetr4 of the source faces6 5t is e7*ected that the machined constructionof GROMDs model and the use of the *rotection grid ma4 im*roe this characteristiceentuall4 ma>ing N di fulfil the re9uirements of 50O @/"@ for a .ider fre9uenc4 range62here .ere t.o biggest concerns related to the tests6 2he first one .as the *h4sical integrit4of the louds*ea>er units the e7*ected long duration of the tests at high *o.er out*ut coulddegrade the 9ualit4 of the source itself causing ariations on sound *o.er s*ectra during thelong term stabilit4 inestigation6 2he influence of the tem*erature increase at the louds*ea>er coil could cause out*ut ariations eentuall4 oiding the results6 2he second one .as there*roducibilit4 of the test set$u*6 2he choice of the test *arameters that could influence themeasurement results and a .a4 to control them in an a**ro*riate .a4 .as carefull4 discussed b4 the authors before starting the measurements6  2he main goal .as to determine an engineering *rocedure that could be du*licated and under .hich the sound *o.er leel : L W ; emitted b4 the source, as .ell as its s*ectrum could bemeasured b4 a standardised method6 2he tests .ere re*eated at least H times at each source *osition in order to alidate is re*eatabilit4 of the *rocedure and the source stabilit462he tests described in the ne7t sections .ere carried out to determine four basiccharacteristics of the source its sound *o.er s*ectrum at a gien in*ut oltage :so callednominal test oltage;, the relationshi* bet.een the sound *o.er leel and the in*ut oltage,the long term stabilit4 of the sound *o.er s*ectrum and the *ossible influence of the sourceeleation aboe the test room floor on its sound *o.er leel and s*ectrum6 2.2 E%uip&ent and Test Set'up 2he e9ui*ment used for the tests re*orted herein are listed on table 6 2he "HH anal4ser .asused to generate a *in> noise signal that .as then am*lified b4 the *o.er am*lifier to driethe dodecahedron6 2he driing signal .as monitored b4 channel ' of the %# anal4ser toma>e sure the e7citation signal .ould not ar4 during the sound *o.er measurements6 Table 1: List of equipment used for the tests. <9ui*mentModelMultimeter P HI%AFunction Generator P HH"#ARamdom 5ncidence Micro*hones 3arson )ais "88/# channel mico*hone multi*le7er 3arson )ais Model ""#Real 2ime Anal4ser Norsonic R2A %#Real 2ime Fre9uenc4 Anali(er '&J 24*e "HHumidit4 & 2em*erature meter Vaisala MH1alibrator '&J 24*e "HReference 0ound 0ource '&J t4*e "#0trobosco*e General Radio 8@General Radio 8@Po.er am*lifier 0tudio R A'0%## 2.(Test Description 2he sound *o.er measurements .ere made using the com*arison method *roceduresdescribed on the international standard 50O HI BHC6 A fi7ed arra4 of si7 micro*hones .as *ositioned inside one of 5NM<2RODs reerberant rooms and a reference sound source t4*e"# .as used for the com*arison measurements6 2hree source *ositions .ere used in eachsound *o.er measurement6For each source *osition the signals from the @ micro*hones in the fi7ed arra4 .ere fed intochannel A of the %# anal4ser through the ""# multi*le7er6 One micro*hone *osition .asta>en as reference and used to calibrate the anal4ser in*ut6 All other micro*hones .ere thenindiiduall4 calibrated and each offset recorded for further correction of the collected data62he reerberant room used for the tests has a olume of /@m H  and three source *ositionscould be mar>ed .ith no need for changing micro*hone *ositions6  2he channel ' of the %F# anal4ser .as calibrated using a & >G( *ure tone :sine .ae;generated .ith the am*litude of #6& V RM0s  b4 the HH&"#A function generator6 2his signal .asfed to the *o.er am*lifier .hich .as ad=usted to delier & V RM0s  on its out*ut6 All oltagereadings .ere made .ith the HFI%A multimeter6 2he am*lified signal used to drie thedodecahedron .as also connected to channel ' of the %F# R2A6 2his channel reference .asset to &# $@  V RM0s 6 After this calibration *rocedure the %FA channel ' indicated &"# d' re &# $@ V RM0s  at the &6#>G( &KH octae band62he 5NM<2RO *rotot4*e used for the tests is e9ui**ed .ith t.ele 0<3<N5+M coa7iallouds*ea>ers model 10AF#F6 2his louds*ea>er model has a *a*er cone and is rated H8L RM0 62he louds*ea>ers are connected in a series$*arallel arrange that results in a nominalim*edance of 86HH Ohms6 2his im*edance .as used to calculate the electrical RM0 *o.er of the dodecahedron based on the indiidual rated *o.er of the louds*ea>er6 2he nominal RM0 electrical *o.er of the louds*ea>ers .as determined b4 the manufacturer according to the 'ra(ilian standard N'R&#H#H6 5n the tests re*orted on the *roduct datasheet, the louds*ea>ers are continuousl4 drien at the RM0 *o.er for t.o hours in a F8 litresealed bo76 Once dodecahedron has t.ele louds*ea>er units assembled in a much smaller olume and the heat dissi*ation could reall4 be a *roblem the driing oltage to thedodecahedron .as limited a fraction of its calculated RM0 oltage6 5t .as e7*ected thatdoing so the tem*erature increase .ould be >e*t .ithin reasonable alues, minimi(ing the *ossible distortions associated .ith the resulting modifications of the electromagneticcharacteristics of the louds*ea>ers6 2he theoretical RM0 electrical *o.er of 5NM<2ROsdodecahedron *rotot4*e is F"# L .hich corres*onds to a driing oltage of FI6H V RM0 65n order to *resere the integrit4 of the sound source and considering a long term stabilit4 test.as *lanned to last u* to F hours and H# minutes .ith continuous o*eration, the driingoltage .as limited to &#V RM06  2his oltage corres*onds to F68? of the rated RM0 electrical *o.er of the dodecahedron6 2he relationshi* bet.een the in*ut oltage and the sound *o.er emitted b4 the sound source.as inestigated b4 com*aring the sound *o.er s*ectrum of the dodecahedron .hen drien b4 four different oltages, sho.n on table "6 2hese oltages .ere set b4 ad=usting the *o.er am*lifier to a fi7ed gain and ar4ing the *in> noise leel on its in*ut6 2hese *rocedure .aschosen due to the difficult4 of controlling the am*lifier gain6 As a dra.bac>, the driingoltage could onl4 be set in fi7ed ste*s on the "&HH anal4ser .hich resulted in the aboeuneenl4 rounded alues of the driing oltage6 Table 2: Input voltages  Nominal Voltage:V RM0 ;Real Voltage:V RM0 ;Voltage leel:d' re &# $@ V;#6@#68/I  115,5 &6@&6@"%  124,2 86@86@"H  135,0 &#/6II"  139,8  )uring each test run the Leq  leels and s*ectra .ere measured for both in*ut oltage andout*ut sound *ressure6 2he Leq  oltage leels aried from run to run .ithin a K$ &d' range62he difference bet.een the nominal in*ut leel and the actual measured leel for each run.as used to normali(e the corres*onding Leq  sound *ressure leel to e7actl4 the same in*utoltage62he 5NM<2RO unit .as also submitted to an endurance test in order to ealuate its long termstabilit46 )uring this test the in*ut oltage leel .as >e*t constant and set to the nominalalue of &#V RM06  2he source height .as set to &6HIm aboe the room floor6 )uring theendurance test the source .as turned on continuousl4 for Fh H#min and four sound *o.er measurements .ere *erformed6 2he time interals bet.een source actiation and eachmeasurement start are sho.n on table H6 Table 3: Measurement intervals Measurement 5nde72ime before measurement start &&# mim"&h &# mimHF hFF h H# mim2he last test *erformed .as the sound *o.er measurement on 5NM<2RODs dodecahedron atdifferent heights6 2he source stand .as ad=usted in three different *ositions aboe the roomfloor6 2he in*ut oltage .as >e*t constant and the sound *o.er .as measured after &#minutes of continuous o*eration for each height6 2he source .as turned off bet.een eachmeasurement to allo. *ro*er cooling6 2he three different height are sho.n on table F6 Table 4: Soure heights 0ource heights #68Im#6/Im&6HIm (. MEASUREMENT RESU)TS 2he 5NP< source .as tested .ith the in*ut signal set to &" V RM0 6 2he source height .as >e*tconstant and the srcinall4 su**lied tri*od .as used to hold the source6 2hree measurementruns at three different *osition each .ere made .ithin t.o consecutie da4s6 2he sound *o.er s*ectra calculated as *er 50O HIF& for each run .ere then aeraged to be *ublished asthe source characteristic s*ectrum6 2he com*arison of this s*ectrum .ith the corres*ondings*ectrum obtained for the 5NM<2RO source is *lotted in figure &62he *lot on figure " sho.s ho. the sound *o.er s*ectrum changed .hen the 5NM<2ROsource .as drien b4 four different oltages at a constant height6 5t is clear that the differents*ectra =ust shift u* .ith the oltage increase, >ee*ing its sha*e unchanged6
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks