HamCall (October 1991)

home *** CD-ROM | disk | FTP | other *** search

/ HamCall (October 1991) / HamCall (Whitehall Publishing)(1991).bin / bcast / audvid / crown000.txt < prev next >

Wrap

Text File | 1990-10-14 | 20KB | 472 lines

TESTIN╟ MICROPHONE╙ WIT╚ TIM┼ DELA┘ SPECTROMETRY b∙ Brucσ Bartlett Crowε International¼ Elkhart¼ Indiana Froφ AE╙ Preprin⌠ 213│ (B-1)¼ 76tΦ Conventioε oµ thσ Audi∩ ì Engineerinτ Society¼ 198┤ Octobe≥ 8-11¼ Ne≈ York« Thσ ì figure≤ referreΣ t∩ iε thσ tex⌠ arσ iε thi≤ preprint. Abstract║ Therσ i≤ ß new¼ cost-effectivσ anΣ efficien⌠ wa∙ ì t∩ measurσ microphones║ Timσ Dela∙ Spectrometr∙ (TDS)« Aε ì examplσ oµ ß TD╙ measuremen⌠ facilit∙ i≤ described« Topic≤ ì covereΣ arσ TD╙ theory¼ test-rooφ design¼ tes⌠ procedures¼ ì measuremen⌠ examples¼ anΣ disadvantage≤ anΣ advantage≤ oµ ì TD╙ vs« anechoic-chambe≥ methods. ░ INTRODUCTION Conventiona∞ anechoic-chambe≥ measurement≤ oµ ì microphone≤ requirσ costl∙ fiberglas≤ wedges¼ cablσ floorinτ ì anΣ ß room-suspensioε system« B∙ contrast¼ ß TD╙ ì measuremen⌠ installatioε require≤ littlσ o≥ n∩ acoustica∞ ì treatmen⌠ o≥ vibratioε isolation« Al∞ that'≤ needeΣ i≤ aε ì adequate-sizσ empt∙ rooφ anΣ ß TD╙ analyzer. Somσ othe≥ time-selectivσ measuremen⌠ method≤ tha⌠ ì eliminatσ thσ neeΣ fo≥ anechoiπ condition≤ arσ thσ gateΣ ì FFT¼ tone-burst¼ impulse¼ anΣ cross-spectruφ method≤ [1]« ì Thσ TD╙ techniquσ developeΣ b∙ Heyse≥ [2▌ offer≤ superio≥ ì signal-to-noisσ rati∩ [1]« Tha⌠ i≤ becausσ i⌠ use≤ ß ì higher-energ∙ tes⌠ signa∞ anΣ filter≤ ou⌠ backgrounΣ noise. A≤ aε examplσ oµ ho≈ thσ TD╙ systeφ caε bσ pu⌠ int∩ ì practice¼ thi≤ repor⌠ describe≤ ß TD╙ tes⌠ facilit∙ whicΦ ì ha≤ beeε iε usσ fo≥ tw∩ year≤ fo≥ prototypσ anΣ productioε ì testing. ▒ TD╙ THEORY 1.▒ Thσ filter-trackeΣ sine-wavσ sweep Beforσ describinτ thσ tes⌠ facility¼ let'≤ briefl∙ ì explaiε ho≈ TD╙ work≤ iε ß microphone-measuremen⌠ ì arrangement« Thσ TD╙ analyze≥ generate≤ ß sine-wavσ ì frequenc∙ swee≡ whicΦ i≤ playeΣ througΦ ß loudspeaker« Thσ ì changσ iε frequenc∙ i≤ linea≥ witΦ time« Thσ microphonσ ì unde≥ tes⌠ pick≤ u≡ thσ sounΣ generateΣ b∙ thσ loudspeaker¼ ì anΣ thσ microphonσ signa∞ i≤ filtereΣ b∙ ß bandpas≤ filte≥ ì tha⌠ track≤ thσ generateΣ swee≡ (Fig« 1)« Thσ trackinτ filte≥ i≤ no⌠ iε synπ witΦ thσ generateΣ ì frequenc∙ sweep« I⌠ i≤ time-offse⌠ t∩ compensatσ fo≥ thσ ì propagatioε dela∙ oµ sounΣ travelinτ froφ speake≥ t∩ ì microphone« Fo≥ example¼ a⌠ thσ instan⌠ ß 1000-H· tonσ ì reache≤ thσ microphonσ througΦ thσ air¼ thσ tracking-filte≥ ì cente≥ frequenc∙ i≤ se⌠ t∩ 100░ Hz. 1.▓ Rejectioε oµ reflections No≈ supposσ tha⌠ thσ loudspeaker'≤ sounΣ reflect≤ ofµ ß ì wal∞ anΣ enter≤ thσ microphonσ afte≥ ß certaiε delay« B∙ ì thσ timσ thσ reflectioε enter≤ thσ microphone¼ thσ trackinτ ì filte≥ wil∞ havσ swep⌠ t∩ ß highe≥ frequenc∙ thaε thσ ì reflectioε (Fig« 2)« Iµ thσ filte≥ bandwidtΦ i≤ sufficientl∙ ì narrow¼ thσ reflectioε i≤ rejecteΣ o≥ filtereΣ out« N∩ ì reflectioε signal≤ arσ receiveΣ b∙ thσ TD╙ analyzer« Iε ì othe≥ words¼ aε anechoiπ measuremen⌠ ha≤ beeε madσ iε aε ì ordinar∙ room. Iε aε anechoiπ chamber¼ sounΣ strike≤ thσ sound- ì absorbinτ materia∞ oε thσ interio≥ surface≤ anΣ i≤ absorbeΣ ì s∩ tha⌠ littlσ o≥ n∩ reflectioε occurs« WitΦ ß TD╙ system¼ ì therσ i≤ n∩ acoustiπ treatment¼ s∩ reflection≤ d∩ occur« ì Bu⌠ the∙ arσ rejecteΣ b∙ thσ sweeping¼ narrowbanΣ filte≥ iε ì thσ TD╙ analyzer. 1.│ Filte≥ bandwidth Thσ bandwidtΦ oµ thσ trackinτ filte≥ caε bσ preset« Thσ ì wide≥ thσ bandwidth¼ thσ greate≥ thσ "timσ window.ó Thi≤ i≤ ì ß rangσ oµ timσ ove≥ whicΦ signal≤ arσ accepteΣ b∙ thσ ì analyzer« Thσ relatioε betweeε timσ window¼ bandwidth¼ anΣ ì swee≡ ratσ i≤ ╘ ╜ B/╙ wherσ ╘ ╜ widtΦ oµ timσ windo≈ iε ì seconds╗ ┬ ╜ bandwidtΦ iε Hz¼ anΣ ╙ ╜ swee≡ ratσ iε Hz/seπ ì [2,3]. 1.┤ Spacσ window Sincσ sounΣ travel≤ ß certaiε distancσ withiε ß timσ ì interval¼ thσ timσ windo≈ correspond≤ t∩ ß "spacσ window"« ì Thσ spacσ windo≈ i≤ aε ellipsoiΣ spacσ arounΣ thσ speake≥ ì anΣ microphone¼ insidσ oµ whicΦ sounΣ reflection≤ arσ ì includeΣ iε thσ measurement« SounΣ reflection≤ originatinτ ì outsidσ thσ spacσ windo≈ arσ excludeΣ froφ thσ measurement« ì Actually¼ the∙ arσ attenuateΣ │ d┬ a⌠ thσ edgσ oµ thσ spacσ ì windo≈ ellipsoid¼ anΣ b∙ greate≥ amount≤ outsidσ that. è Thσ speake≥ anΣ microphonσ arσ a⌠ thσ fociΘ oµ thσ ì ellipsoiΣ (Fig« 3)« Iµ thσ fociΘ (speake≥ anΣ microphone⌐ ì arσ separateΣ b∙ ß distancσ tha⌠ i≤ smal∞ compareΣ t∩ thσ spacσ windo≈ (say¼ ▓ fee⌠ apar⌠ fo≥ ß 10-foo⌠ spacσ window)¼ ì thσ spacσ windo≈ ellipsoiΣ approximate≤ ß sphere. Iε tha⌠ case¼ i⌠ i≤ simplσ t∩ calculatσ thσ requireΣ ì space-windo≈ sizσ t∩ excludσ reflections« Yo⌡ measurσ thσ ì distancσ froφ thσ microphonσ t∩ thσ neares⌠ surface¼ anΣ se⌠ ì thσ spacσ windo≈ smalle≥ thaε twicσ tha⌠ distance« Fo≥ example¼ supposσ thσ floo≥ i≤ thσ neares⌠ surface¼ ì anΣ i⌠ i≤ ╡ fee⌠ froφ thσ microphone« Thσ spacσ windo≈ ì shoulΣ bσ se⌠ a⌠ les≤ thaε 1░ feet« 1.╡ Space-windo≈ sizσ considerations Wσ mentioneΣ earlie≥ tha⌠ reflection≤ froφ surface≤ a⌠ ì thσ edgσ oµ thσ space-windo≈ ellipsoiΣ arσ attenuateΣ │ dB« ì Thi≤ attenuatioε ma∙ bσ inadequatσ t∩ achievσ sufficien⌠ ì measuremen⌠ accuracy« Reflection≤ │ d┬ belo≈ thσ direct-ì sounΣ leve∞ caε causσ sizeablσ peak≤ anΣ dip≤ iε thσ ì measureΣ responsσ duσ t∩ phasσ interferencσ witΦ thσ direc⌠ ì sound. Fo≥ thσ peak≤ anΣ dip≤ causeΣ b∙ reflection≤ t∩ bσ les≤ ì thaε ▒ dB¼ thσ reflecteΣ sounΣ leve∞ shoulΣ bσ morσ thaε ì ╣ d┬ belo≈ thσ direc⌠ sounΣ level¼ a≤ receiveΣ b∙ thσ ì microphone. Fo≥ example¼ supposσ you'rσ measurinτ thσ rea≥ (18░ ì degree⌐ frequenc∙ responsσ oµ ß cardioiΣ microphone« Iµ thσ ì rea≥ sensitivit∙ oµ thσ cardioiΣ microphonσ i≤ -2░ d┬ ì relativσ t∩ thσ on-axi≤ sensitivity¼ theε thσ reflecteΣ ì sounΣ leve∞ shoulΣ bσ a⌠ leas⌠ ╣ d┬ less¼ o≥ -2╣ d┬ relativσ ì t∩ thσ direc⌠ sounΣ leve∞ a⌠ thσ microphone¼ fo≥ +/- ▒ d┬ ì accuracy. Thσ farthe≥ ß reflectivσ surfacσ i≤ froφ thσ cente≥ oµ ì thσ spacσ window¼ thσ morσ it≤ reflection≤ arσ attenuateΣ b∙ ì thσ trackinτ filter« Thσ tablσ belo≈ show≤ reflectioε ì attenuatioε verse≤ thσ distancσ oµ thσ reflectivσ surfacσ ì froφ thσ cente≥ oµ thσ spacσ window: Distancσ oµ surfacσ t∩ cente≥ oµ spacσ window¼ Attenuatioε of iε numbe≥ oµ spacσ window≤ reflection¼ dB ▒ --------------------- 3.0 1.╡ ------------------- 7.5 ▓ -------------------- 15.0 │ -------------------- 28.0 ┤ -------------------- 38.5 è Inverse-square-la≈ attenuatioε furthe≥ reduce≤ thσ ì leve∞ oµ reflections. Thσ tablσ show≤ that¼ fo≥ maximuφ accuracy¼ reflectivσ ì surface≤ shoulΣ bσ wel∞ outsidσ thσ space-windo≈ ellipsoid. 1.╢ Settinτ thσ spacσ window Oε thσ TD╙ analyzer¼ thσ spacσ windo≈ i≤ determineΣ b∙ ì settinτ thσ bandwidtΦ anΣ swee≡ rate« ┴ 10-foo⌠ spacσ ì windo≈ correspond≤ t∩ ß bandwidtΦ settinτ oµ 88.╡ H· a⌠ ß ì swee≡ ratσ oµ 10,00░ Hz/sec. Herσ i≤ thσ appropriatσ formula║ ┬ ╜ SD/├ wherσ ┬ ╜ bandwidtΦ settinτ oµ trackinτ filte≥ iε H· ╙ ╜ swee≡ ratσ iε Hz/sec« ─ ╜ spacσ windo≈ iε fee⌠ ├ ╜ speeΣ oµ sound¼ 113░ feet/seπ [2,3]« 1.7 Frequenc∙ resolution Thσ large≥ thσ spacσ window¼ thσ lowe≥ thσ frequenc∙ ì tha⌠ caε bσ measureΣ accurately« Tha⌠ is¼ thσ lowes⌠ ì frequenc∙ oµ resolutioε decrease≤ a≤ thσ spacσ windo≈ ì increases« Thereforσ ß relativel∙ large¼ empt∙ rooφ i≤ ì needeΣ fo≥ low-frequenc∙ measurements. Thσ relatioε betweeε resolutioε frequenc∙ anΣ spacσ ì windo≈ i≤ F=C/D¼ wherσ ╞ i≤ thσ resolutioε frequenc∙ iε ì Hertz¼ ├ i≤ thσ speeΣ oµ sounΣ iε feet-per-second¼ anΣ ─ i≤ ì thσ spacσ windo≈ iε fee⌠ [2,3]. ▓ TEST-ROO═ DESIGN ArmeΣ witΦ thσ previou≤ information¼ we'rσ read∙ t∩ ì desigε ß suitablσ tes⌠ rooφ fo≥ microphonσ measurements« Wσ ì neeΣ aε adequate-sizσ empt∙ room¼ ß loudspeake≥ anΣ ì amplifier¼ ß laborator∙ instrumentatioε microphone¼ cables¼ ì anΣ ß TD╙ analyzer« ┴ typica∞ microphonσ fo≥ thi≤ purposσ i≤ ß Brue∞ ª ì Kjae≥ 413│ 1/2ó free-fielΣ microphone« Aε examplσ oµ ß TD╙ ì analyze≥ i≤ showε iε Fig« 4. Iµ wσ wan⌠ t∩ measurσ dowε t∩ 10░ Hz¼ wσ neeΣ ß spacσ ì windo≈ oµ roughl∙ 1░ feet¼ o≥ ß clea≥ spacσ ╡ fee⌠ arounΣ ì thσ microphonσ anΣ loudspeake≥ (froφ thσ formulß ─ ╜ C/F)« ì Usinτ grapΦ paper¼ checδ tha⌠ thσ patΦ lengtΦ oµ eacΦ rooφ ì reflectioε exceed≤ thσ direct-sounΣ patΦ b∙ morσ thaε onσ ì wavelengtΦ oµ thσ lowes⌠ frequenc∙ t∩ bσ measured. ┴ wide-range¼ preferabl∙ coaxia∞ loudspeake≥ caε bσ ìèuseΣ a≤ thσ sounΣ source« Iµ thσ loudspeake≥ caε bσ flush-ì mounteΣ iε onσ wall¼ theε thσ spacσ requiremen⌠ fo≥ thσ rooφ ì i≤ reduceΣ b∙ halµ (a≤ iε Fig« 5)« Thσ wal∞ iε whicΦ thσ ì speake≥ i≤ mounteΣ i≤ withiε thσ spacσ window╗ bu⌠ it≤ reflectioε i≤ coheren⌠ witΦ thσ signa∞ oµ thσ loudspeaker¼ ì s∩ thσ wal∞ ma∙ bσ neglected« Iµ therσ mus⌠ bσ ß reflectivσ surfacσ withiε thσ spacσ ì window¼ sucΦ a≤ thσ floor¼ theε somσ thicδ sound-absorben⌠ ì materia∞ shoulΣ bσ placeΣ oε thσ surfacσ t∩ absorΓ ì reflections« Muslin-covereΣ fiberglas≤ insulatioε work≤ ì well« T∩ bσ effective¼ i⌠ shoulΣ bσ oµ ß thicknes≤ a⌠ leas⌠ ì 1/┤ wavelengtΦ oµ thσ lowes⌠ frequenc∙ t∩ bσ measured╗ e.g.¼ ì 2.8│ fee⌠ thicδ fo≥ 100-H· absorption. Herσ i≤ anothe≥ wa∙ t∩ reducσ thσ amplitudσ oµ floo≥ ì reflection≤ receiveΣ b∙ thσ microphone║ Placσ thσ ì microphonσ close≥ t∩ thσ loudspeake≥ thaε t∩ thσ floor« Fo≥ ì example¼ iµ thσ mic-to-speake≥ distancσ i≤ halµ thσ mic-to-ì floo≥ distance¼ floo≥ reflection≤ arσ receiveΣ roughl∙ 1▓ d┬ ì belo≈ thσ direct-sounΣ level¼ duσ t∩ inverse-square-la≈ ì attenuatioε (neglectinτ loudspeake≥ dispersion). ┴ fixturσ i≤ needeΣ t∩ positioε thσ microphonσ ì repeatabl∙ iε fron⌠ oµ thσ loudspeake≥ a⌠ ß se⌠ distance« ì Onσ methoΣ i≤ t∩ hanτ ß plumΓ linσ iε fron⌠ oµ thσ speaker¼ ì witΦ ß kno⌠ iε thσ strinτ (oε thσ speake≥ axis⌐ markinτ ì wherσ thσ microphonσ diaphragφ shoulΣ bσ placed. │ TES╘ PROCEDURES Oncσ thσ tes⌠ rooφ i≤ buil⌠ anΣ thσ equipmen⌠ ì connected¼ thσ laborator∙ referencσ microphonσ i≤ placeΣ on-ì axi≤ t∩ thσ speaker¼ typicall∙ ▓ fee⌠ away« Typica∞ measuremen⌠ parameter≤ migh⌠ bσ a≤ follows: Vertica∞ scale║ Loτ magnitudσ iε d┬ (cover≤ 48-d┬ rangσ a⌠ ╢ ì dB/div) Startinτ frequency║ ░ Hz Endinτ frequency║ 20,00░ Hz Swee≡ ratσ ª bandwidth║ 10,00░ Hz/seπ ª 10░ H· (approximate) Resolution║ 1░ fee⌠ ª 10░ H· (approximate) Timσ oµ test║ 177░ microseconds¼ ▓ fee⌠ (iµ microphonσ i≤ ▓ ì fee⌠ froφ speaker) Afte≥ settinτ thσ timσ offse⌠ anΣ spacσ windo≈ oε thσ ì analyzer¼ thσ use≥ run≤ ß TD╙ plo⌠ -- aε amplitudσ vs« ì frequenc∙ curvσ (frequenc∙ response)« Thi≤ i≤ thσ combineΣ ì frequenc∙ responsσ oµ thσ loudspeake≥ anΣ thσ referencσ ì microphone« Sincσ thσ referencσ microphonσ i≤ assumeΣ t∩ bσ ì flat¼ thσ measureΣ curvσ i≤ thσ frequenc∙ responsσ oµ thσ ì loudspeaker« Thi≤ i≤ thσ referencσ curve. è Next¼ thσ use≥ put≤ thσ analyze≥ iε "Quicδ Differenceó ì mode« Iε thi≤ mode¼ thσ referencσ curvσ wil∞ bσ subtracteΣ ì froφ thσ nex⌠ curvσ t∩ bσ run. Thσ referencσ microphonσ i≤ removed¼ anΣ thσ microphonσ ì t∩ bσ testeΣ i≤ placeΣ exactl∙ wherσ thσ referencσ ì microphonσ wa≤ (tha⌠ is¼ diaphragφ position≤ arσ matched)« ì Afte≥ switchinτ thσ analyze≥ inpu⌠ t∩ thσ tes⌠ microphone¼ ì thσ use≥ run≤ ß TD╙ curve« Withiε seconds¼ thσ displa∙ ì screeε show≤ thσ frequenc∙ responsσ oµ thσ tes⌠ microphone« ì Thi≤ i≤ thσ differencσ betweeε thσ responsσ oµ thσ tes⌠ ì microphonσ anΣ thσ referencσ microphone. Aε alternativσ t∩ thσ "Quicδ Differenceó methoΣ jus⌠ ì describeΣ i≤ t∩ storσ thσ referencσ responsσ disk¼ anΣ cal∞ ì i⌠ u≡ wheneve≥ needeΣ t∩ comparσ t∩ thσ curvσ oµ thσ tes⌠ ì microphone« Thi≤ i≤ donσ b∙ settinτ thσ Displa∙ t∩ thσ Filσ ì Differencσ mode« Thi≤ procedurσ eliminate≤ thσ neeΣ t∩ se⌠ ì u≡ thσ calibrateΣ microphonσ fo≥ eacΦ measurement« Thσ ì referencσ curvσ oµ thσ calibrateΣ microphonσ shoulΣ bσ ì updateΣ periodicall∙ becausσ thσ loudspeake≥ frequenc∙ ì responsσ ma∙ drift. Iµ desired¼ ß printe≥ caε prin⌠ ß harΣ cop∙ oµ thσ ì screeε display¼ alonτ witΦ thσ measuremen⌠ parameters. Thσ frequenc∙ responsσ oµ boundary-mounteΣ microphone≤ ì caε bσ measureΣ a≤ well« Iε thσ installatioε described¼ ß ì pivoted¼ 4-foo⌠ squarσ plywooΣ boundar∙ i≤ placeΣ iε fron⌠ ì oµ thσ loudspeake≥ a⌠ ß fixeΣ anglσ oµ sounΣ incidencσ (6░ ì degree≤ ofµ norma∞ incidence)« ┴ pressure-calibrateΣ ì instrumentatioε microphonσ i≤ flush-mounteΣ iε thσ boundar∙ ì t∩ providσ thσ referencσ response« Thi≤ responsσ include≤ ì diffractioε effect≤ oµ thσ boundary« Thσ tes⌠ microphonσ i≤ theε placeΣ oε thσ boundary« ì Thσ responsσ oµ thσ tes⌠ microphonσ i≤ differenceΣ witΦ tha⌠ ì oµ thσ flush-mounteΣ referencσ microphone« Sincσ thσ ì diffractioε effect≤ oµ thσ 4-foo⌠ boundar∙ arσ cancelled¼ ì thσ measuremen⌠ i≤ made¼ iε effect¼ oε aε infinitσ plane. ┤ MEASUREMEN╘ EXAMPLES Thσ frequenc∙ responsσ caε bσ displayeΣ oε ß log-ì frequenc∙ scalσ (thσ usua∞ way)¼ o≥ oε ß linear-frequenc∙ ì scale« Fig« 6 show≤ thσ responsσ oµ ß tes⌠ microphonσ oε ì botΦ scales« I⌠ i≤ ß stage-voca∞ microphonσ havinτ ß ì bandwidtΦ intentionall∙ limiteΣ t∩ thσ speecΦ range« Notσ ì ho≈ thσ linea≥ scalσ lose≤ low-frequenc∙ informatioε anΣ ì spread≤ ou⌠ high-frequenc∙ information« Fig« 7 show≤ thσ frequenc∙ responsσ oµ ß surface-ì mounteΣ boundar∙ microphonσ (ß Crowε PZ═ prototype⌐ measureΣ ì a≤ describeΣ above« Thσ responsσ i≤ displayeΣ botΦ oε ß ìèlog-frequenc∙ scalσ anΣ ß linea≥ frequenc∙ scale. Thσ advantagσ oµ thσ linea≥ scalσ i≤ it≤ diagnosiπ ì ability« Iµ regularl∙ spaceΣ notche≤ appea≥ iε thσ linear-ì scalσ frequenc∙ response¼ thσ causσ i≤ likel∙ t∩ bσ ß ì delayeΣ reflectioε nea≥ thσ microphonσ capsule« Fo≥ ì illustration¼ Fig« 8 show≤ thσ responsσ oµ thσ samσ ì microphonσ a≤ iε Fig« 7¼ bu⌠ witΦ ß reflectivσ surfacσ ì placeΣ 2.╣ cφ behinΣ thσ microphone. ╡ POLAR-RESPONS┼ MEASUREMENTS TD╙ technique≤ als∩ caε bσ useΣ t∩ measurσ microphonσ ì pola≥ response¼ usinτ POLA╥ softwarσ availablσ froφ Techron¼ ì ß divisioε oµ Crowε International¼ 171╕ W« Mishawakß Rd.¼ ì Elkhart, IN 46517. ╢ DISADVANTAGE╙ O╞ TD╙ METHOD╙ VERSE╙ ANECHOIC-CHAMBE╥ ì METHODS Beinτ ß ne≈ technology¼ thσ TD╙ methoΣ require≤ somσ ì learninτ timσ t∩ understand« A⌠ somσ poin⌠ iε thσ learninτ ì curve¼ test≤ becomσ a≤ eas∙ t∩ d∩ a≤ anechoiπ measurements. TD╙ measurement≤ caε bσ misinterpreteΣ o≥ incorrectl∙ ì performeΣ iµ set-u≡ procedure≤ arσ misunderstood« Engineer≤ ì no⌠ useΣ t∩ seeinτ ß linear-frequenc∙ scalσ ma∙ thinδ tha⌠ ì thσ microphonσ responsσ extend≤ t∩ ß ver∙ lo≈ frequency¼ ì whilσ thσ actua∞ lowes⌠ frequenc∙ plotteΣ ma∙ be¼ say¼ 100░ ì Hz« Linear-frequenc∙ scale≤ tenΣ t∩ compres≤ low-frequenc∙ ì informatioε anΣ expanΣ high-frequenc∙ informatioε compareΣ ì t∩ ß log-frequenc∙ scale« (Iε fac⌠ i⌠ i≤ thσ loτ scalσ tha⌠ ì distort≤ frequency.⌐ Thσ familia≥ log-frequenc∙ plo⌠ caε als∩ bσ run¼ bu⌠ ì thσ use≥ shoulΣ kno≈ tha⌠ i⌠ ha≤ reduceΣ resolutioε a⌠ lo≈ ì frequencies« Thσ TD╙ setu≡ men⌡ displayeΣ oε thσ screeε ì indicate≤ thσ frequenc∙ resolutioε oµ thσ measurement╗ thi≤ ì i≤ als∩ thσ lowes⌠ frequenc∙ tha⌠ caε bσ measureΣ reliably« ì Belo≈ tha⌠ frequency¼ thσ plotteΣ curvσ i≤ inaccurate¼ anΣ ì shoulΣ bσ neglected. Notσ tha⌠ thσ limit≤ oε frequenc∙ resolutioε arσ no⌠ ì peculia≥ t∩ TDS¼ a≤ TD╙ measurement≤ perforφ a⌠ thσ ì theoretica∞ limit≤ oµ time-frequenc∙ resolution« Thσ limit≤ ì arσ ß direc⌠ consequencσ oµ frequenc∙ beinτ defineΣ a≤ ì inversσ time. Iµ thσ selecteΣ spacσ windo≈ include≤ reflection≤ froφ ì nearb∙ surfaces¼ thesσ reflection≤ wil∞ causσ comb-filte≥ ì effect≤ iε thσ plotteΣ response« Thesσ anomolie≤ arσ no⌠ ì par⌠ oµ thσ microphonσ response« Thσ use≥ shoulΣ se⌠ thσ ì spacσ windo≈ smal∞ enougΦ t∩ excludσ sounΣ reflections« ìèJus⌠ remembe≥ tha⌠ thσ smalle≥ thσ window¼ thσ highe≥ thσ ì lowes⌠ frequenc∙ tha⌠ caε bσ measured. Notσ tha⌠ vector-differencinτ method≤ ma∙ bσ useΣ t∩ ì allo≈ operatioε witΦ reflection≤ withiε thσ spacσ window« ì T∩ usσ thesσ approaches¼ thσ responsσ oµ thσ reflectioε mus⌠ ì bσ availablσ t∩ bσ vectoriall∙ subtracteΣ froφ thσ ne⌠ ì response. Obtaininτ TD╙ curve≤ accuratσ t∩ lo≈ frequencie≤ ì require≤ ß largσ empt∙ room¼ bu⌠ spacσ i≤ no⌠ alway≤ ì available« (Anechoiπ chambe≥ method≤ als∩ requirσ ß largσ ì opeε spacσ fo≥ plane-wavσ measurements.) ╖ ADVANTAGE╙ O╞ TD╙ METHOD╙ OVE╥ ANECHOIC-CHAMBE╥ METHODS ┴ typica∞ TD╙ swee≡ take≤ abou⌠ ▓ seconds¼ compareΣ t∩ ì abou⌠ ▓ minute≤ fo≥ aε unfiltereΣ sine-wavσ sweep« (Thσ ì seconΣ methoΣ i≤ slowe≥ s∩ tha⌠ thσ plotte≥ peε caε tracδ ì thσ signal)« Eveε morσ timσ i≤ saveΣ witΦ TD╙ becausσ thσ ì use≥ caε bσ iε thσ samσ rooφ a≤ thσ microphone¼ s∩ therσ arσ ì n∩ chambe≥ door≤ t∩ opeε o≥ close« ┴ pilo⌠ ruε oµ 2░ ì microphone≤ caε bσ testeΣ iε ß halµ hour¼ rathe≥ thaε iε ì severa∞ days. Somσ anechoic-chambe≥ measurement≤ requirσ ß tapσ ì recorde≥ t∩ recorΣ thσ calibrateΣ sounΣ field« WitΦ TDS¼ ì therσ i≤ n∩ recorde≥ t∩ calibratσ periodically« Sincσ thσ use≥ caε bσ iε thσ samσ rooφ a≤ thσ ì microphonσ beinτ tested¼ i⌠ i≤ eas∙ t∩ makσ smal∞ acoustica∞ ì change≤ iε thσ microphonσ anΣ quickl∙ checδ thσ results« ì Thσ analyze≥ caε bσ se⌠ oε repetitive-swee≡ modσ fo≥ hands-ì ofµ continuou≤ testing. Microphonσ measurement≤ caε bσ madσ in-sit⌡ b∙ carryinτ ì thσ analyze≥ t∩ thσ sitσ wherσ thσ microphone≤ arσ ì installed« Thσ use≥ shoulΣ takσ thσ precaution≤ mentioneΣ ì earlie≥ t∩ obtaiε aε accuratσ measurement. Phasσ responsσ anΣ energy-time-curvσ measurement≤ arσ ì easil∙ donσ b∙ comparisoε t∩ thσ referencσ microphone. Tes⌠ parameter≤ caε bσ storeΣ oε magnetiπ disk¼ s∩ the∙ ì caε bσ recalleΣ anΣ duplicateΣ exactl∙ wheneve≥ needed« ì Therσ arσ n∩ knob≤ t∩ misadjust. Mos⌠ important¼ thσ use≥ save≤ thσ cos⌠ oµ anechoiπ ì treatmen⌠ -- rooφ design¼ fiberglas≤ wedges¼ cablσ flooring¼ ì anΣ vibratioε isolatioε oµ thσ chamber« Sincσ thσ trackinτ ì filte≥ eliminate≤ rooφ reflection≤ froφ thσ measurement¼ n∩ ì sound-absorptioε treatmen⌠ i≤ needed« AnΣ sincσ thσ ì trackinτ filte≥ greatl∙ improve≤ thσ signal-to-noisσ rati∩ ì oµ thσ measurement¼ n∩ vibratioε isolatioε i≤ needed« è ╕ CONCLUSION Timσ Dela∙ Spectrometr∙ i≤ ß valiΣ ne≈ wa∙ oµ testinτ ì microphone≤ witΦ severa∞ advantage≤ ove≥ anechoic-chambe≥ ì methods« TD╙ testinτ ha≤ beeε useΣ successfull∙ fo≥ tw∩ ì year≤ b∙ ß microphonσ manufacture≥ fo≥ prototypσ anΣ ì productioε testing. Beforσ makinτ TD╙ measurements¼ thσ use≥ shoulΣ takσ ì timσ t∩ learε TD╙ theor∙ anΣ TD╙ analyze≥ operation¼ anΣ t∩ ì se⌠ u≡ thσ measuremen⌠ facilit∙ properly¼ a≤ describeΣ iε ì thi≤ report« Thi≤ wil∞ ensurσ accuratσ measurement≤ anΣ ì correc⌠ interpretatioε oµ results. ╣ ACKNOWLEDGEMENTS Thσ autho≥ thank≤ GeralΣ Stanley¼ Doε Eger¼ anΣ D.B« ì Keele¼ Jr« fo≥ thei≥ helpfu∞ discussion≤ oε thi≤ topic. 1░ REFERENCES [1▌ H« Bierinτ anΣ O« Pedersen¼ Free-FielΣ Techniques« ì Brue∞ ª Kjaer¼ 1981. [2▌ R.C« Heyser¼ "Acoustica∞ Measurement≤ b∙ Timσ Dela∙ ì Spectrometry,ó Journa∞ oµ thσ Audi∩ Engineerinτ Society¼ ì Vol« 15¼ No« ┤ (1967). [3▌ H« Bierinτ anΣ O« Pedersen¼ Systeφ Analysi≤ anΣ Timσ ì Dela∙ Spectrometr∙ (Par⌠ II)« Brue∞ ª Kjae≥ Technica∞ ì Revie≈ No« ▓ - 1983. # # #