Research evaluates role of density in acoustical insulation performance formefunction ” Preventing noise intrusion from speech and music is simplified with optimal density of drywall partition cavity insulation. ‘by. Staniey Rater, Ascectursl Constuctin Manager, Acoust ted Sates Gypaur> Compay ny factors iniluence the acoustical performance of gypsum board partition systems. Air cavities, resilient members, sealant details, and various types of acoustical insulation are same of the variables that have a significant elfect. USG has conducted extensive research into one of these important factors — acoustical insulation density—to determine the optimum density for THERNArIiR Sound Anenuation Fite Blankers, The Tiegssaiae® Insulation in densities rom 2 Ib ‘eu. t0 12 Nveust. and thicknesses irom 110 12 in, TreRsearsce Insulation is not currently manutactured in densities less than ‘earch included evaluations of mineraltiber es used about 2.0 Ib cuit, 30 the evaluation for lower er slass-fiber material at 0.7 and 0.8 fb ‘ew. Tess were Coneducted with variety OF partition dsigns, The best overall acoustic petormance was achieved with anlineralfiber blanket in the 2.5 19 3.0 Ib-cu.it, density ange. Densities above 3h cust, provide more sound attenuation at high frequencies, but lows-ire periormance starts to suiier at higher densities, Mid and hngh-ieequency performance yes down siewicaatly as insulation density tals below 23 Ts.eu The conclusive to nignuiacture standaed Tee arnt Sourd Attenuation Fire Blankets s$AF6 wth a nominal density (2.5 th.cust. sas based on thy acoustic research, Cost.to beni ates, fee periormance gn handling chaeacterstes Dryas partitions sith Titesiaris SAFB's inthe sad cavities consistently outperfou those that include ‘mineral-fiber insulation with densities Jess chan 2.5 Ib cust. In fact, tests conducted at Riverbank Acoustical Laboratories showed that partitions with 2.5 Ib cu.ft, THeasaracR SAFB's can provide STC ratings up to 4 points higher than those ia which low-density glass fiber 1/2 to 1 in, thicker is used. The acoustical benefits of cavity insulation as well as the effects of insulation density vary with frequency, the basic partition design and the insulation thickness. The effect of insulation density on acoustical performance must be determined for each partition design Partitions for sound isolation Three primary sound source les typically occur in buildings (1) speech or spevchlike, (21 music, and (3) machinery or rmchanical equipment, With any of these sound sources, the degree of privacy that will be achieved with a given pactition ean be evaluated simply. One adds the sound pressure level of the background noise in the space subject to intrusive sound and the sound atlenuation provided by the separating structure and then compares this su ga che sound pressure level of the ‘source. This technique is used i the following sections 10 show the importance of insulation density when isolating spoech and then music. To simpy the discussion below its assumed that no ilgaking paths are involved and thatthe Laboratory sound transmission loss the sound attenuation thats actually achieved! teantmursd on nex pase)RESEARCH EVALUATES INSULATION DENSITY ROLEcontinued Density and isolating speech ‘The critical frequencies requiring isolation when speech is the sound source are from 250 Hz (approximately middle "C") through 4.000 He (see Fig. 1 on next page). The dot field in Fig 1 isa graphic representation of male speech (cased voice level ina 100-59. enclosed office—approx. 68 dB}. The dot fields, used for graphic calculation of intelligibility by the ‘Aniculation Index (Al method, Speech at any given loudness level allsin a 30 dB loudness envelope (vertical scale since diferent words and letters are produced with diferent energy levels. The number of dats on each vertical Hine gives a pictorial view of the relative contribution ofthat 1-actave band on speech ineligibility. The most eneray loudest isin the 250 to 1,000 Hz frequency range. Nole, however, the Table. Sound isolation advaetage ia decibels 2B} of 2.5 10 3 0b seu.t, THERMAFIBER Sound Atenuation fire Blankets (SAFB) over 0.7 00.8 eu. glossliberasulaion (CFI in gypsum board (GBI p3etiions denseness ofthe dot array above 1,000 Hz. These are the frequencies that give clarity to speech, so hearing only a small portion of the speech energy in these frequencies can result in loss of speech privacy. Deywall partitions, especially single- layer systems, tend to be weak sound isolators inthe 2,000 ta 4,000 Hz frequency range. due tothe thickness of the gypsum board and its natural physical characteristics. This weakness zone can be seen in Fig. 1 The A and B curves ave the sum of the RC-20 noise level assumed for his example and the sound transmission loss of two diferent parition designs employed here. The A curve is fora partion consisting of 358ST studs 24 in-o.c. with a single layer of 5/8-in. Sreeraock Brand Feecave "C” Gypsum Panels on each side. The B curve partition isthe same except Partition design | ‘Octave band center frequencies in Hz* sTc Tsfaton deny onyvaableimexchiesd | 928250 aD Egg Stan 4000 | improvement ssn « | «s | « |=] | ors oo] fe | 8 fo] a | me fg Ge -sn 2 Nem ~oe | fo | o | oe | a 5 Ste GS 3 eon cos fos | ae | oe | a3 | 20 Stn | Stein | 1 em o | foo | o foo | a > | GF 3.in, | c | Gite | T T a 5 SN 3m o |. | 2 foo | ow | Gres Goan oa zon vo | ow [ow | | oo | ow Je Grete | Gee t + sadn i. mf as to A re | Gos acm | |3+in., 2.5 Ib Just, THeRarigeR SAFB'S have been added to the cavity. The measure of intelligibility (Al of the partion ‘without the SAFB's can be determined simply by counting the ts above the A curve. There ate about 48 dots above this curve which translates to approximately 70% sentence intelligibility for chese conditions, There are about 15 dots above the B curve which translates ta approximately 10% sentence intelligibility. In order fo the speech tobe totally inaudible, the A or 8 curve would have ta be 5to 10.d8 above alle dots Adding Teese SAFB's improves performance about 8 {dB across the frequency spectrum for this particular design The improvement with ther designs can be more or less, The raph illustrates the need to maximize partition performance inthe 250-4,000 Hz frequencies when isolating speech. Table 1 shows the improvement in performance attributable tw the density ofthe cavity insulation, THERMAR BCR SAFB’S consistently provide higher sound isolatios than low-density lass fiber over the entre 250-4,000 Hz frequency range. with 10 5 dB improvementin the critical 2,000 Hz octave being, ily typical, ‘A 310 5 dB improvement appears to be small until itis ‘apared withthe improvements that ypically can be ceved in the 500 to 2,500 Hz range by other methods. For “example, air space depth and cavity insulation thickness must bbe doubled or the numberof layers of gypsum board increased bby 50% to achieve 2 106 d8 impavemeats in the 500 10 2,500 Hz irequency range. Itshould be noted, however, that increasing cavity depth, insulation thickness and the number cf gypsum board layers can provide much greater impzovement than this below 250 Hz so these techniques play a critical role in the design of high-performance partitions One of the advantages ofthe improvement due i insulation density is that its adie fo the improvements that can be achieved with avher methods, Density and isolating music Music contains substantial energy in the whole audible frequency range and is generally much louder than speech, The top curve Fig, 11s typical design spectrum for 90 48 ‘orchestea music. Most people would judge this as “love! forchestra music, The graph shows thatthe privacy achieved with a single Laver partition with THExsariete SAFB’S & curve) need not be increased very much in the 50010 1,000 He frequencies 0 start isolating the music source, but it must be impeovexd significantly above 1,000 Hz and below 250 Mz. The greatest reves ate in the 63 He and 2,000 Hz octave bands, Creating, the equovalent of nea separate walls, and increasing the umber at insets of gypsum board, the cavity depth and the slatwon thickness are some techniques that caa be used t0 rove acaustic performance, However, above 1,000 Hz. a Pertormance plateau is quickly reached that is dificult to ‘exceed with simple techniques. This is where the density of the Fig. nolan Speech and Music, Dot field reverent 68.68 speech J Asclation Index method o determining speech privacy. Kolotion "A" represents sum of C.20 backround noise and sound eansmssion loss provided by partion consisting of 3-S8in. meal Su, 2410. 0¢,. with single-layer Sitin, SHEETROCK Brand Cypsum Panes each side, bolation “81s for same partion except with in. THERMAFIBER Sound Aengation Fire Blankets in stud cavity "ELDAR pena , th i . au 2 HEHE TPE . JEERNT AUTEN AH] Hh: f { beid “TT TTT 7 rAd i LULLED, ONT, BF, TOO 449 73. 0 610 1000 ToD 2500 400 4300 31 3's0\ g0\129- 9) 315. 50 800 1250000 31305000 8000 ‘Soesthied octave band center fequenciesin Mz (cs) Insulation is one ofthe keys to success in isolating music with yPsum panitions. One pacition tested at Riverbank Acoustical Laboratories consisted of 605120 USG Studs, wo layers of 5/8in, SHEcrROCK Brand FktecoDe "C” Gypsum Panels on one side, RC-1 Resilient Channels and 2 layers of the same type panels on the other side and 5-in. SAFB'S inthe cavity. This high-performance system provided 2-48 more isolation inthe 250 and 500 Hz octaves and almost 6 d8 more isolation above 1,600 He than when 6rin, low-density fiber was used. The optimum density forcavity insulation in the 31 t0 125 Hz frequencies appears to be in the 0.7 103.0 Ibvcust, range, Ivthin this feequency zone, insulation at 0.7 Ib.icu:i. may somtimes be 1 or 2.JB better in certain 1/3 octaves while a 3 Ibcu.f insulation may be 1 or 2 dB better at other 1/3 octaves. This varies with insulation thickness and the partition design Comparative listening tests on the receiving side of test partitions emphasized the importance of isolating the upper frequencies when the source is music. I high-frequency ‘solation is substantially out of balance with mid-frequency isolation, the mid-feequency harmonics can be heard, along with the high-frequeney fundamentals, a8 a unnatural clephone quality, highly intelligible sound. However, ithe ‘vant om est 3807RESEARCH EVALUATES INSULATION DENSITY ROLEIcontinued partition can attenuate the sound so thatthe intrusive music sounds natural and unfiltered (similar to a stereo played at low volume), the intrusion is not nearly as objectionable, This result ean be created by a partition witha sound transmission Foss curve tat, starting at about 31 H2, rises at a nominal rate fof about 12 dB per octave to 400 or 500 He, gradually flattens out in the next octave and then remains elatively flat above 1,000 Hz, Except for special facilities such as recording, studios and music buildings, itis seldom economically feasible to totally isolate loud music, especially when the background noise level is very low in the adjacent spaces. Therefore, a parition with these general characteristics isthe best choice when isolating music. Insulation density is an important resource to use in achieving this goal The sound isolation of unamplified speech is discussed in the previous section. However, the lation requirements for ‘can approach those of music. The low: frequency component of speech can be boosted significantly through amplifcation. This low-frequency component does ‘not contribute te speech intelligibility, but it can be heard through a narttinn that provides litle isolation in the low frequencies. A partition designed to isolate music is also an excellent design for isolating amplified speech, Density and electrical outlets ‘Another sound: solation problem isthe transmission of sound through electrical outlets on opposite sides of a partition High-frequency sound is easly ransmitted through small ‘openings inthe outlet boxes themselves ori joints between the boxes and the gypsum board. The joint between the outlet boxes and the gypsum board should always be seated with an appropriate caulking material, such as USG Acoustical Sealant In high-perlarmance Sound isolating partitions itis Usually necessary o seal the back of outlet boxes as well, It takes vigilant inspection during partition construction to assure that the back sides of all outlet boxes are well sealed, and it is difficult 0 verify or correct ater the partition is buil, The problem created by inadequate seating can be minimized by offsetting the outlets and using the cavity insulation 952 sound attenuation medium. (Underwriters Laboratories, inc. requires a minimum 2 et 10 maintain the fie cating.) The greater the ost, the farther sound must, travel through the cavity insulation in order for ito be ‘eansmitted through inadequately sealed outlet boxes, Fig, 2 shows that sound tansmission loss theough 3 poraus material increases significantly as density increases. Low- density glass ber must be nearly twice as thick gs the standard Trckwasibice SAFB's to provide the same attenuation (curves C and 0. Fig. 2), The way to handles the acoustics of ‘outlets should always be decided by the project acoustician since so many variables are volved Summary Drywall parution systems with cavity insulation in the 2.5 10 Fras. 2 Pantin USA, Fig.2, {Blect of acoustics insulation thickness and density on sound tanemition los A 0 i o wie} +4 < © é Too 350. WW 6101000. “Toon 25003000 135 200-415-500" aU 1250" 2000 $150.50 (One-third octave baad center frequencies in He Curve A5.5-n. 4. but, THERMAFIBER SAFB special density) j *RALTLBS-S1 Curve Btn, 2.8 1b.u, THERMAFIBER SAFB standard deny) RALTUBA I Curve boi, Jeu, Glas Fiber RAL TUBS Curve D Son, 288 cui, THERMAFIBER SAFB standard densivy RACTUBS IO Curve ES: 7 Iaveuit Gloss Fiber RALTUBS.29| *RAL = Riverbank Acoustica Laboraares, TL = Tansmssin Loss Test Conductedin 1H, 3 = Text Nome 4.0 lb Jcu.t. density range consistently outperform those with lower density glass-liber insulation, The 2.5 IbJeu.ft. insulation provides better results when isolating ether speech or music and provides a distinct advantage in minimizing the negative tetiects ofthe sound leaks associated with electrical outlets. WTheRarioce SAFB's (standard 2.5 Ib /ou.t. density! are substituted in a partition tested with the commonly-used 0.7 10 0.81b.cuA.glasssber insulation, the sound transmission loss will always be superior oral least equal tothe tested pastition. The Texsuaiice SAFB's have the added advantage of a higher melt-point thaa glass fiber, so they offer superior fie endurance. Consequently it cannot be assumed that iow density glass fiber can be substituted for THHeks aise SAFB's in arated partition without significantly reducing the acoustic or fire performance. o C UNITED STATES GYPSUM COMPANY ~ BUILDING AMERICA CCopyng 195, Unt Sate Grr Company TERMI AFI, SICTTROCK,PIRECODE RC-T an CSG ae Unto States Gaps Compu, Int sth vher Dower age He