“SECRETARY OF LABOR,Complainant,v.STEWART-WARNER CORPORATION,Respondent.OSHRC Docket No. 80-5966_ORDER_The Parties’ stipulation and settlement agreement is approved. Thisorder is issued pursuant to a delegation of authority to the ExecutiveSecretary. 41 Fed. Reg. 37173 (1976), amended at 44 Fed. Reg. 7255 (1979).FOR THE COMMISSIONRay H. Darling, Jr.Executive SecretaryDated: November 18, 1987————————————————————————SECRETARY OF LABOR,Complainant,v.STEWART-WARNER CORPORATION,a Corporation,Respondent.OSHRC Docket No. 80-5966_DECISION AND ORDER_Appearance:FRANCIS X. LILLY, Esq., Solicitor of Labor,JOHN SECARAS, Regional Solicitor,CHRISTOPHER D. WOLFE, Esq., andJAMES L. BOWERS, Esq., for Raymond J. Donovan, Secretaryof Labor, U.S. Department of Labor, Complainant.Columbus R. Gangoni, Jr., Esq., and Gerald C. Peterson, Esq.,David B. Lowe, Esq., Winston & Strawn, Chicago, Illinois,Attorney for Stewart-Warner Corporation, Respondent.Mark B. Bigelow, Esq., Chicago, Illinois, Attorney forUnited Workers Association, U.E.: Local 1154, Authorized EmployeeRepresentative.BOBRICK, JudgeThis proceeding was commenced pursuant to Section 10(c) of theOccupational Safety and Health Act of 1970, 29 U.S.C. 651, et seq.,(hereinafter referred to as the \”Act\”), wherein Respondent,Stewart-Warner Corporation, contested a Citation issued by Complainant,Raymond J. Donovan, Secretary of Labor, U.S. Department of Labor,hereinafter the \”Secretary.\”[[1\/]]The Citation charged Respondent with a repeat violation of the 1980Occupational Safety and Health Regulations 29 C.F.R. 1910.95(b)(1),entitled Occupational Noise Exposure [[2\/]] (hereinafter the \”NoiseStandard\”), in Departments 2, 6, and 75 of its Chicago plant._FACTS OF CASE_On July 27, 1980, in response to a complaint made by an employee ofRespondent alleging exposure to excessive noise levels, Complainant,through one of its compliance officers from the Occupational Safety andHealth Administration (hereinafter \”compliance officer\”), began aninspection of Respondent’s facility in Chicago. Such safety and healthinspection ended August 7, 1980.Respondent, at the inspected work site, manufactured or fabricated,among other things, a broad variety of alemite (lubrication) systems,instruments, and related fittings such as automobile hood latch devices,automobile door handles, large oil mist lubricating systems, airless andconventional spray equipment, fluid dispensing pumps, electrical pumps,miscellaneous lubrication fittings, and various speed pressure andtachometer, meters and gauges. (Tr. 2604-2607, Ex. C-25).The products manufactured by Respondent are precision made and involveexacting tolerances. The manufacturing work requires a high degree oftechnology, a wide variety of engineering expertise and highly skilledmachining operations. The inspection concluded by the Secretary in July1980, was not the first inspection of Respondent’s workplace whereinexcessive noise levels were found to exist. In 1975 and again in 1977,safety and health inspections had revealed excessive noise levels towhich employees were exposed. The 1977 inspection resulted in theissuance of a Serious Citation which alleged violations of 29 C.F.R. ?1910.95(b)(1), in Departments 1, 2, 6 and 75 of Respondent plant (Ex.C-20); the Citation was not contested and became a final order of theCommission. From time to time requests for modification of abatementwere filed; on March 20, 1979 a hearing was held as to abatement datechanges, at which time a final abatement date of August 25, 1979, wasagreed to by the parties (Ex. C-19, 20, 21).As a result of the current 1980 inspection, a Repeat Citation was issuedto Respondent. This Repeat Citation alleged the existence of excessivenoise levels, in violation of the Noise Standard, in Departments 2, 6.and 75. This Citation was contested and is the subject matter of theinstant case.Looking to the individually cited areas we first find Department 2.Department 2 is located on the first floor of Respondent’s facility. Itis a large open area having exposed concrete walls, floor and ceiling;approximately 71 punch presses are located in this space. Of the 71punch presses, approximately 41 are referred to as \”secondary presses\”and 21 are \”automatic presses.\” An automatic press is one wherein theoperator loads raw materials into a feeding mechanism and parts are thenautomatically produced with little employee interaction. The operatormerely loads the press with the raw materials and monitors the partsbeing made to insure that they are being correctly made and properlyexpelled from the machine. A secondary press is one in which theoperator loads an individual piece of raw material or partially producedpart into the press, activates the press, and then monitors that thepress properly pressed and expelled the part from its dies (Tr. 2616,2619). Approximately 115 employees are employed in Department 2 (Ex.C-24). The presses with their component high air pressure systems arethe main source of the noise levels in this department.Department 6 is located on the fourth floor of Respondent’s facility andoccupies approximately 1,200 square feet of space having exposedconcrete walls, floor, and ceiling. Within Department 6 are located 35screw machines of which there are three basic types. A screw machine isa machine in which round metal bars (stock) are fed through rotatingdies which then fabricate small machined components. These screwmachines are referred to as multi-spindled machines wherein thefabrication of parts is done simultaneously on several of the metal barstock inserted into the feed portion of the machine. The operation ofthe screw machines represents the basic source of the noise levels inDepartment 6 (Ex. C-24). Approximately 42 employees are employed inDepartment 6 (Ex. C-24).Department 75 occupies approximately 12,000 square feet of floor spaceon the third floor of the facility, such area also having exposedconcrete walls, floor, and ceiling. Located within this department are74 multispindled screw machines, benches and grinders and otherequipment. The screw machines, known as Davenports, fabricate smallmachined components on multi-spindled dies from a number of bars ofmetal being fed into one end of the machine. Approximately 67 employeesare assigned to Department 75 (Tr. 327, Exs. C-13, 24). The screwmachines appear to be the basic source of the noise levels in this workarea.During relevant times, the Respondent had administered audiometricexaminations to its employees. These tests were analyzed by aconsultant, engaged by Respondent, with written codes being assigned tothe tested employee to indicate whether the individual should bemedically referred to an ear specialist. The audiometric exams togetherwith the evaluations were routinely stored in the company safe whilenever being provided to any of Respondent’s employees. No actions orreferrals were ever made to doctors based upon the results of theexaminations even though the analysis done of these exams recommendedthat a significant number of employees be seen by a medical ear doctor.(Tr. 76, 77, 78, 96, 97, 2090, 2209, 2211, 2229, 4309, 4310, 4311).Before discussing the legal and factual issues, the arguments concerningthe application or enforcement of the Noise Standard, and the economic,and technological feasibility of engineering noise controls proposed forthose cited areas where employees were allegedly exposed to excessivenoise levels, it will be necessary to first review several proceduralmatters raised by Respondent. It will also be necessary to dispose ofcertain factual challenges to the sampling and measuring made by theSecretary’s compliance officer to record the noise levels in theworkplace, and to review certain challenged procedures followed by thecompliance officer in taking her noise level readings._FINDINGS OF FACT AND CONCLUSIONS OF LAW __I. CHALLENGE TO COMMISSION JURISDICTION_Respondent argues that the Commission lacks jurisdiction over thismatter in that it perceives the Secretary’s action to be a continuationof the former Citation proceeding with its successive Petitions forModification of Abatement proceedings; and that the Secretary’s actionsherein should not be the subject matter of a new citation (Ex. C-20,21). Respondent maintains that a settlement agreement had been reachedbetween the parties and, under such settlement agreement, Respondent haddetermined that further controls were not feasible in accordance withsaid agreement. Respondent thus reaches the conclusion that this action,in effect, should be an enforcement action of the terms of thesettlement agreement and not a newly instituted citation proceeding.Respondent extrapolates on this position arguing that the settlementagreement had not become a final order of the Commission and that theagreement did not specifically grant the Commission jurisdiction toenforce its terms. To bolster its argument, Respondent points out thatthe past Citation, together with its Petition for Modification ofAbatement was made part of these proceedings by their introduction intoevidence. Accordingly, goes Respondent’s agreement, the current actionis an alleged enforcement action of the terms of the settlementagreement, the Commission having no jurisdiction over any matterconnected with the alleged 1977 violations since the settlementagreement disposed of same.While Respondent’s argument had never before been raised in thisproceeding, except in its post-trial brief, it could thus be considereduntimely. We nonetheless, review same, finding Respondent’s position tobe without any factual or legal merit.The scenario portrayed by Respondent in this argument does not fit thefacts of this case and in any event hardly makes any sense at all. Inthe instant matter, we find that the Complainant, on the basis of aninspection in July and August of 1980, issued a Repeat Citation foralleged excessive noise levels to which Respondent’s employees wereexposed in Departments 2, 6, and 75. While Respondent urges that theproof of its argument is demonstrated by the fact that the 1977Citation, and the Petition for Modification of Abatement proceedingswere brought into this matter to enforce their terms, we find this notto be the case at all. This matter deals with the Secretary presentlylitigating the propriety of the Citations issued to Respondent onSeptember 12, 1980 (Ex. C-74). The prior Citations, referred to byRespondent as the central core of its argument, were introduced intothese proceedings for the sole purpose of showing knowledge on behalf ofthe Respondent, feasibility of abatement, and proper classification ofthe instant violation as repeat. The Repeat Citation, involved in thiscase, by its terms, lists violations occurring on July 24, 1980, and thecontest filed by Respondent vested this Commission with jurisdiction._II. THE CITATION WAS NOT ISSUED WITHIN 180 DAYS OF THE ALLEGED VIOLATION._Respondent submits that under Section 9(c) of the Act, and what itreferred to as its implementing Regulation, 29 C.F.R. ? 1903.14, acitation to be valid and enforceable, must be issued with reasonablepromptness, and in any event not more than 180 days after the date ofthe alleged violation. Additionally, Respondent tells us that for acitation to be valid it must be in writing, describing the nature of thealleged violation, _and_ _must_ _state_ the following:\”that the issuance of a citation does not constitute a finding that aviolation of the Act has occurred unless there is a failure to contestas provided for in the Act or, if contested, unless the citation isaffirmed by the Review Commission.\” 29 C.F.R. ? 1903.14(e).Respondent argues that since the above cited clause was not in theinstant Citation, and that a citation containing such a clause was neverissued within 180 days of the alleged violation, the Citation isimproper and must be dismissed. In support of this argument, Respondentcites several cases holding that regulations duly adopted by an agencyhave force and effect in law and that an agency is bound by its ownregulations. Cf., _Union of Concerned Scientists v. AEC_, 499 F.2d 1069,1082 (D.C. Cir. 1974); _Service v. Dulles_, 364 U.S. 363, 388 (1957);_City of Santa Clara v. Andrus_, 572 F.2d 660, 669 n. 5 (9th Cir.),_cert._ _denied_, 99 S. Ct. 176 (1978); _Husky Oil Co. v. Department ofEnergy_, 582 F.2d 644 (TECA. 1978).We believe Respondent’s argument fails for two reasons. First, thegrounds alleged by Respondent for avoiding the Citation are not includedin the Act. Second, after reviewing the face of the Citation we findthat the Secretary, in issuing it, had done so with substantialcompliance with the statute and regulations, and as such, no prejudiceto the company resulted by such an omission in the Citation as chargedby Respondent. _Chicago Bridge and Iron v. Occupational Safety andHealth Review Commission_, 535 F.2d 371 (7th Cir. 1976)._III. THE \”CITATION\” WAS IMPROPER_Respondent argues that the issuance of a Citation classified as \”Repeat\”deprived it of its due process rights because the Citation wasmischaracterized as repeat, and that by reason of this, it was unable tointroduce evidence that the original Citation (1977) was invalid.Respondent further argues that in order to establish a \”repeat\”violation, the Secretary was required to show that abatement of theprior alleged violation had been accomplished, citing _Braswell MotorFreight Lines, Inc._, 5 BNA OSHC 1469, 1977. Respondent argues thatsince there was no evidence that it had \”abated\” the alleged violationin Departments 2, 5 and 75, the Citation must be dismissed.Respondent’s argument misses the mark. The instant case deals with theissuance of a Repeat Citation. A Repeat Citation has been denied both bythe Review Commission and various U.S. Courts of Appeals as a violationoccurring when there was a Commission final order against the sameemployer for a substantially similar violation. _Potlatch Corp._, 7 BNAOSHC 1061, 1063, 1979 CCH OSHD 23,294; _George Hyman Construction Co. v.OSHRC_, 582 F.2d 834 (4th Cir. 1978); _Todd Shipyards Corp. v. Secretaryof Labor and OSHRC_, 566 F.2d 1327 (9 Cir. 1981); _Kent NowlinConstruction Company, Inc. v. OSHRC_, 648 F.2nd 1278 (10 Cir. 1981);_Bunge Corporation v. Secretary of Labor and OSHRC_, 638 F.2nd 831 (5thCir. 1981). There is no case law or statutary provision predicating theexistence of a repeat violation on an employers abatement efforts.In the instant case there can be little doubt that the Secretary has metthe requirements for issuance of a Repeat Citation. In 1977 Respondentwas issued, at its same workplace, a Citation for the same violation asthat found in the present case (Ex. C-20). That 1977 Citation became afinal order of the Commission when Respondent chose not to contest.Accordingly, we find the instant Repeat Citation a valid and fullyenforceable citation. The particular facts surrounding the violation in1977 was found not relevant to the instant violation._IV. LEGALITY OF THE MEASUREMENT OF NOISE LEVELSAND EXPOSURE TO ALLEGED EXCESSIVE NOISE LEVELS_The Secretary maintains that it’s compliance officer had correctlymeasured sound levels existing at Respondent’s plant and had alsocorrectly measured levels of exposure of Respondent’s employees to suchnoise levels. The Secretary further maintains that these measurementsshow that employees in Departments 2, 6, and 75 were exposed to noiselevels in excess of the permissible levels set forth in the NoiseStandard. The noise levels were measured by the compliance officerthrough use of sound level meters, which measured the noise levels at aparticular instant in time, and through what is known as \”dosimeters,\” adevice used to measure accumulated exposure to sound levels over aperiod of time.Respondent on the other hand argues that there is art ambiguity in theNoise Standard and that this, coupled with Complainants method ofdetermining existing sound levels, resulted in error requiring thevacation of the Citation. Respondent also argues that the noise readingswere unreliable and erroneously measured. This last argument will bediscussed separately._A. Measuring Mixed Sound Levels_Respondent argues that Table G-16 of the Noise Standard (See n. 2) failsto give adequate notice that mixed sound levels can be used to determineemployee exposure. Respondent maintains that the method of determiningexisting sound levels through use of a dosimeter, which measuresaccumulated sounds from different source of varying levels over a giventime period, under the language of the Noise Standard, isconstitutionally infirm since the Noise Standard does not require anemployer to account for accumulated mixed levels of noise exposure.Next, Respondent argues the inclusion of impact noise levels existing atthe plant site was unconstitutionally included in Complainant’s findingsof employee noise exposure since the Noise Standard, as interpreted byRespondent, does not provide for the inclusion of impact noise inmeasuring noise exposure.With respect to Respondent’s first argument, that the noise regulationis not clear and concise in its language, not giving fair warning of theprohibitive conduct addressed by the standard, i.e., dealing with\”accumulative noise, levels,\” we find this argument without substance orsupport. The claim of ambiguity perceived by Respondent rests upon whatit characterizes as a failure within the Standard to give adequatenotice that \”accumulative mixed sound levels\” could be used to determineemployee exposure. In support of this argument Respondent points tofootnote 1 [[3\/]] of the Standard which in applicable part reads:\”When the daily noise exposure is composed of two or more periods ofnoise exposure of different levels, their combined effect _should_ beconsidered, rather than the individual effect of each.\”The constitutionally offending enforcement of this Standard, asperceived by Respondent, develops from the inclusion of the term\”should\” in footnote 1. which Respondent argues, renders the entireStandard advisory or unenforceable. Respondent cites _Pittsburgh-DesMoines Steel Coal_, CCH 1977-78 OSHD 21,807 (R.C. 1977), _aff’d subnom_; _Marshall v. Pittsburgh-Des Moines Steel Coal_, 584 F.2d 638 (3rdCir. 1978).We find Respondent’s interpretation of the standard ill founded sincethe standard is unambiguous on its face in establishing that employeeexposure to certain sound levels, for certain prescribed period of time,is actionable requiring abatement efforts by the employer. The standardclearly deals with total noise to which employees are exposed, and thisis quite evident in the Standard which speaks of \”sound (levels)exceeding those listed in Table G-16.\”The absence of any qualifiers to the terms \”sound levels\” and \”noiseexposure\” clearly and practically implies that noise levels means totalnoise levels to which employees are exposed regardless of intensity,source or modulation. This is sensible since it would be impossible todraft a regulation which would address each one of a myriad of differentconceivable sources and levels of noise. Secondly, having a standardwhich deals with noise levels predicated on intensity, rather thanfrequency, modulation or source, is a more recognizable standard to dealwith in determining whether remedial action should be taken or not sinceit deals with one factor, that being total noise level exposure.Respondent’s reliance on the existence of some sort of constitutionalquagmire rooted in footnote 1 to the Standard is misplaced since thefootnote is advisory in nature; furthermore it is overshadowed by theobjectively clear mandate found in the basic text of the standard,revealing to any reasonable and practical person, the potential threatto ear injury and the need for employee protection. _American CanCompany_, 10 BNA OSHC 1305, 1982 CCH OSHD 25,899; _Ryder Truck Lines,Inc. v. Peter J. Brennan_, 497 F.2d 230 (5th Cir. 1974). We further seethat the application of the Standard to the facts of the case, i.e.,high noise levels, employee exposure, noise induced hearing loss andtechnical methods of abatement, makes this remedial regulation a mostsimple and unambiguous one, finding easy application and interpretationto the potential hazards involved. _Diebold, Inc. v. OSHRC_, 585 F.2d1327 (6th Cir. 1978); _Vance Construction v. Donovan_, 723 F.2d 410 (5thCir. 1984). While the cited Circuit Court decisions are a sufficientbasis alone to reject Respondent’s argument, we also find that theCommission has consistently and repeatedly held that the Noise Standardis not unforceably vague but is a fully enforceable standard. _SherwinWilliams_, 11 BNA OSHC 2105, CCH 1984 OSHD 26,986; _Sun Ship, Inc._, 11BNA OSHC 1028, 1979 CCH OSHD 26,353; _American Can_, _supra_._B. Inclusion of Impact Noise_Respondent argues that fair warning is not incorporated in the Standardalerting employers that impact and impulse noise will be included intotal noise readings. Respondent points to the last sentence of footnote1 which provides \”Exposure to impulse or impact noise should not exceed140 dB peak sound pressure level,\” to establish the proposition thatinclusion of impact and impulse noise in a sound level meter reading isunconstitutional. Respondent argues that, since the footnote indicatesthat impact noise should not exceed 140 dB, it thus should not beincluded in a scale measurement of existing noise in the workplace.I fail to find any logic to this argument for several reasons. First.remedial legislation and regulation, such as the Noise Standard, shouldbe read in the manner applicable to its intended purpose and notantagonistic towards the conduct to which it is addressed. Secondly, theproposition forwarded by Respondent does not make sense since to a largeextent noise sources are impact and impulse in nature and represent thebulk of noise produced by machinery, as is the case at Respondent’s ownwork site (Ex. C-23, Fig. 2, 3, and 13). If one were to followRespondent’s argument, most all excessive noise levels would not fallwithin the occupational noise exposure standard. Lastly, Respondent’sreading of footnote 1 of the Noise Standard, does not square with theplain reading of the provision itself and this presents sufficient basisalone to reject its argument. Accordingly, we find no affirmity in theStandard precluding its enforcement. _V. ACCURACY AND RELIABILITY OF __SOUND LEVEL MEASUREMENTS MADE BY THE SECRETARY_As a major part of the Secretary’s _prima_ _facie_ case, it is necessarythat he demonstrate that employees were exposed to sound levels inexcess of those specified in Table G-16 [[4\/]] of the Noise Standard. Inorder to meet this burden, sound level readings of the work area anddosimeter readings of employee exposure were taken and recorded by theSecretary’s Industrial Hygienist\/compliance officer. The noisemeasurements were then entered into this record for the purpose ofestablishing noise level exposure of Respondent’s employees in the citedareas.Respondent raises certain procedural and evidentiary challenges to theaccuracy, reliability and application of those sound leveldeterminations made by the Secretary. Also Respondent challenges theNoise Standard’s application, _vis-a-vis_, the Secretary’s sound levelmeasurements and methods by where they were taken.First, Respondent, in defense to the Citation, argues that theSecretary’s method of measuring noise level exposure, used to reach hisconclusions that there existed excessive noise exposure of employees,deprived Respondent of due process since there exists an absence of anyclear language in the Noise Standard that give an employer fair warningof how the sound levels are to be determined or measured. Respondentmore specifically argues that the Noise Standard’s language withinFootnote 1 of the Standard, to wit,\”when the daily noise exposure is composed of two or more periods ofnoise exposure of different levels, their combined effect _should_ beconsidered rather than the individual effect of each,\”makes the collection of different noise levels optional.Respondent points to the use of the word \”should\” found in Footnote 1,arguing that this makes the provision advisory in nature rather thanmandatory, and accordingly is unenforceable by its own terms. Respondentrelies upon the case of _Pittsburgh-Des Moines Steel Co._, 5 BNA OSHC1420, 1977-1978 CCH OSHD ? 21,804 (No. 13708, 1977), _aff’d_ _sub nom_,_Marshall v. Pittsburgh-Des Moines Steel Coal_, 584 F.2d 638 (3rd Cir.1978). Enlarging on this argument, Respondent concludes that an employershould not be compelled to accumulate mixed sound levels to determineemployee exposure. Respondent extrapolates from this conclusion theproposition that since dosimeter readings measure varying degrees ofnoise levels, as well as impact and continuous noise levels, theinclusion and acceptance of dosimeter readings as evidence of employeeexposures deprives it of due process.We fail to follow Respondent’s argument since we find no ambiguity inthe standard’s requirements or application, particularly with regard tonoise measurements, and accordingly find Respondent’s arguments withoutmerit.Looking to the Noise Standard, maximum noise exposure for employees isclearly and unambiguously set forth in Table G-16. The Noise Standardbecomes actionable upon the existence of excessive noise level exposureby employees. Whatever the provisions of Footnote 1, it is clear it hasno effect on the unambiguous application of the noise exposure valuesset forth in Table G-16 showing permissible noise exposures. Nothing inFootnote 1 alters the mandate of the Standard that noise levels inexcess of the permissible noise exposure values be eliminated througheither feasible engineering or administrative controls. We find thereading of Noise Standard clear and concise, having easy understandingand application. Respondent misreads the purpose of Footnote 1, with itscomponent calculations, in relationship to the mandated permissiblenoise level exposures. As mentioned earlier, Footnote 1 is an advisorynote dealing with a method of noise calculation and not with basic noiselevel exposure limits; additionally the noise formula goes with the textof that note.We find no due process problem, as proposed by Respondent, sinceFootnote 1 presents nothing more than directions on how to use alonghand method of computing noise level exposures when different levelsof sound are present in the workplace. Footnote 1 is an acousticalengineering formula for calculating varying noise measurements.Apparently at the time the Noise Standard was written, Footnote 1 wasneeded in that dosimeters had not been invented (Tr. 744, 745). With theadvent and availability of accurate and reliable dosimeters, Footnote 1is now all but ignored or forgotten by acoustical engineers orindustrial hygienists who might find it necessary to determine employeeexposure to sound levels since the dosimeter does the calculationsautomatically and accurately (Tr. 1153-1157). Furthermore, sinceFootnote 1 has no application in determining or calculating noiseexposure as a matter of law, and in this case as a matter of fact, wefail to find the perceived error as alleged.Respondent next argues that the Standard fails to provide fair warningto employers that impact or impulse noise should be included in noisereadings. Again, Respondent refers to language of Footnote 1 whichreads, \”exposure to impulse or impact noise should not exceed 140 dBpeak sound pressure levels.\” Alluding this provision to the complianceofficer’s operating handbook (OSHA Industrial Hygiene Manual) dealingwith the measuring of noise, Respondent maintains that the dosimeterreadings should not be followed or used as a means of determiningemployee exposure since the dosimeter measured both impact andcontinuous noise levels.We again find Respondent’s argument without merit. The referencedprovision of Footnote 1 gives the end limits of noise exposure at 140decibel; that is, no employee is allowed such an exposure for any momentor time period. This coupled with the well established principle that acompliance officer’s manual has little relevance since it in no wayaffords a party procedural or substantive rights, leaves Respondent’sargument without any credence. Moreover, the failure to follow suchguidelines, if any existed, is not grounds for vacating a citation._General Electric Co._, 7 BNA OSHC 2183, 1980 CCH OSHD ? 24,502; _PAFEquipment Co._, 7 BNA OSHC 1209, 1979 CCH OSHD ? 23,421; _Limach Co._, 6BNA OSHC 1244, 1977-1978 CCH OSHD ? 22,467; _FMC Corp._, 5 BNA OSHC1707, 1977-1978 CCH OSHD ? 22,060; _Combustion Engineering, Inc._, 5 BNAOSHC 1943, 1977-1978 CCH OSHD ? 22,241; _Concerned Residents of BuckhillFalls v. Grant_, 537 F.2d 2938 (3rd Cir. 1976).Respondent next challenges the specific noise readings taken by thecompliance officer used to establish the existing noise levels to whichemployees were exposed, and which were the basis for issuance of theCitation. Respondent argues that both the sound level meter and thedosimeters used to take the readings were inaccurate, unreliable,improperly used or otherwise produced erroneous data. Respondent lists avariety of reasons why both the sound level meter and the dosimeter didnot accurately measure sound levels and noise level exposures, namelyfailure to calibrate the dosimeter prior to inspection, failure to usewind screens on microphones, the existence of moisture and wind in theworkplace, improper instruction to employees wearing dosimeters, faultymonitoring of employees wearing dosimeters by the compliance officer,failure to take noise readings in lunchroom, and the inclusion of impactand continuous noise levels in the dosimeter readings.We have reviewed the record most carefully, and find more thansufficient evidence to support a finding that the sound level meterreadings and the dosimeter readings were reliable and accurate and fullyusable as a source to establish the existence of excessive noise levelsin the workplace.Looking to the compliance officer’s conduct in use of the sound levelmeters and the dosimeters, the evidence clearly establishes that thecompliance officer calibrated the sound level meter and dosimeter bothbefore and after use in Respondent’s workplace to insure their accuracy(Tr. 2263). That the dosimeter and the sound level meters were properlyoperating and accurately recording sound levels and sound levelexposures is most clear and convincing in the record and was neverchallenged by any evidence submitted by Respondent (Tr. 2290, 2291,2293, 2296, 2297, 2298, 2316, 2317, 2428, 2429, 2432, 2433, 2483).Additionally, we find that the compliance officer did instruct employeesin the proper use of the dosimeter and did monitor the dosimeterperformance as well as the employees’ wearing of the dosimeter to insureit had been done correctly. When a dosimeter was found to be used oroperated in a manner not consistent with good operational procedures, orused in a manner inconsistent with the instructions given by thecompliance officer, the reading of that dosimeter was disregarded (Tr.2264, 2265, 2312, 2435, 2514, 2515, 2516, 2537, 2538). The evidence isclear that the compliance officer fully understood the importance ofproper pre and post inspection calibration of both the sound level meterand dosimeter and did so when she used these instruments (Tr. 2483, 2481).Additional support for the reliability and accuracy of the noisereadings made by the compliance officer is found in the testimony of theSecretary’s expert witness as well as Respondent’s expert witness, bothof whom were acoustical engineers. Each of these experts testified thattheir own individual noise readings were consistent with the noisereadings taken by the compliance officer (Tr. 918, 4547, 4549,4560-4563, 4594, 4595, 4624). While there is sufficient evidence in thisrecord to show that the compliance officer’s readings, standing alone,were reliable and accurate, the fact that both parties’ technicalconsultants, (who were intimately connected with the measurement andanalysis of sound and the operation of sound measuring equipment) hadfound that their measurements were totally consistent with thecompliance officer’s noise measurements established beyond any possiblechallenge that the sound level and sound exposure measurements of thecompliance officer, introduced into evidence, were solid, unrefutable,and totally reliable in establishing sound levels existing inRespondent’s workplace and sound exposures of employees.Respondent forwards one last argument concerning the validity of thedosimeter readings. This argument deals with the ability of thedosimeter to accurately record sound levels of varying intensity over aperiod of time. Respondent argues that dosimeters used to measure soundlevels in a typical metal fabricating environment, will record andindicate noise levels 5 to 13 decibels of sound (hereinafter \”dBA\”)[[5\/]] higher than the sound levels actually existing in the workplace.Respondent maintains that this situation results from the dosimeter’sinability to record simultaneously both impact and impulse noise. As abasis for its argument, Respondent directs our attention to Chapter VIof OSHA’s Industrial Hygiene Manual,[[6\/]] and statements by thecompliance officer that the dosimeter would record impulse noise but notimpact noise (Tr. 244). Respondent also utilizes the testimony of theSecretary’s expert, George Kamperman, an acoustical engineer, to supportthe proposition that the dosimeters used in a typical metal fabricatingnoise environment would produce results 5 to 13 dBA higher or six timesthe expected noise relative to the Noise Standards 5 dBA \”exchange rate\”[[7\/]] (Tr. 1168, 1171). Respondent thus asserts that this dosimetererror coupled with a standard plus\/minus 2 dBA error inherent in everydosimeter results in erroneous readings from two to six times theexpected noise levels in relation to OSHA’s 5 dBA \”exchange rate.\”While the record might show some inconsistency in the prolongedtestimony concerning the accuracy of the dosimeter [[8\/]] there appearsto be no question that the dosimeter is an instrument that canaccurately measure and record total noise exposure over a specificperiod of time. I reached this conclusion from the lengthy testimony ofMr. George Kamperman, an acoustical engineer, whose experience in thesematters was significant. He showed a substantial involvement in thedevelopment and use of sound and vibration measuring devices, inaddition to the development and production of the dosimeter itself (Tr.740, 741, 743). I find his testimony persuasive concerning the reliableuse of sound level meters and dosimeters and an the fact that thesedevices gave equivalent answers in the measurement of sound levels,albeit that the format of such sound readings may be different (Tr. 743,745, 746). I find Mr. Kamperman’s testimony to be reliably supported bythe record, that the type of noise generated at Respondent’s metalworking plant was such that a dosimeter could accurately measure andrecord the levels of noise exposure of Respondent’s employees in theworkplace (Tr. 747, 918, 1153, 1154, 1156, 1157, 1174, 1183).As above discussed, we found the method employed by the complianceofficer in utilizing the sound level meter and dosimeter to be correctand in accordance with standard operating procedures aimed at accuratelyrecording sound levels and sound level exposures. Now having determinedthat dosimeters indeed accurately record sound levels in excess of TableG-16 of the Noise Standard, as determined by the Secretary. It followsthen that the Secretary’s allegations, as contained in the Citation andComplaint, that employees were subjected to sound levels exceeding thosepermissible levels listed in Table G-16 of the Noise Standard aresupported by significant evidence in this record and. as a matter offact and law, will be sustained.This now brings us to the next issue as to whether there existedfeasible administrative or engineering controls capable of reducing theexcessive noise levels to those levels allowable in Table G-16, of theStandard, and whether Respondent was obligated to install such controls. VI. THE REQUIREMENT FOR USE OF FEASIBLE ENGINEERING OR ADMINISTRATIVE_CONTROLS TO REDUCE EXCESSIVE SOUND LEVELS_We find the language of the Noise Standard unequivocal in its mandatethat once noise levels, to which employees are exposed, exceed thoselevels listed in Table G-16 of the Standard the employer must utilizefeasible engineering or administrative controls to reduce the excessivesound levels to permissible levels. A considerable body of case law hasevolved concerning the interpretation of the term \”feasible engineeringcontrols.\” Within this body of law developed a difficult and changingconcept concerning the economic component of technological feasibility.The application of this concept of economic feasibility has over theyears taken on different meanings within the Commission. Withoutbelaboring this decision with the complete dialectic history of thedefinitional evolution of economic feasibility and technologicalfeasibility, we will start at that point of interpreting the NoiseStandard with the case of _Continental Can_, 4 BNA OSHC 1541, 1976-1977CCH OSHD ? 21,009 (No. 3976, 1976).In the _Continental Can_ case, there evolved what has come to be knownas the \”cost-benefit\” theory. This theory dealt with determining whethercontrols were feasible or not, whether there was a realisticconsideration of the hazards presented by excessive noise, and whetherthe health benefits to employees from noise reduction through the use ofengineering controls justified the cost to the employer. The rationaleof this decision was that the Act recognized that perfect safety andhealth could not be achieved, and that a balancing of costs and benefitsmust be made. (_Also_ _see_, _Castle & Cooke Foods_, 5 BNA OSHC 1435,1977-78 CCH OSHD 854, _aff’d_ 692 F.2d 641 (9th Cir. 1982).Propelled by the U.S. Supreme Courts decision in _American TextileManufacturers’ Institute, Inc., v. Donovan_, 101 S. Ct. 2478 (1981),(hereafter \”_ATMI_\”), the Commission in _Sun Ship, Inc._, 11 BNA OSHRC1028, 1983 CCH OSHD 26,353, appeal dismissed No. 83-3081 (3rd Cr. 1983),found Continental Can to be an erroneous decision and set aside thecost-benefit theory of economic feasibility. The Commission held thatthe Supreme Court in _ATMI_ had determined that when interpretingstandards the term \”feasible\” meant \”capable of being done\” or \”thatwhich is achievable\” and that Congress intended employee health tooutweigh \”all\” other considerations save those making the attainment ofthis benefit \”unachievable\” (101 U.S. at 2490). The Commission’s conceptof engineering and economic feasibility looked to specific costconsiderations and, whether such costs would be so high as to threatenthe economic viability of an industry. The Commission placed upon theSecretary the burden of proving feasibility and that the cost of theengineering and administrative controls would not threaten the citedemployer’s long-term profitability and competitiveness, or that theemployer’s inability to afford the cost of controls resulted from theemployer’s lagging behind its industry in providing safety and healthprotection for employees. [[9\/]] Thus, in _Sun Ship_ we find theCommission fairly well tailoring its definition of technological andeconomic feasibility along the lines articulated in the _ATMI_ case.Looking to the most recent of Commission cases (issued July 20,1984) wesee the Commission requiring The reexamination of the economicfeasibility ruling contained in _Sun Ship_. _Sherwin-Williams Co._, 11BNA OSHC 2105, 1984 CCH OSHD ? 26,986. The Commission in_Sherwin-Williams_, determined that the cost benefit analysis containedin _Continental Can_ and _Castle & Cooke_ cases still had application)in noise cases regardless of the holding in _ATMI_. The Commissionreached this decision by holding that the _ATMI_ decision had limitedapplication to standards promulgated under Section 6(b)(5) of the Act,quoting _Industrial Union Department v. American Petroleum Institute_,448 U.S. 607, 642 (1980). The Commission specifically stated that theSupreme Court had expressly reserved judgment on the promulgation ofstandards under Section 3(8) of the Act, 29 (U.S.C. ? 652(a) (_ATMI_,452 U.S. at 513 (note 32).That being the case, the Commission concluded it was not required by_ATMI_ to abandon the cost-benefit approach of _Continental Can_.Significantly the Commission found support in bringing back thecost-benefit theory in the rulings of the Fifth, Ninth and SeventhCircuit Courts, citing _Marshall v. West Point Pepperell,Inc._, 588F.2nd 979 (5th Cir. 1979); _Donovan v. Castle & Cooke Food_, 692 F.2d641 (9th Cir. 1982); _International Harvester Co. v. OSHRC_, 628 F.2d982 (7th Cir. 1980); _Turner Co. v. Secretary of Labor_, 561 F.2d 82(7th Cir. 1977).We now reach that point, in this decision at least, where we mustdetermine what test or theory of economic feasibility is to be applied,and by what method we must apply that particular theory to the facts ofthe case. This is not without some difficulty since the Commission inreturning to the \”cost-benefit\” theory, distinguishing the _ATMI_ and_Sun Ship_ cases from its own basic ruling, came to rely on variousCircuit Court cases to reach a synergistic decision balancing its ownrational set out in _Sherwin-Williams_ with that of the variouscircuits. Accordingly, while we are constrained now to apply the\”cost-benefit\” theory in reaching the economic feasibility issue, wemust look to the circuits for guidance in the application of thattheory. In this regard we, in the instant matter, are particularly boundto the rulings of the Seventh Circuit.In _International Harvester_, _supra_ the Seventh Circuit’s evaluationof economic feasibility of noise controls, like the Commission, requiredthat all relevant costs and benefit factors be considered. However,unlike the Commission, the Court pointed out that in comparing costs andbenefits, benefits to employees must be weighted more heavily in thecalculus than costs to the employer (628 F.2d at 988). Departing fromthe ruling of the Seventh Circuit, a; well as other circuits, we findthe Commission in _Sherwin-Williams_, in reaching its own cost analysis,comparing the cost of earplugs to the cost of engineering controls withpredictable results assured. Furthermore, unlike the Commission holding,the 7th Circuit held that the fact that controls are expensive does notmean they are not economically feasible under the cost-benefit theory,citing _Turner Co. v. Secretary of Labor_, 561 F.2d 82 at 988 (7th Cir.1977).The Seventh Circuit, in _International Harvester_, held, that since theuse of personal protective equipment could not prevent _all_ employeesfrom experiencing hearing loss and hearing difficulties, the benefits toemployees from implementation of engineering noise controls would besubstantial (at 989). The Court, like the Commission, considered thefact that substantial noise reductions could be obtained through the useof engineering controls, and that such substantial noise reduction wouldprevent hearing loss. However, the Court in recognizing the remedialpurpose of the Act, after considering the availability of personalprotective equipment which had a miniscule cost in relation to the costof implementation of feasible engineering controls, found the benefitsof engineering controls such that they were held economically feasible.It should be noted that the Commission, in _Sherwin-Williams_, inreaching its conclusion as to whether engineering controls wereeconomically feasible, within the meaning of the Noise Standard,specifically looked to the holdings of the various Circuit Courts, i.e.,Fifth, Sixth, Seventh and Ninth to determine if their costs werejustified or not when measured against their expected benefits, 11 OSHC,at p. 2109, 1984 CCH OSHD 26,986, at p. 34,703. Accordingly, it will benecessary for us to review those considerations of the Seventh Circuitas to \”expected benefits\” in making the correct balance under thecost-benefit theory, and then juxtapose those considerations against theCommission’s own application of the cost-benefit theory in order toarrive at a decision which comports with applicable and controlling law.It must be said at this point that, looking to the facts of this case,and then applying controlling Circuit rulings, the result to be arrivedat will be somewhat different than one would initially expect under theholding of the _Sherwin-Williams_ case. The Commission in adopting thecost-benefit theory, unlike the Circuit rulings. placed great emphasison the cost component in the cost-benefit formula, comparing costs ofengineering controls to costs of earplugs. This can be seen from thelanguage of _Sherwin-Williams_ wherein the Commission states:\”The burden of producing evidence then returns to the Secretary \”whomust establish that the benefit of the proposed engineering controlsjustifies their relative cost in comparison to other abatement methods.\”1984 CCH OSHD 26,986, at p. 34,072\”In this case, the cost per employee is even greater than that in_Continental Can_ and _Castle & Cooke_. . .here, the total cost is manytimes higher and-significant continuing costs will be incurred. . .\”1984 CCH OSHD, 26,986 at page 34,703). . .The Seventh Circuit in contrast to the Commission, in discussing costs,has consistently held that relative cost comparison should not to bemade between personal protective equipment and engineering controlsstating \”furthermore in comparing costs and benefits, benefits toemployees weigh more heavily in the calculus than cost to the employer.\”See _International Harvester Co._, _supra_ at 988; _RMI Co._, _supra_ at572. We find a dichotomy in the term \”benefits\” as understood by theCommission and as understood by those courts which have, from time totime, defined the cost-benefit theory. The Commission in_Sherwin-Williams_ has given to the term \”benefits\” a performanceingredient focusing on the perceived capabilities of personal protectiveequipment [[10\/]] as opposed to the definitive capabilities ofengineering controls. In _Sherwin-Williams_, this \”capability\”ingredient translated into employers being given the option to chooseeither engineering controls or a program of personal protectiveequipment in complying with the basic terms of the Standard. It is inthis context that the Commission understands the term benefit, a conceptquite apart from that ascribed to by the various Circuit Courts who havespoken on this issue.The Commission, in arriving at their understanding of \”benefits\” lookedto the least costly means or measures of interpreting ostensiblecompliance with the Standard, all but ignoring the considerable benefitswhich would be obtained by the employees through the more expensivemethod of lowering sound levels. The Commission’s approach to defining\”cost benefits\” is to do so in terms of the least costly proposedabatement method, without taking into consideration the absence orshortfalls in health and safety benefit to employees as is easily seenby the evidence they considered and did not consider, and as noted inthe Commission’s conclusions that\”the testimony of employees reveals that shortcomings exist inSherwin-Williams’ personal protective equipment program but they do notaffect the conclusion that the wearing of earplugs by Sherwin-Williams’employees reduced sound levels experienced by employees within TableG-16 limits. While engineering controls would significantly reduce noiseexposure, the Secretary has failed to demonstrate that the benefits thatwill be achieved by the controls justify the cost of implementing them.\”1984 CCH OSHD at page 34,702.Significantly the Commission’s conclusions in _Sherwin-Williams_followed unrebutted testimony of employees that the wearing of earplugsand earmuffs produced \”discomfort, difficulty in communication, hearingloss, difficulty in determining the source of sounds, and earinfections…that employees took earplugs out during the workday andaltered the plugs to make them more comfortable.\” 11 BNA OSHC at 2107,1984 CCH OSHD, at p. 34,699. Additionally, three experts in the field ofaudiology bolstered the testimony of the employees, in that eachindicated that the employer’s program of personal protective equipmentdid not protect the hearing of employees and that employees would sufferhearing loss if they continued to work in the sound levels existing atSherwin-Williams’ worksite. The Commission’s emphasis on the totaldollar expenditure as the controlling factor in applying the\”cost-benefit\” theory was pointed out by Commission Cleary in hisdissent stating:\”By its decision the majority virtually eliminates engineering controlsfrom the Noise Standard. This is neither consistent with the languageand purpose of the Act nor with the Commission power.\”1984 CCH OSHD, at pages 34,703, 34704Thus, we see the Commission’s application of the cost-benefit theory, asabove discussed, presenting a significant departure from the SeventhCircuit application, as well as these other Circuit rulings on thisissue. The Circuits have been unanimous in their view that in applyingthe \”cost-benefit\” theory one must emphasize the benefits a particularmode of compliance with the Standard will give employees, and not thecheapest and least expensive way possible for the employer to gaincompliance with the Standard. In _International Harvester Co._, _supra_,the court noted, in finding engineering controls preferable to the lessexpensive earplug method,\”the benefit to _employees_ from implementation of engineering noisecontrols will be substantial. The use of personal protective equipmenthas not prevented hearing loss and hearing difficulties among theseemployees. Implementation of engineering controls will bring all 80employees within permissible noise exposure level\” (at 989). (Emphasisadded).Likewise the Fifth and Sixth Circuits (cited by the Commission inreaching its interpretation) follows the Seventh Circuit in holding thatthe Noise Standard reflects the determination that the preferable methodof abating hazardous noise exposure is through technological oradministrative controls such as the installation of damping material tomachine surfaces and work walls. (_Marshall v. West Point Pepperell,Inc._, 588 F.2d 979 at 981) (5th Cir. 1979); _RMI Co. v. Secretary ofLabor_, 594 F.2nd 566 (6th Cir. 1979). The courts have found thatnumerous medical and physiological studies document the serious effectsof excessive noise exposure on employees, (_West Point Pepperell, Inc._,_supra_, at 981, footnote 4). Of particular significance is the courtsrecognition of the Noise Standard’s mandate that only when engineeringor administrative controls are not feasible, would workers be provided,and required to wear, personal protective equipment (_West PointPepperell, Inc._, at 981). Likewise in _RMI Co. v. Secretary of Labor_,_supra_ the court quoting _Turner Co._, 561 F.2d 82 (7th Cir. 1977),held that the cost of proposed controls was to be balanced against theproposed benefits flowing therefrom in order that resources would beallocated in priority to the degree of harm established and that\”Controls will not necessarily be infeasible merely because they areexpensive.\” (at 572) Thus we see somewhat of a uniform approach by thevarious Circuits, including the Seventh, that in applying the costbenefit theory it is the benefits to employees that must be compared tothe costs rather than the making of a comparison of costs of personalprotective _vis-a-vis_ the costs of engineering controls.Looking to both Commission and Circuit case law, we find ourselves withdistinct differing interpretations as to the application of thecost-benefit theory. We find the Seventh Circuit application showing anemphasis towards employee health and safety considerations,[[11\/]]whereas the Commission, in _Sherwin-Williams_, while recognizing thatemployees were experiencing hearing loss and having troubles with thepersonal protective equipment program, nonetheless choosing personalprotective equipment as the most expedient means of complying with theNoise Standard. In opting for expediency the Commission, in_Sherwin-Williams_, garnered to the employer the economic benefits of acheaper means of compliance, subordinating the protective benefitsassociated with the more costly engineering controls — all the whileemployees experiencing hearing loss, ear infections, and other safetyrelated dangers in the high noise environment where they were obligatedto wear earplugs. The Seventh Circuit, quite apart from the Commission,specifically found engineering controls to be economically feasible byreason of the very fact that they benefited those employees who didexperience hearing loss and other hearing difficulties, and that itrepresented a means which would bring _all_ employees within permissiblenoise exposure levels. _International Harvester Co. v. OSHRC_, _supra_at 989.We find ourselves with two distinct approaches to the application of thecost- benefit theory, each having significant differences and departuresfrom one another — all of which leave us in a difficult position infashioning an Order in this case. Looking to guidance from the courts,we are assured of the general principle that an agency is charged withthe responsibility of formulating law or policy, while the courts bearthe final responsibility for interpreting that law or policy. _BeverlyEnterprises v. NLRB_, 727 F.2d 591 (6th Cir. 1984). We are bound byCommission precedent to apply the \”cost-benefit\” theory, as mandated by_Sherwin-Williams_, disregarding the former \”feasibility test\” of _SunShip_. We are likewise bound to follow the dictates of the CircuitCourt’s appellate jurisdiction, and not depart from their judgmentalinterpretation and application of the rules promulgated by theCommission, _Beverly Enterprises v. NLRG_, _supra_; _Smith Steel CastingCompany v. Donovan_, 725 F.2d 1032 (5th Cir. 1984).Accordingly, we find ourselves constrained to follow the case law of theCircuit in which these proceedings have taken place, and are required toapply the courts reading of the cost benefit theory within the Standard._Smith Steel Casting Co. v. Donovan_, _supra_; _Beverly Enterprises v.NLRB_, _supra_.Looking to the test established by the Seventh Circuit, the benefits toemployees will be weighed against the cost to the employer, and allrelevant cost and benefit factor will enter into the determination ofeconomic feasibility. However, unlike the Commission, employee benefitswill weigh heavier on the scale than costs to employer, and we will lookto an abatement method that leaves no employee behind, unprotected, andpossibly sustaining injury. _RMI Co. v. Secretary of Labor_, _supra_ at572, _International Harvester Co. v. OSHRC_, _supra_, at 988 and 989. Aswill be more fully discussed below, applying this test, it is clear thebenefits to employees from mandating the installation of engineeringcontrols will be found to decisively outweigh their costs to theemployer, and will be required of the employer where foundtechnologically feasible._VII. BENEFITS UNDER COST BENEFIT THEORY_The litmus test in determining economic feasibility of proposedengineering controls under the cost-benefit theory requires that wecarefully analyze those benefits which inure to the employees from theimposition upon the employer of the requirement to provide suchengineering controls, and weight them against cost of such controls tothe employers. At the same time there must be a realistic recognition ofthe capabilities of a program of personal protective equipment and itsability to meet the health hazard associated with excessive noiselevels. Selection of the most effective method, in terms of employeeprotection after each method has been found to be a viable alternative,would be the last step toward compliance with the Standard. — Forpractical purposes we will compare the two methods.First in taking inventory of employee benefits obtained throughengineering controls and through personal protective equipment, it isimportant that we fully understand the nature of the health hazardinvolved. We find, and the record herein demonstrates, that excessivenoise levels represent an insidious occupational health hazardmanifesting itself in sensory neural hearing loss. It is insidiousbecause the,\”hazards to the auditory system are not dramatic like hazards to vision- onset is slow and deterioration is gradual. Often the process is faradvance before it is noticed and only then is the extreme socialdebilitation of deafness recognized – too late for either prevention orcorrective measures to help much.\”(Respondent Exhibit 10, p. 59)In opting for a particular means of protection it is particularlyimportant to choose the most effective or proven means to protect theemployees since unrealized expectations of a particular touted mode ofemployees protection when found wanting, will have allowed for thehazard to cause hearing loss to employees. Hearing loss induced throughthe excessive noise levels is irreversible since it involves damage tothe neurological auditory sensors (Tr. 3900). The loss of this vitalsensory system has a significant adverse impact and creates a handicapin a person to a degree closely comparable in magnitude to the handicapassociated with loss of sight (Tr. 3898-3899). Indeed it could be saidthat a person who is blind can function much better in society than aperson who has experienced a hearing loss (Tr. 3897). Clearly thequality of life of a person experiencing noise induced hearing loss issignificantly impaired (Tr. 3898, 3899).Employees who begin to experience hearing loss due to excessive noiselevels in the workplace will continue to do so if they remain in thehigh noise level (Tr. 3900). It is only through the reduction of noiselevels to approximately 80 dBA that there exists no risk to employeesfrom the pernicious and insidious health hazard associated with suchexposure to high noise levels (Tr. 3901, 3908).In assessing the benefits to employees and weighing them against thecost of engineering controls, vis-a-vis the costs for a program ofpersonal protective equipment, we must compare the scope of protectionafforded the employees from each mode and how each provides thatnecessary quantum of protection from the hazards involved._International Harvester Co._, _supra_; _RMI Co. v. Secretary of Labor_,_supra_.The evidence establish that noise levels in Respondent’s plant wereexcessive, generally averaging in the upper 90’s dBA range, (97, 98, 99dBA) and at times reaching 113 dBA (Tr. 2267-2284). It is important thenthat in resolving the \”cost-benefit\” issues herein we select a means ofhazard abatement which would reasonably assure all of Respondent’semployees protection from the dangerous high noise level. The SeventhCircuit found this test to be a particularly important one in choosing ameans of employee protection; it was on this basis that the court foundengineering noise controls feasible under the cost-benefit theory(notwithstanding their significant costs), rejecting a program ofpersonal protective equipment as a suitable method of compliance._International Harvester Co._, _supra_ at 989.[[12\/]] We first look tothe levels of protection obtained from use of a program of personalprotective equipment.The evidence in this case unequivocally established that the advertisedexpectations for noise attenuation obtained from use of personalprotective equipment did not correlate with its actual on-the-jobperformance as a means of noise hazard protection. Comprehensive studiesshowed that in actual use personal protective equipment did not providethe extent of protection that it was designed or held out to provide.Looking to Respondent’s Exhibit R-10, we find that attenuation factorsfor earplugs varying greatly from their advertised and mean attenuationexpectations. Indeed it appeared undisputed that the results of recentstudies indicated that hearing protectors, worn in real life conditions,produced attenuation results, for a given work force, in a quantum farless than the manufacturers stated estimates of noise attenuation (Tr.3862), Ex. R-10; also see _Sherwin Williams Company_ at 34,705, footnote5, Commissioner Cleary dissenting) [[13\/]]. In the instant case we findthat Respondent provided its employees with two types of earplugs, thesebeing \”Swedish Wool\” and the \”E-A-R\” earplug (Tr. 405). UtilizingExhibit R-10, we are able to evaluate the actual degree of employeeprotection provided to Respondent’s workforce by use of these earplugs.In the case of the earplug known as \”Swedish Wool\” under the trade nameof Bilson, (Exhibit R-10, page 17), we find that while the majority ofthose employees, who correctly wore the earplug, received a meanattenuation of 18.4 decibels of noise attenuation, a significant portionof the employees would receive only 13.5 and 8.4 decibels of noiseattenuation (Ex. R-10, page 10, line 5). Applying these attenuationfactors to noise levels existing at the workplace, we find that theneural component of some of the employees’ ears, while wearing theearplug, were exposed to noise levels from 105 decibels to 97 decibelsin the higher levels, and 90 to 82 in the lower levels. Looking to thedata for the \”E-A-R\” earplug, we find that while the majority ofemployees would receive 31.5 decibels of attenuation, if correctly worn,a significant portion of the employee population would receive 24.1 and16.5 attenuation. This would result in employees, who wore the earplug,being exposed to sound levels of 89 to 97 decibels in the higher end ofnoise exposure.It should be noted that cited attenuation figures apply to thoseindividuals who correctly comply with Respondent’s plan of personalequipment. Those employees who, knowingly or unknowingly, incorrectlyuse or insert the plugs, or who by reason of inability due todiscomfort, ear infection, or personal reasons, fail to wear theirpersonal protective equipment throughout the work day, will of coursereceive a small modicum of protection or no protection at all. [[14\/]]Consistent with this finding that not all employees who wore earplugswere receiving adequate noise protection was the evidence, of record,that Respondent’s employees were experiencing neurological damage totheir ears from the excessive noise levels, and that Respondent’s methodof choice of complying with the Noise Standard, its program of personalprotective equipment, was not capable of providing that quantity ofprotection to all employees necessary to arrest their continued loss ofhearing.This learned fact, of course, is not to be unexpected since the evidenceestablished the existence of gaps or lapses in the program such as whenemployees, while wearing the \”E-A-R\” earplug and receiving greater noiseattenuation benefits, found that in order to carry out their jobfunctions they needed, from time to time, take the earplug out tocommunicate in the workshop or on the telephone (Tr. 2218).Additionally, the inserting and removing of earplugs seemed to present aproblem to employees by reason of the oil mist in the air and thedeposits of oil on the earplug and within the ear (Tr. 2218, 2219).Consistent with the evidence of difficulties in use of earplugs was thetestimony of employees who reported hearing loss since working atRespondent’s workplace (Tr. 2091, 2213, 2232, 2237, 3876).The phenomena of lower expectations in the capability of personalprotective equipment seems consistent with other documentary evidencewhich showed that employees were experiencing noise induced hearingloss, or were tending toward that condition. Looking to Respondent’s ownaudiometric examinations, many employees were showing signs or trendstowards noise induced hearing loss. (Ex. C-75, 76, 77, 80, 81).While Respondent’s expert in the field of audiology disputed the factthat employees were experiencing signs and trends of hearing loss,resulting from excessive noise level, to any degree exceeding that ofthe general population, I specifically find the conclusions ofRespondent expert, in this regard, without any credence orpersuasiveness since the method employed by Respondent’s expert toarrive at these conclusions were Without foundation and appearedcontrary to the basic facts of this case. Looking to the testimony ofRespondent’s expert concerning his calculations of the percentage ofhearing loss within Respondent’s work force, what he calculated andconsidered to be a normal rate of hearing loss among employees, ascompared to the general population, would within 20 years leave each andevery employee within that work force stone deaf. One could hardlycharacterize such a result as normal, as did Respondent expert witness(Tr. 3442, 3667, 3679, 3685, 3796-3801).Using Respondent’s own audiometric test results we do find credible thefact that 28.6% of the employees in Department 75 showed adverse changesin hearing sensitivity, 14.9% of employees in Department 2 showed trendsand shifts towards hearing impairment, and in Department 6, 6.3% showedsuch shifts in hearing (Tr. 3705-3706, 3707, 3708, Ex. C-77, 78, 79, 80,81). Apparently on a plant wide basis, 17% of employees showed shifts intheir hearing (EX. R-28, Tr. 3705).Significantly, the data introduced at trial concerning the audiograms,and the extent employees were experiencing adverse shifts in theirhearing, is consistent with the testimony of the employees themselvesconcerning problems developing in their hearing while employed atRespondent’s work place. Employees testified that since being employedwith Respondent their hearing had deteriorated and had presented themwith problems in communicating with others (Tr. 2091, 2213, 2232, 2237).We find that substantial evidence has amply demonstrated the fact that aprogram of personal protective equipment (earplugs, ear muffs) is notable to afford to all of Respondent’s employees that requisite quantityof protection, under the Noise Standard, from the hazards associatedwith high noise levels in the workplace. The courts have mandated thatsuch a program is unacceptable as compliance with the Noise Standardunder the cost benefit theory. _International Harvester_, _supra_; _WestPoint Pepperell, Inc._, _supra_.While the dollar costs of such a program of personal protectiveequipment are clearly nominal when compared to the costs of engineeringcontrols, effective protection of the entire work force is a benefitlost with such economies. When looking to the efficacy of engineeringcontrols, however, we are assured of one controlling fact, that beingthat the noise exposure of _all_ employees will not exceed that levelobtained through use of noise controls designed to lower the excessivenoise levels. As recognized by Respondent’s own expert, a good hearingconservation program aimed at conserving human hearing includes the useof engineering noise controls to lower overall noise levels (Tr. 3765,3888). Once engineering controls are in place there should be no reasonwhy an employee should sustain hearing impairment as a result ofexcessive workplace noise (Tr. 3888). This is a benefit that just cannot be overlooked and one the courts have used to justify reasonablecosts expenditures for feasible engineering controls. Furthermore thecourts have recognized that the employer is obligated to take thosesteps necessary to protect those workers who predictably will not farewell under a program of earplugs or ear muffs. (_InternationalHarvester_, _supra_ at 989; pp. 35 to 39, _supra_).It is important to be aware of the fact that once the insidious natureof high noise levels is eliminated through use of engineering noisecontrols no employee in the work force will be adversely affected. Wefurther know that employees showing potential risk of hearing loss, oncein the lowered noise levels will now enjoy no threat to their hearingand any hearing loss experienced will be arrested (Tr. 3901, 3873, 3900).When considering that employees are working in an environment of noiselevels up to and exceeding 100 decibels the use of earplugs hardly seemsthe type of benefit envisioned by the Noise Standard; its mandaterequires realistic employee protection from the hazards of high noiselevels. The various Circuit Courts have likewise reached this sameconclusion in interpreting the standard. _International Harvester Co. v.OSHRC_, _supra_; _RMI Company v. Secretary of Labor_, 594 F.2d 566 (6thCir. 1979); _Marshall v. West Point Pepperell, Inc._, _supra_.It has not been disputed that none of the adverse conditions effectingemployees would exist if noise levels were lowered through the use ofengineering controls since _all_ employees in the workplace would havean exposure no greater than that of the lower noise levels obtainedthrough the utilization of the engineering controls. We view this as abenefit to employees that well justifies the cost associated withfeasible engineering controls and is a result mandated by the Act, asset forth in its stated purpose assuring assure \”every working man orwoman\” safe and healthful working conditions (29 U.S.C. ? 651(b)).It must be emphasized at this point that in Section 2(b) of the Act,Congress specifically declared it to be its purpose to assure so far aspossible \”_every working man and woman_\” safe and healthful workingconditions 29 USC 865(b). Thus following this language we would expectan understanding that _all_ employees are entitled to adequateoccupational safety and health protection. Accordingly, a means ofcompliance with safety or health regulations that garners to some, oreven most, employees the required occupational safety and healthprotection is inconsistent with the stated purpose of the Act if it doesnot protect all employees. Such a means of compliance is not acceptableas compliance with the purposes of the Act, regardless of whatevertheory of \”cost-benefit\” is applied, _RMI Company_, _supra_ at 572;_International Harvester_, _supra_ at 989. Exhibit R-10, together withthe testimony of employees, the audiometric examinations, and reports ofRespondent’s expert show large gaps in employee noise protection makingthe program of earplugs unacceptable as a means of compliance with theNoise Standard, as required by the Act and as interpreted by the casescited.One last point concerning the comparative benefits of engineeringcontrols and personal protective equipment. A key adjunct to theeffectiveness of a personal protective equipment program is therequirement to monitor the work force’s hearing with follow-up to anydifficulties brought out by such monitoring; the monitoring of thehearing health of employees is done through audiometric testing. In theinstant case we find Respondent did administer audiometric examinationsto its employees working in the hazard area. However, for such programto be operated correctly, as an objective monitor of the effectivenessof the personal protective equipment, employees who are shown by theexamination results not to be doing well should be removed from thenoise hazard area so as to forstall any further loss in their hearing.They are then to be referred out to a doctor for consultation.Further, for a program of personal protective equipment to provide thebenefits of effective employee protection from the noise hazard (inparticular the type of benefit envisioned by the Commission asarticulated in _Sherwin-Williams_, _supra_), the monitoring programitself had to be fully effective, administered correctly, andconscientiously applied in order to have any semblance of effectivenesscomparable to engineering noise controls (Tr. 3438, 3764, 3765, 3888,3889, 3890, 3891). Thus, if the audiometric testing facet of the programof personal protective equipment is found ineffective, so must theentire program of personal protective equipment be evaluated. Accordingto Respondent’s own expert, the issuance to employees of earplugs alonecannot be considered an effective hearing conservation program designedto adequately protect employees from the hazards of high noise levels(Tr. 3764, 3767, 3888, 3890, 3891).In the instant case we find Respondent’s program of personal protectiveequipment a failure (not only from its inability to adequately protectall employees in the workplace as earlier discussed), but from the factthat the monitoring of the employees success in use of the earplugs wassuperficial at best. and otherwise a dismal failure. Respondent didadminister audiometric examinations to employees but that was as far asit went. The record herein is replete with evidence showing that thedata contained in the audiometric examinations was all but ignored. Theevidence showed that the audiometric tests were given solely toestablish the fact that such examinations were given, and not for thepurpose of monitoring the health of these employees exposed to the highnoise level. Nor were the audiometric examinations used to monitor theeffectiveness of the earplug program in the workplace (Tr. 76, 86, 87, 96).Employees were not told of the results of the audiometric exams, eventhough these examinations indicated hearing loss. What instead occurredwas for Respondent to lock away the exam results in the safe of itsGeneral Manager. While Respondent’s expert industrial audiologist wouldreview the examination results, and issue a printed report callingattention to the fact that employees were experiencing shifts in theirhearing, these reports, like the audiometric examination, were alsolocked away in the safe and not used in any manner whatsoever includingthe monitoring of adversely affected employees. The evidence is repletethat the monitoring program was a sham, designed solely to have itappear as though there was some sort of an act of compliance with theNoise Standard by Respondent — and it was nothing more then that (Tr.74, 76, 77, 78, 86, 96, 97, 99, 107, 401, 435, 2090, 2209, 2210, 2211,2229, 2230, 4311, 4309, 4313, 4314).Indeed the individual who was responsible for employee safety andhealth, the Manager of Security and Safety, admittedly did not know themeaning and purpose of the information given to Respondent by its ownaudiometric’s expert. Nor was he one bit interested in understandingwhat the audiometric exams as a whole showed (Tr. 87). He merely viewedthe program of audiometric exams, with its computer analysis printout,as a means of complying with the hearing conservation program mandatedunder the OSHA regulations (Tr. 96). In accordance with this view,Respondent had placed all the documents in a safe in the GeneralManager’s office never to be seen by any employee or member ofRespondent’s supervision again (Tr. 96, 97). Even if the audiometricexaminations and analysis reports were to reveal an employee going stonedeaf nothing would be done and the reports would be filed away neveragain to see the light of day (Tr. 100-107). Accordingly, we mustconclude from this, that as a matter of fact, Respondent’s program ofpersonal protective equipment did not offer the same requisite degree ofsafety and protection to employees that would have been available toemployees had Respondent installed engineering controls to lower theexcessive noise levels to safe levels._SUMMARY_We find the facts to squarely reveal that a program of personalprotective equipment is not able to adequately protect all employees inthe workplace from the hazards of excessive noise. We further find thatthe facts show that only through use of feasible engineering noisecontrols would all employees be assured they were receiving adequateprotection from the harmful effects of excessive noise levels.Additionally, the lowering of excessive noise levels through use offeasible engineering controls would prevent those employees who hadexperienced hearing loss, or who showed signs or trends of hearing loss,from experiencing further damage to their hearing. Thus, as a matter offact, we find that the benefits to employees from the installation offeasible engineering controls justifies the reasonable costs associatedwith the installation of such controls.We find that personal protective equipment, with their lesser costs, cannot provide the same degree of success in protection from the harmfuleffects of high noise levels as would engineering controls. We find, asa matter of fact, that the use of personal protective equipment, withall its economies of cost, not preventing hearing loss among allemployees, a result which could be obtained through use of engineeringcontrols.The evidence has established that the use of personal protectiveequipment did not prevent _all_ employees from experiencing hearingloss, that employees complained of hearing difficulties, that employeeswho experienced hearing difficulties were continued in employment in thehigh noise area, and under these circumstances the Seventh Circuit (aswell as all other Circuits) find that the reasonable costs ofengineering noise controls is well justified, — to which we are bound._International Harvester, Inc._, _supra_; _RMI Company v. Secretary ofLabor_, _supra_. The fact that the cost of the engineering noisecontrols is far greater than the cost of personal protective equipmentdoes not alter any conclusions herein which would make engineeringcontrols economically unfeasible, as has currently been determined bythe Commission. We are thus left with the inescapable legal conclusionthat, under the facts of this case, after application of thecost-benefit theory, feasible engineering control are required to lowernoise level in Respondent’s workplace._VIII. TECHNOLOGICAL FEASIBILITY OF ENGINEERING NOISE CONTROLS_We now turn to the issue of whether there were available to Respondentfeasible engineering noise controls capable of lowering sound levels inthe cited workplace. This issue, generally referred to as \”technologicalfeasibility,\” deals with the availability of engineering controlscapable of lowering sound levels while at the same time being fullycompatible with the manufacturing and maintenance processes associatedwith the machinery that is the subject of the noise suppressionactivity. As part of the technological feasibility issue, there existsan economic component dealing with the reasonableness of the cost forsuch controls as balanced against the amounts of noise reduction thatcan be obtained from such controls. (We suggest this is where the thrustof the \”cost-benefit\” determination should reside).With respect to the economic component of technological feasibility, theproposed noise control should incorporate a _reasonable balance_ betweenits noise suppression ability, the number of employees who will beaffected by the installation of such controls, and its ultimate cost. Ifa reasonable balance is met between these three factors, then thecontrols must be considered technologically feasible if they areotherwise able to lower noise levels without harming the manufacturingprocess. Establishing that the engineering controls are technologicallyfeasible, and that there exists a reasonable balance between these threementioned factors of noise suppression ability, the number of employeesaffected and ultimate cost, is the responsibility of the Secretary whomust preponderate on these issues. _Sherwin-Williams Company_, _supra_;_Donovan v. Castle and Cooke Food_, 692 F.2d 641(9th Cir. 1982); _RMICompany v. Secretary of Labor_, _supra_ at 574.Before turning to the individual facts involving technologicalfeasibility, it is important to sort out what evidence was foundcredible and reliable and what was not. Considerable evidence andtestimony has been submitted in this record on the issue of feasibilityof engineering controls. In evaluating the persuasiveness of thisevidence, we find that both the Secretary’s and Respondent’s acousticalengineering experts were highly credible witnesses whose views we foundpersuasive in reaching resolution on these issues. Complementing theviews of the experts was the testimony received from those employees whoboth operated and maintained the machines which are the subject matterof the action. The evidence received from these witnesses hadconsiderable impact upon the ultimate conclusions concerning thefeasibility of engineering noise controls. We, however, do not find thesame reliability or degree of persuasiveness in the testimony given bytwo of Respondent’s witnesses concerning the proposed engineeringcontrols. We specifically refer to the testimony of Mr. Prem Rai and Mr.Ted Gabryszewski. We find their testimony generally unreliable, biasedand inconsistent with that credible testimony received from theoperators and maintenance employees who, on a daily basis, worked withthe punch presses and screw machines.With respect to the observations of Mr. Rai, and his calculationsconcerning the estimated stockups made of each screw machine per day,together with the time calculated necessary to gain access to the screwmachines through the doors on the proposed acoustical enclosures. wefind no support in this record for their reliability. His conclusionswere at odds with the credible testimony of those employees who operatedand serviced the screw Machines and who were intimate with the day today operation of the machines. We find the testimony of Mr. Rai and Mr.Gabryszewski not reconcilable with the credible evidence as a whole, andgive no persuasive element to it. Continuing along this line. we findthe proposition contained in Respondent’s \”domino theory,\” whereemployees would open noise enclosures on a sequential basis in order tohear their machines operating, to be inconsistent with the facts of thecase and contrary to acoustical principles established by the expert inrecommending such noise controls.Since we disregarded the general conclusions reached by Mr. Rai and Mr.Gabryszewski as without foundation or reliability, it follows then thatthose opinions of Respondent’s acoustical engineering expert which werebased upon the conclusions and calculations of Mr. Rai andMr.Gabryszewski, and not of the experts own personal knowledge, mustlikewise be disregarded. With this in mind, we can now review theappropriateness of the engineering controls proposed by the Secretary asfeasible engineering controls._Department 75_Department 75 contains approximately 72 Davenport automatic screwmachines (Tr. 144, 2170, Exhibit R-8). These machines fabricate smallmachined components from bar stock fed into the receiving end of themachine, The parts produced are utilized in the assembly of Respondent’svarious product lines (Tr. 2168, 2612, Exhibits R-8, C-13, AE-1).Screw machine operators generally are assigned the responsibility forthe operation of two screw machines during an eight-hour shift (Tr.2169). Approximately 67 employees are assigned to Department 75 duringthe two shifts the Department is in operation (Tr. 327, Exhibit C-24).On the average, approximately 50 of the 72 machines are in operation ona given day (Tr. 1847, 2178). The screw machine operators’ duties aresuch that most of the workday is spent near his machines within theDepartment (Tr. 2172, Exhibit C-15).The Davenport screw machine is a noisy machine, having sound levelsfrequently exceeding the OSHA standard. On the average, operators areexposed to 99-102 decibel during their workday (Exhibits R-8, C-74). Theengineering noise control proposed for the Davenport screw machine is acomplete acoustical enclosure of the machine (Tr. 481, 1840, 2458). TheSecretary’s expert witness, whose testimony I find credible an thissubject, testified that the noise enclosure for a Davenport screwmachine is a state of the art, off-the-shelf item easily available toRespondent. He had seen these enclosures used in 50 to 100 othermanufacturing facilities (Tr. 909). The acoustical noise enclosure for ascrew machine is not an unusual engineering solution, but is one veryfamiliar to the industry and one that will significantly achieve noisereductions bringing the noise level within compliance with the NoiseStandard (Tr. 913-916). Indeed, it is possible that noise levels couldbe brought to levels of 85 decibels in this workplace (Tr. 2036, 2037).The cost for such acoustical enclosure is approximately $4,000-6,000,plus a like amount for installation (Tr. 916). We find the Secretary’sexpert’s conclusions consistent with the experience of the Respondent’sown personnel who had constructed a prototype shroud to cover theDavenport stock tubes, and this was found to be very effective inreducing noise levels (Exhibit C-43). Interestingly, Respondent declinedthe installation of noise enclosures, since it felt that such actionwould needlessly comply with the Act (Tr. 1555, 1556, Exhibits C- 43,C-45, C-46). Additionally, in 1972 the Respondent had begun orderingacoustical enclosures for screw machines, five of which were installedin 1974 (Tr. 1821, Exhibits C-18, C-78, p. 68, C-49, p.9). Respondentlater changed its supplier of acoustical noise enclosures to the GTSafety Equipment Company (Tr. 1844-1846, 2183, 2793). At the time of thehearing, approximately 24 screw machines were equipped with enclosuresin Department 75, producing significant noise reductions (Tr. 1843, 3124). We find that the acoustical enclosures are compatible with themanufacturing and maintenance evolutions carried on at Respondent’sworkplace. These enclosures are designed specifically for the type ofmachine used by Respondent and with appropriate venting are fullycompatible with all activities carried on at Respondent’s workplace.(Tr. 907, 909, 912, 1116, 1140, Exhibit C-23). Indeed, Respondent’s ownacoustical expert admitted that he had recommended acoustical enclosuresas one means of reducing noise levels to less than 10 decibels atanother facility having 54 Davenport screw machines (Tr. 4587, 44489,4592, 4593). Respondent’s expert witness further testified that awell-designed acoustical enclosure would reduce noise levels by at least20 decibels (Tr. 4575), and when enclosures were correctly used noiselevels in Department 75 would be below 90 dBA (Tr. 4436).As mentioned above, we do not find any reliance in the testimony of Mr.Prem Rai, who indicated that enclosures would have to be opened forapproximately 16 minutes per day, thus degrading their use subjectingemployees to exposure levels of 105 decibels for 32 minutes per day (Tr.3641, 3642, 4505, 4506, 4574, 4618-4620). This testimony was notconsistent with the testimony of the employees who indicated thatenclosures would be opened while the machine was running, once every 20minutes for approximately 15 seconds to check parts (Tr. 4719-4626), andonce or twice per hour for two or three seconds to make a tooladjustment (Tr. 4702). The testimony of the employees, which I findcredible, indicated that the total amount of time for which theenclosures would have to be open to check parts or make tool adjustmentwould be less than 16 minutes per day for both machines (Tr. 4719-4725).The conclusions of Respondent’s engineering expert that with the use ofthe enclosures employees would be exposed to excessive noise levels for38 minutes per day is not found credible since it was based oninaccurate or biased data (Tr. 4563).The overwhelming credible evidence established that noise enclosures area feasible engineering control available for Respondent’s screw machinesin Department 75. It is not difficult reaching this conclusion sinceRespondent has had experience with enclosing with enclosing screwmachines in its Johnson City, Tennessee plant, wherein the enclosure of14 machines had reduced noise levels to 92 decibels (Tr. 1775-1777).Finally, new Davenport screw machines were purchased, or planned forpurchase, with acoustical enclosures (Tr. 346, 347, 1983, 1984, C-78 p.85). When carefully viewing the record, it would be extremely difficultto say acoustical noise enclosures for screw machines in Respondent’sworkplace were not technologically feasible.We do not find the evidence submitted by the Respondent, at allpersuasive that there would be additional costs associated withenclosing the machines. We find that acoustical enclosures impose nodifficulty or additional cost of labor to employees who on the machines,and that the enclosures require little maintenance. We believe currentestimated cost of $5,000 per enclosure, plus an installation cost ofbetween $1,200-1,400 per machine to be an economically feasible and costeffective outlay of capital to achieve compliance with the NoiseStandard and Act (Tr. 1822, 1861, 1863, 1864). Significant noisereduction will be obtained and a significant number of employees willbenefit from the lowered noise levels. It is clear that the financialburden to Respondent, when compared with the abatement of the healthhazards to the employees as earlier described, makes these costsreasonable and the imposition of the engineering controls required underthe Act and the Noise Standard._Department 2_Department 2 contains approximately 70 presses, of which 41 aresecondary presses and 25 are automatic (Tr. 2616, 2824). Approximately49 employees are employed in this department as operators and setters(Tr. 2617). An automatic press operator’s duties include monitoring thetransportation of materials for the press, clearing parts from thepress, and stocking the press with new materials (Tr. 2618). A secondarypress operator’s duties include loading the part into press, activatingthe press and ensuring that the part clears the die area (Tr. 2619).Also employed in this Department are die setters and other maintenanceemployees. Department 2 operates two shifts. Each employee works fivedays per week, approximately five to eight employees work on the weekend(Tr. 2086). The parts produced in Department 2 are removed from thepress by either a blowoff mechanism, a mechanical ejection method, or insome instances the part drops out of the bottom of the die (Tr. 2084).The Secretary’s expert determined that the greatest source of noise inDepartment 2 was the air ejection system used to remove and transportparts from the dies (Exhibit C-23). This fact was known to StewartWarner (Exhibit C-55). Respondent, in August of 1973 and October of1979, performed studies which identified jet airblasts as the majorsource of excessive sound levels (Exhibits C-49, C-55). The Secretary’sexpert concluded that control of the air blasts could reduce employeeexposure by at least 5 dBA at cost of $25 to $40 per air nozzle (Tr.763, 765). To demonstrate the capability of noise suppression that couldbe achieved through use of air mufflers, the expert had performedseveral experiments wherein he insolated noise occurring from the jetairblasts; such experiments revealed that noise levels could drop from101.5 dBA to 93.8 dBA (Tr. 753, 754). The costs for controlling airnoise appears extremely reasonable in view of the results which could beobtained in noise reduction, and the number of employees who wouldbenefit therefrom.The air ejection system created noise in two distinct manners. The airejection system was continually on during the stamping process. Noisewas generated from the airjet, and from the air hitting the die onclosing. The noise emanating from these sources could be controlledthrough use of a valving method allowing the air blast to be only whenthe part in the machine was ready to expelled. When the machine wascycling the air valve would discontinue the flow of air, thuseliminating significant amounts of noise.A second method of noise control dealing with the air ejection systeminvolved the use of multiple air ejection nozzles to be substituted forone single nozzle for removal and transfer of materials from the die tothe to the part bin (Tr. 762, Exhibit C-23).The inclusion of air ejector mufflers was also found to be a significantnoise control device. Respondent’s own supervisory employees and otherwitnesses, who operated the presses recognized that the most efficientmeans of air noise reduction in Department 2 would be through the use ofmufflers on the air ejector system (Tr. 2621, Exhibits C-78, p. 60,C-67, C-12, p.3). It was estimated that installation of such mufflerscould provide greater air thrust requiring lower air pressure in thesystem. This could result in 76 percent less air consumption with asignificant energy cost savings (Exhibit C-11, p. 6).Air noise problems could be eliminated with the use of air mufflers, thevalving of continuously on air, and the use of a multiple air ejectionsystem. Offsetting the cost of these controls is the fact thatsignificant energy used in the current air ejection system could bereduced through the use of noise controls that require less air pressureenergy. This would result in significant energy savings to theRespondent. We find that the control of air noise, as above described,to be a feasible engineering control with advantageous economicovertones justifying its use. It is clear the benefits to employees bysignificantly lowered noise levels, through these noise controlsmethods, well justify their reasonable costs and are found to berequired of Respondent under the Noise Standard.An alternative method of noise control recommended by the Secretary’sexpert was the use of a mechanical ejection system (Tr. 760, ExhibitC-23). A mechanical ejection system would be so designed that as the diein the press separated a series of linkages which serve to automaticallyremove and expel the part from the die without resort to a noisy airejection system (Tr. 765). While the mechanical ejector systemrecommended by the Secretary may be a possible technologically feasiblecontrol, we find that the Secretary did not establish this fact by apreponderance of evidence. The proof supplied by the Secretary on thispoint amounted to nothing more than the general recommendations of itsexpert, providing no backup information demonstrating the feasibility,workability or _cost_ of this recommended control.The Secretary’s expert’s general conclusions are not sufficient toadequately carry the burden of proof necessary in on the issue of themechanical ejector’s technical feasibility. Indeed, the Secretary’sexpert was totally unaware of the size, shape weight, number of dies,finish requirements or tolerances needed to adequately demonstrate thatmechanical ejectors would be economically or technically feasible forinstallation (Tr. 1104, 1105). This information was absolutely essentialto such a determination (Tr. 4423, 4424). Accordingly, thisrecommendation is not found to be feasible under the proofs submitted.The last method of noise control recommended by Secretary’s expert dealtwith the partial and full enclosures of the punch presses (Tr. 1107,Exhibit R-4, Tr. 950, 951). It was conceded, however, that therecommended use of full and partial enclosures was not, with certainty,the most cost effective method of reducing employee exposure to noise(Tr. 951, 952). Additionally, this record is absent any evidence of astudy of work routines, employee operations, or machine operationsnecessary to establish or demonstrate that enclosures were feasible andwere compatible with the work processes and employee operations.The Respondent’s expert had concluded that a full enclosure for a manualpress was not technologically feasible (Tr. 4426). Throughout hisprofessional career he had never seen a manual press with a fullenclosure; this included visits to at least 250 plants (Tr. 4426). Ifind Respondent’s expert’s opinions and observations, that a fullenclosure for a manual press Is not feasible, persuasive and reliable(Tr. 4426). With respect to full enclosures for automatic machines,Respondent’s expert indicated that such a recommendation might beconsidered on a machine by machine basis, but that there wasinsufficient information developed to pursue any knowledgeable decision(Tr. 4428). Respondent’s expert found the same true with respect topartial enclosures, indicating that a determination of feasibilityrequired much more Information than presently was available in theengineering report of the Secretary’s acoustical engineer experttogether with his testimony (Tr. 4429, 4430). Again we agree withRespondent’s expert and accept these conclusions as controlling herein.We can only conclude that while we do not find partial or fullenclosures for automatic presses infeasible. there is insufficientevidence in this record to indicate that such controls are feasible, andas such we find that the Secretary has not carried his burden indemonstrating, through a preponderance of evidence, the feasibility ofthese proposed noise controls. Accordingly, such controls at this pointwould not be imposed upon the Respondent._Department 6_The manufacturing equipment located in Department 6 consists of 28 screwmachines, of which there are three different makes (Tr. 2199). As inDepartment 75, operators would run two machines during an eight-hour day(Tr. 2199). The automatic screw machine operator would set up andoperate the machines, performing all tasks necessary for their efficientoperation (Exhibit C-15). Basically, the machines located in Department6 operated generally the same as the Davenport machines in Department 75(Tr. 2200, 2201, 2203). As with the screw machines in Department 75enclosures were found to be an appropriate noise control device. Theenclosures for Department 6’s screw machine were off-the-shelf itemsrepresenting current technology. The enclosures were designed to betotally compatible with the manufacturing maintenance processes of thescrew machines. The Secretary’s expert calculated the cost of enclosuresas between $6,000-8,000 piece (Tr. 779). The lower cost would beindicative of an enclosure bought with the new machine while the $8,000figure represented the price for a retrofit enclosure purchased for amachine currently at the plant (Tr. 779). As with the Davenport machine,the manufacturer of the Acme-Gridly machine, a type used Department 6,specifically made screw machine acoustical enclosures at cost rangingfrom $4,150- $5,600 (Exhibit C-18, p. 7). Noise reductions of between12-18 decibels were available with this type enclosure (Exhibit C-18, p.8). Significantly, Industrial Sound Control, a manufacturer of screwmachine acoustical enclosures, had produced noise enclosures ranging incost from $2,500-$4,000. Such enclosures enabled customers to meet allOSHA requirements (Exhibit C-16, p. 15). The costs of enclosures appearreasonable when compared with amount of noise reduction obtained andnumber of employees affected.In September 1979, the National Acme Company informed Respondent thattotal acoustical enclosures capable of reducing sound levels toapproximately 85 dBA were available for its Acme-Gridly machines(Exhibits C-70, R-22, pp. 3, 6). While the Respondent indicated that itwould not be feasible to place noise enclosures on these machines sincethey would be retired soon, this argument is rejected since there was noindication of the length of the useful life of the machines and how longRespondent intended to use them (Tr. 2778, 2779). It appeared that thesemachines still could be used for an indefinite period of time (Tr.3602). Moreover, we found that Respondent made no representations thatany machines would be retired, or that new machines were on order, orthat there were plans to order machines (Tr. 3602, 3603, 3064).We find the Secretary’s expert quite persuasive with respect to screwmachine acoustical enclosure as a feasible engineering control.Furthermore, it is clear from the evidence that noise enclosures are astandard state of the art engineering noise control fully compatiblewith the screw machine’s operation and manufacturing process. As earliermentioned, we do not accept the calculations or conclusions of witnessMr. Rai concerning the nonfeasibility of these controls. We particularlyfind this evidence unreliable. In those instances where Respondent’sacoustical engineering expert’s conclusions were predicted on Mr. Rai’sstudy, we cannot find the same degree of persuasiveness we have foundwith some of his other opinions.We find the recommended noise enclosures for those screw machines thathave a useful life of at least three years from the date hereof to be atechnically and economically feasible control whose benefits toemployees health require that Respondent obtain and install same forDepartment 6.The Secretary’s expert also proposed for Department 6 the use ofacoustical tile of baffling for the walls and ceilings (Tr. 774). Withrespect to baffling, this is a commercially available device in therecommended size of 2 foot by 4 foot panels, ? inch in thickness, andhaving a plastic coating designed to resist oil and other air suspendedcontaminants. The Secretary’s expert estimated the installed cost of thebaffling at S1.00 to $1.50 per square foot (Tr. 775, 1111, Exhibit C-23,p. 3). The expert estimated that the noise reduction that could beachieved with this application would range between 2 and 5 dBA, theleast noise reduction, 2 dBA, occurring between two operating screwmachines that were not equipped with acoustical enclosures (ExhibitC-23, p. 3, Tr. 775, 776).The record shows that the Secretary’s expert spend a minimum amount oftime in Department 6 developing his conclusion that baffles and ceilingtiles were an appropriate engineering noise control. We find thiswitness not very familiar with the physical condition of Department 6.He did not know the specifics concerning the existing fire andsprinkling system located near the ceilings, the location of thelighting and electrical systems. nor was he familiar with other pipingfound in the Department. The various piping systems did have asignificant impact upon the feasibility of the acoustical bafflingrecommendation.Respondent’s acoustical engineering expert, on the other hand, performedan in-depth comprehensive sound measuring analysis of the workarea ofDepartment 6 (Tr. 4508- 4523). His skillful review of the acoustics ofDepartment 6 revealed to him that most operators of equipment wouldreceive no more than 1 to 2 decibels of noise reduction. Additionally,he found that the fire extinguishing piping, as well as the electricaland lighting piping and fixtures would significantly interfere with theplacement of noise baffles (Tr. 4649, 4537-4539). He further found thatceiling tile would not be as effective a noise suppressant as noisebaffles (Tr. 4534). Respondent’s expert estimated that the installedcost of baffling would be approximately $2.50 per square foot, for atotal cost of $180,000 (Tr. 4669, 4670).We are inclined to find the testimony of Respondent’s expert morepersuasive on the point of acoustical baffles and titles than theSecretary’s expert. Respondent’s expert conducted a much more in-depthanalysis of the acoustics of Department 6, and thus we find hisconclusions better placed. We find that, at this given time, without theinstallation of noise enclosures for the screw machines the use ofceiling tile or baffles is not a feasible engineering control. We reachthis conclusion for two reasons. First, looking to the economic aspectof proposed noise control we find the expenditure of approximately$180,000 to obtain only one or two decibels of noise reduction not areasonable economic trade off since the benefit in noise reductionexperienced by the employees does not balance reasonably with the costincurred. This is evidenced even more so with the use of ceiling tilesince it produces even less noise suppressant.Secondly, we find that in the case of baffles, as a matter of fact,their installation would be compatible with the fire sprinkler system orthe electrical lighting and piping in the workplace. To install thebaffling below the fire sprinkling system or the lighting system woulddegrade the fire fighting system and present a serious bumping hazard toemployees, in addition to interfering with movement of material andequipment through the workplace. The suggested use of sound absorptionfoam on the walls and ceilings, we find to be both a health and firehazard (Tr. 4532, 1115).Lastly, open airjets in Department 6 were found to be a significantnoise source. The installation of valving or mufflers on this noisesource was not disputed in these proceedings, and apparently wassomething that amounted to no more than required maintenance work. Bothexperts were in agreement that the noise source was something that waseasily controlled with appropriate mufflers, valving or maintenance (Tr.4673)._XI. CONCLUSIONS AND SUMMARY_We find the evidence clear, convincing and reliable that excessive noiselevels, as reported by the compliance officer, existed in the workplace.We further find that these excessive noise levels had a significantlyadverse effect on the hearing of those employees exposed. We findCircuit case law mandating that in weighing costs against benefits thebenefits to employees, in the reduction of noise levels throughengineering controls, must weight heavier in the balance than the costs.Accordingly, under the \”cost-benefit\” theory, feasible engineering noisecontrols was the mandated method of abatement of the noise hazard inRespondent’s workplace.In the instant case, we find that the engineering controls, foundtechnologically feasible, also came with a favorable economicfeasibility component (i.e., a reasonable balance between the cost ofthe controls, the amount of noise reduction obtained, and the number ofemployees affected) and that they provided to employees health benefitswhich could not otherwise inure to them through the use of personalprotective equipment, a means of abatement far less costly in terms ofdollars and cents.The evidence established that employees in the workplace were not faringsatisfactorily with the use of personal protective equipment, and thathearing loss or trends towards hearing loss, which employeesexperienced, could be arrested with the imposition upon the employer ofthe obligation to install the more costly engineering control methods.The Secretary had determined that certain engineering controls weretechnologically feasible and we agree in part with that determination.In those instances where we agree with the Secretary’s conclusions, wedeal with standard state of the art, uncomplicated but effective, noisecontrols. This case does not deal with exotic engineering devicescommanding significantly expensive materials and engineering services.The noise controls found feasible in this case are standardoff-the-shelf items having current competitive market prices.We find that acoustical noise enclosures for screw machines inDepartment 75 and Department 6 will assure significant reductions ofnoise levels to and below the levels mandated by the Noise Standard.These controls, when viewed from an economic standpoint, present a wellbalanced expenditure of funds when comparing costs and noise abatementproperties. Indeed, the Respondent had programmed for the purchase andinstallation of these controls, and in fact had some in place,demonstrating their effectiveness. Both parties’ experts in this caseconcurred that acoustical enclosures for screw machines were feasibleengineering controls for this Respondent (when each was fully andrealistically aware of the operations carried on by the operators of themachines). Accordingly, we find it to be the obligation of the employerto utilize and install acoustical noise enclosures on all its screwmachines, that have a useful life of three years from the date hereof,in Departments 6 and 75.We find that the control of air noise, a significant noise source inDepartment 2, a realistic and cost effective noise abatement method. Thecontinuous flow of high pressure of air, as a means of ejecting partsfrom punch press dies and sending them into bins, created high levels ofnoise. The installation of waiving, which would control the flow of airat the precise time of injecting the part. the use of mufflers, and theuse of multiple jets, represent a we’ll recognized state of the artnoise control. The cost for such valving and mufflers, again. is notexpensive since they are off-the-shelf items competitively priced.Additionally, their use would represent significant savings to theemployer by reason of the conservation of energy used for high pressure air.We find that the Secretary did not adequately preponderate on the issueof technological feasibility of enclosures, partial enclosures, ceilingtile and acoustical panels. The Secretary failed to provide adequatetechnical and economic data to make a case for their use. The same istrue with respect to mechanical ejectors; the record was totally absentsufficient technical and cost information or data to indicate that thiscontrol was indeed a feasible engineering noise control having areasonable balance in terms of noise suppression and costs._X. ABATEMENT DATES_In terms of abatement time, the evidence established that all the screwmachines can be enclosed within an 18 month period. Acousticalenclosures on screw machines are readily available from themanufacturers of the screw machines as well as other sources;installation time would be consistent with the 18 month abatementperiod. Installation of all controls associated with noise sources fromthe air ejection systems can be performed within six months sincecomponents needed are readily available from commercial sources andeasily obtained. Installation time is nominal._XI. NATURE OF VIOLATION_The condition alleged in the Citation is one that dates back asignificant number of years. The Respondent had been cited for theidentical violation in 1977. That Citation had been affirmed; indeed theRespondent had on several occasions sought abatement date modifications.The current citation cites the same violation, at the same location, asthe earlier 1977 Citation, and accordingly will be affirmed as a RepeatCitation, (see pp. 9 and 10, _supra_)._XII. PENALTY_The evidence has demonstrated that the hazard associated with noiseexposure is significant and serious (pp. 33 and 34, _supra_). We findemployees not doing well in the workplace where excessive noise levelsexist (pp. 38 and 39, _supra_). We find that this condition has existedfor a significant number of years, and that the condition was well knownby Respondent’s supervision.The evidence established that Respondent’s program of personalprotective equipment was the basic means of protection for employeesfrom the hazards associated with noise levels. Respondent’s monitoringof the employees’ success in obtaining realistic protection through useof personal protective equipment was found to be indifferent borderingon nonexistence. There essentially was no monitoring of the health ofits employees (pp. 42 to 44, _supra_). Respondent’s monitoring ofemployees hearing through use of audiometric examinations was a shamdesigned only to show ostensible compliance with the Act, and not at allused to monitor whether its program of personal protective equipmentprogram was working or was harmful to those employees who relied upon itfor protection from the excessive noise levels.We find that Respondent’s noncompliance with the Noise Standardrepresents a serious departure from its obligation to provide a safe andhealthful workplace for its employees. as required under the Act.Accordingly, we find the imposition of a penalty in the amount of $1,000to be appropriate after considering those factors in Section 17(j) ofthe Act._ORDER_Based upon the foregoing Findings of Fact and Conclusions of Law, andfor those reasons set out in my Decision and Order, and for good causeshown, it is ORDERED that1. Repeat Citation Number 1 is hereby AFFIRMED, with abatementrequirements and dates as setforth in my Decision and Order.2. A penalty of $1,000 is ASSESSED.Edward A. BobrickJudge, OSHRCDated: December 31, 1984Chicago, IllinoisFOOTNOTES:[[1\/]] Jurisdiction of the parties and the subject matter herein isconfirmed upon the Occupational Safety and Health Review Commission bySection 10(c) of the Act.[[2\/]] ? 1910.95 _Occupational Noise Exposure_.(a) Protection against the effects of noise exposure shall be providedwhen the sound levels exceed those shown in Table C-16 when measured onthe A scale of a standard sound level meter at slow response. . .(b)(1) When employees are subjected to sound [levels] exceeding thoselisted in Table G-16, feasible administrative or engineering controlsshall be utilized. If such controls fail to reduce sound levels withinthe levels of Table G-16, personal protective equipment shall beprovided and used to reduce sound levels within the levels of the table.(2) If the variations in noise level involve maxima at intervals of 1second or less, it is to be considered continuous.Table G-16 — Permissible Noise Exposures [[1\/]]Duration per day, hours \tSound leveldBA slow response8 \t906 \t924 \t953 \t972 \t1001? \t1021 \t105? \t110? or less \t115[[3\/]] When the daily noise exposure is composed of two or more periodsof noise exposure of different levels, their combined effect should beconsidered, rather than the individual effect of each. If the sum of thefollowing fractions: C_1 \/T_1 + C_2 \/T_2 [+. . .+] C_n \/T_n exceedsunity, then the mixed exposure should be considered to excess the limitvalue. C_n indicates the total time of exposure at a specified noiselevel and T_n indicates the total time of exposure permitted at that level.Exposure to impulsive or impact noise should not exceed 140 dB peaksound pressure level.[[4\/]] ? 1910.95 _Occupational noise exposure_.(a) Protection against the effects of noise exposure shall be providedwhen the sound levels exceed those shown in Table G-16 when measured onthe A scale of a standard sound level meter at slow response. . .[[5\/]] The A added to dB (decibel) indicates that the decibel of soundwas measured on the A scale of a standard sound meter at slow response,as required in 29 C.F.R. 1910.95(a). Additionally, decibels the basicunit of measurement of sound levels, are recorded on sound level metersaccording to several scales. On the A scale, the meter is more sensitiveto higher pitched tones than those of a lower pitch, just as the humanear is. The \”slow\” response is another setting of the instrument bywhich it averages out high level noises of brief duration (such ashammering), rather than responding to the individual impact noises. SeeU.S. Dep’t of Labor, Guidelines to the Department of Labor’sOccupational and Noise Standards p. 3 (1971). _Marshal v. West PointPepperil_, 588 F.2d 979 at 982 n. 5 (5th Cir. 1979).[[6\/]] \”Impulse [or impact] noise _shall_ be measured in the peak modeof a precision (Type 1) SLM [as opposed to the non-precision Type 2instrument used by OSHA in this case] set to flat or unweightedresponse. Where flat response is not available, the C weighted networkis used.\”[[7\/]] A term meaning a rate at which noise intensity or strengthdoubles with every 5 dBA increase in noise levels.[[8\/]] The Secretary’s expert Mr. George Kamperman was queried as tocertain statements made by him at an acoustical engineering conferencein Canada, to the effect that dosimeters do read higher than the actualsound levels (Tr. 1152, 1168-1170). While at times the testimony givenon this subject was confusing. I find that Mr. Kamperman adequatelyexplained the rationale behind these statements, as well as his laterstatements that the dosimeter would indeed give accurate readings byreason of the scientific formula contained in Table G-16 (Tr.1152-1154). In response to questions by the Judge to clear up theapparent conflict in his testimony, Mr. Kamperman explained that thedosimeter would take inconsistencies of sound into account when itautomatically calculated the G-16 formula on sound level measurementsover periods of time (Tr. 1154), and that the sound level metercomponent of the dosimeter would read noise accurately, and inparticular it would read and record accurately the type of noise foundin Respondent’s workplace (Tr. 747, 1096, 1152, 1153, 1154, 1156, 1157,1174, 1182, 1183).[[9\/]] Be it as it may, this ruling seemed inconsistent with theholdings that financial burdensomeness was not a defense for employersin occupational safety and health matters. _Industrial Union Department,AFL-CIO, et al. v Hodgson_, 499 F.2d 467 (D.C. Cir. 1974); _AmericanFederation of Labor, etc. v. Brennan_, 530 F.2d 109, at 122, 123 (C.A.3, 1975).[[10\/]] As will be discussed below, a program of personal protectionequipment (earplugs) will not provide noise hazard protection toemployees as advertised, expected or generally understood, anddefinitely not the same protection that is associated with engineeringnoise controls (pp. 35 to 39 infra).[[11\/]] The same is true with respect to the holding of the 5th, 6th,and 9th Circuit.[[12\/]] We believe the language of the Seventh Circuit is well worthrepeating as it plays a decisive role in bringing understanding to thelegitimate priorities within the cost-benefit theory in terms ofemployee protections; and in that it fits precisely into theseproceedings by reason of the facts of this case. The court held,\”the benefit to _employees_ from implementation of engineering noisecontrols will be substantial. The use of personal protective equipmenthas not prevented hearing loss and hearing difficulties among theseemployees. Implementation of engineering controls will bring _all_ 80employees within permissible noise exposure levels\” (at 989). (Emphasisadded)[[13\/]] \”…Moreover, earplugs and muffs have significant limitations.They are subject to employee resistance because they are uncomfortable.Not all employees can benefit from them. The Secretary put on experttestimony that noise reduction achieved by earplugs in actual use isless than half the attenuation achieved in laboratory tests. This hasbeen recently borne out by a NIOSH-CDC study to the same effect. HearingProtectors Field Measurement, _Morbidity and Mortality Weekly Report_,607 (Nov. 19, 1982).[[14\/]] While not covered in this record, earplugs worn throughout theworkday tend to naturally work free of the ear canal through normalmandibular movements in talking or chewing. This degrades the sealbetween the earplug and ear canal allowing noise to enter the inner ear.”