Citation: 51 Fed. Reg. 150 1986

Content downloaded/printed from HeinOnline (http://heinonline.org)Fri Feb 12 15:00:19 2016

-- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at http://heinonline.org/HOL/License

-- The search text of this PDF is generated from uncorrected OCR text.

150 Federal Register / Vol. 51, No. 1 /Thursday, January 2, 1980 / Rule*s and RegulationsEIOnMENTALPROTECTIO

ENVIRONMENTAL PROTECTIONAGENCY

40 CFR Part 60

IAD-FRL-2915-51

Standards of Performance for NewStationary Sources; Basic OxygenProcess Furnaces

AGENCY: Environmental ProtectionAgency (EPA).ACTION: Final rule.

SUMMARY: On January 20,1983,amendments to the standards ofperformance for primary emissions frombasic oxygen process furnaces {BOPF's)(40 CFR Part 60, Subpart N) wereproposed, together with standards ofperformance for secondary emissionsfrom basic oxygen process steelmakingfacilities (40 CFR Part 60, Subpart Na).This action promulgates theamendments to Subpart N, which areapplicable to any BOPF constructed,reconstructed, or modified after June 11,1973. This action also promulgatesSubpart Na, which is applicable to anytop-blown BOPF and to any hot metaltransfer station or skimming stationused for a bottom-blown or top-blownBOPF, for which construction,reconstruction, or modificationcommenced after January 20, 1983.

These standards implement sectionill of the Clean Air Act and are basedon a determination that iron and steelplants cause, or contribute significantlyto, air pollution that may reasonably beanticipated to endanger public health orwelfare. The intended effect of thesestandards is to require all new,modified, and reconstructed BOPF's tocontrol primary and secondaryemissions to the level achievablethrough use of the best demonstratedsystem of continuous emissionreduction, considering costs, nonairquality health and environmentalimpacts, and energy requirements.EFFECTIVE DATE: January 2, 1986.

Under section 307(b)(1) of the CleanAir Act, judicial review of the actionstaken by this notice is available only bythe filing of a petition for review in theU.S. Court of Appeals for the District ofColumbia Circuit within 60 days oftoday's publication of this rule. Undersection 307(b)(2) of the Clean Air Act,the requirements that are the subject oftoday's notice may not be challengedlater in civil or criminal proceedingsbrought by the EPA to enforce, theserequirements.ADDRESS: Background InformationDocument. The background informationdocument (BID) for the promulgated

standards may be obtained from theU.S. EPA Library (MD-35), ResearchTriangle Park, North Carolina 27711,telephone number (919) 541:-2777. Pleaserefer to "Basic Oxygen ProcessFurnaces-Background Information forPromulgated Standards" {EPA-450/3-82-005b). The BID contains: (1) Asummary of all the public commentsmade on the proposed amendedstandards along with responses to thecomments, (2) a summary of the changesmade to the standards since proposal,and (3) a final environmental impactstatement (EIS) for the final standards.

Docket. Docket number A-79-6,containing information considered indevelopment of the promulgatedstandards, is available for publicinspection between 8:00 a.m. and 4:00p.m., Monday through Friday, at EPA'sCentral Docket Section (LE-131), WestTower Lobby, Gallery 1, 401 M Street,S.W., Washington, D.C. 20460. Areasonable fee may be charged forcopying.FOR FURTHER INFORMATION CONTACT:.Mr. Doug Bell, Standards DevelopmentBranch, Emission Standards andEngineering Division (MD-13), U.S.Environmental Protection Agency,Research Triangle Park, North Carolina27711, telephone (919) 541-5624.SUPPLEMENTARY INFORMATION:

BackgroundThe new source performance

standards (NSPS's) for BOPF's werepromulgated on March 8, 1974 (39 FR9318). The standards of performancelimited mass emissions of particulatematter from both open hood and closedhood primary emission control systemsto no greater than 50 mg/dscm (0.022 gr/dscfQ. An opacity limit was promulgatedon. April 13, 1978, as a supplement to themass standard (43 FR 15602). Theopacity of exhaust gases from primaryemission control devices was limited toless than 10 percent, except that anopacity greater than 10 percent but lessthan 20 percent could occur once persteel production cycle.

In 1979, the Natural ResourcesDefense Council, Inc.; Friends of theEarth, Inc.; and the Group Against Smogand Pollution petitioned the UnitedStates Court of Appeals for the Districtof Columbia for the regulation of fugitiveor "secondary" emissions not capturedby the BOPF primary control system.Alftost simultaneously, the results of the4-year review of Subpart N were,announced, including the Agency'sintention to revise Subpart N to regulatesecondary emissions (44 FR 17460,March 21, 1979). Noting this action, the ,Court dismissed the suit and approved

the anticipated schedule for proposingand promulgating secondary emissionstandards. Amendments to Subpart Nwere proposed on January 20, 1983.

The proposed amendments werecontained in two Subparts of 40 CFRPart 60-Subparts Na and N. Asproposed, Subpart Na appliedexclusively to secondary emissions fromBOPF's (including top-blown andbottom-blown furnaces) and hot metaltransfer stations or skimming stationsfor which construction, modification, orreconstruction commenced after January20,1983. The proposed secondarystandards limited visible emissions fromthe BOPF shop roof monitor (or otherbuilding openings) to an opacity nogreater than 10 percent during theoperation of a top-blown BOPF, exceptthat an opacity greater than 10 percentbut less than 20 percent was permittedonce per steel production cycle. Visibleemissions from bottom-blown furnaceswere limited to an opacity no greaterthan 30 percent, except that an opacitygreater than 30 percent but less than 60percent was permitted twice per steelproduction cycle. The proposedstandards required thatcompliance withthe roof monitor visible emissionstandards be determined by ReferenceMethod 9, based on a 3-minute average.Mass emissions from any device usedsolely to collect secondary emissionswere limited to 23 mg/dscm (0.010 gr/dscf}. An opacity limit of 5 percent forsecondary emission collection deviceswas also proposed. Under the proposedstandards, roof-mounted electrostaticprecipitators (RMESP's) used to controlsecondary emissions were exempt fromthe mass standard and the opacity limitfor secondary emission collectiondevices, as were any devices used tocollect both primary and secondaryemissions.

Subpart N contains standards ofperformance that regulate primaryemissions of particulate matter fromBOPF's constructed, reconstructed, ormodified after June 11, 1973. On January20, 1983, amendments also wereproposed for Subpart N BOPF'sconstructed, reconstructed, or modifiedbetween June 11, 1973, and January 20,1983. These proposed amendments didnot revise the existing emission limit of50 mg/dscm (0.022 grldscf) for massemissions of particulate matter from theprimary emission control device or theexisting opacity limit for visibleemissions exiting the primary emissioncontrol device. The existing standardlimits visible emissions exiting theprimary emission control device to anopacity of less than 10 percent, exceptthat an opacity greater than 10 percent

Federal Register / Vol. 51, No.1 / Thursday, 'janiary "2,'1986, / Rules and.:Reglations 151

but less than 20 percent may occur onceper steel production cycle. Additionally,no changes were proposed to theexisting monitoring or reportingrequirements. The proposedamendments did revise the test methodand procedure institutions for samplingof mass emissions by Reference *Method5. The existing standard required asampling rate of at least 0.9 dscm/hr(0.53 dscf/min); the proposedamendments required a minimumsample volume of at least 0.9 dscm (32'dscf). Except for this sample volumerequirement, no other changes to theexisting procedures for determining'compliance were proposed for theseSubpart N BOPF's.

Also on January 20, 1983, amendments,to Subpart N were proposed to regulateBOPF's constructed, reconstructed, ormodified after January 20, 1983. Theseproposed standards adjusted the massemission limits and the test methodsand procedures for determiningcompliance. For BOPF's with a closedhood primary emission control system,the proposed amendments limited massemissions of particulate matter to nomore than 68 mg/dscm (0.030 gr/dscf),as measured for the primary oxygenblow. For BOPF's with an open hoodprimary emission control system, massemissions of particulate matter werelimited to no more than 50 mg/dscm(0.022 gr/dscf), as measured for theprimary oxygeh blow. Even though ashorter testing period was proposed,these mass emission limits werecomparable to the existing standard.

Data presented in Table 4-12 of theBID suggest 1hat a lower concentrationlimit for open hood furnaces may bepossible. These data show performancelevels better than performancemeasured during testing to support theexisting NSPS limit of 50 mg/dscm (0.022gr/dscf). Determination of theequipment design, operating criteria,and process variables responsible forproducing lower emissionconcentrations from the open hoodfurnaces will be examined during thenext 4-year NSPS review forconsideration of a change in theconcentration limit.

Primary emission control systemperfdrmance data for one plant in Table4-9 of the BID suggest that operation ofventuri scrubbers at high pressure drops,approaching 90 inches of water, couldachieve lower emission concentrationsthan those required by the existing orproposed rule (-60 inches). It is truethat pressure drop is a major factor;however, other factors such as gas flowrate, throat design, and liquid-to-gasratio will also affect outlet

concentration. A further factor to beconsidered is that increased pressuredrop requires differently sized6quipment to apply more scrubbingenergy. A cost comparison for onemodel system converted from 60-inch to80-inch pressure drop showedincremental cost effectiveness of $0,000per ton of additional particulateremoved. On a cost-effectiveness basis,it appears that a reduction in requiredemission concentration based on higherscrubber pressure drop observed in oneplant is inappropriate.

The proposed amendments to the testmethods and procedures for sampling ofmass emissions by Reference Method 5specified that sampling for each runmust continue for an integral number ofprimary oxygen blows with a totalduration of at least 60 minutes. Aminimum sample volume requirement of0.9 dscm (32 dscf) also was proposed.

A proposed definition for "primaryoxygen blow" also was added to§ 60.141. The proposed definitiondefined "primary oxygen blow" as theperiod in the steel production cycle of aBOPF during which a high volume ofoxygen-rich gas is introduced to the bathof molten iron by means of a lanceinserted from the top of the vessel orthrough tuyeres in the bottom or thebottom and sides of the vessel. Reblowswere excluded from the proposeddefinition, as was the introduction ofnitrogen through tuyer~s in the bottomor bottom and sides of the vessel.

No changes were proposed to theexisting opacity standard for primaryemission control devices. Under theexisting standard, visible emissionsfrom control devices for open or closedhood primary control systems werelimited to no more than 10 percentopacity, except that an opacity greaterthan 10 percent but less than 20 percentcould occur once per steel production-cycle. No changes to the test methodsand procedures for determiningcompliance with the opacity limit wereproposed.

The Final StandardsThe final secondary emission

standards (Subpart Na) apply to anynew, modified, or reconstructed top-blown BOPF and to any new, modified,or reconstructed hot metal transferstation or skimming station used with abottom-blown or a top-blown BOPF, forwhich construction commences afterJanuary 20,1983. Visible emissions fromshop roof monitors (or other buildingopenings) are limited to an opacity nogreater than 10 percent during the steelproduction cycle of a top-blown BOPF,and during hot metal transfer andskimming for a bottom-blown BOPF

except that an opacity greater than 10percent but less than 20 percent mayoccur once per steel production cycle.Visible emissions from the shop roofmonitor during the furnace cycle (i.e.,the steel production cycle excluding hotmetal transfer and skimming) of abottom-blown BOPF are not subject tothe roof monitor opacity standard. Asproposed, Reference Method 9, withdata reduction procedures for 3-minuteaverages, will be used to determinecompliance with the secondary emissionopacity standards.

The numerical emission limit reflectsthe performance of an open hoodprimary system used also for secondaryemission control, in addition to localhooding for hot metal transfer andskimming emissions, which isconsidered best demonstratedtechnology (BDT) for the capture ofsecondary emissions from the affectedfacilities. Preliminary investigationindicates that this limit also could bemet by using fume suppression systemsto control hot metal transfer emissions,in addition to other controls onsecondary emissions. Fume suppressionappears to achieve emission reductionsat costs significantly below those of theBDT on which the standards are based.The Administrator does not considerfume suppression systems to beadequately demonstrated to serve as thebasis for the standards. However, theAdministrator encourages sources todevelop fume suppression systems thatcan meet the requirements of thestandards. Section 111(j) of the CleanAir Act authorizes the Administrator togrant innovative technology waivers forsuch purposes..

To ensure the collection of secondaryemissions from affected facilities, thefinal standards limit mass emissionsfrom devices used solely for thecollection of secondary emissions to 23mg/dscm (0.01 gr/dscf). This numericallimit is based on the performance of abaghouse that is considered BDT'forsecondary emission collection. Anopacity limit of 5 percent for thesecondary emission collection devicealso is included to ensure properoperation and maintenance of theequipment. These emission limits alsoapply only to top-blown BOPF's and tohot metal transfer stations or skimmingstations used with a bottom-blown or a

'top-blown BOPF.Under the promulgated standards,

fume suppression systems are exemptfrom the mass and opacity limits forsecondary emission collection devices,as are devices used for the collection ofprimary and secondary emissions, EPAhad proposed to exempt RMESP's from

152 Federal Register / Vol. 51, No. 1 / Thursday, January 2, 198& / Rules and Regulations

the limits for a secondary controldevice, based on a judgment that anRMESP that met only the roof monitor'opacity standard "would most likely beas good or better than BDT" (48 FR 2667,col. 3). EPA now believes that such ajudgment is premature. In particular.such an RMESP might have a greaterflow volume, and thus greater emissions,than BDT. EPA is therefore notpromulgating any such exemption.However, RMESP's are acceptable asemission control devices if they meetboth the mass and opacity standards.because testing of RMESP's may beimpractical, the Agency will considerwaiving the mass. test provided that theRMESP meeis the opacity standard andis found to be properly engineered andinstalled.

The monitoring requirements in thepromulgated standards have beenrevised since proposal to require adevice (or devices) for the continualmonitoring and recording of exhaustventilation rates, or levels of exhaustventilation, for each duct of thesecondary emission control systemduring each phase of each steelproduction. cycle. The device (ordevices) must be placed at a location (orlocations) near each capture point of thesecondary emission control system or in,an alternative location (or locations)approved in advance by theAdministrator. New provisions alsohave been added for the operation of astrip chart recorder, should thisequipment be used as a recordingmechanism.

A new provision also has been addedto the final standards that requires thesemiannual reporting of allmeasurements (as indicated by themonitoring device) over any 3-hourperiod that average more than 10percent below the average levelsmaintained during the most recentcompliance test for mass emissions fromthe secondary emission collectiondevices. The accuracy of themeasurements may be considered whenmeasurement results are reported. Also,when a scrubber primary emissioncontrol device is, used to collectsecondary emissions, the promulgatedstandards require the continualmonitoring and recording of scrubberpressure drop during each phase of eachfurnace production cycle. Again, allmeasurements over any 3-hour periodthat average more than 10 percent belowthe average levels maintained during themost recent performance. test must bereported semiannually.

Data currently are not available toestablish a separate mass standardspecifically for secondary emissions

collected by a scrubber primaryemission control device. Prior to thenext 4-year review of Subparts N andNa, the Agency will collect and evaluatetest data that may become availabledocumenting the performance ofprimary emission, control devices duringsecondary operations.

Section 6.144a, which specifies testmethods and procedures, also has beenrevised since proposal. Instructions for*aggregating visible emissions frommultiple building openings160.144a(b)(1)] have been deleted fromthe final rule. Sample volumerequirements have been increased fromthe proposed level of 2.27 dscm. (80 dscf)to a minimum of 5.67 dscm (200 dscf) foreach run. A new provision has beenadded to the final rules that allowssmaller sample volumes, subject toapproval by the Administrator, whennecessitated by process variables orother factors. Proposed instructions fordetermining compliance by ReferenceMethod 2 for devices that monitorexhaust ventilation rates have beenrevised to require 12 pairs of readingsfor each duct of the secondary emissioncapture system. Comparableinstructions for determining complianceby Reference Method 2 also have beenadded for devices that monitor thelevels of exhaust ventilation and recordonly step changes when a set point isreached.

Compliance provisions of § 60.145ahave been revised since proposal toallow an owner or operator of a BOPFshop that normally operates twofurnaces with overlapping cycles to shutdown one furnace during compliancetesting for both mass and visibleemission standards. A new provisionhas been added to § 60.145a thatrequires the owner or operator tooperate the furnace being tested atexhaust ventilation rates or levels foreach duct of the secondary emissioncontrol system that are appropriate forsingle-furnace operation. Following thecompliance test, the owner or operatormust operate the secondary emissioncontrol system at exhaust ventilationrates or levels (for each duct of thesystem) that are no lower than 90percent of the values established duringthe most recent compliance test. Aspreviously noted, all measurements (asindicated by the monitoring device) thataverage more than 10 percent below theaverage levels maintained during themost recent compliance test for massemissions from the secondary emissioncollection device must be included in asemiannual report. Although the reportwould not be considered evidence of aviolation of the emission limit, it could

indicate possible improper operation ormaintenance of the collection device.

In the final rule, the language of§ 60.145a has been changed to requirethat visible emission observations forboth hot metal transfer and skimmingemissions begin with the startup ofoperation and terminate 3' minutes aftercompletion of the operation. A newprovision has been added that requiresall visible emission observations to beidentified and recorded in conjunctionwith the starting and stopping times ofregulated operations in the steelproduction cycle. Instructions fordetermining compliance with the 5-percent opacity standard and the 0.010-gr/dscf mass standard for secondaryemission collection devices also havebeen added to § 60.145a.

The visible emissions observationlimit proposed for hot metal transfer andskimming was based on roof monitoropacity observations in a top-blownfurnace shop, as is the'final rule. Somedifficulties were encountered in the datacollection process for hot. metal transferand skimming sources. An. overlapping

,equipment operating schedule causedemissions from more than one vessel tobe mixed, resulting in higher opacitiesthan would have been measured for asingle vessel. As a result, the opacitystandard for these sources may behigher than necessary. In ourengineering judgment, however, thisdata collection difficulty does not impairthe achievability of the standardbecause there: are no significant processdifferences between hot metal transferand skimming in bottom-blown furnace'shops as opposed. to top-blown furnaceshops. Likewise, equipment used toperform these processes is notnecessarily different in bottom-blownversus top-blown shops. As part of the4-year NSPS review process, EPA willcollect more data for hot metal transferand skimming sources and will revisethe standard if the new data indicatesuch a revision is appropriate.

Other c rifying changes to SubpartNa since proposal include severalrevisions to definitions. The definition of"hot metal transfer" has been changedto include all transfer operations, andthe definition of "secondary emissions"has been revised to include specificallyhot metal transfer and skimmingemissions. Also included in thisdefinition are particulate matteremissions that escape from openings inthe primary emission control system,such as lance hole openings, or fromgaps or tears in the ductwork of theprimary emission control system or fromleaks in hoods. Ih addition, thedefinition of "'steel production cycle"

Federal Register / Vol. 'f," No. '1 / Thutisd'dy,- January 2, 1986 J "Rules and Regulations

has been expanded to includepreheating (when used) and vesselturnup, as well as turndown, duringsampling operations. New definitionsalso have been added for "primaryemissions," "primary emission controlsystem," "secondary emission controlsystem," and "fume suppressionsystem." Finally, the definition of"startup" has been deleted because adefinition is now included in theGeneral Provisions for 40 CFR Part 60.

A number of changes have been madesince proposal to the amendments to theexisting standard for primary emissions.The title of the existing standard hasbeen revised to read "Subpart N-Standards of Performance for PrimaryEmissions from Basic Oxygen ProcessFurnaces for Which Construction isCommenced After June 11, 1973." Thelanguage and format of the existingstandard also have been clarified toindicate clearly the applicability ofrequirements for BOPF's constructed,reconstructed, or modified after June 11,1973, but on or before January 20, 1983,and for BOPF's constructed,reconstructed, or modified after January20, 1983.

No changes to the existing mass andvisible emission limits for the primaryemission control device have been madesince proposal for Subpart N BOPF'sconstructed, reconstructed, or modifiedafter June 11, 1973, but on or beforeJanuary 20,1983. Except for anincreased sample volume requirement,Subpart N BOPF's will continue todetermine compliance with these limitsas prescribed by the existing standards.However, the proposed minimumsample volume requirement of 0.9 dscm(32 dsco specified in § 60.144(b)(1) formass sampling by Reference Method 5has been increased to a minimum of 1.5dscm (53 dscf). It should be noted thatthe minimum sample volume of 9.9 dscm(32 dscf) proposed under subparagraph -(b)(1) was the result of a FederalRegister misprint; a minimum samplevolume of 0.9 dscm (32 dscfl wasintended.

Since proposal on January 20, 1983,EPA has become aware of a differencein furnace cycle time between top-blownand bottom-blown furnaces that mayimpact measured emissionconcentrations for those facilitiesregulated under Subpart N. In general,bottom-blown furnaces have shorteroxygen blowing periods than top-blownfurnaces. The testing provisions forthose BOPF's constructed,reconstructed, or modified after June 11,1973, but on or before January 20, 1983,allow testing from the beginning of theoxygen blow, or scrap preheat if

practiced, until just prior to tapping. Thebulk of the particulate emissions occurduring the oxygen blow, however. Witha shorter oxygen blowing period androughly the same length non-blowingperiod as top-blown furnaces, thebottom-blown furnaces are all6wed agreater proportion of test time duringnon-blowing periods, thus tending todilute the measured particulateconcentration more than is allowed fortop-blown furnaces. EPA will studywhether different sampling times wouldbetter reflect BDT and intends topropose revise sample periodrequirements and mass standards, asmay be appropriate for top-bottom andbottom-blown furnaces under Subpart Nas part of the 4-year NSPS review cycle.

The final rules continue themonitoring requirements of the existingprimary standards for Subpart N BOPF'swith certain minor changes. Under§ 60.143(b)(2), a monitoring device "isrequired for the continuousmeasurement of the water supplypressure to the control equipment; thepressure sensor or tap for the devicemust be located close to the waterdischarge poinL The existing standardsthet* state that the Administrator may beconsulted for approval of alternativelocations for the pressure sensor or tap.Under the final rules, alternativelocations for the pressure sensor or tapmust be approved in advance by theAdministrator. The word "continuous"also wa' changed to "continually" toavoid confusion with General Provisionrequirements relating to continuousmonitoring systems, as defined in 40CFR 60.2.

Additionally, § 60.143 was amendedto reduce the frequency of reportingrequirements from quarterly tosemiannually. Under the final rules,each owner or operator must reportsemiannually any measurements (asindicated by the monitoring device) overany 3-hour period that average morethan 10 percent below the average levelsmaintained during the most recentcompliance test for mass emissions. Theaccuracy of these measurements may beconsidered when measurement resultsare reported. Although reporting of thesemeasurements would not be consideredevidence of noncompliance with the -

standards, the values reported mayindicate a potential operation ormaintenance problem with the devicethat necessitates a complianceinspection, particularly if a repeatedpattern is reported.

A number of changes also were madeto the definitions for terms applicable toSubpart N BOPF's. The final rulesinclude a revised definition of "steel

production cycle" for Subpart N BOPF'sconstructed, reconstructed, or modifiedafter June 11, 1973, but on or beforeJanuary 20,1983. For these BOPF's, thecurrent definition of "steel productioncycle" has been expanded to includesampling (vessel turndown and turnup)operations. A new definition also wasadded for "primary emissions."

"Primary emissions" means particulatematter emissions from the BOPFgenerated during the steel productioncycle and captured by the primaryemission control system. Also, thedefinition of "startup" was deleted fromthe final rule because a definition is nowincluded in the General Provisions of 40CFR Part 60. The proposed definition of"basic oxygen process furnace" has notbeen revised since proposal-

The final amendments to Subpart Nfor BOPF's tonstructed, reconstructed,or modified after January 20, 1983,continue the proposed requirementswith few exceptions. No changes to theadjusted limits for mass emissions fromopen or closed hood primary controlsystems have been made since proposal.The final rules require that after January20, 1983, particulate emissions from anynew, modified, or reconstructed BOPFwith an open hood primary system belimited to 50 mg/dscm (0.022 gr/dscf), asmeasured for the primary blow. For anyBOPF constructed, modified, orreconstructed after January 20, 1983,particulate emissions from any BOPF forwhich closed hooding is the primarycontrol method are limited to 68 mg/dscm (0.030 gr/dscf), as measured forthe primary blow. Although thesampling period (i.e., the primary blow)is different than the existing standard,the level of control required by therevised standards is comparable to thelevel of control required under theexisting standard. With the exception ofan increased sample volumerequirement, no changes to theReference Method 5 test methods andprocedures for determining compliancewith the adjusted mass emission limitshave been made since proposal. Theproposed minimum'sample volumerequirement of 0.9 dscm (32 dscfo hasbeen increased to a minimum of 1.5dscm (53 dscf). As proposed, shortersampling times and smaller samplevolumes are permitted whennecessitated by process variables orother factors, subject to approval by theAdministrator.

A new operating requirementpertaining to the operation of theprimary emission control system duringreblows has been added since proposal.BOPF's required to meet a massemission limit based on the primary

153

154 Federal Register / Vol, 51, No. 1 / Thursday; January 2, 1986 / Rules and Regulations

oxygen blow are required to operate thegas cleaning device during any reblowin a manner identical to operationduring the primary oxygen blow. Thisrequirement applies during compliancetesting and during routine operationfollowing any compliance test. Thisprovision was added to ensure thatmass emissions generated during areblow are controlled at the same levelas mass emissions occurring during theprimary oxygen blow. Under typicaloperating conditions, no significantincrease in electrical powerconsumption for'the primary gascleaning system would occur due to thisrequirement.

No changes have been made to theexisting opacity standard for theprimary control device. The existingstandard limits visible emissions fromcontrol devices for open or closed hoodprimar'y control systems to no more than10 percent opacity, except that anopacity greater than 10 percent but lessthan 20 percent may occur once per steelproduction cycle. Compliance with theopacity standard would be determinedwith the use of Reference Method 9, asprescribed under the existing standards.

The final amendments includechanges in definitions of termsapplicable to BOPF's constructed,reconstructed, or modified after January20, 1983. Although the definition of"primary oxygen blow" has not beenrevised since proposal, the definition of"steel production cycle" has beenexpanded to include samplingoperations (vessel turndown and turnup)and deslagging operations.

BOPF's constructed, reconstructed, ormodified after January 20, 1983 also aresubject to the amended monitoring andreporting requirements. As discussedpreviously, § 60.143 has been clarified torequire advance approval by theAdministrator for alternative locationsfor the monitoring device pressuresensor or tap. The final standards alsoreduce the frequency of reportingrequirements from quarterly tosemiannually.

Summary of Environmental, Energy, andEconomic Impacts

The impacts of the secondaryemission standards have not changedsince proposal. Withdrawal of the roofmonitor opacity standard for bottom-blown furnaces would not affect thenationwide environmental, energy, oreconomic impacts because the impactsof the standards are based onprojections for increased capacity forexisting top-blown furnaces. Minorrevisions to the estimatedenvironmental, energy, and economicimpacts are included in Docket No. A-

79-6 as item IV-B-7. These changescontain a revised discussion of thewater pollution impacts of RegulatoryAlternative III when wet or semiwetelectrostatic precipitators (ESP's) areused in lieu of other primary controlsystems. These comments also indicatethat the gas cooling and conditioningsystem used with an ESP must bedesigned as a total evaporation systembecause the steel industry effluentregulation (40 CFR Part 420, May 27,1982) specifies no discharge from thesesystems. With these changes, theanalysis of nationwide environmental,energy, and economic impacts inVolume I of the BID is now consideredthe final EIS for the promulgatedstandards.

Briefly, the standards would reducesecondary particulate emissions fromBOPF facilities for which construction isexpected to commence during the period1981 through 1986 by about 2,527 tonsper year, assuming the use of open hoodcontrol. If closed hood controls wereused, secondary emissions would bereduced by about 2,718 tons per year. Noadverse water, solid waste, or noiseimpacts would result from theimplementation of the secondaryemission standards, although the levelof solid waste from BOPF facilitieswould increase by about 2 percent withopen hood controls and by about 2.5percent with closed hood controls, dueto the control of secondary emissions.

The cumulative capital costassociated with the, secondary emissionstandards through the first 5 yearswould be about $18.2 million, assumingthe use of open hood controls. Theannualized cost would be about $5.6million per year. If closed hooding wereused as a means of compliance, thecapital costs through the first 5 yearswould be about $29.4 million, whileannualized costs would be about $3.8million per year. In each case, the priceof finished steel would increase byabout 0.2 percent by the end of the fifthyear.

Further analysis of secondaryemission control technologies indicatesthat the cost effectiveness of the controltechnologies that can be used to complywith the final standards ranges from$1,665 to $3,727 per ton of particulatematter removed for furnace emissionsand from $1,755 to $3,100 per ton ofparticulate matter removed for hot metaltransfer stations. In addition, newBOPF's may be able to employ lowercost, innovative hot metal transfercontrols. However, the aggregated costeffectiveness of $2,245 per ton ofparticulate matter for secondaryemission controls has not been revisedsince proposal.

Electrical energy requirements forBOPF control .systems that commenceconstruction during the period 1981through 1986 would increase by about 60percent, or 22.8 million kWh per year,assuming the use of open hood control.

Public Participation

During development of the proposedstandards, trade associations, plantpersonnel, equipment manufacturersand vendors, environmental groups, andother interested parties suppliedinformation and data for input onvarious aspects of the standards. Thestandards recommended for proposalwere discussed at meetings of theNational Air Pollution ControlTechniques Advisory Committee(NAPCTAC) held December 2, 1980, andSeptember 22-23, 1981. These meetingswere open to the public and eachattendee was given an opportunity tocomment on the recommendedstandards. The standards wereproposed January 20, 1983. A publichearing on the proposed standards wasnot held because it was not requestedby commenters or other interestedparties. The public comment periodextended from January 20,1983, to April5, 1983. A total of five public commentswere received. All written commentshave been considered and, whereappropriate, changes to the proposalhave been incorporated in the finalstandards.

Significant Comments and Changes tothe Proposed Amendments-

Comments on the proposedamendments and revisions werereceived from four industryrepresentatives and one individual withexpertise in BOPF emission testing. Adetailed discussion of these commentsand responses can be found in the BIDfor the promulgated standardsreferenced in the ADDRESSES sectionof this preamble. The summary ofcomments and responses in the BIDserves as the basis for the changes thathave been made to the proposedstandards. The major comments andresponses are summarized in thispreamble under the following headings:Reference Method 9 and the 3-MinuteAverage, Best Demonstrated Technologyfor Bottom-Blown Furnaces,Concentration Standard for SecondaryEmission Collection Devices, Roof-Mounted Electrostatic Precipitators, andFume Suppression Systems.

Reference Method 9 and the 3-MinuteAverage

The majority of the industryrepresentatives' comments reflect their

Federal Register / Vol. 51, No. 1 / Thursdfiy, January 2, 1986 / Rules and Regulations

concern regarding the use of a 3-minutedata reduction procedure for analysis ofvisible emission data by ReferenceMethod 9. The commenters noted thatMethod 9 requires that an opacity valueconsist of the average of 24 consecutivevisible emission observations taken at15-second intervals and that theproposed standards alter thisrequirement to produce an opacity valuebased on an average of 12 consecutivevisible emission observations.

Two commenters contended that thepublic was not afforded sufficientopportunity to comment on the proposedchanges. One commenter furtherasserted that the proposed changes arenot supported by any analysis of theeffects of the changes on the scientificreliability of Method 9. This commenterconcluded that the changes constitute adhoc rulemaking and noted that suchprocedures have been invalidated byDonner Hanna Coke Corp. v. Castle, 464F. Supp. 1295 (W.D.N.Y. 1979). Thecommenter believes that "The 'DonnerHanna' Court decision clearlydemonstrates that the test method usedto evaluate compliance with a regulationis as important as the regulation itself.Merely specifying the test method is notenough when the accuracy and precisionof the method are unknown in theprescribed application."

The commenters also indicated thatthe proposed 3-minute averagecontradicts previously articulated policyabout how Method 9 should be used. Insupport of this, one commenter notedthat "In previous instances, EPA hassuggested that there is no known basisfor altering the procedures andmethodology established under Method9. For example, on November 24, 1982,while reconsidering the Illinois Stateimplementation plan (SIP), theAdministrator stated:

There is no means, using Method 9, toaccount for plumes less than 6 minutes induration (noncontinuous). There is also nomeans, using the averaging techniques ofMethod 9. to account for exemption periodsother than 6 minutes or for aggregation of anyduration. (47 FR 5300, 5302).The second commenter noted that "Inthe past, EPA has recognized that[Method 91 cannot be used forintermittent non-stack sources."

In responding to these comments, wefirst examined the question of whetherthe public has had opportunity tocomment on the proposed changes. Weconclude that both the general publicand the steelmaking facilities withBOPF's have had ample opportunity toparticipate in and comment on theproposed changes. Prior to proposal,representatives of public interest groups

and steel companies were providedseveral opportunities to comment on the3-minute average. These opportunitiesare detailed in the response to comment2.5.1 in the BID for the final standards.At proposal,-the preamble to theregulations discussed testing and datareduction procedures used to establishthe visible emission standards and theprocedures that would be followed todetermine compliance with the visibleemission standards (48 FR 2662).Comments received on these proceduresare being taken into account in thesefinal standards.

In addition to the information providedto the public before proposal, ananalysis of the effects of the proposedchanges on the scientific reliability ofMethod 9 was included in the docketthat accompanied the proposed rule (1i-B-92). The information in this analysisdemonstrated that the precision andaccuracy of the proposed changes to theMethod 9 data reduction procedure forvisible emission observations of BOPF'sare equivalent to the precision andaccuracy in the current Method. (Thisanalysis is discussed in detail in the.BID.) Moreover, no new data. orinformation on the visible emissionstandards that contradicts the findingsof our analysis was included with thepublic comments. Therefore, the publichas been provided sufficient opportunityto comment on the proposedperformance testing and data analysisprocedures.

We note that in citing the Illiriois SIPreconsideration, the commentersuggested that statements about the useof Method 9, in its current form, shouldlimit how the Method might be modified.This suggestion is, however, withoutmerit. Rather, the validity ofmodifications to standards and testmethods (and the relationship betweenthese) set forth in 40 CFR Part 60 mustbe evaluated and must be supported ontheir own merits.

As to the question'of whetherReference Method 9 can be used forintermittent nonstack sources, it shouldbe noted that since its promulgationDecember 23, 1971 (36 FR 24895),Method 9 has been amended to observevisible emissions from both controldevice exhaust stacks and nonstacksources (39 FR 39872, November 12,1974). (See, for example, subparts AAand BB.) Indeed, the Method, atparagraph 2.1, specifically includesprocedures to be used in determiningvisible emissions from nonstack sourcessuch as roof monitors. Often, bothcontrol devices and nonstack sourcessuch as roof monitors intermittently emitplumes of varying opacity; i.e., visibleemissions vary with process or control

device operation and are observed onlyduring part of the time the emissionsource is operating. Also, emissions areobserved only when they exceed thevisibility threshold level of about 2percent opacity. Reference Method 9'canbe applied accurately to both categoriesof emission points by following theprocedures of paragraph 2.1 and byrecording visible emission observationsof zero percent opacity when no visibleemissions are evident.

The passage from Method 9 cited bythe Court in Donner Hanna does notstate that Method 9 is inappropriate forcharacterizing visible emissions from allpoints of intermittent emissions:EPA recognizes that certain types of opacityviolations that are intermittent in naturerequire a different approach in applying'theopacity standards .... (39 FR 39873)

The passage cited by the Court indicatesthat Method 9 (with the 6-minuteaveraging technique) may notcharacterize the performance of captureor control technologies adequately incertain cases. This can be illustrated bythe hypothetical example of a sourcethat exhibits zero opacity for 23 out of 24readings and 25 percent on the 24threading. On a 6-minute average basis,the opacity would be calculated as 1percent. On a 15-second basis, opacityWould simply be 23 periods of zero andI period of 25 percent. It is obvious thatthis latter method of presenting the datais much more descriptive of the emissioncharacteristics in this case than is the 6-minute 1-percent opacity average. In thebroader context, this hypotheticalexample illustrates how shorteraveraging periods more accuratelyreflect the character of short-durationplumes. (It also shows how generalopacity standards that are commonlyused in SIP's do not account for plumesof short duration.) For this reason, ashorter averaging time and a'highernumerical visible emission standardwere proposed for BOPF's. A 6-minuteaveraging period could have beenselected to analyze BOPF visibleemission data. If this option had beenselected and implemented, thenumerical level of the standard wouldhave been about one-half the numericallevel proposed in conjunction with a 3-minute averaging period. However,BOPF secondary emissions are typicallyof short duration, and a 3-minuteaveraging period more closelycharacterizes the performance of thecapture systems used to control theemissions than a 6-minute averagingperiod.

Finally, it is agreed that the testmethod used to determine compliance

155I

156 Federal Register / Vol. 51,,No. I / Thursday, January 2, 1986 / Rules and Regulations

with a standard is very important inestablishing the stringency or effect of astandard. For that reason, the basis forthe standard, including not only the database but also both the test method usedto develop the data base and the testmethod used for compliance, must beconsidered in establishing a standard.

In conclusion, it is noteworthy thatthis proposed rulemaking does not affectMethod 9 data reduction procedures orapplication of Method 9 to all sourcesregulated in 40 CFR Part 60. Rather, itonly establishes procedures for BOPF'sthat commence construction,modification, or reconstruction afterJanuary 20, 1983. Use of a 3-minuteaverage was proposed in paragraph 2.5,"Test Methods and Procedures" inSubpart Na, rather than as amendmentsto Reference Method 9. This was donebecause 3-minute average opacityvalues are considered appropriate forand are intended to be-applicable onlyfor BOPF shop emissions. In this way,Method 9 as it applies to other sourcesis not affected by this rulemaking. Thisapproach is consistent with the intendeduse of the test methods in Appendix A(40 CFR Part 60). The intent is detailedin the introduction to Appendix A,which states, in part:

Within each standard of performance, asection titled 'Test Methods and Procedures'is provided to (1) identify the test methodsapplicable to the facility subject to therespective standard and (2) identify anyspecial instructions or conditions to befollowed when applying a method to therespective facility. Such instructions ... areto be used either in addition to, or as asubstitute for, procedures in a referencemethod.Therefore, it is appropriate to alter-after analysis and public comment-thedata reduction procedures for BOPFvisible emission observations within theprovisions of Subpart Na;

Several commenters also questionedthe scientific reliability of the proposed3-minute averages for Reference Method9 opacity computations of visibleemissions from BOPF's. The commenterscontended that the modified averagingprocedure will produce average opacityvalues that are less accurate and moresubject to error than the 6-minuteopacity values. The commenters statedthat the proposed secondary standardsrequire the improper use of ReferenceMethod 9 for observing roof monitorvisible emissions because the methodwas originally promulgated forobservation of stacks, with specifieddata reduction procedures for 6-minuteaverages. The commenters stated thatquantitative conclusions regarding theaccuracy of the method were valid onlyfor emissions from siacks. One

commenter also maintained thatmodifications to Method 9 should beevaluated thoroughly to determine theaccuracy and precision of anydeviations and recommended thatMethod 9, unmodified, be used for theproposed regulation or that the proposedroof monitor opacity standards bedeleted until an appropriate method isdeveloped.

We share the commenters' concernsthat knowledge of the precision andaccuracy of a test method-andconsideration of these factors for thepublic record-is important indeveloping and enforcing standards.Prior to responding to these concerns, itis appropriate to review the generalprocedure followed to establish astandard of performance and tounderstand the role of the test method toestablish and determine compliancewith the standard. Typically, a standardis expressd as a numerical emissionlimit that quantitifies the performance ofBDT for emission control. A data base isgathered to establish an emission limitthat is achievable for the emissionsource being regulated. The data base isobtained with either an existing or newtest method that has been devised forthe pollutant and source being regulated.If the method used to develop the database differs from the method that wouldbe used to determine compliance, themathematical relationship betweenthese methods must be known. In eithercase, the test method is proposed andpromulgated according to theprocedures outlined in section 307(d) ofthe Clean Air Act as amended is foundeither in Appendix A or in the subpartsof 40 CFR Part 60. In the proposed andfinal standards, the accuracy andprecision of the test method aredocumented and considered carefully.

When the visible emission standardswere developed for BOPF's, a largenumber of visible emission observationswere gathered according to theprocedures of Reference Method 9.These data, which consist of more than100 hours of observations, indicate thatplumes from the roof monitors of BOPFshops are of short duration and that theperformance of BDT for controllingthese emissions is best characterizedwith an average opacity value based ona shorter, 3-minute averaging periodinstead of the 6-minute averaging periodspecified in the data reductionprocedures contained in paragraph 2.5 ofthe reference method.

Therefore, a visible emission standardwas proposed that would be based on.the BOPF visible emission data baseand on average opacity valuescalculated from 12 consecutive 15-second visible emission observations

recorded by following the procedures ofReference Method 9. The proposed andfinal performance test procedures arealso based on observing and recordingvisible emissions with ReferenceMethod 9 and on calculating a 3-minuteopacity average from the visibleemission data.

Before 3-mihute rather than 6-minuteaverages were proposed, the question ofwhether 3-minute averages are lessaccurate or precise than 6-minuteaverages was addressed. (Note that theaccuracy and precision of 6-minuteaverages have been established and arenot an issue here.) This involvedanalyzing and comparing the frequencyof occurrence of differences betweenMethod 9 observations (observer mean)and mean values calculated fromtransmissometer readings (instrumentmean) for both 3- and 6-minuteaveraging periods, This analysis, whichwas included in the docket prior toproposal (II-B-92), is based on a July1976 report that also was included in thedocket prior to proposal (II-A-22). Theresults of this analysis are summarizedin Table 2-3 of the BID for the finalstandards.

The results are reported as thepercentage of the total number ofmeasurements that fall within aparticular range of observer error. Forexample, 36.5 percent of the totalnumber of 3-minute averages calculatedfor the smoke generator black smokehad an observer error of between 0 and5 percent opacity. Note that onlypositive errors are reported in the table.Examination of this table reveals thatthe distribution of errors for 3-minuteaveraging is about the same as for 8-minute averaging (e.g., for generatorblack smoke, 36.5 percent of thecaluclated 3-minute averages had anobserver error of 0 to 5 percent opacityas compared to 35.3 percent of thecalculated -minute averages). Thus, theaverage opacities calculated on a 3-minute basis are no more subject toerror than the average opacitiescalculated on a 6-minute basis.

With regard to the applicability ofMethod 9 to plumes from both stacksand roof monitors, a review of opacitytheory demonstrates that quantitativeconclusions about the accuracy andprecision are equally valid for each. Themajor factors influencing plume opacityare: particle characteristics (particle sizedistribution and refractive index),particulate concentration, thebackground against which the emissionsare viewed, the observer's positionrelative to the sun, and the light pathlength through the emission plume.Particle characteristics and particulate

Federal Register / Vol. 51, No. 1 I Thursday, Janiary 2, 1986'/ Rules and Regulafions

concentration are determined by theprocess operation and the emissioncontrol technology. In the steel industry,plumes are released at elevated points,such as stacks and roof monitors. As aresult, the background for reading theopacity of visible emissions is the same.for both types of sources, generallyconsisting of sky, horizon, or otherstructures. Thus, the ability to readopacity of visible emissions from roofmonitors compared to the ability to readvisible emissions from stacks is notinfluenced by'background.

While the geometry of stacks and roofmonitors does differ, Reference Method9 explicitly accounts for opacityreadings from roof monitors. Stacks aregenerally circular, and, as a result, ihepath length through the plume isessentially the same in all directions.Roof monitors tend to be rectangularwith a long and short dimension. Thelight path length through the plume isdifferent depending on whether theobserver sights along the long dimensionor the short dimension. When theopacity of visible emissions from roofmonitors is read, paragraph 2.1 ofReference Method 9 specifically requiresthat observations be takenapproximately perpendicular to the longdimension of the roof monitor (i.e.,across the short dimension of theplume), which ensures that observedopacity is minimized. When the visibleemission standards for BOPF's weredeveloped, visible emissionobservations were taken from existingfurnace shops with typical roof monitordesigns. We have no data or otherinformation that indicates that futureroof monitor designs would be modifiedto cause plume path length, and thusobserved opacity, to' increase. Thus, theeffect of path length on opacity wastaken into account during developmentof the standards, and paragraph 2.1 ofReference Method 9 ensures thatcompliance with the standards isdetermined by reading plumes acrossthe shorter path length.

The position of the visible emissionobserver with respect to the sun and theplume does affect perceived opacity dueto the light-scattering effects of plumes.However, this light-scatteringinterference is nullified as the observeris positioned with his back to the sun asdescribed in paragraph 2.1 of theReference Method. This paragraphprovides explicit instructions thatposition the observer with respect toboth stacks andirectangular openingssuch as roof monitors. Thus, the effect ofposition on opacity is taken into accountduring development of visible emissionstandards.

The ability to read the opacity ofvisible emissions, therefore, does notdepend on whether these emissions arereleased from stacks or roof monitors.Reference Method 9 is as applicable toplumes from roof monitorsas it is toplumes from stacks.

After proposal of Subpart Na,statistical analyses of visible emissionobservations recently taken from roofmonitors and stacks were performed tovalidate this conclusion further. Theseanalyses are summarized in amemorandum entitled "Opacity Error forDifferent Averaging Times" (IV-B-6),which reports the results of statisticalanalyses of visible emissions fromfugitive emission sources. The fugitiveemissioi data were obtained mostlyfrom iron and steel sources with BOPFshops and roof monitors being majorcontributors to the data base.

One statistical analysis examined theprecision of observations of the opacityof a fugitive emission plume madesimultaneously by two visible emissionobservers. For average opacity valuescalculated with a 6-minute averagingperiod, the standard deviation was 2.1percent opacity with 93 percent of theruns having a difference betweenobserver readings less than or equal to7.5 percent opacity. For average opacityvalues calculated with a 3-minuteaveraging period, the standard deviationwas essentially the same-2.4 percentopacity with 92 percent of the runshaving a difference between observerreadings less than or equal to 7.5 percentopacity. This analysis of visible •emission observations of fugitiveemissions therefore supports theconclusion that between-observer -precision is the same for averageopabity values calculated with 3- and 6-minute averaging periods.

As shown in Table 2-4 of the BID forthe final standards, the stack standarddevations range from 4.3 to 9.8 percentfor 3- and 6-minute averages comparedto the 2.1 percent and 2.4 percentreported above for the fugitive sources.Thus, it is conlcuded that the variability

* between observers in reading opacityfrom roof monitors is similar to thatbetween observers reading opacity fromstacks.

In summary, the 3-minute average wasselected because it better characterizesthe brief duration of BOPF visible -emissions. We conclude that the publicwas afforded a sufficient opportunity tocomment. We also conclude thatReference Method 9 is valid andscientifically reliable for application toBOPF shop roof monitors. In addition,the scientific reliability of the 3-minuteaverage, as compared to the 6-minute

average, was considered and analyzedfor public review prior to proposal of thestandards. Further analysis confirmsthat no significant difference existsbetween the accuracy and precision of3-minute averages and the accuracy andprecision of 6-minute averages.Consequently, no changes were made inthe final rules regarding The use of a 3-'minute data reduction procedure fordetermining compliance with visibleemission standards by ReferenceMethod 9.Best Demonstrated Technology forBottom-Blown Furnaces

One commenter criticized theselection of the proposed roof monitoropacity limit for bottom-blown furnacesin that the level of the proposedstandards does not reflect BDT. Insupport of his contention, thecommenter pointed to the failure of thesystem to achieve 90 percent capture ona consistent basis during the emissiontesting program due to mechanicalproblems (e.g., bell damper failures).The commenter also alleged that thesecondary emission control system waspoorly designed and maintained. As analternative to selecting the Republicsystem as BDT for controlling secondaryemissions from bottom-blown furnaces,the commenter recommended that a newregulatory alternative be considered forbottom-blown furnace control. Thesuggested alternative would consist of afurnace enclosure with hot metaladdition and tapping hoods exhaustedto a collection device capable ofhandling 600,000 acfm or the gasvolumes handled by the control systemsfor top-blown furnaces, as demonstratedat the Bethlehem and J&L plants.

The opacity limit proposed for bottom-blown vessels was consistent with datafor bottom-blown furnaces that wereavailable at proposal. The effect of thecommenter's recommendations wouldbe to establish an emission limit at alevel that has not been observed inpractice for boftom-blown furnaces. it isagreed that the performance of controlson other sources, such as top-blownfurnaces, suggests that betterperformance should be achievable forbottom-blownfurnaces and raises doubtas to whether the proposed limits reflectthe performance of BDT. The commenteralso correctly implied that caputre andcontrol of emissions from these vesselshas not been demonstrated at a levelconsistent with top-blown vessels. Amore stringent limit based on transfer oftechnology was not consideredwarranted in the case of bottom-blownvessels because of the wide range ofconditions over which the vessels

157 ,

158 Federal .gster./oVol. 51,. No. a / Thursday; January 2, 1986 / Rules and Regulations

operate and the correspondingly widerange of emissions from the vessels.Based on this comment and furtherreview, it has been determined thatbetter control will likely becomeavailable before the next 4-year reviewof the primary and secondary standardsand that the-proposed standards, ifpromulgated, would probably not reflectBDT* In particular, it should be notedthat revised State regulations applicableto existing bottom-blown BOPF's thathave been adopted pursuant to therequirements of 40 CFR Part 52 are muchmore stringent than the proposedstandards. Because of these findings, theproposed roof-monitor opacity limit forbottom-blown vessels is beingwithdrawn from the final rules.

The requirement for limiting theopacity of visible emissions produced byhot metal transfer and skimming inbottom-blown furnace shops, however,is being retained. There are nosignificant process differences betweenhot metal transfer and skimming inbottom-blown furnace shops as opposedto top-blown furnace shops. Likewise,equipment used to perform theseprocesses are not necessarily differentin bottom-blown versus top-blownshops. Therefore, it is EPA's judgmentthat there'is no significant difference inemission potential, either controlled oruncontrolled. For the above reasons, theroof monitor opacity limit for hot metaltransfer and skimming in top-blownfurnace shops is extended to apply tohot metal transfer and skimming inbottom-blown furnace shops. Theprovisions of Subpart Na, § 60.145a,allow the owner or operator of anaffected facility to suspend shopoperations not subject to Subpart Naduring compliance testing. Theseprovisions provide a means of avoidinginterference from bottom-blown furnaceemissions when compliance testing of.hot metal transfer and skimmingoperations is conducted. By similarreasoning, the proposed concentrationand opacity standards for secondaryemission collection devices in bottom-blown furnace shops also are retained.

Withdrawal of the roof monitoropacity standard does require thedevelopment and consideration of arevised Regulatory Alternative I, thealternative selected as the basis of theproposed standards. With the deletionof the bottom-blown furnace roofmonitor opacity standard, RegulatoryAlternative II consists of BDT captureand collection controls for secondaryemissions from top-blown furnaces, inaddition to BDT capture and collectioncontrols for hot metal transfer stationsand skimming stations used with

bottom-blown or top-blown furnaces.Withdrawal of the roof monitor opacitystandards for bottom-blown furnaceswould not affect the environmental,energy, or economic impacts estimatedfor the proposed standards. No impactwould result because industry growthforecasts for the period 1981 through1985 (upon which the impacts werebased) project only the expansion ofixisting top-blown furnace capacity.Consequently, Regulatory Alternative 11remains the alternative selected as thebasis of the promulgated standards.

Concentration Standard for SecondaryEmission Collection Device

One commenter criticized the level ofthe proposed mass emission limit of 23mg/dscm (0.010 gr/dscf) for secondaryemission collection devices, assertingthat the standard did not reflect theperformance level achievable forbaghouses, which is BDT for thisapplication. The commenter argued thata properly designed, operated, andmaintained baghouse controllingemissions from hot metal transfer,skimming, and other operations canachieve an outlet emission rate of 0.005gr/dscf.

The numerical emission limit of 0.010gr/dscf proposed for secondaryemission collection devices is based onthe data available at the time ofproposal. The data base for the emissionlimit, as proposed, includes data fromhot metal transfer, partial buildingevacuation for BOPF's, and full-cycledata for electric arc furnaces (EAF's).Other operations that generatesecondary BOPF emissions are notincluded in the data base. While thecommenter correctly stated that certainoperations can be controlled to achievean emission rate of 0.005 gr/dscf, controlof all pertinent operations has not beendemonstrated adequately for BOPF's atthat level.

The data supporting the proposedstandard do not support a lower limit. Itis agreed that new data, coupled withthe transfer of technology from EAFs,might support the lower limitrecommended by the commenter.However, without the benefit of publiccomment and because it is not clearwhether other data reflect technologydifferent from the data on which the0.010-gr/dscf limit is based, the finalrules do not revise the standard forsecondary emission collection devices.Revision of the standard to reflect thehigher performance level will beconsidered during the next 4-yearreview of Subparts N and Na.

Roof-Mounted Electrostatic Precipitators

One commenter questioned theconclusion that performance of RMESP'smay be equivalent to or better than BDTfor secondary emission collection whenno data were available to assess RMESPperformance. The commenter alsoquestioned why a Japanese RMESP wasnot evaluated to support the proposedstandard. This commenter pointed outthat a baghouse can handle secondaryemissions containing hard-to-precipitatekish particles and significant amounts offine particulate emitted in large burstsmore efficiently than can an RMESP.Although the commenter supported theproposed RMESP waiver for innovativetechnology, he suggested that theconclusion that the RMESP may beequivalent to (or better than) BDT at alower cost should be verified.

Section 111(j) of the Clean Air Actprovides that an owner or operator of anaffected facility may request a waiverfrom one or more requirements of thestandards to encourage the use of theinnovative control system. For an owneror operator to obtain a waiver, theremust be findings that the technology hasnot been demonstrated adequately andthat it has either a substantial likelihoodof achieving greater emission reductionthan required by the NSPS or ofachieving equivalent emission reductionat lower cost (including energy andnonair environmental costs). Once awaiver has been issued, its terms mustinclude: (1) Assurance that the sourceemissions will not prevent attainmentand maintenance of national ambient airquality standards (NAAQS's); (2)

* assurance that the technology willfunction properly; (3) a time limit fortesting the technology not to exceed 7years from issuance or 4 years fromstartup, or until the technology provesunworkable; and (4) a restriction to thatportion of the source on which thetechnology is used.

The commenter is correct in notingthat no data were available to quantifythe performance, costs, and otherimpacts associated with the use ofRMESP's. However, as indicated in theproposed rulemaking, there werequalitative indications that RMESP'scould be at least as efficient as, andmore cost effective than, capture hoodsducted to a particulate collection device,which is the technololgy on which thenumerical standard is based. Some ofthis qualitative information wasobtained from RMESP's in Japan.-Unfortunately, the Japanese companiesusing the RMESP's would not permit thequantitative evaluation of their controltechnologies.

Federal Register / Vol. 51* No. 1 / Thursday,'Jaruary 2, 1986 [Rules and Regulations9

EPA had proposed to exemptRMESP's from the limits for a secondarycontrol device, based on a judgment thatan RMESP that met only the roofmonitor opacity standard "would mostlikely be as good or better than BDT" (48FR 2667, col. 3). EPA now believes thatsuch a judgment is premature. Inparticular, such as RMESP might have agreater flow volume, and thus greateremissions, than BDT. EPA is thereforenot promulgating any such exemption.

EPA recognizes that it may be difficultto test compliance with the massemission limit in § 60.142a(a)(1) (23 mg/dscm) using Method 5 as prescribed by§ 60.144a(a](4), when an RMESP is used.If such cases arise, source owners oroperators may be able to demonstrate toEPA's satisfaction by other means thatthe source is in compliance with thestandard, under 40 CFR 60.8(a)(4).Means of demonstrating compliancemight include opacity observations,observations of plume volume, limitedmass test data, engineering evaluationof control systemeffectiveness, or somecombination thereof.

Under the terms of any waivergranted, provisions would be included toensure'that any RMESP installed tocomply with the NSPS would bedesigned and operated properly. Datacollected during the term of the waiverwould be used to verify RMESPperformance and costs. Should theperformance and costs be demonstratedas equivalent to or better than BDT, theinclusion of provisions specific toRMESP's may be considered during thenext 4-year review of Subpart Na.

Fume Suppression SystemsTwo commenters pointed out that,

while the proposed standards do notrequire the use of a specific technologyto meet the proposed emission limits, asteel producer could experiencedifficulties in permit approval if theproducer desired to implement atechnology not examined or discussed inthe preamble or BID for the proposed orpromulgated rule. In this regard, bothcommenters pointed to fumesuppression, a relatively newtechnology, which could provide thepotential for eliminating or reducing theneed for particulate capture techniques.One commenter stated that the cost toimplement and operate this technologycan be substantially less than for othercontrol alternatives, with fumesuppression providing an equivalentperformance. Accordingly, thecommentefs recommended that thistechnology be examined as a viablealternative control technique.

Data regarding fume suppression werenot available when Volume I of the BID

was being prepared. Since proposal,.however, new information and datahave been obtained to document theperformance and cost of this system.These data indicate the fumesuppression systems may provideperformance equivalent to that of othercontrol alternatives at a lower cost.

Suppression techniques reduce ironoxide fuming above molten metal or slagbaths by inhibiting oxidation of the ironat the surface of the bath. Oxidation isinhibited by blanketing the bath surfacewith flame, steam, wet sand, or wet slag.To date, these techniques have beenapplied experimentally to: (1) Blastfurnace iron tapping, (2) blast furnaceslag tapping, (3) open hearth tapping, (4)electric furnace tapping, (5) hot metaltransfer, (6) BOPF charging, (7) BOPFtapping, and (8)'BOPF lance holeemissions.

Detailed test data to quantify theperformance of suppression techniqueson facilities regulated by 40 CFR Part 60,Subpart N or Na, are not yet available.However, preliminary data quantifyingthe performance of flaine suppression (atype of fume suppression) on a BOPF hotmetal transfer station at U.S. Steel'sGary plant indicate that opacity isreduced significantly with the use offlame suppression. These data areprovided in Table 2-2 of the BID for thefinal standards. Thus, it appears thatflame suppression is capable ofachieving emission reduction at leastequal to requirements of the proposedregulation during hot metal transfer.

Moreover, because fume suppressioneliminates the need for hot metaltransfer capture and collection devices,emission reductions may be achieved atcosts significantly below those of theBDT on which the standards are based.The cost effectiveness (dollars per ton of,particulate matter removed) of the fumesuppression technique applied tosecondary emissions from hot metaltransfer operations is estimated at $633/ton. This can be compared to a costeffectiveness of $3,100/ton for abaghouse collecting only hot metaltransfer emissions. Based onconsideration of the preliminaryperformance data and costs of fumesuppression, an owner or operator maychoose to install this system to controlhot metal transfer emissions. Thestandards of performance for hot metaltransfer stations are expressed asnumerical emission limits and do notpreclude the use of any controltechnology, so long as it can achievecompliance with the numerical limit.

Additionally, an innovativetechnology waiver may be granted to anowner or operator desiring to install afume suppression system. Under the

terms of an innovative technologywaiver, the owner or operator would begranted an extended time period to meetthe roof monitor visible emissionstandards. Because fume suppressionsystems eliminate the need foradditional collection equipment, theyare exempt from the mass and opacitystandards for secondary emissioncollection devices. Provisions would beincluded in any waiver t~o ensure that afume suppression system installed tomeet the NSPS would be disigned andoperated properly. Data collected duringthe term of the waiv'er would be used todocument the performance and costs offume suppression for particulateemissions control. Should theperformance and costs be equivalent toor better than BDT, the inclusion ofprovisions specific to fume suppresionsystems would be considered during thenext 4-year review of Supart Na.

The final standards also contain adefinition for fume suppression systems.A fume suppression system is defined as"the equipment comprising any systemused to inhibit the generation ofemissions from steelmaking facilities bymeans of an inert gas, flame, or steamblanket applied to the surface of molteniron or steel."

Information Requirement Impacts

The final standards for primary andsecondary emissions (Subparts N andNa) require no reports in addition tothose required under the GeneralProvisions of 40 CFR Part 60. TheGeneral Provisions contain notificationrequirements, which enable the Agencyto keep abreast of facilities subject tothe standards; contain requirements forthe conduct and reporting of initialperformance tests; and require reports ofexcess emissions. Analysis of thesereporting requirements indicates theyare both necessary and reasonable,considering the savings in time andresources required for effectiveenforcement. In the absence of thesereporting requirements, effectiveenforcement of the regulation wouldrequire frequent individual inspectionsand tests.

The Paperwork Reduction Act (PRA)of 1980 (Pub. L. 96-511) requiresclearance from the Office ofManagement and Budget (OMB) ofcertain public reporting andrecordkeeping requirements beforepromulgation of this rulemaking.Information collection requirementsassociated with this regulation (thoseincluded in 40 CFR Part 60, Subpart A,Subpart N, and Subpart Na) have beenapproved by OMB under the provisionsof the PRA of 1980, 44 U.S.C. 3501 et seq.

0159

160 Federal Register / Vol., 51,, No. 1 / Thursday, January 2, 1986 / Rules and Regulations

and have been assigned OMB controlnumber 2060-.0029.

Regulatory Flexibility AnalysisThe Regulatory Flexibility Act of 1980

requires the identification of potentiallyadverse impacts of Federal regulationsupon small business entities. The Actspecifically requires the completion of aregulatory flexibility analysis in thoseinstances where small business impactsare possible. Determination of the needto perform a regulatory flexibilityanalysis is based upon the considerationof three factors: (1) The maximum sizeof a small business; (2) the number ofsmall businesses affected;,and (3) theexpected economic impacts. Thesestandards are not subject to theRegulatory Flexibility Act of 1980because no small businesses will beaffected. No impacts on smallgovernments or small organizations areanticipated because they also will notbe affected. This determination isdiscussed in the preamble and BID forthe proposed standards.

DocketThe docket is an organized and

complete file of all the informationconsidered in the development of thisrulemaking. The docket is a dynamic filebecause material is added throughoutthe rulemaking development. The docketsystem is intended to allow members ofthe public and industries involved in therulemaking to identify and locatedocuments readily so they canparticipate effectively in the rulemakingprocess. Along with the-statement ofbasis and purpose of the proposed andpromulgated standards and responses tosignificant comments, the-contents ofthe docket will serve as the record incase of judicial review, except forinteragency review materials [section'307(d}[7)(Af].Miscellaneous

The effective date of this regulation isJanuary 2,1988. Section 111 of the CleanAir Act provides that standards ofperformance or revisions thereofbecome effective upon promulgation andapply to affected facilities, constructionor modification of which wascommenced after the date of proposal.

The promulgation of these standardswas preceded by a review of thestandards of performance for BOPF's (40CFR Part 60, Subpart N), which wascompleted in 1979 (44 FR 17460, March21, 1979). The review of the primarystandard resulted in the conclusion thatsecondary emissions from BOPF'srepresent a major air pollution source.Clarifying revisions to the primarystandard were also recommended. In

addition, publication of thesepromulgated standards was preceded byconsultation with appropriate advisorycommittees, independent experts, andFederal departments and agencies, inaccordance with section 117.

This regulation will be reviewel 4years from the date of promulgation asrequired by the Clean Air Act. Thisreview will include an assessment ofsuch factors as the need for integrationwith other programs, the existence ofalternative methods, enforceability,improvements in emission controltechnology, and reporting requirements.

Section 317 of the Clean Air Actrequires the Administrator to prepare aneconomic impact assessment of"revisions (of new source performancestandards) which the Administratordetermines to be substantial . . ."[section 317(a)]. An economic analysisof the standard was prepared for theproposed rulemaking. ["StandardsSupport and Environmental ImpactStatement, Volume 1: ProposedStandards of Performance for BasicOxygen Process Furnances" (EPA-450/3-82--005a)]. The nationwide economicimpacts and the aggregated costeffectivenes of the secondary emissioncontrol standards have not been revisedsince proposal.

Under Executive Order 12291, aregulation considered "major" is subjectto the requirement of a RegulatoryImpact Analysis. This regulation is not"major" because: (1) The nationalannualized compliance costs, includingcapital charges resulting from thestandards, total less than $100 million;(2) the amended standards do not causea major increase in prices or productioncosts: and (3) the standards do not causesignificant adverse effects on domesticcompetition, employment, investment,productivity, innovation, or competitionin foreign markets.

This regulation was submitted toOMB for review as required byExecutive Order 12291. Any commentsfrom OMB and any response to. thosecomments are included in Docket A-79-6. The docket is available for publicinspection at EPA's Central DocketSection, West Tower Lobby, Gallery 1,Waterside Mall, 401 M Street, SW.,Washington, DC 20460.

Pursuant to the provisions of 5 U.S.C.605(b), I hereby certify that the proposedrule will not have a significant economicimpact on a substantial number of smallentities.List of Subjects in 40 CFR Part 60

Air pollution control,Intergovernmental relations, Reportingand recordkeeping requirements,

Incorporation by reference, Basicoxygen process furnaces.

Dated: December 22. 1985.-Lee M. Thomas,Administrator

PART 60-[AMENDED]

40 CFR Part 60 is amended as follows:1. The authority citation for 40 CFR -

Part 60 continues to read as follows:Authority: Secs. 101,111, 114, 116, 301,

Clean Air Act as amended (42 U.S.C. 7401,7411, 7414, 741. 7601).

2. The title of 40 CFR Part 60, SubpartN, is revised to read as follows:

Subpart N--Standards of Performancefor Primary Emissions from BasicOxygen Process Furnances for WhichConstruction Is Commenced AfterJune 11, 1973

3. In § 60.141, paragraph (a) is revised,existing paragraphs (b) and (c) aredeleted, and new paragraphs (b), (c) and(d) are added to read as follows:

§ 60.141 Definitlons.

(a) "Basic oxygen process furnace"(BOPF) means any furnace with arefractory lining in which molten steel isproduced by charging scrap metal,molten iron, and flux materials or alloy"additions into.a vessel and introducing ahigh volume of oxygen-rich gas. Openhearth, blast, and reverberatoryfurnaces are not included in thisdefinition.

(b) "Primary emissions" meansparticulate matter emissions from theBOPF generated during the steelproduction cycle and captured by theBOPF primary control system.

(c) "Primary oxygen blow" means theperiod in the steel production cycle of aBOPF during which a high volume ofoxygen-rich gas is introduced to the bathof molten iron by means of a lanceinserted from the top of the vessel orthrough tuyeres in the bottom or throughthe bottom and sides of the vessel. Thisdefinition does not include anyadditional or secondary oxygen blowsmade after the primary blow or theintroduction of nitrogen or other inertgas through tuyeres in the bottom orbottom and. sides of the vessel.

(d) "Steel production cycle" means theoperations conducted within the BOPFsteelmaking facility that are required toproduce each batch of steel and includesthe following operations: scrap charging,preheating (when used), hot metalcharging, primary oxygen blowing,sampling (vessel turndown and turnup),additional oxygen blowing (when used),

Federal Register / Vol. 51, No. 1 / Thdrsday, January 2, 1986 / Rules and Regulations 161

tapping, and deslagging. This definitionapplies to an affected facilityconstructed, modified,.or reconstructedafter January 20, 1983. For an affectedfacility constructed, modified, orreconstructed after June 11, 1973, but onor before January 20,1983, "steelproduction cycle" means the operationsconducted within the BOPF steelmakingfacility that are required to produce'each batch of steel and includes thefollowing-operations: scrap charging,preheating (when used), hot metalcharging, primary oxygen blowing,sampling (vessel turndown and turnup),additional oxygen blowing (when used),and tapping.

4. In § 60.142, the introductory text ofparagraph (a) is revised and paragraphs(b) and (c) are added to read as follows:

§ 60.142 Standard for particulate matter.(a) Except as provided under

paragraph (b) of this section, on andafter the date on which the performancetest required to be conducted by § 60.8is completed, no owner or operator.subject to the provisions of this subpartshall discharge or cause the dischargeinto the atmosphere from any affectedfacility any gases which:

(1) * * *(2) * * *(b) For affected facilities constructed,

modified, or reconstructed after January20,1983, the following limits shall apply:

(1) On or after, the date on which theperformance test under § 60.8 is requiredto be completed, no owner or operatorof an affected facility for which openhooding is the method for controllingprimary emissions shall cause to bedischarged to the atmosphere any gasesthat:

(i) Contain particulate matter inexcess of 50 mg/dscm (0.022 gr/dscf), asmeasured for the primary oxygen blow.

(ii) Exit from a control device not usedsolely for the collection of secondaryemissions, as defined in § 60.141a, andexhibit 10 percent opacity or greater,except that an opacity greater than 10percent but less than 20 percent mayoccur once per steel production cycle.

(2) On or after the date on which theperformance test required by § 60.8 iscompleted, no owner or operator of anaffected facility for which closed 'hooding is the method for controllingprimary emissions shall cause to bedischarged into the atmosphere anygases that:

(i) Contain particulate matter in,excess of 68 mg/dscm (0.030 gr/dscf), asmeasured for the primary oxygen blow.

(ii) Exit from a control device not usedsolely for the collection of secondaryemissions, as defined in*§ 60.141a, and

exhibit 10 percent opacity or greater,except that an opacity greater than 10percent but less than 20 percent mayoccur once per steel production cycle.

(c) On and after the date on which theperformance test required by § 60.8 iscompleted, each owner or operator of anaffected facility subject to paragraph (b)of this section shall operate the primarygas cleaning system during any reblowin a manner identical to operationduring the primary oxygen blow.

5. In § 60.143, paragraphs (b)(2) and (c)are revised to read as follows: "

§ 60.143 Monitoring of operations.(b) * * *(1) ** *

(2) A monitoring device for thecontinual measurement of the watersupply pressure to the controlequipment. The monitoring device is tobe certified by the manufacturer to beaccurate within _5 percent of thedesign water supply pressure. Themonitoring device's pressure sensor orpressure tap must be located close to thewater discharge point. TheAdministrator must be consulted forapproval in advance of'selectingalternative locations for the pressuresensor or tap.* * * * *

(c) Any owner or operator subject tothe requirements of paragraph (b) of thissection shall report to the Administrator,on a semiannual basis, allmeasuremefits over any 3-hour periodthat average more than 10 percent belowthe average levels maintained during themost recent performance test conductedunder § 60.8 in which the affectedfacility demonstrated compliance withthe mass standards under § 60.142(a)(1),(b)(1)(A) or (b)(2)(A). The accuracy ofthe respective measurements, not toexceed the values specified inparagraphs (b)(1) and (b](2) of thissection, may be taken into considerationwhen determining the measurementresults that must be reported.

6. In § 60.144, paragraph (b) is revisedto read as follows:

§ 60.144 Test methods and procedures.* * * * *

(b) For Method 5, the sampling timeshall be as follows:

(1) For affected facilities thatcommenced construction, modification,or reconstruction on or before January20,1983, the sampling for each run shallcontinue for an integral number of steelproduction cycles with total duration ofat least 60 minutes. A cycle shall start atthe beginning of either the scrap preheator the oxygen blow and shall terminateimmediately prior to'tapping. The

minimum sample volume shall be atleast 1.5 dscm (53 dscf). Shortersampling times and smaller samplevolumes, when necessitated by processvariables or other factors, may beapproved by the Administrator.

(2) For affected facilities thatcommence construction, modification, orreconstruction after January 20 1983, thesampling for each run shall continue foran integral number of primary oxygenblows, with total duration of at least 60minutes. The minimum sample volumeshall be at least 1.5 dscm (53 dscf).Shorter sampling times, and smallersample volumes, when.necessitated byprocess variables or other factors, may.be approved by the Administrator.

7. Sections 60.143 and 60.144 areamended to include the followingstatement at the end of each section:(Approved by the Office of Management andBudget under control number 2060-0029)

8. By adding a new Subpart Na asfollows:

Subpart Na-Standards of Performance forSecondary Emissions From Basic OxygenProcess Steelmaking Facilities for WhichConstruction Is Commenced After January20, 1983Sec.60.140a Applicability and designation of

affected facilities.60.141a Definitions.60.142a Standards for particulate matter.60.143a Monitoring of operations.60.144a Test methods and procedures.60.145a Compliance provisions.

Subpart Na-Standards ofPerformance for Secondary EmissionsFrom Basic oxygen ProcessSteelmaking Facilities for WhichConstruction Is Commenced AfterJanuary 20, 1983

§ 60.140a Applicability and designation ofaffected facilities.

(a) The provisions of this subpartapply to the following affected facilitiesin an iron and steel plant: top-blownBOPF's and hot metal transfer stationsand skimming stations used withbottom-blown or top-blown BOPF's.

(bJ This subpart applies to any facilityidentified in paragraph (a) of this sectionthat commences construction,modification, or reconstruction afterJanuary 20,1983.

(c) Any BOPF subject to theprovisions of this subpart is subject tothose provisions of Subpart N of thisPart applicable to affected facilitiescommencing construction, modificationor reconstruction after January 20, 1983.

6ederal Register / Vol. 51, No. 1 / Thursday, January 2, 1986 / Rules and Regulations

§ 60.141a Definitions.All terms in this subpart not defined

below are given the same meaning as inthe Clean Air Act as amended or inSubpart A of this part.

"Basic Oxygen process furnace"(BOPF means any furnace with arefractory lining in which molten steel isproduced by charging scrap metal,molten iron, and flux materials or alloyadditions into a vessel and byintroducing a high volume of oxygen-rich gas. Open hearth, blast, andreverberatory furnaces are not includedin this definition.

"Bottom-blown furnace" means anyBOPF in which oxygen and othercombustion gases are introduced to thebath of molten iron through tuyeres inthe bottom of the vessel or throughtuyeres in the bottom and sides of thevessel.

"Fume suppression system" means theequipment comprising any system usedto inhibit the generation of emissionsfrom steelmaking facilities with an inertgas, flame, or steam blanket applied tothe surface of molten iron or steel."Hot metal transfer station" means

the facility where molten iron is emptiedfrom the railroad torpedo car or hotmetal car to the shop ladle. Thisincludes the transfer of molten iron fromthe torpedo car or hot metal car to amixer (or other intermediate vessel). andfrom a mixer (or other intermediatevessel) to the ladle. This facility is alsoknown as the reladling station or ladletransfer station.

"Primary oxygen blow" means theperiod in the steel production cycle of aBOPF during which a high volume ofoxygen-rich gas is introduced to the bathof molten iron by means of a lanceinserted from the top of the vessel. Thisdefinition does not include anyadditional, or secondary, oxygen blowsmade after the primary blow.

"Primary emission control system"means the combination of equipmentused for the capture and collection ofprimary emissions (e.g., an open hoodcapture system used in conjunction witha particulate matter cleaning devicesuch as an electrostatic precipitator or aclosed hood capture system used inconjunction with a particulate mattercleaning device such as a scrubber).

"Primary emissions" meansparticulate matter emissions from theBOPF generated during the steelproduction cycle which are captured by,and do not thereafter escape from, theBOPF primary control system.

"Secondary emission control system"means the combination of equipmentused for the capture and collection ofsecondary emissions (e.g., (1) an openhood system for the capture and

collection of primary and secondaryemissions from the BOPF, with localhooding ducted to a secondary emissioncollection device such as a'baghouse forthe capture and collection of emissionsfrom the hot metal transfer andskimming station; or (2) an open hoodsystem for the capture and collection ofprimary and secondary emissions fromthe furnace, plus a furnace enclosurewith local hooding ducted to asecondary emission collection device,such as a baghouse, for additionalcapture and collection of secondaryemissions from the furnace, with localhooding ducted to a secondary emissioncollection device, such as a baghouse,for the capture and collection ofemissions from hot metal transfer andskimming station; or (3) a furnaceenclosure with local hooding ducted to asecondary emission collection devicesuch as a baghouse for the capture andcollection of secondary emissions from aBOPF controlled by a closed hoodprimary emission control system, withlocal hooding ducted to a secondaryemission collection device, such as abaghouse, for the capture and collectionof emissions from hot metal transfer andskimming stations).

"Secondary emissions" meansparticulate matter emissions that are notcaptured by the BOPF primary controlsystem, including emissions from hotmetal transfer and skimming stations.This definition also includes particulatematter emissions that escape fromopenings in the primary emission controlsystem, such as from lance holeopenings, gaps or tears in the ductworkof the primary emission control system,or leaks in hoods.

"Skimming station" means the facilitywhere slag is mechanically raked fromthe top of the bath of molten iron.

"Steel production cycle" means theoperations conducted within the BOPFsteelmaking facility that are required toproduce each batch of steel, includingthe following operations: scrap charging,preheating (when used), hot metalcharging, primary oxygen blowing,sampling (vessel turndown and turnup),additional oxygen blowing (when used),tapping, and deslagging. Hot metaltransfer and skimming operations for thenext steel production cycle are alsoincluded when the hot metal transferstation or skimming station is anaffected facility.

"Top-blown furnace" means anyBOPF in which oxygen is introduced to,the bath of molten iron by means of anoxygen lance inserted from the top ofthe vessel.

§ 60.142a Standards for particulate matter.(a) Except as provided under

paragraphs (b) and (c) of this section, onand after the date on which theperformance test under § 60.8 is requiredto be completed, no owner or operatorsubject to the provisions of this subpartshall cause to be discharged into theatmosphere from any affected facilityany secondary emissions that:

(1) Exit from the BOPF shop roofmonitor (or other building openings) andexhibit greater than 10 percent opacityduring the steel production cycle of anytop-blown BOPF or during hot metaltransfer or skimming operations for anybottom-blown BOPF; except that anopacity greater than 10 percent but lessthan 20 percent may occur once per steelproduction cycle.

(2) Exit from a control device usedsolely for the collection of secondaryemissions from a top-blown BOPF orfrom hot metal transfer or skimming fora top-blown or a bottom-blown BOPFand contain particulate matter in excessof 23 mgtdscm (0.010 gr/dscf).

(3) Exit from a control device usedsolely for the collecton of secondaryemissions from a top-blown BOPF orfrom hot metal transfer or skimming fora top-blown or a bottom-blown BOPFand exhibit more than 5 percent opacity.

(b) A fume suppression system usedto control secondary emissions from anaffected facility is not subject toparagraphs (a)(2) and (a)(3) of thissection.

(c) A control device used to collectboth primary and secondary emissionsfrom a BOPF is not subject toparagraphs (a)(2) and (a)(31 of thissection.

§ 60.143a Monitoring of operations.(a] Each owner or operator of an

affected facility shall install, calibrate,operate, and maintain a monitoringdevice that continually measures andrecords for each steel production cyclethe various rates or levels of exhaustventilation at each phase of the cyclethrough each duct of the secondaryemission capture system. Themonitoring device or devices are to beplaced at locations near each capturepoint of the secondary emission capturesystem to monitor the exhaustventilation rates or levels adequately, orin alternative locations approved inadvance by the Administrator.

(b) If a chart recorder is used, theowner or operator shall use chartrecorders that are operated at aminimum chart speed of 3.8 cm/hr (1.5in./hr).

(c) All monitoring devices are to becertified by the manufacturer to be

__ I I all I16'

Federal Register / VoL 51, No. T f Thursday, January Z 1986 / Rules and Regulations 163

accurate to within ±10 percentcompared to EPA Reference Method 2.The owner or operator shall recalibrateand check thedevice(s) annually and atother times as the Administrator mayrequire, in accordance with the writteninstructions of the manufacturer and bycomparing the device against EPAReference Method 2.

(d) Each owner or operator subject tothe requirements of paragraph (a) of thissection shall report on A semiannualbasis all measurements of exhaustventilation rates or levels over any 3-hour period that average more than 10percent below the average rates orlevels of exhaust ventilation maintainedduring the most recent performance testconducted under § 60.8 in which theaffected facility demonstratedcompliance with the standard under§ 60.142a (a)(2). The accuracy of therespective measurements, not to exceedthe values specified in paragraph (c) ofthis section, may be considered whendetermining the measurement resultsthat must be reported.

(e) If a scrubber primary emissioncontrol device is used to collectsecondary emissions, the owner oroperator shall report on a semiannualbasis all measurements of exhaustventilation rate over any 3-hour periodthat average more than 10 percent belowthe average levels maintained during themost recent performance test conductedunder § 60.8 in which the affectedfacility demonstrated compliance withthe standard under § 60.142(a)(1).(Approved by the Office of Management andBudget under control number 200-0029)§ 60.144a Test methods and procedures.

(a) The reference methods inAppendix A of this part, except asprovideOi under § 60.8(b) and as notedbelow, shall be used to determine.compliance with § 60.142a as follows:

(1) Method I for sample and velocitytraverses;

(2) Method 2 for volumetric flow rate;(3) Method 3 for gas analysis;(4) Method 5 for concentration of

particulate matter and associatedmoisture content; and

(5) Method 9 for visible emissionsexcept as provided in paragraph (b) ofthis section.

,(b) For Method 9, the folowinginstructions for recording observationsand reducing data shall apply instead ofSections 2.4 and 2.5 of Method 9:

(1) Section 2.4. Opacity observationsshall be recorded to the nearest 5percent at 15-second intervals. Duringthe initial performance test conductedpursuant to § 60.8, observations shall bemade and recorded in this manner for aminimum of three steel production

cycles. During any subsequentcompliance test, observations may bemade for any number of steel productioncycles, although, where conditionspermit, observations will generally bemade for a minimum of three steelproduction cycles.

(2) Section 2.5. Opacity shall bedetermined as an average of 12consecutive observations recorded at.15-second intervals. For each steelproduction cycle, divide theobservations recorded into sets of 12consecutive observations. Sets need notbe consecutive in time, and iii no caseshall two sets overlap. For each set of 12observations, calculate the average bysumming the opacity of 12 consecutiveobservations and dividing this sum by12.

(c) For the sampling of secondaryemissions by Method 5, the sampling foreach run is to continue for a sufficientnumber of steel production cycles toensure a total sample volume of at least5.67 dscm (200 dscf) for each run.Shorter sampling times and smallersample volumes, when necessitated byprocess variables or other factors, maybe approved by the Administrator.Sampling is to be conducted only duringthe steel production cycle.

(d) For the monitoring and recordingof exhaust ventilation rates or levelsrequired by § 60.143a(a), the followinginstructions for Reference Method 2shall apply:

( ) For devices that monitor andrecord the exhaust ventilation rate,compare velocity readings recorded bythe monitoring device against thevelocity readings obtained by Method 2.Take Method 2 readings at a point orpoints that would properly characterizethe monitoring device's performanceand that would adequately reflect thevarious rates of exhaust ventilation.Obtain readings at sufficient intervals toobtain 12 pairs of readings for each ductof the secondary emission capturesystem. Compare the averages of thetwo sets to determine whether the 'monitoring device velocity is within ±10percent of the Method 2 average.'

(2) For devices that monitor the levelof exhaust ventilation and record onlystep changes when a set point rate isreached, compare step changes recordedby the monitoring device against thevelocity readings obtained by Method 2.Take Method 2 readings at a point orpoints that would properly characterizethe performance of the monitoringdevice and that would adequatelyreflect the various rates of exhaustventilation. Obtain readings at sufficientintervals to obtain 12 pairs of readingsfor each duct of the secondary emissioncapture system. Compare the averages

of the two sets to determine whether themonitoring device step change is within±10 percent of the setpoint rate.(Approved by the Office of Management andBudget under control number 2060-0029)

§ 60.145a Conpliance provisions.(a) When determining compliance

with mass and visible emission limitsspecified in § 60.142a(s)(2) and§ 60.142a(a)(3), the owner or operator ofa BOPF shop that normally operates twofurnaces with overlapping cycles mayelect to operate only one furnace. If anowner or operator chooses to shut downone furnace, he shall be allowed areasonable time period to adjust hisproduction schedule before thecompliance tests are conducted. Theowner or operator of an affected facilitymay also elect to suspend shopoperations not subject to this subpartduring compliance testing.

(b) During compliance testing for massand visible emission standards, if anowner or operator elects to shut downone furnace in a shop that normallyoperates two furnaces with overlappingcycles, the owner or operator shalloperate the secondary emission controlsystem for the furnace being tested atexhaust ventilation rates or levels foreach duct of the secondary emissioncontrol system that are appropriate forsingle-furnace operatin. Following thecompliance test, the owner or operatorshall operate the secondary emissioncontrol system at exhaust ventilationrates or levels for each duct of thesystem that are no lower than 90 percentof the exhaust ventilation valuesestablished during the most recentcompliance test.

(c) For the purpose of determiningcompliance with visible and massemission standards, a steel productioncycle begins when the scrap or hot metalis charged to the vessel (whicheveroperation occurs first) and terminates 3minutes after slag is emptied from thevessel into the slag pot. Consecutivesteel production cycles are not requiredfor the purpose of determiningcompliance. Where a hot metal transferor skimming station is an affectedfacility, the steel production cycle alsoincludes the hot metal transfer orskimming operation for the next steelproduction cycle for the affected vessel.Visible emission observations for bothhot metal transfer and skimmingoperations begin with the start of theoperation and terminate 3 minutes aftercompletion of the operation.

(d) For the purpose of determiningcompliance with visible emissionstandards specified in § 60.142a(a)(1)and (a)(3), the starting and stopping

164 Federal Register./ Vol. 51, No. 1 / Thursday, January 2, 1986 / Rules and Regulations

times of regulated process operationsshall be determined and the starting andstopping times of visible emissions datasets shall be determined accordingly.

(e) To determine compliance with'§ 60.142a(a)(1), select the data setsyielding the highest and setond highest3-minute average opacities for each steelproduction cycle. Compliance isachieved if the highest 3-minute average

for each cycle observed is less than 20'percent and the second highest 3-minuteaverage is 10 percent or less.

(f) To determine compliance with§ 60.142(a)(2), determine theconcentration of particulate matter inexhaust gases exiting the secondaryemission collection device withReference Method 5. Compliance isachieved if the concentration of

particulate matter 'does not exceed 23mg/dscm (0.010 gr/dscf).

(g) To determine compliance with§ 60.142a(a)(3), construct consecutive 3-minute averages for each steelproduction cycle. Compliance isachieved if no 3-minute average is morethan 5 percent.[FR Doc. 85-30892 Filed 12-31-85; 8:4,5 am!WWNG cl smx 560'5