§ 63.9621 - What test methods and other procedures must I use to demonstrate initial compliance with the emission limits?

§ 63.9621 What test methods and other procedures must I use to demonstrate initial compliance with the emission limits for particulate matter?

(a) On or before January 25, 2021, for affected sources that commenced construction or reconstruction on or before September 25, 2019, you must conduct each performance test that applies to your affected source according to the requirements in § 63.7(e)(1) and paragraphs (b) and (c) of this section. After January 25, 2021, for affected sources that commenced construction or reconstruction on or before September 25, 2019, and after July 28, 2020, or upon start-up, which ever date is later, for affected sources that commenced construction or reconstruction after September 25, 2019, you must conduct each performance test that applies to your affected source, including the initial performance tests for mercury required in § 63.9620(k)(1) and the initial performance tests for hydrogen chloride and hydrogen fluoride required in § 63.9620(l), under normal operating conditions of the affected source. The owner or operator may not conduct performance tests during periods of malfunction. The owner or operator must record the process information that is necessary to document operating conditions during the test and include in such record an explanation to support that such conditions represent normal operation. Upon request, the owner or operator shall make available to the Administrator such records as may be necessary to determine the conditions of performance tests. You must also conduct each performance test that applies to your affected source according to the requirements in paragraphs (b) and (c) of this section.

(b) For each ore crushing and handling affected source and each finished pellet handling affected source, you must determine compliance with the applicable emission limit for particulate matter in Table 1 to this subpart by following the test methods and procedures in paragraphs (b)(1) through (3) of this section.

(1) Except as provided in § 63.9620(e), determine the concentration of particulate matter in the stack gas for each emission unit according to the test methods listed in paragraphs (b)(1)(i) through (v) of this section.

(i) EPA Method 1 or 1A in appendix A–1 to part 60 of this chapter to select sampling port locations and the number of traverse points. Sampling ports must be located at the outlet of the control device and prior to any releases to the atmosphere.

(ii) EPA Method 2, 2A, 2C, 2D, or 2F in appendix A–1 to part 60 of this chapter or EPA Method 2G in appendix A–2 to part 60 of this chapter, as applicable, to determine the volumetric flow rate of the stack gas.

(iii) EPA Method 3A or 3B in appendix A–2 to part 60 of this chapter to determine the dry molecular weight of the stack gas. The voluntary consensus standard ANSI/ASME PTC 19.10–1981 (incorporated by reference-see § 63.14) may be used as an alternative to the manual procedures (but not instrumental procedures) in EPA Method 3B.

(iv) EPA Method 4 in appendix A–3 to part 60 of this chapter to determine the moisture content of the stack gas.

(v) EPA Method 5 or 5D in appendix A–3 to part 60 of this chapter or EPA Method 17 in appendix A–6 to part 60 of this chapter to determine the concentration of particulate matter.

(2) Each EPA Method 5, 5D, or 17 performance test must consist of three separate runs. Each run must be conducted for a minimum of 1 hour. If any measurement result is reported as below the method detection limit, use the method detection limit for that value when calculating the average particulate matter concentration. The average particulate matter concentration from the three runs will be used to determine compliance, as shown in Equation 1 of this section.

Where:

C_i = Average particulate matter concentration for emission unit, grains per dry standard cubic foot, (gr/dscf);

C₁ = Particulate matter concentration for run 1 corresponding to emission unit, gr/dscf;

C₂ = Particulate matter concentration for run 2 corresponding to emission unit, gr/dscf; and

C₃ = Particulate matter concentration for run 3 corresponding to emission unit, gr/dscf.

(3) For each ore crushing and handling affected source and each finished pellet handling affected source, you must determine the flow-weighted mean concentration of particulate matter emissions from all emission units in each affected source following the procedure in paragraph (b)(3)(i) or (ii) of this section.

(i) If an initial performance test is conducted on all emission units within an affected source, calculate the flow-weighted mean concentration of particulate matter emissions from the affected source using Equation 2 of this section.

Where:

C_a = Flow-weighted mean concentration of particulate matter for all emission units within affected source, (gr/dscf);

C_i = Average particulate matter concentration measured during the performance test from emission unit “i” in affected source, as determined using Equation 1 of this section, gr/dscf;

Q_i = Average volumetric flow rate of stack gas measured during the performance test from emission unit “i” in affected source, dscf/hr; and

n = Number of emission units in affected source.

(ii) If you are grouping similar emission units together in accordance with § 63.9620(e), you must follow the procedures in paragraphs (b)(3)(ii)(A) through (C) of this section.

(A) Assign the average particulate matter concentration measured from the representative unit, as determined from Equation 1 of this section, to each emission unit within the corresponding group of similar units.

(B) Establish the maximum operating volumetric flow rate of exhaust gas from each emission unit within each group of similar units.

(C) Using the data from paragraphs (b)(3)(ii)(A) and (B) of this section, calculate the flow-weighted mean concentration of particulate matter emissions from the affected source using Equation 3 of this section.

Where:

C_a = Flow-weighted mean concentration of particulate matter for all emission units within affected source, gr/dscf;

C_k = Average particulate matter concentration measured during the performance test from the representative emission unit in group “k” of affected source “a,” as determined using Equation 1 of this section, gr/dscf;

Q_k = Sum of the maximum operating volumetric flow rates of stack gas from all similar emission units within group “k” of affected source, dscf/hr; and

m = Number of similar emission unit groups in affected source.

(c) For each ore dryer affected source and each indurating furnace affected source, you must determine compliance with the applicable emission limit for particulate matter in Table table 1 to this subpart by following the test methods and procedures in paragraphs (c)(1) through (32) of this section.

(1) Determine the concentration of particulate matter for each stack according to the test methods listed in paragraphs (c)(1)(i) through (v) of this section.

(i) EPA Method 1 or 1A in appendix A–1 to part 60 of this chapter to select sampling port locations and the number of traverse points. Sampling ports must be located at the outlet of the control device and prior to any releases to the atmosphere.

(ii) EPA Method 2, 2A, 2C, 2D, or 2F in appendix A–1 to part 60 of this chapter or EPA Method 2G in appendix A–2 to part 60 of this chapter, as applicable, to determine the volumetric flow rate of the stack gas.

(iii) EPA Method 3A or 3B in appendix A–2 to part 60 of this chapter to determine the dry molecular weight of the stack gas. The voluntary consensus standard ANSI/ASME PTC 19.10–1981 (incorporated by reference-see § 63.14) may be used as an alternative to the manual procedures (but not instrumental procedures) in EPA Method 3B.

(iv) EPA Method 4 in appendix A–3 to part 60 of this chapter to determine the moisture content of the stack gas.

(v) EPA Method 5 or 5D in appendix A–3 to part 60 of this chapter to determine the concentration of particulate matter.

(2) Each EPA Method 5 or 5D performance test must consist of three separate runs. Each run must be conducted for a minimum of 1 hour. If any measurement result is reported as below the method detection limit, use the method detection limit for that value when calculating the average particulate matter concentration. The average particulate matter concentration from the three runs will be used to determine compliance, as shown in Equation 1 of this section.

(d) For each indurating furnace subject to the initial performance testing under § 63.9620(k)(1) or (l), you must determine compliance with the applicable emission limits for mercury, hydrogen chloride and hydrogen fluoride in tables 2 and 3 to this subpart by following the test methods and procedures in paragraphs (d)(1) through (9) of this section. You are not required to complete the initial performance test for mercury emissions when you are using a CEMS in accordance with paragraph (e) of this section.

(1) The furnace must be operated at or above 90 percent of capacity throughout the duration of the performance testing. If testing cannot be performed at or above 90 percent of capacity, you must provide an explanation for the lower production rate in your performance test plan. The lower production rate must be approved by the Administrator prior to beginning performance testing. For indurating furnaces that comply with the mercury emissions limit in table 2 to this subpart by adjusting the activated carbon injection rate based on the taconite pellet production rate, you must complete the performance testing for mercury in accordance with the provisions in § 63.9634(n).

(2) Use the methods specified in paragraphs (c)(1)(i) through (iv) of this section to select sampling port locations and the number of traverse points and to determine the volumetric flow rate, dry molecular weight, and moisture content of the stack gas.

(3) Determine the concentration of mercury for each stack using Method 29 or Method 30B in 40 CFR part 60, appendix A, or the voluntary consensus standard ASTM D6784–16 (incorporated by reference, see § 63.14). For Method 29 and ASTM D6784–16, the sample volume must be at least 1.7 dry standard cubic meters (dscm) (60 dry standard cubic feet) per run. For Method 30B, each test run must be at least one hour in duration.

(4) Determine the concentration of hydrogen chloride and hydrogen fluoride for each stack using Method 26A in 40 CFR part 60, appendix A. Each test must consist of three separate runs. The minimum sample volume must be at least 2 dscm per run.

(5) During each stack test run, determine the weight of taconite pellets produced and calculate the emissions rate of each pollutant in pounds of pollutant per long ton (lb/LT) of pellets produced for each test run. The weight of taconite pellets produced must be determined by measurement using weigh hoppers, belt weigh feeders, or weighed quantities in shipments, or calculated using the bulk density and volume measurements. If any measurement result for any pollutant is reported as below the method detection limit, use the method detection limit as the measured emissions level for that pollutant when calculating the emission rate. If the furnace has more than one stack, calculate the total emissions rate for each test run by summing the emissions across all stacks, as shown in Equation 4.

Where:

E_f,i = Emissions rate for test run “i” for all emission stacks on indurating furnace “f”, lb/LT of pellets produced,

C_s = Emission rate for stack “s” measured during test run “i” on indurating furnace “f”, lb/dscf,

Q_s = Average volumetric flow rate of stack gas measured at stack “s” during test run “i” on indurating furnace “f”, dscf/hour;

P_f = Pellets produced in indurating furnace “f” during the stack test, LT; and

n = Number of emissions stacks on furnace “f”.

(6) Calculate the average emissions rate for each furnace using the three test runs, as show in Equation 5 of this section.

Where:

E_f = Average emission rate for indurating furnace “f”, lb/LT of pellets produced,

E₁ = Emissions rate for run 1 for indurating furnace “f”, lb/LT of pellets produced,

E₂ = Emissions rate for run 2 for indurating furnace “f”, lb/LT of pellets produced, and

E₃ = Emissions rate for run 3 for indurating furnace “f”, lb/LT of pellets produced.

(7) For each indurating furnace constructed or reconstructed on or after May 15, 2023, determine compliance with the applicable mercury emission limit in table 2 to this subpart by calculating the average emissions rate from the three test runs performed on the furnace using Equations 4 and 5 of this section.

(8) For each indurating furnace constructed or reconstructed before May 15, 2023, you must determine compliance with the applicable mercury emission limit in accordance with the procedures specified in either paragraph (d)(8)(i) or (ii) of this section.

(i) Determine compliance with the mercury limit for individual furnaces in table 2 to this subpart by calculating the average mercury emissions rate for each affected indurating furnace using Equations 4 and 5 of this section, or

(ii) Determine compliance with the mercury limit for groups of indurating furnaces in table 2 to this subpart in accordance with the method in § 63.9623(d).

(9) Determine compliance with the applicable hydrogen chloride and hydrogen fluoride emission limits in table 3 to this subpart by calculating the average emissions rate for each indurating furnace for the three test runs performed on the furnace using Equations 4 and 5 of this section.

(e) For each indurating furnace using mercury CEMS to demonstrate compliance with the applicable emission limits for mercury, you must determine compliance with the applicable mercury limit in table 2 to this subpart by using a 30-day rolling average of the 1-hour arithmetic average CEMS data, including CEMS data during startup and shutdown as defined in this subpart. The mercury CEMS must be installed, calibrated, maintained, and operated as accordance with the requirements in § 63.9631(j).

[68 FR 61888, Oct. 30, 2003, as amended at 85 FR 45492, July 28, 2020; 89 FR 16427, Mar. 6, 2024]