(PMD), zirconia (ZRDO) or electrochemical type (ECS).

(4) Hydrocarbon (HC) analysis. (i) The hydrocarbon analyzer must be of the heated flame ionization (HFID) type.

(ii) If the temperature of the exhaust gas at the sample probe is below 190 °C, the temperature of the valves, pipework, and so forth, must be controlled so as to maintain a wall temperature of 190 °C ± 11 °C. If the temperature of the exhaust gas at the sample probe is above 190 °C, the temperature of the valves, pipework, and so forth, must be controlled so as to maintain a wall temperature greater than 180 °C.

(iii) The oven must be capable of maintaining temperature within 2 °C of the set point.

(iv) Fuel and burner air must conform to the specifications in §89.312–96. (v) The percent of oxygen interference must be less than 3 percent, as specified in § 89.319-96(d).

(5) Oxides of nitrogen (NOx) analysis. (i) This analysis device must consist of the subsequent items, following the sample probe, in the given order:

(A) Pipework, valves, and so forth, controlled so as to maintain a wall temperature above 60 °C.

(B) A NO2 to NO converter. The NO2 to NO converter efficiency must be at least 90 percent.

(C) An ice bath or other cooling device located after the NOx converter. (D) A (CLD).

chemiluminescent detector

(ii) The quench interference must be less than 3.0 percent as measured in § 89.318-96.

(b) Other gas analyzers yielding equivalent results may be used with advance approval of the Administrator.

(c) The following requirements must be incorporated in each system used for testing under this subpart.

(1) Carbon monoxide and carbon dioxide measurements must be made on a dry basis (for raw exhaust measurement only). Specific requirements for the means of drying the sample can be found in §89.309-96(e).

(2) Calibration or span gases for the NOx measurement system must pass through the NO2 to NO converter.

(d) The electromagnetic compatibility (EMC) of the equipment must be on

a level as to minimize additional errors.

(e) Gas drying. Chemical dryers are not an acceptable method of removing water from the sample. Water removal by condensation is acceptable. A water trap performing this function and meeting the specifications in §89.308– 96(b) is an acceptable method. Means other than condensation may be used only with prior approval from the Administrator.

§ 89.310-96 Analyzer accuracy and specifications.

(a) Measurement accuracy-general. The analyzers must have a measuring range which allows them to measure the concentrations of the exhaust gas sample pollutants with the accuracies shown in Table 3 in Appendix A of this subpart.

(1) Response time. The analyzer response time must be measured and accounted for before recording of data begins.

(2) Precision. The precision of the analyzer must be, at worst, ±1 percent of full-scale concentration for each range used at or above 100 ppm (or ppmC) or ±2 percent for each range used below 100 ppm (or ppmC). The precision is defined as 2.5 times the standard deviation(s) of 10 repetitive responses to a given calibration or span gas.

(3) Noise. The analyzer peak-to-peak response to zero and calibration or span gases over any 10-second period must not exceed 2 percent of full-scale chart deflection on all ranges used.

(4) Zero drift. The analyzer zero-response drift during a 1-hour period must be less than 2 percent of full-scale chart deflection on the lowest range used. The zero-response is defined as the mean response including noise to a zero-gas during a 30-second time interval.

(5) Span drift. The analyzer span drift during a 1-hour period must be less than 2 percent of full-scale chart deflection on the lowest range used. The analyzer span is defined as the difference between the span-response and the zero-response. The span-response is defined as the mean response including noise to a span gas during a 30-second time interval.

(c) Emission measurement accuracy— Bagged sampling. (1) Good engineering practice dictates that exhaust emission sample analyzer readings below 15 percent of full-scale chart deflection should generally not be used.

(2) Some high resolution read-out systems, such as computers, data loggers, and so forth, can provide sufficient accuracy and resolution below 15 percent of full scale. Such systems may be used provided that additional calibrations are made to ensure the accuracy of the calibration curves. If a gas divider is used, the gas divider must conform to the accuracy requirements specified in §89.312–96(c). The following procedure for calibration below 15 percent of full scale may be used:

(i) Span the full analyzer range using a top range calibration gas meeting the accuracy requirements of § 89.312–96(c).

(ii) Generate a calibration curve according to, and meeting the requirements of, §§ 89.319-96 through 89.323-96.

(iii) Select a calibration gas (a span gas may be used for calibrating the CO2 analyzer) with a concentration midway between the two lowest calibration gases or non-zero gas divider increments. This gas must be "named" to an accuracy of ±2.0 percent of NIST gas standards, or other standards approved by the Administrator.

(iv) Using the calibration curve fitted to the points generated in paragraphs (c)(2)(i) and (ii) of this section, check the concentration of the gas selected in paragraph (c)(2)(iii) of this section. The concentration derived from the curve must be within ±2.3 percent (±2.8 percent for CO2 span gas) of the original named gas concentration.

(v) Provided the requirements of paragraph (c)(2)(iv) of this section are met, use the gas divider with the gas selected in paragraph (c)(2)(iii) of this section and determine the remainder of the calibration points. Fit a calibration curve per §§ 89.319-96 through 89.322-96 of this chapter for the entire analyzer range.

(d) Emission measurement accuracycontinuous sampling. Analyzers used for continuous analysis must be operated such that the measured concentration falls between 15 and 100 percent of fullscale chart deflection. Exceptions to these limits are:

(1) The analyzer's response may be less than 15 percent or more than 100 percent of full scale if automatic range change circuitry is used and the limits for range changes are between 15 and 100 percent of full-scale chart deflection;

(2) The analyzer's response may be less than 15 percent of full scale if:

(i) Alternative (c)(2) of this section is used to ensure that the accuracy of the calibration curve is maintained below 15 percent; or

(ii) The full-scale value of the range is 155 ppm (or ppmC) or less.

§ 89.311-96 Analyzer calibration frequency.

(a) Prior to initial use and after major repairs, bench check each analyzer (see § 89.315-96).

(b) Calibrations are performed as specified in §§ 89.319-96 through 89.324– 96.

(c) At least monthly, or after any maintenance which could alter calibration, the following calibrations and checks are performed.

(1) Leak check the vacuum side of the system (see § 89.316-96).

(2) Check that the analysis system response time has been measured and accounted for.

(3) Verify that the automatic data collection system (if used) meets the requirements found in Table 3 in appendix A of this subpart.

(4) Check the fuel flow measurement instrument to insure that the specifications in Table 3 in appendix A of this subpart are met.

(d) Verify that all NDIR analyzers meet the water rejection ratio and the CO2 rejection ratio as specified in § 89.318-96.

(e) Verify that the dynamometer test stand and power output instrumentation meet the specifications in Table 3 in appendix A of this subpart.

§ 89.312-96 Analytical gases.

(a) The shelf life of all calibration gases must not be exceeded. The expiration date of the calibration gases stated by the gas manufacturer shall be recorded.

(b) Pure gases. The required purity of the gases is defined by the contamination limits given below. The following gases must be available for operation:

(1) Purified nitrogen (Contamination ≤ 1 ppm C, ≤ 1 ppm CO, ≤ 400 ppm CO2, ≤0.1 ppm NO)

(2) Purified oxygen (Purity 99.5 percent vol O2)

(3) Hydrogen-helium mixture (40 ± 2 percent hydrogen, balance helium) (Contamination ≤ 31 ppm C, ≤ 400 ppm

CO)

(4) Purified synthetic air (Contamination ≤1 ppm C, ≤1 ppm CO, ≤ 400 ppm CO2, ≤ 0.1 ppm NO) (Oxygen content between 18-21 percent vol.)

(c) Calibration and span gases. (1) Calibration gas values are to be derived from NIST Standard Reference Materials (SRM's) or other standardized gas samples and are to be single blends as listed in the following paragraph.

(2) Mixtures of gases having the following chemical compositions shall be available:

Cз Hз and purified synthetic air (dilute measurements);

C3 Hз and purified nitrogen (raw measurements);

CO and purified nitrogen;

NOx and purified nitrogen (the amount of NO2 contained in this calibration gas must not exceed 5 percent of the NO content);

CO2 and purified nitrogen

(3) The true concentration of a span gas must be within ±2 percent of the NIST gas standard. The true concentration of a calibration gas must be within ±1 percent of the NIST gas standard. The use of precision blending devices (gas dividers) to obtain the required calibration gas concentrations is acceptable, provided that the blended gases are accurate to within ±1.5 percent of NIST gas standards, or other gas standards which have been approved by the Administrator. This accuracy implies that primary gases used (or blending) must be "named" to an accuracy of at least ±1 percent, trace

able to NIST or other approved gas standards. All concentrations of calibration gas shall be given on a volume basis (volume percent or volume ppm). (4) The gas concentrations used for calibration and span may also be obtained by means of a gas divider, either diluting with purified N2 or diluting with purified synthetic air. The accuracy of the mixing device must be such that the concentration of the diluted gases may be determined to within ±2 percent.

gas

(d) Oxygen interference check gases shall contain propane with 350 ppmC +75 ppmC hydrocarbon. The concentration value shall be determined to calibration tolerances by chromatographic analysis of total hydrocarbons plus impurities or by dynamic blending. Nitrogen shall be the predominant diluent with the balance oxygen.

(e) Fuel for the FID shall be a blend of 40 percent ±2 percent hydrogen with the balance being helium. The mixture shall contain less than 1 ppm equivalent carbon response; 98 to 100 percent hydrogen fuel may be used with advance approval of the Administrator.

(f) Hydrocarbon analyzer burner air. The concentration of oxygen must be within 1 mole percent of the oxygen concentration of the burner air used in the latest oxygen interference check (%O1⁄2 I). If the difference in oxygen concentration is greater than 1 mole percent, then the oxygen interference must be checked and, if necessary, the analyzer adjusted to meet the %O2 I requirements. The burner air must contain less than 2 ppmC hydrocarbon.

§ 89.313-96 Initial calibration of analyzers.

(a) Warming-up time. The warming-up time should be according to the recommendations of the manufacturer. If not specified, a minimum of two hours shall be allowed for warming up the analyzers.

(b) NDIR and HFID analyzer. The NDIR analyzer shall be tuned and maintained according to the instrument manufacturer's instructions. The combustion flame of the HFID analyzer shall be optimized in order to meet the specifications in §89.319-96(b)(2).

(c) Zero setting and calibration. (1) Using purified synthetic air (or nitrogen), the CO, CO2, NOx, and HC analyzers shall be set at zero.

(2) Introduce the appropriate calibration gases to the analyzers and the values recorded. The same gas flow rates shall be used as when sampling exhaust.

(d) Rechecking of zero setting. The zero setting shall be rechecked and the procedure described in paragraph (c) of this section repeated, if necessary.

§ 89.314-96 Pre- and post-test calibration of analyzers.

Each operating range used during the test shall be checked prior to and after each test in accordance with the following procedure. (A chronic need for parameter adjustment can indicate a need for instrument maintenance.):

(a) The calibration is checked by using a zero gas and a span gas whose nominal value is between 80 percent and 100 percent of full-scale, inclusive, of the measuring range.

(b) After the emission test a zero gas and the same span gas will be used for rechecking. The analysis will be considered acceptable if the difference between the two measuring results is less than 2 percent of full scale.

§ 89.315-96 Analyzer bench checks.

(a) Prior to initial use and after major repairs verify that each analyzer complies with the specifications given in Table 3 in appendix A of this subpart.

(b) If a stainless steel NO2 to NO converter is used, condition all new or replacement converters. The conditioning consists of either purging the converter with air for a minimum of 4 hours or until the converter efficiency is greater than 90 percent. The converter must be at operational temperature while purging. Do not use this procedure prior to checking converter efficiency on in-use converters.

§ 89.316-96 Analyzer leakage and response time.

(a) Vacuum side leak check. (1) Any location within the analysis system where a vacuum leak could affect the test results must be checked.

(2) The maximum allowable leakage rate on the vacuum side is 0.5 percent of the in-use flow rate for the portion of the system being checked. The analyzer flows and bypass flows may be used to estimate the in-use flow rates.

(3) The sample probe and the connection between the sample probe and valve V2 (see Figure 1 in appendix B of this subpart) may be excluded from the leak check.

(b) Pressure side leak check. The maximum allowable leakage rate on the pressure side is 5 percent of the in-use flow rate.

(c) The response time shall be accounted for in all emission measurement and calculations.

§ 89.317-96 NOx converter check.

(a) Prior to its introduction into service, and monthly thereafter, the chemiluminescent oxides of nitrogen analyzer shall be checked for NO2 to NO converter efficiency. Figure 2 in appendix B of this subpart is a reference for the following paragraphs.

(b) Follow good engineering practices for instrument start-up and operation. Adjust the analyzer to optimize performance.

(c) Zero the oxides of nitrogen analyzer with zero-grade air or zero-grade nitrogen.

(d) Connect the outlet of the NOx generator to the sample inlet of the oxides of nitrogen analyzer which has been set to the most common operating range.

(e) Introduce into the NOx generator analyzer-system an NO-in-nitrogen (N2) mixture with an NO concentration equal to approximately 80 percent of the most common operating range. The NO2 content of the gas mixture shall be less than 5 percent of the NO concentration.

(f) With the oxides of nitrogen analyzer in the NO mode, record the concentration of NO indicated by the analyzer.

(g) Turn on the NOx generator O2 (or air) supply and adjust the O2 (or air) flow rate so that the NO indicated by the analyzer is about 10 percent less than indicated in paragraph (b)(5) of this section. Record the concentration of NO in this NO+O2 mixture.

(h) Switch the NOx generator to the generation mode and adjust the generation rate so that the NO measured on the analyzer is 20 percent of that measured in paragraph (b)(5) of this section. There must be at least 10 percent unreacted NO at this point. Record the concentration of residual NO.

(i) Switch the oxides of nitrogen analyzer to the NOx mode and measure total NOx. Record this value.

(j) Switch off the NOx generator but maintain gas flow through the system. The oxides of nitrogen analyzer will indicate the NOx in the NO+O2 mixture. Record this value.

(k) Turn off the NOx generator O2 (or air) supply. The analyzer will now indicate the NOx in the original NO-in-N2 mixture. This value should be no more than 5 percent above the value indicated in paragraph (b)(4) of this section.

(1) Calculate the efficiency of the NOx converter by substituting the concentrations obtained into the following equation:

the measured gas and in CLD instruments by the interfering gas quenching the radiation. The interference checks described in this section are to be made initially and after any major repairs that could affect analyzer performance.

(b) CO analyzer water and CO2 interference checks. Prior to its introduction into service and annually thereafter, the NDIR carbon monoxide analyzer shall be checked for response to water vapor and CO25

(1) Follow good engineering practices for instrument start-up and operation. Adjust the analyzer to optimize performance on the most sensitive range to be used.

(2) Zero the carbon monoxide analyzer with either zero-grade air or zerograde nitrogen.

(3) Bubble a mixture of 3 percent CO2 in N2 through water at room temperature and record analyzer response.

(4) An analyzer response of more than 1 percent of full scale for ranges above 300 ppm full scale or more than 3 ppm on ranges below 300 ppm full scale requires corrective action. (Use of conditioning columns is one form of corrective action which may be taken.)

(c) NOx analyzer quench check. The two gases of concern for CLD (and HCLD) analyzers are CO2 and water vapor. Quench responses to these two gases are proportional to their concentrations and, therefore, require test techniques to determine quench at the highest expected concentrations experienced during testing.

(1) NOx analyzer CO2 quench check. A CO2 span gas having a concentration of 80 percent to 100 percent of full scale of the maximum operating range used during testing shall be passed through the CO2 NDIR analyzer and the value recorded as a. It is diluted approximately 50 percent with NO span gas and then passed through the CO2 NDIR and CLD (or HCLD), with the CO2 and NO values recorded as b and c respectively. The CO2 shall then be shut off and only the NO span gas passed through the CLD (or HCLD) and the NO value recorded as d. Percent CO2 quench shall be calculated as follows and shall not exceed 3 percent: