(a) Each CEMS must be installed, operated, and maintained according to the applicable Performance Specification (PS) of 40 CFR part 60, appendix B.
(b) You must conduct a performance evaluation of each CEMS according to the requirements in Sec. 63.8, according to the applicable Performance Specification of 40 CFR part 60, appendix B, and according to paragraphs (b)(1) through (4) of this section.
(1) If your CEMS monitor generates a single combined response value for VOC (examples of such detection principles are flame ionization, photoionization, and non-dispersive infrared absorption), but it is not a flame ionization analyzer, you must use PS 8 to show that your CEMS is operating properly.
(i) Use EPA Test Method 25A of 40 CFR part 60, appendix A, to do the relative-accuracy test PS 8 requires.
(ii) Calibrate the reference method with propane.
(iii) Collect a 1-hour sample for each reference-method test.
(2) If you continuously monitor VOC emissions using a flame ionization analyzer, then you must conduct the calibration drift test PS 8 requires, but you are not required to conduct the relative-accuracy test PS 8 requires.
(3) If you continuously monitor VOC emissions using gas chromatography, you must use PS 9 of CFR part 60, appendix B, to show that your CEMS is operating properly.
(4) You must complete the performance evaluation and submit the performance evaluation report before the compliance date that is specified for your source in Sec. 63.2133.
(c) Calibrate the CEMS with propane.
(d) Set the CEMS span at not greater than 5 times the relevant emission limit, with 1.5 to 2.5 times the relevant emission limit being the range considered by us to be generally optimum.
(e) You must monitor VOC concentration in fermenter exhaust at any point prior to dilution of the exhaust stream.
(f) Each CEMS must complete a minimum of one cycle of operation (sampling, analyzing, and data recording) for each successive 30-minute period within each batch monitoring period. Except as specified in paragraph (g) of this section, you must have a minimum of two cycles of operation in a 1-hour period to have a valid hour of data.
(g) The CEMS data must be reduced to arithmetic batch averages computed from two or more data points over each 1-hour period, except during periods when calibration, quality assurance, or maintenance activities pursuant to provisions of this part are being performed. During these periods, a valid hour of data shall consist of at least one data point representing a 30-minute period.
(h) You must have valid CEMS data from at least 75 percent of the full hours over the entire batch monitoring period.
(i) For each CEMS, record the results of each inspection, calibration, and validation check.
(j) You must check the zero (low-level) and high-level calibration drifts for each CEMS in accordance with the applicable PS of 40 CFR part 60, appendix B. The zero (low-level) and high-level calibration drifts shall be adjusted, at a minimum, whenever the zero (low-level) drift exceeds 2 times the limits of the applicable PS. The calibration drift checks must be performed at least once daily except that they may be performed less frequently under the conditions of paragraphs (j)(1) through (3) of this section.
(1) If a 24-hour calibration drift check for your CEMS is performed immediately prior to, or at the start of, a batch monitoring period of a duration exceeding 24 hours, you are not required to perform 24-hour-interval calibration drift checks during that batch monitoring period.
(2) If the 24-hour calibration drift exceeds 2.5 percent of the span value (or more than 10 percent of the calibration gas value if your CEMS is a gas chromatograph (GC)) in fewer than 5 percent of the checks over a 1-month period, and the 24-hour calibration drift never exceeds 7.5 percent of the span value, then the frequency of calibration drift checks may be reduced to at least weekly (once every 7 days).
(3) If, during two consecutive weekly checks, the weekly calibration drift exceeds 5 percent of the span value (or more than 20 percent of the calibration gas value, if your CEMS is a GC), then a frequency of at least 24-hour interval calibration checks must be resumed until the 24-hour calibration checks meet the test of paragraph (j)(2) of this section.
(k) If your CEMS is out of control, you must take corrective action according to paragraphs (k)(1) through (3) of this section.
(1) Your CEMS is out of control if the zero (low-level) or high-level calibration drift exceeds 2 times the limits of the applicable PS.
(2) When the CEMS is out of control, take the necessary corrective action and repeat all necessary tests that indicate that the system is out of control. You must take corrective action and conduct retesting until the performance requirements are below the applicable limits.
(3) During the batch monitoring periods in which the CEMS is out of control, recorded data shall not be used in data averages and calculations, or to meet any data availability requirement established under this subpart. The beginning of the out-of-control period is the beginning of the first batch monitoring period that follows the most recent calibration drift check during which the system was within allowable performance limits. The end of the out-of-control period is the end of the last batch monitoring period before you have completed corrective action and successfully demonstrated that the system is within the allowable limits. If your successful demonstration that the system is within the allowable limits occurs during a batch monitoring period, then the out-of-control period ends at the end of that batch monitoring period. If the CEMS is out of control for any part of a particular batch monitoring period, it is out of control for the whole batch monitoring period.