A. Daily and Weekly Data and System Review Check for CEMS

These recommendations are not based on any regulatory requirements, but on gathered knowledge and experience collected from many CEM operators. Each day perform this quick review of available information on the status of the CEM system so you can determine if anything needs immediate attention, attention sometime today or no attention at all.
  1. Each day find the reports which are generated automatically overnight. Look these reports over and respond as needed. These reports typically include:
    • a. Most Part 60 regulations require a CEMS to go through a calibration check once per day, typically every 24-hours. §60.13 (d)(1) does not specifically say that the unit or source has to be operating. Check to see what your permit or state agency is expecting.
    • b. The results from a calibration check can trigger a gaseous monitor to be considered to be “Out-Of-Control” (OOC) one of two different ways. [4.3.1 of Appendix F].
      • i. For gas monitors (SO2, NOX, CO2, O2 and CO, but not opacity), when the calibration error is > than 2 times (2X) the Performance Specification (PS) for 5 consecutive days, from the conclusion of the calibration check on the 5th day until a follow up calibration check is successfully performed and the results are shown to be < 2X the PS. OR
      • ii.Any gaseous monitor which shows that the calibration error was 4X the PS for a single calibration check, then from the conclusion of the calibration check at 4X the PS until a follow up calibration check is successfully performed and the results are shown to be < 4X the PS. Additionally, the data from that monitor is deemed OOC going back to the last good calibration check with results < 4X.
    • c. For an opacity monitor, which is not a gaseous monitor, it also requires a calibration check at least once per day, and usually it is scheduled to occur every 24 hours through the data system. The opacity monitor is considered to be OOC when:
      • i.If either the zero drift or upscale drift check exceeds twice (2X) the applicable drift specification found in PS-1 for any one day.
    • d. What are the Performance Specifications?
      • i. SO2 PS = 2.5% of the span. 2X PS = 5.0% and 4X PS = 10.0%.
      • ii.NOx PS = 2.5% of the span. 2X PS = 5.0% and 4X PS = 10.0%.
      • iii. O2 or CO2 PS = 0.5% using |R-A|. 2X PS = 1.0% and 4X PS = 2.0%.
      • iv. CO PS = 5% of the span. 2X PS = 10.0% and 4X PS = 20.0%.
      • v. Opacity PS = ± 2.0%. 2X PS = ± 4.0%.
  2. Usually each morning, the data system has been configured to not only perform these calibration checks, but to generate reports which show how the system was performing and what the emissions have been for the past day. These reports included:
    • a. A Calibration Error report for all of the monitors. This report shows the results for the calibration check which ran automatically this morning.
    • b. An Average report showing the hourly averages for the data from the monitors from yesterday.
    • c. A second Average report showing the derived (calculated) hourly values from yesterday.
    • d. A “Process Now History” report, which documents any error encountered when the various Process Now sequence ran earlier today.
    • e. Other reports might be generated as well, showing extended averages (3-hour, 24-hour, 30-day, etc.) or year to date totals, or the results of other calculations.
    • f. If your system has an opacity monitor, then the system should also generate an “Opacity Matrix Report”, which lists all of the 6-minute averages from the previous day.
    • g. If your unit started up within the past 24 hours, verify the system performed a calibration check on all of the monitors within the 8-hour startup grace period.
  3. Review the status of the CEM system with the control room operator and see if there are any CEM system alarm up and present.
  4. Hourly data validation. In general, one data point in each of the 15-minute quadrants that the unit or source operated. §60.13 (h)(2) (i) Except as provided under paragraph (h)(2)(iii) of this section, for a full operating hour (any clock hour with 60 minutes of unit operation), at least four valid data points are required to calculate the hourly average, i.e., one data point in each of the 15- minute quadrants of the hour.
    (ii) Except as provided under paragraph (h)(2) (iii) of this section, for a partial operating hour (any clock hour with less than 60 minutes of unit operation), at least one valid data point in each 15-minute quadrant of the hour in which the unit operates is required to calculate the hourly average.
    (iii) For any operating hour in which required maintenance or quality-assurance activities are performed:
    (A) If the unit operates in two or more quadrants of the hour, a minimum of two valid data points, separated by at least 15 minutes, is required to calculate the hourly average; or <br />
    (B) If the unit operates in only one quadrant of the hour, at least one valid data point is required to calculate the hourly average
  5. Opacity six minute averages - §60.13 (h)(1) Six-minute opacity averages shall be calculated from 36 or more data points equally spaced over each 6-minute period.
 
On Thursday or Friday, make a trip to the shelter and make sure everything looks good. Look over the air clean up system and make sure it’s in good shape. Double check the pressure in the calibration bottles and make sure there > 500 psi to get the calibration checks through Monday.

 

Download the pdf document here.

Jon Konings, Senior Regulatory and Reporting Engineer, ESC    Published: 10/1/2016 5:10:39 PM

Questions about the ESC Das?

We'd love to answer your questions.

Contact Us

New to Data Acquisition Systems?

Check out our helpful animation that teaches the basics of

What is a DAS?