    Comments for Draft Appendix F (Part 60)  -  Quality Assurance Procedure
                                  Procedure 6
                                                      Richard A. Hovan Sr.
                                                      rich@epasystem.com
                                                      (512) 592-1417
                                                      June 19, 2013

2.0 Definitions: There is no reference to 40CFR Part60 standards for a Cylinder Gas Audit (CGA). This in itself is a part of Appendix F. This must be added.
2.5 Dynamic Spiking: "known concentration of HCL" how is this addressed? If this is going to be a means to provide daily checks for the system then a Traceable Calibration Gas must be used. Any other means should have the same rigorous QA requirements. This specifically addresses the use of calibrators. Any device that is used on a daily basis to perform a daily system check (by definition of "good Science") must perform the minimum daily checks of the rest of the system.
2.6 Liquid Evaporative Standard: Liquid Evaporative Standard or for that matter any "standard" must be provided with its own performance and test method standard; just like portable analyzers where required to meet CTM 030 &034 which were written specifically for defined applications for a portable.  As a minimum a Conditional Test Method (CTM) must be written and posted on the EPA website for state accessibility and use when employing an alternative means of validation such as Dynamic Spiking in place of traditional traceable cylinder gas testing.					              In addition, if a piece of hardware is being used that is integrated into the CEM system, by EPA definition it is then part of the system and it must meet all QA/QC procedures of the system. With that in consideration, then as a minimum, a Performance Specification should be required for the Calibration Gas Generator. As a minimum a daily Traceable Zero Gas should be applied to any kind of "calibrator."
2.8 Reference Gas Concentrations: As referenced in Table 4 of PS-18 (rev3); Zero is referenced as LOD, in reality a Traceable Zero Gas (TZG) is available and NIST has a test method in place to provide gas suppliers to create and supply TZG. 	                                                                    	                          As additional comments for this section, I respectfully disagree with the "NA" for Zero gas for Spiking. If Spiking is to be applied then it should meet the same criteria as cylinder gas checks and at a minimum have two (2) points, the first being Zero (TZG). NOTE: The biggest issue with any calibrator is that they are affected by local climate conditions which include temperature, humidity and ambient pollutant conditions. With respect to the later; filtered (carbon) air is used for the blending but there is no method of checking the filtered ambient air (typically it is replaced on a routine time schedule). The problem with this methodology is that the ambient air levels of pollutants vary day by day, especially for gases such as HCl and Hg, where just mixing the coal in the yard could easily produce high transient pollutant conditions.                                                        If Spiking is to be used as a means of validation then a Zero Gas spike must be incorporated.
5.2 Concentration Accuracy Auditing Requirements:  ".... Once each calendar quarter by dynamic spiking (DSA), cylinder gas audits (CGA) or other acceptable alternative..." implies that DSA is the standard which is absolutely incorrect. The fact is that DSA is a new alternative method and PS-18 is actually just defining it. Based on the existing Appendix F standards CGA's are the `gold standard' and this should be reflected in this document, not contradicting itself. I would propose that this statement read "...Once each calendar quarter by cylinder gas audit (CGA) or other acceptable alternative such as dynamic spiking audits (DSA)..."                                                                                                               NOTE: if the present language is left as is, EPA can expect that there will be a large number of requests from industry and analyzer manufacturers that this same method be acceptable for standard gases such as NOX, SO2, CO, CO2, and even O2. If this is the case then EPA should expect a very labor intensive process of changing all the existing Performance Specifications to reflect this.
      5.2.2.1 Perform a CGA: "... known concentrations within the following ranges..." the table should be as follows to conform to the standards as defined by the existing Appendix F:                  
                                  Audit Point
                                  Audit Range
                                       1
                    Zero (TZG preferred) or 10-20% of span
                                       2
                                40-60% of span
                                       3
                                70-100% of span
      
      5.2.3.1 Preform a CGA with DSA: reference Appendix A to PS-18 of Appendix B  -  this does not exist in the present (rev3) of PS-18. Therefore my comment will be directed simply to the fact that DSA must meet the same 3 point criteria as a standard cylinder gas method, i.e. 3 points as follows:
                                  Audit Point
                                  Audit Range
                                       1
                    Zero (TZG preferred) or 10-20% of span
                                       2
                                40-60% of span
                                       3
                                70-100% of span

5.5 Criteria for Optional QA Test Frequency: this entire section should be removed from the document. Simple logic dictates this for the following reasons. The method of monitoring HCl has been documented by the EPA as a "higher degree of difficulty" to measure than regular pollutants such as NOX, SO2, and CO and the concept of allowing this measurement to have less restrictive testing has no good engineering logic.                                             NOTE: if the present language is left as is, EPA can expect that there will be a large number of requests from industry and analyzer manufacturers that this same method be acceptable for standard gases such as NOX, SO2, CO, CO2, and even O2. If this is the case then EPA should expect a very labor intensive process of changing all the existing Test Methods to reflect this.
      5.5.1 "If the CEMS is in control for eight consecutive quarters that include a minimum of two RATAs, the source owner or operator may revise their auditing procedures to use CGA or DSA each quarter for eight subsequent quarters." As stated in 5.5 comment, this should be removed.
      5.5.2 "The source owner of operator must perform at least one RATA that meets the acceptance criteria every 2 years." They must perform every year. As stated in 5.5 comments, this should be removed.
5.6 "...owner or operator fails a RATA, CGA or DSA..." this should be removed as it is specific reference to 5.5 and 5.5 should be removed as defined above.

						

                                       
