E [C]/R Incorporated
                  Providing Environmental Technical Support Since 1989




MEMORANDUM   

DATE:		February 11, 2011

TO:		Melanie King, EPA OAQPS/SPPD/ESG

FROM:	Tanya Parise, EC/R, Inc.

SUBJECT:	Summary of the January 26, 2011 Meeting with API Regarding CPMS Requirements in the RICE NESHAP

--------------------------------------------------------------------------------


INTRODUCTION

      The American Petroleum Institute (API) requested this meeting to follow-up on earlier meetings with EPA.  Summaries of the two previous meetings between EPA and API on this issue are available in the docket (EPA-HQ-OAR-2008-0708-0693 and 0696).  API wanted to discuss the specific issues it has with the continuous parameter monitoring system (CPMS) requirements in the National Emission Standards for Hazardous Air Pollutants (NESHAP) for stationary reciprocating internal combustion engines (RICE).  The most recent amendments to the NESHAP for stationary RICE were published in the Federal Register on August 20, 2010 (75 FR 51570).  
      
ATTENDEES

Melanie King, EPA
Jeff Adams, BP
Michael Horowitz, EPA (by phone)
Jeff Panek, IES
Peter Westlin, EPA
James Harrison, Exterran
Barrett Parker, EPA
Rebecca Rentz, Bracewell & Giuliani (by phone)
Jason Dewees, EPA
Tom Monahan, ExxonMobil (by phone)
Colin Boswell, EPA
John Wagner, API (by phone)
Tanya Parise, EC/R
Clay Freeberg, Chevron (by phone)
Matt Todd, API
Jenny Yang, Marathon (by phone)
John Admire, DCP Midstream
Bill Wehrum, Hunton & Williams (by phone)
Rex War, DCP Midstream
Jim McCarthy, IES (by phone)














SUMMARY OF MEETING 

	API expressed that it wished to have a comprehensive discussion with EPA to explore in detail possible solutions to the issues related to the CPMS requirements.  API provided a rough agenda for the meeting consisting of the following, which is loosely followed in this meeting summary:

   * Introduction and Purpose
   * Major Source CPMS Performance Specifications
         o Temperature Monitoring
         o Data Availability, Validity and Averaging
   * Area Source Monitoring and Compliance Assurance Concerns
         o Unmanned rural area source technical challenges and issues
         o Alternatives to continuous monitoring at area sources
	
	API reiterated that it has raised significant issues in earlier meetings related to the CPMS requirements in the rule and it is particularly urgent to resolve these issues as they relate to major sources.  API's request for an administrative stay on the CPMS requirements was granted; however, the stay will run out shortly and API indicated it is primarily focusing on trying to resolve the CPMS issues in time for major sources to comply with the rule.  

Temperature Monitoring

	Language currently in the rule requires that sources conduct a temperature calibration check at least every 90 days, and API indicated that there is a lot of uncertainty about that requirement.  API expressed that the thermocouple either works or not, and although over time it might fail, it is more important, in API's opinion, to verify the performance of temperature measurement equipment as opposed to calibrating the thermocouple.  EPA said that redundant thermocouples could be used for the calibration and further said that it seems like it would be appropriate to allow the owner/operator the discretion to determine what temperature calibration means to use.  API noted that the language in the rule might have been clearer if the requirement stated that sources needed to "verify performance."  Further, API said that if sources have the flexibility in their monitoring plans to define how to verify devices this might be something that would be appropriate in covering a variety of different scenarios.  API also expressed that the requirement to conduct performance checks quarterly is excessive because temperature measurement technology is stable.  API added that the personnel needed to conduct performance checks would be different than the normal staff/operator and would necessitate a technical expert.  

	EPA asked what API proposed as an alternative period.  API responded that though the temperature might drift, it is generally stable and therefore proposed a year, reiterating again that quarterly/every 3 months is too much in API's opinion.  


Data Availability

	API argued that the 100 percent data capture requirement is too high and is higher than what is required in other rules.  EPA noted that some situations are excluded from data capture. For instance, data during malfunctions, calibrations checks, and repairs are not required to be collected.  Therefore, since these are exceptions, it is not a 100 percent data capture, EPA said.  The goal is to keep the monitoring system operating at all times when the equipment is running, EPA emphasized.  API added that it may have periods without data for a number of reasons, e.g., loss of power, and may have deviations beyond its control that it might not be able to document.  In other words, there may be situations where data is missing, but the cause for the data gap may remain unknown.  API was concerned that missed 15-minute periods could lead to Title V permit deviations and asked EPA to allow compliance to be demonstrated without having all periods.  EPA asked what API recommended for an alternative.  In response, API proposed that perhaps two 15-minute periods would be an appropriate data capture amount to represent 1 hour.  API further recommended that EPA look at the 4-hour rolling period only and consider allowing a certain number of data gaps.  Specifically, API suggested that a valid 4-hour average consist of 3 hours worth of data or that a valid 1-hour period consist of 30 minutes of data.  

Area Sources

	API asked for more flexibility in demonstrating catalyst activity for area sources under the rule and recommended an assurance method to indicate that the catalyst is working in conjunction with carbon monoxide (CO) portable measurements.  API indicated that it believes such methods would provide a good indication of catalyst performance.  EPA asked what API specifically recommended and API suggested a work practice standard or a limit with a work practice standard for compliance purposes.  API expressed that it would like to move away from reference method stack testing (Method 320/FTIR) since there is no portable option to measure formaldehyde.  EPA noted that the Engine Manufacturers Association (EMA) has asked EPA to consider using total hydrocarbons (THC) as an indicator of catalyst performance or having THC as a limit.  API indicated that it has reviewed the information from the EMA, but said that the information from EMA focused on THC because, according to API, EMA did not believe that EPA was open to the idea of allowing the use of CO.  In fact, according to API, CO is more indicative of catalyst activity and how well formaldehyde is being reduced than THC.  Further, API added that portable analyzers do not measure THC, it is more problematic to measure THC emissions, and methane dominates the THC measurement in engines burning natural gas.  API clarified that it was not suggesting that CO be used as a surrogate; only as a pollutant that would show catalyst activity after which the source would follow-up with a performance test or replacement of the catalyst.  

	EPA asked about the frequency of performance testing, specifically what API thought of quarterly and semi-annual tests.  API responded that it would prefer annual testing, but was okay with semi-annual testing as well, adding that coordinating the timing of performance testing becomes important with a quarterly test schedule.

	API expressed concerns with the pressure checks, which are required to be performed monthly, because during these checks, operators also have to conduct a load determination, which is especially problematic at area sources.  API made the point that if a CO test is being done, that would essentially do the same thing as what the pressure drop is intended to do, i.e., indicate catalyst activity.  

	As it relates to area sources, API reiterated from previous meetings with EPA that there are problems with the CPMS requirements because of communication issues, end-of-line power issues, and so on. 

	The meeting closed with API indicating that it would provide specific material to EPA in short order with recommendations on CPMS language that API believes would be appropriate.


