Multi-Sector Air Pollutants Regulations, response to comments: part 2

Part 2 - Engines


3.1 The Department received a number of comments from stakeholders about the use of the word “original”, to describe engines manufactured before the coming into force date. Stakeholders also expressed their concerns on the types of documents that were required to prove that an engine is “original”. These documents may no longer be available and thus additional flexibility was requested to allow owners and operators to demonstrate that an engine is “original” using any acceptable document. Finally, it was suggested that engines be considered “original” by default unless proven to be “modern”, whereas the proposed Regulations considered all engines as “modern” until proven to be “original”.

Response: The regulatory text has changed as a result of these comments. The term “original” is replaced by “pre-existing” to provide clarity. The definition of “pre-existing” no longer specifies which documents are needed; however, engines are still considered modern unless they are proven to be “pre-existing”. Typically, regulatees should have to prove that their engines are eligible to meet less stringent limits, not more stringent ones. Had the proposed modifications been incorporated, less stringent emission requirements would have applied by default. The burden would have been placed on enforcement officers to prove that an engine is modern and subject to more stringent limits. This proposal would have resulted in higher oxides of nitrogen (NOX) emissions.

3.2 Stakeholders commented on the definition of “rated brake power”. In the proposed Regulations, the rated brake power was to be demonstrated using the information on the nameplate provided by the engine manufacturer or, in the absence of the name plate, another document provided by either the engine manufacturer or a governmental organization that also indicated the serial number of the engine. Engine manufacturers have indicated that the rated brake power can be determined based on engine make and model and the date of manufacture, so the serial number is not always necessary to make this determination.

Response: The regulatory text has changed as a result of these comments. The Regulations do not prescribe which document must be used as proof and instead allows operators to use the rated power as designated by the manufacturer either on the engine name plate or otherwise.


3.3 Stakeholders requested that engines used in the aluminium, petroleum-refining and forest product sectors or engines driving generators be excluded from the scope of the Regulations. Other stakeholders supported more comprehensive coverage and indicated that all modern engines should be subject to the Regulations: not only those used by industrial sectors, but also engines used for commercial, institutional and agricultural applications.

Response: The Regulations cover modern engines located at industrial facilities, regardless of whether the engine drives a generator, a compressor or a pump. This decision is the result of extensive discussions with industry, provincial and territorial governments and non-governmental environmental organisations as part of the Base-Level Industrial Emission Requirement (BLIER) discussions. The Air Quality Management System (AQMS) framework did not address emissions from, nor consult with, commercial, institutional and agricultural users of engines. For modern engines, the performance standards are comparable to current United States Environmental Protection Agency New Source Performance Standards (U.S. EPA NSPS) for Stationary Spark Ignition Internal Combustion Engines (which cover non-industrial engines as well), which have been adjusted for Canadian conditions such as the weather and the location of engines. They are also equivalent to British Columbia’s requirements, the jurisdiction with the most stringent limit for engines in Canada.


3.4 Stakeholders commented that a two-year delay should be granted to meet the requirements of the Regulations when engines are acquired from other owners. This would allow time for the new owner or operator to retrofit engines to meet the yearly average for the expanded fleet, schedule performance testing and register the newly-acquired engines. In the proposed Regulations, the responsible person had to ensure that all engines in their fleet met the requirements, regardless of when they were acquired.

Response: The regulatory text has changed as a result of these comments. The changes provide the flexibility needed to determine the dates on which performance testing is required for recently-acquired engines and to retrofit engines that are not already equipped with control technology. While a two-year delay would have created problems with the administration of the Regulations by allowing a significant number of engines to avoid the regulatory requirements, resulting in increased emissions, the following flexibilities have been added:

  1. When an engine is acquired, any performance test that would normally have been required may be postponed, but must be completed no later than the end of the three-month period following the acquisition. This was added to give the responsible person time to transfer records and to schedule testing for newly-acquired engines.
  2. A responsible person who chooses the yearly average option has three months to designate the newly-acquired engines as belonging to a subgroup and to assign the emission-value to be used to calculate the yearly average, both of which will be retroactively effective as of the date of acquisition.
  3. Newly-acquired engines that have never been equipped with an emission control system that ensures emissions ≤ 10 g/kWh have a nine-month grace period from all engine requirements in the Regulations. If the engine is not retrofitted and registered by the end of the grace period, the engine may be found out of compliance retroactively. This provides owners and operators the time needed to identify the engines requiring modifications, as well as to perform those modifications.

3.5 The proposed Regulations established less stringent obligations for engines identified as low-use by the owner or operator. The election of an engine as low-use engine could only be done once to avoid the possibility of repeatedly switching between regular-use and low-use designations, thereby taking advantage of less stringent obligations even when using an engine on a regular basis. Stakeholders suggested that an owner or operator should be allowed to designate an engine as low-use engine once every three years. Stakeholders also suggested that the designation of an engine as low-use be allowed at any time during the year, provided that the engine was used as low-use for the first portion of that year.

Response: The Department has changed the regulatory text as a result of these comments to provide additional flexibility. The regulatory text now allows a responsible person to designate an engine as low-use once every three years, and this designation may occur at any time during the year. By limiting the frequency of designation to three years, concerns about switching between engine usage types are minimized.

3.6 The proposed Regulations offered owners or operators the option to replace one or more pre-existing engines that have been removed from their group with one or more modern replacement engines or with one replacement turbine or one replacement electric motor, in order to meet their yearly average limit. For each replacement that would take place, there was a time limit of one year, starting from the removal of the pre-existing engine from the group, in order to include the modern engines, the turbine or the electric motor. For each replacement that would take place, the proposed Regulations required that the modern replacement engines have a total power equal to or lower than the total power of the engines being removed from the group. This limit was designed to avoid situations where a large modern engine replaces a small pre-existing engine, thereby reducing the yearly average without a real reduction in emissions overall.

Stakeholders requested the removal of this limitation on the power of modern replacement engines included in the yearly average.

Response: The Department has changed the regulatory text as a result of these comments. The term “replacement units” now includes all modern engines, turbines, or electric motors that replace pre-existing engines in a group for the purpose of the yearly average. Although there is now a limit of maximum power for all replacement units, there is more flexibility in how to meet this limit. The Regulations no longer have a power limit for each individual replacement that would take place. Instead the requirements apply to all replacements, considered in combination rather than as individual actions. In effect, the concept of a power “bank” has been introduced. Each time a pre-existing engine is removed from the group, its rated power is added to the “bank” The total rated power of all replacement units may not exceed the rated power in the “bank”. The Department also removed the time limit of one year for the replacement of a pre-existing engine, as this power limit must be met at all times for all replacements.

The replacement provisions in the Regulations are designed to provide owners or operators the flexibility to manage their group of pre-existing engines while not compromising the environmental benefits of the Regulations. Removal of the power limit would have allowed owners or operators to meet their compliance obligations without reducing the emissions from their group of pre-existing engines.

3.7 Stakeholders suggested that the default emission-values for replacement turbines be aligned with the emission standards agreed to by the AQMS Turbine BLIER working group. It is planned that these emission standards will be included in future guidelines for combustion turbines. This working group was composed of provincial, territorial, federal, industry and non-governmental organizations (NGO) representatives. In the proposed Regulations, replacement turbines used higher emission-values based on the emission standards of the 1992 Canadian Council of the Minister of the Environment (CCME) National Emission Guidelines for Stationary Combustion Turbines (December 1992 - CCME EPC/AITG-49E).

Response: The regulatory text has changed as a result of these comments. The default emission-values for replacement turbines are replaced by values corresponding to the emission standards developed by the Turbine BLIER working group and are used in the calculation of the yearly average for turbines capable of meeting those emission standards (determined using manufacturer information).

The emission standards for combustion turbines were developed based on the specifications of currently manufactured turbines. As some existing turbines may not meet the BLIER NOX standards for new turbines, the inclusion of older combustion turbines in the yearly average is restricted.

3.8 Stakeholders commented on the testing requirements introduced in the proposed Regulations. The regulatory text proposed that performance testing be done in accordance with U.S. EPA methodologies, but not at the same frequencies. The frequency of these tests depended on whether the engine was lean-burn or rich-burn. The comments reflected opposing positions and suggestions included: less frequent testing of engines that have passed two previous tests, alignment of testing frequency with the U.S. EPA in order to have the same testing frequency for rich-burn and lean-burn engines, an increase to the testing frequency for lean-burn engines and exemptions from testing for engines in remote locations due to the challenge of transporting calibration gases to these sites. Some stakeholders also suggested less frequent performance testing if owners or operators choose to implement more frequent, but less rigorous, emission checks between performance tests.

Response: The regulatory text has changed as a result of these comments. The Department developed a solution which offers testing options to industry without compromising the environmental benefits of the Regulations.

The approach for rich-burn engines provides default and optional testing frequencies. The default is unchanged from what was published in the proposed Regulations: performance testing is required every nine months or 4,380 operating hours, whichever comes first. Optionally, performance testing may be extended to 36 months or 8,760 operating hours, whichever comes first, provided that quarterly (less rigorous) emission checks are performed. The emission check requirements were based on consultations with manufacturers of portable analyzers and testing experts. These changes reduce the number of performance tests by approximately 20% compared to what was proposed in Canada Gazette, Part I (CGI). Performance tests are more onerous than emission checks as they take longer to be performed and the methodology is more rigorous and requires more expertise. Emission checks are already conducted by some companies to make sure that engines are properly operated and maintained.

The approach for lean-burn engines is also a compromise to address comments that recommended increasing the testing frequency for lean-burn engines and comments that suggested less frequent testing for all engines. Unlike the CGI text, pre-existing lean-burn engines subject to the per-engine limit must be tested. Also, a baseline test is required for pre-existing lean-burn engines, prior to the assignment of a default emission value of 4g/kWh, to demonstrate that the engine is capable of emitting less than this value. The frequency of subsequent tests for lean-burn engines remains unchanged (every 36 months or 17,520 operating hours, whichever comes first). In addition, emissions monitoring of NOX, oxygen and CO must be performed annually for lean burn engines with a rated power of at least 375 kW, replacing the annual measurement of the oxygen concentration in the exhaust gases that was proposed in CGI.

Though this approach does not align with the testing frequencies used by the U.S. EPA, experts in Canada and the U.S. consulted by the Department indicated that the approach in the Regulations addresses the varying ability of rich-burn and lean-burn engines to maintain low emissions.

As for the exemption for testing engines in remote locations, no change was made to the regulatory text because several testing methodologies are permitted by the Regulations, some of which are already suitable for testing in remote sites as a sample can be extracted and brought to a laboratory with no need to transport calibration gases to these sites.

Operation and Maintenance

3.9 In the proposed Regulations, the responsible person had to install a non-resettable hour meter on each engine to be considered as low-use and on each engine included in the calculation of the yearly average. Stakeholders requested more options by which to determine the number of operating hours in a year. It was suggested that operator logs or supervisory control and data acquisition (SCADA) systems should be allowed to determine the number of operating hours and that, in a case where calculation of operating hours was not possible, a default number of hours could be used for all engines included in a yearly average calculation. It was also requested that less frequent readings be required, moving from two to one per year, if a non-resettable hour meter was used.

Response: The regulatory text has changed as a result of these comments. The Department will now accept data from several sources for determining engine operating hours. When calculating the yearly average, new accepted sources of information include operator logs, automated data acquisition systems such as SCADA systems and replacement of hours with a number of days. This additional flexibility will not compromise the environmental benefits of the Regulations as the operating hours are only used to calculate the yearly average, giving more importance to engines that are used more often and the sources of information added are sufficiently accurate for this purpose.

For low-use engines, the Department has only added SCADA systems as a new accepted source of information. More accurate measurements are needed as engines must operate during less than 1,314 hours in each period of three calendar years (not including emergency operation) in order to be considered as “low use”. Also, for enforcement purposes, more verification is needed to ensure that an engine deemed to be a low-use engine is, in fact, eligible to be exempt from a number of other provisions in the Regulations.

If non-resettable hour meters are to be used, one annual reading, as close as possible to being one year apart, can now be used to provide the data needed, which will impose less of a burden to industry than the twice per year requirement of CGI.

Registry, Reporting and Recording of Information

3.10 Stakeholders commented that the last owner or operator of an engine within a reporting year should be the one responsible for reporting the emissions of that engine for the entire year. This was suggested in order to reduce the administrative burden generated by the submission of multiple reports for the same engine.

Response: No changes to the regulatory text arose as a result of this request because this would create both legal and compliance issues. Making changes to accommodate this request would not significantly reduce the administrative burden, as the information to be submitted for an engine is the same regardless of who submits the report. The owner or operator who conducts a performance test on an engine during a year must report the result in the annual report for that engine, regardless of when the performance test occurs. In the case where the owner or operator of a modern engine changes partway through the year, any owner or operator who has not conducted a performance test or an emission check on the engine during the year would only indicate that they have no information to report on this engine.

3.11 Some stakeholders expressed their concern that it is sometimes impossible to find the serial number of an engine. As a solution, it was suggested that the Department should assign a unique government-issued identification number to all engines. In the proposed Regulations, the serial number was needed to identify each engine in the engine registry.

Response: The Department has made changes to address these comments. The Department will issue a unique alphanumeric identifier for engines that do not have a known serial number. Either the serial number or this unique identifier must be indicated on a name plate permanently affixed to the engine in a visible location.

Cost Benefit Analysis

3.12 Industry stakeholders commented that modern engines projected in the CGI analysis was high and that they have a surplus of engines in storage which would be used to replace engines at their end of life.

Response: In the CGI analysis, no surplus engines were accounted for in the analysis. Based on information provided by industry, the analysis is updated to assume that an additional 10% of the engine inventory is held in storage. It is assumed these surplus engines are used for engine replacement as required and that they have the same emission performance characteristics as those they replace. As a result, fewer modern engines are expected to be purchased since surplus engines are available to be installed.

3.13 Industry commented that the number of modern engines projected to replace pre-existing engines in the CGI analysis was high given the installation dates and lifespan of the pre-existing engines.

Response: Assumptions used to determine the demand for modern engines to replace pre-existing ones have been revised. In the CGI analysis, pre-existing engines in the inventory were assumed to be replaced at a constant rate, resulting in a projection of 2,801 modern engines to replace pre-existing engines. This assumption has been updated based on sales data provided by engine manufacturers in summer 2011 that shows approximately 66% of the pre-existing engines were installed after 1995 and are unlikely to be replaced prior to 2035. The updated analysis assumes that only 34% of the pre-existing engines in use are expected to be replaced at a constant rate based on the useful life of each engine model. In the updated analysis, it is also assumed that engines would be replaced with pre-existing surplus equipment in storage before a new one is purchased. As a result, the updated projection for modern engines is 277.

3.14 Comments were received regarding the useful life of engines provided by Accurata Inc. of 20, 40 and 60 years based on engine speed in revolutions per minute (RPM). Stakeholders indicated they had many engines approaching 40 years of age that they did not intend to replace.

Response: The useful life of engines depend on many factors, including but not limited to engine obsolescence, fatigue, catastrophic failure, and maintenance practices. Due to insufficient data on these specific factors at the engine level, no changes have been made to the useful life assumption. Also, during a September 2014 meeting with industry stakeholders, it was generally agreed that the Accurata Inc. assumptions about engine life were acceptable based on averages. The useful life of low speed, medium speed, and high speed engines is retained for the analysis, which is 60, 40, and 20 years, respectively and are replaced at a rate of 1.7%, 2.5%, and 5%, respectively.

3.15 Stakeholders commented that they were not realizing a 5% fuel savings attributed to engines retrofitted with rich-to-lean engine management systems to operate at 2.7g NOX/kWh, and that rich-burn engines do not consume 19-29% more fuel than modern lean-burn engines with pre-combustion chambers.

Response: For CGI, Accurata Inc. provided an estimate of fuel savings of 10% for engines retrofitted with rich-to-lean engine management systems to operate at 2.7g NOX/kWh. The CGI Engine cost benefit analysis (CBA) used a more conservative value of 5% fuel savings. Based on a review of technical expert reports and information gathered through BLIER working group discussions, the current analysis retains the assumption of 5% fuel savings on average.

To account for cases in which rich burn engines are replaced by modern lean burn engines with pre-combustion chambers to meet regulatory requirements, the CBA has been updated by reducing fuel savings from 19%-29% range used in CGI to 10%-24% range. Modifications to the fuel savings range used in the analysis is based on a review of studies and manufacturing specifications.

3.16 One comment questioned higher than expected health benefit estimates for Saskatchewan given the rural location of equipment covered by the Regulations.

Response: A large proportion of the NOX emissions from the regulated equipment occur in remote rural locations.  However, atmospheric dispersion and transformation of NOX is such that emissions even in remote regions or occurring up-wind can impact air quality across a very wide geographic area.  Under the Regulations, emission reductions will be most significant in rural areas where the majority of the regulated equipment is located.  As a result, air quality improvements will be most significant in those same areas.  Air quality is, however, expected to improve across most of the province.  In higher density regions of the province, even small air quality improvements can result in significant health benefits. The majority of the health benefits are expected to be experienced in Alberta, followed by Saskatchewan. However, ambient air quality is affected by air emissions which can travel across provincial borders, following the prevailing winds that blow typically from West to East. As a result, Saskatchewan is able to benefit from the reduction of NOX emissions in Alberta.

3.17 Industry commented that since pipelines no longer operate as monopolies, and must compete with other pipelines and modes of transportation, it should not be assumed that costs can be passed on to customers.

Response: Pipeline tolls are regulated by either the National Energy Board for interprovincial and international pipelines or provincial energy boards for intra-provincial pipelines. Firms are generally allowed to recover increased operating or regulatory costs and achieve a regulated rate of return on pipeline assets. Therefore, the Department notes there is a possibility for firms to pass on costs.

3.18 Comments and information on capital costs were received from stakeholders who commented that capital cost for rich-to-lean engine management system technology should be higher and that for non-selective catalytic reduction (NSCR) it should be lower. A cost range of $150k-$250k, was provided for installing rich-to-lean engine management system technology. A cost range of $60k-$120k was also provided for installing NSCR. These costs were based on a breakdown of materials and labour required to purchase and install the technologies.

Response: In the CGI analysis, capital costs used were taken from a report prepared for the Department by Calgary-based consultancy Accurata Inc. For similar engine models, the Accurata cost estimates for equipping a sample engine model with a rich-to-lean engine management system ranged from $125k to $160k, and the cost to convert to NSCR ranged from $125k to $165k. The Department shared the cost breakdown with industry in September, 2014 and received cost estimate information in response.

Based on the cost estimate information provided by industry, the Department has increased cost estimates for rich-to-lean engine management system technology by 40% relative to CGI estimates due to higher assumed costs related to the existing control system. NSCR capital cost estimates have also been reduced by 40% based on lower cost estimates for exhaust system modifications and field installation. These cost changes are applied to all engine model retrofit options in the inventory. The changes bring the capital cost estimates for rich-to-lean engine management system technology and NSCR into the ranges provided by stakeholders.

3.19 Stakeholders commented that the annual incremental maintenance costs for a NSCR system operating at 2.7g/kWh were very high and the cost of waste gate repair was omitted in the cost estimation for rich-to-lean engine management system.

Response: Incremental maintenance costs for NSCR systems were estimated to range from $20k to $28k in the CGI Engines CBA. This range was provided by Accurata Inc. based in part on the assumption of quarterly on-site engine tuning. This estimate has been decreased by 50% relative to CGI estimates based on a review of other studies and assuming that some engine tuning could be conducted in conjunction with other routine maintenance tasks. In addition, annual maintenance cost savings for rich-to-lean engine management systems, assumed to be $15k in the CGI, have been decreased by $10k per unit to include the material and labour costs for waste gate repair. Inclusion of waste gate repair brings maintenance savings down to $5k per retrofit in this analysis.

3.20 Multiple comments expressed a view that the costs and number of hours allocated to administrative requirements were low.

Response: The Department has reviewed administrative requirements of the Regulations and consulted with industry to provide more information regarding the calculation of administrative costs. The resulting changes in administrative costs include the addition of ongoing activities related to submission, verification and certification of information to be provided in submissions to the engine registry and annual reports.

Other changes include addition of costs related to the familiarization with the administrative requirements of the Regulations, maintaining and submitting records, and a reallocation of costs to calculate performance test results and emission averages. As a result, time assumed to be required to complete administrative requirements is increased compared to the time allocated in CGI. Wage rates have also been adjusted for some tasks to align with technical staff wages rather than administrative staff. The total impact of the above changes is an increase in the total administrative cost by $6.6M.

3.21 In the CGI engines CBA, 65.6M MMBtu of processed gas was estimated to be saved as a result of the Regulations. A constant natural gas price of $4/MMBtu was used to monetize the quantity of fuel saved. Industry commented that the assumption of a constant natural gas price of $4/MMBtu should be revised. The National Energy Board 2013 Energy Market Assessment indicates that the natural gas price in the reference case is assumed to increase from U. S. $3.90/MMBtu in 2013 to U. S. $6.20/MMBtu in 2035 (in 2012 dollars).

Response: The updated analysis estimates 84.6M MMBtu of fuel savings; however, own-use fuel reported by industry is typically categorized as raw gas, rather than processed gas. Monetizing raw gas is a challenge given that limited information is available to determine the value added between the point where fuel savings occur and the point of sale. As a result, fuel savings benefits are not monetized in this analysis. This change could omit significant benefits and as such, monetized benefits should be considered conservative.

3.22 A firm requested that compliance costs to small business be mitigated.

Response: The Department has assessed this request. Following calls to about 250 small businesses in the oil and gas sector during summer 2014 and several outreach efforts since fall 2012, the Department has not identified any small businesses that would be affected by the requirements for pre-existing engines. Recognizing the possibility that some small businesses that own or operate engines above the size threshold have not self-identified or that larger engines may be owned or operated by small businesses in the future, an explicit exclusion from the requirements for pre-existing engines has been added to the Regulations for small businesses. Costs are mitigated for small businesses that meet the exclusion thresholds and submit the required information in order to exclude their pre-existing engines from requirements.

Page details

Date modified: