Effectiveness of Refueling Vapor Recovery Still Up In The Air

Joe Totten hunts for test data to technically support the current US government policy on Onboard Refueling Vapor Recovery and Stage II systems.

Wanted: Data. Real data. Reliable data. And lots of it. Data that would show how well the federally mandated refueling vapor recovery systems work in the real world of vehicle refueling. A world made up of divergent gasoline dispensing systems, located in areas with drastically varying climatic conditions and characterized by wide-ranging fuel storage and turnover volumes. A world in which decisions (and supporting data) affecting mandates and related investments in vapor recovery equipment are scrutinized by myriad experts, each with their own interests and strong opinions about how things ought to be.

The want-ad language at left might well describe the wishes of many people who are responsible for, or affected by, the refueling vapor recovery dilemma. The dilemma is how to implement the federally mandated double-barreled attack on vehicle refueling vapors: Onboard Refueling Vapor Recovery (ORVR) equipment on vehicles and Stage II vapor recovery systems at service stations.

Had either one of these weapons been mandated exclusive of the other, the regulatory agencies and affected industry segments still would have been faced with difficult tasks. These tasks include not only regulating the design and use of needed vapor recovery systems, but also establishing the technology and equipment for in-use testing and monitoring of the systems to ensure compliance and evaluate whether the mandated objectives are being achieved.

The enforcement and compliance tasks were made doubly onerous when the same piece of federal legislation mandated both ORVR and Stage II vapor recovery. Starting with the inclusion of ORVR equipment in model year 1998 passenger vehicles, these two different technologies must coexist at least until 2010, and probably well beyond that year, before Stage II will no longer be required. And to make the tasks even harder, when ORVR-equipped vehicles are refueled at some stations with Stage II systems, the overall effectiveness of the vapor recovery effort can be significantly reduced (more on this point later).

Why the data push?
For any regulatory enforcement program to work effectively and efficiently, the program must have certain essential components. Based on my experience in assessing the effectiveness of federal regulatory programs for the US General Accounting Office, I can tell you what I looked for in a good enforcement program. It should include the following:

1. A clearly identified objective, such as preventing work-related accidents, or, in our immediate case, limiting refueling vapor emissions into the atmosphere. 
2. Uniform mandatory standards regarding the equipment, devices, systems and procedures that must be complied with to achieve the stated objective. 
3. An inspection program, including inspectors, equipment and test procedures, for determining whether or not affected items or operations comply with the standards. 
4. Authority to enforce compliance through civil or criminal penalties and through “shut-down” or “cease and desist” orders. 
5. Feedback and analysis of inspection and testing results to (1) measure the extent of compliance with the standards; (2) evaluate whether compliance with the standards is accomplishing the stated objectives; and (3) identify the need for improving any of the foregoing program components (including the standards themselves and the compliance testing procedures and equipment) to better accomplish the overall purpose of the program.

Item five (above) is where the need for actual performance data becomes critical. Are the mandates for vapor recovery devices, equipment and procedures being complied with? Are the devices, equipment and procedures actually accomplishing what they are supposed to do? While EPA and CARB may have reasonably good answers to the first question, data that would answer the second question is currently not available and, as things are now going, may not be available for years to come.

In the meantime, however, more and more vehicles are being equipped with mandated ORVR systems which, by model year 2006, will be mandated for 100 percent of new passenger cars and light trucks. The mandates for Stage II systems will continue at least until the ORVR-equipped vehicles are “in widespread use.” Today, after three decades of debate and one decade of mandates, the actual effect of ORVR and Stage II systems on refueling vapor emissions is unknown and will remain unknown indefinitely.

Update on EPA testing of ORVR
As reported in my Nov/Dec article on the ORVR, EPA officials identified the lack of in-use testing of ORVR-equipped vehicles as the major issue on ORVR’s effectiveness (“ORVR: Just Where Is It Taking Us?” page 25). They said that such tests will be required in “several years.”

In February 1999, I asked EPA officials why in-use testing was not started immediately after ORVR systems were mandated (beginning with 40 percent of model year 1998 passenger cars). I also asked when EPA planned to start such testing. Agency officials advised in March that EPA still had not performed any performance audits or recall testing on ORVR equipped vehicles and that:

  “We do not reveal our compliance testing plans in advance. We would prefer that manufacturers be prepared for any type of audit at all times.

“Although in-use testing will be required in several years, certification testing is performed before the introduction of the vehicle into the market. EPA-in use testing begins 3 years after the introduction of a certain model-year vehicle. As stated above, EPA’s recall program has an excellent record of assuring in-use compliance.”

While agreeing that the lack of in-use testing of ORVR-equipped vehicles is a major issue, EPA officials seem to be placing at least as much, if not more, importance on certification testing. As indicated in the above quote, certification testing is done before the vehicles are introduced into the market. While certification testing may be useful and important, such testing does not prove that the ORVR equipment will perform adequately in actual field conditions. This is particularly important now in view of the different kinds of dispensing systems being used today—conventional (non-Stage II) systems and different kinds of Stage II systems. Also, since EPA testing procedures for ORVR systems do not provide for testing with Stage II systems, it is not clear how “assuring in-use compliance” via EPA’s recall program will provide the needed data.

The lack of in-use testing of ORVR-equipped vehicles leaves much to be assumed about their effectiveness. In this regard, I wonder just how EPA will execute its authority to eventually phase out the Stage II vapor recovery mandates.

Will “widespread use” equal vapor recovery?
The legislation mandating ORVR and Stage II (the Clean Air Act Amendments of 1990) authorizes EPA to revise or waive the application of Stage II requirements for areas classified as serious, severe or extreme for ozone after the required ORVR systems are in widespread use throughout the motor vehicle fleet.

Given that the “required” ORVR equipment must limit refueling vapor emissions to 0.2 grams of hydrocarbons per gallon of fuel dispensed, I asked EPA officials:

  • If they agreed that EPA cannot revise or waive the Stage II requirements until it determines that the ORVR-equipped vehicles are in widespread use and that the ORVR equipment is working to limit refueling vapors to the specified limit.
  • How EPA will determine whether the ORVR equipment in widespread use is limiting the refueling vapor emissions to the specified limit.
  • If these determinations by EPA will involve only non-Stage II dispensing systems and, if so, how EPA will know what happens to refueling vapors when ORVR equipped vehicles are refueled at service stations equipped with Stage II systems.
  • If, upon EPA’s decision to revise or waive the Stage II requirements, the states will be required to make the same revisions or waivers.

In response, EPA officials provided the following answers:

  “We agree that EPA cannot revise or waive the Stage II requirements for specified areas until it determines that ORVR-equipped vehicles are in widespread use. EPA is assured that the equipment works effectively at the time of certification because of the tests that must be performed before certification. EPA’s recall program has an excellent track record of assuring in-use compliance and we have no reason to believe that the recall process will work any differently for ORVR vehicles.”

“Our assessment of the effectiveness of ORVR will include data from certification, recall and manufacturer in-use tests.”

“Currently, EPA’s test procedure is performed using conventional nozzles. However, EPA is still examining the possibility of requiring tests to be performed on Stage II nozzles.”

“States can still use Stage II as an additional pollution prevention strategy; however, SIP [State Implementation Plan] credits will likely depend on the fleet turnover of ORVR vehicles for different areas of the country.”

Several things about the above answers bother me: First, the answers avoid the question of whether EPA can legally revise or waive Stage II requirements without determining that ORVR-equipped vehicles are, in fact, limiting refueling vapor emissions to the specified level.

Second, EPA seems to have unwavering faith that the certification tests (performed before the vehicles are marketed) assure that the ORVR equipment will work effectively.

Third, EPA’s general references to its recall program, as having an excellent track record of assuring “in-use compliance,” leave room for speculating: will the in-use compliance tests go so far as to actually test the equipment’s performance under real, in-the-field conditions? Or will it stop with a determination that the ORVR equipment on the in-use vehicle meets design and manufacturing specifications? Does EPA’s reference to “manufacturer in-use tests” mean what it infers?

Fourth, I find little comfort in knowing that EPA is only “still examining the possibility” of requiring ORVR tests to be performed on Stage II nozzles. In this regard, EPA said that some data on ORVR/Stage II interaction was presented at CARB’s November 10, 1998 enhanced vapor recovery workshop and that EPA or CARB “may collect more data if needed.” EPA did not comment on the nature of this data, its value to EPA or its specific bearing on the issues discussed in this article.

Fifth, my suspicion that EPA is not planning to insist on proof of ORVR effectiveness is reinforced by EPA’s comment that state credits will likely depend on fleet turnover of ORVR vehicles.

Importance of in-use testing data
In my Nov/Dec article, I described several issues raised by Ed Hasselmann, Bart Scowley and Wolf Koch on the effectiveness of ORVR systems when refueling is via the different types of dispenser systems. While not unanimous on all of the issues, these three vapor recovery experts, as well as EPA officials, identified the absence of in-use testing data as a major issue.

On the various other issues raised by Hasselmann, Scowley and Koch, EPA officials said that we did not give them sufficient time to comment before we went to press with the Nov/Dec article and that they would have liked to have given PE&T readers the whole picture on these issues. For this article, we again presented the issues to EPA officials and, for each issue, asked them to (1) comment on whether the issue still exists, (2) provide data or references that show how the issue was resolved or (3) describe the actions needed to resolve the issue, who is to take the actions and when. Each issue is enumerated below (page numbers refer to the Nov/Dec issue of PE&T) and is followed by EPA’s comments. As you consider the issues and EPA comments, consider whether reliable test data on the in-use performance of ORVR-equipped vehicles would help resolve the lingering concerns.

Issue #1: Questionable ability of ORVR systems to regenerate the carbon canisters for any length of time when refueling vapors (as distinguished from evaporative vapors) are collected. (Page 29)

EPA response: EPA is not aware of any data that substantiates this claim. Auto manufacturers, as part of the certification process, are required to project the efficiency of the emission control system at 100,000 miles. For ORVR systems, this can be done a variety of ways including actual refueling tests for 100,000 miles. A number of manufacturers have tested the durability of their ORVR systems by refueling for 100,000 miles and experienced minimal degradation of the ORVR systems as indicated by their durability factors that are applied to certification data.

Issue #2: Possible damage to ORVR systems when conventional nozzles fail to shut off. Conventional nozzles sometimes fail to shut off, especially when the car’s tank is being “topped off.” (Page 29)

EPA response: EPA is not aware of any data that substantiates this claim.

Issue #3: Impractical field-testing capability. The only available means of testing ORVR systems’ performance is a very expensive laboratory shed test. Testing at service stations is not practicable because of the location of the canisters on the vehicle. (Page 29)

EPA response: It is true that a simple method of performing representative testing of ORVR at gas stations has not been developed. This does not mean that it is impossible to develop a simple in-use screening.

Issue #4: Questionable ORVR performance with balance Stage II systems. When the nozzle is depressed, the vapor valve is opened so that the storage tank is vented to the atmosphere, not only as the nozzle is being attached to and removed from the car’s fillneck, but also during the refueling itself, because of gaps between the nozzle faceplate and the fillneck. (Page 53)

EPA response: EPA does not consider this to be a problem since, in general, most balance systems have a slight negative pressure when refueling vehicles, which prevents vapors from being vented into the atmosphere. Furthermore, gaps between the nozzle and the faceplate of the automobile will create a possible leak situation with or without ORVR.

Issue #5: Balance Stage II systems create fugitive emissions when used with ORVR cars unless the balance systems are required to pass periodic tightness tests and storage tanks are equipped with pressure/vacuum vent valves. (Page 53)

EPA response: EPA believes this is true with or without ORVR. Based on testing conducted in California in the fall of 1998, EPA believes that there will be no increase in emissions when ORVR vehicles are refueled at balance Stage II stations.

Issue #6: Incompatibility between ORVR and assist Stage II equipment that creates pressure and fills the pipes and UST with fresh air, developing new vapors that enter the atmosphere through the vapor vent valves or as fugitive emissions through the inevitable breaches in the system. (Page 53)

EPA response: EPA supports CARB’s efforts to resolve the potential problem described in Issues #6 and #7. EPA believes that the environmental impact of this potential problem is small enough that it does not require an immediate resolution (because ORVR vehicles currently comprise less than three percent of the cars and trucks on the road).

Issue #7: Because of the fugitive emissions created as described in Issue #6, CARB will require that Stage II assist operators either install a vent processor or control the vapor pump operation during the refueling of ORVR cars. Also, CARB is proposing a new certification test procedure that will measure the amount of air returned by assisted systems. Those returning more than 50 percent fresh air relative to dispensed liquid will fail unless they have a vent processor. It appears that CARB will decertify existing equipment after the new rules are finalized. (Page 53)

EPA response: (See EPA comments on Issue #6 above.)

Author’s note: As we were going to press, PE&T learned that the CARB proposed procedures mentioned in Issue #7 (above) (TP-201-2D) had been withdrawn from rulemaking and that the Board directed the Executive Officer and staff to “further consider the extent of the possible impacts of the interaction between vehicles equipped with onboard refueling vapor recovery systems and phase II vapor recovery systems, and to present the Board with an appropriate regulatory solution, if necessary, in consideration of the impacts on the environment and on the economy of the state.”

In commenting on the above issues for this article, neither EPA nor CARB mentioned this action or the reasons for it.

Issue #8: Possible ORVR equipment maintenance problems and the tendency of owners to either ignore the problems or to disable the systems. (Page 53)

EPA response: EPA is not aware of any data that substantiates this claim. ORVR is an extension of the evaporative system which has been in place in vehicles for several years and requires no maintenance for the useful life of the vehicle. Our experience with evaporative systems tells us that it is not likely that there will be any maintenance problems. Because of the nature of the ORVR/evaporative system, it is virtually impossible to disable, nor is there any benefit to disabling the evaporative or ORVR system.

In my opinion, in-use test data would go a long way towards proving or disproving most of the eight issues enumerated above. The prospects for the development of such data are not enhanced by the fact that, as discussed under Issue #3, the only available means of testing ORVR systems’ performance is a very expensive laboratory shed test. While stating that it is not impossible to develop a simple in-use screening method, EPA officials did not say whether EPA would be pursuing the development of such a method.

Adding vapor to the vapor problem
To obtain additional information and viewpoints on the issues enumerated above, I contacted two additional experts on vapor recovery equipment: Robert W. McDowell, PE, Director of Engineering and Manufacturing, Hasstech, Inc., and Tedmund P. Tiberi, President of ARID Technologies, Inc.

Both Bob McDowell and Ted Tiberi believe that these eight issues are reasons for concern and offer the following additional observations. On the subject of carbon canister durability, McDowell says he is not aware of any publicly available test data on the effectiveness of carbon canisters past the 100,000-mile point of equivalent usage. Many people today are keeping vehicles on the road far past this level. The majority of automobile manufacturers today advertise that no tune-ups are needed before 100,000 miles, inferring their expectations that the vehicles will stay on the road much longer.

Both McDowell and Tiberi say that they addressed the issue of additional vapor emissions resulting from refueling ORVR-equipped vehicles at certain types of Stage II dispenser systems. According to McDowell, most Stage II systems cause storage tanks to remain in a slight positive pressure which increases when ORVR vehicles are refueled, even with a nominal vapor to liquid (V/L) ratio of 1: 1. This, he says, means that if the system is not pressure tight, hydrocarbon emissions are constantly occurring and are worsened when ORVR-equipped vehicles are refueled, the magnitude depending on the size of the leaks in the system.

Ted Tiberi goes even further. He says that, no matter what kind of dispensing equipment is involved and no matter whether ORVR equipment is being used or not, the refueling process generates what he refers to as “evaporative vapor” emissions from the storage tanks. The evaporative vapors are in addition to the “refueling vapors” displaced from a vehicle’s fuel tank. Tiberi says that evaporative vapors, left unchecked, will eventually escape into the atmosphere via the tank vent system or breaches in the vapor system. He says further that evaporative vapors are generated in much greater volumes when ORVR vehicles are fueled at stations equipped with Stage II systems, particularly some of the assist systems. Tiberi says that, while the ORVR systems may do their job by limiting the refueling vapor emissions, the volume of evaporative vapors that can escape from storage tanks as a result of the process can be far greater than the uncaptured refueling vapor from vehicle tanks.

Both McDowell and Tiberi are involved in technologies that are designed to limit the emissions of evaporative vapors from storage tanks. Tiberi discusses one of these technologies in detail in his article, “Membranes, Molecules and the Science of Permeation,” on page 30.

EPA on evaporative vapors
In view of the information provided by Bob McDowell and Ted Tiberi, we asked EPA officials whether (1) EPA has any information about the extent to which ORVR systems cause the creation of additional evaporative vapor emissions, and (2) EPA believes that evaporative vapors, as described above, is an issue that needs to be addressed in the mandated vapor recovery program.

In response, EPA officials again referred to CARB’s presentation of data on this subject at its November 1998 workshop (see page 17). adding that:

  “For assist systems, we support CARB’s efforts to quantify the environmental impact of this potential problem, and to resolve this potential problem, if necessary. As stated above, we believe that the environmental impact of this potential problem is small enough that it does not require an immediate solution (because ORVR vehicles currently comprise less than 3% of the cars and trucks on the road). EPA will address the issue if it is determined to be a problem.

“As previously stated, we believe that there will be no increase in emissions when ORVR vehicles are refueled at balanced Stage II stations.”

The EPA position as quoted above does not recognize the fact that, somewhere down the road, ORVR-equipped vehicles will be in “widespread use.” EPA officials seem content to wait to see what CARB decides to do about this and the other issues concerning vapor recovery. Further, EPA’s position that there will be no increase in emissions when ORVR vehicles are refueled at balance Stage II stations is valid only if, as stated by Wolf Koch in his July article in PE&T: “(1) balance systems are required to comply with periodic tightness tests; and (2) balance systems are required to utilize a pressure/vacuum (P/V) vent valve, now a local option in California.”

CARB testing awaits procedure approval
In February, CARB officials said that CARB has conducted some testing to determine the impact of ORVR systems on the overall effectiveness of Phase II (Stage II) systems. They did not say what the testing had shown or what conclusions were reached. They did say, however, that “more testing is needed to further analyze the potential problems” and that “a test procedure for determining the effectiveness of systems with ORVR-equipped vehicles has been developed and is scheduled to go to the Board in December 1999.”

CARB officials did not address the emission of evaporative vapors other than those created by ORVR/Stage II interactions.

Conclusion
Neither EPA nor CARB has yet produced in-use test protocols or results that would show whether or not ORVR-equipped vehicles and Stage II systems are, in fact, working to limit vapor emissions as intended by the federal mandates. So, the actual impact that ORVR and Stage II are having on refueling vapor emissions is unknown.

In commenting on my conclusion, EPA officials disagreed, stating that the impact is known and that “preliminary data was presented in CARB’s November 10, 1998 workshop in Sacramento. A final report is expected in a month or so.” So far as I could find out, CARB has not issued, and does not plan to issue, such a report. As shown in the section immediately above, CARB officials now say that more testing is needed and that they are planning to send a test procedure to the Board in December 1999.

Even if refueling vapors are recovered, the creation of additional evaporative vapors reduces the overall effectiveness of the vapor recovery effort. This fact needs to be included in any assessment of vapor recovery efforts.

Substantial resources have gone into years of debates, mandates, enforcement and compliance efforts to control refueling vapor emissions. Footing the bill for these efforts, as taxpayers, automobile purchasers and gasoline consumers, the public should have assurance that the mandates are warranted and are working. At this time, such assurances still seem to be a long way off.

Discuss