CRC study finds some adverse results from use of mid-level ethanol blends in MY 2001-2009 engines; DOE and ethanol industry say study significantly flawed
The Coordinating Research Council (CRC) released results of a two-year study conducted by FEV into the effects of intermediate-level (E15 and E20) ethanol blends (earlier post) that found that two popular gasoline engines used in light-duty automotive applications of vehicles from model years 2001 through 2009 failed with mechanical damage when operated on intermediate-level ethanol blends under the test conditions.
In response, the US Department of Energy (DOE) and the Renewable Fuels Association charged that the study was fundamentally flawed. In a post on the DOE website, Patrick Davis, Program Manager of the DOE Vehicle Technologies Program wrote that:
Today, a research organization for jointly-funded work by the auto and oil industries, called the Coordinating Research Council (CRC), released a report on the effects of E15 and E20, or gasoline mixed with up to 15 or 20 percent ethanol, respectively, on vehicle engines. The study claims mechanical damage and suggests degraded engine performance, emissions and durability on conventional vehicles from the use of E15 or E20 fuel. We believe the study is significantly flawed.
The Coordinating Research Council (CRC) is a non-profit organization that directs, through committee action, engineering and environmental studies on the interaction between automotive/other mobility equipment and petroleum products. The Sustaining Members of CRC are the American Petroleum Institute (API) and a group of automobile manufacturer members (Chrysler, Ford, General Motors, Honda, Mitsubishi, Nissan, Toyota, and Volkswagen). CRC research programs are managed by five technical committees (Advanced/Vehicle/Fuel/Lubricants, Atmospheric Impacts, Emissions, Performance, and Aviation.)
Background. Growth Energy, an ethanol industry trade group, petitioned the EPA in March 2009 to raise the limit on ethanol in gasoline from 10 to 15 percent. In June 2008, EPA outlined testing needed for the agency to approve a waiver, and EPA requirements were consistent with test plans developed by the auto and oil industries. The CRC was working with EPA and DOE on a multi-year suite of tests on the effects of higher blends of ethanol. This testing included more than $14.5 million of research sponsored by the auto and oil industries, and $40 million of testing sponsored by the federal government.
Before those tests were completed in October 2010 and January 2011, EPA granted partial waivers to allow the introduction of E15 into the marketplace for use in model year 2001 and later vehicles. (Earlier post.) EPA’s decision was based largely on a DOE study of the effects of E15 on durability of catalytic converters, the primary pollution control system in a vehicle.
Three main engine wear mechanisms are impacted by ethanol fuels: abrasive wear, adhesive wear and corrosion. Material changes have made current engines and vehicles robust to ethanol concentrations in gasoline of up to 10 volume percent, and flex-fuel vehicles (E85) are designed to handle the higher concentrations. At blend ratios higher than 10% for non-flex-fuel cars, however, automakers were concerned about potential problems with engine valve and valve seat wear and catalyst durability.
The CRC test. The objective of this durability study was to identify possible engine component wear caused by additional ethanol content in the fuel using an engine test cycle employed by an original equipment manufacturer (OEM) member of CRC to test for engine durability. The CRC Engine Durability study took duplicates of eight different vehicle model engines spanning 2001-2009 model years.
The engines were tested with E20, and then, as appropriate, E15 and E0, for 500 test cycles, corresponding to 500 hours, with monitoring at regular intervals. To test the effect of ethanol on in-use engine durability, vehicles with engines of various valvetrain types were chosen. FEV and the CRC project panel agreed to test eight vehicle types which represented a selection of various valvetrain type engines in popular light-duty automotive applications in non-FFVs from model year 2001 through 2009.
Because the valvetrain design has an impact on allowable system wear, the study evaluated engines with a variety of valve designs:
- Mechanical valvetrains that do not use hydraulic lash adjusters;
- Hydraulic lash adjusters with a small allowable travel;
- Hydraulic lash adjusters with a large allowable travel;
- Non-premium valve seat materials that were not designed for ethanol fuels;
- Engines with small valves leading to low seat loads; and
- Engines with large valves leading to high seat loads.
FEV used accelerated testing to reduce test times and expose failures, if any. Accelerated testing is a standard practice in the automotive industry for producing faster results and higher expected performance from engines/vehicles. The severity helps reduce test time and compensate for the inherently small sample size associated with durability tests, according to FEV.
While extended light-load operation is an option, it greatly increases the duration, sample size, and cost of test programs. Given the lack of recent automotive experience with extended light-load operation and the industry emphasis on rapid implementation of intermediate-level ethanol blends, FEV determined accelerated testing to be the best approach for this study.
The engine durability protocol used for the evaluations consisted of 500 transient hours with each cycle 60 minutes in duration. The test cycle is a standard engine durability cycle from a local OEM except that the maximum engine speed was limited to 3500 rpm.
FEV limited the speed for two reasons:
the lower maximum speed significantly reduces the test severity making it more likely that the test engines will complete test without failures unrelated to the test objective; and
high speed testing can conceal valve seat wear issues by increasing oil pullover through the positive crankcase ventilation (PCV) and lubricating the valve seats.
The entire 500 hours were run with the respective fuel (E20, E15, or E0). FEV concluded that the test cycle used should correlate with ~100,000 miles of vehicle usage.
|FEV’s durability cycle. Click to enlarge.|
The different types of vehicles of various engine configurations, sizes, valvetrain types and mileage were tested with E20, then on E15 if they failed on E20, and then on E0 if they failed on E15. Vehicles which passed the test on E20 were not retested on lower ethanol blends.
“Pass” and “Fail” criteria for five different categories were determined at the beginning of the program and were assessed on each engine after completion of the durability test. These five categories are:
- emissions during the FTP75 test;
- diagnostic trouble codes (DTCs);
- valve clearance;
- compression; and
FEV deemed an engine was deemed to have failed the test if it failed in at least one of these five categories.
|Overall results. Click to enlarge.|
Each chosen engine was tested in duplicate on each fuel. Eight different vehicle types (two samples of each type) were tested with E20. Results of the E20 testing are as follows:
- three vehicle types (five vehicle samples) failed the durability testing on E20;
- three other vehicle types (four vehicle samples) did not pass all specified criteria after the 500 hour durability test, but were waived after a detailed review of the data with the respective OEM contact.
When an engine failed the durability test on E20, another set of duplicate vehicles with the same engine type was procured from the used car market and scheduled for durability testing with E15. When an engine failed on E15, then another set of duplicate vehicles with the same engine type was procured from the used car market and scheduled for durability testing with E0. In total, 28 engines from eight different vehicle types were tested during this study (16 on E20, 6 on E15 and 6 on E0).
Different types of failures were observed throughout the testing. After completion of all testing and detailed review of the experienced failure modes, FEV drew the following conclusions:
Out of eight different tested engine types, one had a design that was (in retrospect) inappropriate for the test cycle, two failed on E20 and E15, and five passed on E20 and by assumption E15 and E0.
Out of the two failed tested engine types, both successfully completed the reference testing on E0.
There is an 11% chance that all three E15 failures (two with one vehicle type and one with another) would have occurred if failure were independent of ethanol. The results for E20 are the same. Combining the E15 results with the E20 results, there is a 7% chance that all six failures (two E15 and two E20 with one vehicle type and one E15 and one E20 with another) would have occurred with ethanol containing fuels if failure were independent of ethanol.
For the failed engine which also failed on E0 reference fuel, the failures can not be directly linked to the ethanol content. The design of the engine interacting with the test cycle is the primary reason cited by the OEM maker for the observed failures.
The observed failures do not show that specific valvetrain types are more or less sensitive to ethanol content.
The majority of the failures can be linked to issues with valve seats, either related to material or wear/deformation.
DOE response. In his response on the DOE website, Davis charged that the choice of test engines, test cycle, limited fuel selection, and failure criteria of the CRC program resulted in unreliable and incomplete data, which severely limits the utility of the study. More specifically:
The CRC failed to establish a proper control group. Only three out of the eight engines were tested with straight gasoline containing no ethanol (E0), and one of those three failed the CRC’s test.
No engines were tested with E10 fuel, the de facto standard gasoline for all grades, which represents more than 90% of gasoline available in the US market. Even though E10 fuel has been in the market for over 30 years and is used in all current conventional gasoline vehicles and small non-road engines, it was not part of the CRC test program.
The CRC employed a test cycle designed specifically to stress the engine valve train. This test cycle was developed specifically for this study and thus there is no experience base for how to interpret results from the testing.
The CRC used the arbitrary criterion of 10 percent engine leakdown (a diagnostic test in which an engine cylinder is pressurized with compressed air, and the rate at which the cylinder loses pressure is measured) to determine if an engine “failed.” This is not a standard previously employed by either industry or federal agencies during testing, nor as a criterion for any warranty claims. DOE testing has shown that it is not reliable indicator of durability issues.
The CRC decided to select several engines already known to have durability issues, including one that was subject to a recall involving valve problems when running on E0 gasoline and E10.
It is no surprise that an engine having problems with traditional fuels might also “fail” with E15 or E20 ethanol-blended fuels—especially using a failure criterion chosen to demonstrate sensitivity to ethanol and operated on a cycle designed to stress the valves.—Pat Davis
In contrast, Davis noted, the DOE’s own study on E15 use found:
An inspection of critical engine components, such as valves, and did not uncover unusual wear that would be expected to impact performance.
Rather than using an aggressive test cycle intended to severely-stress valves, the Energy Department program was run using a cycle more closely resembling normal driving.
The Energy Department testing program was run on standard gasoline, E10, E15, and E20.
The Energy Department test program comprised 86 vehicles operated up to 120,000 miles each using an industry-standard EPA-defined test cycle (called the Standard Road Cycle).
The resulting data showed no statistically significant loss of vehicle performance (emissions, fuel economy, and maintenance issues) attributable to the use of E15 fuel compared to straight gasoline.
The Energy Department test program also showed that 10% engine leakdown is not a reliable indicator of vehicle performance. In the Energy Department program, there were vehicles found to exceed 10% leakdown for all fuels, including vehicles running on E0 and E10. There was no correlation between fuel type and leakdown, and high leakdown measurements did not correlate to degradation in engine or emissions performance.
Final Report CRC Project: CM-136-09-1B Intermediate-Level Ethanol Blends Engine Durability Study