Blog

FAQ: In-use NOx emissions from diesel passenger cars

As the story concerning in-use NOx emissions from diesel passenger vehicles continues to unfold, a number of questions concerning the ICCT’s past and on-going research into discrepancies between chassis dynamometer test emissions and on-road test emissions, have come up repeatedly.

In response, we’ve put together this FAQ sheet, which we’ll try to update as needed. See also our press statement concerning the U.S. EPA’s notice of violation of the Clean Air Act to Volkswagen, here.

What is the diesel NOx problem, and how long has it been known?

The scientific community has been aware of high in-use NOX emissions from diesel vehicles for several years. High emissions tend to happen during instances of demanding driving (e.g. sudden accelerations or going uphill) that are part of normal operation but that are not adequately captured by the emissions tests performed in the laboratory. This is especially true of the European certification procedure, which is based on the New European Driving Cycle (NEDC) and has very low acceleration and engine load.

Why test diesel cars in the U.S., which has such a small diesel market?

Because not only are the U.S. emissions standards most stringent but, since the in-use compliance aspects of the U.S. regulations are more robust, it was reasonable to hypothesize that diesel cars sold in the U.S. market would be clean — i.e., on-road emissions would conform to the regulatory limits. If that hypothesis proved true, that would help to demonstrate the practical possibility of building and marketing clean diesel cars in any market, but especially in Europe, where the diesel share of the new-car market is 53%.

How was the initial ICCT research on U.S. diesel passenger cars conceived and funded?

The in-use testing project that ICCT conducted in collaboration with West Virginia University developed from a larger, on-going research program in Europe. Early conceptual discussions among the ICCT, the European Commission’s Joint Research Centre (JRC), and the California Air Resources Board (CARB) eventually led to ICCT submitting a proposal for funding a research project to the European Commission in September 2012. That proposal was rejected in November 2012, at which point the ICCT determined to pursue a smaller-scale project using its own funds. (ICCT is funded primarily by foundations, including Mercator, Hewlett, and Climateworks, and through projects for government agencies.) In November 2012, the ICCT issued a request for proposals to conduct in-use emissions tests in the U.S. from qualified consultants with access to a portable emissions monitoring systems (PEMS). In early 2013, ICCT contracted with West Virginia University’s Center for Alternative Fuels, Engines and Emissions to conduct on-road testing; CARB offered to test the same vehicles on the official certification tests in its El Monte facility, to provide baseline emissions data. The project was funded at US$70,000. While the U.S. project proceeded, the ICCT was simultaneously working with the European Commission JRC and other EU stakeholders on related research into real-world NOx emissions from European diesel passenger vehicles and participating in the technical working groups in charge of amending Euro 6 regulations.

How were the U.S. vehicles selected and obtained?

Two vehicles were rented from commercial rental agencies, and one from a private individual. The make and model were determined by availability; there are very few diesel vehicles available through rental agencies in the U.S. (To illustrate: one of the vehicles initially selected was a Mercedes-Benz, but when the time came to begin the on-road testing the rental agency had sold that car, and there was no replacement at any rental agency in the State of California.) Another important criterion was that the vehicles not be brand new, but also have low mileage (less than 15,000 miles). Lastly and most importantly, the vehicles had to represent a mix of NOx control technology and mainstream and upscale vehicles. The VW Jetta had a lean-NOx trap (LNT), while the Passat and the BMW X5 had selective catalytic reduction (SCR) systems.

How and where were the U.S. diesel cars tested?

The on-road emissions were tested using a portable emissions measurement system, which provides a continuous stream of vehicle data signals including pollutant emission rates, velocity, vehicle position, exhaust temperature, and weather conditions. PEMS data were gathered over a variety of pre-defined test routes representing diverse driving conditions pertinent to major United States population centers located in California. The city-driving route selected was the “Los Angeles Route Four” loop (LA4), which was originally used in developing the original FTP vehicle certification cycle, with some minor modifications at locations where the traffic pattern or roads have changed since the FTP’s development. Additionally, the VW Passat was operated over an extended distance of nearly 2,500 miles, predominantly in highway driving conditions between California and Washington State.

Are high NOx emissions a sign that a vehicle has a “defeat device”?

No. As noted above, NOx emissions from diesel engines are known to “spike” when the driving conditions are more demanding (e.g., sudden acceleration, uphill driving). Furthermore, the certification cycle currently used in Europe (the New European Driving Cycle, NEDC) represents unrealistically mild driving, which makes the regulated Euro 6 NOx limit (80 mg/km over the NEDC test) less stringent in practice than it is on paper. The fact that some vehicles certified to Euro 6 have been found to have high in-use NOx emissions likely speaks more to the shortcomings of the current emissions certification protocol in the EU. Test results from European market vehicles that we have analyzed provide no evidence of the use of defeat devices by any manufacturer. Either a real-world test program properly structured to detect defeat devices or an extensive investigation, such as the one carried out by EPA and CARB, would be required to uncover a “defeat device.”

Shouldn’t the malfunction indicator light warn if the vehicle is emitting more than allowable amounts of a pollutant?

No. The on-board diagnostics signal equipment malfunctions—i.e., components not operating as designed—not high emissions as such.

What should be done in the future to improve diesel vehicle NOx emissions?

In Europe specifically, an effective implementation of the impending real-driving emissions (RDE) test would be a major achievement. The European Commission will phase in RDE testing in two steps, with increasing levels of stringency. It is widely expected that the initial step of conformity factors (applicable from September 2017 onward) will lie around a value of 2 for NOx emissions from diesel passenger cars—i.e., these vehicles will still be allowed to emit about twice the regulated Euro 6 emission limit of 80 mg/km during the on-road test, effectively making this the first time that the Euro standards will be changed to raise an emission limit instead of lowering it. Moreover, since RDE does not include cold-start emissions, the allowed increase will be substantially higher than is indicated by the conformity factor. The second step of RDE, likely to apply from September 2019 onward, should bring conformity factors close to 1. This compromise should address the urgent problem of keeping Euro 6 diesel passenger cars with weak on-road NOx control from being awarded emissions type-approval certificates in the EU.

It should be noted that the RDE will be conducted on pre-production vehicles, and a manufacturer could change calibrations on production vehicles. The RDE is not intended or sufficient to detect defeat devices on production vehicles.

More generally, the effectiveness of the EPA and CARB reaction upon being presented with inexplicable in-use emissions test results illustrates a simple but essential point, applicable in any jurisdiction: regulations must include compliance and enforcement provisions, and regulatory agencies must be given both the legal authority and the technical resources to make those provisions real, not mere decoration.

Didn’t the ICCT test a BMW X3 which had very high NOx emissions?

No. The references circulating in the press since a story that appeared in Auto BILD (first here, then corrected here) are based on testing done by Emissions Analytics in the U.K. — to be precise, on a single on-road test. ICCT purchased that data, along with much more data, from Emissions Analytics as part of the basis for a not-yet-completed analysis that will be published as an addendum to our October 2014 paper, Real-world emissions from modern diesel cars. A single on-road test does not provide enough data for any meaningful analysis, which is one reason why the dataset for the forthcoming study is so large. The BMW vehicles tested by ADAC for our most recent paper, NOx control technologies for Euro 6 diesel passenger cars, performed better than the average on NEDC and WLTC test cycles.