Lably
Skip to content

Volkswagen emissions scandal

2010s diesel emissions scandal involving Volkswagen

  • 66 min read
  • 5 chapters
Lateral view VW Golf TDI clean diesel at the 2010 Washington Auto Show
Mariordo Mario Roberto Duran Ortiz · CC BY-SA 3.0

Related on Lably

Through Vehicle emissions control, Open source and Carbon dioxide

Background

Introduction

In 2013, the International Council on Clean Transportation (ICCT) commissioned the West Virginia University Center for Alternative Fuels Engines and Emissions (WVU CAFEE) to test on-road emissions of diesel cars sold in the U.S. Researchers at WVU CAFEE, who conducted live road tests in California using a Japanese on-board emission testing system, detected additional nitrogen oxide (NOx) emissions from two out of three tested vehicles, both made by Volkswagen. In May 2014, ICCT published WVU CAFEE's findings and reported them to the California Air Resources Board (CARB) and the United States Environmental Protection Agency (EPA).

In September 2015, the EPA announced that Volkswagen had violated the Clean Air Act by installing unlawful software into their diesel vehicles. Regulators in multiple countries began to investigate the automaker, and its stock price fell in value by a third in the days immediately after the news. Volkswagen Group CEO Martin Winterkorn resigned, and the head of brand development Heinz-Jakob Neusser, Audi research and development head Ulrich Hackenberg, and Porsche research and development head Wolfgang Hatz were suspended.

Volkswagen announced plans in April 2016 to spend €16.2 billion (US$18.32 billion at April 2016 exchange rates) in relation to the scandal, and agreed in June 2016 to pay up to $14.7 billion to settle civil charges in the United States. In January 2017, Volkswagen pleaded guilty to criminal charges and signed an agreed Statement of Facts, which set out how the company's management asked engineers to develop the defeat devices, because its diesel models could not pass US emissions tests without them, and deliberately sought to conceal their use. In April 2017, a US federal judge ordered Volkswagen to pay a $2.8 billion criminal fine for "rigging diesel-powered vehicles to cheat on government emissions tests".

Winterkorn was charged in the United States with fraud and conspiracy on 3 May 2018. As of 1 June 2020, the scandal had cost VW $33.3 billion in fines, penalties, financial settlements and buyback costs. Government and civil actions were taken in the U.S. and the European Union, where most of the affected vehicles were located. While these vehicles remained legal to drive in both regions, consumer groups and governments sought to ensure that Volkswagen had compensated affected owners, as it had been required to do in the United States.

The scandal raised awareness over the higher levels of pollution emitted by all diesel-powered vehicles from a wide range of car makers, which under real-world driving conditions exceeded legal emission limits. A study conducted by ICCT and ADAC showed the biggest deviations from Volvo, Renault, Jeep, Hyundai, Citroën and Fiat, resulting in investigations opening into other diesel emissions scandals. A discussion was sparked on the topic of software-controlled machinery being generally prone to cheating, and a way out would be to open source the software for public scrutiny.

NOx and Volkswagen Diesel anti-pollution system

In general, diesel engines have better fuel efficiency and less carbon dioxide (CO2) emission than petrol engines, but they emit 20 times more nitrogen oxide (NOx) unless somehow treated. Three-way catalytic converters, which have been very effective at reducing NOx in petrol engine exhaust, do not function well for them. As NOx are harmful air pollutants, regulators in the United States and Europe have implemented increasingly stringent NOx emission standards for passenger cars since the early 2000s.

To deal with this problem, in 2005 Volkswagen licensed Mercedes' urea-based selective catalytic reduction (SCR) system called BlueTec for future diesel engine development. While effective at reducing NOx, an SCR system like Bluetec was expensive, high-maintenance and required more space than other methods, making it unsuitable for Volkswagen's compact cars such as Golf or Jetta. Some managers at Volkswagen rejected BlueTec, and preferred to develop their own inexpensive "lean NOx trap" system. In 2007, Volkswagen canceled the licensing deal for BlueTec and announced that it would use its own pollution control technology.

Volkswagen chose the "lean NOx trap" system for its turbo-diesel Golf and Jetta models, but the solution did not work well as it required a fuel-rich exhaust gas in the purification process and fuel economy suffered as a result. Despite these issues, the company continued to market its diesel technology as an affordable and eco-friendly innovation, but this impression projected to the public did not reflect the reality. In reality, the system failed to combine lower fuel consumption with compliant NOx emissions, and Volkswagen chose around 2006 to program the Engine Control Unit (ECU) to switch from lower fuel consumption and high NOx emissions to low-emission compliant mode when it detected an emissions test, particularly for the EA 189 engine. This caused the engine to emit NOx levels above limits in daily operation, but comply with US NOx standards when being tested, constituting a defeat device. In 2015 the news magazine Der Spiegel reported that at least 30 people at management level in Volkswagen knew about the deceit for years, which Volkswagen denied in 2015.

Starting in the 2009 model year, Volkswagen Group began migrating its light-duty passenger vehicle's turbocharged direct injection (TDI) diesel engines to a common-rail fuel injection system. This system allows for higher-precision fuel delivery using electronically controlled fuel injectors and higher injection pressure, theoretically leading to better fuel atomization, better air/fuel ratio control, and by extension, better control of emissions.

Volkswagen described the diesel engines as being as clean as or cleaner than US and Californian requirements, while providing good fuel economy and performance. Due to the good fuel economy provided by its diesel fleet, in 2014 Volkswagen was registered with a Corporate Average Fuel Economy (CAFE) of 34 mpg‑US (6.9 L/100 km; 41 mpg‑imp). The low emissions levels of Volkswagen vehicles tested with the defeat device in operation enabled the company to receive green car subsidies and tax exemptions in the US.

Underlying U.S. and EU emission standards

The Volkswagen and Audi cars identified as violators had been certified to meet either the US EPA Tier 2 / Bin 5 emissions standard or the California LEV-II ULEV standard. Either standard requires that nitrogen oxide emissions not exceed 0.043 grams per kilometre (0.07 g/mi) for engines at full useful life which is defined as either 190,000 kilometres (120,000 mi) or 240,000 kilometres (150,000 mi) depending on the vehicle and optional certification choices.

This standard for nitrogen oxide emissions is among the most stringent in the world. For comparison, the contemporary European standards known as Euro 5 (2008 "EU5 compliant", 2009–2014 models) and Euro 6 (2015 models) only limit nitrogen oxide emissions to 0.18 grams per kilometre (0.29 g/mi) and 0.08 grams per kilometre (0.13 g/mi) respectively. Defeat devices are forbidden in the EU. The use of a defeat device is subject to a penalty.

Nature reported in 2015 that 20 percent of European city dwellers were exposed to unhealthy levels of nitrogen dioxide, and that in London, where diesel road traffic was responsible for 40 percent of NOx emissions, 3,000 deaths per year could be attributed to air pollution. A Channel 4 documentary in January 2015 referred to the UK government moving to a CO2 emission band system for road tax, which favoured diesel power, as the "great car con", with Barry Gardiner MP, former member of the Blair government, stating that the policy, which lowered CO2 emissions yet increased NOx pollution, was a mistake.

Early warnings 1998–2013

In 1998, a Swedish researcher criticized the New European Driving Cycle standard for allowing large emission differences between test and reality. The Washington Post also reported that in the late 1990s, EPA engineers at Virginia Testing Laboratory had built a system called ROVER, designed to test a car's emissions on the road. The project was shut down in 2001, despite preliminary tests indicating gaps between emissions from lab tests and real world tests of about 10 to 20 percent.

In 2011, the European Commission's Joint Research Centre published a report which found the average on-road NOx emission of all tested diesel vehicles to be 0.93 ± 0.39 g/km, and that of tested Euro 5 diesel vehicles to be 0.62 ± 0.19 g/km. Those numbers substantially exceeded the respective Euro 3–5 emission limit. In 2013, the research center then warned:

Sensors and electronic components in modern light-duty vehicles are capable of 'detecting' the start of an emissions test in the laboratory (e.g., based on acceleration sensors or not-driven/not-rotating wheels). Some vehicle functions may only be operational in the laboratory, if a predefined test mode is activated. Detecting emissions tests is problematic from the perspective of emissions legislation, because it may enable the use of defeat devices that activate, modulate, delay, or deactivate emissions control systems with the purpose of either enhancing the effectiveness of these systems during emissions testing or reducing the effectiveness of these systems under normal vehicle operation and use. While the use of defeat devices is generally prohibited, exceptions exist in cases where it is necessary to protect the engine against damage and to ensure safe vehicle operation (EC, 2007). These exceptions leave room for interpretation and provide scope, together with the currently applied test procedure, for tailoring the emissions performance [...].

The European Commission and European governments could not agree upon who was responsible for taking action. In the United Kingdom, the Department for Transport received a report from the International Council on Clean Transportation (ICCT) in October 2014, which stated there was a "real world nitrogen oxides compliance issue" with diesel passenger cars. The UK's DEFRA research indicated a significant reduction in NOx and particulate matter from 1983 to 2014. Respirable suspended particles with a diameter of 10 micrometres – also known as PM10 (including diesel particulates) – halved since 1996 despite the increased number and size of diesel cars in the UK.

ICCT's emission testing, 2013–14

In late 2012, the independent body International Council on Clean Transportation (ICCT) was seeking proposals from research facilities to study on-road emissions of so-called "clean diesel" cars sold in the US. Researchers at West Virginia University's Center for Alternative Fuels Engines and Emissions (WVU CAFEE) submitted a proposal to ICCT, and were awarded a US$70,000 grant for a study where they had to test US-spec diesel vehicles with two different emission reduction technologies: lean NOx trap (LNT) and selective catalyst reduction (SCR). John German, co-lead of the US branch of ICCT, said the idea for this "very ordinary" test came from Peter Mock, managing director ICCT in Europe, and they chose to put US vehicles through on-the-road tests because their emissions regulations are more stringent than those in the European Union. The ICCT expected the cars to pass, and thought they would be able to use the results to demonstrate to Europeans that it was possible to run diesel cars with cleaner emissions.

In early 2013, researchers at WVU CAFEE began testing emissions from the three diesel cars: a Volkswagen Passat, a Volkswagen Jetta, and a BMW X5, on the road in California. They wanted to test a Mercedes as well, but could not obtain one. They used a portable emissions measurement system, making it possible to collect real world driving emissions data, for comparison with laboratory dynamometer testing. The three vehicles were all certified at a California Air Resources Board facility before the tests as falling below the emissions limits when using the standard laboratory testing protocols. They put 2,400 kilometres (1,500 mi) on the Jetta and X5. For their final test, they wanted to put even more mileage on the Passat and drove it from Los Angeles to Seattle and back again, virtually the entire West Coast of the United States, over 3,200 kilometres (2,000 mi). The BMW was "at or below the standard ... with exception of rural-up/downhill driving conditions". But the researchers found that under real-world driving conditions the Jetta exceeded US emissions limits "by a factor of 15 to 35" while the Passat exceeded the limit "by a factor of 5 to 20".

The emissions far exceeded legal limits set by both European and US standards. One of the testers said, "... we did so much testing that we couldn't repeatedly be doing the same mistake again and again." John German said the deceit required more effort than merely adding some code to the engine software, as the code would also have to be validated. The US test results confirmed the ICCT's findings in Europe. ICCT also purchased data from two other sources. The new road testing data and the purchased data were generated using Portable Emissions Measurement Systems (PEMS) developed by multiple individuals in the mid-late 1990s and published in May 2014. The West Virginia scientists did not identify the defeat device, but they reported their findings in a study they presented to the EPA and CARB in May 2014. In May 2014 Colorado's RapidScreen real-world emissions test data reinforced the suspected abnormally high emissions levels. After a year-long investigation, an international team of investigators identified the defeat device as a piece of code labelled "acoustic condition" which activated emissions-curbing systems when the car's computer identified it was undergoing a test.

EPA Notice of Violation, 2015

On 18 September 2015, the US Environmental Protection Agency served a Notice of Violation (NOV) of the Clean Air Act on Volkswagen Group alleging that Volkswagen and Audi automobiles equipped with 2-litre TDI diesel engines, and sold in the US between 2009 and 2015, had an emissions-compliance "defeat device" installed, and ordered a recall of 482,000 vehicles. A Notice of Violation is a notification to the recipient that the EPA believes it has committed violations and is not a final determination of liability.

Volkswagen's "defeat device" is specially-written engine-management-unit firmware that detects "the position of the steering wheel, vehicle speed, the duration of the engine's operation, and barometric pressure" when positioned on a dynamometer using the FTP-75 test schedule. These criteria very closely match the EPA's required emissions testing protocol which allowed the vehicle to comply with emissions regulations by properly activating all emissions control during testing. The EPA's NOV alleged that under normal driving conditions, the software suppressed the emissions controls, allowing better fuel economy, at the expense of emitting up to 40 times more nitrogen oxides than allowed by law.

Intelligence agencies, 2015

In February 2017, Der Spiegel reported that in February 2015, former Israeli diplomat Avi Primor had shown Ferdinand Piëch, the Volkswagen chairman of the board at the time, a document in which US agencies warned CEO Martin Winterkorn early about the manipulation. During this meeting at the end of February 2015, Primor introduced Piëch to his friend Yuval Diskin, who after retiring from directing the Israeli secret service of the Interior Shin Bet, had founded a cybersecurity company. Shin Bet apparently knew about the scandal early. Primor confirmed that the meeting took place, but both Primor and Diskin denied tipping off Piech. In early March 2017, Piech asked Winterkorn whether there had been a warning by US agencies, which Winterkorn denied.

Sign in to keep reading

4 more chapters are free with a Lably account.

More in Environment