First look: Results of the German transport ministry's post-VW vehicle testing

In September 2015, shortly after it became known that Volkswagen applied illegal defeat devices in some of its diesel vehicles, the German Ministry for Transport (BMVI) commissioned a series of laboratory and on-road vehicle tests in order to determine whether other manufacturers were also making use of similar devices. Now, after months of silence, the results of the German testing program were published on a Friday afternoon—a traditional time for announcing news but avoiding the spotlight.

The testing was coordinated by the Ministry and the German vehicle type-approval agency, the KBA, but the tests were carried out by several technical service providers that have the necessary laboratories and Portable Emissions Measurement Systems (PEMS). The vehicles were obtained from rental agencies or purchased from vehicle dealers.

In total, 56 vehicles (53 different vehicle models) were tested. Among those were 32 Euro 6 passenger cars, 24 Euro 5 passenger cars and 5 light commercial (N1) vehicles. All vehicles were run through eight tests:

Laboratory tests

  • Test 1: Cold-start NEDC, with pre-conditioning, as usually used for type-approval
  • Test 2: Same as Test 1, but without pre-conditioning, i.e., with a warm engine
  • Test 3: NEDC at 10 degrees Celsius, i.e., at a lower temperature than for type-approval

On-road tests

  • Test 4: NEDC on the road, using PEMS, following the normal NEDC speed trace
  • Test 5: NEDC backwards, i.e., first driving the extra-urban/highway part and then the urban part of the cycle
  • Test 6: Following the NEDC but with all velocities 10% higher than usual
  • Test 7: Following the NEDC but with all velocities 10% lower than usual
  • Test 8: Following the Real Driving Emissions (RDE) test procedure that was introduced in the EU as an additional element for type-approval from 2016 onwards

As can be seen from the definition of the test runs, the main focus of the German program was on varying the speed trace followed. This was the trigger used by the Volkswagen defeat device to switch into an alternative exhaust aftertreatment calibration. Test #3, where the test temperature is varied, was added part way through the course of the program, to cover one additional aspect of how vehicles might be triggered to behave differently under different testing conditions. Whether the tests defined for the German program are sufficient to uncover all potential types of defeat devices remains open to question.

According to the BMVI/KBA report, it was possible to verify that the four Volkswagen Euro 5 cars tested (Polo, Golf Plus, Beetle, Passat) use an illegal defeat device that drastically reduces the exhaust aftertreatment as soon as there is a deviation from the normal NEDC speed trace. No similar defeat device—that is, one based on test cycle recognition—was found for other manufacturers (the report is actually quite carefully worded here, so for an outside reader it is not entirely clear whether there might be more information forthcoming, as some media reports suggest). Also, no similar defeat device was found in the Volkswagen Euro 6 vehicles tested.

But the report makes it quite clear that all vehicle manufacturers are using defeat devices as those are defined in article 3 of European Commission regulation 715/2007. Emissions increased drastically in a number of the test vehicles even when the testing conditions changed only slightly. Among the worst offenders were two Renault Kadjar models that emit up to 18 times more nitrogen oxides (NOx) when tested on the road than in the laboratory, a Suzuki Vitara (14 times higher), a Land Rover (13 times higher), a Dacia Sandero (12 times higher) and an Opel Zafira (11 times higher). (See chart.) In comparison, the VW models with a defeat device look almost harmless, with on-road emissions “only” 2–6 times higher than in the laboratory.


Overview of the laboratory and on-road test results of the vehicle testing program by the German Ministry for Transport.

On average, the maximum measured on-road NOx emission level for the Euro cars was at 491 mg/km, more than 6 times the regulatory emissions limit of 80 mg/km. This average figure is close to our own assessment of the on-road performance of Euro 6 diesel cars. For comparison, a modern Euro VI heavy-duty truck emits about 180 mg/km of NOx on the road and a Euro 6 gasoline car only about 50 mg/km. But it should be noted that the BMVI/KBA testing program also found vehicles with relatively low NOx emissions, even when tested on the road. For example, the Euro 6 versions of the VW Touran, VW Passat, and Audi A3 remained below the 80 mg/km limit in all of the BMVI/KBA test runs.

The Ministry for Transport shared the test results with manufacturers months before they were publicly released, and asked them to explain the high emissions. Generally, the manufacturers claim that at lower temperatures they have to reduce the exhaust aftertreatment of their vehicles to prevent damage to the engine For some manufacturers, “lower temperatures” means below 20 ˚C, or 68 ˚F (Alfa Romeo, Fiat, Hyundai, Jeep, Opel); others define it as below 17 ˚C (Audi, Nissan, Porsche). The ministry did not find the explanations universally persuasive: the report classifies 22 of the 53 vehicle models tested as having extraordinarily high NOx emissions without a satisfactory explanation provided by the manufacturer. As a result, vehicle manufacturers (Audi, Daimler, Opel, Porsche and Volkswagen) will “voluntarily” recall a total of 630,000 Euro 5 and Euro 6 diesel cars that were type-approved in Germany in order to improve their on-road exhaust aftertreatment.

In addition, the Minister for Transport, Alexander Dobrindt, announced that from now on manufacturers requesting type approval from the KBA will have to declare whether there is a defeat-like device installed in the vehicle and explain in detail how it works and why the vehicle cannot function without it. This would be an important improvement in the vehicle approval process and is in line with one of the key recommendations from our recent briefing paper on defeat devices in the EU and US regulations. Of course, it remains to be seen how this new procedure is applied in practice and whether other EU member states follow Germany’s example.

Vehicle testing