New ‘real world’ tests could still allow illegally polluting cars on roads

Publication date: 13th September 2017

New so-called ‘real world’ emissions tests prompted by the VW scandal still fail to capture the actual impact diesel vehicles are having on air pollution in towns and cities, a new investigation by Greenpeace has revealed.


Independent testing of two new diesel cars [1] was carried out by leading experts Emissions Analytics (EA) on behalf of Greenpeace. The results showed that during testing along urban roads both cars emitted more toxic nitrogen pollutants (NOx) than their own Real Driving Emissions (RDE) results showed, breaking the legal limit of 168mg/km.


Two popular diesel cars – the VW Golf and Vauxhall Insignia – were tested on well-used commuter routes into and out of London during the morning and evening rush hours. The VW failed to stay under the 168mg/km limit on both the morning and evening tests and was almost three times the limit when results were analysed for the most congested section of the route [1a]. The Vauxhall insignia passed both morning and evening tests but failed on the most congested section.   


The most concerning examples for each of the cars occurred during the congested section of the route with emissions of harmful nitrogen pollutants (NOx) 42% higher for the Vauxhall Insignia and 118% higher for the VW Golf during that section than levels detected in the urban section of those cars’ RDE tests [2].


This shows that during stop-start traffic during rush hour, these cars fail to meet the legal standard pointing to potential flaws in the RDE test and loopholes that allows companies to test at any time of day, therefore avoiding rush hour altogether if they choose to.


Paul Morozzo, Greenpeace UK Clean Air Campaigner, said:

“The RDE tests should leave the auto industry no room to hide their cars’ real emissions but our investigation suggests this is not the case. These new tests are not ‘real’ enough to ensure the most polluting cars are kept off our roads. That car companies are allowed to avoid rush hour traffic when testing in urban areas is a major flaw.


“Instead of wasting more time and money hiding behind tests that still don’t reflect what’s happening in the real world, car companies should switch from diesel to electric and hybrid technology. Ministers cannot rest on their laurels either – these tests do not solve the problem of air pollution which makes a ban on new diesels long before 2040 even more crucial.”


Since September 1, all new car models must pass the RDE test [3] as part of the type approval process [4] before being put out on the market. The new testing regime requires cars to be driven on the road, rather than in a lab as with previous tests, in three different settings – urban, rural, and motorway. To pass, the cars must emit no more than 168mg/km nitrogen dioxide (NOx) in both the urban section of the test, and overall (when all three results are averaged).


Thanks to industry lobbying [5], the legal limit for NOx emissions in the RDE test has already been more than doubled compared with the previous 80mg/km lab testing limit. Added to this, the new testing rules contain a number of loopholes, including, crucially, that cars can be tested at any time of day. This means that companies can avoid rush hour traffic if they choose, potentially hiding the poorer performance of cars when driven in conditions that people are most likely to be driving in.


The cars tested, their urban RDE results (obtained by Unearthed, Greenpeace’s investigative news platform), and the results from the EA tests are as follows:


Vauxhall Insignia

RDE Test results obtained from Vauxhall: 128.6mg/km NOx and 140.2mg/km NOx

Emissions Analytics results: 111mg/km NOx (pm rush hour) and 107mg/km NOx (am rush hour)

EA results from the most congested part of the route: 199mg/km NOx


VW Golf 1.6

RDE Test results obtained from VW: 203.4mg/km NOx

Emissions Analytics results: 311mg/km NOx (pm rush hour) and 226mg/km NOx (am rush hour)

EA results from the most congested part of the route: 445mg/km NOx


Nick Molden, CEO, Emissions Analytics said:

“The results of the test demonstrate what we expected to be the case based upon our wider testing program*. Our testing showed that while improvements are present under some conditions, the results also show that further improvements may be required for both the testing regime and for diesels to manage challenges like London’s rush hour.


“Other Emissions Analytics testing shows diesels can be cleaner in normal urban and motorway driving. However, the pace of change, or lack of, continues to threaten the future of diesel.”



Full RDE documents and Emissions Analytics data can be viewed here:…ions-independent/


Images can be downloaded here:


Footage can be viewed here:


For more information, contact Alison Kirkman on 07896 893154 or email




Notes to editors:

[1] Three cars were tested overall, the third being a BMW Mini ONE D but RDE data could not be obtained for the model tested and so results have not been included for that car.

For information, the results for the Mini from the Emissions Analytics tests were as follows:
Emissions Analytics overall results: 214mg/km NOx (pm rush hour) and 85mg/km NOx (am rush hour)
EA results from the most congested part of the route : 116mg/km NOx


Cars were tested twice during the morning rush hour driving from Perivale near Wembley to Old street along the A40 and A501, with the overall result presented by Emissions Analytics as an average of those two tests, and twice during the evening rush hour driving from Goswell Rd to Great North Road Finchley along Holloway Rd and Archway Rd again, with the result presented by Emissions Analytics as an average of those two tests.


[1a] The most congested part of the route was a three-mile stretch between Edgware Road and Angel, Islington, along the A501.


[2] We compared the highest urban section RDE test result (140.2mg/km) for the Vauxhall with the congested section of the Emissions Analytics test (199mg/km), and it showed that harmful nitrogen pollutants (NOx) were 42% higher than levels detected in the RDE test results for that car.  (If you compared the lowest urban section RDE result for the Vauxhall with the Emissions Analytics congested section, it would be even higher). Comparing the only urban RDE test result obtained from VW for the VW Golf 1.6 (203.4mg/km) with the congested section of the EA test for that car (445mg/km) showed the VW producing 118% higher NOx.




[4] Type approval is the confirmation that production samples of a design will meet specified performance standards. The specification of the product is recorded and only that specification is approved.



Euro 6 legislation was put in place in 2007 with a target of 80mg, which was to come into force in September 1st 2014 for new type approvals (a significantly new car, with design changes) and Sept 1st 2015 for all new cars (those coming off the production line that are the same as older models). So the car industry has known 80mg was the target since 2007 and has had to comply with Euro 6 cars since 2014.


When RDE tests were designed, car companies successfully lobbied for the pass limit to be raised to 168mg/km for a number of years – a 2.1 conformity factor.

– For new type approvals, RDE came into force 1st Sept 2017.

– For all new registrations RDE will come into force 1st Sept 2019.

– For new types RDE goes down to 1.5x on Jan 1st 2020 and for new registrations down to 1.5x on Jan 1st 2021.