Do we still need to mind the MPG gap?

The gap between official miles per gallon and real world mpg has grown to 22%. This is up 5% since we first started testing fuel economy almost three years ago.

The average official combined miles per gallon of the 459 passenger cars we have tested is 57 and this is increasing by approximately 1.7 mpg per year. Real world miles per gallon (TMPG) on the other hand, which averages 44 mpg, remains flat thus causing an increase in the gap of about two percentage points each year as can be seen in the graph below.


MPG vs. Engine size

The graph below shows that broadly speaking the gap grows as the engine size reduces. If you buy a five litre car you will not get great mpg but at least it will be consistent with the salesman’s patter and most likely your expectations. However, if you are shopping for a frugal run-around you are better off looking at the one to three litre engines which give the best absolute performance as well as a lower divergence from official figures than the super minis.


Fuel economy by engine type

Our data also shows that petrol engines, as expected, have worse fuel economy than diesels but interestingly the gap to official is also larger. And, manuals return a better fuel economy than automatics but automatics have a smaller gap between official and real-world figures.

mgp gap

More MPG

We’ll be looking at MPG in more detail in next month’s newsletter, including an analysis of the manufacturer leader board. The published results will be anonymised but OEMs are welcome to email me if they would like to find out how they sit within the table.

* The original Transport & Environment report Mind The Gap can be found here

Do eco tyres really save you money?

Preliminary tests have shown that ratings on tyre labels are not telling the full story. At mid-range speeds, an F-rated tyre performs as well as a B-rated tyre for fuel economy.

We tested two contrasting sets of 175/70 R14 tyres on the road. One set was a standard tyre with B-rating for fuel economy and the other had an F-rating. The test route incorporated a range of steady-state speeds from 40mph to 70mph on tarmac in consistent ambient temperatures.

The B rated tyre was superior in the 40-70 mph range by an average of 3.8% mpg and 3.4% less CO2. There isn’t much in it at the mid-range speed but a performance gap opens up at 55mph and by the time you get to 70mph the fuel economy has improved by 12.9%.

Thus a consumer buying B rated tyres is unlikely to notice a fuel economy benefit if the journeys they customarily make are mainly urban. Whereas a consumer heading up and down the motorway each day should enjoy an improvement.

Now this was an unashamedly quick and dirty investigation but it does demonstrate that the relationship between rolling resistance and fuel economy is not linear and that to bring real improvements to the way tyres are bought and sold manufacturers need to adopt more sophisticated models.

The current tyre labelling system, made mandatory by the EU in November 2012, is not working. In a report compiled by the National Tyre Distributors Association (NTDA) and LANXESS, the manufacturers of high-tech rubber for tyres, it was found that one year on 93% of tyre retailers said customers never or only occasionally requested information on the label and only 30% knew that tyres affect fuel consumption.

We think manufacturers need better models to translate rolling resistance calculations into fuel economy effects. Improved, independently verified testing and labelling, perhaps with a monetary quantification of the typical benefit would provide a tangible benefit that the consumer would welcome.

March 2014 – Transatlantic testing

This month, in preparation for a presentation we are giving at the 24th CRC Real World Emissions Conference in San Diego, we have been making a detailed examination of the data we are collecting in America.

In most respects the emissions testing we perform in the USA is identical to the work we do in the UK. We use the same equipment, the same methodology (with some adjustments to account for factors such as the widespread use of air conditioning in California) and even our technicians travel between sites to ensure continuity in our processes.

In the UK, the data is published by What Car? magazine under the brand True MPG and in the USA it is published by Motor Trend magazine as Real MPG.

However, despite consistency in the testing process there are marked differences in the results. One of the most noticeable differences between the UK and the USA is that the statutory figures provided by the Environmental Protection Agency (EPA) are a closer match to real world figures than those generated using the New European Drive Cycle (NEDC). As can be seen in this graph, Real MPG is within 1% of the EPA combined figure compared with an average of 18% below statutory in the UK.

The test cycle in the USA was improved in 2008 and now involves five tests: the city, highway, high speed (up to 80mph), hot (with air con) and cold at 20°F (-7°C). The total distance of the five tests is 43.9 miles and takes 1 hour 35 minutes to complete, compared with 6.8 miles and 19 minutes 40 seconds in the UK. In the US, 15% of new models are tested by the EPA to check the manufacturer figures, and failure to come within 3% of the published result can lead to a hefty fine.

Some more comparisons between Emissions Analytics’s data from the UK and USA can be seen in the table below.

transatlantic testing

It would appear from the test data gathered to date in the USA that the EPA figures are well calibrated to average driving, although variations in the real world can lead to divergence from this by up to 20%. We are now running at full speed in the USA and will be testing upwards of 250 passenger cars per year. It will be interesting to see if a gap between statutory and real world fuel economy starts to appear as the pressure to deliver the best fuel economy label grows.

February ’14 – NOx in the News

NOx in the news

In the press last week was the news that the European Commission has launched legal proceedings against the UK for failing to deal with air pollution. Britain was supposed to meet EU limits set out in the Air Quality Directive by 2010 but the government has said these levels will not be reached until 2020 in most areas and in London it is likely that they will not be met until 2025.

The main cause of these air-borne contaminants regulated by the EU is diesel engines, but why is Britain so far from the target?

A real-world view of NOx

Although auto manufacturers have introduced a number of modifications to meet the ever tightening controls of NOx emissions, a study conducted by Imperial College London and Emissions Analytics, on Euro 5 light-duty diesels, shows the real-world figures exceed Euro 5 standards threefold in most instances.


In the graph above you can see that all the cars in the sample failed to meet Euro standard 4 or 5 and, only three reached Euro standard 3. It can also be seen that the real-world average NOx emissions is considerably higher than the limits set out in the regulations. This is the same situation we see with fuel efficiency when we measure cars for True MPG, comparing their statutory mpg figures to performance in the real world. However, due to high levels of NOx being produced during stop-start driving, such as in traffic, the resulting gap between regulated and real-world air pollution is even more pronounced.

The reasons for Britain breaching EU regulations are many and complex; both NOx and miles per gallon standards are calculated using the New European Drive Cycle, the shortcomings of which have been widely reported and are supported by Emissions Analytics’ large volume of real-world data. Others have voiced concerns regarding the number of monitoring stations and the use of modelled data in EU Air Quality Directive compliance assessments. What it is clear is that real-world data has an important part to play in policy making.

Euro 6 and beyond…

With the introduction of Euro 6 demanding a drop in NOx of 80 per cent on the previous standard, EA and Imperial are continuing their study to see what the impact of this new ruling will be. Emissions Analytics is also developing a new traffic simulation model which will calculate the effect of speed and congestion on fuel economy, as well greenhouse gas (CO2) and air pollution (NOx and CO) components underpinned by the data from its real-world test of more than 400 models of passenger car.

Speed Demonisation

Changing speed limits on the UK’s roads is hardly out of the news at the moment. For instance, various London boroughs, including Camden and Islington, have recently announced a reduction in the speed limit from 30mph to 20mph to improve road safety. In 2011, the government briefly toyed with the idea of increasing motorway speed limits from 70mph to 80mph. Now they are looking at the possibility of reducing sections of the network to 60mph to alleviate congestion.

Although the Highways Agency politely declined our offer to provide data on the effect of speed on fuel economy and emissions for their M1 consultation, I have decided to share it with you instead as I think it makes for interesting reading.

Every car has an optimum speed for maximum fuel efficiency but what is the range between models and what difference does it make? By mining our data, gathered during tests on more than 500 passenger cars, we decided to find out.

The table below shows that the average optimum speed for the top five selling cars in the UK (2011) is 46mph over an 8mph range.

You can also see that for these same vehicles a reduction in speed from 70mph to 60mph improves fuel economy by an average of 22%, but this varies from 15% on the Ford Focus 1.6 petrol and VW Golf 1.6 diesel to 34% on the Vauxhall Corsa 1.3 diesel.

speed demonisation

In a separate study we looked at the effect of reducing the speed limit from 30mph to 20mph and while this reduced CO2 emissions, the impact on CO2, NOx and particulates, due to changes in driving style, warrants further investigation.

While reducing speed may deliver one objective it can have a number of knock-on effects; every car has an optimum speed, 60mph is better than 70mph for fuel economy but 20mph is not necessarily better than 30mph for all tailpipe emissions. Factor in other considerations such as air quality, congestion and road safety and the picture becomes even more complex. What can be concluded however, is that robust data should be the cornerstone of any proposed changes to the rules of our roads.

Sign up here to receive this newsletter

A year of data from Emissions Analytics

Emissions Analytics specialises in the analysis and interpretation of complex data sets. So here is a festive round-up of a few of our own numbers for the year:

  • True MPG results supplied to What Car? for 186 vehicles in 2013
  • Launch of Real MPG for Motor Trend magazine in the USA, 71 cars tested so far
  • 16,000 miles driven on real roads in the UK and US
  • 41,000 miles driven at the track for clients’ testing projects at home and abroad
  • An average of 185 grams of carbon dioxide emitted per kilometre
  • Largest engine tested this year was 12.9 litres
  • The smallest was 875cc
  • The average MPG was 44.5, with the worst just 20.5 and the best 71.7 (a hybrid).

Merry Christmas and a Happy New Year from the Emissions Analytics team.

DPF Regeneration Mysteries

There are two types of diesel particulate filter (DPF) regeneration which can occur without external intervention by the driver or a mechanic. The first is passive regeneration which happens when high exhaust temperatures reached during intensive use (normally either long spells of motorway driving or hill climbing) automatically burn off the trapped diesel particles. The other is active regeneration of the DPF where the ECU is programmed to initiate post combustion fuel injection to increase the exhaust temperature and burn away diesel particles when the filter becomes around 45 per cent full.

If either type of regeneration occurs whilst we are conducting an on-road emissions test we have to scrap the test and start again. So, we keep a careful eye on the exhaust temperature, which we measure using our PEMS equipment, to monitor spikes which indicate regeneration. As a result of this process we have recorded the regeneration activity, in terms of both occurrences and emissions output, and the results are not what you might expect.

One thing we have noticed is that regeneration will often happen after the extra-urban phase of the test, during the urban phase. Thus the opportunity for passive regeneration has been missed and active regeneration takes place instead.

Why does this matter? Well, firstly there is the cost to the operator as active regeneration requires the engine to work harder to reach the required temperature and this reduces fuel economy by as much as 5mpg. But not only that, as can be seen from the graph below, active regeneration has a significant and negative impact on NOx emissions.


Graph by Joseph Ruxton, Imperial College London

NOx, a major source of air pollution, can as much as double during regeneration due to late fuel injection and increased engine temperatures. If this happens in town and city environments rather than on the motorway, it could have an even greater negative impact on human health. Although it should be noted that the negative effective of the rise in NOx is offset somewhat by the primary NO2 reduction the regeneration is designed to achieve.

Another observation on active regeneration concerns the new cars we test. Because every vehicle What Car? receives to test drive and review is passed to Emissions Analytics who then tests it to record the data which powers True MPG, many of the cars we are testing are very new, with only a few hundred miles on the clock, if that. Yet we are finding that active DPF regeneration can on occasionally take place during the course of the urban phase of our test. It is unlikely that the DPF is 45 per cent full on such a new vehicle, so what is the trigger for regeneration? We haven’t found the answer to this yet, so if you have any ideas please do get in touch.

More still needs to be understood on the impact of DFP regeneration in real-world conditions and Emissions Analytics is working with Imperial College London to provide data and analysis to investigate the subject further. What is clear is that the anomalies described above, combined with the issue of illegal removal of DPFs from older vehicles, means that it is likely that more discussion on this subject will be required in the near future.

Please feel free to use or share this information with a credit to Emissions Analytics.