Weighing the cost of a new van

EA is in the business of improving the quality and availability of data in the automotive sector and so we will soon be introducing on-road testing for light commercial vehicles in the UK, helping consumers make informed decisions about which vehicle to buy with an easy to use miles per gallon calculator.

A major contributing factor to fuel economy for vans and pickup trucks is the load on board. So, in addition to our standard fuel economy test, we will include repeat cycles with the vehicle fully laden at its maximum payload. This will enable the buyer to consider how they expect to use the vehicle and find out what effect this will have on their fuel consumption.

In a 2014 report, the RAC Foundation found that the carriage of equipment, and delivery and collection of goods accounted for 79% of the mileage of light commercial vehicles in the UK. However, this is unquantified in the official tests.

The regulators would argue that testing vans and trucks unladen provides a level playing field, allowing for comparisons between vehicles. Yet in this sector of the market, carrying capacity varies within a CO2 class. The fully laden Ford Transit Courier can carry 112kg (247lb) more than the Vauxhall Corsavan but both record CO2 of 103g/km. In theory they should give the same fuel economy – but only if driven empty!

USA perspective

Although the EPA’s five-cycle test regime produces official figures which, on average, more closely match real-world performance than in Europe, the test still does not account for a fully-loaded vehicle.

Thus the question facing the consumer, who is purchasing a vehicle for its intended purpose, to carry stuff, is the same on both sides of the Atlantic – how much will it really cost to run?

We have been addressing this issue from our Motor Trend test centre in Los Angeles, looking into the impact of variable payload. For example, when we packed the Nissan NV200 with bags of cement increasing its payload to 1132lb (513kg) – its maximum is 1470lb (667kg), and drove it on our 88-mile (142km) city and highway test route, we saw a reduction in fuel economy from 25mpg to 22mpg (US gallons). Read more. Based on the average distance driven of 15,000 miles, that equates to an additional 82 gallons (310l) of fuel per year, which multiplied by the 8 million or so new commercial vehicles registered each year is a lot of extra gas.

Vans and pickups are an increasingly important sector for UK and US drivers, so this data will help us understand their real-world performance. Watch this space for trends and observations as UK testing gets underway in March 2015.

Ploughing ahead with tractor testing

Working with Emissions Analytics, engineering researchers at Harper Adams University have developed a new method to accurately measure the exhaust emissions of agricultural machinery.

Traditionally, machines such as tractors are tested in controlled environments such as a soil hall, not taking into account the many variables from real-life scenarios, which result in transient engine loads.

With support from CLAAS and EA, the postgraduate students used PEMS equipment quite literally in the field. Two engine exhaust after treatment systems commonly used by tractors to reduce emissions were tested; a CLAAS Axion 830 with a selective catalytic reduction system and a CLAAS Arion 650 using exhaust gas recirculation combined with a diesel particulate filter.

Both tractors were trialled in three conditions – using a dynamometer to add a variable load to the PTO shaft; road testing at high speed with a fully laden trailer; and in a field with a subsoiler and power harrow. As with road vehicles the CO, CO2, NOx and THC data was combined with vehicle data such as GPS position, engine load and speed.

The project was designed to investigate the differences between theoretical CO2 conversion factors and real-world factors for non-road mobile machinery; and to demonstrate the feasibility of using PEMS equipment for such a task. The results, analysed by Miles Metcalfe with supervision from PhD student Rob Fillingham, were written up for his MEng Agricultual Engineering dissertation for which he was awarded a first class.

Metcalfe demonstrated that the assumed linear relationship between engine load and CO2 conversion factor was in fact better suited to an exponential curve, and that by using the traditional conversion factors supplied by DEFRA, CO2 is in fact being over estimated as there is a significant difference of p<0.001 between the DEFRA value and the PEMS result.

Observations made, such as an increase in CO2 conversion factors whilst the tractor was turning at lower engine speeds or loads at the field’s headlands, show that this innovative collaboration between Emissions Analytics and the agricultural sector has the potential to allow farmers to not only save fuel but reduce damaging emissions by using data gathered from real agricultural practices. And, as regulations change for N-RMM, the  power of PEMS to measure compliance is evident.

Real Driving Emissions – are you ready?

Emissions Analytics was recently asked to write an article for Automotive World’s Megatrends magazine. This month’s newsletter is a summary of that article which considers the potential impact of the proposed changes to the New European Drive Cycle.

Emissions Analytics’ data resource, from tests on more than 800 vehicles, is transforming the economics of obtaining emissions data for OEMs who are tasked with understanding and acting upon the proposed legislative changes concerning Real Driving Emissions (RDE) and the move towards the World Harmonised Light Test Cycle and Procedures (WLTC/P).

The new testing system, developed by global representatives for the United Nations Economic Commission for Europe is due to be finalised in the spring of 2015. This test cycle is more representative of real-world driving and the test procedures should be more robust than those associated with the New European Drive Cycle (NEDC).

One of the reasons for the proposed change is the growing gap between the number of miles per gallon certified during the NEDC test and the fuel economy achievable by real drivers on the road which can be seen clearly below.

Using Emissions Analytics’ real-world data, the blue line on the graph below illustrates how this gap is growing, at about two percentage points per year, and is likely to continue to expand if left unchecked. With the introduction of the WLTC/P in 2017, we predict the gap will close by between half and two-thirds (shown below in green), depending on how stringent the final protocol is.

Although this will bring the European divergence closer it will not equal the USA variance, where the more stringent five-cycle system is in operation.

Bold line: EA data 
Dashed line (historical): Other sources of data
Dashed line (future): Model based prediction by EA

The European Parliament and European Commission have proposed this new test be introduced in 2017, although there are challenges and opposition from some parts of the automotive industry that would like longer to adapt to the changes.

One of the challenges facing OEMs is the profile of the NEDC replacement, the WLTC. The International Council on Clean Transportation has estimated that the effect on the EU CO2 target value will be an increase of around 5-8%. Emissions Analytics believes the increase could be higher than this.  If nothing were changed in the targets, OEMs would need to deliver further efficiencies in their vehicles, and consumers in some countries could find themselves paying more vehicle tax.

There is a methodology under development for translating the existing NEDC results into WLTC, but this is still work-in-progress and has limitations. What is clear is that forewarning of how current vehicles perform on the test can bring significant benefits to the engineers developing the vehicles which will be on the road when the WLTC/P is adopted. This is why some manufacturers use Emissions Analytics’ data to ensure compliance and to stay competitive, benchmarking their own progress against that of their closest rivals.

In times when manufacturers are under increasing pressure to be open and honest about their vehicles’ true in-use performance, plus with the imminent legislative changes which will formalise this requirement, there has never been a greater need for a reliable and robust source of data which can offer the insight and intelligence needed.

Emissions Analytics will shortly be making their data available to the automotive sector directly via a new, subscription based software platform called RDEanalytics. More will follow on this in a later newsletter but for a sneak preview or to find out more email us now.

Hybrid efficiency put to the test

Despite common perception, the advantage of hybrids over frugal diesels is often illusory, if judged solely on fuel economy. Having tested over 30 hybrids in the UK and US, Emissions Analytics is able to analyse the data to understand how they really perform.

To illustrate the point we have taken a sample of 10 vehicles tested since 2013 – two standard hybrids versus 8 diesels – from the real-world fuel economy testing we conduct with What Car? in the UK. Each has an engines in the 1.5 to 2.2 litre range, power up to 150bhp, two-wheel drive and with hatchback, saloon or estate body style. The table shows the sample, ranked by fuel economy with the best MPG at the top:

While hybrids deliver good fuel economy in real driving, they can be eclipsed by up to 10mpg by some non-hybrid diesels. And that is after having taken into account any net changes in battery charge levels, to ensure that the hybrids are not penalised over our cycle. For certain driving patterns however, hybrids may still be the better option. Over our complete dataset of more than 500 vehicles in the UK, we can quantify how average MPG changes under congestion and aggressive and fast driving.

What this data shows is that hybrids suffer much less than their ICE equivalents under congested urban driving: on average a 3% penalty compared to 7%. In contrast, by doubling the average rate of acceleration the MPG falls by more for hybrids, especially diesel hybrids.

Comparing motorway driving to town driving, all types of vehicle show better MPG on the former, but the difference between hybrids and ICE vehicles is dramatic – typically because the downsized engines found in the hybrids are less suited for high speed motorway cruising.

Even more than their tolerance of congestion, the value of hybrids may be in their pollutant emissions, as even the cleanest diesels typically exceed the regulated values of NOx. In a recent report by the International Council on Clean Transportation, which analysed data from Emissions Analytics, the average exceedance was seven times for the latest Euro 6 diesel cars.

This compares to petrols, which generally meet the regulated NOx standards, even in real-world driving. Carbon monoxide is higher for the petrols, but again within the regulated values. Therefore, petrol hybrids have the benefit over ICE diesels in their effect on air quality, made even better as a proportion of urban driving will be on battery, with zero emissions. Although not included here, plug-in hybrids can show this pattern even more strongly.

In summary, hybrids deliver good but not best-in-class fuel economy, but they are typically the cleanest, and if you are a light-footed, congested town driver, they are ideal.

Beware the danger of downsizing

The old adage that good things come in small packages is not necessarily true in the automotive sector, at least not when it comes to engine size.

Looking at the data in the table below, based on real-world tests from over 500 model variants, it can be seen that engines under one litre have the greatest variance from their official figures. As well as achieving 36% less than you expect in terms of fuel economy, you will only achieve 3mpg more than the average of all the cars we’ve tested, which includes thirsty performance cars.

For maximum fuel economy you should look for a 1-3 litre engine, as these will return around 45-46mpg. And, to avoid being too disappointed with the result, pick a 2-3 litre vehicle as it will be only 15% worse than you were told you could achieve.

To understand why these ‘eco’ engines perform so poorly on the road, it’s necessary to look at the miles per gallon penalty of accelerating.

The graph shows that the smaller the engine, the larger the penalty acceleration has on mpg. And because the NEDC test has relatively few episodes of acceleration and those that it does have are of a gentle nature, these small engines appear to perform well and receive a high mpg result. That is, until they are taken on the road and driven by real drivers when every rev suddenly burns through the fuel.

Now contrast the US data, where the EPA five test cycle contains much harsher boundaries and you can see that although there are large engines and an overall lower average mpg, the variance to the statutory figures is very small. You may not get as much from your tank – average European fuel economy is over 40% better than in the US – but at least you know how much not much is.

In fact, the US is at a much earlier stage in the trend towards downsizing than Europe, evidenced by the fact that so far we have tested no vehicles in category below 1 litre.  Perhaps the more strenuous EPA tests will put a limit on such downsizing – or, at least, downsizing without down-weighting.

Measuring the manufacturers: The EA league table

As promised last month we are going to examine our fuel economy data in more detail, particularly looking at how different manufacturers perform.

Of the 459 vehicles included in this study, spread across 26 manufacturers, the average real-world mpg by OEM ranges from 31 to 54mpg, with a mean of 44mpg. The deficit from the official figures ranges from 7% to 23%, with the average 22% below NEDC.

The graph, which ranks manufacturers (excluding manufacturers where we have tested fewer than five of their vehicles) according to the extent of their variance from official figures, reveals some interesting observations. Three of the OEMs towards the right of the graph with lower mpg are performance-car-only manufacturers, specifically OEMs 19, 23 and 26. If we exclude these, then the range of real-world mpg shrinks to 40 to 54mpg. By comparison the NEDC figures for these vehicles range from 50 to 63mpg, almost the same spread but 20% higher.

There is then one outlier that does worse than average (OEM 10) and two that are significantly better (OEMs 21 and 25). These manufacturers may have a model range which causes this skewing effect, for instance if they have few big cars in their model mix their average mpg could appear better than other manufacturers. When these three OEMs are also removed from the analysis, the miles per gallon narrows even more from 42 to 49mpg.

So, despite the remaining 20 manufacturers representing a variety of nationalities, technology types and target audiences, their vehicles have surprisingly similar performances in the real world. However, the variance from the official figures still ranges from -13% to -24%.

Although I promised to keep this anonymous, I feel confident that the top performer won’t mind being mentioned. Therefore, I can reveal that OEM 26, which has the smallest variance from the official combined mpg, is Porsche. It also has the second lowest absolute performance, perhaps not surprising given its model line up.

Our consultancy service allows OEMs to see their vehicles in direct relation to their competitors. Contact us if you would like to find out more.

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.

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.

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.

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  robin.north@imperial.ac.uk

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.