Fighting fire with fuel: Why the EU should cap the level of fuel enrichment in passenger car engines

It is summer 2019, and you are enjoying an afternoon with your friends around a campfire in the woods. The day starts well, when a sudden gust of wind fans the flames onto a patch of dry grass, which rapidly blazes towards the edge of the clearing and into the nearby forest. In a desperate attempt to halt the fire, you grab the glass of precious champagne next to you and throw it onto the flames with little effect. In this situation, one can immediately see there are probably a few things that could have been done differently.

In the field of vehicular emissions, some things could also have been better anticipated. The European Commission’s Joint Research Centre (JRC) recently presented new analysis of real-driving emissions (RDE) from publicly available sources. Among other things, they looked at harmful carbon monoxide (CO) emissions, for which there is currently no on-road limit.

The JRC found that certain spark-ignition (e.g. gasoline) vehicles emit several times the laboratory Euro 6 CO limit of 1 gram per kilometer, which is twice the compression-ignition (e.g. diesel) vehicle limit. Vehicles equipped with 1.6-liter or smaller engines tended to emit more CO, with one vehicle emitting as much as 10 times the Euro 6 limit when tested on the motorway. Are these results surprising?

Not really. The culprit is fuel enrichment, a strategy commonly used to boost performance and prevent engine damage from hot exhaust gases. During high power demand, combustion and the exhaust are cooled by deliberately throwing precious fuel (as opposed to champagne) onto the flames. Although this strategy can be permissible under current regulations, the process presents further engineering issues and results in increased CO emissions, which are a precursor of tropospheric ozone formation that contributes to global air pollution, and are harmful to human health.

The figure below shows an example of elevated CO emissions from a 2016 Citroën C3 1.2L dual fuel (gasoline-liquid petroleum gas) tested over the RDE during a 75 km long section of rural and motorway driving. The bulk of emissions occurred during uphill driving and acceleration events (with up to 1.4 gram of CO per second), coinciding with the highest power demand.

motorway and rural sections of RDE test

Average engine power across European manufacturers has steadily increased over the last decade, while the need for better fuel economy has simultaneously led to the adoption of smaller, lighter engines with fewer cylinders. This is known as downsizing. So, what are the ways to achieve more with less?

One option is to “boost” the engine by pushing in more fresh air and, consequently, burning more fuel. However, one drawback to that technology is the risk of abnormal combustion, or engine “knock,” which could destroy the engine in seconds if boost is not properly applied.

Alternately, you could choose to increase fuel flow by approximately 10% more than the ideal amount for clean combustion. This would only yield around a 3% increase in power, but it can be done without increasing hardware cost. However, the small increase in power is accompanied by a large increase in emissions. The extra injected fuel will not burn completely due to insufficient oxygen, leading to the formation of partial combustion products and reduced catalyst efficiency. The incomplete combustion is emitted almost entirely as additional CO from the engine, and the reduced catalyst efficiency increases tailpipe hydrocarbon emissions and further increases CO emissions. But what if, despite taking this questionable route, the result is still not enough power?

With turbocharged or supercharged engines you can use higher boost to increase air and fuel flow, but to avoid the destructive “knock” you will have to modify another element: the ignition timing. Retarding the spark timing makes the engine less efficient, lowers the combustion temperature (thus reducing knock), and makes the exhaust temperature hotter. That would seem fairly counter-intuitive, but to increase power, one must once again sacrifice efficiency using even more fuel and air, a vicious cycle that does not end here. The exhaust temperature can get dangerously close to the material limit (roughly 980°C), creating another risk of engine damage. So, what can you do if one still requires yet more power?

One ostensibly simple solution would be to add yet more fuel to cool the exhaust, in the order of up to forty-five percent in some cases. However, this extra fuel does not release its energy content through combustion. Instead, the fuel is simply taking some of the heat while vaporizing. Such waste translates into reduced real-world fuel economy and dramatically increased emissions.

Available engineering and regulatory solutions

The latest revision of the EU regulation compels manufacturers to declare fuel enrichment strategies during type-approval, such as any Auxiliary Emission Strategy (AES). Such strategies could be deemed legal by type-approval bodies if manufacturers can justify the requirement as a necessary means of engine protection.

The good news is that solutions already exist to reduce the use of fuel enrichment. These include direct fuel injection into the combustion chamber, enhanced charge air cooling, exhaust gas recirculation, electrically assisted or variable geometry turbochargers, liquid-cooled exhaust manifolds, and electric boost through hybridization. Injecting water is also a particularly interesting solution, as it is a renewable resource that is much cheaper than fuel and is six times more efficient at cooling than fuel when vaporized. In addition to these strategies, limiting the engine power is always a viable, albeit less popular option.

Instead of blessing the practice of dumping lots of extra fuel into the engine, the regulatory solution is to limit on-road CO emissions during RDE testing. The testing protocol sets limits for the driving dynamics of a valid test using the product of the vehicle speed with its positive acceleration. The figure below shows that RDE covers a wider range of dynamic conditions than the WLTC laboratory test.

chart of RDE dynamic boundary compared to WLTC

Returning to the analogy of the campfire that threatened to burn out of control: What could have you done differently? Of course, one could anticipate and guard against the fire risk with a bucket of water, rather than wasting glasses of champagne as a last minute and highly inadequate control measure.

But one could also remove the risk entirely by determining in advance a more suitable fire spot, in the same way that vehicle manufacturers would not have to implement additional engine protection measures if they simply design the initial performance target for real-world driving. In other words, the engine should be “rightsized” for real-world driving as opposed to blindly “downsized” for type approval testing.

You could also argue that the forest ranger, the de-facto ‘regulator’ in this scenario, may share responsibility by not clearly defining a reasonable set of pre-emptive safety conditions. In the United States, the Tier 3 regulation not only limits CO emissions in multiple laboratory tests, but also confines the source of the problem by requiring manufacturers to cap the level of fuel enrichment during their most power-intensive US06 test. Rather than allowing manufacturers to put themselves in situations where they are always chasing brushfires with champagne glasses, regulators should stop permitting the manufacturers to set broad fires in the first place. Surely prevention is always better than the cure.