How Hyundai made its fuel economy error
Could other automakers? claims also be off the mark?
One of the major automotive news stories of recent weeks, at least in terms of mass-media coverage, has been the revelation that Hyundai and Kia overstated the fuel economy of some of their 2011 and 2012 model-year vehicles.
In a joint statement, the two revealed that: ?Procedural errors at the automakers? joint testing operations in Korea led to incorrect fuel consumption ratings for select vehicle lines.?
As a result of these corrections, the average combined fleet fuel consumption ratings for each of the companies increase by 0.3 L/100 km for the 2013 model year.
Perhaps equally surprising as the error, both companies voluntarily announced they will reimburse buyers of the affected models for the additional cost of fuel used annually, based on the difference between the initial and revised fuel-consumption figures.
That compensation will be calculated individually to reflect the kilometres driven by each owner annually and the price of gasoline in his/her region, so long as the current owners keep their vehicles.
In addition the amount will be bumped by 15 per cent to help compensate for any inconvenience experienced, according to Hyundai Canada CEO Steve Kelleher.
Of course, the announcement of the error almost immediately triggered cries of ?foul? and at least two class-action lawsuits in the U.S.
But to those familiar with all the intricacies of the emissions and fuel economy testing procedures ? the two are tightly related ? the question may not be so much about how such a thing could happen as why it doesn?t happen more often.
Perhaps the most important thing to understand in explaining the error is that fuel consumption ratings published by Natural Resources Canada and the Environmental Protection Agency (EPA), in the U.S., as well as advertised by automakers, are not determined by actually measuring fuel consumption.
For vehicles using carbon-based fuels (such as gasoline, diesel, ethanol, natural gas, propane, etc.) they?re calculated by determining the amount of carbon in the exhaust during selected segments of the tailpipe emissions testing process.
Those tests aren?t conducted on the road but in a laboratory with the test vehicle driving rollers connected to a dynamometer, programmed to simulate the load the vehicle would encounter at the same speed in real-world driving.
As a parallel, think of an exercise bike on which you can adjust the load you have to overcome when pedalling.
In order to determine what load should be programmed into the dynamometer at various speeds, tests must be conducted on prototypes representative of each vehicle or range of vehicles to determine actual road load over the same speed range.
It?s at that stage of testing that the Hyundai/Kia errors originated. With a procedure called coast-down testing.
In a coast-down test, the subject vehicle is accelerated to a speed of about 120 km/h, then allowed to coast down, in neutral, to a speed of 8 km/h.
During the coast-down process, time and speed are monitored and recorded continuously and from those data, the total drag force acting on the vehicle can be calculated. And from the drag force, the power required to keep the car rolling at any given speed.
Once the drag force is known, the power required to drive the vehicle, and accordingly the required dynamometer load at any speed within the test range, can also be calculated.
Coast-down tests must be conducted multiple times, on a smooth, level tarmac or concrete surface and in opposite directions to ensure that the resulting data are statistically valid.
There are strict constraints on all aspects of the test, from temperature and wind direction and speed to tire pressure and tread wear to the break-in on the vehicle itself.
They?re defined by an SAE (Society of Automotive Engineers) procedure (J2263), but as with any kind of regulation there is some room for interpretation. In this case, as it turned out, the interpretations made by Hyundai/Kia?s engineers did not align precisely with those of the EPA.
That fact became apparent only when the EPA conducted its own test audit on one or more of the subject vehicles and came up with different results.
Typically, manufacturers conduct their own fuel-consumption tests and submit the results to EPA for approval. EPA says it conducts audit tests on only about 15 per cent of all the vehicles it certifies.
Once the exact cause of the differences was apparent, Hyundai and Kia revised their internal procedures and announced the correction to the public.
All of which raises the question, are there other automakers whose internal interpretations differ from the EPA?s?