Regulators to crack down on phoney fuel figures

NEW car buyers are not the only people to notice that the gap between a new car’s rated fuel efficiency and its performance in the real world is becoming larger.

According to a speaker at this week’s Cars of Tomorrow Conference, car-makers become more sophisticated about how they prepare their new models for testing by national authorities, the results achieved have become unachievable for regular drivers, The authorities even suspect that some car-makers are actually inserting software into the on-board computers in their cars that recognises when the cars are being tested and sets everything to full lean, and limits rates of acceleration.

There is now a movement amongst the regulators to change the testing procedure so that the results better reflect performance in the hands of normal drivers, said Anup Bandivadekar, the passenger vehicles program director at the International Council on Clean Transportation (ICCT).

The ICCT is a non-profit organisation that offers unbiased research and scientific analysis to environmental regulators around the world.

“It’s like a cat and mouse game. The regulators say I know you are doing this, but then the car companies have pretty smart engineers too,” he said.

This game reached the headlines in November last year when Hyundai and Kia were forced by the US Environmental Protection Agency to admit that they had overstated the mileage achieved by one third of the cars they were selling in the US. It turns out none of their cars could achieve the claimed 40 miles per gallon (5.88 litres/100kms).

“We don’t know right now, but we are trying to understand whether the car knows if it is on a test dynamometer, and we think there are cars that know,” Dr Bandivadekar said.

“(The car would know) because when you are on a dynamometer, essentially the rear wheels aren’t going anywhere, so your front wheels are spinning, and you are putting in some loads.” “We don’t understand it fully, so I hesitate to make any categorical statement, but I think that cars know that they are being tested and, when you test them, you get a better result.” In order to counter this possibility, and to make the actual test more realistic, authorities are considering changing the test procedures.

“Right now the testing cycle contains really smooth acceleration, very steady cruising, smooth deceleration. But people do all kinds of things with their cars,” he said.

He admitted a test cycle cannot be designed to emulate the worst driving habits, but he said the procedure could be more accurate.

“You could build a test cycle that has greater acceleration/deceleration, more stop and go behavior that is more reflective. I think that is going to happen.” Engineers at the car companies have refined their approach to efficiency testing to get the maximum result, right down to making sure the car and the rig are at exactly the right temperature to get a maximum result.

Left: Anup Bandivadekar



“There are a million little things that the guys who are doing type approvals of the vehicles can do. Their job is to ask how can I get the best rating for my car, so they are fine-tuning.

“They are looking at the test procedures and saying I can play around, I’m going to push it to the limit. I’m not going to break any laws, but I am going to the limit.” For instance, Dr Bandivadekar said it was common practice to fit test vehicles with the lowest rolling resistance tyre available in the range.

“In Hyundai’s case, for example, the issue was the road load coefficient. You want to input what the weight of the vehicle is going to be and what the aerodynamic is going to be.

“Essentially, there are certain coefficients for aerodynamics and rolling resistance, so you could put on the best tyres. But large volumes of your cars may not necessarily have those same tyres.

“If your testing procedure is not making sure that the highest volume component is there, rather than a show-piece thing, it will vary from jurisdiction to jurisdiction.

“With Hyundai, it seems like those road load coefficients on rolling resistance and aerodynamics and weight were a bit optimistic in order to hit their targets.” Dr Bandivadekar said it was possible authorities may switch to a different system of rating cars, where tests would result in a rating, or fuel consumption/emissions that the vehicle must never exceed, a so-called “not to exceed” standard.

He said heavy trucks in the US are already rated on a “not to exceed” basis and that European authorities were “seriously considering” adopting this approach when the Euro 6 standards are introduced in 2016.

“You want the comfort of knowing the emissions are reasonably below the standard limit, that you are not just barely meeting the standard in the real world, that you are doing better than that.

“That means you are leaving room for deterioration that happens in practice. Especially with these post-combustion devices there is deterioration, and you want to leave room for that and right now it is not clear whether there is that much room, so we’ll see.”