The regulation that governs testing under the Worldwide Harmonised Light Vehicles Test Procedure (WLTP) is a mammoth document at 589 pages.
There has been a fair bit of media interest recently in the work of the Australian Automobile Association (AAA) testing EVs in the real world and how real-world range often falls short of the WLTP rating that is provided by manufacturers.
In this article, I’m going to specifically discuss the test procedure for determining EV range in a battery EV and why it is so hard to replicate in real world driving. There are many more aspects to WLTP such as testing fuel consumption for ICE vehicles, calculating EV energy consumption, how PHEVs are tested, etc. but The Driven tells me I don’t have 589 pages to cover it all.
What is the point of these standardised tests?
The point of WLTP and the earlier test standards like the New European Drive Cycle (NEDC) is to test vehicles in a way that is objective, repeatable and provides a fair way to compare vehicles.
Of course, it’s helpful if it can also be representative of what drivers can expect in practice, but that is very difficult to achieve with so many variables at play in real world driving.
Indeed, it’s quite possible to have two vehicles with an identical WLTP range that will perform very differently under a different set of test conditions than those specified in the standard (for example, in very cold weather).
That’s just a limitation of the test procedure. It’s possible that we could do it differently, but we don’t, and that is probably to keep things simple for consumers who have to digest this information.
How does WLTP test the range of an EV?
WLTP has the concept of a test cycle. In a test cycle, the car is put on a dynamometer (dyno) and driven for 30 minutes covering 23.25 km at specific and varying speeds as set out in the standard.
During that 30 minutes, the car goes through four phases representing different driving styles: low speed (urban), medium speed (suburban), high speed (rural roads), and extra-high speed (motorways).
The top speed is 131 km/h. The dyno sets the resistive force of the rollers to simulate resistance due to aerodynamic drag. Importantly, it does not simulate climbing hills. All four phases assume the road is flat. That’s a big assumption.
There are other test conditions that are carefully controlled for replicability. The test chamber is conditioned to 23°C and the vehicle is left in the chamber for 12 hours to “soak” so that all of the components (including the battery) will be at a uniform 23°C. The weight of the passenger and other loads are carefully controlled, and the heating/cooling functions are turned off.
An interesting rule exists regarding regenerative braking. The test procedure requires the car to be set on the regenerative braking mode that is the default one from the factory. Even though the test cycle has no hills, there are continual changes in speed and when slowing, some of this energy can be recovered with regenerative braking.
Conceptually, it’s pretty simple to determine the EV range. The car is fully charged and repeatedly put through the 30 minute test cycle on the dyno until the car can’t sustain the necessary speeds (eg, it goes into turtle mode) or it stops. Suppose that a car was to drive exactly 20 test cycles before the battery is drained. That would give a WLTP range of 23.25 km × 20 = 465 km.
An example of a dynamometer (“dyno”)
CC-BY-SA-2.0 licensed image from Wikipedia
Why can’t I achieve the WLTP range?
The most important thing to understand from the test cycle is that highway driving in Australia is closest in driving style to the extra-high speed phase. The other three phases all entail driving that is slower and more varied.
EV drivers all understand that the detriment of stop-start driving in city traffic is considerably reduced by regenerative braking. This is counterintuitive for people used to ICE vehicles achieving better fuel economy on the highway than around town.
The issue, as I see it, is that many new EV drivers bought their vehicle with an eye on the range printed on the windscreen sticker. They judge the range performance of their EV on a long highway drive where range is most important to them.
Unfortunately, this is comparing apples and oranges. You can hopefully see that a driver will struggle to ever achieve the WLTP range on a highway trip because the entire journey characteristically looks like the one phase of the test cycle that is least favourable to EVs.
As I mentioned earlier, there are also a lot of variables that are tightly controlled in the WLTP test procedure. When you drive at high speed on a highway, things that will differ from the test cycle include terrain, the presence of headwinds or tailwinds, air temperature (different battery chemistries will behave differently in cold weather), more or less weight in the vehicle, rougher or wet road surfaces, whether you’re using cabin heating or cooling, and even the regenerative braking mode you prefer.
It is common and unsurprising that EV drivers often achieve a higher range than the WLTP rating in urban driving and a lower range on the highway since the test procedure aims to model a mix of the two modes.
Unfortunately, it’s not easy to estimate the range assuming the entire test cycle was completed using one phase only. For that, it’s probably best to rely on a test drive, the experience of existing owners of a model, or the testing undertaken by the likes of the AAA.
Ben Elliston is an independent energy researcher and the ACT branch chair of AEVA.
