One of the big draws at the 2014 Paris Motor Show was the unveiling of the Jaguar XE. Critically acclaimed, the compact executive sedan took centre stage on the Jaguar stand, lauded for its design, fuel efficiency and its potential for the Jaguar brand.
The BMW 3 Series contender is the first Jaguar to use the company’s new aluminium modular vehicle architecture. It is also the first car in its class to make intensive use of aluminium – the material accounts for three-quarters of its structure.
Less well known about the XE is that it is the first car to have moved into production without a prototype build for aerodynamic testing. Instead, digital simulation was used for all aerodynamic testing, as well as key thermal management applications such as heat exchanger performance, grille sizing (with active shutters), brake cooling and aeroacoustic wind noise design evaluation.
The work was carried out using simulation software developed by Exa Corporation, a US$54.5m Burlington, Massachusetts-based software company headed up by Steve Remondi, President and Chief Executive.
The use of simulation software in automotive R&D is on the rise as the technology becomes more reliable, and viable when compared to physical testing. Crash testing, for example, is by its very nature expensive, and so OEMs are exploring ways to take those tests into the virtual world. Despite this new and growing acceptance of simulation software, switching from the physical to the virtual requires a major leap of faith on the part of the OEM. The only way to ensure that level of trust is to prove the simulation software’s accuracy over time.
Exa’s development work on the XE was the culmination of many years’ collaboration between the two companies. “We’ve been working with JLR for almost 15 years, and we’ve moved through the development process,” Remondi tells Megatrends. “Jaguar used our software to influence the design process and brought the drag down to just under 100 grams per kilometre. With an aerodynamic Cd rating of 0.26, it’s the most efficient drag Jaguar has ever developed. And it was done with our software.”
Although Remondi is able to confirm that Exa worked extensively on the Renault EOLAB concept car, which was also launched at the 2014 Paris motor show, he will not allow himself to be drawn on other production cars in development along the lines of the XE programme. Nonetheless, he is proud of Exa’s customer portfolio: “Most of the top OEMs use our product at some level. JLR is one of the more advanced, using it on more disciplines than many others. Jaguar is pushing the envelope, and pushing our envelope faster, but that’s mainly because it’s a more aggressive and agile company.”
With a limited number of wind tunnels available to vehicle manufacturers, the cost and wind tunnel time are at a premium. In addition, OEMs spend vast sums of money on the physical prototypes. Switching from physical prototypes to simulation has considerable benefits, explains Remondi. Cost aside, these include the ability to make design changes earlier in the development cycle, and the ability to involve all relevant parties throughout the OEM in every stage of the vehicle’s development by sharing real time data.
(Image credit: NASA Ames/Dominic Hart)
“By using digital simulation, you can move the vehicle up into the studio, and have ongoing conversations about every part of the car every day. As the designer works on the car, the software provides feedback on how it performs, not only to the designer, but to any other parties looking at the software at the time, no matter where they are located.”
All of the aerodynamics and fluid simulations are calculated by the software, which results in countless lines of code. To facilitate analysis of that code, the software also creates a visual display. “We present a realistic lifelike view that means the data can be explained to designers and marketing and sales and management so that they understand what’s going on.”
To develop a level of trust with a vehicle manufacturer that permits a switch from physical to virtual prototyping requires years of running the software alongside physical prototypes. The leap of faith from physical to virtual has been made by Jaguar; but what role can simulation software play among the mainstream brands? “On mainstream brands, it’s equally important,” says Remondi. “There are US pick-up trucks that have extra thick side glass to reduce wind noise. That means a higher cost of materials and thus lower profitability, because they didn’t get the wind noise right during the early stages of development.”
Assessing the interior noise of a vehicle can help an OEM to cut down on unnecessary material usage for sound dampening. “Wind noise is one of the top customer complaints. It’s often caused by the A-pillar vortex. We can calculate exactly the pressure fluctuations, transfer that to the interior and play precisely what that wind will sound like inside the car. JLR uses this heavily on this car because it wants a quiet interior. That’s very important to Jaguar customers, Jaguar has done a lot of work on this up front because changing the A-pillar or the hood means changing the front cowl, and the transition to the windshield, and from the windshield to the side glass. That has to be done up front, because it’s the most critical aspect of the car. Traditionally, the NVH guys, the noise guys, get the car last.”
For reasons of cost, finding a solution for a higher end car is less complex than for mainstream models. “It’s often easier to solve a problem if you have premium brands that can absorb the costs. And this applies to everything. Fuel economy matters in every car in every segment. Now noise matters in every car, in every segment. Jaguar is going to have a low interior, and it will use premium materials to achieve that. But other brands still have to work on it, because if they don’t work at it at all, they’ll be worst in class. Wind noise shows up in your JD Power scores. So if you’re worst in class, no matter what class you’re in, that’s going to hurt sales.”
In addition to working with car manufacturers, Exa works with truck OEMs. “We worked on the SuperTruck programme in the US. The EPA in the US allows our truck customers to certify to government fuel economy standards based on our results.”
The SuperTruck programme involves a whole range of different stakeholders in addition to the OEMs. “We’re doing a lot of work with the trailer manufacturers as well as the tractor guys. And there are third parties and even fleets involved. The big fleets even have the economic incentive to now look at their own devices and their own configurations. We showed them that we could make a truck almost double its fuel economy, mainly by using aerodynamic devices.”
A key difference between using a physical prototype model in a wind tunnel and using simulation software is that when developing with simulation software, a final prototype is built to confirm behaviour, not discover behaviour. Traditionally, changes have been left to the end of the development cycle. “Most car manufacturers will tell you that the design is supposed to end at a certain point, and the engineering work is supposed to end when they release the tooling and the design to manufacturing,” says Remondi. “And the costs keep going up because this is when they’re testing and finding problems.”
The real world is full of unpredictable conditions, such as road bumps and incorrectly inflated tyres. Remondi grins; the software, he says, can also account for unpredictable road surfaces, weather and environmental conditions. “We have the ability to do surface roughness, vary tyre conditions, we can even do upstream turbulence if we have measurements over what turbulence comes to the car from the upstream traffic. We can plug that into the system too and do everything required for full real world predictions.”
The XE is proof that the best aerodynamics can be achieved through virtual prototyping. Looking to the future, what else could this technology enable OEMs to do? “The next thing we’re doing is fuel economy over the entire drive cycle.” At the moment, drive cycle testing involves fixed test rigs and human drivers driving cars for tens of thousands of miles on test tracks. Switching such tests to simulation software would enable an OEM to run numerous simultaneous tests to assess differing outcomes, further increasing efficiency. “With simulation, OEMs get a better car. No panics, no rushes at the end at the end, and still they get a better car. That’s ultimately what we’re after.”
Martin Kahl
