www.theverge.com
Last October, the Chevrolet Corvette ZR1 set a top speed of 233 mph on consecutive runs around a closed track. Thats not the fastest street-legal production car in the world: the Bugatti Chiron Super Sport 300+ will, as the name implies, top 300 mph. Whats special about the ZR1, then? Its $174,995 starting price may sound expensive, but its a steal compared to the Bugatti, which costs somewhere north of $4 million. The ZR1 is officially the worlds fastest production car available for less than $1 million. The ZR1 achieved that speed on a massive test track in Papenburg, Germany, a place where the banking is so steep that the drivers suffered through 1.7 vertical Gs on the turns. Thats just one number out of an endless series of figures that the team behind that record-setting run calculated well in advance, tapping into simulations usually reserved for more utilitarian jobs, like figuring out how steep a grade a Silverado can tow up before blowing a gasket.Here, the only number that really mattered was top speed a figure that, in simulation, differed from reality by less than half of one percent. This is how they did it.Front aerial shot of the Chevrolet Corvette ZR1 coupe in Competition Yellow with optional ZTK Performance Package on the race track. Preproduction model shown. Actual production model may vary.Everyday HeroThe Corvette has always been a machine punching above its weight in the competitive ring of international performance cars. I dont mean that literally, as it is often the heaviest option within sports car shootouts. But, when compared to competitors from Porsche, Ferrari, and Lamborghini, the Corvette is the perennial value option, an everyman American choice out there making the worlds best look a little bit silly.The ZR1, though, elevates that formula to another level with more power and better handling. While the code ZR1 first appeared as an engine upgrade on the third-generation Corvette in the early 1970s a limited option hardly anyone picked it wasnt until the C4 of the 90s that the ZR1 appeared as the top-shelf trim of the Corvette. The Corvette is the perennial value option, an everyman American choice out there making the worlds best look a little bit silly.The first ZR1 was a revelation, compared favorably to period icons like the Ferrari 348 despite costing half as much. The latest ZR1 looks set to continue that legacy while raising the performance bar to new heights. You already know the things top speed, delivered by a whopping 1,064-horsepower, 5.5-liter twin-turbocharged V8. Everything else has been upgraded to match, including carbon ceramic brakes, wheels made of carbon fiber, and sticky Michelin Pilot Sport Cup 2 R tires.Its a machine built to satisfy anyones need for speed. While its starting price means its hardly a good fit for every budget, its still outperforming machines that cost much, much more.The months of work leading up to the test run began with one simple question: where in the world can you safely test a car like this?Members of the Corvette team, including General Motors President Mark Reuss on the track in Papenburg, Germany. Preproduction models shown. Actual production model may vary.The Venue Engineers at General Motors knew early that the new ZR1 was going to be the companys fastest production car ever. The first problem was finding a safe place to find out exactly how fast. While GM has plenty of facilities of its own, including a 4.5-mile circle track in Milford, Michigan, the team behind the record-breaking run knew they needed somewhere bigger.The High-Speed Oval at ATP Automotive Testing in Papenburg is certainly big. The track is 7.6 miles in total length, each straight running 2.5 miles before the asphalt turns back on itself. At 233 mph, those straights fly by in less than 30 seconds. Its the perfect playground for this sort of testing, but since it isnt a GM-owned facility, the team wouldnt have infinite time just three hours spread over Friday and Saturday mornings. The idea was to optimize everything in advance and calculate every detail from car alignment to tire pressure, leaving nothing to chance on the day of. Go do this, and well break some records, Jason Kolk, simulation integration lead at GM, told me.With the location selected, the simulations could properly begin. They started with a virtual model of every element of the car, which the team called a virtual vehicle a digital twin to simulate aerodynamics and engine output as accurately as possible. We even have a virtual ECU and controllers, said Ping Hwang, GMs performance systems integration engineer.In a regular production car, these digital twins are primarily used to track changes during development and test reliability and efficiency. The same physics model can be applied to towing, to range, to different temperatures, said Wesley Haney, senior calibration engineer at GM. It isnt a single simulation that handles all this, but rather a suite, including off-the-shelf solutions like CarSim mixed with custom options written in C, Python, and even Fortran, a programming language that dates back to the 1950s and will surely outlive us all. Despite the digital horsepower provided by GMs high-performance computing centers in Michigan, its safe to say this application pushed all those tools to the limits. We didnt actually have a tire model that was run at that speed. Like, how much is the tire going to expand that speed? Kolk said. Fuel consumption can also be simulated, and there was a novel factor at play here: thrust. Running full-throttle at redline, the ZR1 generates 37 pounds of force from the engine exhaust alone. While thats nothing compared to a SpaceX engine, every little bit helps when youre trying to break records.Beyond horsepower and thrust, aerodynamic drag is a huge factor. But Corvette engineers had already designed a low-drag aerodynamic package ideal for this application, ditching the ZR1s big wing in exchange for speed. So, no tweaks were needed there. Early LapsThe team first applied the simulations to runs on GMs own circle track at Milford, where the car was soon lapping for real at 225 mph, already 13 mph faster than the official top speed of the ZR1s previous generation. Believe it or not, that was unwelcome news. According to the simulations, the car should have run faster. Jake Hedrick, virtual propulsion engineer at GM, started poring over the data. After further investigation, he determined it wasnt the simulation that was off, but the car itself. What we were seeing is that, analytically, we were about 80 horsepower down, Hedrick said. That was due to multiple factors, but mainly driven by the spark calibration.Spark here refers to ignition timing, when the engine controller in the car tells the spark plugs to fire. If its too early, the explosive force of combustion fights the rotation of the engine. If its too late, you dont get complete combustion of the fuel before the exhaust valve opens and the next cycle begins.We even have a virtual ECU and controllers.With this real-world data, the team optimized the engine mapping, unlocking the missing power the simulation said should have been there. They also analyzed things like ride height and alignment, too. We found if we lowered the car just a hair, we could get a tenth of a mile per hour, Kolk said. But the team decided to leave that on the table due to time constraints. That just left the question of weather, the one factor that the team couldnt control. They could, though, look at historic trends. The team analyzed years of historic conditions to find the ideal time for their run: cool temperatures and low wind. Wind is bad because, to set the record, the cars speed is averaged over two runs in opposite directions. Any help you get from a tailwind will be more than negated when headed in the opposite direction. But our wind on Saturday was almost nonexistent. It was really ideal conditions, Kolk said, setting the stage for speed. The Big DayThe simulations had factored in everything possible and predicted everything they couldnt calculate. The team even determined the optimal place on the track where the driver should open the throttle wide for the run, or go to WOT as the team put it. But there was still the human variable. None of the simulations told the team just how stable or unstable the car would feel when going around the 50-degree banked turns at over 200 mph. The driver would need to put their foot to the floor well before they got to the straightaway.Would the driver feel a sense of like, I cant go to the pedal yet? Kolk said. Or, if they went WOT too early, would that add a lot of heat to the system?Chris Barber, lead development engineer on the Corvette ZR1 and one of the drivers at the track, said that everything felt right from the beginning. When we did our first actual 200-mph-plus test, all the drivers were so confident that we just went for it, he said.It was amazing to just go to wide open throttle well over 200 miles an hour, Barber said, describing what it was like to go screaming through the turn at three times the average highway speed limit and then putting his right foot to the floor. Youre almost looking sideways at the sky... then once you get onto the straightaway, it just whizzes by so fast. It was amazing to just go to wide open throttle well over 200 miles an hour.But there was a last-minute issue. During Fridays test runs, the car was losing speed, coming up short of projections. The ZR1s V8 engine is designed to run at 8,000 RPM. In the perfect world of the simulation, thats fast enough to hit the maximum velocity of the Corvette. However, reality isnt so smooth. With the tires skipping over bumps, the revs would spike, causing the car to hit the rev limiter. Once you touch that limit, it wants to shut the party down, and it effectively ends the run, Haney said.To fix this, the team raised the rev limit by 100 rpm only in sixth gear, giving just enough headroom for the engine to bring the car to its top speed without stumbling. That raised redline, plus the revised engine timing, will all be part of the production ZR1, which will include a driving mode aptly called Top Speed. With that, everything was set. But instead of relying on a test driver to make that final run, GM president Mark Reuss got the nod. The team wanted to show that it didnt take a professional to extract this kind of performance from the machine. The final result, with Reuss at the wheel, was an average of 233.29 mph. How does that figure compare to the simulation? When the team took the exact weather conditions and plugged them into the sim, the number was 232.2, just 1.09 mph off. The team thinks the difference comes down to the tire diameter increasing at speed, something the simulation cant handle not yet, at least. Kolk said that result was proof that the simulations are absolutely on the money, which will help engineers optimize more pedestrian products to solve equally challenging problems. Aerodynamic and thermal performance are obviously key when youre trying to top 230 mph on the track, but its even more important in some of the most complicated on-road challenges, like increasing the range of heavy-duty EVs when towing. We do all this work up front so we dont have to worry about one system holding everything else back from achieving something really great, Kolk said. Better simulations mean engineering teams can develop cheaper, lighter parts more quickly, iterating in the digital world rather than the physical.So, while driving simulators are a good thing for gaming, they can certainly help you out in the real world, too. That was by far my high score, Barber said.See More:
