Power play, p.18
Power Play, page 18
One engineer recalled using the time on the flight to ask Musk his opinion on the suspension characteristics for the Model S, a subject they had been debating. Because Tesla was building their car from scratch, such questions were purely up to them. Was the car’s handling going to be sporty, like a BMW, or more giving, like a Lexus? Musk paused, looking directly at his engineer. “I’m going to sell a fuck load of cars, so whatever suspension you need so I can sell a fuck load of cars—that’s the suspension I want.”
Maybe the engineer caught Musk on a bad day. Or maybe it got to the heart of how Musk ran triage. Running two complex operations, there was only so much maniacal obsession he could offer. He would give flippant answers until precisely the moment when something became the object of his focus, at which point he gave every fuck. In such a world, you’re delegated to and given full authority—until he turns his attention on you and your little fiefdom.
This engineer decided the best way to keep a career at Tesla was to avoid more travel with Musk—best not to fly too close to the sun.
As Rawlinson’s team worked to develop the Model S, one of its senior leaders, whom Rawlinson had inherited from his predecessor, presented to Musk a plan for helping the team prioritize the vehicle’s features. He said GM and Ford engaged in similar processes; if Ford was developing the Fusion, it would gather all of the data it could on competing cars, rank each function, then decide which attribute it wanted to exceed, making trade-offs when needed.
Musk listened to the senior leader for about twenty minutes before cutting him off. “This is the stupidest thing I’ve ever seen,” he said, before walking out. “Don’t ever show it to me again.” Musk didn’t want to prioritize one thing over another, he wanted to prioritize everything.
About a week later, the manager was gone. This wasn’t a new occurrence; the team had seen a lot of turnover, much of it attributable to either the company’s dire financial situation or the need to move engineering prowess from Detroit closer to headquarters. Musk gathered the departed leader’s remaining colleagues together.
“Look they’re good engineers, but just not good enough for the team,” he said.
* * *
—
While Rawlinson set up shop in Hawthorne, JB Straubel and his battery team were among those still in Silicon Valley. Tesla had moved into its new headquarters in Palo Alto in 2010, ahead of the IPO. Far from Rawlinson and his staff, Straubel’s team was developing its own culture. By comparison it was a bastion of stability. Many of the recruits he’d hired in the early days through his connections at Stanford had stayed around, growing into new roles or developing even greater expertise in the world of batteries.
The success of the Roadster coupled with Kurt Kelty’s persistence began to open doors in Japan. Straubel had chastised Kelty for continuing to stop by Panasonic in Japan every couple of months, especially after they had found in Sanyo a willing partner for the Roadster (not to mention Panasonic’s letter indicating that they had no intention of doing business with Tesla). But Kelty still believed Panasonic’s batteries were superior enough to be worth the trouble, that each cell could hold more energy than what they were getting with Sanyo.
It was in pursuit of their business that, in 2009, Kelty had found himself in a small meeting room at his former employer, sitting with Naoto Noguchi, the president of the company’s unit responsible for battery cells. The walls were yellowish from Noguchi’s chain-smoking. Kelty sat on his heels at a traditional short-legged Japanese table, trying to balance his laptop computer as he presented. Data from continued testing and real-world results of the Roadster allowed Kelty to demonstrate how their battery pack system was working. In particular, he was able to say that no Roadster had caught fire from thermal runaway.
In a major breakthrough for Kelty, Noguchi agreed to provide sample cells for testing. He and Yoshi Yamada, who ran Panasonic’s U.S. operations, visited Tesla’s headquarters that year. Their interest came as Panasonic was beginning to acquire control of Sanyo (they would take a majority stake in the company in December of 2009).
And just like that, the tide turned for Tesla and Panasonic. The manufacturer became excited to be part of a high-profile Silicon Valley startup and was on board to provide cells for the Model S. More than that, Kelty and Straubel wanted Panasonic to invest in Tesla, which was still hungry for funds. The Japanese company agreed to kick in $30 million.
For its successes, Straubel’s team still had some maturing to do. Tesla was no longer a little startup converting Lotus cars into Roadsters, leaning on the experience of their UK counterparts. Tesla was planning to build thousands of Model S sedans a year—on its own. The team got a taste of what that would be like when they became responsible for building the battery packs for the Roadsters. Then more so when they began building components that Tesla was going to sell to Daimler and Toyota, as part of the lifesaving deals Musk had signed a year earlier. Those deals, and the way they would influence the company’s culture, would play a larger role in the success of the company than even the money would.
When Akio Toyoda and Elon Musk celebrated an arrangement that included the two automakers working together on electric cars, the details remained far from complete. The idea was that the startup would do for Toyota’s popular RAV4 compact sport utility vehicle what it had begun doing for Daimler’s Smart cars: providing electric powertrains. But for the teams charged with implementing the agreement, things were less clear. To make the Smart car, Straubel had reconfigured the Roadster’s powertrain to fit into the tiny two-seater. Now, Musk wanted him and his small team to do it for a new, larger vehicle, while also continuing to work on battery packs for Daimler, and also developing a new powertrain for the Model S.
Straubel’s team assumed they would just be handing over powertrains to their business partners. The Toyota team, which was inexperienced in dealing with lithium-ion battery technology, thought it was getting help designing an electric RAV4 from the ground up. Some on the Tesla team wondered whether Toyota was secretly trying to steal its technological know-how.
There were cultural differences, too, as the first meeting between the Tesla and Toyota teams made clear. Greg Bernas of Toyota, who had been the chief engineer on another Toyota vehicle, arrived with a newly purchased book on the basics of electric cars. One of Tesla’s engineers, meanwhile, carried around a harmonica to play between meetings.
It took months of haggling before they were ready to get to work. Because the teams only had twenty months to complete the vehicle, Toyota agreed to use an old RAV4 platform, avoiding all of the testing and research that the larger automaker’s internal rules would require for putting out a car on a new platform. Toyota wasn’t accustomed to the way Tesla made on-the-fly changes—to its battery pack, for instance, and the computer software controlling it. As the teams managed cold-weather testing in Alaska, they struggled to diagnose why the prototype vehicle was vibrating on slick roads. Tesla engineers tapped on their laptops, adjusting the vehicle’s traction-control algorithms. They resolved the problem in a few hours of work rather than taking the data back to the lab for further review.
While that kind of speed impressed Toyota, the executives weren’t happy with the quality of the product they were receiving. A show car delivered for the Los Angeles auto show in 2011 left executives fuming. A Toyota manager was at a University of Michigan football game tailgate party when he got word from his crew in LA about the poor quality of the SUV delivered by Tesla the weekend ahead of the show. The SUV looked sloppy; that kind of inattention to detail wasn’t going to fly for Toyota, especially when they would be presenting it to the press and public. He dialed up a Tesla manager. “What the fuck is this?” he barked before demanding that the Tesla engineers meet him in LA the next day to fix the mess.
One big source of tension emerged over how Tesla validated its powertrains. To Toyota’s dismay, Tesla engineers told them that they took their suppliers’ word that parts were up to snuff, instead of conducting quality control tests to ensure durability for real-world use. That was a major no-no in the established world of car building.
For all the headaches, Tesla’s team was getting a useful primer on how to develop a powertrain that could not only deliver muscle, but that could last in the real world. Toyota’s handholding with the RAV4 was having unintended benefits—Straubel and his colleagues took what they learned and poured it straight into the Model S.
* * *
—
Tucked away in a complex of industrial buildings next to the SpaceX headquarters in Hawthorne, Elon Musk opened a new design studio for Tesla. He picked an old airplane hangar that had been converted into a basketball gym a few years earlier. The proximity meant he could easily sneak out of SpaceX to see what chief designer Franz von Holzhausen’s team was working on.
In 2011, even as the Model S remained in the distance, they began to consider the next vehicle in Tesla’s lineup. Musk had long touted his goal of a third-generation car aimed at the masses, but there were too many obstacles to bringing that to market yet. Selling just 20,000 Model S sedans a year wouldn’t fuel enough growth—in revenue or brand exposure.
So the Tesla team considered alternatives. They could use the Model S’s platform to make variation cars, such as a van or a sport utility vehicle. This would allow Tesla to get savings in parts and tooling, spreading the development costs over a greater number of vehicles sold. It was something that larger automakers had done for years—much in the way that the Roadster, eight years earlier, had been built on the platform of the Lotus Elise as a means to reduce Tesla’s costs (while defraying some of Lotus’s).
That memory should have been instructive. Musk had dictated so many changes to the Elise in his design of the Roadster that costs soared far beyond what had been expected. His demands for the vehicle that was to share the Model S’s platform threatened to repeat his mistake.
The next car, which Musk had taken to calling the Model X, needed to be a family vehicle with three rows of seats. Musk’s personal experience weighed heavily on the debate surrounding the new car. It wasn’t lost on the designers and engineers that Musk’s own five children were getting bigger; he had firsthand experience with sport utility vehicles. They knew this because he complained about it a lot.
He had a few clear ideas. For one, he wanted to make it easier to put kids in the second row. A minivan’s sliding door might open further than an SUV’s, but still it was a struggle to put a kid into a car seat without bumping your head on the roof, especially for Musk, at over six feet. For another, there was the third row. Musk recounted to the team how he used his SUV. He first placed his youngest kids in safety seats in the second row, so he could attend to them from the front. His older twins went in the third row. But he didn’t want Tesla’s vehicle to be like the Audi Q7, where you had to do gymnastics to get back there past the second seat.
The team imagined an SUV with the overall curves of the Model S, but with rear doors that opened upward like bird wings (a bit like the DeLorean in Back to the Future). These would allow for the largest opening. This way Musk could stand up straight while installing his kids. In theory, such doors could create a huge opening to enter the SUV’s back two rows. They created mock-ups of the design so Musk could physically see and touch the rear doors. At first, the opening wasn’t big enough, he declared. He wanted it to be as effortless as a magic carpet ride to get into the third row. The concept car they sketched out kept getting longer and longer, to accommodate a wider rear-door opening.
Development of a car can be influenced by many factors. At Tesla’s major competitors, designers and engineers might watch in agony as various levels of bureaucracy, from marketing to finance, weighed in on a project, not to mention senior leaders and their last-minute suggestions. At Tesla, it was becoming clear that the decider was Musk and Musk alone. He had flexed his muscles, in an early instance, when Martin Eberhard didn’t take him seriously about the quality of the dashboard. With the Model S, his personal preferences had likewise been written all over the car. Musk has a long torso, sitting higher in the seat than a typical driver. Consequently, he pushed the team to hang the sun visors in a way that, engineers worried, wouldn’t be useful for most drivers. He rarely carried anything but a phone, his assistant trailing him with whatever he might need. So he had no interest in cluttering a center console area with cubbies. Instead, the team laid a strip of carpet between the car’s two front seats, with little walls that formed a sort of gutter.
Even the placement of the external charge port was influenced by Musk, specifically the layout of his garage. Most American drivers park nose first; the design team figured that placing the charge port at the front of the car made intuitive sense. But Musk wanted it at the rear of the car: that’s where it best aligned with his home charger.
The cars were being built in Musk’s image. It remained to be seen whether car buyers would share that vision.
* * *
—
Even with the IPO the prior year and the partnerships with major automakers, it was painfully clear in 2011 that Tesla wouldn’t have enough money to get the Model S out, not without yet another infusion of money. That year, Tesla had grown to more than 1,400 employees, mostly in northern California, and the company was racing to fill a factory with expensive tools to build the Model S, all while sales of the Roadster came to an end that year after completing their contracted run. That meant the only money coming in was from the deals with Daimler and Toyota.
Rawlinson’s team was looking for ways to save pennies. They would run countless computer simulations to understand the car’s aerodynamics and how it would affect performance. Then they’d rent Chrysler’s wind tunnel, to the tune of $150,000 or $200,000, going in on a Saturday night and staying overnight until 6 a.m. on Monday to see how the car worked in the real world.
The team struggled with getting a sunroof that met Musk’s standards. Suppliers were either asking for more than Tesla could afford or else offering up compromised versions. Musk got so angry that he ordered his team to hire sunroof designers away from the suppliers and make the part in-house, figuring it would be cheaper to do it themselves.
Still Rawlinson progressed with the vehicle to the point where they were ready for the all-important crash tests. Each car cost $2 million to make, and he sent those gems barreling into objects, destroying them. They discovered that the welding of the car’s aluminum wasn’t holding as intended; the structure was falling to pieces on impact. It needed a redesign, something that added time and money—and increased pressure on Rawlinson, with Musk hovering over him at each test.
Rawlinson wasn’t the only one who felt it. A dark humor bonded Musk’s senior executive team. Musk, who split his time between Hawthorne and Palo Alto, would hold an executive committee meeting at the start of the day every Tuesday that would often stretch into the lunch hour. Depending on Musk’s mood, the running joke among the team was Musk’s lunch plans: Who would he be devouring this week?
Many around the table could see that, more and more often, Rawlinson was on the menu; he drew Musk’s increasing irritation. Those working for Rawlinson in Hawthorne couldn’t miss it either, overhearing him on calls with the CEO, then feeling the fire of his outbursts afterward. Musk’s anger with Rawlinson erupted at one point. Musk, with a rugby player frame, towered over Rawlinson during a disagreement. “I don’t believe you!” Musk screamed as he jammed a pointed finger toward Rawlinson’s chest. “I don’t believe you!”
The relationship was continuing to take a toll on Rawlinson, who also privately worried about his ailing mother back in the UK. Her health had worsened, and with no one else to help, he was trying to make arrangements for her from the other side of the world.
Musk had his own home troubles. His relationship with Talulah Riley was on the rocks. After getting married in 2010, they had been living apart for the past several months. In his rare free time, Musk could be found in the basement of his 20,000-square-foot French Nouveau mansion in Bel Air playing BioShock, a dystopian video game based upon the ideology of Ayn Rand. At Tesla’s Christmas party that year, Musk could be seen passed out on a pool table in a room whose entrance was blocked by his brother, Kimbal.
Rawlinson and Musk squabbled over a lot of issues, but none greater than one that dated back to the very beginning of the Model S, when Henrik Fisker, the designer turned Tesla rival, was still at the helm and had churned out what Tesla employees derided as the “White Whale.” The problem was tied to the placement of the battery pack at the vehicle’s floor, which added height to the vehicle. To address its bulbous appearance, Tesla designer Franz von Holzhausen had stretched the car out, allowing the battery cells to be better distributed beneath. This made for a lower roof line and proportions consistent with a sleek sedan, rather than when Fisker was trying to fit everything into a typical midsized car footprint. But Musk remained concerned about the roof line rising too high. He rode Rawlinson to make the battery pack as slim as possible. Rawlinson worried that going too thin would leave the pack vulnerable to road debris piercing the underside of the vehicle. They fought over millimeters. Eventually, Rawlinson relented. He told Musk he’d cut as much off the bottom as the CEO wanted. He lied.
Rawlinson’s team was hard at work on the Model X, too, with an idea for how to make the miraculous rear doors a reality. They had studied the “gull-wing” doors of the Mercedes, a similar feature to what Musk had demanded, but they concluded they needed something stronger. Their doors would be much larger; they’d need to be dual-hinged, not only rising outward but folding in at their halfway point, like a falcon hovering in flight. They settled on a hydraulic system, which would automatically lift the door rather than requiring a passenger to do so. To test it out, they welded a sample door onto a car frame, behind the SpaceX building, and expectantly they pushed the button.
