Tesla's wild ride (pg. 3)
Despite their differences, the two men initially worked well together. Both are technical guys who attack problems relentlessly. Eberhard's style is to question every engineering assumption. He'd corner people in the hall and debate them on the merits of air-cooled vs. liquid-cooled battery packs. "Technically he is just brilliant, and he has a tenacity that is unbelievable," says Laurie Yoler, a venture capitalist and an early Tesla board member and investor. "He is the guy you want around in those early days when you have naysayers all around." The team was solving huge technical problems, from battery cooling and load balancing to the power electronics. But if he didn't like an idea, Eberhard could also be very insulting. When an early member of the marketing team suggested putting solar panels on the roof of Tesla's new headquarters in San Carlos, Calif., Eberhard's response was, "Why the fuck would we do that?" (Eberhard says now that the company simply couldn't afford it.)
At the time, Musk's primary job was running SpaceX, but he and Eberhard talked practically every day. On some weekends both men would continue the conversation over dinner at one or the other's home with their families. Using what he was learning about rockets, Musk would constantly suggest materials that could shave a few pounds from the chassis or the body. Says Wright: "There were signs that he wanted to fiddle in the details, but it wasn't enough to make me run screaming. Musk is a technically savvy guy who wanted to help."
The highlight of board meetings for the tech-obsessed group were the show-and-tell portions, with little focus on the bottom line. "Martin would come in all excited and talk about one breakthrough or another," Yoler says. "Then we would go out to the shop and look at the latest electric motor or battery pack prototype." In his role as chairman, Musk would ask technical questions and then offer his own suggestions. As the car progressed, staffers began to realize that a green light from Eberhard was not sufficient. "The question always had to be asked," says Tarpenning, "'What will Elon think of that?'"
As time went on, Musk became more and more comfortable pulling rank. Jessica Switzer, who ran marketing at Tesla until the car's official launch in 2006, recalls persuading Eberhard to spend $30,000 on focus groups to test the car's logo, look, and feel. A few weeks later Musk killed the project without explanation. With Eberhard's approval, Switzer hired people from a PR firm in Detroit to drum up publicity in the automotive press before the car's launch. Musk promptly fired them. She later learned that Musk didn't want to spend money on marketing before the car was finished and figured his own involvement and the car itself would drum up more than enough PR.
When it came to design, Musk's vision - building the Next Great American Car Company - soon came into conflict with Eberhard's goal of getting a cool electric sports car to market quickly and relatively cheaply. The Lotus Elise chassis on which the Roadster was based had a high doorsill, a feature that makes entering the car tricky if you are not careful. Getting out is even harder. It took several attempts for Musk's wife to get out of an early Roadster prototype while wearing a dress. So Musk ordered the engineers to lower the doorsill two inches, thereby losing much of the cost savings that come from using a crash-tested off-the-rack chassis. "Have you tried getting out of an Elise?" asks Musk. "It's like you have to be a contortionist."
And rather than use the fiberglass body panels from the Elise that Eberhard had suggested, Musk insisted on carbon fiber, a lighter, stronger, and "cooler" material, in his opinion. He then went on to redesign the headlights and the door latches. After riding for a weekend in an early Roadster model and taking a beating in the standard Lotus seats, he insisted that custom seats be developed. Every change meant additional cost and time. "I always argued that we would sell exactly as many cars whether the door latches were push-button or electronic, whether the body panels were carbon fiber or fiberglass," Eberhard says. "All the nicer, cooler, faster stuff increased risk."
But Musk got his way, in large part because he was putting more and more of his own money into Tesla. He led Tesla's $12 million second round of financing in the fall of 2005, and also convinced some of his high-powered friends, including Google founders Sergey Brin and Larry Page and eBay employee No. 2, Jeff Skoll, to invest in later rounds. To date, he has personally put in $55 million of the $145 million Tesla has raised.
Musk, who is precise in his sentences, laughs easily, and if fired up will literally leap from his chair to punctuate a comment, admits he poked his nose into everything. "I was very insistent on things during the design phase, and it is true those things cost money," he says, "but you can't sell a $100,000 car that looks like crap." Unfortunately, while the exterior of the Tesla was designed and redesigned to meet Musk's exacting specifications, there was one very big problem: Two months before the car was set to debut in the summer of 2006, it still didn't have a production-ready transmission.
As Hollywood heavy hitters like Michael Eisner and Governor Arnold Schwarzenegger mingled with well-heeled car buffs at the Roadster's public unveiling at the Santa Monica Airport in July 2006, it looked like a slam-dunk for Tesla. But behind the scenes company execs were sweating. Electric motors have the advantage of being lightning fast from a standing start. But to get to the top speed that Tesla had promised (125 mph), they needed either a more powerful drive train or a second gear that could send the car speeding beyond 100 mph.
Problem was, Tesla's engineering team didn't yet have the experience to build a more powerful drive train, and no one had come up with a two-speed transmission that could go from 13,000 rpm to 7,000 rpm and survive for more than a few thousand miles before it wore out. Eberhard was inclined to stay on schedule, get cars on the road by sticking with one gear, and offer a Roadster that topped out at 110 mph.
Instead Musk launched the search for a supplier that could deliver a two-speed transmission. "Why did DeLorean fail?" Musk asks. "Because it was a shitty sports car. It may have looked cool, but it had the acceleration of a Honda Civic. That's what our car would have been with the motor we had and the power electronics we had connected to a single speed."
Meanwhile Eberhard was spending more and more time basking in the glow of the clean-tech crowd. He was the face of Tesla, the voice on its blog. He became a regular on the conference circuit and even starred in his own BlackBerry "innovators" ad. But at least four board members, including Musk, were growing concerned that Eberhard didn't have a firm grasp of the company's increasingly complex finances and supply chain. At an executive staff meeting at Tesla's San Carlos headquarters in June 2007, Eberhard grew visibly agitated, according to Straubel and others, as Tom Colson, head of manufacturing, went through a cost analysis of the Roadster put together by one of the company's VC backers.
In Tesla's own prospectus for its most recent round of funding, dated April 12, 2007, it had estimated the cost of building the car at $65,000, dropping as production ramped up. But just two months later, the VCs now believed the average cost was going to be well north of $100,000 for the first 50 cars and would decrease only slightly as more cars were built. "If this is true," Eberhard told Colson and the room, "you and I are both fired."
Eberhard doesn't dispute that things seemed to be heading south, but he says that for months he had been asking the board to hire a CFO and a COO and it wouldn't approve his choices. "I raised my hand and said, 'I am drowning, please help me,'" he recalls. Board member Yoler says that Eberhard could have hired anyone he liked but was holding out for Elon's approval, which never came.
In fact, there was another search going on at the same time. Tesla was moving from its development stage to an operational stage, where costs and schedules were taking precedence. According to board members (all venture capitalists at the time, except for Musk's brother), that wasn't a good fit for Eberhard. Even his old friend Tarpenning saw the problem. "You reach a point where the same people who are running the company when it has three people are not the same people who are running the company when it has 300," he says. "Both of us had been around the Valley long enough where we knew that to be the case." Eberhard himself agreed to join a board subcommittee to search for his own replacement. Dozens of candidates were interviewed and rejected; in the meantime Eberhard remained optimistic that Tesla would be able to hit its Aug. 27 production date.
But according to Darryl Siry, Tesla's head of sales, marketing and service, that wasn't going to happen. In June, he says, Lotus factory officials began warning that the late-August launch wasn't realistic. Eberhard persisted, saying in staff meetings that "we're dead if we miss that launch," Siry says. Yet Tesla hadn't even released all the car's specs to the parts suppliers. Two suppliers, Xtrac and then Magna, failed to get the two-speed transmission to work. Still, Tesla was ramping up spending as if it was going to start production in late August, putting $469,696, for example, into stereo and navigation gear that never made it into the cars and instead was sold back to the distributors at a loss. "Elon was pushing for early shipping all the time - it wasn't me," says Eberhard. "I resisted that spending, but Elon insisted."
Even though the Roadster project was wildly off course, between January and June 2007 there were monthly board meetings about designs for showrooms and a parallel project to find a site - they picked Albuquerque - to build a factory for Tesla's second model car, a $59,000 sedan code-named Whitestar. "Either senior management just wasn't paying attention, or they were hoping it would work itself out and they could fix it later," says one board member. "They were running line items on cost, irrespective of where milestones were on development of the car and the supply chain, as if they were not related."