September 2008 Archives

Tesla Motors Seattle Road Show

| No Comments
Tesla Motors kicked off their first visit to Seattle with an owners Town Hall meeting last night. It was an informal Q&A discussion. In attendance were about a dozen of the thirty-something Washington state Tesla owners and a half dozen Tesla representatives. We chatted about all things Tesla, from production status to possible Seattle store locations. Here's my summary.

Production Status

Signature #38 has entered production and they have delivered into the first few Signature 100 owners. The production rate is about ten per week. They expect to step that up to twenty per week by the end of this year. The first batch of the 10-per-week production will arrive in California next week and they will begin installing drivetrain 1.5 in those vehicles.

Ramping up production turns out to be very complex. It's not just about the assembly process in Hethel, but also about coordinating all of the parts suppliers. For some suppliers it's easy to increase delivery rate, but it's more challenging for others and everyone has to be on board to meet the production rate.

The production rate is impacted by a variety of things, some planned, some not. The Hethel plant shuts down for two weeks in August and for another two weeks around Christmas. There are other scheduled stops. If there's a problem in any stage of the assembly process, output stops until the problem is resolved. Depending on the nature of the problem, they might be able to process the backlog, increasing the output rate to maintain the average rate, or time may be lost.

The Production Timeline

Starting when a car enters production at Hethel, it takes about three weeks to finish the glider. There is approximately one week of painting, one week of assembly line, and one week of interior, testing, and finishing. Right now, it's actually taking about four weeks for the whole process, but three weeks is the goal.

Then there is shipping. It takes about six weeks to transport by ship from Hethel to Menlo Park. Air freight is much faster and much more expensive (and burns more carbon which may be an issue for many Tesla owners). Right now, all of the gliders are being airfreighted and when they switch to ships they may offer an airfreight option at additional cost. (Maybe somewhere in the $4,000 to $7,000 range, TBD.)

When the glider gets to Menlo Park, the motor, ESS, and PEM are installed and the whole car is tested, then prepped for delivery. This takes two more weeks.

So, right now it's about six weeks plus shipping time from start of production to delivery, with a goal of getting that down to five weeks plus shipping time. That's eleven weeks, nearly three months, when the cars are shipped by boat, assuming no problems with suppliers or the assembly process.

Seattle Tesla Store Location

Darryl Siry spent the day scouting store locations and found three promising locations in Seattle, two in the South Lake Union area and one on Capitol Hill. There was vocal support for finding a suitable Bellevue location, but their real estate agent didn't show Darryl anything good there.

The greater Seattle area is split in half by Lake Washington, creating a commuting nightmare between Seattle and "the Eastside." Seattle is the big city with high population density, a big downtown area and many businesses, including Amazon and Starbucks. Bellevue is a growing city with its own downtown area. Surrounding that are the suburbs and the area's largest employers: Microsoft (in Redmond) and Boeing (in Everett, Renton, and Kent). Google has a small presence on both sides of the lake.

Crossing the lake during morning and evening rush hour is terrible, as traffic across the two bridges clogs up and slows to a crawl. So traveling the few miles from Seattle to the Eastside can be a big pain and will be a barrier to making a casual trip to the Tesla Store if you're on the wrong side of the lake.

Tesla has a tough choice in finding a location that stands out, isn't crazy expensive and is convenient for both current owners and prospective Roadster and Model S owners. Eventually, they will have a presence on both sides of the lake, but we all want a visible, accessible location on our side. The conversation was pretty similar to the discussions on the owners forum for New York, Chicago and Miami.

The Mobile Connector

If the mobile connector is powered with a standard household outlet (120V/15A circuit, drawing 12A) the Roadster charges at a rate of about 8 miles of range per hour of charging, or more than 30 hours to charge a fully depleted pack. Connected to a 240V/50A circuit, drawing 30A, the rate jumps to about 40 miles of range per hour of charging, or about 7 hours to charge a fully depleted battery.

However, it turns out that there are regulatory roadblocks to Tesla Motors selling an EV power cable with a 240V/50A connector. Basically, it's illegal.

The existing EV owner community knows that the best way to charge their car on the road is to find an RV park and use one of their 240V/50A plugs, so obviously RVs have these types of connectors. When we had our RAV4-EV charger installed in our garage, we had the electrician install a 240V/50A outlet and put the corresponding plug on the charger. So, if we ever want to take it on the road, we just unhook the charger from the wall, put it in the back and map out stops at RV parks. No problem.

Unfortunately, there are laws that restrict the types of plugs that EV manufacturers can sell, but these laws don't apply to RVs. Tesla Motors considers it absolutely essential that owners have a way to charge from these 240V/50A sources and will find us a solution, but it's not here yet.

Realistic Range

The range for the Roadster with drivetrain 1.5 is 240 on the highway EPA cycle, 250 on the city cycle and 244 for the mixed city/highway range. The EPA highway range uses a 55 mph cycle with a 10% allowance for A/C.

There's been a bunch of discussion about the real-world range when driving the Roadster at actual freeway speeds. Martin Eberhard got a lot of owners nervous by noting that he drove it aggressively for 125 miles and had 30 miles of range left. The firmware in Martin's Roadster uses a very conservative estimate for remaining range, so the actual range would have been higher than 155 had he kept driving.

Tesla is continuing to improve the firmware and the range estimate. Newer firmware will show three range estimates based on (1) your current instantaneous power usage, (2) average power usage over the past 30 miles, and (3) expected power usage by the EPA highway cycle.

Obviously, it's important for Roadster drivers to have a good idea of how much charge remains in the car. Having an overly conservative estimate isn't much better than having an overly optimistic estimate, and the fact is the remaining range depends on how you are driving. By providing the three numbers, drivers get an idea of how far they can go if they keep driving the way they have been, or if they drop down to more conservative EPA-style driving.

In addition to the remaining range estimate, there are multiple levels of low battery warning. The first level puts a limit on torque, which will limit acceleration but not speed. In the last warning stage, a limit will be placed on top speed. When the battery is nearly depleted, there will be a warning of imminent power loss and you'll need to pull over and stop before the car just stops. At that point, you can turn the car off then back on again to enter "extended range" mode. Doing that will discharge the batteries low enough that battery life may be impacted, but there will be plenty of warning before things get to that stage.

Extrapolating from their personal experience, they estimate a range of about 200 miles driving at a sustained 75 mph, and a range of 150 miles driving very aggressively. They will be doing some more methodical long-range testing with the new drivetrain and will let us know what they find out. (They were planning to do some long range testing during the drive up to Seattle, but the car left later than expected and may not have had a chance to do so.)

On a personal note, Cathy and I have been driving our RAV4-EV as our primary daily vehicle for about two months now. At first we were nervous about its 100-mile range (80 miles without getting into the yellow warning level on the charge gauge.) After getting used to driving an EV, even that short range is way more than we need for daily driving. We don't even bother charging it every night unless we're below 50% charge or we expect to travel farther than our usual trips into Issaquah and Bellevue. We've only had one time when we didn't charge overnight and had a surprise trip the next day and couldn't take the EV.

The only reason we don't charge it every night is because the RAV4-EV charger isn't so smart and fills the battery pack all the way up, which causes excess heating, thus potentially reducing battery life. The Tesla Roadster has a much smarter charger that allows you to choose between a charge level that's the best for long-term battery life, or a full charge for maximum driving range. With the Roadster we'll charge it every night and have way more range than we'll ever need, and we'll only charge it all the way up on the rare occasion when we drive to Portland or eastern Washington.

Some people have long commutes, but for our fairly typical needs, an 80-mile range is plenty, and a 244-mile range will be total overkill. We really love not having to go to the gas station, we just charge overnight and have the full range in the morning.

Model S

The Model S will be a pure electric vehicle, not a range-extended serial hybrid. There was a brief period when they were planning to offer an RE-EV option, but that passed and they are back on the pure EV track.

They have the exterior design and Darryl says it looks great. It is technically a hatchback design, but in a good way, perhaps something along the lines of the Mercedes-Benz CLS-Class. (They in fact have a Model S test mule based on a CLS.) It will have a flat battery pack, as opposed to the Roadster's boxy ESS. Their production goal is 20,000 per year, worldwide.

When pressed about when we would get to see it, they only say it will be "soon." The plan is to show it first to existing Tesla customers, then the press, then the public by spring of 2009.

The Model S factory location isn't final yet, it still has to go through the environmental impact study and approval process. There are the usual wildlife issues that happen pretty much everywhere. Their goal is to have cars rolling out by late 2010, but that's an aggressive schedule that is vulnerable to various issues including delays resulting from the environmental impact study.

Current customers will get the first opportunity to order the Model S, with multiple slots open to each Roadster owner, including a way to give a slot to their closest friends.

Reporting iPhone Bugs

| No Comments
I've been an iPhone developer since early June. Lots of people are talking about what a pain it is to deal with Apple as an iPhone developer, from the stifling NDA to the slow and capricious App Store approval process. The thing that's bugging me the most right now is the bug reporting and resolution process. Here's how it goes.

Step 1: Write some code. See that it fails.

Step 2: Spend hours trying to figure out what you did wrong. Look at the sparse documents and sample code. You can't ask questions of fellow developers because of the NDA, so it's all up to you. Eventually, you decide that the bug isn't your fault.

Step 3: Report the bug to Apple. Explain briefly that when you do X, Y happens, and Y is not good. Explain in some amount of detail so they know what you're talking about, but don't spend your entire day working on the report because surely someone else has already reported this bug.

Step 4: Wait. Budget at least a week. Meanwhile, you want your app to do this thing that doesn't work, so either put that feature on hold or try to find some other way to do something that accomplishes approximately the same thing.

Step 5: Get a piece of email from Apple saying that they need sample code that demonstrates the problem.

Step 6: You don't want to send them your entire app, because it's much more complex than what is needed to reproduce the problem and you've spent weeks or months on it and don't really like the idea of packing it up and shipping the whole thing to Apple. So, now you get to spend some ridiculous amount of time writing a minimally complex app that demostrates the problem. This takes anywhere from an hour to an entire day, possibly even more.

Step 7: Add the sample code to your bug report.

Step 8: Wait. Again with the waiting. By now, you've probably found some other way to do whatever isn't working, or yanked that feature, or put the whole app on hold to work on something else.

Step 9: A week or two later, you get another note from Apple saying that engineering looked at your bug report and sample code and decided they already have the bug in the database. So, now your bug is closed and you have no direct way to track the status of the other bug. Maybe the other bug was actually submitted after you reported the bug, but you're left out in the cold anyway.

Step 10: Don't bother waiting. Apple's not going to contact you again. No one will ever look at your bug report or tell you what happened with the issue.

Step 11: After you've gone through this process enough times that you have a list of these things, try sending email to Apple asking how those issues are going.

Step 12: Wait. I have no idea how long this step takes, or if I'll ever hear back. It's been two weeks since I sent email asking for an update.

Now I have more bugs I'd like to report, but frankly my enthusiasm is waning. Why should I spend hours or days of my time to report issues to Apple, just to have them trick me into writing sample code, dupe my bug, then never send me an update?

Apple should want these bug reports. The bugs are there, they are slowing down development of iPhone apps, and it's probably not just me. But this whole process is so broken, I'm probably not the only iPhone developer that is not reporting the bugs that are getting in my way.

About this Archive

This page is an archive of entries from September 2008 listed from newest to oldest.

July 2008 is the previous archive.

October 2008 is the next archive.

Find recent content on the main index or look in the archives to find all content.