Euro 6 regs in full effect; ACEA launches clean diesel campaign in support of technology-neutral policies
Toyota to focus on hybrids, Safety Sense at IAA

Tesla Roadster 3.0 battery upgrade: $29,000 for ~78 kWh, 331 miles

Tesla Motors has made the promised battery upgrade for its Roadster available via its online shop. For a price of $29,000, with a $5,000 downpayment, customers will receive an new battery pack that Tesla says stores roughly 40% more energy than the original battery, with an increase in total range of more than 35%. Since the original Roadster featured a 56 kWh pack and 245 miles of range, presumably the new pack is ~78 kWh with a resulting range of ~331 miles (533 km).

There is a “slight” increase in the battery weight, Tesla noted, but Tesla has managed to deliver the great increase in capacity within the same footprint.

The cost per kWh (~$372) is higher than a Model S battery due to the upgrades being almost entirely hand-built, and low-volume (only 2 or 3 per week). It also includes additional work to remove, upgrade, and reinstall the power electronics module (PEM), Tesla said.

Per the original announcement (earlier post), the Roadster 3.0 upgrade comprises three main elements: the higher capacity battery pack, enabled by higher capacity cells; a retrofit aero kit to improve aerodynamics; and a further reduction in rolling resistance. News and pricing of the remaining two elements (aero kit and RR) has not surfaced yet.

Tesla said then that it expected that the combination of the three would deliver a 40-50% improvement on range between the original Roadster and Roadster 3.0—i.e., between 343 and 368 miles. Tesla also said that there is a set of speeds and driving conditions under which Roadster 3.0 can drive more than 400 miles (645 km).



70kwh and around $414 kWh according to this:

Account Deleted

So once again Tesla has proven that they are still the only car company in the world that makes cars that get better over time >after< you buy them. Tesla's software gets better through regular OTA upgrades of all systems and the battery packs can be replaced with even larger future battery packs. Musk has said he expects to offer about 5kwh more on the Model S battery pack for every 1 to 2 years.

Every other car maker sell you a car that cannot be upgraded physically and has no software updates either. I almost get angry at thinking about how slow the old auto industry is at innovating and how much they think they innovate when in fact they don't when compared to Tesla. With the old auto industry it is mostly about marketing talking things up when in fact there is very little substance in the few innovations they do make such as improving an obsolete technology like the combustion engine or introducing a complex, costly and certain to fail fuel cell vehicle.

Eventually Tesla's success will turn the industry around so OTA upgrades will become standard and every carmaker will have BEVs for sale. Then again the industry is about to change completely going from selling cars to selling miles using a fleet of autonomous people and cargo vehicles likely operated, maintained, cleaned, fueled and owned by the car makers.


If they are upgrading from the old 2.9A 18650 Panasonic cell to the newer 4.0A version, they will add about 52Kg to the mass of the system while keeping the same number of batteries.

@Henrik, the cars get better over time provided you pay for it.
The s/w updates are presumably to fix bugs as well as improve the system. It is alright for Tesla to do this as they have a fairly small user base, but the big boys like Ford and VW have to be more cautious as they have so many cars on the roads.
+ they just aren't used to s/w based cars yet.


(1) For about the same expense you can upgrade your 2009 era Vette much fatser and more fuel efficient(same vintage as the Roadster) with an LS-7 crate engine and some transmission mods. "Pay $30,000 and make your old car better" is not disruptive, and the only difference is that is a factory-sold mod rather than a performance shop upgrade.

(Wow: why don't factories do this? Because it is not a profitable enough enterprise, and it won't be for Tesla, either.)

(2)"they just aren't used to s/w based cars yet"
Because a Model S has a large flat screen display, it appears to be "software based" when other cars are not. To accept that, you must ignore the following:
The most complex glass cockpit, fly-by-wire and interconnected fighter aircraft in the world, the F-22, operates with about 5M lines.
The Boeing 787, whose advanced electrical system has over 100 microprocessors just for the Secondary Power Distribution, operates with about 15M lines.
Windows 8 needs about 50M.

A competitive premium auto (think A8, LS600, Panamera, etc.) has over 100 microcontrollers and requires about 100M lines of code.

Why is it that the aviation industry was more than a decade behind automotive in creating inexpensive but robust communication networks? Why is it that automotive needs drove the creation of Time Triggered Protocol that saw its first implementation on the 787 YEARS after being common in automobiles? Why are some of the most demanding microprocessor needs based in automotive applications?

This OTA business seems to have the entire semi-technical world completely spoofed. NOTHING keeps the rest of the industry from doing OTA software maintenance except the sheer number, geographic distribution, and age of cars, as mahonj points out, and liability fears. Tesla has quite literally orders of magnitude less to lose than the "big boys", and they are willing to take the risk, just like they do not bother to lock out browser use when rolling. There is no way for me to prove this, but I would wager an enormous sum of money that best of breed PHEVs like the '16 90-series Volvo SUV with significant automated driving features have more computing power and more code than the Model S.

Account Deleted

Come on. Microsoft updates and patches 1,500 million computers on a daily to monthly basis. They span 20 years old computers to brand new ones across all countries in over 500,000 different hardware configurations. Google does it with over 1 billion phones and Apple with over 500 million phones and computers. I got Win 10 a month ago and have since monitored how often it updates. It does virus and malware updates many times per day, win store apps once a day and system updates and bug fixes about once a week. That system will be rolled out to about 1000 million devises in 11 more months. Whatever the auto industry does it can be done. The backend data centers are not a problem. Computers are evolving fast. My first Macintosh had a 20Mbyte hard drive and now I have a 20 Terabyte (external, but still more compact than my Macintosh) drive. It is 1,000,000 times larger in about 30 years! This development continues so whatever you think is big now will be absolutely nothing in 20 years.

I agree that luxury cars like Tesla and Volvo X90 has much more software and systems to update than a Smartphone or a PC and it shows when Tesla needs several hours for a full system update. Tesla also now has nearly 100,000 Model S that get full system OTA updates and various cloud services but it is tiny by comparison to Apple, Microsoft, Netflix, Google, Facebook, etc. When the autonomous technology become standard on Tesla's cars by 2020 to 2025 the computing power and software in their cars will need to get much larger. Probably 10 to 100 times more processing power and 10 times more code and ten to 100 times faster internet connections.

My point is that the auto industry is entering an era where software probably matters more than physical designs and also require more engineering hours. OTA updates will become a necessity as you cannot physically recall every car whenever you find a serious software bug. That will happen all the time looking ahead. You need to fix it ASAP without much costs and you do that by OTA updates.

Nick Lyons

Hmmm. For $29K I could get a high-end Subaru Forester or VW Golf TDI Sportwagen (my particular daydreams...). Obviously some people have money to burn, and I'm sure this upgrade is a much better value than dropping $30K in a drunken weekend in Vegas, for instance.

The bigger question in my mind is: why is Tesla devoting resources for this? Their core mission and undivided focus should be getting the Model X and (more importantly) Model III to market timely. Same question for 'ludicrous mode'--is it really worth diverting engineering resources to these projects when they have bigger fish to fry?

Based on my experience managing software projects in high-tech companies, my advice to Tesla: Don't get distracted by cool but non-essential products and features--focus, focus, focus on the core goal and mission.


Could it be that their customers requested or showed interest.

Could it be an excercise. A small'trial type production run spun from future model pack development. Incurring minimal R&D and assuming it is model specific, could be a job for trainees and process designers in house?

That would make sense as there is a limited liability exposure and a few / hundred test drivers happy to participate.

Free data collection? Why not.


@Davemart, and others commenting on the price and comparison to other purchases: It's a custom, hand built upgrade for silly rich people. The hand built (with something like 10,000+ connected wires probably has a LOT to do with the price of that battery pack. :)

"The price of the battery upgrade is $29,000, including all labor and logistics, which is equal to Tesla's expected cost. It is not our intention to make a profit on the battery pack. The reason the cost per kWh is higher than a Model S battery is due to the almost entirely hand-built, low-volume (only 2 or 3 per week) nature of Roadster battery packs. It also includes additional work to remove, upgrade, and reinstall the power electronics module (PEM.)"

Direct from Tesla's website:

Nick Lyons

@Arnold: Could it be that their customers requested or showed interest.

Sure. But the project gains zero new customers and zero profit--they are doing it at cost. And there is always opportunity cost. If this project advances strategic goals somehow, more power to them, but it smacks of a vanity project to me.

Tesla needs market share--more customers, and more not-so-rich customers is what they need if they want to meet their stated goal of being a mainstream auto manufacturer. Catering to the whims of a small subset of Roadster owners is not strategic, IMHO.

Account Deleted

It is mostly about showing some respect for the customers you have and by doing that also sending a message to would be Tesla customers that if they buy a Tesla they will not be left behind without being able to maintain their cars. I admit there may also be an element of vanity at play. The Roadster is the first ever long-range performance BEV made. It crushed stereotype expectations that electric cars where slow and ugly. The car is historic and because of that it will still be a collector's item in year 2115. The Model S will probably not.

About focus. I think Tesla is quite focused. That battery pack did not take a lot of time to make. Tesla will not enter new cooperation with other automakers and let existing deals expire like the one with Benz and Toyota. I am still worried those falcon doors on the Model X could spell more trouble. They did take a lot of time to develop and they were not strictly necessary at this point in Tesla's history. Could have been added later. On the other hand they are unique and may become a massive marketing success.

BEVs that are great and inexpensive will not come until we get that self-driving technology that enable Tesla's and others to operate a global fleet of autonomous taxis and cargo trucks. Model 3 being a small car costing at least 35,000 USD is still not for the masses. You can get great gasser SUV for that money. So we need self-driving cars to cut the capital cost of BEVs by letting them drive 100,000 miles per year instead of 15,000.

Nick Lyons

Henrik: I am still worried those falcon doors on the Model X could spell more trouble. They did take a lot of time to develop and they were not strictly necessary at this point in Tesla's history.

Bingo. Hopefully, the falcon doors won't turn into albatross doors for Tesla. And on a personal note, even if I could afford the Model X, those doors would be a show stopper for me. I keep cars for at least 10 years, and as I become a frugal old codger, I imagine how well those doors might be working down the road.

The comments to this entry are closed.