Nissan appears to have done it. They're even making a profit off of their Leafs. I think it just points to Fiat's incompetence and being late to the game. Then again, Nissan designed the Leaf from the ground up as an electric car, just like how Tesla designed their Model S.
In the 1970's my father converted his 1966 Pontiac Tempest LeMans to run on propane. The engine was originally designed to run on the "white gas" that was leaded 110 octane with 11:1 compression. During the gasoline crisis he did the conversion, also in part because the car ran like crap once the high octane gasoline was no longer available. The conversion involved putting in bronze valve guides, the tank, a converter that uses the engine coolant to heat the propane and the propane carbeurator which sits inline with the gasoline carberuator. The engine runs great on propane though there's about 20% less milage and a bit less power running propane compared to gasoline due to the 20% lower energy density of the fuel. As a test, running Mobile One oil he went 80,000 miles without changing the oil (only the filter) and the oil was clear when he drained it and the engine was still at the tight end of the factory spec. Propane is a lot easier on the engine than gasoline.
He also switched the car to use an electric fuel pump when it was switched to gasoline (which was extremely rare).
One nice thing was that while camping we could fill the propane tanks for the stove from the car which has a 26 gallon tank. I think propane is also safer than natural gas since it does not require a very high pressure.
He did not have to compromise running on propane other than the fact that there was a bit less power due to the fact that propane burns at a lower temperature than gasoline. The car drove better at high altitude since the propane carberuator automatically compensated for the lower air density.
For many years propane was much cheaper than gasoline as well, so the conversion paid for itself many times over. He still has the car though he rarely drives it any more.
I still use my HP Laserjet 4M/Plus I bought years ago. The thing was built like a tank. Once in a while I need to do a rebuild which is cheap but the thing just works. New HP stuff is crap. I remember being unable to print PDFs with the newer HP printers because they would crash so I had to print them at home on my ancient printer.
Find me an American who's qualified for the job and they'll be hired on the spot. My employer has had a very hard time finding qualified people. The key word here is qualified. I don't care what race, sex, sexual orientation or anything else, as long as you wear pants and can code I'm happy. The problem is that there are far too few Americans graduating with Computer Engineering degrees. I need someone who is good at writing software and understands hardware. When I code I have several documents open at the same time, usually datasheets and schematics.
A standard CS degree doesn't have the hardware background and I'm sorry to say that a lot of code I've seen from EEs is utter crap. Finding someone good at both seems to be damned near impossible.
It often takes months to find someone qualified for a position. Too often the person I interview doesn't even have decent C programming skills, let alone understand things like CPU cache, virtual memory, multi-core programming (up to 48 * 2 right now) or how to interact with complex hardware devices.
I don't want someone who just adds a quick fix or hack. I want someone who does it right, even going so far as to reengineer code if it will make it better (i.e. easier to maintain, more portable, etc.) The code I get from vendors who design boards for us almost always is crap. The first thing I do is throw it out and start over. I try and maintain strict separation between board specific code, our code and the standard U-Boot bootloader code. I've ripped out almost all board-specific code from the common code and put it where it belongs along with minimizing the amount of board-specific code that is required.
I can confirm this. I have interviewed a lot of people for trying to fill positions and finding a good qualified person is damned near impossible, regardless of race. The best candidate I interviewed was a trans-gendered Russian who we made an offer to (I guess she decided to stay where she was or had a better offer). I've interviewed some white people who clearly are not qualified as well as some Indians who are very well qualified.
If you know how to work on U-Boot or embedded programming (multi-core RISC experience a major plus) I'm desparately looking for someone to help out since I'm totally overwhelmed. I'm working with 48-core NUMA chips, 10G/40G networking, PHYs, Linux kernel stuff, I2C devices and a lot more. I work on everything from XHCI to AHCI, NAND, EMMC/SD, SPI and just about everything else imaginable. Want to play with running an application or Linux on 96 cores? Generally not much assembly language, though I consider MIPS assembly to be quite elegant compared to the horrendous mess that X86 has turned into. I'm always getting new boards to support, new PHYs, new I2C devices, SATA port multipliers and just about anything imaginable. Virtual memory and low-level CPU experience a plus (we run U-Boot in virtual memory to simplify things). We're also working on ARM64.
I too work with a lot of Indians. It all depends on the person. There are a number of people I work with who are extremely talented and others who are not who usually don't last all that long. It depends on the company. For example, if I see a resume for someone coming from certain well-known Silicon Valley companies my expectations immediately drop significantly.
One problem they have in Indian universities is that they typically do not have the hands-on experience one can get at a good US university. While I was in labs with oscilloscopes, programming FPGAs and wire-wrapping CPUs such things are not available in universities in much of the world (they stopped wirewrapping soon after I finished the class).
Now if I could only find a highly qualified person to help me work on U-Boot on multi-core MIPS64 chips (48 cores * 2 NUMA)) and new boards I'd be happy, regardless of if they're Indian or anything else.
At my job we have openings that we find extremely difficult to fill with local talent just because the demand is so great and the number of people with the skills is small. Sadly there are far fewer Americans graduating with computer engineering degrees than there are job openings to fill. They'd rather get degrees as English majors or MBAs or other easier degrees then complain when they can't get work after graduating.
Tesla isn't doing it because they decided to go after a different market with a car with a reasonable range. They also have economies of scale. Their batteries should also scale nicely downward since most of the cost is in the 18650 cells.
Part of Fiat's cost is that they are trying to shoehorn an EV drivetrain into a gasoline powered car. There are a lot of changes possible with an EV as Tesla has shown that can reduce the cost of manufacturing. Designing an EV from the ground up reduces cost. After all, Nissan's Leaf is profitable, though Nissan really needs to address a few major problems (like their lack of active cooling for the batteries).
After having bought a Tesla Model S I find it difficult to go back to an internal combustion engine. The responsiveness and smoothness are unparalleled with throw you back in the seat instant acceleration.
I see no problem making EVs for well under 50K and making a profit on them.
If Tesla can sell their cars with an 85KWh battery that gets 265 miles for $73,570 and average over 25% margins then there is no reason why a car that is quite a bit smaller with significantly less range shouldn't be able to be made for under $40K and be profitable. It all comes down to how much is invested in the technology. Tesla spent years perfecting their technology. They spent a lot of R&D optimizing their batteries, both for performance, reliability, capacity and cost. While everyone else is screwing around with expensive LiFePO prismatic batteries, Tesla was able to work with 18650 cells using a different chemistry. They did a lot of things to cut the cost of the 18650 cells and focused on how to get safety and long life out of the cells while getting much higher energy density. Now they're focusing on their gigafactory to further reduce the battery cost.
Tesla also went with an induction motor which is cheaper to manufacture than the standard synchronous motor everyone else uses since there are no rare-earth magnets in it nor strong magnetic fields to deal with during manufacturing.
A Fiat 500e needs only a fraction of Tesla's battery capacity and a smaller inverter and electric motor so it should be cheaper. Maybe they should do what Toyota did and either license Tesla's technology or have Tesla build the drive train and battery like they did for the Toyota electric Rav 4.
I see tons of EV cars in California. Leafs are everywhere.
If Tesla can make a full-sized sedan with a 265 mile range (85KWh battery) for $73,570 while averaging a 25% profit margin there's no reason why Fiat shouldn't be able to make a profit selling a much smaller car with a much smaller battery and a much smaller range. Perhaps they should invest in Tesla's gigafactory to bring down cost and/or license Tesla's batteries like Toyota did. If they're like Nissan then they only need around 1/4 the capacity of Tesla's 85KWh battery pack, a smaller electric motor and a smaller inverter. If it cost Fiat $46,650 per-car then they're doing something wrong. They're probably using those more expensive, lower energy dense LiFePo prismatic batteries that everyone except Tesla is using.
It's not that it's 0.43mm per year, but one needs to look at the rate of change. It may be 0.43mm/year now, but in 10 years it may be 4.3mm/year or higher. The problem is that all of these things we are seeing are increasing in a non-linear mode.
I alreay use a product called Stick'n'Find. It's like Tile except it has a user-replacable battery and is available now. It also supports Android. It also provides a SDK and will sense temperature.
They are not standard lithium-ion batteries. For example, the cell design is different in order to cut cost. They do not have the normal dimple on top nor do they contain the protection circuit. The protection is provided by a low melting point wire that is bonded to the battery that doubles as a fuse. They are also not exclusive to Panasonic since they also work with Sanyo.
The chemistry is also a bit different. The batteries are automotive grade. They are designed to handle more temperature extremes and for rapid charging as well as for longer life than a laptop cell. As for Tesla owning the chemistry I heard that during one of the quarterly earnings conference calls. They did a lot of research and testing for their batteries. Arguably most of their technology lies in their battery management and pack designs but they also have a number of key patents in the actual cell design.
I had a good talk with a friend of mine who works there in R&D. He was describing it all a compromise between three things: reliability, performance and capacity. Choose two of the three.
They have some interesting patents as well, such as a metal air battery design.
Actually they don't make much on carbon credits any more and haven't for a while. Their margins are already over 25% on their cars so they are making money on their cars.
Tesla has stated that they don't make all that much any more from ZEV credits. They're doing fairly well. They had a bit of a loss this last quarter due to their rapid expansion (which is what they should be doing). The GAAP accounting penalizes them quite a bit for the way it deals with leases which is totally unrealistic. It prevents them from recording any income from the leases for quite some time which is why the big investors look at the non-GAAP numbers.
It all depends on the battery. Tesla claims that rapid charging has no impact on the life of their battery and they are rated for at least 3000 full charge/discharge cycles, which for an 85KWh battery pack is over 600,000 miles. Also, typically on the Tesla owners don't need to charge to 100% or drain it close to 0% since it has a decent range.
The Leaf, by comparison, is another story. The Leaf lacks active battery cooling and in hot climates the batteries have been dying. In Arizona people are reporting 40% range loss in two years and rapid charging does impact the life of the batteries.
Here's one where a Tesla was rear-ended by a semi. The Tesla was able to drive away, despite being knocked forward around 100'. The semi wasn't going anywhere.
They use a custom 18650. The cells are not normal 18650 cells. They use that format for cost reasons. Elon has said that ideally they would prefer a slightly larger format.
The reason they use so many small cells has to do with safety. The battery packs are designed so that a few cells dying won't have a noticeable impact on the pack. It also adds to the safety.
As it is, their energy density is much higher than their competitors who use LiFo prismatic cells.
Actually the AC system should last a lot longer too. The compressor is electric and fully sealed and since there's nothing mounted to the engine with flexible hoses the chance of it leaking is reduced significantly. There's no compressor clutch either. You also get rid of the power steering pump since that is now electric. The factory told me the electric motor in my Model S is lubricated for 12 years and it's easy to remove by just undoing some bolts.
While I haven't seen the compressor on my MS the one on my Prius was totally sealed, though for some inexplicable reason they still mounted the electric AC compressor to the engine block and used flexible hoses for the AC.
The dealers don't like EVs because it gets rid of one of their biggest sources of revenue, service. EVs don't need oil changes or belt changes. The complexity of the drive train is significantly reduced. Even the brakes will last a lot longer due to regenerative braking. About the only thing dealers will have to still do is rotate the tires and do the basic inspections. There's no spark plugs, air filters, fuel filters, fuel pump, belts or the myriad of other mechanical parts that wear down over time. There's no transmission even. There's no fuel injectors, exhaust system or anything else.
There are only about a dozen moving parts in the drive train on my Tesla model S. The electric motor drives a 9.71:1 gear reduction and a differential that goes to the wheels. That's it. At the factory I was told that the electric motor is lubricated for 12 years.
The AC and power steering are electric so they will wear much better than the belt-driven pumps and compressors that have clutches.
The parts that can break are also much easier to get to without an engine in the way. If the motor does need to be worked on it's relatively simple to pull out the entire drive train. The National Geographic special on the Tesla factory shows them installing it in under 5 minutes. See https://www.youtube.com/watch?... The power steering, AC compressor and coolant pumps and whatnot can be easily reached by just removing a panel and/or pulling out the plastic frunk module.
Actually Tesla does own the IP to their batteries. Panasonic manufactures them and provided some IP, but Tesla has their own unique chemistry and cell design. Tesla's cells are automotive grade yet are much higher energy density than the LiFe cells other manufacturers use. They also developed their own battery management support. A quick search turns up a lot of battery related patents that Tesla owns. They do a lot of active battery research as well. You cannot buy the Tesla cells from Panasonic even though Panasonic manufactures them.
It was the 3.11 kernel (OpenSuSE 13.1 with all the up-to-date patches). Normally I run XFS which for me has always been rock solid and extremely stable and reliable (even in cases where I had major issues occur with RAID). I have NEVER had a single issue with XFS and I have been running it for years. The tools for it are top-notch which is the main reason I run it. I also usually back up my systems. Since it was the only drive in the system and only 250GB I did not separate/home from root. Fortunately after fsck I was able to recover most of the data but the root directory was totally trashed and all the recovered data was under lost+found.
I tried btrfs a year ago but gave up because the performance made it unusable. My mail server (Cyrus) stores each email as a separate file. I gave up after waiting an hour to untar my email onto a SSD, something that takes only seconds with XFS or EXT4.
In this case it was a server provided by IT. I was working out of state living out of a hotel room. Since I was in a huge rush with the project there was no time to set it up to back up, nor did I have anything to back up on to.
Slashdot Mirror
Nissan appears to have done it. They're even making a profit off of their Leafs. I think it just points to Fiat's incompetence and being late to the game. Then again, Nissan designed the Leaf from the ground up as an electric car, just like how Tesla designed their Model S.
In the 1970's my father converted his 1966 Pontiac Tempest LeMans to run on propane. The engine was originally designed to run on the "white gas" that was leaded 110 octane with 11:1 compression. During the gasoline crisis he did the conversion, also in part because the car ran like crap once the high octane gasoline was no longer available. The conversion involved putting in bronze valve guides, the tank, a converter that uses the engine coolant to heat the propane and the propane carbeurator which sits inline with the gasoline carberuator. The engine runs great on propane though there's about 20% less milage and a bit less power running propane compared to gasoline due to the 20% lower energy density of the fuel. As a test, running Mobile One oil he went 80,000 miles without changing the oil (only the filter) and the oil was clear when he drained it and the engine was still at the tight end of the factory spec. Propane is a lot easier on the engine than gasoline.
He also switched the car to use an electric fuel pump when it was switched to gasoline (which was extremely rare).
One nice thing was that while camping we could fill the propane tanks for the stove from the car which has a 26 gallon tank. I think propane is also safer than natural gas since it does not require a very high pressure.
He did not have to compromise running on propane other than the fact that there was a bit less power due to the fact that propane burns at a lower temperature than gasoline. The car drove better at high altitude since the propane carberuator automatically compensated for the lower air density.
For many years propane was much cheaper than gasoline as well, so the conversion paid for itself many times over. He still has the car though he rarely drives it any more.
I still use my HP Laserjet 4M/Plus I bought years ago. The thing was built like a tank. Once in a while I need to do a rebuild which is cheap but the thing just works. New HP stuff is crap. I remember being unable to print PDFs with the newer HP printers because they would crash so I had to print them at home on my ancient printer.
Find me an American who's qualified for the job and they'll be hired on the spot. My employer has had a very hard time finding qualified people. The key word here is qualified. I don't care what race, sex, sexual orientation or anything else, as long as you wear pants and can code I'm happy. The problem is that there are far too few Americans graduating with Computer Engineering degrees. I need someone who is good at writing software and understands hardware. When I code I have several documents open at the same time, usually datasheets and schematics.
A standard CS degree doesn't have the hardware background and I'm sorry to say that a lot of code I've seen from EEs is utter crap. Finding someone good at both seems to be damned near impossible.
It often takes months to find someone qualified for a position. Too often the person I interview doesn't even have decent C programming skills, let alone understand things like CPU cache, virtual memory, multi-core programming (up to 48 * 2 right now) or how to interact with complex hardware devices.
I don't want someone who just adds a quick fix or hack. I want someone who does it right, even going so far as to reengineer code if it will make it better (i.e. easier to maintain, more portable, etc.) The code I get from vendors who design boards for us almost always is crap. The first thing I do is throw it out and start over. I try and maintain strict separation between board specific code, our code and the standard U-Boot bootloader code. I've ripped out almost all board-specific code from the common code and put it where it belongs along with minimizing the amount of board-specific code that is required.
I can confirm this. I have interviewed a lot of people for trying to fill positions and finding a good qualified person is damned near impossible, regardless of race. The best candidate I interviewed was a trans-gendered Russian who we made an offer to (I guess she decided to stay where she was or had a better offer). I've interviewed some white people who clearly are not qualified as well as some Indians who are very well qualified.
If you know how to work on U-Boot or embedded programming (multi-core RISC experience a major plus) I'm desparately looking for someone to help out since I'm totally overwhelmed. I'm working with 48-core NUMA chips, 10G/40G networking, PHYs, Linux kernel stuff, I2C devices and a lot more. I work on everything from XHCI to AHCI, NAND, EMMC/SD, SPI and just about everything else imaginable. Want to play with running an application or Linux on 96 cores? Generally not much assembly language, though I consider MIPS assembly to be quite elegant compared to the horrendous mess that X86 has turned into. I'm always getting new boards to support, new PHYs, new I2C devices, SATA port multipliers and just about anything imaginable. Virtual memory and low-level CPU experience a plus (we run U-Boot in virtual memory to simplify things). We're also working on ARM64.
I too work with a lot of Indians. It all depends on the person. There are a number of people I work with who are extremely talented and others who are not who usually don't last all that long. It depends on the company. For example, if I see a resume for someone coming from certain well-known Silicon Valley companies my expectations immediately drop significantly.
One problem they have in Indian universities is that they typically do not have the hands-on experience one can get at a good US university. While I was in labs with oscilloscopes, programming FPGAs and wire-wrapping CPUs such things are not available in universities in much of the world (they stopped wirewrapping soon after I finished the class).
Now if I could only find a highly qualified person to help me work on U-Boot on multi-core MIPS64 chips (48 cores * 2 NUMA)) and new boards I'd be happy, regardless of if they're Indian or anything else.
At my job we have openings that we find extremely difficult to fill with local talent just because the demand is so great and the number of people with the skills is small. Sadly there are far fewer Americans graduating with computer engineering degrees than there are job openings to fill. They'd rather get degrees as English majors or MBAs or other easier degrees then complain when they can't get work after graduating.
Tesla isn't doing it because they decided to go after a different market with a car with a reasonable range. They also have economies of scale. Their batteries should also scale nicely downward since most of the cost is in the 18650 cells.
Part of Fiat's cost is that they are trying to shoehorn an EV drivetrain into a gasoline powered car. There are a lot of changes possible with an EV as Tesla has shown that can reduce the cost of manufacturing. Designing an EV from the ground up reduces cost. After all, Nissan's Leaf is profitable, though Nissan really needs to address a few major problems (like their lack of active cooling for the batteries).
After having bought a Tesla Model S I find it difficult to go back to an internal combustion engine. The responsiveness and smoothness are unparalleled with throw you back in the seat instant acceleration.
I see no problem making EVs for well under 50K and making a profit on them.
If Tesla can sell their cars with an 85KWh battery that gets 265 miles for $73,570 and average over 25% margins then there is no reason why a car that is quite a bit smaller with significantly less range shouldn't be able to be made for under $40K and be profitable. It all comes down to how much is invested in the technology. Tesla spent years perfecting their technology. They spent a lot of R&D optimizing their batteries, both for performance, reliability, capacity and cost. While everyone else is screwing around with expensive LiFePO prismatic batteries, Tesla was able to work with 18650 cells using a different chemistry. They did a lot of things to cut the cost of the 18650 cells and focused on how to get safety and long life out of the cells while getting much higher energy density. Now they're focusing on their gigafactory to further reduce the battery cost.
Tesla also went with an induction motor which is cheaper to manufacture than the standard synchronous motor everyone else uses since there are no rare-earth magnets in it nor strong magnetic fields to deal with during manufacturing.
A Fiat 500e needs only a fraction of Tesla's battery capacity and a smaller inverter and electric motor so it should be cheaper. Maybe they should do what Toyota did and either license Tesla's technology or have Tesla build the drive train and battery like they did for the Toyota electric Rav 4.
I see tons of EV cars in California. Leafs are everywhere.
If Tesla can make a full-sized sedan with a 265 mile range (85KWh battery) for $73,570 while averaging a 25% profit margin there's no reason why Fiat shouldn't be able to make a profit selling a much smaller car with a much smaller battery and a much smaller range. Perhaps they should invest in Tesla's gigafactory to bring down cost and/or license Tesla's batteries like Toyota did. If they're like Nissan then they only need around 1/4 the capacity of Tesla's 85KWh battery pack, a smaller electric motor and a smaller inverter. If it cost Fiat $46,650 per-car then they're doing something wrong. They're probably using those more expensive, lower energy dense LiFePo prismatic batteries that everyone except Tesla is using.
It's not that it's 0.43mm per year, but one needs to look at the rate of change. It may be 0.43mm/year now, but in 10 years it may be 4.3mm/year or higher. The problem is that all of these things we are seeing are increasing in a non-linear mode.
I alreay use a product called Stick'n'Find. It's like Tile except it has a user-replacable battery and is available now. It also supports Android. It also provides a SDK and will sense temperature.
They are not standard lithium-ion batteries. For example, the cell design is different in order to cut cost. They do not have the normal dimple on top nor do they contain the protection circuit. The protection is provided by a low melting point wire that is bonded to the battery that doubles as a fuse. They are also not exclusive to Panasonic since they also work with Sanyo.
The chemistry is also a bit different. The batteries are automotive grade. They are designed to handle more temperature extremes and for rapid charging as well as for longer life than a laptop cell. As for Tesla owning the chemistry I heard that during one of the quarterly earnings conference calls. They did a lot of research and testing for their batteries. Arguably most of their technology lies in their battery management and pack designs but they also have a number of key patents in the actual cell design.
http://www.teslamotorsclub.com...
http://www.teslamotorsclub.com...
I had a good talk with a friend of mine who works there in R&D. He was describing it all a compromise between three things: reliability, performance and capacity. Choose two of the three.
They have some interesting patents as well, such as a metal air battery design.
Actually they don't make much on carbon credits any more and haven't for a while. Their margins are already over 25% on their cars so they are making money on their cars.
You don't need to exceed gasoline in terms of energy density since the efficiency of an electric motor is much higher.
Tesla has stated that they don't make all that much any more from ZEV credits. They're doing fairly well. They had a bit of a loss this last quarter due to their rapid expansion (which is what they should be doing). The GAAP accounting penalizes them quite a bit for the way it deals with leases which is totally unrealistic. It prevents them from recording any income from the leases for quite some time which is why the big investors look at the non-GAAP numbers.
It all depends on the battery. Tesla claims that rapid charging has no impact on the life of their battery and they are rated for at least 3000 full charge/discharge cycles, which for an 85KWh battery pack is over 600,000 miles. Also, typically on the Tesla owners don't need to charge to 100% or drain it close to 0% since it has a decent range.
The Leaf, by comparison, is another story. The Leaf lacks active battery cooling and in hot climates the batteries have been dying. In Arizona people are reporting 40% range loss in two years and rapid charging does impact the life of the batteries.
Here's one where a Tesla was rear-ended by a semi. The Tesla was able to drive away, despite being knocked forward around 100'. The semi wasn't going anywhere.
http://www.teslamotorsclub.com...
They use a custom 18650. The cells are not normal 18650 cells. They use that format for cost reasons. Elon has said that ideally they would prefer a slightly larger format.
The reason they use so many small cells has to do with safety. The battery packs are designed so that a few cells dying won't have a noticeable impact on the pack. It also adds to the safety.
As it is, their energy density is much higher than their competitors who use LiFo prismatic cells.
Actually the AC system should last a lot longer too. The compressor is electric and fully sealed and since there's nothing mounted to the engine with flexible hoses the chance of it leaking is reduced significantly. There's no compressor clutch either. You also get rid of the power steering pump since that is now electric. The factory told me the electric motor in my Model S is lubricated for 12 years and it's easy to remove by just undoing some bolts.
While I haven't seen the compressor on my MS the one on my Prius was totally sealed, though for some inexplicable reason they still mounted the electric AC compressor to the engine block and used flexible hoses for the AC.
The dealers don't like EVs because it gets rid of one of their biggest sources of revenue, service. EVs don't need oil changes or belt changes. The complexity of the drive train is significantly reduced. Even the brakes will last a lot longer due to regenerative braking. About the only thing dealers will have to still do is rotate the tires and do the basic inspections. There's no spark plugs, air filters, fuel filters, fuel pump, belts or the myriad of other mechanical parts that wear down over time. There's no transmission even. There's no fuel injectors, exhaust system or anything else.
There are only about a dozen moving parts in the drive train on my Tesla model S. The electric motor drives a 9.71:1 gear reduction and a differential that goes to the wheels. That's it. At the factory I was told that the electric motor is lubricated for 12 years.
The AC and power steering are electric so they will wear much better than the belt-driven pumps and compressors that have clutches.
The parts that can break are also much easier to get to without an engine in the way. If the motor does need to be worked on it's relatively simple to pull out the entire drive train. The National Geographic special on the Tesla factory shows them installing it in under 5 minutes. See https://www.youtube.com/watch?...
The power steering, AC compressor and coolant pumps and whatnot can be easily reached by just removing a panel and/or pulling out the plastic frunk module.
Actually Tesla does own the IP to their batteries. Panasonic manufactures them and provided some IP, but Tesla has their own unique chemistry and cell design. Tesla's cells are automotive grade yet are much higher energy density than the LiFe cells other manufacturers use. They also developed their own battery management support. A quick search turns up a lot of battery related patents that Tesla owns. They do a lot of active battery research as well. You cannot buy the Tesla cells from Panasonic even though Panasonic manufactures them.
It was the 3.11 kernel (OpenSuSE 13.1 with all the up-to-date patches). Normally I run XFS which for me has always been rock solid and extremely stable and reliable (even in cases where I had major issues occur with RAID). I have NEVER had a single issue with XFS and I have been running it for years. The tools for it are top-notch which is the main reason I run it. I also usually back up my systems. Since it was the only drive in the system and only 250GB I did not separate /home from root. Fortunately after fsck I was able to recover most of the data but the root directory was totally trashed and all the recovered data was under lost+found.
I tried btrfs a year ago but gave up because the performance made it unusable. My mail server (Cyrus) stores each email as a separate file. I gave up after waiting an hour to untar my email onto a SSD, something that takes only seconds with XFS or EXT4.
In this case it was a server provided by IT. I was working out of state living out of a hotel room. Since I was in a huge rush with the project there was no time to set it up to back up, nor did I have anything to back up on to.
I have two others. One of which bricked itself after two weeks. So far I haven't had any problem with the replacement but it gets backed up daily.