This is Henrik's Typepad Profile.
Join Typepad and start following Henrik's activity
Join Now!
Already a member? Sign In
Recent Activity
There is no shortage of lithium and there will never be. End of story. One ton of lithium carbonate is enough to make 2000 kwh of lithium batteries that will cost at least 200,000 USD assuming a very low battery price of 100 USD per kwh. So a price of 7,000 USD for a ton of lithium carbonate is not going to make lithium batteries expensive. Lithium prices could easily double once more and still not be a problem for battery producers like Tesla. 1kwh of lithium battery contains about 0.5 kg of lithium carbonate that currently cost 3.5 USD. Musk said 1kwh of lithium battery contains about 80 USD worth of raw materials when bought at London Commodity Exchange. Musk aims to make his Giga factory so efficient it only needs 20 USD in costs to turn the needed 80 USD of raw materials into 1kwh of finished cells. Interview with Musk 24 min into video he say there is 80 USD worth of raw materials in 1kwh of lithium battery. It is a good thing that lithium has increased in price to 7000 USD per ton because that leads to more production and investment in new production capacity for lithium carbonate which is needed. Consider just one of the world’s easy to mine lithium reserves in Argentina. There are 128 million ton of minable lithium carbonate in Argentina so we could make the batteries needed for 2560 million cars with a 100kwh battery in each car (100kwh requires 50kg of lithium carbonate so 20 cars per ton of lithium carbonate and 20*128=2560 million cars). There are about 1,500 million vehicles globally so Argentina alone already has all the lithium the world needs for its current vehicle fleet. However, eventually resources like argentines lithium brine will be mined up and other sources of lithium will evolve. The most promising is to make lithium from salt from desalination facilities. Such production does not exist today but there are endless amounts of lithium in salt from seawater (like trillions of trillions of BEVs) so it is not a resource that will ever run out like oil will. And any country with access to seawater can make it. Moreover, lithium can be recycled so the same lithium can continue to power the same stock of vehicles forever. Not so with oil and gas.
Anyone still doubt we will see fully autonomous cars by 2020 everywhere?
Perhaps an industry will emerge for retrofitting old non-autonomous cars with autonomous tech so that they become self-driving. That could make them more wanted and keep demand for oil up for a longer time. Another possibility is that some oil exporting countries collapses under Islamist extremist infighting and that this will keep oil prices higher. That would have happened long ago for several countries had it not been for repeated use of Russian and US military power to stabilize the existing Islamic regimes. It is impossible to predict what will happen politically in this regard. However, it is easy to predict that full autonomy tech will make new BEVs more economical than new gassers in terms of total cost per mile driven. BEVs last much longer than gassers and cost less to fuel and maintain than gassers. However, they cost more to make than gassers so full autonomy is needed for BEVs to take full advantage of their lower operating cost and drive them 100k miles per year for 10 years. Full autonomy is also needed for BEVs to solve their remaining range anxiety issues and charging issues.
Just noted that the Model S P100D’s 315 miles range is a massive 45 miles better than the 270 miles Model S P90D it can be compared to. Must said only 200 packs of the 100kwh kind can be made currently per week so the even longer range model S 90D with 294 miles range will not get the 100kwh pack until a few more month from now. The upcoming Model S 100D should therefore get 340 miles range. Harvey you can easily get more miles than the 315 miles if you stay below 60 mph and abstain from hard accelerations that will give you 400 miles. So if you have bad weather the day you absolutely have to drive 300 miles non-stop for 5 hours you can. Have you ever done that (5 hours without stopping)? I did drive from Copenhagen to Brussels once in my life and back the next day but I stopped for lunch each time. I could easily do that 571 miles trip in a Model S 90D charging at lunch break and one more resting stop. But I will never do it again 11 hours in a car is not fun at all. What you want - a 140kwh battery and 500 miles range - is most probably not in any significant demand for the extra price it will cost over a 100kwh 340 miles Model S 100D. I think Tesla will keep making longer range BEV options until they see no significant demand for more range and then they will stop developing better range battery packs. Anyway self-driving cars are coming by 2020 and then the range debate for EVs will be over as it will no longer be relevant.
ECI i3 still has a small gas tank outside of California where it is sold. The "rotting" gas is a real problem for PHEVs with large battery packs. To me it is just another reason that PHEVs will not work. They still pollute and they are expensive. Better focus all resources on long-range self driving BEVs and scrap all R&D for combustion engines and complex transmissions.
Alain there will be some substitution effects. I think that self-driving BEVs will enter the market so massively after 2020 on a global scale that gasoline will drop in price and old non-autonomous gassers will sell for nearly zero USD thereby destroying the demand for new non-autonomous gassers. The poor will end up with the least attractive and unsafe vehicles: the old gassers that are nearly free apart from some maintenance and insurance that will also be less when the vehicle cost less to replace in case of an accident.
Affordability begins late next year for Tesla with Model 3. It will also become fully self driving so the owner can make money renting it out when not using it himself. In that way 20 cents per mile should be possible for a Model 3 and that is all cost including insurance, maintenance, amortization and energy. This is nearly half the cost of the cheapest new gassers on the market that are more like 35 cents per miles everything included. However, it depends on Tesla being able to deliver on the fully self driving capability. I do not think that Tesla’s vehicles will be fully self driving and approved for driving people without driver license until 2020. I think Tesla’s fully self driving hardware is ready at launch of Model 3 in late 2017 and their software is ready by 2019 and the data documenting safety in order to get the legal authorization to operate Tesla’s vehicles as 100% driverless vehicles will be ready by 2020. For people not wanting to buy a 35k USD model 3 transportation is still available at about 20 cents per miles simply by using other people’s Teslas as well as the Tesla fleet that Tesla owns and make available for transportation in areas where there is otherwise not enough privately owned Tesla’s to meet the demand for Transportation at any time of the day. Tesla’s CTO Straubel said they did the 100kWh battery pack using the same cells as in previous packs but that the pack is reorganized to fit more cells. At the Giga fab opening he said they will begin production of Model 3’s battery pack using a new larger cell by around February 2017. At about May 2017 they will start making battery packs at the giga fab for model S and X also using the new larger cell made at the giga fab. My bet is that by summer 2017 Tesla could start selling a Model S and X with a 110 kwh pack and 350 miles EPA range. You can be sure there will never be a 500kw FCV with 350 miles rage as it is not technical feasible unless you have hydrogen tanks at the passenger seats as well as in the trunk area.
This is good. It gives those automakers that can't make their own system a chance to still sell cars after 2020. By 2020 all new luxury cars will be fully autonomous level 4/5. By 2025 all new cars sold will be fully autonomous. There will be zero demand for non-autonomous cars on a global scale. Auto makers that are late to this game will bankrupt. Either you are among the first to sell fully autonomous cars or you will bankrupt. I still expect a huge market for highly discounted used old cars without full autonomy. They will be for the absolut poorest.
Interesting what about price? Only for China?
I may add that for high definition 3D mapping you will use light sensitive black and white stereo cameras so they can function at night like lidar. Cameras will require more computing power and more advanced software than a lidar sensor making the same quality 3D maps because the lidar signal is more simple than the camera signal. However, if you got the computing power and has solved the software issues then Lidar has no advantages over cameras and they cost much more.
Mahonj when the electric rage is high in a PHEV you need a smaller gas tank in order to be sure the gas is used up frequently enough for the car to work. Old gasoline will destroy the gas engine. That is why. When Tesla launches model 3 the market for small PHEV cars will largely disappear I expect so problem solved. BEVs with 200 plus miles range will cost the same as PHEVs and then the demand for PHEVs will go away.
Lidar is not mission critical for self-driving cars. Stereo cameras can make 3D maps just as Lidar can and in the same detail. Even radar can make 3D maps but not yet in the same detail. Please explain why anyone thinks Lidar is mission critical for autonomous cars. Only advantage I see with Lidar is that it takes less GPU power to create a 3D map using Lidar than using cameras. However, if computing power is not a limiting factor to get the job done then Lidar is not mission critical.
Is the Rover’s autopilot OTA updatable or not and does it collect accident data?
This is the clearest statement about intent to make fully autonomous vehicles that I have seen from any of the old auto-makers. And the best part of it is that Ford believes they can have a volume product ready by 2021. GM’s CEO has said they need 10 years or until 2025 for full-autonomy cars so Ford is clearly more ambitions than GM on this all important issue. I also like that Ford realizes that fully self driving vehicles are by far the most important thing that will happen in the auto-industry since the invention of the mass produced Ford T with a combustion engine. What Ford is not so clear about is the plan for how to get from where they are now with 10 test vehicles to a mass produced fully autonomous car that is approved for autonomous taxi driving. Tesla’s plan for this is clearer IMO. Tesla will keep improving its autopilot offered for all of its cars until it can drive 10 times safer in all circumstances than the average human driver. Then they will use their massive fleet data on accidents with autopilot engaged to get approval to operate their vehicles without driver oversight, i.e. as fully autonomous vehicles. In terms of statistics you need about 6 billion miles per unique driving environment (like USA) to scientifically prove that your system drives 10 times safer than the average human driver. 6 billion miles can be driven by 400,000 cars that drive 15,000 miles per year each. Ford, Tesla, Google, GM, Uber and everyone else also need that statistical documentation before they can get the permission to operate their cars without human drivers in control of the car. Tesla has realized this but I am not sure the old auto-industry or even Google get it yet. Driven around with 100 test vehicles will not be enough by any means. We need 400,000 vehicles to make a proper statistics. Also Ford will stick with LIDAR. This is not the right call because Lidar cannot see color as needed to see a traffic light and Lidar cannot see in heavy rain or fog as radar can. Ford’s vehicles will therefore also need cameras and radar. Tesla’s next autopilot hardware will get more cameras and radar but no Lidar as it is not needed for anything as cameras and radar can do all that Lidar can do and them some more like color and heavy fog driving. My point is that Lidar is not needed for making fully autonomous cars but cameras are and radar are needed for giving autonomous cars abilities that human drives do not have because we cannot see in heavy rain or fog.
I am beginning to get a little hope back that there is light at the end of the tunnel with regard to the old auto-industry actually beginning to be serious about using resources to make long range and driverless BEVs that can compete with what Tesla has and are developing. To really change the world as fast as possible in that direction we need more than one corporation (read Tesla) to do all the hard work. I like that VW has now realized that long-range BEVs have more cabin and trunk space than similar powered gassers or FCV because batteries and electric motors are more compact than the alternative drive trains although BEVs are still more heavy.
We need zero emission engines that are sustainable not another planet destroyer like this gasser. It is time we start to phase out polluting tech by banning the registration of new non-zero emission vehicles for people and corporations in city areas. Then keep expanding these zero emission zones for new vehicle registrations until all new non-zero emission vehicles are banned altogether from 2025. It will still take another 20 years or until about 2045 before all vehicle air pollution is eliminated as there is a large fleet of gassers that needs to be worn out and scrapped if a total ban on new vehicles is effective by 2025. It is either that or a global warming apocalypse with mass extinction of species that cannot adapt fast enough to a warming planet. Another advantage of simply banning polluting tech is that we do not need subsidies or taxes that are costly to administrate and enforce. A ban is far cheaper to implement and enforce.
Harvey you should listen to ECI because he is right. Even if the cost of electricity for electrolysis of renewable hydrogen was an imaginative 0 USD per kwh hydrogen would still cost about 7-10 USD per kg at the hydrogen filling station because of the extreme cost of equipment and maintenance that is needed to make and store hydrogen at such a station. It will never be able to compete with gasoline or electricity for BEVs. Davemart would argue that this can’t be true because most of the old auto-makers are investing in hydrogen and they can’t be wrong about the long term cost potential about hydrogen. Problem is that they are wrong and that we live in a world were stupidity rules in even the most improbable places like the management of our largest corporations. However, in the long term it does not matter much. Market competition (read natural selection of the fittest) will bankrupt those firms that stick with uneconomical technology. You will see for yourself in a few years. The auto companies that copies Tesla’s business plan will prosper and the rest will perish.
Maarten you can drive a car and you do not have lidar eyes but simple video sight so we don’t need lider for fully autonomous cars. Science teams like lidar because its resolution is low (meaning less data input) and it works natively in 3D (meaning you do not need complex software to make dual cameras make a 3D image) and it work just as good with or without daylight. Camera sensors can do the same with the right software, more computing power and light sensitive sensors. Camera sensors will always cost less and have much more range and better resolution than lidar. I think radar will see more use in fully autonomous cars as they can see in heavy rain and fog which cameras and Lidar cannot. You could drive at full speed in heavy rain or fog without risk with a high resolution radar which would be valuable for taxi services and fully autonomous semis that need to operate as many hours as possible during the year. PS a laser mosquito killer will definitely use 3D cameras as Lidar does not have the resolution to see and recognize a mosquito.
It does not work in real life. SOFCs are heavy and bulky and that is why they can only fit 5kw in it. That is not enough. You can hardly drive 1 hour on the highway before the 24kwh battery is depleted and then you need to wait 5 hours to charge it again every hour or 70 miles. SOFCs are also really expensive and they take time to heat up before the make electricity. It is not like flipping a switch. But they are highly durable. Electrek took note that Musk believe his Giga factory will hit less than 100 USD per kwh at the cell level by 2020. Now add 30 USD per kwh for packing, cooling and power electronics and we get 130 USD at the pack level by 2020 including a standard 25% gross profit margin. The model 3 will get a 55 kwh pack to do min 215 miles range enough for most because of home charging and supercharging. So battery cost is only 55*130 = 7150 USD. A 5kw SOFC cost more and it is only 1/5 of the needed power to keep a car like that going on the highway.
Google has a big problem when it comes to commercializing its self-driving tech. In order to validate the safety of a self-driving system with regard to the all important death statistics you need to log about 6 billion miles per year. Google’s problem is that this will require about 400,000 cars driving 15,000 miles per year. Tesla will get that by 2018 and UBER could get it by outfitting all of their driver’s cars with self-driving tech. Google needs a plan for how they will get their tech into 400,000 cars for validation testing. I do not see anyone wanting to help Google with that. The old automakers can easily do it be selling 400,000 cars with the needed tech once they have developed it. However, they will do it with their in-house developed tech because they all can (it is not rocket science anymore because we now have the needed processor and sensor tech to make it happen) and because it is corporate suicide not to have control over the self-driving tech. Without the fully self-driving ability you will not be able to sell a single car by 2025 so you need to develop and control that tech in-house. IMO the only option there is for Google is to buy a car company like Volvo or Jaguar and start outfitting all of their cars with self-driving tech. Since these cars are also mostly gassers Google will also need to transform this car making company into a BEV maker because all self-driving cars will eventually be BEVs as they have significantly lower cost per miles driven when operated as self-driving cars that drive over 50,000 miles per year.
I think the Tesla pickup will be much like the F350 in size and towing ability but it will come with two or three battery pack options say 110kwh, 130kwh and 150kwh. 200kwh is not needed I think. If you really need it just load a 100kwh Tesla powerpack or two at the back of the pickup and connect it or load 10 of them at a gooseneck and connect it. I would not be surprised to see Tesla launch it only when it can be fully autonomous as that would guarantee its commercial success. A pickup that could get you heavy equipment without spending time to drive that pickup would be really attractive for any construction site or farmer. I am fully convinced Tesla can have such a beast ready by 2020 or 2022 at the latest.
LIDAR cost up to 80,000 USD a piece, car radars cost up to 200 USD a piece, car camera censors cost up to 20 USD a piece, and ultrasonic sensors up to 5 USD a piece. Nobody use LIDAR for production vehicles. Not a single production car in the world has Lidar because of the cost. We need fully autonomy capable hardware in production cars by next year so that the software can start its development and be rolled out using OTA so that in turn fully autonomous cars are a reality by 2019 and documented to be much safer than human drivers by 2020 so that we can change the law and allow autonomous vehicles without human driver controls. We also need 360 degree vision, redundancy and low air drag meaning no sensor racks on top of cars. All that means at least 8 camera, radar or lidar sensors per car and possible 12 or 16 per car. Lidar is prohibitively costly currently and will be so for many years ahead. We need fully self-driving cars in 2 to 3 years from now not in 10 years from now so using Lidar is currently a wrong decision for the R&D teams.
I think the most interesting news is that Tesla and Mobileye is parting. I admit I did not see that coming. It could have been really bad news with regard to Tesla’s ability to launch Model 3 on schedule. However, the following statement from Musk has convinced me it is not. I quote “This was expected and will not have any material effect on our plans. MobilEye’s ability to evolve its technology is unfortunately negatively affected by having to support hundreds of models from legacy auto companies, resulting in a very high engineering drag coefficient. Tesla is laser-focused on achieving full self-driving capability on one integrated platform with an order of magnitude greater safety than the average manually driven car.” Other news I have read have said that Tesla has hired top talent for chip design from AMD among others. Apparently Tesla is designing some of its own sensors and processors for the next generation autopilot hardware that will go into Model S and X first and then Model 3 at launch. Another, thing I read at electric is that the Giga factory will use a unique new cell format. Instead of the standard 18650 they will use a new cylindrical form factor and unique chemistries as well. What all this tells me is that Musk is unhappy with the speed of how traditional suppliers work. Musk wants the innovation to go much faster and using standard components will just slow down the development of Tesla’s products. So Tesla does it alone for the key technologies that will enable the transition to a zero carbon economy on a global scale. If others will not do it Tesla will do it and they are not waiting for anyone to get on board. Also I read that Tesla’s autopilot version 8 should enable Tesla cars to automatically exit highways along with major interface improvements. Tesla goal with autopilot is to make it 10 times safer than manual driving in all situations and achieve that goal as fast as possible. Tesla believes they are currently twice as safe (It is a belief because the data are not yet extensive enough to prove it with regard to the death statistics. We need at least 50 deaths to prove it). The beta designation for the auto pilot will not be removed until the autopilot is fully autonomous and conclusively documented to be 10 times safer than the average human driver. That would translate into 1 death per 900 million miles versus 1 death per 90 million miles as it currently is in the US.
Problem is that nearly nobody wants to buy the current plug-ins. They are less than 1% of total sales. Only the Tesla Model S and Model X are in demand with over 20% of total sales in their market segment at the places they are for sale and can be serviced. All other plug-ins currently sucks when compared to similar priced gassers and that is why they don’t sell. Everybody should copy Tesla’s Master Plan especially the piece about making long-range fully autonomous BEVs that the owners can rent out to others using a Smartphone app when not using the vehicles themselves. That will make it easy to get 100,000 miles per year on your BEV and benefit fully from the BEV’s very low marginal cost per mile driven. Long-range BEVs are expensive to buy but cheap to drive in terms of fuel and maintenance and they last much longer than gassers. So you need to get as many miles on a BEV as possible to minimize the total cost per mile driven. This is why full autonomy tech is needed before BEVs can become mass market. It is the only way to get enough miles on the BEV to make it hugely economical versus the gassers.