This is Bob Wallace's Typepad Profile.
Join Typepad and start following Bob Wallace's activity
Join Now!
Already a member? Sign In
Bob Wallace
Recent Activity
You are right, Dave, none of us know. But the people who are very knowledgeable about batteries are consistently telling us that EVs push out other vehicles when battery prices drop low enough. They consistently talk about battery price, not pack price. We have Navigant Research telling us that the material cost for the Panasonic cells that Tesla uses is about $70/kWh and the finished price should drop to about $100/kWh. That is consistent with what I've been reading for some time. Citigroup recently cited $230/kWh as the key mark where battery storage wins out over conventional generation and puts the fossil fuel incumbents into terminal decline. UBS has stated that the $230/kWh mark will be reached by the broader market within two to three years, and will likely fall to $100/kWh. (Tesla is already at $180/kWh based, apparently, on their purchasing volume.) Tesla/Panasonic is predicting a 30% cost drop when their new factory is running. That's under $130/kWh. Based on what multiple sources are telling us we need no new chemistry to make EVs dominate. All we apparently need is large scale production (economies of scale). Look at this graphic, Dave - At $250/kWh PHEVs and hybrids are goners. At $150/kWh ICEVs need fuel for less than $2/gallon to compete. I recognize that this is data that some people don't want to accept. But if one can't show how the data is wrong and throws it away then they are not behaving as a rational player, but a "believer".
There are multiple long range freight options. H2 FCEV is one. Electrify our rail system and use it rather than long haul trucks. Use battery trucks for "the last mile". (The Tras-Siberian railway is electric and runs about twice the width of the US.) Someone calculated that three Tesla S battery packs would power a loaded 18 wheeler 100 miles. With a few more years of capacity increase we could probably have 200 mile 18-wheelers and use battery swapping. Pull into a swapping bay and drive out charged in less than two minutes. We've got trucks running on overhead wires similar to urban electric trolleys. If enough power could be delivered it might be possible to install wires every few miles along our major highways and let trucks charge up and then run on batteries in between overheads. Super-capacitors would likely come into play here. South Korea has full sized buses running on wireless charging embedded under about 10% of the 18 mile route. Perhaps enough power could be delivered this way to keep trucks rolling and charging. I suspect we'll see several ideas tried out over the next few years. I'm kind of partial to electrified rail. That would take a lot of traffic off our roads and greatly reduce road damage, saving us a lot in highway expansion and repair. We're going to be freeing up rail space as we phase out coal and petroleum.
We don't need seasonal storage. We need storage for a few days at a time. The number of days is an unknown and will remain so as we build out our grids and interconnect them. Right now it looks like batteries will be the affordable answer for perhaps a few days storage. If we need power out further then we'll likely turn to some sort of dispatchable generation such as natural gas, biofuel or H2. So far this year (end of week 44) Germany would have needed longer term storage during one week. That power could have easily been put away during the previous few weeks. By the following week wind and solar were back up and producing closer to the annual average. During even the low week wind and solar produced over half of average annual output. Where the cutoff is between "3" and "7" will change over time. Cheaper storage will extend the "3". Cheaper dispatchable generation will lower the "7". Whether it would make sense for Germany to over-build wind/solar, buy hydro from Sweden, buy stored hydro from Switzerland or Austria, buy wind from the UK, or buy solar from southern Europe needs to be considered along with storage calculations. As we extend our grids we lower variability and allow for shared storage.
"Companies such as VW don't know, and so are developing both batteries and fuel cells." Let's turn the clock back 5, 6, 7 years. At that time it was believed by many that we had to develop a replacement for oil because of Peak Oil. It was also clear to others that we needed to get off oil because of climate change. Batteries weren't very good. Fuel cells seemed, I would imagine, as likely the answer as batteries. I suspect people assumed the H2 would be clean and not reformed methane. Several car companies started developing both FCEVs and EVs. In some companies their group think probably favored one solution over the other. Four years ago Nissan introduced the Leaf. The Tesla S hit the market only a bit over two years ago. I suspect the Leaf, Tesla and other EVs made knees knock in the companies that picked the FCEV route, but they plunged ahead. Before long we'll see FCEVs up for sale. We'll see independent reviews and some owner experience reports. Over a couple of years we'll probably get a very good idea if FCEVs have a future. Perhaps they do. I can't find an objective reason why the market would swing to FCEVs but perhaps there's a factor that has yet to be identified.
Citigroup last week cited $230/kWh as the key mark where battery storage wins out over conventional generation and puts the fossil fuel incumbents into terminal decline. UBS, in a report based around a discussion with Navigant research, says the $230/kWh mark will be reached by the broader market within two to three years, and will likely fall to 100/kWh. And it predicts that the market for battery storage will grow 50-fold by 2020, mostly in helping households and businesses consumer more of their solar output, but also at grid scale and with electric vehicles. So here are some highlights gleaned from the UBS discussion with Navigant: Navigant estimates the cost of materials going into a battery at the Tesla Gigafactory on a processed chemical basis (not the raw ore) is $69/kWh [this metric is per kW per hour of operation]. The cost of the battery is only ~10-20% higher than the bill of materials – suggesting a potential long-term competitive price for Lithium Ion batteries could approach ~$100 per kWh. Tesla currently pays Panasonic $180/kW for their batteries, although conventional systems still selling for $500-700/kWh. But Navigant says that the broader market place will reach the levels Tesla is paying in the next two to three years.
I checked resale prices for 2011 Nissan Leafs and Toyota Camrys. About the same based on mileage. Fuel savings for a 13,000 mile driver is about $1,200 a year. ($3.50 gas and 12 cent electricity.) Add some more in for oil changes and higher maintenance costs for the Camry. The MSRP for the Leaf is ~29,010 and $21,510 after federal subsidy. The MSRP for the Camry is $22,970. The Camry is a bit larger, but if you're someone who is considering a Leaf then you're probably talking a commute and errand car, not a long trip car for the family. And I would expect the driving experience of the Leaf somewhat better.
"Batteries are prohibitively expensive for any significant use as backup power for the grid." Let's examine that for a moment. EOS Energy Storage's zinc hybrid cathode battery specs are - $160/kWh 10,000 cycles 30 year calendar life 75% efficiency Now let's assume 6% financing over a 20 year term. That means an annual payment of $13.80 per kWh. (I'm leaving out the cost of battery charger and inverter. Thirty year financing would lower the annual payment.) Cycle that battery every day (4c/kWh electricity with efficiency loss = 5.3c/kWh input costs) and the stored electricity would cost 9.1 cents. It's unlikely H2 with it's high loss rate could touch that. The cost of storing electricity for a week jumps to 32 cents. That's less than the average cost of gas peaker power in California (49 cents). Storing electricity for a year in an EOS battery would cost 1,385 cents. Obviously that's too expensive. There is probably some point in terms of number of days of storage where H2 would be less expensive than batteries. But batteries are pretty much the best choice for short term storage. -- I'll suggest that the cost of battery charger and inverter can be ignored. These batteries will operate for grid smoothing as well as storage and the BoS costs would likely be paid through their grid regulation role. One starts with a grid regulation system and hooks up more batteries for storage. The additional cost is a parking space and some cable.
First,I'd have to understand why an EV has to have a 300 mile range. Some people may take off on long drives with a baloney sandwich and an empty bottle to pee in, but most people stop and take a break after a few hours.
"If batteries go down a lot, and other things don't, then they will be much cheaper than the other things." It's a very safe bet that battery prices will go down. Nothing other than increased production levels is needed to bring down battery prices. No new inventions. No magic pixie dust. Just more automation and larger scale material purchasing. Toyota says that hydrogen may eventually go down from 17 cents a mile to 10 cents a mile. Switching to "clean" H2 would be more expensive. Being agnostic is fine. But try to not define closing your eyes to clearly verified facts as agnosticism. That's the sort behavior that believers engage in as a way to protect their beliefs.
Harvey, I think this is not a real problem - "(1) long time to fully charge (30+ minutes)" The only time an EV driver is going to use a rapid charger is on a long drive day. Once we've got 200 mile range EVs that means starting with 200 miles, stopping 30 min, driving 180 miles, stop 30 drive 180. 560 miles with one hour spent charging. Eat a meal and pee/get something to drink during those charging stops. A FCEV driver will stop and refill, using about 10 minutes. Then almost all will stop for a meal and again to pee/get something to drink. They will spend maybe 40 minutes not driving, arriving at destination 20, even 30 minutes ahead of the EV driver. To save that half hour on a 550 mile drive they will spend $55.50 to $93.59 for hydrogen (Toyota's numbers). The EV driver will spend $16.50 (also Toyota's number). Then there's the rest of the year. 13,000 miles at 17 cents costs $2,210. At 10 cents $1,300. At 3 cents (for the EV) $390. The FCEV driver, if they refill with 20 miles left in their tank will stop 46 times a year to refill. Interrupting ones drive, getting out, hooking up, swiping, unhooking, getting back in. 15 minutes? That's 11.5 hours vs the multi-use time spent only on long trip days by the EV driver.
Let's set advocacy aside. Work the numbers for FCEVs with their H2 from reformed methane. (That will help with nitrous oxide but not with CO2 emissions. The H2 really needs to come from renewable electricity but that would be more expensive.) Work the numbers for EVs charged with renewable electricity. Advocates may buy a few EVs or FCEVs simply because they are advocates, but the vast majority of the market is going to respond to cost. As I read this article Toyota is saying that at first it will cost 17 cents per mile to drive a FCEV. (300 miles for $50.) And that cost may eventually drop to 19 cents per mile. (300 miles for $30.) Toyota also says in that article that it will cost 3 cents per mile to drive an EV. (300 miles for $9.60.) And to add a bit of perspective, driving a 50 MPG ICEV on average priced $3.50 per gallon gas costs 7 cents per mile. 3 - 7 - 10 - 17. Which do you think most people would choose?
Cost per mile to drive. That's the critical thing here. If FCEVs can't be operated for considerably less than ICEVs while costing more how are economies of scale reached? There may be some advocates willing to pay a premium to both purchase and operate a FCEV. But are there the 100,000 to 500,000 numbers it takes to create real price drops? Then, assuming there's some way other than large scale manufacturing to bring purchase price down to ICEV level, why would someone purchase a car that costs more per mile to operate than another option? (Don't look to "greenies". H2 from reformatted NG isn't green.)
Let me share this, just read it a few minutes ago - "Speaking at the JP Morgan Auto Conference in New York, Toyota’s senior vice president Bob Carter said that Department of Energy estimates suggest that a full tank of compressed hydrogen will cost around $50. This will fall to $30 in time, however. Toyota’s ‘mass production’ fuel cell car will have a range of 300 miles when in arrives in California next summer. Refueling will takes minutes, while the Japanese giant says it has modeled “specific locations” that will enable the majority of owners to reach a station in just six minutes. By comparison, nationwide fuel economy figures indicate that the average driver pays $44.50 to travel 300 miles while owners of the Toyota Prius, with its EPA rating of 50 mpg, pay just $21." I can't determine if the $50/$30 price for H2 includes road tax or not. Let's assume, out of caution, it does. That says that right now it's going to cost more than twice as much per mile to drive an H2 FCEV than to drive an efficient ICEV. Why would someone pay as much or more for a FCEV and then 2x+ as much to drive it? Even when we work the price down to Toyota's assumed lowest price it's still going to be cheaper to drive an ICEV. Unless FCEVs can be built in high numbers prices are very unlikely to come down. And unless there are a lot of FCEVs on the road the investment for a H2 infrastructure won't be available.
In a UCS survey 56% of US drivers reported having a place to plug in where they parked. About 14% (16%) could use an existing outlet where they park at work. Expanding work/school outlets is likely the best route for reaching the other ~44%. It's going to be easier to install strings of charge points in parking lots. (Head to head parking means only one "post" per four cars.) That, and apartments/condo and parking garage that are already installing outlets. Some locales are already requiring charge spots for new construction.
The EIA/McKinsey number of about $250/kWh (in the linked figure) says battery price, not pack price. But let's assume that's incorrect and it's pack price. When the new Tesla/Panasonic giga factory is up and running the cost of cells-batteries is expected to fall to about $130/kWh. Doubling that number to reach pack price still kills PHEVs. And, as you state, Tesla battery packs are now likely in the $250 to $300/kWh range. At the low end of that range PHEVs are finished. The giga factory should knock $50 off the price bringing the upper end into PHEV-killing range. The material cost for lithium-ion batteries is about $70/kWh. We are heading toward $100/kWh batteries. As the price moves down capacities are likely to continue increasing. That means that we will need fewer pounds of batteries to power an EV a mile which decreases the amount of kWh storage that needs to be purchased. EVs become even more price competitive. There appears to be some point out there, this year or a very few years from now, at which PHEVs are priced out. Regardless of whether the range extender is an ICE or a fuel cell.
Whit, I don't know what the future will hold for fuel cells. I'm not seeing a likely future for them with personal transportation. Or even large scale transportation for that matter. Battery swapping for large trucks would likely beat out fuel cells. Where I can see a possible future for fuel cells is "the last 10%" - deep backup for the grid. As storage prices drop it will likely be most price efficient to run the grid with mostly wind, solar and storage. But storing out past a couple days of low wind and solar input gets expensive. We will probably need some sort of dispatchable generation that can be turned on a handful of times a year and run for a few days at most. Whatever that generation is it needs to have the right combination of low capex and easy to store, not expensive fuel. H2 is harder to store than, say, bio-oil. And harder to store than synthetic fuels. Oil and other liquid fuels can be stored in simple, non-pressurized tanks or even "holes in the ground". Will fuel cells be cheaper than turbines that can burn oil? The overnight cost of a CCNG plant is pretty low, just over $1/watt. Capex is very important because these generators will be sitting idle most of the time, thus will have few hours per year to recapture fixed costs. Will it be cheaper to produce H2 using renewable energy or to produce bio-oil or some sort of synthetic fuel? H2 back to electricity is very lossy. Is there a biofuel or synthetic fuel that can be used in fuel cells as efficiently as combined-cycle plants? I'd say we're a few years away from understanding what sort of future fuel cells might have.
Dave, yes, a PHEV with a H2 fuel cell range extender would mean few trips to fill up but the discussion is about FCEVs, not PHEVs. The problem for PHEVs (gas, diesel, or H2) is that once battery prices drop below about $250/kWh then PHEVs are no longer competitive. And it seems that Tesla may now be paying Panasonic only $180/kWh for batteries. htge-costs-plunge-below100kwh-19365tp:// I don't have any firm opinion on wireless vs. plug in charging at this time. I can easily see many people moving to wireless for 'ordinary' charging. But I've not read about a demonstration of wireless rapid charging - the sort of power exchange done at a Tesla supercharger. In the future we might do some of both, wireless charge for normal days and plug in for speedy charges on long drive days. I can't see spending big money for a Tesla robotic home charger. Why not just install a wireless charger for a few hundred dollars and a small amount per mile to cover the inefficiency of wireless charging? A loss of less than 10% from what I've read. Would increase electricity use from 0.3 kWh/mile to 0.33 kWh/mile or a third of a penny. Less than $50 a year for a 13,000 mile driver.
No, E-P, I avoid you as much as possible. And while I agree that we may see autopiloted cars charge themselves that's a ways off. if the city streets are ever to provide electric charging for the majority of vehicles kept ungaraged, there is no way that that can be done with charging posts and wires That is obviously incorrect. All it requires is enough charging posts and connecting cables. Since it is not necessary to charge all EVs every night then we wouldn't even need 100% coverage, just a smart assignment system so that the cars that need charging would be able to plug in. I'm not saying that would be the best, most economical solution, just a workable one.
If anyone is working behind the scenes to promote FCEVs it's most likely the natural gas industry. Our H2 supply will come from methane. It won't be green energy, but another opportunity for the gas industry to sell product. -- I'm not saying that I think anyone is necessarily pushing H2 FCEVs in secret. I think what has happened is that a few years back batteries didn't look promising and H2 looked like a reasonable way to replace petroleum. Car companies set up divisions to develop FCEVs and these operations developed power and momentum. EVs have appeared to be viable only in the last 2 or 3 years. Nissan Leaf sales started in the last two weeks of 2010 and the first Tesla Model S deliveries were in June 2012. Probably less than two years ago was it obvious that EVs were for real and coming on strong. FCEV programs were underway and probably operating with a certain amount of denial about what they were facing. I think what we're seeing is the continued growth of an evolutionary branch that will, with a little more time, wither and die. But that's my opinion. I'm very happy to be proven wrong if FCEVs can prove to provide us a cheaper way off petroleum and high CO2 emissions. (Moving to methane does not count.)
"Like a car with combustion engine, refueling takes no more than around three minutes." While the advantage of FCEV cars is that they can be refilled faster than EVs this claim of "three minutes" is less than totally honest. To refill in three minutes one first has to interrupt their drive and go to a filling station. Sometimes that may be a convenient as pulling over as you drive past. For many early FCEV drivers it would mean leaving your desired route and driving some significant distance. Then you've got to get out, hook up, swipe your card, fill for three minutes, unhook, get back in, and return to your route. While something similar will happen for rapid charging an EV the EV driver will only need to do that a handful of times per year when on very long drives. The FCEV drive will make 50 or more fueling stops per year. What little time savings is gained on very long driving days is more than outweighed by the hours of refilling that happens during the rest of the year.
"I guess the first robotic home chargers for Tesla will cost 3 to 5k USD as they are low volume items. If Tesla can get to 30k unit sales per year or more the price could go down maybe in half." I would imagine the cost of "robotic charging" could drop to "very low". Imagine this - We require all cars to have rapid charge port at the same spot on the vehicle - 14" aft of the center of the front wheel and 18" off the ground / wherever. We install 'parking guides'. Semi-flexible slots for front wheels. Until the car is all the way forward in the slot the charger will not operate. The charging process is automated. First the car signals that it wants a charge and that its charge port is open. Then an arm from the charger slides straight out and the end engages with the car plug. By using appropriate materials and allowing a small amount of flexibility in the arm end no sophisticated control would be necessary. (Think - plug at the bottom of a funnel.) Adding systems to make sure the access door was open and the car correctly positioned would be cheap. (Think really cheap digital camera and image analyzing software - simpler than face detection found in cheap cameras.) Metal, a short run of flexible cable, a single motor to extend and retract the arm. The way to make the systems very cheap is to eliminate the need to 'think' and adjust. Make them 'stick and go'.
EVs (and PHEVs) will likely be 'dispatchable loads' which the grid will use to smooth out supply peaks as we add more renewables to the grid. On average an EV will need to charge less than 3 hours a day with a normal 240 vac feed. By allowing the grid to determine actual time of charging EVs can be brought on line or taken off line as desired. A very likely solution for those who have no place to charge where they park at night is to provide them charging during the day. Install grid-controlled, low cost outlets in work and school parking lots. Plugged in cars become daytime dispatchable loads, useful for regulating daytime peaks. Dispatchable loads are important to and valuable for grids. They reduce the amount of capacity that has to be built. They minimize storage needs. They can help keep prices stable and avoid the use of expensive peaking power.
Here's your problem, Roger. Studies find that once batteries fall enough in price hybrids and FCEVs are no longer price competitive. Citigroup (IIRC) set that price point at $230/kWh. Apparently Tesla is now paying Panasonic $180/kWh for their batteries. As for making a profit. Tesla, like any other company that stays in business sells product for more than it costs to produce. For a luxury item the margin might be higher because the market will tolerate it. The markup on a $100,000 car might be $25,000. The markup on a $30,000 car might be $4,500. But if the company sells six $30k cars they make more than selling one $100k car. And there are far, far more people who are willing to pay $30k for a car than there are willing to pay $100k for a car. The ratio might be 1,000:1.
No, I don't think I have the full picture which is why I asked you to explain the route to affordable nuclear. So far what I hear said is: 1. Build new type reactor. 2. 3. 4. 5. 6. Nuclear electricity becomes cheap enough to compete. I want you to fill in the missing steps of how nuclear gets from >$0.11/kWh to ~0.05/kWh. If you can adequately do that then I'll have some more of the picture. I'm pretty sure cost of fuel is not the answer. And if by not having to refuel for 20 years meant an almost 100% CF that wouldn't do it. Moving from 90% CF to 100% CF would drop the price only a penny or so.
" Smaller Model III with 200-mi range has less popular appeal, and will not bring much profit." A 200 mile range EV for under $30,000 will not be popular? A Tesla for less than the cost of the average new car ($32,046) in the US? The profit margin will be less for the "3" but it will sell as fast as Tesla can crank them out.