Toyota beginning on-road testing of new SiC power semiconductor technology; hybrid Camry and fuel cell bus
29 January 2015
|SiC PCU under the hood of the Camry hybrid test vehicle. Click to enlarge.|
Toyota will begin the on-road testing of silicon carbide (SiC) power semiconductors in Japan this year, using a Camry hybrid prototype and a fuel cell bus. The tests will evaluate the performance of the SiC technology, which could lead to significant efficiency improvements in hybrids and other electric-drive vehicles. (Earlier post.)
Power semiconductors are found in power control units (PCUs), which are used to control motor drive power in hybrids and other vehicles with electric powertrains. PCUs play a crucial role in the use of electricity, supplying battery power to the motors during operation and recharging the battery using energy recovered during deceleration. At present, power semiconductors account for approximately 20% of a vehicle’s total electrical losses; raising the efficiency of the power semiconductors is a promising way to increase powertrain efficiency.
|SiC PCU. Click to enlarge.|
Compared to silicon, SiC power semiconductors lose 1/10 the power and drive frequency can be increased by a factor of ten. This enables the coil and capacitor, which account for approximately 40% of the size of the PCU, to be reduced in size. Through use of SiC power semiconductors, Toyota aims to improve hybrid vehicle (HV) fuel efficiency by 10% under the Japanese Ministry of Land, Infrastructure, Transport and Tourism’s (MLIT) JC08 test cycle and reduce PCU size by 80% compared to current PCUs with silicon-only power semiconductors.
The technologies behind these SiC power semiconductors were developed jointly by Toyota, Denso Corporation, and Toyota Central R&D Labs., Inc. as part of the results of a broader R&D project in Japan (the R&D Partnership for Future Power Electronics Technology under consignment from the New Energy and Industrial Technology Development Organization). The three had announced their joint development efforts in May 2014.
In the Camry hybrid prototype, Toyota is installing SiC power semiconductors (transistors and diodes) in the PCU’s internal voltage step-up converter and the inverter that controls the motor. Data gathered will include PCU voltage and current as well as driving speeds, driving patterns, and conditions such as outside temperature.
By comparing this information with data from silicon semiconductors currently in use, Toyota will assess the improvement to efficiency achieved by the new SiC power semiconductors. Road testing of the Camry prototype will begin (primarily in Toyota City) in early February 2015, and will continue for about one year.
|Brief presentation on Toyota’s newly developed silicon carbide (SiC) power semiconductor for use in automotive power control units from the May 2014 announcement. Click to enlarge.|
Similarly, on 9 January, Toyota began collecting operating data from a fuel cell bus currently in regular commercial operation in Toyota City. The bus features SiC diodes in the fuel cell voltage step-up converter, which is used to control the voltage of electricity from the fuel cell stack.
Data from testing will be reflected in development, with the goal of putting the new SiC power semiconductors into practical use as soon as possible.
|Camry SiC-equipped test vehicle. Click to enlarge.|
A lot of progress is independent on whether the vehicle is FCEV, PHEV or a BEV.
They use a lot of the same equipment regardless of where the electrons come from.
Recently discussion elsewhere regarding the projected higher range Renault and Zoe cars assumed that to get double the range, double the energy from the battery was needed, so folk were guessing at 48kwh for the pack for the new Leaf in 2016-7.
For instance a large part of VW's plan to increase range for the E-Golf comes from other improvements as well as battery specific energy gains, see page 17 here:
So drive train improvements, aerodynamics and weight are also important.
LG have said that they can make a battery good enough to enable a 200 mile range BEV by 2016 for $35k.
On a trend where batteries are decreasing from perhaps $400kwh in 2014 to $180kwh in 2025, which are the sort of figures often quoted aside from whatever Tesla is up to, then by 2017 we might be on $300 kwh.
For a 48kwh pack that comes to nearly $15k, which is pushing it.
My guess would be that the new Leaf and Zoe will have packs in the area of 40kwh, and the rest of the range aside from the perpetual optimism in such things is from other improvements.
Silicon carbide power components are a part of this sort of improvement.
Incidentally since a 40kwh pack would still cost $12k, that explains Toyota's reluctance to press on with BEVs with present technology.
They see them as most suitable for city cars, and since those start well under $20k then a £12k battery pack does not make them truly economic.
Nissan are hoping that people will be prepared to compromise cross country ability to a degree, and so they can sell cars which are a bit more upmarket and pricier than pure city cars.
Posted by: Davemart | 29 January 2015 at 03:38 AM
Yeah, its all based on compromises and balance of the car. If you tie most of your cost into a battery and make it close to the parity price of ICE you're probably going to be lacking a majority of features, or be lacking range. So good observation.
A PHEV with 16-25kwh battery is a all I'd ever need really. If batteries get to $300 or even $200/kwh I would expect fleet wide adoption for hybrid adoption. I would in most cases be in electric drive, save money on fuel, and have only a small cost premium on the vehicle. Which could be made back rather quickly, rather than a long range EV where the cost premium is very high and it would take much longer to pay back.
Posted by: CheeseEater88 | 29 January 2015 at 06:06 AM
Fair enough, but perhaps it should be noted that PHEV batteries are high power, and the costings are different to those in BEVs.
So $300kwh for BEV batteries does not mean that those in a PHEV will only cost that.
LG Chem have designed a pretty good pack for the next generation Volt though, so we are getting there, and progress is being made on PHEV batteries as well as BEV ones.
BTW, the specific energy increase for both PHEV and BEV batteries seems to be of the order of 80%, taking them up to the same kind of figure as that already employed in the Tesla.
Raising that much is tough indeed, as Musk has noted.
Posted by: Davemart | 29 January 2015 at 06:22 AM
Yes, but if the pack is large enough lets say a 25kwh one, it doesn't have to be that specialized of a chemistry. All you have to have is enough instantaneous energy to drive your motor, more cells found in larger batteries allow for an increased power output using normally less than ideal chemistries for a hybrid application.
16kwh is fairly large compared to the 5kwh batteries we find in older hybrids. Where the chemistry had to cycle rapidly. The larger the pack, the less cycling, and the less power strained they are. So high specific energy batteries could find a home in PHEVs they just have to come down in price a bit more.
I will say, the volt makes a very compelling PHEV, that's at 16kwh, taking it to 25kwh would bring it to leaf EV range, it would also allow the use of cheaper chemistries.
A 20% boost on top of the range would mean 100miles could be attainable in EV driving. If the system could be kept under a $10,000 premium, payback could happen in under 150k miles, if under $15,000 it may take 200K miles.
If they went with a smaller 16kwh pack, they could probably see returns as a little quicker, but as you said, the batteries are of a more expensive chemistry to obtain the needed power output.
Posted by: CheeseEater88 | 29 January 2015 at 06:46 AM
I will say, as soon as payback comes in around 75K miles or less, that is the point of mass adoption of that technology. Most owners want to see the benefit during the time they own their vehicle. (5 years from new is a common occurrence)
So here is hoping!
Posted by: CheeseEater88 | 29 January 2015 at 06:52 AM
Fair enough again.
I suppose I really changed the subject somewhat, to more typical PHEVs, as the Volt is the only one planned AFAIK which is going for such a big pack.
Clearly at that level the chemistry can be much more like that in a BEV.
For much of the world outside of NA the ~22 miles AER of typical PHEVs really is much more optimum for costs.
In the UK for instance the average annual mileage is not much greater than the AER over a year, and that will include some long runs where the AER won't help much on them, even if it were 40 miles or so.
Those figures are slightly misleading, as the average for new cars is considerably higher, but a very big AER is not worth paying for for most people in the world as the petrol engine can simply take over, and costs held down with the smaller pack.
Posted by: Davemart | 29 January 2015 at 07:21 AM
RE: Optimum electric range for PHEVs: With so many tradeoffs to consider, it becomes a complex formula to determine what the optimum battery size for a PHEV is. If reduction in fleet GHG emissions is the goal, smaller packs with lower cost leading to larger market share might justify a shorter all-electric range. By the same token, lowering fuel usage/emissions of popular, higher-consumption vehicles (SUVs, CUVs) would have a greater impact than replacing an already-efficient compact car with a Chevy Volt (where is my CrossVolt??)
Posted by: Nick Lyons | 29 January 2015 at 07:54 AM
Chinese ZEV city car regulations mean that 50km on the NEDC cycle will be where the majority of PHEVs are pitched.
Increasing battery specific energies may give interesting options though.
Mercedes are going to need it just to hit the 50km target, and are confident that they will have the batteries to do it, up from the present ~30km.
VW and Audi are already there with their PHEV designs, which means that when better batteries are available in 2016-7 they certainly could if they choose up the AER to something similar to that of the first generation Volt, perhaps as an option or in North America only.
Anton Wahlman who goes to the big trade events and conferences has been told by VW that they won't up the pack size.
However VW in the past have been less than forthcoming about what they will do, until they darn near have the cars on the forecourt, so I would not rule it out.
Posted by: Davemart | 29 January 2015 at 08:15 AM
For myself and other holdouts 40miles EV is a wonderful number, if I could get an SUV with 40miles EV range and a small efficient(~30mpg highway) gas generator for a $5,000-$8,000 premium over a comparable vehicle. Icing on the cake would be the same $5000-$8,000 premium with a 25kwh battery and ~70+EV miles.(probably will never happen)
As the premium gets lower over time, we'll see more and more manufactures jump on board.
One problem with long range EVs with ICE generators would be if they have too long of range the generator is just lugged around for almost no reason most of the time. The ICE generators do need to run and be maintained every so often, if you can get a good balance between EV range >40 and less than 100 I think there could be massive amounts of EV miles without much compromise or overhead. I think 60miles EV under most conditions would be low enough to have the generator run often enough to heat up completely, and run a full closed loop cycle run its emission monitors, and then shut down. It probably has to clear the gas out once a year or two... So maybe $50 a year in gas for most users, the rest would be electric (around $561)... saving around $700-$800 a year if you look at maintenance, fuel, and other factors. (more as the price of gas goes up, I used $2.50/gal, if it hits over 3.50/gal you could save over $1,250 a year)
So really, if you could get the EV with a range extender for a $5000 premium with 40 mile range it would pay off for most owners quickly, until then, EVs are going to have a slow go in America.
Posted by: CheeseEater88 | 29 January 2015 at 08:24 AM
Another upside of the PHEV is the infrastructure allows for an easier transition to full EVs, so its win win win for everyone. Reduced emissions, reduced fuel costs, a reason to invest in battery research, infrastructure investments in electric charging, and battery production would hit mass production levels, rather than be small niche items.
Posted by: CheeseEater88 | 29 January 2015 at 08:30 AM
VW's idea of a PHEV FCEV means that you would not have the same problems of souring, and the leakage rate from a hydrogen tank is miniscule.
Of course there is not a hydrogen station on every corner, but their design with large batteries topped off with a FCEV for long range would not need many to provide cross country capability.
Another option is a fuel cell RE, which would solve cold weather range reduction problems by both providing cabin heating and keeping the batteries at optimum temperature.
PowerCell is in the late stage of design for a diesel fuel truck fuel cell, for ancillary power, but it would seem that they may be capable of development into a car RE.
Posted by: Davemart | 29 January 2015 at 08:54 AM
Yeah, I just see PHEV ICE vehicles as a stepping stone for H2, basically the type generator could be interchangeable between FC and ICE. Figure every advance that affects EVs FCs and PHEV ICEs.
As batteries get better, it make improvements on any EV type. So any improvements gained benefit almost all of the industry.
Hydrogen FCs and ICEs are going to slug it out for a long time, Renewable liquid fuels are going to keep the ICE competitive for a long time... but yeah, I would love a FC over a ICE,
I see near term(now till 2020) that companies will push PHEVs like the volt into several model ranges. FCs should be gaining a foothold in midsized models, nothing spectacular, a niche.
midterm(2020-2025) I see Hybrids in every Light duty size. H2 Stations should start popping up in large metropolitan areas outside of CA. FCVs transition from niche to mainstream in locations that can support them. Still a low adoption rate all things considered, but gaining market share on pure EVs. It won't pass it but it will get closer.
Long term (2035+) I see a phasing out of most ICEs(only ones left will be in larger vehicles especially those that tow/work), but all will be hybrid.
Most compact/subcompact vehicles will be pure EVs with ranges like that of the Tesla model S today. Mainstream adoption of H2, such that you can travel to most every city without much concern on filling up. FC vehicles will probably have passed up pure EVs at this point. (unless a miracle batter come into existence)
Posted by: CheeseEater88 | 29 January 2015 at 09:39 AM
There are so many options that 'picking winners' simply is not possible at this stage.
Of course lithium air batteries would provide completely new capabilities, but so would on board reformation, or methanol etc fuel cells, as they avoid the need for expensive high pressure storage and provide much greater density energy storage.
My own preference is always to take weight out where possible, as transport is basically about moving mass over distance, so the less mass the better.
So I would love to see through the road, on the move, inductive or magnetic resonant charging, which would avoid the need for either massive battery packs or fuel cells and hydrogen storage.
All the technologies we are developing will have a part to play though, and they are all as big an advance on combustion engines as they were over having the streets buried deep in horse dung.
Posted by: Davemart | 29 January 2015 at 10:00 AM
Nick, I am hoping that a Volt style hybrid comes to a crossover, or SUV like edge/explorer... I don't expect a expedition sized vehicle, but that would be great if offered in PHEV... You are very right in pointing out the diminishing returns from already efficient compacts/subcompacts. (the cost benefits are abysmal too, a larger vehicle would benefit from that 40miles EV much better)Also, its very easy to put a $10,000 system on a large vehicle like a truck and not have it affect the buyers too much(smaller percentage of the purchase price), also the returns on that investment are likely to be much faster than on a compact.
That's why ford put aluminum on its large vehicle, greater return on investment.(that and it was mainly to stay within its weight class and keep its bragging rights.)
Posted by: CheeseEater88 | 29 January 2015 at 10:05 AM
Where did you get the $400/kWh battery price in 2014? Last year, Nissan was offering replacement batteries for $230/kWh and Elon has repeatedly said that prices would be below $200/kWh when the Gigafactory starts up.
Posted by: DaveD | 29 January 2015 at 06:47 PM
To be fair DaveD, Nissan could be displacing some of the cost of the battery. As suggested in this article:
But yes, it is a great sign that Nissan is confident that they could make money even at that low price in the future if it isn't already making money today.
I hope that these low cost batteries mean much larger and capable PHEVs like SUVs, trucks and large sedans all areas that need the added efficiency the most, and where it can be easily integrated because of space and existing MSRP.
Posted by: CheeseEater88 | 29 January 2015 at 08:49 PM
I forgot to mention, I think Davemart was initially thinking smaller capacity PHEVs like those in Fords and Toyota's hybrids... the chemistries are very expensive for the KwHs compared to a Leaf, a Model S, or a Bolt.
Posted by: CheeseEater88 | 29 January 2015 at 08:52 PM
'Cette évolution est liée aux progrès technologiques des constructeurs asiatiques, et se trouve favorisée par la forte baisse des prix des batteries du fait des gains de productivité et d’importantes surcapacités industrielles. Selon Bernstein, le prix moyen des batteries est tombé fin 2014 à 400 dollars par kilowattheure, contre 1.000 en 2010, et devrait encore chuter à 180 dollars en 2025. De quoi permettre aux constructeurs d’augmenter le nombre de modules par pack, sans que la facture s’envole'
I've seen similar figures floated a lot though, and Nissan tell us that the replacement battery pack is selling at a loss.
They presumably figure by the time that they need to provide many of them, then they can get a lot nearer the cost and not loose too much overall.
Posted by: Davemart | 30 January 2015 at 12:20 AM
Nissan made many changes to the Leaf in the 2013 MY, including replacing the aluminum doors and hood with steel; moving the charger to become a part of the front wheel drive module; replacing the electric emergency brake with a floor type pedal e brake; etc., all in all reducing the weight by 170 lbs. Drivers are experiencing about a 5-8% increase in range.that reduces the car to about (3300 - 170) 3,130 lbs...a decided step in the right direction. Further improvements are expected as Nissan vapor so far has spec'd 200 miles of range as a target; perhaps more efficient electronics will be included in the new models.
Posted by: Lad | 30 January 2015 at 12:23 AM
I've just re-read my source, and it is more authoritative than I had thought.
Without wishing to be rude to the lady, it is 'straight from the horse's mouth':
'Béatrice Foucher, directrice du programme électrique de Renault,'
So for Nissan/Renault, $400 kwh in 2014 it is then.
The mid point of around $300 kwh in around 2016-7 in line with a hoped for decrease to $180 kwh in 2025 that they give is of course purely my guesstimate.
Posted by: Davemart | 30 January 2015 at 01:03 AM
While engaging in battery price speculations you might want to remember that Navigant Research reported a few months ago that Tesla was paying Panasonic $180/kWh. Now.
BTW, that is not a misquote in the linked article. I've read the original research paper.
Furthermore it is expected that battery prices will fall another 30% once the Tesla/Panasonic gigafactory opens. That takes price to below $130/kWh by 2017.
There's no magic sauce that gets the Panasonic prices so low, it's volume manufacturing. Economy of scale. Other companies (L G Chem, BYD, for example) are also going bit with major increases in manufacturing capacity. Tesla won't be the only company with access to very affordable batteries.
A 200 mile range EV could likely be built with a 50 kWh battery pack. That's $6,500 at $130/kWh.
What would a best guess be for a Camry engine, transmission along with cooling, fuel and exhaust systems?
Posted by: Bob Wallace | 30 January 2015 at 06:52 AM
Or some of us might prefer it if not every thread was about the magic of Tesla.
Posted by: Davemart | 30 January 2015 at 07:27 AM
Yes, but he's also talking about prices that Panasonic is providing today. If we're going to speculate on battery costs, why assume worst case rather than look at the best case situations that are already occurring?
GM didn't announce the Bolt without some type of hope that by 2017 these numbers were attainable and Nissan is letting it leak that they'll have 200+ mile range on the Leaf in that time frame as well.
These things are not happening if they think they'll be paying $400/kWh in the next two years. To assume those prices seems pretty negative based on directions we're seeing from the people in a position to know.
Posted by: DaveD | 30 January 2015 at 11:24 AM
Meanwhile, a 7-gallon gasoline fuel tank costs around $50.
Using SiC power controller that is significantly downsized from current controller, along with compact solid-state battery that can now be placed under the hood, a future HEV will be a no-compromise vehicle with respect to trunk & luggage space, and will be very practical.
Even with the continued use of bulky NiMh batteries like now, some battery can be put under the hood, while the rest can be put between the front seats, thus sparing the rear trunk for uncompromised luggage space.
Thus, a future Prius with 60-EPA-MPG, or a future Camry hybrid with 50-EPA-MPG, enabled by higher efficiency of SiC controller, and with an uncompromised luggage space, using a $50-7-gallon gasoline tank instead of a $6,500 battery pack, will prove to be the most practical vehicle of all.
Posted by: Roger Pham | 30 January 2015 at 02:27 PM
If indeed, Tesla is paying $180 per kWh for battery right now, then why did other car mfg's not lineup at Panasonic's door to buy the same battery? Hobby-grade Lithium polymer batteries are costing $800-900 per kWh now, all made in China.
But, a 60-EPA-MPG Prius is significant in another way:
Hydrogen engine. Indeed, with direct injection running at ultra-lean mode, a hydrogen engine can be 20% more efficient than current gasoline engine. A 60-mpg Prius can attain 72 mpg using hydrogen.
Thus, the FCEV Mirai can be equipped with Prius' engine and hybrid power train, with 4-kg of H2 in 2 tanks smaller than the 5-kg capacity of the Mirai right now. At 72 mpg, the range would be 288 miles using 4kg of H2. 4kg of H2 can be stored in tanks costing $1,600 instead of in $16,000 of batteries, 80 kWh batteries at $200 per kWh.
Posted by: Roger Pham | 30 January 2015 at 02:50 PM