Toyota and Denso develop SiC power semiconductor for power control units; targeting 10% improvement in hybrid fuel efficiency
20 May 2014
|
|
| Toyota intends to leverage the benefits of high frequency and high efficiency of SiC power semiconductors to enable PCU downsizing of 80%. Click to enlarge. |
Toyota Motor Corporation, in collaboration with Denso Corporation (Denso) and Toyota Central R&D Labs., Inc. (Toyota CRDL), has developed a silicon carbide (SiC) power semiconductor for use in automotive power control units (PCUs). Toyota will begin test driving vehicles fitted with the new PCUs on public roads in Japan within a year.
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.
|
|
| Left:PCU with silicon power semiconductors (Production model). Right:PCU with SiC power semiconductors (Future target). Click to enlarge. |
In hybrids and other vehicles with an electrified powertrain, PCUs supply electrical power from the battery to the motor to control vehicle speed, and also send electricity generated during deceleration to the battery for storage. PCUs contain multiple power semiconductors, which are usually made of silicon; these power semiconductors account for more than 25% of semiconductors used in hybrids, according to Toyota.

|
|
| Above. Role of power semiconductors. During conduction, a portion of current is lost as heat. Loss also occurs when the current switches on and off.The main cause of loss in silicon power semiconductors is due to the “tail” current that keeps flowing briefly after turn-off. By contrast, conduction loss in SiC power semiconductor is small, and the current doesn’t tail during switching. This means loss is small, and higher efficiency can be achieved. Below. Because switching loss is small in SiC power semiconductors, high frequency drive is possible. In the current PCUs, coils and condensers, which temporarily hold electricity, take up 40% of the PCU. The high frequency control of SiC allows the downsizing of coils and condensers within the PCU. Source: Toyota. Click to enlarge. |
|
However, PCUs account for approximately 25% of the total electrical power loss in hybrid electric vehicles, with an estimated 20% of the total loss associated with the power semiconductors alone.
Therefore, a key way to improve fuel efficiency is to improve power semiconductor efficiency, specifically by reducing resistance experienced by the passing current. Due to the importance of power semiconductors, Toyota has been developing them in-house since launching the first-generation Prius gasoline-electric hybrid in 1997, and has focused on increasing their efficiency.
As an example, the Si power semiconductors used in the current third-generation Prius has one-quarter the power loss of those in the first generation.
As SiC enables higher efficiency than silicon alone, Toyota CRDL and Denso began basic research in the 1980s, with Toyota participating from 2007 to jointly develop SiC semiconductors for practical use.
|
|
| Toyota and Denso achieved the high efficiency SiC transistors through adoption of a trench structure. Source: Toyota. Click to enlarge. |
Toyota has installed the jointly developed SiC power semiconductors in PCUs for prototype hybrids, and test driving on test courses has confirmed a fuel efficiency increase exceeding 5% under the JC08 test cycle.
In December last year, Toyota established a clean room for dedicated development of SiC semiconductors at its Hirose Plant, which is a facility for research, development and production of devices such as electronic controllers and semiconductors.
In addition to improved engine and aerodynamic performance, Toyota is positioning high efficiency power semiconductors as a key technology for improving fuel efficiency for hybrids and other vehicles with electrified powertrains. Going forward, Toyota will continue to boost development activities aimed at early implementation of SiC power semiconductors.
Toyota will exhibit the technology at the 2014 Automotive Engineering Exposition, to be held from 21-23 May at the Pacifico Yokohama convention center in Yokohama.
SiCs have been around for a while but high operation temperature, high power SiCs are relatively new.
A 90% reduction in volume and (xx%) reduce in weight is important for all electrified vehicles and special for new aircraft with on-board batteries.
The 5% to 10% gain in HEVs efficiency will be welcomed by all future HEV users. Will it be possible to update current HEVs?
Posted by: HarveyD | 20 May 2014 at 08:01 AM
Two things make me happy when I see Toyota (and their partner/suppliers) chipping away at the boring details of electrification, from HVAC components to coolant pumps to motor control architectures. First, of course, it's uplifting to see all the ancillary things that enable electrification at all levels, from microhybrids to full BEVs. Since they started Project G21 20 years ago – the initiative that would become the Prius – Toyota technical leadership has consistently plowed money into improvements throughout the vehicle that make electrification at a minimum transparent to the driving experience, if not in fact more pleasant.
The second thing that makes me happy is the silence of all those TM conspiracy theorists who insist that everything from TM's decision to stick to NiMH batteries for hybrids to their CA-TX move proves Toyota complicity with Big Oil to prevent the electrification of the automobile. Right... TM's investment in electrification at all levels – especially industrialization – dwarfs anyone else in the industry.
Before the chorus starts up telling me how SiC has been here forever, I know. I still had lots of hair and could stay standing in an amateur boxing ring back when SiC was “just around the corner” for the first time. Only five years ago when I was trying to buy devices of a size for a serious (over 100kW) purpose, they still pretty much came with a data sheet per each manufactured lot. MHI, who actually manufacture the devices, has only in the past few years built an integrated motor/controller set for long-term road testing. Denso does not have a habit of announcing something that doesn't end up in production for half-a-decade, so I look forward to seeing some SiC in a nearby Prius/Camry/Avalon very soon.
Posted by: Herman | 20 May 2014 at 08:17 AM
@Herman:
We have been happy exclusive Toyota users (10+ Toyotas in our family) for the last 20+ years but with the exclusion of its HEVs, Toyota has often been slow to adapt to new technologies.
Toyota seems to be a strong believer of mini step by step improvements for their product line and is very slow to change what has been working well and is often one or two steps behind others.
Consequently, Toyota's units have less early design change problems and continues to build very reliable units. We keep our Toyotas for an average of 15 years with no major failures.
Posted by: HarveyD | 20 May 2014 at 08:45 AM
Harvey, if you are in manufacturing, forgive me for sounding pedantic. You cannot go into a cutting-edge manufacturer -- very high yield, constant cost improvement, ever-reduced material use and footprint -- and not be surrounded with the tools of the Toyota Production System (TPS). TM is utterly dedicated to the rhythmic, invariable, and constantly improving production of anything they make. It doesn't necessarily mean they aren't making new technologies happen, but if it isn't productionized and reliable in the field they won't do it. When the Hirose power electronics line first opened there was truly nothing like it; you can see from this article they plan to do this yet again.
EVERYTHING that makes TM electrification work, from the economical production of small motor/controllers for HVC and Power Steering to the main drive inverters, is applicable to the BEV.
(Note to the correct critics who will point out Prius-related flaws that resulted in major recalls. All TRUE. But when you look into the nature of their failures, you can see that it was not because TPS didn't work --- it's because core principles were not applied.)
That Toyota can profitably manufacture a profoundly reliable automobile with all of the mechanical and electrical content of the Prius is still impressive. It is a generational feat. I believe (but do not know any more than anyone else) that they are not confident this can be done in large-scale non-NiMH batteries soon enough to commit to the technology. However, remember that G21 was known only to a very small inner circle of TM leadership for years. Are they are doing something similar now with batteries? Quite possibly.
Posted by: Herman | 20 May 2014 at 10:18 AM
The most interesting tidbit that's not mentioned here is something Toyota should be crowing about from the rooftops: I believe they developed this tech for the Le Mans LMP1 entry! They have Denso's name all over that car and you can beat your sweet hiney that's not a coincidence.
Right now, they look like favorites to win at Le Mans this year so maybe they're waiting to do some big marketing around that event???
Regardless, THIS is why I love racing. It drives tech forward at a much faster pace than business as usual could ever do.
Posted by: DaveD | 20 May 2014 at 11:42 AM
Sigh...oh for an edit button. I mean to say "you can bet your sweet hiney" Hey, I was trying to be polite and not say the bad words LOL
Posted by: DaveD | 20 May 2014 at 11:52 AM
SiC is better known as carborundum, still a champion for resisting heat stress and metallic crystal phase changes over rapidly varying temperatures. Think cutting tools. In the worst scenario, the carbon will probably precipitate out as droplets of 'solvent' (which is why wood saws are dirty -- the carbon lubricates)) or the crystals will shear and twin to close defects. There must be some very sophisticated chemistry to explore, as how the n or p holes are made, the extent of purity required, etc. The tetrahedral lattices of Sic put them in a league with zeolites, which were I think the most underrrated chemical engineering achievement in the last 40 years. Modernization of the refinery business would not have occurred without them.
Are we maybe too fascinated in lithium and graphite, and missing where the big research money is going? And which countries' money it is?
Posted by: kalendjay | 20 May 2014 at 04:10 PM
Herman, I'm very pro-Toyota but I have to admit that we had to do with 3 and 4-speed auto transmissions while most others had 6 and 7 speed units etc.
BNW will be using light weight materials soon but Toyota will wait another 5 years or so. The same can be said about improved batteries etc.
Same applies to updated designs, Toyota is know to be one generation behind etc
Posted by: HarveyD | 20 May 2014 at 04:11 PM
I think we're missing another big point here: This could allow EVs to increase their range by 5-10% without adding more batteries. It also reduces cooling system needs and overall weight of the system.
Smaller, lighter, more aerodynamic for the overall vehicle (less cooling needs...hence less air intake needed), simpler system design, extends range. What's not to love? OK, the price :-)
But hopefully that will come down at some point to be relevant.
Posted by: DaveD | 20 May 2014 at 06:13 PM
@Harvey,
Toyota's fuel cell tech and FCV are the most advanced in the world, soon to come, and soon, Toyota's SiC tech for inverter will top the world again.
The Prius, the world's most practical and the best-selling EV's, has high aluminum content, including the feather-weight hood, the brakes, and the front structure etc..
When one considers that a FCV is a BEV using a 10-10-0 battery technology (>10x the energy density, <1/10th the cost per kWh of storage, and available NOW), PLUS Toyota's significant lead in HEV market penetration, then Toyota is quite ahead of the pack.
Posted by: Roger Pham | 22 May 2014 at 01:25 AM
Roger,
I know you like FCVs, but I think you're a bit over the top there, mate. LOL
Show me which fuel cell you're going to buy and what it's cost is. I'd love to see that comparison with a battery pack. :-)
Posted by: DaveD | 22 May 2014 at 11:15 AM
DaveD....FCs may make economic sense for extended range electrified vehicles. (at least until 5-5-5 batteries are available) The new Hyundai FCEV @$499/month including free H2 is a good very first example.
I'm very pro-Toyota but Hyundai (and many others) have been moving faster with technology and design changes in the last 5 years or so. To remain in the number one place, Toyota will have to move faster on post-NiMH batteries, BEVs, FCEVs, light weight bodies, etc. VW is getting very close and may take the lead in 2015 or so.
Posted by: HarveyD | 23 May 2014 at 09:00 AM
Harvey,
I don't judge any of these things based on some lease price (BEVs OR FCVs). I want to see them for sale, not gimmicks to make them look attractive for early adopters.
I can buy a BEV. I can buy my own electric motor and batteries and it's easy to find the price if I want to build my own. Show me where I can buy an FCV or any of the FC components and what the price is.
Until you do that, don't come telling me about comparing prices :-)
Posted by: DaveD | 23 May 2014 at 12:05 PM
Harvey,
I don't judge any of these things based on some lease price (BEVs OR FCVs). I want to see them for sale, not gimmicks to make them look attractive for early adopters.
I can buy a BEV. I can buy my own electric motor and batteries and it's easy to find the price if I want to build my own. Show me where I can buy an FCV or any of the FC components and what the price is.
Until someone can do that, they can't talk about comparing prices :-)
Posted by: DaveD | 23 May 2014 at 12:23 PM
Harvey,
I don't judge any of these things based on some lease price (BEVs OR FCVs). I want to see them for sale, not gimmicks to make them look attractive for early adopters.
I can buy a BEV. I can buy my own electric motor and batteries and it's easy to find the price if I want to build my own. Show me where I can buy an FCV or any of the FC components and what the price is.
Until someone can do that, they can't talk about comparing prices :-)
Posted by: DaveD | 23 May 2014 at 12:24 PM
Harvey,
I don't judge any of these things based on some lease price (BEVs OR FCVs). I want to see them for sale, not gimmicks to make them look attractive for early adopters.
I can buy a BEV. I can buy my own electric motor and batteries and it's easy to find the price if I want to build my own. Show me where I can buy an FCV or any of the FC components and what the price is.
Until someone can do that, they can't talk about comparing prices :-)
Posted by: DaveD | 23 May 2014 at 12:36 PM
@DaveD,
"Be patient, grasshopper..."
Quoting from:
http://www.japantimes.co.jp/news/2014/05/22/business/corporate-business/toyota-cheers-plan-u-s-hydrogen-stations/#.U4Ay2tJdX_o
"LOS ANGELES – Toyota Motor Corp.’s North American chief said California’s plan for more than 50 hydrogen fuel stations within two years is making executives feel “bullish” about the outlook for its Camry-sized fuel cell car due in 2015...
Just based on the product, based on the infrastructure we see that will be in place in California, we’re much more bullish on fuel cell,” Lentz said. “I’m more bullish on fuel cell than I was on EV. I think the cost of fuel cells is going to be substantially less than in an EV,” he said, referring to electric vehicles.
Toyota, which is also providing at least $7.2 million to a California startup to operate hydrogen stations, has said fuel cell autos will be an easier zero-emissions option for many consumers than battery-only vehicles owing to range, performance and refueling time matching gasoline. Honda Motor Co. has said it will offer a revamped hydrogen sedan in the state in 2015, and Hyundai Motor Co. will lease a fuel cell version of its Tucson sport utility vehicle, the first of which arrived in the state Tuesday."
Posted by: Roger Pham | 23 May 2014 at 10:54 PM
LOL That's all well and good Roger. But it's still smoke and mirrors.
We'll see when we get a price for purchasing the fuel cell vehicle or the component parts. Until then, it's all feel good press releases: The world is happy. Unicorns love us all. Elves are going to bake us cookies and brownies are going to repair our shoes while we sleep. :)
Hey, maybe it will, maybe it won't. But right now it's just lots of opinions and hopes. At least we don't have long to wait and find out the realities now!
Posted by: DaveD | 24 May 2014 at 03:21 PM
David...I don't know about your area but around here cars monthly rental fees are much the same from one manufacturer to another, i.e about 0.9% to 1% of the vehicle total value.
Normally, a $500/month rented vehicle is worth about $45K o $55K. There is a definite relationship between rental fees and total value.
Posted by: HarveyD | 26 May 2014 at 05:23 PM
Harvey, not even close on new tech a company is trying to push. Or do you want to tell me that the $600 leases on the FCX Clarity were actually covering the cost of those vehicles? LOL
Posted by: DaveD | 28 May 2014 at 02:44 PM
Harvey,
By the way, please remember that a Honda spokesman said that the FCX Clarity cost between $1-$2 million and wouldn't get more specific. Yet that $600 lease was supposed to represent a real world relationship to purchasing price? :)
No, it didn't. I'm not criticizing the FCX or FCV's in general. I'm just pointing out that these early programs have no way to defray their costs with volumes so they pick a lease price for differing reasons and those prices have ZERO relationship with the actual cost of the car.
So I'm still back to my original statement: Show me where I can BUY an FCV or it's major components, and we'll start comparing prices to BEVs. Until then, we're all talking about Rainbows vs Unicorns.
Posted by: DaveD | 28 May 2014 at 02:48 PM
One last thing and I'll shut up. But in the interest of fairness, the same issues were true of the EV1 when GM leased them back in the 90's. There was ZERO correlation between the actual cost of those vehicles and what they leased them for.
As a BEV fan boy, I WILL point out an interesting fact: So...the FCVs are now where the BEVs were in the 1990's!
I'm sorry, I just could NOT restrain myself! :)
Posted by: DaveD | 28 May 2014 at 02:54 PM
@DaveD
Manufacturers have two choices when selling or renting new products:
1. Keep the price low and limit production to limit losses. (that's what many electrified vehicle manufacturers are doing)
2. Try to get close to the real initial high cost and rely on future mass production cost reduction to lower sale prices. (That's what Nissan, Toyota and a few others did with their early electrified vehicles)
Eventually, marketing method 2) may prevail, specially from manufacturers who are seriously aiming at electrified vehicles mass production.
Tesla is a separate limited production case story. Their marketing may change with their new Model E, X etc supposedly aiming for the mass markets.
Posted by: HarveyD | 02 June 2014 at 10:31 AM