By on January 30, 2015

Toyota SiC Power Control Unit 01

Toyota unveiled its plans Wednesday to trial a new hybrid system using silicon carbide power semiconductors that could find its way into hybrids and EVs.

The trial will compare the new silicon carbide semiconductors with silicon units currently found in many a hybrid’s, FCV’s and EV’s power control unit, which are linked to a 20 percent loss in overall electric power. The aim is to increase powertrain efficiency by mitigating said losses through the new semiconductors.

The test subjects will be a Camry hybrid prototype and a fuel-cell bus. The bus trial began in early January, when Toyota started data collection on a setup featuring SiC diodes in the FCV’s voltage step-up converter. Meanwhile, the Camry — whose SiC components are in both the hybrid’s voltage step-up and inverter — will hit the road for a year beginning early next month. Both tests are being carried out in Toyota City, Japan.

Toyota hopes to have the SiC technology — developed in a partnership with Denso Corporation and Toyota Central R&D Labs, Inc. — ready for practical use as soon as possible.

Get the latest TTAC e-Newsletter!

Recommended

21 Comments on “Toyota Unveils Silicon Carbide Semiconductor Trial...”


  • avatar
    schmitt trigger

    Silicon Carbide (SiC) has been the holy grail of power electronics for over a decade now.
    The material itself is a significantly better semiconductor material than plain old Silicon, which is the mainstay of electronics.

    Having said that, SiC device manufacturing is much more difficult than Si itself, and the issues in the recent past have been related to microscopic defects.

    But I’m old enough to remember the early transistor radios that used Germanium. Ge was inferior to Silicon, which at the time its usage was mostly limited to military projects.

    Then manufacturing breakthroughs occurred, prices dropped, availability surged, and the rest as they say, is history.

  • avatar
    SCE to AUX

    Maybe Toyota can use this in their EV program. Or not.

  • avatar
    bumpy ii

    The big thing here is that if SiC conductors can significantly reduce the switching losses, then full-time series hybrids become the most practical and efficient configuration.

    • 0 avatar
      YellowDuck

      bumpy could you explain this in more detail? Very interested. A series hybrid is one where the ICE only ever drives a generator, to charge the battery, correct? How does this SiC technology make that arrangement more efficient? I don’t think I understand what “switching losses” are.

      • 0 avatar
        SCE to AUX

        “Switching” power supplies convert direct current (DC) to alternating current (AC), or vice-versa. They do this by rapidly switching the power on and off, with corresponding changes in polarity.

        Such a device is needed for DC batteries to power an EV motor, which is almost always AC.

        Switchers are generally very efficient, but even 5-10% losses hurt efficiency, and increase cooling requirements.

        • 0 avatar
          wumpus

          I was going to say: switching really doesn’t seem like a big place for losses, but:

          1. Getting 5-10% means really sweating the efficiency. You should be able to get that while charging (both from mains and from an extender engine), but getting that out on the drive motor[s] is another story. Still wildly better than what an IC does.

          2. Those 10% losses really add up when you realize that you lose them going from the charger to the battery, then the battery to the motor. Note that I’m pretty sure the battery isn’t all that efficient, but supercaps (for hybrids or burst power) pretty much are 100% efficient.

      • 0 avatar
        shaker

        The existing silicon FET transistor switches are extremely low resistance (.005 Ohms or so) when fully switched on, and are passing DC current, but when switched on and off rapidly, they spend a long time in a “resistive” state, which wastes power. The SiC material spends much less time acting resistive, so less power (heat) is generated in the device itself, and more delivered to the motor.
        As an added benefit, the SiC device can operate efficiently at higher temperatures, thus the devices can be smaller, and less heat-dissipation is needed, (resulting in a more compact inverter design).
        These increases in efficiency can lead to smaller battery packs, better regenerative energy recovery, more range etc.

        This tech, combined with supercapacitors and the next generation of battery tech will make EV’s even more viable for everyday use.

        Now, if we can only make our power grid a sustainable, efficient way to feed the next generation of EV’s… ?

  • avatar
    PrincipalDan

    The picture (I don’t know if it is the angle or the lighting) but the picture makes it look like a modern take on the mythical 100 mpg carburetor.

    • 0 avatar
      ckb

      The difference is that SiC has decades of research and testing that indicates what performance gains should be realized vs the urban legends supporting a 100mpg carburetor.

  • avatar
    YellowDuck

    Okay, another stupid question. There are energy losses converting AC from the grid to DC for battery storage, then again in converting DC to AC to run the motor (and presumably also when converting AC back to DC for storage from regenerative braking). Why not just use a DC motor?

    Thanks for educating me here.

    • 0 avatar
      Felis Concolor

      The big problem with DC power is its frightening ability to jump gaps and create its own short circuit even when fuses and breakers blow. This is why high current DC fuses are filled with inhibiting gel, which flows between the broken contacts to prevent runaway conditions. AC can employ lots of tricks with voltage, frequency and amperage to provide safe delivery of very high power over very thin wires, and its self-snuffing nature means open circuits of a fraction of an inch are in a safe condition.

      I’m not a DC electrician, but if there are any out there who can explain with greater clarity the whys and hows of AC vs DC power, I’m certain many here would appreciate it.

      And speaking anecdotally, there’s a damn good reason the world runs on Tesla’s alternating currents despite Edison’s push to have direct current become the standard.

    • 0 avatar
      bumpy ii

      You can use DC motors (and a lot of homebrew electric car conversions do), but manufacturers prefer AC motors for a variety of reasons: wider powerband, more sophisticated motor control, less maintenance, and perhaps most importantly AC controllers fail off (the motor just stops), while DC controllers fail on (the motor runs away at full speed).

  • avatar
    HerrKaLeun

    the missing (and crucial) part of thsi story is: what do those do in a car? Electricity goes from battery to motor and these are somewhere in between doing what? Like voltage regulators or inverters?

    • 0 avatar
      Scoutdude

      In short they are the “switches” that turn the power to the motor on and off.

    • 0 avatar
      schmitt trigger

      If you apply the full battery voltage, it will go “full throttle”. The DC motor controllers allow you to regulate the voltage and thus the speed of the motor.

      DC motors however, use brushes which wear out and cause sparks that are an EMI nightmare. Moreover the heavy current is carried by the rotor, which is difficult to cool in a sealed motor. Machining the rotor piece adds significant cost.

      Nevertheless, DC motors dominated the traction-motor segment for over 100 years for the simple fact that batteries only store DC current.

      On the other hand, AC motors are brushless. No EMI to worry about. The high current windings are on the stator (outside) which may be easily cooled.
      But AC cannot be stored, and the control algorithm is significantly more complex, requiring voltage AND frequency changes.
      The AC motor controller converts the relatively constant DC voltage from the battery, into a variable frequency and voltage three-phase power.

      Enter microcontrollers and Silicon semiconductors.
      Microchips have now the brainpower to properly calculate the voltage and frequency vectors to smoothly control a motor across its speed/load range, which includes -very importantly- the regenerative range.

      But microchips are only that, micro. They cannot, by themselves control tens of thousands of watts. They need the help of power semiconductor, in the form of MOSFETs or better yet IGBTs.
      These power semiconductors have traditionally been Silicon, and the article explains that they will now use Silicon Carbide, which holds the promise to yield improved efficiencies.

      • 0 avatar
        wumpus

        This hasn’t quite been true since the 20th century.

        “DC motors” are brushless and have been for quite some time. They really aren’t DC, but are hit with bursts of power at the appropriate time. In many ways, the circuit looks a lot like a [classic] spark plug circuit. The power fills an inductor, the switch closes, and the power goes into the motor (instead of the sparkplug). Brushless motors tend to be cheaper than inductor motors (the types in nearly all volume [for volumes of Tesla or higher] electric cars) but generally don’t allow charging while breaking. Ideally, the Tesla “3” series will have a big brushless in the back with a current “D” motor in the front (for braking).

        “AC motors” (induction motors) are pretty much the same as 20th century AC motors. The difference is that modern motor controlers generate the AC on the fly so the frequency of the AC matches the speed of the motor (this is done both for efficiency and to get more powerful motors to start at all. I think a classic “directly connected to mains” AC motor was limited to ~30hp, and that was with 3-phase (optimized for motors)).

  • avatar
    CarnotCycle

    I was not aware how high losses could get in DC/AC conversion in mobile units. Given the power going through them, the implication of 20% loss translates to kilowatts of heat needing dissipation from a pretty small component – those things must get hot.

  • avatar
    honda_lawn_art

    Dr. Hibbert: ‘Was it carbon based, or silicone based?
    Homer: ‘The second one, xylophone.’

  • avatar
    jdash1972

    In an industrial application a “drive” is used to operate a squirrel cage induction motor at different speeds. The same motor connected to normal 60Hz AC power will run at a speed determined by the number of poles the motor was built with. 7200 / number of poles = motor rpm. Most motors are 3600, 1800 and 1200, slower motors with more poles are generally less common. An AC variable frequency drive takes the incoming 60Hz AC power and rectifies and filters it using a capacitor bank, to a single phase DC bus. Transistors are used to reconvert this dC power to AC of any desired frequency. You observe a voltage to frequency ratio when driving the motor so the operational rpm range of the motor is limited. This is exactly what’s being done in a Tesla, except they start with DC from a battery. Those power transistors, sometimes called IGBT’s or insulated gate bipolar transistors, rapidly switch the power to synthesize a pure sine wave at a frequency to achieve the desire motor rpm. Every step in his conversion has losses, just as the transistors have a voltage drop across their junctions and dissipate this power as heat. So big heat sink and whatever cleaver means to keep he electronics cool. Medium voltage drives have huge losses, the air conditioner is as large as the drive itself in most cases. This is what I do for a living.

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • Scoutdude: I have not seen #2 yet. Looks interesting but it doesn’t really address the total problem. What...
  • APaGttH: Almost $30K for a Corolla if you tick all the boxes…and no more grunt under the hood. $26K to enter...
  • Michael S6: 29 k is GTI money
  • EBFlex: Why does anyone reply to EBFlex? He’s obnoxious, aggressively ignorant, and has never added an ounce of...
  • ToolGuy: Oh hey Peter. “GM is awesome, and Cadillac represents GM at its finest.” Agree or disagree?

New Car Research

Get a Free Dealer Quote

Who We Are

  • Matthew Guy
  • Timothy Cain
  • Adam Tonge
  • Bozi Tatarevic
  • Chris Tonn
  • Corey Lewis
  • Mark Baruth
  • Ronnie Schreiber