Could Bidirectional Switches Become an Industry Standard?
Bidirectional switches offer advantages over conventional power switches and could become standard in the industry.
For well over a century, engineers and scientists have been developing more advanced semiconductor power switches aimed at improving the functioning of all kinds of devices. Conventional power switches have generally been unidirectional, thus requiring at least two separate switches to turn those devices on or off and block voltage or current in either or both directions.
Bidirectional switches can be used in applications like electric vehicles, renewable energy generation, and energy storage, among others. Image used courtesy of Adobe Stock
Use Cases for Bidirectional Switches
New advancements in technology have enabled the development of bidirectional switches capable of handling the on and off positions in both directions. These bidirectional switches offer several advantages over conventional power switches, such as cost savings, improved efficiency, reduced conduction losses, and the ability to create smaller devices.
Those advantages are only the tip of the iceberg when it comes to what will be possible as bidirectional switches become a mainstay in the industry. In particular, some of the strongest use cases for bidirectional switches are in applications like electric vehicles, renewable energy generation, energy storage, solid-state circuit breakers, and motor drives.
Before discussing the specific technical aspects of bidirectional switches, it helps to understand some comparisons with the most common conventional power switches. Bidirectional switches may be seen as the next logical step in the evolution of power semiconductor switches.
Image used courtesy of Adobe Stock
The Evolution of Power Semiconductor Switches
Of course, the two most commonly used types of power semiconductor switches today are the metal oxide semiconductor field effect transistor (MOSFET) and the insulated gate bipolar transistor (IGBT).
The MOSFET combines a resistor and a diode, the two previous steps in the evolution of power semiconductors. It also adds a switch that chooses between the resistor and diode modes of operation. When the switch is open, the MOSFET functions as a diode by blocking voltage in one direction and conducting as a diode in the other.
When the switch is closed, the diode is bypassed, and the MOSFET functions as a resistor. When the semiconductor acts as a resistor, it can turn on and off rapidly, with the only limit imposed by the speed of the switch. The switch in a MOSFET is built into the device's surface and is voltage-controlled.
As a resistor, the MOSFET's voltage drop is determined by the doping level and thickness of the P- section. Heavier doping and less thickness provide a lower voltage drop. However, as a diode, the maximum voltage the MOSFET can block with the switch open while acting like a reverse-biased diode is limited by the doping level and thickness of the device.
Boosting the doping to reduce the resistance leads to a reduced ability to block voltage. As a result, the MOSFET's resistance and efficiency are related to its ability to block voltage. MOSFETs capable of blocking higher voltages also have higher on resistance.
On the other hand, an IGBT structure includes an additional doping layer at the bottom of the device, which has a significant impact on its behavior. When the IGBT's switch is closed, the device's resistance drops, enabling it to conduct much higher levels of current with reduced voltage drop.
Essentially, the device conducts like a forward-biased diode although the diode junction imposes a minimum voltage drop of roughly 0.7 volts for silicon devices. An additional voltage drop occurs with the switch, which again is built into the device's surface.
IGBTs capable of high voltage levels can conduct much higher currents at lower voltage drops than MOSFETs with comparable voltage ratings. However, these higher currents come with a cost. High-voltage IGBTs turn off much more slowly than MOSFETs do.
IGBTs come with an inherent conflict between low on-state voltage drops and the turnoff time. Longer turn-off times result in higher switching losses. As a result, continued development of new IGBTs focuses on minimizing the trade-off between conduction losses and switching losses.
However, newer bidirectional switches aim to solve all the problems highlighted above with MOSFETs, IGBTs, and other conventional power solutions. For example, Ideal Power's B-TRAN (bi-directional bi-polar junction transistor) has the same three-layer structure as the IGBT, except that it has a control switch on each side.
Bidirectional switches are capable of blocking both positive and negative voltage and conducting current in both directions. In the case of B-TRAN, it can also be used in unidirectional applications like voltage source inverters or battery chargers.
MOSFETs and IGBTs are purely unidirectional power semiconductor devices and can't be used as bidirectional switches. Using these conventional power semiconductor devices in bidirectional applications requires either two MOSFETs or two IGBTs and two diodes, which must be connected in a common-emitter configuration, quadrupling the number of parts needed for bidirectional power converters.
However, bidirectional switches can take the place of both devices and the diodes due to its bidirectional design, requiring one-fourth the number of components. They also cut the number of high-voltage switches required in a circuit in half.
B-TRAN bidirectional switches. Image used courtesy of Ideal Power
Benefits of Bidirectional Switches
While the most obvious advantage of bidirectional switches is reducing the number of parts required, that's only the beginning. Of course, using fewer parts means lower costs involved in building any application requiring bidirectional power switching.
Additionally, using fewer parts enables engineers to reduce the size of whatever it is they are designing. Over the years, a key trend in electronics design is shrinking the size of devices, and implementing bidirectional switches that require fewer components enables the next step in shrinking device size.
However, the benefits of bidirectional power switches go beyond the basics of cost reduction and smaller device sizes. In a recent whitepaper, Ideal Power explained the more technical benefits that can be enjoyed by engineers who utilize bidirectional power switches in place of conventional switching options. While the findings apply specifically to the company's B-TRAN applications with solid-state circuit breakers, it's possible that similar benefits could be accessed in future bidirectional power switching solutions from other manufacturers.
For example, circuit breakers paired with B-TRAN devices were found to offer forward voltage drop and conduction characteristics that significantly reduce power loss. Compared to a bidirectional IGBT solid-state circuit breaker module, B-TRAN reduced the voltage drop more than fourfold. An IGBT-based bidirectional switch recorded a voltage drop of 2.75 volts when the load current was positive, versus a drop of 0.6 volts for the B-TRAN breaker.
The B-TRAN switch also reduced the amount of power loss. At a load current of 200 amperes, the switch lost about 150 watts, versus 1100 watts and 1500 watts for two other comparable solutions.
In a matrix converter, B-TRAN reduces the number of required switching devices to nine, versus 18 silicon carbide MOSFETs, silicon MOSFETs, or reverse-blocking IGBTs or 36 devices utilizing IGBTs with fast diodes. B-TRAN reduces the amount of power lost versus those conventional solutions by 72% to 78%.
For example, in a three-phase 300-ampere load, IGBT-based bidirectional switches lose 7000 to 9000 watts of power, while the B-TRAN-based system loses about 1900 watts of power.
The Future of Switching Solutions
Thomas Edison invented the first circuit breaker in the late 1800s, while John Henry Holmes was the first to use a switch on a mechanism in 1884. Of course, we've come a long way since then. Of course, as with every form of technology, scientists and engineers are always working toward improvements.
Bidirectional switches and their applications with solid-state circuit breakers seem to be the next natural steps in the evolution of this critical electrical component. With so many benefits to be enjoyed from bidirectional switches over traditional power switching solutions, it seems likely that bidirectional solutions could one day become standard in the world of engineering design.
In fact, some of today’s most advanced technologies may only be able to move forward when bidirectional switches become standard. For example, power semiconductors account for about 20% of the total electric power losses in hybrid EVs and potentially even larger losses in electric vehicles. Replacing conventional switching solutions with bidirectional switches could boost EV ranges by 7% to 10%, depending on the type of switch used and the vehicle it’s applied to.
Conventional switching solutions result in significantly larger power losses in applications like EV chargers. Today, AC-DC converters using IGBT or silicon MOSFET switches are standard in EV chargers. However, bidirectional switches could reduce power losses by 50%, give or take, depending on the type of switches used.
Implementing bidirectional switches in EV chargers could reduce the amount of time it takes to charge a vehicle. This becomes even more critical as bidirectional charging becomes more commonplace. Bidirectional charging allows vehicle owners to earn a little money by sending some of their vehicle’s excess power back to the grid or by using that excess power in their home.
Renewable energy solutions could also benefit from bidirectional switches. For example, standard IGBT-based designs for the inverters used in solar panels tend to offer up to 97% efficiency. However, bidirectional switches could increase these efficiencies to 99%, thus generating more usable electricity at lower costs and resulting in reduced thermal management costs and potentially less expensive inverter designs.
Energy storage solutions can also benefit from bidirectional switches. As energy costs rise, energy storage solutions are also becoming increasingly popular among homeowners with solar panels or other renewable energy sources for their homes. Implementing bidirectional switches in these energy storage solutions would boost the value of storage systems.
Other newer technologies that could benefit from bidirectional switches are those found in IT infrastructure and any other use cases that involve the use of large amounts of electricity. By reducing conduction losses, bidirectional switches could result in sizable savings on electricity costs. In the case of data centers, wasted energy also increases the complexity of the cooling system required to keep them from overheating.
Ultimately, any technology requiring voltage and current conduction/blocking in both directions could benefit from bidirectional switches, so it may be only a matter of time before they become standard.
Bidirectional switches are already an industry standard and widely-used in many applications. Today they are implemented with a pair of FETs back-to-back. Monolithic lateral GaN HEMTs can also have dual-gates, making them a nearly ideal AC switch. The big benefit, besides very low losses, is the cost is lower than competing Silicon FETs, because GaN uses the same merged drain-structure with twin gates, making the die-size far smaller than discrete two-transistor solution. As the GaN AC switch becomes more widely available, it will replace back-to-back FETs or IGBTs with a higher-performance, lower-cost transistor.