ON Semiconductor Corporation (ON) Earnings Call Transcript & Summary

Good morning, everyone, and welcome to today's webinar on design considerations of silicon carbide-based solid-state circuit breakers. I'm Kris Costers with onsemi, and I will be your moderator today. In today's webinar, you will learn more about the advantages onsemi's EliteSiC MOSFETs, gate driver design and UL and IEC safety standard requirements for SSCB applications. At the end of the webinar, we will be holding a live Q&A session to answer any questions you may have. [Operator Instructions] Now let's meet to this presenter. Prasad Paruchuri is a Technical Marketing Team Lead of the industrial PSG business unit at onsemi, where he is driving innovative EliteSiC family products for industrial applications. Prasad brings over 25 years of semiconductor experience in areas that span from application development and system architecting to technical marketing and strategy development. Now without further ado, let's start our webinar.

My name is Prasad Paruchuri, PSG Technical Marketing Team Lead. Today, we are going to present a solid-state circuit breaker applications based on silicon carbide MOSFETs.

My name is Hunter Freberg, and I'm a PSG Technical Marketing Engineer, and I'll be covering some of the silicon carbide portfolio that can be utilized for solid-state circuit breaker applications.

We will just describe you the circuit breakers in general. As last 10 or 20 years or even 50 years back, we would then use the mechanical circuit breakers. And to get the -- for high-performance applications, new electronic circuit breakers are used by some of the customers. And there are many advantages of solid-state circuit breakers. Some of them are intelligent networking, hard clash mitigation. The response time is 10x faster than the mechanical breakers and dynamic current adjustment or programmable current adjustment, and then you can load shed certain particular loads using intelligent network. So this slide shows you a block diagram of the solid-state circle breaker. So the main blocks on these solid-state circuit breakers are input circuit where you have air gap. And if you are doing a current sensing through a current transformer and if you are using a GFCI breaker function also that control IC for the GFCI breaker, then power switches are connected, back-to-back connection like a common source configuration. For a 20 amps breaker, we may need like a 6x discrete backwash now this particular device. And we can also use the simple gate driver using NCD57080. And then for intelligence purpose, we are implementing RSL15, which is a Bluetooth device for communication purposes, and also, it had a microcontroller, which can be reused for some control functions for the circuit break application. And the bottom block is basically RW power supply to drive the gate drivers. And we need to understand why the silicon carbide FETs are suitable for solid-state circuit breaker applications. The main advantages of using the silicon carbide FETs are what are the major functions needed for the solid-state circuit breaker application we need to understand. The conduction losses are key design parameters in the SSCB application, and the conduction losses determine the operating junction temperature of these power switches. Silicon carbide technology will allow to operate maximum junction temperature of 174 degrees centigrade. Theoretically, the die of the silicon carbide FET can work even 200 degrees 300 degrees. But when we use the discrete components, packaging materially is the limitation of that maximum operating temperature. Then another advantage of the silicon carbide FETs are RDSon ratio from case temperature of -- at 25 degrees to the case temperature of 125 degrees is around 1.5x. And the other one is silicon carbide FETs will provide low RDSon from 650 volts to 170 volts range. And these devices are commercially available from onsemi and multiple other vendors, too. Silicon carbide FETs also allow parallel devices and share the current equally. This will help you to -- by paralleling multiple devices, we can get low RDSon as like a 2-millivolt 3-millivolt type of devices. And correspondingly, we can reduce the conduction losses. For high-current applications, for example, greater than 80 amps, we may need silicon carbide waste modules. In relation to the silicon carbide FET technology, silicon carbide JFET technology also suitable for solid-state circuit breaker application, but we need to add a cascode silicon FET to implement in this solid-state circuit breaker application. For medium-voltage solid-state circuit breaker application, for example, 2 kV to 20-kilovolt devices, these modules are also good target, and silicon carbide JFETs are also -- may be useful in future applications. Now we described about -- before going to the design calculations after selection of the silicon carbide FETs for this application, we've done the simulation. And in the simulation, what we've done is we paralleled the couple of devices like we are back-to-back connected with a common source and then paralleled multiple after. In this example, we just shown the 2 switches are parallel, but we may need more switches parallel to get to the low RDSon. And one of the key parameters of the [ securities RG ] also because in high-frequency [ routine ] application, we normally use [ RG ] in the range 1 ohm, 2 ohm or 3 ohm. But in case of solid-state circuit breaker applications, we need a high [ RG ] value, maybe 4.7 ohms minimum, in some cases, 10 ohm, 22 ohms. There are other customers using even 40-some ohms. It depends upon the maximum turnoff time they would like to allow. In relation to the solid-state switches, we also need to add [ DAS ] number. [ DAS ] number circuit consists of [ TBS ] and also [indiscernible] and also an MOV. In some cases, to reduce the voltage spikes across the individual switches, we also need to add [ RC ] snubbers. Then we also done a simulation of whether the gate driver voltage of 0 to 18 volts is good enough for this type of application. And we've also done a simulation for that. And planar silicon carbide FETs have threshold voltage of minimum 1.8 and max maybe 2.8 volts around that, maybe up to 3 volts. And also, the circuit need to withstand high-peak currents like 5 kiloamps during the short-circuit condition. And the tripping current, what we've done the simulation is for 280 amps, and of course, the gate voltage -- minimum gate voltage is 0 volts. And we've done the complete simulation with 0 to 8 volts and also incorporated some random noise on the gate signal and see whether it will malfunction or it is okay to work on it. This is the schematic we used to do the simulations. So this is back-to-back switches, and this is a gateway circuit and then the gateway [ RG ] value. In this case, we just chose 10-ohm devices. And externally to control it, we also have an extra [ RG ] also incorporated. Then the 0 to 18 volts gate drive source is provided to turn on and turn off the gate signal. Then again, we do have the sample circuit here. And based on the simulation, we observed that 0 to 18 volts is good enough for this application. We don't need negative gate voltage because negative gate voltage is [indiscernible] topology, where you have high side and low side and you need to avoid the cost conduction. While in this particular application, as both switches are on at the same time, like in the past 2 half cycle, the tough FET will be turned on and due to the native cycle, the work on FET will be turned down and then corresponding [indiscernible] diodes are to complete the path. And here, we are also describing you 2 major UL standards and the safety standards required for solid-state circuit breaker applications. And these are the UL 498 and UL 1077 are the major safety standards we need to meet for the U.S. market. Then for ICE, the European market, we also need IEC 60898 and [ actively ] 1458 needed. And then there's other U.S. standard also available. I don't go very detail into these standards. And thanks to [ Professor ] Hopkins, who actually prepared this slide as part of his presentation at [indiscernible] effect tutorials, and I took the slide because this is more a standards related, nothing to do with onsemi. So we took the information from Dr. Hopkins' presentation. Then regarding the UL safety standards. These are the 2 main standards, the UL 489 and UL 1077. The main difference between these 2 standards are physical differences. They are required pole spacing. And the minimum pole spacing required is like 1 inch under 8 and 2 inch for the over surface. Similarly for the UL 1077, we have 3/18-inch and 0.5 inch. Then similarly, minimal terminal spacing, you have 0.5 inch for 130 volts and 1 inch for 600 volts, [ 3/8 ]. For commercial application, 3-way 32 inch up to 300 volts and 1/4-inch terminals. Their maximum temperature rates at their terminals are 50 degrees to 120 degrees Fahrenheit. Then similarly, overload condition, they will do the 50 cycles at 6x the rated current are 150 amps minimum, whichever is higher. Then similarly, for UL 1077, we do have 50 cycles with 1.5x rated current. Then we have the endurance test, and we also have the short-circuit test. The short-circuit test is one of the critical factor because it needs to withstand 5,000 amps peak current during the short-circuit rating. So the short circuit is limited by the wires, cables connected to the breaker. And this slide is also we have taken from Dr. Hopkins' presentation. So from onsemi, we have 2 different solutions available based on the output current requirement and the voltage. And onsemi have a discrete solution targeting for 25-amp applications or maybe 40-amp applications. And onsemi developed different simulation schematic and work with multiple customers to use discrete FETs in the 25 amps application. And this is the block back on what we have. And in one of the applications, we paralleled 3 or 4 switches for each switch location. So for back to back, we need 6 to 8 devices. So the total RDSon with 12 milliohms -- 15 milliohms max and 12 milliohms typical. So with that, we are able to get 6 milliohms time-to-time condition. And similarly, if you're looking for a current above [ 80 amps ], you may need a module. And onsemi developed the module for test market, whether any customers are interested on it. And on the next slide -- in the couple of slides later, we will show you this module, what -- how it performed and what is RDSon and those type of things. And we implemented a F2 package module, and we implement the 20-milliohm, first-generation FETs in this technology because these samples we developed a couple of years back. And we are waiting for some customers who have an interest to use these type of modules. Then we can qualify the module and go into production if there is any interest coming from the customers. Solid-state circuit breaker simulation and design guidelines. And we try to do the simulation using M2 technology, 650 volts and M3S 350 volts silicon carbide FETs for 120-volts AC application. And we are measured on the -- using M2 technology parts in the 20-amps application. Then based on that, we calculate losses and simulations. And then we repeated the same for M3S technology base on 12 milliohms and 8 milliohms. So these are the 3 part numbers, what we have, NTBG015N65SC1, which is in production today, the typical RDSon is 12.43 milliohm. And here, the [ 015 ] is the max value we've shown. In general, most of the FETs, we normally show the typical value, but this particular device, somehow it is designated as 15. But the typical RDSon value is 12.43. And then [ RTH junction ] case is a 0.30 K per watt as per the data sheet. Then we created a simulation models for the new M3S technology for 12 milliohms and 8 milliohms. And RDSon typically is 14.77 milliohms, and this is a high side for the model. We have -- don't have the die still not available to test and what accuracy we will get it. Based on that, we will modify those values. And also, we have 8-milliohm part. We got [ 9.81 ] milliohm typical and junction to case is 0.24 K per watt here. Then these are the RDSon variation with respect to the 12 volts, 15 volts and 18 volts. So onsemi recommends to drive the gateway voltage at 18 volts to get the RDSon. Then similarly, for RDSon, with respect to junction temperature, you can see here this is for M2 technology. And you can see from 25 degrees to 125 degrees, you are having like 1.1 to 2x. So these are the design calculations for the 20 amps breaker, where we use the 4 switches, 4 FETs for the 8-switch location. Then this is a junction capacity -- junction to case thermal resistance for this device. And then what we measured in the application is these are double measurements. The case [ small ] temperature observed is 83.8 degrees centigrade and ambient temperature measured is 59.6, which are roughly 60 degrees. And that is the required by our customer to have the maximum operating case temperatures. And the RDSon at 12.4 milliohms and then when we pass the current through the 5 amps because these are 20 amps and then 4 devices are parallel, so each device carries 5 amps. Then the power loss calculation for each device, each is RDSon multiplied by [ I square ], and that will give you 0.31 watts. Then we calculate the [ R, theta ] case to ambient. And then based on this calculation, we'll get you 78.1 K per watt. Then from this, we will calculate junction temperature. So this is a form of junction temperature. And based on that, we will have the junction temperature of each device is around 83.9 degrees C. But based on the PCB layout and the location of these devices, maybe a slight variation of the temperature of the junction may be possible. But -- so the layout is very critical when -- especially when we are using the discrete design. And for -- we also simulated with the same M2 device based on the PLECS models. And these -- for the single switch, what will be the power losses, we calculated here. And based on that, we use the VGS 18 volts and get resistance of 10 ohm, and we got the temperature -- the T -- ambient T is 59.6 degrees centigrade. Then for continuous current of 20 amps and we simulated for all these 3 devices with the models. The first one is we have the measurement and then also simulated it. For the second and third devices, these are simulation only for now. So based on the simulations, what we observed is for M2 technology-based devices when we use for device in parallel per switch location, ambient temperature is on the bottom one -- on the top. And then case temperature is in the middle, and the junction temperature is in the bottom. So you have 15.6 degree ambient, case temperature is 83.35, and the junction temperature [ 83.43 ]. So the case temperature and the junction temperature are very close. One of the reason is power loss in this application is very small. So there is not much difference between the case and junction temperatures. While similarly when we implement M3S technology, the 12 milliohms, and both of the typical values are very close to 12 to 14 milliohms. And but we observed that the temperature of the KAC is slightly higher. That is one of the reason is M3S RDSon variation with respect to high temperature is slightly higher on the ratio compared to the M2 technology. And similarly, when we replaced 12 milliohm part with 8 milliohm, we will have even higher temperature, which is like [ 98.63 ] degrees, and the junction temperature is also very close. So in all these 3 applications, we see the maximum junction temperature is well below 100 degrees, with ambient temperature -- operating temperature close to 60 degrees. So as TJ max of these silicon carbide FETs are 175 degrees. So these are well within the limits. Even if you consider the PCB layout [indiscernible] temperatures, we want to keep it with 130 degrees. So even then, we have -- 30% margin is available. So all these 3-part numbers can be used. M2 MOSFET technology and the M3S 650 volts technology of the products. And the comparison is on the D2PAK MOSFETs, the 7-lead version. And the M3 second generation have a TJ of 83.4 centigrade and a power loss of 2.48 watts. And we measured and compared with the simulation, and then they are both very close to each other while for M3S 12 milliohm and M3S 8 milliohm parts, we have done the simulations only. And based on that, we will see that TJ max we observed is 90 degrees, 95 C for the 12-milliohm part and 98 degrees -- 98.69 degrees centigrade for 8-milliohm part. And if you can see the below the RDSon of the various devices for NTBG015M65SC1, it is a 12.43 milliohm. And while the similar 12-milliohm part from M3S technology is 14.77. And you can see the RDSon at 125 degrees is a little bit higher. And similarly, for 8 milliohm, we have the 9.81 milliohm. And we have paralleled the 4 devices with 12-milliohm device for M2 technology and also similarly for M3S technology. While for 8 milliohm part of M3S, we only paralleled 3 devices per switch location. Here, explaining the 4 FETs and [ 4 FETs ] scheme. So the current to the each device in the 20-amps application for here is a 5 amps and with the same RDSon for 12 milliohm. M3S also is a 5 amps, while in case of the 8 milliohms, each device has 6.66 amps. And then power loss for each device, these are 0.31 watt and 0.37 watt and 0.435 watts. And the power loss total are also listed here. And as the power loss is so small, the junction temperature is very close to the case temperature. The next one, we are talking about how we can reduce the transient voltage suppression when we use these bidirectional switches. Transient suppression design is a critical part of the solid-state circuit breaker design. The key is to have sufficient energy storage, anticipating elements so that the line becomes deenergized quickly and the halt and supply are isolated. The entire voltage across the supply becomes the voltage across the solid-state circuit breaker. This is just prior to isolator activation. Suppression can be broadly divided into different layers. Snubber circuits, that means [ axis ] number across [indiscernible] each switch. Then transient voltage suppressors like TVSs, MOVs, zener diodes and transorbs. And spark gaps like -- think neon tube light bulbs. These are the commonly used approaches of suppression. And on this slide, also, we have taken the information from presentation from Dr. Hopkins. He presented this presentation during [ FI ] conference. And the way the -- how the [ axis ] number and MOV circuit works. With MOVs, solid-state circuit breaker design is more complicated, and MOV ratings are concept SSCB design. As the switch opens, the energy is rooted into the snubber. And the snubber voltage exceeds the SSCB rated voltage, [ are past ] voltage rating, the MOV begins dropping energy. As the highly nonlinear MOV drastically drops resistance and drops the majority of energy, the snubber maintains nearly constant voltage. When the MOV [ big voltage ] begins declining, the snubber capacity charges into MOV until the MOV recovers to a high resistance. So the isolator opens after -- and the snubber completes to be charging 3 bleed resistor circuit. The final design is about cost of energy capacity. The snubber energy capacity offsets MOV energy capacity. The design requests and optimization of snubber and TVS materials for the particular application, our product profile that fits the [indiscernible] that is isolated [indiscernible]. I'm not going to add any slides related to isolated [indiscernible] and how you choose those things because of time limitation, but Dr. Hopkins presented on these particular things also in detail. Maybe you can go through the -- his presentation as needed. Now I will hand over the presentation to Hunter.

Thank you, Prasad. So as I said at the beginning, I'll essentially just be giving a brief overview of our EliteSiC technology from the discrete side and then a little bit of a road map for some of the devices that we have seen to release. So just kind of a brief overview for some of the applications driving the silicon carbide MOSFET and diode market in addition to things like solid-state circuit breakers, on the industrial side, you have solar inverters, UPSs, energy storage systems, EV charging. On the automotive side, you have things like traction inverters and onboard chargers, anything that's high voltage, high power encompassed into electric vehicles. The advantages for onsemi and our EliteSiC technology in our approach largely boil down to 3 different components. It's going to be our proven quality, our robust planar designs and our manufacturing. So whether that's through the in-process control and burden, defect scanning and 100% avalanche testing of all dies during the manufacturing process, which results in no drift in our threshold or parameters and a high reliability gate oxide, or to our manufacturing approach, our vertically integrated manufacturing, we either call powder to products or substrate to systems, we control every part of the supply chain. We have best-in-class design tools, our physical and scalable, accurate simulation models, a wealth of application notes and design guides as well as a new online system design tool that you could see some of the prior information that was presented on -- we utilize that tool for. And then lastly, just our innovative technologies. Our third-generation is our latest silicon carbide technologies for our devices that are available, again, in automotive and industrial grade, have a wide offering and standard and custom power products for modules. And we have a large portfolio of voltages and RDSons available for [ TO-247-3L and 4L ] packages. So again, for our vertically integrated manufacturing, what we call substrate to systems, we go all the way from the boule manufacturing itself into the fab, epitaxial process and manufacturing cleaner devices, have planar available today for our whole portfolio, but we will be releasing trench devices eventually as well as migrating from 6-inch to 8-inch wafers. We do in-house packaging of our devices and our modules, and again, a full portfolio of diodes and MOSFETs and die-only and metal options and, again, in automotive and industrial qualified devices. For the modules, a full portfolio of hybrid that would be IGBT with silicon carbide diodes or full silicon carbide modules having silicon carbide MOSFETs and silicon carbide diodes in a number of different package configurations, whether single or dual cooling and direct and indirect. And again, deep applications and system know-how for automotive and industrial applications. So to speak a little bit more on our silicon carbide boule manufacturing. We have outstanding [ sic ] crystal growth technology, which is, again, a key material for next-generation semiconductors that significantly improve efficiency in electric vehicles, EV charging and energy infrastructure and, again, our solid-state circuit breakers. We utilize GTAT, GT Advanced Technologies, which was founded in 1994 and headquartered in Hudson, New Hampshire. They have some additional offices located elsewhere and 119 employees globally. Since March of 2020, we've been partnered with GTAT. However, as of the end of 2021, we have fully acquired GTAT and, therefore, are a fully vertically integrated company, again, from powder all the way to our end products. So for the silicon carbide MOSFET portfolio, for the discretes that we have that could be utilized in the solid-state circuit breaker applications, we have our 650-volt silicon carbide FETs and our M2 family available in RDSons ranging from 15 to 95 milliohms and TO-247 3-lead, 4-lead for our through-hole devices as well as a D2PAK 7-lead and a [ toll and a power A8 ] for the Surface Mount devices. We will be releasing 650-volt silicon carbide MOSFETs in our M3 generation technology. As you can see, those will range from 8 to 32 milliohms with a 18-volt gate dry voltage. So much improved RDSon with our third-generation technology and, again, available on TO-247 3-lead, 4-lead for the through-hole and Surface Mount D2PAK 7-lead. We will also be releasing TO leadless devices in 16, 23 and 32 milliohm RDSon as well as the top cool Surface Mount DPAK devices. We have 900-volt devices available in our M2 family. Those are in 20 milliohm and 60 million RDSon values. And again, in TO-247 3-lead, 4-lead and the Surface Mount D2PAK 7-lead. And for our 950-volt silicon carbide MOSFETs for our M3 family, we'll be migrating to our third generation. Those will be available, ranging from 11 milliohms to 45 milliohms RDSon in the same package as mentioned previously. For our 1,200-volt devices, we have devices available from our very first generation, our M1 family, ranging from 20 milliohms to 160 milliohms, again at a TO-247 3-lead, 4-lead, Surface Mount D2PAK 7-lead, and we also have die-options available in that generation. For our third generation, again, same packages as previously, ranging from 14 milliohms to 65 milliohms RDSon. We will be releasing -- or we are sampling 1,200-volt DPAK devices at the moment in 65 and 40 milliohms. And later, we'll be sampling 30 and 22. And we will eventually have M3 die to replace our M1 die at the moment. And lastly, for our discretes, we have our 1,700-volt first-generation devices in a 28-milliohm RDSon and a 1-ohm or 960-milliohm RDSon values. So for the performance comparison of our 1,200-volt M3 devices, you can see on this slide between our first generation comparable or similar device, 20 milliohms to 22. With our third generation and as well as a 21-milliohm device from 1 of our lead competitors, their latest third generation, you can see from M1 to M3, we have substantial improvements. So I'll draw your attention to the figures of merit, the figure of merit 1, which is our RDSon by our gate charge. And our figure of merit 2, which is our RDSon by our energy of our output capacitor. These are 31% and 44% lower, respectively. You can also see, if you look a little bit lower, the improvements in switching losses. And while it's not explicitly stated, you can see that the difference between the values from our competitor A to our third generation, we are quite a bit better than one of our lead competitors for 1,200-volt devices. For our 650-volt third-generation silicon carbide MOSFETs, we have some data to share on those devices. So you can see us compared to our 650-volt M2 devices. In the first column, our M3S 650-volt devices in the second and then 3 of our lead competitors as well in their respective 650-volt similar RDSon devices. You can see again the figures of merit that we mentioned previously. Two big ones are going to be our RDSon by our gate charge again and our RDSon by the energy output capacitor, quite a bit lower than both our second-generation 650-volt devices as compared to our third generation as well as a lot of our competitors. And I'll hand it back to Prasad to cover some of the silicon carbide MOSFET modules that we have available for these solid-state circuit breaker applications.

Yes. onsemi have multiple modules available using the gel-filled modules these are. And we have F1 module, F2 molecule and F5 modules. Similarly, onsemi have Q0, Q1, Q2 module. These 3 modules are comparable with some of the configuration like a half bridge, full bridge, not 6-pack devices as the solid-state circuit breaker require a special configuration and I'm not giving any part numbers or anything of these modules to you. But onsemi have the capability to do your custom modules. And to target a high-current application like 3-phase AC input and current rating in the range of 200 amps applications, maybe based on the new technologies, the current can be extended to maybe 300 to 400 amps, too. And onsemi developed a module in F2 package with 16 die. And we used the [ bolder ] M1 technology because these samples have been like a couple of years back based on the request from one of the customer for their R&D purposes. And this is the layout we used in that design like common [indiscernible] input, output. Then you have a common source configuration. And there are 8 die in parallel for each switch location. And with the 20 milliohm, we are able to get 8 die so the 20 by 8, close to like 2.5 milliohms RDSon. And for time to time, you will have like a 5 ohms. Per medium voltage 2-kilovolt solid-state circuit breaker applications, we can consider using onsemi's Q2 pin module. And this module contains 3 -- 2 to 3 milliohms 1,200-volt silicon carbide MOSFETs. There are multiple dies parallel to achieve the 2 to 3 milliohms per switch. So the resulting module is 3.6 kV and 6 to 9 milliohms RDSon based on rated temperate. We can connect the source of the last 2 FETs for common source configuration. So the midpoint connections of each module are not needed for this design. However, the R6 number circuit can be implemented across these things as needed. To develop a higher-voltage module, onsemi developing a 2 kV product using M3S technology, and we will see the 2-, 3-kilovolt devices available in 2024. By implementing the 2 kV type of devices, we can able to develop a 6.6 kV devices, or by using the existing module after Q2, we can [ reconnect ] 3 modules on the left side to get to 10-kV device. And the other 3 modules on the right side of this picture, you will get back-to-back 10-kV devices. Totally, you need 6 module to configure as 10 kilovolts SSCB application. Then we will get into the gate drivers, how you can implement the gate drivers for these applications. As we are using a common source configuration, one channel -- single-channel gate driver is good enough to drive the volt [indiscernible], but you can use a nonisolated gate driver theoretically because it is with respect to the source. But for practical conditions and to avoid the any noise and those things, we recommend to use isolated gate driver. As you are looking for the higher voltage [ lines ], that will help you to use isolated gate drivers. onsemi have isolated gate drivers with a 5 kV isolation and functional voltage of up to 1,300 volts. And onsemi have also 2 channel and other multiple intelligent, high-performance drivers, too. So these are the single-channel gate drivers. What we have 57000 and 57001, both are -- these are in production today. Then a version of the device, 57100 is also going to be released in the couple of months, probably Q4. The main difference with the device the VCC max of capability up to 30 volts. Then we do have very simple devices like 8-pin die gate drivers. If you want to use it, like a 57080 is one of the examples we can use. And especially, we can use a split gate driver for these type of applications. Then we do have the 2 channel gate drivers here. And then this is the one other driver we will recommend, the 57100 are the existing product, 57000 which is in production in the last couple of years, while the 57100 will be released end of the year. And we are sampling as we speak in coming months. And this device also have a high kV. It's 10 voltage, and the working voltage of 1,400 volts, and it has a very high CMTI and the propagation delay of 66 nanocoulombs. And for the over current protection, based on the circuit, you can use a DSAT function here. In this example, if you use this gate driver, or alternatively, you can use the current sense in between the source pins and you measure the current sense -- current through that -- through a current sense amplifier and then feed into your microcontroller RSL15, which is a Bluetooth device. Then we do also have a new development, which is -- with more production functions involved and those devices is in development as we speak. And major advantage of this function is it had all the functions we've talked about in the previous device. All those are available. In relation to that, there is a bidirectional communication for temperature protection is also available. There are 2 current sensors, and you can sense the temperatures and send the -- based on the current limit, you can shut down the gate driver and send the false signal to the secondary side. And you can also send analog [indiscernible] signal corresponding to the junction temperature of these 2 [indiscernible]. And one of the major advantage of these driver is as in the solid-state circuit breaker, we are paddling multiple devices. And then the spread of this area occupied by these multiple devices are high, so there is a possibility we can incorporate 2 different NTC devices at the 2 locations of the PC board or the module, and we can able to utilize that temperature information also feed into the microcontroller to do any protection or any other functions we want to implement it. These are the various portions of the same device. And I don't go into the details on that. Then there is also a simple device is also available from our other division that those are [ NCP501152 ], and then there is another device [ NCP501157 ], and these devices also can be used. And there are a couple of slides, which describe you other onsemi parts available for these designs. And we have limited time. So that's one of the reasons we would like to -- I will go very fast on these slides. And one of the device, what we have is the NCS37014. This is a GFCI controller, and a GFCI function is more implemented in most of the solid-state circuit breaker applications. Even though self-test is not needed, but this device is a latest device and it have very low portion current, and we recommend to use this device for your SSCB application with GFCI function. And this is a simple block background for a mechanical circuit breakers where GFCI breaker is implemented. And we are producing this device in very high volume, in tens of millions or 20 million units to 1 particular customer and then a couple of other customers, too. Then this is a simple schematic for self-test GFCI breaker application. Now we also need an onsemi power supply for these gate drivers to power the gate drivers. So onsemi can recommend it, but nonisolated but configuration to generate from 120 volts AC input or 230 volts AC input to a 15 volts or 18 volts gate drive requirements of this application. And we have 2 different part numbers available here, and you can develop like 5 to 10 watts type of power supply with this design. Then we also have another part with similar to the previous portion, but this particular device in relation to the main [indiscernible] with 18 volts, you also have LDO integrated. The LDO integrated have a programmable 3.3 and 5 volts and can deliver up to 50 milliamps and this is good -- 100 amps, sorry, 100 milliamps can develop this LDO, which can power the RSL15 Bluetooth device and the microcontroller part of the section. And this is the block diagram of the TL output [indiscernible] power switcher, and these are the [indiscernible] boards are also available for this design. With this, I will conclude my presentation. We can have questions and answers for the next 10 to 15 minutes. Thank you.