In recent years, fire accidents caused by battery charging have occurred frequently in China, and the difficulty of industry regulation has increased. Poor quality batteries, modified batteries, and future plans for lithium battery recycling and reuse are facing great challenges.The country has clearly stipulated by law that regardless of the size of the battery capacity (including common specifications such as 48V/20A/960W), it is prohibited to park or charge in areas of high-rise civil buildings (height>24 meters). With the breakthrough of artificial intelligence technology, home robots are able to perceive, understand, and respond to environmental changes more intelligently, and the gradual integration of robots into households is the trend. The only source of power for home service robots is the "high rate and large capacity" lithium battery. Taking Tesla Optimus as an example, it is equipped with a 2300W lithium battery with a range of 5 hours. To cope with the safety management of battery charging and discharging in this application scenario, higher requirements and technical challenges are proposed compared to the safety standards of new energy vehicle batteries. New material technologies are needed to adapt to the product ecology of the artificial intelligence era and develop more advanced specialized AI ASIC chips to meet this scenario.

The four models of gallium nitride (GaN) ASIC intelligent fast charging chips (including CT-3602, CT1020, CT1007, and CT-1901) released by Zhongke Wireless Semiconductor this time are customized fast charging chips for robots through the differential design of "GaN ASIC chip+GaN power transistor" integration. The chip includes four specifications of 1000W, 200W, 100W, and 65W, and has a built-in fast charging protocol. In response to the real-time monitoring of battery status in the 1000W overcharging scenario, it forms a full cycle charging closed-loop monitoring and management of the electronic migration process through artificial intelligence algorithm technology and high sampling rate ADC monitoring. It is a specialized intelligent overcharging ASIC chip solution designed for high-power and high-capacity battery charging application scenarios such as home robots, industrial robots, medical robots, etc. It monitors the voltage, current, battery thermal runaway, overcharge protection, current mutation, and poor quality batteries during the battery charging process to reduce the risk of electronic migration caused by abnormal currents and the disconnection response under extreme working conditions, in order to reduce the charging safety of "high rate and large capacity batteries" in the home environment.

图1.CT-3602, CT1020, CT1007 and CT-1901 series fast charging chips

The intelligent charging chip technology of this robot has obtained multiple invention patents.

Technical advantages：

(CT-3602, CT1020, CT1007, and CT-1901) "Gallium Nitride ASIC Chip+GaN HEMT" design, built-in overcharging protocol, battery monitoring and management technology, and ADC high-speed sampling monitoring technology, real-time monitoring and management of electronic migration risks during battery charging process, accurately meeting the core requirements of embodied robots for safe charging in home environments.

1. Gallium Nitride Encapsulation: ASIC chip+GaN HEMT bonding, high frequency and efficiency, module volume reduced by 30%, low internal resistance characteristics reduce conduction loss, LLC design structure system efficiency reaches over 98%, charging heat reduced by 25%, perfectly adapting to the dual requirements of space compactness and low heat generation of robots.

2. Ultra low power consumption: Extend standby "silent time"

The starting current is only 4 μ A, and it automatically enters Burst Mode (pulse mode) when under light/no load. The standby power consumption is as low as 0.3mW. This feature significantly reduces the energy consumption of the robot in non working state, extends the "silent standby time", and reduces the need for frequent energy replenishment, especially suitable for scenarios such as inspection and warehousing that require long-term standby.

3. EMI optimization: ensuring stable operation of sensors

By built-in frequency jitter (cycle 4ms ± 8%), 300ns front edge blanking (eliminating switch glitches), and synchronous ramp compensation (enhancing CCM stability), the EMI peak is 10dB μ V lower than similar chips to extend battery life.

4. AI protection: Full scene charging protection

Integrate cycle by cycle OCP (peak current limit), VCC overvoltage protection (OVP), overcurrent protection, current surge protection, undervoltage lockout (UVLO), output short circuit protection, over temperature protection (OTP, triggered at 150 ℃/released at 120 ℃) soft start function and multimodal artificial intelligence technology to monitor the charging closed-loop formed during the battery migration process and perform self inspection on the battery after sudden disconnection. The detection data is uploaded to the client APP in real time through the protocol interface, and human intervention is given to sudden stories in the charging state in real time.

Joining hands with Zhongke Semiconductor to inject "chip" power into embodied robots

Provided a full scene power system chip solution for embodied robots, from local to whole machine. For detailed specifications, sample applications, or technical support, please visit the official website of Zhongke Semiconductor[ http://www.ctunite.com/ ]Or contact the technical service team：

Øemail：sales1@ctunite.com

Øaddress：321 Jin Building, Design Industry Park, No. 3838 Nanshan Avenue, Nanshan District, Shenzhen