Application of 8-bit microcontroller in SoC

How does the 8051 keep up with today's performance requirements?

The earliest 8051 was introduced by Intel in 1980. It allowed each instruction to be executed across 12 clock cycles, minimizing the need for hardware resources. Ten years later, Dallas Semiconductor ( today's Maxim) improved this architecture with a new design that removed redundant bus actions, allowing typical instructions to take only four clock cycles; they also introduced a compatible microcontroller that could directly replace the 8051 , instantly increasing the performance of existing systems by three times.

Silicon Laboratories ' 8 -bit microcontroller family uses a new proprietary design to implement the 8051 instruction set architecture that provides maximum instruction throughput while maintaining full object code compatibility , resulting in the C8051F CPU being a " hard wired" implementation rather than the original micro -coded design. The new design maps the instruction set onto a two-stage basic processing pipeline to increase throughput while maintaining an 8- bit program memory width. This approach has resulted in a family of new devices that can execute the vast majority of instructions in 1 or 2 clock cycles, outperforming the original 8051 design by 20 to 25 times. With this performance, engineers can support applications that would otherwise require a more expensive 16- or 32- bit microcontroller using a lower-cost 8- bit product.

What functions can be integrated into microcontrollers today without sacrificing performance?

In almost all electronic systems, engineers want to integrate the most functionality into the smallest space. This approach has many well-known advantages, including reduced part count, fewer inventory components, lower manufacturing costs, and potentially higher performance and reliability. Therefore, when evaluating the microcontroller selected for a specific application, it is important to consider these advantages from the perspective of the total solution cost, not just the price of a single part.

On-chip flash memory has become increasingly popular in recent years as prices have fallen; all but the most price-sensitive, high-volume applications now consider it worth paying a small price for the greater flexibility over one-time-programmable devices.

In-system debug is also a desired feature. Since they can eliminate the expensive emulators that were previously required, the application cost of new components will be reduced, and designers will be freer to choose the most suitable components for specific applications. Although 32- bit microcontrollers and digital signal processors have widely provided in-system debug functions, 8 -bit products rarely provided in-system debug functions before Silicon Laboratories launched the C8051F component series . Precision oscillators, analog-to-digital converters, and digital-to-analog converters are the most commonly required analog parts in the system. Temperature measurement functions, voltage references, and comparators are also commonly used. In terms of digital peripherals, the most commonly used standards for external communication include UART and SPI , I2C , USB , and CAN interfaces. In addition, functions such as timers and programmable counter arrays are added.

Integrating ADCs and DACs into microcontrollers often results in performance degradation compared to solutions using discrete components, particularly in terms of linearity and signal-to-noise ratio for analog functions; even so, there are new components that are now as good as solutions using best-in-class components or other products with built-in 16- bit, 1 MSPS ADCs.

As mentioned earlier, 8- bit microcontrollers are now able to provide peak outputs of up to 100 MIPS . Figure 1 is a functional circuit diagram of such components. It is the C8051F120 provided by Silicon Laboratories , which has 128 kbytes of built-in flash memory,

8.25 kbytes RAM , 12 -bit ADC, 12 -bit DAC and various digital peripherals including UART , SPI bus, I2C bus, timer modules and external memory interface. This component is designed to support computationally intensive mixed-signal embedded applications that require high-performance ADCs and DACs. It also provides in-circuit debugging capabilities.

Figure 1 : The integration of analog and digital functions allows 8- bit microcontrollers to provide functionality close to that of a system-on-a-chip

Typical Application: High-Speed ​​Battery Charger

After considering existing products, we will describe an application in which a mixed-signal microcontroller can provide the advantages we discussed in the high-speed charging circuit for lithium-ion batteries. [page]

Batteries must be charged quickly, safely, and efficiently. The best charging method usually includes three stages: a low current regulation stage to minimize early self healing and avoid premature termination of the charging process; a constant current charging stage to provide most of the power; and a constant voltage stage/end of charging, which is usually the longest stage. During the charging process, some of the electrical energy is converted into heat. When the battery reaches full power, all the energy sent to it will become heat, which can cause danger and harm, so the battery temperature must be monitored to avoid damage.

To determine whether the battery is fully charged, most lithium-ion chargers keep the battery voltage fixed and monitor the minimum current value. Figure 2 shows the charging curve, which is the most effective curve obtained by implementing a buck converter. This buck converter is a switching regulator that uses an inductor or transformer as an energy storage device and then sends energy from the input to the output in the form of individual packets. The feedback loop regulates the energy transfer through the transistor to maintain a constant voltage or current within the load range of the circuit.

Figure 2 : Charging curve of lithium-ion battery

The appropriate charging circuit can be designed using the 8- bit microcontroller C8051F300 and its built-in analog-to-digital converter, flash memory, pulse width modulator, temperature sensor, and accurate clock circuit. Figure 3 shows the block diagram of this type of charger. The built-in high-speed analog-to-digital converter on the chip provides accurate charging voltage monitoring function to avoid overcharging and provide maximum charging effectiveness and battery life. The built-in comparator and PWM on the chip provide the necessary functions for the implementation of high-speed buck converters, so that they only require a very small external inductor. The built-in temperature sensor on the chip provides a stable driving voltage that can be used to determine the battery temperature; if necessary, it can also use an external resistive temperature sensing component. Finally, this microcontroller also provides configuration and program setting functions, which enables it to support different types of batteries, helping customers reduce parts inventory and speed up the time to market for new products.

Figure 3 : Functional block diagram of a Li-ion battery charger

in conclusion

8 -bit microcontrollers are still quite active. Due to the increasingly rich analog and digital peripheral integration, 8 -bit microcontrollers can now provide functions close to system-on-chip for many common applications. When these components are based on the 8051 architecture, engineers will feel very familiar with them, making design and development work simpler, faster and less expensive.