Benefits of Digital Control for Power Conversion

Last modified by Microchip on 2024/01/23 16:53

The control of Switch Mode Power Supplies (SMPSs) has traditionally been done in the analog domain. In recent years, low-cost, high-performance Digital Signal Controllers (DSCs), such as Microchip's dsPIC33 DSC have opened many opportunities for designing digital power supplies. There are many benefits of implementing digital techniques into power conversion applications. The implementation of complex power conversion topologies, such as resonant and quasi-resonant converters. Digital control fully supports these advanced topologies, including phase-shifted full-bridge, and LLC-resonant converters to achieve very high efficiency and power density. As a result, digital control provides many options to optimize the efficiency of power supplies over the entire range of operation.

Because many analog controller blocks can be integrated into the dsPIC® Digital Signal Controller (DSC), bill of materials (BOM) costs and system complexity can be reduced while increasing system performance.

analog vs digital power supply graphic

Figure 1

By using advanced and sophisticated control algorithms, it's now possible to dynamically change the control parameters and adapt the system behavior to the input and load changes. The possibility of dynamically switching on and off phases in multi-phase systems is only one benefit among many others we'll discuss in the next few sections. Many key components of a power supply can now be consolidated into one chip. Figure 1 shows the differences between analog and digital power supply components.

Back to top

Increased Efficiency

The dsPIC Digital Signal Controller (DSC) has the advantage of having on-board high-performance peripherals, such as the Pulse Width Modulation (PWM) generator, the Analog-to-Digital (ADC) and the comparator, and the high-performance dsp engine. The combination of these features allows power supply engineers to implement many digital algorithms into a single dsPIC Digital Signal Controller (DSC). As a result, system performance is increased with higher power efficiency.

The power dissipated in a switch is the product of the voltage and the current flowing through it when open or closed. If the switch is operated when the current is zero, the dissipated power will also be zero. Any traditional control loops that are done in the analog domain are easily implemented in the digital world. The higher-efficiency digital power supply reduces the size of inductors, capacitors, transformers, and heat sinks, resulting in reduced costs.

Digital power supplies provide tremendous versatility in optimizing efficiency at multiple operating points. For the PFC boost converter, switching losses can be reduced at lighter loads by operating the converter at a lower switching frequency. Due to the lighter load, the magnetics will still be able to handle the lower switching frequencies. If an interleaved PFC converter is implemented, one phase can be turned OFF at light loads. Similarly, for a phase-shifted full-bridge converter, extra switching losses can be eliminated at light loads by turning OFF switching of the synchronous MOSFETs and using the body diodes instead. Another example is in a buck converter application. Synchronous buck converters are typically preferred for high current outputs. However, the use of the synchronous MOSFET causes circulating currents at light loads, which in turn causes higher losses. Therefore, the synchronous/free-wheeling MOSFET in a buck converter can be disabled when the converter operates in Discontinuous Current mode.

Many new techniques can be implemented in the digital domain that are almost impossible or too complex to be implemented in the analog domain. One example is the implementation of non-linear control systems. It is also possible to change the control loop in real-time, adapting the behavior of the system to dynamically change the system operating conditions, for instance, during changing input and output load conditions.

Back to top

Power Density

One of the leading power conversion market trends is the increase in the power density, i.e., more and more power that is generated by components that are increasingly getting smaller.

As previously mentioned, different control loops can be implemented into a single dsPIC DSC chip. This allows different functions to be implemented into the controller, thus reducing component counts. The dsPIC DSC provides benefits because of its capability of integrating several different functions, removing several analog controller components. It is possible to implement many control algorithms in the dsPIC DSC, and the ability to control more than one converter with other auxiliary functions, e.g., system supervision and communication, without interfering with the performances of the converter control loops. The overall design can be designed with less margin in the component values.

Back to top

Power Management

In many applications, the power converter is not a stand-alone product, but a part of a bigger system. In this case, the overall system requirements should be considered from the very beginning of the design process. The versatility of the digital solution using a dsPIC DSC allows us to do so.

In large power conversion systems, high-level control of the converters makes it possible to manage power conversion so that it is generated only when it is needed and its distribution can also be optimized.

Power management is also useful in small and limited systems. Designers are more and more commonly requested to supply a communication link to the converter so that the main processor in the system is aware of the power conversion status and can also make strategic decisions in the case of a fault in the power generation train.

A typical analog power supply will accomplish its power management requirements using a housekeeping MCU, which contains a dedicated microcontroller, and supporting circuitry for sequencing, monitoring, and communications, as shown in Figure 2.

benefits digital power management graphic

Figure 2

Back to top


An analog solution is hard-wired, while the dsPIC DSC is fully programmable. Code programmability is particularly essential during the production cycle. At the end of production, testing can be performed by downloading a test program into the chip. This adds the versatility of creating several different tests to check for functionality and verify performance. Moreover, component measurement can be measured and taken into account for the tolerances of the passive components, then compensate for them in firmware.

Whenever the tests are over, the final application code can be downloaded. Flash programmability allows users to change the final version of the code, to keep track of possible bugs, or to customize last-minute customer requests after the overall design is complete. Lastly, code can be written to create a platform that can be used as the starting point for other products.

Back to top

Monitoring and Protection

A key requirement in power systems is the management of fault conditions. In a power supply system, there are some typical checks/monitoring that are performed to ensure the correct behavior of the entire system, e.g., over-temperature, over-current, and over-voltage.

Temperature sensing and over-temperature events are relatively slow events. Over-voltage checks must be done at a much faster rate compared to temperature. Current monitoring may require very high-performance hardware if current must be checked cycle-by-cycle. In this case, either the ADC or comparator can be used. The dsPIC DSC internal comparators (up to four) can be used to replace any external stand-alone comparators since they are fast enough to cope with the highest frequency PWM signals.

The fault management peripheral in the dsPIC DSC is fully programmable, allowing power supply designers to select between the cycle-by-cycle monitoring. For example. when a fault event is detected, the PWM output is disabled, and then at the beginning of the following PWM period, the PWM signal automatically restarts.

Back to top


The ability to communicate with other companion chips on the same board has become increasingly popular in power conversion designs. Communications open a wide range of additional features for the power converter, where the parameters may be configured and changed in run-time and real-time, either locally or from a remote central unit. The behavior of the converter can also be changed dynamically if needed in code. New code releases can be uploaded locally or remotely to upgrade the system, fix bugs, or add functionality.

A large and complex system may give to rise several different needs, for instance, the interconnection between the unit and the central monitoring/supervision center (housekeeping circuitry). This means that the unit must be equipped with the capability of communication. Some standard buses are commonly used in the market, e.g., PMBUS. The number and type of data that can be exchanged on this communication link depend on the final application.

Digital power supplies, on the other hand, eliminate the need for this additional circuitry because all of the system parameters are already measured by the dsPIC DSC. These parameters can be stored in the DSC’s memory and transmitted to the remote system using on-chip communication peripherals such as SPI, I2C™, UART, or CAN. Any modifications to the system operation can also be made by a simple software routine without additional hardware. Local or remote data logging can be used in those systems where the quality of the produced power must be monitored and reported; data can be recorded for the operational conditions and the fault events. Power conversion designers can easily select the communication links that best fit the application. A customized protocol can be designed on top of the communication link that matches the designer's needs.

Back to top