Microchip dsPIC® Digital Signal Controllers (DSC) offer a range of auxiliary functions for the power conversion designer.
The following sections describe the definition of fault in power conversion systems, and its characteristics in the analog and digital domains.
Faults are unexpected functional conditions where some system parameters are outside their predefined ranges. Typical examples are:
- System over-current and over-voltage.
- MOSFET over-current
- MOSFET output over-voltage
- Drive supply under/over-voltage
- AC current sense offset
- DC reference under/over-voltage
Should such an event occur, the power conversion application must be able to monitor the fault occurrence and immediately shut down the output PWM signals in order to prevent damage or even destruction of the unit, or from causing harm to the operator.
If a power supply is not correctly protected from accidental operating conditions, for instance, the input over-voltage, output over-current or over-temperature, the unit can be damaged. The damage can not only result in repair and replacement costs, but may also endanger the operator’s life from a safety standpoint. Because fault events are unexpected functional conditions where one or more values in the system are below or above some predefined threshold, the fundamental characteristic of a fault is that it must be managed as fast as possible. For this reason, http://www.microchip.com/ParamChartSearch/chart.aspx?branchID=8183& | dsPIC® Digital Signal Controllers (DSC)]]] monitor faults continuously during system start-up.
Analog World Faults
Fault conditions can be managed in analog system by analog PWM chip. A dedicated pin can be used to detect the over-current condition. It's common that the fault threshold is hard-wired and cannot be changed by the user, which leads to inflexibility.
Sometimes there is the need to add an external comparator to detect the over-current/voltage event. This increases the total system cost and complexity. In terms of fault detection, many PWM chips have fault input pins. The effect of such an active pin is to shut down the PWM output or to restart the soft start sequence. In many cases, comparator chips are used to detect the signal crossing the preset threshold.
Digital World Faults
This section describes the rationale and advantages of the digital approach to handling the fault problem. The figure below shows the control software flowchart for an SMPS AC/DC design. The fault check routine is used to check for faults that have occurred in the system. If a fault has occurred, the system has to be shut down. To do this, the fault loop is called and used to disable all active modules, such as the ADC and PWM. The PWM module on the dsPIC33FJ16GS504 has built-in fault inputs that ensure a fast PWM shutdown in order to prevent damage to the system and downstream electronics. After the PWM is shut down, the program execution jumps to the fault loop.
The internal comparator(s) can be used to detect the fault event. The comparator threshold is set by an internal DAC. This allows the DAC to dynamically change the threshold if needed. When the comparator trips, the PWM outputs are switched off. This means that the designer can define what level should be present on the PWM pins during these events. This is a “hardware reaction” of the chip, meaning no CPU intervention is required. This ensures a fast response to the fault events. In addition to monitoring, digital fault operation is also capable of generating an interrupt to the CPU, enabling the designer to determine what should be done after the event has occurred. We'll explore the fault detection actions in detail in the Fault Detection article.