Adaptive Load Fault Detection in High-Frequency Switching Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-frequency and narrow pulse-width switching systems pose challenges in accurately detecting load faults due to transient errors from capacitive and inductive loading, leading to false fault detection and potential system damage.
Innovation Solution
A method involving an error signal monitoring system that sets an error flag during a specified observation frame, particularly in high-frequency conditions, to differentiate between transient and actual fault conditions, using a fast mode error filter and adaptive threshold determination to avoid false load fault indicators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simple current detection techniques are used in capacitively loaded systems, then the initial current drive at the onset of a pulse is high enough to approximate short circuit or load fault conditions, but false load fault detection occurs
Solution Approach 1:
The patent applies preliminary action by establishing an observation window before the actual fault detection measurement. The system waits for a predetermined time period after pulse onset before evaluating current levels, allowing transient startup currents to subside. This preliminary waiting period prevents false detection of capacitive loading as fault conditions, thereby improving measurement precision without compromising system reliability.
Solution Approach 2:
The patent extracts the transient observation period from the fault detection process. By separating the initial transient phase from the measurement phase, the system can ignore harmful transient currents and focus only on sustained fault conditions. This extraction of the problematic transient period eliminates false detections while maintaining accurate fault detection capability.
2Measurement precision
If current detection is monitored over an extended period of time to exceed initial startup transients, then false load fault detection is reduced, but output fault detection becomes inadequate when pulse width is less than or on the order of the predetermined output fault detection time
Solution Approach 1:
The patent applies dynamics by making the observation window duration adaptive rather than fixed. The system dynamically adjusts the observation period based on the actual pulse width and operating conditions. This allows the system to use shorter observation windows for narrow pulses, maintaining fast detection capability, while using longer windows for wider pulses to filter transients, thus resolving the contradiction between accuracy and response time.
Solution Approach 2:
The patent changes the parameter of observation window duration based on operating conditions. By varying this time parameter adaptively, the system optimizes fault detection for different pulse widths. When pulses are narrow, the observation window is shortened to enable timely detection; when pulses are wide, the window is extended to filter transients, thereby maintaining both speed and accuracy across different operating scenarios.
3Speed
If output current sensor reports load fault condition immediately, then fast fault response is achieved, but transient errors from capacitive and inductive loading cause false fault detection
Solution Approach 1:
The patent uses preliminary action by implementing a brief observation period before triggering fault detection. This preliminary waiting phase allows transient currents from capacitive and inductive loading to settle, ensuring that only genuine fault conditions are detected. The system maintains fast response by keeping this preliminary period short, thus achieving both speed and precision.
Solution Approach 2:
The patent introduces an intermediary observation window between the current sensor and the fault detection logic. This intermediary phase acts as a buffer that filters out transient errors while preserving genuine fault signals. The observation window mediates between immediate detection and delayed detection, providing optimal balance between response speed and detection accuracy.
Data Source
AI summary
In one embodiment, a method for sensing an output fault condition is disclosed. The method includes monitoring an error signal that indicates an output fault condition, and monitoring an input signal having a duration. An error flag is set if a fast switching mode is detected and if the error signal is asserted within a specified time interval during the input signal duration.