Adaptive SEL Current Surge Mitigation via Signature Vector Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to effectively and rapidly distinguish between standard power surges and single event latchup (SEL) current surges in integrated circuits, leading to potential damage due to inadequate detection and response times, especially in environments exposed to energetic particles.
Innovation Solution
A method involving adaptive current control that monitors signature vectors to differentiate between standard and non-standard current surges by comparing detected signature components to predetermined values, using a sensing circuit and adaptive current control circuit to selectively cycle power and prevent damage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional current threshold comparators are used to detect SEL events, then current surge detection is provided, but the system cannot distinguish between normal power transients and true SEL transients
Solution Approach 1:
The current transient detection is segmented into multiple characteristic dimensions: magnitude threshold, rise time threshold, and decay time threshold. By dividing the detection process into these separate criteria, the system can evaluate each aspect independently and combine them for more accurate SEL identification, reducing false positives from normal transients.
Solution Approach 2:
The system changes the detection parameters from a single current threshold to multiple parameters including rise time and decay time characteristics. This multi-parameter approach allows the system to distinguish SEL events (which have specific transient characteristics) from normal power transients (which have different temporal profiles), thereby improving detection accuracy while maintaining reliability.
2Reliability
If simple current-limiting regulators are used, then current protection is provided, but the system cannot determine when current surge corresponds to SEL
Solution Approach 1:
The system implements feedback by continuously monitoring the current transient characteristics and comparing them against predefined SEL signatures. The comparator provides feedback about whether the transient matches SEL patterns, enabling the system to make informed decisions about device protection while accurately identifying SEL events.
Solution Approach 2:
The system performs preliminary characterization of SEL transient signatures during normal operation or setup. By establishing reference profiles of what SEL transients look like in advance, the system can quickly compare future events against these known patterns, making SEL identification straightforward without complex real-time analysis.
3Reliability
If existing commercial SEL protection products are used, then current threshold control is provided, but response time is insufficient to prevent latent damage
Solution Approach 1:
The system rushes through the detection and response process by using dedicated hardware comparators that can evaluate current transients and trigger protection within nanoseconds. This hardware-based approach skips the delays associated with software processing, enabling response times fast enough to prevent latent damage while maintaining device protection.
4Adaptability or versatility
If commercial off-the-shelf integrated circuit devices are used, then device availability is improved, but internal power subdivision for SEL detection is not possible
Solution Approach 1:
The system introduces an intermediary SEL detection circuit that sits between the power supply and the commercial off-the-shelf device. This intermediary component monitors the current transients without requiring any internal modification to the protected device, thereby maintaining device compatibility while adding the necessary SEL detection capability through an external mediator.
Data Source
AI summary
Method for SEL mitigation involves determining one or more base sets of signature vector components for each of a plurality of signal loading conditions experienced by a protected device in an operating state, each set of base signature vector components together comprising a base signature vector. The method further involves monitoring signature vector components for the protected device to determine a detected signature vector which is comprised of a set of detected signature vector components. The detected signature vector is compared to a dynamically selected base signature vector which is associated with the device state and signal loading condition which are currently active to differentiate between the occurrence of standard current surges associated with normal operation of the protected device and a non-standard current surge.