BGA Solder Joint Longevity Estimation Using Telemetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating solder-joint longevity in ball grid arrays (BGAs) are unreliable due to stochastic nature of signals from canary devices, leading to false and missed alarms, and require additional circuits that consume solder joints, causing issues with system reliability and maintenance.
Innovation Solution
A computer system generates a service action based on a longevity index value calculated using thermal and vibration telemetry data and a current longevity model, which incorporates accelerated failure testing and field failure data, allowing for robust and reliable estimates of solder-joint longevity without the need for additional circuits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If canary devices are used to monitor solder joint integrity, then reliability estimation is enabled, but device complexity increases due to additional circuits and solder joint consumption
Solution Approach 1:
The system uses existing system operations and built-in sensors to generate monitoring data without requiring dedicated canary devices or additional monitoring circuits. The computer system itself serves its own monitoring needs by utilizing operational telemetry data that is already being collected during normal system operation.
Solution Approach 2:
Existing system sensors and telemetry infrastructure are repurposed to serve dual functions: normal system operation and solder joint reliability monitoring. The same thermal and vibration sensors used for system management are also used to generate longevity estimates, eliminating the need for separate monitoring hardware.
2Reliability
If canary devices are deployed to monitor solder joints, then longevity estimation is possible, but solder joint quantity available for actual connections is reduced
Solution Approach 1:
The system monitors its own solder joint longevity using existing operational data without consuming any solder joints for monitoring purposes. All monitoring is performed through analysis of telemetry data from sensors already present in the system, requiring no additional physical test structures.
Solution Approach 2:
Instead of using physical canary solder joints as proxies, the system creates a virtual model of solder joint longevity by analyzing operational telemetry data. This digital twin approach allows monitoring without physical consumption of solder joints.
3Productivity
If monitoring is based solely on canary device signals, then continuous monitoring is achieved, but measurement precision decreases due to stochastic nature of signals leading to false and missed alarms
Solution Approach 1:
The system combines multiple data sources including thermal telemetry, vibration telemetry, power cycle data, and operational load information to estimate solder joint longevity. This multi-parameter approach averages out stochastic variations in any single measurement and provides a more robust and accurate reliability estimate.
Solution Approach 2:
The system continuously updates longevity estimates as new telemetry data becomes available during system operation. The model learns from accumulated operational data and refines its predictions over time, improving measurement precision through continuous feedback from actual system behavior rather than relying on static canary device readings.
Data Source
AI summary
A method for generating a service action for a computer system is described. During the method, a longevity index value for a packaging technology in the computer system is calculated using a longevity model and thermal and vibration telemetry data that is collected in the computer system. This longevity model may be based on accelerated failure testing of the packaging technology, field failures of the packaging technology in a group of computer systems that includes the computer system, and/or thermal and vibration telemetry data for the group of computer systems. Furthermore, using the longevity index value, the service action for the computer system is determined. Based on the longevity index value, remedial as action, e.g., repairs to the computer system, may be scheduled and performed.