TL;DR
This paper presents a novel multi-sensor Prognostic and Health Management (PHM) system for detecting defects in the ferrite phase shifter cementing process. By establishing a quantitative model relating cementing strength to key process parameters (temperature, time, and pressure) and employing on-line monitoring, the system significantly improves the reliability and reduces defects in this critical manufacturing step.
What Problem Does This Paper Address?
The manufacturing of ferrite phase shifters involves a crucial cementing process that bonds microwave dielectric rods and ferrite bars. A common failure mode in this process is insufficient cementing strength, leading to defects like fractures. Existing methods for detecting these defects are often limited, and there's a need for a more robust, on-line approach to ensure product reliability. This paper addresses the challenge of identifying and mitigating these cementing process defects by leveraging PHM principles and multi-sensor data.
Methods
The authors developed a PHM-based approach to detect defects in the ferrite phase shifter cementing process. Their methodology involved:
- Identifying Key Process Parameters: Through analysis and Differential Scanning Calorimetry (DSC) tests, they identified curing temperature, curing time, and curing pressure as the most influential parameters affecting cementing strength.
- Developing a Quantitative Model: They established a relational model linking cementing strength to these parameters using the Avrami formula, which describes the kinetics of thermosetting resin curing. This model was refined using DSC data and extensive process tests.
- Sensitivity Analysis: A sensitivity analysis was performed on the process parameters to determine their impact on cementing strength, leading to the identification of critical on-line detection criteria.
- Establishing a Multi-Sensor PHM System: Based on the derived criteria, a system was built incorporating temperature and pressure sensors for on-line monitoring of the cementing process.
- On-line Detection, Diagnosis, and Control: The PHM system was designed to detect deviations from optimal conditions, diagnose potential defects, and enable adjustments (e.g., extending curing time or modifying pressure) to maintain product quality.
Datasets
The study utilized data from:
- Differential Scanning Calorimetry (DSC) tests: Performed at four different temperatures (115°C, 125°C, 135°C, and 145°C) to understand the exothermic reaction kinetics of the adhesive.
- Process tests for cementing: 28 groups of tests were conducted with varying temperatures (four levels) and curing pressures (seven levels, 5-35 N). This generated data on cementing strength under different conditions.
- Actual production data: Bonding strength data was collected before and after implementing the on-line detection method to assess its effectiveness.
Key Insights and Ideas
- Cementing Strength is Highly Sensitive to Process Parameters: The research clearly demonstrates that curing temperature and pressure have a significant impact on the final cementing strength, with pressure showing a particularly strong influence.
- Optimal Pressure Window Exists: The study found that cementing strength increases with pressure up to a certain point (around 15 N) and then decreases, highlighting the importance of operating within a specific pressure range.
- PHM for Manufacturing Process Control: Applying PHM principles, traditionally used for product health monitoring, to a manufacturing process itself offers a powerful way to ensure quality and reduce defects proactively.
- Avrami Formula as a Foundation: The Avrami formula provides a robust theoretical framework for modeling the curing kinetics and relating it to observable process parameters and ultimately, product strength.
- On-line Detection is Feasible and Effective: The developed multi-sensor system successfully enabled on-line detection and control, leading to a measurable improvement in product qualification rates.
Key Results
- Improved Qualification Rate: After implementing the on-line detection method, the qualification rate of ferrite phase shifters increased from approximately 85% to over 95%.
- Reduced Process-Related Defects: The system successfully identified and helped rectify potential defects related to curing temperature and pressure. Out of 242 samples, only 12 were sub-quality, and the method detected 24 out of 26 potential process-related defects, achieving a detection probability of 92.3%.
- Established Optimal Process Conditions: The study identified optimal process conditions for cementing: curing temperature of 135°C, curing time of 2 hours, and curing pressure between 15 N and 16 N.
- Quantitative Model Developed: A comprehensive quantitative model relating cementing strength to curing temperature, time, and pressure was established, validated by experimental data.
Limitations and Future Work
- Limitations of On-line Detection: The authors acknowledge that their current on-line detection method cannot detect material defects such as cracks, uneven surfaces in ferrite materials, or pollution/aging in resin adhesives.
- Future Work: The paper suggests that future research will focus on developing indirect on-line non-destructive testing methods to address these undetected defects, particularly those related to material properties that are difficult to observe with conventional sensors. The complexity, cost, and implementation feasibility of new methods will be key considerations.
Summary updated