{"title":"SAC305焊料PCB在不同温度和振动载荷条件下的RUL估计","authors":"P. Lall, Tony Thomas, K. Blecker","doi":"10.1109/ITherm45881.2020.9190530","DOIUrl":null,"url":null,"abstract":"Remaining Useful Life (RUL) estimation of electronic packages for different conditions of vibration loads and temperatures have various applications in scheduling maintenance and component replacement effectively to reduce the cost of the same. In this study, SAC305 alloy is used as the solder alloy, and the RUL is estimated using different particle filtering and time-series analysis techniques. The test board is a lead-free SAC305 daisy chain CABGA package which is subjected to different temperatures 25oC, 55oC, 100oC and 155oC for two vibration acceleration levels of 5g and 10g. The vibration of the test board is carried out to its first natural frequency for all conditions of temperature and vibration load. Strain signals are acquired using data acquisition and signal amplifying unit from four separate locations of the test board at a frequent time interval during vibration as the parameter used for predicting failure. In-situ measurements of resistance of the packages are also measured to identify the failure of the packages during vibration. The strain signals acquired at regular intervals during vibration at different locations of the board are used to find the feature vectors that can predict failure. Principal component analysis (PCA) is used as the data reduction technique for both time and frequency-based features of the strain signal. Feature vectors are estimated from the time, frequency, and spectral content of the strain signal using different multivariate statistical techniques. The variations in the feature vectors for different conditions of temperature and load is studied by combining all the feature vector data together and analyzing it for different patterns. The correlation of the same is studied to understand the changes in the feature vectors with different conditions. The two major feature vectors that can predict the failure includes frequency and spectral content from 500 Hz to 2000 Hz of the strain signal and the instantaneous frequency of the whole strain signal.","PeriodicalId":193052,"journal":{"name":"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"RUL Estimations of SAC305 Solder PCB's under Different Conditions of Temperature and Vibration Loads\",\"authors\":\"P. Lall, Tony Thomas, K. Blecker\",\"doi\":\"10.1109/ITherm45881.2020.9190530\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Remaining Useful Life (RUL) estimation of electronic packages for different conditions of vibration loads and temperatures have various applications in scheduling maintenance and component replacement effectively to reduce the cost of the same. In this study, SAC305 alloy is used as the solder alloy, and the RUL is estimated using different particle filtering and time-series analysis techniques. The test board is a lead-free SAC305 daisy chain CABGA package which is subjected to different temperatures 25oC, 55oC, 100oC and 155oC for two vibration acceleration levels of 5g and 10g. The vibration of the test board is carried out to its first natural frequency for all conditions of temperature and vibration load. Strain signals are acquired using data acquisition and signal amplifying unit from four separate locations of the test board at a frequent time interval during vibration as the parameter used for predicting failure. In-situ measurements of resistance of the packages are also measured to identify the failure of the packages during vibration. The strain signals acquired at regular intervals during vibration at different locations of the board are used to find the feature vectors that can predict failure. Principal component analysis (PCA) is used as the data reduction technique for both time and frequency-based features of the strain signal. Feature vectors are estimated from the time, frequency, and spectral content of the strain signal using different multivariate statistical techniques. The variations in the feature vectors for different conditions of temperature and load is studied by combining all the feature vector data together and analyzing it for different patterns. The correlation of the same is studied to understand the changes in the feature vectors with different conditions. The two major feature vectors that can predict the failure includes frequency and spectral content from 500 Hz to 2000 Hz of the strain signal and the instantaneous frequency of the whole strain signal.\",\"PeriodicalId\":193052,\"journal\":{\"name\":\"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)\",\"volume\":\"12 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ITherm45881.2020.9190530\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 19th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ITherm45881.2020.9190530","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
RUL Estimations of SAC305 Solder PCB's under Different Conditions of Temperature and Vibration Loads
Remaining Useful Life (RUL) estimation of electronic packages for different conditions of vibration loads and temperatures have various applications in scheduling maintenance and component replacement effectively to reduce the cost of the same. In this study, SAC305 alloy is used as the solder alloy, and the RUL is estimated using different particle filtering and time-series analysis techniques. The test board is a lead-free SAC305 daisy chain CABGA package which is subjected to different temperatures 25oC, 55oC, 100oC and 155oC for two vibration acceleration levels of 5g and 10g. The vibration of the test board is carried out to its first natural frequency for all conditions of temperature and vibration load. Strain signals are acquired using data acquisition and signal amplifying unit from four separate locations of the test board at a frequent time interval during vibration as the parameter used for predicting failure. In-situ measurements of resistance of the packages are also measured to identify the failure of the packages during vibration. The strain signals acquired at regular intervals during vibration at different locations of the board are used to find the feature vectors that can predict failure. Principal component analysis (PCA) is used as the data reduction technique for both time and frequency-based features of the strain signal. Feature vectors are estimated from the time, frequency, and spectral content of the strain signal using different multivariate statistical techniques. The variations in the feature vectors for different conditions of temperature and load is studied by combining all the feature vector data together and analyzing it for different patterns. The correlation of the same is studied to understand the changes in the feature vectors with different conditions. The two major feature vectors that can predict the failure includes frequency and spectral content from 500 Hz to 2000 Hz of the strain signal and the instantaneous frequency of the whole strain signal.