Decorrelated Feature Space and Neural Nets Based Framework for Failure Modes Clustering in Electronics Subjected to Mechanical Shock

2012 ◽  
Vol 61 (4) ◽  
pp. 884-900 ◽  
Author(s):  
P. Lall ◽  
P. Gupta ◽  
K. Goebel
Keyword(s):  
Failure Modes ◽  
Feature Space ◽  
Neural Nets ◽  
Mechanical Shock

KL Transform and Neural-Net Based Framework for Failure Modes Classification in Electronics Subjected to Mechanical-Shock

2011 ◽  
Author(s):  
Pradeep Lall ◽  
Prashant Gupta ◽  
Kai Goebel
Keyword(s):  
Failure Mode ◽  
Failure Modes ◽  
Learning Algorithm ◽  
Back Propagation ◽  
Feature Space ◽  
System Level ◽  
Neural Net ◽  
Mechanical Shock ◽  
Time Frequency ◽  
Damage Initiation

Electronic systems under extreme shock and vibration environments including shock and vibration may sustain several failure modes simultaneously. Previous experience of the authors indicates that the dominant failure modes experienced by packages in a drop and shock frame work are in the solder interconnects including cracks at the package and the board interface, pad cratering, copper trace fatigue, and bulk-failure in the solder joint. In this paper, a method has been presented for failure mode classification using a combination of Karhunen Loe´ve transform with parity-based stepwise supervised training of a perceptrons. Early classification of multiple failure modes in the pre-failure space using supervised neural networks in conjunction with Karhunen Loe´ve transform is new. Feature space has been formed by joint time frequency analysis. Since the cumulative damage may be accrued under repetitive loading with exposure to multiple shock events, the area array assemblies have been exposed to shock and feature vectors constructed to track damage initiation and progression. Error Back propagation learning algorithm has been used for stepwise parity of each particular failure mode. The classified failure modes and failure regions belonging to each particular failure modes in the feature space are also validated by simulation of the designed neural network used for parity of feature space. Statistical similarity and validation of different classified dominant failure modes is performed by multivariate analysis of variance and Hoteling’s T-square. The results of different classified dominant failure modes are also correlated with the experimental cross sections of the failed test assemblies. The methodology adopted in this paper can perform real-time fault monitoring with identification of specific dominant failure mode and is scalable to system level reliability.

Anomaly-Detection and Prognostication of Electronics Subjected to Shock and Vibration

2009 ◽  
Author(s):  
Pradeep Lall ◽  
Prashant Gupta ◽  
Arjun Angral ◽  
Jeff Suhling
Keyword(s):  
Experimental Data ◽  
Frequency Analysis ◽  
Health Monitoring ◽  
Solder Ball ◽  
Failure Modes ◽  
Feature Space ◽  
Time Frequency Analysis ◽  
Time Frequency ◽  
Fault Mode ◽  
Electronic Assemblies

Failures in electronics subjected to shock and vibration are typically diagnosed using the built-in self test (BIST) or using continuity monitoring of daisy-chained packages. The BIST which is extensively used for diagnostics or identification of failure, is focused on reactive failure detection and provides limited insight into reliability and residual life. In this paper, a new technique has been developed for health monitoring and failure mode classification based on measured damage precursors. A feature extraction technique in the joint-time frequency domain has been developed along with pattern classifiers for fault diagnosis of electronics at product-level. The Karhunen Loe´ve transform (KLT) has been used for feature reduction and de-correlation of the feature vectors for fault mode classification in electronic assemblies. Euclidean, and Mahalanobis, and Bayesian distance classifiers based on joint-time frequency analysis, have been used for classification of the resulting feature space. Previously, the authors have developed damage pre-cursors based on time and spectral techniques for health monitoring of electronics without reliance on continuity data from daisy-chained packages. Statistical Pattern Recognition techniques based on wavelet packet energy decomposition [Lall 2006a] have been studied by authors for quantification of shock damage in electronic assemblies, and auto-regressive moving average, and time-frequency techniques have been investigated for system identification, condition monitoring, and fault detection and diagnosis in electronic systems [Lall 2008]. However, identification of specific failure modes was not possible. In this paper, various fault modes such as solder inter-connect failure, inter-connect missing, chip delamination chip cracking etc in various packaging architectures have been classified using clustering of feature vectors based on the KLT approach [Goumas 2002]. The KLT de-correlates the feature space and identifies dominant directions to describe the space, eliminating directions that encode little useful information about the features [Qian 1996, Schalkoff 1972, Theodoridis 1998, Tou 1974]. The clustered damage pre-cursors have been correlated with underlying damage. Several chip-scale packages have been studied, with leadfree second-level interconnects including SAC105, SAC305 alloys. Transient strain has been measured during the drop-event using digital image correlation and high-speed cameras operating at 100,000 fps. Continuity has been monitored simultaneously for failure identification. Fault-mode classification has been done using KLT and joint-time-frequency analysis of the experimental data. In addition, explicit finite element models have been developed and various kinds of failure modes have been simulated such as solder ball cracking, trace fracture, package falloff and solder ball failure. Models using cohesive elements present at the solder joint-copper pad interface at both the PCB and package side have also been created to study the traction-separation behavior of solder. Fault modes predicted by simulation based pre-cursors have been correlated with those from experimental data.

The Mechanical Reliability of a Commercial MEMS Microphone

2013 ◽  
Vol 662 ◽  
pp. 551-555
Author(s):  
Wen Xiao Fang ◽  
Qin Wen Huang
Keyword(s):  
Recent Work ◽  
Random Vibration ◽  
Failure Modes ◽  
Mechanical Tests ◽  
Vibration Test ◽  
Mechanical Shock ◽  
Mechanical Reliability ◽  
Constant Acceleration ◽  
Shock Test ◽  
Mems Microphone

This paper describes our recent work on the mechanical reliability of a commercial MEMS microphone by performing three mechanical tests, i.e. a constant acceleration test, a mechanical shock test, and a random vibration test, according to the standard of Mil-Std-883. We find that, the studied MEMS part of the microphone can survive a stress limit above 20000g. Two failure modes, i.e. the breaking of diaphragm and the backplate and the delamination of the electrode from the backplate are revealed in the three tests.

Fault-Mode Classification of Solid State Luminaires Using Bayesian Probabilistic Models

2014 ◽  
Author(s):  
Pradeep Lall ◽  
Junchao Wei ◽  
Peter Sakalaukus
Keyword(s):  
Solid State ◽  
Probabilistic Models ◽  
Failure Modes ◽  
Feature Space ◽  
Luminous Flux ◽  
Color Temperature ◽  
Test Technique ◽  
Fault Mode ◽  
Decision Boundaries

A new method has been developed for assessment of the onset of degradation in solid state luminaires to classify failure mechanisms by using metrics beyond lumen degradation that are currently used for identification of failure. Luminous Flux output, Correlated Color Temperature Data on Philips LED Lamps has been gathered under 85°C/85%RH till lamp failure. The acquired data has been used in conjunction with Bayesian Probabilistic Models to identify luminaires with onset of degradation much prior to failure through identification of decision boundaries between lamps with accrued damage and lamps beyond the failure threshold in the feature space. In addition luminaires with different failure modes have been classified separately from healthy pristine luminaires. It is expected that, the new test technique will allow the development of failure distributions without testing till L70 life for the manifestation of failure.

Failure Mode Clustering is Electronic Assemblies Using Sammon’s Mapping With Supervised Training of Perceptrons

2011 ◽  
Author(s):  
Pradeep Lall ◽  
Prashant Gupta ◽  
Kai Goebel
Keyword(s):  
Failure Mode ◽  
Solder Ball ◽  
Failure Modes ◽  
Feature Vector ◽  
Feature Space ◽  
Image Correlation ◽  
Neural Net ◽  
Sammon’S Mapping ◽  
Electronic Assemblies ◽  
Diagnostic Detection

An anomaly detection and failure mode classification method has been developed for electronic assemblies with multiple failure modes. The presented prognostic health management method targets the pre-failure space of the electronic assembly life to trigger repair or replacement of impending failures. Presently, health monitoring systems focus on reactive diagnostic detection of failure modes. Examples of diagnostic detection include the built in self test and on-board diagnostics. In this paper, damage pre-cursors from time-spectral measurements of the electronic assemblies has been measured under applied vibration and shock stimulus. The time-evolution of spectral content of the damage pre-cursors has been studied using joint time frequency analysis in a full-field manner on the printed circuit assembly. Frequency moments have been used to build a feature vector. Evolution of the feature vector with damage initiation and progression has been studied under shock and vibration. The feature vector from multiple locations in the board assemblies has been mapped into a de-correlated feature space using Sammon’s mapping. Several chip-scale packages have been studied, with SAC305 and SAC405 leadfree second-level interconnects. Transient strain has been measured during the drop-event using digital image correlation and high-speed cameras operating at 100,000 fps. Continuity has been monitored simultaneously for failure identification. In addition, explicit finite element models have been developed and various kinds of failure modes have been simulated such as solder ball cracking, trace fracture, package falloff and solder ball failure. The neural net has been trained using simulated data-sets created from error-seeded models with specific failure modes. The neural net has then been used to identify and classify the failure modes in board assemblies experimentally. Supervised learning of multilayer neural net in conjunction with parity has been used to identify the hard-separation boundaries between failure mode clusters in the de-correlated feature space. The assemblies have been cross-sectioned to verify the identified failure modes. Cross-sections indicate that the experimentally measured failures modes correlate well with the position of the cluster in the de-correlated feature space.

Constitutive Behavior of SAC Leadfree Alloys at High Strain Rates

2011 ◽  
Author(s):  
Pradeep Lall ◽  
Sandeep Shantaram ◽  
Mandar Kulkarni ◽  
Geeta Limaye ◽  
Jeff Suhling
Keyword(s):  
Constitutive Model ◽  
Solder Alloy ◽  
Failure Modes ◽  
High Strain ◽  
Image Correlation ◽  
Constitutive Behavior ◽  
Strain Rates ◽  
Mechanical Shock ◽  
High Strain Rates ◽  
Test Technique

Electronic products are subjected to high G-levels during mechanical shock and vibration. Failure-modes include solder-joint failures, pad cratering, chip-cracking, copper trace fracture, and underfill fillet failures. The second-level interconnects may be experience high-strain rates and accrue damage during repetitive exposure to mechanical shock. Industry migration to leadfree solders has resulted in proliferation of a wide variety of solder alloy compositions. Few of the popular tin-silver-copper alloys include Sn1Ag0.5Cu and Sn3Ag0.5Cu. The high strain rate properties of leadfree solder alloys are scarce. Typical material tests systems are not well suited for measurement of high strain rates typical of mechanical shock. Previously, high strain rates techniques such as the Split Hopkinson Pressure Bar (SHPB) can be used for strain rates of 1000 per sec. However, measurement of materials at strain rates of 1–100 per sec which are typical of mechanical shock is difficult to address. In this paper, a new test-technique developed by the authors has been presented for measurement of material constitutive behavior. The instrument enables attaining strain rates in the neighborhood of 1 to 100 per sec. High speed cameras operating at 300,000 fps have been used in conjunction with digital image correlation for the measurement of full-field strain during the test. Constancy of cross-head velocity has been demonstrated during the test from the unloaded state to the specimen failure. Solder alloy constitutive behavior has been measured for SAC105, and SAC305 solders. Constitutive model has been fit to the material data. Samples have been tested at various time under thermal aging at 25°C and 125°C. The constitutive model has been embedded into an explicit finite element framework for the purpose of life-prediction of leadfree interconnects. Test assemblies has been fabricated and tested under JEDEC JESD22-B111 specified condition for mechanical shock. Model predictions have been correlated with experimental data.

Survivability of MEMS Accelerometer Under Sequential Thermal and High-G Mechanical Shock Environments

2015 ◽  
Author(s):  
Pradeep Lall ◽  
Amrit Abrol ◽  
Lee Simpson ◽  
Jessica Glover
Keyword(s):  
Failure Modes ◽  
Performance Enhancement ◽  
Research Work ◽  
Harsh Environments ◽  
Mechanical Shock ◽  
X Rays ◽  
Mems Devices ◽  
Mems Accelerometer ◽  
Capacitive Accelerometer ◽  
Mems Accelerometers

Reliability data on MEMS accelerometers operating in harsh environments is scarce. Micro-electro-mechanical systems (MEMS) are used in a variety of military and automotive applications for sensing acceleration, translation, rotation, pressure and sound. This research work focuses on dual axis MEMS accelerometer reliability in harsh environments. Structurally an accelerometer behaves like a damped mass on a spring. Commercially there are three types of accelerometers namely piezoelectric, piezoresistive and capacitive depending on the components that go into the fabrication of the MEMS device. Previously, majority of concentration was focused on an effective internal design, performance enhancement of CMOS-MEMS accelerometers and packaging techniques Cheng [2002], Qiao [2009], Lou [2005], and Weigold [2001]. Studies have also been conducted to obtain an enhanced inertial mass SOI MEMS process using a high sensitivity accelerometer Jianbing [2013], Chen [2005]. There have been prior test(s) conducted on MEMS accelerometers, Jiang [2004], Cao [2011], Chun-Sun [2009], Lou [2009], Tanner [2000] and Yang [2010] but the availability of data on reliability degradation of such devices in harsh environments Brown [2003] is almost little to none which thereby generates the importance of this work and also makes way for a whole new path involving the reliability assessment techniques for MEMS devices. Concentration of our work is primarily on the reliability of this accelerometer upon sequential exposure to harsh environment(s) and drop-shock. Reliability of accelerometers in high G environments is unknown. The effects of these pre-conditions along with the drop test condition has been studied and analyzed. In this piece of research work, a test vehicle with a MEMS accelerometer, ADXL278 dual axis capacitive accelerometer, has been tested under high/low temperature exposure followed by subjection to high-g and low-g shock loading environments. The test boards have been subjected to mechanical shocks using the method 2002.5, condition G, under the standard MIL-STD-883H test. The stress environment and the test condition used for this paper are 1500g and 70g respectively where 70g is the full scale range output of ADXL278 in the drop direction with pulse duration set to 0.5millisecond. The deterioration of the accelerometer output has been characterized using the techniques of Mahalanobis distance and Confidence intervals. Scanning Electron Microscopy (SEM) has been used to study the different failure modes inside of the accelerometer, which were potted and polished and later de-capped. Furthermore, the non-destructive evaluations of the MEMS accelerometer have been demonstrated through X-rays and micro-CT scans.

The Response of Electrostatic MEMS Structure under Mechanical Shock and Electrostatic Forces

2013 ◽  
Vol 427-429 ◽  
pp. 120-123
Author(s):  
Xiang Guang Li ◽  
Qin Wen Huang ◽  
Yun Hui Wang ◽  
Yu Bin Jia ◽  
Zhi Bin Wang
Keyword(s):  
Failure Modes ◽  
Electrostatic Force ◽  
Short Circuit ◽  
Superposition Method ◽  
Mechanical Shock ◽  
Mems Devices ◽  
Electrostatic Forces ◽  
Mode Superposition Method ◽  
Electrostatic Mems ◽  
Micro Cantilever

For MEMS devices actuated by electrostatic force, unexpected failure modes can be hardly predicted when the electrostatic force coupled with the shock. A response model is established when a micro cantilever subjected to electrostatic force and mechanical shock. First, based on the theory of transverse forced vibration in vibration mechanics, the equation of motion under shock and electrostatic fore is presented. Then the reduced order model is gained after simplifying by mode superposition method. The computing results indicate that: the shock amplitude and duration are the key factors to affect the reliability of the device; the shock load and electrostatic forces make the threshold voltage much lower than the anticipated value. The micro cantilever may collapse to the substrate even at a voltage far lower than the pull-in voltage. This early dynamic pull-in instability may cause some failures such as short circuit, adhesion or collision damage.

Sign in / Sign up

Export Citation Format

Share Document