Embed Size (px)
Citation preview
Medical Components Reliability
Specifications Meeting – MLCC
Phase I wrap up and Phase II
introduction
Anthony PrimaveraNovember 14, 2008
Background
2
OverviewOverview
• Medical Electronics Trends• Growth Drivers• Technology Trends• Technology Integration Plans
– Key Technology Gaps– Initial Projects
• Component Reliability Specifications Project– Project Overview– Opportunities
• Medical Reliability for MLCC (multi-layer ceramic capacitor) Project Update
3
Medical Electronics Trends Medical Electronics Trends -- GrowthGrowth• Impact to national economy on healthcare will
force the need for less expensive systems.– By 2010 there will be an expected 40 million persons in
the U.S. over 65 (U.S. Census Bureau)– 74 Million elders in US, 1.2 Billion world wide by 2025– US spending is currently 15-16% of federal budget on
health care. – By 2025, there will be 1 retiree for every 3 workers.– Higher spending will be required if current model is
continued to be followed. • long term impact is an international trend toward home
health technologies and preventive health care
4
Medical Electronics Trends Medical Electronics Trends -- Growth Growth
Source: Bureau of Labor Statistics, pub in CNN/Money.com, “Where the jobs will be Greatest employment growth is likely to be in service industries, according to new labor study.” By Jeanne Sahadi, Feb 13, 2004.
• Medical related job demand fastest growing• Inadequate supply of labor
• Costs are critical• Rising hospital care costs, • Escalation in the number of un-
insured, • Shorter healthcare giver – patient
interaction time.
• Growing “Consumer Medical Electronics Market”• home diagnostic equipment,
wearable patient monitoring equipment, etc.
$0
$20
$40
$60
$80
$100$Bn
GEOGRAPHIC DISTRIBUTION
Americas
$66Bn5% of
Electronics Industry
2007
5.1% CAAGR2007-2013
$89Bn
$60Bn
$66Bn
54%
Japan
8%
Europe
25%
Asia/ROW
13%
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013
Kc78.273mw-med
5
Medical Electronics Trends Medical Electronics Trends -- Market GrowthMarket Growth
DISRUPTIVE TECHNOLOGIES / Consumer Medical
Products???
6
Medical Technology DriversMedical Technology Drivers
Category DescriptionMedical TherapyTherapy managementPatient EligibilityAvailability
SOLUTION Hardware & Software
SUPPORT Technology Access
MARKET
Before Today (2006) Future (2012)Necessity Reactive ProphylacticCaregiver Caregiver/patient AutomatedCriticality driven Comfort driven Wellness/risk driven
Developed Nations Developing Nations
Local Global Global <--> g'local
DeviceTherapy specific
SystemLimited convergence
Multi-systemIntegrated / convergent
Newly formed markets, new technologies are in developmentNewly formed markets, new technologies are in developmentPREPRE--COMPETETIVE INDUSTRY COLLABORATION IS VITALCOMPETETIVE INDUSTRY COLLABORATION IS VITAL
Implantable Defibrillators—US Annual–350,000 people newly indicated for this therapy –100,000 + defibrillator implants per year
Shock Shock
DeliveredDelivered
Tachy Arrhythmia Therapy Extending the lives of people whose
hearts beat too fast
Medical Market: Example Technology
Heart Failure Therapy
– 5 year mortality rates as high as 50%
– Affects > 14M people (US, Europe & Japan)
TINESTINES
STEROIDSTEROID COLLARCOLLAR
GUIDE WIREGUIDE WIRE
ELECTRODEELECTRODE
Medical Market: Example Technology
Lead Technology
Pacemakers – Currently the largest medical device market
US Annual– 650,000 patients diagnosed with this condition annually– 300,000 implants annually
Digital Health• In the past 10 years, growth, innovation and
miniaturization have lead to major advances in medical electronics manufacturing and the therapies they deliver.
• Patient care enhancement• New and Unique Medical Products• Monitor Systems• Sensor Technology• Improved Diagnostics
• Wireless technology for data transfer • Instant and remote monitoring• Power can be transmitted via RF signals• Can off-load computing and data storage to remote host system
outside the device.
Digital Health
• Miniaturization / Nano• MEMS• Self or bio powered systems• Localized measurement – ex lab on a chip
• And many more….
• All these advancements lead to complex problems involving materials management, energy constraints, data security, reliability and above all patient safety.
Stents
PTCA Systems
IntravascularBrachytherapy
Atherectomy
Stents
PTCA Systems
IntravascularBrachytherapy
Atherectomy
Areas of Opportunity in Medical Market
Pacemakers
AICDs
Leads
AblationCatheters
Pacemakers
AICDs
Leads
AblationCatheters
Pacemakers
AICDsLeads
AAA Systems
PeripheralStents
Neurovascular
AAA Systems
PeripheralStents
Neurovascular
AAA Systems
PeripheralStents
Neurovascular
ENDO--VASCULARSOLUTIONS
Beating Heart BypassSurgery
Minimally-Invasive
Vein Harvesting
Beating Heart BypassSurgery
Minimally -Invasive
Vein Harvesting
Beating Heart BypassSurgery
Minimally -Invasive
Vein Harvesting
CARDIACCARDIACSURGERYSURGERY
EQUIPMENT
FOR
SURGERY
DATA TRANSFER AND ANALYSIS DATA TRANSFER AND ANALYSIS DATA TRANSFER AND ANALYSIS
IMPLANTABLEPRODUCTS
Coclear devices
Pin and drug
VASCULARINTERVENTION
Beating Heart BypassSurgery
Minimally-Invasive
Vein Harvesting
Beating Heart BypassSurgery
Minimally -Invasive
Vein Harvesting
MRI
SONOGRAM
BLOOD
ANALYZER
-
DIAGNOSTIC
AND
MONITORING
Challenges to Implanted Device Assembly
•What make us so different?
•Small, smaller, smallest….•Severe constraints on space, weight, shape, power consumption, bodily fluids etc.
•Sealed device, must be right the first time, and last for years and years.
•Volumes can’t drive supply chain.
Server & telecom Driving ThisLarge I/O BGAs
Implantable devices need that!
Challenges: 1 – Design Limits
Space is our Enemy!No, not that space
This Space
Free space inside a device is, wasted on air
14
Medical Technology Trends Medical Technology Trends -- SectorsSectors
• Global trends defined within three medical sectors– Diagnostic imaging devices and large scale equipment, e.g.,
Ultrasound, MRI, etc. – Portable products (those devices that are easily transported) – Implanted products (those devices implanted in a human body)
Some product solutions will necessarily consist of combinations of all three categories of devices.
• Main differences– Product size, features and form factor– Energy type, source and usage– Reliability requirements– Regulatory issues– Supply chain
MEDICAL APPLICATIONS ARE VARIEDSIZE/FUNCTIONALITY DRIVES DIFFERENCES IN TECHNOLOGY GAPS
15
Gap Identification Gap Identification -- DiagnosticsDiagnostics
Gaps 2006 2008 2010 2012
DesignHigh density sensor arraysIntegrated detection-processing architecturesLow power image detection and processingHigh speed image acquisition
MaterialsMagnetic susceptibility <100 ppmLow temperature assemblyThermal managementConnector-less assembly
Data Processing and TransferHigh signal fidelity (low loss, high integrity)High bandwidthUltra-low data/bit error rates
Diagnostic Imaging Systems
Green = No Gap Issues or Resolved Yellow = Known Gap Mitigation Techniques Red = No Known Solution – Development Required
Interconnect Density
16
Gap Identification Gap Identification -- PortablesPortables
GapsReliability
User Interoperability and Wireless
RF Traffic
Standards for Components Used in Portable Medical Devices
Materials & Processing
Targeted Nanoparticles
Design
Technical convergence at the device/component level
Ubiquitous remote programming
Novel energy sources
Point-of-care clinical biomarkers
Portable Medical Devices
2012201020082006 2012201020082006
Green = No Gap Issues or Resolved Yellow = Known Gap Mitigation Techniques Red = No Known Solution – Development Required
RF Traffic ReliabilityComponent Reliability
Nanomaterials
Remote programmingEnergy SourcesClinical Biomarkers
17
Gap Identification Gap Identification -- ImplantablesImplantables
GapsReliability
Standard implantable device use conditions
Medical component test methods
Medical component reliability standards
Materials & Processing
RoHS compliant components
RoHS compatible implantable device processes
Nano materials for implantables
Advanced materials for implantables
Reaching physical limitations of interconnect technology
MEMS packaging for implantables
Design
MEMS sensors
Higher capacity/alternative energy sources
Higher energy density, higher voltage charge delivery capacitors
Smaller volume (size) components
Active Implantable Medical Devices
2012201020082006 2012201020082006
Analytical Tools & Methods
Advanced integrated modeling tools (electrical, thermal, mechanical, optical, chemical)
Advanced micro measurement tools
Business
Optimized supply chain
Long development cycle
Product life cycle management
Green = No Gap Issues or Resolved Yellow = Known Gap Mitigation Techniques Red = No Known Solution – Development Required
Component Reliability
RoHS
Nanomaterials
Interconnect densityMEMs packaging
Advanced design
Optimized Supply ChainProduct Life cycle
18
Potential ProjectsPotential Projects
Common between multiple medical product sectors:• Component reliability: Medical Electronics
Components Reliability Specifications Project• RoHS compliance for medical electronics• Interconnect density: Advanced Printed Wiring
boards• RF Reliability: Impact of increasing RF traffic in
increasing clinical and home environments
ONGOING FOCUS ON COMPONENT RELIABILITY PROJECT
Medical Component
Reliability Specifications
Project
20
Medical Component Reliability Specifications ProjectMedical Component Reliability Specifications Project
Object: • To leverage industry knowledge to create a minimum set of
requirements for electronics components for application in life critical applications
• This will allow component suppliers access to the entire industry by providing commonly accepted accelerated testing, extrapolation analysis, materials and processes
• Medical device manufacturers will achieve proven quality, reliability and consistency from these components
21
Medical Component Reliability Specifications ProjectMedical Component Reliability Specifications Project
Specific Deliverables:• Test and Extrapolation Methodologies
– Sampling Population Assessment– Range and Conditions of Applicability– Test Methodologies and Criteria– Medical Grade Guidelines
• FMEA of MLCC (Multi-Layer Ceramic Capacitors) failures• Use Conditions for Life Critical Medical Components• Review of existing and related Standards and Test Methods• Preliminary Test Results for MLCC devices at NIST• Creation for Test DOE matrix for use in Medical Reliability for
MLCC Project
22
Medical Components Reliability Specifications Project
Project Members (SOW)
23
Medical Components Reliability Specifications Project
Project Suppliers
24
Medical Component Reliability Project ProcessMedical Component Reliability Project Process
Supplier Methods, Processes, Risks
• Raw Materials (e.g. for capacitors Ceramic, Tantalum, Terminations)
Existing Standards• Supplier Plan• Component Reliability and
Assessment
Use Conditions• Manufacturing Process & Testing• Storage (before / after assembly)• RoHS Compliant Requirements • Operating (In Use)
Accelerated Testing and Extrapolation Methodology• Sampling/population assessment• Conditions of Applicability• Test Methodology and Criteria
CO
MPO
NEN
T R
ELIA
BIL
ITY
PRO
JEC
T TE
AMInputs OUTPUTS
Medical Grade Specifications
25
SubgroupsSubgroupsUse Condition
Objective: – Establish use conditions of life critical implantable medical devices using published literature
Output:– Extensive literature based definition of post-implant and pre-implant operating conditions - C
thermal, mechanical, electrical, environmental and biocompatibility conditionsLifetime Prediction
Objective: – Assess the contribution of each Use Condition, and combinations thereof, to enable
measurement of realistic mechanisms Accurate prediction, Consistent reportingOutput:
– Define what constitutes failure - Identify key parameter(s) controlling life times– Determine acceleration/extrapolation mechanisms - Develop methodology for experimental tests
LOT HOMOGENEITYObjective:
– Develop sampling and testing methodologies to minimize, and essentially eliminate, the introduction of defective Multi-Layer Ceramic Chip (MLCC)
Output: – Failure Mode and Effects Analysis requirements
– Statistical methods to establish quality (initial) performance– Statistical methods to establish reliability (time dependent) performance– Process capability assessment tools / process control– Acceptance criteria (within lot and lot-to-lot)
Project Deliverables
Test and Extrapolation
Methodologies
28
Test and Extrapolation MethodologiesTest and Extrapolation Methodologies
Main goal is to determine guidelines and methods to assess component reliability as related to implantable medical or other life critical applications.
Proposed solution was to develop an understanding of the expected failure modes and mechanisms, use conditions, and comparative rate of each type of failure.
Once established, test methods could be developed to hopefully accelerate the failures and be used to improve reliability of those component types that a correlation between failures and acceleration methods could be established.
Phase 1 of the MLCC component project established the data used to create the working test DOE matrix.
Phase 2 of the project will result in execution of the test DOE and establishment of the correlation between accelerated tests and failure modes. The test results will be used to generate the guidelines and recommended test methods determined to be suitable for implantable or life critical MLCC’s and subsequently other device.
Medical Failure Modes
and Effects Analysis
(FMEA) of MLCC
Failures
30
Failure Modes and Effects Analysis (FMEA)Failure Modes and Effects Analysis (FMEA)• FMEA
Procedure by which each potential failure mode for a given component within the system is analysed to determine the effects on the system, potential risk to the patient or user, failure causes, and associated prevention controls
– Failure Mode• The Failure Mode is the manner by which component failure is
observed or characterized
– Potential Effect(s)• A Potential Effect is the consequence a Failure Mode has on the
safety or functionality of the device
31
FMEA ExampleFMEA ExampleComponent Function Failure Mode Potential Effects(s) S F D Risk
IndexPotential Cause(s)
Preventio n/Control
sMedical Grade MLCC
Metal Layers – Terminations
Unsolderable plating material
Open circuit 3 2 1 6
Cold solder joint 2 2 3 12Defects in plating material
Delamination 3 2 3 18
Open circuit 3 2 1 6Defects in barrier layer Increased ESR 1 1 4 4
Voiding 1 2 2 4Poor Cohesion/Adhesion of the underlying material Ag or Cu
Open circuit 3 2 1 6
Poor shear strength
2 2 2 8
Delamination 3 2 3 18Oxidation of terminations
Poor solderability 3 2 1 6
Mechanical Dimensions out of spec Lack of mechanical fit
2 1 1 2
Open circuit 3 2 1 6Short circuit 3 1 1 3Potential arcing path
3 1 2 6
Termination bandwidth out of spec
Open circuit 3 2 1 6
Short circuit 3 1 1 3
Use Conditions for Life Critical
Medical Components
33
Use Conditions TableUse Conditions Table
Defined By:
• Mechanical• Thermal• Electrical• Environmental• Biocompatibility• Physical
34
Use Conditions Mechanical Table ExampleUse Conditions Mechanical Table Example
Review of existing and
related Standards and Test Methods
36
Examples of Standards and Test Methods ReviewedExamples of Standards and Test Methods ReviewedAEC
– ZERO DEFECTS GUIDELINE – AEC-Q004
– STRESS TEST QUALIFICATION FOR PASSIVE COMPONENTS – AEC – Q200
– FAILURE MECHANISM BASED STRESS TEST QUALIFICATION FOR INTEGRATED CIRCUITS – AEC – Q004
HDBK– MILITARY HANDBOOK - ENVIRONMENTAL STRESS SCREENING (ESS) OF ELECTRONIC EQUIPMENT -
DOD – HDBK – 344(USAF) MILITARY HANDBOOK
– DEPARTMENT OF DEFENSE HANDBOOK ENVIRONMENTM-J STRESS SCREENING PROCESS FOR ELECTRONIC EQUIPMENT THIS HANDBOOK – MIL- HDBK-2164A
– MILITARY HANDBOOK – ENVIRONMENTAL STRESS SCREENING (ESS) OF ELECTRONIC EQUIPMENT - MIL-HDBK-344Ai
GEIA - Government Electronics & Information Technology Association– ANSI/GEIA-STD-0003 Procedures for Long Term Storage and Electronic Devices
– ANSI/GEIA-STD-0005-1 Performance Standard for Aerospace and High Performance Electronic
Systems Containing Lead-free Solder
– ANSI/GEIA-STD-0005-2 Standard for Mitigating the Effects of Tin Whiskers in Aerospace and
High Performance Electronic Systems
Others
Medical Reliability
Focus on MLCC (Phase II)
38
Medical Reliability for MLCC Project (Phase II)Medical Reliability for MLCC Project (Phase II)Scope of Work: • Determination of accelerated life test methods of long term leakage and
break down failures of MLCC s • Continuation of Component Reliability Project • Specific deliverables of Phase II include the following:
– Creation of a test vehicle• Design of Test board• Fabrication of Test vehicle• Population of test board with functional MLCC’s
– Creation of fixtures and test equipment cables and peripherals at NIST– Testing of DOE variables from Phase I at NIST Boulder Facility– Completion of screening experiments at NIST. – Collection of Data and Data Mining resulting failures for trends and insight– Failure analysis of Test output “Failures as defined in Phase I”– NIST coordinate Failure Analysis Suppliers– Phase II Report
2500X
Ideal Construction:Free of VoidsGood Wetting and FilletNo detachment at CeramicUniform Coverage of Ag / Cu layers
Ideal Construction:Complete Barrier Layer (Ni)Formation of Ni3Sn4, Cu6Sn5
Question?
Do small blemishes cause reliability concerns?
“Undesirable Characteristics” in discrete components
•Plating Issues
•Many components have plating issues or shortcomings that lead to latent defects, manufacturing yield loss and scrap
•Possible Failure Locations– UBM or base layer plating– Barrier layer to base layer– Finish plating to barrier layer– Solder to barrier layer
Examples of Plating Issues
• Incomplete base layers
• Leads to the following possible defects
– Voiding– Weak solder adhesion– Exposure of base Ceramic– Micro Cracking– Layer Separation– Electrical Failure or Leakage
Although plating defects can occur anywhere, they are more common at the corners. Here the solder has breached the barrier and base layers at the corner of this capacitor
Poor coverage of the plating on the leads to short shelf life, yield loss, scrap, poor adhesion of the solder, or latent defects
Gaps in plating Ni barrier High Correlation to voiding
Examples of Plating Issues
• Poor Base Layer Adhesion
• Leads to the following possible defects
– Weak solder adhesion– Exposure of base Ceramic– Micro Cracking– Layer Separation– Electrical Failure or Leakage– Peeling of plating– Full Failure of Solder Joint
Internal DelaminationDifficult to detect
Examples of Plating Issues
•Voiding and base material adhesion issues
Examples of Plating Issues
• Poor Base Layer Adhesion• Examples – 200C reflow Sn/Pb
Good Plating Separated plating
Base layer / Barrier layer Issues
• Base Layer Separation Issues
• Cracks and separation of barrier metals can lead to full failure of the joint following subsequent assembly and handling
Base layer / Barrier layer Issues
• Base Layer Voiding
• Missing base layer areas could potentially lead to electrical problems, voids and outgassing.
0402 Capacitor
Base layer / Barrier layer Issues
• Base Layer Voiding
• Voiding can be at interface or within the base layer itself.
1206 Capacitor
Base layer / Barrier layer Issues
• Base Layer Voiding
• Missing base layer areas could potentially lead to electrical problems, voids and outgassing.
0402 Capacitor
Internal Layer Uniformity
Ceramic Cracking
• Pre Existing flaws and cracks in components can lead to post assembly failures. Especially true in large ceramic capacitor and resistor arrays
Ceramic Cracking
• Corner Cracks• Can lead to post assembly
plating failures at termination
Ceramic Cracking
• Corner Cracks• Sometime they are extremely difficult to detect or
notice in standard microscopy. Example
1206 Capacitor – Bright Field 1206 Capacitor – Dark Field
Plating Failure / Cracking
• Poor adhesion can result in separation and failure near the bottom heel fillet region. This is made worse by design limitations on PCB real estate
DOE Formation• Input from multiple signals (Field, MFG yield, Supplier Returns/ Rejects)
– Unique environmental concerns (ie, forming gas, temp, IC levels)
– Mechanism based – from Use Conditions
– Creation of Failure Pareto by NIST (NIST combined all data generically)
DOE Formation - continued
• Outputs from Sub-groups combined with failure Pareto
• Groups looked for areas Unique to Medical or areas that are known factors to cause failures
• Key differences– Reflows – multiple– Burn in – temp / bias– End use environment – mechanical high cycle low strain applications– Slightly elevated but constant temp (37C)– Shipping and storage conditions similar to portable– Constant “On” 24/7 usage– Sensitivity to flaws and small defects– Severity of failure (Could result in loss of therapy, explant or patient
harm/death)
Implanted
Medical Device Reliability Focus on MLCC
Grady White, Damian Lauria, Andy Slifka, Elizabeth Drexler Materials Reliability Division
Outline1.
Failure source identification (Pareto development)
2.
Experimental parameters3.
Experimental setup and procedure
4.
Preliminary results5.
Beginning Phase II
Pareto Development
Goal:Identify field failure sources for capacitors to obtain enough information for evaluation of proposed accelerated test parameters and procedures
Approach:-
Gather failure information from multiple (4) companies
-
Construct common terminology for all companies
-
Sort failure information into similar categories
Company A
Category % of Total
decreasing importance Complaints
Stress crack Xxx
Flex Cracking Xxx
Chipouts Xxx
Packaging Xxx
Dielectric Breakdown Xxx
Termination Defects Xxx
Physical Damage Xxx
Thermal Shock Crack Xxx
Dielectric Sheet Defects Xxx
Administrative Xxx
Poor Wetting Xxx
Mechanical Damage Xxx
Delaminations Xxx
Margins Xxx
Inconclusive Xxx
Company A (cont)
Category % of Total
decreasing importance Complaints
Plating Defects Xxx
Aging Xxx
Cap out of Tolerance Xxx
Wrong Chip Thickness Xxx
Disconnected Electrodes Xxx
Tombstoning Xxx
Pb/Sn
Termination Issues Xxx
Electrode Defects Xxx
Termination Defects Xxx
Green Chip Damage Xxx
Surface Contamination Xxx
Peeling Termination Xxx
Spattering Xxx
Glue Dot Adhesion Xxx
DF Xxx
Company B
Complaints by Root Cause Number %
Board Flex 46 Xxx
Microcrack 29 Xxx
Customer Application 19 Xxx
Design Problem 16 Xxx
Al Contamination 13 Xxx
Inhomogeneity 12 Xxx
Thermomech. Overstress 11 Xxx
Insufficient Termination 11 Xxx
No Foundings 8 Xxx
Pre-Term Damage 7 Xxx
Silicon Contamination 5 Xxx
Mechanical Damage 4 Xxx
Leakage Site 4 Xxx
Mechanical Impact 4 Xxx
Print Defect 4 Xxx
Not Determined 4 Xxx
Organic Contamination 3 Xxx
Void 3 Xxx
Thermal Overstress 3 Xxx
Contamination During B… 3 Xxx
Contamination in Termination 3 Xxx
Exposed Electrodes 2 Xxx
Electro-mech. Damage 2 Xxx
High Destruction 2 Xxx
Overloading 2 Xxx
Company C
Failure Mode Description Device #
% of Device
Total %
Failure
Leaky Ceramic Cap Xxx Xxx Xxx
Degraded, out of spec Ceramic Cap Xxx Xxx Xxx
Cracked Ceramic Cap Xxx Xxx Xxx
Loose Parts Ceramic Cap Xxx Xxx Xxx
Shorted Ceramic Cap Xxx Xxx Xxx
LeakyCeramic Cap Array Xxx Xxx Xxx
CrackedCeramic Cap Array Xxx Xxx Xxx
Solder joint, crackedCeramic Cap Array Xxx Xxx Xxx
Degraded, out of specCeramic Cap Array Xxx Xxx Xxx
DelaminatedCeramic Cap Array Xxx Xxx Xxx
Foreign material presentCeramic Cap Array Xxx Xxx Xxx
ShortedCeramic Cap Array Xxx Xxx Xxx
Solder joint, missingCeramic Cap Array Xxx Xxx Xxx
Solder joint, openCeramic Cap Array Xxx Xxx Xxx
Leaky Tantalum Cap Xxx Xxx Xxx
Degraded, out of spec Tantalum Cap Xxx Xxx Xxx
Shorted Tantalum Cap Xxx Xxx Xxx
Solder joint, electrically intermittent Tantalum Cap Xxx Xxx Xxx
Solder joint, open Tantalum Cap Xxx Xxx Xxx
Foreign material present Tantalum Cap Xxx Xxx Xxx
Loose Parts Tantalum Cap Xxx Xxx Xxx
Misassembly Tantalum Cap Xxx Xxx Xxx
Open, electrically Tantalum Cap Xxx Xxx Xxx
Undetermined Tantalum Cap Xxx Xxx Xxx
LeakyTantalum Cap Array Xxx Xxx Xxx
Overstress, otherTantalum Cap Array Xxx Xxx Xxx
UndeterminedTantalum Cap Array Xxx Xxx Xxx
Failure Mode Description Device #
% of Device
Total %
Failure
Company D
Failure Mode %
Solderability Xxx
Cracks/chipouts Xxx
End termination damage Xxx
Leakage Xxx
Capacitance value Xxx
1.
The details of the failure analysis, the terminology and the interpretation of the capacitor failures all depend upon the perspective of the company providing the failure data.
2.
The identification of the sources of capacitor failure also depends upon company perspective.3.
The identifications of various failures are not always orthogonal
Therefore, combining the four sets of data into one Pareta
is not straightforward.
Conclusions:
Table I A B C D
Cracks 31.2 21.2 1.1 33.8
Leakage 11.1 9.6 81.6 0.5
Termination defects 6.7 6.4 46.6
Solder issues 3.8 0 0.9 18.7
Mechanical treatment 8.2 3.6
Customer application 9.7 36.8 0.1
Undetermined 1.8 1.8 0.2
Drift out of specifications 3.1 0 1.6 0.5
Electrodes 1.5 0.9 0.6
After communicating with the four different companies that have provided data, the various failure mechanisms were combined into
9 categories (numbers in Table are %):
Table II A B C D
INTRINSIC
Cracks 31.2 21.2 1.1 33.8
Leakage 11.1 9.6 81.6 0.5
Termination defects 6.7 6.4 46.6
Solder issues 3.8 0 0.9 18.7
Drift out of specifications 3.1 0 1.6 0.5
Electrodes 1.5 0.9 0.6
EXTRINSIC
Customer application 9.7 36.8 0.1
Mechanical treatment 8.2 3.6
UNKOWN
Undetermined 1.8 1.8 0.2
The previous table has been sorted into intrinsic capacitor, extrinsic capacitor (e.g., device design and fabrication, handling), and undetermined failures
Cracks vs. dielectric?
Experimental Test Setup
Test Parameters:
Precondition atmosphere: air, forming gas
Vibration: none, 3 g
Precondition humidity: nothing, 85% RH/85
oC
Environmental
Vendors: 3 -
randomized
Sources
Temperature: RT, 85 oC, 125
oC, 150
oC
Frequency: DC, 1500 Hz
Voltage: 5 V, 100 V
Electrical
-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0 20 40 60 80 100 120 140 160 180 200
Temperature, C
Hea
t Flo
w, J
Current status 2: DSC measurement of 4 capacitors
Phase change occurs at about 125 oC. Calls into question the use of 125
oC
and 150 oC
temperatures for accelerated tests.
Test Equipment
Circuit elements:
- 0.1 μF capacitors
-1 MΩ
resistors
-
Power sources
-
Sorenson DC power supply
-
Keithley
263
-
Keithley
33120A (1,500 Hz)
-
96 V batteries
-
6 V batteries
Special filtered 110 V line source
Measurement apparatus:-National Instrument USB-6259
-
Single ended measurements on 32 cap coupon
-Agilent 34980A Multifunction Switch Unit- 6 34921A armature switching board-
Each board can make differential
measurements on a 32 cap coupon-
Tests on thermocouple and on precision
resistor with battery show +/-
10 nV
scatter
Test Chambers:-Associated Environmental Systems Temperature and RH chamber-Refurbished Screening Systems QRS-410T Temperature and Vibration chamber
Event Detector:
1.
Continuous –
not periodic –
monitoring2.
Rapid response
3.
Detection threshold must be pre-defined –
slow trends and sub-threshold events not detected
Series Resistor:
VRV
C
R
IdealMeasure the leakage current by measuring the voltage drop across a known resister.
Sampling process rather than continuous monitoring:-High sampling rate means we miss less but we get too much data to analyze-
Low sampling rate means we see long term
trends but most of the time we are not observing the voltage
There is no way that we will routinely catch sub-ms events using a sampling approach. If enough events occur during a measurement, we will observe an excursion.
We will see long term trends that an event detector would miss.
Series Resistor:
VRV
C
R
IdealMeasure the leakage current by monitoring the voltage drop across a know (i.e., 1 MΩ) resistor.
Measurement concern:
RC > 1 GΩ, maybe > 100 G
Therefore, VR < 0.1% and maybe <0.001% of V
i.e., it won’t take much noise to mask the signal
Series Resistor:
VRV
C
R VRV
RV
Cideal
R
RL
Less IdealIdeal
( ) ⎟⎟⎠
⎞⎜⎜⎝
⎛+=+=
++
=
++
==
V
RV
V
R
V
LV
VL
L
RR
RVRR
RRV
RR
RR
V
RRRRR
Vii
1
1statesteady
reachedhasConceii L
−== 0
-0.00025
-0.00020
-0.00015
-0.00010
-0.00005
0.000000 50 100 150 200 250 300 350 400 450 500
Arb. Units
Vol
tage
(V) Cap 1
Cap 2Cap 3Cap 4Cap 5
Close-up view of the beginning and end regions
showing the change. Note that the ordinate
axis is non-physical -
it was chosen simply to
have something to plot the voltages against while
avoiding the 8-day time gap.
Slope
-0.00004
-0.00002
0
0.00002
0.00004
0.00006
0.00008
0.0001
0 1 2 3 4 5 6
Cap #
V/un
it tim
e
BeginningEnd
Preliminary results with differential measurements made with the
NI board looked very good.
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
0.005
0 50 100 150 200 250 300 350
Time (h)
VR
If a 32 capacitor coupon is tested by the NI board, the measurements must be conducted in a single-ended, rather than a differential, mode. Under these conditions, noise from the ground overwhelmed the signals of interest.
Single ended measurements made on as-received, thermally cycled, or indented capacitors detected nothing but noise.
Measurements made on a single capacitor the had been pretreated by thermal cycling through the Curie point 10 times. The applied voltage during the test was 100 VDC.
To eliminate this problem, all measurements of test boards are being made in differential mode using the Agilent system. To further reduce the possibility of noise, DC measurements are being made with batteries as a voltage source.
Information on Boards:Cable # Coupon V T(
oC) Vib Pre-Treat Mfg Reflow
1
4 96 DC
125
no
no
B
32
5 96 DC
125
no
no
C
23
7 5 AC
125
no
no B
34
73 6 DC
125
no
no
B
25
16 5 AC
125
no
no
A
16
34 6 DC
125
no
no
C
1
Experimental parameters used on the coupons currently being evaluated
96 V, DC: Coupon 4
0
0.00001
0.00002
0.00003
0.00004
0.00005
0.00006
0.00007
9/29/20080:00
10/4/20080:00
10/9/20080:00
10/14/20080:00
10/19/20080:00
10/24/20080:00
10/29/20080:00
Time at Temperature
StD
ev (V
R/V0
)
Cap 22Cap 6
96 V DC: Coupon 5
0
0.00001
0.00002
0.00003
0.00004
0.00005
0.00006
0.00007
9/29/20080:00
10/4/20080:00
10/9/20080:00
10/14/20080:00
10/19/20080:00
10/24/20080:00
10/29/20080:00
Time at Temperature
StDe
v (V
R/V
0)
Cap 16Cap 30
Summary:
1.
To reduce noise, differential measurements are being made2.
Capacitor behavior vs. t appears to be different between manufacturers and for different applied voltages at 125 oC
3.
The power law behavior we expected to see, corresponding to damage accumulation has not yet appeared.
4.
The scatter in the data increases linearly with time for both voltages 5.
At 96 V, outliers appear, increasing in number and magnitude with time (corresponding to clusters of events?).
DOE Formation - continued• Phase 1 testing
– As a result of the DOE formation, FMEA, Pareto, and Use Conditions, preliminary experiments were performed on test boards / components. Initial testing was performed to give input to phase II PCB design, electrical hardware setup and criteria for failure.
– Measured as received and pre-stressed components. Subjected to high temp and bias to determine leakage breakdown.
– Performed at NIST Boulder
• DOE Completion (128 Cell Matrix)
– Optimized to balance variables (Texas Instruments performed the analysis using Jump Software)
– Main variables are: (Temp, Vibration, forming gas, Voltage, Voltage type AC/DC, number of reflows)
86
MLCC DOE Project (Phase II)MLCC DOE Project (Phase II)Objective: • The Medical Component Reliability team determined X number of
environmental conditions were wanted to investigate in the DOE:– Temperature– Voltage Level– Type of Voltage– Atmosphere Precondition– Humidity Precondition– Vendor– Number of Reflows
• 128 run DOE is required to determine all main affects and two way interactions
– Team chose a 0.1uf 16V X7R MLCC in an 0603 package– 128 Serialized coupons were built to support DOE. Each coupon having 32
components for electrical testing and 3 components for mechanical testing
Medical Reliability Focus on MLCC
(Phase II)
Jerry Peasley Micro Systems
Engineering Inc.November 14, 2008
Medical Reliability Focus on MLCC
(Phase II)
Jerry Peasley November 14, 2008
89
Medical Reliability for MLCC Project (Phase II)Medical Reliability for MLCC Project (Phase II)
Objective: Establish a mechanistic test strategy for lot based anomalous behavior detection of MLCC capacitors
Approach:Designed experiment to identify key factors (based on FMEA and Use Conditions, part of Phase I)Custom test vehicle design and fabrication - assembly of MLCC components from multiple suppliers using industry standard methodsDevelopment of automated testing infrastructure (fixtures, equipment, cables and peripherals) over 15k data points per test instanceTest and analyze data
o Failure Analysis with Supplierso Identification of test methodology with consensuso Phase II Report
90
MLCC Project DOE MLCC Project DOE –– StatusStatus
• Electrical Test– All 128 coupons have completed initial electrical test– Parameters tested:
• Insulation Resistance• Capacitance• Dissipation Factor
• Post Reflow Test is also completed on all 128 coupons– Attend November 14, 2008 meeting at FDA to hear preliminary
results
91
MLCC Project DOE MatrixMLCC Project DOE Matrix
Variables and Levels based on FMEA and Medical use conditions
Couponnumber= row
number X1=Vibration X2=Temperature X4=Voltage X6=Cyclic_Voltage X5=Atmosphere_Precondition X3=Humidity_Precondition X7=Vendor X8=Number_of_reflows1 YES 125 C 96 volts 1500,50% Biphasic Forming Gas NONE Vendor C 3 reflows2 YES 85 C 6 volts DC NONE NONE Vendor A 3 reflows3 YES 125 C 6 volts DC Forming Gas NONE Vendor C 3 reflows4 NO 125 C 96 volts DC NONE NONE Vendor B 3 reflows5 NO 125 C 96 volts DC NONE NONE Vendor C 2 reflows6 YES 85 C 6 volts DC Forming Gas 85 / 85 Vendor C 3 reflows7 NO 125 C 6 volts 1500,50% Biphasic NONE NONE Vendor B 3 reflows8 YES 85 C 6 volts DC NONE 85 / 85 Vendor C 2 reflows9 YES 125 C 6 volts 1500,50% Biphasic NONE 85 / 85 Vendor A 3 reflows
10 YES 125 C 96 volts DC Forming Gas 85 / 85 Vendor B 3 reflows. . . . . . . . .. . . . . . . . .
128 YES 85 C 96 volts DC Forming Gas NONE Vendor C 2 reflows
Focus area
92
Initial Test SetupInitial Test Setup
Design
Automated Test
Test Rack
Nearly 15,000 Datapoints
Assembly- Multiple suppliers- Multiple reflows
93
Test ParametersTest Parameters
• MLCC – 0.1uf 16V ±10% X7R• Capacitance and Dissipation Factor
• 1khz 1Vrms• DC Leakage
• 16V – measured at 10, 15, 30, 45 and 60 seconds• Converted to IR
• Reflow Environment
94
Capacitance (Farads)Capacitance (Farads)
8.75E-08
9.25E-08
9.75E-08
1.03E-07
1.08E-07
1.13E-07
A B C
All Data Grouped by Vendor
Df
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
A B C
95
Dissipation Factor (%)Dissipation Factor (%)
All Data Grouped by Vendor
Outlier
All Data Grouped by Vendor with outliers removed
Df
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
A B C
96
Dissipation Factor (%)Dissipation Factor (%)
All Data All Data with outliers removed
97
Insulation ResistanceInsulation ResistanceIR
1.00E+08
1.00E+09
1.00E+10
1.00E+11
1.00E+12
1.00E+13
A B C
Outliers
Insulation Resistance (ohms)Insulation Resistance (ohms)
98
All Data All Data with outliers removed
Insulation Resistance (ohms)Insulation Resistance (ohms)
99
Next StepsNext Steps
• Complete DOE and analysis at NIST Boulder Facility• Complete Failure Analysis with Suppliers• Identify key test for anomalous behavior detection• Publish Phase II report
