Failure into the Design Process September 15, 2016 9000 Virginia - - PowerPoint PPT Presentation

▶
failure into the design
SMART_READER_LITE
LIVE PREVIEW

Failure into the Design Process September 15, 2016 9000 Virginia - - PowerPoint PPT Presentation

Dec 21, 2023 126 likes •610 views

Implementing Physics of Failure into the Design Process September 15, 2016 9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com Nathan Blattau, Ph.D. Senior Vice President of DfR Solutions, has been

slide-1
SLIDE 1

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Implementing Physics of Failure into the Design Process

September 15, 2016

slide-2
SLIDE 2

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Nathan Blattau, Ph.D.

Senior Vice President of DfR Solutions, has been involved in the packaging and reliability of electronic equipment for more than ten years. His specialties include best practices in design for reliability, robustness of Pb-free, failure analysis, accelerated test plan development, finite element analysis, solder joint reliability, fracture, and fatigue mechanics of materials.

2

slide-3
SLIDE 3

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Role of Modeling in Design

  • Nobody can afford to repeatedly test and redesign to create reliable, cost

effective products

  • Working with models allows an interdisciplinary design team to create a more

reliable design smarter, faster & cheaper!

3

slide-4
SLIDE 4

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Simulation & Modeling

4

  • Performing thermal, mechanical &

electrical simulations & extracting the results into a time-to-failure prediction

slide-5
SLIDE 5

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Physics of Failure (PoF)

5

  • PoF Definition: Use of science to capture an understanding of failure

mechanisms & evaluate useful life under actual operating conditions

  • Using PoF, design, perform, and interpret the results of accelerated life

tests

  • Starting at design stage
  • Continuing through lifecycle of the product
  • Start with standard industry specifications
  • Modify or exceed them
  • Tailor test strategies specifically for product design & materials, use

environment, and reliability needs

slide-6
SLIDE 6

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Physics of Failure Definitions

6

  • Failure of a physical device or structure attributed to
  • Gradual or rapid degradation of the material(s) in the device
  • In response to the stress or combination of stresses the device is exposed

to, such as:

  • Thermal, Electrical, Chemical, Moisture, Vibration, Shock, Mechanical Loads . . .
  • Failures May Occur:
  • Prematurely
  • Gradually
  • Erratically
slide-7
SLIDE 7

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Using Physics of Failure During the Design Stage

7

  • Design the product for robustness and to meet the

environmental requirements

  • Vibration
  • Mechanical Shock
  • Thermal Cycling
slide-8
SLIDE 8

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Designing for Mechanical Loads

8

  • Unlike other materials, solder is a poor engineering

material

  • Extreme dimensional variations
  • Presence of voids is normal and expected
  • Is constantly being subjected to inelastic deformations

under thermal cycling and shock

  • You would never use steel, titanium or aluminum under these

types of conditions, unless you want it to fail

  • During vibration we need to prevent inelastic deformations

(plasticity)

  • This makes the field of electronics reliability unique
slide-9
SLIDE 9

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Vibration Fatigue

9

  • Due to the high number of cycles we need to avoid

inelastic deformations at all cost

  • Inelastic deformations (plasticity and creep) are low

cycle fatigue (< 100,000 cycles)

  • During vibration cycles accumulate quickly
  • Example, 100 Hz vibration – 100 cycles per second
  • Time to accumulate 100,000 cycles, 16.67 minutes
slide-10
SLIDE 10

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Designing for Vibration

10

  • Octave rule: the PCB natural frequencies should be at

least 2X the chassis natural frequencies to prevent coupling

  • Recommended reading Steinberg’s Vibration Analysis
  • f Electronic Equipment
  • If the chassis resonant frequency is close to the PCB then

there can be significant amplification of the PCB deformations

slide-11
SLIDE 11

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Natural Frequency

11

  • Do it by hand
  • Limited shapes
  • Simple support

conditions

  • Use FEA to handle

complex shapes and boundary conditions

Vibration Analysis of Electronic Equipment David S. Steinberg

slide-12
SLIDE 12

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Using Sherlock or FEA for Vibration During Design Phase

12

  • There are many factors that can be adjusted to modify

the natural frequency response of the printed circuit board

  • Component placement
  • Boundary conditions (mount points)
  • PCB properties (thickness, laminate)
slide-13
SLIDE 13

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Boundary Conditions Component Mass – Heatsinks Printed Circuit Board Properties Typically, the higher natural frequency the more robust the design

13

slide-14
SLIDE 14

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Boundary Conditions

14

  • Chassis typically have

lower natural frequencies than circuit boards

  • Usually looking for circuit

boards having natural frequencies greater than 150 Hz

  • Natural frequency should

not coincide with peaks in the expect vibration input

23 Hz is too low for most applications, need to make changes

slide-15
SLIDE 15

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Boundary Conditions

15

  • Current, PCI-E type
  • Add an additional

mount at high deflection area

Add more support

slide-16
SLIDE 16

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Modifying Boundaries in Sherlock

16

slide-17
SLIDE 17

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com 17

  • Almost a 4X

increase with

  • ne additional

mount

  • Changing

heatsink from Copper to Aluminum

slide-18
SLIDE 18

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com 18

  • Heatsink from

Copper to Aluminum

  • 82 to 93Hz
slide-19
SLIDE 19

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Additional Mount Point and Increasing PCB Thickness

19

1.544 to 1.65 mm NF increases to 108.5 Hz

slide-20
SLIDE 20

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Conduct Physics of Failure Assessment

20

  • Once an acceptable NF is achieved
  • During vibration the assembly is assumed to deform

elastically so that strain on the PCB is proportional to the strain in the solder and leads

  • This allows PCB strain to be used to make fatigue

predictions

slide-21
SLIDE 21

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Vibration Fatigue

21

  • HCF failures typically occur in the lead or solder joint

Component Motion Board bending

slide-22
SLIDE 22

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com 22 22

  • Lifetime under vibration

is divided into two regimes

  • Low cycle fatigue (LCF)
  • High cycle fatigue (HCF)
  • LCF is driven by inelastic strain

(Coffin-Manson)

  • HCF is driven by elastic strain

(Basquin)

 

b f f e

N E 2   

 

c f f p

N 2   

  • 0.5 < c < -0.7; 1.4 < -1/c > 2
  • 0.05 < b < -0.12; 8 > -1/b > 20
slide-23
SLIDE 23

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com Harmonic Steinberg D.S. Vibration analysis for electronic equipment. John Wiley & Sons, 2000. Random MIL-STD-810G Figure 514.6C-1 US Highway truck vibration exposure 1 hour is equivalent to 1000 miles

Typical Vibration Levels

23

Enter profile into Sherlock

slide-24
SLIDE 24

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Physics of Failure Results – Random Vibration

24

slide-25
SLIDE 25

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Random Vibration Results

25

  • U1 is failing because

it is in areas of high bending (red)

  • Further design

changes are necessary to get this board to survive the expected field environment

  • Additional mount

points, stiffeners, etc..

slide-26
SLIDE 26

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Mechanical Shock

26

  • Very similar to vibration
  • The higher the board stiffness

(Natural Frequency is directly related) the more robust with regards to mechanical shock

  • Lower component mass, Increase board

thickness

  • Due to today’s low profile

surface mount components, shock failures are primarily driven by board flexure

  • BGAs don’t care about in-plane

shock, unless it causes the board to bend

  • Shock tends to be an overstress

event (though, not for car doors)

  • Failure distribution is ‘random’
slide-27
SLIDE 27

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

In Plane Shock

27

slide-28
SLIDE 28

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Design - Board Thickness Effects

Board thickness 1.575 mm 0.97 mm displacement Board thickness 1.836 mm 0.68 mm displacement

28

Board thickness 2.285 mm 0.41 mm displacement

slide-29
SLIDE 29

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Using Sherlock During Design for Thermal Cycling Fatigue

29

Examples

  • Package selection
  • Printed circuit board properties
  • Solder pad design
  • Plated through hole
slide-30
SLIDE 30

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Predictive Models: Physics of Failure (PoF)

30

  • Modified Engelmaier for Pb-free Solder (SAC305)
  • Semi-empirical analytical approach
  • Energy based fatigue
  • Determine the strain range (Dg)
  • C is a correction factor that is a function of dwell time and

temperature, LD is diagonal distance, a is coefficient of thermal expansion (CTE), DT is temperature cycle, h is solder joint height

T h L C

s D

D D  D a g

slide-31
SLIDE 31

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Predictive Models: Physics of Failure (PoF)(cont.)

31

  • Determine the shear force applied to the solder joint
  • F is shear force, L is length, E is elastic modulus, A is the area, h

is thickness, G is shear modulus, and a is edge length of bond pad

  • Subscripts: 1 is component, 2 is board, s is solder joint, c is bond

pad, and b is board

  • Takes into consideration foundation stiffness and both

shear and axial loads

 

                         D   a G G A h G A h A E L A E L F L T

b c c c s s s

9 2

2 2 1 1 1 2

 a a

slide-32
SLIDE 32

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Predictive Models – Physics of Failure (PoF)(cont.)

32

  • Determine the strain energy dissipated by the

solder joint

  • Calculate cycles-to-failure (N50), using energy

based fatigue models

 

1

0019 .

D   W N f

s

A F W  D   D g 5 .

slide-33
SLIDE 33

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Thermal Cycling Design – Component Choices

33

  • Plastic components typically perform better than ceramic

components

  • Smaller components usually perform better than larger

components

  • Larger solder joints (pad size, thickness) perform better
slide-34
SLIDE 34

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

2512 resistor or two 1206 size resistors?

34

2512 1206

slide-35
SLIDE 35

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Bond Pad Influence 2512 Resistor

35

  • Increase bond

pad length from 1.28 mm to 2.00 mm

  • 862 cycles
  • 1296 cycles
  • 54% increase in

life

slide-36
SLIDE 36

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

PCB Influence 2512 Resistor

36

  • Decrease PCB CTE

from 16 ppm to 15 ppm

  • 1296 cycles
  • 1632 cycles
  • 26% increase in

life

  • 14 ppm - 2117

cycles, 63% increase over baseline

slide-37
SLIDE 37

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Plated Though Holes

37

  • Design influences
  • PCB thickness
  • Plating thickness
  • Hole diameter
slide-38
SLIDE 38

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

IPC TR-579

38 

Round Robin Reliability Evaluation of Small Diameter (<20 mil) Plated Through Holes in PWBs

Activity initiated by IPC and published in 1988

Objectives

Confirm sufficient reliability

Benchmark different test procedures

Evaluate influence of PTH design and plating (develop a model)

slide-39
SLIDE 39

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

  • Determine applied stress applied (σ)
  • Determine strain range (∆ε)
  • Apply calibration constants
  • Strain distribution factor, Kd(2.5 –5.0)
  • PTH & Cu quality factor KQ(0 –10)
  • Iteratively calculate cycles-to-failure (Nf50)

Plated Through Hole Via Barrel Cracking Fatigue Life Based On IPC TR-579

39

slide-40
SLIDE 40

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

PoF Durability/Reliability Risk Assessments PCB Plated Through Hole Via Fatigue Analysis

40

When a PCB experiences thermal cycling the expansion/ contraction in the z-direction is much higher than that in the x-y plane. The glass fibers constrain the board in the x-y plane but not through the thickness. As a result, a great deal of stress can be built up in the copper via barrels resulting in eventual cracking near the center of the barrel

slide-41
SLIDE 41

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

IMEC Plated Through Hole Fatigue Model

41

  • Alternative model
  • Better accuracy when predicting

fatigue of large plated through holes

  • Less reliance on correction coefficients
  • TR-579 can be overly conservative in

certain cases

Kd and Kq are IPC-TR-579 correction factors

slide-42
SLIDE 42

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Plating Thickness

42

20 microns – 1621 cycles 30 microns – 2166 cycles

slide-43
SLIDE 43

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Printed Circuit Board Thickness

43

1.6 mm – 2166 cycles 2.0 mm – 1623 cycles

slide-44
SLIDE 44

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Printed Circuit Board Expansion (z-axis)

44

60 ppm – 1293 cycles 45 ppm – 2984 cycles

slide-45
SLIDE 45

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Accuracy of PoF-Based Models

45

  • Once a physics of failure

model has been developed and validated, it typically displays accuracy similar to a validated finite element model

slide-46
SLIDE 46

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com 46

Tradi aditio tional al Reli liab abil ility ity Grow

  • wth

th in Prod

  • duct

uct Deve evelopment lopment

Empir irical ical “TRIAL & ERROR” Method to Demonstrate Stati atistica tical l Confide fidence nce

Today, This Reactive Approach Is Not Enough!

  • Testing doesn’t truly simulate actual usage.
  • Can not afford the time or money to test to high reliability.
  • Problems found too late for effective corrective action, quick fixes often used.
  • Testing more parts & more/longer tests “seen as only way” to increase reliability.

DESIGN - BUILD - TEST - FIX (D-B-T-F) 6) REPEAT 3-5 Until Nothing Else Breaks Or You Run Out Of Time/Money.

Yes No

4) Faults Detected ? 5) Fix Whatever Breaks. 2) Build 3) Test 1) Design

46

slide-47
SLIDE 47

9000 Virginia Manor Rd Ste 290, Beltsville MD 20705 | 301-474-0607 | www.dfrsolutions.com

Traditional Reliability Growth in Product Development

Empirical “TRIAL & ERROR” Method to Demonstrate Statistical Confidence

47

Implementing Physics of Failure gets you through the loop the first time

47

DESIGN - BUILD - TEST - FIX (D-B-T-F)

Yes No

4) Faults Detected ? 5) Failure Analysis 2) Build 3) Test 1a) Design 1b) PoF 6) Fix and PoF