burn-in & test socket workshop - testconx · professional services are needed costly-1. page 7...
TRANSCRIPT
March 4 - 7, 2001Hilton Mesa Pavilion Hotel
Mesa, Arizona
Sponsored By The IEEE Computer SocietyTest Technology Technical Council
Burn-in & TestSocket Workshop
IEEE IEEE
COMPUTERSOCIETY
COPYRIGHT NOTICE• The papers in this publication comprise the proceedings of the 2001BiTS Workshop. They reflect the authors’ opinions and are reproduced aspresented , without change. Their inclusion in this publication does notconstitute an endorsement by the BiTS Workshop, the sponsors, or theInstitute of Electrical and Electronic Engineers, Inc.· There is NO copyright protection claimed by this publication. However,each presentation is the work of the authors and their respectivecompanies: as such, proper acknowledgement should be made to theappropriate source. Any questions regarding the use of any materialspresented should be directed to the author/s or their companies.
Session 7Wednesday 3/07/01 8:00AM
Advancements In Socket Products
PASS ChipMin Cho Inabata America Corporation
Evaluating Elastomer For High Density BGA SocketIla Pal - Ironwood Electronics, Inc.
Low Cost Thermal Management Using Compliant Thermal Interface MaterialsNancy Dean Honeywell Electronic Materials (Presenter)
Kenichiro Fukuyama - Honeywell Electronic Materials
Technical Program
Page 1
2001 Burn-in and Test Socket Workshop 2001 Burn-in and Test Socket Workshop
Manufactured by
Corporation
PASS CHIPPASS CHIPPASS CHIP
Distributed by
Corporation Inabata America
March 4 - 7, 2001
Presented By : Min Cho
Page 2 inabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabata
MissionBased on years of experience and survey, we havesuccessfully invented the most effective, simple, and leastcostly tools for electrical contact cleaning: PASS CHIPproducts.
To be a unique providerof the best solution for thecontact problems whichinevitably arise with allkinds of electronic devicesto be seen everywheretoday.
To contribute toward moreproductivity yielded by moreeffective and less costly productionin the electronics industriesworldwide, resulting in higherquality goods and higher cost-performance that can morecomfortably surround and furtherenrich the daily lives of people.
Page 3 inabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabata
Market Outlook -1The following facts will dramatically expand the market of contact cleaners.
Semiconductors are in huge demand
The IT revolution iscreating a large varietyof applications ofpowerful semiconductorproducts.
PCs, Mobile Phones,Video Cameras, DVDplayers, CommunicationDevices, HomeAppliances, etc...
Estimation of monthly worldwideproduction of semiconductors
Year 2000 2003 2005Number of Unit 30,000 60,000 100,000(in Millions)
This creates a need for morefrequent contact cleaning!
Page 4 inabatainabatainabatainabatainabatainabatainabatainabata
Market Outlook -2Lighter, more compact packages of semiconductorswill inevitably follow.
It is a fact: Electronic commodity goods areproduced lighter, more compact, more ergonomicwith each new generation. Newer types ofsemiconductors, such as FBGA & CSP willbecome more prevalent.
This creates the need for more advancedcontact cleaning methods.
Page 5 inabatainabatainabatainabatainabatainabatainabatainabata
Market Outlook -3More productivity and more effectiveness inevery process of production are needed
As the usage of mass-production increases, our testingprocess must become more effective.An efficient contact cleaning operation should be easily donewithout a burden of cost.
This creates the need for an easier &cheaper method for contact cleaning
Page 6 inabatainabatainabatainabatainabatainabatainabatainabata
Before
Ultrasonic-Wave cleaningThe outside cleaning companyusually needs 4-5 days for cleaningtime, and this requires substitutes ofburn-in or final testing board
Costly, time consuming,troublesome in-outhandling
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPThe semiconductor industry has been looking for abetter solution of contact cleaning for years.The following methods of contact cleaning arecurrently widely used:
Professional services are needed
Costly
-1
Page 7 inabatainabatainabatainabatainabatainabatainabatainabata
Before
Brush cleaningManual brushing takes alot of time
Time consuming,less productive
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Air blow cleaningSolder residues cannotalways be completelyremoved
Less effective
-2
Page 8 inabatainabatainabatainabatainabatainabatainabatainabata
Before
Replacement of contact pins instead of cleaning
Re-manufacturing thecontact pins is costly
Expensive
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Replacing the pinsrequires a professionalservice
Costly, time consuming
-3
Page 9 inabatainabatainabatainabatainabatainabatainabatainabata
With
Proven effectiveness in cleaning
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
" PASS CHIP " has created a perfect solution forcontact cleaning to eliminate all sort of headachesthat keep annoying production managers.
Major manufacturers in the semiconductor industry havethoroughly evaluated "PASS CHIP" models and been wellsatisfied with its effectiveness in cleaning.
-1
Page 10 inabatainabatainabatainabatainabatainabatainabatainabata
With PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Evaluation Test of Cleaning operation
Package Used : TSOP2
Burn-in Boards : 48 sockets x 66boards
Evaluation method :
-2
Comparedoccurrence ofdefective devicesbefore and afterPASS Chipapplication
Page 11 inabatainabatainabatainabatainabatainabatainabatainabata
With PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Evaluation Test Results (66 Boards) Before Cleaning After Cleaning
Board Good Defective Ratio Good Defective Ratio1 42 6 12.5% 48 0 0.0%2 42 6 12.5% 48 0 0.0%3 45 3 6.3% 48 0 0.0%4 46 2 4.2% 47 1 2.1%5 43 5 10.4% 47 1 2.1%6 44 4 8.3% 48 0 0.0%7 40 8 16.7% 47 1 2.1%
64 37 11 22.9% 45 3 6.3%65 39 9 18.8% 46 2 4.2%66 43 5 10.4% 48 0 0.0%
Total 2667 503 15.9% 3109 59 1.8%
-3
Page 12 inabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabatainabata
With
Ease of usage
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
"PASS CHIP" can be easily placed on the contact pinsmanually or by a vacuum pad, and requires NO professionalservices for effective cleaning.
Low cost
Cleaning your contactors with a disposable "PASS CHIP"requires no additional costs.
-4
Page 13 inabatainabatainabatainabatainabatainabatainabatainabata
With
Short cleaning time
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
One or two wipes with "PASS CHIP" can instantlyremove the solder residues or dirt from the contactpins, which results in shorter down time and a moreproductive testing process.
-5
Page 14
1. Flat sheet model for TSOP / QFP type of socket(Single coated or Double coated)
2. Cubic sheet model for TSOP / QFP type of socket
3. Pad model for CSP / BGA type of socket
inabatainabatainabatainabatainabatainabatainabatainabata
Products for IC Socket contact cleaning
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Useful both in a burn-in and a final test process as well as in awriting process of SRAM or Flash Me
Products-1
Available to any type of socket, i.e. OPENTOP, LID or LIDLESS.
Available to any customer specifications.
Page 15 inabatainabatainabatainabatainabatainabatainabatainabata
Product for Probe Card contact cleaning
PASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIPPASS CHIP
Available to any customer specifications
Products-2
Product for DIMM contact cleaningAvailable to any customer specifications
1
Evaluating Elastomerfor High Density BGA
Socket
Ila Pal
2
Agenda Socket design Tolerance analysis Elastomer properties Electrical characteristics Mechanical characteristics Transmission line theory Conclusions
3
Socket Design
Socket Base
Target P CB with S MT pads
SocketTop Assembly
Compression Screw
Compression plate
BGA Package
Z-Axis Elastomer
Machine Screw
Socket Lid
Machine Screw
Alignment plate
4
Alignment Plate Design
A
C
B
D
FE
A = Ball diameterB = Hole to edge of alignment plateC = Hole to PCB padD = PCB pad to PCB padE = Ball to ballF = Edge of ball to edge of alignment plate
Alignment plate
Solder ball
PCB pad
5
Z-Axis Conductive Wire on Elastomer
Low-resistance (<0.01Ω) connector 30 Micron diameter gold plated brass wires Wire density - 0.1mm or less 63° angled wire embedded in silicone Insulation resistance: 1000 MΩ @500V DC 50mA per filament Operating temperature: -35° to 100° C
6
SEM Picture of Elastomer
7
Solder Ball on Elastomer
0.75mm
8
Elastomer before Compression
9
Elastomer after Compression
10
Surface Effect on Solder Ball
11
Resistance Test Setup
R
Gold plated pin electrode
Elastomer
Test board
Compress
Gold plated pin electrode
HP 4338A
12
Compression Vs Resistance
0 10 2030 40 50 60 70 80
100
0.0 0.1 0.2 0.3 0.4 0.5
Compression (mm)
Cont
inui
ty R
esist
ance
(mΩ
)
2.0 T
1.0 T
0.5 T
13
Compression Load Test Setup
Compression point
Load Cell Compression rate 0.5mm/min
Flat boards
Elastomer
Compression length = T Compression Load
14
Compression Vs Load
0 20 40 60 80
100 120 140 160 180
0 0.05 0.1 0.15 0.2 0.25 0.3 35 0.4 0.45 0.5
Compression (mm)
Com
pres
sion
Load
(gf/
mm
2 )
0.5 T
1.0 T
2.0 T
15
Reliability Test Setup
0.5mm round Gold plated pin
Au Plated Electrode
Elastomer thickness 1mm
Compression 0.3mm
R
Compression cycle
Compression 1 Sec
Open 1 Sec
16
Endurance Characteristics
-100
-80
-60
-40
-20
0
20
40
60
80
100
1 20000 40000 60000 80000 100000 The number of Compression
Cha
nge
of R
esist
ance
(mΩ
)
17
Test Setup for Measuring Current
φ1.0mm Au plated electrode
Au sheet electrode
Thermal couple Elastomer
Compress
18
Current Carrying Capacity
25 26 27 28 29
30 31 32 33 34 35
0 1 2 3 4 5 6 7 8 9 10 Time (min)
Tem
pera
ture
(deg
C)
10.0A 5.0A
19
Transmission Line Types
PARALLEL-PLATE TRANSMISSION LINE
TWO-WIRE TRANSMISSION LINE COAXIAL TRANSMISSION LINE
2a
2bD
2a
w
d
20
Equivalent Circuit of Transmission Line
+
-∆ z
v (z , t )
R L
CG
∆ z
∆ z∆ z
∆ z
i( z , t) i ( z + , t)∆ z
v ( z + , t)∆ z
P a r a m e t e r P a r a lle l- P la te L in e
T w o - W ir e L in e C o a x ia l L in e U n it
R sR
w2
πaR s
+ba
R s 112π
Ω /m
L wd
µ
πµ
c o sh -1
aD2
πµ
2ln
ab
H /m
G dw
σ πσ
c o sh -1
aD2
2 π σ
ln
ab
S /m
C dw
∈ π ∈
c o sh -1
aD2
2 π ∈
ln
ab
F /m
Source: Fundamentals of Engineering Electromagnetics - David K. Cheng
21
Results
For 0.3mm thick elastomer: Capacitance = 0.006 pF Inductance = 0.06 nH Resistance < 0.010 Ω
22
Surface Mount Socket
Solderless Socket
23
Conclusions Simple design of socket Easy manufacturability of components Better electrical and mechanical characteristics Low compression force per solder ball Damage free solder balls Fast, dense and durable High frequency IC prototyping applications Field upgradeable systems applications
ITherm20005/24/00
NFD
Low Cost Thermal Management using Compliant Thermal Interface Materials
Nancy DeanKenichiro Fukuyama
Honeywell Electronic Materials
BiTS20011/05/01
SIA Roadmap - Projected Power Dissipation
0255075
100125150175200
1995 2000 2005 2010 2015
Sing
le C
hip
Pack
age
Pow
er (W
)
Cost PerformanceHigh Performance
Power Dissipation at Burn-in and Test increases
- Thermal Management becomes more of an issue.
BiTS20011/05/01
Thermal Resistance Calculations
Die (device)
Heat Sink
Tjunction
Tambient
THS base
Tdie backside
θHeat Sink
θInterface
θDie
Total Thermal Resistance =(Tjunction - Tambient)/Power
High interface thermal resistance means Heat Sink thermalresistance must be lower (i.e., heat sink larger or more expensive)
BiTS20011/05/01
Heat Sink Thermal Resistance
As a general rule, heat transfer from a heat sinkscales with fin area.
Heat spreading on heat sink base becomesimportant for large heat sinks.
High density fins (large number of tall fins per unitlength) can be expensive to manufacture.
Want to minimize heat sink size to allow greaterpacking of sockets.
Lowering thermal resistance in the rest of thesystem allows a less expensive heat sink to beused.
BiTS20011/05/01
Why is an Interface Material Needed?
< 2% of area is in metal tometal contact, even forvery smooth surfaces.
Heat conduction acrossair gap is poor.
Real Surfaces are not smooth
BiTS20011/05/01
Why is an Interface Material Needed?
Gaps in interface dueto non-planarity.
Gaps can be quitelarge.
Conduction acrossgap is poor.
Surfaces are not Planar
Thermal resistance of InterfaceMaterial is
θθθθ = (Tsink - T die)/Q
BiTS20011/05/01
Use of a Thermal Interface Material During Burn-in
Ideally, you would use no interface material No need to clean No need to replace material after each cycle.
But, Then Dont have cushioning for die. Usually must polish heat sink base. More of the thermal budget will be occupied by interface
thermal resistance →→→→ larger heat sink.
Compromise by using an interface material thatlasts many cycles, performs well under burn-inconditions but does not leave a residue.
BiTS20011/05/01
Characterizing interface materials for Burn-in Use
Desired Properties Typical available Information
Low Thermal resistance Thermal resistance at one pressure
High Mechanical Compliance (sometimes) mechanical compliance data.
Robust enough to last severalhundred + cycles
No cycling info
Leaves no residue on die thatmust be cleaned.
No residue info
Easy to use and apply Some use info
Withstands high temperatureburn-in environment with nodegradation in properties fromlower temperature
No measurement of high temperatureproperties.
BiTS20011/05/01
Materials Used in Characterization Tests
Material DescriptionA Silicone Gel material with conductive filler,
0.010 thickB Elastomeric Pad with thermally conductive
filler, 0.020 thickC Graphite foil, 0.020 thickD Composite of aligned carbon fibers in polymeric
matrix [2], 0.020 thickD1 Improved version of material D, 0.020 thick
BiTS20011/05/01
00.05
0.10.15
0.20.25
0.30.35
0.40.45
0.5
0 50 100 150 200 250 300
Test Pressure (psi)
Ther
mal
Res
ista
nce
(Cin
2/W
)Pressure Sensitivity of Thermal Perf.
Increasing Pressure
Decreasing Pressure
Cut Bar Test
Ensure Material Performs Well at pressure in burn-in
BiTS20011/05/01
Benefits of Having a Compliant Interface
Force across die surface and thermalcontact can be more uniform.
Die surface is cushionedwill not scratch
Can absorb tolerances if die height fromdevice to device is different
BiTS20011/05/01
BiTS20011/05/01
Mechanical Compliance
Many material suppliers report some mechanical compliance.
Need enough compliance at socket pressure to overcomemechanical tolerance stackups.
0
5
10
15
20
25
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007
Displacement (in)
Appl
ied
Pres
sure
(psi
)
Material A
Material B
Material C
Material D
BiTS20011/05/01
Thermal Performance At Elevated Temperature
In general, you want thermalperformance reported as afunction of pressure.
Some materials may performdifferently at elevatedtemperatures, socharacterization should bedone at use temperature.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 10 20 30 40 50Pressure (psi)
Ther
mal
Res
ista
nce
(C in
2/W
)
Material A (70C)
Material A (130C)
Mat. D 70 C
Mat. D 130C
LowerTemperature
HigherTemperature(separation ofparticles).
BiTS20011/05/01
Filmy Residue left on Die
(e)
Residue Left on die can causeproblems with downstreamprocesses marking bonding thermal interface application
Residue must be cleaned, ifpresent.
Metric for characterizing residueis a visual since amount ofresidue is usually too little todetect otherwise.
Some materials leave an oily film.
BiTS20011/05/01
Flake type Residue Left on DIe
Some interface materialslost chunks of material onthe die Chunks could migrate off
die and cause problemselsewhere.
Interface materialperformance will degrade asthe material falls apart.
BiTS20011/05/01
Residue Left on Die
InterfaceMaterial
300 pressurecycles at roomtemperature
20 tempcycles,25C-150C
Bake48 hours at150°°°°C
Material A Residue Residue Residue, sticksto die.
Material B Residue Residue ResidueMaterial C No Residue Flaking Residue,
FlakingMaterial D No Residue No Residue Residue
Poor adhesionMaterial D1 No Residue No Residue No Residue
Good adhesion
BiTS20011/05/01
Material Robustness
Want interface material tolast many cycles without adegradation in performance. Cycle several hundred to
several thousand on-offpressure cycles.
Use application temperature Use application pressure
Cycling tests can be difficultor time consuming to make try to correlate other tests
methods or materialsproperties.0.00
0.020.040.060.080.100.120.140.160.18
0 200 400 600 800 1000
Cycles
Ther
mal
Res
ista
nce
(Cin
2 /W)
Sample 1, Material D1
Sample 2, Material D1
Vertical Motionon and off Die
Heat Sink
Heater
ThermocoupleHoles
Interface Material
BiTS20011/05/01
Robustness - Material Strength
Do a tensile test onmaterial pull a 0.5 wide strip to
breaking. Record load at failure
Tests had similar resultsat room temperature andelevated temperature
Results did not correlatewith robustnessbehavior.
Tensile Test
0
1
2
3
4
Material D Material B
Poun
ds Room Temp150 C
BiTS20011/05/01
Robustness - Abrasion Resistance
Cycling, particularly witha clamshell socket, ismore represented byabrasion.
Standard abrasion testsare too severe.
Tried to automate anabrasion tests using asoft, foam brush. Materials that tended to
flake or tear in cycling failedhere
0.100
Interface Material
Instron (vertical motion)
Heat Sink
Foam Brush
BiTS20011/05/01
Cost Advantages in Using an compliant interface
Use lower thermal resistance heat sinkSmaller heat - can put more sockets per boardReduce fin height - reduce distance between
boards/racksReduce heat sink aspect ratio (fin packing) leads
to a less expensive heat sink
Yield improvementsmechanical cushioning of die will reduce
breakage or damage.
BiTS20011/05/01
Summary
Using an interface material for a burn-in thermalsolution may enable lower cost heat sinks to be used.
Burn-in is a severe environment for an interfacematerial.
Conventional interface tests do not tell everything thatis needed to predict success in a high temperature,repetitive assembly environment.
Conventional tests that are of interest are Thermal resistance as a function of pressure Mechanical compliance Bake tests
BiTS20011/05/01
Summary
New characterization methods that did correlatewith burn-in suitability High temperature bake in contact with silicon die abrasion test high temperature thermal resistance
Characterization methods were used to optimize amaterial (D) for burn-in use Maximize mechanical cycling minimize residue maintain good thermal performance at elevated
temperatures
BiTS20011/05/01