31
March 3-6 , 2002 Hilton Phoenix East/Mesa Hotel Mesa, Arizona Sponsored By The IEEE Computer Society Test Technology Technical Council Burn-in & Test Socket Workshop IEEE IEEE COMPUTER SOCIETY

Burn-in & Test Socket WorkshopFirewire USB2 Infiniband SerialATA SDRAM DDR RDRAM DDR2 AGP AGP2x AGP4x AGP8x PCI-X PCI PCI 2 FSB FSB FSB FSB FSB SerialATA I/O Data Rate (Gbps) M e m

  • Upload
    others

  • View
    9

  • Download
    1

Embed Size (px)

Citation preview

March 3-6 , 2002Hilton Phoenix East/Mesa 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 2002BiTS 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.

Burn-in & Test SocketWorkshop

Keynote AddressTuesday 3/05/02 8:00PM

“Tooling: The Other ‘Capital’ Equipment”

Steven B. StraussITTO Manager

Intel Corporation

Technical Program

11

Steven B StraussIntel Test Tooling

Operation ManagerChandler, Arizona

Tooling:Tooling:The OtherThe Other

“Capital” Equipment“Capital” Equipment

22

Key Messages:Key Messages:• We have just experienced one of the worst

economic downturns in the history of thisindustry

• Technical challenges continue toaccelerate

• Increasing costs have driven changes inCapital Equipment for Test

• The Tooling Supply industry has notchanged to meet customer needs insolutions, cost, or capability

• It’s time

33Source: VLSI Research

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

0255075

100125150175200225250275300325350375400

Billio

n D

olla

rs ($

B)

-100%-90%-80%-70%-60%-50%-40%-30%-20%-10%0%10%20%30%40%50%60%

Gro

wth

Rat

e (%

)

Chip Sales $BChip Sales % ChangeCAGR 1958-2006

2001 = $141B down 31%2002 = $170B up 21%

15% CAGR1958-2006

The economic world we live in:The economic world we live in:Semiconductor Industry CyclesSemiconductor Industry Cycles

44

The economic world we live in:The economic world we live in:Lessons Learned?Lessons Learned?

TIME

GROWTH

EnablingTechnologies

IrrationalExuberance

RationalGrowth

Turbulence

Are we here?

55

“If the automobile industry advanced“If the automobile industry advancedas rapidly as the semiconductoras rapidly as the semiconductorindustry, a Rolls Royce would getindustry, a Rolls Royce would get500,000 miles per gallon of gas and500,000 miles per gallon of gas andit would be cheaper to throw itit would be cheaper to throw itaway than park it.”away than park it.”

– Gordon E. Moore– Gordon E. Moore Chairman Emeritus, Intel Corporation Chairman Emeritus, Intel Corporation

66

The Technical World we live in:The Technical World we live in:Extending Moore’s LawExtending Moore’s Law

4004400480088008

8080808080858085

80868086 286286386386

486486PentiumPentium®®

Pentium IIPentium II

Pentium IIIPentium III

1.8B1.8B

425M425M

0.0010.001

0.010.01

0.10.1

11

1010

100100

1,0001,000

10,00010,000

19701970 19801980 19901990 20002000 20102010

TransistorsTransistors(MT)(MT)

~2B Transistors~2B Transistors

Pentium 4Pentium 4

Transistors doubling every 2 years toward a billion transistorsTransistors doubling every 2 years toward a billion transistors

~2004: 5GHz, 80W

77

The Technical world we live in:The Technical world we live in:The Volume LifecycleThe Volume Lifecycle

No time for Mistakes - Get It Right The First TimeNo time for Mistakes - Get It Right The First Time

0.350.35µµmm(Act.)(Act.)

0.250.25µµmm(Act.)(Act.)

0.180.18µµmm(Act.)(Act.)

UnitUnitVolumeVolume

Quarters into ProductionQuarters into Production1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 1010

0.130.13µµmm(Est.)(Est.)

88

• Issue– How do you

deliver a lot ofnoise free power,quickly

– Power in = HeatOut

• Capabilitiesneeded�� 2X improvement in2X improvement in

capacitance andcapacitance andinductance needed /inductance needed /generationgeneration

�� Need to optimize theNeed to optimize thecomplete silicon tocomplete silicon totester power deliverytester power deliverysolutionsolution

PII / PIIIPentium

486386286

8086

80858080

800840040.01

0.10

1.00

10.00

100.00

???

1970 1980 1990 2000 2010Year

Icc

(am

p)

The Technical World we live in:The Technical World we live in:Power DeliveryPower Delivery

P4

99

How rapidly can we innovate new thermalsolutions / interface materials ?

0.10.1

11

1010

100100

’71’71 ’74’74 ’78’78 ’85’85 ’92’92 ’00’00 ’04’04 ’08’08

PowerPower(Watts)(Watts)

4004400480088008

8080808080858085

80868086286286

386386486486

PentiumPentium®®

processorsprocessors

The Technical World we live in:The Technical World we live in:Power Density will get worsePower Density will get worse

0000

50505050

100100100100

150150150150

200200200200

250250250250

300300300300

W/c

m2

Tod

ayT

oday

Tod

ayT

oday

+1yr

+1yr

+1yr

+1yr

+2yr

+2yr

+2yr

+2yr

+4yr

+4yr

+4yr

+4yr

AveAveAveAveAveAveAveAve

Hot spotHot spotHot spotHot spotHot spotHot spotHot spotHot spot

1010

CPU Packaging ChallengesCPU Packaging ChallengesC

ompl

exity

/ Pe

rfor

man

ce

25 MHz 2.0+ GHz

WIREBOND FLIPCHIP5 WATTS 100+ WATTS

ORGANIC

SINGLE CHIP

PPGA

PLGAOLGA

FCxGA1

FCxGA2ORGANIC

MODULAR

INT1

Slot M

Slot2

Slot 1

INT2

INT3

FCxGA4

• Current & power are increasing• Voltage scaling lowers noise margins• Frequency push for performance > 3 GHz• Form-factors continue to segment

1111

Communications ProductsCommunications ProductsPackaging ChallengesPackaging Challenges

CostReduce

vfBGA

µµµµBGA

.5mm Ball Pitch

.8 mm Thick

DensityIncrease

DensityIncrease

1.2 mm Thick

SCSP

DieDensity

> 2 DieLarger Body

FCBGA

FC-CSP

Increase Body & I/O

CapabilityIncrease

Mixed-Stack

CapabilityIncrease

Backside Stack

Ultra-Thin Die

TSOP

DensityIncrease

IntegratedComponents and

Substrates

WLPLead/Halogen Free

Future possibilities …Stacked package MCM ?Through Silicon vias ?

1212

The Technical World we live in:The Technical World we live in:Device I/O TrendsDevice I/O Trends

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0USB

Firewire USB2

Infiniband

SerialATA

SDRAMDDR

RDRAM

DDR2

AGP AGP2x

AGP4x

AGP8x

PCI-XPCI PCI 2

FSB

FSB

FSB

FSBFSB

SerialATAI/O

Dat

a R

ate

(Gbp

s)

Memory Interface

Peripheral Interfaces

FSB

High

Spe

ed S

eria

l

BSBBSB

BSB

1313

The Manufacturing World we live in:The Manufacturing World we live in: Test Flow Revolution Test Flow Revolution

• The test industry has changed– Implemented advanced DFT to

manage test complexity– Eliminate capability treadmill– Enable parallel test in complex

designs• Migrate significant percentage of test

content to less expensive structuraltest

– Enables capital cost reduction– Deliver state-of-the-art capabilities– Simplified tester hardware designs– Optimized content and flow

Fab

Structural Wafer Sort

Assembly

STDBI

Structural Class

Functional Class

PPV

Optimization

in progress

1414

Capital vs. ToolingCapital vs. Tooling

• Historically Capital Equipment got all the attention• But… We are buying less and less Test Capital• And…Tooling is becoming a bigger and bigger %

$

OtherToolingCapital

Burn InFunctional

Test PPVStructural Test

CPU

B

CPU

C

CPU

A

CPU

B

CPU

C

CPU

A

CPU

B

CPU

C

CPU

A

CPU

B

CPU

C

CPU

A

1515

What is Tooling?What is Tooling?

Big Iron Tester Interfaces:PCB’s, Contactors, Pogo Pins

1616

What is Tooling?What is Tooling?

Wafer Test: Probe Cards, Probes,Space transformers

1717

Big Boards and Sockets: Burn in, Massively parallel test and fusing

What is Tooling?What is Tooling?

1818Box or Use Based Testing

What is Tooling?What is Tooling?

1919

Provides a temporary Thermal Mechanical and/orElectrical interface to the DUT

Is custom to productsIs dependent on and customized to: Packagingform factors, Electrical and Thermal requirements

and Device FunctionHas demand that is driven by product ramp cycles

and is renewed and upgraded by product,package etc…

Provides supplier and supply chain challenges likeno other commodity

Is a technology, development and HVM enabler!

Tooling:Tooling:

2020

What does it takes to deliver tooling in this environment ?What does it takes to deliver tooling in this environment ?

RFI Suppression with different via types (1000mils spacing or 8 vias)

-30

-20

-100

10

20

30

40

4.8 5 5.2 5.4 5.6 5.8 6 6.2

Frequency (GHz)

Max

E-f

ield

(V/m

)

No Via ThickPTH (20mils FHS)ThinPTH (10mils FHS)Thick Filled Via (20mils FHS)Thin Filled Via (10mils FHS)

PCB Technology / DesignMethodologies

Socket/Contactor Design andFabrication

High Speed/Data Rate & PowerDelivery Modeling/Simulation

C21b

C21a

L1 L2R1 RM2

1

System Integration

RF/EMC Modeling &Simulation

ContactTechnology

Electrical & TestCircuit Design

REQUIREDTOOLING

COMPETENCIES

Thermal Control /Mechanical Actuation

2121

Tooling Suppliers areTooling Suppliers arenot keeping up!not keeping up!

2222

The Treadmill Challenge in ToolingThe Treadmill Challenge in Tooling• NPI increasing at 40%

to 50% per year–More custom designs

• Steep Product Ramps -GIRTFT

• Product Health is a biglever

• Tooling and Servicecosts are increasing asan overall percentageof test cost

Product A Product B

Test

Cos

t / U

nit

TOOLING

Depreciation Tooling Support Labor

2323

Tooling suppliersTooling suppliers• Are there any “turn key” tooling suppliers?

The tooling industry is fragmented• The Capital Analogy:

– Buy capital equipment – One stop shopping– You don’t buy the pieces from 2-3 suppliers!

• Lets buy a Burn In board– 1 supplier for design– 1 supplier for PCB manufacturing– 1 supplier for sockets

• Who stands by the final product? The Customer ?The Tooling Supply Chain Needs to

provide Solutions, not components

2424

• Tooling suppliers fall into 2 categories– Component manufacturers– Integrators or Assemblers

• A typical tooling supply chain contains 2-4poorly synchronized suppliers–Design–Custom component design and manufacturing–Assembly or Integration–HVM support

This industry infrastructure will not achievethe requirements of the next generations

Today’s Tooling Supply ChainToday’s Tooling Supply Chain

2525

ToolingToolingToolingTooling Lead TimeLead TimeLead TimeLead Time

0

5

10

15

Goal < 2 weeksGoal < 4 weeks

Supplier Lead Time

Complex Assemblies

Sub Components

Wee

ksW

eeks

2626

The Next GenerationThe Next GenerationTooling Supply ChainTooling Supply Chain

• Is proactively on the “treadmill”– Technically– Economically– Logistically

• Enables fungible designs that last 2-3 processgenerations

• Is in sync with the specific technologies of thecustomers

• Provides turn key solutions• Has 2 - 4 weak lead times• Is low, low, low cost and continues to drive costs

down

2727

What it takesWhat it takes• Evolution will not yield these goals!• The scaling treadmill that the industry has relied

upon needs to be replaced by disruptivetechnologies

• If you want to survive you must:– invest in disruptive technologies– cannibalize your current ones– acknowledge that your business model will be

completely different in 2 years.

Can you do this? If not you won’t survive!About ½ of you will be around in 2 years

Will you be one of them?

2828

““We sometimes refer to the definition of insanity asWe sometimes refer to the definition of insanity asdoing the same thing the same way over and overdoing the same thing the same way over and overagain, hoping for a different result. If you want toagain, hoping for a different result. If you want to

achieve different results -- achieve different results -- betterbetter resultsresults -- you have-- you haveto do things differently. Success is not accidental. Itto do things differently. Success is not accidental. Ithappens because people plan carefully and they layhappens because people plan carefully and they laythe necessary groundwork to get the right result inthe necessary groundwork to get the right result in

the end.the end.””--- - Craig BarrettCraig Barrett President and CEO, Intel Corporation President and CEO, Intel Corporation December 29, 2000 December 29, 2000