session 4 - testconx.org · session 4 march 11 - 14, 2007 paper #1 3 trends in contact technologies...

BiTS Workshop 2007 Archive

Session 4

ARCHIVE 2007

TRENDS IN CONTACT TECHNOLOGIES

“Next Generation Contact Technology For Semiconductor Test” Valts Treibergs, Jason Mroczkowski

Everett Charles Technologies STG

“’Off-set’ Pin Contact Innovation - An Effective Contact Solution to Pb-Free Devices for MT8704iHF (Multitest) Test Handler”

Ariel Sabellon, Eugene F. Batilo Cypress - Philippines

“Braided Electrical Contact Element (BeCe)”

Che-Yu Li Che-Yu Li and Company, LLC

“Elastomeric Interconnects - Reliable Enough for Production Test?”

Frank Bumb, Jack Pereschuk Phoenix Test Arrays Nick Langston, Sr.

Antares Advanced Test Technologies

COPYRIGHT NOTICE

The papers in this publication comprise the proceedings of the 2007 BiTS Workshop. They reflect the authors’ opinions and are reproduced as presented , without change. Their inclusion in this publication does not constitute an endorsement by the BiTS Workshop, the sponsors, BiTS Workshop LLC, or the

authors.

There is NO copyright protection claimed by this publication or the authors. However, each presentation is the work of the authors and their respective companies: as such, it is strongly suggested that any use

reflect proper acknowledgement to the appropriate source. Any questions regarding the use of any materials presented should be directed to the author/s or their companies.

All photographs in this archive are copyrighted by BiTS Workshop LLC. The BiTS logo and ‘Burn-in & Test Socket Workshop’ are trademarks of BiTS Workshop LLC.

20072007Session 4

March 11 - 14, 2007

Paper #1

1

Trends In Contact Technologies

BiTS 2007

NEXT GENERATION CONTACT TECHNOLOGY

FOR FINE PITCH SEMICONDUCTOR TEST

Valts Treibergs – R&D Engineering ManagerJason Mroczkowski – Product Specialist

Semiconductor Test Group - MN

2BiTS 2007

Agenda • Challenges of Fine-pitch Probe

Architectures (0.4mm and below)• Current Fine-pitch Probe

Architectures– Performance Attributes

• Next-generation Probe Architecture– Performance Attributes– Scalability

• Proper Probe Selection to Fit the Application

20072007Session 4

March 11 - 14, 2007

Paper #1

2


3BiTS 2007

Challenges With Current Fine Pitch Probe Architectures

• Probe Z Axis Compliance– Fine pitches typically dictate the need for long probes

• Low spring forces - very fine springs required• Higher contact resistance (Rc)• Low current carrying capacity (CCC)• Low bandwidth, high inductance

– Some short probe designs exist, but have limited compliance– Probes tend to be very fragile

• Internal Resistance Consistency - Biasing– need consistent contact between plunger(s) and barrel

components throughout compression• Tip Geometries

– Limits to DUT tip style, excessive PCB wear due to point loading

Cres

4BiTS 2007

Current Probe Architectures for Fine Pitch

• Double Ended Spring Probe (4 Piece)– Non-biased plunger– Biased plunger

• Single Ended Spring Probe (3 Piece)• External Spring Probe (3 Piece)• Next Generation Cantilever-biased

Spring Probe (4 Piece)Note: Other variants are possible and widely used

20072007Session 4

March 11 - 14, 2007

Paper #1

3


5BiTS 2007

Double-Ended Probe - Unbiased • Original spring probe design• Four piece construction• Little plunger and barrel

interaction - no designed biasing mechanism

• Plating integrity can be very difficult to control inside barrel

• Probes widely available - low cost

6BiTS 2007

Performance Attributes• Disadvantages

– Long -> high inductance– Most variable contact

resistance

0

20

40

60

80

100

120

140

160

180

200

0.000

0.024

0.076

0.118

0.162

0.210

0.258

0.302

0.352

0.392

0.440

0.486

0.532

0.584

0.626

0.672

0.718

0.766

0.726

0.674

0.626

0.586

0.540

0.494

0.446

0.398

0.358

0.308

0.260

0.218

0.164

0.118

0.078

0.032

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

5

10

15

20

25

30

35

40

Forc

e (g

)

∑

⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

=

−

−

−

−

PCBPlunger

Plunger

PlungerBarrel

Barrel

BarrelPlunger

Plunger

PlungerDUT

Contact

R

R

RR

RR

R

R

Resistance Variation

HIGH

LOW

20072007Session 4

March 11 - 14, 2007

Paper #1

4


7BiTS 2007

Double-Ended Probe - Biased

NORMAL FORCE

BIAS FORCE

BIAS ANGLE

• Improved spring probe design• Four piece construction• Angled surface biases plunger to

barrel for improved contact• Critical: Proper plating inside barrel

- smooth surfaces

Inside barrel

surface

8BiTS 2007

Performance Attributes

• Disadvantages– Long -> high inductance– Internal part wear

• Advantages– Improved contact

resistance

0

20

40

60

80

100

120

140

160

180

200

0.000

0.002

0.080

0.164

0.244

0.326

0.410

0.492

0.574

0.656

0.738

0.822

0.902

0.984

1.066

1.146

1.230

1.316

1.242

1.162

1.080

1.000

0.916

0.834

0.752

0.668

0.588

0.504

0.420

0.336

0.254

0.172

0.090

0.008

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

5

10

15

20

25

30

35

40

45

Forc

e (g

)

∑

⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−

−

−

−

PCBPlunger

Plunger

PlungerBarrel

Barrel

BarrelPlunger

Plunger

PlungerDUT

Contact

R

R

RR

R

R

RR


HIGH

LOW

20072007Session 4

March 11 - 14, 2007

Paper #1

5


9BiTS 2007

Single-Ended Probe - Biased • 3 Piece architecture• Internal spring• Angled surface biases plunger to

barrel for improved contact

NORMAL FORCE

BIAS FORCE

BIAS ANGLE

Inside barrel

surface

Note bias scrub marks

10BiTS 2007

Performance Attributes• Low Rc• Good CCC• Internal barrel plating

challenging• Good BGA solution• Fine pitch version is force

and compliance limited• Very difficult to

manufacture

0

20

40

60

80

100

120

140

160

180

200

0.000

0.024

0.070

0.116

0.162

0.208

0.254

0.298

0.344

0.388

0.434

0.482

0.528

0.570

0.616

0.664

0.710

0.754

0.716

0.672

0.628

0.584

0.538

0.492

0.446

0.402

0.354

0.308

0.266

0.220

0.174

0.126

0.082

0.038

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

10

20

30

40

50

60

Forc

e (g

)

∑

⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−

−

−

PCBPlunger

Plunger

BarrelPlunger

Barrel

BarrelDUT

Contact

R

R

RRR

R


HIGH

LOW

20072007Session 4

March 11 - 14, 2007

Paper #1

6


11BiTS 2007

External Spring Probe – Spring Biased • 3 Piece architecture• External spring• Natural spring bend biases plunger to

barrel for improved contact• Less internal wear than bias-plunger

60.000°

SPRING INDUCEDMOMENT NORMAL FORCE

BIAS FORCE BIAS FORCE

12BiTS 2007

Performance Attributes• Larger spring volume

- higher force to DUT at fine pitch

• Good compliance• Long life


HIGH

LOW0

20

40

60

80

100

120

140

160

180

200

0.000

0.010

0.0480.0

860.1

220.1

620.1

980.2

360.2

740.3

140.3

520.3

940.4

340.4

740.5

120.5

540.5

920.6

300.5

960.5

580.5

260.4

880.4

500.4

120.3

700.3

320.2

920.2

540.2

160.1

800.1

420.1

040.0

660.0

28

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

5

10

15

20

25

30

35

40

45

Forc

e (g

)

∑

⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−

−

−

PCBPlunger

Plunger

BarrelPlunger

Barrel

BarrelDUT

Contact

R

R

RRR

R

20072007Session 4

March 11 - 14, 2007

Paper #1

7


13BiTS 2007

Next-Gen Probe - Cantilever Biased

• 4 piece architecture - barrel-less

• Quad-cantilever arm biased• Bias force is independent of

spring force• High compliance to test height

ratio• Dual springs - 30g+ force at

fine pitch (.4mm)GeminiTM

ECT Patent Pending

14BiTS 2007

Performance Attributes

• External plating surfaces -> stable resistance throughout stroke

• Quad-cantilever arms guaranteed to be in contact in at least 2 locations in any condition

NORMAL FORCE

INDEPENDENTCANTIVEVERBIAS FORCE

20072007Session 4

March 11 - 14, 2007

Paper #1

8


15BiTS 2007

Resistance Attributes

0

20

40

60

80

100

120

140

160

180

200

00.0

240.0

46 0.070.0

920.1

160.1

360.1

620.1

840.2

06 0.230.2

520.2

760.2

960.3

180.3

420.3

640.3

740.3

560.3

340.3

140.2

94 0.270.2

480.2

240.2

040.1

780.1

560.1

340.1

12 0.090.0

680.0

440.0

22

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

10

20

30

40

50

60

Forc

e (g

)

0

20

40

60

80

100

120

140

160

180

200

0.000

0.020

0.044

0.066

0.088

0.112

0.132

0.158

0.178

0.200

0.224

0.246

0.270

0.292

0.314

0.338

0.358

0.370

0.354

0.330

0.310

0.290

0.266

0.244

0.220

0.200

0.178

0.154

0.132

0.110

0.086

0.064

0.040

0.018

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

10

20

30

40

50

60

Forc

e (g

)

500K Cycles

0 Cycles


HIGH

LOW

∑

⎟⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−

−

−

PCBLower

Lower

LowerUpper

Upperl

UpperDUT

Contact

RR

R

R

RR

2

16BiTS 2007

INSERTION LOSS (dB)

5.0

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00Frequency (GHz)

Inse

rtio

n Lo

ss (d

B)

GEMINI 0.4mm INLINE

Bandwidth

RF Comparison – vs. Double Ended

Lm

Ls

Rs

Ls

Rs

Csout

Csin

Csin

Csout

IN(DUT)

OUT(BOARD)

Ccout

Ccin

INSERTION LOSS (dB)

5.0

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00

Frequency (GHz)

Inse

rtio

n Lo

ss (d

B)

[email protected] Vespel

20072007Session 4

March 11 - 14, 2007

Paper #1

9


17BiTS 2007

Design Flexibility - Scalability• Kelvin QFN

– 0.15 mm force/sense tip spacing

– 0.45mm PCB spacing

– 30g– Cantilever

biased• BGA Kelvin

0

20

40

60

80

100

120

140

160

180

200

00.0

04 0.03

0.056

0.084

0.108

0.132

0.168 0.2

0.228

0.254

0.282

0.318 0.3

50.3

740.4

060.4

320.4

540.4

560.4

440.4

220.3

940.3

660.3

340.3

020.2

720.2

440.2

140.1

88 0.15

0.124

0.088

0.064

0.032

Deflection (mm)

resi

stan

ce (m

Ohm

s)

0

5

10

15

20

25

30

35

Forc

e (g

)

18BiTS 2007

Proper Probe Selection Requires Understanding of the Test Application

• DUT performance characteristics– RF, power, resistance sensitivity, geometry– Test program pass/fail criteria

• Handler requirements– Alignment accuracy, available force, Z-stack

variability• Cost

– Cost per probe – you get what you pay for– Cost over lifetime of socket - COO– Cost of test-cell down-time for socket

maintenance– Test floor support

20072007Session 4

March 11 - 14, 2007

Paper #1

10


19BiTS 2007

Probe Choice is Still Application Dependent + -

Double-Ended Unbiased

• Lowest Cost • Compliance • Long Life

• Rc Most Unstable • Low Force At Fine Pitch • High Inductance

Double-Ended Biased

• Low Cost • Compliance

• More Stable Rc • Decreased Life • High Inductance • Low Force

Single-Ended Biased

• Medium Cost • Low Rc • Good CCC

• Impedance Mismatch At Rf Frequencies

• Low Force / Compliance • Difficult To Scale To Fine

Pitch External Spring Biased

• Long Life • Good Compliance • Higher Forces At Finer

Pitches • Low Inductance

• Medium High Cost

Gemini Cantilever Biased

• Longest Life • Low Rc • Near 50Ω At .5mm Pitch • Low Inductance • 30+g DUT Force At Finest

Pitches • Good Compliance • Optimal for Kelvin

• Highest Cost • Array Pitch > Inline Pitch

20BiTS 2007

Thank You – Questions?

20072007Session 4

March 11 - 14, 2007

Paper #2

1


BiTS 2007 1

Ariel SabellonEugene Batilo

Nicolas LeeKif LohAnthony Buendia

Shunsuke SasakiToshio Kazama

Affiliations

“Off-set” Pin Contact Innovation -An Effective Contact Solution toPb-Free Devices for MT8704iHF

(Multitest) Test Handler

2007 Burn-in and Test Socket WorkshopMarch 11 - 14, 2007

BiTS 2007 2

Agenda

• Pb-Free Performance and Challenges toMT8704iHF Handler with Spring LedgeContact Design

• Innovative Contactor Design Conversionwithout forcing to change the current set-up(DUT Boards, Lead Supports, etc)

• New Contactor Design Performance on Pb-Free Device for MT8704iHF Handler

20072007Session 4

March 11 - 14, 2007

Paper #2

2


BiTS 2007 3

• Low Yield with High Recovery Rate• Poor Contact• Early Breakage of Contact Pins• Accumulation of Mold Debris - Contamination• Mechanical Wear

Pb-Free Performance to MT8704iHFHandler with Spring Ledge Contact

BiTS 2007 4

3K insertions 6K insertions

150K insertions (Cleaning Required)

Spring Ledge ContactPenetration needed

Pb-Free Test Challenges

20072007Session 4

March 11 - 14, 2007

Paper #2

3


BiTS 2007 5

MT8704 SPRING LEDGE CONTACT DAILY PERFORMANCE EVALUATION RESULTS on NiPd SSOP (300mils)

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

6105

1450

3

6105

1473

4M

6105

1472

2

6105

1472

2N

6105

1505

0N1

6105

1505

1M1

6105

1505

1N

6105

1532

3M

6105

1505

1

6105

1505

1N1

6105

1565

4LOT NUMBERS

YIEL

D

0100002000030000400005000060000700008000090000100000

# O

F UN

ITS

(%) FPY RECOVERY (%DELTA) TOTAL QTY. OUT CUM INSERTS (%) FINAL YIELD

SPRING LEDGE REPLACED @ 47K CUM INSERTIONS DUE TO BROKEN CONTACTPIN ISSUE

Pb-Free Performance to MT8704iHFHandler with Spring Ledge Contact

BiTS 2007 6

BEFORE (Spring Ledge) AFTER (Microcontactor ®)

• Low FPY on NiPd (Lead Free) device• Short contact life span due to early breakage of pins

• High FPY on NiPd (Lead Free) device• Better contact life span (20x higher than the current contact)

MT8704 New Socket Design Assembly

20072007Session 4

March 11 - 14, 2007

Paper #2

4


BiTS 2007 7

CONTACT BLOCK

UPPER SOCKET ASSEMBLY

INTERCONNECT BOARD

LOWER SOCKET ASSEMBLY

DEVICE

DUT BOARD

MT8704 New Socket Design Assembly

BiTS 2007 8

INTERCONNECT BOARD

FREE STATE CONTACT STATE

DEVICE

Microcontactor ®

MT8704 Off-Set Contact Design

20072007Session 4

March 11 - 14, 2007

Paper #2

5


BiTS 2007 9

BASE MATERIAL: STEELPLUNGER PLATING: Pd AlloySPRING FORCE: 35gF

Microcontactor ® Pin Mechanical Construction

BiTS 2007 10

Microcontactor ® REPLACED @ 942K CUM INSERTIONS DUE TO WEAR-OUT and DROP of YIELD

Pb-Free Performance to MT8704iHFHandler with Microcontactor ®

20072007Session 4

March 11 - 14, 2007

Paper #2

6


BiTS 2007 11

MT8704 Contact Performance Comparison (S-Ledge vs. Microcontactor ®

BiTS 2007 12

$ Value Pin Insert = 0.00068 vs 0.00400

Microcontactor ®Spring Ledge

NHK

Cost Analysis (ROI)

20072007Session 4

March 11 - 14, 2007

Paper #2

7


BiTS 2007 13

Conclusion• Microcontactor ® is better than spring ledge

for lead-free (NiPdAu) applications• Penetration is required to ensure better

contact on a lead-free (NiPdAu) materials• Microcontactor ® insertions are higher

compare to S-Ledge contact (47k vs. 942k)• Improved FPY from 87.60% to 97.50%• Reduced delta from 10.40% to 1.40%• Improve $ value inserts per pin from

0.0040 down to 0.000680.

20072007Session 4

March 11 - 14, 2007

Paper #3

1


BRAIDED ELECTRICALCONTACT ELEMENT

(BeCe)Che-Yu Li

Che-Yu Li and Company, LLC

BiTS WORKSHOP 2007

Braided Electrical Contact Element (BeCe) 2

OUTLINE

• Motivation• Braided Electrical Contact Element

– Design– Performance

• Interposer• Conclusions

20072007Session 4

March 11 - 14, 2007

Paper #3

2



MOTIVATION

• Requirements– Fine Pitch– Low Resistance and Inductance – Low Contact Force and High

Compliance– High Durability and Reliability– High Service Temperature


MOTIVATION

POGO PIN– Large Pitch

HELICAL SPRING– High Resistance

ANISOTROPICALLY CONDUCTIVE FILM

– Low Compliance

CURRENT CONTACT TECHNOLOGIES

20072007Session 4

March 11 - 14, 2007

Paper #3

3



BRAIDED ELECTRICAL CONTACT ELEMENT

All Dimensions in mils

Plated Tip

BeCe Braiding

BeCe: A Conductive Braided Wire Stand-alone Structure of Short Cylindrical Form


DESIGN PRINCIPLES

• Fine pitch:– Smaller core diameter and wire diameter

• Low bulk resistance:– Multiple parallel conductors

• Low inductance:– Short height allowed by high elastic compliance

• High elastic compliance and low contact force:– Helical design, and low modulus, and high yield

strength wire

20072007Session 4

March 11 - 14, 2007

Paper #3

4



DESIGN PRINCIPLES (cont.)

Low contact resistance and high reliability:– Multiple contact tips

• High durability:– Vacuum melted wire stock

• High service temperature:– High melting temperature of wire


SINGLE PIN TEST DATA

BeCe (Copper Alloy)• Low Force

- <2g/mil

• Low Resistance- <10milliohms

• High Compliance- >25% of uncompressed

height

18X40 BeCe Copper Alloy

05

10

152025

0 0.005 0.01 0.015 0.02 0.025

inches

gram

s

run 1

run 2

run 3

18X40 BeCe Copper Alloy

01020304050

0 5 10 15 20 25

grams

mill

iohm

s

run 1

run 2

run 3

20072007Session 4

March 11 - 14, 2007

Paper #3

5



SINGLE PIN TEST DATA

12X40 UEC

0

5

10

15

20

25

0 0.005 0.01 0.015 0.02 0.025

inches

gram

s run 1run 2run 3

12X40 UEC

0

10

20

30

40

50

0 5 10 15 20 25

gramsm

illio

hms run 1

run 2run 3

BeCe (Copper Plated Stainless Steel)

• Low Force- <2g/mil

• Low Resistance- <15milliohms

• High Compliance- >30% of uncompressed

height


PERFORMANCE OF CONVENTIONAL 2 MM POGO PIN

Initial travel to Recommended Travel 11.8mils092706 sample2 1a

Pogo Pin 2.00-1

0

5

10

15

20

25

0 0.004 0.008 0.012 0.016

inches

gram

s

run 1run 2run 3run 4run 5


Pin 2.00-1

0100200300400500600700800900

1000

0 5 10 15 20 25

grams

mill

iohm

s



Pogo Pin 2.00-1

0

5

10

15

20

25

0 0.004 0.008 0.012 0.016

inches

gram

s



Pogo Pin 2.00-1

0100200300400500600700800900

1000

0 5 10 15 20 25

grams

mill

iohm

s


20072007Session 4

March 11 - 14, 2007

Paper #3

6



STABILITY OF CONTACT RESISTANCE

BeCe (Copper Alloy)

18X40 mil copper alloy BeCe, 2X2 array, Cycled at 8mils displacement,Resistance data at 10mils displacement

Just 2 m !!


DURABILITY OF BeCe ARRAY

>1M cycles @ near elastic limit

20072007Session 4

March 11 - 14, 2007

Paper #3

7



SUMMARY OF ATTRIBUTES

• Normal Elastic Compliance up to 30% of Uncompressed Height and Average Contact Force of 15 Grams per BeCe or Less

• 10 m or Less Total Resistance Per BeCe Contacting Solder Bump or Contact Pad, or Soldered

• 10 GHz or More Frequency Capability

• Demonstrated 1M or More Touchdowns and 50K or More Touchdowns Between Cleaning


SUMMARY OF ATTRIBUTES• Service Temperature of >250° C.

• High Reliability: >1M cycles @ near elastic limit

• Connection Pitch to 10 mils or Less With I/O Counts to 5000 or More With SolderableEnds, or Wire-Bondable at One End

• Low Cost Manufacturing

20072007Session 4

March 11 - 14, 2007

Paper #3

8



FR-4 BASED INTERPOSER

1. BeCe Fixed in Carrier

2. BeCe Movable in Carrier

3. PCB Technology used

Upper Spacer

CarrierLower Spacer

BeCe

Adhesive

Adhesive

BeCeGuide Tube

Carrier


FR-4 BASED INTERPOSER

Populated With 1247 BeCe Contacts, Fixed in Carrier

18X50 BeCe, 1247 I/O, .040” Pitch1.65” X 1.25” Overall

20072007Session 4

March 11 - 14, 2007

Paper #3

9



SOCKET ADAPTOR & ALIGNMENT

• Simple Design• Two Alignment

Pins Required• Adaptor

Illustration

PACKAGE SIDE

TEST BOARD SIDEAlignment PinAdaptor Pin

Interposer Opening


AREA ARRAY MULTIPLE PIN TEST DATA

Multi-pin test results1247 18X50 SS/CU BeCe Interposer

0.000.020.040.060.080.10

0 20 40 60 80 100

Multi-Pin test BeCe Number

BeC

e Re

sist

ance

O

hms

20072007Session 4

March 11 - 14, 2007

Paper #3

10



SOLDERED BeCe

• Illustration

• To package for test before reflow

• To probe card• Conventional

Manufacturing tools


THERMAL MANAGEMENT

• Redistribution

• Thermal path to thermal management system

• PCB Technology used

Outer Most BeCe

To Thermal Management System

20072007Session 4

March 11 - 14, 2007

Paper #3

11



CONCLUSIONS

• BeCe is extendable and can meet next generation test and burn-in needs

• The manufacturing of BeCe contacts and Interposers is low cost and suitable for small lots of varied foot prints

• Qualification tools for BeCe contacts and interposers are fully developed

20072007Session 4

March 11 - 14, 2007

Paper #4

1


2007 Burn-in and Test Socket Workshop March 11 - 14, 2007

Frank Bumb &Jack PereschukPHOENIX TEST

ARRAYS, LLC

Nicholas Langston, Sr.ANTARES ADVANCED

TEST TECHNOLOGY

Elastomeric Interconnects-Reliable enough for production?

4/6/2007 2

Current technologies on fumes?

• Some think so. Even today.• Lead lengths vs. performance requires

– 10Gb/s + data rates– 30 ps – edge rates– Interconnect signal paths of 1mm to 0.5mm

• (This paper will focus only on the needs of high performance Final test)

20072007Session 4

March 11 - 14, 2007

Paper #4

2


4/6/2007 3

Incumbent technology• Spring Probes

-As they get smaller:– challenge both makers and users– pitch reduction difficult– Increasingly fragile– Carry less current – Reach Bandwidth limits– Ultimately, can’t meet

lower inductances and the need for speed

4/6/2007 4

Elastomers: …to the rescue?

Elastomerics offer:– Very low profile compliant paths

• Low resistance • Low inductance• Low capacitance• low force possible

– Highest possible performance metrics• High bandwidth

20072007Session 4

March 11 - 14, 2007

Paper #4

3


4/6/2007 5

“Metal Filled Polymers” (MFP)

• “Elastomers”– Most formed as sheets– Most vertical path– No Individual conductor assembly– Individual conductors not serviceable– Cost can be lower if volumes higher

4/6/2007 6

Focus by structure• Elastomeric Interconnects with conductive metal

paths: Metal Filled Polymers (MFP)– Generic

• Wires arrayed & embedded in Elastomer (E)• Cohesively stacked particles co-molded in E

– Dedicated Circuits• Dispersed particles in E (MFP)• Dispersed particles in E (MFP) & on carrier

20072007Session 4

March 11 - 14, 2007

Paper #4

4


4/6/2007 7

Focus on function• Elastomeric Interconnects with conductive

metal paths– We will consider:

• How they are made• How they work• mechanical differences • Electrical similarities • Behavioral issues over time• Generic Life span data

4/6/2007 8

How they are made

• Metals– Powders– Particles– Wires

• Elastomer– Flexible adhesive

binder (matrix) – Silicone– Epoxy– Synthetic Rubber

+

Metal Filled Polymers (MFP)

20072007Session 4

March 11 - 14, 2007

Paper #4

5


4/6/2007 9

Compression forces particles together,forms temporary ‘solid’ conductor

How they work

Not Conductive

Minimally Conductive

Reliably Conductive lowest resistance

x / 3 X*rated travel

4/6/2007 10

Type: Generic 1

• Gold Wires arrayed, embedded in Polymer– BENEFITS– ISSUES?

20072007Session 4

March 11 - 14, 2007

Paper #4

6


4/6/2007 11

• Metal spheres arrayed, embedded in Polymer

– BENEFITS– ISSUES?

Type: Generic 2

4/6/2007 12

• Randomly arranged particles in Polymer


Metal Filled 1 Defined leads

20072007Session 4

March 11 - 14, 2007

Paper #4

7


4/6/2007 13


• Randomly arranged particles in Polymer

• Includes topside protective layer


4/6/2007 14

• Randomly arranged particles in Polymer -on a Carrier

• includesIntegrated topside protective layer

– BENEFITS?– ISSUES?


20072007Session 4

March 11 - 14, 2007

Paper #4

8


4/6/2007 15

Data Capture

• History: 600+test sequences – 110 Million+ hits – Over 4 years – ~ test length:

150K – 250K hits(some up to 700K)

• Conditions– Room Temp– Pneumatic Drive– Set to 35 PSI– 5500 hits/ hr

24/7– Hardstops

required– Automated data

capture

4/6/2007 16

DATA CAPTURE Hardware

Handler simulator

PC & HP DVM

20072007Session 4

March 11 - 14, 2007

Paper #4

9


4/6/2007 17

Test Setup 1Schematic of test setup -shown exploded-

Tests are conducted under 30-35% compression

“DUT”

Test “load board”

M+ M-

Elastomeric Element

Protective circuit

4/6/2007 18

Test Setup 2

Schematic of test setup -shown compressed

M+ M-

Elastomeric Element

Protective circuit

20072007Session 4

March 11 - 14, 2007

Paper #4

10


4/6/2007 19

The First PlungeElastomeric Interconnect:

Single Plunge (new)

0.0000.0250.0500.075

0.1000.125

18 72 126 180 234 288 342 396ms

Ohm

s

4/6/2007 20

& the Last PlungeElastomeric Interconnect:

Single Plunge (new vs worn)

0.000

0.025

0.050

0.075

0.100

0.125

18 72 126 180 234 288 342 396ms

Ohm

s

1390K

20072007Session 4

March 11 - 14, 2007

Paper #4

11


4/6/2007 21

Elastomer: Wear Curves vs Life

0.000

0.050

0.100

0.150

0.200

0.250

18 72 126 180 234 288 342 396ms

Ohm

s

198K195K293K390K

Test Results: Wear

4/6/2007 22

Test Results: the Front edge

Elastomer: Leading Edge Curves vs Life

0.000

0.050

0.100

0.150

0.200

0.250

0.300

18 36 54 72 90ms

Ohm

s

198K195K293K390K

20072007Session 4

March 11 - 14, 2007

Paper #4

12


4/6/2007 23

Again, the Front edgeZoomed in:

Leading Edge Curves vs Life

0.000

0.025

0.050

0.075

0.100

54 72 90ms

Ohm

s

198K195K293K390K

4/6/2007 24

Test Results 5 (Reliability)

Typical Elastomeric Life Plot 250K +

0.000

0.020

0.040

0.060

0.080

0

40,0

00

80,0

00

120,

000

160,

000

200,

000

240,

000

280,

000

320,

000

360,

000

Actuations

Ohm

s

20072007Session 4

March 11 - 14, 2007

Paper #4

13


4/6/2007 25

Test Results 6Characteristic Behavior vs. time

stopped, rest 10 minutes, restarted

0.000

0.010

0.020

1

81,3

01

162,

601

243,

901

325,

201

406,

501

4/6/2007 26

Failure: what happens• Resistance Rise:

– Strain through accumulated use (wear): Supporting structures soften with use

– Excessive Overactuation = strain (tear)– Prolonged exposure to heat & compression

causes reformation: mechanical compression set, loss of resilience

– Contamination of Dut side Surfaces:

20072007Session 4

March 11 - 14, 2007

Paper #4

14


4/6/2007 27

Failure: What doesn’t happen • Sudden electrical changes are rare

• Opens are rare - Surface contamination is the usual culprit –some are cleanable

• Shorts almost never - Redistribution of conductor material

4/6/2007 28

Do’s and Don’ts of care– Do keep them clean– Remove pressure when not in use– ...Especially at high temp– Alcohol Free– Store Appropriately

20072007Session 4

March 11 - 14, 2007

Paper #4

15


4/6/2007 29

Values: Electrical Lows & Highs

• Short path = “low…”– Low profile– Low resistance – Low inductance– Low capacitance– low force

• Short path = “high…”– High bandwidth– High Current– High

“performance”

REVIEW

4/6/2007 30

Value Offsets:• In Automation-

Coplanarity, Actuation, critical

• Devil‘s in the Details-design & materials

• Elastomerics: can be damaged-misinsertion, careless handling

• High data rates for automation at amoderate cost

• Easy Replacement and Maintenance

• ComingPitch Reductions seen as easier

• Service is rare

REVIEW

20072007Session 4

March 11 - 14, 2007

Paper #4

16


4/6/2007 31

Review (Reliability)

Typical Elastomeric Life Plot 250K +

0.000

0.020

0.040

0.060

0.0800

40,0

00

80,0

00

120,

000

160,

000

200,

000

240,

000

280,

000

320,

000

360,

000

Actuations

Ohm

s

REVIEW

4/6/2007 32

Bottom line

At 10-40 GHz+ -and beyond,The technology is here.

Elastomeric Element Lifetimes of 250K to 400K are expectable

session 4 - testconx.org · session 4 march 11 - 14, 2007 paper #1 3 trends in contact technologies...

Documents