Atsushi MinePhill Mai

JEM3000 Laurelview Court

Fremont, CA 94538

Comprehensive Approach to Control Contact Resistance Instability and Improve First Pass

Yield of Bumped Devices

2005 SouthWest WorkshopJune 5 to 8, 2005

Jerry J. Broz, Ph.D. International Test Solutions

5690 Riggins CourtReno, Nevada 89502

Joe FoerstelSean Chen

Altera Corporation101 Innovation Drive San Jose, CA 95134

06/07/2005 SouthWest Test Workshop 2005 2

Overview

• Introduction• Objectives / Approach• Methodology Overview• Implementation / Characterization• Summary

06/07/2005 SouthWest Test Workshop 2005 3

Evaluation Goals

• To determine the initial time zero path resistance of the VS crown probe card, and monitor that same path resistance after various amounts of die had been sorted.

• To compare the performance of current standard VS flat tip technology against the new VS crown tip in terms of wafer yield.

06/07/2005 SouthWest Test Workshop 2005 4

VS-Series Probe (Crown)

• Newly developed spring probe design• Achieves precise probe position and planarity

Crown Tip Shape

Guide Plate Stopper

VS Series

Picture of the Probe

06/07/2005 SouthWest Test Workshop 2005 5

Probe Mark

Contact Concept

Image of the Contact Point

Crown Probe Tip

Stable CRES by lower contact force with

minimal probe mark.

The ridge of the crown probe tip translates vertical

force into radial force.

06/07/2005 SouthWest Test Workshop 2005 6

VS Crown Tip Probe Mark Images

06/07/2005 SouthWest Test Workshop 2005 7

VS Crown Tip Probe Mark Images

06/07/2005 SouthWest Test Workshop 2005 8

Microscope Images of VS Crown Tip Probe Marks

06/07/2005 SouthWest Test Workshop 2005 9

Probe Mark Size Comparison

• Crown Tip Marks vs. Flat Tip Bump Deformation

06/07/2005 SouthWest Test Workshop 2005 10

Side View Comparison Of Probe Marks

Flat Tip Crown Tip

50 um 100 um

50 um

100 um

140 um

130 um dia.

140 um dia.Speced.

111 um high measured

Actual probe mark location

VSC 2 Probe Head

Bump

Die Surface

06/07/2005 SouthWest Test Workshop 2005 11

VS Crown Probe After Sorting 12 Wafers

Non-destructive on-line cleaning is neededto keep the contact surfaces debris free.

VS Crown Probe VS Crown Probe

VS Crown Probe VS Crown Probe VS Crown Probe

06/07/2005 SouthWest Test Workshop 2005 12

Path Resistance Measurement

• Shorted Probe Card PCB used to zero out the resistance measurements of Test Head and Test Head Cables.

• VS crown tip probe card used to probe a shorted wafer, using 150 um OD.

• Resistance measurements taken at 3 minute intervals, while resting on a die, no Z-up or Z-down in between.

• Zero out resistance measurements subtracted out to acquire actual path resistance.

06/07/2005 SouthWest Test Workshop 2005 13

Yield Comparison

• Use VS flat tip and VS crown tip to probe the same wafers.

• Perform selective resort (resorting bad dice only) with both technologies to achieve maximum yield for comparison.

• Inspect VS crown tip probe marks. • Inspect VS crown tip probes to compare bump

residue build-up. • Measure Planarity and Contact Resistance using

PRVX.

06/07/2005 SouthWest Test Workshop 2005 14

Testing Parameters

• Same tester and prober used throughout the experiments.

• VS Flat Tip parameters– Probing: 175um OD– Cleaning: 50um OD, 12 times every 50 dice

• Cleaning medium – 3M 1um lapping film

• VS Crown Tip parameters– Probing: 150um OD– Cleaning: 100um OD, once every wafer

• Cleaning medium – Probe Polish 99, filled cleaning polymer• Due to tip shape requirements a lapping film cannot be used

06/07/2005 SouthWest Test Workshop 2005 15

Physical Path Measured for Resistance

EpoxyEpoxy

Probe PinProbe Head

6.3 mil Copper Wire

PCB

Cross-sectional View

Shorted Wafer

Test Head

Connectors

Path Resistance measured outlined by Blue Arrows.

Signal

Ground

06/07/2005 SouthWest Test Workshop 2005 16

Path Resistance, 1st Die Probed

VS Crown Tip Second Resistance ReadingSingle TD, Four Readings over 3 min. intervals, 1st Die Probed

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0 50 101 152 202

Tester Channel

Res

ista

nce

(ohm

)

T0 T1 T2 T3

AVG Resistance: 1.075 ohm

06/07/2005 SouthWest Test Workshop 2005 17

Path Resistance, 2nd Die Probed

Double TD, Four Readings at 3 min. Intervals, 2nd Die Probed

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0 50 101 152 202

Tester Channel

Res

ista

nce

(ohm

)

T0 T1 T2 T3

AVG Resistance: 0.998 ohm

VS Crown Tip Second Resistance Reading

06/07/2005 SouthWest Test Workshop 2005 18

Path Resistance, 2754th Die Probed

Resistance Reading After Probing 2754 DieSingle TD, Four readings over 3 min. Intervals

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0 50 101 152 202

Tester Channel

Res

ista

nce

(ohm

)

T0 T1 T2 T3

AVG Resistance: 1.103 ohm

06/07/2005 SouthWest Test Workshop 2005 19

Average Path Resistance

• 25 Die Tested in a row, with no CleaningVS Crown Tip Average Path Resistance

25 Die in a row, no cleaning

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 D14 D15 D16 D17 D18 D19 D20 D21 D22 D23 D24 D25

Die # on Shorted Wafer

Res

ista

nce

(ohm

)

AVG MAX

open found on 1 pin on die 18 (removed from data)

06/07/2005 SouthWest Test Workshop 2005 20

Mechanical Performance Characterization

• Bench-top Materials Testing System – Assess cleaning material performance.– Evaluate applied load characteristics of probe.

High ResolutionCamera

Precision Z-stage

High MagnificationOblique Lens

06/07/2005 SouthWest Test Workshop 2005 21

Mechanical Performance Characterization

• High resolution and video imaging• Synchronized load vs. overtravel data acquisition

High Magnification and High Resolution Imaging

50-gram Load Cell with Cleaning Material Installed onto Platen

High MagnificationOblique Lens

Probes

VS Crown Tip 5 Probes on Probe Polish 150-um Z-overtravel

06/07/2005 SouthWest Test Workshop 2005 22

Probe Contact with BumpC

ompr

essi

on L

oad

(gra

ms)

1 x VS Crown Tip Probe at 150um OTon Solder Bump Material

VS Crown Tip Probes on Bump 150-um Z-overtravel

VS Crown Tip Probes on Bump 150-um Z-overtravel

Play Video

06/07/2005 SouthWest Test Workshop 2005 23

Probe Clean to Visualize Penetration

Crown tip penetratesinto polymer layer

Spring Engages

Com

pres

sion

Loa

d (g

ram

s)

1 x VS Crown Tip Probe at 150um OTon Probe Clean Material

VS Crown Tip Probe Cleaning 150-um Z-overtravel

VS Crown Tip Probes on Probe Clean 150-um Z-overtravel

Play Video

06/07/2005 SouthWest Test Workshop 2005 24

5 Probes on Probe Polish 99C

ompr

essi

on L

oad

(gra

ms)

Crown tip penetratesinto polymer layer

Spring Engages

5 x VS Crown Tip Probe at 150um OTon Probe Polish Material

VS Crown Tip Probes on Probe Polish 99 150-um Z-overtravel

VS Crown Tip Probes on Probe Polish 99 150-um Z-overtravel

Play Video

06/07/2005 SouthWest Test Workshop 2005 25

VS Crown Tip Probes After Online Cleaning

Cleaned Once per Wafer Cleaned Once per Wafer

VS Crown Probe VS Crown Probe

06/07/2005 SouthWest Test Workshop 2005 26

Yield Comparison

2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24

wafer #

yiel

d (d

pw)

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

reco

very

(%)

WS1 RS1 RS%

VSCC2 Split VSC Split

Final Yield EqualGood Dice Per Wafer Equal

VS Crown Tip vs. VS Flat Tip

VS Crown Tip VS Flat Tip

Final Yield EqualGood Die per Wafer Equal

06/07/2005 SouthWest Test Workshop 2005 27

Planarity Reading from PRVX

-1.500

-1.000

-0.500

0.000

0.500

1.000

1.500

1 26 51 76 101 126 151 176 201 226 251 276 301

Virtual Pad #

Mils

VS Crown Point Probe Card Planarity, I/O Only, 2754 Die Sorted

06/07/2005 SouthWest Test Workshop 2005 28

Contact Resistance Reading from PRVX

0.000

0.500

1.000

1.500

2.000

2.500

3.000

3.500

1 26 51 76 101 126 151 176 201 226 251 276 301

Virtual Pad #

Ohm

s

VS Crown Point Probe Card Contact Resistance, I/O Only, 2754 Die Sorted

06/07/2005 SouthWest Test Workshop 2005 29

Summary• VS crown tip probe path resistance is on the same order as

standard VS flat.• VS crown tip Path Resistance holds stable after 2500+ die

sorted and non-destructive cleaning only after each wafer.• VS crown tip is able to achieve maximum yield at first sort,

with lower resort recovery.• Probe marks generated by VS crown tip show minimal

disturbance to the bump structure, compared to VS flat tip and other vertical probing technologies.

• On-line cleaning with Probe Polish 99 was effective in keeping the crown tip clean without affecting the tip geometry in order to maintain consistent yield.

• Planarity remained at +/-1mil after probing 12 wafers.

06/07/2005 SouthWest Test Workshop 2005 30

Acknowledgements• Patrick Mui – Engineering Manager, JEM-America• Altera Manufacturing Engineering• JEM-Japan Test Engineering• ITS Applications Engineering

06/07/2005 SouthWest Test Workshop 2005 31

SWTW – 2005

Thank you for your attention

Questions ???