Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus...
Transcript of Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus...
![Page 1: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/1.jpg)
Case studies of Case studies of Wafer Sort Floor Problems Wafer Sort Floor Problems
Darren JamesDarren JamesSWTW Committee SWTW Committee
MemberMember
![Page 2: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/2.jpg)
Applying What We Learned• We’ve completed Gage R&R Study• We know we have a solid metrology tool• We know the limits of our metrology tool
• Now What?
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 22
![Page 3: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/3.jpg)
Where to apply metrology• Process monitor• New process validation
– Test at Temperature – Pad Shrink– New Technology (i.e. copper pads)
• Probe card qualification• Probe technology comparison• New equipment qualification• Maintenance monitoring• Tool evaluation and selection
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 33
![Page 4: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/4.jpg)
PROCESS MONITOR
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 44
![Page 5: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/5.jpg)
August 09
Fab Process Issue
A process issue, pad discoloration, was found during the experiment. WaferWoRx300 picked up the discoloration of the pads as extremely large scrub marks. This was reported back to the fab.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 55
![Page 6: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/6.jpg)
Real-time Process Feedback
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 66
Pietzschmann, et al, SWTW 2005
![Page 7: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/7.jpg)
Controlling the sort process
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 77
![Page 8: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/8.jpg)
NEW PROCESS VALIDATION
•Test at Temperature•Pad Shrink
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 88
![Page 9: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/9.jpg)
Thermal Movement of Probe Card
dependency on: - probe card construction- used stiffener material- PCB properties
(1) Z movement probe arrayrange: up to 110umtime: over 2h preheatingdirection: -Z or/and +Z
(2) X & Y material expansionrange: up to 15um / 150mmtime: < 20min of preheatingdirection: star like
(3) X or Y drift probe arrayrange: up to 35umtime: over 5h preheatingdirection: X or Y
(4) Single needle movementrange: 3…15umtime: 1st h of preheatingdirection: each needle different
-100
-75
-50
-25
0
25
50
75
0 10 20 30 40 50 60 70 80
Time [min]
Time [min]
Z-po
sitio
n [
m]
Pietzschmann, et al, SWTW 2005
![Page 10: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/10.jpg)
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1010
• Comparison of Y Wafer Scale between 2 wafers. Probed at 88°C (L) and 150°C (R).
• Y Wafer Scale identifies whether the die-to-die step size is accurate in the Y axis.
• An 8x1 probe card array was used to probe this wafer (outlined by black box).
• 88°C has an error of approx 7µ (aqua-orange), 150°C of approx 33µ (light blue-orange).
Thermal Movement of Probe Card
![Page 11: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/11.jpg)
Thermal Drift
Marks were drifting off of the pads during the probing process. A thermal drift was found on WaferWoRx300, changes to the prober settings helped to correct this issue.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1111
![Page 12: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/12.jpg)
Advanced Optical Analysis - Offset/scaling @ temp Assessment
hot cold
clearance X ~12mclearance Y ~ 9m
clearance X ~16mclearance Y ~14m
Scrub X/Y-Position Error Scrub X/Y-Position Error
Probe Card Verification at Temperature
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1212
Pietzschmann, et al, SWTW 2005
![Page 13: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/13.jpg)
PMA scrub measurements w/prober errors removed
PCA – scrub measurements
Validating Pad Shrink
Stacking all of the scrub marks to evaluate total sort process CpK
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1313
Evaluating the probe card by removing prober and setup error
![Page 14: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/14.jpg)
max. 8.4mavg. 4.0mstdev. 1.9m
significant improvement in probing process stability
2,5µm
First examples had to be reworked in every case
Spec ≥7µm
first delivery after feedback
Pad Edge Distance Worst Case
11,5µmSpec ≥7µm
max. 19.5mavg. 10.9mstdev. 5m
7,5µm
scrub mark analysis
3,5µm
0µm 9µm
12,5µm
7,5µm
Test at Temperature – Thermal movement analysis
Scrub X/Y-Position Error– Avg. per die
Scrub X/Y-Position Error– Avg. per die
Probe card verification
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1414
Pietzschmann, et al, SWTW 2005
![Page 15: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/15.jpg)
PROBE CARD QUALIFICATIONPROBE CARD QUALIFICATION
June 7 to 10, 2009June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test WorkshopIEEE SW Test Workshop 151515
![Page 16: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/16.jpg)
Blank/un-patterned wafer analysisVector view of scrub X-Y position
Vector view
New Probe Technology Validation
• Evaluating a new probe card
• Plot shows a thermal scaling affect
• Card did not pass incoming Q/A
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1616
![Page 17: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/17.jpg)
Comparison Major Scrub Lengths
probe card with new stiffener
X 1min PMI inspection
XX
X
Xdiff. 4m
diff. 20m
diff. 11m
diff. 8m
5min manual operator stop
125°C
old probe card
XX
X
Xdiff. 17m
diff. 18mdiff. 22m
125°C
diff. 18m
same product
Validating Probe Card Design
Thermal effects caused by probe card stiffener.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1717
New stiffener design provided better performance
![Page 18: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/18.jpg)
Loaded Planarity Deflection
Automated Deflection Test
• Deflection of probe card from no over travel until all probes are touching ~5 microns
• Low probes no deflection• High probes drift up as the probe card is over travelled
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1818
![Page 19: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/19.jpg)
Probe Card Deflection Graph
Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with card deflection.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 1919
![Page 20: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/20.jpg)
PROBE TECHNOLOGY
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2020
![Page 21: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/21.jpg)
Comparing Probe Card Technologies
Comparing different probe card technologies and validating which on will work with new smaller pad sizes.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2121
Pietzschmann, et al, SWTW 2005
![Page 22: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/22.jpg)
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2222
MEMS card Quad DUTCantilever Dual DUT
Comparing Probe Card Technologies
Comparing different probe card technologies.This plot shows scrub length.
![Page 23: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/23.jpg)
August 09
Card to Card same prober
Test was ran to determine variance between two different types of cantilever probecards. The same prober, prober settings, same time and temperature were used. Obvious difference found in theamount of pad damage between probe tip types. Less damage to pads while maintaining good contact with die is better.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2323
![Page 24: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/24.jpg)
EQUIPMENT COMPARISON
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2424
![Page 25: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/25.jpg)
prober to prober
Chuck tilt is present Card planarityCard to Chuck parallelism
Double touch is occurring during probe mark inspect on prober
Test was ran to determine prober to prober variance. The same card, same prober settings, same time and temperature were used. Chuck tilt and head stage to chuck parallelism were found and fixed.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2525
![Page 26: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/26.jpg)
August 09
Translation Effects
Note the change not only in area of the same probe mark from one td to the next, but also in shape and direction of scrub. This is due to effects of deflection and translation.
Some marks were failing probe mark inspection on the edges of the die. After running WaferWorx300, translation was determined to be the leading factor.
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2626
![Page 27: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/27.jpg)
MAINTENANCE MONITOR
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2727
![Page 28: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/28.jpg)
Using the Data to Address Equipment Problems
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2828
![Page 29: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/29.jpg)
Maintenance Validation
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 2929
![Page 30: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/30.jpg)
Scaling & Stepping Errors
Stage Stepping Accuracy
Stage Stepping
0
2
4
6
8
10
Y Scale X Scale
Mic
rons
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 3030
![Page 31: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/31.jpg)
Setup Validation
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 3131
![Page 32: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/32.jpg)
Probe Card Setup Errors
Operator Setupresults in X-Axis Offset Error
![Page 33: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/33.jpg)
Prober Setup Error
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 3333
![Page 34: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/34.jpg)
TOOL EVALUATION
![Page 35: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/35.jpg)
Failure Signature Example – chuck movement (theta)
standard vertical probe card : all needles scrub in one direction
This movement reduces the scrub length= online cleaning necessary
new vertical probe card : half of the needle scrub in opposite direction
problem overlap zones
highestfailure rate
Consistent pad damage = lowerfailure rate
Inconsistent pad damage = higherfailure rate
Prober vs Probe Card
Pietzschmann, et al, SWTW 2005
![Page 36: Case studies of Wafer Sort Floor Problems · 2017-03-26 · Plotting actual position versus expected position, shift is seen and Z loading increases causing pins to translate up with](https://reader033.fdocuments.in/reader033/viewer/2022060411/5f1086577e708231d4498988/html5/thumbnails/36.jpg)
Acknowledgements• Frank Pietzschmann
• Equipment used to gather the data
June 7 to 10, 2009June 7 to 10, 2009 IEEE SW Test WorkshopIEEE SW Test Workshop 3636