Fusion utilization update 7
0 likes178 views
This document summarizes tester utilization performance from weeks 210-230 and 240-245. It provides utilization percentages for different tester systems, with Fusion FT showing an improvement from 65.6% to 77.6%. The document identifies key issues for different tester types, such as contact problems for Fusion WT, and shows graphs comparing utilization for Fusion before and after using a new pogo ring design. It also discusses actions to address prober accuracy problems, such as classifying probers based on accuracy and inspecting machines.
Fusion utilization update 7
- 1. Starting Point :Tester Utilization Performance (av. week 210-230 as presented in AIMS 28-08-2002) Tester Installed Planning Operation Overall Overall System Base Utilization Utilization Utilization Utilization in PSK from ACS from ACS from ACS in Output # % % % % Fusion FT 13 84.1% 77.6% 65.6% 49.7% Fusion WT 21 99.2% 66.7% 66.2% 49.0% Catalyst FT 14 74.2% 65.6% 48.7% 52.7% Catalyst WT 4 ? 74.2% ? 65.6% ? 48.7% ? 52.7% VQ FT 37 91.1% 68.0% 62.1% 63.5% VQ WT 17 ? 91.1% ? 68.0% ? 62.1% ? 63.5% Turbo FT 26 94.4% 88.0% 83.2% 75.0% Turbo WT 20 82.5% 85.5% 71.7% 64.2% Total/Av. 152 89.1% 74.1% 66.3% 61.1% Target Operational Utilization: - Immature Testers/Applications > 80% - Mature Testers/Applications > 90% Division, Content owner, Date(in numbers), ISO No 1
- 2. Current Tester Utilization Performance (av. week 240-245) Tester Installed Planning Operation Overall Overall System Base Utilization Utilization Utilization Utilization in PSK from ACS from ACS from ACS in Output # % % % % Fusion FT ? ? ? ? ? Fusion WT ? 28.5 ? 99.4 ? 78.6 ? 78.2 ? 0 Catalyst FT ? ? ? ? ? Catalyst WT ? ? ? ? ? VQ FT ? ? ? ? ? VQ WT ? ? ? ? ? Turbo FT ? ? ? ? ? Turbo WT ? ? ? ? ? Total/Av. ? ? ? ? ? Current Operational Utilization: - Quite improved compared to first half 2002 by ongoing improvement-actions. - Target Operational Utilization can be > 80% for immatutr testers/ applications. Division, Content owner, Date(in numbers), ISO No 2
- 3. Fusion utilization Planning Utilization (a) Operation Utilization (b) Overall Tester Utilization (a)x(b) Overall Tester Utilization from Output 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 Wk210~wk230 Wk231~wk245 20.0 Avg. Util. : 68.51% Avg. Util. : 77.39% 10.0 0.0 WK210 WK212 WK214 WK215 WK216 WK217 WK218 WK219 WK220 WK221 WK222 WK223 WK224 WK226 WK227 WK228 WK229 WK231 WK232 WK233 WK234 WK235 WK236 WK238 WK239 WK240 WK241 WK243 WK244 WK245 WK211 WK213 WK225 WK230 WK237 WK242 Division, Content owner, Date(in numbers), ISO No 3
- 4. Fusion utilization 2002-10-07~2002-10-27 Utilization Efficiency Prod.(%) Idle.(%) System .(%) Type.(%) Prober.(%) Contact.(%) Sm all_D.(%) Unknow n_D.(%) Schedule_D.(%) N-Prod.(%) 65.49% 83.26% 78.65% 0.21% 1.25% 4.40% 0.39% 0.50% 0.01% 2.03% 0.55% 12.00% Fusion Utilization 2002-10-07 ~ 2002-10-27 Unknown-down Non-Production Prod.(%) 2.03% 12.00% Type down 4.40% Idle.(%) System.(%) Type.(%) Prober.(%) Contact.(%) System down 1.25% Small_D.(%) Unknown_D.(%) Schedule_D.(%) Production 78.65% N-Prod.(%) Division, Content owner, Date(in numbers), ISO No 4
- 5. WT Utilization Pilots in PSK (Top 5 key issues to be tackled) Fusion WT Catalyst WT VQ WT Operational Utilization av. wk 210-230 66.7% av. wk 210-230 ..% av. wk 210-230 ..% av. wk 240-245 78.6% Key Issues: 1.Docking .4.1% 1.Contact ....% 1.Prober ....% problems accuracy Info. From FTF 2.Prober ..3% 2.Type down ....% accuracy 3.System hang .0.3% Finished. 4.Prober-Wafer ....% 3.Prober-Wafer ....% 2.Prober-Wafer ....% contacting contacting contacting 5. Abrasive paper 2% Operational Utilization Target 86.0% Target 80.0% Target 80.0% Division, Content owner, Date(in numbers), ISO No 5
- 6. Fusion New Design Pogo Ring Contact Problem is a key issue Under Survey. in Fusion of WTF. New design of Contact is better than Pogo Ring by Before, testing is more Engineering Stable ? Group. Division, Content owner, Date(in numbers), ISO No 6
- 7. Fusion New Design Pogo Ring Use new pogo ring New pogo ring is better. Use old pogo ring Trace the same type in the same days,but the results are different Division, Content owner, Date(in numbers), ISO No 7
- 8. Fusion New Design Pogo Ring Fusion#06 Utilization 2002-1007~2002-1107 The re-test rate of 120.00% new pogo ring is lower Fusion#34 Utilization 2002-1007~2002-1107 100.00% Than old one. 120.00% 80.00% Yield 100.00% 60.00% Utilization 80.00% RetestRateYield 40.00% 60.00% Utilization 20.00% 40.00% RetestRate 0.00% 20.00% 185 116 139 162 208 24 47 70 93 1 0.00% 103 120 137 154 35 69 86 18 52 1 Division, Content owner, Date(in numbers), ISO No 8
- 9. Fusion New Design Pogo Ring Machine Number Installation date Description Remark Use from 10/17 till 11/14 then new pogo ring Fusion#34 2002/10/17 with prober move to Fusion#03 in 11/15 Fusion#02 2002/11/1 in use ** Fusion#07 2002/11/1 in use ** Fusion#05 2002/11/5 in use ** Fusion#09 2002/11/14 in use ** Fusion#03 2002/11/15 in use ** Total use 5 sets in site 2002/11/18 Division, Content owner, Date(in numbers), ISO No 9
- 10. Fusion New Machine stop & Inspection Design Pogo Ring ⇒ Utilization lost ⇒ Prober accuracy problem New docking. Old docking. Ideal New docking:0.783 Utilization Expectation:0.95 distribution. Target:17% Discuss with prober New Docking will increase Vendor - Undergoing! Utilization effectively Division, Content owner, Date(in numbers), ISO No 10
- 11. Fusion New Design Pogo Ring Fusion07 New Pogo Ring Installation days :11/01 Trace type: PCF8811BW1_b Utilization Improvement: 0.685 -> 0.827 = > improve 14.2% From Operation Utilization: 0.786 -> 0.827 = > improve 4.1% Improvement Estimation: >= 4.1% Division, Content owner, Date(in numbers), ISO No 11
- 12. Prober accuracy problem: Current spec. survey •Prober accuracy +/- 4 um •Probe card alignment +/- 6 um (epoxy) (New probe card alignment) +/- 5 um (epoxy) •Bump placement accuracy +/- 4 um •Human variation +/- 2 um ? How to ensure probe How to measure Accumulated shift at worst case: 4+6+4+2 = 16 um ! for 32 um bump width ensure accuracy? prober How to accuracy? card Bump placement PCF8813 bump width: 40 um accuracy? Division, Content owner, Date(in numbers), ISO No Pitch 12
- 13. Prober accuracy verification : Alignment Error Condition: Wafer : Standard wafer Temp : 30 degree C Probing mode : Horizontal (From wafer top) Reference Die : Wafer center die Die size : 82x82 mil Rotation Error = 0.0015413 Degrees no:4090u-73 Prober X range Y range Cpk(X) Cpk(Y) Measure Data 8.330 9.460 0.896 0.744 After Rotation 5.387 6.210 1.437 0.947 Division, Content owner, Date(in numbers), ISO No 13
- 14. Prober accuracy verification – Wafer to wafer accuracy error Condition: • Prober :4090u-13 • Wafer type: 40913 • Die size : X 139.643 mil , Y 122.831 mil • Temperature : 30 degree C • Load wafer to chuck profile alignment • Train reference target on die of wafer center for measurement • Probed and measured the reference target position • Repeat measured wafer to wafer for 25 times. Unload wafer reload profile alignment probing & measurement Repeat measured 25 times load & unload Wafer to wafer average X shift 0.43 Wafer to wafer average Y shift 0.41 Wafer to wafer Max X shift 1.96 Wafer to wafer Max Y shift 1.72 25 wafers X shift range -4.16 ~ 8.30 = 12.46 25 wafers Y shift range -3.41 ~ 7.61 = 11.02 25 wafers average X range 6.49 25 wafers average Y range 6.17 Max X shift range 8.08 Min X shift range 5.71 Max Y shift range 8.81 Min Y shift range 5.24 Average Cp(X) 1.15 Average Cp(Y) 1.10 Division, Content owner, Date(in numbers), ISO No 14
- 15. Prober accuracy verification – (Die size V.S accuracy) Condition: X,Y within ± 4 um •Prober: X out of 4090u-57 spec •Wafer: Y out of spec Glass Mask 82Size : 82 x 82 Die X 82 471.5 :X 82 Die Size 471.5 x 82 82 X 471.5 Die Size : 82 x 471.5 X & Y out 6 6 6 •Temp: of spec 4 4 4 2 2 2 30 degree C -6 -4 -2 0 0 2 4 6 -6 -4 -2 0 0 2 4 6 -6 -4 -2 0 0 2 4 6 •Probing mode: -2 -2 -2 -4 -4 -4 From wafer top -6 -6 -6 Die size (mil) 82 X 82 471.5 X 82 82 X 471.5 (Horizontal) X min shift -4.62 -3.15 -3.34 X max shift 3.35 4.33 3.46 •Reference Die: X mean -0.9064 0.5133 -0.5233 Wafer center die Y min shift -3.41 -4.32 -4.53 Y max shift 4.86 3.36 3.28 Y mean 0.4553 -1.0811 -0.8071 X range 7.97 7.48 6.80 Y range 8.27 7.68 7.81 Division, Content owner, Date(in numbers), ISO No 15
- 16. Prober accuracy problem - action Use CMS tool to classify prober to 4 levels: If (1)Cp > 1.33, class: A -- Prober Quantity: 21 (2) 1.0 < Cp <= 1.33, class: B (3) Cp <= 1.0, class: C (4) Prober accuracy more than 8 um, class: D Contact and discuss with Vendor, Machine stop & Inspection Survey & Action are undergoing ! Time reduce, then Utilization will go up Deservedly!! Division, Content owner, Date(in numbers), ISO No 16
- 17. Action – (4) New Material Abrasive Paper Probe Tip Abrasive Side Result: Clean effect Cleaning Layer is better than Cushion Layer before Liner Paper Cleaning method CL type cleaning film B2L type cleaning film Yield (particle size: 3.0μm) (particle size: 1.0μm) Tester FUS-10 78% 73% Before : 18 mil 1 um FUS-34 79% 76% Now : 18 mil 3 um FUS-35 78% 70% Division, Content owner, Date(in numbers), ISO No 17
- 18. Division, Content owner, Date(in numbers), ISO No 18