Product / Process Change Notification

Transcription

1 Product / Process Change Notification N A Dear Customer, Please find attached our INFINEON Technologies PCN: Introduction of new tester platform for final test for PROFET BTS432E2 E3062A, BTS442E2 E3062A in PG-TO263-5 and BTS432E2 and BTS442E2 in PG-TO220-5 Important information for your attention: Please respond to this PCN by indicating your decision on the approval form, sign it and return to your sales partner before 30. November Infineon aligns with the widely-recognized JEDEC STANDARD JESD46, which stipulates: Lack of acknowledgement of the PCN within 30 days constitutes acceptance of the change. Your prompt reply will help Infineon Technologies to assure a smooth and well executed transition. If Infineon does not hear from your side by the due date, we will assume your full acceptance to this proposed change and its implementation. Your attention and response to this matter is greatly appreciated. Infineon Technologies AG Postal Address Headquarters: Am Campeon 1-12, D Neubiberg, Phone +49 (0) Chairman of the Supervisory Board: Wolfgang Mayrhuber Management Board: Dr. Reinhard Ploss (CEO), Dominik Asam, Dr. Helmut Gassel, Jochen Hanebeck Registered Office: Neubiberg Commercial Register Amtsgericht München HRB Page 1 of 2

2 Product / Process Change Notification N A Products affected: Sales Name SP N OPN Package Detailed Change Information: Subject: BTS432E2 E3062A SP BTS432E2E3062ABUMA1 PG-TO263-5 BTS432E2 SP BTS432E2BKSA1 PG-TO220-5 BTS442E2 E3062A SP BTS442E2E3062ABUMA1 PG-TO263-5 BTS442E2 SP BTS442E2BKSA1 PG-TO220-5 Introduction of new final test platform Reason: Phase out the Incal i9472 (AOT) tester platform Infineon Technologies Sdn. Bhd., Melaka, Malaysia Description: Old New Incal i9472 Teradyne µflex Product Identification: No change in product identification. Traceability will be ensured via bau number and date code. Impact of Change: The tester platform verification is performed via theadvanced Measurement System Analysis (AMSA) methodology, which includes a test result correlation and a bin sorting verification. No impact on quality and reliability. Processes are optimized to meet identical product performance according to specification. Attachments: 2_cip16067 Time Schedule: Final qualification report: Available on request First samples available: N.A. Intended start of delivery: October 2018 or earlier based on customer approval. If you have any questions, please do not hesitate to contact your local Sales office Page 2 of 2

3 Customer Information Package (CIP) for PCN A New tester platform for PROFET products BTS432E2 E3062A, BTS442E2 E3062A in PG-TO263-5 and BTS432E2 and BTS442E2 in PG-TO220-5 final test at Infineon Technologies Malaysia Sdn. Bhd., Melaka, Malaysia restricted

4 Content General information: Scope of change page 3 Tester platform change release procedure page 4 Product type reference matrix (PROFET ) and results page 5 Tester platform change release procedure for reference product page 6 type BTS432E2 and BTS442E2 Pass/Fail Analysis: Overview of Pass/Fail correlation results page 7-8 Test Stability Analysis: General information on AMSA procedure page 9 Principle of AMSA diagram page Results from AMSA test procedure for BTS432E2 (Cold Insertion) page Assessment of test parameters BTS432E2 (Cold Insertion) page Results from AMSA test procedure for BTS432E2 (Warm Insertion) page 20 Assessment of test parameters BTS432E2 (Warm Insertion) page Results from AMSA test procedure for BTS442E2 (Cold Insertion) page Assessment of test parameters BTS442E2 (Cold Insertion) page Results from AMSA test procedure for BTS442E2 (Warm Insertion) page 45 Assessment of test parameters BTS442E2 (Warm Insertion) page Conclusion of AMSA results page 58 2

5 Scope of change: In line with the company strategy to phase out the Incal i9472 (AOT) tester platform at IFMY, Infineon Technologies, Melaka, Malaysia, conversion projects for existing products tested at this platform is conducted. The Teradyne MicroFlex tester has been qualified as the new test platform for several products including the PROFET BTS432E2 and BTS442E2. 3

6 Tester platform change release procedure: The full test package correlation was executed on the BTS432E2 E3062A (PG-TO263-5) and BTS442E2 E3062A (PG-TO263-5). A single test package was created with an optimized correlation flow due to similarity for both devices (BTS432E2 & BTS442E2). Same test hardware (DUT-Board, Tester, Handler) Same test concept and test methodology Same functionality (PROFET ) Same wafer technology Same package and package technology (PG-TO263-5 and PG-TO220-5) All referenced product types were correlated by a Golden Device correlation & buy-off at all production test temperatures. 4

7 Customer Information Package (CIP) for A Reference matrix for PROFET Products: Reference Product BTS432E2 (TO263) BTS442E2 (TO263) BTS432E2 (TO263) BTS442E2 (TO263) Sales Name Package BuyOff Lots Correlation Reference Tester Platform Result BTS432E2 PG-TO263-5 AMSA 3 * 1k lots Incal i9472 Pass BTS442E2 PG-TO263-5 AMSA 1 * 1k lots Incal i9472 Pass BTS432E2 PG-TO220-5 GD - Incal i9472 Pass BTS442E2 PG-TO220-5 GD - Incal i9472 Pass 5

8 Tester platform change release procedure for reference product type BTS432E2 and BTS442E2: The evaluation of both test platforms was performed via 2 step approach: 1. A general pass/fail correlation based on 3 different lots for BTS432E2 (3 x 1k devices) and 1 lot for BTS442E2 (1 x 1k device) was performed in full productive mode on Incal i9471 and Teradyne Microflex test platform at Cold and Hot temperatures. Hereby a comparative study of every fail device was done and reported pass/fail changes were analyzed in detail. 2. Gage r&r (GrR) procedure was performed using the AMSA tool to validate tester stability with statistical tools. Hereby the Reproducibility and Repeatability of the measurement system is checked by repeated testing and comparison of measured values and binning (device by device, test by test). 6

9 Overview of Pass/Fail correlation results: BTS432E2: AOT uflex Pass / Fail Bin Flip Correlation Cold -43C Qty In Pass Fail Yield Qty In Pass Fail Yield 0 VC642895G % VC643246G % VC648915G % AOT uflex Pass / Fail Bin Flip Correlation Hot 150C Qty In Pass Fail Yield Qty In Pass Fail Yield 0 VC642895G % VC643246G % VC648915G % Remarks: VC643246G02 2 failures at IBS12 parameter with AOT rejects failing also at Flex test platform No quality risk. All electrical pass/fails were verified at both test platforms with test flow: AOT: Cold Hot (starting with 1000 units at first insertion and removing rejects) FLEX: Cold Hot (starting with 1000 units at first insertion and removing rejects) Rejects collected after final insertion has good correlation at both platforms. No risk of failed unit being delivered to customer. 7

10 Overview of Pass/Fail correlation results: BTS442E2: AOT uflex Pass / Fail Bin Flip Correlation Cold -43C Qty In Pass Fail Yield Qty In Pass Fail Yield 0 VC643247G % AOT uflex Pass / Fail Bin Flip Correlation Hot 150C Qty In Pass Fail Yield Qty In Pass Fail Yield 0 VC643247G % All electrical pass/fails were verified at both test platforms with test flow: AOT: Cold Hot (starting with 1000 units at first insertion and removing rejects) FLEX: Cold Hot (starting with 1000 units at first insertion and removing rejects) Rejects collected after final insertion has good correlation at both platforms. No risk of failed unit being delivered to customer. 8

11 General information on AMSA procedure The Advanced Measurement System Analysis (AMSA) methodology, which is based on the Gage r&r procedure, is a fast and efficient tool to: analyse and assess the test performance identify test instabilities (Repeatability, Reproducibility, Bin Flipping) evaluate the impact of the measurement process on the overall yield 9

12 Principle of AMSA diagram Gage r&r method: The Gr&R method is a widely used and recognized procedure to assess the statistical properties of reproducibility and repeatability of measurement systems. Based on this statistical procedure a Gr&R value can be calculated giving an indication on the variability of the test system. The variability is classified as excessive if Gr&R > 25%. Hereby reproducibility and repeatability are defined as: Reproducibility: is the variation in the average of the measurements made by different appraisers using the same measuring equipment when measuring the identical characteristic of the same part. Repeatability: is the variation in measurement obtained with one measurement instrument when used several times by an appraiser while measuring the identical characteristic on the same part 10

13 Principle of AMSA diagram The uncertainty of the measurement by using different hardware sets and possible offset of different test equipment is expressed by Gr&R. Therefore the test limits are guard banded compared to specification limits by corresponding Sigma value of determined Gr&R to ensure the specification limits, which is shown by below graph. LSL (low test spec min limit) (guard band) g Test no - name propability plots of all runs (offset) o HSL (high test limit) spec max (guard band) g Condition: test guard band g > test offset o Setup `1` = old tester Setup `2` = new tester Figure: Graphs showing repeated test values (blue, green) for the test parameter incl. the low and high test and specification (spec) limits. 11

14 AMSA Review Product: BTS432E2 12

15 Results from AMSA test procedure for BTS432E2 Within the AMSA procedure, the stability of test system was validated using 10 reference devices which was tested 2x on the Incal i9472 tester and 2x on two Teradyne Microflex testers at Hot (150 deg) and Cold (-43 deg) temperatures The test runs revealed the following results with GrR > 25% Cold 1. Test #755, VDT+ (Slew rate on) 2. Test #613, VBU+ (Undervoltage restart of charge pump) 3. Test #617, VBU- (Undervoltage shutdown) 4. Test #760, VDT- (Slew rate off) 5. Test #639, VI_D (Input threshold hysteresis) 6. Test #700, T+ (Turn-on time) 13

16 BTS432 Cold: 1. Test #755, VDT+ (Slew rate on) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. However both limits will be adjusted Revise limit New LSL on µflex : 1 V/µs New USL on AOT : 2 V/µs 14

17 BTS432 Cold: 2. Test #613, VBU+ (Undervoltage restart of charge pump) offset due to change in test method (pos. ramp vs binary search). There is no risk at LSL as it is not defined in datasheet. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 3.5V New USL on AOT 6.5V 15

18 BTS432 Cold: 3. Test #617, VBU- (Undervoltage shutdown) offset due to change in test method (neg. ramp vs binary search). There is no risk at LSL as the trend is shifted nearer to LSL (closer gap). As for USL, the new test method is similar to characterization of device design and there is no risk of delivering out of spec units to customer. Revise limit New USL on µflex: 4.2V New LSL on AOT: 3.2V 16

19 BTS432 Cold: 4. Test #760, VDT- (Slew rate off) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 1.9 V/µs New USL on AOT: 4.2V/µs 17

20 BTS432 Cold: 5. Test #639, VI_D (Input threshold hysteresis) offset due to change in test method for VI+ and VI- (pos. & neg. ramp vs binary search) causing a higher calculated delta value (VI_D = VI+ - VI-). There are no minimum and maximum values defined in datasheet for the delta parameter (typ. 0.5V). There is low risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 0.25V New USL on AOT: 0.9V 18

21 BTS432 Cold: 6. Test #700, T+ (Turn-on time) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. There is no risk at LSL as the trend is shifted nearer to LSL (closer gap). As for USL, there is possible risk due to shifted trend towards LSL. Revise limit New USL on µflex: 230µs New LSL on AOT: 71µs 19

22 Results from AMSA test procedure for BTS432E2 The test runs revealed the following results with GrR > 25% Hot 1. Test #639, VI_D (Input threshold hysteresis) 2. Test #617, VBU- (Undervoltage shutdown) 3. Test #760, VDT- (Slew rate off) 4. Test #709, T- (Turn-off time) 5. Test #604, VBO- (Overvoltage restart) 6. Test #245, ILK (Initial peak short circuit current limit) 7. Test #613, VBU+ (Undervoltage restart of charge pump) 8. Test #626, VI+ (Input turn-on threshold voltage) 9. Test #632, VI- (Input turn-off threshold voltage) 10.Test #755, VDT+ (Slew rate on) 20

23 BTS432 Hot: 1. Test #639, VI_D (Input threshold hysteresis) offset due to change in test method for VI+ and VI- (pos. & neg. ramp vs binary search) causing a higher calculated delta value. Trend shifted out of USL. There are no minimum and maximum values defined in datasheet for the delta parameter (typ. 0.5V). There is no risk of delivering out of spec units to customer. USL and LSL revised for trend shift. New USL on µflex: 0.75V New LSL on µflex: 0.37V No changes for AOT 21

24 BTS432 Hot: 2. Test #617, VBU- (Undervoltage shutdown) offset due to change in test method (neg. ramp vs binary search). There is no risk at LSL as the trend is shifted nearer to LSL (closer gap). Low risk at USL due to different test method used. Revise limit New USL on µflex: 3.9V New LSL on AOT: 3V 22

25 BTS432 Hot: 3. Test #760, VDT- (Slew rate off) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 1.15 V/µs New USL on AOT: 2.2 V/µs 23

26 BTS432 Hot: 4. Test #709, T- (Turn-off time) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. There is no risk at LSL as the trend is shifted nearer to LSL (closer gap). As for USL, there is possible risk due to shifted trend towards LSL. Revise limit New USL on µflex: 70µs New LSL on µflex: 23µs New LSL on AOT: 37µs 24

27 BTS432 Hot: 5. Test #604, VBO- (Overvoltage restart) offset. Reason for this offset is due to change in test method (neg. ramp vs binary search). No risk for LSL as new trend is shifted closer to LSL (DS min. 42V). For USL, there is no max limit defined in datasheet. No risk of delivering out of spec units to customer. None 25

28 BTS432 Hot: 6. Test #245, ILK (Initial peak short circuit current limit) offset. Reason for this offset is due to difference in parasitic inductance across both test platforms. No risk at LSL due to closer gap with new trend. Repeatability is also improved. No risk at USL as there is no max limit defined in datasheet at hot. None 26

29 BTS432 Hot: 7. Test #613, VBU+ (Undervoltage restart of charge pump) offset due to change in test method (pos. ramp vs binary search). There is low risk at LSL as it is not defined in datasheet. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on uflex: 3.5V No changes on AOT 27

30 BTS432 Hot: 8. Test #626, VI+ (Input turn-on threshold voltage) The high GrR value is caused by tight limits. (offset of <0.1V). Reason for this offset is due to change in test method for VI+ (pos. ramp vs binary search). There is low risk at LSL due to the trend shift. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. None 28

31 BTS432 Hot: 9. Test #632, VI- (Input turn-off threshold voltage) The high GrR value is caused by tight limits. (offset of <0.1V). Reason for this offset is due to change in test method for VI- (neg. ramp vs binary search). There is no risk at LSL due as the trend shifts closer to LSL. There is low risk at USL as the limit has 0.5V guard band from the datasheet max limit None 29

32 BTS432 Hot: 10. Test #755, VDT+ (Slew rate on) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. Low risk at LSL due to known test hardware differences. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on uflex: 0.6 V/µs New USL on AOT: 1.8V/us 30

33 AMSA Review Product: BTS442E2 31

34 Results from AMSA test procedure for BTS442E2 Within the AMSA procedure, the stability of test system was validated using 20 reference devices which was tested 2x on the Incal i9472 tester and 2x on one Teradyne Microflex testers at Hot (150 deg) and Cold (-43 deg) temperatures The test runs revealed the following results with GrR > 25% Cold 1. Test #755, VDT+ (Slew rate on) 2. Test #617, VBU- (Undervoltage shutdown) 3. Test #639, VI_D (Input threshold hysteresis) 4. Test #700, T+ (Turn-on time) 5. Test #709, T- (Turn-off time) 6. Test #613, VBU+ (Undervoltage restart of charge pump) 7. Test #760, VDT- (Slew rate off) 8. Test #245, ILK (Initial peak short circuit current limit) 9. Test #632, VI- (Input turn-off threshold voltage) 10.Test #750, TILK (Short circuit shutdown delay after input pos. slope) 11.Test #604, VBO- (Overvoltage restart) 12.Test #626, VI+ (Input turn-on threshold voltage) 32

35 BTS442 Cold: 1. Test #755, VDT+ (Slew rate on) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. Low risk at LSL due to test hardware differences. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 0.8 V/µs New USL on AOT 1.5V/µs 33

36 BTS442 Cold: 2. Test #617, VBU- (Undervoltage shutdown) offset due to change in test method (neg. ramp vs binary search). There is no risk at LSL as the trend is shifted nearer to LSL (closer gap). Low risk at USL due to shifted trend. Revise limit New USL on µflex: 4.2V New LSL on AOT 3V 34

37 BTS442 Cold: 3. Test #639, VI_D (Input threshold hysteresis) offset due to change in test method for VI+ and VI- (pos. & neg. ramp vs binary search) causing a higher calculated delta value (VI_D = VI+ - VI-) Trend shifted towards USL. There is no minimum and maximum value defined in datasheet for the delta parameter (typ. 0.5V). No risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 0.25V New USL on AOT: 0.78V 35

38 BTS442 Cold: 4. Test #700, T+ (Turn-on time) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. No risk at LSL. Low risk at USL due to shifted trend. Revise limit New USL on µflex: 265 µs New LSL on AOT: 130µs 36

39 BTS442 Cold: 5. Test #709, T- (Turn-off time) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. No risk at LSL. Low risk at USL due to shifted trend. Revise limit New USL on µflex: 68 µs New LSL on AOT: 19 us 37

40 BTS442 Cold: 6. Test #613, VBU+ (Undervoltage restart of charge pump) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. Low risk at LSL due to wider gap. No risk at USL. Revise limit New LSL on µflex: 3.5V New USL on AOT: 5.7V 38

41 BTS442 Cold: 7. Test #760, VDT- (Slew rate off) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. Low risk at LSL. No risk at USL. Revise limit New LSL on µflex: 1.4 V/µs New USL on AOT: 4.3V/µs 39

42 BTS442 Cold: 8. Test #245, ILK (Initial peak short circuit current limit) The high GrR value is caused by improved test stability. Reason for this offset is due to difference in parasitic inductance across both test platforms. 3k study shows good correlation. No risk None 40

43 BTS442 Cold: 9. Test #632, VI- (Input turn-off threshold voltage) The high GrR value is caused by tight limits. (offset of <0.1V). Reason for this offset is due to change in test method for VI- (neg. ramp vs binary search). There is no risk at LSL as the trend shifts closer to LSL. The USL is not guaranteed in Datasheet. None 41

44 BTS442 Cold: 10. Test #750, TILK (Short circuit shutdown delay after input pos. slope) The high GrR value is caused by difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is no risk for LSL due to trend shift. Low risk for USL due to wider gap. Revise limit New USL at uflex: 280µs New LSL at AOT: 100us 42

45 BTS442 Cold: 11. Test #604, VBO- (Overvoltage restart) offset. Reason for this offset is due to change in test method (neg. ramp vs binary search). No risk for LSL (DS min. 42V). No risk for USL (no max limit defined in datasheet). None 43

46 BTS442 Cold: 12. Test #626, VI+ (Input turn-on threshold voltage) The high GrR value is caused by tight limits. (offset of <0.1V). Reason for this offset is due to change in test method for VI+ (pos. ramp vs binary search). Low risk at LSL due to shift towards USL None 44

47 Results from AMSA test procedure for BTS442E2 The test runs revealed the following results with GrR > 25% Hot 1. Test #617, VBU- (Undervoltage shutdown) 2. Test #639, VI_D (Input threshold hysteresis) 3. Test #709, T- (Turn-off time) 4. Test #750, TILK (Short circuit shutdown delay after input pos. slope) 5. Test #604, VBO- (Overvoltage restart) 6. Test #632, VI- (Input turn-off threshold voltage) 7. Test #755, VDT+ (Slew rate on) 8. Test #760, VDT- (Slew rate off) 9. Test #626, VI+ (Input turn-on threshold voltage) 10.Test #700, T+ (Turn-on time) 11.Test #600, VBO+ (Overvoltage shutdown) 12.Test #100, RO12 (On-state resistance) 45

48 BTS442 Hot: 1. Test #617, VBU- (Undervoltage shutdown) offset due to change in test method (neg. ramp vs binary search). No risk at LSL. Low risk at USL due to trend shift. Revise limit New USL on µflex: 3.9V New LSL on AOT: 3.5V 46

49 BTS442 Hot: 2. Test #639, VI_D (Input threshold hysteresis) offset due to change in test method for VI+ and VI- (pos. & neg. ramp vs binary search) causing a higher calculated delta value (VI_D = VI+ - VI-) Trend shifted towards USL. There is no minimum and maximum value defined in datasheet for the delta parameter (typ. 0.5V). No risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 0.37V New USL on µflex: 0.85V New USL on AOT:0.65V 47

50 BTS442 Hot: 3. Test #709, T- (Turn-off time) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the switching behavior of the device. No risk at LSL. Low risk at USL due to trend shift. Revise limit New USL on µflex: 85µs New LSL on µflex: 25µs New LSL on AOT: 53µs 48

51 BTS442 Hot: 4. Test #750, TILK (Short circuit shutdown delay after input pos. slope) offset due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. No risk at LSL. Low risk at USL. Revise limit New USL on µflex: 380µs New LSL on AOT: 265us 49

52 BTS442 Hot: 5. Test #604, VBO- (Overvoltage restart) offset due to change in test method (neg. ramp vs binary search). No risk at LSL. Low risk at USL due to trend shift. None 50

53 BTS442 Hot: 6. Test #632, VI- (Input turn-off threshold voltage) offset due to change in test method for (neg. ramp vs binary search). Low risk at USL due to shift towards LSL None 51

54 BTS442 Hot: 7. Test #755, VDT+ (Slew rate on) offset. Reason for this offset is due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New LSL on µflex: 0.5 V/µs New USL on AOT: 1.6 V/µs 52

55 BTS442 Hot: 8. Test #760, VDT- (Slew rate off) offset. Reason for this offset is due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise LSL New LSL on µflex: 0.7 V/µs New USL on AOT: 2 V/µs 53

56 BTS442 Hot: 9. Test #626, VI+ (Input turn-on threshold voltage) offset. Reason for this offset is due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. None 54

57 BTS442 Hot: 10. Test #700, T+ (Turn-on time) offset. Reason for this offset is due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. Revise limit New USL on µflex: 305µs New LSL on AOT: 150µs 55

58 BTS442 Hot: 11. Test #600, VBO+ (Overvoltage shutdown) offset. Reason for this offset is due to difference in parasitic capacitance and inductance across both test platforms affecting the behavior of the device. There is possible risk at LSL due to the faster response at new test platform Flex. The new trend is shifted closer to USL and therefore, there is no risk of delivering out of spec units to customer. None 56

59 BTS442 Hot: 12. Test #100, RO12 (On-state resistance) Improved stability at new test platform None - 57

60 Conclusion of AMSA results AMSA Report: The Gr&R result shows good correlation between the 2 tester platforms (Incal i9472 & Teradyne Microflex). The assessment of parameters with Gr&R > 25% proves high quality of the Teradyne Microflex test platform. 58

61