Automotive Devices: Quad No- Lead (QFN) Technology with Inspectable Solder Connections

Transcription

1 Automotive Devices: Quad No- Lead (QFN) Technology with Inspectable Solder Connections FTF-SDS-F0026 Dwight Daniels Package Engineer A P R TM External Use

2 Agenda Wettable Lead Ends / Definition & Alternate Names QFN Packages / Intro to Features & Lead End Options Freescale Evaluations / Inspectable Solder Joint Study Solder Joint Life / How Good are They? Conclusion External Use 1

3 Purpose: Wettable Lead Ends A wettable lead end on a QFN is to promote wetting of solder to the ends of the QFN terminals during PCB assembly in order to form a solder fillet which is inspectable by AOI (Automated Optical Inspection). PROMOTE WETTING OF SOLDER wettable does not guarantee wetting SOLDER FILLET presence and shape are dependent on many factors of the PCB assembly process INSPECTABLE BY AOI assumes expertise and equipment capability at the PCB assembly facility External Use 2

4 Wettable Lead Ends A.K.A. Nomenclature: WETTABLE FLANK (WF) the flank (end) of the lead/terminal is created to be wettable INSPECTABLE JOINT (IJ) the lead/terminal ends are created to be wettable to promote a visible, inspectable joint/fillet SOLDERABLE LEAD END the lead/terminal end is created to allow solder to readily wet to it SIDE SOLDERABLE the side of the package (I.O.W. lead ends) is created to promote solder wetting SOLDER FILLET the resulting formation/presence of the inspectable solder joint In the case of the 1 st 4 -- A rose by any other name is still a rose. External Use 3

5 Agenda Wettable Lead Ends / Definition & Alternate Names QFN Packages / Intro to Features & Lead End Options Freescale Evaluations / Inspectable Solder Joint Study Solder Joint Life / How Good are They? External Use 4

6 QFN Packages (Quad Flat-pack No-leads) QFN / Package Structure Package Characteristics / Pro s &Con s Lead End Variants / Options & Creation External Use 5

7 Anatomy of a QFN Package Small Footprint Due to no lead extension Saw QFN Half-Etch Features Locking features, etc. Terminal - Design Variations Standard pull back e-type - exposed lead end Wettable plated lead end feature Exposed pad Good thermal efficiency Added mechanical interface wirebond flip chip Chip-on-Lead configuration No pad (typically) Accommodate a larger die:package ratio External Use 6

8 QFN Package Characteristics: General Small body sizes <2x2mm up to ~ 12x12mm (sweet spot 3x3mm to 8x8mm) Relatively low lead count Thin package height 1mm max height is typical Single sided mold no parting line Lead frame aligns with the seating plane of the package body Exposed flag is a standard feature (typically) Good thermal efficiency Solderable feature to improve mechanical robustness & board life. Leads/terminals contained within the physical outline of the body No bent/skewed leads No Trim/Form/Excise required No lead structure which extends beyond the body Small footprint No lead compliance Traditionally difficult to inspect solder joints after board assembly Inspection requirements are market dependent This is why we re here. External Use 7

9 QFN Terminal Design Variants * Standard Lead End Plastic Mold Compound (EMC) fills in the half-etched* portion of the lead end Exposed Lead End (e-type) Increased terminal surface area Potentially wettable terminal end surface Wettable Lead End (WF or IJ*) Wettable terminal end surface Used for applications which require AOI (Automated Optical Inspection) * Wettable Flank or Inspectable Joint -- or -- side solderable or External Use 8

10 QFN Terminal Design Variants * Standard Lead End Plastic Mold Compound (EMC) fills in the half-etched* portion of the lead end Plated Surface Shown in Gray Exposed Lead End (e-type) Increased terminal surface area Potentially wettable terminal end surface Wettable Lead End (WF or IJ*) Wettable terminal end surface Used for applications which require AOI (Automated Optical Inspection) * Wettable Flank or Inspectable Joint -- or -- side solderable or External Use 9

11 QFN Wettable Lead End Variants Wettable Lead End methods of formation Mechanical Removal: created by removing material from the package during the assembly process, usually by mechanical saw. - Often referred to as Step Cut Etching: typically created by the half etch process during lead frame manufacturing - Often referred to as Dimple Plating After Singulation: created after singulation in assembly by depositing wettable material on the lead end - Often performed by Electroless Sn plating External Use 10

12 Wettable Lead End Physical Comparison Step Cut 2-pass saw process Dimple Leadframe ½-etch process External Use 11

13 Wettable Lead End Step Cut Formation Step Cut 2-pass saw process Step-Cut saw blade 1 st pass saw Singulation saw blade 2 nd pass saw External Use 12

14 Wettable Lead End Variants Pros & Cons Mechanical (step cut) Freescale s preferred option Design Flexibility can be applied to existing leadframe designs Created using existing QFN assembly manufacturing processes IJ features are full lead width IJ features can be greater than half the lead frame thickness Adds nominal cycle time to the assembly process Requires plating process in assembly - even if using preplated leadframes Etched (dimple) Created during the normal leadframe manufacturing flow Equally compatible with standard Cu or pre-plated lead frames Half-etched features are limited to roughly half the leadframe thickness Features etched during leadframe manufacturing can t be full lead width Saw singulation of narrow, halfetched features are difficult to saw singulate cleanly - Imposes a practical minimum lead pitch for saw singulated QFN package outlines Consumes more of the surface area of the lead/terminal Post-Singulation (e-less) Design Flexibility can be applied to existing leadframe designs Maximizes wettable height & width for any given lead frame thickness Adds nominal cycle time to the assembly process Requires additive process in assembly - even if using preplated leadframes Usually created using electroless plating - Not a standard QFN assembly process - E-less plating can be difficult to maintain - Can complicate handling in assembly manufacturing. External Use 13

15 QFN & Wettable Lead Ends Wrap-up SMALL BODY, SMALL FOOTPRINT GOOD THERMAL EFFICIENCY VERSITILE MANUFACTURING CONFIGURATIONS VARIOUS TERMINAL DESIGN OPTIONS Including WF/IJ/SS options STEP CUT freescale PREFERRED OPTION FOR IJ/WF compatability with existing manufacturing robust performance More about this later External Use 14

16 Agenda Wettable Lead Ends / Definition & Alternate Names QFN Packages / Intro to Features & Lead End Options Freescale Evaluations / Inspectable Solder Joint Study Solder Joint Life / How Good are They? External Use 15

17 Inspectable Joint Step Cut Evaluation & Inspection Results Freescale Evaluation / Design & Set-up Representative Solder Joints / Good vs. Bad or Ugly? Inspection / Results and Comparisons External Use 16

18 Step Cut PCB Assembly Evaluation Study Factors for Board Assembly: Factors were chosen based on former board mount studies of QFN-IJ (wettable lead-end) packages. 2 package outlines: 5x5mm, 32ld QFN-IJ 8 DOE legs (step cut) + control cells with e-type leads 7x7mm, 48ld QFN-IJ 8 DOE legs (step cut) + control cells with e-type leads & punch dimple 2 PCB pad designs for each body size: Nominal lead size mm (under the package) + 0.3mm (extension beyond the package) Nominal lead size mm (under the package) + 0.6mm (extension beyond the package) 2 solder paste materials: Both SAC305 (Sn96.5/Ag3/Cu0.5), 89% metal load, ROL0, no clean Aging vs. No-Aging: 16hr, 155C aging bake prior to PCB assembly vs. No aging pre-bake External Use 17

19 Step Cut PCB Assembly Evaluation Board Assembly Study: overview of the numbers 768 study samples (step cut units): plus 288 control units 5x5mm, 32ld QFN-IJ 8 DOE legs (step cut). 2 control cells with e-type leads 7x7mm, 48ld QFN-IJ 8 DOE legs (step cut). 2+2 control cells with e-type & punch dimple step cut solder joints: plus control solder joints mm pad extension mm pad extension control parts (all with 0.6mm pad extension) 16 DOE Study Populations: plus 6 control cells: Temp Cycles (and climbing): External Use 18

20 PCB Assembly Process Factors Successful formation of inspectable solder joints is dependent on many factors of the PCB assembly process. PCB PAD DESIGN Pad length/extension (beyond package outline), feature clearance, consistent PCB quality of finish SOLDER PASTE MATERIAL flux system, alloy, quality/uniformity, compatibility (with overall process) PCB ASSEMBLY PROCESS OPTIMIZATION placement/alignment accuracy, dispense volume, reflow profile & conditions STENCIL DESIGN & MATERIAL material, thickness, aperature, release characteristics REFLOW CONDITIONS oven type, # of zones, temperature control, oven atmosphere, board layout Note: PCB assembly for this study was performed at an independent, 3 rd party contract board assembly provider no pre-assembly optimization was performed. External Use 19

21 PCB Assembly Process Factors Package dimensions Package Vehicle 1: Body size: 5x5mm Body Thickness: 0.85mm Leads: 0.4 x 0.5mm pitch Exposed Pad: 3.6 x 3.6mm Package Vehicle 2: Body size: 7x7mm Body Thickness: 0.85mm Leads: 0.4 x 0.5mm pitch Exposed Pad : 5.05 x 5.05mm Note: In this study the exposed pads of all packages were soldered to the PCB External Use 20

22 Step Cut PCB Assembly Evaluation Pad & Stencil Design Details: 2 PCB pad designs for each body size: Nominal lead size mm (under the package) + 0.3mm (extension beyond the package) Nominal lead size mm (under the package) + 0.6mm (extension beyond the package) PCB Pads Under Package Package Lead e.g. 0.4 x 0.25mm Exterior to Nominal Package Outline mm wide 0.45mm mm pad extension 0.3 mm pad extension Solder Paste 0.25 mm wide 0.35mm + [0.55 / 0.25] mm based on pad design for 0.6 / 0.3 pads External Use 21 Pictorial representations are NOT to scale

23 PCB Assembly Solder Fillet Formation 0.3mm pad design 0.6mm pad design External Use 22

24 PCB Assembly Solder Fillet Formation 0.3mm pad design 0.6mm pad design External Use 23

25 PCB Assembly Solder Fillet Formation (cross-sections) 0.3mm pad 0.6mm pad External Use 24

26 PCB Assembly Solder Fillet Formation (cross-sections) 0.3mm pad 0.6mm pad 200µm 200µm 200µm 200µm 200µm External Use 25

27 PCB Assembly Solder Fillet Formation comparison: Step Cut & Dimple - punch Step Cut 200µm solder joint formation solder joint cross-sections Dimple 200µm External Use 26

28 Step Cut PCB Assembly Evaluation Solder fillet formation by cell AND Manual vs. AOI Inspection results External Use 27

29 Step Cut PCB Assembly Evaluation Solder fillet formation vs. DOE Factors & Manual vs. AOI Inspection results Pad design is a dominant factor for consistent fillet formation This result is consistent with past Freescale fillet formation studies There is general alignment between manual & automated inspection methods - specifically, complete agreement that pad design is the key factor Presence of a fillet and interpretation of the shape of the fillet are somewhat subjective Note: Manual inspection & AOI confirm no missing fillets on all control lots. All control lots were assembled on pads with 0.6mm pad extension. Control cell 5-A-6-n exhibited relatively high occurrence of convex fillets External Use 28

30 PCB Assembly AOI Failure Mode Short pads are more susceptible to variation in fillet formation due to placement &/or solder printing process tolerance. This solder joint is formed and reliable, but AOI is all but certain to interpret the joint as missing. AOI failure AOI pass External Use 29

31 PCB Assembly AOI Failure Mode Poor alignment on a short pad is more likely to form a solder fillet that AOI will reject. pin 25 Note: Mitigation strategy for poor fillet formation is process optimization and increased (longer) pad length with accompanying paste print increase. External Use 30

32 Comments on Common Solder Joint Inspection Methods: AOI (Automated Optical Inspection) Faster and more repeatable than human eye or X-Ray Desirable for high reliability applications, such as automotive Relatively complex set-up and programming for interpretation of lighting X-Ray Relatively slow Preferred for identifying voids and defects located under the package Voids do not correlate well with early solder joint life failures Defects (such as electrical shorts) under the package are not common Human Eye Inspection SLOW! And subject to fatigue, distraction, boredom, hunger, restlessness, etc. Better at judging if a solder joint is formed when the fillet is not optimal External Use 31

33 PCB Assembly Evaluation Results - Wrap Up Board assembly of units, in 22 distinct populations, was performed using common industry methods & pre-selected BOM, for the purpose of studying inspectability & reliability of QFN packages with step cut IJ feature. GOOD PROCESS YIELD especially considering the process wasn t optimized >99% process yield hard failures (opens/shorts due to print errors) ALIGNMENT BETWEEN HUMAN-EYE INSPECTION & AOI agreement that we achieved 100% solder fillet formation on long (0.6mm) pads agreement that all discrepent or marginal fillets occurred on short (0.3mm) pads poor agreement on which solder joints (fillets) were considered visual fails SOLDER WETTING & WETTED HEIGHT excellent wetting to the lead ends wetted solder height consistently >140um increased variability in solder fillet formation/shape on shorter (0.3mm) pads SUCCESSFUL AUTOMATED OPTICAL INSPECTION capable of performing relatively fast inspection requires expertise to set-up and program properly likely to reject reliable solder joints if board assembly process exhibits significant variation External Use 32

34 Agenda Wettable Lead Ends / Definition & Alternate Names QFN Packages / Intro to Features & Lead End Options Freescale Evaluations / Inspectable Solder Joint Study Solder Joint Life / How Good are They? External Use 33

35 Solder Joint Life Joint Life / Study Results Feature vs. Feature / Where does the weakness lie Fillet formation / How this factor influences outcome External Use 34

36 Solder Joint Test Study Conditions Solder Joint Life Testing (SJRT) performed at internal Freescale lab. DAISY CHAIN CONFIGURATION ALL UNITS/BOARDS INSITU MONITORING real-time event detection/recording. AIR-to-AIR TEMPERATURE CYCLE Temperature range = -40C / 125C approx 1 hour cycles 15minute dwells & 15minute transitions ALL BOARDS TEMP CYCLED SIMULTANEOUSLY All SJRT printed circuit boards are cycling in the same chamber External Use 35

37 Step Cut Solder Joint Reliability Testing Results to date 5x5mm 32ld: 5x5mm Step cut DOE - SJRT 4333 cycles CELL paste pad ext aging Cell 1st Fail # Failed % Failed 7x7mm 48ld: 7x7mm Step cut DOE - SJRT Step Cut 51 A 0.6 No % Step Cut 52 A 0.6 Yes % Step Cut 53 A 0.3 No % Step Cut 54 A 0.3 Yes % Step Cut 55 B 0.6 No % Step Cut 56 B 0.6 Yes % Step Cut 57 B 0.3 No % Step Cut 58 B 0.3 Yes % 4333 cycles CELL paste pad ext aging Cell 1st Fail # Failed % Failed Step Cut 71 A 0.6 No % Step Cut 72 A 0.6 Yes % Step Cut 73 A 0.3 No % Step Cut 74 A 0.3 Yes % Step Cut 75 B 0.6 No % Step Cut 76 B 0.6 Yes % Step Cut 77 B 0.3 No % Step Cut 78 B 0.3 Yes % Note: e-type control cells show similar SJRT performance (all control cells = 0.6pads & no bake). 3 of 4 e-type control cells have 0 failures at 4333 cycles. 1 control cell (e-5-a-3-n) has 9 failures 1 st fail at 2796 cycles. External Use 36

38 Solder Joint Reliability Test results as of 4333 cycles A-6-n A-6-b A-3-n A-3-b B-6-n B-6-b 3459 * 5-B-3-n *The single solder joint from cell 57, labeled as missing based on human-eye inspection, survived at least 3459 cycles during board life testing (-40/125C). 5-B-3-b External Use 37

39 Board Life Charts 5x5mm study cells (w/enough fails to chart) External Use 38

40 Board Life Charts 5x5mm study cells (w/enough fails to chart) Outlier early life failure probable manufacturing artifact due to non-optimized PCB assembly process External Use 39

41 Solder Joint Reliability Test results as of 4333 cycles A-6-n A-6-b A-3-n A-3-b 7-B-6-n B-6-b B-3-n B-3-b External Use 40

42 Board Life Charts 7x7mm study cells (w/enough fails to chart) External Use 41

43 Board Life Charts 7x7mm study cells (w/enough fails to chart) Outlier early life failure probable manufacturing artifact due to non-optimized PCB assembly process External Use 42

44 Step Cut PCB Assembly Evaluation Solder Joint Reliability vs. DOE Factors (based on failures after 4.3K cycles) All PCB Assembly cells Solder: A vs. B Pad: 0.6 vs Aging: y vs. n Solder paste system shows the largest impact on solder joint life Pad size may be a factor, but likely due to board assembly variation rather than actual weakness related to the presence or quality of the fillet. Body size: 5x5 vs. 7x Note: e-type control cells show similar SJRT performance and response to DOE factors. 3 of 4 e-type control cells have 0 failures at 4333 cycles. 1 control cell (e-5-a-3-n) has 9 failures 1 st fail at 2796 cycles. External Use 43

45 Agenda Wettable Lead Ends / Definition & Alternate Names QFN Packages / Intro to Features & Lead End Options Freescale Evaluations / Inspectable Solder Joint Study Solder Joint Life / How Good are They? CONCLUSION External Use 44

46 Conclusion Freescale IJ Offering / Preferences & Availability Practicality / In the hands of the customers External Use 45

47 Freescale Inspectable Joint (IJ) Features The step cut (mechanical removal by saw) is Freescale s primary solution for QFN-IJ (QFN packages with inspectable joint lead-end features) DEMONSTRATED MANUFACTURABILITY Repeatable results for our customers as well as for Freescale ROBUST SOLDER JOINT RELIABILITY SJR >2000+ cycles (-40/125C) assumes a well designed/optimized PCB assembly process SUPERIOR AVAILABILITY/COMPATABILITY Adaptable to the widest array of QFN outlines and manufacturing lines The dimple (etched) IJ feature is offered by Freescale as an alternative QFN-IJ solution for use cases which cannot use the step cut due to specific application requirements. External Use 46

48 Practical Considerations What Matters Consistent formation of inspectable solder joints requires a well designed and optimized board assembly process. DESIGN: Matters One of the more significant factors in producing good, consistent inspectable solder joints PCB pad design, feature clearances, finish quality, PCB construction, etc. PROCESS : Matters Many factors of the board assembly process influence successful formation of inspectable solder joints Material choices consistently prove to be a significant factor in successful formation of inspectable solder joints as well as solder joint life. Choose a good quality, reliable solder paste material system & supplier AOI: Customer s choice Capability & repeatability are highly dependent on set up, programming, lighting, etc. Likely to reject reliable solder joints if PCB design and assembly process aren t optimized WF / IJ FEATURE TYPE: Doesn t really matter Step-cut or half-etched or e-type all offer excellent solder joint life, Step-cut or half-etched or e-type all have good potential for formation of inspectable joints Wettability of the lead end does not guarantee wetting to the lead end or an inspectable joint External Use 47

49 THANK YOU External Use 48

50 Freescale Semiconductor, Inc. External Use