Burn-in & Test Socket Workshop

Transcription

1 Burn-in & Test Socket Workshop March 6-9, 2005 Hilton Phoenix East / Mesa Hotel Mesa, Arizona ARCHIVE TM

2 Burn-in & Test Socket Workshop TM COPYRIGHT NOTICE The papers in this publication comprise the proceedings of the 2005 BiTS Workshop. They reflect the authors opinions and are reproduced as presented, without change. Their inclusion in this publication does not constitute an endorsement by the BiTS Workshop, the sponsors, BiTS Workshop LLC, or the authors. There is NO copyright protection claimed by this publication or the authors. However, each presentation is the work of the authors and their respective companies: as such, it is strongly suggested that any use reflect proper acknowledgement to the appropriate source. Any questions regarding the use of any materials presented should be directed to the author/s or their companies. The BiTS logo and Burn-in & Test Socket Workshop are trademarks of BiTS Workshop LLC.

3 tm Burn-in & Test Socket Workshop Session 2 Monday 3/07/05 10:30AM Technical Program SOCKET DESIGN ADVANCEMENTS Development Of A Production Worthy Kelvin Contact Test Socket Fred Megna Micronics Japan Co., LTD Valen Burd Microchip Corporation Topless Burn-in Socket A Customer s Socket Standardization (follow-up to the keynote speaker 2004) Holger Hoppe Infineon Technologies AG The Case For A Universal Socket Footprint John Mendes IBM Microelectronics

4 Development of a Production Worthy Kelvin Contact Test Socket 2005 Burn-in and Test Socket Workshop March 6-9, 2005 Valen Burd Microchip Corporation 2355 W. Chandler Blvd Chandler, AZ Fred Megna Micronics Japan Co. LTD (Phoenix office) 1201 S. Alma School Rd. Suite 7550 Mesa, AZ TM

5 Static Device Tolerance Body Dimension (mm) Nominal Body Dimension Body Tolerance 12.00% 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% (% Dut Body) 6x6 qfn 28 5x5 qfn 32 11/12/2002March 7, x4 qfn 24 4x3 dfn 12 3x3dfn10 Dut Body 3x3 dfn8 2x2 dfn 6 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 2

6 Existing J-Element Device guide Device outline Contact width Dut Pad width Sense/Force j-elements are located side by side. TWCC is limited by dut pad width and device outline tolerance. 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 3

7 TWCC Specification CASE- A CASE- B CASE- C Socket PKG Minimum pad width (mm) PKG outline tolerance (+/- mm) Pad location tolerance (+/- mm) Socket guide clearance (mm) Socket guide tolerance (mm) J-Element location tolerance (+/- mm) Maximum isolation film width (mm) Minimum contact width (mm) (TWCC) /12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 4

8 Existing Design 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 5

9 New: Kelvin B-Type J-Element Device guide Device outline Contact width Device guide Device outline Dut Pad width DUT Pad width Sense/Force j-elements are located front to back. TWCC is limited by dut pad length which is typically 1.5x width. 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 6

10 J-Element Existing J-Kelvin New: Kelvin B-Type 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 7

11 Kelvin B Design 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 8

12 Operation 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 9

13 Current Capacity Kelvin B (R) Kelvin B(L) Delta ( o C) (A) 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 10

14 Kelvin B Specifications J-Element Type Kelvin B Path Length 1.8mm Inductance 0.58nH Contact Resistance 0.03 Ohms Contact Force 30 grams Minimum Pitch 0.5mm Current 3.0A Overtravel 0.2mm 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 11

15 Production Initial set up o C = 100% yield 1000 RT = 100% yield o C = 100% yield Comparison Yield (Delta) =6.3% Increase Maintenance Pin Drop Cleaning 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 12

16 Initial Production 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 13

17 14K cycles Production 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 14

18 43K cycles Production 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 15

19 Cleaning 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 16

20 Initial Production 14K cycles 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 17

21 Initial Production 43K cycles 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 18

22 Longevity and Maintenance Rolling contact Single elastomer 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 19

23 Future Plans Additional Beta Site Manufacturing Robustness Cleaning Rebuilding Additional Packages QFN 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 20

24 Valen Burd his idea Acknowledgements Microchip US Engineering Resource Microchip Thailand Manufacturing Support MJC Japan Design Development Support 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 21

25 Questions??? 11/12/2002March 7, 2005 BiTS 2005 Development of a Production Worthy Kelvin Contact Test Socket (Burd/Megna) 22

26 BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Topless Burn-in Socket - a Customer s s Socket Standardization (follow-up to the key notes speaker 2004) Holger Hoppe Infineon Technologies AG holger.hoppe@infineon.com Holger Hoppe Page 1.

27 Introduction BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Burn-in sockets (and boards) in memory mass production: still very cost intensive Constant demand decreasing cost per piece longer socket (and board) life time more universality for increasing package variety high parallelism of sockets on burn-in boards Holger Hoppe Page 2. socket size as small as possible New FBGA burn-in socket concepts needed!

28 A g e n d a BiTS Burn-in & Test Socket Workshop March 6-9, 2005 State-of-the-art Open top FBGA burn-in socket The new Topless approach The route to standardization Conclusions Holger Hoppe Page 3.

29 A g e n d a BiTS Burn-in & Test Socket Workshop March 6-9, 2005 State-of-the-art Open top FBGA burn-in socket The new Topless approach The route to standardization Conclusions Holger Hoppe Page 4.

30 State-of of-the-art Open top FBGA burn-in socket in socket (I) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Overview: 10 different parts (in total ~ 70) Guide Latch (2x) Slider Spring B (2x) Contacts (nx) Cover Spring A (4x) Insulator 1 Insulator 2 Locator Holger Hoppe Page 5. (PCB)

31 State-of of-the-art Open top FBGA burn-in socket in socket (II) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Holger Hoppe Page 6.

32 State-of of-the-art Open top FBGA burn-in socket Functionality in socket (III) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Open latches and contacts while pressing down cover Rough and fine alignment of dropping device inside the socket Hold down device in right position while closing the cover Contact all balls Holger Hoppe Page 7.

33 State-of of-the-art Open top FBGA burn-in socket How can the socket be simplified? in socket (IV) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Every single socket contains same features and mechanisms e.g. Opening and Alignment. Loader / Unloader are much more precise today (accuracy ~0.1mm). Move functionality from the socket to the loader. Reduce number of parts saving costs and space. Holger Hoppe Page 8.

34 A g e n d a BiTS Burn-in & Test Socket Workshop March 6-9, 2005 State-of-the-art Open top FBGA burn-in socket The new Topless approach The route to standardization Conclusions Holger Hoppe Page 9.

35 The new approach Topless (I) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Coverless Open socket directly with the loader head Remove cover Remove 4 springs Loader head to be aligned precisely on socket Slider and Latches directly actuated by loader head Holger Hoppe Page 10.

36 The new approach Topless (II) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Guideless Use perfect accuracy of loader Preciser guarantees accurate position of device at loader head Removing the guide opens the socket for wide range of package sizes Fine alignment only by top plate Holger Hoppe Page 11. Air flow over device improved Coverless + Guideless = Topless (patents pending)

37 The new approach Topless (III) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Holger Hoppe Page 12.

38 A g e n d a BiTS Burn-in & Test Socket Workshop March 6-9, 2005 State-of-the-art Open top FBGA burn-in socket The new Topless approach The route to standardization Conclusions Holger Hoppe Page 13.

39 The route to standardization (I) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Internal Initial ideas about Topless Patent registrations Lab tests + general investigations First Trial: SUCCESS Kick Off with Supplier 2 and loader supplier Production release: SUCCESS? First real Topless Trial: SUCCESS GITop Specification 11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05 02/05 03/05 External Supplier 3 Supplier 2 Project start: new socket development Project start: new socket development Fine adjustments Supplier 1 Holger Hoppe Page 14. Project start: feasibility study with modified open top socket Start new socket development

40 The route to standardization (II) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Supplier 1 Supplier 2 Supplier 3 Infineon Specification GITop Holger Hoppe Page 15. Loader/Unloader Supplier

41 The route to standardization (III) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Supplier 1 Supplier 2 Supplier 3 Infineon Specification Infineon NEW Specification GITop Holger Hoppe Page 16. Loader/Unloader Supplier

42 The route to standardization (IV) BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Holger Hoppe Page 17.

43 A g e n d a BiTS Burn-in & Test Socket Workshop March 6-9, 2005 State-of-the-art Open top FBGA burn-in socket The new Topless approach The route to standardization Conclusions Holger Hoppe Page 18.

44 Conclusions (I) Topless Burn Topless Burn-in socket BiTS Burn-in & Test Socket Workshop March 6-9, 2005 New loader/unloader generation allows new socket concept. New socket concept is very promising. Cover for socket actuation is not needed anymore. Guide for device alignment is not needed anymore. Reduced socket price. One socket can handle many devices. Final production release still open (expected mid of March 05). Holger Hoppe Page 19.

45 Conclusions (II) Standardization process Holger Hoppe Page 20. BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Quite difficult to combine 3 supplier technologies (+ 1 Loader manufacturer) Each supplier is trying to introduce and prevail its own technology and ideas. It needs a lot of communication, each major change has to be forwarded and aligned with all other parties. Avoid more than 2 suppliers! Try to determine all facts in your specification and avoid any softening of YOUR specification. BUT: At the end of the day you ll get 2 or more compatible, cheap, well performing products according your wishes (?) and a good competition between your suppliers!

46 BiTS Burn-in & Test Socket Workshop March 6-9, 2005 Thank You! Yamaichi Electronics Thanks to Enplas Semiconductor Texas Instruments Questions? Holger Hoppe Page 21.

47 Systems and Technology Group The Case for a Universal Socket Footprint 2005 Burn-in and Test Socket Workshop March 6-9, 2005 Burn-in & Test Socket Workshop TM John Mendes IBM Corporation Essex Junction VT

48 Agenda Introduction: Socket Footprints Vendor s Galore Application Diversity Board Layout Configurations Direct (Drop-In) Field Replacement Footprint Overlay 2 BiTS /12/2002March 7, 2005

49 Introduction Socket Footprints Are like noses everyone has one Come in all shapes and sizes Not compatible from device to device Not compatible from vendor to vendor Anyone wonder why? Designs are almost always vendor driven Little attention paid to this subject 3 BiTS /12/2002March 7, 2005

50 Vendor s Galore Multiple Vendors commonality Unique footprint design Able to modify existing footprints Competitive pricing scales Manufacturer Sources 4 BiTS /12/2002March 7, 2005

51 Application Diversity Electrical Performance High Frequency/Bandwidth (> 3GHZ) Low Noise/Crosstalk Contact Inductance Mechanical Performance Compliance Insertion Cycles Cleaning MTBF (Mean Time Between Failure) MTBM (Mean Time Between Maintenance) 5 BiTS /12/2002March 7, 2005

52 Board Layout Configurations Board Area Single vs. Multi DUT Device Decoupling Thru-Via and Probe Pad Placement Component Placement Restrictions Balanced/Matched Network Circuitry Handler Kit Configuration 6 BiTS /12/2002March 7, 2005

53 Direct (Drop-In) Field Replacement Conditions invoking changes Test Application Device Physical Characteristics Field Engineering Methods Tester or Manufacturing Equipment Commodity Price Flexibility 7 BiTS /12/2002March 7, 2005

54 Footprint Overlay Socket types Device Pockets Mounting Topology Multi-DUT Quad Octal Working with vendors 8 BiTS /12/2002March 7, 2005

55 LGA Footprint Overlay (Cont ) Socket Types CSP BGA QFN ubga PGA PLCC QFP MLF 9 BiTS /12/2002March 7, 2005

56 Footprint Overlay (Cont ) QFN 5x5 Example Device Pad Layout Vendor1 Layout Vendor2 Layout Vendor1 and Vendor2 Overlay Combined Socket Overlay Device Pocket Overlay (zoom) 10 BiTS /12/2002March 7, 2005

57 Conclusion Benefits Universal Hardware Compatible with existing Handler Kits One footprint for all devices Interchangeable between vendors Same device pad layout configurations Offset vs. linear contacts Flexible field replacement Hardware Cost reduction 11 BiTS /12/2002March 7, 2005

58 Conclusion (cont ) Possible Drawbacks Technical Feasibility Geometries too complex Costly to implement New tooling / Complex machining No buy-in by vendors Forgetaboudit Large Physical Aspects ASICS, PowerPC, etc. Limited Application BGA s, LCC s, MLF s, QFN s 12 BiTS /12/2002March 7, 2005

59 THANK YOU 13 BiTS /12/2002March 7, 2005

60 QFN 5x5 Example Layout 14 BiTS /12/2002March 7, 2005

61 Board Pad Layout 15 BiTS /12/2002March 7, 2005

62 Vendor1 Board Pad Layout 16 BiTS /12/2002March 7, 2005

63 Vendor2 Board Pad Layout 17 BiTS /12/2002March 7, 2005

64 Vendor1 and Vendor2 Overlay 18 BiTS /12/2002March 7, 2005

65 Combined Socket Overlay 19 BiTS /12/2002March 7, 2005

66 Device Pocket Overlay (zoom) 20 BiTS /12/2002March 7, 2005