0.62 0imapsne.org/virtualCDs/2019/2019 Posters/E11.pdfProduct Level Design Optimization for Pad...

1
Product Level Design Optimization for Pad Cratering Reliability During Drop Impact Huayan Wang 1 , Jing Wang 1 , Jiefeng Xu 1 , Vanlai Pham 1 , Ke Pan 1 , Seungbae Park 1 , Hohyung Lee 2 , Gamal Refai-Ahmed 2 1 State University of New York at Binghamton, 2 Xilinx Inc BINGHAMTON UNIVERSITY STATE UNIVERSITY OF NEW YORK Mechanical Engineering Optomechanics & Physical Reliability Lab. * Professor Seungbae Park Objectives Results Background Methodology Conclusion Drivers of pad cratering Finer pitch Brittle laminates Stiffer solders Large heat sink Mitigation methods of pad cratering Solder mask defined Larger pad Limitations on board flexure Component location Edge bonding More compliant solder To evaluate various factors’ influence to the product level pad cratering risk SMD vs. NSMD Edge bonding effect Top plate material Heat sink weight Case SMD/NSMD Edge Bonding Top Plate Material Heatsink 1(Ref) NSMD No Al 1 2 SMD \ \ \ 3 \ Yes \ \ 4 SMD Yes \ \ 5 \ \ Stainless Steel \ 6 \ \ Zinc Alloy \ 7 \ \ \ 1.75 Design of experiments (\: same as reference case) Max stress on PCB in assembly process Max PCB stress after reflow Max PCB stress after top & bottom plates assembly Max PCB stress after heat sink assembly Max stress on PCB during drop Free drop method Input-G method Input-D method Direct acceleration input method 0 5 10 0 500 1000 Acceleration (m/s^2) Time (ms) Acceleration 88.23G Duration: 8.6ms Schematic of failure site observed during drop test Finite Element Modeling Sub modeling to get max stress on PCB Max PCB stress change in assembly process Max PCB stress in drop test Max PCB stress in assembly process 1.00 0.86 0.70 0.62 1.00 1.00 1.00 Ref Case2 Case3 Case4 Case5 Case6 Case7 0.0 0.5 1.0 Max Von Mises Cases SMD decreases the stress by 14% Edge bonding decreases the stress by 30% Combination of both can decrease the stress by 38% Top plate material and heatsink weight has no influence to the PCB stress Max PCB stress in drop test Heat sink weight has the most significant influence to the PCB stress during drop Edge bonding can effectively decrease the PCB stress during drop With the presence of edge bonding, the SMD’s mitigation to the PCB stress is neglectable Stiffer top plate can reduce the PCB stress during drop Both solder mask define and edge bonding can reduce PCB stress in assembly process Heat sink weight is the most inflectional factor to the PCB stress during drop With the presence of the edge bonding, SMD and NSMD yield the same max PCB stress Rigid reinforce plates can reduce the max PCB stress during drop Edge bonding, stainless steel plates and light heat sink design gives the best pad cratering reliability

Transcript of 0.62 0imapsne.org/virtualCDs/2019/2019 Posters/E11.pdfProduct Level Design Optimization for Pad...

Page 1: 0.62 0imapsne.org/virtualCDs/2019/2019 Posters/E11.pdfProduct Level Design Optimization for Pad Cratering Reliability During Drop Impact Huayan Wang1, Jing Wang1, Jiefeng Xu1, Vanlai

Product Level Design Optimization for Pad Cratering Reliability During

Drop ImpactHuayan Wang1, Jing Wang1, Jiefeng Xu1, Vanlai Pham1, Ke Pan1, Seungbae Park1, Hohyung Lee2, Gamal Refai-Ahmed2

1State University of New York at Binghamton, 2Xilinx Inc

BINGHAMTONU N I V E R S I T YSTATE UNIVERSITY OF NEW YORK

Mechanical

Engineering

Optomechanics & Physical Reliability Lab.

* Professor Seungbae Park

Objectives

Results

Background

Methodology

Conclusion

Drivers of pad cratering

Finer pitchBrittle laminatesStiffer soldersLarge heat sink

Mitigation methods of pad cratering

Solder mask defined

Larger pad

Limitations on board

flexure

Component location

Edge bonding

More compliant solder

To evaluate various factors’ influence to the product level pad cratering

risk

SMD vs. NSMD

Edge bonding effect

Top plate material

Heat sink weight

Case SMD/NSMD Edge BondingTop Plate Material

Heatsink

1(Ref) NSMD No Al 12 SMD \ \ \3 \ Yes \ \4 SMD Yes \ \

5 \ \ Stainless Steel \

6 \ \ Zinc Alloy \7 \ \ \ 1.75

Design of experiments (\: same as reference case)

Max stress on PCB in assembly process

Max PCB stress after reflow

Max PCB stress after top & bottom

plates assembly

Max PCB stress after heat sink

assembly

Max stress on PCB during drop

Free drop method

Input-G method

Input-D method

Direct acceleration

input method 0 5 10

0

500

1000

Accele

ration (

m/s

^2)

Time (ms)

Acceleration

88.23G

Duration: 8.6ms

Schematic of failure site observed during drop test

Finite Element Modeling

Sub modeling to get max stress on PCB

Max PCB stress change in

assembly process

Max PCB stress in

drop test

Max PCB stress in assembly process

1.00

0.86

0.700.62

1.00 1.00 1.00

Ref Case2 Case3 Case4 Case5 Case6 Case70.0

0.5

1.0

Max V

on M

ises

Cases

SMD decreases the stress by 14%

Edge bonding decreases the stress by

30%

Combination of both can decrease the

stress by 38%

Top plate material and heatsink weight

has no influence to the PCB stress

Max PCB stress in drop test

Heat sink weight has the

most significant influence to

the PCB stress during drop

Edge bonding can effectively

decrease the PCB stress

during drop

With the presence of edge

bonding, the SMD’s

mitigation to the PCB stress

is neglectable

Stiffer top plate can reduce

the PCB stress during drop

• Both solder mask define and edge bonding can reduce PCB stress in assembly process

• Heat sink weight is the most inflectional factor to the PCB stress during drop

• With the presence of the edge bonding, SMD and NSMD yield the same max PCB stress

• Rigid reinforce plates can reduce the max PCB stress during drop• Edge bonding, stainless steel plates and light heat sink design gives the

best pad cratering reliability