Controlling the Solder Joint Reliability of eWLB Packages in Automotive Radar Applications Using a Design for Reliability Approach

Author(s):  
M. Niessner ◽  
G. Haubner ◽  
W. Hartner ◽  
S. Pahlke

A DfR (Design for Reliability) approach which is systematically based on simulation, sensitivity analysis and experimental validation is applied for identifying, understanding and controlling the key factors which determine the solder joint reliability of eWLB (Embedded Wafer Level Ball Grid Array) packages that carry embedded 77 GHz dies and sit on hybrid PCB (Printed Circuit Board) stacks. The hybrid stack investigated in this work is characteristic to automotive RADAR (Radio Detection And Ranging) applications and consists of one low-loss RF (Radio Frequency) layer and several FR4 layers. In line with previous work [1], the mechanical material properties of the low-loss RF laminate material are found to be the key factor. Simulation is used to systematically screen for mechanical properties which are favorable for achieving a high solder joint reliability on the unconstrained PCBs used for standardized solder joint reliability testing. A simplified virtual assessment of PCBs constrained by the mounting in system module housings is done. Both simulation and experimental results show that RF laminate materials with low Young’s modulus are the class of materials which allows for the highest solder joint reliability for all the conditions investigated in this study.

Author(s):  
John Lau ◽  
Yida Zou ◽  
Sergio Camerlo

The creep analyses of solder-bumped wafer-level chip-scale package (WLCSP) on printed circuit board (PCB) subjected to temperature cycling loading are presented. Emphasis is placed on the effects of PCB thickness on the solder joint reliability of the WLCSP assembly. Also, the effects of crack-length on the crack tip characteristics such as the J-integral in the WLCSP solder joint are studied by the fracture mechanics method. Finally, the effects of voids on the crack growth in the WLCSP solder joint are investigated.


Author(s):  
Jefferson Talledo

Leadframe-based packages are commonly used for semiconductor power devices. With these packages, heat dissipation is much better compared with laminate substrated-based packages. However, the solder joint reliability requirement under thermal cycling condition is also higher and this is what makes the development of a power package challenging. One of the usual requirements from customers is that there should be no solder joint failure up to 2,000 thermal cycles. This paper presents the thermomechanical simulation of a power leadframe package that was conducted to improve its solder joint reliability. Board level solder joint cycle life was predicted using finite element analysis and the result was validated with actual solder life result from board level reliability evaluation. Since available solder prediction equation was for the characteristic life (63.2% accumulative failure), using the normalized characteristic life was implemented for predicting the number of cycles to first failure of the solder joint connection and the approach showed good agreement with the actual result. Results also indicated that the choice of epoxy mold material and the type of PCB (printed circuit board) have a significant contribution to the solder joint reliability performance.


Author(s):  
Jefferson Talledo

Solder joint reliability is very important to ensure that an integrated circuit (IC) semiconductor package is functional within its intended life span as the solder joint establishes electrical connection between the IC and the printed circuit board (PCB). Solder fatigue failure or crack under thermal cycling is one of the common problems with board-mounted packages. There are several factors or package characteristics that have impact on solder fatigue life like package size and material properties of the package components. This paper presents a thermo-mechanical modeling of a leadframe-based semiconductor package to study the impact of lead sidewall solder coverage and corner lead size on the solder joint reliability. Finite element analysis (FEA) technique was used to calculate the solder life considering 50% and 100% package lead sidewall solder coverage as well as smaller and larger critical corner leads of the package. The results of the analysis showed that higher lead sidewall solder coverage and larger lead could significantly increase solder life. Therefore, ensuring lead sidewall solder wettability to have higher solder coverage is beneficial. The study also reveals that packages with side wettable flanks are not only enabling high speed automated optical inspection required for the automotive industry, but they are also providing improved solder joint reliability.


2020 ◽  
Vol 2020 (1) ◽  
pp. 000001-000004
Author(s):  
Dae-Suk Kim ◽  
Karthikeyan Dhandapani

Abstract An updated solder joint reliability (SJR) modeling methodology under thermal cycling (TC) is proposed and implemented for the diagonal solder crack path case as well as the SJR correlation of wafer-level package (WLP) and fan-out wafer-level package (FOWLP) data, which have the conventional solder failure mode around the under-bump metallization (UBM). First, two critical element layers near by the UBM layer and the printed circuit board (PCB) Cu pad are defined as the percentage of the total solder height in order to differentiate the critical element size around the UBM and the PCB Cu pad. Secondly, a crack path evaluation (CPE) method is developed for the gradual selection of the elements from the highest creep strain energy density (SED) value up to the predefined volume. The conventional solder crack path at the package interface or the diagonal solder crack path can be analyzed depending on the package technology because the critical solder elements are selected depending on the SED level and the failure path. The proposed SJR modeling method successfully demonstrates the diagonal solder crack path selection and further improves the SJR correlation of WLP and FOWLP.


2013 ◽  
Vol 2013 (1) ◽  
pp. 000039-000044
Author(s):  
Gary Gu ◽  
Jon Chadwick ◽  
Daniel Jin

Applications of Wafer Level Packages (WLPs) have shown tremendous growth in the rapid developing smartphones and other portable electronic devices. The technology trends lead to smaller chip size, low cost, and more integrated functions, but also face higher reliability requirements due to the reduced number of solder bumps as well as smaller bump size and height. New assembly technologies such as flexible phone board and conformal coating also brought up new thermo-mechanical reliability challenges. Based on 3D finite element modeling, the current studies focus on solder joint reliability of WLPs and compared between flex based and traditional rigid based WLP assemblies. Conformal coated and underfilled WLPs as well as some bump parameters are also studied. The parametric studies were carried out in ANSYS and all models were created by using APDL (ANSYS Parametric Design Language) scripts. Each simulation starts from stress free status set at solder reflow temperature and were subjected to thermal cyclic load between −40 and +125°C with ramp and dwell time. Creep strain was considered for solder alloys and kinematic plastic hardening was considered for other elastic-plastic materials. The solder fatigue life is estimated by using modified Coffin-Manson equation and was compared with available thermal cycling test data. The results show that underfill is still the most effective option and conformal coating can play an important role if the right material is selected. Bump parameters such as height, which have certain effects on the solder reliability on WLP-on-Rigid, have limited impact on WLP-on-Flex assembly.


Sign in / Sign up

Export Citation Format

Share Document