Instron Corporation

825 University Avenue h Norwood, MA 02062-2643

Tel: +1-781-828-2500 h Fax: +1-781-575-5750
www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints

E.H. Wong1,2, R. Rajoo1, Y.W. Mai2, S.K.W. Seah3, K.T. Tsai4, L.M. Yap4
1
Institute of Microelectronics, [email protected]
2
University of Sydney, Centre for Advanced Materials Technology
3
National University of Singapore, Dept of Mechanical & Production Engineering
4
Instron Singapore Pte Ltd

Abstract High strain rate tends to raise the fracture toughness of ductile
materials while lowers the fracture toughness of brittle materials
This paper presents a summary of the fundamental theories [13]. Thus, the reliability of interconnections at drop impact will
behind board level drop impact covering the dynamics of drop be governed more by the fracture toughness of the brittle
impact assembly, dynamics of PCB, as well as interconnection intermetalics than by its bulk creep strength, as in the case of
stress. This is followed by a comprehensive study of the fracture temperature cycling. It follows that the solder alloys and pad
characteristics of solder interconnections under high-speed finishing that are ideal for temperature cycling may not be ideal
impact using a newly developed Instron Micro Impactor which for drop impact. While JEDEC prescribed board level drop test
provides both the fracture strength as well as fracture energy of standard may be used to evaluate the impact performance of
impact. The test matrix consists of five solder alloys, four pad solders and pad finishing, it is a very costly test and the shock and
finishing, three thermal histories, and two solder mask designs noise accompanying the test can be extremely annoying. Simpler
forming a total of 120 combinations. The test has highlighted and cheaper tests have been explored for investigating the impact
weakness in NSMD design and caution on SnAgCu solder when performance of interconnection joint, with potential as a routine
used in drop impact application. quality control test in component manufacturing environment.
Component level impact shear tests have been reported using
1. Introduction split Hopkinson bar technique [14] as well as miniature Charpy
test [15]. The former has reported inferior impact fracture
Miniaturisation of portable electronic products has contributed to strength of SnAgCu solder compared to eutectic SnPb solder
the vulnerability of board level interconnections to drop impact. while the later has reported superior impact fracture energy of
This has prompted a new JEDEC board level drop impact test the SnAgCu solder.
standard [1]. Drop experiments at the product level have
provided useful understanding of the dynamic response of the This paper presents a summary of the fundamental theories
PCB as a function of product type and impact angle [2-4]. behind the board level drop impact test, follows by a
Numerical simulation [5] and experimental validation [6,7] has comprehensive study of the characteristics of solder interconnects
confirmed differential flexing between the PCB and the IC under high speed impact using a newly developed Instron Micro
packages as the primary driver for the failure of board level Impactor which provides both the fracture strength as well as
interconnections in drop impact. Computational models have fracture energy of impact.
been used extensively to model the dynamics of the PCB as well as
the interconnection stresses due to board level drop impact [8-10,
5]. Analytical solutions that provide physical insights to the
dynamics of PCB [11] and the interconnection stresses have also
been presented [12,5,6].

The difference is measurable TM

1
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

2. Fundamental Mechanics of Board For the initial conditions y(0)=Y(0)=0 and y(0)=Y(0)=-Vo,
Level Drop Impact as in the case of a board level drop impact, the solutions to the
equations of motion for the base can be shown to be
2.1 Dynamics of Drop Impact Test Assembly (4)
Fig 1a illustrates a typical board level drop test. In the typical
drop impact test, the base structure is prescribed with a half-sine
acceleration pulse of a defined amplitude and duration. In the and its acceleration expressed in terms of K, M, and the free-fall
experimental set-up, this prescribed acceleration pulse is height H is given by
achieved by manipulating the fall height and the stiffness of the
(5)
shock pad This is usually carried out haphazardly. An
understanding of the mechanics of this event will be useful for the
conduct of the experiment. This can be achieved by modeling the
drop assembly together with the shock pad simplistically as a It becomes clear that in carrying out the board level drop impact
combination of two spring-mass systems as illustrated in Fig 1b. experiment, one shall first select the shock pad with the
appropriate stiffness to produce the desired duration for base
PCB m
acceleration, follow by varying the free-fall height to produce the
k
Connector V desired amplitude of base acceleration.
Base y M

V
2.2 Dynamics of PCB
Shock Pad
Guide Rod
Y The base structure and the connectors are typically made of metal
K
whose longitudinal stiffness is much higher than the flexural
stiffness of the PCB. The half-sine acceleration pulse will
(a) (b)
therefore transmit to the PCB through the base and the
Figure 1: Typical board level drop impact (a) schematic (b) modelled
as 2 spring-mass systems
connectors with little distortion. As such, the analysis of a board
level drop impact may be reduced to modeling the PCB alone
where m and k are the representative mass and stiffness of the subjected to the half-sine acceleration pulse applied at its points
PCB respectively; M is the mass of the base structure, and K is the of mounting to the connectors.
stiffness of the shock pad. The equation of motion of the system
is given by eq(1) The dynamics of the PCB subjected to a half-sine acceleration
(1) pulse applied at its support may be analysed by modelling the
PCB as an Euler-Bernoulli beam simply supported at its ends as
depicted in Fig 2. The equation of motion is given by eq (6)
and the frequencies of the system can be shown to be given by
(2) (6)

L
y(x,t)
For a typical board level drop impact set up, Kk and eq(2) w(x,t)
(7)
is reduced to
t Y(t) t
(3)
y x

Figure 2: PCB modelled as an Euler-Bernoulli beam

The difference is measurable TM

2
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

where b2=Db/h and Db=EI is the flexural stiffness of the (10b)

PCB as a beam. The bending moment and acceleration along
the PCB as a function of time have been derived as a
superposition of flexing modes [11] and are presented in eq(7a)
and eq(7b) respectively.

(7a)

where p2=Dp/h and Dp=Eh3/12(1-2) is the flexural stiffness

of PCB as a plate; a and s are the length and width-to-length
(7b) ratio of the PCB respectively.

These equations have been implemented into a simple numerical

program and have been validated against FE analysis [11]. The
response spectrum of bending moment and acceleration at the
The function T(R,t) is given by eq(8). centre of PCB, modelled as a thin plate with s=1, is illustrated

(8)

where 1 is the fundamental flexing frequency of the PCB, and

R=n/ is the ratio of n-mode PCB flexing frequency to the Figure 3: Response spectrum of acceleration and bending moment at the
centre of PCB
frequency of the base acceleration.
in Fig 3.
Similar expression for PCB modelled as a thin rectangular plate 2.3 Interconnection stress due to board bending
simply supported at its edges is provided, without elaboration, in The bending moment in the PCB induces flexing of the PCB. But
eq(9 & 10) [11]. the IC component that is attached to the PCB does not acquire
(9)
the same curvature as the PCB due to the different boundary
condition. This differential flexing curvature induces stresses
on the solder interconnection. This is illustrated in Fig 4(a). The
(10a) stress in the interconnection may be modelled analytically using
beam on elastic foundation technique [12] whereby the
interconnections are modelled as a continuous elastic foundation
(Fig 4b) that transmits only axial force across the PCB and the IC
component. The bending moment, Mb, is applied along the
edges of the PCB. The uniform axial stress in a discrete
interconnection at distance from the middle of the IC
component can be shown to be
(11)

The difference is measurable TM

3
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

where is a characteristic parameter, ka is the PCB Strain trough due

PCB fibre strain/

to higher mode of PCB
axial stiffness of the elastic foundation per unit length per unit

connectivity
Electrical
flexing
width, De is the effective flexural stiffness of the PCB and IC
package, given by 1/De = 1/Dpcb + 1/Dpkg, Ar the ratio of the
Electrical open
package area to the total area of the interconnections, p the
Electrical closed
interconnection pitch, 3=sinh(x)cos(x), 2 4 6 8 10
4=cosh(x)sin(x) Time (ms)
Figure 5: One-to-one correspondence of PCB flexing and
electrical connectivity

Solder Interconnection 3. Impact Characterisation of Solder Joints

PCB
(a) The structural integrity of the interconnection under drop impact
IC Component depends not only on the magnitude of the driving force, but also

Mb Mb the resistance of the materials. Component level material testing
ka is highly desirable, especially for quality assurance, due to its low
(b) cost. However, the prevailing industry practice of solder ball
Lp shearing is inappropriate due to its slow speed which tends to
Figure 4: (a) Deformation of solder interconnection due to differential
induce failure only in the bulk solder while not evaluating the
flexing; (b) beam on elastic foundation model more critical intermetallic (IMC) region.

The inertial stress due to the accelerations of the IC package and The characterisation work by [14] using split Hopkinson bar
its solder interconnections can be computed simply using technique has provided evidence of ductile-bulk solder to brittle-
DAlemberts Principle. It can be readily shown that the IMC failure transition for pristine ternary Pb-free solder ball on
interconnection stress due to board bending is 2 orders of electroless Ni-P/Au (ENIG) pad finishing when the shear speed
magnitude higher than that due to acceleration [5, 11]. The role was raised from 1 mm/min to 3 m/s. At the shear speed of 3 m/s,
of flexing has been demonstrated in an experiment using a single the Sn3.8Ag0.7Cu solder alloy exhibited only a third of the
daisy-chained CSP mounted at the centre of a PCB subjected to fracture strength of Sn37Pb solder alloy. Similar ductile-bulk to
standard drop test [6,7]. The fibre strain measurement on the brittle-IMC failure was also reported by [15] using miniature
PCB shows the flexing of PCB consisting of fundamental mode Charpy when the shear speed was raised from 0.2 mm/s to 1 m/s.
superimposed with the next higher mode (Fig 5). The electrical At the shear speed of 1 m/s, Sn3Ag0.5Cu solder, on both
connectivity became open after the PCB has flexed more than Cu-organic surface preservative (OSP) and ENIG pad finishing,
30% of the amplitude corresponds to the fundamental bending was found to have a third higher fracture energy than the
mode. This was interrupted by momentary closure of the circuit Sn37Pb solder alloy.
that corresponds in time with the higher bending mode. The
one-to-one correspondence of electrical connectivity and the fibre A possible reason for the discrepancy is that the two works did not
strain endorses board bending as the dominant failure driver. measure the same characteristic one measuring fracture
strength while another measuring fracture energy. The following
works investigate the component level characteristics of solder
balls in details using Instron Micro Impactor which gives both
fracture strength and fracture energy.

The difference is measurable TM

4
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

3.1 Experiment
The sample consists of 400m solder ball on organic substrate.
The test matrix consists of 5 solder alloys, 4 pad finishes, 3
thermal histories, and 2 solder mask designs forming a total of
120 combinations. Each combination consists of at least 11
specimens, 5 of which were evaluated under static shearing and
the rest under impact shearing. The details are tabulated in
Table 1.
Solder alloys
Sn37Pb

Sn3.5Ag

Sn3.8Ag0.7Cu

Sn2.5Ag0.8Cu0.5Sb

Sn58Bi
Pad finishes
Electroless Ni-P/Au (ENIG) Figure 6: Static fracture strength of SMD (a) pristine (b) aged and TC

Hot air Sn37Pb solder levelling (HASL)

Organic surface preservative (OSP) Impact Fracture Strength & Fracture Modes

Immersion tin (Tin)
The impact fracture load of SMD_pristine system together with
Thermal
As-reflow (pristine)
histories
Thermal aging at 1250C for 125 hrs (Aged) fracture modes are depicted in Fig 7a. Observations: (i) The

Temperature cycle -400C to 1250C with 15 mins impact fracture strength of solder joint is generally higher than
dwell and 15 mins ramp for 500 cycle (TC) its static fracture strength. (ii) The impact fracture strength
Mask design
Solder mask defined with 500m pad and
300m pad opening (SMD)
exhibits higher dispersion in data. (iii) SnAg solder exhibits the

Non-solder mask defined with 300m pad and highest while BiSn solder the lowest impact fracture strength. (iv)
500m mask opening (NSMD) SnPb solder exhibits predominantly bulk failure failure usually
Shear speed
50m/s using Dage BT100 (Static) occurs just beneath the shear tool. Pb-free solder, on the other

600 m/s using Instron Micro Impactor
(Impact)
hand, exhibit predominantly a mixture of IMC and bulk failure
failure usually occurs along the bond interface. Note that in most
cases even though the failure surface appeared visually as brittle
3.2 Results IMC, this was at times accompanied by traces of tin picked up by
Static Fracture Strength EDS. The definition of IMC + bulk use here therefore
Fig 6a depicts the static shear fracture load for SMD design, encompasses failure in the IMC as well as between IMC and the
pristine condition, (SMD_pristine) for the full combination of solder interface.
solder alloys and pad finishes and Fig 6b depicts the percentage
variation of fracture load after thermal aging and temperature The effect of thermal treatment is depicted in Fig 7b. Note that
cycling (TC) respectively. The following observations may be while aging induces diverse response from various solder alloys,
made: (i) There is no significant differences among the solder TC tends to reduce the impact fracture strength of all solder
alloys except for SnBi solder which exhibits significantly higher alloys. Note again the huge reduction in impact fracture
fracture strength. (ii) Both aging and TC reduces the static strength for the specific combination of BiSn solder with HASL
fracture strength, though it varies significantly with respect to pad pad finishing.
finishing. (iii) The specific combination of SnBi solder with HASL
pad finish exhibits a huge reduction in fracture strength. This is
believed to be attributed to the formation of low-melting point
ternary SnBiPb eutectic phase [16] when Bi from solder ball
alloys with SnPb from pad finishing.

The difference is measurable TM

5
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

Figure 7 Impact fracture strength of SMD (a) pristine (b) aged and TC Figure 8 Impact fracture strength of NSMD (a) pristine (b) aged and TC

The fracture load of NSMD-pristine system is depicted in Fig 8. Force-Time characteristic

Observations: (i) The fracture strength is generally lower than The force-time history of the impact is fully registered by the
that of SMD_pristine similar observation was also made for Micro Impactor. The force-time characteristics of typical
static shear. (ii) The fracture strength fluctuates wildly with pad SMD_pristine system are depicted in Fig 9. With known velocity
finishing. (iii) The filure mode is predominantly pad-lift which of the impact head, the force-time characteristic can be readily
could explain the lower fracture strength. (iv) Interestingly, the transformed into force-displacement characteristic and the
fracture strength seems to follow, despite vaguely, the reverse energy consumed in fracturing the solder ball can be computed.
trend of the melting temperature of the solder alloys. (v) The It is observed that the SnPb solder while not giving absolutely
system has in general benefited from thermal treatment, but high fracture load tends to give high fracture displacement with
the response differs significantly between solder alloys and resultant high fracture energy.
pad finishing. 6
3 1
Force (N)

(a)
4 2
2 4
5
0
0 0.05 0.1 0.15 0.2 0.25 0.3
-2
8
2 (b)
Force (N)

6
4 1
4
3
2 5
0
-2 0 0.1 0.2 0.3
Time (ms)
1SnPb, 2SnAg, 3SnAgCu, 4SnAgCuSb, 5BiSn
Figure 9 Typical force-time characteristics for SMD_pristine system
(a) ENIG (b) OSP

The difference is measurable TM

6
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

Impact Fracture Energy

The fracture energy of SMD system is depicted in Fig 10.
Observations: (i) SnPb solder exhibits the highest fracture energy
follows by SnAg solder with BiSn solder exhibiting the least
fracture energy. (ii) The dispersion of data is much more than
that observed for fracture strength. (iii) Thermal treatment gave
rise to diverse effect for the system.

Figure 11 Summary SMD_solder (a) pristine (b) aged (c) TC

Fig 12 presents the average value (over all solder alloys) of static
fracture strength, impact fracture strength, and impact fracture
energy for individual pad finishing normalised by the
Figure 10 Impact fracture energy of SMD (a) pristine (b) aged and TC corresponding average value of ENIG finishing. The finishing are
arranged in orders of increasing impact fracture strength. In
general, OSP and Tin finishing seems to outperform ENIG and
Summary of Results HASL in terms of both impact fracture strength and impact
Fig 11 presents the average value (over all pad finishing) of static fracture energy.
fracture strength, impact fracture strength, and impact fracture
energy for individual solder normalised by the corresponding
average value of SnPb solder. The solders are arranged in orders
of increasing impact fracture strength. Observations: (i) SnAgCu
solder while having comparable impact fracture strength as SnPb
solder, has much lower impact fracture energy. (ii) SnAg solder
seems to have the best impact characteristics, especially after
thermal treatment. (iii) BiSn solder while having the highest
static fracture strength has the lowest impact fracture strength
and fracture energy.

Figure 12 Summary SMD_pad (a) pristine (b) aged (c) TC

The difference is measurable TM

7
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

Fig 13 presents the same for NSMD_pristine system.

Observations: (i) There is no clear winner among the solders
which is understandable in view of the fact that most of the
failure is in the core material - manifested as pad-lift. (ii) Tin
and HASL finishing seems to perform consistently poorer.

Figure 13: Summary NSMD_pristine system (a) solder (b) pad finishing

Fractographs
Figure 14: Fractographs of SMD_pristine SnPb_OSP subjected to (a) static
Fig 14 presents the typical fractographs of SMD_pristine system shear, (b) impact shear; SnAgCu_OSP subjected to (c) static shear, (d)
for SnPb_OSP and SnAgCu_OSP subjected to static and impact impact shear
shearing. Note (i) the ductile bulk solder failure in the
SnPb_OSP system in both static and impact shearing, and (ii)
the ductile bulk solder to brittle IMC transition in the
SnAgCu_OSP system with increased shearing speed. Fig 15
presents the typical fractographs of NSMD_pristine system for
the same solder_finishing and loadings. Bulk solder failure
were observed in both systems under static shearing while
core material failure were observed in both systems under
impact shearing.

Figure 15: Fractographs of NSMD_pristine SnPb_OSP subjected to

(a) static shear, (b) impact shear; SnAgCu_OSP subjected to (c) static shear,
(d) impact shear

The difference is measurable TM

8
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

4. Discussions SMD Vs NSMD

Compared to SMD system, NSMD system exhibits significantly
Fracture Strength Vs Shear Speed lower impact fracture strength and energy. This is partly
The fracture characteristic of the solder joint with respect to shear attributed to the absence of reinforcement from the solder mask
speed may be illustrated schematically in Fig 16. The fracture in the NSMD system and partly due to the stress concentration
strength of the ductile bulk solder starts low but increases with along the edge of the pad that led to cracking and cratering of
shear speed while that of the brittle IMC or IMC-solder bond (for underneath core material. Such failure is going to be more
SMD system) or the core material (for NSMD system) starts high prevalent with decreasing pad size and is potentially a limiting
but decreases with shear speed. The fracture strength of solder factor to decreasing interconnection pitch.
joint follows that of the bulk solder at low speed and jumps to
that of IMC or IMC-solder (or core materials for NSMD system) SnAgCu Vs SnPb
after the transition shear speed. It is clear that the fracture In view of the pad-lift failure nature of NSMD system,
strength of the IMC and IMC-solder bond depends both on the comparison between SnAgCu and SnPb solders were made only
types of solder as well as the pad finishing. In the case of fine using SMD system. While there is little difference between the
interconnection pitch when the solder volume is small, the impact fracture strength of both solders, the impact toughness of
fracture strength of the bulk solder may also depend to a small SnAgCu solder is much lower than that of SnPb solder. While it is
degree on the pad finishing through diffusion of pad metallurgy not clear at this point the relative role of fracture strength and
into the bulk solder. fracture energy in the drop impact reliability of solder
interconnection, it is reasonable to believe that the drop impact
Fracture strength of ductile bulk solder
reliability of SnAgCu solder is inferior to that of SnPb solder.
Fracture Strength

Among the solders evaluated, it appears that SnAg solder possess

Fracture strength of brittle IMC the best characteristics for drop impact application.
or IMC-solder bond

Fracture strength of core The Ideal Pad Finishing

material (NSMD system)
It appeared that OSP and Tin finishing compares favourably with
Shear speed HASL and ENIG finishing for SMD system. On the other hand,
Transition shear speed
OSP and ENIG appear to be favoured for NSMD system. Since
Figure 16: Summary SMD_solder (a) pristine (b) aged (c)
failure predominantly occurs in the core materials for the NSMD
system, the poorer performance of Tin and HASL finishing is
Fracture Strength & Fracture Energy believe to be attributed to the high processing temperature that
The bulk solder, IMC, IMC-solder bond, and the core material of has resulted in degradation of the core materials. This argument
the organic laminate shall have adequate fracture strength and may drawn support from the evidence that the impact fracture
impact toughness (expressed in terms of fracture energy) such strength of solder alloys appeared to decline with solders that
that the board level interconnection does not experience have increasing melting (hence processing) temperature
catastrophic failure in just a single drop. At the same time, high (Fig 8). However, in view of the large standard deviation of
fracture strength and fracture toughness are also believed to be the data, one needs to treat these deduction with care until
desirable for the board level interconnection to survive the further supporting evidence.
hundreds of flexing cycles experienced in the 30 repeated drops
specified in JEDEC Std JESD22-B111 [1].

The difference is measurable TM

9
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

Component level QC test 6. Acknowledgements

It is clear that the component level ball impact shear test can not
introduce a similar state of loading on the board level The authors would like to acknowledge the excellent work of
interconnection as in a JEDEC prescribed board level drop impact internship student Raj Kumar of Ngee Ann Polytechnic.
test. However being simpler and cheaper to perform, it has the
potential as a quality control tool for component manufacturing. Symbols
As a QC tool, it is only important that the impact test is able to a, L, Lp: Length of plate, beam, half length of the IC package
induce failure in the weakest element as observed in the JEDEC Ao, to: Amplitude, duration of half-sine base acceleration
test. For example, if the observed weakest element in the JEDEC Ar: Ratio of package area to the area of the interconnection
test for SnPb solder is the bulk solder while that in the Pb-free Db, Dp: Flexural stiffness of beam, plate
solder is the IMC bond interface, the component level impact Dpcb, Dpkg, De: Flexural stiffness of PCB, IC package, effectiveness
shear test shall be able to evaluate these weak elements stiffness of PCB and package modeled as beam
qualitatively by producing similar fracture mode. In the same E, h, I, , : Modulus, thickness, moment of inertia, density,
argument, the appropriate impact shear speed shall be the speed Poisson ratio of PCB/package
that can induce failure in the weakest element. H: Free-fall height
ka, k, K: Axial stiffness of continuous interconnection, linear
5. Conclusion stiffness of PCB, linear stiffness of shock pad
m, M: Concentrated mass of PCB, base
A summary of the fundamental mechanics behind a board level Mb, M(x,t), M(x,y,t): Bending moment in beam, plate
drop impact test has been presented with analytical equations p: Interconnection pitch
provided for the dynamics of the JEDEC board level drop test, the S()): Axial stress in the discrete interconnection
dynamics of the PCB, as well as the interconnection stress. Vo: Impact velocity
A comprehensive study of the impact characteristics of the solder w(x,t), w(x,y,t): Deflection of beam, plate with respect base
ball at component level has been performed using the newly y(x,t), z(x,y,t): Displacement of beam, plate with respect to an
developed Micro Impactor that provides both the fracture absolute reference
strength and fracture energy. The following conclusions has Y(t), Z(t): Displacement of base
been drawn: n: The nth angular modal frequency of vibrating body
h SMD design is stronger than NSMD design.
:: Angular frequency of input base acceleration
h The impact characteristic of Sn3.8Ag0.7Cu solder is inferior
while the Sn3.5Ag solder is superior to that of Sn37Pb solder. References
h OSP and Tin pad finishing performed well for SMD
system while OSP and ENIG finishing performed well for [1] JEDEC Standard JESD22-B111, Board level drop test
NSMD system. method of components for handheld electronic products.
h Both impact strength and impact energy are important [2] S.K.W. Seah, et al., "Mechanical response of PCBs in
to prevent catastrophic and cyclic failure of board level portable electronic products during drop impact," Proc 4th
interconnection. EPTC, Singapore, pp. 120-125, 2002.
h The drop impact reliability of the solder joint shall be [3] C.T. Lim, et al., Drop impact survey of portable electronic
evaluated at the speed that induces the appropriate products, Proc 53rd ECTC, 2003.
failure mode. [4] E.H. Wong, et al., Tackling board level drop impact
reliability of electronic interconnections, Interpack 2003.

The difference is measurable TM

10
 www.instron.com

Drop Impact: Fundamentals and Impact Characterisation of Solder Joints (cont.)

[5] E.H. Wong, et al., Drop impact test Mechanics & physics
of failure, Proc 4th EPTC, pp. 327-333, 2002.
[6] E.H. Wong, et al., Fundamentals of drop impact,
presented in JEDEC working group meeting, Dec 2003.
[7] S.K.W. Seah, at al., Experiments and failure drivers in drop
impact of portables, JEDEX, San Jose, 2004
[8] T. Sogo, et al., Estimation of fall impact strength for BGA
solder joints, Proc. ICEP, pp. 369-373, 2001.
[9] L. Zhu, Submodeling technique for BGA reliability analysis
of CSP packaging subjected to an impact loading, Proc.
IPACK, 2001.
[10] J. Wang, et al., Modeling solder joint reliability of BGA
packages subject to drop impact loading using
submodelling, Proc. Abaqus Conference, 2002.
[11] E.H. Wong, Dynamics of board level drop impact, ASME
Trans. JEP, to be published in Jul issue 2005.
[12] E. Suhir, On a paradoxical phenomenon related to beams
on elastic foundation: Could external compliant leads
reduce the strength of a surface-mounted device?, J.
Applied Mechanics, 55:818, 1988.
[13] M.A. Meyers, Dynamic behavior of materials,
John Willey, 1994.
[14] D.M. Williamson, et al., Spall, quasi-static and high strain
rate shear strength data for electronic solder materials,
Internal report Cavendish Laboratory No. SP 1113,
Oct 2002.
[15] M. Date, et al., Impact reliability of solder joints, Proc 54
ECTC, pp. 668-674.
[16] J. Bath, et al., Research update: lead-free solder
alternatives, Circuit Assembly, May 2000 issue.

The difference is measurable TM

11