IPC-7095C-2013.pdf - 第148页

Hence, for better solder joint yields and solder joint reli- ability , the backward compatibility reﬂow soldering proﬁle depicted in Figure 8-23 should be used. During this reﬂow proﬁle, the SnAgCu solder ball also melts…

with the help of special microscopes such as the endo-

scope, peripheral solder joints can be viewed. SnAgCu sol-

der joint microstructures are multiphase microstructures

and the surface of the solder joints appears rough. Figure

8-22 shows a typical SnAgCu BGA solder joint. This is

much different from typical SnPb BGA solder joints which

usually have a shiny surface. Ball

8.5.5.6 Transition Lead-Free Technologies The transi-

tion from a total tin/lead soldering system to a total lead-

free soldering system is not going to happen overnight.

There may be an interim phase where tin/lead and lead-free

solders will co-exist on board assemblies. This transition

phase entails that the impact of lead in the SAC solder be

evaluated for impact on solder joint yields and reliability.

The likely lead-free board assemblies during this transition

phase are listed in Table 8-5.

The ﬁrst possible lead-free board assembly listed above is

‘‘forward compatibility.’’ The board assembly soldering

process for forward compatibility assemblies has been con-

verted to lead-free technology with a change in the solder

paste composition and the reﬂow soldering proﬁle to match

this change. However, some components, such as BGAs

that are soldered on the board, will still have tin/lead sol-

der due to the component supplier’s lead-free roadmap

having a conversion date later than the board assembler’s

conversion date. This results in the tin/lead solder joints of

the BGA-type components being ‘‘contaminated’’ by the

lead replacement metals in the lead-free solder paste.

The second possible lead-free board assembly listed above

is ‘‘backward compatibility.’’ Backward compatibility sce-

narios arise when component suppliers introduce the lead-

free components, but not all board assemblers that use

these components have converted their board assembly

lines to lead-free technology. These assemblers will still be

soldering the lead-free components with eutectic tin/lead

solder paste using tin/lead reﬂow soldering proﬁles. A tin/

lead component would obviously be preferable in this case,

but the component suppliers may, due to economic reasons,

not want to carry two component line items, one tin/lead

and one lead free, for the same device. The solder joints

formed with this combination of tin/lead solder paste and

lead-free solder ball will have a mixed composition. When

soldering a Ball Grid Array package with SnAgCu lead-

free solder balls using tin/lead solder paste, two different

scenarios arise based on the reﬂow proﬁle used. The two

reﬂow proﬁles compared are shown in Figure 8-23 with the

total lead-free reﬂow proﬁle also shown for comparison.

The tin/lead reﬂow proﬁle, which is illustrative of the pro-

ﬁle used today for tin/lead assemblies, does not exceed the

melting point of the BGA’s SnAgCu solder ball. When a

reﬂow proﬁle is unable to make lead-free solder ball melt

or dissolve, the solder joint yield and/or solder joint fatigue

life may suffer. The tin/lead solder paste deposited on the

lands of the solder balls melts but the SnAgCu solder balls

are still not molten. In such a material combination, the

lead may likely diffuse through the grain boundaries of the

solder ball. How high the lead from the tin/lead solder dif-

fuses up the SnAgCu solder ball will depend on how high

the reﬂow temperature gets and for how long the tin/lead

solder is molten. As shown in Figure 8-24, which depicts a

micrograph of a cross-section of a SnAgCu solder ball of a

BGA package soldered to a board using a reﬂow proﬁle

which is unable to melt or dissolve SnAgCu solder ball, the

resulting solder joint microstructure is inhomogeneous.

This deleteriously impacts the solder joint reliability.

Yield impact on such solder joint is also deleterious due to

two reasons. One is due to the poor self-alignment of the

BGA during reﬂow soldering because the solder ball does

not become molten. This creates a potential for open joints

when the component is misaligned to some extent during

or after the Placement process step. Secondly, the lack of

‘‘ball collapse’’ may cause open solder joints from a lack of

contact between the solder paste deposit and the solder

ball.

Figure 8-22 Endoscope Photo of a SnAgCu BGA Solder

Ball

Table 8-5 Types of Lead-Free Assemblies Possible

Deﬁnition

Component

Termination/

BGA Ball Solder Paste

Board

Surface

Finish

Forward

compatibility

contains lead lead-free

leaded or

lead-free

Backward

compatibility

lead-free 63Sn37Pb

leaded or

lead-free

Total

lead-free

lead-free lead-free lead-free

January 2013 IPC-7095C

133

Provided by IHS under license with IPC

Not for Resale

No reproduction or networking permitted without license from IHS

--`,,```,,,,````-`-`,,`,,`,`,,`---

Hence, for better solder joint yields and solder joint reli-

ability, the backward compatibility reﬂow soldering proﬁle

depicted in Figure 8-23 should be used. During this reﬂow

proﬁle, the SnAgCu solder ball also melts, and lead from

the molten tin/lead solder paste mixes thoroughly with

molten SnAgCu solder ball and generates a homogeneous

structure of a lead-rich phase in the tin matrix. Such a

microstructure is shown in Figure 8-25.

Moreover, since the SnAgCu solder ball melts and col-

lapses, the self-aligning process and coplanarity reduction

also occur, thereby enhancing solder joint yields of the

BGA.

8.5.5.7 Ball Replacement Due to limited availability of

components with SnPb balls, some functional designs may

require ball replacement of the BGA component to allow

common processing with tin lead production assemblies.

This practice, while commonly employed, has drawbacks

due to the additional thermal cycles required for the ball

removal and subsequent replacement. Additionally, the

resulting interface alloy at the component to ball attach-

ment site may not achieve the desired or expected eutectic

properties. The process, however, does allow the use of

limited component types to comply with tin lead use strat-

egies for performance, contract mandates, and/or compat-

ibility with legacy or production designs that are suscep-

tible to the additional temperature excursion required for

lead free production processes. The sphere replacement

IPC-7095c-8-23

Figure 8-23 Comparison of Reﬂow Soldering Proﬁles for Tin/Lead, Backward Compatibility and Total Lead-Free Board

Assemblies

300

250

200

150

100

150 200 250

300

350 400

Tmelting for SnAgCu

Tmelting for SnPb

SnAgCu Balls do NOT Melt

Not good for backward compatibility

SnAgCu BAlls Melt

Will be OK for backward compatibility

Typical Lead Free Solder (SnAgCu) Profile

Temperature, Deg C

Time, Seconds

Figure 8-24 Micrograph of a cross-section of a BGA

SnAgCu solder ball, assembled onto a board with tin/lead

solder paste using the standard tin/lead reﬂow soldering

proﬁle. The SnAgCu solder ball does not melt; black/grey

interconnecting ﬁngers are lead-rich grain boundaries; rod

shape particles are Ag3Sn IMCs; grey particles are Cu6Sn5

IMCs.

Figure 8-25 Micrograph of a cross-section of a BGA

SnAgCu solder ball, assembled onto a board with tin/lead

solder paste using a backward compatibility reﬂows sol-

dering proﬁle. The SnAgCu solder ball has melted.

IPC-7095C January 2013

134

Provided by IHS under license with IPC

Not for Resale

No reproduction or networking permitted without license from IHS

--`,,```,,,,````-`-`,,`,,`,`,,`---

process avoids paste alloy, PCB surface ﬁnish, and compo-

nent ﬁnish compatibility issues that can be very complex to

balance for successful production results.

Reballing of BGA components involves removal of the

original balls with a localized heat source such as a

vacuum desoldering tool, blade tip or solder wick. The

component surface is then cleaned and prepared for attach-

ment of new spheres that are located with a ﬁxture that

aligns to the original pattern of the component. The spheres

are then attached to the component using a typical rework

heat source such as a convection reﬂow oven, or the ﬁxture

with component and replacement spheres can be placed in

a reﬂow oven. As an alternate when many parts are to be

reballed, one can use automated laser systems that will pick

and place balls from a tray and then reﬂow them in place

on the BGA substrate using the laser beam.

In either case, the heat source must be controlled using a

thermal proﬁle appropriate for the component and new ball

alloy. Once the balls are attached, the BGA package is

removed from the ﬁxture and the balls can be examined

and cleaned in preparation for installation and soldering

with the remainder of the assembly components. As with

all BGA component processes, moisture and ESD sensitiv-

ity must be considered. BGA balls are commonly available

in a wide range of solder alloys and dimension but they

must be selected for compatibility with other process ele-

ments such as the board surface plating and solder paste. It

is also critical to maintain the component design spacing

and pitch through proper ball size and placement.

Detailed procedures for BGA Reballing are available in

IPC-7711/21B, ‘‘Rework, Modiﬁcation and Repair of Elec-

tronic Assemblies.’’

8.6 Design for Reliability (DfR) Process As a general

recommendation, the Design for Reliability procedures

detailed in IPC-D-279 should be followed.

Appropriate DfR measures to improve reliability for grid

array components take one of two forms, which are best

employed in combination for improved reliability margins.

These measures are:

1. CTE-tailoring to reduce the global expansion mismatch.

2. Increasing attachment compliancy to accommodate the

global expansion mismatch by increasing the solder

joint height (stand-off).

3. Further, a DfR procedure aiming at high-reliability

could also include:

4. Eliminate the effect of the global expansion mismatch

by mechanically coupling the component and the sub-

strate with an appropriate underﬁll.

5. Choosing a soft die attach to reduce the impact of the

low CTE of the die (2.7 to 2.8 ppm/°C) on both the

global and local thermal expansion mismatch.

CTE-tailoring involves choosing the materials or material

combinations of the multilayer board and/or the compo-

nents to achieve an optimum CTE. An optimum CTE for

active components dissipating power is 1-3 ppm/°C

(depending on the power dissipated) with the multilayer

board having the larger CTE, and 0 ppm/°C for passive

components. Of course, since an assembly has a multitude

of components, full CTE optimization cannot be achieved

for all components—it needs to be for the components with

the largest threat to reliability. For military applications

with the requirement of hermetic and thus ceramic compo-

nents, CTE-tailoring has meant the CTE-constraining of

the multilayer boards with such materials as Kevlar™ and

graphite ﬁbers, or copper-Invar-copper and copper-

molybdenum-copper planes. Such solutions are too expen-

sive for most commercial applications for which glass-

epoxy or glass-polyimide are the materials of choice for the

multilayer boards. Thus, CTE-tailoring has to take the form

of avoiding larger size components that are either ceramic

(CGAs, MCMs), plastic with Alloy 42 lead frames

(TSOPs, SOTs, or plastic with rigid bonded silicon die

(PBGAs).

Increasing attachment compliancy for leadless solder

attachments means increasing the solder joint height (C4,

C5, shimming, gluing, 10Sn90Pb balls, 10Sn90Pb col-

umns) or switching to a leaded attachment technology. For

leaded attachments, increasing lead compliancy can mean

changing component suppliers to those having lead geom-

etries promoting higher lead compliancy or switching to

ﬁne-pitch technology.

The DfR process needs to emphasize a physics-of-failure

perspective without neglecting the statistical distribution of

failures. The process might involve the following steps:

• Identify Reliability Requirements – expected design life

and acceptable cumulative failure probability at the end

of this design life.

• Identify Loading Conditions – use environments (e.g.,

IPC-SM-785) and thermal gradients due to power dissipa-

tion, which may vary and produce large numbers of mini-

cycles (Energy Star).

• Identify/Select Assembly Architecture – part and substrate

selections, material properties (e.g., CTE), and attachment

geometry.

• Assess Reliability – determine reliability potential of the

designed assembly and compare to the reliability require-

ments using the approach shown here, a ‘‘Figure of

Merit’’ approach, or some other suitable technique; this

process may be iterative.

• Balance Performance, Cost and Reliability Requirements.

8.7 Validation and Qualification Tests Performance test

methods and qualiﬁcation requirements are speciﬁed in

IPC-9701; IPC-9701A includes guidelines for lead-free

January 2013 IPC-7095C

135

Provided by IHS under license with IPC

Not for Resale

No reproduction or networking permitted without license from IHS

--`,,```,,,,````-`-`,,`,,`,`,,`---