Plastic Package - Altera Corporation

Altera Corporation 1

Plastic Package

Reliability & Testing

June 1999, ver. 1 Application Note 113

A-AN-113-01

Introduction

Designers expect Altera

devices to perform in a variety of environments

and conditions. Reliable devices decrease design and maintenance costs

and increase product life cycles. Altera submits its devices to a series of

strenuous tests to ensure that they provide maximum value to end users.

Device reliability is generally defined as the probability that a device will

perform its intended function under specified operating conditions

throughout its life. As a result, industry-wide reliability guidelines were

established.

Altera devices exceed the reliability requirements established by the

Electronic Industries Association (EIA) and Joint Electron Device

Engineering Council (JEDEC). JEDEC qualification tests ensure that

Altera devices meet or exceed these reliability standards. Tests are

performed under extreme conditions and the failure rate can be

extrapolated to typical conditions using equations that describe the

acceleration of failure mechanisms. This application note describes some

of the tests used for device reliability testing, including:

■

Stress-test-driven qualifications

■

Moisture/reflow sensitivity

■

Temperature cycling

■

Humidity bias testing

■

HAST testing

■

Autoclave testing

For more information on Altera device reliability, see the

Altera Device

Reliability Report

, which is available from Altera literature services.

EIA/JEDEC

Standards

EIA/JEDEC standards identify testing requirements that range from

general to specific. For example, all plastic package testing must follow

general guidelines specified by JESD 47, but only certain types of device

packages may be required to undergo HAST testing. The pyramid in

Figure 1 illustrates the hierarchy for EIA/JEDEC testing procedures and

serves as the model for Altera testing requirements.

2 Altera Corporation

AN 113: Plastic Package Reliability & Testing

Figure 1. EIA/JEDEC Testing Hierarchy

As new technologies are developed, Altera engineers rely on the

EIA/JEDEC testing hierarchy to qualify new semiconductor

architectures. New architectures may include smaller process geometries,

an increased number of die layers, increased gate and I/O pin counts,

decreased voltage levels, and changes in package material composition.

These design changes affect the reliability of a device. As a result, the

device may require a different set of tests as technologies advance.

The Stress-Test-Driven Qualification of Integrated Circuits (JEDEC

Std. 47) guideline determines which tests a new design must undergo and

helps product engineers identify and correct flaws that may arise. The

guideline also identifies sample size requirements and testing

qualifications for new devices to help expose process flaws.

For more information on EIA/JEDEC standards, visit the JEDEC web site

http://www.jedec.org

Moisture/

Reﬂow

Sensitivity

Altera uses moisture/reflow sensitivity testing (J-STD-020A) guidelines

to inform customers of device sensitivity to moisture-induced stress. After

testing, devices are assigned sensitivity levels. These classification levels

are useful in determining the proper storage and handling of Altera

devices to prevent thermal and mechanical damage during solder reflow

or repair. Moisture sensitivity levels determine the length of time a device

can be exposed to humidity. See Table 1.

JESD 47

Stress-Test-Driven

Qualification

of Integrated Circuits

J-STD-020A

Moisture/Reflow Sensitivity

Classification for Non-Hermetic

Solid State Surface Mount Devices

JESD22-A110-A

HAST Testing

JESD22-A104-A

Temperature

Cycling

JESD22-A102-B

Accelerated Moisture

Resistance —

Unbiased Autoclave

JESD22A-101-B

Steady State

Temperature

Humidity Bias

Life Test

General

Guidelines

Specific Testing

Requirements

Altera Corporation 3

AN 113: Plastic Package Reliability & Testing

Note:

(1)Condition percentages denote relative humidity levels.

The device classification applies to all surface mount packages: ball-grid

array (BGA), small-outline integrated circuit (SOIC), plastic J-lead chip

carrier (PLCC), plastic thin quad flat pack (TQFP), plastic quad flat pack

(PQFP), and power quad flat pack (RQFP) device packages. Because of

their composition, these device packages could be damaged during solder

reflow.

The vapor pressure of moisture trapped inside of a plastic package

increases when the package is exposed to high solder reflow

temperatures. This pressure can lead to internal delamination, internal

cracks, bond damage, wire necking, bond lifting, die lifting, thin film

cracking, or cratering beneath the bonds. In extreme cases, the package

emits an audible crack, also known as the “popcorn” effect.

To minimize the chances of device damage, Altera recommends using a

100

convection reflow system that is capable of maintaining the reflow

profiles required by the J-STD-020A standard.

For more information on reflow soldering, see

Application Note 81 (Reflow

Soldering Guidelines for Surface-Mount Devices

Small, thin devices can reach body temperatures greater than 220 ˚C when

reflow-soldered to boards profiled for larger devices. To compensate for

this difference, some small packages have been reclassified to withstand

235 ˚C. Table 2 defines the transition thickness/volume of packages that

can reach 235 ˚C when reflow soldered to boards with larger devices.

Table 1. Moisture Sensitivity Levels

Level Floor Life Soaking Requirements

Time Conditions

(1)

Time (Hours) Conditions

(1)

1 Unlimited

≤

30 ˚C/85

168 85 ˚C/85

2 1 Year

≤

30 ˚C/60

168 85 ˚C/60

2a 4 Weeks

≤

30 ˚C/60

672 30 ˚C/60

3 168 Hours

≤

30 ˚C/60

192 30 ˚C/60

4 72 Hours

≤

30 ˚C/60

96 30 ˚C/60

5 48 Hours

≤

30 ˚C/60

72 30 ˚C/60

5a 24 Hours

≤

30 ˚C/60

48 30 ˚C/60

6 Time On Label

≤

30 ˚C/60

Time On Label 30 ˚C/60

4 Altera Corporation

AN 113: Plastic Package Reliability & Testing

Note:

(1) VPR phase will not exceed 219

For more information on moisture/reflow sensitivity classification for

plastic packages, refer to

JEDEC Standard J-STD-020A

Package reliability can also be affected by the rate at which a particular

device reaches its testing temperature. Table 3 lists classification reflow

profiles for two types of reflow convection methods: full convection and

infra-red convection.

Note:

(1) Altera recommends using the full convection method for production.

(2) See Table 2.

Temperature

Cycling

Temperature cycling (JESD22-A104-A) is performed to determine the

resistance of Altera devices to high and low temperature extremes. This

environmental stress test is designed to simulate the extensive changes in

temperature to which devices and packages may be exposed. To pass the

test, devices must not show signs of damage such as cracking, chipping,

or breaking. Table 4 shows the classifications for Altera devices.

Table 2. Small Device Reflow Conditions

Conditions Package Thickness

≥

2.5 mm & All

BGA Packages

Package Thickness

< 2.5 mm & Package

Volume

≥

350 mm

Package Thickness

< 2.5 mm & Package

Volume < 350 mm

Convection Temp. 220 +5/-0 ˚C Convection 220 +5/-0 ˚C Convection 235 +5/-0 ˚C

VPR Peak Temp. 215-219 ˚C 215-219 ˚C N.A.

(1)

IR/Convection Temp. 220 +5/-0 ˚C 220 +5/-0 ˚C 235 +5/-0 ˚C

Package Type All BGA, FineLine BGA

packages All other packages T44, Q44, T100, T32

packages

Table 3. Classification Reflow Profiles

Condition Full or Infra-Red Convection

(1)

Average ramp-up rate (183 ˚C to peak) 1 - 3 ˚C/second

Preheat temperature 125 ˚C (

25 ˚C) 120 seconds maximum

Temperature maintained above 183 ˚C 60 to 150 seconds

Time within 5 ˚C of actual peak

temperature 10 to 20 seconds

Peak temperature range

(2)

Ramp-down rate 6 ˚C/second maximum

Time 25 ˚C to peak temperature 6 minutes maximum

Altera Corporation 5

AN 113: Plastic Package Reliability & Testing

Temperature cycling may also expose weaknesses in packages that are

composed of different materials. If device packages are composed of

different materials, expansion and contraction of the package materials

can occur at different rates when exposed to temperature extremes.

Figure 2 shows an example of the differences in the thermal coefficients of

expansion between two materials with identical lengths.

Figure 2. The Effects of Thermal Coefﬁcient of Expansion on Two Materials

Because the materials are in contact with each other, the actual expansion

of the two materials is a composite of the two. Both materials expand to

the same length when placed together, which places extra stress on

material 1. The materials act similarly during compression. As a result,

shear stresses can occur at the joint between the two materials, which

results in the passivation of metal layers.

Table 4. Temperature Cycling Classifications

Classification Low Temperature (˚C) High Temperature (˚C)

A –55 (+0, –10) +85 (+0, –10)

B –55 (+0, –10) +125 (+0, –10)

C –65 (+0, –10) +150 (+0, –10)

D –65 (+0, –10) +200 (+0, –10)

F –65 (+0, –10) +175 (+0, –10)

G –40 (+0, –10) +125 (+0, –10)

H –55 (+0, –10) +150 (+0, –10)

Material 1

Material 2

Material 1

Material 2

Expansion Caused by

Temperature Increase

Individual

Material Expansion

Effective Expansion of

Bonded Materials

6 Altera Corporation

AN 113: Plastic Package Reliability & Testing

When devices are repeatedly cycled between hot and cold temperature

extremes, the shear stress of the materials change from compressive to

tensile and back. If the temperature extremes are great enough, cracking

and/or shear displacement of the materials could occur, resulting in

displacement of the metal traces below the packaging layers. See Figure 3.

Figure 3. Passivation Cracking & Layer Displacement

During this test, shear stresses are greatest at the corners of the die and are

generally greater for larger die. Cracks in packaging can run down into

the device, eventually breaking or severing metal layers or polysilicon

below the passivation layer. Displaced metal traces cause leaking current

or short circuits when metal lines are moved against other lines or power

buses. To avoid these failures, Altera extensively tests devices following

the procedure of test JESD22-A104-A.

To avoid these failures, Altera extensively tests devices following the

procedures of test JESD22-A104-A. By following the device classifications

shown in Table 4 on page 5, you can mimimize the chance of this type of

failure.

Humidity Bias

Testing

The Steady-State Temperature Humidity Bias Life Test (JESD22A-101-B)

evaluates the reliability of Altera devices in humid environments. The test

uses temperature, humidity, and electrical bias to accelerate the

penetration of moisture through device packages. The test is performed at

85 ˚C and 85

relative humidity with applied bias. This test can uncover

flaws in device package material composition or package assembly.

The test is carried out in a temperature-humidity test chamber capable of

maintaining continuous temperature and relative humidity levels. It must

also provide electrical connections to the devices under test in a specified

biasing configuration. The test is performed under extreme conditions.

The failure rate can be exrapolated to typical conditions using equations

that describe acceleration of failure mechanisms.

Altera uses two types of electrical biasing for testing: continuous bias and

cycled bias.

Passivation

Material 2

Material 1

Altera Corporation 7

AN 113: Plastic Package Reliability & Testing

In continuous electrical bias, DC bias is applied continuously. Continuous

bias is more severe than cycled bias when die temperatures are

≤

10 ˚C

higher than the chamber’s ambient temperature or if die temperatures are

not known when the power dissipation of the device is less than 200 mW.

If the power dissipation of the device exceeds 200 mW, you should

recalculate the die temperature (power

thermal resistance). If this

number is more than 5 ˚C than the test specification, then you should use

cycled bias. If the die temperature exceeds the chamber’s ambient

temperature by more than 5 ˚C, include the temperature difference in the

test results because acceleration of failure mechanisms will be affected.

In cycled electrical bias, DC voltage applied to the devices is interrupted

periodically with frequency and duty cycles. If biasing causes the

temperature to rise 10 ˚C above the chamber’s ambient temperature, the

cycled bias is more severe than continuous bias when optimized for a

specific device type. Heating caused by power dissipation drives

moisture away from the die and slows moisture-related failure. Cycled

bias lets moisture collect on the die during the off periods. Cycling

electrical bias with one hour on and one hour off is optimal for most

devices.

HAST Testing

Highly Accelerated Stress Testing (HAST) (JESD22-A110-A) evaluates the

reliability of Altera devices in humid environments. Like humidity bias

testing, HAST testing uses a combination of electrical bias, high

temperature, and high humidity. However, HAST testing also includes

high pressure to accelerate corrosion-type failures. The test is performed

at 130 ˚C and 85

relative humidity with applied bias. HAST testing can

uncover flaws in packaging material, seals, and joints between the

package material and pins.

The electrical bias patterns used for HAST testing are similar to the ones

used for humidity bias testing. The bias is used for powering the devices

inside a HAST chamber. Wherever possible, an alternating sequence of

and ground is used so that signal pins that are biased with one

voltage are placed closest to signal pins that are biased with the opposite

voltage. For HAST testing, V

is set to the minimal condition.

Specially designed boards are used in HAST chambers. Ordinary boards

cannot be used because of severe HAST chamber conditions. For testing,

Altera uses special polyimide printed circuit boards (PCBs) with buried

traces.

8 Altera Corporation

AN 113: Plastic Package Reliability & Testing

Autoclave Testing

Autoclave (JESD22-A102-B) testing uses a temperature/humidity

environment with no applied electrical bias to accelerate failure

caused by metallization corrosion. This test subjects devices to

saturated steam at 121 ˚C and 15 PSIG pressure. This pressure is

“gauge” pressure, and is twice the atmospheric pressure of 15 PSI.

The pressure can also be represented as two atmospheres. Autoclave

testing tests the resistance of passivation layers on the die and

exposes flaws in package design.

High pressure in autoclave testing drives water vapor through

device packages until it reaches the die surface. Bond pads and

poorly passivated metals can corrode. Several other failures can also

occur. For example, non-volatile memory—using charge storage on

floating gates (flash and EEPROM devices)— could be subject to

charge loss from the floating gates. This charge loss is caused by

moisture penetrating through passivation layers. If the moisture

reaches the floating gate, it can quickly leak the stored charge off the

floating gate. Altera uses the results of autoclave testing to minimize

device failure.

Other failure mechanisms that can be aggravated are external

package damage, shorts, or leakage paths. Improper

rinsing/cleaning following lead finish can result in contamination

migrating between the leads. This contamination causes leakage

paths. Tin and lead migration, via dendritic growth from one lead to

another, is another possible failure mechanism for autoclave testing.

Conclusion

To ensure a consistent level of quality, Altera subjects its plastic

packaged devices to a series of EIA/JEDEC tests. These tests are

designed to accelerate failure in devices, uncovering any process or

design flaws in Altera devices. Altera uses the results of these tests to

improve packages and minimize field failures. The information is

useful to help designers match their stress and environment

requirements and to establish guidelines for handling, soldering,

classifying, and storing Altera devices.

Altera is a trademark of Altera Corporation in the United States and other countries. Altera acknowledges the

trademarks of other organizations for their respective products or services mentioned in this document. Altera

products are protected under numerous U.S. and foreign patents and pending applications, maskwork rights,

and copyrights. Altera warrants performance of its semiconductor products to current specifications in

accordance with Altera’s standard warranty, but reserves the right to make changes to any products and

services at any time without notice. Altera assumes no responsibility or liability arising out

of the application or use of any information, product, or service described herein except as

expressly agreed to in writing by Altera Corporation. Altera customers are advised to

obtain the latest version of device specifications before relying on any published

information and before placing orders for products or services.

101 Innovation Drive

San Jose, CA 95134

(408) 544-7000

http://www.altera.com

Applications Hotline:

(800) 800-EPLD

Customer Marketing:

(408) 544-7104

Literature Services:

(888) 3-ALTERA

lit_req@altera.com

9 Altera Corporation

Printed on Recycled Paper.

AN 113: Plastic Package Reliability & Testing