Reliability Qualification Report
SGA-9189
Products Qualified by Similarity
March, 2001
522 Almanor Ave., Sunnyvale, Ca 94086 Phone: (800) SMI-MMIC http://www.stanfordmicro.com
SGA-9289
Introduction
Stanford Microdevices' SGA-9189 is a high performance amplifier designed for operation
from DC to 3500 MHz. With optimal matching at 2 GHz, TOI = 39 dBm and P1dB=26
dBm. This RF device uses the latest Silicon Germanium Heterostructure Bipolar
Transistor (SiGe HBT) process. The process has a V
BCEO
=8V and an f
T
=25 GHz. The
SGA-9189 is cost-effective for applications requiring high linearity, even at moderate
biasing levels.
Fabrication Technology
These amplifiers are manufactured using a Silicon Germanium Heterojunction Bipolar
Transistor (HBT) technology. This patented self-aligned emitter, double poly HBT
process has been in production by our foundry since 1998. The process has been
successfully used for a wide range of RFIC products including GSM PAs, DECT front
end transceivers, LNAs & VCOs. This process offers comparable performance to GaAs
HBTs with the added advantages of mature and high producible Silicon wafer
processing.
Package Type
The SGA-9189 power amplifier is packaged in a plastic encapsulated SOT-89 package
that is assembled using a highly reproducible automated assembly process. The die is
mounted using an industry standard thermally and electrically conductive silver epoxy.
The die is mounted directly to the exposed paddle to provide a low thermal resistance
path for heat conduction out of the package.
SGA-9189 Reliability Qualification Report
Figure 1 : Photograph of SOT 89 Encapsulated Plastic Package
Qualification Methodology
The Stanford Microdevices Qualification Procedure consists of a series of tests designed to
stress various potential failure mechanisms. This testing is performed to ensure that
Stanford Microdevices products are robust against potential failure modes that could arise
from the various die and package failure mechanisms stressed. The qualification testing is
based on JEDEC test methods common to the semiconductor industry. For the tests
requiring full RF and DC testing to determine PASS/FAIL, the manufacturing test
specifications are used as the PASS/FAIL criteria.
Qualification By Similarity
A device can be qualified by similarity to previously qualified products provided that no
new potential failure modes/mechanisms are possible in the new design as compared
to the qualified design. The following product has been qualified by similarity to SGA-9189
SGA-9189 Reliability Qualification Report
Group Test Description
Test Standard
Quantity
In
Quantity
Out
Results
A0
Moisture Preconditioning / IR Reflow Simulation JEDEC 22-A113B
650
650
PASS
A1
Temp Cyle -65c to 150c 1000 cycles
JEDEC 22-A104A
120
120
PASS
A2
High Temp Operating Life 1000 hrs.
JEDEC 22-A108A
189
189
PASS
C
Autoclave 121c 15 PSI 96 hrs.
JEDEC 22-A102B
80
80
PASS
E
Solderbility Steam Age
JEDEC 22-B102C
44
44
PASS
G
High Temp Storage 1000 hrs T= 150C
JEDEC 22-A103A
83
83
PASS
Qualification Results
Table 1: Qualification Test Results for SGA-9189.
SGA-9289
SGA-9189 Reliability Qualification Report
Junction Temperature Determination
One key issue in performing the qualification testing is to accurately determine the
junction temperature of the device. Stanford Microdevices uses a 3
m spot size infrared
camera that allows a device to be measured at its normal operational parameters. The
3
m spot size allows for very good resolution compared to the heated area of the
transistor, which in this case is approximately 1-2
m. The results for the device, running
at maximum operational current of 180mA, a device voltage of 5.0V, and a base plate
lead temperature of 85.5
o
C.
Tlead = 85C
Id = 180mA
Vd = 5.00V
Tj = 125C
Figure 2: Infrared Thermal Image of SGA 9189, Vd = 5.0V, Id = 180mA, Tlead = 85C
SGA-9189 Reliability Qualification Report
Operational and Accelerated Life Testing
The purpose of the operational life test is to statistically show that the product operated at
125
o
C will be reliable. This is accomplished by operating several hundred devices for a
total test time of 1000 hours. The results for this test are expressed in device hours which
are calculated by multiplying the total number of devices passing by the number of hours
tested.
Operational Life Test Results
The results for SGA-9189 High Temperature Operating Life Test are as follows:
Group
Test Time
(hours)
Quantity
In
Quantity
Out
Device
Hours
A2
1000
189
189
189,000
Table 2: High Temperature Operational Life Test Results
Accelerated Life Test Results
The following data demonstrates the results from accelerated life tests performed on the
Stanford 4A SiGe HBT Process. The test was performed on 77 units running at a peak
junction temperature of 195
o
C. The test exceeded 10,000 hours (1.14 years) with no
failures. The FIT rate / MTTF calculation can be found below. The FIT rates were
generated assuming 1 failure. In reality, there were no failures, making this a very
conservative calculation.
Stanford Microdevices Process 4ASiGe HBT
FIT Rate / MTTF Calculation
SGA Series Devices
**Stanford Microdevices does not assume any liability arising from the use of this data.
*Ea = 0.7 eV
Parameters
Junction Temp C FIT Rate MTTF (hrs)
55
0.053
1.89E+10
125
4.136
2.42E+08
*The Ea of 0.7eV is conservative, 0.85eV is the activation energy for electomigration which is assumed to be the primary failure mechanism for
the SiGe process.
PDF 
ZIP