Download the PDF file
MCM/Hybrid
Performance
Defining performance for today’s military systems is
extremely critical. Historically, the military overspecified performance of
MCMs/Hybrids used in military applications. In addition, the military mandated
the use of military-qualified parts.
Today, non-military and military parts are often used together. This presents a potential risk to the
application and makes defining satisfactory operation more difficult than in the
past. To assist in developing performance criteria for MCMs/Hybrids, performance
specification MIL-PRF-38534 should be used.
MCM/Hybrid
Time
MCM/Hybrids, qualified to MIL-PRF-38534 and listed in
QML-38534, are designed to last up to 20 years. On the other hand, non-military
parts are designed to last from 12 months up to 5 years. The part’s package has
a significant effect on the part’s life. Packages that dissipate heat
effectively and/or protect the internal circuitry from the environment last
longer. The environmental categories detailed in the Scope section of this
document/database establish timed-based requirements for part usage.
MCM/Hybrid
Operating Conditions
With the inclusion of non-military parts in military
applications, operating conditions have become more critical than in the past.
The environmental categories detailed in the Scope section of this
document/database establish operating condition requirements for product
usage.
Reliability
Typical failure rates for deposited circuit elements
are shown in Table 1.
Because Hybrids and MCMs typically are significantly
higher in unit cost than standard components, an alternative to lengthy testing
of many devices is to perform a product assessment. RELTECH Assessment
Guidelines are a viable alternative to the expensive testing used in the past,
as it requires only a relatively few devices, with testing tailored to the
expected application, and for the specific product technology.
When passive structures are used in microwave
applications, the primary reliability issues associated with transmission lines
are electromigration and adhesion of the metal to the substrate. Electromigration is the movement of metal
atoms along a metallic strip, due to momentum exchange with electrons. The metal atoms that migrate along the line
tend to accumulate at grain boundaries.
The accumulation of metal can create fingers of metal that can short the
device. In the other possible scenario,
failure of the metal strip to adhere properly to the substrate can cause the
line to break, creating an “open” circuit.
It could also cause the line to break off connection with an active
component in the device circuit.
Element
|
Temperature (°C)
|
15
|
60
|
76
|
110
|
136
|
|
0.0005
|
0.001
|
0.0015
|
0.002
|
0.0025
|
|
0.001
|
0.0015
|
0.0025
|
0.006
|
0.025
|
Wire Bonds
|
|
|
0.000005
|
0.00002
|
0.0001
|
0.001
|
0.006
|
|
0.00001
|
0.00001
|
0.00001
|
0.00001
|
0.00005
|
|
0.00001
|
0.00001
|
0.00001
|
0.00001
|
0.00001
|
|
0.000005
|
0.000005
|
0.000006
|
0.000008
|
0.00001
|
|
|
|
0.0001
|
0.0003
|
0.0009
|
0.0027
|
0.007
|
|
0.005
|
0.01
|
0.03
|
0.09
|
0.27
|
|
0.0001
|
0.0003
|
0.0009
|
0.0027
|
0.007
|
Microcircuits
|
|
Quadgate or
equivalent
|
0.002
|
0.0036
|
0.018
|
0.082
|
0.24
|
|
0.004
|
0.0072
|
0.036
|
0.164
|
0.48
|
|
0.0125
|
0.0225
|
0.1125
|
0.512
|
1.5
|
|
0.025
|
0.0459
|
0.225
|
1.02
|
3.0
|
|
0.050
|
0.09
|
0.45
|
2.04
|
6.0
|
|
0.1
|
0.2
|
1.2
|
3.5
|
15
|
|
0.3
|
18.006
|
3.8
|
5.0
|
20
|
Table 1. MCM Circuit Element Base Failure Rates (%/1000 hr.)
Single crystal gallium arsenide (GaAs) has the
following properties that are relevant to transmission line design:
Dielectric constant 12.8
Loss tangent 0.00006
(high resistivity GaAs)
Surface roughness 1/40
micron
Thermal conductivity 32
watts/m-0K
Power dissipation in vias is minimal and therefore,
does not limit device performance. Active components, and in some cases thin
film resistors, tend to be the elements that require power derating.
Air bridges, designed to limit the current density to
less than 5x105 A/cm2 , show no degradation at
1750C. Mechanical tests of air bridges have shown no failures at
stress levels up to 30,000 g’s (centrifuge test) and under temperature cycling
tests from –65 to +1500
C.