14
Noise Parameter Applications Information
F
min
values at 2 GHz and higher are based on
measurements while the F
mins
below 2 GHz have been
extrapolated. The F
min
values are based on a set of
16 noise gure measurements made at 16 di erent
impedances using an ATN NP5 test system. From these
measurements, a true F
min
is calculated. F
min
represents
the true minimum noise gure of the device when the
device is presented with an impedance matching network
that transforms the source impedance, typically 50Ω, to
an impedance represented by the re ection coe cient
o
. The designer must design a matching network that
will present
o
to the device with minimal associated
circuit losses. The noise gure of the completed ampli er
is equal to the noise gure of the device plus the losses
of the matching network preceding the device. The
noise gure of the device is equal to F
min
only when the
device is presented with
o
. If the re ection coe cient
of the matching network is other than
o
, then the noise
gure of the device will be greater than F
min
based on
the following equation.
NF = F
min
+ 4 R
n
|
s
–
o
|
2
Zo (|1 +
o
|
2
)(1 –
s
|
2
)
Where R
n
/Z
o
is the normalized noise resistance,
o
is
the optimum re ection coe cient required to produce
F
min
and
s
is the re ection coe cient of the source
impedance actually presented to the device. The losses
of the matching networks are non-zero and they will
also add to the noise gure of the device creating a
higher ampli er noise gure. The losses of the matching
networks are related to the Q of the components and
associated printed circuit board loss.
o
is typically fairly
low at higher frequencies and increases as frequency is
lowered. Larger gate width devices will typically have a
lower
o
as compared to narrower gate width devices.
Typically for FETs, the higher
o
usually infers that an
impedance much higher than 50Ω is required for the
device to produce F
min
. At VHF frequencies and even
lower L Band frequencies, the required impedance can
be in the vicinity of several thousand ohms. Matching to
such a high impedance requires very hi-Q components
in order to minimize circuit losses. As an example at
900 MHz, when airwwound coils (Q> 100) are used for
matching networks, the loss can still be up to 0.25 dB
which will add directly to the noise gure of the device.
Using muiltilayer molded inductors with Qs in the 30 to
50 range results in additional loss over the airwound coil.
Losses as high as 0.5 dB or greater add to the typical 0.15
dB F
min
of the device creating an ampli er noise gure
of nearly 0.65 dB. A discussion concerning calculated
and measured circuit losses and their e ect on ampli er
noise gure is covered in Avago Application 1085.
Reliability Data
Nominal Failures per million (FPM) 90% con dence Failures per million (FPM)
for di erent durations for di erent durations
Channel (FITs) 1 year 5 year 10 year 30 year (FITs) 1 year 5 year 10 year 30 year
Temperature 1000 1000
(
o
C) hours hours
100 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
125 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 11
140 <0.1 <0.1 <0.1 <0.1 160 <0.1 <0.1 6 160 9.3K
150 <0.1 <0.1 2 140 26K <0.1 0.3 780 8800 131K
160 <0.1 <0.1 920 21K 370K <0.1 67 24K 120K 520K
180 <0.1 4400 450K 830K 1000K 21 53K 590K 850K 1000K
NOT
recommended
Predicted failures with temperature extrapolated from failure distribution and activation energy data of higher temperature
operational life STRIFE of PHEMT process
