6
RS is perhaps the easiest to
measure accurately. The V-I curve
is measured for the diode under
forward bias, and the slope of the
curve is taken at some relatively
high value of current (such as
5 mA). This slope is converted
into a resistance Rd.
0.026
RS = Rd – ––––––
If
RV and CJ are very difficult to
measure. Consider the impedance
of CJ = 0.16 pF when measured at
1 MHz — it is approximately
1MΩ. For a well designed zero
bias Schottky, RV is in the range of
5 to 25 KΩ, and it shorts out the
junction capacitance. Moving up
to a higher frequency enables the
measurement of the capacitance,
but it then shorts out the video
resistance. The best measurement
technique is to mount the diode in
series in a 50 Ω microstrip test
circuit and measure its insertion
loss at low power levels (around
-20 dBm) using an HP8753C
network analyzer. The resulting
display will appear as shown in
Figure 7.
INSERTION LOSS (dB)
3
-40
FREQUENCY (MHz)
-10
-25
3000
-20
10 1000100
-35
-30
-15
50 Ω
50 Ω
0.16 pF
50 Ω
50 Ω9 KΩ
Figure 7. Measuring C
J
and R
V
.
At frequencies below 10 MHz, the
video resistance dominates the
loss and can easily be calculated
from it. At frequencies above
300 MHz, the junction capacitance
sets the loss, which plots out as a
straight line when frequency is
plotted on a log scale. Again,
calculation is straightforward.
LP and CP are best measured on
the HP8753C, with the diode
terminating a 50 Ω line on the
input port. The resulting tabula-
tion of S11 can be put into a
microwave linear analysis
program having the five element
equivalent circuit with RV, CJ and
RS fixed. The optimizer can then
adjust the values of LP and CP
until the calculated S11 matches
the measured values. Note that
extreme care must be taken to
de-embed the parasitics of the
50 Ω test fixture.
Detector Circuits
When DC bias is available,
Schottky diode detector circuits
can be used to create low cost RF
and microwave receivers with a
sensitivity of -55 dBm to
-57 dBm.[1] These circuits can take
a variety of forms, but in the most
simple case they appear as shown
in Figure 8. This is the basic
detector circuit used with the
HSMS-285x family of diodes.
In the design of such detector
circuits, the starting point is the
equivalent circuit of the diode, as
shown in Figure 6.
Of interest in the design of the
video portion of the circuit is the
diode’s video impedance— the
other four elements of the equiv-
alent circuit disappear at all
reasonable video frequencies. In
general, the lower the diode’s
video impedance, the better the
design.
VIDEO
OUT
RF
IN
Z-MATCH
NETWORK
VIDEO
OUT
Z-MATCH
NETWORK
RF
IN
The situation is somewhat more
complicated in the design of the
RF impedance matching network,
which includes the package
inductance and capacitance
(which can be tuned out), the
series resistance, the junction
capacitance and the video
resistance. Of these five elements
of the diode’s equivalent circuit,
the four parasitics are constants
and the video resistance is a
function of the current flowing
through the diode.
26,000
RV ≈ ––––––
IS + Ib
where
IS = diode saturation current
in µA
Ib = bias current in µA
Saturation current is a function of
the diode’s design,[2] and it is a
constant at a given temperature.
For the HSMS-285x series, it is
typically 3 to 5 µA at 25°C.
Saturation current sets the detec-
tion sensitivity, video resistance
and input RF impedance of the
zero bias Schottky detector diode.
[1] Agilent Application Note 923, Schottky Barrier Diode Video Detectors.
[2] Agilent Application Note 969, An Optimum Zero Bias Schottky Detector Diode.
Figure 8. Basic Detector Circuits.