7
100 pF
HSMS-285A
VIDEO
OUT
RF
INPUT
WIDTH = 0.017"
LENGTH = 0.436"
WIDTH = 0.078"
LENGTH = 0.165"
TRANSMISSION LINE
DIMENSIONS ARE FOR
MICROSTRIP ON
0.032" THICK FR-4.
Figure 17. 2.45 GHz Matching
Network for the HSMS-285A
Series.
0.030" PLATED THROUGH HOLE,
3 PLACES
0.094" THROUGH, 4 PLACES
FINISHED
BOARD
SIZE IS
1.00" X 1.00".
MATERIAL IS
1/32" FR-4
EPOXY/
FIBERGLASS,
1 OZ. COPPER
BOTH SIDES.
NOTE THAT
THE BACK SIDE
OF THE BOARD
IS A GROUND
PLANE.
Figure 18. Physical Realization.
0
0
0
1
1.000
0.900
0.100
0.00 REF.
0.670
0.330
0.00 REF.
0
0
#2-56 TAP
0.40 MIN.,
4 PLACES
#2-56 TAP
THROUGH,
4 PLACES
MATERIAL:
0.250" H.H.
BRASS PLATE
Figure 19. Mounting Plate.
Figure 20. Test Detector.
CHIP CAPACITOR, 20 TO 100 pF
HSMS-2850
VIDEO OUT
RF IN
FREQUENCY (GHz): 2.3-2.6
Figure 21. Input Impedance.
R
2.3
-20
FREQUENCY (GHz)
2.45
0
-10
-15
2.6
-5
Figure 22. Input Return Loss.
2.45 GHz Detector Circuit
At 2.45 GHz, the RF impedance of
the HSMS-285A series is closer to
the line of constant susceptance
which passes through the center of
the chart, resulting in a design
which is realized entirely in
distributed elements — see
Figure 17.
In order to save cost (at the
expense of having a larger circuit),
an open circuit shunt stub could
be substituted for the chip
capacitor. On the other hand, if
space is at a premium, the long
series transmission line at the
input to the diode can be replaced
with a lumped inductor.
A possible physical realization of
such a network is shown in
Figure 18.
This board is mounted on the
brass or aluminum mounting plate
shown in Figure 19.
Two SMA connectors (E.F.
Johnson 142-0701-631 or
equivalent), a high-Q capacitor
(ATC 100A101MCA50 or
equivalent), miscellaneous
hardware and an HSMS-285B are
added to create the test circuit
shown in Figure 20.
The calculated input impedance
for this network is shown in
Figure 21.
The corresponding input match is
shown in Figure 22. As was the
case with the lower frequency
design, bandwidth is more than
adequate for the intended RFID
application. Note that this same
design applies to the HSMS-286A
series when it is used with 3 to
5
μ
A of external bias.
A word of caution to the designer
is in order. A glance at Figure 21
will reveal the fact that the circuit
does not provide the optimum
impedance to the diode at
2.45GHz. The temptation will be
to adjust the circuit elements to
achieve an ideal single frequency
match, as illustrated in Figure 23.
This does indeed result in a very
good match at midband, as shown
in Figure 24.
However, bandwidth is narrower
and the designer runs the risk of a
shift in the midband frequency of
his circuit if there is any small
deviation in circuit board or diode
characteristics due to lot-to-lot
variation or change in temper-
ature. The matching technique
illustrated in Figure 21 is much
less sensitive to changes in diode
and circuit board processing.