ADRF5026 Data Sheet
Rev. 0 | Page 10 of 13
APPLICATIONS INFORMATION
EVALUATION BOARD
The ADRF5026-EVALZ evaluation board is a 4-layer evaluation
board. The outer copper (Cu) layers are 0.5 oz (0.7 mil) plated
to 1.5 oz (2.2 mil) and are separated by dielectric materials.
Figure 15 shows the ADRF5026-EVALZ evaluation board stack up.
1.5oz Cu (2.2mil) 1.5oz Cu (2.2mil)
0.5oz Cu (0.7mil)
0.5oz Cu (0.7mil)
1.5oz Cu (2.2mil)
RO4003
1.5oz Cu (2.2mil)
TOTAL THICKNESS
~62mil
W = 14mil
G = 7mil
T = 2.2mil
H = 8mil
16767-015
Figure 15. Evaluation Board Stack Up
All RF and dc traces are routed on the top copper layer, whereas
the inner and bottom layers are ground planes that provide a
solid ground for the RF transmission lines. The top dielectric
material is 8 mil Rogers RO4003, which offers optimal high
frequency performance. The middle and bottom dielectric
materials provide mechanical strength. The overall board
thickness is 62 mil, which allows 2.4 mm RF launchers to be
connected at the board edges.
16767-016
Figure 16. Evaluation Board Layout
The RF transmission lines are designed using a coplanar
waveguide (CPWG) model with a width of 14 mil and a ground
spacing of 7 mil, and have a characteristic impedance of 50 Ω.
For optimal RF and thermal grounding, as many plated through
vias as possible are arranged around transmission lines and
under the exposed pad of the package.
The RF input and output ports (RFC, RF1, and RF2) are connected
through 50 Ω transmission lines to the 2.4 mm launchers (J1,
J2, and J3, respectively). These high frequency RF launchers are
connected by contact and are not soldered to the board.
The thru calibration line, THRU CAL, can calibrate out the
board loss effects from the ADRF5026-EVALZ evaluation board
measurements to determine the device performance at the pins
of the IC. Figure 17 shows the typical board loss for the
ADRF5026-EVALZ evaluation board at room temperature, the
embedded insertion loss, and the de-embedded insertion loss
for the ADRF5026.
0
–8
–7
–6
–5
–4
–3
–2
–1
0 5 10 15 20 25 30 35 40 45
INSERTION LOSS (dB)
FREQUENCY (GHz)
DEEMBEDDED IL
EMBEDDED IL
THRU LOSS
16767-017
Figure 17. Insertion Loss vs. Frequency
Two power supply ports are connected to the VDD and VSS test
points, and the ground reference is connected to the GND test
point. On the supply traces, VDD and VSS, a 100 pF bypass
capacitor filters high frequency noise. Additionally, unpopulated
component positions are available for applying extra bypass
capacitors.
Two control ports are connected to the EN and CTRL test
points. There are provisions for the resistor capacitor (RC) filter
to eliminate dc-coupled noise, if needed by the application.
The ADRF5026-EVALZ evaluation board schematic is shown in
Figure 18.
G
N
D
R
F
2
G
N
D
G
N
D
G
N
D
G
N
D
G
N
D
R
F
1
G
N
D
G
N
D
GND
GND
RFC
GND
GND
GND
EN
VSS
VSS
CTRL
VDD
EN
CTRL
VDD
1
2
3
4
5
678910
11
12
13
14
15
1617181920
ADRF5026
RF2
RFC
RF1
R4
0
R3
0
C10
100pF
C7
100pF
THRU_CAL
16767-018
Figure 18. Simplified Evaluation Board Schematic
Data Sheet ADRF5026
Rev. 0 | Page 11 of 13
16767-025
Figure 19. Evaluation Board Component Placement
Table 6. Evaluation Board Components
Component Description
RF1_A, RFC_A, RF2_A End launch connectors, 2.4 mm
VDD_A, VSS_A, CTRL_A, EN_A, GND_A Through-hole mount test points
C7, C10 100 pF capacitors, 0402 package
R3, R4 0 Ω resistors, 0402 package
U2 ADRF5026 SPDT switch
PROBE MATRIX BOARD
The probe matrix board is a 4-layer board. This board also uses
an 8 mil Rogers RO4003 dielectric. The outer copper layers are
0.5 oz (0.7 mil) plated to 1.5 oz (2.2 mil). The RF transmission
lines were designed using a CPWG model with a width of
14 mil and a ground spacing of 7 mil to have a characteristic
impedance of 50 Ω.
1.5oz Cu (2.2mil) 1.5oz Cu (2.2mil)
0.5oz Cu (0.7mil)
0.5oz Cu (0.7mil)
1.5oz Cu (2.2mil)
RO4003
1.5oz Cu (2.2mil)
TOTAL THICKNESS
~62mil
W = 14mil
G = 7mil
T = 2.2mil
H = 8mil
16767-019
Figure 20. Probe Matrix Board Stack Up
Figure 20 and Figure 21 show the stack up and the layout,
respectively, of the probe matrix board. Measurements are made
using GSG probes at close proximity to the RF pins. Probing
reduces the reflections caused by mismatch arising from
connectors, cables, and board layout, resulting in a more
accurate measurement of insertion loss and return loss. Signal
coupling between the RF probes limits the isolation measurement.
The ADRF5026-EVALZ evaluation board is used for making
isolation measurements.
RF traces for a through-reflect-line (TRL) calibration are
designed on the probe matrix board. Board loss is compensated
for by using a nonzero line length at calibration. The actual
board duplicates the same layout in matrix form, which allows
multiple devices to assemble at once. Insertion loss and return
loss measurements are made on this board, while isolation
measurements are made on the ADRF5026-EVALZ evaluation
board.
RFC
VSS
EN
CTRL
VDD
RF2
EPAD
RF1
16767-020
7mil
7mil
14mil 8mil
10mil
Figure 21. Probe Matrix Board Layout
ADRF5026 Data Sheet
Rev. 0 | Page 12 of 13
Narrow-Band Impedance Matching
5G mmWave Frequencies
Narrow-band impedance matching on the RF transmission
lines can improve return loss and insertion loss for a targeted
frequency range. The impedance matched circuit, highlighted
in Figure 22, achieves a flat insertion loss response of 2.4 dB
from 28 GHz to 43 GHz (see Figure 23). The dimensions of the
50 lines are 14 mil trace width and a 7 mil gap. To implement
this impedance matched circuit, an 8 mil trace with a width of
5 mil is inserted between the pin pad and the 50 Ω trace.
Table 7, Figure 23, Figure 24, and Figure 25 show the measured
performance of ADRF5026 on the impedance matched circuit
on the probe matrix board.
16767-021
RFC
VSS
EN
CTR
L
VDD
RF2
EPAD
RF1
7mil
7mil
5mil
8mil
8mil
10mil
14mil
Figure 22. Impedance Matched Circuit
Table 7. Impedance Matched Parameters
Parameter Test Condition Typ Unit
Insertion Loss See Figure 23
Between RFC and RF1/RF2 100 MHz to 18 GHz 1.3 dB
18 GHz to 26 GHz 2.0 dB
26 GHz to 35 GHz 2.4 dB
35 GHz to 40 GHz 2.4 dB
40 GHz to 44 GHz 2.5 dB
Return Loss See Figure 24
RFC and RF1/RF2 (On) 100 MHz to 18 GHz 17 dB
18 GHz to 26 GHz 10 dB
26 GHz to 35 GHz 7 dB
35 GHz to 40 GHz 9 dB
40 GHz to 44 GHz 15 dB
RF1/RF2 (Off) See Figure 24
100 MHz to 18 GHz 18 dB
18 GHz to 26 GHz 17 dB
26 GHz to 35 GHz 18 dB
35 GHz to 40 GHz 12 dB
40 GHz to 44 GHz 7 dB
0
–5
–4
–3
–2
–1
0 5 10 15 20 25 30 35 4540
INSERTION LOSS (dB)
FREQUENCY (GHz)
16767-022
Figure 23. Insertion Loss vs. Frequency, with Impedance Matching
0
–40
–35
–30
–25
–20
–15
–10
–5
0 5 10 15 20 25 30 35 40 45
RETURN LOSS (dB)
FREQUENCY (GHz)
RF1/RF2 OFF
RF1/RF2 ON
RFC
16767-023
Figure 24. Return Loss vs. Frequency, with Impedance Matching
0
–110
–70
–80
–90
–100
–60
–50
–40
–30
–20
–10
0 5 10 15 20 25 30 35 40 45
ISOLATION (dB)
FREQUENCY (GHz)
RF1 TO RF2
RFC TO RF1/RF2
16767-024
Figure 25. Isolation vs. Frequency, with Impedance Matching

ADRF5026BCCZN

Mfr. #:
Manufacturer:
Analog Devices Inc.
Description:
RF Switch ICs High isolation SP DT,40GHz,fast switch
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union

Products related to this Datasheet