Broadband RF port: LF to 500 MHz
Conversion gain: 3.7 dB
Noise figure: 12.2 dB
Input IP3: 22.7 dBm
Input P
LO drive: 0 dBm
Differential high impedance RF input port
Single-ended, 50 Ω LO input port
Single-supply operation: 5 V @ 98 mA
Power-down mode
Exposed paddle LFCSP: 3 mm × 3 mm
APPLICATIONS
Cellular base station receivers
ISM receivers
Radio links
RF instrumentation
: 8.3 dBm
1dB
LF to 500 MHz
AD8342
FUNCTIONAL BLOCK DIAGRAM
VPDC PWDN EXRB COMM
9
1
0
1
1
1
2
COMM
13
RFCM
14
RFIN
15
VPMX
16
VPLO LOCM LOIN COMM
AD8342
1
BIAS
2
Figure 1.
3
COMM
8
IFOP
7
6
IFOM
5
COMM
4
05352-001
GENERAL DESCRIPTION
The AD8342 is a high performance, broadband active mixer.
It is well suited for demanding receive-channel applications
that require wide bandwidth on all ports and very low intermodulation distortion and noise figure.
The AD8342 provides a typical conversion gain of 3.7 dB with
an RF frequency of 238 MHz. The integrated LO driver presents
a 50 Ω input impedance with a low LO drive level, helping to
minimize the external component count.
The differential high impedance broadband RF port allows for
easy interfacing to both active devices and passive filters. The
RF input accepts input signals as large as 1.6 V p-p or 8 dBm
(relative to 50 Ω) at P
1dB
.
The open-collector differential outputs provide excellent balance and can be used with a differential filter or IF amplifier,
such as the AD8369 or AD8351. These outputs can also be converted to a single-ended signal through the use of a matching
network or a transformer (balun). When centered on the VPOS
supply voltage, the outputs may swing ±2 V differentially.
The AD8342 is fabricated on an Analog Devices proprietary,
high performance SiGe IC process. The AD8342 is available in a
16-lead LFCSP. It operates over a −40°C to +85°C temperature
range. An evaluation board is also available.
Rev. 0
Information furnished by Analog Devices is believed to be accurate and reliable.
However, no responsibility is assumed by Analog Devices for its use, nor for any
infringements of patents or other rights of third parties that may result from its use.
Specifications subject to change without notice. No license is granted by implication
or otherwise under any patent or patent rights of Anal og Devices. Trademarks and
registered trademarks are the property of their respective owners.
VS = 5 V, TA = 25°C, fRF = 238 MHz, fLO = 286 MHz, LO power = 0 dBm, ZO = 50 Ω, R
nated into 100 Ω through a 2:1 ratio balun, unless otherwise noted.
Table 1.
Parameter Conditions Min Typ Max Unit
RF INPUT INTERFACE
Return Loss Hi-Z input terminated with 100 Ω off-chip resistor 10 dB
Input Impedance
DC Bias Level Internally generated; port must be ac-coupled 2.4 V
OUTPUT INTERFACE
Output Impedance Differential impedance, frequency = 48 MHz 10||0.5 kΩ||pF
DC Bias Voltage Supplied externally 4.75 VS 5.25 V
Power Range Via a 2:1 impedance ratio transformer 13 dBm
LO INTERFACE
Return Loss 10 dB
DC Bias Voltage Internally generated; port must be ac-coupled VS − 1.6 V
POWER-DOWN INTERFACE
PWDN Threshold 3.5 V
PWDN Response Time Device enabled, IF output to 90% of its final level 0.4 µs
Device disabled, supply current <5 mA 4 µs
PWDN Input Bias Current Device enabled −80 µA
Device disabled +100 µA
POWER SUPPLY
Positive Supply Voltage 4.75 5 5.25 V
Quiescent Current
VPDC Supply current for bias cells 5 mA
VPMX, IFOP, IFOM Supply current for mixer, R
VPLO Supply current for LO limiting amplifier 35 mA
Total Quiescent Current VS = 5 V 85 98 113 mA
Power-Down Current Device disabled 500 µA
Frequency = 238 MHz (measured at RFIN with RFCM acgrounded)
= 1.82 kΩ 58 mA
BIAS
= 1.82 kΩ, RF termination = 100 Ω, IF termi-
BIAS
1||0.4 kΩ||pF
Rev. 0 | Page 3 of 20
AD8342
AC PERFORMANCE
VS = 5 V, TA = 25°C, LO power = 0 dBm, ZO = 50 Ω, R
unless otherwise noted.
Table 2.
Parameter Conditions Min Typ Max Unit
RF FREQUENCY RANGE
LO FREQUENCY RANGE
IF FREQUENCY RANGE
1
1
1
50 500 MHz
High side LO 60 850 MHz
10 350 MHz
CONVERSION GAIN fRF = 460 MHz, fLO = 550 MHz, fIF = 90 MHz 3.2 dB
f
INPUT 1 dB COMPRESSION POINT fRF = 460 MHz, fLO = 550 MHz, fIF = 90 MHz 8.5 dBm
f
= 238 MHz, fLO = 286 MHz, fIF = 48 MHz 8.3 dBm
RF
LO TO IF OUTPUT LEAKAGE LO power = 0 dBm, fLO = 286 MHz −27 dBc
LO TO RF INPUT LEAKAGE LO power = 0 dBm, fLO = 286 MHz −55 dBc
2× LO TO IF OUTPUT LEAKAGE
LO power = 0 dBm, f
IF terminated into 100 Ω and measured with a differential probe
RF TO IF OUTPUT LEAKAGE RF power = −10 dBm, fRF = 238 MHz, fLO = 286 MHz −32 dBc
IF/2 SPURIOUS RF power = −10 dBm, fRF = 238 MHz, fLO = 286 MHz −70 dBc
1
Frequency ranges are those that were extensively characterized; this device can operate over a wider range. See the H section for details. igh IF Applications
= 1.82 kΩ, RF termination 100 Ω, IF terminated into 100 Ω via a 2:1 ratio balun,
BIAS
= 461 MHz, fLO = 550 MHz,
RF2
= 89 MHz each RF tone −10 dBm
IF2
= 239 MHz, fLO = 286MHz,
RF2
= 47MHz each RF tone −10 dBm
IF2
= 410 MHz, fLO = 550 MHz, f
RF2
= 188 MHz, fLO = 286 MHz, f
RF2
= 238 MHz, fLO = 286 MHz
RF
= 90 MHz,
IF1
= 48MHz,
IF1
22.2 dBm
22.7 dBm
50 dBm
44 dBm
−47
dBm
Rev. 0 | Page 4 of 20
AD8342
SPUR TABLE
VS = 5 V, TA = 25°C, RF and LO power = 0 dBm, fRF = 238MHz, fLO = 286MHz, ZO = 50 Ω, R
IF terminated into 100 Ω via a 2:1 ratio balun.
Note: Measured using standard test board. Typical noise floor of measurement system = −100 dBm.
Supply Voltage, VS 5.5 V
RF Input Level 12 dBm
LO Input Level 12 dBm
PWDN Pin VS + 0.5 V
IFOP, IFOM Bias Voltage 5.5 V
Minimum Resistor from EXRB to COMM 1.8 kΩ
Internal Power Dissipation 650 mW
θJA 77°C/W
Maximum Junction Temperature 135°C
Operating Temperature Range −40°C to +85°C
Storage Temperature Range −65°C to +150°C
ESD CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on the
human body and test equipment and can discharge without detection. Although this product features
proprietary ESD protection circuitry, permanent damage may occur on devices subjected to high energy
electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance
degradation or loss of functionality.
Stresses above those listed under Absolute Maximum Ratings
may cause permanent damage to the device. This is a stress
rating only; functional operation of the device at these or any
other conditions above those indicated in the operational section of this specification is not implied. Exposure to absolute
maximum rating conditions for extended periods may affect
device reliability.
Rev. 0 | Page 6 of 20
AD8342
M
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
X
M
N
M
M
C
I
F
O
F
P
C
R
V
R
4
3
5
6
1
1
1
1
PIN 1
INDICATOR
1VPLO
2LOCM
AD8342
3LOIN
TOP VIEW
(Not to Scale)
4COMM
5
6
M
M
O
M
F
O
I
C
Figure 2. 16-Lead LFCSP
Table 5. Pin Function Descriptions
Pin No. Mnemonic Function
1 VPLO Positive Supply Voltage for the LO Buffer: 4.75 V to 5.25 V.
2 LOCM
3 LOIN
AC Ground for Limiting LO Amplifier. Internally biased to Vs − 1.6 V. AC-couple to ground.
LO Input. Nominal input level 0 dBm. Input level range −10 dBm to +4 dBm (relative to 50 Ω). Internally
biased to Vs − 1.6 V. AC-couple.
4, 5, 8, 9, 13 COMM Device Common (DC Ground).
6, 7 IFOM, IFOP Differential IF Outputs (Open Collectors). Each requires dc bias of 5.00 V (nominal).
10 EXRB
Mixer Bias Voltage. Connect resistor from EXRB to ground. Typical value of 1.82 kΩ sets mixer current to
nominal value. Minimum resistor value from EXRB to ground = 1.8 kΩ. Internally biased to 1.17 V.
11 PWDN Connect to Ground for Normal Operation. Connect pin to VS for disable mode.
12 VPDC Positive Supply Voltage for the DC Bias Cell: 4.75 V to 5.25 V.
14 RFCM AC Ground for RF Input. Internally biased to 2.4 V. AC-couple to ground.
15 RFIN RF Input. Internally biased to 2.4 V. Must be ac-coupled.
16 VPMX Positive Supply Voltage for the Mixer: 4.75 V to 5.25 V.
12 VPDC
11 PWDN
10 EXRB
9 COMM
8
7
P
M
O
M
F
I
O
C
05352-002
Rev. 0 | Page 7 of 20
AD8342
TYPICAL PERFORMANCE CHARACTERISTICS
VS = 5 V, TA = 25°C, RF power = −10 dBm, LO power = 0 dBm, ZO = 50 Ω, R
100 Ω via a 2:1 ratio balun, unless otherwise noted.
6
= 1.82 kΩ, RF termination 100 Ω, IF terminated into
BIAS
6
5
IF = 48MHz
4
GAIN (dB)
3
IF = 10MHz
2
1
50
100 150 200 250 300 350 400 450 500
IF = 140MHz
IF = 90MHz
RF FREQUENCY (MHz)
Figure 3. Conversion Gain vs. RF Frequency
5
IF = 48MHz
4
3
IF = 140MHz
GAIN (dB)
2
1
0
–15–10–505
LO LEVEL (dBm)
IF = 90MHz
Figure 4. Gain vs. LO Level, RF Frequency = 238 MHz
Figure 21. Input IP2 vs. RF Frequency (Second RF = R F - 50 MHz)
60
58
IF = 10MHz
IF = 90MHz
IF = 140MHz
LO LEVEL (dBm)
IF = 48MHz
INPUT IP2 (dBm)
56
54
52
50
48
46
44
42
40
–15
–13 –11 –9–7–5–3–113
550
05352-010
05352-029
5
40
30
INPUT IP2 (dBm)
20
10
0
10
50100150200250300350
IF FREQUENCY (MHz)
05352-011
Figure 24. Input IP2 vs. IF Frequency (Second RF = R F - 50 MHz)
60
58
56
54
52
50
48
INPUT IP2 (dBm)
46
44
42
40
4.755.254.854.955.055.15
VPOS (V)
05352-030
Figure 22. Input IP2 vs. LO Level, f
14.0
13.5
13.0
12.5
12.0
NOISE FIGURE (dB)
11.5
11.0
100 150 200 250 300 350 400 450 500
50550
RF FREQUENCY (MHz)
= 238 MHz, ,f
RF
RF2
Figure 23. Noise Figure vs. RF Frequency, IF Frequency = 48 MHz
=188MHz
05352-016
Rev. 0 | Page 11 of 20
Figure 25. Input IP2 vs. Vpos, f
f
= 188 MHz, fLO = 286 MHz
RF2
16
14
12
10
8
6
NOISE FIGURE (dB)
4
2
0
60110160210260310
10
RF = 238MHz
IF FREQUENCY (MHz)
= 238 MHz,
RF1
RF = 460MHz
05352-017
Figure 26. Noise Figure vs. IF Frequency
AD8342
16
15
30
25
NORMAL
MEAN = 12.25
STD. DEV. = 0.14
NF PERCENTAGE
14
13
NF (dB)
12
11
10
–15
–13 –11 –9–7–5–3–113
Figure 27. Noise Figure vs. LO Power, f
5.0
4.5
4.0
3.5
3.0
2.5
GAIN (dB)
2.0
1.5
1.0
0.5
0
1.8
2.02.22.42.62.83.03.23.4
Figure 28. Gain vs. R
61
59
57
55
53
51
INPUT IP2 (dBm)
49
47
45
1.8
2.02.22.42.62.83.03.23.4
NF = 140MHz
NF = 10MHz
LO POWER (dBm)
NF = 90MHz
NF = 48MHz
RF
5
= 238 MHz
R
(kΩ)
BIAS
, RF Frequency = 238 MHz, LO Frequency = 286MHz
BIAS
R
(kΩ)
BIAS
05352-018
05352-024
05352-037
20
15
PERCENTAGE
10
5
0
11.8
11.9 12.0 12.1 12.2 12.3 12.4 12.5 12.6 12.7
Figure 30. Noise Figure Distribution, f
NOISE FIGURE (dB)
= 238 MHz, fLO = 286 MHz
RF
12.8
05352-023
BIAS
105
100
95
90
85
80
75
,
SUPPLY CURRENT (mA)
05352-015
05352-036
30
25
INPUT IP3
20
15
NOISE FIGURE
10
5
NOISE FIGURE AND INPUT IP3 (dBm)
0
1.83.0
2.02.22.42.62.8
R
BIAS
CURRENT
(kΩ)
Figure 31. Noise Figure, Input IP3 and Supply Current vs. R
f
= 238 MHz, f
RF1
10
9
8
7
6
5
4
INPUT P1dB (dBm)
3
2
1
0
1.8
2.02.22.42.62.83.03.23.4
= 239 MHz, fLO = 286 MHz
RF2
R
(kΩ)
BIAS
Figure 29. Input IP2 vs. R
, fRF = 238 MHz (Second RF = RF – 50MHz),
BIAS
f
= 286 MHz
LO
Rev. 0 | Page 12 of 20
Figure 32. Input P1dB vs. R
, fRF = 238 MHz, fLO = 286 MHz
BIAS
AD8342
V
0
–10
–20
–30
–40
–50
LEAKAGE (dBc)
–60
–70
–80
–90
50
250450650850
LO FREQUENCY (MHz)
Fig ure 3 3. LO t o RF Leak age v s. LO Fr eque ncy, LO Po wer = 0 dBm
05352-021
120
100
80
60
40
SUPPLY CURRENT (mA)
20
0
–40
–20020406080
TEMPERATURE (°C)
Figure 36. Supply Current vs. Temperature
05352-034
0
–5
–10
–15
–20
–25
–30
FEEDTHROUGH (dBc)
IF = 10MHz
–35
–40
–45
50
100 150 200 250 300 350 400 450 500
IF = 48MHz
RF FREQUENCY (MHz)
Figure 34. RF to IF Feedthrough, RF Power = −10 dBm
0
–5
–10
–15
–20
–25
–30
FEEDTHROUGH (dBc)
–35
–40
–45
50
150250350450550650750
850
LO FREQUENCY (MHz)
Fig ure 3 5. LO t o IF Fe edth roug h vs. LO Fr eque ncy, LO Po wer = 0 dBm
550
05352-020
05352-035
0
–2
–4
–6
–8
–10
–12
RETURN LOSS (dB)
–14
–16
–18
60
160260360460560660760
LO FREQUENCY (MHz)
Fig ure 3 7. LO R etur n Loss vs. LO Frequ ency
100pF
POS
VPOS
RF IN
100pF0.1µF
1
2
VPDC PWDNEXRB COMM
13
1nF
100Ω
1nF
100pF0.1µF
100pF0.1µF
COMM
RFCM
14
15
RFIN
16
VPMX
VPLOLOCMLOINCOMM
1
1nF
1
1
AD8342
2
1nF
1
LO IN
1.82kΩ
0
3
COMM
IFOP
IFOM
COMM
9
8
7
6
5
4
Figure 38. Characterization Circuit Used to Measure TPC Data
TC2-1T
100pF 0.1µF
860
05352-059
IF OUT
(50Ω)
VPOS
05352-058
Rev. 0 | Page 13 of 20
AD8342
CIRCUIT DESCRIPTION
The AD8342 is an active mixer optimized for operation within
the input frequency range of near dc to 500 MHz. It has a differential, high impedance RF input that can be terminated or
matched externally. The RF input can be driven either singleended or differentially. The LO input is a single-ended 50 Ω
input. The IF outputs are differential open-collectors. The mixer
current can be adjusted by the value of an external resistor to
optimize performance for gain, compression, and intermodulation, or for low power operation. Figure 39 shows the basic
blocks of the mixer, including the LO buffer, RF voltage-tocurrent converter, bias cell, and mixing core.
The RF voltage to RF current conversion is done via a resistively
degenerated differential pair. To drive this port single-ended,
the RFCM pin should be ac-grounded while the RFIN pin is
ac-coupled to the signal source. The RF inputs can also be
driven differentially. The voltage-to-current converter then
drives the emitters of a four-transistor switching core. This
switching core is driven by an amplified version of the local
oscillator signal connected to the LO input. There are three
limiting gain stages between the external LO signal and the
switching core. The first stage converts the single-ended LO
drive to a well-balanced differential drive. The differential drive
then passes through two more gain stages, which ensures a limited signal drives the switching core. This affords the user a
lower LO drive requirement, while maintaining excellent distortion and compression performance. The output signal of these
three LO gain stages drives the four transistors within the mixer
core to commutate at the rate of the local oscillator frequency.
The output of the mixer core is taken directly from its open
collectors. The open collector outputs present a high impedance
at the IF frequency. The conversion gain of the mixer depends
directly on the impedance presented to these open collectors. In
characterization, a 100 Ω load was presented to the part via a
2:1 impedance transformer.
The device also features a power-down function. Application of
a logic low at the PWDN pin allows normal operation. A high
logic level at the PWDN pin shuts down the AD8342. Power
consumption when the part is disabled is less than 10 mW.
EXTERNAL
BIAS
RESISTORVPDCPWDN
BIAS
VPLO
IFOP
IFOM
05352-040
RFIN
RFCM
Figure 39. Simplified Schematic Showing the Key Elements of the AD8342
TO
V
I
LO
INPUT
As shown in Figure 40, the IF output pins, IFOP and IFOM, are
directly connected to the open collectors of the NPN transistors
in the mixer core so the differential and single-ended impedances looking into this port are relatively high—on the order of
several kΩ. A connection between the supply voltage and these
output pins is required for proper mixer core operation.
IFOP IFOM
LOIN
RFCMRFIN
COMM
Figure 40. AD8342 Simplified Schematic
05352-041
The AD8342 has three pins for the supply voltage: VPDC,
VPMX, and VPLO. These pins are separated to minimize or
eliminate possible parasitic coupling paths within the AD8342
that could cause spurious signals or reduced interport isolation.
Consequently, each of these pins should be well bypassed and
decoupled as close to the AD8342 as possible.
The bias for the mixer is set with an external resistor (R
BIAS
)
from the EXRB pin to ground. The value of this resistor directly
affects the dynamic range of the mixer. The external resistor
should not be lower than 1.82 kΩ. Permanent damage to the
part could result if values below 1.8 kΩ are used. This resistor
sets the dc current through the mixer core. The performance
effects of changing this resistor can be seen in the Typical Performance Characteristics section.
Rev. 0 | Page 14 of 20
AD8342
AC INTERFACES
The AD8342 is designed to downconvert radio frequencies (RF)
to lower intermediate frequencies (IF) using a high or low-side
local oscillator (LO). The LO is injected into the mixer core at a
frequency higher or lower than the desired input RF. The
− f
difference between the LO and the RF , f
− fLO (low side) is the intermediate frequency, fIF. In addition
f
RF
LO
RF,
(high side) or
to the desired RF signal, an RF image is downconverted to the
+ f
desired IF frequency. The image frequency is at f
when
LO
IF
driven with a high side LO . When using a broadband load, the
conversion gain of the AD8342 is nearly constant over the
specified RF input band (see Figure 3).
The AD8342 is designed to operate over a broad frequency
range. It is essential to ac-couple RF and LO ports to prevent dc
offsets from skewing the mixer core in an asymmetrical manner, potentially degrading noise figure and linearity.
The RF input of the AD8342 is high impedance, 1 kΩ across the
frequency range shown in Figure 41. The input capacitance
decreases with frequency due to package parasitics.
2.001.00
1.75
1.500.75
1.25
1.000.50
0.75
RESISTANCE (kΩ)
0.500.25
0.25
00
0
100M 200M 300M 400M 500M 600M 700M 800M 900M
FREQUENCY (Hz)
1G
CAPACITANCE (pF)
05352-042
Figure 41. RF Input Impedance
The matching or termination used at the RF input of the
AD8342 has a direct effect on its dynamic range. The characterization circuit, as well as the evaluation board, uses a 100 Ω
resistor to terminate the RF port. This termination resistor in
shunt with the input stage results in a return loss of better than
−10 dBm (relative to 50 Ω). Table 4 shows gain, IP3, P1dB, and
noise figure for four different input networks. This data was
measured at an RF frequency of 250 MHz and at an LO
frequency of 300 MHz.
Table 4. Dynamic Performance for Various Input Networks
The RF port can also be matched using an LC circuit, as shown
in Figure 42.
50Ω
100nH
3.6pF
1kΩ
(1000 + j0) Ω
Z
f
Z
L
O
MAIN
= 50Ω
= 250MHz
05352-043
Figure 42. Matching Circuit
Impedance transformations of greater than 10:1 result in a
higher Q circuit and thus a narrow RF input bandwidth. A 1 kΩ
resistor is placed across the RF input of the device in parallel
with the device internal input impedance, creating a 500 Ω load.
This impedance is matched to as close as possible to 50 Ω for
the source, with standard components using a shunt C, series L
matching circuit (see Figure 43).
50.0
25.0
50.0
100.0
100.0
200.0
3
200.0
2
500.0
500.0
1
05352-044
25.0
Q = 3.0
10.0
4
10.0
Point 1(1000.0 + j0.0)Ω Q = 0.0 at 250.000 MHz
Point 2(500.0 + j0.0)Ω Q = 0.0 at 250.000 MHz
Point 3(55.6 − j157.2)Ω Q = 2.8 at 250.000 MHz
Point 4(55.6 − j0.1)Ω Q = 0.0 at 250.000 MHz
Rev. 0 | Page 15 of 20
Figure 43. LC Matching Example
AD8342
IF PORT
The IF port comprises open-collector differential outputs. The
NPN open collectors can be modeled as current sources that are
shunted with resistances of ~10 kΩ in parallel with capacitances
of ~1 pF.
The specified performance numbers for the AD8342 were
measured with 100 Ω differential terminations. However, different load impedances may be used where circumstances dictate. In general, lower load impedances result in lower conversion gain and lower output P1dB. Higher load impedances
result in higher conversion gain for small signals, but lower IP3
values for both input and output.
If the IF signal is to be delivered to a remote load, more than a
few millimeters away at high output frequencies, avoid unintended parasitic effects due to the intervening PCB traces. One
approach is to use an impedance transforming network or
transformer located close to the AD8342. If very wideband output is desired, a nearby buffer amplifier may be a better choice,
especially if IF response to dc is required. An example of such a
circuit is presented in Figure 45, in which the AD8351 differential amplifier is used to drive a pair of 75 Ω transmission lines.
The gain of the buffer can be independently set by appropriate
choice of the value for the gain resistor, R
50
45
40
35
30
25
20
RESISTANCE (kΩ)
15
10
5
0
100M 200M 300M 400M 500M 600M 700M 800M 900M
AD8342
COMM
IFOP
IFOM
COMM
0
V
+
S
8
R
7
6
R
5
FREQUENCY (Hz)
Figure 44. IF Port Impedance
C
F
100ΩR
C
F
+
+
AD8351
G
–
.
G
0.5
0.4
0.3
0.2
0.1
CAPACITANCE (pF)
0
–0.1
–0.2
1G
05352-045
V
S
Tx LINE ZO = 75Ω
Tx LINE ZO = 75Ω
Z
The high input impedance of the AD8351 allows for a shunt
differential termination to provide the desired 100 Ω load to the
AD8342 IF output port.
It is necessary to bias the open-collector outputs using one of
the schemes presented in Figure 47 and Figure 48. Figure 47
illustrates the application of a center tapped impedance transformer. The turns ratio of the transformer should be selected to
provide the desired impedance transformation. In the case of a
50 Ω load impedance, a 2-to-1 impedance ratio transformer
should be used to transform the 50 Ω load into a 100 Ω differential load at the IF output pins. Figure 48 illustrates a differential IF interface where pull-up choke inductors are used to bias
the open-collector outputs. The shunting impedance of the
choke inductors used to couple dc current into the mixer core
should be large enough at the IF operating frequency so it does
not load down the output current before reaching the intended
load. Additionally, the dc current handling capability of the
selected choke inductors needs to be at least 45 mA. The selfresonant frequency of the selected choke should be higher than
the intended IF frequency. A variety of suitable choke inductors
are commercially available from manufacturers such as Murata
and Coilcraft. Figure 46 shows the loading effects when using
nonideal inductors. An impedance transforming network may
be required to transform the final load impedance to 100 Ω at
the IF outputs. There are several good reference books that
explain general impedance matching procedures, including:
• Chris Bowick, RF Circuit Design, Newnes, Reprint Edition,
1997.
• David M. Pozar, Microwave Engineering, Wiley Text Books,
Analysis and Design, Prentice Hall, Second Edition, 1996.
90
120
60
150
L
210
50MHz
500MHz
500MHz
30
330
REAL
CHOKES
0180
50MHz
IDEAL
CHOKES
V
+
S
Z
Ω
0
0
1
=
L
Figure 45. AD8351 Used as Transmission Line Driver and Impedance Buffer
Rev. 0 | Page 16 of 20
05352-046
240
270
300
05352-049
Figure 46. IF Port Loading Effects Due to Finite Q Pull-Up Inductors
(Murata BLM18HD601SN1D Chokes)
AD8342
V
+
S
AD8342
8
COMM
IFOP
IFOM
COMM
7
6
5
ZL
2:1
Ω
0
0
1
=
I
F
OUT
Z
Ω
0
5
=
O
05352-047
Figure 47. Biasing the IF Port Open-Collector Outputs
Using a Center-Tapped Impedance Transformer
V
+
S
AD8342
8
COMM
IFOP
IFOM
COMM
R
C
F
IF
+
7
6
5
Z
R
F
V
+
S
OUT
Ω
0
0
1
=
L
IF
OUT
C
IMPEDANCE
TRANSFORMING
NETWORK
–
Z
L
05352-048
Figure 48. Biasing the IF Port Open-Collector Outputs
Using Pull-Up Choke Inductors
The AD8342 is optimized for driving a 100 Ω load. Although
the device is capable of driving a wide variety of loads, to maintain optimum distortion and noise performance, it is advised
that the presented load at the IF outputs is close to 100 Ω. The
linear differential voltage conversion gain of the mixer can be
modeled as
RGAv×=
LOADm
where:
g
1
G
m
R
is the single-ended load impedance.
LOAD
is the transistor transconductance and is equal to
g
m
1810/R
R
is 15 Ω.
e
The external R
m
+=1
π
BIAS
Rg
em
.
resistor is used to control the power dissipa-
BIAS
tion and dynamic range of the AD8342. Because the AD8342
has internal resistive degeneration, the conversion gain is primarily determined by the load impedance and the on-chip
degeneration resistors. Figure 49 shows how gain varies with IF
load. The external R
resistor has only a small effect. The
BIAS
most direct way to affect conversion gain is by varying the load
impedance. Small loads result in lower gains while larger loads
increase the conversion gain. If the IF load impedance is too
large it causes a decrease in linearity (P1dB, IP3). In order to
maintain positive conversion gain and preserve SFDR performance, the differential load presented at the IF port should
remain in the range of ~ 100 Ω to 250 Ω.
30
25
20
15
10
VOLTAGE GAIN (dB)
5
0
101000
100
IF LOAD (Ω)
MODELED
MEASURED
05352-057
Figure 49. Voltage Conversion Gain vs. IF Loading
LO CONSIDERATIONS
The LOIN port provides a 50 Ω load impedance with commonmode decoupling on LOCM. Again, common-grade ceramic
capacitors provide sufficient signal coupling and bypassing of
the LO interface.
The LO signal needs to have adequate phase noise characteristics and low second-harmonic content to prevent degradation
of the noise figure performance of the AD8342. An LO plagued
with poor phase noise can result in reciprocal mixing, a mechanism that causes spectral spreading of the downconverted signal, limiting the sensitivity of the mixer at frequencies close-in
to any large input signals. The internal LO buffer provides
enough gain to hard-limit the input LO and provide fast switching of the mixer core. Odd harmonic content present on the LO
drive signal should not impact mixer performance; however,
even-order harmonics cause the mixer core to commutate in an
unbalanced manner, potentially degrading noise performance.
Simple lumped element low-pass filtering can be applied to help
reject the harmonic content of a given local oscillator, as shown
in Figure 50. The filter depicted is a common 3-pole Chebyshev,
designed to maintain a 1-to-1 source-to-load impedance ratio
with no more than 0.5 dB of ripple in the pass band. Other filter
structures can be effective as long as the second harmonic of the
LO is filtered to negligible levels, for example, ~30 dB below the
fundamental.
AD8342
LOIN3COMM
LOCM
L2
1.28R
2πf
2
L
L
C3 =
c
R
S
LO
SOURCE
C1 =
f
- FILTER CUTOFF FREQUENCY
C
1.864
2
πf
R
c
L
C1C3
FOR RS= R
L2 =
Figure 50. Using a Low-Pass Filter to Reduce LO Second Harmonic
4
R
L
1.834
2
πf
R
c
L
05352-050
Rev. 0 | Page 17 of 20
AD8342
HIGH IF APPLICATIONS
In some applications it may be desirable to use the AD8342 as
an up-converting mixer. The AD8342 is a broadband mixer
capable of both up and down conversion. Unlike other mixers
that rely on on-chip reactive circuitry to optimize performance
over a specific band, the AD8342 is a versatile general-purpose
device that can be used from arbitrarily low frequencies to several GHz. In general, the following considerations help to ensure optimum performance:
• Minimize ac loading impedance of IF port bias network.
• Maximize power transfer to the desired ac load.
• For maximum conversion gain and the lowest noise per-
formance reactively match the input as described in the
IF Port section.
• For maximum input compression point and input intercept
points resistively terminate the input as described in the
IF Port section.
As an example, Figure 51 shows the AD8342 as an upconverting mixer for a WCDMA single-carrier transmitter design. For this application, it was desirable to achieve −65 dBc
adjacent channel power ratio (ACPR) at a −13 dBm output
power level. The ACPR is a measure of both distortion and
noise carried into an adjacent frequency channel due to the
finite intercept points and noise figure of an active device.
Because a WCDMA channel encompasses a bandwidth of
almost 5 MHz, it is necessary to keep the Q of the matching
circuit low enough so that phase and magnitude variations are
below an acceptable level over the 5 MHz band. It is possible
to use purely reactive matching to transform a 50 Ω source
to match the raw ~1 kΩ input impedance of the AD8342.
However, the L and C component variations could present
1
1
AD8342
2
1nF
1
3
1970MHz
OSC
1.82kΩ
0
1nF
COMM
IFOP
IFOM
COMM
VPOS
9
4
100pF
8
34nH
VPOS
34nH
100pF
ETC1-1-13
1nF
1nF
2140MHz OUT
7
6
5
production concerns due to the sensitivity of the match. For
this application, it is advantageous to shunt down the ~1 kΩ
input impedance using an external shunt termination resistor
to allow for a lower Q reactive matching network. The input is
terminated across the RFIN and RFCM pins using a 499 Ω
termination. The termination should be as close to the device as
possible to minimize standing wave concerns. The RFCM is
bypassed to ground using a 1 nF capacitor. A dc blocking capacitor of 1 nF is used to isolate the dc input voltage present on
the RFIN pin from the source. A step-up impedance transformation is realized using a series L shunt C reactive network.
The actual values used need to accommodate for the series L
and stray C parasitics of the connecting transmission line segments. When using the customer evaluation board with the
components specified in Figure 51, the return loss over a 5 MHz
band centered at 170 MHz was better than 10 dB.
External pull-up choke inductors are used to feed dc bias into
the open-collector outputs. It is desirable to select pull-up choke
inductors that present high loading reactance at the output
frequency. Coilcraft 0302CS series inductors were selected due
to their very high self-resonant frequency and Q. A 1:1 balun
was ac-coupled to the output to convert the differential output
to a single-ended signal and present the output with a 50 Ω
ac loading impedance.
The performance of the circuit is shown in Figure 52. The average ACPR of the adjacent and alternate channels is presented
vs. output power. The circuit provides a 65 dBc ACPR at
−13 dBm output power. The optimum ACPR power level can be
shifted to the right or left by adjusting the output loading and
the loss of the input match.
–60
–62
05352-052
–64
–66
ACPR (dBc)
–68
–70
–250
–20–15–10–5
OUTPUT POWER (dBm)
Figure 52. Single Carrier WCDMA ACPR Performance of Tx Up-Conversion
Circuit (Test Model 1_64)
ADJACENT
CHANNELS
ALTERNATE
CHANNELS
05352-053
Rev. 0 | Page 18 of 20
AD8342
EVALUATION BOARD
An evaluation board is available for the AD8342. The evaluation board is configured for single-ended signaling at the IF output port via a
balun transformer. The schematic for the evaluation board is presented in Figure 53.
R9
0Ω
PWDN
LOCM LOIN
1
C7
11
2
R8
10kΩ
DUT
10
EXRB
3
INLO
COMM
COMM
C8
1000pF
R6
1.82kΩ
9
COMM
IFOP
IFOM
COMM
4
C11
100pF
8
7
6
5
OPEN
Z2
OPEN
R10
0Ω
Z1
OPEN
R12
OPEN
Z3
R11
0Ω
Z4
OPEN
R16
0Ω
C10
100pF
R15
0Ω
R3
OPEN
T1
34
2
16
TC2-1T
R4
OPEN
C9
0.1µF
100Ω TRACES,
NO GROUND PLANE
VPOS
RF_IN
VPOS
PWDN
L1
0Ω
C14
OPEN
VPOSGND
0.1µF
0.1µF
C2
C5
PWDN
VPOS
0.1µF
W1
C12
50Ω
TRACE
R5
100Ω
R1
0Ω
R2
0Ω
R7
0Ω
1000pF
1000pF
C4
1000pF
C6
1000pF
C13
100pF
12
VPDC
13
COMM
C1
14
RFCM
C3
15
RFIN
16
VPMX
VPLO
1000pF
Figure 53. Evaluation Board
Table 6. Evaluation Board Configuration Options
Component Function Default Conditions
R1, R2, R7,
C2, C4, C5, C6, C10
C12, C13, C14, C9
Supply decoupling. Shorts or power supply decoupling resistors and filter
capacitors.