+7.5 dB to +55.5 dB
Low output-referred noise: 48 nV/√Hz typical
Active input impedance matching
Optimized for 10-bit/12-bit ADCs
Selectable output clamping level
Single 5 V supply operation
AD8332 and AD8334 available in lead frame chip scale package
APPLICATIONS
Ultrasound and sonar time-gain controls
High performance AGC systems
I/Q signal processing
High speed, dual ADC drivers
GENERAL DESCRIPTION
The AD8331/AD8332/AD8334 are single-, dual-, and quadchannel ultralow noise, linear-in-dB, variable gain amplifiers
(VGAs). Optimized for ultrasound systems, they are usable as a
low noise variable gain element at frequencies up to 120 MHz.
Included in each channel are an ultralow noise preamplifier
(LNA), an X-AMP® VGA with 48 dB of gain range, and a
selectable gain postamplifier with adjustable output limiting.
The LNA gain is 19 dB with a single-ended input and
differential outputs. Using a single resistor, the LNA input
impedance can be adjusted to match a signal source without
compromising noise performance.
The 48 dB gain range of the VGA makes these devices suitable
for a variety of applications. Excellent bandwidth uniformity is
maintained across the entire range. The gain control interface
provides precise linear-in-dB scaling of 50 dB/V for control
voltages between 40 mV and 1 V. Factory trim ensures excellent
part-to-part and channel-to-channel gain matching.
Rev. E
Information furnished by Analog Devices is believed to be accurate and reliable. However, no
responsibility is assumed by Anal og Devices for its use, nor for any infringements of patents or ot her
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 Analog Devices.
Trademarks and registered trademarks are the property of their respective owners.
AD8331/AD8332/AD8334
FUNCTIONAL BLOCK DIAGRAM
IPLOPLON
LNA
INH
LMD
+
19dB
–
LNA VCM
BIAS
AD8331/AD8332/AD8334
–
48dB
ATTENUATOR
+
VGA BIAS AND
INTERPOL ATOR
Figure 1. Signal Path Block Diagram
60
50
40
30
20
GAIN (dB)
10
0
–10
100k1M10M100M1G
V
GAIN
V
GAIN
V
GAIN
V
GAIN
V
GAIN
V
GAIN
FREQUENCY (Hz)
Figure 2. Frequency Response vs. Gain
Differential signal paths result in superb second- and thirdorder distortion performance and low crosstalk.
The VGA’s low output-referred noise is advantageous in driving
high speed differential ADCs. The gain of the postamplifier can
be pin selected to 3.5 dB or 15.5 dB to optimize gain range and
output noise for 12-bit or 10-bit converter applications. The
output can be limited to a user-selected clamping level,
preventing input overload to a subsequent ADC. An external
resistor adjusts the clamping level.
The operating temperature range is −40°C to +85°C. The
AD8331 is available in a 20-lead QSOP package, the AD8332 is
available in 28-lead TSSOP and 32-lead LFCSP packages, and
the AD8334 is available in a 64-lead LFCSP package.
Power-Up Response Time V
V
HILO GAIN RANGE INTERFACE (PIN HILO)
Logic Level to Select HI Gain Range 2.25 5 V
Logic Level to Select LO Gain Range 0 1.0 V
Input Resistance 50 kΩ
OUTPUT CLAMP INTERFACE
(PIN RCLMP; HI OR LO GAIN)
Accuracy
HILO = LO R
HILO = HI R
MODE INTERFACE (PIN MODE)
Logic Level for Positive Gain Slope 0 1.0 V
Logic Level for Negative Gain Slope 2.25 5 V
Input Resistance 200 kΩ
POWER SUPPLY (PIN VPS1, PIN VPS2,
PIN VPSV, PIN VPSL, PIN VPOS)
Supply Voltage 4.5 5.0 5.5 V
Quiescent Current per Channel
AD8331 20 25 mA
AD8332, AD8334 20 29 mA
Power Dissipation per channel No signal
AD8331 125 mW
AD8332, AD8334 145 mW
Power-Down Current
AD8332 (VGA and LNA Disabled) 50 300 600 μA
AD8331 (VGA and LNA Disabled) 50 240 400 μA
LNA Current
AD8331 (ENBL) Each channel 7.5 11 15 mA
AD8332, AD8334 (ENBL) Each channel 7.5 12 15 mA
VGA Current
AD8331 (ENBV) 7.5 14 20 mA
AD8332, AD8334 (ENBV) 7.5 17 20 mA
PSRR V
1
All dBm values are referred to 50 Ω.
2
The absolute gain refers to the theoretical gain expression in Equation 1.
3
Best-fit to linear-in-dB curve.
4
The current is limited to ±1 mA typical.
= 30 mV p-p 300 μs
INH
= 150 mV p-p 4 ms
INH
= 2.74 kΩ, V
CLMP
= 2.21 kΩ, V
CLMP
= 1 V p-p (clamped) ±50 mV
OUT
= 1 V p-p (clamped) ±75 mV
OUT
= 0 V, f = 100 kHz −68 dB
GAIN
Rev. E | Page 6 of 40
Page 7
AD8331/AD8332/AD8334
ABSOLUTE MAXIMUM RATINGS
Table 2.
Parameter Rating
Voltage
Supply Voltage (VPSn, VPSV, VPSL, VPOS) 5.5 V
Input Voltage (INHn) VS + 200 mV
ENB, ENBL, ENBV, HILO Voltage VS + 200 mV
GAIN Voltage 2.5 V
Power Dissipation
AR Package
1
0.96 W
CP-20 Package (AD8331) 1.63 W
CP-32 Package (AD8332) 1.97 W
RQ Package
1
0.78 W
CP-64 Package (AD8334) 0.91 W
Temperature
Operating Temperature Range −40°C to +85°C
Storage Temperature Range −65°C to +150°C
Lead Temperature (Soldering 60 sec) 300°C
θ
JA
AR Package
CP-20 Package
CP-32 Package
RQ Package
CP-64 Package
1
Four-layer JEDEC board (2S2P).
2
Exposed pad soldered to board, nine thermal vias in pad—JEDEC, 4-layer
board J-STD-51-9.
3
Exposed pad soldered to board, 25 thermal vias in pad—JEDEC, 4-layer
board J-STD-51-9.
1
2
2
1
3
68°C/W
40°C/W
33°C/W
83°C/W
24.2°C/W
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.
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.
Rev. E | Page 7 of 40
Page 8
AD8331/AD8332/AD8334
PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS
LMD
INH
VPSL
LON
LOP
COML
VIP
VIN
MODE
GAIN
1
2
3
4
5
6
(Not to Scale)
7
8
9
10
PIN 1
INDICAT OR
AD8331
TOP VIEW
20
19
18
17
16
15
14
13
12
11
COMM
ENBL
ENBV
COMM
VOL
VOH
VPOS
HILO
RCLMP
VCM
03199-003
Figure 3. 20-Lead QSOP Pin Configuration (AD8331)
Table 3. 20-Lead QSOP Pin Function Description (AD8331)
Pin No. Mnemonic Description
1 LMD LNA Signal Ground
2 INH LNA Input
3 VPSL LNA 5 V Supply
4 LON LNA Inverting Output
5 LOP LNA Noninverting Output
6 COML LNA Ground
7 VIP VGA Noninverting Input
8 VIN VGA Inverting Input
9 MODE Gain Slope Logic Input
10 GAIN Gain Control Voltage
11 VCM Common Mode Voltage
12 RCLMP Output Clamping Level
13 HILO Gain Range Select (HI or LO)
14 VPOS VGA 5 V Supply
15 VOH Noninverting VGA Output
16 VOL Inverting VGA Output
17 COMM VGA Ground
18 ENBV VGA Enable
19 ENBL LNA Enable
20 COMM VGA Ground
TA = 25°C, VS = 5 V, RL = 500 Ω, RS = RIN = 50 Ω, RFB = 280 Ω, CSH = 22 pF, f = 10 MHz, R
−4.5 dB to +43.5 dB gain (HILO = LO), and differential output voltage, unless otherwise specified.
60
50
40
30
20
GAIN (dB)
10
0
–10
00.20.40.60.81.0 1.1
Figure 7. Gain vs. V
ASCENDING GAIN MODE
DESCENDI NG GAI N MODE
(WHERE AVAILABLE)
GAIN
HILO = HI
HILO = LO
V
(V)
GAIN
and MODE (MODE Available on AC Package)
2.0
1.5
1.0
0.5
0
–0.5
GAIN ERROR (dB)
–1.0
–1.5
–2.0
00.20.40.60.81.0 1.1
Figure 8. Absolute Gain Error vs. V
2.0
1.5
1.0
0.5
0
–0.5
GAIN ERROR (dB)
–1.0
–1.5
–2.0
00.20.40.60.81.0 1.1
Figure 9. Absolute Gain Error vs. V
1MHz
–40°C
10MHz
V
GAIN
30MHz
50MHz
70MHz
V
GAIN
(V)
at Three Temperatures
GAIN
(V)
at Various Frequencies
GAIN
03199-007
+25°C
+85°C
03199-008
03199-009
50
40
30
20
PERCENT OF UNI TS (%)
10
0
–0.5 –0.4 –0. 3 –0.2 –0. 100.1 0. 2 0.3 0.40.5
25
20
15
10
5
0
25
20
15
PERCENT OF UNI TS (%)
10
5
0
Figure 11. Gain Match Histogram for V
50
40
30
20
10
GAIN (dB)
0
–10
–20
100k1M10M100M500M
Figure 12. Frequency Response for Various Values of V
= ∞, CL = 1 pF, VCM pin floating,
CLMP
SAMPLE SIZE = 80 UNITS
V
= 0.5V
GAIN
GAIN ERROR (dB)
Figure 10. Gain Error Histogram
SAMPLE SIZE = 50 UNITS
V
= 0.2V
GAIN
V
= 0.7V
GAIN
0.01
–0.17
–0.15
–0.13
–0.11
–0.09
–0.07
CHANNEL TO CHANNEL G AIN MATCH (d B)
–0.05
V
V
V
V
V
V
FREQUENCY (Hz)
–0.03
GAIN
GAIN
GAIN
GAIN
GAIN
GAIN
–0.01
= 1V
= 0.8V
= 0.6V
= 0.4V
= 0.2V
= 0V
0.03
0.05
0.07
0.09
0.11
= 0.2 V and 0.7 V
GAIN
03199-010
03199-011
0.13
0.15
0.17
0.19
0.21
03199-012
GAIN
Rev. E | Page 12 of 40
Page 13
AD8331/AD8332/AD8334
60
50
40
30
20
GAIN (dB)
10
0
V
V
V
V
V
V
GAIN
GAIN
GAIN
GAIN
GAIN
GAIN
= 1V
= 0.8V
= 0.6V
= 0.4V
= 0.2V
= 0V
0
V
= 1V p-p
OUT
–20
V
= 1.0V
GAIN
V
V
GAIN
GAIN
= 0.7V
= 0.4V
–40
–60
CROSSTALK (d B)
–80
–100
AD8332
AD8334
–10
100k1M10M100M500M
FREQUENCY (Hz)
Figure 13. Frequency Response for Various Values of V
, HILO = HI
GAIN
03199-013
30
V
= 0.5V
GAIN
20
10
0
GAIN (dB)
–10
–20
–30
100k1M10M100M500M
= RS = 75Ω
R
IN
RIN = RS = 100Ω
= RS = 200Ω
R
IN
R
= RS = 500Ω
IN
R
FREQUENCY (Hz)
= RS = 1kΩ
IN
RIN = RS = 50Ω
03199-014
Figure 14. Frequency Response for Various Matched Source Impedances
30
V
= 0.5V
GAIN
R
=
∞
FB
20
10
0
GAIN (dB)
–10
–20
–30
100k1M10M100M500M
Figure 15. Frequency Response, Unterminated LNA, R
FREQUENCY (Hz)
= 50 Ω
S
03199-015
–120
100k1M10M100M
FREQUENCY (Hz)
03199-016
Figure 16. Channel-to-Channel Crosstalk vs.
Frequency for Various Values of V
50
45
40
35
30
1µF
25
COUPLING
20
GROUP DELAY (ns)
15
10
5
0
100k1M10M100M
0.1µF
COUPLING
FREQUENCY (Hz)
GAIN
03199-017
Figure 17. Group Delay vs. Frequency for Two Values of AC Coupling
20
HI GAIN
10
0
–10
–20
20
LO GAIN
10
OFFSET VOLTAGE (mV)
0
–10
–20
00.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
V
GAIN
(V)
T = +85°C
T = +25°C
T = –40°C
T = +85°C
T = +25°C
T = –40°C
03199-018
Figure 18. Representative Differential Output Offset Voltage vs.
at Three Temperatures
V
GAIN
Rev. E | Page 13 of 40
Page 14
AD8331/AD8332/AD8334
R
IN
R
FB
= 100Ω,
= 549Ω
50j
–50j
RIN = 50Ω,
R
FB
= 200Ω,
R
IN
R
FB
R
= 200Ω
IN
R
IN
= 270Ω
= 1.1kΩ
= 500Ω
= 1kΩ
R
IN
f
= 100kHz
RIN = 50Ω
100j
–100j
FB
R
= 100Ω
IN
03199-022
35
30
25
20
15
% TOTAL
10
5
0
SAMPLE SIZE = 100
0.2V < V
49.650.550.450.350.250.150. 049.949.849.7
GAIN
< 0.7V
Figure 19. Gain Scaling Factor Histogram
100
SINGLE ENDED, PIN VOH OR VOL
R
=
∞
L
10
1
OUTPUT IMPEDANCE (Ω)
GAIN SCALING FACTOR
25j
R
= 6kΩ,
IN
R
=
∞
FB
0Ω17Ω
RIN = 75Ω,
R
= 412Ω
FB
03199-019
–25j
Figure 22. Smith Chart, S11 vs. Frequency,
0.1 MHz to 200 MHz for Various Values of R
20
VIN = 10mV p-p
15
10
5
0
GAIN (dB)
–5
0.1
100k100M10M1M
Figure 20. Output Impedance vs. Frequency
10k
RFB = 6.65kΩ, CSH = 0pF
1k
100
INPUT IMPEDANCE (Ω)
10
100k100M10M1M
R
R
= 1.1kΩ, CSH = 1.2pF
FB
= 412Ω, CSH = 12pF
R
FB
Figure 21. LNA Input Impedance vs.
Frequency for Various Values of R
FREQUENCY (Hz)
= 3.01kΩ, CSH = 0pF
FB
R
= 549Ω, CSH = 8.2pF
FB
FREQUENCY (Hz)
R
=∞, CSH = 0pF
FB
RFB = 270Ω, CSH = 22pF
and C
FB
–10
= 75Ω
R
03199-020
–15
100k1M10M100M500M
FREQUENCY (Hz)
Figure 23. LNA Frequency Response, Single Ended, for Various Values of R
IN
03199-023
IN
20
15
RFB =
∞
10
5
0
GAIN (dB)
–5
–10
03199-021
–15
100k1M10M100M500M
FREQUENCY (Hz)
03199-024
Figure 24. Frequency Response for Unterminated LNA, Single Ended
SH
Rev. E | Page 14 of 40
Page 15
AD8331/AD8332/AD8334
500
f
= 10MHz
400
300
200
100
OUTPUT REFERRED NOISE (nV/ Hz)
0
LO GAIN
HI GAIN
00.20.40.60.81.0
AD8332
AD8334
AD8331
V
GAIN
(V)
Figure 25. Output-Referred Noise vs. V
2.5
RS = 0, RFB =∞, V
HILO = LO OR HI
2.0
GAIN
= 1V,
03199-025
GAIN
1.00
RS = 0, RFB =∞,
V
= 1V, f = 10M Hz
GAIN
0.95
0.90
0.85
0.80
0.75
0.70
0.65
INPUT NOISE (nV/ Hz)
0.60
0.55
0.50
–50–30–101030507090
TEMPERATURE ( °C)
Figure 28. Short-Circuit, Input-Referred Noise vs. Temperature
10
f = 5MHz, RFB =∞,
= 1V
V
GAIN
03199-028
1.5
INPUT NOISE (nV/ Hz)
1.0
0.5
100k1M10M100M
FREQUENCY (Hz)
Figure 26. Short-Circuit, Input-Referred Noise vs. Frequency
100
10
1
INPUT NOISE (nV/ Hz)
0.1
00.20.40.60.81.0
RS = 0, RFB =∞,
HILO = LO OR HI, f = 10MHz
V
(V)
GAIN
Figure 27. Short-Circuit, Input-Referred Noise vs. V
GAIN
1
INPUT NOISE (nV/ Hz)
03199-026
0.1
1101001k
Figure 29. Input-Referred Noise vs. R
7
INCLUDES NOISE OF VGA
6
5
4
3
NOISE FI GURE (dB)
2
1
03199-027
SIMULATION
0
501001k
Figure 30. Noise Figure vs. R
R
THERMAL NOISE
S
ALONE
SOURCE RESIST ANCE (Ω)
R
= 50Ω
IN
R
= 75Ω
IN
= 100Ω
R
IN
= 200Ω
R
IN
R
=
∞
FB
SOURCE RESIST ANCE (Ω)
for Various Values of R
S
03199-029
S
03199-030
IN
Rev. E | Page 15 of 40
Page 16
AD8331/AD8332/AD8334
–
–
–
35
30
25
20
15
NOISE FI GURE (dB)
10
5
HILO = L O, R
HILO = HI, R
HILO = LO, R
HILO = HI, RIN = 50Ω
=
∞
FB
=
∞
IN
IN
f = 10MHz, RS = 50ΩPREAMP LIMITED
= 50Ω
HARMONIC DIST ORTIO N (dBc)
30
–40
–50
–60
–70
–80
f = 10MHz,
= 1V p-p
V
OUT
HILO = HI , HD2
HILO = HI , HD3
HILO = L O, HD2
HILO = L O, HD3
0
0 0.10.20.30.40.50.60.70.80.91.01.1
Figure 31. Noise Figure vs. V
30
25
20
15
10
NOISE FI GURE (dB)
HILO = LO, R
5
0
1015202530354045505560
= 50Ω
IN
HILO = L O, R
V
(V)
GAIN
f = 10MHz, RS = 50Ω
HILO = HI, RIN = 50Ω
HILO = HI, R
=
∞
FB
GAIN (dB)
GAIN
=
∞
FB
Figure 32. Noise Figure vs. Gain
0
G = 30dB,
V
= 1V p-p
OUT
–10
–20
–30
–40
–50
–60
–70
HARMONIC DISTORTION (dBc)
–80
–90
1M10M100
FREQUENCY (Hz)
HILO = L O, HD2
HILO = LO , HD3
HILO = HI, HD2
HILO = HI, HD3
Figure 33. Harmonic Distortion vs. Frequency
03199-031
03199-032
03199-033
–90
0200018001600140012001000800600400200
Figure 34. Harmonic Distortion vs. R
40
f = 10MHz,
= 1V p-p
V
OUT
–50
–60
–70
HARMONIC DIST ORTIO N (dBc)
–80
–90
0 10203040
Figure 35. Harmonic Distortion vs. C
20
f = 10MHz,
GAIN = 30d B
–40
–60
HILO = HI, HD2
–80
HARMONIC DIST ORTIO N (dBc)
–100
01234
R
LOAD
HILO = LO, HD3
HILO = HI, HD2
C
LOAD
HILO = L O, HD2
V
(V p-p)
OUT
(Ω)
LOAD
HILO = L O, HD2
HILO = HI, HD3
(pF)
LOAD
HILO = LO, HD3
HILO = HI , HD3
Figure 36. Harmonic Distortion vs. Differential Output Voltage
03199-034
03199-035
50
03199-036
Rev. E | Page 16 of 40
Page 17
AD8331/AD8332/AD8334
0
V
= 1V p-p
OUT
–20
INPUT RANGE
LIMITED WHEN
–40
HILO = LO
–60
–80
DISTORTION (dBc)
–100
–120
00.1 0. 2 0.3 0.4 0. 5 0.6 0.70.8 0. 9 1.0
HILO = HI, HD3
HILO = L O, HD2
V
(V)
GAIN
Figure 37. Harmonic Distortion vs. V
HILO = L O, HD3
HILO = HI, HD2
, f = 1 MHz
GAIN
0
V
= 1V p-p
OUT
–20
INPUT RANGE
LIMITED WHEN
–40
HILO = LO
–60
HILO = LO, HD2
HILO = L O, HD3
03199-037
0
V
= 1V p-p COMPOSITE (
OUT
G = 30dB
–10
–20
–30
–40
–50
IMD3 (dBc)
–60
–70
–80
–90
1M10M100M
f
+
f
)
1
2
FREQUENCY (Hz)
HILO = LO
HILO = HI
Figure 40. IMD3 vs. Frequency
40
10MHz HILO = HI
35
30
25
20
1MHz HILO = HI
1MHz HILO = LO
10MHz HILO = LO
03199-040
–80
DISTORTION (dBc)
HILO = HI , HD3
–100
–120
00.1 0. 2 0.3 0.4 0. 5 0.6 0.70.8 0. 9 1.0
V
GAIN
(V)
Figure 38. Harmonic Distortion vs. V
HILO = HI, HD2
, f = 10 MHz
GAIN
10
f = 10MHz
0
–10
HILO = HI
–20
INPUT POWER (dBm)
–30
–40
00.1 0. 2 0.3 0.4 0. 5 0.6 0.70.8 0. 9 1.0
V
GAIN
HILO = LO
(V)
Figure 39. Input 1 dB Compression vs. V
GAIN
15
OUTPUT I P3 (dBm)
10
5
V
= 1V p-p COMPOSITE (
03199-038
0
010.90. 80.70.60. 50. 40.30.20. 1
OUT
Figure 41. Output Third-Order Intercept vs. V
2mV
100
90
10
0
50mV10ns
03199-039
Figure 42. Small Signal Pulse Response, G = 30 dB,
V
GAIN
f
+
f
)
1
2
(V)
GAIN
03199-041
.0
3199-042
Top: Input, Bottom: Output Voltage, HILO = HI or LO
Rev. E | Page 17 of 40
Page 18
AD8331/AD8332/AD8334
(V p-p)
OUT
V
(V)
OUT
V
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
–1
0
0545403530252015105
4
3
2
1
0
Figure 46. Clamp Level vs. R
G = 40dB
INPUT
HILO = HI
R
R
CLMP
R
CLMP
R
CLMP
HILO = LO
(kΩ)
CLMP
= 48.1kΩ
R
CLMP
= 7.15kΩ
= 2.67kΩ
03199-046
0
CLMP
= 16.5kΩ
20mV
100
90
10
0
500mV10ns
Figure 43. Large Signal Pulse Response, G = 30 dB,
HILO = HI or LO, Top: Input, Bottom: Output Voltage
2
G = 30dB
1
INPUT
(V)
0
OUT
V
CL = 0pF
CL = 10pF
CL = 22pF
CL = 47pF
3199-043
–1
INPUT IS NOT TO SCALE
–2
–5050403020100–10–20–30–40
TIME (ns)
Figure 44. Large Signal Pulse Response for Various Capacitive Loads,
= 0 pF, 10 pF, 20 pF, 50 pF
C
L
500mV
200mV400ns
3199-045
Figure 45. Pin GAIN Transient Response,
Top: V
, Bottom: Output Voltage
GAIN
–2
–3
CLMP
03199-047
3199-048
03199-044
–4
–30–20–100 1020304050607080
TIME (ns)
Figure 47. Clamp Level Pulse Response for 4 Values of R
200mV
100
90
10
0
100ns
Figure 48. LNA Overdrive Recovery, V
= 0.27 V VGA Output Shown
V
GAIN
0.05 V p-p to 1 V p-p Burst,
INH
Rev. E | Page 18 of 40
Page 19
AD8331/AD8332/AD8334
1V
100
90
10
0
Figure 49. VGA Overdrive Recovery, V
= 1 V VGA Output Shown Attenuated by 24 dB
V
GAIN
1V
100
90
10
0
Figure 50. VGA Overdrive Recovery, V
= 1 V VGA Output Shown Attenuated by 24 dB
V
GAIN
2V
200mV1ms
Figure 51. Enable Response, Top: V
ENB
INH
INH
, Bottom: VB
100ns
3199-049
4 mV p-p to 70 mV p-p Burst,
100ns
3199-050
4 mV p-p to 275 mV p-p Burst,
3199-051
, V
= 30 mV p-p
OUT
INH
2V
1V1ms
Figure 52. Enable Response, Large Signal,
, Bottom: V
B
Top: V
ENB
0
–10
–20
–30
–40
PSRR (dB)
–50
–60
–70
–80
100k1M10M100M
VPSV, V
GAIN
, V
= 150 mV p-p
OUT
INH
VPS1, V
= 0.5V
FREQUENCY (Hz)
GAIN
VPS1, V
= 0.5V
GAIN
Figure 53. PSRR vs. Frequency (No Bypass Capacitor)
140
V
= 0.5V
GAIN
QUIESCENT S UPPLY CURRENT (mA)
130
120
110
100
90
80
70
60
50
40
30
20
–40100806040200–20
AD8334
AD8332
AD8331
TEMPERATURE ( °C)
Figure 54. Quiescent Supply Current vs. Temperature
= 0V
3199-052
03199-053
03199-054
Rev. E | Page 19 of 40
Page 20
AD8331/AD8332/AD8334
TEST CIRCUITS
MEASUREMENT CONSIDERATIONS
Figure 55 through Figure 68 show typical measurement
configurations and proper interface values for measurements
with 50 Ω conditions.
FERRITE
BEAD
0.1µF
120nH
22pF
0.1µF
Figure 55. Gain and Bandwidth Measurements
Short-circuit input noise measurements are made using
The input-referred noise level is determined by dividing the
output noise by the numerical gain between Point A and Point B
and accounting for the noise floor of the spectrum analyzer.
The gain should be measured at each frequency of interest and
with low signal levels because a 50 Ω load is driven directly. The
generator is removed when noise measurements are made.
NETWO RK ANALYZE R
50Ω50Ω
18nF
270Ω
INH
LMD
NETWORK ANALYZER
0.1µF
DUT
0.1µF
INOUT
237Ω
237Ω
28Ω
28Ω
1:1
03199-055
Figure 62.
50Ω50Ω
INOUT
10kΩ
18nF
FERRITE
10kΩ
BEAD
120nH
22pF
0.1µF
0.1µF
INH
LMD
DUT
0.1µF
0.1µF
237Ω
237Ω
28Ω
28Ω
1:1
03199-056
Figure 56. Frequency Response for Various Matched Source Impedances
FERRITE
BEAD
120nH
22pF
Figure 57. Frequency Response for Unterminated LNA, R
NETWORK ANALYZER
0.1µF
INH
LMD
0.1µF
DUT
50Ω50Ω
0.1µF
0.1µF
INOUT
237Ω
237Ω
28Ω
28Ω
1:1
03199-057
= 50 Ω
S
Rev. E | Page 20 of 40
Page 21
AD8331/AD8332/AD8334
R
R
K
NETWORK ANALY ZE
50Ω50Ω
INOUT
10kΩ
18nF
LNA
0.1µF
18nF
INH
LMD
0.1µF
AND
10µF
0.1µF
AND
10µF
270Ω
DUT
0.1µF
0.1µF
VGA
237Ω
237Ω
237Ω
237Ω
28Ω
28Ω
28Ω
28Ω
1:1
03199-058
50Ω1:1
03199-059
FERRITE
BEAD
120nH
22pF
0.1µF
AND
10µF
INH
LMD
0.1µF
Figure 58. Group Delay vs. Frequency for Two Values of AC Coupling
NETWORK
ANALYZER
FERRITE
BEAD
OUT
120nH
50Ω
Figure 59. LNA Input Impedance vs. Frequency in Standard and Smith Chart (S11) Formats
0.1µF
22pF
NETWORK ANALY ZE
50Ω50Ω
INOUT
LNA
0.1µF
0.1µF
0.1µF
VGA
0.1µF
0.1µF
237Ω
237Ω
28Ω
28Ω
1:1
03199-060
FERRITE
BEAD
120nH
22pF
0.1µF
INH
LMD
0.1µF
Figure 60. Frequency Response for Unterminated LNA, Single Ended
NETWOR
FERRITE
BEAD
120nH
22pF
18nF
270Ω
0.1µF
0.1µF
INH
LMD
DUT
0.1µF
0.1µF
237Ω
237Ω
1:1
28Ω
28Ω
Figure 61. Short-Circuit, Input-Referred Noise
IN
ANALYZER
50Ω
03199-061
Rev. E | Page 21 of 40
Page 22
AD8331/AD8332/AD8334
–
SPECTRUM
ANALYZER
0.1µF
0.1µF
BA
50Ω
IN
1:1
3199-062
SPECTRUM
ANALYZER
22pF
GAIN
INH
LMD
0.1µF
49.9Ω
50Ω
SIGNAL GE NERATOR
TO MEASURE G AIN
DISCONNECT F OR
NOISE MEAS UREMENT
0.1µF
1Ω
FERRITE
BEAD
120nH
Figure 62. Noise Figure
18nF
270Ω
50Ω
LPF
SIGNAL
GENERATOR
–6dB
0.1µF
22pF
0.1µF
AD8332
INH
LMD
0.1µF
0.1µF
1kΩ
1kΩ
28Ω
28Ω
–6dB
1:1
50Ω
IN
03199-063
Figure 63. Harmonic Distortion vs. Load Resistance
SPECTRUM
IN
ANALYZER
50Ω
03199-064
50Ω
LPF
SIGNAL
GENERATOR
–6dB
0.1µF
22pF
0.1µF
18nF
AD8332
INH
LMD
270Ω
0.1µF
0.1µF
237Ω
237Ω
28Ω
28Ω
–6dB
1:1
Figure 64. Harmonic Distortion vs. Load Capacitance
50Ω
50Ω
SIGNAL
GENERATORS
+22dB
+22dB
6dB
–6dB
FERRITE
COMBINER
–6dB
18nF
270Ω
BEAD
0.1µF
120nH
22pF
INH
LMD
0.1µF
Figure 65. IMD3 vs. Frequency
DUT
0.1µF
0.1µF
237Ω
237Ω
28Ω
28Ω
1:1
–6dB
SPECTRUM
ANALYZER
INPUT
50Ω
03199-065
Rev. E | Page 22 of 40
Page 23
AD8331/AD8332/AD8334
E
18nF
INH
LMD
270Ω
DUT
0.1µF
0.1µF
237Ω
28Ω
237Ω
28Ω
50Ω
FERRITE
BEAD
120nH
22pF
0.1µF
0.1µF
OSCILLOSCOP
50Ω
IN
1:1
3199-066
Figure 66. Pulse Response Measurements
OSCILLOSCOPE
18nF
INH
LMD
270Ω
0.1µF
DUT
0.1µF
TO PIN GAIN
OR ENxx
255Ω
255Ω
DIFF
PROBE
GENERATOR
NETWORK
ANALYZER
CH1
PULSE
50Ω
CH2
9.5dB
03199-067
FERRITE
BEAD
120nH
22pF
50Ω
RF
SIGNAL
GENERATOR
0.1µF
0.1µF
Figure 67. GAIN and Enable Transient Response
50Ω
PROBE
DIFF
INOUT
PROBE
POWER
3199-068
FERRITE
BEAD
120nH
22pF
50Ω
RF
SIGNAL
GENERATOR
0.1µF
18nF
0.1µF
270Ω
INH
LMD
TO POWER
PIN(S)
0.1µF
DUT
0.1µF
50Ω
255Ω
255Ω
Figure 68. PSRR vs. Frequency
Rev. E | Page 23 of 40
Page 24
AD8331/AD8332/AD8334
VINV
O
VCM
V
V
V
V
THEORY OF OPERATION
OVERVIEW
The following discussion applies to all part numbers. Figure 69,
Figure 70, and Figure 71 are functional block diagrams of the
AD8331, AD8332, and AD8334, respectively.
INH
LMD
INH1
LMD1
LMD2
INH2
IPLOPLON
V
MID
+
LNA
–
LNA
BIAS
–
ATTENUAT OR
–48dB
+
VGA BIAS AND
INTERPOLATOR
AD8331
ENBL
ENBVGAIN
Figure 69. AD8331 Functional Block Diagram
+19dB
LNA 1
LNA 2
LON1 LOP1
BIAS
(V
MID
)
IP1
IN1
–
+
+
–
ATTENUATOR
–48dB
VGA BIAS AND
INTERPOLAT OR
ATTENUATOR
–48dB
AD8332
LON2 LOP2
VIP2
VIN2
ENB
Figure 70. AD8332 Functional Block Diagram
GAIN
INT
CM1
V
MID
V
MID
VCM2
GAIN
INT
21dB
21dB
3.5dB/
15.5dB
CLAMP
3.5dB/
15.5dB
CLAMP
PA21dB
PA1
PA2
HIL
VOH
VOL
RCLMP
MODE
HILO
VOH1
VOL1
GAIN
VOL2
VOH2
RCLMP
LON1 LOP1VIP1VIN1 EN12
INH1
LMD1
LMD2
INH2
LON2
LOP2
VIP2
VIN2
MODE
VIN3
VIP3
03199-069
LOP3
LON3
INH3
LMD3
LMD4
INH4
LNA 1
LNA 2
LNA 3
LNA 4
LNA
BIAS
LNA
BIAS
–
ATTENUATO R
–48dB
+
VGA BIAS AND
INTERPOL ATOR
+
ATTENUATO R
–48dB
–
GAIN UP/
DOWN
–
ATTENUATO R
–48dB
+
VGA BIAS AND
INTERPOL ATOR
+
ATTENUATO R
–48dB
–
AD8334
Figure 71. AD8334 Functional Block Diagram
Each channel contains an LNA that provides user-adjustable
input impedance termination, a differential X-AMP VGA, and a
programmable gain postamplifier with adjustable output voltage
limiting.
3199-070
external components.
INH
LMD
Figure 72 shows a simplified block diagram with
VIN
48dB
ATTENUATOR
VIP
BIAS AN
INTERPOLATO
PRE-AMPLIFIER
19dB
+
LNA
–
BIAS
)
(V
MID
LON
LOP
CM1
V
MID1
V
MID2
V
MID3
V
MID4
VCM4EN34VIN4VIP4LON4 LOP4
SIGNAL PATH
V
VCM
D
R
21dB
GAIN
INT
21dB
21dB
GAIN
INT
21dB
21dB
MID
GAIN
INTERFAC E
CLAMP
PA1
PA2
PA3
PA4
CLAMP34
3.5dB/15. 5dB
HILO
POST-
AMP
CLAMP
CLMP12
VOH1
VOL1
GAIN12
HILO
VOL2
VOH2
VCM2
VCM3
VOH3
VOL3
GAIN34
VOL4
VOH4
CLMP34
03199-071
VOH
VOL
RCLMP
GAIN
03199-072
Figure 72. Simplified Block Diagram
Rev. E | Page 24 of 40
Page 25
AD8331/AD8332/AD8334
The linear-in-dB gain-control interface is trimmed for slope
and absolute accuracy. The gain range is 48 dB, extending from
−4.5 dB to +43.5 dB in HI gain and +7.5 dB to +55.5 dB in LO
gain mode. The slope of the gain control interface is 50 dB/V,
and the gain control range is 40 mV to 1 V. Equation 1 and
Equation 2 are the expressions for gain.
GAIN (dB) = 50 (dB/V) × V
− 6.5 dB, (HILO = LO) (1)
GAIN
or
GAIN (dB) = 50 (dB/V) × V
The ideal gain characteristics are shown in
60
50
40
30
+ 5.5 dB, (HILO = LO) (2)
GAIN
Figure 73.
HILO = HI
LOW NOISE AMPLIFIER (LNA)
Good noise performance relies on a proprietary ultralow noise
preamplifier at the beginning of the signal chain, which minimizes
the noise contribution in the following VGA. Active impedance
control optimizes noise performance for applications that benefit
from input matching.
A simplified schematic of the LNA is shown in
is capacitively coupled to the source. An on-chip bias generator
establishes dc input bias voltages of 3.25 V and centers the
output common-mode levels at 2.5 V. A Capacitor C
same value as the Input Coupling Capacitor C
from the LMD pin to ground.
C
FB
VPOSLOP
I
0
Figure 74. INH
is connected
INH
R
FB
I
0
LON
LMD
of the
20
GAIN (dB)
10
0
–10
00.20.40.60.81.0 1.1
ASCENDING GAIN MODE
DESCENDING GAIN MODE
(WHERE AVAILABLE)
Figure 73. Ideal Gain Control Characteristics
HILO = LO
V
(V)
GAIN
03199-073
The gain slope is negative with the MODE pulled high (where
available):
GAIN (dB) = −50 (dB/V) × V
+ 45.5 dB, (HILO = LO) (3)
GAIN
or
GAIN (dB) = −50 (dB/V) × V
+ 57.5 dB, (HILO = HI) (4)
GAIN
The LNA converts a single-ended input to a differential output
with a voltage gain of 19 dB. If only one output is used, the gain
is 13 dB. The inverting output is used for active input impedance
termination. Each of the LNA outputs is capacitively coupled to
a VGA input. The VGA consists of an attenuator with a range of
48 dB followed by an amplifier with 21 dB of gain for a net gain
range of −27 dB to +21 dB. The X-AMP gain-interpolation
technique results in low gain error and uniform bandwidth, and
differential signal paths minimize distortion.
The final stage is a logic programmable amplifier with gains of
3.5 dB or 15.5 dB. The LO and HI gain modes are optimized for
12-bit and 10-bit ADC applications, in terms of output-referred
noise and absolute gain range. Output voltage limiting can be
programmed by the user.
C
INH
R
S
INH
C
SH
Figure 74. Simplified LNA Schematic
Q1Q2
I
0I0
LMD
C
LMD
03199-074
The LNA supports differential output voltages as high as 5 V p-p
with positive and negative excursions of ±1.25 V, about a
common-mode voltage of 2.5 V. Because the differential gain
magnitude is 9, the maximum input signal before saturation is
±275 mV or +550 mV p-p. Overload protection ensures quick
recovery time from large input voltages. Because the inputs are
capacitively coupled to a bias voltage near midsupply, very large
inputs can be handled without interacting with the ESD protection.
Low value feedback resistors and the current-driving capability
of the output stage allow the LNA to achieve a low input-referred
voltage noise of 0.74 nV/√Hz. This is achieved with a current
consumption of only 11 mA per channel (55 mW). On-chip
resistor matching results in precise single-ended gains of 4.5×
(9× differential), critical for accurate impedance control. The
use of a fully differential topology and negative feedback
minimizes distortion. Low HD2 is particularly important in
second harmonic ultrasound imaging applications. Differential
signaling enables smaller swings at each output, further
reducing third-order distortion.
Rev. E | Page 25 of 40
Page 26
AD8331/AD8332/AD8334
Active Impedance Matching
The LNA supports active impedance matching through an external
shunt feedback resistor from Pin LON to Pin INH. The input
resistance, R
, is given by Equation 5, where A is the single-
IN
ended gain of 4.5, and 6 kΩ is the unterminated input impedance.
R
×
R
R
C
is needed in series with RFB because the dc levels at Pin LON
FB
FB
=
IN
1
k6
A
+
and Pin INH are unequal. Expressions for choosing R
of R
and for choosing CFB are found in the Applicat ions section.
IN
C
and the ferrite bead enhance stability at higher frequencies
SH
k6
FB
=
k33
(5)
R
+
FB
in terms
FB
where the loop gain is diminished and prevent peaking. Frequency
response plots of the LNA are shown in
Figure 23 and Figure 24.
The bandwidth is approximately 130 MHz for matched input
impedances of 50 Ω to 200 Ω and declines at higher source
impedances. The unterminated bandwidth (when R
= ∞) is
FB
approximately 80 MHz.
Each output can drive external loads as low as 100 Ω in addition
to the 100 Ω input impedance of the VGA (200 Ω differential).
Capacitive loading up to 10 pF is permissible. All loads should
be ac-coupled. Typically, Pin LOP output is used as a singleended driver for auxiliary circuits, such as those used for
Doppler ultrasound imaging, and Pin LON drives R
.
FB
Alternatively, a differential external circuit can be driven from
the two outputs in addition to the active feedback termination.
In both cases, important stability considerations discussed in
the
Applications section should be carefully observed.
The impedance at each LNA output is 5 Ω. A 0.4 dB reduction
in open-circuit gain results when driving the VGA, and 0.8 dB
with an additional 100 Ω load at the output. The differential
gain of the LNA is 6 dB higher. If the load is less than 200 Ω on
either side, a compensating load is recommended on the
opposite output.
LNA Noise
The input-referred voltage noise sets an important limit on
system performance. The short-circuit input voltage noise of
the LNA is 0.74 nV/√Hz or 0.82 nV/√Hz (at maximum gain),
including the VGA noise. The open-circuit current noise is
2.5 pA/√Hz. These measurements, taken without a feedback
resistor, provide the basis for calculating the input noise and
noise figure performance of the configurations in
Figure 75.
Figure 76 and Figure 77 are simulations extracted from these
results, and the 4.1 dB NF measurement with the input actively
matched to a 50 Ω source. Unterminated (R
= ∞) operation
FB
exhibits the lowest equivalent input noise and noise figure.
Figure 76 shows the noise figure vs. source resistance, rising at
low R
, where the LNA voltage noise is large compared to the
S
source noise, and again at high R
due to current noise. The
S
VGA’s input-referred voltage noise of 2.7 nV/√Hz is included in
all of the curves.
+
V
IN
–
+
V
IN
–
+
V
IN
–
Figure 75. Input Configurations
7
6
5
4
3
NOISE FI GURE (dB)
2
1
SIMULATION
0
501001k
Figure 76. Noise Figure vs. R
Active Matched and Unterminated Inputs
7
INCLUDES NOISE OF VGA
6
5
4
3
NOISE FI GURE (dB)
2
1
SIMULATION
0
501001k
Figure 77. Noise Figure vs. R
UNTERMINATED
R
IN
R
S
V
OUT
RESISTIVE TERMINATION
R
IN
R
S
R
S
ACTIVE IM PEDANCE MATCH
R
R
S
RIN=
RESISTIVE TERMINAT ION
= 50Ω
R
IN
R
= 75Ω
IN
R
= 100Ω
IN
= 200Ω
R
IN
R
FB
for Various Fixed Values of RIN, Actively Matched
S
R
IN
R
FB
1 + 4.5
INCLUDES NOISE OF VGA
(R
= RIN)
S
ACTIVE IM PEDANCE MATCH
UNTERMINATED
RS (Ω)
=
∞
RS (Ω)
V
FB
V
for Resistive,
S
OUT
OUT
03199-075
03199-076
03199-077
Rev. E | Page 26 of 40
Page 27
AD8331/AD8332/AD8334
The primary purpose of input impedance matching is to
improve the system transient response. With resistive termination,
the input noise increases due to the thermal noise of the
matching resistor and the increased contribution of the LNA’s
input voltage noise generator. With active impedance matching,
however, the contributions of both are smaller than they would
be for resistive termination by a factor of 1/(1 + LNA Gain).
Figure 76 shows their relative noise figure (NF) performance. In
this graph, the input impedance was swept with R
to preserve
S
the match at each point. The noise figures for a source impedance
of 50  are 7.1 dB, 4.1 dB, and 2.5 dB, respectively, for the
resistive, active, and unterminated configurations. The noise
figures for 200  are 4.6 dB, 2.0 dB, and 1.0 dB, respectively.
Figure 77 is a plot of the NF vs. RS for various values of RIN,
which is helpful for design purposes. The plateau in the NF for
actively matched inputs mitigates source impedance variations.
For comparison purposes, a preamp with a gain of 19 dB and
noise spectral density of 1.0 nV/√Hz, combined with a VGA
with 3.75 nV/√Hz, yields a noise figure degradation of
approximately 1.5 dB (for most input impedances), significantly
worse than the AD8332 performance.
The equivalent input noise of the LNA is the same for singleended and differential output applications. The LNA noise figure
improves to 3.5 dB at 50 Ω without VGA noise, but this is
exclusive of noise contributions from other external circuits
connected to LOP. A series output resistor is usually
recommended for stability purposes when driving external
circuits on a separate board (see the
Applications section). In
low noise applications, a ferrite bead is even more desirable.
VARIABLE GAIN AMPLIFIER
The differential X-AMP VGA provides precise input
attenuation and interpolation. It has a low input-referred noise
of 2.7 nV/√Hz and excellent gain linearity. A simplified block
diagram is shown in
GAIN
g
m
VIP
VIN
Figure 78.
GAIN INTERPOLATOR
(BOTH CHANNEL S)
6dB
R
2R
Figure 78. Simplified VGA Schematic
48dB
POSTAMP
+
–
POSTAMP
03199-078
X-AMP VGA
The input of the VGA is a differential R-2R ladder attenuator
network with 6 dB steps per stage and a net input impedance of
200 Ω differential. The ladder is driven by a fully differential
input signal from the LNA and is not intended for single-ended
operation. LNA outputs are ac-coupled to reduce offset and
isolate their common-mode voltage. The VGA inputs are biased
through the ladder’s center tap connection to VCM, which is
typically set to 2.5 V and is bypassed externally to provide a
clean ac ground.
The signal level at successive stages in the input attenuator
falls from 0 dB to −48 dB in 6 dB steps. The input stages of the
X-AMP are distributed along the ladder, and a biasing interpolator,
controlled by the gain interface, determines the input tap point.
With overlapping bias currents, signals from successive taps
merge to provide a smooth attenuation range from 0 dB to
−48 dB. This circuit technique results in excellent, linear-in-dB
gain law conformance and low distortion levels and deviates
±0.2 dB or less from the ideal. The gain slope is monotonic with
respect to the control voltage and is stable with variations in
process, temperature, and supply.
The X-AMP inputs are part of a gain-of-12 feedback amplifier
that completes the VGA. Its bandwidth is 150 MHz. The input
stage is designed to reduce feedthrough to the output and to
ensure excellent frequency response uniformity across gain
setting (see
Figure 12 and Figure 13).
Gain Control
Position along the VGA attenuator is controlled by a singleended analog control voltage, V
, with an input range of
GAIN
40 mV to 1.0 V. The gain control scaling is trimmed to a slope
of 50 dB/V (20 mV/dB). Values of V
beyond the control
GAIN
range saturate to minimum or maximum gain values. Both
channels of the AD8332 are controlled from a single gain
interface to preserve matching. Gain can be calculated using
Equation 1 and Equation 2.
Gain accuracy is very good because both the scaling factor and
absolute gain are factory trimmed. The overall accuracy relative
to the theoretical gain expression is ±1 dB for variations in
temperature, process, supply voltage, interpolator gain ripple,
trim errors, and tester limits. The gain error relative to a best-fit
line for a given set of conditions is typically ±0.2 dB. Gain
matching between channels is better than 0.1 dB (
Figure 11
shows gain errors in the center of the control range). When
V
< 0.1 or > 0.95, gain errors are slightly greater.
GAIN
Rev. E | Page 27 of 40
Page 28
AD8331/AD8332/AD8334
The gain slope can be inverted, as shown in Figure 73 (available
in most versions). The gain drops with a slope of −50 dB/V
across the gain control range from maximum to minimum gain.
This slope is useful in applications, such as automatic gain
control, where the control voltage is proportional to the
measured output signal amplitude. The inverse gain mode is
selected by setting the MODE pin HI.
Gain control response time is less than 750 ns to settle within 10%
of the final value for a change from minimum to maximum gain.
VGA Noise
In a typical application, a VGA compresses a wide dynamic
range input signal to within the input span of an ADC. While
the input-referred noise of the LNA limits the minimum
resolvable input signal, the output-referred noise, which
depends primarily on the VGA, limits the maximum
instantaneous dynamic range that can be processed at any one
particular gain control voltage. This limit is set in accordance
with the quantization noise floor of the ADC.
Output and input-referred noise as a function of V
plotted in
Figure 25 and Figure 27 for the short-circuited input
GAIN
are
conditions. The input noise voltage is simply equal to the output
noise divided by the measured gain at each point in the control
range.
The output-referred noise is flat over most of the gain range,
because it is dominated by the fixed output-referred noise of the
VGA. Values are 48 nV/√Hz in LO gain mode and 178 nV/√Hz
in HI gain mode. At the high end of the gain control range, the
noise of the LNA and source prevail. The input-referred noise
reaches its minimum value near the maximum gain control
voltage, where the input-referred contribution of the VGA
becomes very small.
At lower gains, the input-referred noise, and thus noise figure,
increases as the gain decreases. The instantaneous dynamic
range of the system is not lost, however, because the input
capacity increases with it. The contribution of the ADC noise
floor has the same dependence as well. The important
relationship is the magnitude of the VGA output noise floor
relative to that of the ADC.
With its low output-referred noise levels, these devices ideally
drive low voltage ADCs. The converter noise floor drops 12 dB
for every 2 bits of resolution and drops at lower input full-scale
voltages and higher sampling rates. ADC quantization noise is
discussed in the
Applications section.
The preceding noise performance discussion applies to a
differential VGA output signal. Although the LNA noise
performance is the same in single-ended and differential
applications, the VGA performance is not. The noise of the
VGA is significantly higher in single-ended usage, because the
contribution of its bias noise is designed to cancel in the differential
signal. A transformer can be used with single-ended applications
when low noise is desired.
Gain control noise is a concern in very low noise applications.
Thermal noise in the gain control interface can modulate the
channel gain. The resultant noise is proportional to the output
signal level and usually only evident when a large signal is
present. Its effect is observable only in LO gain mode, where the
noise floor is substantially lower. The gain interface includes an
on-chip noise filter, which reduces this effect significantly at
frequencies above 5 MHz. Care should be taken to minimize
noise impinging at the GAIN input. An external RC filter can be
used to remove V
source noise. The filter bandwidth should be
GAIN
sufficient to accommodate the desired control bandwidth.
Common-Mode Biasing
An internal bias network connected to a midsupply voltage
establishes common-mode voltages in the VGA and postamp.
An externally bypassed buffer maintains the voltage. The bypass
capacitors form an important ac ground connection, because
the VCM network makes a number of important connections
internally, including the center tap of the VGA’s differential
input attenuator, the feedback network of the VGA’s fixed gain
amplifier, and the feedback network of the postamplifier in both
gain settings. For best results, use a 1 nF and a 0.1 µF capacitor
in parallel, with the 1 nF nearest to the VCM pin. Separate
VCM pins are provided for each channel. For dc-coupling to a
3 V ADC, the output common-mode voltage is adjusted to
1.5 V by biasing the VCM pin.
POSTAMPLIFIER
The final stage has a selectable gain of 3.5 dB (×1.5) or 15.5 dB
(×6), set by the logic pin, HILO.
diagram.
Gm2
+
Gm1
VCM
–
F1
Gm2
Gm1
Figure 79. Postamplifier Block Diagram
Separate feedback attenuators implement the two gain settings.
These are selected in conjunction with an appropriately scaled
input stage to maintain a constant 3 dB bandwidth between the
two gain modes (~150 MHz). The slew rate is 1200 V/µs in HI
gain mode and 300 V/µs in LO gain mode. The feedback
networks for HI and LO gain modes are factory trimmed to
adjust the absolute gains of each channel.
Figure 79 is a simplified block
F2
VOH
VOL
3199-079
Rev. E | Page 28 of 40
Page 29
AD8331/AD8332/AD8334
V
Noise
The topology of the postamplifier provides constant inputreferred noise with the two gain settings and variable
output-referred noise. The output-referred noise in HI gain
mode increases (with gain) by four. This setting is recommended
when driving converters with higher noise floors. The extra gain
boosts the output signal levels and noise floor appropriately. When
driving circuits with lower input noise floors, the LO gain mode
optimizes the output dynamic range.
Although the quantization noise floor of an ADC depends on a
number of factors, the 48 nV/√Hz and 178 nV/√Hz levels are
well suited to the average requirements of most 12-bit and 10-bit
converters, respectively. An additional technique, described in
the
Applications section, can extend the noise floor even lower
for possible use with 14-bit ADCs.
Output Clamping
Outputs are internally limited to a level of 4.5 V p-p differential
when operating at a 2.5 V common-mode voltage. The postamp
implements an optional output clamp engaged through a
resistor from R
to ground. Table 8 shows a list of
CLMP
recommended resistor values.
Output clamping can be used for ADC input overload
protection, if needed, or postamp overload protection when
operating from a lower common-mode level, such as 1.5 V. The
user should be aware that distortion products increase as output
levels approach the clamping levels, and the user should adjust
the clamp resistor accordingly. For additional information, see
the
Applications section.
The accuracy of the clamping levels is approximately ±5% in LO
or HI mode.
few values of R
Figure 80 illustrates the output characteristics for a
.
CLMP
5.0
4.5
4.0
3.5
3.0
(V)
OL
2.5
,
OH
2.0
V
1.5
1.0
0.5
0
R
=
∞
CLMP
8.8kΩ
3.5kΩ
= 1.86kΩ
R
CLMP
3.5kΩ
8.8kΩ
R
=
∞
CLMP
–3–2–10123
V
(V)
INH
Figure 80. Output Clamping Characteristics
03199-080
Rev. E | Page 29 of 40
Page 30
AD8331/AD8332/AD8334
(
)
V
A
APPLICATIONS
C
LNA—EXTERNAL COMPONENTS
The LMD pin (connected to the bias circuitry) must be bypassed to
ground and signal sourced to the INH pin capacitively coupled
using 2.2 nF to 0.1 F capacitors (see
The unterminated input impedance of the LNA is 6 k. The
user can synthesize any LNA input resistance between 50  and
6 k. R
is calculated according to Equation 6 or selected from
FB
Table 7.
R
k33×
IN
R
=
FB
–k6
(6)
()
R
IN
Table 7. LNA External Component Values for Common
Source Impedances
RIN (Ω) RFB (Nearest STD 1% Value, Ω) CSH (pF)
50 280 22
75 412 12
100 562 8
200 1.13 k 1.2
500 3.01 k None
6 k
∞
When active input termination is used, a decoupling capacitor
(C
) is required to isolate the input and output bias voltages of
FB
the LNA.
Figure 81).
None
LMD
GAIN
1nF
0.1µF
1
LMD2
2
INH2
5V
3
VPS2
4
LON2
5
LOP2
6
COM2
7
10
11
12
13
14
8
9
VIP2
VIN2
VCM2
GAIN
RCLMP
VOH2
VOL2
COMM
*SEE TEXT
1nF0.1µF
1nF0.1µF
LMD1
INH1
VPS1
LON1
LOP1
COM1
VIP1
VIN1
VCM1
HILO
ENB
VOH1
VOL1
VPSV
28
27
26
25
24
23
22
21
20
19
18
17
16
15
1nF
FB
C
*
SH
*
C
FB
*
R
FB
1nF
LNA OUT
0.1µF
1nF
5V
5V
*
*
0.1µF
VGA OUT
VGA OUT
5V
0.1µF
Figure 81. Basic Connections for a Typical Channel (AD8332 Shown)
LN
DECOUPLING
R
RESISTO R
FB
0.1µF
5V
0.1µF
TO EXT
CIRCUIT
LNA
SOURCE
03199-081
The shunt input capacitor, C
, reduces gain peaking at higher
SH
frequencies where the active termination match is lost due to
the gain roll-off of the LNA at high frequencies. The value of
C
diminishes as RIN increases to 500 Ω, at which point no
SH
capacitor is required. Suggested values for C
200 Ω are shown in
Table 7.
for 50 Ω ≤ RIN ≤
SH
When a long trace to Pin INH is unavoidable, or if both LNA
outputs drive external circuits, a small ferrite bead (FB) in series
with Pin INH preserves circuit stability with negligible effect on
noise. The bead shown is 75 Ω at 100 MHz (Murata BLM21 or
equivalent). Other values can prove useful.
Figure 82 shows the interconnection details of the LNA output.
Capacitive coupling between the LNA outputs and the VGA
inputs is required because of the differences in their dc levels
and the need to eliminate the offset of the LNA. Capacitor
values of 0.1 µF are recommended. There is 0.4 dB loss in gain
between the LNA output and the VGA input due to the 5 Ω
output resistance. Additional loading at the LOP and LON
outputs affect LNA gain.
VCM
LNA
DECOUPLING
RESISTO R
VIP
VIN
50Ω
50Ω
100Ω
100Ω
TO EXT
CIRCUIT
5Ω
LON
LNA
C
SH
LOP
5Ω
Figure 82. Interconnections of the LNA and VGA
Both LNA outputs are available for driving external circuits.
Pin LOP should be used in those instances when a single-ended
LNA output is required. The user should be aware of stray
capacitance loading of the LNA outputs, in particular LON. The
LNA can drive 100 Ω in parallel with 10 pF. If an LNA output is
routed to a remote PC board, it tolerates a load capacitance up
to 100 pF with the addition of a 49.9 Ω series resistor or ferrite
75 Ω/100 MHz bead.
Gain Input
The GAIN pin is common to both channels of the AD8332. The
input impedance is nominally 10 MΩ and a bypass capacitor
from 100 pF to1 nF is recommended.
03199-082
Rev. E | Page 30 of 40
Page 31
AD8331/AD8332/AD8334
–
Parallel connected devices can be driven by a common voltage
source or DAC. Decoupling should take into account any
bandwidth considerations of the drive waveform, using the total
distributed capacitance.
If gain control noise in LO gain mode becomes a factor,
maintaining ≤15 nV/√Hz noise at the GAIN pin ensures
satisfactory noise performance. Internal noise prevails below
15 nV/√Hz at the GAIN pin. Gain control noise is negligible in
HI gain mode.
VCM Input
The common-mode voltage of Pin VCM, Pin VOL, and Pin VOH
defaults to 2.5 V dc. With output ac-coupled applications, the
VCM pin is unterminated; however, it must still be bypassed in
close proximity for ac grounding of internal circuitry. The VGA
outputs can be dc connected to a differential load, such as an
ADC. Common-mode output voltage levels between 1.5 V and
3.5 V can be realized at Pin VOH and Pin VOL by applying the
desired voltage at Pin VCM. DC-coupled operation is not
recommended when driving loads on a separate PC board.
The voltage on the VCM pin is sourced by an internal buffer
with an output impedance of 30 Ω and a ±2 mA default output
current (see
Figure 83). If the VCM pin is driven from an
external source, its output impedance should be <<30 Ω and its
current drive capability should be >>2 mA. If the VCM pins of
several devices are connected in parallel, the external buffer
should be capable of overcoming their collective output currents.
When a common-mode voltage other than 2.5 V is used, a
voltage-limiting resistor, R
, is needed to protect against
CLMP
overload.
INTERNAL
2mA MAX
AC GROUNDING F OR
INTERNAL CIRCUI TRY
CIRCUITRY
30Ω
Figure 83. VCM Interface
VCM
100pF
RO << 30Ω
0.1µF
NEW V
CM
03199-083
Logic Inputs—ENB, MODE, and HILO
The input impedance of all enable pins is nominally 25 kΩ and
can be pulled up to 5 V (a pull-up resistor is recommended) or
driven by any 3 V or 5 V logic families. The enable pin, ENB,
powers down the VGA—when pulled low, the VGA output
voltages are near ground. Multiple devices can be driven from a
common source. Consult
Table 3 , Tabl e 4, Ta ble 5, and Tabl e 6
for circuit functions controlled by the enable pins.
Optional Output Voltage Limiting
The RCLMP pin provides the user with a means to limit the
output voltage swing when used with loads that have no
provisions for prevention of input overdrive. The peak-to-peak
limited voltage is adjusted by a resistor to ground, and
Table 8
lists several voltage levels and the corresponding resistor value.
Unconnected, the default limiting level is 4.5 V p-p.
Note that third harmonic distortion increases as waveform
amplitudes approach clipping. For lowest distortion, the clamp
level should be set higher than the converter input span. A
clamp level of 1.5 V p-p is recommended for a 1 V p-p linear
output range, 2.7 V p-p for a 2 V p-p range, or 1 V p-p for
a 0.5 V p-p operation. The best solution is determined
experimentally.
Figure 84 shows third harmonic distortion
as a function of the limiting level for a 2 V p-p output signal.
A wider limiting level is desirable in HI gain mode.
20
V
= 0.75V
GAIN
–30
–40
–50
HD3 (dBc)
–60
–70
–80
1.52.02.53.04.03.54.55. 0
Figure 84. HD3 vs. Clamping Level for 2 V p-p Differential Input
CLAMP LIMIT LEVEL (V p-p)
HILO = LO
HILO = HI
03199-084
Table 8. Clamp Resistor Values
Clamp Resistor Value (kΩ)
Clamp Level (V p-p)
HILO = LO HILO = HI
0.5 1.21
1.0 2.74 2.21
1.5 4.75 4.02
2.0 7.5 6.49
2.5 11 9.53
3.0 16.9 14.7
3.5 26.7 23.2
4.0 49.9 39.2
4.4 100 73.2
Pin HILO is compatible with 3 V or 5 V CMOS logic families. It
is either connected to ground or pulled up to 5 V, depending on
the desired gain range and output noise.
Rev. E | Page 31 of 40
Page 32
AD8331/AD8332/AD8334
X
X
Output Decoupling
When driving capacitive loads greater than about 10 pF, or long
circuit connections on other boards, an output network of
resistors and/or ferrite beads can be useful to ensure stability.
These components can be incorporated into a Nyquist filter
such as the one shown in
Figure 81. In Figure 81, the resistor
value is 84.5 Ω. The AD8332-EVAL incorporates 100  in
parallel with a 120 nH bead. Lower value resistors are permissible
for applications with nearby loads or with gains less than 40 dB.
The exact values of these components can be selected empirically.
An antialiasing noise filter is typically used with an ADC. Filter
requirements are application dependent.
When the ADC resides on a separate board, the majority of
filter components should be placed nearby to suppress noise
picked up between boards and to mitigate charge kickback from
the ADC inputs. Any series resistance beyond that required for
output stability should be placed on the ADC board.
Figure 85
shows a second-order, low-pass filter with a bandwidth of
20 MHz. The capacitor is chosen in conjunction with the 10 pF
input capacitance of the ADC.
OPTIONAL
BACKPLANE
84.5Ω
84.5Ω
0.1µF
0.1µF
1.5µH
1.5µH
158Ω
158Ω
18pF
ADC
03199-085
Figure 85. 20 MHz Second-Order, Low-Pass Filter
DRIVING ADCs
The output drive accommodates a wide range of ADCs. The
noise floor requirements of the VGA depend on a number of
application factors, including bit resolution, sampling rate, fullscale voltage, and the bandwidth of the noise/antialias filter. The
output noise floor and gain range can be adjusted by selecting
HI or LO gain mode.
4V p-p DIFF,
VOH
VOL
48nV/ Hz
187Ω
2:1
187Ω
2V p-p DIFF,
24nV/ Hz
374Ω
LPF
ADC
AD6644
03199-086
Figure 86. Adjusting the Noise Floor for 14-Bit ADCs
OVERLOAD
These devices respond gracefully to large signals that overload
its input stage and to normal signals that overload the VGA
when the gain is set unexpectedly high. Each stage is designed
for clean-limited overload waveforms and fast recovery when
gain setting or input amplitude is reduced.
Signals larger than ±275 mV at the LNA input are clipped to
MODE
Figure 48
-AMP
OVERLOAD
25mV
41dB
24.5dB
5 V p-p differential prior to the input of the VGA.
shows the response to a 1 V p-p input burst. The symmetric
overload waveform is important for applications, such as CW
Doppler ultrasound, where the spectrum of the LNA outputs
during overload is critical. The input stage is also designed to
accommodate signals as high as ±2.5 V without triggering the
slow-settling ESD input protection diodes.
Both stages of the VGA are susceptible to overload. Postamp
limiting is more common and results in the clean-limited
output characteristics found in
cases. The graph in
Figure 87 summarizes the combinations of
Figure 49. Recovery is fast in all
input signal and gain that lead to the different types of overload.
43.5
GAIN (dB)
POSTAMP
OVERLOAD
15mV
LO GAIN
MODE
-AMP
OVERLOAD
25mV
29dB
24.5dB
56.5
GAIN (dB)
POSTAMP
OVERLOAD
4mV
HI GAIN
The relative noise and distortion performance of the two gain
modes can be compared in
Figure 25 and Figure 31 through
Figure 41. The 48 nV/√Hz noise floor of the LO gain mode is
suited to converters with higher sampling rates or resolutions
(such as 12 bits). Both gain modes can accommodate ADC full-
–4.5
1m
INPUT AMP LITUDE (V)
Figure 87. Overload Gain and Signal Conditions
LNA OVERLOAD
0.2750.110m
7.5
1
1m0.2750.110m1
INPUT AMPLITUDE (V)
LNA OVERLOAD
03199-087
scale voltages as high as 4 V p-p. Because distortion performance
remains favorable for output voltages as high as 4 V p-p (see
Figure 36), it is possible to lower the output-referred noise even
further by using a resistive attenuator (or transformer) at the
output. The circuit in
Figure 86 has an output full-scale range of
2 V p-p, a gain range of −10.5 dB to +37.5 dB, and an output
noise floor of 24 nV/√Hz, making it suitable for some 14-bit
ADC applications.
The previously mentioned clamp interface controls the
maximum output swing of the postamp and its overload
response. When the clamp feature is not used, the output level
defaults to approximately 4.5 V p-p differential centered at
2.5 V common mode. When other common-mode levels are set
through the VCM pin, the value of R
should be selected for
CLMP
graceful overload. A value of 8.3 kΩ or less is recommended for
1.5 V or 3.5 V common-mode levels (7.2 kΩ for HI gain mode).
This limits the output swing to just above 2 V p-p differential.
Rev. E | Page 32 of 40
Page 33
AD8331/AD8332/AD8334
V
OPTIONAL INPUT OVERLOAD PROTECTION
Applications in which high transients are applied to the LNA
input can benefit from the use of clamp diodes. A pair of backto-back Schottky diodes can reduce these transients to manageable
levels.
Figure 88 illustrates how such a diode-protection scheme
can be connected.
OPTIONAL
SCHOTTKY
OVERLOAD
CLAMP
231
BAS40-04
FB
0.1µF
2
C
R
C
FB
SH
R
FB
SH
3
4
INH
VPSL
LON
COMM
ENBL
20
19
03199-088
Figure 88. Input Overload Clamping
When selecting overload protection, the important parameters
are forward and reverse voltages and t
BAS40-04 series shown in
Figure 88 has a τrr of 100 ps and VF of
(or τrr). The Infineon
rr
310 mV at 1 mA. Many variations of these specifications can be
found in vendor catalogs.
LAYOUT, GROUNDING, AND BYPASSING
Due to their excellent high frequency characteristics, these
devices are sensitive to their PCB environment. Realizing
expected performance requires attention to detail critical to
good high speed board design.
MULTIPLE INPUT MATCHING
Matching of multiple sources with dissimilar impedances can be
accomplished as shown in
Figure 90. A relay and low supply
voltage analog switch can be used to select between multiple
sources and their associated feedback resistors. An
ADG736
dual SPDT switch is shown in this example; however, multiple
switches are also available and users are referred to the Analog
Devices Selection Guide for switches and multiplexers.
DISABLING THE LNA
Where accessible, connection of the LNA enable pin to ground
powers down the LNA, resulting in a current reduction of about
half. In this mode, the LNA input and output pins can be left
unconnected; however, the power must be connected to all the
supply pins for the disabling circuit to function.
illustrates the connections using an AD8331 as an example.
1
NC
LMD
AD8331
2
NC
INH
C
FB
0.018µF
5
NC
3
VPSL
4
LON
COMM
ENBL
ENBV
COMM
20
19
18
17
Figure 89
5V
A multilayer board with power and ground planes is
recommended with blank areas in the signal layers filled with
ground plane. Be certain that the power and ground pins
provided for robust power distribution to the device are
connected. Decouple the power supply pins with surface-mount
capacitors as close as possible to each pin to minimize impedance
paths to ground. Decouple the LNA power pins from the VGA
supply using ferrite beads. Together with the capacitors, ferrite
beads eliminate undesired high frequencies without reducing
the headroom. Use a larger value capacitor for every 10 chips to
20 chips to decouple residual low frequency noise. To minimize
voltage drops, use a 5 V regulator for the VGA array.
Several critical LNA areas require special care. The LON and
LOP output traces must be as short as possible before connecting to
the coupling capacitors connected to Pin VIN and Pin VIP. R
FB
must be placed near the LON pin as well. Resistors must be
placed as close as possible to the VGA output pins, VOL and
VOH, to mitigate loading effects of connecting traces. Values
are discussed in the
Output Decoupling section.
Signal traces must be short and direct to avoid parasitic effects.
Wherever there are complementary signals, symmetrical layout
should be employed to maintain waveform balance. PCB traces
should be kept adjacent when running differential signals over a
long distance.
5
NC
LOP
6
COML
0.1µF
VIN
0.1µF
MODE
7
8
9
VIP
VIN
MODE
10
GAIN
GAIN
16
VOL
VOUT
15
VOH
14
VPOS5V
13
HILO
RCLMP
VCM
HILO
12
11
VCM
R
CLMP
03199-089
Figure 89. Disabling the LNA
Rev. E | Page 33 of 40
Page 34
AD8331/AD8332/AD8334
18nF
200Ω
50Ω
0.1µF
Figure 90. Accommodating Multiple Sources
SELECT R
INH
LMD
AD8332
ADG736
FB
LNA
1.13kΩ
280Ω
LON
5Ω
LOP
5Ω
03199-090
ULTRASOUND TGC APPLICATION
The AD8332 ideally meets the requirements of medical and
industrial ultrasound applications. The TGC amplifier is a key
subsystem in such applications, because it provides the means
for echolocation of reflected ultrasound energy.
Figure 91 through Figure 93 are schematics of a dual, fully
differential system using the AD8332 and the
high speed ADC with conversion speeds as high as 65 MSPS.
AD9238, 12-bit
Using the EVAL-AD8332/AD9238 evaluation board and a high
speed ADC FIFO evaluation kit connected to a laptop, an FFT
can be performed on the AD8332. With the on-board clock of
20 MHz, minimal low-pass filtering, and both channels driven
with a 1 MHz filtered sine wave, the THD is −75 dB, noise floor
is −93 dB, and HD2 is −83 dB.
HIGH DENSITY QUAD LAYOUT
The AD8334 is the ideal solution for applications with limited
board space.
away from this very compact quad VGA. Note that none of the
signal paths crosses and that all four channels are spaced apart
to eliminate crosstalk.
In this example, all of the components shown are 0402 size;
however, the same layout is executable at the expense of slightly
more board area. The sketch also assumes that both sides of the
printed circuit board are available for components, and that the
bypass and power supply decoupling circuitry is located on the
wiring side of the board.
Figure 94 represents four channels routed to and
Rev. E | Page 34 of 40
Page 35
AD8331/AD8332/AD8334
S3
E
IN2
TP3
(RED)
TP4
L19
SAT
L20
SAT
+5V
120nH FB
120nH FB
FILTER
C67
SAT
TB1
+5V
(BLACK)
TB2
GND
OPTIONAL 4-POLE LOW-PASS
VIN+B
C66
SAT
–B
V
IN
TP5
C50
0.1µF
C49
0.1µF
L12
120nH FB
C80
22pF
+5VLNA
C41
0.1µF
C53
0.1µF
C48
0.1µF
C83
1nF
C69
R27
100Ω
L11
120nH FB
L10
120nH FB
R26
100Ω
C68
1nF
JP5
IN2
C74
1nF
C78
1nF
CFB2
+
C46
1µF
L7
+5VGA
L6
+5VLNA
L17
SAT
JP12
L18
SAT
18nF
RFB2
274Ω
0.1µF
VCM1
JP13
TP2 GAIN
TP7 GND
(R
CLMP
DC2H
C54
0.1µF
C55
0.1µF
JP7
DC2L
C51
R3
JP8
)
0.1µF
AD8332ARU
1
LMD2
2
INH2
3
VPS2
4
LON2
5
LOP2
6
COM2
7
VIP2
8
VIN2
9
VCM2
10
GAIN
11
RCLMP
12
VOH2
13
VOL2
14
COM
+5VGA
C45
0.1µF
LMD1
INH1
VPS1
LON1
LOP1
COM1
VIP1
VIN1
VCM1
HILO
ENB
VOH1
VOL1
VPSV
C85
1nF
28
27
26
+5VLNA
25
24
23
22
21
20
C77
1nF
19
+5VGA
18
17
16
15
C70
0.1µF
JP6
IN1
ENABLE
JP16
DISABLE
120nH FB
120nF FB
22pF
C42
0.1µF
C43
0.1µF
R24
100Ω
L9
L8
R25
100Ω
C79
VCM1
120nH FB
CFB1
18nF
RFB1
274Ω
C59
0.1µF
C58
0.1µF
C56
0.1µF
JP10
L13
+5VGA
HI GAIN
JP10
LO GAIN
JP9
JP17
TP6
C60
0.1µF
OPTIONAL 4-POLE LOW-PASS
L1
SAT
L14
SAT
FILTER
L15
SAT
C64
SAT
SAT
E
L16
S1
IN1
VIN+A
C65
SAT
VIN–A
03199-091
Figure 91. Schematic, TGC, VGA Section Using an AD8332 and AD9238
Rev. E | Page 35 of 40
Page 36
AD8331/AD8332/AD8334
V
ADP3339AKC-3.3
+5V
312
OUT
IN
OUT
TAB
S2
EXT CLOCK
R17
49.9Ω
+3.3VCLK
C86
0.1µF
V
DD
20MHz
GND
SG-636PCE
2
OUT
R1
L5
120nH FB
L4
120nH FB
L3
120nH FB
L2
120nH FB
C31
0.1µF
C30
0.1µF
C29
0.1µF
C1
0.1µF
VIN+_A
V
–_A
IN
C35
0.1µF
C36
0.1µF
R12
1.5kΩ
R4
1.5kΩ
C33
10µF
6.3V
+
R5
33Ω
R6
33Ω
C17
0.1µF
GND
C44
1µF
+
VREF
C38
0.1µF
C34
10µF
6.3V
10µF
C37
0.1µF
1.5kΩ1. 5kΩ
0.1µF
R8
R20
ADCLK
TP 13
8965
C20
0.1µF
JP11JP3
4.7kΩ
TP 12
JP1
1
33Ω
R7
33Ω
R41
DATA
CLK
2
VIN–B
V
+B
IN
C63
0.1µF
+3.3VCLK
R18
499Ω
R16
5kΩ
R19
499Ω
4.7kΩ
C47
+
10µF
6.3V
EXT
JP4
3
2
1
INT
14
OE
3
ADCLK
U5
74VHC04
U5
74VHC04
4312
U5
74VHC04
U5
74VHC04
U6
U5
74VHC04
1213
SPARES
U5
74VHC04
1011
Figure 92. Converter Schematic, TGC Using an AD8332 and AD9238
C32
0.1µF
C39
C16
10µF
0.1µF
C19
1nF
3
6.3V
+3.3VAVDD
C2
+
C61
18pF
C18
1nF
C40
TP9
+
C12
10µF
6.3V
C15
1nF
C62
18pF
R9
0Ω
+3.3VADDIG
C52
10nF
C57
10nF
DNC
DNC
D0_B
D1_B
D2_B
D3_B
D4_B
D5_B
0.1µF
C26
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
C22
0.1µF
AGND
VIN+_A
VIN–_A
AGND
AVDD
REFT_A
REFB_A
VREF
SENSE
REFB_B
REFT_B
AVDD
AGND
VIN–_B
VIN+_B
AGND
AVDD
CLK_B
DCS
DFS
PDWN_B
OEB_B
DNC
DNC
D0_B
D1_B
D2_B
DRGND
DRVDD
D3_B
D4_B
D5_B
C24
1nF
C21
1nF
AVDD
CLK_A
SHARED_REF
MUX_SELECT
PDWN_A
OEB_A
OTR_A
D11_A (MSB)
D10_A
D9_A
D8_A
DRGND
DRVDD
D7_A
D6_A
D5_A
D4_A
U1 A/D CONVERTER AD9238
D3_A
D2_A
D1_A
D0_A
DNC
DNC
DRVDD
DRGND
OTR_B
D11_B (MSB)
D10_B
D9_B
D8_B
D7_B
D6_B
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
ADCLK
R10
0Ω
4.7kΩ
R15
0Ω
R11
100Ω
SHARED
Y
R14
+3.3VADDIG
OTR_A
D11_A
D10_A
D9_A
D8_A
C23
0.1µF
D7_A
D6_A
D5_A
D4_A
D3_A
D2_A
D1_A
D0_A
DNC
DNC
C13
1nF
OTR_B
D11_B
D10_B
D9_B
D8_B
D7_B
D6_B
JP2
REF
N
+3.3VADDIG
C25
1nF
C14
0.1µF
C11
+
10µF
6.3V
03199-092
Rev. E | Page 36 of 40
Page 37
AD8331/AD8332/AD8334
R40
22Ω
22 × 4
18
RP 1
7
2
6
3
54
18
22 × 4
RP2
7
2
6
3
54
18
22 × 4
RP 3
7
2
6
3
54
18
22 × 4
RP 4
7
2
6
3
54
2
4
6
8
10
12
14
HEADER UP MALE NO SHROUD
16
18
20
22
24
2625
2827
30
34
36
38
40
SAM080UPM
1
3
5
7
9
11
13
15
17
19
21
23
29
3132
33
35
37
39
OTR_A
D11_A
D10_A
D9_A
D8_A
D7_A
D6_A
D5_A
D4_A
D3_A
D2_A
D1_A
D0_A
DNC
DNC
DATACLKA
22 × 4
1
RP 9
2
3
4
1
22 × 4
RP 10
2
3
4
1
22 × 4
RP 11
2
3
4
1
22 × 4
RP 12
2
3
4
1
G1
74VHC541
19
G2
2
8
A1
7
3
A2
6
4
A3
5
5
A4
8
6
A5
7
7
A6
6
8
A7
5
9
A8
U10
VCC
GND
20
+
10
18
Y1
17
Y2
16
Y3
15
Y4
14
Y5
13
Y6
12
Y7
11
Y8
C3
0.1µF
C28
10µF
6.3V
+3.3VDVDD
+3.3VDVDD
1
G1
74VHC541
19
G2
8
2
A1
3
7
A2
6
4
A3
5
5
A4
8
6
A5
7
7
A6
8
6
A7
5
9
A8
20
U7
VCC
GND
10
18
Y1
17
Y2
16
Y3
15
Y4
14
Y5
13
Y6
12
Y7
11
Y8
C8
0.1µF
C10
0.1µF
+
C76
10µF
6.3V
OTR_B
D11_B
D10_B
D9_B
D8_B
D7_B
D6_B
D5_B
D4_B
D3_B
D2_B
D1_B
D0_B
DNC
DNC
19
8
1
22 × 4
RP 13
7
2
6
3
5
4
22 × 4
8
1
RP 14
7
2
6
3
5
4
18
22 × 4
RP 15
19
7
2
6
3
5
4
22 × 4
8
1
RP 16
7
2
6
3
5
4
DATACLK
+3.3VDVDD
1
2
3
4
5
6
7
8
9
G1
74VHC541
G2
A1
A2
A3
A4
A5
A6
A7
A8
20
U2
VCC
GND
10
18
Y1
17
Y2
16
Y3
15
Y4
14
Y5
13
Y6
12
Y7
11
Y8
++
C7
0.1µFC90.1µF
C27
10µF
6.3V
+3.3VDVDD
1
2
3
4
5
6
7
8
9
G1
G2
A1
A2
A3
A4
A5
A6
A7
A8
74VHC541
20
VCC
U3
GND
Y1
Y2
Y3
Y4
Y5
Y6
Y7
Y8
C4
10
0.1µFC50.1µFC60.1µF
18
17
16
15
14
13
12
11
+
C75
10µF
6.3V
Figure 93. Interface Schematic, TGC Using an AD8332 and AD9238
22 × 4
18
RP 5
7
2
6
3
54
8
22 × 4
1
RP 6
7
2
6
3
54
18
22 × 4
RP 7
7
2
6
3
54
18
22 × 4
RP 8
7
2
6
3
54
R39
22Ω
42
44
46
4847
52
HEADER UP MALE NO SHROUD
5453
5655
58
60
62
64
66
68
70
72
74
76
78
80
SAM080UPM
41
43
45
4950
51
57
59
61
63
65
67
69
71
73
75
77
79
03199-093
Rev. E | Page 37 of 40
Page 38
AD8331/AD8332/AD8334
CH 1 LNA INPUT
CH 2 LNA INPUT
CH 3 LNA INPUT
64
1
INH2
2
LMD2
3
COM2X
4
LON2
5
LOP2
6
VIP2
7
VIN2
8
VPS2
9
VPS3
10
VIN3
11
VIP3
12
LOP3
13
LON3
14
COM3X
15
LMD3
16
INH3
INH1
COM2
COM3
LMD1
COM1
COM4
INH4
LMD4
20
58 575962 61 6063
LOP1
LON1
COM1X
POWER SUPPLY DECOUPLING
LOCATED O N WIRING SIDE
COM4X
LON4
LOP4
21
22 23 24
VIP1
AD8334
VIP4
50 4956 555154 53 52
VIN1
VPS1
VIN4
VPS4
EN12
EN34
HILO
VCM4
30 31 32
VCM1
VCM3
GAIN12
CLMP12
GAIN34
CLMP34
29
282526 2717 18 19
VCM2
COM12
VOH1
VOL1
VPS12
VOL2
VOH2
COM12
MODE
COM34
VOH3
VOL3
VPS34
VOL4
VOH4
COM34
NC
48
47
46
45
44
43
42
41
40
NC
39
38
37
36
35
34
33
CH 1 DIFFERENTIAL
OUTPUT
CH 2 DIFFERENTIAL
OUTPUT
CH 3 DIFFERENTIAL
OUTPUT
CH 4 DIFFERENTIAL
OUTPUT
CH 4 LNA INPUT
3199-094
Figure 94. Signal Path and Board Layout for AD8334
Rev. E | Page 38 of 40
Page 39
AD8331/AD8332/AD8334
C
Y
OUTLINE DIMENSIONS
9.80
9.70
9.60
0.345
0.341
0.337
PIN 1
0.15
0.05
OPLANARIT
0.10
28
0.65
BSC
0.30
0.19
COMPLIANT TO JEDEC STANDARDS MO-153-AE
1.20 MAX
SEATING
PLANE
15
4.50
4.40
4.30
0.20
0.09
6.40 BSC
141
Figure 95. 28-Lead Thin Shrink Small Outline Package [TSSOP]
(RU-28)
Dimensions shown in millimeters
5.00
BSC SQ
PIN 1
INDICATOR
1.00
0.85
0.80
12° MAX
SEATING
PLANE
TOP
VIEW
BSC SQ
0.80 MAX
0.65 TYP
0.30
0.23
0.18
Figure 97. 32-Lead Lead Frame Chip Scale Package [LFCSP_VQ]
PIN 1
8°
0°
0.75
0.60
0.45
0.010
0.004
COPLANARITY
0.004
0.08
0.60 MAX
25
24
EXPOSED
PAD
(BOTTOM VIEW)
17
16
3.50 REF
0.60 MAX
0.50
4.75
0.05 MAX
0.02 NOM
0.20 REF
COMPLIANT TO JEDEC STANDARDS MO-220-VHHD-2
BSC
0.50
0.40
0.30
COPLANARIT Y
5 mm × 5 mm Body, Very Thin Quad
(CP-32-2)
Dimensions shown in millimeters
2011
1
0.065
0.049
BSC
0.012
0.008
0.025
COMPLIANT TO JEDEC STANDARDS MO-137-AD
0.069
0.053
10
SEATING
PLANE
0.158
0.154
0.150
0.244
0.236
0.228
0.010
0.006
8°
0°
Figure 96. 20-Lead Shrink Small Outline Package [QSOP]
(RQ-20)
Dimensions shown in Inches
PIN 1
32
1
8
9
INDICATOR
3.25
3.10 SQ
2.95
0.25 MIN
THE EXPOSE PAD IS NOT CONNECTED
INTERNALLY. FOR I NCREASED RELIABILITY
OF THE SOL DER JOINTS AND MAXIMUM
THERMAL CAPABIL ITY IT IS RECOMMENDED
THAT THE PAD BE SOLDERED TO
THE GROUND PLANE.
041806-A
0.050
0.016
Rev. E | Page 39 of 40
Page 40
AD8331/AD8332/AD8334
1.00
0.85
0.80
SEATING
PLANE
12° MAX
9.00
BSC SQ
PIN 1
INDICATOR
TOP
VIEW
0.80 MAX
0.65 TYP
0.50 BSC
0.60 MAX
8.75
BSC SQ
0.45
0.40
0.35
0.05 MAX
0.02 NOM
0.20 REF
*
COMPLIANT TO JEDEC S TANDARDS MO-220-VMMD-4
EXCEPT FOR EXPOSED PAD DIMENSION
49
48
33
32
0.60 MAX
EXPOSED PAD
(BOTTOM VIEW)
7.50
REF
0.30
0.25
0.18
64
17
PIN 1
INDICATOR
1
*
4.85
4.70 SQ
4.55
16
THE EXPOSE PAD IS NOT CONNECTED
INTERNALL Y. FOR INCREASED RELIABIL ITY
OF THE S OLDER JOI NTS AND MAXI MUM
THERMAL CAPABILITY I T IS RECO MMENDED
THAT THE PAD BE SOLDERED TO
THE GROUND PL ANE.
031706-A
Figure 98. 64-Lead Lead Frame Chip Scale Package (LFCSP_VQ)
9 mm × 9 mm Body, Very Thin Quad
(CP-64-1)
Dimensions shown in millimeters
ORDERING GUIDE
Model Temperature Range Package Description Package Option
AD8331ARQ –40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
AD8331ARQ-REEL –40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
AD8331ARQ-REEL7 –40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
AD8331ARQZ
AD8331ARQZ-RL
AD8331ARQZ-R7
AD8331-EVAL Evaluation Board with AD8331ARQ
AD8332ACP-R2 –40°C to +85°C 32-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-32-2
AD8332ACP-REEL –40°C to +85°C 32-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-32-2
AD8332ACP-REEL7 –40°C to +85°C 32-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-32-2
AD8332ACPZ-R7
AD8332ACPZ-RL
AD8332ARU –40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
AD8332ARU-REEL –40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
AD8332ARU-REEL7 –40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
AD8332ARUZ
AD8332ARUZ-R7
AD8332ARUZ-RL
AD8332-EVAL Evaluation Board with AD8332ARU
EVAL-AD8332/AD9238 Evaluation Board with AD8332ARU and AD9238 ADC
AD8334ACPZ-WP
AD8334ACPZ-REEL
AD8334ACPZ-REEL7
AD8334-EVAL Evaluation Board with AD8334ACP
–40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
–40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
–40°C to +85°C 20-Lead Shrink Small Outline Package (QSOP) RQ-20
–40°C to +85°C 32-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-32-2
–40°C to +85°C 32-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-32-2
–40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
–40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
–40°C to +85°C 28-Lead Thin Shrink Small Outline Package (TSSOP) RU-28
–40°C to +85°C 64-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-64-1
–40°C to +85°C 64-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-64-1
–40°C to +85°C 64-Lead Lead Frame Chip Scale Package (LFCSP_VQ) CP-64-1
Rev. E | Page 40 of 40
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.