31 MHz, Dual Programmable Filters
and Variable Gain Amplifiers
Data Sheet
ADRF6516
Rev. C Document Feedback
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 Analog Devices.
Trademarks and registered trademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700 ©2010–2017 Analog Devices, Inc. All rights reserved.
Technical Support www.analog.com
FEATURES
Matched pair of programmable filters and VGAs
Continuous gain control range: 50 dB
Digital gain control: 15 dB
6-pole Butterworth filter: 1 MHz to 31 MHz
in 1 MHz steps, 1 dB corner frequency
Preamplifier and postamplifier gain steps
IMD3: >65 dBc for 1.5 V p-p composite output
HD2, HD3: >65 dBc for 1.5 V p-p output
Differential input and output
Flexible output and input common-mode ranges
Optional dc offset compensation loop
SPI programmable filter corners and gain steps
Power-down feature
Single 3.3 V supply operation
APPLICATIONS
Baseband IQ receivers
Diversity receivers
ADC drivers
Point-to-point and point-to-multipoint radio
Instrumentation
Medical
FUNCTIONAL BLOCK DIAGRAM
ENBL
VPSD
COMD
LE
CLK
DATA
SDO
COM
VPS
OPP1
OPM1
COM
GAIN
VOCM
COM
OPM2
OPP2
INP1 INM1 VPS
ADRF6516
COM VICM OFS1 VPS
COM INP2 INM2 VPS COM OFDS OFS2 VPS
SPI
09422-001
Figure 1.
GENERAL DESCRIPTION
The ADRF6516 is a matched pair of fully differential, low noise
and low distortion programmable filters and variable gain
amplifiers (VGAs). Each channel is capable of rejecting large
out-of-band interferers while reliably boosting the desired signal,
thus reducing the bandwidth and resolution requirements on the
analog-to-digital converters (ADCs). The excellent matching
between channels and their high spurious-free dynamic range
over all gain and bandwidth settings make the ADRF6516 ideal
for quadrature-based (IQ) communication systems with dense
constellations, multiple carriers, and nearby interferers.
The filters provide a six-pole Butterworth response with 1 dB
corner frequencies programmable through the SPI port from
1 MHz to 31 MHz in 1 MHz steps. The preamplifier that precedes
the filters offers a SPI-programmable option of either 3 dB or 6 dB
of gain. The preamplifier sets a differential input impedance of
1600 Ω and has a common-mode voltage that defaults to VPS/2
but can be driven from 1.1 V to 1.8 V.
The variable gain amplifiers that follow the filters provide 50 dB
of continuous gain control with a slope of 15.5 mV/dB. Their
maximum gains can be programmed to various values through
the SPI. The output buffers provide a differential output impedance
of 30 Ω and are capable of driving 2 V p-p into 1 kΩ loads. The
output common-mode voltage defaults to VPS/2, but it can be
adjusted down to 700 mV by driving the high impedance
VOCM pin. Independent, built-in dc offset compensation loops
can be disabled if fully dc-coupled operation is desired. The
high-pass corner frequency is defined by external capacitors on
the OFS1 and OFS2 pins and the VGA gain.
The ADRF6516 operates from a 3.15 V to 3.45 V supply
and consumes a maximum supply current of 360 mA when
programmed to the highest bandwidth setting. When disabled,
it consumes <9 mA. The ADRF6516 is fabricated in an advanced
silicon-germanium BiCMOS process and is available in a 32-lead,
exposed paddle LFCSP. Performance is specified over the −40°C
to +85°C temperature range.
ADRF6516 Data Sheet
Rev. C | Page 2 of 29
TABLE OF CONTENTS
Features .............................................................................................. 1
Applications ....................................................................................... 1
Functional Block Diagram .............................................................. 1
General Description ......................................................................... 1
Revision History ............................................................................... 2
Specifications ..................................................................................... 3
Timing Diagrams .......................................................................... 5
Absolute Maximum Ratings ............................................................ 6
ESD Caution .................................................................................. 6
Pin Configuration and Function Descriptions ............................. 7
Typical Performance Characteristics ............................................. 8
Register Map and Codes ................................................................ 15
Theory of Operation ...................................................................... 16
Input Buffers ............................................................................... 16
Programmable Filters ................................................................. 16
Variable Gain Amplifiers (VGAs) ............................................ 17
Output Buffers/ADC Drivers ................................................... 17
DC Offset Compensation Loop ................................................ 17
Programming the Filters and Gains ......................................... 18
Noise Characteristics ................................................................. 18
Distortion Characteristics ......................................................... 19
Maximizing the Dynamic Range.............................................. 19
Key Parameters for Quadrature-Based Receivers .................. 20
Applications Information .............................................................. 21
Basic Connections ...................................................................... 21
Supply Decoupling ..................................................................... 21
Input Signal Path ........................................................................ 21
Output Signal Path ..................................................................... 21
DC Offset Compensation Loop Enabled ................................ 21
Common-Mode Bypassing ....................................................... 21
Serial Port Connections ............................................................. 22
Enable/Disable Function ........................................................... 22
Error Vector Magnitude (EVM) Performance ........................... 22
EVM Test Setup .......................................................................... 22
Effect of Filter Bandwidth on EVM ......................................... 22
Effect of Output Voltage Levels on EVM ................................ 23
Effect of COFS Value on EVM ..................................................... 23
Evaluation Board ............................................................................ 24
Evaluation Board Control Software ......................................... 24
Schematics and Artwork ........................................................... 25
Outline Dimensions ....................................................................... 29
Ordering Guide .......................................................................... 29
REVISION HISTORY
8/2017—Rev. B to Rev. C
Change to Figure 4 ........................................................................... 7
Updated Outline Dimensions ...................................................... 29
Changes to Ordering Guide Section ............................................ 29
2/2012—Rev. A to Rev. B
Changes to Figure 57 ...................................................................... 24
Changes to Figure 58 ...................................................................... 25
Added Figure 59 .............................................................................. 26
Changes to Figure 60 and Figure 61 ............................................. 27
Changes to Table 6 .......................................................................... 27
9/2011—Revision A: Initial Version
Data Sheet ADRF6516
Rev. C | Page 3 of 29
SPECIFICATIONS
VPS = 3.3 V, TA = 25°C, ZLOAD = 1 kΩ, digital gain code = 111, unless otherwise noted.
Table 1.
Parameter Test Conditions/Comments Min Typ Max Unit
FREQUENCY RESPONSE
Low-Pass Corner Frequency, f
C
6-pole Butterworth filter, 0.5 dB bandwidth
1
31
MHz
Step Size 1 MHz
Corner Frequency Absolute
Accuracy
Over operating temperature range ±15 % fC
Corner Frequency Matching Channel A and Channel B at same gain and
bandwidth settings
±0.5 % fC
Pass-Band Ripple 0.5 dB p-p
Gain Matching Channel A and Channel B at same gain and
bandwidth settings
±0.1 dB
Group Delay Variation From midband to peak
Corner Frequency = 1 MHz 135 ns
Corner Frequency = 31 MHz
ns
Group Delay Matching Channel A and Channel B at same gain
Corner Frequency = 1 MHz 5 ns
Corner Frequency = 31 MHz 0.2 ns
Stop-Band Rejection
Relative to Pass Band 2 × fC 30 dB
5 × fC 75 dB
INPUT STAGE INP1, INM1, INP2, INM2, VICM pins
Maximum Input Swing
At minimum gain, V
GAIN
= 0 V
V p-p
Differential Input Impedance 1600 Ω
Input Common-Mode Range 0.4 V p-p input voltage, HD3 > 65 dBc 1.1 1.65 1.8 V
Input pins left floating VPS/2 V
VICM Output Impedance 7 kΩ
GAIN CONTROL GAIN pin
Voltage Gain Range VGAIN from 0 V to 1 V −5 +45 dB
Gain Slope 15.5 mV/dB
Gain Error VGAIN from 300 mV to 800 mV 0.2 dB
OUTPUT STAGE OPP1, OPM1, OPP2, OPM2, VOCM pins
Maximum Output Swing At maximum gain, RLOAD = 1 kΩ 2 V p-p
HD2 > 65 dBc, HD3 > 65 dBc
V p-p
Differential Output Impedance 30 Ω
Output DC Offset Inputs shorted, offset loop disabled 35 mV
Output Common-Mode Range 0.7 1.65 2.8 V
VOCM pin left floating VPS/2 V
VOCM Input Impedance 23 kΩ
NOISE/DISTORTION
Corner Frequency = 1 MHz
Output Noise Density Gain = 0 dB at fC/2 −141 dBV/√Hz
Gain = 20 dB at f
C
/2
dBV/√Hz
Gain = 40 dB at fC/2 −112 dBV/√Hz
Second Harmonic, HD2 250 kHz fundamental, 1.5 V p-p output voltage
Gain = 5 dB 82 dBc
Gain = 40 dB 68 dBc
Third Harmonic, HD3
250 kHz fundamental, 1.5 V p-p output voltage
Gain = 5 dB 71 dBc
Gain = 40 dB 56 dBc
ADRF6516 Data Sheet
Rev. C | Page 4 of 29
Parameter Test Conditions/Comments Min Typ Max Unit
IMD3 f1 = 500 kHz, f2 = 550 kHz, 1.5 V p-p composite
output voltage
Gain = 5 dB 61 dBc
Gain = 35 dB 42.5 dBc
IMD3 with Input CW Blocker f1 = 500 kHz, f2 = 550 kHz, 1.5 V p-p composite
output, gain = 5 dB; blocker at 5 MHz, 10 dBc
relative to two-tone composite output voltage
40 dBc
Corner Frequency = 31 MHz
Output Noise Density Midband, gain = 0 dB −143.5 dBV/√Hz
Midband, gain = 20 dB −139 dBV/√Hz
Midband, gain = 40 dB
dBV/√Hz
Second Harmonic, HD2 8 MHz fundamental, 1.5 V p-p output voltage
Gain = 5 dB 68 dBc
Gain = 40 dB 70 dBc
Third Harmonic, HD3 8 MHz fundamental, 1.5 V p-p output voltage
Gain = 5 dB 55 dBc
Gain = 40 dB 75 dBc
IMD3 f1 = 14 MHz, f2 = 15 MHz, 1.5 V p-p composite
output voltage
Gain = 5 dB 55 dBc
Gain = 35 dB 77.5 dBc
IMD3 with Input CW Blocker
f1 = 14 MHz, f2 = 15 MHz, 1.5 V p-p composite
output, gain = 5 dB; blocker at 150 MHz, 10 dBc
relative to two-tone composite output voltage
dBc
DIGITAL LOGIC
LE, CLK, DATA, SDO, OFDS pins
Input High Voltage, VINH >2 V
Input Low Voltage, VINL <0.8 V
Input Current, IINH/IINL <1 µA
Input Capacitance, CIN 2 pF
SPI TIMING LE, CLK, DATA, SDO pins (see Figure 2 and Figure 3)
fSCLK 1/tSCLK 20 MHz
tDH DATA hold time 5 ns
tDS DATA setup time 5 ns
tLH LE hold time 5 ns
tLS LE setup time 5 ns
tPW CLK high pulse width 5 ns
tD CLK to SDO delay 5 ns
POWER AND ENABLE VPS, VPSD, COM, COMD, ENBL pins
Supply Voltage Range 3.15 3.3 3.45 V
Total Supply Current ENBL = 3.3 V
Corner frequency = 31 MHz 360 mA
Corner frequency = 1 MHz
mA
Disable Current ENBL = 0 V 9 mA
Disable Threshold 1.6 V
Enable Response Time Delay following ENBL low-to-high transition 20 µs
Disable Response Time Delay following ENBL high-to-low transition 300 ns
Data Sheet ADRF6516
Rev. C | Page 5 of 29
TIMING DIAGRAMS
WRITE BIT MSB - 2
B2
LSB
t
DS
t
DH
t
LH
t
LS
t
PW
t
CLK
NOTES
1. THE FIRST DATA BIT DETERMINES WHETHER THE PART IS WRITING TO OR READING FROM THE INTERNAL 8-BIT REGISTER. FOR A WRITE
OPERATION, THE FIRST BIT SHOULD BE A LOGIC 1. THE 8-BIT WORD IS THEN WRITTEN TO THE DATA PIN ON CONSECUTIVE RISING
EDGES OF THE CLOCK.
CLK
LE
DATA B3 B7 MSB
B4 B5 B6
09422-003
Figure 2. Write Mode Timing Diagram
DON’ T CAREDON’ T CAREREAD BIT DON’ T CARE DON’T CARE DO N’T CARE DO N’T CARE DON' T CARE
B2LSB
CLK
LE
DATA
SDO
NOTES
1. T HE FI RS T DAT A BIT DE TERM INES WHETHER THE PART I S WRITI NG TO O R RE ADING FROM THE INTE RNAL 8-BIT RE GI S TER. FOR A READ
OPERATION, THE FIRST BIT SHOULD BE A LOGIC 0. THE 8-BIT WORD IS THEN REGISTERED AT THE SDO PIN ON CONSECUTIVE FALLING EDGES
OF THE CLOCK.
B3 B4 B5 B6 B7 MSB
DON’ T CARE
DON’ T CARE
tDS tDH
tLH
tLS
tPW
tCLK
tD
09422-004
Figure 3. Read Mode Timing Diagram
ADRF6516 Data Sheet
Rev. C | Page 6 of 29
ABSOLUTE MAXIMUM RATINGS
Table 2.
Parameter Rating
Supply Voltages, VPS, VPSD 3.45 V
ENBL, OFDS, LE, CLK, DATA, SDO
VPSD + 0.5 V
INP1, INM1, INP2, INM2 VPS + 0.5 V
OPP1, OPM1, OPP2, OPM2 VPS + 0.5 V
OFS1, OFS2 VPS + 0.5 V
GAIN VPS + 0.5 V
Internal Power Dissipation 1.25 W
θ
JA
(Exposed Pad Soldered to Board)
37.4°C/W
Maximum Junction Temperature 150°C
Operating Temperature Range −40°C to +85°C
Storage Temperature Range −65°C to +150°C
Lead Temperature (Soldering 60 sec) 300°C
Stresses at or above those listed under Absolute Maximum
Ratings may cause permanent damage to the product. This is a
stress rating only; functional operation of the product at these
or any other conditions above those indicated in the operational
section of this specification is not implied. Operation beyond
the maximum operating conditions for extended periods may
affect product reliability.
ESD CAUTION
Data Sheet ADRF6516
Rev. C | Page 7 of 29
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
09422-002
VPSD
COMD
LE
CLK
DATA
SDO
COM
VPS
OPP1
OPM1
COM
GAIN
VOCM
COM
OPM2
OPP2
COM
INP2
INM2
VPS
COM
OFDS
OFS2
VPS
ENBL
INP1
INM1
VPS
COM
VICM
OFS1
VPS
ADRF6516
TOP VIEW
(No t t o Scal e)
NOTES
1. CO NNE CT T HE E X P OSE D P ADDLE T O
A LOW IMP E DANCE GROUND P AD.
24
23
22
21
20
19
18
17
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
32
31
30
29
28
27
26
25
Figure 4. Pin Configuration
Table 3. Pin Function Descriptions
Pin No. Mnemonic Description
1 VPSD Digital Positive Supply Voltage: 3.15 V to 3.45 V.
2 COMD Digital Common. Connect to external circuit common using the lowest possible impedance.
3 LE Latch Enable. SPI programming pin. TTL levels: VLOW < 0.8 V, VHIGH > 2 V.
4 CLK SPI Port Clock. TTL levels: VLOW < 0.8 V, VHIGH > 2 V.
5 DATA SPI Data Input. TTL levels: VLOW < 0.8 V, VHIGH > 2 V.
6 SDO SPI Data Output. TTL levels: VLOW < 0.8 V, VHIGH > 2 V.
7, 9, 13, 19, 22, 28
COM
Analog Common. Connect to external circuit common using the lowest possible impedance.
8, 12, 16, 25, 29 VPS Analog Positive Supply Voltage: 3.15 V to 3.45 V.
10, 11, 30, 31 INP2, INM2,
INM1, INP1
Differential Inputs. 1600 Ω input impedance.
14 OFDS Offset Compensation Loop Disable. Pull high to disable the offset compensation loop.
15, 26 OFS2, OFS1 Offset Compensation Loop Capacitors. Connect capacitors to circuit common.
17, 18, 23, 24 OPP2, OPM2,
OPM1, OPP1
Differential Outputs. 30 Ω output impedance. Common-mode range is 0.7 V to 2.8 V; default is VPS/2.
20 VOCM Output Common-Mode Setpoint. Defaults to VPS/2 if left floating.
21 GAIN Analog Gain Control. 0 V to 1 V, 15.5 mV/dB gain scaling.
27 VICM Input Common-Mode Voltage. VPS/2 V reference. Use to reference the optimal common-mode drive
to the differential inputs.
32 ENBL Chip Enable. Pull high to enable.
EP Exposed Paddle. Connect the exposed paddle to a low impedance ground pad.
ADRF6516 Data Sheet
Rev. C | Page 8 of 29
TYPICAL PERFORMANCE CHARACTERISTICS
VPS = 3.3 V, TA = 25°C, ZLOAD = 1 kΩ, digital gain code = 111, unless otherwise noted.
–10
–5
0
5
10
15
20
25
30
35
40
45
50
0100 200 300 400 500 600 700 800 900 1000
GAI N ( dB)
V
GAIN
(mV)
BANDWIDTH = 31M Hz
–40°C
VPS = 3.15V, 3.3V, 3.45V
+85°C
VPS = 3.15V, 3.3V, 3.45V
+25°C
VPS = 3.15V, 3.3V, 3.45V
09422-005
Figure 5. In-Band Gain vs. VGAIN over Supply and Temperature
(Bandwidth Setting = 31 MHz)
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
5
10
15
20
25
30
35
40
45
50
110 100
GAI N ( dB)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
09422-006
Figure 6. Gain vs. Frequency over VGAIN (Bandwidth Setting = 31 MHz)
–0.25
–0.20
–0.15
–0.10
–0.05
0
0.05
0.10
0.15
0.20
0.25
0100 200 300 400 500 600 700 800 900 1000
GAIN MISMATCH (dB)
VGAIN (mV)
BANDWIDTH = 31M Hz
09422-007
Figure 7. Gain Matching vs. VGAIN (Bandwidth Setting = 31 MHz)
–5.0
–4.5
–4.0
–3.5
–3.0
–2.5
–2.0
–1.5
–1.0
–0.5
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0100 200 300 400 500 600 700 800 900 1000
GAI N E RROR (dB)
VGAIN (mV)
–40°C
VPS = 3.15V, 3.3V, 3.45V
+25°C
VPS = 3.15V, 3.3V, 3.45V
+85°C
VPS = 3.15V, 3.3V, 3.45V
09422-008
BANDWIDTH = 31M Hz
Figure 8. Gain Conformance vs. VGAIN over Supply and Temperature
(Bandwidth Setting = 31 MHz)
–2
–1
0
1
2
3
4
5
6
7
8
–15
–14
–13
–12
–11
–10
–9
–8
–7
–6
–5
0 5 10 15 20 25 30 35
AMPLITUDE ( dB)
GAIN STEP ( dB)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
DIGITAL G AIN = 111
DIGITAL G AIN = 011
09422-009
Figure 9. Gain Step and Gain Error vs. Frequency
(Bandwidth Setting = 31 MHz, VGAIN = 0 V)
8
9
10
11
12
13
14
–30
–25
–20
–15
–10
–5
0
0510 15 20 25 30 35
GAIN STEP ( dB)
AMPLITUDE ( dB)
FREQUENCY (MHz)
DIGITAL G AIN = 011
DIGITAL G AIN = 000
BANDWIDTH = 31M Hz
09422-010
Figure 10. Gain Step and Gain Error vs. Frequency
(Bandwidth Setting = 31 MHz, VGAIN = 0 V)
Data Sheet ADRF6516
Rev. C | Page 9 of 29
–20
–15
–10
–5
0
5
10
15
20
0 5 10 15 20 25 30 35 40
OP1dB (dBV)
GAI N ( dB)
DIGITAL G AIN = 111
BANDWIDTH = 31M Hz
DIGITAL G AIN = 000
09422-011
Figure 11. Output P1dB vs. Gain at 15 MHz (Bandwidth Setting = 31 MHz)
–10
–5
0
5
10
15
20
25
30
35
40
110 100
GAI N ( dB)
FREQUENCY (MHz)
09422-012
Figure 12. Frequency Response vs. Bandwidth Setting (Gain = 30 dB),
Log Scale
–10
–5
0
5
10
15
20
25
30
35
40
010 20 30 40
50 60 70 80 90 100
GAI N ( dB)
FREQUENCY (MHz)
09422-013
Figure 13. Frequency Response vs. Bandwidth Setting (Gain = 30 dB),
Linear Scale
20
22
24
26
28
30
32
34
36
38
40
246810 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
–40°C, VPS = 3.15V, 3.3V, 3.45V
+25°C,
VPS = 3.15V, 3.3V, 3.45V
+85°C,
VPS = 3.15V, 3.3V, 3.45V
GAI N ( dB)
V
GAIN
(mV)
BANDWIDTH = 31M Hz
09422-014
Figure 14. Frequency Response over Supply and Temperature
(Bandwidth Setting = 31 MHz, Gain = 30 dB)
0
100
200
300
400
500
600
700
800
900
1000
0.3 330
GROUP DELAY (ns)
FREQUENCY (MHz)
BW = 1MHz
BW = 5MHz
BW = 10MHz
BW = 20MHz
BW = 31MHz
50
GAI N = 20dB
09422-015
Figure 15. Group Delay vs. Frequency (Gain = 20 dB)
–2.0
–1.5
–1.0
–0.5
0
0.5
1.0
1.5
2.0
0.3 330
GROUP DELAY MISMATCH ( ns)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
GAI N = 40dB
GAI N = 20dB
09422-016
Figure 16. Group Delay Matching vs. Frequency
(Bandwidth Setting = 31 MHz)
ADRF6516 Data Sheet
Rev. C | Page 10 of 29
–5
–4
–3
–2
–1
0
1
2
3
4
5
0.2 0.4 0.6 0.8 1.0 1.2 1.4
GROUP DELAY MISMATCH ( ns)
FREQUENCY (MHz)
BANDWIDTH = 1M Hz
GAI N = 20dB
GAI N = 0dB
09422-017
Figure 17. IQ Group Delay Matching vs. Frequency
(Bandwidth Setting = 1 MHz)
–0.50
–0.25
0
0.25
0.50
0510 15
FREQUENCY (MHz)
FREQUENCY (MHz)
20 25 30
00.5 1.0 1.5 2.0 2.5 3.0
AMPLITUDE MISMATCH (dB)
BANDWIDTH = 30M Hz BANDWIDTH = 1M Hz
09422-018
Figure 18. IQ Amplitude Matching vs. Frequency
0
10
20
30
40
50
60
70
80
90
0510 15 20 25 30 35 40 45
HD2 AT 16M Hz ( dBc)
GAI N ( dB)
+25°C, VPS = 3.3V
+25°C, VPS = 3.15V
+25°C, VPS = 3.45V
+85°C, VPS = 3.3V
+85°C, VPS = 3.15V
+85°C, VPS = 3.45V
–40°C, V P S = 3.3V
–40°C, V P S = 3.15V
–40°C, V P S = 3.45V
09422-019
Figure 19. HD2 vs. Gain over Supply and Temperature
(Bandwidth Setting = 31 MHz, 1.5 V p-p, 8 MHz CW Fundamental Output)
30
40
50
60
70
80
90
0510 15 20 25 30 35 40 45
HD2 AT 16M Hz ( dBc)
GAI N ( dB)
VOCM = 0.9V
VOCM = 1.2V
VOCM = 1.4V
VOCM = 1.65V
09422-020
Figure 20. HD2 vs. Gain over Output Common-Mode Voltage
(Bandwidth Setting = 31 MHz, 1.5 V p-p, 8 MHz CW Fundamental Output)
0
10
20
30
40
50
60
70
80
90
0 5 10 15 20 25 30 35 40 45
HD3 AT 24M Hz ( dBc)
GAI N ( dB)
+25° C, VPS = 3.3V
+25° C, VPS = 3.15V
+25° C, VPS = 3.45V
+85° C, VPS = 3.3V
+85° C, VPS = 3.15V
+85° C, VPS = 3.45V
–40°C, VPS = 3.3V
–40°C, VPS = 3.15V
–40°C, VPS = 3.45V
09422-022
Figure 21. HD3 vs. Gain over Supply and Temperature
(Bandwidth Setting = 31 MHz, 1.5 V p-p, 8 MHz CW Fundamental Output)
30
40
50
60
70
80
90
0510 15 20 25 30 35 40 45
HD3 AT 24M Hz ( dBc)
GAI N ( dB)
VOCM = 0.9V
VOCM = 1.2V
VOCM = 1.4V
VOCM = 1.65V
09422-023
Figure 22. HD3 vs. Gain over Output Common-Mode Voltage
(Bandwidth Setting = 31 MHz, 1.5 V p-p, 8 MHz CW Fundamental Output)
Data Sheet ADRF6516
Rev. C | Page 11 of 29
50
55
60
65
70
75
80
0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50
HARMO NI C DISTORTION (dBc)
VICM (V)
GAI N = 0dB, HD2
GAI N = 0dB, HD3
GAI N = 10dB, HD2
GAI N = 10dB, HD3
09422-024
Figure 23. HD2 and HD3 vs. Input Common-Mode Voltage
(Bandwidth Setting = 31 MHz, 0.4 V p-p Input Level)
0510 15 20 25 30 35 40 45 50
OIP3 (dBV)
GAI N ( dB)
DIGITAL GAIN = 000
DIGITAL GAIN = 111
BANDWIDTH = 31M Hz
f1 = 14M Hz , f2 = 15MHz
09422-025
0
5
10
15
20
25
30
35
40
45
Figure 24. In-Band OIP3 vs. Gain (Bandwidth Setting = 31 MHz)
0 5 10 15 20 25 30 35 40 45 50
OIP3 (dBV)
GAI N ( dB)
BANDWIDTH = 31M Hz
f1 = 14M Hz , f2 = 15MHz
DIGITAL GAIN = 111
–40°C
+85°C
09422-026
0
5
10
15
20
25
30
35
40
45
+25°C
Figure 25. In-Band OIP3 vs. Gain over Temperature
(Bandwidth Setting = 31 MHz)
10
20
30
40
50
60
70
80
90
100
110
00.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
IM D3 ( dBc)
COMPOSITE OUTPUT VOLTAGE (V p-p)
GAI N = 30dB
GAI N = 20dB
GAI N = 10dB
GAI N = 0dB
09422-027
Figure 26. In-Band Third-Order Intermodulation Distortion
(Bandwidth Setting = 31 MHz, Digital Gain = 000)
0
10
20
30
40
50
60
70
80
90
100
IM D3 ( dBc)
GAI N = 40dB
GAI N = 30dB
GAI N = 20dB
GAI N = 10dB
GAI N = 0dB
00.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
COMPOSITE OUTPUT VOLTAGE (V p-p)
09422-028
Figure 27. In-Band Third-Order Intermodulation Distortion
(Bandwidth Setting = 31 MHz, Digital Gain = 111)
–170
–160
–150
–140
–130
–120
–110
–100
–90
–80
–70
–60
–50
–40
–30
–20
–10
0
10
20
30
40
50
60
70
–55 –45 –35 –25 –15 –5 515 25 35 45 55 65
IMD2 AT 15M Hz ( dBV)
INPUT LEVEL AT 115MHz AND 130MHz ( dBV/TONE)
OUT-O F-BAND I IP2
PREAMP GAIN = 3d B
PREAMP GAIN = 6d B
2:1 SLOPE
BANDWIDTH = 31M Hz
09422-029
Figure 28. Out-of-Band IIP2, IMD2 Tone at Midband
(Bandwidth Setting = 31 MHz)
ADRF6516 Data Sheet
Rev. C | Page 12 of 29
–160
–150
–140
–130
–120
–110
–100
–90
–80
–70
–60
–50
–40
–30
–20
–10
0
10
–55 –50 –45 –40 –35 –30 –25 –20 –15 –10 –5 05
IMD3 AT 15M Hz ( dBV)
INPUT LEVELAT 115MHz AND 215MHz (dBV/TONE)
3:1 SLOPE
BANDWIDTH = 31M Hz
OUT-O F-BAND I IP3
PREAMP GAIN = 3d B
PREAMP GAIN = 6d B
09422-030
Figure 29. Out-of-Band IIP3, IMD3 Tone at Midband
(Bandwidth Setting = 31 MHz)
10
15
20
25
30
35
40
45
50
55
60
–20 –10 010 20 30 40 50
NF ( dB)
GAI N ( dB)
DIGITAL G AIN = 000
DIGITAL G AIN = 100
DIGITAL G AIN = 110
DIGITAL G AIN = 111
09422-031
Figure 30. Noise Figure vs. Analog Gain over Digital Gain
(Bandwidth Setting = 31 MHz, Noise Figure at 1/2 Bandwidth)
20
25
30
35
40
45
50
–5 515 25 35 45
NF ( dB)
GAI N ( dB)
09422-032
1MHz
2MHz
4MHz
8MHz
16MHz
31MHz
Figure 31. Noise Figure vs. Gain over Bandwidth Setting
(Digital Gain = 111, Noise Figure at 1/2 Bandwidth)
–150
–155
–160
–130
–135
–140
–145
–110
–115
–120
–125
–20 –10 010 20 30 40 50
OUTPUT NOISE DENSITY (dBV/√Hz)
GAI N ( dB)
09422-033
DIGITAL G AIN = 000
DIGITAL G AIN = 100
DIGITAL G AIN = 110
DIGITAL G AIN = 111
Figure 32. Output Noise Density vs. Analog Gain over Digital Gain
(Bandwidth Setting = 31 MHz, Measured at 1/2 Bandwidth)
–150
–140
–145
–120
–125
–130
–135
–100
–105
–110
–115
–5 515 25 35 45 50
010 20 30 40
OUTPUT NOISE DENSITY (dBV/√Hz)
GAI N ( dB)
09422-034
1MHz
2MHz
4MHz
8MHz
16MHz
31MHz
Figure 33. Output Noise Density vs. Gain over Bandwidth Setting
(Digital Gain = 111, Measured at 1/2 Bandwidth)
–150
–145
–140
–135
–130
–125
–120
–115
–110
–105
–100
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
OUTPUT NOISE DENSITY (dBV/√Hz)
FREQUENCY (MHz)
GAI N = 20dB
GAI N = 0dB
GAI N = 40dB BANDWIDTH = 1M Hz
DIGITAL G AIN = 111
09422-052
Figure 34. Output Noise Density vs. Frequency
(Bandwidth Setting = 1 MHz, Digital Gain = 111)
Data Sheet ADRF6516
Rev. C | Page 13 of 29
–150
–145
–140
–135
–130
–125
–120
–115
–110
515 25 35 45 55 65 75 85 95
OUTPUT NOISE DENSITY (dBV/√Hz)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
DIGITAL G AIN = 111
GAI N = 20dB
GAI N = 0dB
GAI N = 40dB
09422-051
Figure 35. Output Noise Density vs. Frequency
(Bandwidth Setting = 31 MHz, Digital Gain = 111)
–45 –40 –35 –30 –25 –20 –15 –10 –5
OUTPUT NOISE DENSITY AT 15MHz (dBV/√Hz)
BLO CKE R LEV E L AT 150MHz (dBV rms)
09422-037
–145
–140
–135
–130
–125
–120
–115
–110
–105
–100
–95
–90
GAI N = 20dB
GAI N = 0dB
GAI N = 40dB
BANDWIDTH = 31M Hz
DIGITAL G AIN = 111
Figure 36. Output Noise Density vs. Input CW Blocker Level
(Bandwidth Setting = 31 MHz, Blocker at 150 MHz)
–40
–20
0
20
40
500
1000
1500
2000
2500
0 5 10 15 20 25 30
C
IN
(pF)
R
IN
(Ω)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
09422-038
Figure 37. Input Impedance vs. Frequency
(Bandwidth Setting = 31 MHz)
10
15
20
25
30
35
0
10
20
30
40
50
510 15 20 25 30
L
SERIES OUT
(n H)
R
SERIES OUT
(Ω)
FREQUENCY (MHz)
09422-039
BANDWIDTH = 31M Hz
Figure 38. Output Impedance vs. Frequency
(Bandwidth Setting = 31 MHz)
0
20
40
60
80
100
120
0 5 10 15 20 25 30
ISOLATION (dB)
FREQUENCY (MHz)
GAI N = 20dB
GAI N = 40dB
09422-040
GAI N = 0dB
BANDWIDTH = 31M Hz
Figure 39. Channel Isolation, Output to Output, vs. Frequency
(Bandwidth Setting = 31 MHz)
325
330
335
340
345
350
355
360
365
0510 15 20 25 30 35
I
SUPPLY
(mA)
BANDWIDTH ( M Hz )
DIGITAL G AIN = 000
DIGITAL G AIN = 111
09422-041
Figure 40. Current Consumption at Minimum and Maximum Digital Gain
vs. Bandwidth (Bandwidth Setting = 31 MHz, Gain = 30 dB)
ADRF6516 Data Sheet
Rev. C | Page 14 of 29
352
354
356
358
360
362
364
366
368
370
–40 –20 020 40 60 80 100
I
SUPPLY
(mA)
TEMPERATURE ( °C)
DIGITAL G AIN = 000
DIGITAL G AIN = 111
09422-042
BANDWIDTH = 31M Hz
Figure 41. Current Consumption at Minimum and Maximum Digital Gain
vs. Temperature (Bandwidth Setting = 31 MHz, Gain = 30 dB)
28MHz S I GNAL = 60mV p-p
TO 600mV p-p
200ns/DIV
20dB G AIN STEP
VGAIN = 750mV TO 450mV
BANDWIDTH = 31M Hz
09422-143
Figure 42. Gain Step Response
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
CMRR (dB)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
GAI N = 20dB
GAI N = 40dB
09422-144
Figure 43. Common-Mode Rejection Ratio (CMRR) vs. Frequency
(Bandwidth Setting = 31 MHz)
Data Sheet ADRF6516
Rev. C | Page 15 of 29
REGISTER MAP AND CODES
The filter frequency, preamplifier gain, postamplifier gain, and VGA maximum gain can be programmed using the SPI interface. Table 4
provides the bit map for the internal 8-bit register of the ADRF6516. The preamplifier, postamplifier, and VGA maximum gain code bits
(Bits[B3:B1]) are referred to elsewhere in this data sheet as Digital Gain Code 000 through Digital Gain Code 111.
Table 4. Register Map
MSB LSB
B8 B7 B6 B5 B4 B3 B2 B1
Filter frequency code Preamplifier gain
code
Postamplifier gain
code
VGA max gain
code
See Table 5 0 = 3 dB
1 = 6 dB
0 = 6 dB
1 = 12 dB
0 = 22 dB
1 = 28 dB
Table 5. Frequency Code vs. Corner Frequency Lookup Table
5-Bit Binary Frequency Code
1
Corner Frequency (MHz)
00000 No signal
00001
1
00010 2
00011 3
00100 4
00101 5
00110 6
00111 7
01000 8
01001 9
01010 10
01011 11
01100
12
01101 13
01110 14
01111 15
5-Bit Binary Frequency Code
1
Corner Frequency (MHz)
10000 16
10001
17
10010 18
10011 19
10100 20
10101 21
10110 22
10111 23
11000 24
11001 25
11010 26
11011 27
11100
28
11101 29
11110 30
11111 31
1 MSB first.
ADRF6516 Data Sheet
Rev. C | Page 16 of 29
THEORY OF OPERATION
The ADRF6516 consists of a matched pair of buffered, program-
mable filters followed by a cascade of two variable gain amplifiers
and output ADC drivers. The block diagram of a single channel
is shown in Figure 44.
The programmability of the bandwidth and of the pre- and post-
filtering gain through the SPI interface offers great flexibility
when coping with signals of varying levels in the presence of
noise and large, undesired signals nearby. The entire differential
signal chain is dc-coupled with flexible interfaces at the input
and output. The bandwidth and gain setting controls for the two
channels are shared, ensuring close matching of their magnitude
and phase responses. The ADRF6516 can be fully disabled
through the ENBL pin.
3dB/6dB
PREAMP
3dB/6dB
11dB/14dB
6dB/12dB
1MHz TO 31M Hz
PROG.
FILTERS 25dB
VGA
6dB/12dB
ADC
DRIVER
BASEBAND
INPUTS BASEBAND
OUTPUTS
GAIN AND FILTER
PROGRAMMING
SPI BUS ANALOG
GAI N CONT ROL
15mV/dB
25dB
VGA
OUTPUT
COMMON-MODE
CONTROL
SPI
INTERFACE
09422-046
Figure 44. Signal Path Block Diagram for a Single Channel of the ADRF6516
Filtering and amplification are fundamental operations in any
signal processing system. Filtering is necessary to select the
intended signal while rejecting out-of-band noise and inter-
ferers. Amplification increases the level of the desired signal
to overcome noise added by the system. When used together,
filtering and amplification can extract a low level signal of
interest in the presence of noise and out-of-band interferers.
Such analog signal processing alleviates the requirements on
the analog, mixed signal, and digital components that follow.
INPUT BUFFERS
The input buffers provide a convenient interface to the sensitive
filter sections that follow. They set a differential input impedance
of 1600 Ω and float to a common-mode voltage near VPS/2. The
inputs can be dc-coupled or ac-coupled. If using direct dc coupling,
the common-mode voltage presented to the inputs must be
approximately VPS/2 to maximize the input swing capacity.
For a 3.3 V supply, the common-mode voltage can range
from 1.1 V to 1.8 V while maintaining a >65 dBc HD3 for a
400 mV p-p input signal. The VICM pin provides the optimal
midsupply common-mode voltage and can be used as a refer-
ence for the driving circuit. The VICM voltage is not buffered
and must be sensed at a high impedance point to prevent it
from being loaded down.
The input buffers in both channels can be configured simul-
taneously for a gain of 3 dB or 6 dB through the SPI (see the
Register Map and Codes section). When configured for a 3 dB
gain, the buffers support a 400 mV p-p differential input level
with ~70 dBc harmonic distortion. For a 6 dB gain setting, the
buffers support 280 mV p-p inputs.
PROGRAMMABLE FILTERS
The integrated programmable filter is the key signal processing
function in the ADRF6516. The filters follow a six-pole Butter-
worth prototype response that provides a compromise between
band rejection, ripple, and group delay. The 0.5 dB bandwidth is
programmed from 1 MHz to 31 MHz in 1 MHz steps via the serial
programming interface (SPI), as described in the Programming
the Filters and Gains section.
The filters are designed so that the Butterworth prototype filter
shape and group delay responses vs. frequency are retained for
any bandwidth setting. Figure 45 and Figure 46 illustrate the
ideal six-pole Butterworth magnitude and group delay
responses, respectively. The group delay, τg, is defined as
τg = −∂φ/∂ω
where:
φ is the phase in radians.
ω = 2πf (the frequency in radians/sec).
Note that for a frequency scaled filter prototype, the absolute
magnitude of the group delay scales inversely with the band-
width; however, the shape is retained. For example, the peak
group delay for a 28 MHz bandwidth setting is 14× less than
for a 2 MHz setting (see Figure 46).
0
–20
–40
–60
–80
–100
–120
–140
–160
–1801M 10M 100M 1G
RELATI V E M AGNI TUDE (Hz )
FREQUENCY ( Hz )
09422-043
Figure 45. Sixth-Order Butterworth Magnitude Response for 0.5 dB
Bandwidths Programmed from 2 MHz to 29 MHz in 1 MHz Steps
Data Sheet ADRF6516
Rev. C | Page 17 of 29
500
400
300
200
100
0
–100
100k 1M 10M 100M
GRO UP DE LAY ( ns)
FREQUENCY ( Hz )
BW = 2MHz BW = 28MHz
14×
09422-044
Figure 46. Sixth-Order Butterworth Group Delay Response for
0.5 dB Bandwidths Programmed to 2 MHz and 28 MHz
The corner frequency of the filters is defined by RC products,
which can vary by ±30% in a typical process. Therefore, all the
parts are factory calibrated for corner frequency, resulting in
a residual ±15% corner frequency variation over the −40°C to
+85°C temperature range. Although absolute accuracy requires
calibration, the matching of RC products between the pair of
channels is better than 1% by observing careful design and
layout practices. Calibration and excellent matching ensure
that the magnitude and group delay responses of both channels
track together, a critical requirement for digital IQ-based
communication systems.
VARIABLE GAIN AMPLIFIERS (VGAs)
The cascaded VGAs are based on the Analog Devices, Inc.,
patented X-AMP® architecture, consisting of tapped 25 dB
attenuators followed by programmable gain amplifiers. The
X-AMP architecture generates a continuous linear-in-dB
monotonic gain response with low ripple. The analog gains of
both cascaded VGA sections are controlled through the high
impedance GAIN pin with an accurate slope of 15 mV/dB.
The gain response shown in Figure 47 shows the GAIN pin
voltage range and the absence of gain foldback at high VGAIN.
By changing the gains of both VGAs simultaneously, a more
gradual variation in noise and distortion is achieved. The fixed
gain following each of the variable gain sections can also be pro-
grammed to two different values to maximize dynamic range.
50
–10
0.3
–0.3
0
GAI N ( dB)
GAI N E RROR (dB)
V
GAIN
(V)
0
10
20
30
40
0.500.25 0.75 1.00 1.501.25 1.75 2.00 2.25 2.50 2.75 3.00
–0.2
–0.1
0
0.1
0.2
09422-049
15mV/dB
Figure 47. Linear-in-dB Gain Control Response of the X-AMP VGA Cascade
Showing Consistent Slope and Low Error
OUTPUT BUFFERS/ADC DRIVERS
The low impedance (30 Ω) output buffers of the ADRF6516
are designed to drive either ADC inputs or subsequent amplifier
stages. They are capable of delivering up to 1.5 V p-p composite
two-tone signals into 1 kΩ differential loads with >65 dBc
IMD3. The output common-mode voltage defaults to VPS/2,
but it can be adjusted from 700 mV to 2.8 V without loss of
drive capability by presenting the VOCM pin with the desired
common-mode voltage. The high input impedance of VOCM
allows the ADC reference output to be connected directly. Even
though the output common-mode voltage is adjustable and the
offset compensation loop can null the accumulated dc offsets
(see the DC Offset Compensation Loop section), it may still be
desirable to ac couple the outputs by selecting the coupling cap-
acitors according to the load impedance and desired bandwidth.
DC OFFSET COMPENSATION LOOP
In many signal processing applications, no information is
carried in the dc level. In fact, dc voltages and other low
frequency disturbances can often dominate the intended signal
and consume precious dynamic range in the analog path and
bits in the data converters. These dc voltages can be present
with the desired input signal or can be generated inside the
signal path by inherent dc offsets or other unintended signal-
dependent processes such as self-mixing or rectification.
Because the ADRF6516 is fully dc-coupled, it may be necessary
to remove these offsets to realize the maximum signal-to-noise
ratio (SNR). This can be achieved with ac coupling capacitors at
the input and output pins; however, large value capacitors with
low impedance values are required because the high-pass corners
must be <10 Hz. To address the issue of dc offsets, the ADRF6516
provides an offset compensation loop that nulls the output differ-
ential dc level, as shown in Figure 48. If the compensation loop
is not required, it can be disabled by pulling the OFDS pin high.
ADRF6516 Data Sheet
Rev. C | Page 18 of 29
GAIN
FROM
FILTERS
COFS OFSx
OFDS
50dB
VGA OUTPUT ADC
DRIVER
BASEBAND
OUTPUTS
09422-050
Figure 48. Offset Compensation Loop Operates Around the VGA
and Output Buffer
The offset compensation loop creates a high-pass corner, fHP,
that is superimposed on the normal Butterworth filter response.
Typically, fHP is many orders of magnitude lower than the lowest
programmed filter bandwidth so that there is no interaction
between them. Setting fHP is accomplished with capacitors,
COFS, from the OFS1 and OFS2 pins to ground. Because the
compensation loop works around the VGA sections, fHP is also
dependent on the total gain of the cascaded VGAs. In general,
the expression for fHP is given by
fHP (Hz) = 6.7 × (Post Filter Linear Gain/COFS (µF))
where Post Filter Linear Gain is expressed in linear terms, not
in decibels (dB), and is the gain following the filters, which
excludes the preamplifier gain of 1.4 (3 dB) or 2 (6 dB).
Note that fHP increases in proportion to the gain. For this reason,
COFS must be chosen at the highest operating gain to guarantee
that fHP is always below the maximum limit required by the system.
PROGRAMMING THE FILTERS AND GAINS
The 0.5 dB corner frequencies for both filters and the gains of
the preamplifiers and postamplifiers are programmed simulta-
neously through the SPI port. An 8-bit register stores the 5-bit
code for corner frequencies of 1 MHz through 31 MHz, as well
as the 1-bit codes for the preamplifier gain, the VGA maximum
gain, and the postamplifier gain (see Table 4). The SPI protocol
not only allows frequency and gain codes to be written to the
DATA pin, but it also allows the stored code to be read back via
the SDO pin.
The latch enable (LE) pin must first go to a Logic 0 for a read or
write cycle to begin. On the next rising edge of the clock (CLK),
a Logic 1 on the DATA pin initiates a write cycle, whereas a
Logic 0 on the DATA pin initiates a read cycle. In a write cycle,
the next eight CLK rising edges latch the desired 8-bit code, LSB
first. When LE goes high, the write cycle is completed and the
frequency and gain codes are presented to the filter and ampli-
fiers. In a read cycle, the next eight CLK falling edges present
the stored 8-bit code, LSB first. When LE goes high, the read
cycle is completed. Detailed timing diagrams are shown in
Figure 2 and Figure 3.
NOISE CHARACTERISTICS
The output noise behavior of the ADRF6516 depends on the gain
and bandwidth settings. Figure 49 and Figure 50 show the total
output noise spectral density vs. frequency for different band-
width settings and VGA gains.
–150
–145
–140
–135
–130
–125
–120
–115
–110
515 25 35 45 55 65 75 85 95
OUTPUT NOISE DENSITY (dBV/√Hz)
FREQUENCY (MHz)
BANDWIDTH = 31M Hz
DIGITAL G AIN = 111
GAI N = 20dB
GAI N = 0dB
GAI N = 40dB
09422-051
Figure 49. Total Output Noise Density with a 31 MHz Corner Frequency
for Three Different Gain Settings
–150
–145
–140
–135
–130
–125
–120
–115
–110
–105
–100
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
OUTPUT NOISE DENSITY (dBV/√Hz)
FREQUENCY (MHz)
GAI N = 20dB
GAI N = 0dB
GAI N = 40dB BANDWIDTH = 1M Hz
DIGITAL G AIN = 111
09422-052
Figure 50. Total Output Noise Density with a 1 MHz Corner Frequency
for Three Different Gain Settings
Both the filter sections and the VGAs contribute to the total noise
at the output. The filter contributes a noise spectral density profile
that is flat at low frequencies, peaks near the corner frequency, and
then rolls off as the filter poles roll off the gain and noise. The
magnitude of the noise spectral density contributed by the filter,
expressed in nV/√Hz, varies inversely with the square root of the
bandwidth setting, resulting in a total integrated noise in nV that is
nearly constant with bandwidth setting. At higher frequencies,
after the filter noise rolls off, the noise floor is set by the VGAs.
Each of the X-AMP VGA sections used in the ADRF6516 con-
tributes a fixed and flat noise spectral density to its respective
output, independent of the gain setting. Because the VGAs are
cascaded in the ADRF6516, the total noise contributed by the
VGAs at the output increases gradually with higher gain. This
is apparent in the noise floor variation at high frequencies at
different VGA gain settings.
Data Sheet ADRF6516
Rev. C | Page 19 of 29
The exact relationship depends on the programmed fixed gain of
the amplifiers. At minimum gain, only the last VGA contributes
to the −144 dBV/√Hz minimum noise floor, which is equivalent
to 63 nV/√Hz. At lower frequencies within the filter bandwidth
setting, the VGAs translate the filter noise directly to the output
by a factor equal to the gain following the filter.
At low values of VGA gain, the noise at the output is the flat
spectral density contributed by the last VGA. As the gain
increases, more noise from the filter and first VGA appears at
the output. Because the intrinsic filter noise density increases
at lower bandwidth settings, it is more pronounced than it is
at higher bandwidth settings. In either case, the noise density
asymptotically approaches the limit set by the VGAs at the
highest frequencies. For other values of VGA gain and bandwidth
setting, the detailed shape of the noise spectral density changes
according to the relative contributions of the filters and VGAs.
Because the noise spectral density outside the filter bandwidth
is limited by the VGA output noise, it may be necessary to use
an external, fixed-frequency, passive filter prior to analog-to-
digital conversion to prevent noise aliasing from degrading the
signal-to-noise ratio. A higher sampling rate relative to the maxi-
mum required ADRF6516 corner frequency setting reduces the
order and complexity of this external filter.
DISTORTION CHARACTERISTICS
The distortion performance of the ADRF6516 is similar to its
noise performance. The filters and the VGAs contribute to the
overall distortion and signal handling capabilities. Furthermore,
the front end must also cope with out-of-band signals that can be
larger than the in-band signals. These out-of-band signals are
filtered before reaching the VGA. It is important to understand
the signals presented to the ADRF6516 and to match these
signals with the input and output characteristics of the part.
When the gain is low, the distortion is typically limited by the
input section because the output is not driven to its maximum
capacity. When the gain is high, the distortion is likely limited
by the output section because the input is not driven to its
maximum capacity. An exception to this is when the input is
driven with a small desired signal in combination with a large
out-of-band signal. In this case, the out-of-band signal may
drive the input to distort. As long as the input is not overdriven,
the out-of-band signal is removed by the filter. A high VGA
gain is still needed to raise the small desired signal to a higher
level at the output. The overall distortion introduced by the part
depends on the input drive level, including the out-of-band
signals, and the desired output signal level.
As noted in the Input Buffers section, the input section can
handle a total signal level of 400 mV p-p for a 3 dB preamplifier
gain and 280 mV p-p for a 6 dB preamplifier gain with >70 dBc
harmonic distortion. This includes both in-band and out-of-band
signals.
To distinguish and quantify the distortion performance of the
input section, two different IP3 specifications are presented. The
first is called in-band IP3 and refers to a two-tone test where the
signals are inside the filter bandwidth. This is exactly the same
figure of merit familiar to communications engineers in which
the third-order intermodulation level, IMD3, is measured.
To quantify the effect of out-of-band signals, a new out-of-band
(OOB) IIP3 figure of merit is introduced. This test also involves
a two-tone stimulus; however, the two tones are placed out-of-
band so that the lower IMD3 product lands in the middle of the
filter pass band. At the output, only the IMD3 product is visible
because the original two tones are filtered out. To calculate the
OOB IIP3 at the input, the IMD3 level is referred to the input
by the overall gain. The OOB IIP3 allows the user to predict the
impact of out-of-band blockers or interferers at an arbitrary
signal level on the in-band performance. The ratio of the desired
input signal level to the input-referred IMD3 at a given blocker
level represents a signal-to-distortion limit imposed by the out-
of-band signals.
MAXIMIZING THE DYNAMIC RANGE
The role of the ADRF6516 is to increase the level of a variable
in-band signal while minimizing out-of-band signals. Ideally,
this is achieved without degrading the SNR of the incoming
signal or introducing distortion to the incoming signal.
The first goal is to maximize the output signal swing, which can
be defined by the ADC input range or the input signal capacity
of the next analog stage. For the complex waveforms often encoun-
tered in communication systems, the peak-to-average ratio, or
crest factor, must be considered when selecting the peak-to-peak
output. From the selected output signal and the maximum gain
of the ADRF6516, the minimum input level can be defined.
Lower signal levels do not yield the maximum output and suffer
a greater degradation in SNR.
As the input signal level increases, the VGA gain is reduced from
its maximum gain point to maintain the desired fixed output
level. The output noise, initially dominated by the filter, follows
the gain reduction, yielding a progressively better SNR. At some
point, the VGA gain drops sufficiently that the VGA noise
becomes dominant, resulting in a slower reduction in SNR from
that point. From the perspective of SNR alone, the maximum
input level is reached when the VGA reaches its minimum gain.
ADRF6516 Data Sheet
Rev. C | Page 20 of 29
Distortion must also be considered when maximizing the dynamic
range. At low and moderate signal levels, the output distortion
is constant and assumed to be adequate for the selected output
level. At some point, the input signal becomes large enough that
distortion at the input limits the system. The maximum tolerable
input signal depends on whether the input distortion becomes
unacceptably large or the minimum gain is reached.
The most challenging scenario in terms of dynamic range is the
presence of a large out-of-band blocker accompanying a weaker
in-band desired signal. In this case, the maximum input level is
dictated by the blocker and its inclination to cause distortion.
After filtering, the weak desired signal must be amplified to the
desired output level, possibly requiring maximum gain. Both
the distortion limits associated with the blocker at the input and
the SNR limits created by the weaker signal and higher gains are
present simultaneously. Furthermore, not only does the blocker
scenario degrade the dynamic range, it also reduces the range
of input signals that can be handled because a larger part of the
gain range is used to simply extract the weak desired signal
from the stronger blocker.
KEY PARAMETERS FOR QUADRATURE-BASED
RECEIVERS
The majority of digital communication receivers makes use of
quadrature signaling, in which bits of information are encoded
onto pairs of baseband signals that then modulate in-phase (I)
and quadrature (Q) sinusoidal carriers. Both the baseband and
modulated signals appear quite complex in the time domain with
dramatic peaks and valleys. In a typical receiver, the goal is to
recover the pair of quadrature baseband signals in the presence
of noise and interfering signals after quadrature demodulation.
In the process of filtering out-of-band noise and undesired inter-
ferers and restoring the levels of the desired I and Q baseband
signals, it is critical to retain their gain and phase integrity over
the bandwidth.
The ADRF6516 delivers flat in-band gain and group delay,
consistent with a six-pole Butterworth prototype filter, as
described in the Programmable Filters section. Furthermore,
careful design ensures excellent matching of these parameters
between the I and Q channels. Although absolute gain flatness
and group delay can be corrected with digital equalization,
mismatch introduces quadrature errors and intersymbol inter-
ference that degrade bit error rates in digital communication
systems.
Data Sheet ADRF6516
Rev. C | Page 21 of 29
APPLICATIONS INFORMATION
BASIC CONNECTIONS
Figure 51 shows the basic connections for a typical ADRF6516
application.
SUPPLY DECOUPLING
A nominal supply voltage of 3.3 V must be applied to the supply
pins. The supply voltage must not exceed 3.45 V or drop below
3.15 V. Each supply pin must be decoupled to ground with at
least one low inductance, surface-mount ceramic capacitor of
0.1 µF placed as close as possible to the ADRF6516 device.
The ADRF6516 has two separate supplies: an analog supply and
a digital supply. Take care to separate the analog and digital
supplies with a large surface-mount inductor of 33 µH. Decouple
each supply separately to its respective ground through a 10 μF
capacitor.
INPUT SIGNAL PATH
Each signal path has input buffers, accessed through the
INP1, INM1, INP2, and INM2 pins, that set a differential input
impedance of 1600 Ω. These inputs sit at a nominal common-
mode voltage around midsupply.
The inputs can be dc-coupled or ac-coupled. If using direct
dc coupling, the common-mode voltage, VCM, can range from
1.1 V to 1.8 V. The VICM pin can be used as a reference common-
mode voltage for driving a high impedance sensing node of the
preceding cascaded part (VICM has a 7 kΩ impedance).
For example, the high impedance VOCM input pin of the
ADRF6806 quadrature demodulator can be directly connected
to the VICM pin of the ADRF6516. This gives the ADRF6806
the optimal common-mode voltage reference to drive the
ADRF6516.
OUTPUT SIGNAL PATH
The low impedance (30 Ω) output buffers are designed to drive
a high impedance load, such as an ADC input or another amplifier
stage. The output pins—OPP1, OPM1, OPP2, and OPM2—sit
at a nominal output common-mode voltage of VPS/2, but can
be driven to a voltage of 0.7 V to 2.8 V by applying the desired
common-mode voltage to the high impedance VOCM pin.
DC OFFSET COMPENSATION LOOP ENABLED
When the dc offset compensation loop is enabled via the OFDS
pin, the ADRF6516 can null the output differential dc level. The
loop is enabled by pulling the OFDS pin low to ground. The
offset compensation loop creates a high-pass corner frequency,
which is proportional to the value of the capacitors that are
connected from the OFS1 and OFS2 pins to ground. For more
information about setting the high-pass corner frequency, see
the DC Offset Compensation Loop section.
COMMON-MODE BYPASSING
The ADRF6516 common-mode pins, VICM and VOCM, must
be decoupled to ground. Use at least one low inductance,
surface-mount ceramic capacitor with a value of 0.1 μF to
decouple the common-mode pins.
09422-053
VPSD
COMD
LE
CLK
DATA
SDO
COM
VPS
OPP1
OPM1
COM
GAIN
VOCM
COM
OPM2
OPP2
COM
INP2
INM2
VPSCOM
OFDS
OFS2
VPS
ENBL
INP1
INM1VPSCOM
VICM
OFS1
VPS
ADRF6516
VPS
VPSD
0.1µF
VPS
VPS
VPS
VPS
VPS
OUT P UT1 (+ )
INPUT1 (–)
0.1µF
INPUT1 (+ )
INPUT2 (+ )
INPUT2 (–)
OUT P UT1 (–)
OUT P UT2 (–)
OUT P UT2 (+ )
0.1µF
0.1µF
0.1µF
0.1µF
0.1µF
0.1µF
SERIAL
CONTROL
INTERFACE
VPS
0.1µF
Figure 51. Basic Connections
ADRF6516 Data Sheet
Rev. C | Page 22 of 29
SERIAL PORT CONNECTIONS
The ADRF6516 has a SPI port to control the gain and filter band-
width settings. Data can be written to the internal 8-bit register
and read from the register. It is recommended that low-pass RC
filtering be placed on the SPI lines to filter out any high frequency
glitches. See Figure 58, the evaluation board schematic, for an
example of a low-pass RC filter.
ENABLE/DISABLE FUNCTION
To enable the ADRF6516, the ENBL pin must be pulled high.
Driving the ENBL pin low disables the device, reducing current
consumption to approximately 9 mA at room temperature.
ERROR VECTOR MAGNITUDE (EVM) PERFORMANCE
Error vector magnitude (EVM) is a measure used to quantify
the performance of a digital radio transmitter or receiver by
measuring the fidelity of the digital signal transmitted or
received. Various imperfections in the link, such as magnitude
and phase imbalance, noise, and distortion, cause the
constellation points to deviate from their ideal locations.
In general, a receiver exhibits three distinct EVM limitations
vs. received input signal power. As signal power increases, the
distortion components increase.
• At large enough signal levels, where the distortion compo-
nents due to the harmonic nonlinearities in the device are
falling in-band, EVM degrades as signal levels increase.
• At medium signal levels, where the signal chain behaves
in a linear manner and the signal is well above any notable
noise contributions, EVM has a tendency to reach an opti-
mal level determined dominantly by either the quadrature
accuracy and IQ gain match of the signal chain or the
precision of the test equipment.
• As signal levels decrease, such that noise is a major con-
tributor, EVM performance vs. the signal level exhibits
a decibel-for-decibel degradation with decreasing signal
level. At these lower signal levels, where noise is the
dominant limitation, decibel EVM is directly proportional
to the SNR.
EVM TEST SETUP
The basic setup to test EVM for the ADRF6516 consisted of an
Agilent E4438C used as a signal source and a Hewlett-Packard
89410A vector signal analyzer (VSA) used to sample and calculate
the EVM of the signal. The E4438C IQ baseband differential
outputs drove the ADRF6516 inputs. The I and Q outputs of the
ADRF6516 were loaded with 1 kΩ differential impedances and
connected differentially to two AD8130 differential amplifiers to
convert the signals into single-ended signals. The single-ended
signals were connected to the input channels of the VSA.
EFFECT OF FILTER BANDWIDTH ON EVM
Take care when when selecting the filter bandwidth. In a digital
transceiver, the modulated signal is filtered by a pulse shaping
filter (such as a root-raised cosine filter) at both the transmit
and receive ends to guard against intersymbol interference (ISI).
If additional filtering of the modulated signal is done, the signal
must be within the pass band of the filter. When the corner
frequency of the ADRF6516 filter begins to encroach on the
modulated signal, ISI is introduced and degrades EVM, which
can lead to loss of signal lock.
Figure 52 shows that a digitally modulated QAM baseband
signal with a bandwidth at 9.45 MHz has excellent EVM even
at a filter corner frequency of 8 MHz. Further reduction in the
corner frequency leads to complete loss of lock. As RF input
power was swept, the ADRF6516 attained an EVM of less than
−45 dB over an input power range of approximately 20 dB.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
–25 –20 –15 –10 –5 0 5
GAI N V OLTAGE (V)
EVM (dB)
RF I NP UT POW E R ( dBm)
30MHz
15MHz
10MHz
9MHz
8MHz
GAI N V OLTAGE
09422-054
Figure 52. EVM vs. RF Input Power at Several Filter Corner Settings
(256-QAM, 14 MSPS Signal with α = 0.35; Output Differential Signal Level
Held to 700 mV p-p; OFDS Pulled High)
Figure 53 shows the degradation that a fixed filter corner has
on EVM as the signal bandwidth corner is increased in fine
increments until loss of signal lock occurs.
09422-055
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
3456 7 8910
EVM (dB)
SIGNAL BANDWI DTH CO RNER (MHz)
FILTER BANDW IDT H CORNER
Figure 53. EVM vs. Signal Bandwidth Corner at a Filter Corner of 5 MHz
and a 16-QAM Signal with α = 0.35
Data Sheet ADRF6516
Rev. C | Page 23 of 29
EFFECT OF OUTPUT VOLTAGE LEVELS ON EVM
Output voltage level can affect EVM greatly when the signal is
compressed. When changing the output voltage levels of the
ADRF6516, take care that the output signal is not in compres-
sion, which causes EVM degradation.
Figure 54 show EVM performance vs. RF input power for
several maximum differential I and Q output voltage levels
of 350 mV p-p up to 2.4 V p-p. For the lower maximum differ-
ential output voltage levels, the EVM is less than −45 dB over
approximately 20 dB of input power range.
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
–25 –20 –15 –10 –5 05
EVM (dB)
RF I NP UT POW E R ( dBm)
350mV p-p M AX
700mV p-p M AX
1500mV p-p M AX
2400mV p-p M AX
09422-056
Figure 54. EVM vs. RF Input Power at Several Output Maximum Differential
Voltage Levels (Filter Corner = 10 MHz, OFDS Pulled High)
For the largest tested maximum differential output voltage level
of 2.4 V p-p, the ADRF6516 begins to compress the signal. This
compression causes EVM to degrade, but it still remains below
−40 dB, albeit over a truncated input power range. At the high
end of the input power range, the signal is in full compression
and EVM is large. Given that the gain is near its minimum, the
input signal level must be lowered to bring the output signal out
of full compression and into the proper linear operating region.
EFFECT OF COFS VALUE ON EVM
When enabled, the dc offset compensation loop effectively
nulls any information below the high-pass corner set by the
COFS capacitor. However, loss of the low frequency information
of the modulated signal can degrade the EVM in some cases.
As the signal bandwidth becomes larger, the percentage of
information that is corrupted by the high-pass corner becomes
smaller. In such cases, it is important to select a COFS capacitor
that is large enough to minimize the high-pass corner frequency,
which prevents loss of information and degraded EVM.
Figure 55 shows degradation of the EVM vs. RF input power as
the COFS capacitor value becomes smaller, which increases the
high-pass corner for the dc offset compensation loop.
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
–35 –30 –25 –20 –15 –10 –5 0 5
EVM (dB)
RF I NP UT POW E R ( dBm)
C
OFS
= 1µF
C
OFS
= 220nF
C
OFS
= 1nF
09422-057
Figure 55. EVM vs. RF Input Power at Several COFS Values (Filter Corner = 10 MHz,
256-QAM, 14 MSPS Signal with α = 0.35; Output Differential Signal Level
Held to 700 mV p-p; OFDS Pulled Low)
Figure 56 shows the effect that COFS has on several modulated
signal bandwidths. The modulated bandwidth was swept while
using 1000 pF and 1 µF values for COFS. Total gain was set to
15 dB, so the high-pass filter corner of the 1000 pF capacitor is
26.67 kHz, and the high-pass filter corner of the 1 µF capacitor
is 26.67 Hz. It is recommended that at moderate signal band-
widths, a 1 µF capacitor for COFS be used to obtain the best EVM
when using the dc offset compensation loop.
–50
–45
–40
–35
–30
–25
–20
–15
–10
–5
0
012345678910
EVM (dB)
SIGNAL BANDWI DTH CO RNE R ( M Hz )
COFS = 1µF
COFS = 1000p F
09422-058
Figure 56. EVM vs. Signal Bandwidth Corner with COFS = 1 µF
and COFS = 1000 pF (Filter Corner = 10 MHz)
ADRF6516 Data Sheet
Rev. C | Page 24 of 29
EVALUATION BOARD
An evaluation board is available for testing the ADRF6516.
The evaluation board schematic is shown in Figure 58. Table 6
provides the component values and suggestions for modifying
the component values for the various modes of operation.
EVALUATION BOARD CONTROL SOFTWARE
The ADRF6516 evaluation board is controlled through the
parallel port on a PC. The parallel port is programmed via the
ADRF6516 evaluation software. This software controls the filter
corner frequency, as well as the minimum and maximum gains
for each amplifier in the ADRF6516. For information about the
register map, see Table 4 and Table 5. For information about SPI
port timing and control, see Figure 2 and Figure 3.
After the evaluation software is downloaded and installed, start
the basic user interface to program the filter corner and gain
values (see Figure 57).
To program the filter corner, do one of the following:
Click the arrow in the Frequency Select section of the
window, select the desired corner frequency from the
menu, and click Write Bits.
Click Freq +1 MHz or Freq −1 MHz to increment or
decrement the corner frequency in 1 MHz steps from
the current corner frequency.
To program the preamplifier gain, the VGA maximum gain,
and the postamplifier gain, move the slider switch in the
appropriate section of the window to the desired gain.
The preamplifier gain can be set to 3 dB or 6 dB.
The VGA maximum gain can be set to 22 dB or 28 dB.
The postamplifier gain can be set to 6 dB or 12 dB.
When the user clicks the Write Bits button, a write operation is
executed, immediately followed by a read operation. The updated
information is displayed in the Current Pre-Amp Gain, Current
Frequency, Current VGA Max Gain, and Current Post-Amp
Gain fields.
When the parallel port is updated with a read/write operation,
the current cumulative maximum gain of all the amplifiers is
displayed in the Maximum Gain field. (The analog VGA gain
is not included in this value.)
Because the speed of the parallel port varies from PC to PC,
the Clock Stretch function can be used to change the effective
frequency of the CLK line. The CLK line has a scalar range from
1 to 10; 10 is the fastest speed, and 1 is the slowest.
09422-060
Figure 57. ADRF6516 Evaluation Software
Data Sheet ADRF6516
Rev. C | Page 25 of 29
SCHEMATICS AND ARTWORK
ADRF6516
VPS
P2
VPSD
C4
0.1µF
VPS
VPS
VPS
R3
10kΩ
P4
VPS VICM
C12
0.1µF C14
1000pF
C5
0.1µF
VPS
R12
OPEN
R11
OPEN
R14
OPEN
R13
OPEN
R37
OPEN
R20
0Ω
R19
0Ω
T3
R41
0Ω
R39
OPEN
VPOSD
C1
10µF C2
10µF
L1
L2
33µH
33µH
VPOSDIG_VPOS
VPOS
COMCOMD
C20
100nF
R8
0Ω
C19
100nF
C16
0.1µF R7
0Ω
R5
0Ω
R6
0Ω
C22
R10
0Ω
C21
100nF
100nF
C18
0.1µF
C17
0.1µF
R9
0Ω
OPP1
OPM1_SE_P
OPP2
OPM2_SE_P
R35
0Ω
C24
0.1µF
C23
0.1µF
R51
OPEN
R52
OPEN
R42
0Ω
R38
OPEN
R36
0Ω
T4
R40
OPEN
VPS
C15
0.1µF
C13
1000pF
R32
R46
OPEN
R49
0Ω0Ω
C3
0.1µF
R53
0Ω C7
100nF
C8
100nF
C11
0.1µF
R54
0Ω
R50
0Ω
R44
OPEN
T2
INP2
INM2_SE_P
INM1_SE_P
INP1
R31
0Ω
R45
OPEN
R47
1
3
4
6
2
1
3
4
6
2
0Ω
C6
0.1µF
R56
OPEN
R55
OPEN R57
0Ω
C9
100nF
C10
100nF
R58
0Ω
T1
R48
0Ω
R43
OPEN
LE
CLK DATA
SDO
C29
330pF
R29
100Ω
R30
100Ω
R33
C30
330pF
R34
0Ω
0Ω
R1
10kΩ
C27
0.1µF
VGAIN
VGAIN
VOCM
VOCM
09422-061
VPSD
COMD
LE
CLK
DATA
SDO
COM
VPS
OPP1
OPM1
COM
GAIN
VOCM
COM
OPM2
OPP2
COM
INP2
INM2
VPSCOM
OFDS
OFS2
VPS
INP1
INM1VPSCOM
VICM
OFS1
VPS
ENBL
Figure 58. Evaluation Board Schematic
ADRF6516 Data Sheet
Rev. C | Page 26 of 29
56 55 54 53 52 51 50 49
15 16 17 18 19 20 21 22
1
2
3
4
5
6
7
835
36
37
38
39
40
41
42
PD7_FD15
PD4_FD12
PD6_FD14
PD5_FD13
GND
CLKOUT
GND
VCC
PA5_FIFOARD1
PA2_SLOE
RESET_N
PA3_WU2
PA4_FIFOARD0
PA6_PKTEND
PA7_FLAGD_SCLS_N
GND
VCC
SDA
PB4_FD4
PB3_FD3
PB0_FD0
SCL
PB1_FD1
PB2_FD2
DPLUS
XTALOUT
XTALIN
RDY1_SLWR
AVCC
AVCC
AGND
RDY0_SLRD
CY7C68013A-56LTXC
U4 LE
9DMINUS
10 AGND
11 VCC
12 GND
13 IFCLK
14 RESERVED
23
PB5_FD5
24
PB6_FD6
27
VCC
25
PB7_FD7
26
GND
28
GND
29
30
31
32
33
34
CTL1_FLAGB
PA1_INT1_N
CTL0_FLAGA
CTL2_FLAGC
VCC
PA0_INT0_N
48 47 46 45 44 43
WAKEUP
VCC
PD0_FD8
PD1_FD9
PD3_FD11
PD2_FD10
CLK
DATA
3V3_USB
3V3_USB
3V3_USB
C48
10pF
C49
0.1µF
3V3_USB
3V3_USB
R61
2kΩ
CR2
3V3_USB
R64
0Ω
C37
0.1µF
C45
0.1µF
R62
100kΩ
3V3_USB
Y1
24MHz
3
4 2
1
C54
22pF C51
22pF
1
2
3
4
5
G1
G2
G3
G4
5V_USB
P5
1
2
3
4
5
6
7
8
A0
A1
A2
GND
SDA
SCL
WC_N
VCC
3V3_USB
3V3_USB
24LC64-I_SN
U2 ADP3334
U3
1 8
2
3
4
7
6
5
OUT1
OUT2
FB
NC
IN2
IN1
SD
GND
C47
1.0µF R65
2kΩ
CR1
5V_USB
R69
78.7kΩ
C50
1000pF
R70
140kΩ
C52
1.0µF
3V3_USB
DGND
C35
0.1µF
C42
0.1µF C36
0.1µF
C41
0.1µF
C40
0.1µF C44
0.1µF C46
0.1µF
3V3_USB
R60
2kΩ
R59
2kΩ C38
10pF
C39
0.1µF
SDO
09422-159
Figure 59. USB Evaluation Board Schematic
Data Sheet ADRF6516
Rev. C | Page 27 of 29
09422-062
Figure 60. Top Layer Silkscreen
09422-063
Figure 61. Component Side Layout
Table 6. Evaluation Board Configuration Options
Components Function Default Conditions
C1, C2, C4, C5, C11, C12, C15,
C16, L1, L2, R2
Power supply and ground decoupling. Nominal supply decoupling
consists of a 0.1 µF capacitor to ground.
C1, C2 = 10 µF (Size 1210)
C4, C5, C11, C12, C15, C16 =
0.1 µF (Size 0402)
L1, L2 = 33 µH (Size 1812)
R2 = 1 kΩ (Size 0402)
T1, T2, C3, C6, C7 to C10, R31,
R32, R43 to R58
Input interface. Input SMAs INP1, INM1_SE_P, INP2_SE_P, and INM2
are used to drive the part differentially by bypassing the baluns.
Using only INM1_SE_P and INP2_SE_P in conjunction with the baluns
enables single-ended operation. The default configuration of the
evaluation board is for single-ended operation.
T1 and T2 are 8:1 impedance ratio baluns that transform a single-
ended signal in a 50 Ω system into a balanced differential signal in a
400 Ω system.
R31, R32, R47, R48, R49, and R50 are populated for appropriate balun
interface. R51 to R58 are provided for generic placement of matching
components.
To bypass the T1 and T2 baluns for differential interfacing, remove
the balun interfacing resistors R31, R32, R47, R48, R49, and R50, and
populate R43, R44, R45, and R46 with 0 Ω resistors.
T1, T2 = ADT8-1T+ (Mini-Circuits)
C3, C6 = 0.1 µF (Size 0402)
C7 to C10 = 100 nF (Size 0602)
R31, R32, R47 to R50, R53, R54,
R57, R58 = 0 Ω (Size 0402)
R43 to R46, R51, R52, R55, R56 =
open (Size 0402)
T3, T4, C19 to C24,
R7 to R14, R19, R20,
R35 to R42
Output interface. Output SMAs OPP1_SE_P, OPM1, OPP2, and
OPM2_SE_P are used to obtain differential signals from the part
when the output baluns are bypassed. Using OPP1_SE_P,
OPM2_SE_P, and the baluns, the user can obtain single-ended output
signals. The default configuration of the evaluation board is for
single-ended operation.
T3 and T4 are 8:1 impedance ratio baluns that transform a differential
signal in a 400 Ω system into a single-ended signal in a 50 Ω system.
R7, R8, R9, R10, R19, R20, R35, R36, R41, and R42 are populated for
appropriate balun interface. R7 to R14 are provided for generic
placement of matching components.
To bypass the T3 and T4 baluns for differential interfacing, remove
the balun interfacing resistors R19, R20, R35, R36, R41, and R42, and
populate R37, R38, R39, and R40 with 0 Ω resistors.
T3, T4 = ADT8-1T+ (Mini-Circuits)
C19 to C22 = 100 nF (Size 0402)
C23, C24 = 0.1 µF (Size 0402)
R7 to R10, R19, R20, R35, R36,
R41, R42 = 0 Ω (Size 0402)
R11 to R14, R37 to R40 = open
(Size 0402)
ADRF6516 Data Sheet
Rev. C | Page 28 of 29
Components Function Default Conditions
P2 Enable interface. The ADRF6516 is powered up by applying a logic
high voltage to the ENBL pin (Jumper P2 is connected to VPS).
P2 = installed for enable
P1, C28, C29, R1, R29, R30, R33,
R34
Serial control interface. The digital interface sets the corner
frequency, the preamplifier gain, the postamplifier gain, and the VGA
maximum gain of the device using the serial interface via the LE, CLK,
DATA, and SDO pins. RC filter networks are provided on the CLK and
LE lines to filter the PC signals. CLK, DATA, and LE signals can be
observed via SMB connectors for debug purposes.
P1 = installed
R1 = 10 kΩ (Size 0402)
C28, C29 = 330 pF (Size 0402)
R29, R30 = 100 Ω (Size 0402)
R33, R34 = 0 Ω (Size 0402)
P4, C13, C14, R3 DC offset compensation loop. The dc offset compensation loop is
enabled (low) with Jumper P4. When enabled, the C13 and C14
capacitors are connected to circuit common. The high-pass corner
frequency is expressed as follows:
fHP (Hz) = 6.7 × (Post Filter Linear Gain/COFS (µF))
P4 = installed
C13, C14 = 1000 pF (Size 0402)
R3 = 10 kΩ (Size 0402)
C27
Input common-mode setpoint. The input common-mode voltage
can be set externally when applied to the VICM pin. If the VICM pin is
left open, the input common-mode voltage defaults to VPS/2.
C27 = 0.1 µF (Size 0402)
C18, R6 Output common-mode setpoint. The output common-mode voltage
can be set externally when applied to the VOCM pin. If the VOCM pin
is left open, the output common-mode voltage defaults to VPS/2.
C18 = 0.1 µF (Size 0402)
R6 = 0 Ω (Size 0402)
C17, R5 Analog gain control. The range of the GAIN pin is from 0 V to 1 V,
creating a gain scaling of 15 mV/dB.
C17 = 0.1 µF (Size 0402)
R5 = 0 Ω (Size 0402)
U2, U3, U4, P5
Cypress Microcontroller, EEPROM, and LDO
U2 = Microchip MICRO24LC64
U3 = Analog Devices
ADP3334ACPZ
U4 = Cypress Semiconductor
CY7C68013A-56LTXC
P5 = Mini USB connector
C35, C36, C40, C41, C42, C44,
C46
3.3 V supply decoupling. Several capacitors are used for decoupling
on the 3.3 V supply.
C35, C36, C40, C41, C42, C44, C46
= 0.1 µF (0402)
C48, C49, C45, C56, C57, C58,
R59, R60, R61, R62, R64, CR2
Cypress and EEPROM components. C57, C48 = 10 pF (0402)
C56, C58, C45, C49 = 0.1 µF
(0402)
R59, R60, R61 = 2 kΩ (0402)
R62, R64 = 100 kΩ (0402)
CR2 = ROHM SML-21OMTT86
C47, C50, C52, R65, R69, R70,
CR1
LDO components
C47, C52 = 1 µF (0402)
C50 = 1000 pF (0402)
R65 = 2 kΩ (0402)
R69 = 78.7 kΩ (0402)
R70 = 140 kΩ (0402)
CR1 = ROHM SML-21OMTT86
Y1, C51, C54 Crystal oscillator and components. 24 MHz crystal oscillator. Y1 = NDK NX3225SA-24MHz
C51, C54 = 22 pF (0402)
Data Sheet ADRF6516
Rev. C | Page 29 of 29
OUTLINE DIMENSIONS
3.25
3.10 SQ
2.95
0.80
0.75
0.70
1
0.50
BSC
BOTTOM VIEW
TOP VIEW
PIN 1
INDICATOR
32
9
16
17
24
25
8
0.05 MAX
0.02 NOM
0.20 REF
COPLANARITY
0.08
0.30
0.25
0.18
5.10
5.00 SQ
4.90
0.50
0.40
0.30
0.25 MIN
02-22-2017-A
COMPLIANT TO JEDEC STANDARDS MO-220-WHHD
PKG-003898
SEATING
PLANE
EXPOSED
PAD
SIDE VIEW
PIN 1
INDICATOR AREA OPTIONS
(SEE DETAIL A)
DETAIL A
(JEDEC 95)
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
SECTION OF THIS DATA SHEET.
Figure 62. 32-Lead Lead Frame Chip Scale Package [LFCSP]
5 mm × 5 mm Body and 0.75 mm Package Height
(CP-32-7)
Dimensions shown in millimeters
ORDERING GUIDE
Model1 Temperature Range Package Description Package Option
ADRF6516ACPZ-R7 −40°C to +85°C 32-Lead LFCSP, 7” Tape and Reel CP-32-7
ADRF6516-EVALZ Evaluation Board
1 Z = RoHS Compliant Part.
©2010–2017 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D09422-0-8/17(C)