AD9670 [ADI]
Octal Ultrasound AFE with Digital Demodulator;型号: | AD9670 |
厂家: | ADI |
描述: | Octal Ultrasound AFE with Digital Demodulator |
文件: | 总53页 (文件大小:828K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
Octal Ultrasound AFE
with Digital Demodulator
AD9670
Data Sheet
FEATURES
GENERAL DESCRIPTION
8 channels of LNA, VGA, antialiasing filter, ADC, and digital
demodulator/decimator
Low power
The AD9670 is designed for low cost, low power, small size, and
ease of use for medical ultrasound applications. It contains eight
channels of a VGA with an LNA, a CW harmonic rejection I/Q
demodulator with programmable phase rotation, an antialiasing
filter, an ADC, and a digital demodulator and decimator for data
processing and bandwidth reduction.
150 mW per channel, time gain compensation (TGC) mode,
40 MSPS
62.5 mW per channel, continuous wave (CW) mode;
<30 mW in power-down mode
Each channel features a maximum gain of up to 52 dB, a fully
differential signal path, and an active input preamplifier termination.
The channel is optimized for high dynamic performance and
low power in applications where a small package size is critical.
10 mm × 10 mm, 144-ball CSP_BGA
TGC channel, input referred noise voltage: 0.82 nV/√Hz,
maximum gain
Flexible power-down modes
The LNA has a single-ended-to-differential gain that is selectable
through the serial port interface (SPI). Assuming a 15 MHz noise
bandwidth (NBW) and a 21.6 dB LNA gain, the LNA input SNR
is 94 dB. In CW Doppler mode, each LNA output drives an I/Q
demodulator that has independently programmable phase
rotation with 16 phase settings.
Fast recovery from low power standby mode: <2 μs
Low noise preamplifier (LNA)
Input noise voltage: 0.78 nV/√Hz, gain = 21.6 dB
Programmable gain: 15.6 dB/17.9 dB/21.6 dB
0.1 dB input compression point: 1.00 V p-p/0.75 V p-p/
0.45 V p-p
Flexible active input impedance matching
Variable gain amplifier (VGA)
Attenuator range: 45 dB, linear-in-dB gain control
Postamplifier gain (PGA): 21 dB/24 dB/27 dB/30 dB
Antialiasing filter
Programmable, second-order low-pass filter from 8 MHz to
18 MHz or 13.5 MHz to 30 MHz and high-pass filter
Analog-to-digital converter (ADC)
Signal-to-noise ratio (SNR): 75 dB, 14 bits up to 125 MSPS
Configurable serial low voltage differential signaling (LVDS)
CW mode harmonic rejection I/Q demodulator
Individual programmable phase rotation
Dynamic range per channel: >160 dBFS/√Hz
Close in SNR: 156 dBc/√Hz, 1 kHz offset, −3 dBFS
Digital demodulator/decimator
Power-down of individual channels is supported to increase
battery life for portable applications. Standby mode allows quick
power-up for power cycling. In CW Doppler operation, the
VGA, antialiasing filter, and ADC are powered down. The ADC
contains several features designed to maximize flexibility and
minimize system cost, such as a programmable clock, data
alignment, and programmable digital test pattern generation.
The digital test patterns include built-in fixed patterns, built-in
pseudorandom patterns, and custom user-defined test patterns
entered via the SPI.
I/Q demodulator with programmable oscillator FIR
decimation filter
APPLICATIONS
Medical imaging/ultrasound
Nondestructive testing (NDT)
Rev. A
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 registeredtrademarks arethe property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
Technical Support
©2016 Analog Devices, Inc. All rights reserved.
www.analog.com
AD9670* PRODUCT PAGE QUICK LINKS
Last Content Update: 08/08/2017
COMPARABLE PARTS
View a parametric search of comparable parts.
DESIGN RESOURCES
• AD9670 Material Declaration
• PCN-PDN Information
• Quality And Reliability
• Symbols and Footprints
EVALUATION KITS
• AD9670 Evaluation Board
DOCUMENTATION
Data Sheet
DISCUSSIONS
View all AD9670 EngineerZone Discussions.
• AD9670: Octal Ultrasound AFE with Digital Demodulator
Data Sheet
SAMPLE AND BUY
Visit the product page to see pricing options.
REFERENCE MATERIALS
Press
TECHNICAL SUPPORT
• Industry’s First Octal Ultrasound Receiver with Digital I/Q
Demodulator and Decimation Filter Reduces Processor
Overhead in Ultrasound Systems
Submit a technical question or find your regional support
number.
• Industry’s First Octal Ultrasound Receiver with JESD204B
Serial Interface Reduces Data I/O Routing and Simplifies
Ultrasound System Design
DOCUMENT FEEDBACK
Submit feedback for this data sheet.
• Low Cost, Octal Ultrasound Receiver with On-Chip RF
Decimator and JESD204B Serial Interface
This page is dynamically generated by Analog Devices, Inc., and inserted into this data sheet. A dynamic change to the content on this page will not
trigger a change to either the revision number or the content of the product data sheet. This dynamic page may be frequently modified.
AD9670
Data Sheet
TABLE OF CONTENTS
Features .............................................................................................. 1
CW Doppler Operation............................................................. 33
Digital Demodulator/Decimator.................................................. 35
Vector Profile .............................................................................. 35
RF Decimator.............................................................................. 36
Baseband Demodulator and Decimator.................................. 37
Digital Test Waveforms.............................................................. 38
Digital Block Power Saving Scheme ........................................ 38
Serial Port Interface (SPI).............................................................. 39
Hardware Interface..................................................................... 39
Memory Map .................................................................................. 41
Reading the Memory Map Table.............................................. 41
Reserved Locations .................................................................... 41
Default Values............................................................................. 41
Logic Levels................................................................................. 41
Recommended Startup Sequence ............................................ 41
Memory Map Register Descriptions........................................ 51
Outline Dimensions....................................................................... 52
Ordering Guide .......................................................................... 52
Applications....................................................................................... 1
General Description......................................................................... 1
Revision History ............................................................................... 2
Functional Block Diagram .............................................................. 3
Specifications..................................................................................... 4
AC Specifications.......................................................................... 4
Digital Specifications ................................................................... 7
Switching Specifications .............................................................. 8
Timing Diagrams.......................................................................... 9
Absolute Maximum Ratings.......................................................... 11
Thermal Impedance................................................................... 11
ESD Caution................................................................................ 11
Pin Configuration and Function Descriptions........................... 12
Typical Performance Characteristics ........................................... 15
TGC Mode Characteristics ....................................................... 15
CW Doppler Mode Characteristics ......................................... 19
Theory of Operation ...................................................................... 20
TGC Operation........................................................................... 20
Analog Test Signal Generation ................................................. 33
REVISION HISTORY
2/16—Revision A: Initial Version
Rev. A | Page 2 of 52
Data Sheet
AD9670
FUNCTIONAL BLOCK DIAGRAM
AVDD1 AVDD2
PDWN STBY
DVDD
DRVDD
CWQ+
CWQ–
CWI+
CWI–
LO-A TO LO-H
CWD AND I/Q
DEMODULATOR
LOSW-A TO LOSW-H
LI-A TO LI-H
DOUTA+ TO DOUTH+
DOUTA– TO DOUTH–
14-BIT
ADC
DEMODULATOR/
DECIMATOR
LNA
SERIALIZER
VGA
LVDS
AAF
LG-A TO LG-H
AD9670
8 CHANNELS
FCO+
FCO–
DCO+
DCO–
SERIAL
PORT
INTERFACE
DATA
RATE
MULTIPLIER
LO
REFERENCE
NCO
GENERATION
Figure 1.
Rev. A | Page 3 of 52
AD9670
Data Sheet
SPECIFICATIONS
AC SPECIFICATIONS
AVDD1 = 1.8 V, AVDD2 = 3.0 V, DVDD = 1.4 V, DRVDD = 1.8 V, 1.0 V internal ADC reference, full temperature range (0°C to 85°C), fIN
5 MHz, local oscillator (LO) band mode, RS = 50 Ω, RFB = ∞ (unterminated), LNA gain = 21.6 dB, LNA bias = midhigh, PGA gain = 27 dB,
analog gain control, VGAIN = (GAIN+) − (GAIN−) = 1.6 V, antialiasing filter, low-pass filter (LPF) cutoff = fSAMPLE/3 in Mode I/Mode II,
antialiasing filter LPF cutoff = fSAMPLE /4.5 in Mode III/Mode IV, high-pass filter (HPF) cutoff = LPF cutoff/12.00, Mode I = fSAMPLE = 40 MSPS,
Mode II = fSAMPLE = 65 MSPS, Mode III = fSAMPLE = 80 MSPS, Mode IV = 125 MSPS, radio frequency (RF) decimator bypassed, digital demod-
ulator and baseband decimator bypassed, digital high-pass filter bypassed, low power LVDS mode, unless otherwise noted. All gain setting
options are listed, which can be configured via SPI registers, and all power supply currents and power dissipations are listed for the four mode
settings (Mode I, Mode II, Mode III, and Mode IV), respectively, via slashes in Table 1.
=
Table 1.
Parameter1
LNA CHARACTERISTICS
Gain
Test Conditions/Comments
Min
Typ
Max
Unit
Single-ended input to differential output
Single-ended input to single-ended output
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
15.6/17.9/21.6
9.6/11.9/15.6
1.00
0.75
0.45
1.20
0.90
0.60
2.2
dB
dB
0.1 dB Input Compression Point
1 dB Input Compression Point
V p-p
V p-p
V p-p
V p-p
V p-p
V p-p
V
Input Common Mode (LI-x, LG-x)
Output Common Mode
LO-x
Switch off
Switch on
Switch off
Switch on
RFB = 300 Ω
RFB = 1350 Ω
High-Z
1.5
High-Z
1.5
50
200
6
Ω
V
Ω
V
Ω
Ω
kΩ
pF
LOSW-x
Input Resistance (LI-x)
Input Capacitance (LI-x)
Input Noise Voltage
20
RS = 0 Ω
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
0.83
0.82
0.78
2.6
nV/√Hz
nV/√Hz
nV/√Hz
pA/√Hz
Input Noise Current
FULL CHANNEL (TGC) CHARACTERISTICS
Antialiasing Filter Low-Pass Cutoff
−3 dB, programmable, low band mode
−3 dB, programmable, high band mode
8
13.5
18
30
MHz
MHz
%
In Range Antialiasing Filter
Bandwidth Tolerance
10
Group Delay Variation
Input Referred Noise Voltage
f = 1 MHz to 18 MHz, VGAIN = −1.6 V to +1.6 V
LNA gain = 15.6 dB
LNA gain = 17.9 dB
350
0.96
0.90
0.82
ps
nV/√Hz
nV/√Hz
nV/√Hz
LNA gain = 21.6 dB
Noise Figure
RS = 50 Ω
Active Termination Matched
LNA gain = 15.6 dB, RFB = 150 Ω
LNA gain = 17.9 dB, RFB = 200 Ω
LNA gain = 21.6 dB, RFB = 300 Ω
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
5.6
4.8
3.8
3.2
2.9
2.6
−30
dB
dB
dB
dB
dB
dB
dB
LSB
Unterminated
Correlated Noise Ratio
Output Offset
No signal, correlated/uncorrelated
−100
+100
Rev. A | Page 4 of 52
Data Sheet
AD9670
Parameter1
Signal-to-Noise Ratio (SNR)
Test Conditions/Comments
Min
Typ
69
59
−130
−70
−62
−61
−55
−54
Max
Unit
dBFS
dBFS
dBc/√Hz
dBc
dBc
dBc
dBc
dBc
fIN = 5 MHz at −12 dBFS, VGAIN = −1.6 V
fIN = 5 MHz at −1 dBFS
Close In SNR
Second Harmonic
fIN = 3.5 MHz at −1 dBFS, VGAIN = 0 V, 1 kHz offset
fIN = 5 MHz at −12 dBFS, VGAIN = −1.6 V
fIN = 5 MHz at −1 dBFS, VGAIN = 1.6 V
fIN = 5 MHz at −12 dBFS, VGAIN = −1.6 V
fIN = 5 MHz at −1 dBFS, VGAIN = 1.6 V
Third Harmonic
Two-Tone Intermodulation
Distortion (IMD3)
fRF1 = 5.015 MHz, fRF2 = 5.020 MHz, ARF1 = −1 dBFS,
ARF2 = −21 dBFS, VGAIN = 1.6 V, IMD3 relative to ARF2
Channel-to-Channel Crosstalk
fIN1 = 5.0 MHz at −1 dBFS
Overrange condition2
−60
−55
dB
dB
GAIN ACCURACY
TA = 25°C
Gain Law Conformance Error
−1.6 < VGAIN < −1.28 V
−1.28 V < VGAIN ≤ +1.28 V
1.28 V < VGAIN < 1.6 V
VGAIN = 0 V, normalized for ideal antialiasing filter loss
−1.28 V < VGAIN < +1.28 V, 1 σ
0.4
dB
dB
dB
dB
dB
dB
−1.3
−1.3
+1.3
+1.3
−0.5
Linear Gain Error
Channel-to-Channel Matching
PGA Gain
0.1
21/24/27/30
GAIN CONTROL INTERFACE
Control Range
Control Common Mode
Input Impedance
Gain Range
Differential
GAIN+, GAIN−
GAIN+, GAIN−
−1.6
0.7
+1.6
0.9
V
V
0.8
10
45
14
3.5
750
MΩ
dB
dB/V
dB
ns
Scale Factor
Analog
Digital step size
Analog 45 dB change
Response Time
CW DOPPLER MODE
LO Frequency
fLO = fMLO/M
1
10
MHz
Phase Resolution
Per channel, 4LO3 mode
Per channel, 8LO mode, 16LO mode
CWI+, CWI−, CWQ+, and CWQ−
Per CWI+, CWI−, CWQ+, and CWQ−, each channel
enabled (2 fLO and baseband signal)
45
22.5
AVDD2/2
2.2
Degrees
Degrees
V
Output DC Bias (Single-Ended)
Output AC Current Range
2.5
mA
Transconductance (Differential)
Demodulated IOUT/VIN, per CWI+, CWI−, CWQ+, and
CWQ−
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
RS = 0 Ω, RFB = ∞
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
RS = 50 Ω, RFB = ∞
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
RS = 0 Ω, RFB = ∞
3.3
4.3
6.6
mA/V
mA/V
mA/V
Input Referred Noise Voltage
Noise Figure
1.6
1.3
1.0
nV/√Hz
nV/√Hz
nV/√Hz
5.7
4.5
3.4
dB
dB
dB
Dynamic Range
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
−3 dBFS input, fRF = 2.5 MHz, fLO = 40 MHz, 1 kHz
offset, 16LO mode, 1 channel enabled
164
162
160
156
dBFS/√Hz
dBFS/√Hz
dBFS/√Hz
dBc/√Hz
Close In SNR
−3 dBFS input, fRF = 2.5 MHz, fLO = 40 MHz, 1 kHz
offset, 16LO mode, 8 channels enabled
161
dBc/√Hz
Rev. A | Page 5 of 52
AD9670
Data Sheet
Parameter1
Test Conditions/Comments
Min
Typ
Max
Unit
Two-Tone Intermodulation
Distortion (IMD3)
fRF1 = 5.015 MHz, fRF2 = 5.020 MHz, fLO = 80 MHz,
ARF1 = −1 dBFS, ARF2 = −21 dBFS, IMD3 relative to ARF2
−58
dB
LO Harmonic Rejection
16LO, 8LO, and 4LO modes
I to Q, all phases, 1 σ
I to Q, all phases, 1 σ
Phase I to I, Q to Q, 1 σ
Amplitude I to I, Q to Q, 1 σ
−20
dBc
Degrees
dB
Degrees
dB
Quadrature Phase Error
I/Q Amplitude Imbalance
Channel-to-Channel Matching
0.15
0.015
0.5
0.25
POWER SUPPLY, MODE I/MODE II/
MODE III/MODE IV
AVDD1
AVDD2
DVDD
1.7
2.85
1.3
1.3
1.7
1.8
3.0
1.4
1.8
1.8
1.9
3.6
1.9
1.9
1.9
V
V
V
V
V
mA
Demodulator/decimator enabled
Demodulator/decimator disabled
DRVDD
IAVDD1
TGC mode, LO band mode
148/187/
223/291
CW Doppler mode
4
mA
mA
mA
mA
mA
IAVDD2
TGC mode, no signal, low band mode
TGC mode, no signal, high band mode
CW Doppler mode, 8 channels enabled
RF decimator enabled in Mode III and Mode IV;
demodulator/decimator enabled all modes
230
239
140
156/247/
166/255
IDVDD
IDRVDD
ANSI-644 mode
133/184/
141/146
119/170/
127/169
mA
mA
Low power (IEEE 1596.3 similar) mode, 1 channel per
lane mode
Total Power Dissipation
(Including Output Drivers)
TGC mode, no signal, RF decimator enabled in Mode III
and Mode IV, demodulator/decimator disabled
TGC mode, no signal, RF decimator enabled in Mode III
and Mode IV, demodulator/decimator enabled
1200/1400/
1380/1630
1400/1695/
1570/1900
1345/1555/
1535/2100
1560/1880/
1740/2100
mW
mW
CW Doppler mode, 8 channels enabled
500
mW
mW
mW
Bits
Power-Down Dissipation
Standby Power Dissipation
ADC RESOLUTION
30
50
630
14
ADC REFERENCE
Output Voltage Error
Load Regulation at 1.0 mA
Input Resistance
VREF = 1 V
VREF = 1 V
mV
mV
kΩ
2
7.5
1 For a complete set of definitions and information about how these tests were completed, see the AN-835 Application Note, Understanding High Speed ADC Testing and Evaluation.
2 The overrange condition is specified as 6 dB more than the full-scale input range.
3 The internal LO frequency, fLO, is generated from the supplied multiplier local oscillator frequency, fMLO, by dividing it up by a configurable divider value (M) that can be
4, 8, or 16; the MLO signal is named 4LO, 8LO, or 16LO, accordingly.
Rev. A | Page 6 of 52
Data Sheet
AD9670
DIGITAL SPECIFICATIONS
AVDD1 = 1.8 V, AVDD2 = 3.0 V, DVDD = 1.4 V, DRVDD = 1.8 V, 1.0 V internal ADC reference, full temperature range (0°C to 85°C), unless
otherwise noted.
Table 2.
Parameter1
Temperature Min
Typ
Max
Unit
INPUTS
CLK+, CLK−, TX_TRIG+, TX_TRIG−
Logic Compliance
CMOS/LVDS/LVPECL
Differential Input Voltage42
Input Voltage Range
Input Common-Mode Voltage
Input Resistance (Differential)
Input Capacitance
0.2
GND − 0.2
3.6
V p-p
V
V
kΩ
pF
AVDD1 + 0.2
0.9
15
4
25°C
25°C
MLO+, MLO−, RESET+, RESET−
Logic Compliance
LVDS/LVPECL
Differential Input Voltage2
Input Voltage Range
Input Common-Mode Voltage
Input Resistance (Single-Ended)
Input Capacitance
0.250
GND − 0.2
−0.3
2 × AVDD2
AVDD2 + 0.2
+0.3
V p-p
V
V
kΩ
pF
AVDD2/2
20
1.5
25°C
25°C
LOGIC INPUTS
PDWN, STBY, SCLK, SDIO, ADDRx
Logic 1 Voltage
1.2
1.2
DRVDD + 0.3
0.3
V
V
kΩ
pF
Logic 0 Voltage
Input Resistance3
25°C
25°C
30 (26 for SDIO)
2 (5 for SDIO)
Input Capacitance3
CSB
Logic 1 Voltage
Logic 0 Voltage
DRVDD + 0.3
0.3
V
V
Input Resistance
Input Capacitance
25°C
25°C
26
2
kΩ
pF
LOGIC OUTPUTS
SDIO4
Logic 1 Voltage (IOH = 800 μA)
Logic 0 Voltage (IOL = 50 μA)
GPO0/GPO1/GPO2/GPO3
Logic 0 Voltage (IOL = 50 μA)
DIGITAL OUTPUTS (DOUTx+, DOUTx−)
ANSI-644
1.79
V
V
0.05
0.05
V
Logic Compliance
LVDS
Differential Output Voltage (VOD)
Output Offset Voltage (VOS)
Output Coding (Default)
Low Power, Reduced Signal Option
Logic Compliance
247
1.125
454
1.375
mV
V
Offset binary
LVDS
Differential Output Voltage (VOD)
Output Offset Voltage (VOS)
Output Coding (Default)
150
1.10
250
1.30
mV
V
Offset binary
1 For a complete set of definitions and information about how these tests were completed, see the AN-835 Application Note, Understanding High Speed ADC Testing and
Evaluation.
2 Specified for LVDS and LVPECL only.
3 The typical input resistance and input capacitance values deviate for SDIO; these deviations are noted in the Typ column.
4 Specified for 13 SDIO pins sharing the same connection.
Rev. A | Page 7 of 52
AD9670
Data Sheet
SWITCHING SPECIFICATIONS
AVDD1 = 1.8 V, AVDD2 = 3.0 V, DVDD = 1.4 V, DRVDD = 1.8 V, full temperature range (0°C to 85°C), RF decimator bypassed, digital
demodulator and baseband decimator bypassed, unless otherwise noted.
Table 3.
Parameter1
CLOCK2
Temperature Min
Typ
Max
Unit
Clock Rate
40 MSPS (Mode I)
Full
Full
Full
Full
Full
Full
20.5
20.5
20.5
20.5
40
65
80
125
MHz
MHz
MHz
MHz
ns
65 MSPS (Mode II)
80 MSPS (Mode III)3
125 MSPS (Mode IV)4
Clock Pulse Width High (tEH)
Clock Pulse Width Low (tEL)
OUTPUT PARAMETERS2, 5
Propagation Delay (tPD)
Rise Time (tR) (20% to 80%)
3.75
3.75
ns
Full
Full
Full
Full
Full
Full
Full
Full
Full
25°C
25°C
10.8 − 1.5 × tDCO
10.8 − 1.5 × tDCO
10.8
300
300
tSAMPLE/7
10.8
tFCO + (tSAMPLE/28)
(tSAMPLE/28)
(tSAMPLE/28)
225
10.8 + 1.5 × tDCO
10.8 + 1.5 × tDCO
ns
ps
ps
ns
ns
ns
ps
ps
ps
ns
ns
Fall Time (tF) (20% to 80%)
6
DCO Period (tDCO
)
FCO Propagation Delay (tFCO
DCO Propagation Delay (tCPD
DCO to Data Delay (tDATA
DCO to FCO Delay (tFRAME
Data-to-Data Skew (tDATA-MAX − tDATA-MIN
TX_TRIG to CLK Setup Time (tSETUP
)
)
7
7
)
(tSAMPLE/28) − 300
(tSAMPLE/28) − 300
(tSAMPLE/28) + 300
(tSAMPLE/28) + 300
400
7
)
)
)
1
1
TX_TRIG to CLK Hold Time (tHOLD
)
Wake-Up Time
Standby
Power-Down
ADC Pipeline Latency
APERTURE
25°C
25°C
Full
2
375
16
ꢀs
ꢀs
Clock cycles
Aperture Uncertainty (Jitter)
LO GENERATION
MLO8 Frequency
4LO Mode
8LO Mode
16LO Mode
RESET9 to MLO Setup Time (tSETUP
RESET to MLO Hold Time (tHOLD
25°C
<1
ps rms
Full
Full
Full
Full
Full
4
8
16
1
1
40
80
160
MHz
MHz
MHz
ns
)
tMLO10/2
tMLO10/2
)
ns
1 For a complete set of definitions and information about how these tests were completed, see the AN-835 Application Note, Understanding High Speed ADC Testing and
Evaluation.
2 The clock can be adjusted via the SPI.
3 Mode III must have the RF decimator enabled because the maximum data rate of the baseband demodulator and decimator is 65 MSPS.
4 Mode IV must have the RF decimator enabled because the maximum data rate of the baseband demodulator and decimator is 65 MSPS.
5 Measurements were taken using a device soldered to FR-4 material.
6 In the typical value, tSAMPLE/7, 7 is based on the number of bits (14) divided by 2 because the interface uses double data rate (DDR) sampling.
7 tSAMPLE/28 is based on the number of bits divided by 2 because the delays are based on half duty cycles.
8 MLO refers to the differential signal created via the MLO− pin and the MLO+ pin. This notation is used throughout the data sheet.
9 RESET refers to the differential signal created via the RESET− pin and the RESET+ pin. This notation is used throughout the data sheet.
10 The period of the MLO clock signal is represented by tMLO
.
Rev. A | Page 8 of 52
Data Sheet
AD9670
TIMING DIAGRAMS
ADC Timing Diagram
N – 1
AIN
tA
N
tSETUP
tHOLD
TX_TRIG+
TX_TRIG–
CLK–
tEH
tEL
CLK+
tCPD
DCO–
DCO+
tFRAME
tFCO
FCO–
FCO+
tPD
tDATA
D8
DOUTx–
DOUTx+
MSB
D12
D11
D10
D9
D7
D6
D5
D4
D3
D2
D1
D0
MSB
D12
N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 17 N – 16 N – 16
Figure 2. 14-Bit Data Serial Stream (Default, RF Decimator Bypassed, Demodulator Bypassed, Baseband Decimator Bypassed), 1 Channel/Lane Mode, FCO Mode = Word
CW Timing Diagrams
tMLO
MLO–
MLO+
tHOLD
tSETUP
RESET–
RESET+
Figure 3. CW Doppler Mode Input MLO , Continuous Synchronous RESET Timing, Sampled on the Falling MLO Edge, 4LO Mode
Rev. A | Page 9 of 52
AD9670
Data Sheet
tMLO
MLO–
MLO+
tHOLD
tSETUP
RESET–
RESET+
Figure 4. CW Doppler Mode Input MLO , Continuous Synchronous RESET Timing, Sampled on the Falling MLO Edge, 8LO Mode
tMLO
MLO–
MLO+
tHOLD
tSETUP
RESET–
RESET+
Figure 5. CW Doppler Mode Input MLO , Pulse Synchronous RESET Timing, 4LO/8LO/16LO Mode
tMLO
MLO–
MLO+
tSETUP
tHOLD
RESET–
RESET+
Figure 6. CW Doppler Mode Input MLO , Pulse Asynchronous RESET Timing, 4LO/8LO/16LO Mode
Rev. A | Page 10 of 52
Data Sheet
AD9670
ABSOLUTE MAXIMUM RATINGS
THERMAL IMPEDANCE
Table 4.
Parameter
Rating
Table 5. Thermal Impedance
Symbol Description
AVDD1 to GND
AVDD2 to GND
DVDD to GND
DRVDD to GND
GND to GND
AVDD2 to AVDD1
AVDD1 to DRVDD
AVDD2 to DRVDD
−0.3 V to +2.0 V
−0.3 V to +3.9 V
−0.3 V to +2.0 V
−0.3 V to +2.0 V
−0.3 V to +0.3 V
−2.0 V to +3.9 V
−2.0 V to +2.0 V
−2.0 V to +3.9 V
Value1 Unit
θJA
ΨJB
ΨJT
Junction-to-ambient thermal
resistance, 0.0 m/sec air flow per
JEDEC JESD51-2 (still air)
Junction-to-board thermal
characterization parameter, 0 m/sec
air flow per JEDEC JESD51-8 (still air)
22.0
°C/W
°C/W
°C/W
9.2
Junction-to-top-of-package
0.12
characterization parameter, 0 m/sec
air flow per JEDEC JESD51-2 (still air)
Digital Outputs (DOUTx+, DOUTx−, DCO+,
DCO−, FCO+, FCO−) to GND
−0.3 V to
DRVDD + 0.3 V
1 Thermal impedance results are from simulations. The printed circuit board
(PCB) is JEDEC multilayer. The thermal performance for actual applications
requires careful inspection of the conditions in the application to determine
if they are similar to those assumed in these calculations.
LI-x, LG-x, LO-x, LOSW-x, CWI−, CWI+, CWQ−, −0.3 V to
CWQ+, GAIN+, GAIN−, RESET+, RESET−,
MLO+, MLO−, GPO0, GPO1, GPO2, GPO3
to GND
AVDD2 + 0.3 V
CLK+, CLK−, TX_TRIG+, TX_TRIG−, VREF to GND −0.3 V to
ESD CAUTION
AVDD1 + 0.3 V
SDIO, PDWN, STBY, SCLK, CSB, ADDRx
−0.3 V to
DRVDD + 0.3 V
0°C to 85°C
−65°C to +150°C
150°C
Operating Temperature Range (Ambient)
Storage Temperature Range (Ambient)
Maximum Junction Temperature
Lead Temperature (Soldering, 10 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.
Rev. A | Page 11 of 52
AD9670
Data Sheet
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
1
2
3
4
5
6
7
8
9
10
11
12
A
B
C
LI-E
LI-F
LI-G
LI-H
VREF
RBIAS GAIN+ GAIN–
LI-A
LI-B
LI-C
LI-D
LG-E
LO-E
LG-F
LO-F
LG-G
LO-G
LG-H
LO-H
GND
GND
GND
GND
CLNA
GND
GND
GND
LG-A
LO-A
LG-B
LO-B
LG-C
LO-C
LG-D
LO-D
D
E
F
LOSW-E LOSW-F LOSW-G LOSW-H GND
GND
GND
GND
GND
GND
GND
GND
GND
GND
GND LOSW-A LOSW-B LOSW-C LOSW-D
GND
AVDD1
GND
AVDD2 AVDD2 AVDD2
GND
AVDD1
GND
GND
AVDD1
GND
AVDD2 AVDD2 AVDD2
GND
AVDD1
GND
GND
AVDD1
GND
GND
DVDD
GND
GND
GND
GND
AVDD1
GND
GND
G
H
J
AVDD1
AVDD1
DVDD
CLK– TX_TRIG–
GND
GND
ADDR4 ADDR3 ADDR2 ADDR1 ADDR0
CSB
CLK+ TX_TRIG+ CWQ+
CWI+
AVDD2 MLO+ RESET– GPO3
AVDD2 MLO– RESET+ GPO2
GPO1
GPO0
PDWN
STBY
SDIO
K
L
GND
GND
CWQ–
CWI–
SCLK
DRVDD DOUTH+ DOUTG+ DOUTF+ DOUTE+ DCO+
GND DOUTH– DOUTG– DOUTF– DOUTE– DCO–
FCO+ DOUTD+ DOUTC+ DOUTB+ DOUTA+ DRVDD
FCO– DOUTD– DOUTC– DOUTB– DOUTA– GND
M
Figure 7. Pin Configuration
4
6
10
12
2
8
1
3
5
7
9
11
A
B
C
D
E
F
G
H
J
K
L
M
TOP VIEW
(Not to Scale)
Figure 8. CSP_BGA Pin Location
Rev. A | Page 12 of 52
Data Sheet
AD9670
Table 6. Pin Function Descriptions
Pin No.
Mnemonic
Description
B5, B6, B8, C5 to C8, D5 to D8, E1, E5
to E8, E12, F2, F4, F6, F7, F9, F11, G1,
G3, G5 to G8, G10, G12, H3 to H6, J4,
K1, K2, K4, M1, M12
GND
Ground. Tie these pins to a quiet analog ground.
F1, F3, F5, F8, F10, F12, G2, G9,
G4, G11
AVDD1
DVDD
1.8 V Analog Supply.
1.4 V/1.8 V Digital Supply.
E2 to E4, E9 to E11, J6, K6
B7
AVDD2
CLNA
3.0 V Analog Supply.
LNA External Capacitor.
L1, L12
C1
D1
A1
B1
DRVDD
LO-E
LOSW-E
LI-E
1.8 V Digital Output Driver Supply.
LNA Analog Inverted Output for Channel E.
LNA Analog Switched Output for Channel E.
LNA Analog Input for Channel E.
LNA Ground for Channel E.
LG-E
C2
LO-F
LOSW-F
LI-F
LNA Analog Inverted Output for Channel F.
LNA Analog Switched Output for Channel F.
LNA Analog Input for Channel F.
LNA Ground for Channel F.
D2
A2
B2
LG-F
C3
LO-G
LOSW-G
LI-G
LNA Analog Inverted Output for Channel G.
LNA Analog Switched Output for Channel G.
LNA Analog Input for Channel G.
LNA Ground for Channel G.
D3
A3
B3
LG-G
C4
LO-H
LOSW-H
LI-H
LNA Analog Inverted Output for Channel H.
LNA Analog Switched Output for Channel H.
LNA Analog Input for Channel H.
LNA Ground for Channel H.
D4
A4
B4
LG-H
H1
J1
CLK−
CLK+
Clock Input Complement.
Clock Input True.
H2
J2
TX_TRIG−
TX_TRIG+
ADDR0
ADDR1
ADDR2
ADDR3
ADDR4
DOUTH−
DOUTH+
DOUTG−
DOUTG+
DOUTF−
DOUTF+
DOUTE−
DOUTE+
DCO−
Transmit Trigger Complement.
Transmit Trigger True.
Chip Address Bit 0.
Chip Address Bit 1.
Chip Address Bit 2.
Chip Address Bit 3.
Chip Address Bit 4.
ADC H Digital Output Complement.
ADC H Digital Output True.
ADC G Digital Output Complement.
ADC G Digital Output True.
ADC F Digital Output Complement.
ADC F Digital Output True.
ADC E Digital Output Complement.
ADC E Digital Output True.
Digital Clock Output Complement.
Digital Clock Output True.
H11
H10
H9
H8
H7
M2
L2
M3
L3
M4
L4
M5
L5
M6
L6
DCO+
M7
L7
M8
L8
M9
L9
M10
L10
FCO−
FCO+
Frame Clock Digital Output Complement.
Frame Clock Digital Output True.
ADC D Digital Output Complement.
ADC D Digital Output True.
ADC C Digital Output Complement.
ADC C Digital Output True.
DOUTD−
DOUTD+
DOUTC−
DOUTC+
DOUTB−
DOUTB+
ADC B Digital Output Complement.
ADC B Digital Output True.
Rev. A | Page 13 of 52
AD9670
Data Sheet
Pin No.
M11
L11
K11
J11
K12
J12
H12
B9
Mnemonic
DOUTA−
DOUTA+
STBY
PDWN
SCLK
SDIO
CSB
LG-A
Description
ADC A Digital Output Complement.
ADC A Digital Output True.
Standby Power-Down.
Full Power-Down.
Serial Clock.
Serial Data Input/Output.
Chip Select Bar.
LNA Ground for Channel A.
A9
D9
C9
LI-A
LOSW-A
LO-A
LG-B
LI-B
LOSW-B
LO-B
LG-C
LI-C
LOSW-C
LO-C
LNA Analog Input for Channel A.
LNA Analog Switched Output for Channel A.
LNA Analog Inverted Output for Channel A.
LNA Ground for Channel B.
LNA Analog Input for Channel B.
LNA Analog Switched Output for Channel B.
LNA Analog Inverted Output for Channel B.
LNA Ground for Channel C.
LNA Analog Input for Channel C.
LNA Analog Switched Output for Channel C.
LNA Analog Inverted Output for Channel C.
LNA Ground for Channel D.
LNA Analog Input for Channel D.
LNA Analog Switched Output for Channel D.
LNA Analog Inverted Output for Channel D.
General-Purpose Open Drain Output 0.
General-Purpose Open Drain Output 1.
General-Purpose Open Drain Output 2.
General-Purpose Open Drain Output 3.
B10
A10
D10
C10
B11
A11
D11
C11
B12
A12
D12
C12
K10
J10
K9
J9
J8
K8
K7
J7
A8
A7
A6
LG-D
LI-D
LOSW-D
LO-D
GPO0
GPO1
GPO2
GPO3
RESET−
RESET+
MLO−
MLO+
GAIN−
GAIN+
RBIAS
VREF
Synchronizing Input for LO Divide by M Counter Complement.
Synchronizing Input for LO Divide by M Counter True.
CW Doppler Multiplier Local Oscillator (MLO) Input Complement.
CW Doppler MLO Input True.
Gain Control Voltage Input Complement.
Gain Control Voltage Input True.
External Resistor to Set the Internal ADC Core Bias Current.
Voltage Reference Input/Output.
A5
K5
J5
K3
J3
CWI−
CWI+
CWQ−
CWQ+
CW Doppler I Output Complement.
CW Doppler I Output True.
CW Doppler Q Output Complement.
CW Doppler Q Output True.
Rev. A | Page 14 of 52
Data Sheet
AD9670
TYPICAL PERFORMANCE CHARACTERISTICS
TGC MODE CHARACTERISTICS
Mode I = fSAMPLE = 40 MSPS, fIN = 5 MHz, LO band mode, RS = 50 Ω, RFB = ∞ (unterminated), LNA gain = 21.6 dB, LNA bias = midhigh,
PGA gain = 27 dB, VGAIN = (GAIN+) − (GAIN−) = 1.6 V, antialiasing filter LPF cutoff = fSAMPLE /3, HPF cutoff = LPF cutoff/12.00 (default),
RF decimator bypassed, digital demodulator and baseband decimator bypassed, unless otherwise noted.
2.0
25
20
15
10
5
1.5
1.0
0.5
0°C
0
25°C
85°C
–0.5
–1.0
–1.5
–2.0
0
–1.6
–1.2
–0.8
–0.4
0
0.4
0.8
1.2
1.6
V
(V)
GAIN
GAIN ERROR (dB)
Figure 9. Gain Error vs. VGAIN
Figure 12. Gain Error Histogram, VGAIN = 1.28 V
25
20
15
10
5
20
15
10
5
0
0
GAIN ERROR (dB)
CHANNEL-TO-CHANNEL GAIN MATCHING (dB)
Figure 10. Gain Error Histogram, VGAIN = −1.28 V
Figure 13. Gain Matching Histogram, VGAIN = −1.2 V
35
30
25
20
15
10
5
20
15
10
5
0
0
GAIN ERROR (dB)
CHANNEL-TO-CHANNEL GAIN MATCHING (dB)
Figure 11. Gain Error Histogram, VGAIN = 0 V
Figure 14. Gain Matching Histogram, VGAIN = 1.2 V
Rev. A | Page 15 of 52
AD9670
Data Sheet
1.4
1.2
1.0
0.8
0.6
70
68
66
64
62
60
58
56
54
52
LNA GAIN = 17.9dB
LNA GAIN = 15.6dB
LNA GAIN = 21.3dB
PGA GAIN = 27dB
15 20
0.4
1
50
10
2
3
4
5
6
7
8
9
10
25
30
35
40
45
50
55
FREQUENCY (MHz)
CHANNEL GAIN (dB)
Figure 15. Short-Circuit, Input Referred Noise vs. Frequency
Figure 18. SNR vs. Channel Gain and LNA Gain, AOUT = −1.0 dBFS
–132
74
PGA GAIN = 21dB
PGA GAIN = 21dB
72
–134
–136
–138
–140
–142
–144
–146
70
PGA GAIN = 24dB
68
66
64
PGA GAIN = 27dB
62
60
PGA GAIN = 30dB
58
56
LNA GAIN = 21.3dB
54
–5
0
5
10
15
20
25
30
35
40
45
–5
0
5
10 15 20 25 30 35 40 45 50 55
CHANNEL GAIN (dB)
CHANNEL GAIN (dB)
Figure 16. Short-Circuit, Output Referred Noise vs. Channel Gain,
PGA Gain = 21 dB, VGAIN = 1.6 V
Figure 19. SNR vs. Channel Gain and PGA Gain, AIN = −45 dBm
70
0
SPEED MODE = I (40MSPS)
LO BAND MODE
PGA GAIN = 21dB
–1
68
–2
–3
–4
–5
–6
–7
–8
66
PGA GAIN = 24dB
64
62
60
58
56
54
52
50
PGA GAIN = 27dB
PGA GAIN = 30dB
–9
LNA GAIN = 21.3dB
15 20
–10
10
25
30
35
40
45
50
55
0
5
10
15
20
CHANNEL GAIN (dB)
INPUT FREQUENCY (MHz)
Figure 17. SNR vs. Channel Gain and PGA Gain, AOUT = −1.0 dBFS
Figure 20. Antialiasing Filter Pass-Band Response,
LPF Cutoff = 1/3 × fSAMPLE, HPF = 1/12 × LPF Cutoff
Rev. A | Page 16 of 52
Data Sheet
AD9670
0
0
–10
LNA GAIN = 21.6dB
PGA GAIN = 27dB
MIN V , A = –12.0dBFS
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
GAIN
OUT
–20
MAX V
, A
= –1.0dBFS
GAIN
OUT
–30
–40
–50
V
= –1.2V
GAIN
THIRD ORDER, MIN VGAIN
–60
THIRD ORDER, MAX VGAIN
V
= 0V
GAIN
–70
–80
SECOND ORDER, MIN VGAIN
SECOND ORDER, MAX VGAIN
–90
V
= +1.6V
GAIN
–100
–110
–120
2
3
4
5
6
7
8
9
10
11
–40
–35
–30
–25
–20
–15
–10
–5
0
INPUT FREQUENCY (MHz)
ADC OUTPUT LEVEL (dBFS)
Figure 21. Second-Order and Third-Order Harmonic Distortion vs.
Input Frequency, AOUT = −1.0 dBFS
Figure 24. Second-Order Harmonic Distortion vs. ADC Output Level (AOUT)
0
0
–10
–20
–30
PGA GAIN = 24dB
–10
–20
–30
–40
–50
–60
–40
–50
V
= –1.2V
GAIN
–60
V
= 0V
GAIN
–70
LNA GAIN = 17.9dB
–80
–70
LNA GAIN = 21.6dB
–90
–80
V
= +1.6V
–5
LNA GAIN = 15.6dB
GAIN
–100
–110
–120
–90
–100
10
15
20
25
30
35
40
45
50
–40
–35
–30
–25
–20
–15
–10
0
CHANNEL GAIN (dB)
ADC OUTPUT LEVEL (dBFS)
Figure 22. Second-Order Harmonic Distortion vs. Channel Gain,
OUT = −1.0 dBFS
Figure 25. Third-Order Harmonic Distortion vs. ADC Output Level (AOUT)
A
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
–140
–150
–160
PGA GAIN = 24dB
LNA GAIN = 17.9dB
LNA GAIN = 21.6dB
LNA GAIN = 15.6dB
–100
10
15
20
25
30
35
40
45
100
1k
10k
100k
CHANNEL GAIN (dB)
OFFSET FREQUENCY FROM CARRIER (Hz)
Figure 26. TGC Path Phase Noise,
LNA Gain = 21.6 dB, PGA Gain = 27 dB, VGAIN = 0 V
Figure 23. Third-Order Harmonic Distortion vs. Channel Gain,
AOUT = −1.0 dBFS
Rev. A | Page 17 of 52
AD9670
Data Sheet
8
7
0
–10
fIN1 = 5.0MHZ
fIN2 = 5.01MHZ
6
5
4
3
2
A
A
LEVEL = –1dBFS
LEVEL = –21dBFS
RF1
–20
RF2
–30
–40
1
0
100k
–50
1M
10M
FREQUENCY (Hz)
100M
–60
V
= –1.2V
GAIN
0
–10
–20
–30
–40
–50
–60
–70
–80
–70
–80
–90
–100
–110
–120
V
= +1.6V
GAIN
V
= 0V
GAIN
–90
100k
–40
–35
–30
–25
–20
–15
–10
–5
0
1M
10M
FREQUENCY (Hz)
100M
AMPLITUDE LEVEL (dBFS)
Figure 27. LNA Input Impedance Magnitude and Phase, Unterminated
Figure 29. IMD3 vs. ADC Output Amplitude Level
0
7
6
5
4
3
2
fIN1 = 2.3MHz
fIN2 = 2.31MHz
–10
A
A
R
LEVEL = –1dBFS
LEVEL = –21dBFS
= 50Ω
RF1
RF2
–20
–30
–40
–50
–60
–70
–80
–90
–100
S
R
= 300Ω, 1000Ω
IN
R
= 50Ω
IN
1
0
15
20
25
30
35
40
45
50
2
4
6
8
10
12
14
16
18
20
CHANNEL GAIN (dB)
FREQUENCY (MHz)
Figure 28. IMD3 vs. Channel Gain
Figure 30. Noise Figure vs. Frequency,
RS = RIN = 100 Ω, LNA Gain = 17.9 dB, PGA Gain = 30 dB, VGAIN = 1.6 V
Rev. A | Page 18 of 52
Data Sheet
AD9670
CW DOPPLER MODE CHARACTERISTICS
fIN = 5 MHz, fLO = 20 MHz, 4LO mode, RS = 50 Ω, LNA gain = 21.6 dB, LNA bias = mid-high, all CW channels enabled, phase rotation = 0°.
10
165
160
155
150
145
140
135
130
9
8
7
6
5
4
3
2
1
0
0
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
BASEBAND FREQUENCY (Hz)
0
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
BASEBAND FREQUENCY (Hz)
Figure 31. Noise Figure vs. Baseband Frequency
Figure 32. SNR vs. Baseband Frequency, −3 dBFS LNA Input
Rev. A | Page 19 of 52
AD9670
Data Sheet
THEORY OF OPERATION
MLO–
MLO+
LO
RESET+
CWI+
CWI–
GENERATION
RESET–
R
R
LO-x
FB1
FB2
CWQ+
CWQ–
LOSW-x
T/R
SWITCH
C
S
LI-x
DOUTx+
DOUTx–
PIPELINE
ADC
DEMOD/
DEC
SERIAL
LVDS
ATTENUATOR
–45dB TO 0dB
POST
AMP
FILTER
LNA
15.6dB,
17.9dB,
21.6dB
LG-x
C
SH
21dB,
C
LG
24dB,
27dB,
30dB
GAIN
INTERPOLATOR
NCO
TRANSDUCER
gm
GAIN+ GAIN–
TX_TRIG+ TX_TRIG–
Figure 33. Simplified Block Diagram of a Single Channel
Each channel in the AD9670 contains both a TGC signal path and
a CW Doppler signal path. Common to both signal paths, the
LNA provides four user adjustable input impedance termination
options for matching different probe impedances. The CW
Doppler path includes an I/Q demodulator with programmable
phase rotation needed for analog beamforming. The TGC path
includes a differential X-AMP® VGA, an antialiasing filter, an
ADC, and a digital demodulator and decimator. Figure 33 shows
a simplified block diagram with external components.
Then, calculate the total channel gain using Equation 3.
Channel Gain (dB) = LNAGAIN + VGAATT + PGAGAIN
(3)
In its default condition, the LNA has a gain of 21.6 dB (12×),
and the VGA postamplifier gain is 24 dB. If the voltage on the
GAIN+ pin is 0 V and the voltage on the GAIN− pin is 1.6 V
(45.1 dB attenuation), the total gain of the channel is 0.5 dB if
the LNA input is unmatched. The channel gain is −5.5 dB if the
LNA is matched to 50 Ω (RFB = 300 Ω). However, if the voltage on
the GAIN+ pin is 1.6 V and the voltage on the GAIN− pin is 0 V
(0 dB attenuation), VGAATT = 0 dB. This results in a total gain of
45.3 dB through the TGC path if the LNA input is unmatched or a
total gain of 39.3 dB if the LNA input is matched.
TGC OPERATION
The system gain for TGC operation is distributed as shown in
Table 7.
In addition to the analog VGA attenuation described in Equation 2,
the attenuation level can be digitally controlled in 3.5 dB increm-
ents. In this case, Equation 3 is still valid and the value of VGAATT is
equal to the attenuation level set in SPI Register 0x011, Bits [7:4].
Table 7. Channel Analog Gain Distribution
Section
Nominal Gain (dB)
LNA
15.6/17.9/21.6 (LNAGAIN
)
Attenuator
VGA
−45 to 0 (VGAATT)
21/24/27/30 (PGAGAIN
Low Noise Amplifier (LNA)
)
Filter
ADC
0
0
Good system sensitivity relies on a proprietary ultralow noise LNA
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 impedance matching.
Each LNA output is dc-coupled to a VGA input. The VGA
consists of an attenuator with a range of −45 dB to 0 dB, followed
by an amplifier with 21 dB/24 dB/27 dB/30 dB of gain. The X-AMP
gain interpolation technique results in low gain error and uniform
bandwidth, and differential signal paths minimize distortion.
The LNA inputs, LI-x, are capacitively coupled to the source.
An on-chip bias generator establishes dc input bias voltages of
approximately 2.2 V and centers the output common-mode levels
at 1.5 V (AVDD2 divided by 2). A capacitor, CLG, of the same
value as the input coupling capacitor, CS, is connected from the
LG-x pins into ground.
The linear in dB gain (law conformance) range of the TGC path is
45 dB. The slope of the gain control interface is 14 dB/V, and the
gain control range is −1.6 V to +1.6 V. Equation 1 is the expression
for the differential voltage, VGAIN, at the gain control interface.
Equation 2 is the expression for the VGA attenuation, VGAATT
as a function of VGAIN
The LNA supports three gain settings, 21.6 dB, 17.9 dB, or 15.6 dB,
set through the SPI. Overload protection ensures quick recovery
time from large input voltages.
,
.
V
GAIN (V) = (GAIN+) − (GAIN−)
(1)
(2)
Low value feedback resistors and the current driving capability
of the output stage allow the LNA to achieve a low input referred
noise voltage of 0.78 nV/√Hz (at a gain of 21.6 dB). On-chip
resistor matching results in precise single-ended gains, which
VGAATT (dB) = −14 dB/V (1.6) − VGAIN
Rev. A | Page 20 of 52
Data Sheet
AD9670
are critical for accurate impedance control. The use of a fully
differential topology and negative feedback minimizes distor-
tion. Low second-order harmonic distortion is particularly
important in harmonic ultrasound imaging applications.
R
FB is the resulting impedance of the RFB1 and RFB2 combination
(see Figure 33). Using Register 0x02C in the SPI memory map,
the AD9670 can be programmed for four impedance matching
options: three active terminations and one unterminated option.
Table 8 shows an example of how to select RFB1 and RFB2 for 66 Ω,
100 Ω, and 200 Ω input impedances for LNA gain = 21.6 dB (12×).
Active Impedance Matching
The LNA consists of a single-ended voltage gain amplifier with
differential outputs. The negative output is externally available
on two output pins, LO-x and LOSW-x, that are controlled via
internal switches. This configuration allows the active input
impedance synthesis of three different impedance values (and
an unterminated value) by connecting up to two external
resistances in parallel and controlling the internal switch states
via the SPI. For example, with a fixed gain of 8× (17.9 dB), an
active input termination is synthesized by connecting a feedback
resistor between the negative output pin, LO-x, and the positive
input pin, LI-x. This well known technique is used for interfacing
multiple probe impedances to a single system. The input
resistance (RIN) calculation is shown in Equation 4.
Table 8. Active Termination Example for LNA Gain = 21.6 dB,
RFB1 = 650 Ω, and RFB2 = 1350 Ω
Reg. 0x02C
Value
LO-x
RS (Ω) Switch
LOSW-x
Switch
RFB (Ω)
RIN (Ω)1
00 (default)
100
50
200
N/A2
On
On
Off
Off
Off
On
On
Off
RFB1
100
01
10
11
RFB1 || RFB2 66
RFB2
∞
200
∞
1 See Equation 4.
2 N/A means not applicable.
The bandwidth (BW) of the LNA is greater than 80 MHz.
Ultimately, the BW of the LNA limits the accuracy of the
synthesized RIN. For RIN = RS up to about 200 Ω, the best match
is between 100 kHz and 10 MHz, where the lower frequency
limit is determined by the size of the ac coupling capacitors, and
the upper limit is determined by the LNA BW. Furthermore, the
input capacitance and RS limit the BW at higher frequencies.
Figure 34 shows RIN vs. frequency for various values of RFB.
1k
(RFB1 20 )||(RFB2 20 )30
(4)
RIN
(1 A/2)
where:
FB1 and RFB2 are the external feedback resistors.
20 Ω is the internal switch on resistance.
R
30 Ω is an internal series resistance common to the two internal
switches.
A/2 is the single-ended gain or the gain from the LI-x inputs to
R
= 500Ω, R = 2kΩ
S
FB
the LO-x outputs.
R
R
= 200Ω, R = 800Ω
R
FB can be equal to RFB1, RFB2, or (RFB1 + 20 Ω)||(RFB2 + 20 Ω),
S
S
FB
depending on the connection status of the internal switches.
= 100Ω, R = 400Ω, C = 20pF
FB
SH
100
Because the amplifier has a gain of 8× from its input to its
differential output, it is important to note that the gain, A/2,
is the gain from the LI-x pin to the LO-x pin and that it is 6 dB
less than the gain of the amplifier, or 12.1 dB (4×). The input
resistance is reduced by an internal bias resistor of 6 kΩ in
parallel with the source resistance connected to the LI-x pin,
with the LG-x pin ac grounded. Use Equation 5 to calculate the
required RFB for a desired RIN, even for higher values of RIN.
R
= 50Ω, R = 200Ω, C = 70pF
FB SH
S
10
100k
1M
10M
100M
FREQUENCY (Hz)
Figure 34. RIN vs. Frequency for Various Values of RFB
(Effects of RSH and CSH Are Also Shown)
(RFB1 20 ) || (RFB2 20 ) 30
(5)
RIN
|| 6 k
(1 A/2)
However, as seen for larger RIN values, parasitic capacitance
starts rolling off the signal BW before the LNA produces peaking.
For example, to set RIN to 200 Ω with a single-ended LNA gain of
12.1 dB (4×), the value of RFB from Equation 4 must be 950 Ω,
while the switch for RFB2 is open. If the more accurate equation
(Equation 5) is used to calculate RIN, the value is then 194 Ω
instead of 200 Ω, resulting in a gain error of less than 0.27 dB.
Some factors, such as the presence of a dynamic source resistance,
may influence the absolute gain accuracy more significantly. At
higher frequencies, the input capacitance of the LNA must be
considered. The user must determine the level of matching
accuracy and adjust RFB accordingly.
C
SH further degrades the match; therefore, do not use CSH for
values of RIN that are greater than 100 Ω.
Rev. A | Page 21 of 52
AD9670
Data Sheet
Table 9 lists the recommended values for RFB and CSH in terms
of RIN. CFB is needed in series with RFB because the dc levels at the
LO-x pin and the LI-x pin are unequal.
Figure 36 shows the noise figure as it relates to RS for various
values of RIN, which is helpful for design purposes.
8
R
R
R
R
= 50Ω
= 75Ω
= 100Ω
= 200Ω
IN
IN
IN
IN
7
6
5
4
3
2
1
0
Table 9. Active Termination External Component Values
LNA Gain (dB)
RIN (Ω)
RFB (Ω)
Minimum CSH (pF)
UNTERMINATED
15.6
50
150
90
17.9
50
200
70
21.6
50
300
50
15.6
17.9
21.6
15.6
17.9
21.6
100
100
100
200
200
200
350
450
650
750
950
1350
30
20
10
Not applicable
Not applicable
Not applicable
10
100
(Ω)
1k
R
S
LNA Noise
Figure 36. Noise Figure vs. RS for Various Fixed Values of RIN,
Active Termination Matched Inputs, VGAIN = 1.6 V
The short-circuit noise voltage (input referred noise) is an
important limit on system performance. The short-circuit noise
voltage for the LNA is 0.78 nV/√Hz at a gain of 21.6 dB, including
the VGA noise at a VGA postamplifier gain of 27 dB. These meas-
urements, taken without a feedback resistor, provide the basis
for calculating the input noise and noise figure (NF) performance.
Figure 35 and Figure 36 are simulations of noise figure vs. RS
results with different input configurations and an input referred
noise voltage of 2.5 nV/√Hz for the VGA. Unterminated (RFB = ∞)
operation exhibits the lowest equivalent input noise and noise
figure. Figure 36 shows the noise figure vs. source resistance rising
at low RS—where the LNA voltage noise is large compared to the
source noise—and at high RS due to the noise contribution from
RFB. The lowest NF is achieved when RS matches RIN.
CLNA Connection
Attach a 1 nF capacitor from CLNA (Ball B7) to AVDD2.
DC Offset Correction/High-Pass Filter
The AD9670 LNA architecture is designed to correct for dc offset
voltages that can develop on the external CS capacitor due to
leakage of the T/R switch during ultrasound transmit cycles.
The dc offset correction, as shown in Figure 37, provides a
feedback mechanism to the LG-x input of the LNA to correct
for this dc voltage.
AD9670
R
R
FB1
LO-x
LOSW-x
FB2
Figure 35 shows the relative noise figure performance. With an
LNA gain of 21.6 dB, the input impedance is swept with RS to
preserve the match at each point. The noise figures for a source
impedance of 50 ꢀ are 7 dB, 4 dB, and 2.5 dB for the shunt term-
ination, active termination, and unterminated configurations,
respectively. The noise figures for 200 ꢀ are 4.5 dB, 1.7 dB,
and 1 dB, respectively.
T/R
SWITCH
C
S
LI-x
LNA
LG-x
C
SH
C
15.6dB,
17.9dB,
21.6dB
LG
TRANSDUCER
gm
12.0
DC OFFSET
CORRECTION
10.5
9.0
Figure 37. Simplified LNA Input Configuration
The feedback acts as high-pass filter providing dynamic correction
of the dc offset. The cutoff frequency of the high-pass filter response
is dependent on the value of the CLG capacitor, the gain of the
LNA (LNAGAIN), and the transconductance (gm) of the feedback
transconductance amplifier. The gm value is programmed in
Register 0x120, Bits[4:3]. CS must be equal to CLG for proper
operation.
7.5
SHUNT TERMINATION
6.0
4.5
3.0
ACTIVE TERMINATION
UNTERMINATED
1.5
0
10
100
1k
R
(Ω)
S
Figure 35. Noise Figure vs. RS for Shunt Termination,
Active Termination Matched, and Unterminated Inputs, VGAIN = 1.6 V
Rev. A | Page 22 of 52
Data Sheet
AD9670
AVDD2
249Ω
Table 10. High-Pass Filter Cutoff Frequency, fHP, for CLG = 10 nF
±0.8V DC
31.3kΩ
±1.6V
LNAGAIN
17.9 dB
(mS) 15.6 dB (kHz) (kHz)
=
AD9670
AT 0.8V CM
249Ω
100Ω
Reg. 0x120,
Bits[4:3]
gm
LNAGAIN
=
LNAGAIN
21.6 dB (kHz)
=
GAIN+
0.8V CM
0.01µF
ADA4938-1/
ADA4938-2
100Ω
249Ω
10kΩ
00 (default)
0.5
1.0
1.5
2.0
41
83
133
167
55
83
GAIN–
±0.8V DC
AT 0.8V CM
0.01µF
01
10
11
110
178
220
167
267
330
249Ω
Figure 38. Differential GAIN Pin Configuration
Disable the analog gain control and digitally control the
attenuator using SPI Register 0x011, Bits[7:4]. The control
range is 45 dB, and the step size is 3.5 dB.
For other values of CLG, determine the high-pass filter cutoff
frequency by scaling the values from Table 10 or calculating
based on CLG, LNAGAIN, and gm, as shown in Equation 6.
VGA Noise
fHP (CLG) =
In a typical application, a VGA compresses a wide dynamic
range input signal to within the input span of an ADC. The
input referred noise of the LNA limits the minimum resolvable
input signal, whereas 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 latter limit is set in accordance with the
total noise floor of the ADC.
10 nF
gm
CLG
1
2
× LNAGAIN
×
= fHP (see Table 10) ×
(6)
CLG
Variable Gain Amplifier (VGA)
The differential X-AMP VGA provides precise input attenu-
ation and interpolation. It has a low input referred noise of
2.5 nV/√Hz and excellent gain linearity. The VGA is driven by
a fully differential input signal from the LNA. The X-AMP archi-
tecture produces a linear in dB gain law conformance and low
distortion levels—deviating by only 0.5 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 resulting total gain range is 45 dB, which allows
range loss at the endpoints. The X-AMP inputs are part of a PGA
that completes the VGA. The PGA in the VGA is programmable
to a gain of 21 dB, 24 dB, 27 dB, or 30 dB. This allows the
optimization of channel gain for different imaging modes in the
ultrasound system. The VGA bandwidth is greater than 100 MHz.
The input stage is designed to ensure excellent frequency response
uniformity across the gain setting. For TGC mode, this uniformity
minimizes time delay variation across the gain range.
The output referred noise is a flat 40 nV/√Hz (postamplifier gain =
24 dB) over most of the gain range because it is dominated by the
fixed output referred noise of the VGA. At the high end of the
gain control range, the noise of the LNA and of the source
prevail. The input referred noise reaches its minimum value
near the maximum gain control voltage, where the input referred
contribution of the VGA is miniscule.
At lower gains, the input referred noise and, therefore, the
noise figure, increases as the gain decreases. The instantaneous
dynamic range of the system is not lost, however, because the
input capacity increases as the input referred noise increases. The
contribution of the ADC noise floor has the same dependence. The
important relationship is the magnitude of the VGA output
noise floor relative to that of the ADC.
Gain Control
The analog gain control interface, GAIN , is a differential input.
Gain control noise is a concern in very low noise applications.
Thermal noise in the gain control interface can modulate the
channel gain. The resulting noise is proportional to the output
signal level and is usually evident only when a large signal is
present. Take care to minimize noise impinging at the GAIN
inputs. Use an external RC filter to remove VGAIN source noise. The
filter bandwidth must be sufficient to accommodate the desired
control bandwidth and attenuate unwanted switching noise from
the external DACs used to drive the gain control.
VGAIN varies the gain of all VGAs through the interpolator by
selecting the appropriate input stages connected to the input
attenuator. The nominal VGAIN range is 14 dB/V from −1.6 V to
+1.6 V, with the best gain linearity from approximately −1.44 V
to +1.44 V, where the error is typically less than 0.5 dB. For
VGAIN voltages of greater than +1.44 V and less than −1.44 V, the
error increases. The value of GAIN can exceed the supply
voltage by 1 V without gain foldover.
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.
The AD9670 can bypass the GAIN inputs and control the gain
of the attenuator digitally (see the Gain Control section). This
mode removes any external noise contributions when active
gain control is not needed.
The differential input pins, GAIN+ and GAIN−, can be interfaced
as shown in Figure 38. DC couple the GAIN+ and GAIN− pins,
and drive them to accommodate a 3.2 V full-scale input.
Rev. A | Page 23 of 52
AD9670
Data Sheet
Antialiasing Filter
or 1/6 of the ADC sample clock rate. The cutoff can be scaled to
0.75, 0.8, 0.9, 1.0, 1.13, 1.25, or 1.45 times this frequency through
Register 0x00F. The cutoff tolerance ( 10%) is maintained from
8 MHz to 18 MHz for low band mode and 13.5 MHz to 30 MHz
for high band mode. Table 11 and Table 12 calculate the valid
SPI-selectable low-pass filter settings and expected cutoff
frequencies for the low band and high band mode at the
minimum sample frequency and the maximum sample
frequency in each speed mode.
The filter that the signal reaches prior to the ADC is used to
reject dc signals and to band limit the signal for antialiasing.
The antialiasing filter is a combination of a single-pole high-
pass filter and a second-order low-pass filter. The high-pass
filter can be configured at a ratio of the low-pass filter cutoff.
This is selectable through Register 0x02B.
The filter uses on-chip tuning to trim the capacitors and, in
turn, to set the desired low-pass cutoff frequency and reduce
variations. The default −3 dB low-pass filter cutoff is 1/3, 1/4.5,
Rev. A | Page 24 of 52
AD9670
Data Sheet
Table 11. SPI-Selectable Low-Pass Filter Cutoff Options for Low Band Mode at Example Sampling Frequencies
Address
0x00F,
Bits[7:3]
Sampling Frequency (MHz)
LPF Cutoff
Frequency (MHz)
20.5
40
65
80
125
0 0000
0 0001
0 0010
0 0011
0 0100
0 0101
0 0110
0 1000
0 1001
0 1010
0 1011
0 1100
0 1101
0 1110
1 0000
1 0001
1 0010
1 0011
1 0100
1 0101
1 0110
1.45 × (1/3) × fSAMPLE
1.25 × (1/3) × fSAMPLE
1.13 × (1/3) × fSAMPLE
1.0 × (1/3) × fSAMPLE
0.9 × (1/3) × fSAMPLE
0.8 × (1/3) × fSAMPLE
0.75 × (1/3) × fSAMPLE
9.91
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
8.54
16.67
15.00
13.33
12.00
10.67
10.00
12.89
11.11
10.00
8.89
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
17.33
16.25
20.94
18.06
16.25
14.44
13.00
11.56
10.83
15.71
13.54
12.19
10.83
9.75
Out of tunable
filter range
16.82
1.45 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.25 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.13 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.0 × (1/4.5) × fSAMPLE
0.9 × (1/4.5) × fSAMPLE
0.8 × (1/4.5) × fSAMPLE
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
17.78
16.00
14.22
13.33
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
8.00
Out of tunable
filter range
Out of tunable
filter range
0.75 × (1/4.5) × fSAMPLE Out of tunable
filter range
17.50
1.45 × (1/6) × fSAMPLE
1.25 × (1/6) × fSAMPLE
1.13 × (1/6) × fSAMPLE
1.0 × (1/6) × fSAMPLE
0.9 × (1/6) × fSAMPLE
0.8 × (1/6) × fSAMPLE
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
9.67
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
8.33
16.67
15.00
13.33
12.00
10.67
10.00
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
16.67
8.67
0.75 × (1/6) × fSAMPLE Out of tunable
filter range
Out of tunable
filter range
8.13
15.63
Rev. A | Page 25 of 52
AD9670
Data Sheet
Table 12. SPI-Selectable Low-Pass Filter Cutoff Options for High Band Mode at Example Sampling Frequencies
Sampling Frequency (MHz)
Address 0x00F, LPF Cutoff
Bits[7:3]
Frequency (MHz)
20.5
40
65
80
125
0 0000
1.45 × (1/3) × fSAMPLE
1.25 × (1/3) × fSAMPLE
1.13 × (1/3) × fSAMPLE
1.0 × (1/3) × fSAMPLE
0.9 × (1/3) × fSAMPLE
0.8 × (1/3) × fSAMPLE
0.75 × (1/3) × fSAMPLE
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
19.33
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
0 0001
0 0010
0 0011
0 0100
0 0101
0 0110
0 1000
0 1001
0 1010
0 1011
0 1100
0 1101
0 1110
1 0000
1 0001
1 0010
1 0011
1 0100
1 0101
1 0110
16.67
15.00
27.08
24.38
21.67
19.50
17.33
16.25
20.94
18.06
16.25
14.44
30.00
26.67
24.00
21.33
20.00
25.78
22.22
20.00
17.78
16.00
14.22
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
1.45 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.25 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.13 × (1/4.5) × fSAMPLE Out of tunable
filter range
1.0 × (1/4.5) × fSAMPLE
0.9 × (1/4.5) × fSAMPLE
0.8 × (1/4.5) × fSAMPLE
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
27.78
25.00
22.22
20.83
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
0.75 × (1/4.5) × fSAMPLE Out of tunable
filter range
Out of tunable
filter range
1.45 × (1/6) × fSAMPLE
1.25 × (1/6) × fSAMPLE
1.13 × (1/6) × fSAMPLE
1.0 × (1/6) × fSAMPLE
0.9 × (1/6) × fSAMPLE
0.8 × (1/6) × fSAMPLE
0.75 × (1/6) × fSAMPLE
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
15.71
19.33
16.67
15.00
Out of tunable
filter range
13.54
26.04
23.44
20.83
18.75
16.67
15.63
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Rev. A | Page 26 of 52
Data Sheet
AD9670
Tuning is normally off to avoid changing the capacitor settings
during critical times. The tuning circuit is enabled through the
SPI. It is disabled automatically after 512 cycles of the ADC sample
clock. Initializing the tuning of the filter must be performed after
initial power-up and after reprogramming the filter cutoff
scaling or ADC sample rate. The tuning is initiated in
Register 0x02B, Bit 6.
Clock Input Considerations
For optimum performance, clock the AD9670 sample clock
inputs (CLK+ and CLK−) with a differential signal. This signal
is typically ac-coupled into the CLK+ and CLK− pins via a trans-
former or capacitors. These pins are biased internally and require
no additional bias.
Figure 39 shows the preferred method for clocking the AD9670. A
low jitter clock source, such as the Valpey Fisher oscillator, VFAC3-
BHL-50 MHz, is converted from single-ended to differential using
an RF transformer. The back to back Schottky diodes across the
secondary transformer limit clock excursions into the AD9670
to approximately 0.8 V p-p differential. This helps prevent the
large voltage swings of the clock from feeding through to other
portions of the AD9670, and it preserves the fast rise and fall
times of the signal, which are critical to low jitter performance.
3.3V
A total of four SPI-programmable settings allow the user to vary
the high-pass filter cutoff frequency as a function of the low-
pass cutoff frequency. Two examples are shown in Table 13: one
example is for an 8 MHz low-pass cutoff frequency, and the other
example is for an 18 MHz low-pass cutoff frequency. In both
examples, as the ratio decreases, the amount of rejection on the
low-end frequencies increases. Therefore, making the entire
antialiasing filter frequency pass band narrow can reduce low
frequency noise or maximize dynamic range for harmonic
processing.
®
MINI-CIRCUITS
ADT1-1WT, 1:1Z
0.1µF
0.1µF
XFMR
Table 13. High-Pass Filter Cutoff Options
CLK+
OUT
100Ω
High-Pass Cutoff Frequency
50Ω
Reg. 0x02B[1:0]
AD9670
0.1µF
VFAC3
High-Pass Filter
Cutoff
Low-Pass
Low-Pass
CLK–
Ratio1 Cutoff = 8 MHz Cutoff = 18 MHz
SCHOTTKY
DIODES:
HSM2812
0.1µF
00 (default)
12.00
9.00
6.00
3.00
670 kHz
890 kHz
1.33 MHz
2.67 MHz
1.5 MHz
2.0 MHz
3.0 MHz
6.0 MHz
01
10
11
Figure 39. Transformer-Coupled Differential Clock
If a low jitter clock is available, another option is to ac-couple
a differential PECL signal to the sample clock input pins, as
shown in Figure 40. Analog Devices, Inc., offers a family of
clock drivers with excellent jitter performance, including the
AD9516-0, AD9516-1, AD9516-2, AD9516-3, and AD9516-5
(these five devices are represented by AD9516-x in Figure 40,
Figure 41, and Figure 42), as well as the AD9524.
1 Ratio is the low-pass filter cutoff frequency/high-pass filter cutoff frequency.
Antialiasing Filter/VGA Test Mode
For debug and testing, there is a bypass switch to view the anti-
aliasing filter output on the GPO2 and GPO3 pins. Enable this
mode using SPI Register 0x109, Bit 4. The differential antialiasing
filter output of only one channel can be accessed at a time. The
dc output voltage is 1.5 V (or AVDD2/2) and the maximum ac
output voltage is 2 V p-p.
3.3V
AD9516-x
OR AD9524
VFAC3
OUT
0.1µF
0.1µF
CLK+
CLK
PECL DRIVER
CLK
50Ω*
0.1µF
100Ω
AD9670
0.1µF
ADC
CLK–
240Ω
240Ω
The AD9670 uses a pipelined ADC architecture. The quantized
output from each stage is combined into a 14-bit result in the
digital correction logic. The pipelined architecture permits the
first stage to operate on a new input sample and the remaining
stages to operate on preceding samples. Sampling occurs on the
rising edge of the clock.
*50Ω RESISTOR IS OPTIONAL.
Figure 40. Differential PECL Sample Clock
A third option is to ac couple a differential LVDS signal to the
sample clock input pins, as shown in Figure 41.
3.3V
The output staging block aligns the data, corrects errors, and
passes the data to the output buffers. The data is then serialized
and aligned to the frame and output clocks.
AD9516-x
OR AD9524
VFAC3
OUT
0.1µF
0.1µF
CLK+
CLK
LVDS DRIVER
CLK
50Ω*
0.1µF
100Ω
AD9670
0.1µF
CLK–
*50Ω RESISTOR IS OPTIONAL.
Figure 41. Differential LVDS Sample Clock
Rev. A | Page 27 of 52
AD9670
Data Sheet
130
120
110
100
90
In some applications, it is acceptable to drive the sample clock
inputs with a single-ended CMOS signal. In such applications,
drive CLK+ directly from a CMOS gate, and bypass the CLK− pin
to ground with a 0.1 μF capacitor (see Figure 42).
RMS CLOCK JITTER REQUIREMENT
16 BITS
3.3V
AD9516-x
OR AD9524
14 BITS
12 BITS
0.1µF
VFAC3
OUT
80
CLK
CMOS DRIVER
CLK
OPTIONAL
100Ω
0.1µF
70
50Ω*
CLK+
10 BITS
60
0.125ps
8 BITS
0.25ps
AD9670
50
0.1µF
0.5ps
1.0ps
2.0ps
CLK–
40
0.1µF
30
1
10
100
1000
*50Ω RESISTOR IS OPTIONAL.
ANALOG INPUT FREQUENCY (MHz)
Figure 42. Single-Ended, 1.8 V CMOS Sample Clock
Figure 43. Ideal SNR vs. Analog Input Frequency and Jitter
Clock Duty Cycle Considerations
Power Dissipation and Power-Down Mode
Typical high speed ADCs use both clock edges to generate a variety
of internal timing signals. As a result, these ADCs may be sensitive
to the clock duty cycle. Commonly, a 5% tolerance is required on
the clock duty cycle to maintain dynamic performance characteris-
tics. The AD9670 contains a duty cycle stabilizer (DCS) that retimes
the nonsampling edge, providing an internal clock signal with a
nominal 50% duty cycle. This allows a wide range of clock input
duty cycles without affecting the performance of the AD9670. When
the DCS is on, noise and distortion performance are nearly flat for
a wide range of duty cycles. However, some applications may require
the DCS function to be off. If so, keep in mind that the dynamic
range performance may be affected when operated in this mode.
The power dissipated by the AD9670 is proportional to its
sample rate. The digital power dissipation does not vary sig-
nificantly because it is determined primarily by the DRVDD
supply and the bias current of the LVDS output drivers. The
AD9670 features scalable LNA bias currents (see Table 27,
Register 0x012). The default LNA bias current settings are high.
By asserting the PDWN pin high, the AD9670 is placed into
power-down mode. In this state, the device typically dissipates
5 mW. During power-down, the LVDS output drivers are placed
into a high impedance state. The AD9670 returns to normal
operating mode when the PDWN pin is pulled low. This pin is
only 1.8 V tolerant. To drive the PDWN pin from a 3.3 V logic
level, insert a 1 kΩ resistor in series with this pin to limit the
current.
The duty cycle stabilizer uses a delay-locked loop (DLL) to
create the nonsampling edge. As a result, any changes to the
sampling frequency require approximately eight clock cycles to
allow the DLL to acquire and lock to the new rate.
By asserting the STBY pin high, the AD9670 is placed into a
standby mode. In this state, the device typically dissipates
630 mW. During standby, the entire device, except the internal
references, is powered down. The LVDS output drivers are
placed into a high impedance state. This mode is well suited for
applications that require power savings because it allows the
device to power down when not in use, and then be quickly
powered up. The time to power the device back up is also greatly
reduced. The AD9670 returns to normal operating mode when
the STBY pin is pulled low. This pin is only 1.8 V tolerant. To
drive the STBY pin from a 3.3 V logic level, insert a 1 kΩ resistor
in series with this pin to limit the current.
Clock Jitter Considerations
High speed, high resolution ADCs are sensitive to the quality of the
clock input. The degradation in SNR at a given input frequency (fA)
due only to aperture jitter (tJ) can be calculated as follows:
SNR Degradation = 20 × log10(1/2 × π × fA × tJ)
(7)
In this equation, the rms aperture jitter represents the root mean
square of all jitter sources, including the clock input, analog input
signal, and ADC aperture jitter (see Figure 43).
Treat the clock input as an analog signal when aperture jitter may
affect the dynamic range of the AD9670. Separate power supplies
for clock drivers from the ADC output driver supplies to avoid
modulating the clock signal with digital noise. Low jitter, crystal-
controlled oscillators make the best clock sources, such as the
Valpey Fisher VFAC3 series. If the clock is generated from another
type of source (by gating, dividing, or other methods), retime it
by the original clock during the last step.
In power-down mode, low power dissipation is achieved by
shutting down the reference, reference buffer, PLL, and biasing
networks. The decoupling capacitors on VREF are discharged
when entering power-down mode and must be recharged when
returning to normal operation. As a result, the wake-up time is
related to the time spent in the power-down mode: shorter cycles
result in proportionally shorter wake-up times. To restore the
device to full operation, approximately 375 ꢁs is required when
using the recommended 1 ꢁF and 0.1 ꢁF decoupling capacitors
on the VREF pin and the 0.01 ꢁF decoupling capacitors on the
GAIN pins. Most of this time is dependent on the gain decoupling:
For more information about how jitter performance relates to
ADCs, refer to the AN-501 Application Note and the AN-756
Application Note.
Rev. A | Page 28 of 52
Data Sheet
AD9670
higher value decoupling capacitors on the GAIN pins result in
longer wake-up times.
Advanced Power Control
Not all channels are required during all periods of scanning in
an ultrasound system. Use the POWER_START and POWER_
STOP values in the vector profile to delay the channel startup
and to turn the channel off after a certain number of samples.
These counters are relative to TX_TRIG . The analog circuitry
must power up before the digital circuitry, and the advance time
(POWER_SETUP) for powering up the analog circuitry, before
POWER_START, is set up in Register 0x112, Bits[4:0] (see
Table 27).
A number of other power-down options are available when using
the SPI port interface. The user can individually power down
each channel or put the entire device into standby mode. This
allows the user to keep the internal PLL powered up when fast
wake-up times are required. The wake-up time is slightly depend-
ent on gain. To achieve a 2 ꢁs wake-up time when the device is
in standby mode, 0.8 V must be applied to the GAIN pins.
Power and Ground Connection Recommendations
When connecting power to the AD9670, it is recommended that
two separate 1.8 V supplies be used: one for analog supply
(AVDD1) and one for digital supply (DRVDD). If only one
1.8 V supply is available, route this supply to the AVDD1 pin
first, and then tap the supply off and isolate it with a ferrite
bead or a filter choke preceded by decoupling capacitors for
the DRVDD pin.
POWER_STOP
(PROFILE SPECIFIC)
TX_TRIG±
POWER_START
(PROFILE SPECIFIC)
DIGITAL
POWER
ANALOG
POWER
If the user does not use the digital demodulator and decimator
functions for post ADC processing, the DVDD pin can be tied
to the 1.8 V DRVDD supply. If this is the case, route the DVDD
supply first, tapped the supply off, and isolated it with a ferrite
bead or filter choke preceded by decoupling capacitors for the
DRVDD pin. It is not recommended to use the same supply for
AVDD1, DVDD, and DRVDD.
POWER_SETUP
(SPI SET)
Figure 44. Power Sequencing
Digital Outputs and Timing
The AD9670 differential outputs conform to the ANSI-644
LVDS standard on default power-up. This standard can be
changed to a low power, reduced signal option similar to the
IEEE 1596.3 standard via the SPI using Register 0x015, Bit 7.
This LVDS standard can further reduce the overall power
dissipation of the device by approximately 36 mW.
Use several decoupling capacitors on all supplies to cover both
high and low frequencies. Locate these capacitors close to the
point of entry at the PCB level and close to the device, with
minimal trace lengths.
The LVDS driver current is derived on chip and sets the output
current at each output equal to a nominal 3.5 mA. A 100 Ω
differential termination resistor placed at the LVDS receiver
inputs results in a nominal 350 mV swing at the receiver.
A single PCB ground plane is sufficient when using the AD9670.
With proper decoupling and smart partitioning of the analog,
digital, and clock sections of the PCB, optimum performance
is easily achieved.
The AD9670 LVDS outputs facilitate interfacing with LVDS
receivers in custom ASICs and FPGAs that have LVDS capability
for superior switching performance in noisy environments. Single
point-to-point network topologies are recommended with a
100 Ω termination resistor placed as close to the receiver as
possible. No far-end receiver termination and poor differential
trace routing may result in timing errors. The trace length must
be no longer than 24 inches. Keep the differential output traces
close together and at equal lengths.
Rev. A | Page 29 of 52
AD9670
Data Sheet
An example of the LVDS output using the ANSI-644 standard
(default) data eye and a time interval error (TIE) jitter histogram
with trace lengths of less than 24 inches on regular FR-4 material is
shown in Figure 45 and Figure 46. Figure 47 and Figure 48 show
examples of the trace lengths exceeding 24 inches on regular
FR-4 material. Notice that the TIE jitter histogram reflects the
decrease of the data eye opening as the edge deviates from the
ideal position; therefore, the user must determine whether the
waveforms meet the timing budget of the design when the trace
lengths exceed 24 inches.
EYE: ALL BITS
ULS: 11199/11199
400
300
200
100
0
–100
–200
–300
–400
EYE: ALL BITS
ULS: 11197/11197
400
300
–1.5
–1.0
–0.5
0
0.5
1.0
1.5
TIME (ns)
Figure 47. Data Eye for LVDS Outputs in ANSI-644 Mode with Trace Lengths of
Greater Than 24 Inches on Standard FR-4 Material
200
100
80
70
60
50
40
30
20
10
0
0
–100
–200
–300
–400
–1.5
–1.0
–0.5
0
0.5
1.0
1.5
TIME (ns)
Figure 45. Data Eye for LVDS Outputs in ANSI-644 Mode with Trace Lengths
of Less Than 24 Inches on Standard FR-4 Material
70
60
50
40
30
20
10
0
–300
–200
–100
0
100
200
300
TIME (ps)
Figure 48. TIE Jitter Histogram for LVDS Outputs in ANSI-644 Mode with
Trace Lengths of Greater Than 24 Inches on Standard FR-4 Material
Additional SPI options allow the user to further increase the
internal current of all eight outputs to drive longer trace lengths.
Even though this produces sharper rise and fall times on the
data edges, it is less prone to bit errors and improves frequency
distribution. The power dissipation of the DRVDD supply
increases when this option is used.
In cases that require increased drive current, Register 0x015
allows the user to adjust the drivers from 2.0 mA to 3.72 mA.
Note that this feature requires Bit 3 of Register 0x015 to be
enabled. The drive current can be adjusted for both ANSI-644
and IEEE (low power) mode. See Table 27 for more details.
–150
–100
–50
0
50
100
150
TIME (ps)
Figure 46. TIE Jitter Histogram for LVDS Outputs in ANSI-644 Mode with
Trace Lengths of Less Than 24 Inches on Standard FR-4 Material
Rev. A | Page 30 of 52
Data Sheet
AD9670
The format of the output data is twos complement by default.
Table 14 provides an example of the output coding format. To
change the output data format to twos complement, see the
Memory Map section.
bypassed, the demodulator bypassed, and the baseband decimator
bypassed. Data is clocked out of the AD9670 and must be captured
on the rising and falling edges of DCO , which support double
data rate (DDR) capturing. The frame clock output (FCO )
signals the start of a new output byte and is equal to the sampling
clock rate.
Table 14. Digital Output Coding with RF Decimator Bypassed,
Demodulator Bypassed, and Baseband Decimator Bypassed
A 12-, 14-, or 16-bit serial stream can also be initiated from SPI
Register 0x021, Bits[1:0]. The user can implement different serial
streams and test device compatibility with lower and higher
resolution systems using these modes.
(VIN+) − (VIN−),
Digital Output Mode: Twos
Code Input Span = 2 V p-p (V) Complement (D13 to D0)
16384 +1.00
01 1111 1111 1111
00 0000 0000 0000
11 1111 1111 1111
10 0000 0000 0000
8192
8191
0
0.00
−0.000488
−1.00
When using the SPI, all the data outputs can also be inverted
from their nominal state by setting Bit 2 in the output mode
register (Register 0x014). This is not to be confused with inverting
the serial stream to an LSB first mode. In default mode, as shown
in Figure 2, the MSB is represented first in the data output serial
stream. However, this order this can be inverted so that the LSB is
represented first in the data output serial stream.
Digital data from each channel is serialized based on the
number of lanes that are enabled (see Table 27). The maximum
data rate for each serial output lane is 1 Gbps. For 1 channel/
lane with a 14-bit data stream and an ADC sample clock of
70 MHz, the output data rate is 980 Mbps (14 bits × 70 MHz =
980 Mbps) with the RF decimator bypassed, the demodulator
bypassed, and the baseband decimator bypassed. For higher
sample rates, enabling the RF decimator is required.
Output Zero Stuffing
A zero stuffing feature handles the various decimation rates and
complex (IQ) vs. real samples. As the decimation rates increase,
relatively large amounts of zero stuffing can occur in the output
data stream.
Two output clocks are provided to assist in capturing data from
the AD9670. DCO clocks the output data and is equal to seven
times the sampling clock rate in 14-bit mode with the RF decimator
Rev. A | Page 31 of 52
AD9670
Data Sheet
Table 15. Flexible Output Test Modes
Output Test Mode
Bit Sequence
Subject to
Resolution Select
Pattern Name
Off (default)
Midscale short
+Full-scale short
−Full-scale short
Checkerboard
PN sequence long
PN sequence short
One-/zero-word toggle 11 1111 1111 1111
User input Register 0x019 and Register 0x01A
Digital Output Word 1
Not applicable
Digital Output Word 2
Not applicable
Same
Same
Same
01 0101 0101 0101
Not applicable
Not applicable
00 0000 0000 0000
0000
0001
0010
0011
0100
0101
0110
0111
Not applicable
10 0000 0000 0000
11 1111 1111 1111
00 0000 0000 0000
10 1010 1010 1010
Not applicable
Yes
Yes
Yes
No
Yes
Yes
No
Not applicable
1000
Register 0x01B and Register 0x01C No
There are nine digital output test pattern options available that
can be initiated through the SPI. The test pattern options are
useful when validating receiver capture and timing. See Table 15
for the available output bit sequencing options. Some test patterns
have two serial sequential words and can be alternated in various
ways, depending on the test pattern chosen. Note that some
patterns may not adhere to the data format select option. In
addition, custom user defined test patterns can be assigned in
the user pattern registers (Address 0x019 through Address 0x01C).
All test mode options except pseudorandom number (PN)
sequence short and PN sequence long can support 8- to
SCLK Pin
The SCLK pin is required to operate the SPI port interface. It
has an internal 30 kΩ pull-down resistor that pulls this pin low
and is only 1.8 V tolerant. To drive the SCLK pin from a 3.3 V
logic level, insert a 1 kΩ resistor in series with this pin to limit
the current.
CSB Pin
The CSB pin is required to operate the SPI port interface. CSB
has an internal 70 kΩ pull-up resistor that pulls this pin high
and is only 1.8 V tolerant. To drive the CSB pin from a 3.3 V
logic level, insert a 1 kΩ resistor in series with this pin to limit
the current.
14-bit word lengths to verify data capture to the receiver.
The PN sequence short pattern produces a pseudorandom
bit sequence that repeats itself every 29 − 1 bits, or 511 bits. A
description of the PN sequence short and how it is generated
can be found in Section 5.1 of the ITU-T O.150 (05/96) standard.
The only difference from the standard is that the starting value is a
specific value instead of all 1s (see Table 16 for the initial values).
RBIAS Pin
To set the internal core bias current of the ADC, place a resistor
nominally equal to 10.0 kΩ to ground at the RBIAS pin. Using a
resistor other than the recommended 10.0 kΩ resistor for RBIAS
degrades the performance of the device. Therefore, it is imperative
that at least a 1% tolerance on this resistor be used to achieve
consistent performance.
The PN sequence long pattern produces a pseudorandom bit
sequence that repeats itself every 223 − 1 bits, or 8,388,607 bits.
A description of the PN sequence long and how it generates is
found in Section 5.6 of the ITU-T O.150 (05/96) standard. The
only differences from the standard are that the starting value is
a specific value instead of all 1s, and that the AD9670 inverts the
bit stream (see Table 16 for the initial values). The output sample
size depends on the selected bit length.
VREF Pin
A stable and accurate 0.5 V voltage reference is built in to the
AD9670. This reference is gained up internally by a factor of 2,
setting VREF to 1.0 V, which results in a full-scale differential
input span of 2.0 V p-p for the ADC. VREF is set internally by
default, but the VREF pin can be driven externally with a 1.0 V
reference to achieve more accuracy. However, the AD9670 does
not support ADC full-scale ranges below 2.0 V p-p.
Table 16. PN Sequence
First Three Output Samples
PN Sequence
Short
Long
Initial Value
0x092
0x003
(MSB First, 16-Bit)
When applying the decoupling capacitors to the VREF pin,
use ceramic, low ESR capacitors. Place these capacitors close to
the reference pin and on the same layer of the PCB as the AD9670.
It is recommended that the VREF pin have both a 0.1 ꢁF
capacitor and a 1 ꢁF capacitor that are connected in parallel to
the analog ground. These capacitor values are recommended for
the ADC to properly settle and acquire the next valid sample.
0x496F, 0xC9A9, 0x980C
0xFF5C, 0x0029, 0xB80A
See the Memory Map section for information on how to change
these additional digital output timing features via the SPI.
SDIO Pin
The SDIO pin is required to operate the SPI. It has an internal
30 kΩ pull-down resistor that pulls this pin low and is only 1.79 V
tolerant. If applications require that SDIO be driven from a
3.3 V logic level, insert a 1 kΩ resistor in series with this pin
to limit the current.
General-Purpose Output Pins
The general-purpose output pins, GPO0, GPO1, GPO2, and
GPO3, can be used in a system to provide programmable inputs
to other chips in the system. The value of each pin is set via SPI
Register 0x00E to either Logic 0 or Logic 1 (see Table 27).
Rev. A | Page 32 of 52
Data Sheet
AD9670
Chip Address Pins
CW DOPPLER OPERATION
The chip address pins can be used to SPI address individual
AD9670 devices in a system. When chip address mode is enabled
in Register 0x115, Bit 5 (see Table 27), if the value written to
Bits[4:0] matches the value on the chip address bit pins (ADDR0 to
ADDR4), the device is selected and any subsequent SPI writes
or reads to registers indicated as chip registers are written only
to that device. If chip address mode is disabled, all registers can
be written to, regardless of the value on the address pins.
Each channel of the AD9670 includes a I/Q demodulator. Each
demodulator has an individual programmable phase shifter.
The I/Q demodulator is ideal for phased array beamforming
applications used in medical ultrasound. Each channel can be
programmed for 16 delay states/360° (or 22.5°/step), selectable
via the SPI port. The device has a RESET input that synchro-
nizes the LO dividers of each channel. If multiple AD9670 devices
are used, a common reset across the array ensures a synchronized
phase for all channels. Internal to the AD9670, the individual
Channel I and Channel Q outputs are current summed. If
multiple AD9670 devices are used, the I and Q outputs from
each AD9670 can be current summed and converted to a voltage
using an external transimpedance amplifier.
ANALOG TEST SIGNAL GENERATION
The AD9670 generates analog test signals that can be switched
to the input of the LNA of each channel to be used for channel
gain calibration. The test signal amplitude at the LNA output is
dependent on LNA gain, as shown in Table 17.
Quadrature Generation
Table 17. Test Signal Fundamental Amplitude at the LNA
Output
The internal 0° and 90° LO phases are digitally generated by a
divide-by-M logic circuit, where M = 4, 8, or 16. The internal
divider is selected via SPI Register 0x02E, Bits[2:0] (see Table 27).
The divider is dc-coupled and inherently broadband; the max-
imum LO frequency is limited only by its switching speed. The
duty cycle of the quadrature LO signals must be as close to 50% as
possible for the 4LO and 8LO modes. The 16LO mode does not
require a 50% duty cycle. Furthermore, the divider is implemented
such that the MLO signal reclocks the final flip-flops that generate
the internal LO signals and, thereby, minimizes noise introduced
by the divide circuitry.
LNA Gain
Reg. 0x116, Bits[3:2], At 15.6 dB At 17.9 dB At 21.6 dB
Analog Test Tones
(mV p-p)
(mV p-p)
(mV p-p)
00 (Default)
80
98
119
01
10
160
320
196
391
238
476
Calculate the test signal amplitude at the input to the ADC
given the LNA gain, attenuator control voltage, and the PGA
gain. Table 18 and Table 19 show example calculations.
For optimum performance, the MLO input is driven differentially,
as on the AD9670 evaluation board. The common-mode voltage
on each pin is approximately 1.5 V with the nominal 3 V supply.
It is important to ensure that the MLO source has very low phase
noise (jitter), a fast slew rate, and an adequate input level to
obtain optimum performance of the CW signal chain.
Table 18. Test Signal Fundamental Amplitude at the
ADC Input, VGAIN = 0 V, PGA Gain = 21 dB
LNA Gain
Register 0x116,
Bits[3:2],
Analog Test Tones
At 15.6 dB At 17.9 dB At 21.6 dB
(dBFS)
(dBFS)
(dBFS)
00 (Default)
−29
−28
−26
Beamforming applications require a precise channel to channel
phase relationship for coherence among multiple channels. A
RESET input is provided to synchronize the LO divider circuits in
different AD9670 devices when they are used in arrays. The
RESET input is a synchronous edge triggered input that resets
the dividers to a known state after power is applied to multiple
AD9670 devices. The RESET signal can be either a continuous
signal or a single pulse, and it can be either synchronized with the
MLO clock edge (recommended) or it can be asynchronous. If a
continuous signal is used for the RESET signal, it must be at the
LO rate. For a synchronous RESET , the device can be configured
to sample the RESET signal with either the falling or rising edge
of the MLO clock, which makes it easier to align the RESET
signal with the opposite MLO clock edge. Use Register 0x02E to
configure the RESET signal behavior. Synchronize the RESET
input to the MLO input. Accurate channel-to-channel phase
matching can be achieved via a common clock on the RESET
input when using more than one AD9670.
01
10
−23
−17
−22
−16
−20
−14
Table 19. Test Signal Fundamental Amplitude at the
ADC Input, VGAIN = 0 V, PGA Gain = 30 dB
LNA Gain
Register 0x116,
Bits[1:0],
Analog Test Tones
At 15.6 dB
(dBFS)
At 17.9 dB At 21.6 dB
(dBFS)
(dBFS)
00 (Default)
01
10
−20
−14
−8
−19
−13
−7
−17
−11
−5
Rev. A | Page 33 of 52
AD9670
Data Sheet
Table 20. Phase Select Code for Channel-to-Channel Phase Shift
I/Q Demodulator and Phase Shifter
I/Q Demodulator Phase
(SPI Register 0x02D, Bits[3:0])
The I/Q demodulators consist of double-balanced, harmonic
rejection, passive mixers. The RF input signals are converted
into currents by transconductance stages that have a maximum
differential input signal capability matching the LNA output full
scale. These currents are then presented to the mixers, which
convert them to baseband (RF − LO) and 2× RF (RF + LO).
The signals are phase shifted according to the codes programmed
into the SPI latch (see Table 27). The phase shift function is an
integral part of the overall circuit. The phase shift listed in Table 20
is defined as being between the baseband I or Q channel outputs.
As an example, for a common signal applied to a pair of RF inputs
to an AD9670, the baseband outputs are in phase for matching
phase codes. However, if the phase code for Channel 1 is 0000
and that of Channel 2 is 0001, Channel 2 leads Channel 1 by
22.5°.
Φ Shift
0°
22.5°
45°
67.5°
90°
112.5°
135°
157.5°
180°
202.5°
225°
247.5°
270°
292.5°
315°
0000
0001 (not valid in 4LO mode)
0010
0011 (not valid in 4LO mode)
0100
0101 (not valid in 4LO mode)
0110
0111 (not valid in 4LO mode)
1000
1001 (not valid in 4LO mode)
1010
1011 (not valid in 4LO mode)
1100
1101 (not valid in 4LO mode)
1110
337.5°
1111 (not valid in 4LO mode)
Rev. A | Page 34 of 52
AD9670
Data Sheet
DIGITAL DEMODULATOR/DECIMATOR
The AD9670 contains digital processing capability. Each channel
has three stages of processing available: the RF decimator, the
baseband demodulator, and the baseband decimator. For test
purposes, the input to the demodulator/decimator can be a test
waveform. Normally, this input is the output of the ADC. The
output of the demodulator/decimator is sent to the framer/
serializer for output formatting.
must be read using stream mode. When writing or reading in
stream mode while the SPI configuration is set to MSB first
mode (the default setting for Register 0x000), the write/read
address must refer to the last register address, not the first. For
example, when writing or reading the first profile that spans the
address space between Register 0xF00 and Register 0xF07, with
the SPI port configured as MSB first, the referenced address
must be Register 0xF07 to allow reading or writing the 64
profile bits in MSB mode. For more information, see the AN-877
Application Note, Interfacing to High Speed ADCs via SPI.
The maximum data rate of the baseband demodulator and
decimator is 65 MSPS. Therefore, if the sample of the ADC is
greater than 65 MSPS, the RF decimator (with a fixed rate of 2)
must be enabled. The ADC resolution is 14 bits. The maximum
resolution at the output of the digital processing is 16 bits. Sat-
uration of the ADC is determined after the dc offset calibration to
ensure maximum dynamic range. Depending on the decimation
rate, the loss in output SNR due to truncation to 16 bits is
negligible.
A buffer stores the current profile data. When the profile index
is written in Register 0x10C, the selected profile is read from
memory and stored in the current profile buffer. The profile
memory is read/written in the SPI clock domain. After the SPI
writes the profile index value, the SPI takes 4 SPI clock cycles to
read the profile from RAM and store it in the current profile
buffer. If the SPI is in LSB mode, these additional SPI clock
cycles are provided when the profile index register is written. If
the SPI is in MSB mode, an additional byte must be read or
written to update the profile buffer.
VECTOR PROFILE
To minimize time needed to reconfigure device settings during
operation, the device supports configuration profiles. Up to
32 profiles can be stored in the device. A profile is selected by a
5-bit index. A profile consists of a 64-bit vector, as described in
Table 21. Each parameter is concatenated to form the 64-bit
profile vector. The profile memory starts at Register 0xF00 and
ends at Register 0xFFF. The memory can be written in either
stream or address selected data mode. However, the memory
Updating the profile memory does not affect the data in the
profile buffer. The profile index register must be written to
cause a refresh of the current profile data, even if the profile
index register is written with the same value.
NUMERICALLY
CONTROLLED
OSCILLATOR
–Sin
Cos
BB DECIMATOR
LOW-PASS
FILTER
ADC OUTPUT OR
TEST WAVEFORM
MULTIBAND AAF
DECIMATE BY 2
HIGH-PASS
FILTER
DECIMATOR
I
DC OFFSET
CALIBRATION
FRAMER
SERIALIZER
LOW-PASS
FILTER
DECIMATOR
Q
RF DECIMATOR
BB DEMODULATOR
Figure 49. Simplified Block Diagram of a Single Channel of the Demodulator/Decimator
Table 21. Profile Definition
Field
Bits Description
f
16 Demodulation frequency (fD)
fD = f × fSAMPLE/216, where f = [0,(216 − 1)] and fSAMPLE is the effective sample rate
0x0000: fD = 0 (dc, I = cos(0) = 1, Q = sin(0) = 0)
0x0001: fD = fSAMPLE/216
…
0x8000: fD = fSAMPLE/2
…
0xFFFF (216 − 1): fD = fSAMPLE (216 − 1)/ 216 =−fSAMPLE/216
P
8
Pointer to coefficient block; the coefficients used begin at coefficient P × 8 and continue for M × 8 coefficients, for
example,
0000 0000: points to coefficient 0 and continues M × 8 coefficients
0000 0001: points to coefficient 8 and continues M × 8 coefficients
Rev. A | Page 35 of 52
AD9670
Data Sheet
Field
Bits Description
M
5
3
1
Decimation factor
M = N − 1, where N = decimation factor
0x00: decimate by 1 (no decimation, filtering only)
0x01: decimate by 2
0x02: decimate by 3
…
0x1F: decimate by 32
g
Digital gain compensation
Gain = 2
000: gain = 1 (no shift)
001: gain = 2 (shift by 1)
010: gain = 4 (shift by 2)
…
111: gain = 128 (shift by 7)
HPF Bypass
Digital high-pass filter (HPF) bypass
0 = disable (filter enabled)
1 = enable (filter bypassed)
POWER_START 15
ADC clock counted from TX_TRIG1 signal assertion when the active channels are powered up
Reserved
1
Reserved
POWER_STOP
15
ADC clock counted from TX_TRIG1 signal assertion when the active channels are powered down
1 TX_TRIG indicates the differential signal created via the TX_TRIG− pin and the TX_TRIG+ pin.
10
0
RF DECIMATOR
The input to the RF decimator is either the ADC output data or
a test waveform described in the Digital Test Waveforms section.
The test waveforms are enabled per channel in Register 0x11A
(see Table 27).
LOW BAND FILTER
HIGH BAND FILTER
–10
–20
–30
–40
–50
–60
DC Offset Calibration
Reduce dc offset through a manual system calibration process.
Measure the dc offset of every channel in the system and then
set a calibration value in Register 0x110 and Register 0x111.
Note that these registers are both chip and local registers, meaning
that they are accessed using the chip address and device index.
Bypass the dc offset calibration in Register 0x10F, Bits[2:0].
0
2
4
6
8
10
12
14
16
18
20
FREQUENCY (MHz)
Multiband Antialiasing Filter and Decimate by 2
Figure 50. Antialiasing Filter Frequency Response (Frequency Scale Assumes
fADC = 2 × fDEC = 40 MHz)
The multiband filter is a finite pulse response (FIR). It is program-
mable with low or high band filtering. The filter requires 11 input
samples to populate. The decimation rate is fixed at 2×. Therefore,
the decimation frequency is fDEC = fSAMPLE/2. Figure 50 and Figure 51
show the frequency response of the filter, depending on the mode.
Figure 50 shows the attenuation amplitude over the Nyquist
frequency range. Figure 51 shows the pass band response as
nearly flat.
2
1
0
–1
LOW BAND FILTER
HIGH BAND FILTER
–2
–3
–4
–5
–6
–7
–8
0
2
4
6
8
10
12
14
16
18
20
FREQUENCY (MHz)
Figure 51. Antialiasing Filter Frequency Response (Frequency Scale Assumes
fADC = 2 × fDEC = 40 MHz)
Rev. A | Page 36 of 52
Data Sheet
AD9670
Table 22. Coefficient Memory for M = 4
High-Pass Filter
j\i
7
6
5
4
3
2
1
4
5
6
7
0
3
2
1
0
A second-order Butterworth, high-pass infinite impulse response
(IIR) filter can be applied after the RF decimator. The filter has a
settling time of 2.5 ꢁs and a cutoff of 700 kHz for an encode
clock of 50 MHz. Therefore, if the ADC clock is 50 MHz, the
first 125 samples (2.5 ꢁs/0.02 ꢁs) must be ignored. The filter
can be bypassed or enabled in the vector profile if the filter is
enabled in Register 0x113, Bit 5. If the filter is bypassed by
setting Register 0x113, Bits[5:1], the filter cannot be enabled
from the vector profile.
0
1
2
3
28
29
30
31
27
26
25
24
20
21
22
23
19
18
17
16
12
13
14
15
11
10
9
8
Writes and reads from a coefficient block must begin on a
coefficient block boundary, and an entire coefficient block must
be written or read. After a coefficient block is written, the
coefficient block address automatically increments/decrements
(depending on the LSB/MSB SPI setting in Register 0x000) to
the next coefficient block.
BASEBAND DEMODULATOR AND DECIMATOR
The demodulator downconverts the RF signal to a baseband
Having a direct map between the SPI memory address and
coefficient block address requires a divide by 7, which is not
simple to do in hardware (the address must be mapped within a
single cycle). Therefore, each block is padded to a 16-byte
boundary, but the SPI does not need to shift in these extra
2 bytes when loading coefficient memory sequentially. If the
SPI is configured LSB first, the SPI address bits, Bits[3:0], are all
0s. If the SPI is configured MSB first, the SPI address bits are all
1s. In other words, in LSB mode, the referenced addresses for
the coefficient memory blocks are 0x1000, 0x2000, and so on,
while in MSB SPI mode, the referenced block addresses are
0x100F, 0x200F, and so on.
quadrature signal. The decimator reduces the excess oversampling.
Numerically Controlled Oscillator
The numerically controlled oscillator (NCO) generates I and Q
signals (cos and −sin) for the demodulator. A division of the
effective sample clock generates the oscillator frequency. If the
RF decimator is bypassed, the effective sample clock is the same
as the ADC clock. If the RF decimator is enabled, the effective
clock rate is ½ the ADC sample clock frequency. The divider is
set in the vector profile. The oscillator has a frequency resolution
of 1 kHz. To synchronize different devices, the NCO is reset
upon assertion of the TX_TRIG signal.
Decimation Filter
The coefficient block order and how words/bytes are split across
each other are shown in Table 23. When the SPI is configured
LSB first, C0[0] = B0[0] is written first, and C7[13] = B13[7] is
written last. When the SPI is configured MSB first, C7[13] =
B13[7] is written first, and C0[0] = B0[0] is written last.
The purpose of the decimation filter is to band limit the
demodulated signal prior to decimation. The filter is a
polyphase FIR filter that uses 16 taps per decimation with
symmetrical coefficients. Therefore, there are eight unique,
14-bit coefficients per decimation. The decimation rate and a
pointer to the coefficients used by the filter are set in the vector
profile. Digital gain from 1 to 128 is applied to the filter response.
The digital gain compensation is set in the vector profile.
The position of a coefficient, Cn, in memory is determined
from its index (i, j) by
n = M(1 + i) − (1 + j), if i is even
n = M × i + j, if i is odd
(8)
(9)
The filter is reset upon assertion of the TX_TRIG signal. The
decimation filter takes 32× the decimation input samples or
32 output samples to populate.
where:
M is the decimation factor.
i is the index within the coefficient block from 0 to 7.
j is the decimation phase from 0 to M − 1.
Coefficient Memory
The coefficient memory stores the eight coefficients per
decimation, with a maximum decimation of 32, in a coefficient
memory block. At a maximum decimation of 32, 32 × 8 = 256
coefficients are needed. The coefficient memory is available at
Address 0x1000 to Address 0x1FFF. This is sufficient space to
store up to 2048 coefficients. Each vector profile has a pointer, P, to
the coefficient block within coefficient memory.
Due to symmetry, Coefficient C0 is multiplied by the newest
and oldest samples.
As an example, the coefficient memory for a decimation factor
of M = 4 is shown in Table 22.
The upper 16 bits of the filter output are used as the data output
of the channel. The filter output may have gain applied according
to g, from the vector profile. Additionally, a gain of 4× can be
applied using the filter output gain in Register 0x113, Bit 4.
Coefficients are written using the SPI in stream mode during
startup. Coefficients are written in 14-bit × 8-word = 112-bit
blocks. There are 256 coefficient blocks. The 14-bit × 8-word
coefficients are packed into 14 bytes × 8 bits, as shown in Table 22.
Rev. A | Page 37 of 52
AD9670
Data Sheet
Table 23. Coefficient Block Mapping into SPI Memory Location
Coefficients (8 Words × 14 Bits)
C7[13:0]
C6[13:0]
C5[13:0]
C4[13:0]
69:56
C3[13:0]
C2[13:0]
C1[13:0] C0[13:0]
111:98
97:84
83:70
55:42
41:28
27:14
13:0
SPI Memory (14 Bytes)
B13[7:0] B12[7:0] B11[7:0] B10[7:0] B9[7:0] B8[7:0] B7[7:0] B6[7:0] B5[7:0] B4[7:0] B3[7:0] B2[7:0] B1[7:0]
111:104 103:96 95:88 87:80 79:72 71:64 63:56 55:48 47:40 39:32 31:24 23:16 15:8
B0[7:0]
7:0
DIGITAL BLOCK POWER SAVING SCHEME
DIGITAL TEST WAVEFORMS
To reduce power consumption in the digital block, the
Digital test waveforms can be used in the digital processing block
instead of the ADC output. The digital test waveforms enable is
set in Register 0x11B. Each channel can be individually enabled
demodulator and decimation filter start in an idle state after
running the chip (Register 0x008, Bits[2:0] = 000). The digital
block only switches to a running state when the negative edge of
the TX_TRIG pulse is detected, or with a software TX_TRIG write
(Register 0x10C, Bit 5 = 1).
in Register 0x11A.
Waveform Generator
For testing and debugging, use the programmable waveform
generator instead of ADC data. The waveform generator varies
offset, amplitude, and frequency. The generator uses the ADC
sample frequency, fSAMPLE, and ADC full-scale amplitude, AFULL SCALE
as references. The values are set in Register 0x117, Register 0x118,
and Register 0x119 (see Table 27).
To put the digital block back into the idle state while the rest of
the chip is still running and to save power, enact one of the
following three events: raise the TX_TRIG signal high, write to
the profile index (Register 0x10C, Bits[4:0]), or the power stop
expires if the advanced power control feature is used. Figure 52
illustrates the digital block power saving scheme.
,
x = C + A × sin(2 × π × N)
fSAMPLE n
(10)
CHIP IN POWER-DOWN,
STANDBY,
OR CW MODE
N =
(see Register 0x117)
(11)
64
AFULL SCALE
2x
RUN CHIP
A =
(see Register 0x118)
(12)
(13)
DIGITAL
DEMODULATOR
IDLE
C = AFULL SCALE × a × 2−(13 − b) (see Register 0x119)
TX_TRIG IS HIGH, PROFILE
NEGATIVE EDGE TX_TRIG
INDEX WRITE, OR POWER
Channel ID and Ramp Generator
OR S/W TX_TRIG
STOP EXPIRES
In Channel ID test mode, the output is a concatenated value.
Output Data Bits[6:0] are a ramp. Output Data Bit 7 is 0 in real
data mode or I channel and 1 for Q channel in complex data
mode. Output Data Bits[10:8] are the channel ID such that
Channel A is coded as 000 and Channel B is 001. Output Data
Bits[15:11] are the chip address.
DIGITAL
DEMODULATOR
RUNNING
Figure 52. Digital Block Power Saving Scheme
Filter Coefficients
To check the filter coefficients, the input to the decimating FIR
filter must be a sequence of 1 followed by 0s. The number of zeros
is the decimation rate times the number of taps (16). The output
shifter outputs the LSBs of the filter.
Rev. A | Page 38 of 52
Data Sheet
AD9670
SERIAL PORT INTERFACE (SPI)
The AD9670 serial port interface allows the user to configure
the signal chain for specific functions or operations through a
structured register space provided inside the chip. The SPI
offers the user added flexibility and customization, depending
on the application. Addresses are accessed via the serial port
and can be written to or read from via the port. Memory is
organized into bytes that can be further divided into fields, as
documented in the Memory Map section. For detailed
operational information, see the AN-877 Application Note,
Interfacing to High Speed ADCs via SPI.
the only pins required for communication. Although the device is
synchronized during power-up, caution must be exercised when
using this mode to ensure that the serial port remains synchronized
with the CSB line. When operating in 2-wire mode, it is recom-
mended that a 1-, 2-, or 3-byte transfer be used exclusively. Without
an active CSB line, streaming mode can be entered but not exited.
In addition to word length, the instruction phase determines
whether the serial frame is a read or write operation, allowing
the serial port to be used both to program the chip and to read
the contents of the on-chip memory. If the instruction is a read-
back operation, performing a readback causes the serial data
input/output (SDIO) pin to change direction from an input to
an output at the appropriate point in the serial frame.
Three pins define the serial port interface, or SPI: SCLK, SDIO,
and CSB (see Table 24). The SCLK (serial clock) pin synchro-
nizes the read and write data presented to the device. The SDIO
(serial data input/output) pin is a dual-purpose pin that allows
data to be sent to and read from the internal memory map registers
of the device. The CSB (chip select bar) pin is an active low
control that enables or disables the read and write cycles.
Data can be sent in MSB first mode or LSB first mode. MSB
first mode is the default at power-up and can be changed by
adjusting the configuration register. For more information
about this and other features, see the AN-877 Application Note,
Interfacing to High Speed ADCs via SPI.
Table 24. Serial Port Pins
Pin
Function
HARDWARE INTERFACE
SCLK
Serial clock. Serial shift clock input. SCLK synchronizes
serial interface reads and writes.
Serial data input/output. SDIO is a dual-purpose pin
that typically serves as an input or an output, depending
on the instruction sent and the relative position in the
timing frame.
The pins described in Table 24 constitute the physical interface
between the user programming device and the serial port of the
AD9670. The SCLK and CSB pins function as inputs when
using the SPI. The SDIO pin is bidirectional, functioning as an
input during write phases and as an output during readback.
SDIO
CSB
Chip select bar (active low). This control gates the read
and write cycles.
If multiple SDIO pins share a common connection, ensure that
proper VOH levels are met. Figure 53 shows the number of SDIO
pins that can be connected together and the resulting VOH level,
assuming the same load for each AD9670.
The falling edge of CSB, in conjunction with the rising edge of
SCLK, determines the start of the framing sequence. During an
instruction phase, a 16-bit instruction is transmitted, followed
by one or more data bytes, which is determined by Bit Field W0
and Bit Field W1. An example of the serial timing and its defini-
tions are shown in Figure 54 and Table 25.
1.800
1.795
1.790
1.785
1.780
1.775
1.770
1.765
1.760
1.755
1.750
1.745
1.740
1.735
1.730
1.725
1.720
1.715
During normal operation, CSB signals to the device that SPI
commands are to be received and processed. When CSB is brought
low, the device processes SCLK and SDIO to execute instructions.
Normally, CSB remains low until the communication cycle is
complete. However, if connected to a slow device, CSB can be
brought high between bytes, allowing older microcontrollers
enough time to transfer data into shift registers. CSB can be
stalled when transferring one, two, or three bytes of data. When
W0 and W1 are set to 11, the device enters streaming mode and
continues to process data, either reading or writing, until CSB is
taken high to end the communication cycle. This allows complete
memory transfers without the need for additional instructions.
Regardless of the mode, if CSB is taken high in the middle of a
byte transfer, the SPI state machine is reset and the device waits
for a new instruction.
0
10
20
30
40
50
60
70
80
90
100
NUMBER OF SDIO PINS CONNECTED TOGETHER
Figure 53. SDIO Pin Loading
This interface is flexible enough to be controlled either by serial
PROMs or by PIC microcontrollers, which provides the user with
an alternative to a full SPI controller for programming the
device (see the AN-812 Application Note, Microcontroller-Based
Serial Port Interface (SPI®) Boot Circuit).
In addition to the operation modes, the SPI port can be configured
to operate in different manners. CSB can also be tied low to
enable 2-wire mode. When CSB is tied low, SCLK and SDIO are
Rev. A | Page 39 of 52
AD9670
Data Sheet
tDS
tHIGH
tCLK
tH
tS
tDH
tLOW
CSB
DON’T
SCLK
DON’T
CARE
CARE
DON’T
CARE
DON’T
CARE
R/W
W1
W0
A12
A11
A10
A9
A8
A7
SDIO
D5
D4
D3
D2
D1
D0
Figure 54. Serial Timing Diagram
Table 25. Serial Timing Definitions
Parameter
Timing (ns min)
Description
tDS
tDH
tCLK
tS
12.5
5
40
5
Setup time between the data and the rising edge of SCLK
Hold time between the data and the rising edge of SCLK
Period of the clock
Setup time between CSB and SCLK
tH
2
Hold time between CSB and SCLK
tHIGH
tLOW
tEN_SDIO
16
16
15
Minimum period that SCLK must be in a logic high state
Minimum period that SCLK must be in a logic low state
Minimum time for the SDIO pin to switch from an input to an output relative to the SCLK falling
edge (not shown in Figure 54)
tDIS_SDIO
15
Minimum time for the SDIO pin to switch from an output to an input relative to the SCLK rising
edge (not shown in Figure 54)
Rev. A | Page 40 of 52
Data Sheet
AD9670
MEMORY MAP
READING THE MEMORY MAP TABLE
RESERVED LOCATIONS
Each row in the memory map register table has eight bit
locations. The memory map is roughly divided into three
sections: the chip configuration register map (Address 0x000 to
Address 0x1A1), the profile register map (Address 0xF00 to
Address 0xFFF), and the coefficient register map (Address 0x1000
to Address 0x1FFF). Registers that are designated as local registers
utilize the device index in Address 0x004 and Address 0x005 to
determine to which channels of a device the command is applied.
Registers that are designated as chip registers utilize the chip
address mode in Address 0x115 to determine if the device is
selected to be updated by writing to the chip register.
Undefined memory locations must not be written to except
when writing the default values suggested in this data sheet.
Consider addresses that have values marked as 0 reserved and
write a 0 into their registers during power-up.
DEFAULT VALUES
After a reset, critical registers are automatically loaded with
default values. These values are indicated in Table 27, where an
X refers to an undefined feature.
LOGIC LEVELS
An explanation of various registers follows: “bit is set” is
synonymous with “bit is set to Logic 1” or “writing Logic 1 for
the bit.” Similarly, “bit is cleared” is synonymous with “bit is set
to Logic 0” or “writing Logic 0 for the bit.”
The leftmost column of the memory map indicates the register
address, and the default value is shown in the second rightmost
column. The Bit 7 (MSB) column is the start of the default
hexadecimal value given. For example, Address 0x009, the global
clock register, has a default value of 0x01, meaning that Bit 7 = 0,
Bit 6 = 0, Bit 5 = 0, Bit 4 = 0, Bit 3 = 0, Bit 2 = 0, Bit 1 = 0, and
Bit 0 = 1, or 0000 0001 in binary. This setting is the default for
the duty cycle stabilizer in the on condition.
RECOMMENDED STARTUP SEQUENCE
To save system power during programming, the AD9670
powers up in power-down mode. To start up the device and
initialize the data interface, the SPI commands listed in Table 26
are recommended. At a minimum, the profile memory for an
index of 0 must be written (Registers 0xF00 to Register 0xF07).
If additional profiles and coefficient memory are required, these
can be written after Profile File Memory 0.
For more information on the SPI memory map and other
functions, see the AN-877 Application Note, Interfacing to
High Speed ADCs via SPI.
Rev. A | Page 41 of 52
AD9670
Data Sheet
Table 26. SPI Write Start-Up Sequence Example
Register
0x000
0x002
0x0FF
0x004
0x005
0x113
Write Value
Description
0x3C
Initiates an SPI reset
0x0X (default)
0x01
Sets the speed mode to 40 MHz
Enables speed mode change (required after Register 0x002 writes)
Sets local registers to all channels
Sets local registers to all channels
0x0F
0x3F
0x03
Bypasses the demodulator and decimator; bypasses the RF decimator; enables the digital
high-pass filter
0x011
0xF00
0xF01
0xF02
0xF03
0xF04
0xF05
0xF06
0xF07
0x06 (default)
0xFF
Sets LNA gain= 21.6 dB, sets VGA gain = external, and sets PGA gain = 24 dB
Enables continuous run mode; do not power down channels (POWER_STOP LSB)
Enables continuous run mode; do not power down channels (POWER_STOP MSB)
Powers up all channels, 0 clock cycles after TX_TRIG signal assertion (POWER_START LSB)
Bypasses the digital high-pass filter (POWER_START MSB)
0x7F
0x00
0x80
0x0C
Decimates by 2 (M = 00001); digital gain = 16 (g = 100)
0x00
Points to Coefficient Block 00
0x00
Demodulation frequency = fSAMPLE/8
0x20
Additional profile memory and coefficient memory can be written here
0x10C1
0x014
0x008
0x021
0x199
0x19B
0x188
0x18B
0x18C
0x182
0x10C3
0x00F
0x02B
0x00 (default)
0x00
Sets index profile (required after profile memory writes)
Sets the output data format
TGC run mode2
0x00
0x05
14 bits, 8 lanes, FCO covers the entire frame
Enables automatic clocks per sample calculation
Serial format
0x80
0x50
0x01
Enables the start code
0x27
Sets the start code MSB
0x72
Sets the start code LSB
0x82
Autoconfigures the PLL
0x20
Sets SPI TX_TRIG and index profile2
Sets the low-pass filter cutoff frequency, and mode
Sets the analog LPF and HPF to defaults, tune filters4
0x18
0x40
1 Setting the profile index requires an additional SPI write in SPI MSB mode before the chip is run to complete the current profile buffer update.
2 Running the chip from full power-down mode requires 375 ꢀs wake-up time, as listed in Table 3.
3 Soft TX_TRIG switches the demodulator/decimator digital block to a running state. The soft TX_TRIG may not be needed if a hardware TX_TRIG signal is used to run
the digital block.
4 Tuning the filters requires 512 ADC clock cycles.
Rev. A | Page 42 of 52
Data Sheet
AD9670
Table 27. Memory Map Registers
Addr.
Default
(Hex)
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0 (LSB) Value
Comments
Chip Configuration Registers
0x000 CHIP_PORT_CONFIG
0
LSB first
0 = off
(default)
1 = on
SPI reset
0 = off
(default)
1 = on
1
1
SPI reset
0 = off
(default)
1 = on
LSB first
0 = off
(default)
1 = on
0
0x18
Mirror nibbles so
that LSB or MSB
first mode is set
correctly, regard-
less of shift mode.
An SPI reset reverts
all registers (includ-
ing the LVDS
registers), except
Register 0x000, to
their default
values, and
Register 0x000,
Bits[2:5] are
automatically
cleared.
0x001 CHIP_ID
Chip ID, Bits[7:0]
(AD9670 = 0xA6) (default)
0x7C
0x0X
Default is unique
chip ID, different
for each device;
read-only register.
0x002 CHIP_GRADE
X
X
X
X
Speed mode, Bits[5:4]
(identify device
X
X
X
X
X
X
X
Speed mode is used
to differentiate the
ADC speed power
modes (the user
must update
Reg. 0x0FF to
initiate the mode
setting).
variants of chip ID)
00: Mode I (40 MSPS)
(default)
01: Mode II (65 MSPS)
10: Mode III (80 MSPS)
11: Mode III (125 MSPS)
0x0FF DEVICE_UPDATE
X
X
X
0x00
A write to Reg.
0x0FF (the write
value does not
matter) resets all
default register
values (analog and
ADC registers only,
not LVDS registers
and not Reg. 0x000
or Reg. 0x002,
Bits[5:4]) if
Register 0x002 has
been previously
written since the
last reset/load of
defaults.
0x004 DEVICE_INDEX_2
0x005 DEVICE_INDEX_1
X
X
X
X
X
X
Data
Data
Data
Data
0x0F
0x3F
Bits are set to
Channel H Channel G
0 = off
1 = on
(default)
Channel F Channel E
0 = off
1 = on
(default)
determine which
on-chip device
receives the next
write command.
0 = off
1 = on
(default)
0 = off
1 = on
(default)
Clock
Clock
Data
Data
Data
Data
Bits are set to
channel
DCO
0 = off
1 = on
(default)
channel
FCO
0 = off
1 = on
(default)
Channel D Channel C
Channel B Channel A
determine which
on-chip device
receives the next
write command.
0 = off
1 = on
0 = off
1 = on
0 = off
1 = on
0 = off
1 = on
(default)
(default)
(default)
(default)
0x008 GLOBAL_MODES
X
X
LNA input
imped-
ance
0 = 6 kΩ
(default)
1 = 3 kΩ
X
0
0
Internal power-down mode
000 = chip run (TGC mode)
001 = full power-down (default)
010 = standby
011 = reset all LVDS registers
100 = CW mode (TGC power-down)
0x01
Determines the
generic modes of
chip operation
(global).
0x009 GLOBAL_CLOCK
0x00A PLL_STATUS
X
X
X
X
X
X
X
X
X
X
DCS
0x01
0x00
Turns the internal
DCS on and off
(global).
0 = off
1 = on
(default)
PLL lock
status
0 = not
locked
X
X
X
Monitors the PLL
lock status (read
only, global).
1 = locked
Rev. A | Page 43 of 52
AD9670
Data Sheet
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x00D TEST_IO
User test
mode
0 = con-
X
Reset PN
long gen-
eration
0 = on,
Reset PN
short gen-
eration
0 = on,
Output test mode
0000 = off (default)
0001 = midscale short
0010 = +FS short
0x00
When this register
is set, the test data
is placed on the
output pins in
tinuous,
repeat user
patterns
PN long
running
(default)
1 = off,
PN long
held in
PN short
running
(default)
1 = off,
PN short
held in
0011 = −FS short
place of normal
data (local).
0100 = checkerboard output
0101 = PN sequence long
0110 = PN sequence short
0111 = one-/zero-word toggle
1000 = user input
(1, 2, 3, 4,
1, 2, 3, 4 …)
(default)
1 = single
clock cycle
user pat-
terns, then
zeros (1, 2,
3, 4, 0, 0 …)
reset
reset
1001:1110 = reserved
1111 = ramp output
0x00E GPO
X
X
X
X
General-purpose digital outputs
0x00
0x18
Values placed on
the GPO0 to
GPO3 pins (global).
0x00F FLEX_CHANNEL_
INPUT
Filter cutoff frequency control
0 0000 = 1.45 × (1/3) × fSAMPLE
0 0001 = 1.25 × (1/3) × fSAMPLE
0 0010 = 1.13 × (1/3) × fSAMPLE
BW mode
0 = low
X
X
Antialiasing filter
cutoff (global).
(default,
8 MHz to
18 MHz)
1 = high
(13.5 MHz
to 30 MHz)
0 0011 = 1.0 × (1/3) × fSAMPLE (default)
0 0100 = 0.9 × (1/3) × fSAMPLE
0 0101 = 0.8 × (1/3) × fSAMPLE
0 0110 = 0.75 × (1/3) × fSAMPLE
0 0111 = reserved
0 1000 = 1.45 × (1/4.5) × fSAMPLE
0 1001 = 1.25 × (1/4.5) × fSAMPLE
0 1010 = 1.13 × (1/4.5) × fSAMPLE
0 1011 = 1.0 × (1/4.5) × fSAMPLE
0 1100 = 0.9 × (1/4.5) × fSAMPLE
0 1101 = 0.8 × (1/4.5) × fSAMPLE
0 1110 = 0.75 × (1/4.5) × fSAMPLE
0 1111 = reserved
1 0000 = 1.45 × (1/6) × fSAMPLE
1 0001 = 1.25 × (1/6) × fSAMPLE
1 0010 = 1.13 × (1/6) × fSAMPLE
1 0011 = 1.0 × (1/6) × fSAMPLE
1 0100 = 0.9 × (1/6) × fSAMPLE
1 0101 = 0.8 × (1/6) × fSAMPLE
1 0110 = 0.75 × (1/6) × fSAMPLE
1 0111 = reserved
0x010 FLEX_OFFSET
0x011 FLEX_GAIN
X
X
1
0
0
0
0
0
0x20
0x06
Reserved.
Digital VGA gain control
0000 = GAIN pins enabled (default)
PGA gain
00 = 21 dB
01 = 24 dB (default)
10 = 27 dB
LNA gain
00 = 15.6 dB
01 = 17.9 dB
10 = 21.6 dB
(default)
LNA and PGA gain
adjustment (global).
0001 = 0.0 dB (maximum gain, GAIN pins disabled)
0010 = −3.5 dB
0011 = −7.0 dB
…
11 = 30 dB
1110 = −45.5 dB
1111 = −45.5 dB
0x012 BIAS_CURRENT
X
X
X
X
0
1
PGA bias
0 =100%
(default)
1 = 60%
LNA bias
00 = high
01 = midhigh (default)
10 = midlow
0x09
LNA bias current
adjustment (global).
11 = low
0x013 RESERVED_13
0x014 OUTPUT_MODE
0
0
0
0
0
0
0
0x00
0x01
Reserved.
X
X
X
Output
data
X
Output
data invert
Output data format
00 = offset binary
Data output
modes (local).
enable
0 = enable
(default)
1 =
0 = disable 01 = twos complement
(default)
(default)
10 = gray code
11 = reserved
1 = enable
disable
Rev. A | Page 44 of 52
Data Sheet
AD9670
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x015 OUTPUT_ADJUST
LVDS
1
1
0
LVDS drive
strength
enable
0 = disable
(default)
1 = enable
LVDS drive current
000 = 3.72 mA
001 = 3.5 mA (default)
010 = 3.30 mA
0x61
Data output levels
(global).
output
standard
0 = ANSI
(default)
1 = IEEE
(low
011 = 2.96 mA
100 = 2.82 mA
101 = 2.57 mA
power)
110 = 2.27 mA
111 = 2.0 mA (reduced range)
0x016 FLEX_OUTPUT_
PHASE
X
X
0
DCO
invert
0 =
disable
(default)
1 =
X
X
DCO
0x00
DCO inversion and
course phase
adjustment
phase
adjust
with
respect to
DOUT
(global).
enable
00 = +90°
(default)
01 = 0°
10 = 0°
11 = −90°
0x017 FLEX_OUTPUT_
DELAY
DCO delay
enable
0 = disable
(default)
X
X
DCO clock delay
0x00
DCO delay
(global).
00000: 100 ps (default)
00001 = 200 ps
00010 = 300 ps
…
1 = enable
11101 = 3.0 ns
11110 = 3.1 ns
11111 = 3.2 ns
0x018 FLEX_VREF
X
X
X
X
X
1
0
0
0x04
0x00
Reserved (global).
0x019 USER_PATT1_LSB
B7
B6
B5
B4
B3
B2
B1
B0
User Defined
Pattern 1, LSB
(global).
0x01A USER_PATT1_MSB
0x01B USER_PATT2_LSB
0x01C USER_PATT2_MSB
0x01D USER_PATT3_LSB
0x01E USER_PATT3_MSB
0x01F USER_PATT4_LSB
0x020 USER_PATT4_MSB
0x021 FLEX_SERIAL_CTRL
B15
B7
B14
B6
B13
B5
B12
B4
B11
B3
B10
B2
B9
B1
B9
B1
B9
B1
B9
B8
B0
B8
B0
B8
B0
B8
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x00
User Defined
Pattern 1, MSB
(global).
User Defined
Pattern 2, LSB
(global).
B15
B7
B14
B6
B13
B5
B12
B4
B11
B3
B10
B2
User Defined
Pattern 2, MSB
(global).
User Defined
Pattern 3, LSB
(global).
B15
B7
B14
B6
B13
B5
B12
B4
B11
B3
B10
B2
User Defined
Pattern 3, MSB
(global).
User Defined
Pattern 4, LSB
(global).
B15
0
B14
B13
B12
B11
B10
User Defined
Pattern 4, MSB
(global).
FCO
invert
0 = not
inverted
(default)
1 =
Lane mode
Lane low
rate
0 = normal modulator
(default)
1 = low
sample
frequency (default)
(<32 MHz) 1 = FCO
per sample
FCO rate
with de-
Output word length
00 = 12 bits (default)
01 = 14 bits
LVDS control
(global).
00 = 1 channel/lane
(8 lanes) (default)
01 = 2 channels/lane
(4 lanes)
10 = 4 channels/lane
(2 lanes)
enabled
0 = FCO
per I/Q
10 = 16 bits
11 = reserved
inverted
11 = 8 channels/lane
(1 lane)
(I and Q)
0x022 SERIAL_CH_STAT
X
X
X
X
X
X
X
Channel
power-
down
0x00
Used to power
down individual
channels (local).
1 = on
0 = off
(default)
Rev. A | Page 45 of 52
AD9670
Data Sheet
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x02B FLEX_FILTER
X
Enable
X
X
Bypass
analog
HPF
0 = off
(default)
1 = on
X
Analog high-pass filter 0x00
cutoff
00 = fLP/12.00 (default)
01 = fLP/9.00
Filter cutoff
automatic
low-pass
tuning
1 = on
(self
(global); (fLP = low-
pass filter cutoff
frequency).
10 = fLP/6.00
11 = fLP/3.00
clearing)
0x02C LNA_TERM
X
X
X
X
X
X
X
X
LO-x, LOSW-x
connection
00 = RFB1 (default)
0x00
0x00
LNA active
termination/input
impedance (global).
01 = (RFB1 || RFB2
)
10 = RFB2
11 = ∞
0x02D CW_ENABLE_
PHASE
X
CW
I/Q demodulator phase
0000 = 0° (default)
0001 = 22.5° (not valid for 4LO mode)
0010 = 45°
0011 = 67.5° (not valid for 4LO mode)
0100 = 90°
0101 = 112.5° (not valid for 4LO mode)
0110 = 135°
Phase of
demodulators
(local, chip).
Doppler
channel
enable
0 = off
(default)
0 = on
0111 = 157.5° (not valid for 4LO mode)
1000 = 180°
1001 = 202.5° (not valid for 4LO mode)
1010 = 225°
1011 = 247.5° (not valid for 4LO mode)
1100 = 270°
1101 = 292.5° (not valid for 4LO mode)
1110 = 315°
1111 = 337.5° (not valid for 4LO mode)
0x02E CW_LO_MODE
Partially
enables
LVDS
RESET
Synchro-
RESET
MLO and
RESET
buffer
LO mode
0x00
CW mode
functions (global).
with MLO nous RESET signal
clock
00X = 4LO, 3rd to 5th odd harmonic
rejection (default)
sampling
MLO
clock edge high
polarity
0 = active enable (in
during CW edge
010 = 8LO, 3rd to 5th odd harmonic
rejection
0: LVDS
link dis-
abled
0 = syn-
all modes
except CW
chronous 0 = falling (default)
(default)
011 = 8LO, 3rd to 13th odd harmonic
rejection
(default)
1 = active mode)
low
during CW 1 = asyn- 1 = rising
(default)
1: LVDS
link part-
ially
0 = power-
down
(default)
1 = enable
100 = 16LO, 3rd to 5th odd harmonic
rejection
chronous
101 = 16LO, 3rd to 13th odd harmonic
rejection
11X = reserved
enabled
during
CW. PLL,
FCO, and
DCO are
enabled,
while LVDS
data drivers
are dis-
abled
(switching
activity can
degrade
CW per-
formance)
0x02F CW_OUTPUT
CW output
dc bias
0
0
0
0
0
0
0
0x80
Global.
voltage
0 = bypass
1 = enable
(default)
0x102 RESERVED_102
0x103 RESERVED_103
0x104 RESERVED_104
0x105 RESERVED_105
0x106 RESERVED_106
0x107 RESERVED_107
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
1
0
0
X
0
0
1
0
0
X
0X00
0X00
0x3F
0x00
0x00
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Read
only
0x108 RESERVED_108
0
0
0
0
0
0
0
0
0x00
Reserved.
Rev. A | Page 46 of 52
Data Sheet
AD9670
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
X X
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x109 VGA_TEST
X
VGA/ anti-
aliasing
filter test
mode
enable
0 = off
X
VGA/ antialiasing filter output test mode 0x00
000 = Channel A (default)
001 = Channel B
VGA/ antialiasing
filter test mode
enables antialiasing
filter output to the
GPO2 and GPO3
pins (global).
010 = Channel C
011 = Channel D
100 = Channel E
(default)
1 = on
101 = Channel F
110 = Channel G
111 = Channel H
0x10C PROFILE_INDEX
X
X
Manual
TX_TRIG
0 = off,
use pin
(default)
1 = on,
Profile Index[4:0]
0x00
Index for profile
memory selects
active profile
(global).
auto-
generate
TX_TRIG
(self clears)
0x10D RESERVED_10D
0x10E RESERVED_10E
0x10F DIG_OFFSET_CAL
1
1
0
1
1
0
1
1
0
1
1
0
1
1
1
1
1
1
1
0xFF
0xFF
Reserved.
Reserved.
1
Digital
offset
Digital offset calibration
000 = disable correction, reset
correction value (default)
001 = average 210 samples
010 = average 211 samples
…
Controls digital
offset calibration
enable and the
number of samples
used (global).
calibration
status
0 = not
complete
(default)
1 =
111 = average 216 samples
complete
0x110 DIG_OFFSET_
CORR1
D7
D6
D5
D4
D3
D2
D1
D9
D0
D8
0x00
0x00
Offset correction
LSB (local, chip).
0x111 DIG_OFFSET_
CORR2
D15
D14
D13
D12
D11
D10
Offset correction
MSB (local, chip).
Digital offset calibration (read back if auto calibration is enabled with Register 0x10F; otherwise,
force correction value
Offset correction = [D15:D0] × AFULL SCALE/216
0111 1111 1111 1111 (215 − 1) = +1/2 × AFULL SCALE − 1/216 × AFULL SCALE
0111 1111 1111 1110 (215 − 2) = +1/2 × AFULL SCALE − 2/216 × AFULL SCALE
…
0000 0000 0000 0001 (+1) = 1/216 × AFULL SCALE
0000 0000 0000 0000 = no correction (default)
1111 1111 1111 1111 (−1) = −1/216 × AFULL SCALE
…
1000 0000 0000 0000 (−215) = −1/2 full scale
0x112 POWER_MASK_
CONFIG
X
X
X
X
X
Power up set-up time (POWER_SETUP)
0 0000 = 0
0x02
0x00
Power setup time
is used to set the
power-up time
(global).
0 0001 = 1 × 40/fSAMPLE
0 0010 = 2 × 40/fSAMPLE (default)
0 0011 = 3 × 40/fSAMPLE
…
1 1111 = 31 × 40/fSAMPLE
0x113 DIG_DEMOD_
CONFIG
Digital
high-pass
filter
0 = enable gain
(default)
Decimator
gain scale
0 = no
Decimator and filter
enable
Baseband Demod-
decimator ulator
Enable stages of
the digital
processing
(global).
00 = RF 2× decimator
bypassed (default)
01 = RF 2× decimator
enabled and low band
filter
0 = enable 0 = enable
(default)
1 = bypass 1 = bypass
(default)
(default)
1 = bypass 1 = 4×
gain (shift
decimator
1X = RF 2× decimator
output by enabled and high band
2) filter
0x115 CHIP_ADDR_EN
X
X
Chip
address
mode
Chip address qualifier
0 0000 (default)
(If read, returns the state of ADDR0 to ADDR4 pins)
0x00
Chip address
mode enables the
addressing of
0 = disable
(default)
1 = enable
devices if the value
of the chip address
qualifier equals the
state on the ADDR0
to ADDR4 pins
(global).
Rev. A | Page 47 of 52
AD9670
Data Sheet
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x116 ANALOG_TEST_
TONE
X
X
X
X
Analog test signal
amplitude
(see Table 17 to Table 19)
Analog test signal
frequency
00 = fSAMPLE/4 (default)
01 = fSAMPLE/8
0x00
Analog test tone
amplitude and
frequency (global).
10 = fSAMPLE/16
11 = fSAMPLE/32
0x117 DIG_SINE_TEST_
FREQ
X
X
X
X
X
X
Digital test tone frequency
0 0000 = 1 × fSAMPLE/64
0 0001 = 2 × fSAMPLE/64
…
0x00
0x00
Digital sine test
tone frequency
(global).
1 1111 = 32 × fSAMPLE/64
0x118 DIG_SINE_TEST_
AMP
X
Digital test tone amplitude
0000 = AFULL SCALE (default)
0001 = AFULL SCALE/2
0010 = AFULL SCALE/22
…
Digital sine test
tone amplitude
(global).
1111 = AFULL SCALE/215
0x119 DIG_SINE_TEST_
OFFSET
Offset multiplier (a)
0 1111 = +15
0 1110 =+14
…
Offset exponent (b)
000 = 0 (default)
001 = 1
0x00
Digital sine test
tone offset
(global).
…
0 0000 = 0 (default)
1 1111 = −1
…
111 = 7
1 0000 = −16
Offset = AFULL-SCALE × a × 2 − (13 − b)
Offset range is ~0.5 dB
Maximum positive offset = 15 × 2 − (13 − 7) = +0.25 × AFULL SCALE
Maximum negative offset = –16 × 2 − (13 − 7) ≈ –0.25 × AFULL SCALE
0x11A TEST_MODE_
CH_ENABLE
Channel H Channel G Channel F Channel E Channel D Channel C
Channel B Channel A 0x00
Enable channels
for test mode
(global).
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
enable
0 = off
(default)
1 = on
0x11B TEST_MODE_
CONFIG
X
X
X
X
X
Test mode selection
0x00
Enable digital test
modes (global).
000 = disable test modes (default)
001 = enable digital sine test mode
010 = enable decimator filter test
(output of decimator is the sequence
of filter coefficients)
011 = enable channel ID test mode
16-bit data = digital ramp (7 bits) +
I/Q bit + Channel ID (3 bits) +
Chip Address (5 bits)
100 = enable analog test tone
101 = reserved
110 = reserved
111 = reserved
0x11C RESERVED_11C
0x11D RESERVED_11D
0x11E RESERVED_11E
0x11F RESERVED_11F
0x120 CW_TEST_TONE
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0x00
0x00
0x00
0x00
0x00
Reserved.
Reserved.
Reserved.
Reserved.
0
0
0
CW I/Q
output
swap
LNA offset LNA offset cancellation
cancel-
lation
0 = enable
(default)
1 = disable
CW analog test tone
Sets the frequency
of the analog test
tone to fLO in CW
Doppler mode.
Enables I/Q output
swap. LNA offset
cancellation
transconductance
00 = 0.5 mS (default)
01 = 1.0 mS
override for Reg. 0x116,
Bits[1:0]
0 =
00 = disable override
(default)
01 = set analog test
tone frequency to fLO
1X = set analog test
tone frequency to dc
disable
(default)
1 =
10 = 1.5 mS
11 = 2.0 mS
enable
control (global).
0x180 RESERVED_180
0x181 RESERVED_181
0x182 PLL_STARTUP
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
1
1
0
0
0x87
0x00
0x02
Reserved.
Reserved.
0
0
PLL auto
PLL control
(global).
configure
0 = disable
(default)
1 = enable
0x183 RESERVED_183
0x184 RESERVED_184
0x186 RESERVED_186
0
0
1
0
0
0
0
0
1
0
0
0
0
1
0
1
1
0
1
1
0
0
0x07
0x00
0xAE
Reserved.
Reserved.
Reserved.
0
1
Rev. A | Page 48 of 52
Data Sheet
AD9670
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
0x187 RESERVED_187
0x188 START_CODE_EN
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0x20
Reserved.
Start code 0x01
identifier
Enables start code
identifier (global).
0 = disable
1 = enable
(default)
0x189 RESERVED_189
0x18A RESERVED_18A
0
0
0
0
0
0
0
0
0
0x00
0x00
0x27
Reserved.
Reserved.
0
0
0
0
0
0
0
0x18B START_CODE_MSB B15
B14
B13
B12
B11
B10
B9
B8
Start code MSB
(global).
0x18C START_CODE_LSB
B7
B6
B5
B4
B3
B2
B1
B0
0x72
Start code LSB
(global).
0x190 RESERVED_190
0x191 RESERVED_191
0x192 RESERVED_192
0x193 RESERVED_193
0x194 RESERVED_194
0x195 RESERVED_195
0x196 RESERVED_196
0x197 RESERVED_197
0x198 CLOCK_DOUBLING
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
1
0
1
0
0
0
0
1
0
1
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0x10
0x00
0x18
0x00
0x1C
0x00
0x18
0x00
0x00
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
DCO frequency doubling/divider
1011 = 1/32
1010 = 1/64
1001 = 1/128
1000 = 1/256
0000 = 1 (default)
0001 = 2
DCO frequency
control (global).
0010 = 4
0011 = 8
0100 = 16
0101 = 32
0110 = 64
0111 = 128
1000 = 1/256
1001 = 1/128
1010 = 1/64
1011 = 1/32
1100 = 1/16
1101 = 1/8
1110 = ¼
1111 = ½
0x199 SAMPLE_CLOCK_
COUNTER
Enable
0
0
0
0
0
0
0
0x00
Enables automatic
clocks per sample
calculation (global).
clocks per
sample
auto
calculation
0 = off
(default)
1 = on
0x19A DATA_OUTPUT_
INVERT
X
X
X
X
X
X
X
Invert data 0x00
output
0 = non-
inverted
(default)
1 =
Inverts DOUT
outputs (global).
inverted
0x19B SERIAL_FORMAT
X
Enable
FCO for
start code extra
sample
0 =
disable
1 =
enable
(default)
Enable
FCO for
Enable
FCO rotate
0x70
FCO controls
(global).
FCO con-
tinuously
0 = only
during
0000 = FCO aligned with DOUT
0001 = FCO 1 bit before DOUT
0010 = FCO 2 bits before DOUT
…
1101 = FCO 3 bits after DOUT
1110 = FCO 2 bits after DOUT
1111 = FCO 1 bit after DOUT
sample at
end of
burst
0 = disable 1 = con-
1 = enable tinuous
(default)
burst
(default)
0x19C RESERVED_19C
0x19D RESERVED_19D
0x19E RESERVED_19E
0x19F RESERVED_19F
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0x10
0x00
0x10
0x00
Reserved.
Reserved.
Reserved.
Reserved.
Rev. A | Page 49 of 52
AD9670
Data Sheet
Addr.
(Hex)
Default
Bit 0 (LSB) Value
Register Name
Bit 7 (MSB) Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Comments
Reserved.
Reserved.
0x1A0 RESERVED_1A0
0x1A1 RESERVED_1A1
Profile Memory Registers
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0x00
0x00
0x1000 Coefficient memory
32 × 64 bits
0x00
Global.
to
0x1FFF
Coefficient Memory Registers
0xF00 Profile memory
256 × 112 bits
0x00
Global.
to
0xFFF
Rev. A | Page 50 of 52
AD9670
Data Sheet
Profile Index and Manual TX_TRIG (Register 0x10C)
MEMORY MAP REGISTER DESCRIPTIONS
The vector profile is selected using the profile index in
Register 0x10C, Bits[4:0]. The manual TX_TRIG control in Bit 5
generates a TX_TRIG signal internal to the device. This signal
is asynchronous to the ADC sample clock. Therefore, it cannot
be used to align the data output, advanced power mode, or
NCO reset across multiple devices in the system. The external
pin-driven TX_TRIG control is recommended for systems that
require synchronization of these features across multiple
AD9670 devices.
For more information on the SPI memory map and other
functions, consult the AN-877 Application Note, Interfacing to
High Speed ADCs via SPI.
Transfer (Register 0x0FF)
All registers except Register 0x002 are updated the moment they
are written. Setting Bit 0 of Register 0x0FF high initializes and
updates the speed mode (Address 0x002) and resets all other
registers to their default values. Bit 0 is self clearing. It is
recommended that Register 0x002 and Regoster 0x0FF, Bit 0,
be set at the beginning of the setup SPI writes after the device
is powered up. This avoids rewriting other registers after
Register 0x0FF is set.
Rev. A | Page 51 of 52
AD9670
Data Sheet
OUTLINE DIMENSIONS
10.10
10.00 SQ
9.90
A1 BALL
CORNER
A1 BALL
CORNER
12 11 10
9
8
7
6
5
4
3
2
1
A
B
C
D
E
F
8.80
BSC SQ
G
H
J
0.80
K
L
M
0.60
REF
TOP VIEW
DETAIL A
BOTTOM VIEW
*
1.40 MAX
0.65 MIN
DETAIL A
0.25 MIN
0.50
0.45
0.40
COPLANARITY
0.20
SEATING
PLANE
BALL DIAMETER
*
COMPLIANT WITH JEDEC STANDARDS MO-275-EEAB-1
WITH THE EXCEPTION OF PACKAGE HEIGHT.
Figure 55. 144-Ball Chip Scale Package, Ball Grid Array [CSP_BGA]
(BC-144-1)
Dimensions shown in millimeters
ORDERING GUIDE
Model1
Temperature Range
Package Description
Package Option
AD9670BBCZ
AD9670EBZ
0°C to +85°C
144-Ball Chip Scale Package, Ball Grid Array [CSP_BGA]
Evaluation Board
BC-144-1
1 Z = RoHS Compliant Part.
©2016 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D11041-0-2/16(A)
Rev. A | Page 52 of 52
相关型号:
©2020 ICPDF网 联系我们和版权申明