AD AD9670 Octal ultrasound afe with digital demodulator Datasheet

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
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
10 mm × 10 mm, 144-ball CSP_BGA
TGC channel, input referred noise voltage: 0.82 nV/√Hz,
maximum gain
Flexible power-down modes
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
I/Q demodulator with programmable oscillator FIR
decimation filter
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.
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.
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.
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.
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 registered trademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
©2016 Analog Devices, Inc. All rights reserved.
Technical Support
www.analog.com
AD9670* PRODUCT PAGE QUICK LINKS
Last Content Update: 08/08/2017
COMPARABLE PARTS
DESIGN RESOURCES
View a parametric search of comparable parts.
• AD9670 Material Declaration
• PCN-PDN Information
EVALUATION KITS
• Quality And Reliability
• AD9670 Evaluation Board
• Symbols and Footprints
DOCUMENTATION
DISCUSSIONS
Data Sheet
View all AD9670 EngineerZone Discussions.
• AD9670: Octal Ultrasound AFE with Digital Demodulator
Data Sheet
SAMPLE AND BUY
REFERENCE MATERIALS
Visit the product page to see pricing options.
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
Applications ....................................................................................... 1
Digital Demodulator/Decimator .................................................. 35
General Description ......................................................................... 1
Vector Profile .............................................................................. 35
Revision History ............................................................................... 2
RF Decimator .............................................................................. 36
Functional Block Diagram .............................................................. 3
Baseband Demodulator and Decimator.................................. 37
Specifications..................................................................................... 4
Digital Test Waveforms.............................................................. 38
AC Specifications.......................................................................... 4
Digital Block Power Saving Scheme ........................................ 38
Digital Specifications ................................................................... 7
Serial Port Interface (SPI) .............................................................. 39
Switching Specifications .............................................................. 8
Hardware Interface..................................................................... 39
Timing Diagrams.......................................................................... 9
Memory Map .................................................................................. 41
Absolute Maximum Ratings.......................................................... 11
Reading the Memory Map Table .............................................. 41
Thermal Impedance ................................................................... 11
Reserved Locations .................................................................... 41
ESD Caution ................................................................................ 11
Default Values ............................................................................. 41
Pin Configuration and Function Descriptions ........................... 12
Logic Levels ................................................................................. 41
Typical Performance Characteristics ........................................... 15
Recommended Startup Sequence ............................................ 41
TGC Mode Characteristics ....................................................... 15
Memory Map Register Descriptions ........................................ 51
CW Doppler Mode Characteristics ......................................... 19
Outline Dimensions ....................................................................... 52
Theory of Operation ...................................................................... 20
Ordering Guide .......................................................................... 52
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
LO-A TO LO-H
DVDD
PDWN STBY
DRVDD
CWQ+
CWQ–
CWI+
CWI–
CWD AND I/Q
DEMODULATOR
LOSW-A TO LOSW-H
LI-A TO LI-H
LNA
LG-A TO LG-H
VGA
14-BIT
ADC
AAF
DEMODULATOR/
DECIMATOR
SERIALIZER
LVDS
DOUTA+ TO DOUTH+
DOUTA– TO DOUTH–
AD9670
8 CHANNELS
Figure 1.
Rev. A | Page 3 of 52
FCO+
FCO–
DCO+
DCO–
11041-001
CLK–
DATA
RATE
MULTIPLIER
CLK+
SDIO
CSB
SCLK
GPO0 TO GPO3
SERIAL
PORT
INTERFACE
ADDR0 TO ADDR4
TX_TRIG+
NCO
TX_TRIG–
VREF
RBIAS
REFERENCE
GAIN+
GAIN–
MLO–
MLO+
RESET–
RESET+
LO
GENERATION
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 demodulator 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
0.1 dB Input Compression Point
1 dB Input Compression Point
Input Common Mode (LI-x, LG-x)
Output Common Mode
LO-x
LOSW-x
Input Resistance (LI-x)
Input Capacitance (LI-x)
Input Noise Voltage
Input Noise Current
FULL CHANNEL (TGC) CHARACTERISTICS
Antialiasing Filter Low-Pass Cutoff
In Range Antialiasing Filter
Bandwidth Tolerance
Group Delay Variation
Input Referred Noise Voltage
Noise Figure
Active Termination Matched
Unterminated
Correlated Noise Ratio
Output Offset
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
V p-p
V p-p
V p-p
V p-p
V p-p
V p-p
V
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
20
Ω
V
Ω
V
Ω
Ω
kΩ
pF
0.83
0.82
0.78
2.6
nV/√Hz
nV/√Hz
nV/√Hz
pA/√Hz
RS = 0 Ω
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
−3 dB, programmable, low band mode
−3 dB, programmable, high band mode
8
13.5
f = 1 MHz to 18 MHz, VGAIN = −1.6 V to +1.6 V
LNA gain = 15.6 dB
LNA gain = 17.9 dB
LNA gain = 21.6 dB
RS = 50 Ω
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
No signal, correlated/uncorrelated
−100
Rev. A | Page 4 of 52
18
30
±10
MHz
MHz
%
±350
0.96
0.90
0.82
ps
nV/√Hz
nV/√Hz
nV/√Hz
5.6
4.8
3.8
3.2
2.9
2.6
−30
dB
dB
dB
dB
dB
dB
dB
LSB
+100
Data Sheet
Parameter1
Signal-to-Noise Ratio (SNR)
Close In SNR
Second Harmonic
Third Harmonic
Two-Tone Intermodulation
Distortion (IMD3)
Channel-to-Channel Crosstalk
GAIN ACCURACY
Gain Law Conformance Error
Linear Gain Error
Channel-to-Channel Matching
PGA Gain
GAIN CONTROL INTERFACE
Control Range
Control Common Mode
Input Impedance
Gain Range
Scale Factor
Response Time
CW DOPPLER MODE
LO Frequency
Phase Resolution
Output DC Bias (Single-Ended)
Output AC Current Range
Transconductance (Differential)
Input Referred Noise Voltage
Noise Figure
Dynamic Range
Close In SNR
AD9670
Test Conditions/Comments
fIN = 5 MHz at −12 dBFS, VGAIN = −1.6 V
fIN = 5 MHz at −1 dBFS
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
fRF1 = 5.015 MHz, fRF2 = 5.020 MHz, ARF1 = −1 dBFS,
ARF2 = −21 dBFS, VGAIN = 1.6 V, IMD3 relative to ARF2
fIN1 = 5.0 MHz at −1 dBFS
Overrange condition2
TA = 25°C
−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
Min
Differential
GAIN+, GAIN−
GAIN+, GAIN−
−1.6
0.7
Rev. A | Page 5 of 52
Max
−60
−55
+1.3
−0.5
−1.3
+1.3
0.1
21/24/27/30
0.8
10
45
14
3.5
750
1
45
22.5
AVDD2/2
±2.2
Unit
dBFS
dBFS
dBc/√Hz
dBc
dBc
dBc
dBc
dBc
dB
dB
0.4
−1.3
Analog
Digital step size
Analog 45 dB change
fLO = fMLO/M
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)
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 = ∞
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
−3 dBFS input, fRF = 2.5 MHz, fLO = 40 MHz, 1 kHz
offset, 16LO mode, 8 channels enabled
Typ
69
59
−130
−70
−62
−61
−55
−54
dB
dB
dB
dB
dB
dB
+1.6
0.9
V
V
MΩ
dB
dB/V
dB
ns
10
MHz
Degrees
Degrees
V
mA
±2.5
3.3
4.3
6.6
mA/V
mA/V
mA/V
1.6
1.3
1.0
nV/√Hz
nV/√Hz
nV/√Hz
5.7
4.5
3.4
dB
dB
dB
164
162
160
156
dBFS/√Hz
dBFS/√Hz
dBFS/√Hz
dBc/√Hz
161
dBc/√Hz
AD9670
Parameter1
Two-Tone Intermodulation
Distortion (IMD3)
LO Harmonic Rejection
Quadrature Phase Error
I/Q Amplitude Imbalance
Channel-to-Channel Matching
POWER SUPPLY, MODE I/MODE II/
MODE III/MODE IV
AVDD1
AVDD2
DVDD
DRVDD
IAVDD1
IAVDD2
IDVDD
IDRVDD
Total Power Dissipation
(Including Output Drivers)
Power-Down Dissipation
Standby Power Dissipation
ADC RESOLUTION
ADC REFERENCE
Output Voltage Error
Load Regulation at 1.0 mA
Input Resistance
Data Sheet
Test Conditions/Comments
fRF1 = 5.015 MHz, fRF2 = 5.020 MHz, fLO = 80 MHz,
ARF1 = −1 dBFS, ARF2 = −21 dBFS, IMD3 relative to ARF2
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
Demodulator/decimator enabled
Demodulator/decimator disabled
TGC mode, LO band mode
CW Doppler mode
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
ANSI-644 mode
Low power (IEEE 1596.3 similar) mode, 1 channel per
lane mode
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
CW Doppler mode, 8 channels enabled
Min
Typ
−58
Max
Unit
dB
−20
dBc
Degrees
dB
Degrees
dB
1.9
3.6
1.9
1.9
1.9
V
V
V
V
V
mA
0.15
0.015
0.5
0.25
1.7
2.85
1.3
1.3
1.7
1.8
3.0
1.4
1.8
1.8
148/187/
223/291
4
230
239
140
156/247/
166/255
133/184/
141/146
119/170/
127/169
1200/1400/
1380/1630
1400/1695/
1570/1900
500
mA
mA
mA
mA
mA
mA
mA
1345/1555/
1535/2100
1560/1880/
1740/2100
30
630
14
VREF = 1 V
VREF = 1 V
±50
2
7.5
1
mW
mW
mW
mW
mW
Bits
mV
mV
kΩ
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.
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.
2
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
INPUTS
CLK+, CLK−, TX_TRIG+, TX_TRIG−
Logic Compliance
Differential Input Voltage42
Input Voltage Range
Input Common-Mode Voltage
Input Resistance (Differential)
Input Capacitance
MLO+, MLO−, RESET+, RESET−
Logic Compliance
Differential Input Voltage2
Input Voltage Range
Input Common-Mode Voltage
Input Resistance (Single-Ended)
Input Capacitance
LOGIC INPUTS
PDWN, STBY, SCLK, SDIO, ADDRx
Logic 1 Voltage
Logic 0 Voltage
Input Resistance3
Input Capacitance3
CSB
Logic 1 Voltage
Logic 0 Voltage
Input Resistance
Input Capacitance
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
Logic Compliance
Differential Output Voltage (VOD)
Output Offset Voltage (VOS)
Output Coding (Default)
Low Power, Reduced Signal Option
Logic Compliance
Differential Output Voltage (VOD)
Output Offset Voltage (VOS)
Output Coding (Default)
Temperature
Min
Typ
Max
Unit
3.6
AVDD1 + 0.2
V p-p
V
V
kΩ
pF
2 × AVDD2
AVDD2 + 0.2
+0.3
V p-p
V
V
kΩ
pF
DRVDD + 0.3
0.3
V
V
kΩ
pF
DRVDD + 0.3
0.3
V
V
kΩ
pF
CMOS/LVDS/LVPECL
0.2
GND − 0.2
0.9
15
4
25°C
25°C
LVDS/LVPECL
0.250
GND − 0.2
−0.3
25°C
25°C
AVDD2/2
20
1.5
1.2
25°C
25°C
30 (26 for SDIO)
2 (5 for SDIO)
1.2
25°C
25°C
26
2
1.79
0.05
V
V
0.05
V
454
1.375
mV
V
250
1.30
mV
V
LVDS
247
1.125
Offset binary
LVDS
150
1.10
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
Clock Rate
40 MSPS (Mode I)
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%)
Fall Time (tF) (20% to 80%)
DCO Period (tDCO)6
FCO Propagation Delay (tFCO)
DCO Propagation Delay (tCPD)7
DCO to Data Delay (tDATA)7
DCO to FCO Delay (tFRAME)7
Data-to-Data Skew (tDATA-MAX − tDATA-MIN)
TX_TRIG to CLK Setup Time (tSETUP)
TX_TRIG to CLK Hold Time (tHOLD)
Wake-Up Time
Standby
Power-Down
ADC Pipeline Latency
APERTURE
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)
Temperature
Min
Full
Full
Full
Full
Full
Full
20.5
20.5
20.5
20.5
Full
Full
Full
Full
Full
Full
Full
Full
Full
25°C
25°C
10.8 − 1.5 × tDCO
Typ
Max
Unit
40
65
80
125
MHz
MHz
MHz
MHz
ns
ns
10.8 + 1.5 × tDCO
ns
ps
ps
ns
ns
ns
ps
ps
ps
ns
ns
3.75
3.75
10.8 − 1.5 × tDCO
(tSAMPLE/28) − 300
(tSAMPLE/28) − 300
10.8
300
300
tSAMPLE/7
10.8
tFCO + (tSAMPLE/28)
(tSAMPLE/28)
(tSAMPLE/28)
±225
10.8 + 1.5 × tDCO
(tSAMPLE/28) + 300
(tSAMPLE/28) + 300
±400
1
1
25°C
25°C
Full
2
375
16
μs
μs
Clock cycles
25°C
<1
ps rms
Full
Full
Full
Full
Full
4
8
16
1
1
1
40
80
160
tMLO10/2
tMLO10/2
MHz
MHz
MHz
ns
ns
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–
tEH
CLK–
tEL
CLK+
tCPD
DCO–
DCO+
tFRAME
tFCO
FCO–
FCO+
tPD
MSB
D12
D11
D10
D9
D8
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
DOUTx+
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
11041-003
RESET–
tSETUP
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
11041-002
tDATA
DOUTx–
AD9670
Data Sheet
tMLO
MLO–
MLO+
tSETUP
tHOLD
11041-004
RESET–
RESET+
Figure 4. CW Doppler Mode Input MLO±, Continuous Synchronous RESET± Timing, Sampled on the Falling MLO± Edge, 8LO Mode
tMLO
MLO–
MLO+
11041-105
RESET–
tHOLD
tSETUP
RESET+
Figure 5. CW Doppler Mode Input MLO±, Pulse Synchronous RESET± Timing, 4LO/8LO/16LO Mode
tMLO
MLO–
MLO+
tSETUP
11041-106
RESET–
tHOLD
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
AVDD1 to GND
AVDD2 to GND
DVDD to GND
DRVDD to GND
GND to GND
AVDD2 to AVDD1
AVDD1 to DRVDD
AVDD2 to DRVDD
Digital Outputs (DOUTx+, DOUTx−, DCO+,
DCO−, FCO+, FCO−) to GND
LI-x, LG-x, LO-x, LOSW-x, CWI−, CWI+, CWQ−,
CWQ+, GAIN+, GAIN−, RESET+, RESET−,
MLO+, MLO−, GPO0, GPO1, GPO2, GPO3
to GND
CLK+, CLK−, TX_TRIG+, TX_TRIG−, VREF to GND
SDIO, PDWN, STBY, SCLK, CSB, ADDRx
Operating Temperature Range (Ambient)
Storage Temperature Range (Ambient)
Maximum Junction Temperature
Lead Temperature (Soldering, 10 sec)
Rating
−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
−0.3 V to
DRVDD + 0.3 V
−0.3 V to
AVDD2 + 0.3 V
−0.3 V to
AVDD1 + 0.3 V
−0.3 V to
DRVDD + 0.3 V
0°C to 85°C
−65°C to +150°C
150°C
300°C
Table 5. Thermal Impedance
Symbol
θJA
ΨJB
ΨJT
1
Description
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)
Junction-to-top-of-package
characterization parameter, 0 m/sec
air flow per JEDEC JESD51-2 (still air)
Value1
22.0
Unit
°C/W
9.2
°C/W
0.12
°C/W
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.
ESD CAUTION
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
LI-E
LI-F
LI-G
LI-H
VREF
RBIAS
GAIN+
GAIN–
LI-A
LI-B
LI-C
LI-D
B
LG-E
LG-F
LG-G
LG-H
GND
GND
CLNA
GND
LG-A
LG-B
LG-C
LG-D
C
LO-E
LO-F
LO-G
LO-H
GND
GND
GND
GND
LO-A
LO-B
LO-C
LO-D
GND
GND
GND
GND
LOSW-A LOSW-B LOSW-C LOSW-D
E
GND
AVDD2
AVDD2
AVDD2
GND
GND
GND
GND
AVDD2
AVDD2
AVDD2
GND
F
AVDD1
GND
AVDD1
GND
AVDD1
GND
GND
AVDD1
GND
AVDD1
GND
AVDD1
G
GND
AVDD1
GND
DVDD
GND
GND
GND
GND
AVDD1
GND
DVDD
GND
H
CLK–
TX_TRIG–
GND
GND
GND
GND
ADDR4
ADDR3
ADDR2
ADDR1
ADDR0
CSB
J
CLK+
TX_TRIG+
CWQ+
GND
CWI+
AVDD2
MLO+
RESET–
GPO3
GPO1
PDWN
SDIO
K
GND
GND
CWQ–
GND
CWI–
AVDD2
MLO–
RESET+
GPO2
GPO0
STBY
SCLK
DCO+
FCO+
DOUTD+ DOUTC+ DOUTB+ DOUTA+ DRVDD
DCO–
FCO–
DOUTD– DOUTC– DOUTB– DOUTA–
M
DRVDD DOUTH+ DOUTG+ DOUTF+ DOUTE+
GND
DOUTH– DOUTG– DOUTF– DOUTE–
Figure 7. Pin Configuration
1
2
4
3
6
5
7
10
8
9
12
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
11041-006
L
GND
11041-005
D LOSW-E LOSW-F LOSW-G LOSW-H
Data Sheet
AD9670
Table 6. Pin Function Descriptions
Pin No.
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
F1, F3, F5, F8, F10, F12, G2, G9,
G4, G11
E2 to E4, E9 to E11, J6, K6
B7
L1, L12
C1
D1
A1
B1
C2
D2
A2
B2
C3
D3
A3
B3
C4
D4
A4
B4
H1
J1
H2
J2
H11
H10
H9
H8
H7
M2
L2
M3
L3
M4
L4
M5
L5
M6
L6
M7
L7
M8
L8
M9
L9
M10
L10
Mnemonic
GND
Description
Ground. Tie these pins to a quiet analog ground.
AVDD1
DVDD
AVDD2
CLNA
DRVDD
LO-E
LOSW-E
LI-E
LG-E
LO-F
LOSW-F
LI-F
LG-F
LO-G
LOSW-G
LI-G
LG-G
LO-H
LOSW-H
LI-H
LG-H
CLK−
CLK+
TX_TRIG−
TX_TRIG+
ADDR0
ADDR1
ADDR2
ADDR3
ADDR4
DOUTH−
DOUTH+
DOUTG−
DOUTG+
DOUTF−
DOUTF+
DOUTE−
DOUTE+
DCO−
DCO+
FCO−
FCO+
DOUTD−
DOUTD+
DOUTC−
DOUTC+
DOUTB−
DOUTB+
1.8 V Analog Supply.
1.4 V/1.8 V Digital Supply.
3.0 V Analog Supply.
LNA External Capacitor.
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.
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.
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.
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.
Clock Input Complement.
Clock Input True.
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.
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.
ADC B Digital Output Complement.
ADC B Digital Output True.
Rev. A | Page 13 of 52
AD9670
Pin No.
M11
L11
K11
J11
K12
J12
H12
B9
A9
D9
C9
B10
A10
D10
C10
B11
A11
D11
C11
B12
A12
D12
C12
K10
J10
K9
J9
J8
K8
K7
J7
A8
A7
A6
A5
K5
J5
K3
J3
Data Sheet
Mnemonic
DOUTA−
DOUTA+
STBY
PDWN
SCLK
SDIO
CSB
LG-A
LI-A
LOSW-A
LO-A
LG-B
LI-B
LOSW-B
LO-B
LG-C
LI-C
LOSW-C
LO-C
LG-D
LI-D
LOSW-D
LO-D
GPO0
GPO1
GPO2
GPO3
RESET−
RESET+
MLO−
MLO+
GAIN−
GAIN+
RBIAS
VREF
CWI−
CWI+
CWQ−
CWQ+
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.
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.
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.
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
PERCENTAGE OF UNITS (%)
1.5
GAIN ERROR (dB)
1.0
0°C
0.5
0
25°C
–0.5
85°C
–1.0
20
15
10
5
–1.5
–0.4
0
0.4
0.8
1.2
1.6
VGAIN (V)
0
GAIN ERROR (dB)
Figure 9. Gain Error vs. VGAIN
Figure 12. Gain Error Histogram, VGAIN = 1.28 V
25
20
20
PERCENTAGE OF UNITS (%)
15
10
5
0
15
10
5
GAIN ERROR (dB)
–1.0
–0.9
–0.8
–0.7
–0.6
–0.5
–0.4
–0.3
–0.2
–0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
11041-108
–1.0
–0.9
–0.8
–0.7
–0.6
–0.5
–0.4
–0.3
–0.2
–0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
CHANNEL-TO-CHANNEL GAIN MATCHING (dB)
Figure 10. Gain Error Histogram, VGAIN = −1.28 V
11041-111
PERCENTAGE OF UNITS (%)
11041-110
–0.8
–1.0
–0.9
–0.8
–0.7
–0.6
–0.5
–0.4
–0.3
–0.2
–0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
–1.2
11041-107
–2.0
–1.6
Figure 13. Gain Matching Histogram, VGAIN = −1.2 V
35
20
PERCENTAGE OF UNITS (%)
25
20
15
10
15
10
5
5
0
Figure 11. Gain Error Histogram, VGAIN = 0 V
–1.0
–0.9
–0.8
–0.7
–0.6
–0.5
–0.4
–0.3
–0.2
–0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
CHANNEL-TO-CHANNEL GAIN MATCHING (dB)
Figure 14. Gain Matching Histogram, VGAIN = 1.2 V
Rev. A | Page 15 of 52
11041-112
GAIN ERROR (dB)
11041-109
0
–1.0
–0.9
–0.8
–0.7
–0.6
–0.5
–0.4
–0.3
–0.2
–0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
PERCENTAGE OF UNITS (%)
30
AD9670
Data Sheet
1.4
70
LNA GAIN = 17.9dB
66
64
1.0
SNR (dBFS)
INPUT REFERRED NOISE (nV/√Hz)
68
1.2
0.8
LNA GAIN = 15.6dB
62
60
LNA GAIN = 21.3dB
58
56
54
0.6
52
2
3
4
5
6
7
8
9
10
FREQUENCY (MHz)
11041-008
1
50
10
25
30
35
40
45
50
55
Figure 18. SNR vs. Channel Gain and LNA Gain, AOUT = −1.0 dBFS
74
PGA GAIN = 21dB
PGA GAIN = 21dB
72
–134
70
–136
PGA GAIN = 24dB
68
SNR (dBFS)
OUTPUT REFERRED NOISE (dBc/√Hz)
20
CHANNEL GAIN (dB)
Figure 15. Short-Circuit, Input Referred Noise vs. Frequency
–132
15
11041-011
PGA GAIN = 27dB
0.4
–138
–140
66
64
PGA GAIN = 27dB
62
60
–142
PGA GAIN = 30dB
58
–144
56
5
10
15
20
25
30
35
40
45
CHANNEL GAIN (dB)
54
–5
11041-009
0
10
15
20
25
0
PGA GAIN = 21dB
68
66
40
45
50
55
–2
AMPLITUDE (dBFS)
64
62
PGA GAIN = 27dB
58
PGA GAIN = 30dB
56
35
SPEED MODE = I (40MSPS)
LO BAND MODE
–1
PGA GAIN = 24dB
60
30
Figure 19. SNR vs. Channel Gain and PGA Gain, AIN = −45 dBm
70
54
–3
–4
–5
–6
–7
–8
52
15
20
25
30
35
40
45
50
55
CHANNEL GAIN (dB)
Figure 17. SNR vs. Channel Gain and PGA Gain, AOUT = −1.0 dBFS
–10
0
5
10
15
INPUT FREQUENCY (MHz)
Figure 20. Antialiasing Filter Pass-Band Response,
LPF Cutoff = 1/3 × fSAMPLE, HPF = 1/12 × LPF Cutoff
Rev. A | Page 16 of 52
20
11041-013
–9
LNA GAIN = 21.3dB
11041-010
SNR (dBFS)
5
CHANNEL GAIN (dB)
Figure 16. Short-Circuit, Output Referred Noise vs. Channel Gain,
PGA Gain = 21 dB, VGAIN = 1.6 V
50
10
0
11041-117
LNA GAIN = 21.3dB
–146
–5
–20
–30
–40
THIRD ORDER, MIN VGAIN
–50
THIRD ORDER, MAX VGAIN
–60
–70
SECOND ORDER, MIN VGAIN
–80
–90
–100
SECOND ORDER, MAX VGAIN
2
3
4
5
6
7
8
9
10
11
INPUT FREQUENCY (MHz)
11041-014
PGA GAIN = 24dB
–10
–20
–30
–40
–50
–60
LNA GAIN = 17.9dB
–70
LNA GAIN = 21.6dB
–80
LNA GAIN = 15.6dB
–90
–100
10
15
20
25
30
35
40
45
50
CHANNEL GAIN (dB)
–50
VGAIN = –1.2V
–60
VGAIN = 0V
–70
–80
–90
VGAIN = +1.6V
–100
–110
–120
–40
–35
–30
–25
–20
–15
–10
0
–5
0
–10
–20
–30
–40
VGAIN = –1.2V
–50
–60
VGAIN = 0V
–70
–80
–90
VGAIN = +1.6V
–100
–110
–120
–40
–35
–30
–25
–20
–15
–10
–5
0
ADC OUTPUT LEVEL (dBFS)
Figure 25. Third-Order Harmonic Distortion vs. ADC Output Level (AOUT)
–100
PGA GAIN = 24dB
–10
–110
PHASE NOISE (dBc/√Hz)
–20
–30
–40
LNA GAIN = 17.9dB
–50
LNA GAIN = 21.6dB
–60
LNA GAIN = 15.6dB
–70
–80
–120
–130
–140
–150
–90
–100
10
15
20
25
30
35
40
45
CHANNEL GAIN (dB)
11041-016
THIRD-ORDER HARMONIC DISTORTION (dBFS)
Figure 22. Second-Order Harmonic Distortion vs. Channel Gain,
AOUT = −1.0 dBFS
0
–40
Figure 24. Second-Order Harmonic Distortion vs. ADC Output Level (AOUT)
THIRD-ORDER HARMONIC DISTORTION (dBFS)
0
–30
ADC OUTPUT LEVEL (dBFS)
11041-015
SECOND-ORDER HARMONIC DISTORTION (dBFS)
Figure 21. Second-Order and Third-Order Harmonic Distortion vs.
Input Frequency, AOUT = −1.0 dBFS
–20
–160
100
1k
10k
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
100k
11041-017
HARMONIC DISTORTION (dBFS)
–10
0
–10
11041-122
LNA GAIN = 21.6dB
PGA GAIN = 27dB
MIN VGAIN, AOUT = –12.0dBFS
MAX VGAIN, AOUT = –1.0dBFS
11041-123
0
AD9670
SECOND-ORDER HARMONIC DISTORTION (dBFS)
Data Sheet
Data Sheet
0
8
7
6
5
4
3
2
1
0
100k
–10
–20
1M
10M
FREQUENCY (Hz)
IMD3 (dBFS)
–40
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
100k
100M
–50
–60
–80
–90
–100
–20
10M
FREQUENCY (Hz)
100M
–120
–40
–35
–30
–25
–20
–15
–10
–5
0
AMPLITUDE LEVEL (dBFS)
Figure 29. IMD3 vs. ADC Output Amplitude Level
7
fIN1 = 2.3MHz
fIN2 = 2.31MHz
ARF1 LEVEL = –1dBFS
ARF2 LEVEL = –21dBFS
RS = 50Ω
6
NOISE FIGURE (dB)
–40
–50
–60
–70
RIN = 300Ω, 1000Ω
–80
5
4
3
2
–90
RIN = 50Ω
20
25
30
35
40
CHANNEL GAIN (dB)
45
50
11041-019
–100
15
VGAIN = 0V
11041-127
1M
–30
IMD3 (dBFS)
VGAIN = +1.6V
–110
Figure 28. IMD3 vs. Channel Gain
1
0
2
4
6
8
10
12
14
16
18
20
FREQUENCY (MHz)
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
11041-020
0
VGAIN = –1.2V
–70
Figure 27. LNA Input Impedance Magnitude and Phase, Unterminated
–10
fIN1 = 5.0MHZ
fIN2 = 5.01MHZ
ARF1 LEVEL = –1dBFS
ARF2 LEVEL = –21dBFS
–30
11041-018
PHASE (Degrees)
MAGNITUDE (kΩ)
AD9670
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
9
160
155
SNR (dBc/√Hz)
7
6
5
4
3
2
145
0
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
BASEBAND FREQUENCY (Hz)
Figure 31. Noise Figure vs. Baseband Frequency
130
0
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
BASEBAND FREQUENCY (Hz)
Figure 32. SNR vs. Baseband Frequency, −3 dBFS LNA Input
Rev. A | Page 19 of 52
11041-022
135
1
0
150
140
11041-021
NOISE FIGURE (dB)
8
AD9670
Data Sheet
THEORY OF OPERATION
MLO–
MLO+
RESET+
RESET–
LO-x
RFB2
LOSW-x
S
CWQ+
CWQ–
LI-x
LG-x
CSH
ATTENUATOR
–45dB TO 0dB
LNA
15.6dB,
17.9dB,
21.6dB
CLG
TRANSDUCER
CWI+
CWI–
gm
GAIN
INTERPOLATOR
GAIN+
POST
AMP
FILTER
PIPELINE
ADC
21dB,
24dB,
27dB,
30dB
DEMOD/
DEC
SERIAL
LVDS
DOUTx+
DOUTx–
NCO
GAIN–
TX_TRIG+ TX_TRIG–
11041-023
T/R
SWITCH C
RFB1
LO
GENERATION
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.
TGC OPERATION
The system gain for TGC operation is distributed as shown in
Table 7.
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.
In addition to the analog VGA attenuation described in Equation 2,
the attenuation level can be digitally controlled in 3.5 dB increments. 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
LNA
Attenuator
VGA
Filter
ADC
Then, calculate the total channel gain using Equation 3.
Nominal Gain (dB)
15.6/17.9/21.6 (LNAGAIN)
−45 to 0 (VGAATT)
21/24/27/30 (PGAGAIN)
0
0
Low Noise Amplifier (LNA)
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 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.
VGAIN (V) = (GAIN+) − (GAIN−)
(1)
VGAATT (dB) = −14 dB/V (1.6) − VGAIN
(2)
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.
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 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.
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
Rev. A | Page 20 of 52
Data Sheet
AD9670
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.
RIN 
(RFB1  20 ) || (RFB2  20 )  30 
(4)
(1  A / 2)
where:
RFB1 and RFB2 are the external feedback resistors.
20 Ω is the internal switch on resistance.
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
the LO-x outputs.
RFB can be equal to RFB1, RFB2, or (RFB1 + 20 Ω)||(RFB2 + 20 Ω),
depending on the connection status of the internal switches.
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 IN 
( R FB1  20  ) || ( R FB2  20  )  30 
(1  A / 2)
|| 6 k 
Table 8. Active Termination Example for LNA Gain = 21.6 dB,
RFB1 = 650 Ω, and RFB2 = 1350 Ω
Reg. 0x02C
Value
00 (default)
01
10
11
1
2
LO-x
Switch
On
On
Off
Off
LOSW-x
Switch
Off
On
On
Off
RFB (Ω)
RFB1
RFB1 || RFB2
RFB2
∞
RIN (Ω)1
100
66
200
∞
See Equation 4.
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
RS = 500Ω, RFB = 2kΩ
RS = 200Ω, RFB = 800Ω
100
RS = 100Ω, RFB = 400Ω, CSH = 20pF
RS = 50Ω, RFB = 200Ω, CSH = 70pF
10
100k
1M
10M
FREQUENCY (Hz)
100M
Figure 34. RIN vs. Frequency for Various Values of RFB
(Effects of RSH and CSH Are Also Shown)
(5)
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.
RS (Ω)
100
50
200
N/A2
11041-024
Active Impedance Matching
RFB 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×).
INPUT RESISTANCE (Ω)
are critical for accurate impedance control. The use of a fully
differential topology and negative feedback minimizes distortion. Low second-order harmonic distortion is particularly
important in harmonic ultrasound imaging applications.
However, as seen for larger RIN values, parasitic capacitance
starts rolling off the signal BW before the LNA produces peaking.
CSH 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
Figure 36 shows the noise figure as it relates to RS for various
values of RIN, which is helpful for design purposes.
8
RFB (Ω)
150
200
300
350
450
650
750
950
1350
Minimum CSH (pF)
90
70
50
30
20
10
Not applicable
Not applicable
Not applicable
6
0
10
100
RS (Ω)
1k
Figure 36. Noise Figure vs. RS for Various Fixed Values of RIN,
Active Termination Matched Inputs, VGAIN = 1.6 V
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
RFB1
LO-x
RFB2
LOSW-x
T/R
SWITCH CS
LI-x
LG-x
CSH
LNA
15.6dB,
17.9dB,
21.6dB
CLG
TRANSDUCER
gm
12.0
DC OFFSET
CORRECTION
10.5
Figure 37. Simplified LNA Input Configuration
9.0
NOISE FIGURE (dB)
3
1
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 Ω are 7 dB, 4 dB, and 2.5 dB for the shunt termination, active termination, and unterminated configurations,
respectively. The noise figures for 200 Ω are 4.5 dB, 1.7 dB,
and 1 dB, respectively.
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
10
100
RS (Ω)
1k
11041-025
1.5
0
4
2
LNA Noise
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 measurements, 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.
5
11041-026
RIN (Ω)
50
50
50
100
100
100
200
200
200
NOISE FIGURE (dB)
LNA Gain (dB)
15.6
17.9
21.6
15.6
17.9
21.6
15.6
17.9
21.6
RIN = 50Ω
RIN = 75Ω
RIN = 100Ω
RIN = 200Ω
UNTERMINATED
7
Table 9. Active Termination External Component Values
11041-035
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 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
gm
(mS)
0.5
1.0
1.5
2.0
LNAGAIN =
15.6 dB (kHz)
41
83
133
167
LNAGAIN =
17.9 dB
(kHz)
55
110
178
220
AD9670
LNAGAIN =
21.6 dB (kHz)
83
167
267
330
0.01µF
GAIN–
100Ω
0.01µF
249Ω
ADA4938-1/
ADA4938-2
31.3kΩ
±1.6V
0.8V CM
249Ω
10kΩ
±0.8V DC
AT 0.8V CM
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.
fHP (CLG) =
10 nF
g
1
× LNAGAIN × m = fHP (see Table 10) ×
2 
C LG
C LG
GAIN+
±0.8V DC
100Ω AT 0.8V CM
11041-027
Reg. 0x120,
Bits[4:3]
00 (default)
01
10
11
AVDD2
249Ω
Table 10. High-Pass Filter Cutoff Frequency, fHP, for CLG = 10 nF
VGA Noise
(6)
Variable Gain Amplifier (VGA)
The differential X-AMP VGA provides precise input attenuation 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 architecture 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.
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.
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.
Gain Control
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.
The analog gain control interface, GAIN±, is a differential input.
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 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.
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
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 highpass 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,
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.
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]
0 0000
LPF Cutoff
Frequency (MHz)
1.45 × (1/3) × fSAMPLE
Sampling Frequency (MHz)
20.5
9.91
0 0001
1.25 × (1/3) × fSAMPLE
8.54
0 0010
1.13 × (1/3) × fSAMPLE
0 0011
1.0 × (1/3) × fSAMPLE
0 0100
0.9 × (1/3) × fSAMPLE
0 0101
0.8 × (1/3) × fSAMPLE
0 0110
0.75 × (1/3) × fSAMPLE
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
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
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
Out of tunable
filter range
40
Out of tunable
filter range
16.67
10.67
65
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
10.00
16.25
12.89
20.94
11.11
18.06
10.00
16.25
8.89
14.44
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
17.78
8.00
13.00
16.00
Out of tunable
filter range
Out of tunable
filter range
9.67
11.56
14.22
10.83
13.33
15.71
8.33
13.54
Out of tunable
filter range
16.67
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
12.19
15.00
10.83
13.33
9.75
12.00
8.67
10.67
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
16.67
8.13
10.00
15.63
15.00
13.33
12.00
Rev. A | Page 25 of 52
80
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
16.82
125
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
17.50
AD9670
Data Sheet
Table 12. SPI-Selectable Low-Pass Filter Cutoff Options for High Band Mode at Example Sampling Frequencies
Address 0x00F,
Bits[7:3]
0 0000
LPF Cutoff
Frequency (MHz)
1.45 × (1/3) × fSAMPLE
0 0001
1.25 × (1/3) × fSAMPLE
0 0010
1.13 × (1/3) × fSAMPLE
0 0011
1.0 × (1/3) × fSAMPLE
0 0100
0.9 × (1/3) × fSAMPLE
0 0101
0.8 × (1/3) × fSAMPLE
0 0110
0.75 × (1/3) × fSAMPLE
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
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
20.5
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
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
40
19.33
16.67
15.00
Sampling Frequency (MHz)
65
80
Out of tunable
Out of tunable
filter range
filter range
27.08
Out of tunable
filter range
24.38
30.00
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
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
Rev. A | Page 26 of 52
21.67
26.67
19.50
24.00
17.33
21.33
16.25
20.00
20.94
25.78
18.06
22.22
16.25
20.00
14.44
17.78
125
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
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
15.71
16.00
25.00
14.22
22.22
Out of tunable
filter range
19.33
20.83
13.54
16.67
Out of tunable
filter range
26.04
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
15.00
23.44
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
Out of tunable
filter range
20.83
18.75
16.67
15.63
Data Sheet
AD9670
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 transformer 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, VFAC3BHL-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
0.1µF
Table 13. High-Pass Filter Cutoff Options
1
Ratio1
12.00
9.00
6.00
3.00
High-Pass Cutoff Frequency
Low-Pass
Low-Pass
Cutoff = 8 MHz Cutoff = 18 MHz
670 kHz
1.5 MHz
890 kHz
2.0 MHz
1.33 MHz
3.0 MHz
2.67 MHz
6.0 MHz
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 antialiasing 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.
50Ω 100Ω
CLK+
AD9670
0.1µF
VFAC3
CLK–
0.1µF
SCHOTTKY
DIODES:
HSM2812
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.
3.3V
AD9516-x
OR AD9524
VFAC3
0.1µF
0.1µF
CLK+
CLK
OUT
50Ω*
0.1µF
ADC
AD9670
100Ω
PECL DRIVER
0.1µF
CLK–
CLK
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.
240Ω
240Ω
11041-029
Reg. 0x02B[1:0]
High-Pass Filter
Cutoff
00 (default)
01
10
11
OUT
MINI-CIRCUITS®
ADT1-1WT, 1:1Z
0.1µF
XFMR
11041-028
A total of four SPI-programmable settings allow the user to vary
the high-pass filter cutoff frequency as a function of the lowpass 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.
Clock Input Considerations
*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.
VFAC3
AD9516-x
OR AD9524
0.1µF
0.1µF
CLK+
CLK
OUT
50Ω*
0.1µF
LVDS DRIVER
CLK
100Ω
AD9670
0.1µF
CLK–
*50Ω RESISTOR IS OPTIONAL.
Figure 41. Differential LVDS Sample Clock
Rev. A | Page 27 of 52
11041-030
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.
AD9670
Data Sheet
130
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).
AD9516-x
OR AD9524
0.1µF
CLK
50Ω*
CMOS DRIVER
OPTIONAL
0.1µF
100Ω
SNR (dB)
VFAC3
OUT
110
CLK+
90
14 BITS
80
CLK–
12 BITS
70
50
10 BITS
8 BITS
40
11041-031
0.1µF
*50Ω RESISTOR IS OPTIONAL.
30
Figure 42. Single-Ended, 1.8 V CMOS Sample Clock
1
0.125ps
0.25ps
0.5ps
1.0ps
2.0ps
10
100
ANALOG INPUT FREQUENCY (MHz)
1000
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 characteristics. 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 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.
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)
16 BITS
60
AD9670
CLK
0.1µF
100
11041-033
3.3V
RMS CLOCK JITTER REQUIREMENT
120
(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, crystalcontrolled 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.
For more information about how jitter performance relates to
ADCs, refer to the AN-501 Application Note and the AN-756
Application Note.
The power dissipated by the AD9670 is proportional to its
sample rate. The digital power dissipation does not vary significantly 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.
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.
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 μs is required when
using the recommended 1 μF and 0.1 μF decoupling capacitors
on the VREF pin and the 0.01 μF decoupling capacitors on the
GAIN± pins. Most of this time is dependent on the gain decoupling:
Rev. A | Page 28 of 52
Data Sheet
AD9670
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 dependent on gain. To achieve a 2 μs 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.
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.
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.
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.
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).
TX_TRIG±
DIGITAL
POWER
ANALOG
POWER
POWER_STOP
(PROFILE SPECIFIC)
POWER_START
(PROFILE SPECIFIC)
POWER_SETUP
(SPI SET)
11041-034
higher value decoupling capacitors on the GAIN± pins result in
longer wake-up times.
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.
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.
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
300
200
100
0
–100
–200
–300
–400
ULS: 11197/11197
–1.5
–1.0
–0.5
300
0
TIME (ns)
0.5
1.0
1.5
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
0
–100
70
–200
–300
–1.0
–0.5
0
TIME (ns)
0.5
1.0
1.5
Figure 45. Data Eye for LVDS Outputs in ANSI-644 Mode with Trace Lengths
of Less Than 24 Inches on Standard FR-4 Material
50
40
30
20
10
70
0
–300
60
50
–200
–100
0
TIME (ps)
100
200
300
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
40
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.
30
20
–100
–50
0
TIME (ps)
50
100
150
11041-144
10
0
–150
60
11041-145
–1.5
11041-044
–400
TIE JITTER HISTOGRAM (Hits)
EYE DIAGRAM VOLTAGE (mV)
400
TIE JITTER HISTOGRAM (Hits)
ULS: 11199/11199
11041-045
EYE: ALL BITS
EYE: ALL BITS
400
EYE DIAGRAM VOLTAGE (mV)
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.
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
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.
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.
Table 14. Digital Output Coding with RF Decimator Bypassed,
Demodulator Bypassed, and Baseband Decimator Bypassed
Code
16384
8192
8191
0
(VIN+) − (VIN−),
Input Span = 2 V p-p (V)
+1.00
0.00
−0.000488
−1.00
Digital Output Mode: Twos
Complement (D13 to D0)
01 1111 1111 1111
00 0000 0000 0000
11 1111 1111 1111
10 0000 0000 0000
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.
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
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.
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.
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.
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.
Rev. A | Page 31 of 52
AD9670
Data Sheet
Table 15. Flexible Output Test Modes
Output Test Mode
Bit Sequence
0000
0001
0010
0011
0100
0101
0110
0111
1000
Pattern Name
Off (default)
Midscale short
+Full-scale short
−Full-scale short
Checkerboard
PN sequence long
PN sequence short
One-/zero-word toggle
User input
Digital Output Word 1
Not applicable
10 0000 0000 0000
11 1111 1111 1111
00 0000 0000 0000
10 1010 1010 1010
Not applicable
Not applicable
11 1111 1111 1111
Register 0x019 and Register 0x01A
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
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).
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.
Table 16. PN Sequence
PN Sequence
Short
Long
Initial Value
0x092
0x003
First Three Output Samples
(MSB First, 16-Bit)
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.
Digital Output Word 2
Not applicable
Same
Same
Same
01 0101 0101 0101
Not applicable
Not applicable
00 0000 0000 0000
Register 0x01B and Register 0x01C
Subject to
Resolution Select
Not applicable
Yes
Yes
Yes
No
Yes
Yes
No
No
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.
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.
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.
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 μF
capacitor and a 1 μF 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.
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 synchronizes 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.
Table 17. Test Signal Fundamental Amplitude at the LNA
Output
Reg. 0x116, Bits[3:2],
Analog Test Tones
00 (Default)
01
10
At 15.6 dB
(mV p-p)
80
160
320
LNA Gain
At 17.9 dB
(mV p-p)
98
196
391
At 21.6 dB
(mV p-p)
119
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.
Table 18. Test Signal Fundamental Amplitude at the
ADC Input, VGAIN = 0 V, PGA Gain = 21 dB
Register 0x116,
Bits[3:2],
Analog Test Tones
00 (Default)
01
10
At 15.6 dB
(dBFS)
−29
−23
−17
LNA Gain
At 17.9 dB
(dBFS)
−28
−22
−16
At 21.6 dB
(dBFS)
−26
−20
−14
Table 19. Test Signal Fundamental Amplitude at the
ADC Input, VGAIN = 0 V, PGA Gain = 30 dB
Register 0x116,
Bits[1:0],
Analog Test Tones
00 (Default)
01
10
At 15.6 dB
(dBFS)
−20
−14
−8
LNA Gain
At 17.9 dB
(dBFS)
−19
−13
−7
At 21.6 dB
(dBFS)
−17
−11
−5
Quadrature Generation
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 maximum 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.
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.
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.
Rev. A | Page 33 of 52
AD9670
Data Sheet
I/Q Demodulator and Phase Shifter
Table 20. Phase Select Code for Channel-to-Channel Phase Shift
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°
337.5°
Rev. A | Page 34 of 52
I/Q Demodulator Phase
(SPI Register 0x02D, Bits[3:0])
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
1111 (not valid in 4LO mode)
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.
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. Saturation 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.
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
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.
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.
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
ADC OUTPUT OR
TEST WAVEFORM
MULTIBAND AAF
DECIMATE BY 2
DC OFFSET
CALIBRATION
Cos
HIGH-PASS
FILTER
BB DECIMATOR
I
Q
RF DECIMATOR
LOW-PASS
FILTER
DECIMATOR
LOW-PASS
FILTER
DECIMATOR
FRAMER
SERIALIZER
BB DEMODULATOR
Figure 49. Simplified Block Diagram of a Single Channel of the Demodulator/Decimator
Table 21. Profile Definition
Field
f
Bits
16
P
8
Description
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
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
11041-038
–Sin
AD9670
Data Sheet
Field
M
Bits
5
g
3
HPF Bypass
1
POWER_START
Reserved
POWER_STOP
15
1
15
TX_TRIG indicates the differential signal created via the TX_TRIG− pin and the TX_TRIG+ pin.
10
RF DECIMATOR
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].
LOW BAND FILTER
–10
HIGH BAND FILTER
–20
–30
–40
–50
–60
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)
2
1
0
AMPLITUDE (dBFS)
The multiband filter is a finite pulse response (FIR). It is programmable 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.
11041-039
DC Offset Calibration
0
AMPLITUDE (dBFS)
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).
–1
LOW BAND FILTER
HIGH BAND FILTER
–2
–3
–4
–5
–6
–7
–8
0
2
4
6
8
10
12
FREQUENCY (MHz)
14
16
18
20
11041-040
1
Description
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
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)
Digital high-pass filter (HPF) bypass
0 = disable (filter enabled)
1 = enable (filter bypassed)
ADC clock counted from TX_TRIG1 signal assertion when the active channels are powered up
Reserved
ADC clock counted from TX_TRIG1 signal assertion when the active channels are powered down
Figure 51. Antialiasing Filter Frequency Response (Frequency Scale Assumes
fADC = 2 × fDEC = 40 MHz)
Rev. A | Page 36 of 52
Data Sheet
AD9670
High-Pass Filter
Table 22. Coefficient Memory for M = 4
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 μs 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 μs/0.02 μs) 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.
j\i
0
1
2
3
BASEBAND DEMODULATOR AND DECIMATOR
The demodulator downconverts the RF signal to a baseband
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 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 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.
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.
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.
7
28
29
30
31
6
27
26
25
24
5
20
21
22
23
4
19
18
17
16
3
12
13
14
15
2
11
10
9
8
1
4
5
6
7
0
3
2
1
0
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.
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.
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 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
(8)
n = M × i + j, if i is odd
(9)
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.
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.
Rev. A | Page 37 of 52
AD9670
Data Sheet
Table 23. Coefficient Block Mapping into SPI Memory Location
C7[13:0]
111:98
B13[7:0]
111:104
C6[13:0]
97:84
B12[7:0]
103:96
B11[7:0]
95:88
C5[13:0]
83:70
B10[7:0]
87:80
B9[7:0]
79:72
Coefficients (8 Words × 14 Bits)
C4[13:0]
C3[13:0]
69:56
55:42
SPI Memory (14 Bytes)
B8[7:0] B7[7:0] B6[7:0] B5[7:0] B4[7:0]
71:64
63:56
55:48
47:40
39:32
C2[13:0]
41:28
B3[7:0]
31:24
B2[7:0]
23:16
C1[13:0]
27:14
C0[13:0]
13:0
B1[7:0]
15:8
B0[7:0]
7:0
DIGITAL TEST WAVEFORMS
DIGITAL BLOCK POWER SAVING SCHEME
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
in Register 0x11A.
To reduce power consumption in the digital block, the
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).
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).
x = C + A × sin(2 × π × N)
A=
64
AFULL
2
(see Register 0x117)
(10)
CHIP IN POWER-DOWN,
STANDBY,
OR CW MODE
(11)
RUN CHIP
SCALE
x
(see Register 0x118)
C = AFULL SCALE × a × 2−(13 − b) (see Register 0x119)
(12)
DIGITAL
DEMODULATOR
IDLE
(13)
Channel ID and Ramp Generator
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.
TX_TRIG IS HIGH, PROFILE
INDEX WRITE, OR POWER
STOP EXPIRES
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
NEGATIVE EDGE TX_TRIG
OR S/W TX_TRIG
DIGITAL
DEMODULATOR
RUNNING
Figure 52. Digital Block Power Saving Scheme
11041-048
N=
f SAMPLE  n
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.
Data Sheet
AD9670
SERIAL PORT INTERFACE (SPI)
Table 24. Serial Port Pins
Pin
SCLK
SDIO
CSB
Function
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.
Chip select bar (active low). This control gates the read
and write cycles.
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 definitions are shown in Figure 54 and Table 25.
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.
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
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 readback 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.
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.
HARDWARE INTERFACE
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.
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.
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
0
10
20
30
40
50
60
70
80
90
NUMBER OF SDIO PINS CONNECTED TOGETHER
100
11041-041
Three pins define the serial port interface, or SPI: SCLK, SDIO,
and CSB (see Table 24). The SCLK (serial clock) pin synchronizes 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.
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 recommended that a 1-, 2-, or 3-byte transfer be used exclusively. Without
an active CSB line, streaming mode can be entered but not exited.
VOH (V)
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.
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).
Rev. A | Page 39 of 52
AD9670
Data Sheet
tDS
tS
tHIGH
tCLK
tH
tDH
tLOW
CSB
DON’T
CARE
SDIO
DON’T
CARE
DON’T
CARE
R/W
W1
W0
A12
A11
A10
A9
A8
A7
D5
D4
D3
D2
D1
D0
DON’T
CARE
11041-042
SCLK
Figure 54. Serial Timing Diagram
Table 25. Serial Timing Definitions
Parameter
tDS
tDH
tCLK
tS
tH
tHIGH
tLOW
tEN_SDIO
Timing (ns min)
12.5
5
40
5
2
16
16
15
tDIS_SDIO
15
Description
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
Hold time between CSB and SCLK
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)
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.
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.
For more information on the SPI memory map and other
functions, see the AN-877 Application Note, Interfacing to
High Speed ADCs via SPI.
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.”
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.
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
0x3C
0x0X (default)
0x01
0x0F
0x3F
0x03
Description
Initiates an SPI reset
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
Bypasses the demodulator and decimator; bypasses the RF decimator; enables the digital
high-pass filter
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)
Decimates by 2 (M = 00001); digital gain = 16 (g = 100)
Points to Coefficient Block 00
Demodulation frequency = fSAMPLE/8
0x011
0x06 (default)
0xF00
0xFF
0xF01
0x7F
0xF02
0x00
0xF03
0x80
0xF04
0x0C
0xF05
0x00
0xF06
0x00
0xF07
0x20
Additional profile memory and coefficient memory can be written here
0x10C1
0x00 (default)
Sets index profile (required after profile memory writes)
0x014
0x00
Sets the output data format
0x008
0x00
TGC run mode2
0x021
0x05
14 bits, 8 lanes, FCO covers the entire frame
0x199
0x80
Enables automatic clocks per sample calculation
0x19B
0x50
Serial format
0x188
0x01
Enables the start code
0x18B
0x27
Sets the start code MSB
0x18C
0x72
Sets the start code LSB
0x182
0x82
Autoconfigures the PLL
0x10C3
0x20
Sets SPI TX_TRIG and index profile2
0x00F
0x18
Sets the low-pass filter cutoff frequency, and mode
0x02B
0x40
Sets the analog LPF and HPF to defaults, tune filters4
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.
Running the chip from full power-down mode requires 375 μs wake-up time, as listed in Table 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.
2
3
Rev. A | Page 42 of 52
Data Sheet
AD9670
Table 27. Memory Map Registers
Addr.
Register Name
(Hex)
Chip Configuration Registers
0x000 CHIP_PORT_CONFIG
Bit 7 (MSB)
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0 (LSB)
Default
Value
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
0x001
CHIP_ID
0x002
CHIP_GRADE
X
X
0x0FF
DEVICE_UPDATE
X
X
Speed mode, Bits[5:4]
(identify device
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)
X
X
0x004
DEVICE_INDEX_2
X
X
X
X
0x005
DEVICE_INDEX_1
X
X
0x008
GLOBAL_MODES
X
Clock
channel
FCO±
0 = off
1 = on
(default)
0
0x009
GLOBAL_CLOCK
X
LNA input
impedance
0 = 6 kΩ
(default)
1 = 3 kΩ
X
Clock
channel
DCO±
0 = off
1 = on
(default)
X
X
X
X
0x00A
PLL_STATUS
PLL lock
status
0 = not
locked
1 = locked
X
X
X
X
0x7C
Chip ID, Bits[7:0]
(AD9670 = 0xA6) (default)
Comments
Mirror nibbles so
that LSB or MSB
first mode is set
correctly, regardless of shift mode.
An SPI reset reverts
all registers (including the LVDS
registers), except
Register 0x000, to
their default
values, and
Register 0x000,
Bits[2:5] are
automatically
cleared.
Default is unique
chip ID, different
for each device;
read-only register.
Speed mode is used
to differentiate the
ADC speed power
modes (the user
must update
Reg. 0x0FF to
initiate the mode
setting).
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.
Bits are set to
determine which
on-chip device
receives the next
write command.
Bits are set to
determine which
on-chip device
receives the next
write command.
X
X
X
X
0x0X
X
X
X
X
0x00
Data
Channel H
0 = off
1 = on
(default)
Data
Channel D
0 = off
1 = on
(default)
Data
Channel G
0 = off
1 = on
(default)
Data
Channel C
0 = off
1 = on
(default)
Data
Channel F
0 = off
1 = on
(default)
Data
Channel B
0 = off
1 = on
(default)
Data
Channel E
0 = off
1 = on
(default)
Data
Channel A
0 = off
1 = on
(default)
0x0F
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)
X
X
DCS
0 = off
1 = on
(default)
X
X
X
0x01
Determines the
generic modes of
chip operation
(global).
0x01
Turns the internal
DCS on and off
(global).
0x00
Monitors the PLL
lock status (read
only, global).
Rev. A | Page 43 of 52
0x3F
AD9670
Addr.
(Hex)
0x00D
Register Name
TEST_IO
0x00E
GPO
0x00F
FLEX_CHANNEL_
INPUT
0x010
0x011
FLEX_OFFSET
FLEX_GAIN
0x012
0x013
0x014
Data Sheet
Bit 7 (MSB)
User test
mode
0 = continuous,
repeat user
patterns
(1, 2, 3, 4,
1, 2, 3, 4 …)
(default)
1 = single
clock cycle
user patterns, then
zeros (1, 2,
3, 4, 0, 0 …)
X
Bit 6
X
Bit 5
Reset PN
long generation
0 = on,
PN long
running
(default)
1 = off,
PN long
held in
reset
Bit 4
Reset PN
short generation
0 = on,
PN short
running
(default)
1 = off,
PN short
held in
reset
X
X
X
Bit 3
Bit 2
Bit 1
Bit 0 (LSB)
Output test mode
0000 = off (default)
0001 = midscale short
0010 = +FS short
0011 = −FS short
0100 = checkerboard output
0101 = PN sequence long
0110 = PN sequence short
0111 = one-/zero-word toggle
1000 = user input
1001:1110 = reserved
1111 = ramp output
General-purpose digital outputs
0
0
PGA gain
00 = 21 dB
01 = 24 dB (default)
10 = 27 dB
11 = 30 dB
BIAS_CURRENT
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
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
X
X
1
0
Digital VGA gain control
0000 = GAIN± pins enabled (default)
0001 = 0.0 dB (maximum gain, GAIN± pins disabled)
0010 = −3.5 dB
0011 = −7.0 dB
…
1110 = −45.5 dB
1111 = −45.5 dB
X
X
X
X
1
PGA bias
0 =100%
(default)
1 = 60%
RESERVED_13
OUTPUT_MODE
0
X
0
X
0
Output
data invert
0 = disable
(default)
1 = enable
0
X
0
X
0
Output
data
enable
0 = enable
(default)
1=
disable
BW mode
0 = low
(default,
8 MHz to
18 MHz)
1 = high
(13.5 MHz
to 30 MHz)
Rev. A | Page 44 of 52
X
0
X
Default
Value
0x00
0x00
0x18
Comments
When this register
is set, the test data
is placed on the
output pins in
place of normal
data (local).
Values placed on
the GPO0 to
GPO3 pins (global).
Antialiasing filter
cutoff (global).
0
LNA gain
00 = 15.6 dB
01 = 17.9 dB
10 = 21.6 dB
(default)
0x20
0x06
Reserved.
LNA and PGA gain
adjustment (global).
LNA bias
00 = high
01 = midhigh (default)
10 = midlow
11 = low
0
0
Output data format
00 = offset binary
01 = twos complement
(default)
10 = gray code
11 = reserved
0x09
LNA bias current
adjustment (global).
0x00
0x01
Reserved.
Data output
modes (local).
Data Sheet
Addr.
(Hex)
0x015
AD9670
Bit 7 (MSB)
LVDS
output
standard
0 = ANSI
(default)
1 = IEEE
(low
power)
Bit 6
1
Bit 5
1
Bit 4
0
Bit 3
LVDS drive
strength
enable
0 = disable
(default)
1 = enable
0x016
FLEX_OUTPUT_
PHASE
X
X
0
DCO
invert
0=
disable
(default)
1=
enable
X
0x017
FLEX_OUTPUT_
DELAY
DCO delay
enable
0 = disable
(default)
1 = enable
X
X
0x018
0x019
FLEX_VREF
USER_PATT1_LSB
X
B7
X
B6
X
B5
X
B4
X
B3
0x01A
USER_PATT1_MSB
B15
B14
B13
B12
B11
B10
B9
B8
0x00
0x01B
USER_PATT2_LSB
B7
B6
B5
B4
B3
B2
B1
B0
0x00
0x01C
USER_PATT2_MSB
B15
B14
B13
B12
B11
B10
B9
B8
0x00
0x01D
USER_PATT3_LSB
B7
B6
B5
B4
B3
B2
B1
B0
0x00
0x01E
USER_PATT3_MSB
B15
B14
B13
B12
B11
B10
B9
B8
0x00
0x01F
USER_PATT4_LSB
B7
B6
B5
B4
B3
B2
B1
B0
0x00
0x020
USER_PATT4_MSB
B15
B14
B13
B12
B11
B10
B9
B8
0x00
0x021
FLEX_SERIAL_CTRL
0
FCO
invert
0 = not
inverted
(default)
1=
inverted
Lane low
rate
0 = normal
(default)
1 = low
sample
frequency
(<32 MHz)
0x022
SERIAL_CH_STAT
X
X
FCO rate
with demodulator
enabled
0 = FCO
per I/Q
(default)
1 = FCO
per sample
(I and Q)
X
Lane mode
00 = 1 channel/lane
(8 lanes) (default)
01 = 2 channels/lane
(4 lanes)
10 = 4 channels/lane
(2 lanes)
11 = 8 channels/lane
(1 lane)
X
X
Bit 1
Bit 0 (LSB)
LVDS drive current
000 = 3.72 mA
001 = 3.5 mA (default)
010 = 3.30 mA
011 = 2.96 mA
100 = 2.82 mA
101 = 2.57 mA
110 = 2.27 mA
111 = 2.0 mA (reduced range)
X
DCO
phase
adjust
with
respect to
DOUT
00 = +90°
(default)
01 = 0°
10 = 0°
11 = −90°
DCO clock delay
00000: 100 ps (default)
00001 = 200 ps
00010 = 300 ps
…
11101 = 3.0 ns
11110 = 3.1 ns
11111 = 3.2 ns
1
0
0
B2
B1
B0
X
Rev. A | Page 45 of 52
Bit 2
Default
Value
0x61
Register Name
OUTPUT_ADJUST
Output word length
00 = 12 bits (default)
01 = 14 bits
10 = 16 bits
11 = reserved
X
Channel
powerdown
1 = on
0 = off
(default)
Comments
Data output levels
(global).
0x00
DCO inversion and
course phase
adjustment
(global).
0x00
DCO delay
(global).
0x04
0x00
Reserved (global).
User Defined
Pattern 1, LSB
(global).
User Defined
Pattern 1, MSB
(global).
User Defined
Pattern 2, LSB
(global).
User Defined
Pattern 2, MSB
(global).
User Defined
Pattern 3, LSB
(global).
User Defined
Pattern 3, MSB
(global).
User Defined
Pattern 4, LSB
(global).
User Defined
Pattern 4, MSB
(global).
LVDS control
(global).
0x00
0x00
Used to power
down individual
channels (local).
AD9670
Data Sheet
Addr.
(Hex)
0x02B
Register Name
FLEX_FILTER
Bit 7 (MSB)
X
0x02C
LNA_TERM
0x02D
Bit 5
X
Bit 4
X
Bit 3
Bypass
analog
HPF
0 = off
(default)
1 = on
Bit 2
X
X
Bit 6
Enable
automatic
low-pass
tuning
1 = on
(self
clearing)
X
X
X
X
X
CW_ENABLE_
PHASE
X
X
X
CW
Doppler
channel
enable
0 = off
(default)
0 = on
0x02E
CW_LO_MODE
RESET
with MLO
clock
edge
0 = synchronous
(default)
1 = asynchronous
Synchronous RESET
sampling
MLO±
clock edge
0 = falling
(default)
1 = rising
RESET
signal
polarity
0 = active
high
(default)
1 = active
low
0x02F
CW_OUTPUT
0
0
0x102
0x103
0x104
0x105
0x106
0x107
RESERVED_102
RESERVED_103
RESERVED_104
RESERVED_105
RESERVED_106
RESERVED_107
Partially
enables
LVDS
during CW
0: LVDS
link disabled
during CW
(default)
1: LVDS
link partially
enabled
during
CW. PLL,
FCO, and
DCO are
enabled,
while LVDS
data drivers
are disabled
(switching
activity can
degrade
CW performance)
CW output
dc bias
voltage
0 = bypass
1 = enable
(default)
0
0
0
0
0
0
0
0
0
0
0
0
0x108
RESERVED_108
0
0
Bit 1
Bit 0 (LSB)
Analog high-pass filter
cutoff
00 = fLP/12.00 (default)
01 = fLP/9.00
10 = fLP/6.00
11 = fLP/3.00
Default
Value
0x00
Comments
Filter cutoff
(global); (fLP = lowpass filter cutoff
frequency).
LO-x, LOSW-x
connection
00 = RFB1 (default)
01 = (RFB1 || RFB2)
10 = RFB2
11 = ∞
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°
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)
MLO and
LO mode
RESET
00X = 4LO, 3rd to 5th odd harmonic
buffer
rejection (default)
enable (in
010 = 8LO, 3rd to 5th odd harmonic
all modes
rejection
except CW
011 = 8LO, 3rd to 13th odd harmonic
mode)
rejection
0 = power100 = 16LO, 3rd to 5th odd harmonic
down
rejection
(default)
101 = 16LO, 3rd to 13th odd harmonic
1 = enable
rejection
11X = reserved
0x00
LNA active
termination/input
impedance (global).
0x00
Phase of
demodulators
(local, chip).
0x00
CW mode
functions (global).
0
0
0
0
0
0x80
Global.
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
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
0
0
0
0
0
0
0X00
0X00
0x3F
0x00
0x00
Read
only
0x00
Rev. A | Page 46 of 52
Reserved.
Data Sheet
AD9670
Addr.
(Hex)
0x109
Register Name
VGA_TEST
Bit 7 (MSB)
X
Bit 6
X
Bit 5
X
0x10C
PROFILE_INDEX
X
X
0x10D
0x10E
0x10F
RESERVED_10D
RESERVED_10E
DIG_OFFSET_CAL
1
1
0
1
1
0
Manual
TX_TRIG
0 = off,
use pin
(default)
1 = on,
autogenerate
TX_TRIG
(self clears)
1
1
0
1
1
0
0x110
DIG_OFFSET_
CORR1
DIG_OFFSET_
CORR2
D7
D6
D5
D4
0x111
0x112
POWER_MASK_
CONFIG
0x113
DIG_DEMOD_
CONFIG
0x115
CHIP_ADDR_EN
Bit 4
VGA/ antialiasing
filter test
mode
enable
0 = off
(default)
1 = on
Bit 3
X
Bit 2
Bit 1
Bit 0 (LSB)
VGA/ antialiasing filter output test mode
000 = Channel A (default)
001 = Channel B
010 = Channel C
011 = Channel D
100 = Channel E
101 = Channel F
110 = Channel G
111 = Channel H
Profile Index[4:0]
1
1
Digital
offset
calibration
status
0 = not
complete
(default)
1=
complete
D3
1
1
D2
1
1
1
1
Digital offset calibration
000 = disable correction, reset
correction value (default)
001 = average 210 samples
010 = average 211 samples
…
111 = average 216 samples
D1
D0
D15
D14
D13
D12
D11
D10
D9
D8
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
X
X
X
Power up set-up time (POWER_SETUP)
0 0000 = 0
0 0001 = 1 × 40/fSAMPLE
0 0010 = 2 × 40/fSAMPLE (default)
0 0011 = 3 × 40/fSAMPLE
…
1 1111 = 31 × 40/fSAMPLE
X
X
Digital
Decimator
Decimator and filter
Baseband Demodhigh-pass
gain scale
enable
decimator ulator
filter
0 = no
00 = RF 2× decimator
0 = enable 0 = enable
0 = enable gain
bypassed (default)
(default)
(default)
(default)
(default)
01 = RF 2× decimator
1 = bypass 1 = bypass
1 = bypass 1 = 4×
enabled and low band
gain (shift
filter
decimator
1X = RF 2× decimator
output by enabled and high band
2)
filter
X
X
Chip
Chip address qualifier
address
0 0000 (default)
mode
(If read, returns the state of ADDR0 to ADDR4 pins)
0 = disable
(default)
1 = enable
Rev. A | Page 47 of 52
Default
Value
0x00
Comments
VGA/ antialiasing
filter test mode
enables antialiasing
filter output to the
GPO2 and GPO3
pins (global).
0x00
Index for profile
memory selects
active profile
(global).
0xFF
0xFF
Reserved.
Reserved.
Controls digital
offset calibration
enable and the
number of samples
used (global).
0x00
Offset correction
LSB (local, chip).
Offset correction
MSB (local, chip).
0x00
0x02
Power setup time
is used to set the
power-up time
(global).
0x00
Enable stages of
the digital
processing
(global).
0x00
Chip address
mode enables the
addressing of
devices if the value
of the chip address
qualifier equals the
state on the ADDR0
to ADDR4 pins
(global).
AD9670
Addr.
(Hex)
0x116
Data Sheet
Register Name
ANALOG_TEST_
TONE
Bit 7 (MSB)
X
Bit 6
X
Bit 5
X
0x117
DIG_SINE_TEST_
FREQ
X
X
X
0x118
DIG_SINE_TEST_
AMP
X
X
X
0x119
DIG_SINE_TEST_
OFFSET
0x11A
TEST_MODE_
CH_ENABLE
0x11B
TEST_MODE_
CONFIG
0x11C
0x11D
0x11E
0x11F
0x120
RESERVED_11C
RESERVED_11D
RESERVED_11E
RESERVED_11F
CW_TEST_TONE
0
0
0
0
0
0x180
0x181
0x182
RESERVED_180
RESERVED_181
PLL_STARTUP
0x183
0x184
0x186
RESERVED_183
RESERVED_184
RESERVED_186
1
0
PLL auto
configure
0 = disable
(default)
1 = enable
0
0
1
Channel H
enable
0 = off
(default)
1 = on
X
Bit 4
X
Bit 3
Bit 2
Analog test signal
amplitude
(see Table 17 to Table 19)
Bit 1
Bit 0 (LSB)
Analog test signal
frequency
00 = fSAMPLE/4 (default)
01 = fSAMPLE/8
10 = fSAMPLE/16
11 = fSAMPLE/32
Digital test tone frequency
0 0000 = 1 × fSAMPLE/64
0 0001 = 2 × fSAMPLE/64
…
1 1111 = 32 × fSAMPLE/64
Digital test tone amplitude
0000 = AFULL SCALE (default)
0001 = AFULL SCALE/2
0010 = AFULL SCALE/22
…
1111 = AFULL SCALE/215
Offset exponent (b)
000 = 0 (default)
001 = 1
…
111 = 7
Default
Value
0x00
Comments
Analog test tone
amplitude and
frequency (global).
0x00
Digital sine test
tone frequency
(global).
0x00
Digital sine test
tone amplitude
(global).
Offset multiplier (a)
0 1111 = +15
0 1110 =+14
…
0 0000 = 0 (default)
1 1111 = −1
…
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
Channel G Channel F Channel E Channel D Channel C
Channel B Channel A
enable
enable
enable
enable
enable
enable
enable
0 = off
0 = off
0 = off
0 = off
0 = off
0 = off
0 = off
(default)
(default)
(default)
(default)
(default)
(default)
(default)
1 = on
1 = on
1 = on
1 = on
1 = on
1 = on
1 = on
X
X
X
X
Test mode selection
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
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
CW I/Q
LNA offset LNA offset cancellation
CW analog test tone
output
canceltransconductance
override for Reg. 0x116,
swap
lation
00 = 0.5 mS (default)
Bits[1:0]
0=
0 = enable
01 = 1.0 mS
00 = disable override
disable
(default)
10 = 1.5 mS
(default)
(default)
1 = disable
11 = 2.0 mS
01 = set analog test
1=
tone frequency to fLO
enable
1X = set analog test
tone frequency to dc
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0x00
Digital sine test
tone offset
(global).
0x00
Enable channels
for test mode
(global).
0x00
Enable digital test
modes (global).
0x00
0x00
0x00
0x00
0x00
Reserved.
Reserved.
Reserved.
Reserved.
Sets the frequency
of the analog test
tone to fLO in CW
Doppler mode.
Enables I/Q output
swap. LNA offset
cancellation
control (global).
0x87
0x00
0x02
Reserved.
Reserved.
PLL control
(global).
0
0
0
0x07
0x00
0xAE
Reserved.
Reserved.
Reserved.
0
0
1
X
0
0
0
0
1
0
0
1
1
Rev. A | Page 48 of 52
1
0
1
1
0
0
Data Sheet
AD9670
Addr.
(Hex)
0x187
0x188
Register Name
RESERVED_187
START_CODE_EN
Bit 7 (MSB)
0
0
Bit 6
0
0
Bit 5
1
0
Bit 4
0
0
Bit 3
0
0
Bit 2
0
0
Bit 1
0
0
0x189
0x18A
0x18B
RESERVED_189
RESERVED_18A
START_CODE_MSB
0
0
B15
0
0
B14
0
0
B13
0
0
B12
0
0
B11
0
0
B10
0x18C
START_CODE_LSB
B7
B6
B5
B4
B3
B2
0x190
0x191
0x192
0x193
0x194
0x195
0x196
0x197
0x198
RESERVED_190
RESERVED_191
RESERVED_192
RESERVED_193
RESERVED_194
RESERVED_195
RESERVED_196
RESERVED_197
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
0x199
SAMPLE_CLOCK_
COUNTER
0
0
0
0
0x19A
DATA_OUTPUT_
INVERT
Enable
clocks per
sample
auto
calculation
0 = off
(default)
1 = on
X
X
X
X
X
0x19B
SERIAL_FORMAT
X
0x19C
0x19D
0x19E
0x19F
RESERVED_19C
RESERVED_19D
RESERVED_19E
RESERVED_19F
0
0
0
0
Enable
FCO for
start code
sample
0=
disable
1=
enable
(default)
0
0
0
0
Enable
FCO for
extra
sample at
end of
burst
0 = disable
1 = enable
(default)
0
0
0
0
Enable
FCO continuously
0 = only
during
burst
1 = continuous
(default)
1
0
1
0
0
0
B9
0x00
0x00
0x27
B1
B0
0x72
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
DCO frequency doubling/divider
1011 = 1/32
1010 = 1/64
1001 = 1/128
1000 = 1/256
0000 = 1 (default)
0001 = 2
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
0
0
0
X
X
Invert data
output
0 = noninverted
(default)
1=
inverted
FCO rotate
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
0
0
0
0
Rev. A | Page 49 of 52
0
0
0
0
Default
Value
0x20
0x01
Bit 0 (LSB)
0
Start code
identifier
0 = disable
1 = enable
(default)
0
0
B8
0
0
0
0
0
0
0
0
0x10
0x00
0x18
0x00
0x1C
0x00
0x18
0x00
0x00
Comments
Reserved.
Enables start code
identifier (global).
Reserved.
Reserved.
Start code MSB
(global).
Start code LSB
(global).
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
Reserved.
DCO frequency
control (global).
0x00
Enables automatic
clocks per sample
calculation (global).
0x00
Inverts DOUT
outputs (global).
0x70
FCO controls
(global).
0x10
0x00
0x10
0x00
Reserved.
Reserved.
Reserved.
Reserved.
AD9670
Addr.
(Hex)
Register Name
0x1A0 RESERVED_1A0
0x1A1 RESERVED_1A1
Profile Memory Registers
0x1000 Coefficient memory
to
0x1FFF
Coefficient Memory Registers
0xF00 Profile memory
to
0xFFF
Data Sheet
Bit 7 (MSB)
0
0
Bit 6
0
0
Bit 5
0
0
Bit 4
0
0
Default
Value
0x00
0x00
Comments
Reserved.
Reserved.
32 × 64 bits
0x00
Global.
256 × 112 bits
0x00
Global.
Bit 3
0
0
Rev. A | Page 50 of 52
Bit 2
0
0
Bit 1
0
0
Bit 0 (LSB)
0
0
AD9670
Data Sheet
MEMORY MAP REGISTER DESCRIPTIONS
Profile Index and Manual TX_TRIG (Register 0x10C)
For more information on the SPI memory map and other
functions, consult the AN-877 Application Note, Interfacing to
High Speed ADCs via SPI.
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.
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
A1 BALL
CORNER
10.10
10.00 SQ
9.90
A1 BALL
CORNER
12 11 10 9 8
7 6 5
4
3
2
1
A
B
C
D
8.80
BSC SQ
E
F
G
H
0.80
J
K
L
M
TOP VIEW
0.60
REF
BOTTOM VIEW
DETAIL A
*1.40 MAX
DETAIL A
0.65 MIN
0.25 MIN
0.50
COPLANARITY
0.45
0.20
0.40
BALL DIAMETER
*COMPLIANT WITH JEDEC STANDARDS MO-275-EEAB-1
WITH THE EXCEPTION OF PACKAGE HEIGHT.
01-30-2014-B
PKG-003538
SEATING
PLANE
Figure 55. 144-Ball Chip Scale Package, Ball Grid Array [CSP_BGA]
(BC-144-1)
Dimensions shown in millimeters
ORDERING GUIDE
Model1
AD9670BBCZ
AD9670EBZ
1
Temperature Range
0°C to +85°C
Package Description
144-Ball Chip Scale Package, Ball Grid Array [CSP_BGA]
Evaluation Board
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
Package Option
BC-144-1
Similar pages