AD AD6674-750EBZ 385 mhz bw if diversity receiver Datasheet

385 MHz BW IF Diversity Receiver
AD6674
Data Sheet
FEATURES
APPLICATIONS
JESD204B (Subclass 1) coded serial digital outputs
In band SFDR = 83 dBFS at 340 MHz (750 MSPS)
In band SNR = 66.7 dBFS at 340 MHz (750 MSPS)
1.4 W total power per channel at 750 MSPS (default settings)
Noise density = −153 dBFS/Hz at 750 MSPS
1.25 V, 2.5 V, and 3.3 V dc supply operation
Flexible input range
AD6674-750 and AD6674-1000
1.46 V p-p to 1.94 V p-p (1.70 V p-p nominal)
AD6674-500
1.46 V p-p to 2.06 V p-p (2.06 V p-p nominal)
95 dB channel isolation/crosstalk
Amplitude detect bits for efficient automatic gain control
(AGC) implementation
Noise shaping requantizer (NSR) option for main receiver
function
Variable dynamic range (VDR) option for digital
predistortion (DPD) function
2 integrated wideband digital processors per channel
12-bit numerically controlled oscillator (NCO), up to
4 cascaded half-band filters
Differential clock inputs
Integer clock divide by 1, 2, 4, or 8
Energy saving power-down modes
Flexible JESD204B lane configurations
Small signal dither
Diversity multiband, multimode digital receivers
3G/4G, TD-SCDMA, W-CDMA, GSM, LTE, LTE-A
DOCSIS 3.0 CMTS upstream receive paths
HFC digital reverse path receivers
GENERAL DESCRIPTION
The AD6674 is a 385 MHz bandwidth mixed-signal
intermediate frequency (IF) receiver. It consists of two, 14-bit
1.0 GSPS/750 MSPS/500 MSPS analog-to-digital converters
(ADC) and various digital signal processing blocks consisting of
four wideband DDCs, an NSR, and VDR monitoring. It has an
on-chip buffer and a sample-and-hold circuit designed for low
power, small size, and ease of use. This product is designed to
support communications applications capable of sampling wide
bandwidth analog signals of up to 2 GHz. The AD6674 is
optimized for wide input bandwidth, high sampling rate,
excellent linearity, and low power in a small package.
The dual ADC cores feature a multistage, differential pipelined
architecture with integrated output error correction logic. Each
ADC features wide bandwidth inputs supporting a variety of
user-selectable input ranges. An integrated voltage reference
eases design considerations.
FUNCTIONAL BLOCK DIAGRAM
AVDD1
(1.25V)
AVDD2
(2.5V)
AVDD3
(3.3V)
AVDD1_SR
(1.25V)
DVDD
(1.25V)
DRVDD
(1.25V)
SPIVDD
(1.7V TO 3.4V)
VIN+B
VIN–B
ADC
DIGITALDOWN
DOWNCONVERSION
CONVERSION
DIGITAL
DIGITAL
DOWN
CONVERSION
DIGITAL
DOWNCONVERSION
(×4)
(×4)
(×4)
(×4)
NOISESHAPING
SHAPINGREQUANTIZER
REQUANTIZER
NOISE
(×2)
(×2)
VARIABLEDYNAMIC
DYNAMICRANGE
RANGE
VARIABLE
(×2)
(×2)
TX
OUTPUTS
SIGNAL
MONITOR
DATA ROUTER MUX
FD_B
FAST
DETECT
FD_A
SIGNAL PROCESSING
ADC
JESD204B
HIGH SPEED SERIALIZER
BUFFER
VIN+A
VIN–A
4
SERDOUT0±
SERDOUT1±
SERDOUT2±
SERDOUT3±
BUFFER
CLK+
CLK–
CLOCK
GENERATION
FAST
DETECT
JESD204B
SUBCLASS 1
CONTROL
SIGNAL
MONITOR
PDWN/
STBY
SPI CONTROL
÷2
÷4
÷8
AD6674
AGND
SYSREF± SYNCINB±
SDIO SCLK CSB
DGND DRGND
12400-001
V_1P0
Figure 1.
Rev. C
Document Feedback
Information furnished by Analog Devices is believed to be accurate and reliable. However, no
responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other
rights of third parties that may result from its use. Specifications subject to change without notice. No
license is granted by implication or otherwise under any patent or patent rights of Analog Devices.
Trademarks and registered trademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700 ©2014–2016 Analog Devices, Inc. All rights reserved.
Technical Support
www.analog.com
AD6674* PRODUCT PAGE QUICK LINKS
Last Content Update: 02/23/2017
COMPARABLE PARTS
DESIGN RESOURCES
View a parametric search of comparable parts.
• AD6674 Material Declaration
• PCN-PDN Information
EVALUATION KITS
• Quality And Reliability
• [NO TITLE FOUND] EvalBoard
• Symbols and Footprints
DOCUMENTATION
DISCUSSIONS
Application Notes
View all AD6674 EngineerZone Discussions.
• AN-1371: Variable Dynamic Range
Data Sheet
SAMPLE AND BUY
• AD6674: 385 MHz BW IF Diversity Receiver Data Sheet
Visit the product page to see pricing options.
TOOLS AND SIMULATIONS
TECHNICAL SUPPORT
• AD6674 Delphi Model
• AD9680 / AD6674 IBIS Model
Submit a technical question or find your regional support
number.
REFERENCE MATERIALS
DOCUMENT FEEDBACK
Technical Articles
Submit feedback for this data sheet.
• MS-2714: Understanding Layers in the JESD204B
Specificaton: A High Speed ADC Perspective, Part 1
• MS-2735: Maximizing the Dynamic Range of SoftwareDefined Radio
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.
AD6674
Data Sheet
TABLE OF CONTENTS
Features .............................................................................................. 1
Numerically Controlled Oscillator .......................................... 48
Applications ....................................................................................... 1
FIR Filters ........................................................................................ 50
General Description ......................................................................... 1
General Description ................................................................... 50
Functional Block Diagram .............................................................. 1
Half-Band Filters ........................................................................ 51
Revision History ............................................................................... 3
DDC Gain Stage ......................................................................... 52
Product Highlights ........................................................................... 4
DDC Complex to Real Conversion ......................................... 52
Specifications..................................................................................... 5
DDC Example Configurations ................................................. 53
DC Specifications ......................................................................... 5
Noise Shaping Requantizer (NSR) ............................................... 57
AC Specifications.......................................................................... 6
Decimating Half-Band Filter .................................................... 57
Digital Specifications ................................................................... 8
NSR Overview ............................................................................ 57
Switching Specifications .............................................................. 9
Variable Dynamic Range (VDR) .................................................. 60
Timing Specifications .................................................................. 9
VDR Real Mode.......................................................................... 61
Absolute Maximum Ratings.......................................................... 11
VDR Complex Mode ................................................................. 61
Thermal Characteristics ............................................................ 11
Digital Outputs ............................................................................... 63
ESD Caution ................................................................................ 11
Introduction to JESD204B Interface ........................................ 63
Pin Configuration and Function Descriptions ........................... 12
JESD204B Overview .................................................................. 63
Typical Performance Characteristics ........................................... 14
Functional Overview ................................................................. 64
AD6674-1000 .............................................................................. 14
JESD204B Link Establishment ................................................. 64
AD6674-750 ................................................................................ 18
Physical Layer (Driver) Outputs .............................................. 66
AD6674-500 ................................................................................ 22
JESD204B Tx Converter Mapping ........................................... 68
Equivalent Circuits ......................................................................... 26
Configuring the JESD204B Link .............................................. 68
Theory of Operation ...................................................................... 28
Multichip Synchronization............................................................ 72
ADC Architecture ...................................................................... 28
SYSREF± Setup/Hold Window Monitor ................................. 74
Analog Input Considerations.................................................... 28
Test Modes ....................................................................................... 76
Voltage Reference ....................................................................... 33
ADC Test Modes ........................................................................ 76
Clock Input Considerations ...................................................... 34
JESD204B Block Test Modes .................................................... 76
Power-Down/Standby Mode..................................................... 35
Serial Port Interface (SPI) .............................................................. 79
Temperature Diode .................................................................... 36
Configuration Using the SPI ..................................................... 79
ADC Overrange and Fast Detect .................................................. 37
Hardware Interface..................................................................... 79
ADC Overrange (OR) ................................................................ 37
SPI Accessible Features .............................................................. 79
Fast Threshold Detection (FD_A and FD_B) ........................ 37
Memory Map .................................................................................. 80
Signal Monitor ................................................................................ 38
Reading the Memory Map Register Table............................... 80
SPORT over JESD204B .............................................................. 38
Memory Map Register Table ..................................................... 81
Digital Downconverter (DDC) ..................................................... 41
Applications Information .............................................................. 95
DDC I/Q Input Selection .......................................................... 41
Power Supply Recommendations............................................. 95
DDC I/Q Output Selection ....................................................... 41
Exposed Pad Thermal Heat Slug Recommendations ............ 95
DDC General Description ........................................................ 41
AVDD1_SR (Pin 57) and AGND (Pin 56, Pin 60) ................ 95
Frequency Translation ................................................................... 47
Outline Dimensions ....................................................................... 96
General Description ................................................................... 47
Ordering Guide .......................................................................... 96
DDC NCO + Mixer Loss and SFDR ........................................ 48
Rev. C | Page 2 of 96
Data Sheet
AD6674
REVISION HISTORY
8/2016—Rev. B to Rev. C
Changes to Figure 1........................................................................... 1
Changes to Table 1 ............................................................................ 5
Changes to Table 5 ............................................................................ 9
Changes to Table 8 ..........................................................................12
Added Figure 15; Renumbered Sequentially ...............................15
Added Figure 34 ..............................................................................19
Added Figure 53 ..............................................................................23
Changes to Figure 72 ......................................................................27
Changes to Table 10 ........................................................................32
Changes to Input Clock Driver Section .......................................34
Changes to Clock Jitter Considerations Section .........................35
Changes to Setting Up the NCO FTW and POW Section ........48
Changes to JESD204B Overview Section.....................................63
Changes to Figure 123 Caption and Figure 124 ..........................64
Changes to ADC Test Modes Section...........................................76
Added Datapath Soft Reset Section ..............................................80
Changes to Table 45 ........................................................................81
Changes to Ordering Guide ...........................................................96
4/2015—Rev. A to Rev. B
Changed SPIVDD Range from 1.8 V to 3.3 V to
1.8 V to 3.4 V ..................................................................Throughout
Changes to General Description Section ....................................... 4
Changes to Table 1 ............................................................................ 5
Changes to Table 3 ............................................................................ 8
Changes to Figure 14 ...................................................................... 15
Change to Figure 78 Caption ......................................................... 27
Changes to Table 10 ........................................................................ 29
Changes to Clock Jitter Considerations Section ......................... 32
Added Figure 92; Renumbered Sequentially ............................... 32
Changes to Digital Downconverter (DDC) Section ................... 37
Changes to Table 17 ........................................................................ 46
Changes to Table 23 ........................................................................ 49
Changes to Figure 108 .................................................................... 53
Changes to Figure 116 .................................................................... 56
Changes to Figure 117 and VDR Complex Mode Section ........ 57
Changes to Table 45 ........................................................................ 79
12/2014—Revision A: Initial Version
Rev. C | Page 3 of 96
AD6674
Data Sheet
The analog input and clock signals are differential inputs. The
ADC data outputs are internally connected to four DDCs
through a crossbar mux. Each DDC consists of up to five
cascaded signal processing stages: a 12-bit frequency translator
(NCO), and up to four half-band decimation filters.
Each ADC output is connected internally to an NSR block. The
integrated NSR circuitry allows improved SNR performance in
a smaller frequency band within the Nyquist bandwidth. The
device supports two different output modes selectable via the
SPI. With the NSR feature enabled, the outputs of the ADCs are
processed such that the AD6674 supports enhanced SNR
performance within a limited portion of the Nyquist bandwidth
while maintaining a 9-bit output resolution. NSR is enabled by
default on the AD6674.
Each ADC output is also connected internally to a VDR block.
This optional mode allows full dynamic range for defined input
signals. Inputs that are within a defined mask (based on DPD
applications) are passed unaltered. Inputs that violate this
defined mask result in the reduction of the output resolution.
indicator goes high. Because this threshold indicator has low
latency, the user can quickly turn down the system gain to avoid
an overrange condition at the ADC input. Besides the fast
detect outputs, the AD6674 also offers signal monitoring
capability. The signal monitoring block provides additional
information about the signal being digitized by the ADC.
Users can configure the Subclasss 1 JESD204B-based high speed
serialized output in a variety of two-lane and four-lane
configurations, depending on the DDC configuration and the
acceptable lane rate of the receiving logic device. Multidevice
synchronization is supported through the SYSREF± and
SYNCINB± input pins.
The AD6674 has flexible power-down options that allow significant power savings when desired. All of these features can be
programmed using a 1.8 V capable 3-wire serial port interface
(SPI).
The AD6674 is available in a Pb-free, 64-lead LFCSP, specified
over the −40°C to +85°C industrial temperature range. This
product is protected by a U.S. patent.
With VDR, the dynamic range of the observation receiver is
determined by a defined input frequency mask. For signals
falling within the mask, the outputs are presented at the
maximum resolution allowed. For signals exceeding defined
power levels within this frequency mask, the output resolution
is truncated. This mask is based on DPD applications and
supports tunable real IF sampling, and zero IF or complex IF
receive architectures.
PRODUCT HIGHLIGHTS
Operation of the AD6674 between the DDC, NSR, and VDR
modes is selectable via SPI-programmable profiles.
4.
In addition to the DDC blocks, the AD6674 has several
functions that simplify the AGC function in a communications
receiver. The programmable threshold detector allows
monitoring of the incoming signal power using the fast detect
control bits in Register 0x245 of the ADC. If the input signal
level exceeds the programmable threshold, the fast detect
5.
1.
2.
3.
6.
7.
Rev. C | Page 4 of 96
Wide full power bandwidth supports IF sampling of signals
up to 2 GHz.
Buffered inputs with programmable input termination
eases filter design and implementation.
Four integrated wideband decimation filters and
numerically controlled oscillator (NCO) blocks supporting
multiband receivers.
Flexible SPI controls various product features and
functions to meet specific system requirements.
Programmable fast overrange detection and signal
monitoring.
Programmable fast overrange detection.
9 mm × 9 mm 64-lead LFCSP.
Data Sheet
AD6674
SPECIFICATIONS
DC SPECIFICATIONS
AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, AVDD1_SR = 1.25 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified
maximum sampling rate, 1.0 V internal reference (VREF), AIN = −1.0 dBFS, clock divider = 2, default SPI settings, TA = 25°C, unless
otherwise noted.
Table 1.
Parameter
RESOLUTION
ACCURACY
No Missing Codes
Offset Error
Offset Matching
Gain Error
Gain Matching
Differential Nonlinearity (DNL)
Integral Nonlinearity (INL)
TEMPERATURE DRIFT
Offset Error
Gain Error
INTERNAL VOLTAGE REFERENCE
Voltage
INPUT REFERRED NOISE
VREF = 1.0 V
ANALOG INPUTS
Differential Input Voltage Range
(Internal VREF = 1.0 V)
Common-Mode Voltage (VCM)
Differential Input Capacitance1
Analog Full Power Bandwidth
POWER SUPPLY
AVDD1
AVDD2
AVDD3
AVDD1_SR
DVDD
DRVDD
SPIVDD
IAVDD12
IAVDD22
IAVDD32
IAVDD1_SR2
IDVDD2
IDRVDD2, 3
L = 2 Mode4
ISPIVDD
AD6674-1000
Typ
Max
AD6674-500
Min
Typ
Max
14
Unit
Bits
Guaranteed
0
+0.3
0
+0.3
−6
0
+6
1
+5.1
−0.6
±0.5
+0.7
−4.5
±2.5
+5.0
% FSR
% FSR
% FSR
% FSR
LSB
LSB
Min
14
Full
Full
Full
Full
Full
Full
Full
Guaranteed
0
+0.31
0
+0.23
−6
0
+6
1
+4.5
−0.7
±0.5
+0.8
−5.7
±2.5
+6.9
Guaranteed
0
+0.42
0
+0.41
−6
0
+6
1
+5.2
−0.6
±0.5
+0.8
−3.4
±2.5
+5.0
Full
Full
−14
±13.8
−9
−57
−3
±25
ppm/°C
ppm/°C
Full
1.0
1.0
1.0
V
25°C
2.63
2.48
2.06
LSB rms
Full
−0.31
1.46
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
25°C
Full
1.70
1.94
Min
AD6674-750
Typ
Max
Temp
−0.51
1.46
2.05
1.5
2
1.22
2.44
3.2
1.22
1.22
1.22
1.7
1.25
2.50
3.3
1.25
1.25
1.25
1.8
685
595
125
16
263
200
N/A5
5
1.70
1.94
−0.3
1.46
2.05
1.5
2
1.28
2.56
3.4
1.28
1.28
1.28
3.4
721
677
142
18
292
225
6
Rev. C | Page 5 of 96
1.22
2.44
3.2
1.22
1.22
1.22
1.7
1.25
2.50
3.3
1.25
1.25
1.25
1.8
545
460
125
10
165
190
N/A5
5
2.06
2.06
2.05
1.5
2
1.28
2.56
3.4
1.28
1.28
1.28
3.4
623
572
142
17
217
258
7.0
1.22
2.44
3.2
1.22
1.22
1.22
1.7
1.25
2.50
3.3
1.25
1.25
1.25
1.8
427
398
89
10
139
182
140
5
V p-p
V
pF
GHz
1.28
2.56
3.4
1.28
1.28
1.28
3.4
466
463
100
18
183
237
7
V
V
V
V
V
V
V
mA
mA
mA
mA
mA
mA
mA
mA
AD6674
Parameter
POWER CONSUMPTION
Total Power Dissipation2
Power-Down Dissipation
Standby6
Data Sheet
Temp
Min
AD6674-1000
Typ
Max
Full
Full
Full
3.3
835
1.4
Min
3.6
AD6674-750
Typ
Max
2.8
835
1.4
AD6674-500
Min
Typ
Max
3.1
2.24
710
1.2
2.5
Unit
W
mW
W
1
Differential capacitance is measured between the VIN+x and VIN−x pins (x = A, B).
Measured with a low input frequency, full-scale sine wave.
All lanes running. Power dissipation on DRVDD changes with lane rate and number of lanes used.
4
L is the number of lanes per converter device (lanes per link).
5
N/A means not applicable. At the maximum sample rate, it is not applicable to use L = 2 mode on the JESD204B output interface because this exceeds the maximum
lane rate of 12.5 Gbps. L = 2 mode is supported when the equation ((M × N΄ × (10/8) × fOUT)/L) results in a lane rate that is ≤12.5 Gbps. fOUT is the output sample rate and
is denoted by fS/DCM, where DCM = decimation ratio.
6
Can be controlled by the SPI.
2
3
AC SPECIFICATIONS
AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, AVDD1_SR = 1.25 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified
maximum sampling rate, 1.0 V internal reference, AIN = −1.0 dBFS, clock divider = 2, default SPI settings, TA = 25°C, unless otherwise noted.
Table 2.
Parameter1
ANALOG INPUT FULL SCALE
NOISE DENSITY2
SIGNAL-TO-NOISE RATIO (SNR)3
VDR Mode (Input Mask Not Triggered)
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
NSR Enabled (21% BW Mode)4
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
NSR Enabled (28% BW Mode)4
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
Temp
Full
Full
25°C
Full
25°C
25°C
25°C
25°C
25°C
AD6674-1000
Min Typ
Max
1.7
−154
65.1
67.2
66.6
65.3
64.0
62.4
61.4
57.0
AD6674-750
Min Typ
Max
1.7
−153
65.8
67.3
67.1
66.7
66.2
64.3
63.6
59.9
AD6674-500
Min
67.8
2.06
−153
Unit
V p-p
dBFS/Hz
69.2
69.0
68.6
68.0
64.4
63.8
60.5
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
Typ
Max
25°C
25°C
25°C
25°C
25°C
25°C
25°C
73.8
73.6
73.5
71.9
69.0
68.2
63.6
74.0
73.8
73.7
72.2
71.4
71.0
66.6
75.2
75.2
74.8
74.2
70.3
69.3
65.3
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
25°C
25°C
25°C
25°C
25°C
25°C
25°C
72.4
72.2
72.1
70.5
67.0
66.3
61.9
72.8
72.6
72.5
71.0
70.0
68.9
65.1
72.4
72.4
72.1
71.9
68.3
67.7
64.1
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
Rev. C | Page 6 of 96
Data Sheet
Parameter1
SIGNAL-TO-NOISE-AND-DISTORTION RATIO (SINAD)3
VDR Mode (Input Mask Not Triggered)
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
EFFECTIVE NUMBER OF BITS (ENOB)3
VDR Mode (Input Mask Not Triggered)
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
SPURIOUS FREE DYNAMIC RANGE (SFDR),
SECOND OR THIRD HARMONIC3
VDR Mode (Input Mask Not Triggered)
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
WORST OTHER (EXCLUDING SECOND OR THIRD
HARMONIC)3
VDR Mode (Input Mask Not Triggered)
fIN = 10 MHz
fIN = 170 MHz
fIN = 340 MHz
fIN = 450 MHz
fIN = 765 MHz
fIN = 985 MHz
fIN = 1950 MHz
TWO-TONE INTERMODULATION DISTORTION (IMD)3
AIN1 AND AIN2 = −7.0 dBFS
fIN1 = 185 MHz, fIN2 = 188 MHz
fIN1 = 338 MHz, fIN2 = 341 MHz
CROSSTALK5
FULL POWER BANDWIDTH
AD6674
Temp
25°C
Full
25°C
25°C
25°C
25°C
25°C
25°C
Full
25°C
25°C
25°C
25°C
25°C
25°C
Full
25°C
25°C
25°C
25°C
25°C
25°C
Full
25°C
25°C
25°C
25°C
25°C
25°C
25°C
25°C
25°C
AD6674-1000
Min Typ
Max
65.0
10.5
75
−81
67.1
66.4
65.2
63.8
62.1
61.1
56.0
10.8
10.7
10.5
10.3
10.0
9.8
9.0
88
85
85
82
82
80
68
−95
−94
−88
−86
−81
−82
−75
−87
−88
95
2
1
AD6674-750
Min Typ
Max
65.6
10.4
75
−81
67.1
67.0
66.5
66.1
64.1
63.1
59.0
10.8
10.8
10.7
10.5
10.4
10.2
9.5
85
86
83
82
80
76
68
−95
−89
−83
−82
−85
−83
−80
−85
−83
95
2
AD6674-500
Min
67.6
10.8
80
−82
Typ
Max
69.0
68.8
68.4
67.9
64.2
63.6
60.3
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
11.2
11.1
11.1
11.0
10.4
10.3
9.7
Bits
Bits
Bits
Bits
Bits
Bits
Bits
83
88
83
81
80
75
70
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
−95
−95
−93
−93
−88
−89
−84
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
dBFS
−88
−88
95
2
dBFS
dBFS
dB
GHz
See the AN-835 Application Note, Understanding High Speed ADC Testing and Evaluation, for definitions and for details on how these tests were completed.
Noise density is measured at low analog input frequency (30 MHz).
See Table 10 for recommended device settings to achieve stated typical performance.
4
When NSR is activated on the AD6674-750 and AD6674-1000, the decimating half-band filter is also enabled.
5
Crosstalk is measured at 185 MHz with −1.0 dBFS analog input on one channel and no input on the adjacent channel.
2
3
Rev. C | Page 7 of 96
Unit
AD6674
Data Sheet
DIGITAL SPECIFICATIONS
AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, AVDD1_SR = 1.25 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified
maximum sampling rate, 1.0 V internal reference, AIN = −1.0 dBFS, clock divider = 2, default SPI settings, TA = 25°C, unless otherwise
noted.
Table 3.
Parameter
CLOCK INPUTS (CLK+, CLK−)
Logic Compliance
Differential Input Voltage
Input Common-Mode Voltage
Input Resistance (Differential)
Input Capacitance
SYSTEM REFERENCE INPUTS (SYSREF+, SYSREF−)
Logic Compliance
Differential Input Voltage
Input Common-Mode Voltage
Input Resistance (Differential)
Input Capacitance (Differential)
LOGIC INPUTS (SDIO, SCLK, CSB, PDWN/STBY)
Logic Compliance
Logic 1 Voltage
Logic 0 Voltage
Input Resistance
LOGIC OUTPUT (SDIO)
Logic Compliance
Logic 1 Voltage (IOH = 800 μA)
Logic 0 Voltage (IOL = 50 μA)
SYNC INPUTS (SYNCINB+, SYNCINB–)
Logic Compliance
Differential Input Voltage
Input Common-Mode Voltage
Input Resistance (Differential)
Input Capacitance
LOGIC OUTPUTS (FD_A, FD_B)
Logic Compliance
Logic 1 Voltage
Logic 0 Voltage
Input Resistance
DIGITAL OUTPUTS (SERDOUTx±, x = 0 TO 3)
Logic Compliance
Differential Output Voltage
Output Common-Mode Voltage (VCM)
AC-Coupled
Short-Circuit Current (IDSHORT)
Differential Return Loss (RLDIFF)1
Common-Mode Return Loss (RLCM)1
Differential Termination Impedance
1
Temp
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Min
600
Typ
LVDS/LVPECL
1200
0.85
35
Max
Unit
1800
mV p-p
V
kΩ
pF
2.5
400
0.6
LVDS/LVPECL
1200
0.85
35
1800
2.0
2.5
mV p-p
V
kΩ
pF
CMOS
0.8 × SPIVDD
0
V
V
kΩ
30
CMOS
0.8 × SPIVDD
0
400
0.6
V
V
LVDS/LVPECL/CMOS
1200
0.85
35
1800
2.0
2.5
mV p-p
V
kΩ
pF
CMOS
0.8 × SPIVDD
0
V
V
kΩ
30
Full
Full
360
CML
770
mV p-p
25°C
25°C
25°C
25°C
Full
0
−100
8
6
80
1.8
+100
V
mA
dB
dB
Ω
Differential and common-mode return loss is measured from 100 MHz to 0.75 × baud rate.
Rev. C | Page 8 of 96
100
120
Data Sheet
AD6674
SWITCHING SPECIFICATIONS
AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, AVDD1_SR = 1.25 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified
maximum sampling rate, 1.0 V internal reference, AIN = −1.0 dBFS, clock divider = 2, default SPI settings, TA = 25°C, unless otherwise noted.
Table 4.
Parameter
CLOCK
Clock Rate (at CLK+/CLK− Pins)
Maximum Sample Rate1
Minimum Sample Rate2
Clock Pulse Width High
Clock Pulse Width Low
OUTPUT PARAMETERS
Unit Interval (UI)3
Rise Time (tR) (20% to 80% into 100 Ω
Load)
Fall Time (tF) (20% to 80% into 100 Ω
Load)
PLL Lock Time
Data Rate per Channel (NRZ)4
LATENCY
Pipeline Latency
Fast Detect Latency
Wake-Up Time (Standby)5
Wake-Up Time (Power-Down)5
APERTURE
Aperture Delay (tA)
Aperture Uncertainty (Jitter, tJ)
Out-of-Range Recovery Time
Temp
AD6674-1000
Min
Typ Max
Min
Full
Full
Full
Full
Full
0.3
1000
300
500
500
0.3
750
300
666.67
666.67
4
AD6674-750
Typ
Max
4
AD6674-500
Min
Typ Max
Unit
0.3
500
300
1000
1000
GHz
MSPS
MSPS
ps
ps
4
Full
25°C
100
32
133.33
32
200
32
ps
ps
25°C
32
32
32
ps
25°C
25°C
Full
Full
25°C
25°C
Full
Full
Full
3.125
2
10
12.5
3.125
75
2
7.5
12.5
3.125
75
28
1
12.5
75
28
1
4
530
55
1
2
5
28
1
4
530
55
1
4
530
55
1
ms
Gbps
Clock cycles
Clock cycles
ms
ms
ps
fs rms
Clock cycles
1
The maximum sample rate is the clock rate after the divider.
The minimum sample rate operates at 300 MSPS with L = 2 or L = 1.
3
Baud rate = 1/UI. A subset of this range can be supported.
4
At full baud rate (12.5 Gbps), each ADC outputs data on two differential pair lanes.
5
Wake-up time is defined as the time required to return to normal operation from power-down mode or standby mode.
2
TIMING SPECIFICATIONS
Table 5.
Parameter
CLK± to SYSREF± TIMING REQUIREMENTS
tSU_SR
tH_SR
SPI TIMING REQUIREMENTS
tDS
tDH
tCLK
tS
tH
tHIGH
tLOW
tACCESS
tDIS_SDIO
Test Conditions/Comments
Device clock to SYSREF± setup time
Device clock to SYSREF± hold time
See Figure 4
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 SCLK
Setup time between CSB and SCLK
Hold time between CSB and SCLK
Minimum period that SCLK is in a logic high state
Minimum period that SCLK is in a logic low state
Maximum time delay between the falling edge of SCLK and the
output data valid for a read operation
Time required for the SDIO pin to switch from an output to an
input relative to the SCLK rising edge (not shown in Figure 4)
Rev. C | Page 9 of 96
Min
Typ
Max
117
−96
ps
ps
6
ns
ns
ns
ns
ns
ns
ns
ns
2
2
40
2
2
10
10
10
Unit
10
ns
AD6674
Data Sheet
Timing Diagrams
APERTURE
DELAY
ANALOG
INPUT
SIGNAL
N – 55
SAMPLE N
N+1
N – 54
N – 53
N – 52
N–1
N – 51
CLK–
CLK+
CLK–
CLK+
SERDOUT0–
SERDOUT0+
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
CONVERTER0 MSB
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
CONVERTER0 LSB
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
CONVERTER1 MSB
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
A
B
C
D
E
F
G
H
I
J
CONVERTER1 LSB
SERDOUT1–
SERDOUT1+
SERDOUT2–
SERDOUT2+
SAMPLE N – 55
ENCODED INTO 1
8-BIT/10-BIT SYMBOL
SAMPLE N – 54
ENCODED INTO 1
8-BIT/10-BIT SYMBOL
12400-002
SERDOUT3–
SERDOUT3+
SAMPLE N – 53
ENCODED INTO 1
8-BIT/10-BIT SYMBOL
Figure 2. Data Output Timing (VDR Mode; L = 4; M = 2; F = 1)
CLK–
CLK+
tSU_SR
tH_SR
12400-003
SYSREF–
SYSREF+
Figure 3. SYSREF± Setup and Hold Timing
tHIGH
tDS
tS
tCLK
tDH
tH
tLOW
CSB
SDIO DON’T CARE
DON’T CARE
R/W
A14
A13
A12
A11
A10
A9
A8
A7
D5
Figure 4. Serial Port Interface Timing Diagram
Rev. C | Page 10 of 96
D4
D3
D2
D1
D0
DON’T CARE
12400-004
SCLK DON’T CARE
Data Sheet
AD6674
ABSOLUTE MAXIMUM RATINGS
THERMAL CHARACTERISTICS
Table 6.
Parameter
Electrical
AVDD1 to AGND
AVDD1_SR to AGND
AVDD2 to AGND
AVDD3 to AGND
DVDD to DGND
DRVDD to DRGND
SPIVDD to AGND
AGND to DRGND
VIN±x to AGND
SCLK, SDIO, CSB to AGND
PDWN/STBY to AGND
Operating Temperature Range
Junction Temperature Range
Storage Temperature Range
(Ambient)
Rating
1.32 V
1.32 V
2.75 V
3.63 V
1.32 V
1.32 V
3.63 V
−0.3 V to +0.3 V
3.2 V
−0.3 V to SPIVDD + 0.3 V
−0.3 V to SPIVDD + 0.3 V
−40°C to +85°C
−40°C to +115°C
−60°C to +150°C
Stresses at or above those listed under Absolute Maximum
Ratings may cause permanent damage to the product. This is a
stress rating only; functional operation of the product at these
or any other conditions above those indicated in the operational
section of this specification is not implied. Operation beyond
the maximum operating conditions for extended periods may
affect product reliability.
Typical θJA, ΨJB, and θJC are specified vs. the number of printed
circuit board (PCB) layers in different airflow velocities (in
m/sec). Airflow increases heat dissipation, effectively reducing
θJA and ΨJB. In addition, metal in direct contact with the package
leads and exposed pad from metal traces, through holes, ground,
and power planes reduces the θJA. Thermal performance for
actual applications requires careful inspection of the conditions
in an application. The use of appropriate thermal management
techniques is recommended to ensure that the maximum
junction temperature does not exceed the limits shown in Table 6.
Table 7. Thermal Resistance Values
PCB
Type
JEDEC
2s2p
Board
Airflow
Velocity
(m/sec)
0.0
1.0
2.5
1
θJA
17.81, 2
15.61, 2
15.01, 2
ΨJB
6.31, 3
5.91, 3
5.71, 3
θJC_TOP
4.71, 5
N/A4
N/A4
Per JEDEC 51-7, plus JEDEC 51-5 2s2p test board.
Per JEDEC JESD51-2 (still air) or JEDEC JESD51-6 (moving air).
Per JEDEC JESD51-8 (still air).
4
N/A means not applicable.
5
Per MIL-STD 883, Method 1012.1.
2
3
ESD CAUTION
Rev. C | Page 11 of 96
θJC_BOT
1.21, 5
Unit
°C/W
°C/W
°C/W
AD6674
Data Sheet
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
AVDD1
AVDD2
AVDD2
AVDD1
AGND
SYSREF–
SYSREF+
AVDD1_SR
AGND
AVDD1
CLK–
CLK+
AVDD1
AVDD2
AVDD2
AVDD1
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
AD6674
TOP VIEW
(Not to Scale)
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
AVDD1
AVDD1
AVDD2
AVDD3
VIN–B
VIN+B
AVDD3
AVDD2
AVDD2
AVDD2
SPIVDD
CSB
SCLK
SDIO
DVDD
DGND
NOTES
1. EXPOSED PAD. THE EXPOSED THERMAL PAD ON THE BOTTOM OF THE
PACKAGE PROVIDES THE GROUND REFENCE FOR AVDDx. THIS EXPOSED
PAD MUST BE CONNECTED TO GROUND FOR PROPER OPERATION.
12400-005
FD_A
DRGND
DRVDD
SYNCINB–
SYNCINB+
SERDOUT0–
SERDOUT0+
SERDOUT1–
SERDOUT1+
SERDOUT2–
SERDOUT2+
SERDOUT3–
SERDOUT3+
DRVDD
DRGND
FD_B
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
AVDD1
AVDD1
AVDD2
AVDD3
VIN–A
VIN+A
AVDD3
AVDD2
AVDD2
AVDD2
AVDD2
V_1P0
SPIVDD
PDWN/STBY
DVDD
DGND
Figure 5. Pin Configuration
Table 8. Pin Function Descriptions
Pin No.
Power Supplies
0
1, 2, 47, 48, 49,
52, 55, 61, 64
3, 8, 9, 10, 11,
39, 40, 41, 46,
50, 51, 62, 63
4, 7, 42, 45
13, 38
15, 34
16, 33
18, 31
19, 30
56, 60
57
Analog
5, 6
12
43, 44
53, 54
Mnemonic
Type
Description
EPAD
Ground
AVDD1
Supply
Exposed Pad. The exposed thermal pad on the bottom of the package provides the
ground reference for AVDDx. This exposed pad must be connected to ground for
proper operation. See the Applications Information section for more details.
Analog Power Supply (1.25 V Nominal).
AVDD2
Supply
Analog Power Supply (2.5 V Nominal).
AVDD3
SPIVDD
DVDD
DGND
DRGND
DRVDD
AGND1
AVDD1_SR1
Supply
Supply
Supply
Ground
Ground
Supply
Ground
Supply
Analog Power Supply (3.3 V Nominal).
Digital Power Supply for SPI (1.7 V to 3.4 V).
Digital Power Supply (1.25 V Nominal).
Ground Reference for DVDD.
Ground Reference for DRVDD.
Digital Driver Power Supply (1.25 V Nominal).
Ground Reference for SYSREF±.
Analog Power Supply for SYSREF± (1.25 V Nominal).
VIN−A, VIN+A
V_1P0
Input
Input/DNC
VIN+B, VIN−B
CLK+, CLK−
Input
Input
ADC A Analog Input Complement/True.
1.0 V Reference Voltage Input/Do Not Connect. This pin is configurable through the
SPI as a no connect or an input. Do not connect this pin if using the internal
reference. This pin requires a 1.0 V reference voltage input if using an external
voltage reference source.
ADC B Analog Input True/Complement.
Clock Input True/Complement.
Rev. C | Page 12 of 96
Data Sheet
Pin No.
CMOS Outputs
17, 32
Digital Inputs
20, 21
58, 59
Data Outputs
22, 23
24, 25
26, 27
28, 29
Device Under
Test (DUT)
Controls
14
35
36
37
1
AD6674
Mnemonic
Type
Description
FD_A, FD_B
Output
Fast Detect Outputs for Channel A and Channel B.
SYNCINB−,
SYNCINB+
SYSREF+,
SYSREF−
Input
Active Low JESD204B LVDS Sync Input True/Complement.
Input
Active Low JESD204B LVDS System Reference Input True/Complement.
SERDOUT0−,
SERDOUT0+
SERDOUT1−,
SERDOUT1+
SERDOUT2−,
SERDOUT2+
SERDOUT3−,
SERDOUT3+
Output
Lane 0 Output Data Complement/True.
Output
Lane 1 Output Data Complement/True.
Output
Lane 2 Output Data Complement/True.
Output
Lane 3 Output Data Complement/True.
PDWN/STBY
Input
SDIO
SCLK
CSB
Input/Output
Input
Input
Power-Down Input (Active High). The operation of this pin depends on the SPI
mode and can be configured as power-down or standby. This pin requires an
external 10 kΩ pull-down resistor.
SPI Serial Data Input/Output.
SPI Serial Clock.
SPI Chip Select (Active Low).
To ensure proper ADC operation, connect AVDD1_SR and AGND separately from the AVDD1 and EPAD connection. For more information, see the Applications
Information section.
Rev. C | Page 13 of 96
AD6674
Data Sheet
TYPICAL PERFORMANCE CHARACTERISTICS
AD6674-1000
AVDD1 = 1.25 V, AVDD1_SR = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V,
AIN = −1.0 dBFS, VDR mode (no violation of VDR mask), clock divider = 2, otherwise default SPI settings, TA = 25°C, 128k FFT sample,
unless otherwise noted. See Table 10 for recommended settings.
AIN = –1dBFS
SNR = 67.2dBFS
ENOB = 10.8 BITS
SFDR = 88dBFS
BUFFER CONTROL 1 = 1.5×
–10
–30
AMPLITUDE (dBFS)
–50
–70
–90
–110
–50
–70
–90
0
100
200
300
400
500
FREQUENCY (MHz)
–130
12400-100
–130
0
500
AIN = –1dBFS
SNR = 61.5dBFS
ENOB = 10.1 BITS
SFDR = 82dBFS
BUFFER CONTROL 1 = 6.0×
–20
AMPLITUDE (dBFS)
–50
–70
–90
–40
–60
–80
–100
0
100
200
300
400
500
FREQUENCY (MHz)
–120
12400-101
–130
0
300
400
500
Figure 10. Single Tone FFT with fIN = 765.3 MHz
0
AIN = –1dBFS
SNR = 65.3dBFS
ENOB = 10.5 BITS
SFDR = 85dBFS
BUFFER CONTROL 1 = 3.0×
AIN = –1dBFS
SNR = 60.5dBFS
ENOB = 9.9 BITS
SFDR = 80dBFS
BUFFER CONTROL 1 = 6.0×
–20
AMPLITUDE (dBFS)
–30
200
FREQUENCY (MHz)
Figure 7. Single Tone FFT with fIN = 170.3 MHz
–10
100
12400-104
AMPLITUDE (dBFS)
400
0
–110
–50
–70
–90
–40
–60
–80
–100
–110
–130
0
100
200
300
400
FREQUENCY (MHz)
500
12400-102
AMPLITUDE (dBFS)
300
Figure 9. Single Tone FFT with fIN = 450.3 MHz
AIN = –1dBFS
SNR = 66.6dBFS
ENOB = 10.7 BITS
SFDR = 85dBFS
BUFFER CONTROL 1 = 3.0×
–30
200
FREQUENCY (MHz)
Figure 6. Single Tone FFT with fIN = 10.3 MHz
–10
100
12400-103
–110
Figure 8. Single Tone FFT with fIN = 340.3 MHz
–120
0
100
200
300
400
FREQUENCY (MHz)
Figure 11. Single Tone FFT with fIN = 985.3 MHz
Rev. C | Page 14 of 96
500
12400-105
AMPLITUDE (dBFS)
–30
AIN = –1dBFS
SNR = 64.0dBFS
ENOB = 10.3 BITS
SFDR = 82dBFS
BUFFER CONTROL 1 = 3.0×
–10
Data Sheet
0
0
AIN = –1dBFS
SNR = 59.8BFS
ENOB = 9.6 BITS
SFDR = 79dBFS
BUFFER CONTROL 1 = 8.0×
AMPLITUDE (dBFS)
–20
–40
–60
–80
–100
–60
–80
–100
–120
0
100
200
300
400
500
FREQUENCY (MHz)
–140
12400-107
–120
0
0
200
300
400
500
Figure 15. LTE-FDD 10 MHz Channel FFT with fIN = 230 MHz
90
AIN = –1dBFS
SNR = 57.7dBFS
ENOB = 9.2 BITS
SFDR = 70dBFS
BUFFER CONTROL 1 = 8.0×
–20
100
FREQUENCY (MHz)
Figure 12. Single Tone FFT with fIN = 1293.3 MHz
85
SFDR (dBFS)
–40
SNR/SFDR (dBFS)
AMPLITUDE (dBFS)
AIN = –20dBFS
SNR = 66.5dBFS
–20 ENOB = 10.8 BITS
SFDR = 99dBFS
BUFFER CONTROL 1 = 3.0×
–40
12400-600
AMPLITUDE (dBFS)
AD6674
–60
–80
80
75
70
SNR (dBFS)
–100
0
100
200
300
400
500
FREQUENCY (MHz)
60
700
12400-108
–120
800
850
900
950
1000
1050
1100
SAMPLE RATE (MHz)
Figure 13. Single Tone FFT with fIN = 1725.3 MHz
Figure 16. SNR/SFDR vs. Sample Rate (fS), fIN = 170.3 MHz;
Buffer Control 1 = 3.0×
0
90
AIN = –1dBFS
SNR = 57.0dBFS
ENOB = 9.1 BITS
SFDR = 69dBFS
BUFFER CURRENT = 6.0×
1.5×, SFDR
80
3.0×, SFDR
70
SNR/SFDR (dBFS)
–20
–40
–60
–80
60
50
3.0×, SNR
40
30
1.5×, SNR
20
–100
–120
0
100
200
300
FREQUENCY (MHz)
400
500
Figure 14. Single Tone FFT with fIN = 1950.3 MHz
0
10.3
63.3
100.3 170.3 225.3 302.3 341.3 403.3 453.3 502.3
ANALOG INPUT FREQUENCY (MHz)
Figure 17. SNR/SFDR vs. Analog Input Frequency (fIN);
fIN < 500 MHz; Buffer Control 1 = 1.5× and 3.0×
Rev. C | Page 15 of 96
12400-203
10
12400-109
AMPLITUDE (dBFS)
750
12400-201
65
AD6674
Data Sheet
0
SFDR/IMD3 (dBc AND dBFS)
–20
–40
–60
–80
–100
–20
SFDR (dBc)
–40
IMD3 (dBc)
–60
–80
SFDR (dBFS)
–100
–120
0
100
200
300
400
500
FREQUENCY (MHz)
12400-205
–120
IMD3 (dBFS)
–140
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
INPUT AMPLITUDE (dBFS)
Figure 18. Two-Tone FFT; fIN1 = 184 MHz, fIN2 = 187 MHz
0
20
AIN1 AND AIN2 = –7dBFS
SFDR = 88dBFS
IMD2 = 93dBFS
IMD3 = 88dBFS
BUFFER CONTROL 1 = 4.5×
0
SNR/SFDR (dBc AND dBFS)
–20
–40
–60
–80
–20
SFDR (dBc)
–40
IMD3 (dBc)
–60
–80
–100
SFDR (dBFS)
–100
–120
–120
0
100
200
300
400
FREQUENCY (MHz)
Figure 19. Two-Tone FFT; fIN1 = 338 MHz, fIN2 = 341 MHz
IMD3 (dBFS)
–140
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
500
12400-206
AMPLITUDE (dBFS)
Figure 20. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN)
with fIN1 = 184 MHz and fIN2 = 187 MHz
INPUT AMPLITUDE (dBFS)
12400-208
AMPLITUDE (dBFS)
20
AIN1 AND AIN2 = –7dBFS
SFDR = 87dBFS
IMD2 = 93dBFS
IMD3 = 87dBFS
BUFFER CONTROL 1 = 3.0×
12400-207
0
Figure 21. Two-Tone IMD3/SFDR vs. Input Amplitude (AIN) with fIN1 = 338 MHz
and fIN2 = 341 MHz
Rev. C | Page 16 of 96
Data Sheet
AD6674
3.5
110
100
SFDR (dBFS)
90
3.4
80
SNR (dBFS)
3.3
60
50
40
POWER (W)
SNR/SFDR (dB)
70
SFDR (dBc)
SNR (dBc)
30
L=4
M=2
F=1
3.2
3.1
20
10
3.0
0
90
SFDR
SNR
50
40
30
20
10
0
–50 –40 –30 –20 –10
0
10
20
30
40
50
60
70
80
TEMPERATURE (°C)
90
12400-210
SNR/SFDR (dBFS)
70
60
Figure 23. SNR/SFDR vs. Temperature, fIN = 170.3 MHz
Rev. C | Page 17 of 96
1080
12400-524
1060
1040
1020
980
960
940
920
900
880
860
840
820
800
780
760
740
SAMPLE RATE (MSPS)
Figure 24. Power Dissipation vs. Sampel Rate (fS) (Default SPI)
Figure 22. SNR/SFDR vs. Input Amplitude (AIN), fIN = 170.3 MHz
80
1000
INPUT AMPLITUDE (dBFS)
2.9
720
0
700
–20
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
12400-209
–10
AD6674
Data Sheet
AD6674-750
AVDD1 = 1.25 V, AVDD1_SR = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V,
AIN = −1.0 dBFS, VDR mode (no violation of VDR mask), clock divider = 2, otherwise default SPI settings, TA = 25°C, 128k FFT sample,
unless otherwise noted. See Table 10 for recommended settings.
0
–20
–40
–60
–80
–100
–40
–60
–80
–100
–120
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
–140
12400-219
0
0
FREQUENCY (MHz)
Figure 25. Single Tone FFT with fIN = 10.3 MHz
0
0
–40
–60
–80
–100
–40
–60
–80
–100
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
–140
12400-220
0
0
FREQUENCY (MHz)
Figure 26. Single Tone FFT with fIN = 170.3 MHz
0
0
AMPLITUDE (dBFS)
–40
–60
–80
–100
–40
–60
–80
–100
0
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
Figure 27. Single Tone FFT with fIN = 340.3 MHz
–140
0
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
Figure 30. Single Tone FFT with fIN = 985.3 MHz
Rev. C | Page 18 of 96
12400-230
–120
–120
–140
AIN = −1dBFS
SNR = 63.5dBFS
ENOB = 10.2 BITS
SFDR = 76dBFS
BUFFER CONTROL 1 = 8.5×
–20
12400-221
AMPLITUDE (dBFS)
Figure 29. Single Tone FFT with fIN = 765.3 MHz
AIN = −1dBFS
SNR = 66.7dBFS
ENOB = 10.6 BITS
SFDR = 83dBFS
BUFFER CONTROL 1 = 3.0×
–20
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
12400-229
–120
–120
–140
AIN = −1dBFS
SNR = 64.2dBFS
ENOB = 10.3 BITS
SFDR = 80dBFS
BUFFER CONTROL 1 = 8.5×
–20
AMPLITUDE (dBFS)
AMPLITUDE (dBFS)
Figure 28. Single Tone FFT with fIN = 450.3 MHz
AIN = −1dBFS
SNR = 67.1dBFS
ENOB = 10.7 BITS
SFDR = 86dBFS
BUFFER CONTROL 1 = 2.0×
–20
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
12400-222
–120
–140
AIN = −1dBFS
SNR = 66.2dBFS
ENOB = 10.5 BITS
SFDR = 82dBFS
BUFFER CONTROL 1 = 4.0×
–20
AMPLITUDE (dBFS)
AMPLITUDE (dBFS)
0
AIN = −1dBFS
SNR = 67.3dBFS
ENOB = 10.7 BITS
SFDR = 85dBFS
BUFFER CONTROL 1 = 1.5×
Data Sheet
AD6674
0
–20
–40
AMPLITUDE (dBFS)
–60
–80
–100
–120
0
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
–80
–100
–140
0
0
200
300
400
500
Figure 34. LTE-FDD 10 MHz Channel FFT with fIN = 230 MHz
95
AIN = −1dBFS
SNR = 60.5dBFS
ENOB = 9.6 BITS
SFDR = 71dBFS
BUFFER CONTROL 1 = 8.5×
–20
100
FREQUENCY (MHz)
Figure 31. Single Tone FFT with fIN = 1310.3 MHz
90
–40
SNR/SFDR (dBFS)
AMPLITUDE (dBFS)
–60
–120
12400-231
–140
AIN = –20dBFS
SNR = 67.4dBFS
–20 ENOB = 10.9 BITS
SFDR = 96dBFS
BUFFER CONTROL 1 = 3.0×
–40
12400-601
AMPLITUDE (dBFS)
0
AIN = −1dBFS
SNR = 62.3dBFS
ENOB = 9.8 BITS
SFDR = 68dBFS
BUFFER CONTROL 1 = 8.5×
–60
–80
–100
–120
85
SFDR
80
75
70
FREQUENCY (MHz)
65
525
550
575
600
625
650
675
700
725
750
775
800
12400-223
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
500
12400-225
0
12400-232
SNR
–140
SAMPLE RATE (MSPS)
Figure 35. SNR/SFDR vs. Sample Rate (fS); fIN = 170.3 MHz,
Buffer Control 1 = 3.0×
Figure 32. Single Tone FFT with fIN = 1710.3 MHz
0
–20
95
90
SNR/SFDR (dBFS)
–40
–60
–80
–100
SFDR
85
80
75
70
SNR
–120
65
–140
0
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
12400-233
AMPLITUDE (dBFS)
100
AIN = −1dBFS
SNR = 59.8dBFS
ENOB = 9.5 BITS
SFDR = 68dBFS
BUFFER CONTROL 1 = 8.5×
60
0
50
100
150
200
250
300
350
400
450
FREQUENCY (MHz)
Figure 33. Single Tone FFT with fIN = 1950.3 MHz
Figure 36. SNR/SFDR vs. Analog Input Frequency (fIN);
fIN < 500 MHz; Buffer Control 1 = 3.0×
Rev. C | Page 19 of 96
AD6674
Data Sheet
0
–20
SFDR/IMD3 (dBc AND dBFS)
–20
–40
–60
–80
–100
FREQUENCY (MHz)
0
–80
SFDR (dBFS)
IMD3 (dBFS)
–120
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
INPUT AMPLITUDE (dBFS)
Figure 39. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN)
with fIN1 = 184 MHz and fIN2 = 187 MHz
AIN1 AND AIN2 = −7dBFS
SFDR = 83dBFS
IMD2 = 89dBc
IMD3 = 83dBc
BUFFER CONTROL 1 = 4.5×
0
SFDR/IMD3 (dBc AND dBFS)
–20
–40
–60
–80
–100
–120
SFDR (dBc)
–20
–40
IMD3 (dBc)
–60
–80
–100
SFDR (dBFS)
–120
IMD3 (dBFS)
0
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
FREQUENCY (MHz)
12400-227
AMPLITUDE (dBFS)
IMD3 (dBc)
–60
12400-428
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375
12400-226
0
Figure 37. Two-Tone FFT; fIN1 = 184 MHz, fIN2 = 187 MHz
–140
–40
–100
–120
–140
SFDR (dBc)
Figure 38. Two-Tone FFT; fIN1 = 338 MHz, fIN2 = 341 MHz
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
INPUT AMPLITUDE (dBFS)
Figure 40. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN)
with fIN1 = 338 MHz and fIN2 = 341 MHz
Rev. C | Page 20 of 96
12400-429
AMPLITUDE (dBFS)
0
AIN1 AND AIN2 = −7dBFS
SFDR = 81dBFS
IMD2 = 86dBc
IMD3 = 81dBc
BUFFER CONTROL 1 = 3.0×
Data Sheet
AD6674
120
3.0
SFDR (dBFS)
2.9
2.8
80
POWER (W)
SNR (dBFS)
60
SFDR (dBc)
40
2.6
2.5
SNR (dBc)
20
2.4
2.3
500
0
INPUT AMPLITUDE (dBFS)
12400-430
–5
–10
–15
–20
–25
–30
–35
–40
–45
–50
–55
–60
–65
–70
–75
–80
–85
–90
0
95
90
85
SFDR
80
75
70
–15
10
35
60
TEMPERATURE (°C)
85
12400-228
SNR
65
–40
550
600
650
700
750
SAMPLE RATE (MSPS)
800
850
Figure 43. Power Dissipation vs. Sample Rate (fS); L = 4, M = 2, F = 1 for
fS ≥ 625 MSPS and L= 2, M = 2, F = 2 for fS < 625 MSPS (Default SPI)
Figure 41. SNR/SFDR vs. Input Amplitude (AIN), fIN = 170.3 MHz
SNR/SFDR (dBFS)
2.7
12400-234
SNR/SFDR (dBc AND dBFS)
100
Figure 42. SNR/SFDR vs. Temperature, fIN = 170.3 MHz
Rev. C | Page 21 of 96
AD6674
Data Sheet
AD6674-500
AVDD1 = 1.25 V, AVDD1_SR = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V,
AIN = −1.0 dBFS, VDR mode (no violation of VDR mask), clock divider = 2, otherwise default SPI settings, TA = 25°C, 128k FFT sample,
unless otherwise noted. See Table 10 for recommended settings.
0
0
AIN = −1dBFS
SNR = 68.9dBFS
ENOB = 10.9 BITS
SFDR = 83dBFS
BUFFER CONTROL 1 = 2.0×
–20
–20
–40
–40
AMPLITUDE (dBFS)
–60
–80
–100
–120
–60
–80
–100
0
25
50
75
100
125
150
175
200
225
250
FREQUENCY (MHz)
–140
12400-432
–140
0
75
100
125
150
175
200
225
250
225
250
225
250
Figure 47. Single Tone FFT with fIN = 450.3 MHz
0
0
AIN = −1dBFS
SNR = 68.9dBFS
ENOB = 11.0 BITS
SFDR = 88dBFS
BUFFER CONTROL 1 = 2.0×
–20
AIN = −1dBFS
SNR = 64.7dBFS
ENOB = 10.4 BITS
SFDR = 80dBFS
BUFFER CONTROL 1 = 5.0×
–20
–40
–40
AMPLITUDE (dBFS)
–60
–80
–100
–120
–60
–80
–100
–120
0
25
50
75
100
125
150
175
200
225
250
FREQUENCY (MHz)
–140
12400-433
–140
0
25
50
75
100
125
150
175
200
FREQUENCY (MHz)
Figure 45. Single Tone FFT with fIN = 170.3 MHz
12400-236
AMPLITUDE (dBFS)
50
FREQUENCY (MHz)
Figure 44. Single Tone FFT with fIN = 10.3 MHz
Figure 48. Single Tone FFT with fIN = 765.3 MHz
0
0
AIN = −1dBFS
SNR = 68.5dBFS
ENOB = 10.9 BITS
SFDR = 83dBFS
BUFFER CONTROL 1 = 4.5×
–20
AIN = −1dBFS
SNR = 64.0dBFS
ENOB = 10.3 BITS
SFDR = 76dBFS
BUFFER CONTROL 1 = 5.0×
–20
–40
AMPLITUDE (dBFS)
–40
–60
–80
–100
–120
–60
–80
–100
–120
–140
0
25
50
75
100
125
150
175
200
FREQUENCY (MHz)
225
250
12400-434
AMPLITUDE (dBFS)
25
12400-235
–120
Figure 46. Single Tone FFT with fIN = 340.3 MHz
–140
0
25
50
75
100
125
150
175
200
FREQUENCY (MHz)
Figure 49. Single Tone FFT with fIN = 985.3 MHz
Rev. C | Page 22 of 96
12400-237
AMPLITUDE (dBFS)
AIN = −1dBFS
SNR = 67.8dBFS
ENOB = 10.8 BITS
SFDR = 83dBFS
BUFFER CONTROL 1 = 4.5×
Data Sheet
AD6674
0
0
AIN = −1dBFS
SNR = 63.0dBFS
ENOB = 10.0 BITS
SFDR = 69dBFS
BUFFER CONTROL 1 = 8.0×
–20
–40
AMPLITUDE (dBFS)
–60
–80
–100
–120
–60
–80
–100
0
25
50
75
100
125
150
175
200
225
250
FREQUENCY (MHz)
–140
12400-238
–140
0
25
50
75
100
125
150
175
200
225
Figure 50. Single Tone FFT with fIN = 1310.3 MHz
Figure 53. LTE-TDD 10 MHz Channel FFT with fIN = 230 MHz
0
90
AIN = −1dBFS
SNR = 61.5dBFS
ENOB = 9.8 BITS
SFDR = 69dBFS
BUFFER CONTROL 1 = 8.0×
–20
SFDR
85
SNR AND SFDR (dBFS)
–40
–60
–80
–100
80
75
70
SNR
25
50
75
100
125
150
175
200
225
250
FREQUENCY (MHz)
60
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
520
530
540
550
560
570
580
590
0
12400-239
–140
SAMPLE RATE (MSPS)
12400-442
65
–120
Figure 54. SNR/SFDR vs. Sample Rate (fS), fIN = 170.3 MHz;
Buffer Control 1 = 2.0×
Figure 51. Single Tone FFT with fIN = 1710.3 MHz
100
0
AIN = −1dBFS
SNR = 60.8dBFS
ENOB = 9.6 BITS
SFDR = 68dBFS
BUFFER CONTROL 1 = 8.0×
–20
95
90
SNR/SFDR (dBFS)
–40
–60
–80
–100
85
SFDR
80
75
SNR
70
–120
450
425
60
375
400
FREQUENCY (MHz)
250
350
225
325
200
275
300
175
250
150
225
125
175
200
100
150
75
75
50
100
125
25
50
0
0
25
–140
FREQUENCY (MHz)
Figure 52. Single Tone FFT with fIN = 1950.3 MHz
Figure 55. SNR/SFDR vs. Analog Input Frequency (fIN);
fIN < 500 MHz; Buffer Control 1 = 3.0×
Rev. C | Page 23 of 96
12400-444
65
12400-240
AMPLITUDE (dBFS)
250
FREQUENCY (MHz)
12400-602
–120
475
500
AMPLITUDE (dBFS)
–40
AMPLITUDE (dBFS)
AIN = −20dBFS
SNR = 69.3dBFS
ENOB = 11.2 BITS
SFDR = 100dBFS
BUFFER CONTROL 1 = 2.0×
–20
AD6674
Data Sheet
0
0
AIN1 AND AIN2 = –7dBFS
SFDR = 88dBFS
IMD2 = 94dBFS
IMD3 = 88dBFS
BUFFER CONTROL 1 = 2.0×
–20
–40
–60
–80
–100
50
100
150
FREQUENCY (MHz)
200
250
–80
SFDR (dBc)
Figure 56. Two-Tone FFT; fIN1 = 184 MHz, fIN2 = 187 MHz
IMD3 (dBFS)
–120
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
INPUT AMPLITUDE (dBFS)
12400-445
0
Figure 58. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN) with fIN1 = 184 MHz
and fIN2 = 187 MHz
0
0
AIN1 AND AIN2 = –7dBFS
SFDR = 88dBFS
IMD2 = 88dBFS
IMD3 = 89dBFS
BUFFER CONTROL 1 = 4.5×
–20
–40
–60
–80
–100
SFDR (dBc)
–40
IMD3 (dBFS)
–60
–80
SFDR (dBc)
0
50
100
150
200
FREQUENCY (MHz)
250
12400-446
–100
Figure 57. Two-Tone FFT; fIN1 = 338 MHz, fIN2 = 341 MHz
IMD3 (dBFS)
–120
–90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6
INPUT AMPLITUDE (dBFS)
Figure 59. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN)
with fIN1 = 338 MHz and fIN2 = 341 MHz
Rev. C | Page 24 of 96
12400-448
SFDR/IMD3 (dBc and dBFS)
–20
AMPLITUDE (dBFS)
IMD3 (dBFS)
–60
–100
–120
–120
SFDR (dBc)
–40
12400-447
SFDR/IMD3 (dBc and dBFS)
AMPLITUDE (dBFS)
–20
Data Sheet
AD6674
110
2.40
100
2.35
SFDR (dBFS)
2.30
80
SNR (dBFS)
70
2.25
60
50
POWER (W)
SFDR (dBc)
40
30
SNR (dBc)
20
L=4
M=2
F=1
2.15
2.10
L=2
M=2
F=2
2.05
10
2.00
0
Figure 60. SNR/SFDR vs. Input Amplitude (AIN), fIN = 170.3 MHz
SFDR
85
80
75
65
–40
–15
10
35
60
TEMPERATURE (°C)
85
12400-450
SNR
70
580
560
540
520
500
480
460
440
420
400
380
360
340
SAMPLE RATE (MSPS)
Figure 62. Power Dissipation vs. Sample Rate (fS) (Default SPI)
95
90
1.90
320
0
INPUT AMPLITUDE (dBFS)
12400-449
–5
–10
–15
–20
–25
–30
–35
–40
–45
–50
–55
–60
–65
–70
–75
–80
–85
–90
–20
Figure 61. SNR/SFDR vs. Temperature, fIN = 170.3 MHz
Rev. C | Page 25 of 96
12400-451
1.95
–10
SNR/SFDR (dBFS)
2.20
300
SNR/SFDR (dBc and dBFS)
90
AD6674
Data Sheet
EQUIVALENT CIRCUITS
AVDD3
DVDD
AVDD3
SYNCINB+
1kΩ
LEVEL
TRANSLATOR
VCM
BUFFER
VCM
1kΩ
AVDD3
DGND
AIN
CONTROL
(SPI)
3pF 1.5pF
Figure 67. SYNCINB± Inputs
Figure 63. Analog Inputs
SPIVDD
AVDD1
25Ω
CLK+
SYNCINB± PIN
CONTROL (SPI)
12400-011
VIN–x
ESD
PROTECTED
SPIVDD
1kΩ
SCLK
30kΩ
AVDD1
20kΩ
VCM = 0.85V
12400-012
20kΩ
12400-016
ESD
PROTECTED
25Ω
CLK–
Figure 64. Clock Inputs
Figure 68. SCLK Inputs
AVDD1_SR
SPIVDD
1kΩ
ESD
PROTECTED
20kΩ
CSB
LEVEL
TRANSLATOR
AVDD1_SR
VCM = 0.85V
ESD
PROTECTED
20kΩ
1kΩ
12400-013
SYSREF–
30kΩ
1kΩ
12400-017
SYSREF+
VCM = 0.85V
20kΩ
DVDD
SYNCINB–
AVDD3
20kΩ
12400-015
200Ω
67Ω
28Ω
200Ω
400Ω
10pF
DGND
AVDD3
3pF 1.5pF
200Ω
67Ω
200Ω
28Ω
VIN+x
Figure 65. SYSREF± Inputs
Figure 69. CSB Input
SPIVDD
EMPHASIS/SWING
CONTROL (SPI)
ESD
PROTECTED
DRVDD
SERDOUTx+
x = 0, 1, 2, 3
DRGND
ESD
PROTECTED
DRVDD
DATA–
SERDOUTx–
x = 0, 1, 2, 3
DRGND
SDI
30kΩ
12400-014
OUTPUT
DRIVER
SDIO
SPIVDD
12400-018
DATA+
SDO
1kΩ
Figure 66. Digital Outputs
Figure 70. SDIO
Rev. C | Page 26 of 96
Data Sheet
AD6674
SPIVDD
AVDD2
ESD
PROTECTED
ESD
PROTECTED
FD
JESD LMFC
FD_A/FD_B
V_1P0
TEMPERATURE DIODE
(FD_A ONLY)
ESD
PROTECTED
12400-019
ESD
PROTECTED
FD_x PIN CONTROL (SPI)
V_1P0 PIN
CONTROL (SPI)
Figure 71. FD_A/FD_B Outputs
Figure 73. V_1P0 Input/Output
SPIVDD
ESD
PROTECTED
ESD
PROTECTED
1kΩ
PDWN
CONTROL (SPI)
12400-020
PDWN/
STBY
Figure 72. PDWN/STBY Input
Rev. C | Page 27 of 96
12400-021
JESD SYNC~
AD6674
Data Sheet
THEORY OF OPERATION
The dual ADC cores feature a multistage, differential pipelined
architecture with integrated output error correction logic. Each
ADC features wide bandwidth inputs supporting a variety of
user-selectable input ranges. An integrated voltage reference
eases design considerations.
The AD6674 has several functions that simplify the AGC
function in a communications receiver. The programmable
threshold detector allows monitoring of the incoming signal
power using the fast detect bits of the ADC output data stream,
which are enabled and programmed via Register 0x245 through
Register 0x24C. If the input signal level exceeds the programmable
threshold, the fast detect indicator goes high. Because this
threshold indicator has low latency, the user can quickly lower
the system gain to avoid an overrange condition at the ADC
input.
The Subclass 1 JESD204B-based high speed serialized output
data rate can be configured in one-lane (L = 1) and two-lane
(L = 2) configurations depending upon the sample rate and the
decimation ratio. Multidevice synchronization is supported
through the SYSREF± and SYNCINB± input pins.
ADC ARCHITECTURE
The architecture consists of an input buffered pipelined ADC.
The input buffer is designed to provide a termination impedance to the analog input signal. This termination impedance
can be changed using the SPI to meet the termination needs
of the driver/amplifier. The default termination value is set to
400 Ω. The equivalent circuit diagram of the analog input
termination is shown in Figure 63. The input buffer is
optimized for high linearity, low noise, and low power.
The input buffer provides a linear high input impedance (for
ease of drive) and reduces the kickback from the ADC. The
quantized outputs from each stage are combined into a final
16-bit result in the digital correction logic. The pipelined
architecture permits the first stage to operate with a new input
sample while the remaining stages operate with preceding
samples. Sampling occurs on the rising edge of the clock.
ANALOG INPUT CONSIDERATIONS
driving source. In addition, low Q inductors or ferrite beads can
be placed on each section of the input to reduce high differential capacitance at the analog inputs and, thus, achieve the
maximum bandwidth of the ADC. Such use of low Q inductors
or ferrite beads is required when driving the converter front end
at high IF frequencies. Place either a differential capacitor or
two single-ended capacitors on the inputs to provide a matching
passive network. This ultimately creates a low-pass filter at the
input, which limits unwanted broadband noise. For more information, refer to the AN-742 Application Note, the AN-827
Application Note, and the Analog Dialogue article “TransformerCoupled Front-End for Wideband A/D Converters” (Volume 39,
April 2005) at www.analog.com. In general, the precise values
depend on the application.
For best dynamic performance, match the source impedances
driving VIN+x and VIN−x such that common-mode settling
errors are symmetrical. These errors are reduced by the
common-mode rejection of the ADC. An internal reference
buffer creates a differential reference that defines the span of the
ADC core.
Maximum SNR performance is achieved by setting the ADC
to the largest span in a differential configuration. In the case
of the AD6674, the available span is programmable through
the SPI port from 1.46 V p-p to 2.06 V p-p differential, with
1.70 V p-p differential being the default for the AD6674-1000
and AD6674-750, whereas the default for the AD6674-500 is
2.06 V p-p.
Differential Input Configurations
There are several ways to drive the AD6674, either actively or
passively. However, optimum performance is achieved by
driving the analog input differentially.
For applications where SNR and SFDR are key parameters,
differential transformer coupling is the recommended input
configuration (see Figure 74 and Table 9) because the noise
performance of most amplifiers is not adequate to achieve the
true performance of the AD6674.
For low to midrange frequencies, it is recommended to use a
double balun or double transformer network (see Figure 74) for
optimum performance from the AD6674. For higher
frequencies in the second or third Nyquist zone, it is better to
remove some of the front-end passive components to ensure
wideband operation (see Figure 74 and Table 9).
0.1µF
The analog input to the AD6674 is a differential buffer. The
internal common-mode voltage of the buffer is 2.05 V. The
clock signal alternately switches the input circuit between
sample mode and hold mode. When the input circuit is switched
into sample mode, the signal source must be capable of charging
the sample capacitors and settling within one-half of a clock cycle.
A small resistor, in series with each input, can help reduce the
peak transient current inserted from the output stage of the
R1
R2
BALUN
R2
R1
R3
C1
C2
0.1µF
0.1µF
ADC
R3
C1
NOTES
1. SEE TABLE 9 FOR COMPONENT VALUES.
Figure 74. Differential Transformer Coupled Configuration for AD6674
Rev. C | Page 28 of 96
12400-516
The AD6674 has two analog input channels and two JESD204B
output lane pairs. The AD6674 is designed to sample wide
bandwidth analog signals of up to 2 GHz. The AD6674 is
optimized for wide input bandwidth, high sampling rate,
excellent linearity, and low power in a small package.
Data Sheet
AD6674
Table 9. Differential Transformer Coupled Input Configuration Component Values
Frequency Range
DC to 250 MHz
250 MHz to 2 GHz
DC to 375 MHz
375 MHz to 2 GHz
DC to 500 MHz
500 MHz to 2 GHz
AD6674-750
AD6674-1000
Transformer
ETC1-1-13
BAL0006/BAL0006SMG
ETC1-1-13
BAL0006/BAL0006SMG
ECT1-1-13/BAL0006SMG
BAL0006/BAL0006SMG
Input Common Mode
The analog inputs of the AD6674 are internally biased to the
common mode, as shown in Figure 75. The common-mode
buffer has limited range in that the performance suffers greatly
if the common-mode voltage drops by more than 100 mV.
Therefore, in dc-coupled applications, set the common-mode
voltage to 2.05 V ± 100 mV to ensure proper ADC operation.
Analog Input Controls and SFDR Optimization
IAVDD3 (mA)
250
AVDD3
200Ω
67Ω
200Ω
28Ω
VIN+x
AVDD3
200Ω
67Ω
28Ω
200Ω
400Ω
VCM
BUFFER
C2 (pF)
2
2
2
2
2
Open
AD6674-1000
AND
AD6674-750
200
AD6674-500
150
50
150
250
350
450
550
650
BUFFER CURRENT SETTING
750
850
Figure 76. IAVDD3 vs. Buffer Current Setting in Register 0x018
AIN CONTROL
SPI REGISTERS
(0x008, 0x015,
0x016, 0x018,
0x019, 0x01A,
0x11A, 0x934,
0x935)
12400-517
VIN–x
3pF 1.5pF
C1 (pF)
4
4
4
4
4
Open
100
AVDD3
AVDD3
R3 (Ω)
10
10
10
10
10
0
300
AVDD3
10pF
R2 (Ω)
50
50
50
50
25
25
a high setting of 8.5×. The default setting in Register 0x018 is
3.0× for the AD6674-750 and AD6674-1000, whereas the
default for the AD6674-500 is 2.0×. These settings are sufficient
for operation in the first Nyquist zone. As the input buffer
currents are set, the amount of current required by the AVDD3
supply changes. This relationship is shown in Figure 76. For a
complete list of buffer current settings, see Table 45 for more
details.
The AD6674 offers flexible controls for the analog inputs such
as input termination, input capacitance, buffer current, and
input full-scale adjustment. All of the available controls are
shown in Figure 75.
3pF 1.5pF
R1 (Ω)
10
10
10
10
25
25
12400-341
Device
AD6674-500
Figure 75. Analog Input Controls
Use Register 0x018, Register 0x019, Register 0x01A, Register 0x11A,
Register 0x934, and Register 0x935 to adjust the buffer behavior on
each channel to optimize the SFDR over various input frequencies
and bandwidths of interest.
Input Buffer Control Registers (Register 0x018,
Register 0x019, Register 0x01A, Register 0x934,
Register 0x935, Register 0x11A)
The input buffer has many registers that set the bias currents
and other settings for operation at different frequencies. These
bias currents and settings can be changed to suit the input
frequency range of operation. Register 0x018 controls the buffer
bias current to reduce the effects of charge kickback from the
ADC core. This setting can be scaled from a low setting of 1.0× to
Register 0x019, Register 0x01A, Register 0x11A, and Register 0x935
offer secondary bias controls for the input buffer for frequencies
>500 MHz. Register 0x934 can be used to reduce input capacitance
to achieve wider signal bandwidth but doing so may result in
slightly lower linearity and noise performance. These register
settings do not affect the AVDD3 power as much as Register 0x018
does. For frequencies <500 MHz, it is recommended to use the
default settings for these registers. Table 10 shows the recommended values for the buffer current control registers for
various speed grades.
Use Register 0x11A when sampling in higher Nyquist zones
(>500 MHz for the AD6674-1000). This setting enables the
ADC sampling network to optimize the sampling and settling
times internal to the ADC for high frequency operation. For
frequencies greater than 500 MHz, it is recommended to
operate the ADC core at a 1.46 V full-scale setting irrespective
of the speed grade. This setting offers better SFDR without any
significant decrease in SNR.
Figure 77, Figure 78, and Figure 79 show the SFDR vs. analog input
frequency for various buffer settings (IBUFF) for the AD6674-1000.
Rev. C | Page 29 of 96
Data Sheet
80
The recommended settings shown in Table 10 were used to
collect the data while changing only the contents of Register 0x018.
90
75
85
SFDR (dBFS)
4.5×
3.0×
75
65
65
1.52GHz
1.65GHz
1.76GHz
1.9GHz
1.95GHz
65
60
55
–3
55
INPUT LEVEL (dBFS)
110
160 210 260 310 360
INPUT FREQUENCY (MHz)
410
460
12400-342
60
Figure 80. SNR/SFDR vs. Input Level and Input Frequencies, AD6674-1000
Figure 77. Buffer Current Sweeps, AD6674-1000 (SFDR vs. Input Frequency
and IBUFF); 10 MHz < fIN < 500 MHz; Front-End Network Shown in Figure 74
85
3.0×
4.0×
5.0×
6.0×
80
Figure 81, Figure 82, and Figure 83 show the SFDR vs. analog
input frequency for various buffer settings for the AD6674-500.
The recommended settings shown in Table 10 were used to take
the data while changing the contents of register 0x018 only.
95
75
90
70
85
SFDR (dBFS)
65
60
55
50
80
1.5×
2.0×
2.5×
3.5×
4.5×
75
70
65
45
60
450.3
12400-581
420.3
390.3
360.3
340.7
330.3
301.3
270.3
240.3
Figure 81. Buffer Current Sweeps, AD6674-750 (SFDR vs. Input Frequency and
IBUFF); 10 MHz < fIN < 450 MHz; Front-End Network Shown in Figure 74
75
95
70
90
60
85
SFDR (dBFS)
65
55
50
4.5×
5.5×
6.5×
7.5×
8.5×
4.5×
5.5×
6.5×
7.5×
80
75
70
1701.5
1795.6
INPUT FREQUENCY (MHz)
1889.8
65
Figure 79. Buffer Current Sweeps, AD6674-1000 (SFDR vs. Input Frequency
and IBUFF); 1500 MHz < fIN < 2 GHz; Front-End Network Shown in Figure 74
In certain high frequency applications, the SFDR can be
improved by reducing the full-scale setting, as shown in Table 10.
At high frequencies, the performance of the ADC core is limited by
jitter. The SFDR can be improved by reducing the full-scale level.
60
450.3
480.3
510.3
515.3 610.3 765.3 810.3
INPUT FREQUENCY (MHz)
985.3
1010.3
12400-582
1607.4
12400-453
40
1513.4
210.3
INPUT FREQUENCY (MHz)
80
45
180.3
55
Figure 78. Buffer Current Sweeps, AD6674-1000 (SFDR vs. Input Frequency
and IBUFF); 500 MHz < fIN < 1500 MHz; Front-End Network Shown in Figure 74
170.3
1374.8
150.3
1200.5
95.3
1026.2
125.3
851.9
INPUT FREQUENCY (MHz)
10.3
677.6
12400-452
40
503.4
SFDR (dBFS)
55
–1
–2
12400-454
60
60
SFDR (dBFS)
70
1.5×
70
50
10
75
70
65.3
SFDR (dBFS)
80
80
1.65GHz
1.52GHz
1.76GHz
1.95GHz
1.9GHz
SNR (dBc)
AD6674
Figure 82. Buffer Current Sweeps, AD6674-750 (SFDR vs. Input Frequency and
IBUFF); 450 MHz < fIN < 800 MHz; Front-End Network Shown in Figure 74
Rev. C | Page 30 of 96
AD6674
95
75
90
70
85
65
6.5×
7.5×
8.5×
60
75
Figure 84, Figure 85, and Figure 86 show the SFDR vs. analog
input frequency for various buffer settings for the AD6674-500.
The recommended settings shown in Table 10 were used to take
the data while changing the contents of register 0x018 only.
100
90
80
985.3
70
65
60
55
50
40
1010.3
40
4.0×
5.0×
6.0×
7.0×
8.0×
1205.3
1410.3
1600.3
INPUT FREQUENCY (MHz)
30
1950.3
Figure 86. Buffer Current Sweeps, AD6674-500 (SFDR vs. Input Frequency and
IBUFF); 1 GHz < fIN < 2 GHz; Front-End Network Shown in Figure 74
150.3
180.3
240.3
301.3
340.7
INPUT FREQUENCY (MHz)
390.3
450.3
12400-584
1.0×
1.5×
2.0×
3.0×
4.5×
95.3
1810.3
12400-586
45
50
0
10.3
810.3
75
60
10
510.3
515.3
610.3
765.3
INPUT FREQUENCY (MHz)
80
70
20
480.3
Figure 85. Buffer Current Sweeps, AD6674-500 (SFDR vs. Input Frequency and
IBUFF); 450 MHz < fIN < 1000 MHz; Front-End Network Shown in Figure 74
SNR/SFDR (dBFS)
1950.3
12400-583
1910.3
1810.3
1710.3
1600.3
1510.3
1410.3
1310.3
65
450.3
1110.3
50
1010.3
70
Figure 83. Buffer Current Sweeps, AD6674-750 (SFDR vs. Input Frequency and
IBUFF); 800 MHz < fIN < 2 GHz; Front-End Network Shown in Figure 74
SFDR (dBFS)
80
55
INPUT FREQUENCY (MHz)
4.0×
5.0×
6.0×
7.0×
8.0×
12400-585
SNR/SFDR (dBFS)
80
1205.3
SFDR (dBFS)
Data Sheet
Figure 84. Buffer Current Sweeps, AD6674-500 (SFDR vs. Input Frequency and
IBUFF); 10 MHz < fIN < 450 MHz; Front-End Network Shown in Figure 74
Rev. C | Page 31 of 96
AD6674
Data Sheet
Table 10. AD6674 Performance Optimization for Input Frequencies
Product
AD6674-500
Frequency
(MHz)
DC to 250
250 to 500
500 to 1000
1000 to 2000
AD6674-750
DC to 200
DC to 375
200 to 500
375 to 750
500 to 750
750 to 1000
1000 to 2000
AD6674-1000
DC to 150
DC to 500
500 to 1000
1000 to 2000
1
2
Buffer
Control 1
(0x018)
0x20
(2.0×)
0x70
(4.5×)
0x80
(5.0×)
0xF0
(8.5×)
0x20
(2.0×)
0x40
(3.0×)
0x70
(4.5×)
0xA0
(6.0×)
0xD0
(7.5×)
0xF0
(8.5×)
0xF0
(8.5×)
0x10
(1.5×)
0x40
(3.0×)
0xA0
(6.0×)
0xD0
(7.5×)
Buffer
Control 2
(0x019)
0x60
(Setting 3)
0x60
(Setting 3)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x40
(Setting 1)
0x50
(Setting 2)
0x50
(Setting 2)
0x60
(Setting 3)
0x70
(Setting 4)
Buffer
Control 3
(0x01A)
0x0A
(Setting 3)
0x0A
(Setting 3)
0x08
(Setting 1)
0x08
(Setting 1)
0x09
(Setting 2)
0x09
(Setting 2)
0x09
(Setting 2)
0x08
(Setting 1)
0x08
(Setting 1)
0x08
(Setting 1)
0x08
(Setting 1)
0x09
(Setting 2)
0x09
(Setting 2)
0x09
(Setting 2)
0x09
(Setting 2)
Buffer
Control 4
(0x11A)
0x00 (off)
Buffer
Control 5
(0x935)
0x04 (on)
Input
Full-Scale
Control
(0x030)
0x04
0x00 (off)
0x04 (on)
0x04
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x04 (on)
0x14
0x00 (off)
0x04 (on)
0x14
0x00 (off)
0x04 (on)
0x14
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x00 (off)
0x18
0x00 (off)
0x04 (on)
0x18
0x00 (off)
0x04 (on)
0x18
0x20 (on)
0x00 (off)
0x18
0x20 (on)
0x00 (off)
0x18
Input
Full-Scale
Range
(0x025)
0x0C
(2.06 V p-p)
0x0C
(2.06 V p-p)
0x08
(1.46 V p-p)
0x08
(1.46 V p-p)
0x0A
(1.70 V p-p)
0x0A
(1.70 V p-p)
0x0A
(1.70 V p-p)
0x08
(1.46 V p-p)
0x08
(1.46 V p-p)
0x08
(1.46 V p-p)
0x08
(1.46 V p-p)
0x0A
(1.70 V p-p)
0x0A
(1.70 V p-p)
0x08
(1.46 V p-p)
0x08
(1.46 V p-p)
Input
Capacitance
(0x934)
0x1F
0x1F
0x1F/0x002
0x1F/0x002
0x1F
0x1F
0x1F
0x1F
0x1F
0x1F/0x002
0x1F/0x002
0x1F
0x1F
0x1F/0x002
0x1F/0x002
The input termination can be changed to accommodate the application with little or no impact to ac performance.
The input capacitance can be set to 1.5 pF to achieve wider input bandwidth but results in slightly lower linearity and noise performance.
Rev. C | Page 32 of 96
Input
Termination
(0x016)1
0x0C/0x1C/
0x2C/0x6C
0x0C/0x1C/
0x2C/0x6C
0x0C/0x1C/
0x2C/0x6C
0x0C/0x1C/
0x2C/0x6C
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
0x0E/0x1E/
0x2E/0x6E
Data Sheet
AD6674
Absolute Maximum Input Swing
The absolute maximum input swing allowed at the inputs of the
AD6674 is 4.3 V p-p differential. Signals operating near or at
this level can cause permanent damage to the ADC.
VOLTAGE REFERENCE
A stable and accurate 1.0 V voltage reference is built into the
AD6674. This internal 1.0 V reference sets the full-scale input
range of the ADC. The full-scale input range can be adjusted via
Register 0x025. For more information on adjusting the input
swing, see Table 45. Figure 87 shows the block diagram of the
internal 1.0 V reference controls.
reference voltage. For more information on adjusting the fullscale level of the AD6674, refer to the Memory Map Register
Table section.
The use of an external reference may be necessary, in some
applications, to enhance the gain accuracy of the ADC or
improve thermal drift characteristics. Figure 88 shows the
typical drift characteristics of the internal 1.0 V reference.
1.0010
1.0009
1.0008
V_1P0 VOLTAGE (V)
1.0007
VIN+A/
VIN+B
VIN–A/
VIN–B
FULL-SCALE
VOLTAGE
ADJUST
1.0004
1.0003
1.0002
1.0001
1.0000
V_1P0
0.9998
–50
25
90
Figure 88. Typical V_1P0 Drift
12400-031
V_1P0 PIN
CONTROL SPI
REGISTER
(0x025 AND
0x024)
Figure 87. Internal Reference Configuration and Controls
Register 0x024 enables the user to either use this internal 1.0 V
reference or to provide an external 1.0 V reference. When using
an external voltage reference, provide a 1.0 V reference. The
full-scale adjustment is made using the SPI, irrespective of the
The external reference must be a stable 1.0 V reference. The
ADR130 is a good option for providing the 1.0 V reference.
Figure 89 shows how the ADR130 can be used to provide the
external 1.0 V reference to the AD6674. The grayed out areas
show unused blocks within the AD6674 while the ADR130
provides the external reference.
INTERNAL
V_1P0
GENERATOR
ADR130
1
NC
2
GND SET 5
3
VIN
0.1µF
0
TEMPERATURE (°C)
12400-106
0.9999
INPUT FULL-SCALE
RANGE ADJUST
SPI REGISTER
(0x025 AND 0x024)
INPUT
1.0005
FULL-SCALE
VOLTAGE
ADJUST
NC 6
VOUT 4
V_1P0
0.1µF
FULL-SCALE
CONTROL
Figure 89. External Reference Using the ADR130
Rev. C | Page 33 of 96
12400-032
INTERNAL
V_1P0
GENERATOR
ADC
CORE
1.0006
AD6674
Data Sheet
CLOCK INPUT CONSIDERATIONS
Input Clock Divider
For optimum performance, drive the AD6674 sample clock
inputs (CLK+ and CLK−) with a differential signal. This signal
is typically ac-coupled to the CLK+ and CLK− pins via a
transformer or clock drivers. These pins are biased internally
and require no additional biasing.
The AD6674 contains an input clock divider with the ability to
divide the Nyquist input clock by 1, 2, 4, or 8. The divide ratios can
be selected using Register 0x10B. This is shown in Figure 93.
The maximum frequency at the output of the divider is 1.0 GHz.
Figure 90 shows one preferred method for clocking the
AD6674. The low jitter clock source is converted from a singleended signal to a differential signal using an RF transformer.
0.1µF
CLK+
CLK+
ADC
100Ω
50Ω
12400-035
CLK–
0.1µF
CLK–
÷2
÷4
Figure 90. Transformer Coupled Differential Clock
÷8
Another option is to ac couple a differential CML or LVDS
signal to the sample clock input pins as shown in Figure 91 and
Figure 92.
3.3V
71Ω
33Ω
0.1µF
ADC
0.1µF
Z0 = 50Ω
12400-036
CLK+
CLK–
After programming the desired clock divider settings, changing
the input clock frequency, or glitching the input clock, a datapath
soft reset is recommended by writing 0x02 to Register 0x001.
This reset function restarts all the datapath and clock generation
circuitry in the device. The reset occurs on the first clock cycle
after the register is programmed, and the device requires 5 ms
to recover. This reset does not affect the contents of the memory
map registers.
Figure 91. Differential CML Sample Clock
0.1µF
CLK+
CLK+
0.1µF
CLOCK INPUT
LVDS
DRIVER
100Ω
CLK–
CLK–
50Ω 1
50Ω 1
ADC
0.1µF
150Ω RESISTORS ARE OPTIONAL.
12400-037
0.1µF
CLOCK INPUT
Figure 93. Clock Divider Circuit
The AD6674 clock divider can be synchronized using the
external SYSREF± input. A valid SYSREF± causes the clock
divider to reset to a programmable state. This feature is enabled
by setting Bit 7 of Register 0x10D. This synchronization feature
allows multiple devices to have their clock dividers aligned to
guarantee simultaneous input sampling.
10pF
33Ω
Z0 = 50Ω
REG 0x10B
12400-038
1:1Z
CLOCK
INPUT
The maximum frequency at the CLK± inputs is 4 GHz. This is
the limit of the divider. In applications where the clock input is
a multiple of the sample clock, take care to program the
appropriate divider ratio into the clock divider before applying
the clock signal. This ensures that the current transients during
device startup are controlled.
Figure 92. Differential LVDS Sample Clock
Clock Duty Cycle Considerations
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 clock duty cycle. Commonly, a 5% tolerance is
required on the clock duty cycle to maintain dynamic
performance characteristics. In applications where the clock
duty cycle cannot be guaranteed to be 50%, a higher multiple
frequency clock can be supplied to the AD6674. For example,
the AD6674-1000 can be clocked at 2 GHz with the internal
clock divider set to 2. This ensures a 50% duty cycle, high slew
rate internal clock for the ADC. See the Memory Map section
for more details on using this feature.
Input Clock Divider ½ Period Delay Adjustment
The input clock divider inside the AD6674 provides phase delay
in increments of ½ the input clock cycle. Program Register 0x10C
to enable this delay independently for each channel. Changing
the register does not affect the stability of the JESD204B link.
Clock Fine Delay Adjustment
Adjust the AD6674 sampling edge instant by writing to
Register 0x117 and Register 0x118. Setting Bit 0 of Register 0x117
enables the feature, and Register 0x118, Bits[7:0], set the value
of the delay. This value can be programmed individually for
each channel. The clock delay can be adjusted from −151.7 ps to
+150 ps in ~1.7 ps increments. The clock delay adjustment takes
effect immediately when it is enabled via SPI writes. Enabling the
clock fine delay adjustment in Register 0x117 causes a datapath
reset. However, the contents of Register 0x118 can be changed
without affecting the stability of the JESD204B link.
Rev. C | Page 34 of 96
Data Sheet
AD6674
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) is calculated by
Figure 95 shows the estimated SNR of the AD6674-1000 across
input frequency for different clock induced jitter values. The
SNR can be estimated by using the following equation:
  SNR JITTER  
   SNR ADC 


SNR (dBFS)  10log 10  10   10  10  


SNR = 20 × log 10(2 × π × fA × tJ)
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 specifications. IF
undersampling applications are particularly sensitive to jitter
(see Figure 94).
65
SNR (dBFS)
130
70
RMS CLOCK JITTER REQUIREMENT
120
100
16 BITS
90
14 BITS
80
12 BITS
45
10M
70
10 BITS
60
8 BITS
50
40
30
55
50
1
0.125ps
0.25ps
0.5ps
1.0ps
2.0ps
10
100
ANALOG INPUT FREQUENCY (MHz)
25fs
50fs
75fs
100f s
125f s
150f s
175f s
200f s
100M
1G
INPUT FREQUENCY (Hz)
10G
Figure 95. Estimated SNR Degradation for the AD6674-1000 vs.
Input Frequency and Jitter
1000
POWER-DOWN/STANDBY MODE
12400-039
SNR (dB)
110
60
Figure 94. Ideal SNR vs. Analog Input Frequency and Jitter
Treat the clock input as an analog signal in cases where aperture
jitter may affect the dynamic range of the AD6674. Separate
power supplies for clock drivers from the ADC output driver
supplies to avoid modulating the clock signal with digital noise.
If the clock is generated from another type of source (by gating,
dividing, or other methods), retime it using the original clock at
the last step. See the AN-501 Application Note and the AN-756
Application Note for more in-depth information about jitter
performance as it relates to ADCs.
The AD6674 has a PDWN/STBY pin that can be used to
configure the device in power-down or standby mode. The
default operation is the PDWN function. The PDWN/STBY pin
is a logic high pin. When in power-down mode, the JESD204B
link is disrupted. The power-down option can also be set via
Register 0x03F and Register 0x040.
In standby mode, the JESD204B link is not disrupted and
transmits zeros for all converter samples. This can be changed
using Register 0x571[7] to select /K/ characters.
Rev. C | Page 35 of 96
AD6674
Data Sheet
0.90
TEMPERATURE DIODE
The voltage response of the temperature diode (with SPIVDD =
1.8 V) is shown in Figure 96.
Rev. C | Page 36 of 96
0.80
0.75
0.70
0.65
0.60
–55 –45 –35 –25 –15 –5
5
15 25 35 45 55 65 75 85 95 105 115 125
TEMPERATURE (°C)
Figure 96. Temperature Diode Voltage vs. Temperature
12400-353
The temperature diode voltage can be output to the FD_A pin
using the SPI. Use Register 0x028[0] to enable or disable the
diode. Register 0x028 is a local register. Channel A must be
selected in the device index register (Register 0x008) to enable
the temperature diode readout. Configure the FD_A pin to
output the diode voltage by programming Register 0x040[2:0].
See Table 45 for more information.
0.85
DIODE VOLTAGE (V)
The AD6674 contains a diode-based temperature sensor for
measuring the temperature of the die. This diode outputs a
voltage and serve as a coarse temperature sensor to monitor the
internal die temperature.
Data Sheet
AD6674
ADC OVERRANGE AND FAST DETECT
time. This provides hysteresis and prevents the FD bit from
excessively toggling.
In receiver applications, it is desirable to have a mechanism to
reliably determine when the converter is about to be clipped.
The standard overrange bit in the JESD204B outputs provides
information on the state of the analog input that is of limited
usefulness. Therefore, it is helpful to have a programmable
threshold below full scale that allows time to reduce the gain
before the clip actually occurs. In addition, because input
signals can have significant slew rates, the latency of this
function is of major concern. Highly pipelined converters can
have significant latency. The AD6674 contains fast detect
circuitry for individual channels to monitor the threshold and
assert the FD_A and FD_B pins.
The operation of the upper threshold and lower threshold registers,
along with the dwell time registers, is shown in Figure 97.
The FD_x indicator is asserted if the input magnitude exceeds
the value programmed in the fast detect upper threshold registers,
located in Register 0x247 and Register 0x248. The selected
threshold register is compared with the signal magnitude at the
output of the ADC. The fast upper threshold detection has a
latency of 28. The approximate upper threshold magnitude is
defined by
Upper Threshold Magnitude (dBFS) = 20 log (Threshold
Magnitude/213)
ADC OVERRANGE (OR)
The ADC overrange indicator is asserted when an overrange is
detected on the input of the ADC. The overrange indicator can
be embedded within the JESD204B link as a control bit (when
CSB > 0). The latency of this overrange indicator matches the
sample latency.
The FD_x indicators are not cleared until the signal drops
below the lower threshold for the programmed dwell time. The
lower threshold is programmed in the fast detect lower threshold registers, located in Register 0x249 and Register 0x24A. The
fast detect lower threshold register is a 13-bit register that is
compared with the signal magnitude at the output of the ADC.
This comparison is subject to the ADC pipeline latency but is
accurate in terms of converter resolution. The lower threshold
magnitude is defined by
The AD6674 constantly monitors the analog input level and
records any overrange condition in any of the eight virtual
converters. For more information on the virtual converters,
refer to Figure 102. The overrange status of each virtual
converter is registered as a sticky bit (that is, it is set until
cleared) in Register 0x563. Clear the contents of Register 0x563
using Register 0x562 by toggling the bits corresponding to the
virtual converter to set and reset the position.
Lower Threshold Magnitude (dBFS) = 20 log (Threshold
Magnitude/213)
For example, to set an upper threshold of −6 dBFS, write
0x0FFF to Register 0x247 and Register 0x248; and to set a lower
threshold of −10 dBFS, write 0x0A1D to Register 0x249 and
Register 0x24A.
FAST THRESHOLD DETECTION (FD_A AND FD_B)
The fast detect (FD) bit (enabled in the control bits via
Register 0x559 and Register 0x55A) is immediately set
whenever the absolute value of the input signal exceeds the
programmable upper threshold level. The FD bit is only cleared
when the absolute value of the input signal drops below the
lower threshold level for greater than the programmable dwell
The dwell time can be programmed from 1 to 65,535 sample
clock cycles by placing the desired value in the fast detect dwell
time registers, located in Register 0x24B and Register 0x24C.
See the Memory Map section (Register 0x245 to Register 0x24C in
Table 45) for more details.
UPPER THRESHOLD
DWELL TIME
TIMER RESET BY
RISE ABOVE
LOWER
THRESHOLD
DWELL TIME
FD_A OR FD_B
Figure 97. Threshold Settings for FD_A and FD_B Signals
Rev. C | Page 37 of 96
TIMER COMPLETES BEFORE
SIGNAL RISES ABOVE
LOWER THRESHOLD
12400-040
MIDSCALE
LOWER THRESHOLD
AD6674
Data Sheet
SIGNAL MONITOR
The signal monitor block provides additional information about
the signal being digitized by the ADC. The signal monitor
computes the peak magnitude of the digitized signal. This
information can be used to drive an AGC loop to optimize the
range of the ADC in the presence of real-world signals.
The results of the signal monitor block can be obtained either
by reading back the internal values from the SPI port or by
embedding the signal monitoring information into the
JESD204B interface as special control bits. A global, 24-bit
programmable period controls the duration of the measurement. Figure 98 shows the simplified block diagram of the
signal monitor block.
FROM
MEMORY
MAP
SIGNAL MONITOR
PERIOD REGISTER
(SMPR)
0x271, 0x272, 0x273
DOWN
COUNTER
When the monitor period timer reaches a count of 1, the 13-bit
peak level value is transferred to the signal monitor holding
register, which can be read through the memory map or output
through the serial port (SPORT) over the JESD204B interface.
The monitor period timer is reloaded with the value in the
SMPR, and the countdown is restarted. In addition, the
magnitude of the first input sample is updated in the internal
magnitude storage register, and the comparison and update
procedure, as explained previously, continues.
IS
COUNT = 1?
LOAD
FROM
INPUT
LOAD
LOAD
SIGNAL
MONITOR
HOLDING
REGISTER
COMPARE
A>B
TO SPORT OVER
JESD204B AND
MEMORY MAP
SPORT OVER JESD204B
12400-471
CLEAR
MAGNITUDE
STORAGE
REGISTER
After enabling this mode, the value in the SMPR is loaded into a
monitor period timer that decrements at the decimated clock
rate. The magnitude of the input signal is compared with the
value in the internal magnitude storage register (not accessible
to the user), and the greater of the two is updated as the current
peak level. The initial value of the magnitude storage register is
set to the current ADC input signal magnitude. This comparison
continues until the monitor period timer reaches a count of 1.
Figure 98. Signal Monitor Block
The peak detector captures the largest signal within the
observation period. This period observes only the magnitude of
the signal. The resolution of the peak detector is a 13-bit value,
and the observation period is 24 bits and represents converter
output samples. The peak magnitude is derived by using the
following equation:
Peak Magnitude (dBFS) = 20 log(Peak Detector Value/213)
The magnitude of the input port signal is monitored over a
programmable time period that is determined by the signal
monitor period registers (SMPRs). Only even values of the
SMPR are supported. The peak detector function is enabled by
setting Bit 1 of Register 0x270 in the signal monitor control
register. The 24-bit SMPR must be programmed before
activating this mode.
The signal monitor data can also be serialized and sent over the
JESD204B interface as control bits. These control bits must be
deserialized from the samples to reconstruct the statistical data.
This signal control monitor function is enabled by setting
Bits[1:0] of Register 0x279 and Bit 1 of Register 0x27A.
Figure 99 shows two different example configurations for the
signal monitor control bit locations inside the JESD204B
samples. There are a maximum of three control bits that can be
inserted into the JESD204B samples; however, only one control
bit is required for the signal monitor. Control bits are inserted
from MSB to LSB. If only one control bit is to be inserted (CS = 1),
only the most significant control bit is used (see Configuration 1
and Configuration 2 in Figure 99). To select the SPORT over
JESD204B option, program Register 0x559, Register 0x55A, and
Register 0x58F. See the Memory Map Register Table section for
more information on setting these bits.
Figure 100 shows the 25-bit frame data that encapsulates the
peak detector value. The frame data is transmitted MSB first
with five 5-bit subframes. Each subframe contains a start bit
that can be used by a receiver to validate the deserialized data.
Figure 101 shows the SPORT over the JESD204B signal monitor
frame data with a monitor period timer set to 80 samples.
Rev. C | Page 38 of 96
Data Sheet
AD6674
16-BIT JESD204B SAMPLE SIZE (N' = 16)
EXAMPLE
CONFIGURATION 1
(N' = 16, N = 15, CS = 1)
1-BIT
CONTROL
BIT
(CS = 1)
15-BIT CONVERTER RESOLUTION (N = 15)
15
S[14]
X
14
S[13]
X
13
S[12]
X
12
11
S[11]
X
S[10]
X
10
9
S[9]
X
8
S[8]
X
7
S[7]
X
6
S[6]
X
5
S[5]
X
S[4]
X
4
S[3]
X
3
S[2]
X
2
S[1]
X
1
0
S[0]
X
CTRL
[BIT 2]
X
SERIALIZED SIGNAL MONITOR
FRAME DATA
16-BIT JESD204B SAMPLE SIZE (N' = 16)
14-BIT CONVERTER RESOLUTION (N = 14)
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
S[13]
X
S[12]
X
S[11]
X
S[10]
X
S[9]
X
S[8]
X
S[7]
X
S[6]
X
S[5]
X
S[4]
X
S[3]
X
S[2]
X
S[1]
X
S[0]
X
CTRL
[BIT 2]
X
TAIL
X
SERIALIZED SIGNAL MONITOR
FRAME DATA
Figure 99. Signal Monitor Control Bit Example Configurations
5-BIT SUBFRAMES
5-BIT IDLE
SUBFRAME
(OPTIONAL)
25-BIT
FRAME
IDLE
1
IDLE
1
IDLE
1
IDLE
1
IDLE
1
5-BIT IDENTIFIER START
0
SUBFRAME
ID[3]
0
ID[2]
0
ID[1]
0
ID[0]
1
5-BIT DATA
MSB
SUBFRAME
START
0
P[12]
P[11]
P[10]
P[9]
5-BIT DATA
SUBFRAME
START
0
P[8]
P[7]
P[6]
P5]
5-BIT DATA
SUBFRAME
START
0
P[4]
P[3]
P[2]
P1]
5-BIT DATA
LSB
SUBFRAME
START
0
P[0]
0
0
0
P[] = PEAK MAGNITUDE VALUE
12400-473
EXAMPLE
CONFIGURATION 2
(N' = 16, N = 14, CS = 1)
Figure 100. SPORT over JESD204B Signal Monitor Frame Data
Rev. C | Page 39 of 96
12400-472
1
CONTROL
BIT
1 TAIL
(CS = 1)
BIT
AD6674
Data Sheet
SMPR = 80 SAMPLES (0x271 = 0x50; 0x272 = 0x00; 0x273 = 0x00)
80-SAMPLE PERIOD
PAYLOAD 3
25-BIT FRAME (N)
IDENT.
DATA
MSB
DATA
DATA
DATA
LSB
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
IDLE
80-SAMPLE PERIOD
PAYLOAD 3
25-BIT FRAME (N + 1)
IDENT.
DATA
MSB
DATA
DATA
DATA
LSB
IDLE
IDLE
IDLE
IDLE
IDLE
80-SAMPLE PERIOD
IDENT.
DATA
MSB
DATA
DATA
DATA
LSB
IDLE
IDLE
IDLE
IDLE
IDLE
Figure 101. SPORT over JESD204B Signal Monitor Example with Period = 80 Samples
Rev. C | Page 40 of 96
12400-474
PAYLOAD 3
25-BIT FRAME (N + 2)
Data Sheet
AD6674
DIGITAL DOWNCONVERTER (DDC)
The AD6674 includes four digital downconverters (DDCs) that
provide filtering and reduce the output data rate. This digital
processing section includes an NCO, a half-band decimating
filter, an FIR filter, a gain stage, and a complex to real conversion stage. Each of these processing blocks has control lines that
allow it to be independently enabled and disabled to provide the
desired processing function. The digital downconverter can be
configured to output either real data or complex output data.
The DDCs output a 16-bit stream. To enable this operation, the
converter number of bits, N, is set to a default value of 16, even
though the analog core only outputs 14 bits. In full bandwidth
operation, the ADC outputs are the 14-bit word followed by two
zeros, unless the tail bits are enabled.
DDC I/Q INPUT SELECTION
The AD6674 has two ADC channels and four DDC channels.
Each DDC channel has two input ports that can be paired to
support both real and complex inputs through the I/Q crossbar
mux. For real signals, both DDC input ports must select the
same ADC channel (that is, DDC Input Port I = ADC Channel A
and DDC Input Port Q = ADC Channel A). For complex
signals, each DDC input port must select different ADC
channels (that is, DDC Input Port I = ADC Channel A and
DDC Input Port Q = ADC Channel B).
The inputs to each DDC are controlled by the DDC input selection registers (Register 0x311, Register 0x331, Register 0x351, and
Register 0x371). See Table 45 for information on how to
configure the DDCs.
DDC I/Q OUTPUT SELECTION
ignore all DDC Q output ports. When any of the DDC channels
are set to use complex I/Q outputs, the user must clear this bit
to use both DDC Output Port I and DDC Output Port Q. For
more information, see Figure 110.
DDC GENERAL DESCRIPTION
The four DDC blocks are used to extract a portion of the full
digital spectrum captured by the ADC(s). They are intended for
IF sampling or oversampled baseband radios requiring wide
bandwidth input signals.
Each DDC block contains the following signal processing
stages:




Frequency translation stage (optional)
Filtering stage
Gain stage (optional)
Complex to real conversion stage (optional)
Frequency Translation Stage (Optional)
This stage consists of a 12-bit complex NCO and quadrature
mixers that can be used for frequency translation of both real
and complex input signals. This stage shifts a portion of the
available digital spectrum down to baseband.
Filtering Stage
After shifting down to baseband, this stage decimates the
frequency spectrum using a chain of up to four half-band lowpass filters for rate conversion. The decimation process lowers
the output data rate, which in turn reduces the output interface
rate.
Gain Stage (Optional)
Each DDC channel has two output ports that can be paired to
support both real and complex outputs. For real output signals,
only the DDC Output Port I is used (the DDC Output Port Q is
invalid). For complex I/Q output signals, both DDC Output
Port I and DDC Output Port Q are used.
Due to losses associated with mixing a real input signal down to
baseband, this stage compensates by adding an additional 0 dB
or 6 dB of gain.
Complex to Real Conversion Stage (Optional)
The I/Q outputs to each DDC channel are controlled by the
DDC complex to real enable bit, Bit 3, in the DDC control
registers (Register 0x310, Register 0x330, Register 0x350, and
Register 0x370).
When real outputs are necessary, this stage converts the
complex outputs back to real by performing an fS/4 mixing
operation plus a filter to remove the complex component of the
signal.
The Chip Q ignore bit in the chip mode register (Register 0x200[5])
controls the chip output muxing of all the DDC channels.
When all DDC channels use real outputs, set this bit high to
Figure 102 shows the detailed block diagram of the DDCs
implemented in the AD6674.
Rev. C | Page 41 of 96
AD6674
Data Sheet
ADC
SAMPLING
AT fS
HB1 FIR
DCM = 2
COMPLEX TO REAL
CONVERSION
(OPTIONAL)
Q CONVERTER 1
REAL/I
CONVERTER 2
Q CONVERTER 3
SYSREF±
COMPLEX TO REAL
CONVERSION
(OPTIONAL)
REAL/Q Q
ADC
SAMPLING
AT fS
GAIN = 0dB
OR 6dB
NCO
+
MIXER
(OPTIONAL)
HB1 FIR
DCM = 2
I
HB2 FIR
DCM = BYPASS OR 2
DDC 2
REAL/I
REAL/I
CONVERTER 4
OUTPUT INTERFACE
COMPLEX TO REAL
CONVERSION
(OPTIONAL)
GAIN = 0dB
OR 6dB
HB1 FIR
DCM = 2
HB2 FIR
DCM = BYPASS OR 2
REAL/Q Q
HB3 FIR
DCM = BYPASS OR 2
NCO
+
MIXER
(OPTIONAL)
HB3 FIR
DCM = BYPASS OR 2
I/Q CROSSBAR MUX
I
HB4 FIR
DCM = BYPASS OR 2
DDC 1
REAL/I
REAL/I
REAL/I
CONVERTER 0
SYSREF±
HB4 FIR
DCM = BYPASS OR 2
REAL/I
GAIN = 0dB
OR 6dB
REAL/Q Q
HB2 FIR
DCM = BYPASS OR 2
NCO
+
MIXER
(OPTIONAL)
HB3 FIR
DCM = BYPASS OR 2
I
HB4 FIR
DCM = BYPASS OR 2
DDC 0
REAL/I
Q CONVERTER 5
SYSREF±
SYNCHRONIZATION
CONTROL CIRCUITS
COMPLEX TO REAL
CONVERSION
(OPTIONAL)
HB1 FIR
DCM = 2
REAL/I
CONVERTER 6
Q CONVERTER 7
12400-041
SYSREF±
GAIN = 0dB
OR 6dB
REAL/Q Q
HB2 FIR
DCM = BYPASS OR 2
NCO
+
MIXER
(OPTIONAL)
HB3 FIR
DCM = BYPASS OR 2
I
HB4 FIR
DCM = BYPASS OR 2
DDC 3
REAL/I
SYSREF
Figure 102. DDC Detailed Block Diagram
Figure 103 shows an example usage of one of the four DDC
blocks with a real input signal and four half-band filters (HB4 +
HB3 + HB2 + HB1). It shows both complex (decimate by 16)
and real (decimate by 8) output options.
When DDCs have different decimation ratios, the chip
decimation ratio (Register 0x201) must be set to the lowest
decimation ratio of all the DDC blocks. In this scenario,
samples of higher decimation ratio DDCs are repeated to match
the chip decimation ratio sample rate. Whenever the NCO
frequency is set or changed, the DDC soft reset must be issued.
If the DDC soft reset is not issued, the output may potentially
show amplitude variations.
Table 11, Table 12, Table 13, Table 14, and Table 15 show the
DDC samples when the chip decimation ratio is set to 1, 2, 4, 8,
or 16, respectively. When DDCs have different decimation
ratios, the chip decimation ratio must be set to the lowest
decimation ratio of all the DDC channels. In this scenario,
samples of higher decimation ratio DDCs are repeated to match
the chip decimation ratio sample rate.
Rev. C | Page 42 of 96
Data Sheet
AD6674
ADC
REAL INPUT—SAMPLED AT fS
–fS/2
–fS/3
ADC
SAMPLING
AT fS
REAL
BANDWIDTH OF
INTEREST IMAGE
–fS/4
REAL
BANDWIDTH OF
INTEREST
fS/32
–fS/32
DC
–fS/16
fS/16
–fS/8
FREQUENCY TRANSLATION STAGE (OPTIONAL)
DIGITAL MIXER + NCO FOR fS/3 TUNING, THE FREQUENCY
TUNING WORD = ROUND ((fS/3)/fS × 4096) = +1365 (0x555)
fS/8
fS/4
fS/3
fS/2
I
NCO TUNES CENTER OF
BANDWIDTH OF INTEREST
TO BASEBAND
cos(wt)
REAL
12-BIT
NCO
90°
0°
–sin(wt)
Q
DIGITAL FILTER
RESPONSE
–fS/2
–fS/3
–fS/4
FILTERING STAGE
4 DIGITAL HALF-BAND FILTERS
(HB4 + HB3 + HB2 + HB1)
HB4 FIR
I
HALFBAND
FILTER
Q
HALFBAND
FILTER
BANDWIDTH OF INTEREST
(–6dB LOSS DUE TO
NCO + MIXER)
fS/32
–fS/32
DC
–fS/16
fS/16
–fS/8
HB3 FIR
2
HALFBAND
FILTER
2
HALFBAND
FILTER
HB4 FIR
BANDWIDTH OF
INTEREST IMAGE
(–6dB LOSS DUE TO
NCO + MIXER)
fS/8
HB2 FIR
2
HALFBAND
FILTER
2
HALFBAND
FILTER
HB3 FIR
fS/4
fS/3
fS/2
HB1 FIR
2
HB2 FIR
HALFBAND
FILTER
I
HB1 FIR
2
HALFBAND
FILTER
Q
6dB GAIN TO
COMPENSATE FOR
NCO + MIXER LOSS
COMPLEX (I/Q) OUTPUTS
GAIN STAGE (OPTIONAL)
DIGITAL FILTER
RESPONSE
I
GAIN STAGE (OPTIONAL)
Q
0dB OR 6dB GAIN
COMPLEX TO REAL
CONVERSION STAGE (OPTIONAL)
fS/4 MIXING + COMPLEX FILTER TO REMOVE Q
fS/32
–fS/32
DC
–fS/16
fS/16
–fS/8
I
REAL (I) OUTPUTS
+6dB
+6dB
fS/8
2
+6dB
2
+6dB
I
Q
fS/32
–fS/32
DC
–fS/16
fS/16
DOWNSAMPLE BY 2
I
DECIMATE BY 8
Q
DECIMATE BY 16
0dB OR 6dB GAIN
Q
COMPLEX REAL/I
TO
REAL
–fS/8
fS/32
–fS/32
DC
–fS/16
fS/16
fS/8
Figure 103. DDC Theory of Operation Example (Real Input, Decimate by 16)
Rev. C | Page 43 of 96
12400-042
6dB GAIN TO
COMPENSATE FOR
NCO + MIXER LOSS
AD6674
Data Sheet
Table 11. DDC Samples When Chip Decimation Ratio = 1
HB1 FIR
(DCM1 = 1)
N
N+1
N+2
N+3
N+4
N+5
N+6
N+7
N+8
N+9
N + 10
N + 11
N + 12
N + 13
N + 14
N + 15
N + 16
N + 17
N + 18
N + 19
N + 20
N + 21
N + 22
N + 23
N + 24
N + 25
N + 26
N + 27
N + 28
N + 29
N + 30
N + 31
1
Real (I) Output (Complex to Real Enabled)
HB2 FIR +
HB3 FIR + HB2
HB4 FIR + HB3 FIR +
HB1 FIR
FIR + HB1 FIR
HB2 FIR + HB1 FIR
(DCM1 = 2)
(DCM1 = 4)
(DCM1 = 8)
N
N
N
N+1
N+1
N+1
N
N
N
N+1
N+1
N+1
N+2
N
N
N+3
N+1
N+1
N+2
N
N
N+3
N+1
N+1
N+4
N+2
N
N+5
N+3
N+1
N+4
N+2
N
N+5
N+3
N+1
N+6
N+2
N
N+7
N+3
N+1
N+6
N+2
N
N+7
N+3
N+1
N+8
N+4
N+2
N+9
N+5
N+3
N+8
N+4
N+2
N+9
N+5
N+3
N + 10
N+4
N+2
N + 11
N+5
N+3
N + 10
N+4
N+2
N + 11
N+5
N+3
N + 12
N+6
N+2
N + 13
N+7
N+3
N + 12
N+6
N+2
N + 13
N+7
N+3
N + 14
N+6
N+2
N + 15
N+7
N+3
N + 14
N+6
N+2
N + 15
N+7
N+3
Complex (I/Q) Outputs (Complex to Real Disabled)
HB2 FIR +
HB3 FIR + HB2
HB4 FIR + HB3 FIR +
HB1 FIR
HB1 FIR
FIR + HB1 FIR
HB2 FIR + HB1 FIR
(DCM1 = 2) (DCM1 = 4)
(DCM1 = 8)
(DCM1 = 16)
N
N
N
N
N+1
N+1
N+1
N+1
N
N
N
N
N+1
N+1
N+1
N+1
N+2
N
N
N
N+3
N+1
N+1
N+1
N+2
N
N
N
N+3
N+1
N+1
N+1
N+4
N+2
N
N
N+5
N+3
N+1
N+1
N+4
N+2
N
N
N+5
N+3
N+1
N+1
N+6
N+2
N
N
N+7
N+3
N+1
N+1
N+6
N+2
N
N
N+7
N+3
N+1
N+1
N+8
N+4
N+2
N
N+9
N+5
N+3
N+1
N+8
N+4
N+2
N
N+9
N+5
N+3
N+1
N + 10
N+4
N+2
N
N + 11
N+5
N+3
N+1
N + 10
N+4
N+2
N
N + 11
N+5
N+3
N+1
N + 12
N+6
N+2
N
N + 13
N+7
N+3
N+1
N + 12
N+6
N+2
N
N + 13
N+7
N+3
N+1
N + 14
N+6
N+2
N
N + 15
N+7
N+3
N+1
N + 14
N+6
N+2
N
N + 15
N+7
N+3
N+1
DCM = decimation.
Table 12. DDC Samples When Chip Decimation Ratio = 2
Real (I) Output (Complex to Real Enabled)
HB4 FIR +
HB3 FIR +
HB3 FIR +
HB2 FIR +
HB2 FIR +
HB2 FIR +
HB1 FIR
HB1 FIR
HB1 FIR
(DCM1 = 2)
(DCM1 = 4)
(DCM1 = 8)
N
N
N
N+1
N+1
N+1
N+2
N
N
N+3
N+1
N+1
N+4
N+2
N
N+5
N+3
N+1
N+6
N+2
N
N+7
N+3
N+1
N+8
N+4
N+2
N+9
N+5
N+3
Complex (I/Q) Outputs (Complex to Real Disabled)
HB4 FIR +
HB3 FIR +
HB3 FIR +
HB2 FIR +
HB2 FIR +
HB2 FIR +
HB1 FIR
HB1 FIR
HB1 FIR
HB1 FIR
(DCM1 = 2)
(DCM1 = 4)
(DCM1 = 8)
(DCM1 = 16)
N
N
N
N
N+1
N+1
N+1
N+1
N+2
N
N
N
N+3
N+1
N+1
N+1
N+4
N+2
N
N
N+5
N+3
N+1
N+1
N+6
N+2
N
N
N+7
N+3
N+1
N+1
N+8
N+4
N+2
N
N+9
N+5
N+3
N+1
Rev. C | Page 44 of 96
Data Sheet
AD6674
Real (I) Output (Complex to Real Enabled)
HB4 FIR +
HB3 FIR +
HB3 FIR +
HB2 FIR +
HB2 FIR +
HB2 FIR +
HB1 FIR
HB1 FIR
HB1 FIR
(DCM1 = 2)
(DCM1 = 4)
(DCM1 = 8)
N + 10
N+4
N+2
N + 11
N+5
N+3
N + 12
N+6
N+2
N + 13
N+7
N+3
N + 14
N+6
N+2
N + 15
N+7
N+3
1
Complex (I/Q) Outputs (Complex to Real Disabled)
HB4 FIR +
HB3 FIR +
HB3 FIR +
HB2 FIR +
HB2 FIR +
HB2 FIR +
HB1 FIR
HB1 FIR
HB1 FIR
HB1 FIR
(DCM1 = 2)
(DCM1 = 4)
(DCM1 = 8)
(DCM1 = 16)
N + 10
N+4
N+2
N
N + 11
N+5
N+3
N+1
N + 12
N+6
N+2
N
N + 13
N+7
N+3
N+1
N + 14
N+6
N+2
N
N + 15
N+7
N+3
N+1
DCM = decimation.
Table 13. DDC Samples When Chip Decimation Ratio = 4
Real (I) Output (Complex to Real Enabled)
HB4 FIR + HB3 FIR +
HB2 FIR + HB1 FIR
HB3 FIR + HB2 FIR +
(DCM1 = 8)
HB1 FIR (DCM1 = 4)
N
N
N+1
N+1
N+2
N
N+3
N+1
N+4
N+2
N+5
N+3
N+6
N+2
N+7
N+3
1
Complex (I/Q) Outputs (Complex to Real Disabled)
HB4 FIR + HB3 FIR +
HB2 FIR + HB1 FIR
HB3 FIR + HB2 FIR +
HB2 FIR + HB1 FIR
(DCM1 = 4)
HB1 FIR (DCM1 = 8)
(DCM1 = 16)
N
N
N
N+1
N+1
N+1
N+2
N
N
N+3
N+1
N+1
N+4
N+2
N
N+5
N+3
N+1
N+6
N+2
N
N+7
N+3
N+1
DCM = decimation.
Table 14. DDC Samples When Chip Decimation Ratio = 8
Real (I) Output (Complex to Real Enabled)
Complex (I/Q) Outputs (Complex to Real Disabled)
HB4 FIR + HB3 FIR + HB2 FIR +
HB3 FIR + HB2 FIR + HB1 FIR
HB1 FIR (DCM1 = 16)
(DCM1 = 8)
N
N
N+1
N+1
N+2
N
N+3
N+1
N+4
N+2
N+5
N+3
N+6
N+2
N+7
N+3
HB4 FIR + HB3 FIR + HB2 FIR + HB1 FIR (DCM1 = 8)
N
N+1
N+2
N+3
N+4
N+5
N+6
N+7
1
DCM = decimation.
Table 15. DDC Samples When Chip Decimation Ratio = 16
Real (I) Output (Complex to Real Enabled)
HB4 FIR + HB3 FIR + HB2 FIR + HB1 FIR (DCM1 = 16)
Not applicable
Not applicable
Not applicable
Not applicable
1
Complex (I/Q) Outputs (Complex to Real Disabled)
HB4 FIR + HB3 FIR + HB2 FIR + HB1 FIR (DCM1 = 16)
N
N+1
N+2
N+3
DCM -= decimation.
Rev. C | Page 45 of 96
AD6674
Data Sheet
For example, if the chip decimation ratio is set to decimate by 4,
DDC 0 is set to use HB2 + HB1 filters (complex outputs, decimate
by 4) and DDC 1 is set to use HB4 + HB3 + HB2 + HB1 filters
(real outputs, decimate by 8). DDC 1 repeats its output data two
times for every one DDC 0 output. The resulting output samples
are shown in Table 16.
Table 16. DDC Output Samples When Chip DCM1 = 4, DDC 0 DCM1 = 4 (Complex), and DDC 1 DCM1 = 8 (Real)
DDC Input Samples
N
N+1
N+2
N+3
N+4
N+5
N+6
N+7
N+8
N+9
N + 10
N + 11
N + 12
N + 13
N + 14
N + 15
1
Output Port I
I0 (N)
DDC 0
Output Port Q
Q0 (N)
I0 (N + 1)
Q0 (N + 1)
I0 (N + 2)
Q0 (N + 2)
I0 (N + 3)
Q0 (N + 3)
DCM = decimation.
Rev. C | Page 46 of 96
Output Port I
I1 (N)
I1 (N + 1)
DDC 1
Output Port Q
Not applicable
Not applicable
Data Sheet
AD6674
FREQUENCY TRANSLATION
GENERAL DESCRIPTION
Variable IF Mode
Frequency translation is accomplished by using a 12-bit
complex NCO with a digital quadrature mixer. This stage
translates either a real or complex input signal from an IF to a
baseband complex digital output (carrier frequency = 0 Hz).
NCO and mixers are enabled. NCO output frequency can be
used to digitally tune the IF frequency.
0 Hz IF (ZIF) Mode
The mixers are bypassed, and the NCO is disabled.
The frequency translation stage of each DDC can be controlled
individually and supports four different IF modes using Bits[5:4]
of the DDC control registers (Register 0x310, Register 0x330,
Register 0x350, and Register 0x370). These IF modes are
The mixers and the NCO are enabled in special downmixing by
fS/4 mode to save power.
Test Mode
Variable IF mode
0 Hz IF or zero IF (ZIF) mode
fS/4 Hz IF mode
Test mode
Input samples are forced to 0.999 to positive full scale. The
NCO is enabled. This test mode allows the NCOs to directly
drive the decimation filters.
Figure 104 and Figure 105 show examples of the frequency
translation stage for both real and complex inputs.
NCO FREQUENCY TUNING WORD (FTW) SELECTION
12-BIT NCO FTW = MIXING FREQUENCY/ADC SAMPLE RATE × 4096
I
ADC + DIGITAL MIXER + NCO
REAL INPUT—SAMPLED AT fS
REAL
cos(wt)
ADC
SAMPLING
AT fS
REAL
12-BIT
NCO
90°
0°
COMPLEX
–sin(wt)
Q
BANDWIDTH OF
INTEREST
BANDWIDTH OF
INTEREST IMAGE
–fS/2
–fS/3
–fS/4
–fS/8
fS/32
–fS/32
DC
–fS/16
fS/16
fS/8
fS/4
fS/3
fS/2
–6dB LOSS DUE TO
NCO + MIXER
12-BIT NCO FTW =
ROUND ((fS/3)/fS × 4096) = +1365 (0x555)
POSITIVE FTW VALUES
–fS/32
DC
fS/32
12-BIT NCO FTW =
ROUND (( fS/3)/fS × 4096) = –1365 (0xAAB)
–fS/32
NEGATIVE FTW VALUES
DC
fS/32
Figure 104. DDC NCO Frequency Tuning Word Selection—Real Inputs
Rev. C | Page 47 of 96
12400-043




fS/4 Hz IF Mode
AD6674
Data Sheet
NCO FREQUENCY TUNING WORD (FTW) SELECTION
12-BIT NCO FTW = MIXING FREQUENCY/ADC SAMPLE RATE × 4096
QUADRATURE ANALOG MIXER +
2 ADCs + QUADRATURE DIGITAL REAL
MIXER + NCO
COMPLEX INPUT—SAMPLED AT fS
QUADRATURE MIXER
ADC
SAMPLING
AT fS
I
+
I
I
Q
Q
90°
PHASE
12-BIT
NCO
90°
0°
Q
Q
ADC
SAMPLING
AT fS
Q
Q
I
I
–
–sin(wt)
I
I
+
COMPLEX
Q
+
BANDWIDTH OF
INTEREST
IMAGE DUE TO
ANALOG I/Q
MISMATCH
–fS/3
–fS/4
fS/32
–fS/32
–fS/16
fS/16
DC
–fS/8
fS/8
fS/4
fS/3
fS/2
12-BIT NCO FTW =
ROUND ((fS/3)/fS × 4096) = +1365 (0x555)
POSITIVE FTW VALUES
–fS/32
fS/32
12400-044
–fS/2
DC
Figure 105. DDC NCO Frequency Tuning Word Selection—Complex Inputs
DDC NCO + MIXER LOSS AND SFDR
Setting Up the NCO FTW and POW
When mixing a real input signal down to baseband, 6 dB of loss
is introduced in the signal due to filtering of the negative image.
An additional 0.05 dB of loss is introduced by the NCO. The
total loss of a real input signal mixed down to baseband is
6.05 dB. For this reason, it is recommended that the user
compensate for this loss by enabling the 6 dB of gain in the gain
stage of the DDC to recenter the dynamic range of the signal
within the full scale of the output bits.
The NCO frequency value is given by the 12-bit twos
complement number entered in the NCO FTW. Frequencies
between −fS/2 and +fS/2 (fS/2 excluded) are represented using
the following frequency words:
When mixing a complex input signal down to baseband, the
maximum value each I/Q sample can reach is 1.414 × full scale
after it passes through the complex mixer. To avoid overrange of
the I/Q samples and to keep the data bit-widths aligned with
real mixing, 3.06 dB of loss is introduced in the mixer for
complex signals. An additional 0.05 dB of loss is introduced by
the NCO. The total loss of a complex input signal mixed down
to baseband is −3.11 dB.
The worst case spurious signal from the NCO is greater than
102 dBc SFDR for all output frequencies.
NUMERICALLY CONTROLLED OSCILLATOR
The AD6674 has a 12-bit NCO for each DDC that enables the
frequency translation process. The NCO allows the input
spectrum to be tuned to dc, where it can be effectively filtered
by the subsequent filter blocks to prevent aliasing. The NCO
can be set up by providing a frequency tuning word (FTW) and
a phase offset word (POW).



0x800 represents a frequency of −fS/2.
0x000 represents dc (frequency is 0 Hz).
0x7FF represents a frequency of +fS/2 − fS/212.
The NCO frequency tuning word can be calculated using the
following equation:

mod  f C , f S  

NCO _ FTW  round  212

fS


where:
NCO_FTW is a 12-bit twos complement number representing
the NCO FTW.
fC is the desired carrier frequency in Hz.
fS is the AD6674 sampling frequency (clock rate) in Hz.
mod( ) is a remainder function. For example, mod(110,100) =
10 and for negative numbers, mod(–32,10) = −2.
round( ) is a rounding function. For example, round(3.6) = 4
and for negative numbers, round(–3.4) = −3.
Note that this equation applies to the aliasing of signals in the
digital domain (that is, aliasing introduced when digitizing
analog signals).
Rev. C | Page 48 of 96
Data Sheet
AD6674
For example, if the ADC sampling frequency (fS) is 500 MSPS
and the carrier frequency (fC) is 140.312 MHz, then
Use the following two methods to synchronize multiple PAWs
within the chip.

NCO_FTW =
 mod140.312,500 
round 212
  1149 MHz
500


This, in turn, converts to 0x47D in the 12-bit twos complement
representation for NCO_FTW. The actual carrier frequency,
fC_ACTUAL, is calculated based on the following equation:
fC _ ACTUAL
NCO_ FTW fS
212

 140.26 MHz
A 12-bit POW is available for each NCO to create a known
phase relationship between multiple AD6674 chips or
individual DDC channels inside one AD6674 chip.
The following procedure must be followed to update the FTW
and/or POW registers to ensure proper operation of the NCO:
1.
2.
3.
Write to the FTW registers for all the DDCs.
Write to the POW registers for all the DDCs.
Synchronize the NCOs either through the DDC NCO soft
reset bit (Register 0x300[4]) accessible through the SPI or
through the assertion of the SYSREF± pin.
It is important to note that the NCOs must be synchronized
either through the SPI or through the SYSREF± pin after all
writes to the FTW or POW registers have completed. This is
necessary to ensure the proper operation of the NCO.
NCO Synchronization
Each NCO contains a separate phase accumulator word (PAW)
that determines the instantaneous phase of the NCO. The initial
reset value of each PAW is determined by the POW. The phase
increment value of each PAW is determined by the FTW See
the Setting Up the NCO FTW and POW section for more
information.
Using the SPI. Use the DDC NCO soft reset bit in the DDC
synchronization control register (Register 0x300[4]) to
reset all the PAWs in the chip. This is accomplished by
setting the DDC NCO soft reset bit high and then setting
this bit low. Note that this method can only be used to
synchronize DDC channels within the same AD6674 chip.
Using the SYSREF± pin. When the SYSREF± pin is
enabled in the SYSREF± control registers (Register 0x120
and Register 0x121) and the DDC synchronization is
enabled in the DDC synchronization control register
(Register 0x300[1:0]), any subsequent SYSREF± event
resets all the PAWs in the chip. Note that this method can
be used to synchronize DDC channels within the same
AD6674 chip or DDC channels within separate AD6674
chips.
Mixer
The NCO is accompanied by a mixer. Its operation is similar to
an analog quadrature mixer. It performs the downconversion of
input signals (real or complex) by using the NCO frequency as a
local oscillator. For real input signals, this mixer performs a real
mixer operation (with two multipliers). For complex input
signals, the mixer performs a complex mixer operation (with
four multipliers and two adders). The mixer adjusts its
operation based on the input signal (real or complex) provided
to each individual channel. The selection of real or complex
inputs can be controlled individually for each DDC block using
Bit 7 of the DDC control registers (Register 0x310, Register 0x330,
Register 0x350, and Register 0x370).
Rev. C | Page 49 of 96
AD6674
Data Sheet
FIR FILTERS
Table 17 shows the different bandwidths selectable by including
different half-band filters. In all cases, the DDC filtering stage
on the AD6674 provides <−0.001 dB of pass-band ripple and
>100 dB of stop-band alias rejection.
GENERAL DESCRIPTION
There are four sets of decimate by 2, low-pass, half-band, finite
impulse response (FIR) filters (labeled HB1 FIR, HB2 FIR, HB3
FIR, and HB4 FIR in Figure 102) following the frequency
translation stage. After the carrier of interest is tuned down to
dc (carrier frequency = 0 Hz), these filters efficiently lower the
sample rate, while providing sufficient alias rejection from
unwanted adjacent carriers around the bandwidth of interest.
Table 18 shows the amount of stop-band alias rejection for
multiple pass-band ripple/cutoff points. The decimation ratio of
the filtering stage of each DDC can be controlled individually
through Bits[1:0] of the DDC control registers (Register 0x310,
Register 0x330, Register 0x350, and Register 0x370).
HB1 FIR is always enabled and cannot be bypassed. The HB2,
HB3, and HB4 FIR filters are optional and can be bypassed for
higher output sample rates.
Table 17. DDC Filter Characteristics
ADC
Sample
Rate
(MSPS)
1000
750
500
1
Half Band
Filter
Selection
HB1
HB1 + HB2
HB1 + HB2 +
HB3
HB1 + HB2 +
HB3 + HB4
HB1
HB1 + HB2
HB1 + HB2 +
HB3
HB1 + HB2 +
HB3 + HB4
HB1
HB1 + HB2
HB1 + HB2 +
HB3
HB1 + HB2 +
HB3 + HB4
Real Output
Decimation
Output
Sample
Ratio
Rate
(MSPS)
1
1000
2
500
4
250
Complex (I/Q) Output
Decimation
Output Sample Rate
(MSPS)
Ratio
2
4
8
8
125
1
2
4
500 (I) + 500 (Q)
250 (I) + 250 (Q)
125 (I) + 125 (Q)
Alias
Protected
Bandwidth
(MHz)
385.0
192.5
96.3
Ideal SNR
Improvement1
(dB)
1
4
7
16
62.5 (I) + 62.5 (Q)
48.1
10
750
375
187.5
2
4
8
375 (I) + 375 (Q)
187.5 (I) + 187.5 (Q)
93.75 (I) + 93.75 (Q)
288.8
144.4
72.2
1
4
7
8
93.75
16
46.875 (I) + 46.875 (Q)
36.1
10
1
2
4
500
250
125
2
4
8
250 (I) + 250 (Q)
125 (I) + 125 (Q)
62.5 (I) + 62.5 (Q)
192.5
96.3
48.1
1
4
7
8
62.5
16
31.25 (I) + 31.25 (Q)
24.1
10
PassBand
Ripple
(dB)
<−0.001
Alias
Rejection
(dB)
>100
Ideal SNR improvement due to oversampling and filtering = 10log(bandwidth/(fS/2)).
Table 18. DDC Filter Alias Rejection
Alias Rejection
(dB)
>100
90
85
63.3
25
19.3
10.7
1
Pass-Band Ripple/Cutoff
Point (dB)
<−0.001
<−0.001
<−0.001
<−0.006
−0.5
−1.0
−3.0
Alias Protected Bandwidth for Real
(I) Outputs1
<38.5% × fOUT
<38.7% × fOUT
<38.9% × fOUT
<40% × fOUT
44.4% × fOUT
45.6% × fOUT
48% × fOUT
fOUT = ADC input sample rate ÷ DDC decimation.
Rev. C | Page 50 of 96
Alias Protected Bandwidth for Complex
(I/Q) Outputs
<77% × fOUT
<77.4% × fOUT
<77.8% × fOUT
<80% × fOUT
88.8% × fOUT
91.2% × fOUT
96% × fOUT
Data Sheet
AD6674
HALF-BAND FILTERS
Table 20. HB3 Filter Coefficients
The AD6674 offers four half-band filters to enable digital signal
processing of the ADC converted data. These half-band filters
are bypassable and can be individually selected.
HB3 Coefficient
Number
C1, C11
C2, C10
C3, C9
C4, C8
C5, C7
C6
Table 19. HB4 Filter Coefficients
HB4 Coefficient
Number
C1, C11
C2, C10
C3, C9
C4, C8
C5, C7
C6
Normalized
Coefficient
0.006042
0
−0.049316
0
0.293273
0.500000
Decimal
Coefficient (15-Bit)
99
0
−808
0
4805
8192
0
–20
–40
–60
–80
–100
–120
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NORMALIZED FREQUENCY (× π RAD/SAMPLE)
Figure 107. HB3 Filter Response
0
HB2 Filter
–20
MAGNITUDE (dB)
Decimal Coefficient
(18-Bit)
859
0
−6661
0
38,570
65,536
12400-046
The first decimate by 2, half-band, low-pass, FIR filter (HB4)
uses an 11-tap, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption. The HB4
filter is only used when complex outputs (decimate by 16) or
real outputs (decimate by 8) are enabled; otherwise, it is
bypassed. Table 19 and Figure 106 show the coefficients and
response of the HB4 filter.
MAGNITUDE (dB)
HB4 Filter
Normalized
Coefficient
0.006554
0
−0.050819
0
0.294266
0.500000
The third decimate by 2, half-band, low-pass, FIR filter (HB2)
uses a 19-tap, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption.
–40
–60
The HB2 filter is only used when complex or real outputs
(decimate by 4, 8, or 16) is enabled; otherwise, it is bypassed.
–80
Table 21 and Figure 108 show the coefficients and response of
the HB2 filter.
–100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NORMALIZED FREQUENCY (× π RAD/SAMPLE)
12400-045
–120
Figure 106. HB4 Filter Response
HB3 Filter
The second decimate by 2, half-band, low-pass, FIR filter (HB3)
uses an 11-tap, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption. The HB3
filter is only used when complex outputs (decimate by 8 or 16)
or real outputs (decimate by 4 or 8) are enabled; otherwise, it is
bypassed. Table 20 and Figure 107 show the coefficients and
response of the HB3 filter.
Table 21. HB2 Filter Coefficients
HB2 Coefficient
Number
C1, C19
C2, C18
C3, C17
C4, C16
C5, C15
C6, C14
C7, C13
C8, C12
C9, C11
C10
Rev. C | Page 51 of 96
Normalized
Coefficient
0.000614
0
−0.005066
0
0.022179
0
−0.073517
0
0.305786
0.500000
Decimal
Coefficient (19-Bit)
161
0
−1328
0
5814
0
−19,272
0
80,160
131,072
Data Sheet
0
–20
–20
–40
–40
–60
–80
–60
–80
–100
–100
–120
–120
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NORMALIZED FREQUENCY (× π RAD/SAMPLE)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NORMALIZED FREQUENCY (× π RAD/SAMPLE)
Figure 108. HB2 Filter Response
12400-048
MAGNITUDE (dB)
0
12400-047
MAGNITUDE (dB)
AD6674
Figure 109. HB1 Filter Response
HB1 Filter
DDC GAIN STAGE
The fourth and final decimate by 2, half-band, low-pass, FIR
filter (HB1) uses a 55-tap, symmetrical, fixed coefficient filter
implementation that is optimized for low power consumption.
The HB1 filter is always enabled and cannot be bypassed.
Table 22 and Figure 109 show the coefficients and response of
the HB1 filter.
Each DDC contains an independently controlled gain stage.
The gain is selectable as either 0 dB or 6 dB. When mixing a real
input signal down to baseband, it is recommended that the user
enable the 6 dB of gain to recenter the dynamic range of the
signal within the full scale of the output bits.
Table 22. HB1 Filter Coefficients
HB1 Coefficient
Number
C1, C55
C2, C54
C3, C53
C4, C52
C5, C51
C6, C50
C7, C49
C8, C48
C9, C47
C10, C46
C11, C45
C12, C44
C13, C43
C14, C42
C15, C41
C16, C40
C17, C39
C18, C38
C19, C37
C20, C36
C21, C35
C22, C34
C23, C33
C24, C32
C25, C31
C26, C30
C27, C29
C28
Normalized
Coefficient
−0.000023
0
0.000097
0
−0.000288
0
0.000696
0
−0.0014725
0
0.002827
0
−0.005039
0
0.008491
0
−0.013717
0
0.021591
0
−0.033833
0
0.054806
0
−0.100557
0
0.316421
0.500000
Decimal
Coefficient (21-Bit)
−24
0
102
0
−302
0
730
0
−1544
0
2964
0
−5284
0
8903
0
−14,383
0
22,640
0
−35,476
0
57,468
0
−105,442
0
331,792
524,288
When mixing a complex input signal down to baseband, the
mixer has already recentered the dynamic range of the signal
within the full scale of the output bits, and no additional gain is
necessary. However, the optional 6 dB gain compensates for low
signal strengths. The downsample by 2 portion of the HB1 FIR
filter is bypassed when using the complex to real conversion
stage.
DDC COMPLEX TO REAL CONVERSION
Each DDC contains an independently controlled complex to
real conversion block. The complex to real conversion block
reuses the last filter (HB1 FIR) in the filtering stage along with
an fS/4 complex mixer to upconvert the signal. After upconverting the signal, the Q portion of the complex mixer is no longer
needed and is dropped.
Figure 110 shows a simplified block diagram of the complex to
real conversion.
Rev. C | Page 52 of 96
Data Sheet
AD6674
HB1 FIR
GAIN STAGE
COMPLEX TO
REAL ENABLE
LOW-PASS
FILTER
I
2
0dB
OR
6dB
I
0 I/REAL
1
COMPLEX TO REAL CONVERSION
0dB
OR
6dB
I
cos(wt)
+
90°
fS/4
REAL
0°
–
sin(wt)
LOW-PASS
FILTER
2
Q
0dB
OR
6dB
Q
Q
12400-049
Q
0dB
OR
6dB
HB1 FIR
Figure 110. Complex to Real Conversion Block
DDC EXAMPLE CONFIGURATIONS
Table 23 describes the register settings for multiple DDC example configurations.
Table 23. DDC Example Configurations
Chip
Application
Layer
One DDC
Chip
Decimation
Ratio
2
DDC Input
Type
Complex
DDC
Output
Type
Complex
Bandwidth
Per DDC1
38.5% × fS
No. of Virtual
Converters
Required
2
One DDC
4
Complex
Complex
19.25% × fS
2
Rev. C | Page 53 of 96
Register Settings2
0x200 = 0x01 (one DDC; I/Q selected)
0x201 = 0x01 (chip decimate by 2)
0x310 = 0x83 (complex mixer; 0 dB gain;
variable IF; complex outputs; HB1 filter)
0x311 = 0x04 (DDC I input = ADC Channel A;
DDC Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x200 = 0x01 (one DDC; I/Q selected)
0x201 = 0x02 (chip decimate by 4)
0x310= 0x80 (complex mixer; 0 dB gain;
variable IF; complex outputs; HB2 + HB1 filters)
0x311= 0x04 (DDC I input = ADC Channel A;
DDC Q input = ADC Channel B)
0x314, 0x315= FTW and POW set as required by
application for DDC 0
AD6674
Data Sheet
Chip
Application
Layer
Two DDCs
Chip
Decimation
Ratio
2
DDC Input
Type
Real
DDC
Output
Type
Real
Bandwidth
Per DDC1
19.25%× fS
No. of Virtual
Converters
Required
2
Two DDCs
2
Complex
Complex
38.5%× fS
4
Two DDCs
4
Complex
Complex
19.25% × fS
4
Two DDCs
4
Complex
Real
9.63% × fS
2
Two DDCs
4
Real
Real
9.63% × fS
2
Rev. C | Page 54 of 96
Register Settings2
0x200 = 0x22 (two DDCs; I only selected)
0x201 = 0x01 (chip decimate by 2)
0x310, 0x330 = 0x48 (real mixer; 6 dB gain;
variable IF; real output; HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x05 (DDC 1 I input = ADC Channel B;
DDC 1 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x22 (two DDCs; I only selected)
0x201 = 0x01 (chip decimate by 2)
0x310, 0x330 = 0x4B (complex mixer; 6 dB gain;
variable IF; complex output; HB1 filter)
0x311, 0x331 = 0x04 (DDC 0 I input = ADC
Channel A; DDC 0 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x02 (two DDCs; I/Q selected)
0x201 = 0x02 (chip decimate by 4)
0x310, 0x330 = 0x80 (complex mixer; 0 dB gain;
variable IF; complex outputs; HB2 + HB1 filters)
0x311, 0x331 = 0x04 (DDC I input = ADC
Channel A; DDC Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x22 (two DDCs; I only selected)
0x201 = 0x02 (chip decimate by 4)
0x310, 0x330 = 0x89 (complex mixer; 0 dB gain;
variable IF; real output; HB3 + HB2 + HB1 filters)
0x311, 0x331 = 0x04 (DDC I input = ADC
Channel A; DDC Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x22 (two DDCs; I only selected)
0x201 = 0x02 (chip decimate by 4)
0x310, 0x330 = 0x49 (real mixer; 6 dB gain;
variable IF; real output; HB3 + HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x05 (DDC 1 I input = ADC Channel B;
DDC 1 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
Data Sheet
AD6674
Chip
Application
Layer
Two DDCs
Chip
Decimation
Ratio
4
DDC Input
Type
Real
DDC
Output
Type
Complex
Bandwidth
Per DDC1
19.25% × fS
No. of Virtual
Converters
Required
4
Two DDCs
8
Real
Real
4.81% × fS
2
Four DDCs
8
Real
Complex
9.63% × fS
8
Rev. C | Page 55 of 96
Register Settings2
0x200 = 0x02 (two DDCs; I/Q selected)
0x201 = 0x02 (chip decimate by 4)
0x310, 0x330 = 0x40 (real mixer; 6 dB gain;
variable IF; complex output; HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x05 (DDC 1 I input = ADC Channel B;
DDC 1 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x22 (two DDCs; I only selected)
0x201 = 0x03 (chip decimate by 8)
0x310, 0x330 = 0x4A (real mixer; 6 dB gain;
variable IF; real output; HB4 + HB3 + HB2 + HB1
filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x05 (DDC 1 I input = ADC Channel B;
DDC 1 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x200 = 0x03 (four DDCs; I/Q selected)
0x201 = 0x03 (chip decimate by 8)
0x310, 0x330, 0x350, 0x370 = 0x41 (real mixer;
6 dB gain; variable IF; complex output; HB3 +
HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x00 (DDC 1 I input = ADC Channel A;
DDC 1 Q input = ADC Channel A)
0x351 = 0x05 (DDC 2 I input = ADC Channel B;
DDC 2 Q input = ADC Channel B)
0x371 = 0x05 (DDC 3 I input = ADC Channel B;
DDC 3 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x354, 0x355, 0x360, 0x361 = FTW and POW set
as required by application for DDC 2
0x374, 0x375, 0x380, 0x381 = FTW and POW set
as required by application for DDC 3
AD6674
Data Sheet
Chip
Application
Layer
Four DDCs
Chip
Decimation
Ratio
8
DDC Input
Type
Real
DDC
Output
Type
Real
Bandwidth
Per DDC1
4.81% × fS
No. of Virtual
Converters
Required
4
Four DDCs
16
Real
Complex
4.81% × fS
8
1
2
Register Settings2
0x200 = 0x23 (four DDCs; I only selected)
0x201 = 0x03 (chip decimate by 8)
0x310, 0x330, 0x350, 0x370 = 0x4A (real mixer;
6 dB gain; variable IF; real output; HB4 + HB3 +
HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x00 (DDC 1 I input = ADC Channel A;
DDC 1 Q input = ADC Channel A)
0x351 = 0x05 (DDC 2 I input = ADC Channel B;
DDC 2 Q input = ADC Channel B)
0x371 = 0x05 (DDC 3 I input = ADC Channel B;
DDC 3 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0
0x334, 0x335, 0x340, 0x341 = FTW and POW set
as required by application for DDC 1
0x354, 0x355, 0x360, 0x361 = FTW and POW set
as required by application for DDC 2
0x374, 0x375, 0x380, 0x381 = FTW and POW set
as required by application for DDC 3
0x200 = 0x03 (four DDCs; I/Q selected)
0x201 = 0x04 (chip decimate by 16)
0x310, 0x330, 0x350, 0x370 = 0x42 (real mixer; 6
dB gain; variable IF; complex output; HB4 + HB3
+ HB2 + HB1 filters)
0x311 = 0x00 (DDC 0 I input = ADC Channel A;
DDC 0 Q input = ADC Channel A)
0x331 = 0x00 (DDC 1 I input = ADC Channel A;
DDC 1 Q input = ADC Channel A)
0x351 = 0x05 (DDC 2 I input = ADC Channel B;
DDC 2 Q input = ADC Channel B)
0x371 = 0x05 (DDC 3 I input = ADC Channel B;
DDC 3 Q input = ADC Channel B)
0x314, 0x315, 0x320, 0x321 = FTW and POW set
as required by application for DDC 0.
0x334, 0x335, 0x040, 0x341 = FTW and POW set
as required by application for DDC 1
0x354, 0x355, 0x360, 0x361 = FTW and POW set
as required by application for DDC 2
0x374, 0x375, 0x380, 0x381 = FTW and POW set
as required by application for DDC 3
fS is the ADC sample rate. Bandwidths listed are <−0.001 dB of pass-band ripple and >100 dB of stop-band alias rejection.
The NCOs must be synchronized either through the SPI or through the SYSREF± pin after all writes to the FTW or POW registers have completed. This is necessary to
ensure the proper operation of the NCO. See the NCO Synchronization section for more information.
Rev. C | Page 56 of 96
Data Sheet
AD6674
NOISE SHAPING REQUANTIZER (NSR)
10
0
–10
MAGNITUDE (dB)
–20
–30
–40
–50
–60
–80
0
DECIMATING HALF-BAND FILTER
Half-Band Filter Coefficients
The 19-tap, symmetrical, fixed coefficient half-band filter has
low power consumption due to its polyphase implementation.
Table 24 lists the coefficients of the half-band filter in low-pass
mode. In high-pass mode, Coefficient C9 is multiplied by −1.
The normalized coefficients used in the implementation and
the decimal equivalent values of the coefficients are listed.
Coefficients not listed in Table 24 are 0s.
Table 24. Fixed Coefficients for Half-Band Filter
Coefficient
Number
0
C2, C16
C4, C14
C6, C12
C8, C10
C9
Normalized
Coefficient
0.012207
−0.022949
0.045410
−0.094726
0.314453
0.500000
Decimal Coefficient
(12-Bit)
25
−47
93
−194
644
1024
0.50
Figure 111. Low-Pass Half-Band Filter Response
The half-band filter can also be utilized in high-pass mode. The
usable bandwidth remains at 39.5% of the output sample rate
(19.75% of the input sample clock), which is the same as in lowpass mode). Figure 112 shows the response of the half-band
filter in high-pass mode with an input sample clock of 1000 MHz.
In high-pass mode, operation is allowed in the second and third
Nyquist zones, which includes frequencies from fS/2 to 3 fS/2,
where fS is the decimated sample rate. For example, with an
input clock of 1000 MHz, the output sample rate is 500 MSPS,
fS/2 = 250 MHz, and 3 fS/2 = 750 MHz.
10
0
–10
MAGNITUDE (dB)
The AD6674 decimating half-band filter reduces the input
sample rate by a factor of 2 while rejecting aliases that fall into
the band of interest. For an input sample clock of 1000 MHz,
this reduces the output sample rate to 500 MSPS. This filter is
designed to provide >40 dB of alias protection for 39.5% of the
output sample rate (79% of the Nyquist band). For an ADC
sample rate of 1000 MSPS, the filter provides a maximum
usable bandwidth of 197.5 MHz.
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45
NORMALIZED FREQUENCY (× RAD/SAMPLE)
12400-484
–70
–20
–30
–40
–50
–60
–70
–80
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NORMALIZED FREQUENCY (× π RAD/SAMPLE)
1.0
12400–485
When operating the AD6674 with the NSR enabled, a decimating
half-band filter that is optimized at certain input frequency
bands can also be enabled. This filter offers the user the flexibility
in signal bandwidth process and image rejection. Careful
frequency planning can offer advantages in analog filtering
preceding the ADC. The filter can function either in high-pass
or low-pass mode. On the AD6674-750 and AD6674-1000, this
filter is nonbypassable when the NSR is enabled. The filter can
be optionally enabled on the AD6674-500 when the NSR is
enabled. When operating with NSR enabled, the decimating
half-band filter mode (low pass or high pass) is selected by
setting Bit 7 in Register 0x41E.
Figure 112. High-Pass Half-Band Filter Response
Half-Band Filter Features
NSR OVERVIEW
The half-band decimating filter is designed to provide approximately 39.5% of the output sample rate in usable bandwidth
(19.75% of the input sample clock). The filter provides >40 dB
of rejection. The response of the half-band filter in low-pass
mode is shown in Figure 111 for an input sample clock of
1000 MHz. In low-pass mode, operation is allowed in the first
Nyquist zone, which includes frequencies of up to fS/2, where fS
is the decimated sample rate. For example, with an input clock
of 1000 MHz, the output sample rate is 500 MSPS and fS/2 =
250 MHz.
The AD6674 features an NSR to allow higher than 9-bit SNR to
be maintained in a subset of the Nyquist band. The harmonic
performance of the receiver is unaffected by the NSR feature.
When enabled, the NSR contributes an additional 3.0 dB of loss
to the input signal, such that a 0 dBFS input is reduced to
−3.0 dBFS at the output pins. This loss does not degrade the SNR
performance of the AD6674.
The NSR feature can be independently controlled per channel
via the SPI.
Rev. C | Page 57 of 96
AD6674
Data Sheet
0
21% BW Mode (>75 MHz at 375 MSPS)
–40
–60
–80
–100
0
where:
f0 is the left band edge.
fADC is the ADC sample rate.
TW is the tuning word.
25
50
75
100
AIN = −1dBFS
SNR = 73.9dBFS
ENOB = 11.9 BITS
SFDR = 93dBFS
BUFFER CONTROL 1 = 1.5×
–20
–40
–60
–80
–100
–140
0
where f1 is the right band edge.
0
75
100
–60
–80
100
125
150
175
12400-266
–120
FREQUENCY (MHz)
175
28% BW Mode (>100 MHz at 375 MSPS)
f0 = fADC × 0.005 × TW
–140
150
Figure 115. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 140.3 MHz,
21% BW Mode, Tuning Word = 58
–100
75
125
The second NSR mode offers excellent noise performance
across a bandwidth that is 28% of the ADC output sample rate
(56% of the Nyquist band) and can be centered by setting the
NSR mode bits in the NSR mode register (Address 0x420) to
001. In this mode, the useful frequency range can be set using
the 6-bit tuning word in the NSR tuning register (Address 0x422).
There are 44 possible tuning words (TW, from 0 to 43); each
step is 0.5% of the ADC sample rate.
AIN = −1dBFS
SNR = 74.0dBFS
ENOB = 11.8 BITS
SFDR = 92dBFS
BUFFER ONTROL 1 = 1.5×
–40
50
FREQUENCY (MHz)
Figure 113 to Figure 115 show the typical spectrum that can be
expected from the AD6674 in the 21% BW mode for three
different tuning words.
–20
25
12400-268
–120
f1 = f0 + 0.21 × fADC
50
175
0
where fCENTER is the channel center.
25
150
Figure 114. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 90.3 MHz,
21% BW Mode, Tuning Word = 26 (fS/4 Tuning)
fCENTER = f0 + 0.105 × fADC
0
125
FREQUENCY (MHz)
AMPLITUDE (dBFS)
f0 = fADC × 0.005 × TW
–140
12400-267
–120
The first NSR mode offers excellent noise performance across a
bandwidth that is 21% of the ADC output sample rate (42% of
the Nyquist band) and can be centered by setting the NSR mode
bits in the NSR mode register (Address 0x420) to 000. In this
mode, set the useful frequency range using the 6-bit tuning
word in the NSR tuning register (Address 0x422). There are 59
possible tuning words (TW), from 0 to 58; each step is 0.5% of
the ADC sample rate.
AMPLITUDE (dBFS)
AIN = −1dBFS
SNR = 74.0dBFS
ENOB = 11.8 BITS
SFDR = 95dBFS
BUFFER ONTROL 1 = 1.5×
–20
AMPLITUDE (dBFS)
Two different bandwidth modes are provided; select the mode
from the SPI port. In each of the two modes, the center frequency
of the band can be tuned such that IFs can be placed anywhere
in the Nyquist band. The NSR feature is enabled by default on
the AD6674. The bandwidth and mode of the NSR operation
are selected by setting the appropriate bits in Register 0x420 and
Register 0x422. By selecting the appropriate profile and mode
bits in these two registers, the NSR feature can be enabled for
the desired mode of operation.
Figure 113. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 10.3 MHz,
21% BW Mode, Tuning Word = 0
where:
f0 is the left band edge.
fADC is the ADC sample rate.
TW is the tuning word.
fCENTER = f0 + 0.14 × fADC
where fCENTER is the channel center.
f1 = f0 + 0.28 × fADC
where f1 is the right band edge.
Rev. C | Page 58 of 96
Data Sheet
AD6674
0
AIN = −1dBFS
SNR = 73.0dBFS
ENOB = 11.3 BITS
SFDR = 93dBFS
BUFFER CONTROL 1 = 1.5×
–40
–60
–40
–60
–80
–100
–80
–120
–100
–140
0
25
50
75
100
125
150
175
FREQUENCY (MHz)
–140
0
25
50
75
100
125
FREQUENCY (MHz)
150
175
12400-269
–120
Figure 117. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 90.3 MHz,
28% BW Mode, Tuning Word = 19 (fS/4 Tuning)
0
Figure 116. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 10.3 MHz,
28% BW Mode, Tuning Word = 0
AIN = −1dBFS
SNR = 72.5dBFS
ENOB = 11.3 BITS
SFDR = 94dBFS
BUFFER CONTROL 1 = 1.5×
AMPLITUDE (dBFS)
–20
–40
–60
–80
–100
–120
–140
0
25
50
75
100
125
FREQUENCY (MHz)
150
175
12400-271
AMPLITUDE (dBFS)
–20
–20
AMPLITUDE (dBFS)
0
AIN = −1dBFS
SNR = 72.4dBFS
ENOB = 11.2 BITS
SFDR = 96dBFS
BUFFER CONTROL 1 = 1.5×
12400-270
Figure 116 to Figure 118 show the typical spectrum that can be
expected from the AD6674 in the 28% BW mode for three
different tuning words.
Figure 118. AD6674-750, fCLOCK = 750 MHz, fS = 375 MSPS, fIN = 140.3 MHz,
28% BW Mode, Tuning Word = 43
Rev. C | Page 59 of 96
AD6674
Data Sheet
VARIABLE DYNAMIC RANGE (VDR)
The AD6674 features a VDR digital processing block to allow
up to a 14-bit dynamic range to be maintained in a subset of the
Nyquist band. Across the full Nyquist band, a minimum 9-bit
dynamic range is available at all times. This operation is suitable
for applications such as DPD processing. The harmonic performance of the receiver is unaffected by this feature. When
enabled, VDR does not contribute loss to the input signal but
operates by effectively changing the output resolution at the
output pins. This feature can be independently controlled per
channel via the SPI.
The VDR block operates in either complex or real mode. In
complex mode, VDR has selectable bandwidths of 25% and 43%
of the output sample rate. In real mode, the bandwidth of
operation is limited to 25% of the output sample rate. The
bandwidth and mode of the VDR operation are selected by
setting the appropriate bits in Register 0x430.
When the VDR block is enabled, input signals that violate a
defined mask (signified by gray shaded areas in Figure 119)
result in the reduction of the output resolution of the AD6674.
The VDR block analyzes the peak value of the aggregate signal
level in the disallowed zones to determine the reduction of the
output resolution. To indicate that the AD6674 is reducing
output, the resolution VDR punish bits and/or a VDR high/low
resolution bit can optionally be inserted into the output data
stream as control bits by programming the appropriate value
into Register 0x559 and Register 0x55A. Up to two control bits
can be used without the need to change the converter resolution
parameter, N. Up to three control bits can be used, but if using
three, the converter resolution parameter, N, must be changed
to 13. The VDR high/low resolution bit can be programmed
into either of the three available control bits and simply
indicates if VDR is reducing output resolution (bit value is a 1),
or if full resolution is available (bit value is a 0). Enable the two
punish bits to give a clearer indication of the available
resolution of the sample. To decode these two bits, see Table 25.
Table 25. VDR Reduced Output Resolution Values
VDR Punish Bits[1:0]
00
01
10
11
Output Resolution (Bits)
14
13
12 or 11
10 or 9
The frequency zones of the mask are defined by the bandwidth
mode selected in Register 0x430. The upper amplitude limit for
input signals located in these frequency zones is −30 dBFS. If
the input signal level in the disallowed frequency zones goes
above an amplitude level of –30 dBFS (into the gray shaded
areas), the VDR block triggers a reduction in the output
resolution, as shown in Figure 119. The VDR block engages and
begins limiting output resolution gradually as the signal
amplitudes increase in the mask regions. As the signal
amplitude level increases into the mask regions, the output
resolution is gradually lowered. For every 6 dB increase in
signal level above −30 dBFS, one bit of output resolution is
discarded from the output data by the VDR block, as shown in
Table 26. These zones can be tuned within the Nyquist band by
setting Bits[3:0] in Register 0x434 to determine the VDR center
frequency (fVDR). The VDR center frequency in complex mode
can be adjusted from 1/16 fS to 15/16 fS in 1/16 fS steps. In real
mode, fVDR can be adjusted from 1/8 fS to 3/8 fS in 1/16 fS steps.
Table 26. VDR Reduced Output Resolution Values
Signal Amplitude Violating Defined
VDR Mask
Amplitude ≤ −30 dBFS
−30 dBFS < amplitude ≤ −24 dBFS
−24 dBFS < amplitude ≤ −18 dBFS
−18 dBFS < amplitude ≤ −12 dBFS
−12 dBFS < amplitude ≤ −6 dBFS
−6 dBFS < amplitude ≤ 0 dBFS
Output Resolution
(Bits)
14
13
12
11
10
9
dBFS
0
INTERMODULATION PRODUCTS < –30dBFS
fS
0
INTERMODULATION PRODUCTS > –30dBFS
Figure 119. VDR Operation—Reduction in Output Resolution
Rev. C | Page 60 of 96
fS
12400-124
–30
Data Sheet
AD6674
VDR REAL MODE
VDR COMPLEX MODE
The real mode of VDR works over a bandwidth of 25% of the
sample rate (50% of the Nyquist band). The output bandwidth
of the AD6674 can be 25% only when operating in real mode.
Figure 120 shows the frequency zones for the 25% bandwidth
real output VDR mode tuned to a center frequency (fVDR) of fS/4
(tuning word = 0x04). The frequency zones where the
amplitude may not exceed −30 dBFS are the upper and lower
portions of the Nyquist band signified by the red shaded areas.
The complex mode of VDR works with selectable bandwidths
of 25% of the sample rate (50% of the Nyquist band) and 43%
of the sample rate (86% of the Nyquist band). Figure 121 and
Figure 122 show the frequency zones for VDR in the complex
mode. When operating VDR in complex mode, place I input
signal data on Channel A and place Q input signal data on
Channel B.
dBFS
Figure 121 shows the frequency zones for the 25% bandwidth
VDR mode with a center frequency of fS/4 (tuning word =
0x04). The frequency zones where the amplitude may not
exceed –30 dBFS are the upper and lower portions of the
Nyquist band extending into the complex domain.
dBFS
–30
0
1/8 fS
3/8 fS
1/2 fS
12400-125
–1/2 fS
The center frequency (fVDR) of the VDR function can be tuned
within the Nyquist band from 1/8 fS to 3/8 fS in 1/16 fS steps. In
real mode, Tuning Word 2 (0x02) through Tuning Word 6
(0x06) are valid. Table 27 shows the relative frequency values,
and Table 28 shows the absolute frequency values based on a
sample rate of 737.28 MSPS.
Center
Frequency
1/8 fS
3/16 fS
1/4 fS
5/16 fS
3/8 fS
Upper Band
Edge
1/4 fS
5/16 fS
3/8 fS
7/16 fS
1/2 fS
Table 28. VDR Tuning Words and Absolute Frequency
Values, 25% BW, Real Mode with fS = 737.28 MSPS
Tuning
Word
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
Lower Band
Edge (MHz)
0
46.08
92.16
138.24
184.32
Center
Frequency
(MHz)
92.16
138.24
184.32
230.40
276.48
3/8 fS
1/2 fS
The center frequency (fVDR) of the VDR function can be tuned
within the Nyquist band from 0 to 15/16 fS in 1/16 fS steps. In
complex mode, Tuning Word 0 (0x00) through Tuning Word 15
(0x0F) are valid. Table 29 and Table 30 show the tuning words
and frequency values for the 25% complex mode. Table 29 shows
the relative frequency values, and Table 30 shows the absolute
frequency values based on a sample rate of 737.28 MSPS.
Table 29. VDR Tuning Words and Relative Frequency
Values, 25% BW, Complex Mode
Table 27. VDR Tuning Words and Relative Frequency
Values, 25% BW, Real Mode
Lower Band
Edge
0
1/16 fS
1/8 fS
3/16 fS
1/4 fS
1/8 fS
Figure 121. 25% VDR Bandwidth, Complex Mode
Figure 120. 25% VDR Bandwidth, Real Mode
Tuning
Word
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
0
12400-126
–30
Upper Band
Edge (MHz)
184.32
230.40
276.48
322.56
368.64
Tuning Word
0 (0x00)
1 (0x01)
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
7 (0x07)
8 (0x08)
9 (0x09)
10 (0x0A)
11 (0x0B)
12 (0x0C)
13 (0x0D)
14 (0x0E)
15 (0x0F)
Rev. C | Page 61 of 96
Lower
Band Edge
–1/8 fS
–1/16 fS
0
1/16 fS
1/8 fS
3/16 fS
1/4 fS
5/16 fS
3/8 fS
7/16 fS
1/2 fS
9/16 fS
5/8 fS
11/16 fS
3/4 fS
13/16 fS
Center
Frequency
0
1/16 fS
1/8 fS
3/16 fS
1/4 fS
5/16 fS
3/8 fS
7/16 fS
1/2 fS
9/16 fS
5/8 fS
11/16 fS
3/4 fS
13/16 fS
7/8 fS
15/16 fS
Upper Band
Edge
1/8 fS
3/16 fS
1/4 fS
5/16 fS
3/8 fS
7/16 fS
1/2 fS
9/16 fS
5/8 fS
11/16 fS
3/4 fS
13/16 fS
7/8 fS
15/16 fS
fS
17/16 fS
AD6674
Data Sheet
Table 30. VDR Tuning Words and Absolute Frequency
Values, 25% BW, Complex Mode (fS = 737.28 MSPS)
Lower
Band Edge
(MHz)
−92.16
−46.08
0.00
46.08
92.16
138.24
184.32
230.40
276.48
322.56
368.64
414.72
460.80
506.88
552.96
599.04
Tuning
Word
0 (0x00)
1 (0x01)
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
7 (0x07)
8 (0x08)
9 (0x09)
10 (0x0A)
11 (0x0B)
12 (0x0C)
13 (0x0D)
14 (0x0E)
15 (0x0F)
Center
Frequency
(MHz)
0.00
46.08
92.16
138.24
184.32
230.40
276.48
322.56
368.64
414.72
460.80
506.88
552.96
599.04
645.12
691.20
Table 31. VDR Tuning Words and Relative Frequency
Values, 43% BW, Complex Mode
Upper Band
Edge (MHz)
92.16
138.24
184.32
230.40
276.48
322.56
368.64
414.72
460.80
506.88
552.96
599.04
645.12
691.20
737.28
783.36
Tuning Word
0 (0x00)
1 (0x01)
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
7 (0x07)
8 (0x08)
9 (0x09)
10 (0x0A)
11 (0x0B)
12 (0x0C)
13 (0x0D)
14 (0x0E)
15 (0x0F)
Table 31 and Table 32 show the tuning words and frequency
values for the 43% complex mode. Table 31 shows the relative
frequency values, and Table 32 shows the absolute frequency
values based on a sample rate of 737.28 MSPS. Figure 122 shows
the frequency zones for the 43% BW VDR mode with a center
frequency (fVDR) of fS/4 (tuning word = 0x04). The frequency
zones where the amplitude may not exceed –30 dBFS are the
upper and lower portions of the Nyquist band extending into
the complex domain.
–1/2 fS
0
1/29 fS
1/4 fS
1/2 fS
20/43 fS
Figure 122. 43% VDR Bandwidth, Complex Mode
12400-127
dBFS
Lower Band
Edge (MHz)
–14/65 fS
–11/72 fS
–1/11 fS
–1/36 fS
1/29 fS
7/72 fS
4/25 fS
2/9 fS
2/7 fS
25/72 fS
34/83 fS
17/36 fS
23/43 fS
43/72 fS
31/47 fS
13/18 fS
Center
Frequency
(MHz)
0
1/16 fS
1/8 fS
3/16 fS
1/4 fS
5/16 fS
3/8 fS
7/16 fS
1/2 fS
9/16 fS
5/8 fS
11/16 fS
3/4 fS
13/16 fS
7/8 fS
15/16 fS
Upper Band
Edge (MHz)
14/65 fS
5/18 fS
16/47 fS
29/72 fS
20/43 fS
19/36 fS
49/83 fS
47/72 fS
5/7 fS
7/9 fS
21/25 fS
65/72 fS
28/29 fS
37/36 fS
12/11 fS
83/72 fS
Table 32. VDR Tuning Words and Absolute Frequency
Values, 43% BW, Complex Mode (fS = 737.28 MSPS)
Tuning Word
0 (0x00)
1 (0x01)
2 (0x02)
3 (0x03)
4 (0x04)
5 (0x05)
6 (0x06)
7 (0x07)
8 (0x08)
9 (0x09)
10 (0x0A)
11 (0x0B)
12 (0x0C)
13 (0x0D)
14 (0x0E)
15 (0x0F)
Rev. C | Page 62 of 96
Lower Band
Edge (MHz)
−158.80
−112.64
−67.03
−20.48
25.42
71.68
117.96
163.84
210.65
256.00
302.02
348.16
394.36
440.32
486.29
532.48
Center
Frequency
(MHz)
0.00
46.08
92.16
138.24
184.32
230.40
276.48
322.56
368.64
414.72
460.80
506.88
552.96
599.04
645.12
691.20
Upper Band
Edge (MHz)
158.80
204.80
250.99
296.96
342.92
389.12
435.26
481.28
526.63
573.44
619.32
665.60
711.86
757.76
804.31
849.92
Data Sheet
AD6674
DIGITAL OUTPUTS

INTRODUCTION TO JESD204B INTERFACE
The AD6674 digital outputs are designed to the JEDEC Standard
No. JESD204B serial interface for data converters. JESD204B is
a protocol to link the AD6674 to a digital processing device
over a serial interface with lane rates of up to 12.5 Gbps. The
benefits of the JESD204B interface over LVDS include a reduction
in required board area for data interface routing, and enabling
smaller packages for converter and logic devices.
JESD204B OVERVIEW
The JESD204B data transmit block assembles the parallel data
from the ADC into frames and uses 8B/10B encoding as well as
optional scrambling to form serial output data. Lane synchronization is supported through the use of special control characters
during the initial establishment of the link. Additional control
characters are embedded in the data stream to maintain
synchronization thereafter. A JESD204B receiver is required to
complete the serial link. For additional details on the JESD204B
interface, users are encouraged to refer to the JESD204B
standard.
The AD6674 JESD204B data transmit block maps up to two
physical ADCs or up to eight virtual converters (when DDCs
are enabled) over a link. A link can be configured to use one,
two, or four JESD204B lanes. The JESD204B specification refers
to a number of parameters to define the link and these parameters
must match between the JESD204B transmitter (AD6674
output) and receiver (logic device input).
The JESD204B link is described according to the following
parameters:







L is the number of lanes per converter device (lanes per
link) (AD6674 value = 1, 2, or 4)
M is the number of converters per converter device (virtual
converters per link) (AD6674 value = 1, 2, 4, or 8)
F is the number of octets per frame (AD6674 value = 1, 2,
4, 8, or 16)
N΄ is the number of bits per sample (JESD204B word size)
(AD6674 value = 8 or 16)
N is the converter resolution (AD6674 value = 7 to 16)
CS is the number of control bits per sample (AD6674 value
= 0, 1, 2, or 3)
K is the number of frames per multiframe
(AD6674 value = 4, 8, 12, 16, 20, 24, 28, or 32 )


S is the number of samples transmitted per single converter
per frame cycle (AD6674 value = set automatically based
on L, M, F, and N΄)
HD is high density mode (AD6674 = set automatically
based on L, M, F, and N΄)
CF is the number of control words per frame clock cycle
per converter device (AD6674 value = 0)
Figure 123 shows a simplified block diagram of the AD6674
JESD204B link. By default, the AD6674 is configured to use two
converters and four lanes. Converter A data is output to
SERDOUT0±/SERDOUT1±, and Converter B is output to
SERDOUT2±/SERDOUT3±. The AD6674 allows other
configurations such as combining the outputs of both
converters onto a single lane or changing the mapping of the A
and B digital output paths. These modes are set up through a
quick configuration register in the SPI register map, along with
additional customizable options.
By default in the AD6674, the 14-bit converter word from each
converter is broken into two octets (eight bits of data). Bit 13
(MSB) through Bit 6 are in the first octet. The second octet
contains Bit 5 through Bit 0 (LSB) and two tail bits. The tail bits
can be configured as zeros or a pseudorandom number (PN)
sequence. The tail bits can also be replaced with control bits
indicating an overrange, SYSREF±, signal monitor output, VDR
punish bits, or fast detect output. Control bits are filled and
inserted MSB first, such that enabling CS = 1 activates Control
Bit 2, enabling CS = 2 activates Control Bit 2 and Control Bit 1,
and enabling CS = 3 activates Control Bit 2, Control Bit 1, and
Control Bit 0.
The two resulting octets can be scrambled. Scrambling is
optional; however, it is recommended to avoid spectral peaks
when transmitting similar digital data patterns. The scrambler
uses a self synchronizing, polynomial-based algorithm defined
by the equation 1 + x14 + x15. The descrambler in the receiver
must be a self synchronizing version of the scrambler polynomial.
The two octets are then encoded with an 8B/10B encoder. The
8B/10B encoder works by taking eight bits of data (an octet) and
encoding them into a 10-bit symbol. Figure 124 shows how the
14-bit data is transferred from the ADC, the tail bits are added,
the two octets are scrambled, and the octets are encoded into
two 10-bit symbols. Figure 124 illustrates the default data format.
Rev. C | Page 63 of 96
AD6674
Data Sheet
CONVERTER 0
CONVERTER A
INPUT
ADC
A
CONVERTER B
INPUT
ADC
B
NOISE
SHAPING
REQUANTIZER
MUX/
FORMAT
(SPI
REG 0x561,
REG 0x564)
JESD204B LINK
CONTROL
(L, M, F)
(SPI REG 0x570)
LANE MUX
AND MAPPING
(SPI
REG 0x5B0,
REG 0x5B2,
REG 0x5B3,
REG 0x5B5,
REG 0x5B6)
NOISE
SHAPING
REQUANTIZER
SERDOUT0–,
SERDOUT0+
SERDOUT1–,
SERDOUT1+
SERDOUT2–,
SERDOUT2+
SERDOUT3–,
SERDOUT3+
12400-050
CONVERTER 1
SYSREF±
SYNCINB±
Figure 123. Transmit Link Simplified Block Diagram Showing NSR Mode (Register 0x200 = 0x07)
JESD204B
INTERFACE
TEST PATTERN
(REG 0x573,
REG 0x551 TO
REG 0x558)
JESD204B
LONG TRANSPORT
TEST PATTERN
REG 0x571[5]
SERDOUT0±
SERDOUT1±
SERIALIZER
MSB A13
A12
A11
A10
A9
A8
A7
LSB A6
A5
A4
A3
A2
A1
A0
C2
T
MSB S7
S6
S5
S4
S3
S2
S1
LSB S0
S7
S6
S5
S4
S3
S2
S1
S0
8-BIT/10-BIT
ENCODER
a b
a b c d e f g h i j
i j a b
SYMBOL0
i j
SYMBOL1
a b c d e f g h i j
12400-051
TAIL BITS
0x571[6]
SCRAMBLER
1 + x14 + x15
(OPTIONAL)
OCTET 1
JESD204B SAMPLE
CONSTRUCTION
MSB A13
A12
A11
A10
A9
A8
A7
A6
A5
A4
A3
A2
A1
LSB A0
OCTET 1
OCTET 0
FRAME
CONSTRUCTION
OCTET 0
ADC TEST PATTERNS
(RE0x550,
REG 0x551 TO
REG 0x558)
ADC
JESD204B DATA
LINK LAYER TEST
PATTERNS
REG 0x574[2:0]
C2
CONTROL BITS C1
C0
Figure 124. ADC Output Datapath Showing Data Framing
TRANSPORT
LAYER
SAMPLE
CONSTRUCTION
FRAME
CONSTRUCTION
SCRAMBLER
ALIGNMENT
CHARACTER
GENERATION
8-BIT/10-BIT
ENCODER
PHYSICAL
LAYER
CROSSBAR
MUX
SERIALIZER
Tx
OUTPUT
12400-052
PROCESSED
SAMPLES
FROM ADC
DATA LINK
LAYER
SYSREF±
SYNCINB±
Figure 125. Data Flow
FUNCTIONAL OVERVIEW
Data Link Layer
The block diagram in Figure 125 shows the flow of data through
the JESD204B hardware from the sample input to the physical
output. The processing can be divided into layers that are derived
from the open-source initiative (OSI) model that is widely used
to describe the abstractions layers of communications systems.
These are the transport layer, data link layer, and physical layer
(serializer and output driver).
The data link layer is responsible for the low level functions of
passing data across the link. These include optionally
scrambling the data, inserting control characters for multichip
synchronization/lane alignment/monitoring, and encoding
8-bit octets into 10-bit symbols. The data link layer is also
responsible for sending the ILAS, which contains the link
configuration data, used by the receiver to verify the settings in
the transport layer.
Transport Layer
The transport layer packs the data (consisting of samples and
optional control bits) into JESD204B frames, which are mapped
to 8-bit octets that are sent to the data link layer. The transport
layer mapping is controlled by rules derived from the link
parameters. Tail bits are added to fill gaps where required. Use
the following equation to determine the number of tail bits
within a sample (JESD204B word):
T = N΄ − N − CS
Physical Layer
The physical layer consists of the high speed circuitry clocked at
the serial clock rate. In this section, parallel data is converted
into one, two, or four lanes of high speed differential serial data.
JESD204B LINK ESTABLISHMENT
The AD6674 JESD204B Tx interface operates in Subclass 1 as
defined in the JEDEC Standard No. 204B (July 2011) specification.
The link establishment process is divided into the following
steps: code group synchronization, ILAS, and user data.
Rev. C | Page 64 of 96
Data Sheet
AD6674
Code Group Synchronization (CGS) and SYNCINB±
The ILAS sequence construction is shown in Figure 126. The
four multiframes include the following:
Code group synchronization (CGS) is the process by which the
JESD204B receiver finds the boundaries between the 10-bit
symbols in the stream of data. During the CGS phase, the
JESD204B transmit block transmits /K28.5/ characters. The
receiver must locate /K28.5/ characters in its input data stream
using clock and data recovery (CDR) techniques.


The receiver issues a synchronization request by asserting the
SYNCINB± pin of the AD6674 low. The JESD204B Tx begins
sending /K/ characters. After the receiver has synchronized, it
waits for the correct reception of at least four consecutive /K/
symbols. It then deasserts SYNCINB±. The AD6674 then
transmits an ILAS on the following LMFC boundary.


For more information on the CGS phase, refer to the JEDEC
Standard No. 204B (July 2011), Section 5.3.3.1.
User Data and Error Detection
The SYNCINB± pin operation can also be controlled by the
SPI. The SYNCINB± signal is a differential LVDS mode signal
by default, but it can also be driven single-ended. For more
information on configuring the SYNCINB± pin operation, refer
to Register 0x572. The SYNCINB± pin can also be configured
to run in CMOS (single-ended) mode by setting Bit 4 in
Register 0x572. When running SYNCINB± in CMOS mode,
connect the CMOS SYNCINB signal to Pin 21 (SYNCINB+)
and leave Pin 20 (SYNCINB–) floating.
Initial Lane Alignment Sequence (ILAS)
The ILAS phase follows the CGS phase and begins on the next
LMFC boundary. The ILAS consists of four multiframes, with
an /R/ character marking the beginning and an /A/ character
marking the end. The ILAS begins by sending an /R/ character
followed by 0 to 255 ramp data for one multiframe. On the
second multiframe the link configuration data is sent, starting
with the third character. The second character is a /Q/ character
to confirm that the link configuration data follows. All undefined
data slots are filled with ramp data. The ILAS sequence is never
scrambled.
K K R D
D A R Q C
C D
Multiframe 1: Begins with an /R/ character [K28.0] and
ends with an /A/ character (K28.3).
Multiframe 2: Begins with an /R/ character followed by a
/Q/ (K28.4) character, followed by link configuration
parameters over 14 configuration octets (see Table 33), and
ends with an /A/ character. Many of the parameter values
are of the value − 1 notation.
Multiframe 3: Begins with an /R/ character (K28.0) and
ends with an /A/ character (K28.3).
Multiframe 4: Begins with an /R/ character (K28.0) and
ends with an /A/ character (K28.3).
After the ILAS is complete, the user data is sent. Normally, in a
frame all characters are user data. However, to monitor the frame
clock and multiframe clock synchronization, there is a mechanism
for replacing characters with /F/ or /A/ alignment characters
when the data meets certain conditions. These conditions are
different for unscrambled and scrambled data. The scrambling
operation is enabled by default but can be disabled using the SPI.
For scrambled data, any 0xFC character at the end of a frame is
replaced by an /F/ and any 0x7C character at the end of a
multiframe is replaced with an /A/. The JESD204B Rx checks
for /F/ and /A/ characters in the received data stream and verifies
that they only occur in the expected locations. If an unexpected
/F/ or /A/ character is found, the receiver handles the situation
by using dynamic realignment or asserting the SYNCINB±
signal for more than four frames to initiate a resynchronization.
For unscrambled data, if the final character of two subsequent
frames is equal, the second character is replaced with an /F/ if it
is at the end of a frame, and an /A/ if it is at the end of a multiframe.
Insertion of alignment characters can be modified using the SPI.
The frame alignment character insertion is enabled by default. For
more information on the link controls, see Register 0x571 in the
Memory Map section.
D A R D
D A R D
D A D
START OF
ILAS
START OF LINK
CONFIGURATION DATA
START OF
USER DATA
12400-053
END OF
MULTIFRAME
Figure 126. Initial Lane Alignment Sequence
Table 33. AD6674 Control Characters Used in JESD204B
Abbreviation
/R/
/A/
/Q/
/K/
/F/
1
Control Symbol
K28.0
K28.3
K28.4
K28.5
K28.7
8-Bit Value
000 11100
011 11100
100 11100
101 11100
111 11100
10-Bit Value, RD1 = −1
001111 0100
001111 0011
001111 0010
001111 1010
001111 1000
RD is running disparity.
Rev. C | Page 65 of 96
10-Bit Value, RD1 = +1
110000 1011
110000 1100
110000 1101
110000 0101
110000 0111
Description
Start of multiframe
Lane alignment
Start of link configuration data
Group synchronization
Frame alignment
AD6674
Data Sheet
8B/10B Encoder
The 8B/10B encoder converts 8-bit octets into 10-bit symbols
and inserts control characters into the stream when needed.
The control characters used in JESD204B are shown in Table 33.
The 8B/10B encoding ensures that the signal is dc balanced by
using the same number of ones and zeros across multiple symbols.
The 8B/10B interface has options that can be controlled via the
SPI. These operations include bypass and invert. These options
are intended to be a troubleshooting tool for the verification of
the digital front end (DFE). Refer to the Memory Map section,
Register 0x572[2:1], for information on configuring the 8B/10B
encoder.
The AD6674 digital outputs can interface with custom ASICs
and FPGA receivers, providing superior switching performance
in noisy environments. Single point-to-point network topologies
are recommended with a single differential 100 Ω termination
resistor placed as close to the receiver inputs as possible. The
common mode of the digital output automatically biases itself
to half the DRVDD supply of 1.2 V (VCM = 0.6 V). See Figure 128
for an example of dc coupling the outputs to the receiver logic.
DRVDD
100Ω
DIFFERENTIAL
TRACE PAIR
SERDOUTx+
100Ω
RECEIVER
SERDOUTx–
Digital Outputs, Timing and Controls
OUTPUT SWING = 300mV p-p
The AD6674 physical layer consists of drivers that are defined
in the JEDEC Standard No. 204B (July 2011). The differential
digital outputs are powered up by default. The drivers use a
dynamic 100 Ω internal termination to reduce unwanted
reflections.
Figure 128. DC-Coupled Digital Output Termination Example
If there is no far-end receiver termination, or if there is poor
differential trace routing, timing errors may result. To avoid
such timing errors, it is recommended that the trace length be
less than six inches, and that the differential output traces be
close together and at equal lengths.
Place a 100 Ω differential termination resistor at each receiver
input, which results in a nominal 300 mV p-p swing at the
receiver (see Figure 127). Alternatively, single-ended 50 Ω
termination resistors can be used. When single-ended
termination is used, the termination voltage is DRVDD/2;
otherwise, 0.1 μF ac coupling capacitors can be used to
terminate to any single-ended voltage.
Figure 129 to Figure 131, Figure 132 to Figure 134, and Figure 135
to Figure 137 show examples of the digital output data eye, time
interval error (TIE) jitter histogram, and bathtub curve for one
AD6674 lane running at 10 Gbps, 7.37 Gbps, and 6 Gbps,
respectively. The format of the output data is twos complement
by default. To change the output data format, see the Memory
Map section (Register 0x561 in Table 45).
VRXCM
DRVDD
100Ω
DIFFERENTIAL
0.1µF TRACE PAIR
50Ω
50Ω
De-Emphasis
VCM = VRXCM
De-emphasis enables the receiver eye diagram mask to be met
in conditions where the interconnect insertion loss does not
meet the JESD204B specification. Use the de-emphasis feature
only when the receiver is unable to recover the clock due to
excessive insertion loss. Under normal conditions, it is disabled
to conserve power. Additionally, enabling and setting too high a
de-emphasis value on a short link may cause the receiver eye
diagram to fail. Use the de-emphasis setting with caution
because it may increase EMI. See the Memory Map section
(Register 0x5C1 to Register 0x5C5 in Table 45) for more
information.
SERDOUTx+
100Ω
OR
RECEIVER
0.1µF
OUTPUT SWING = 300mV p-p
Figure 127. AC-Coupled Digital Output Termination Example
12400-054
SERDOUTx–
VCM = DRVDD/2
12400-055
PHYSICAL LAYER (DRIVER) OUTPUTS
PLL
The PLL is used to generate the serializer clock, which operates
at the JESD204B lane rate. The JESD204B lane rate control bit
(Register 0x56E[4]) must be set to correspond with the lane rate.
Rev. C | Page 66 of 96
Data Sheet
AD6674
12400-588
12400-593
Tx EYE
MASK
Figure 129. Digital Output Data Eye, External 100 Ω Terminations at 10 Gbps
Figure 134. Bathtub, External 100 Ω Terminations at 7.37 Gbps
12400-594
12400-589
Tx EYE MASK
Figure 130. Histogram, External 100 Ω Terminations at 10 Gbps
12400-595
12400-590
Figure 135. Digital Output Data Eye, External 100 Ω Terminations at 6 Gbps
Figure 131. Bathtub, External 100 Ω Terminations at 10 Gbps
12400-596
12400-591
Figure 136. Histogram, External 100 Ω Terminations at 6 Gbps
12400-592
Figure 132. Digital Output Data Eye, External 100 Ω Terminations at 7.37 Gbps
Figure 133. Histogram, External 100 Ω Terminations at 7.37 Gbps
Rev. C | Page 67 of 96
Figure 137. Bathtub, External 100 Ω Terminations at 6 Gbps
AD6674
Data Sheet
ADC A
SAMPLING
AT fS
REAL/I
REAL/Q
REAL/I
REAL/Q
I/Q
CROSSBAR
MUX
REAL/I
REAL/Q
REAL/Q
ADC B
SAMPLING
AT fS
REAL/I
REAL/Q
DDC 0
I
I
Q
Q
DDC 1
I
I
Q
Q
DDC 2
I
I
Q
Q
DDC 3
I
I
Q
Q
REAL/I
CONVERTER 0
Q
CONVERTER 1
REAL/I
CONVERTER 2
Q
CONVERTER 3
OUTPUT
INTERFACE
REAL/I
CONVERTER 4
Q
CONVERTER 5
REAL/I
CONVERTER 6
Q
CONVERTER 7
12400-059
REAL/I
Figure 138. DDCs and Virtual Converter Mapping
JESD204B Tx CONVERTER MAPPING
CONFIGURING THE JESD204B LINK
To support the different chip operating modes, the AD6674
design treats each sample stream (real or I/Q) as originating
from separate virtual converters. The I/Q samples are always
mapped in pairs with the I samples mapped to the first virtual
converter, and the Q samples mapped to the second virtual
converter. With this transport layer mapping, the number of
virtual converters are the same whether a single real converter is
used along with a DDC block producing I/Q outputs, or an
analog downconversion is used with two real converters
producing I/Q outputs.
The AD6674 has one JESD204B link. It offers an easy way to
set up the JESD204B link through the quick configuration
register (Register 0x570). The serial outputs (SERDOUT0± to
SERDOUT3±) are considered to be part of one JESD204B link.
The basic parameters that determine the link setup are
Figure 139 shows a block diagram of the two scenarios
described for I/Q transport layer mapping.
I
CONVERTER 0
ADC
REAL
DIGITAL
DOWN
CONVERSION
JESD204B
Tx
If the internal DDCs are used for on-chip digital processing,
the M value represents the number of virtual converters. The
virtual converter mapping setup is shown in Figure 138.


 10 
 M  N '     fOUT 
8
 

Lane Line Rate  
L
L LANES
Q
CONVERTER 1
where:
I/Q ANALOG MIXING
M=2
REAL
Σ
ADC
I
CONVERTER 0
90°
PHASE
Q
f OUT 
JESD204B
Tx
ADC
Q
CONVERTER 1
L LANES
12400-058
I
Number of lanes per link (L)
Number of converters per link (M)
Number of octets per frame (F)
The maximum lane rate allowed by the JESD204B specification
is 12.5 Gbps. The lane rate is related to the JESD204B
parameters using the following equation:
DIGITAL DOWNCONVERSION
M=2
REAL



Figure 139. I/Q Transport Layer Mapping
The JESD204B Tx block for AD6674 supports up to four digital
DDC blocks. Each DDC block outputs either two sample streams
(I/Q) for the complex data components (real + imaginary) or
one sample stream for real (I) data. The JESD204B interface can
be configured to use up to eight virtual converters depending
on the DDC configuration. Figure 138 shows the virtual converters
and their relationship to DDC outputs when complex outputs
are used. Table 34 shows the virtual converter mapping for each
chip operating mode when channel swapping is disabled.
f ADC _ CLOCK
Decimation Ratio
The decimation ratio (DCM) is the parameter programmed in
Register 0x201.
Use the following steps to configure the output:
1.
2.
3.
4.
5.
6.
Power down the link.
Select the quick configuration options.
Configure detailed options.
Set output lane mapping (optional).
Set additional driver configuration options (optional).
Power up the link.
If the lane rate calculated is less than 6.25 Gbps, select the low
lane rate option by programming a value of 0x10 to Register 0x56E.
Rev. C | Page 68 of 96
Data Sheet
AD6674
See the Example 1: ADC with DDC Option (Two ADCs + Four
DDCs) section and the Example 2: ADC with NSR Option
(Two ADCs + NSR) section for two examples describing which
JESD204B transport layer settings are valid for a given chip
mode.
Table 35 and Table 36 show the JESD204B output configurations supported for both N΄ = 16 and N΄ = 8, respectively, for a
given number of virtual converters. Take care to ensure that the
serial lane rate for a given configuration is within the supported
range of 3.125 Gbps to 12.5 Gbps.
Table 34. Virtual Converter Mapping
No. of
Virtual
Converters
Supported
1
2
2
4
4
8
1 to 2
1 to 2
Chip
Operating
Mode
(Register
0x200[3:0])
One DDC
mode (0x1)
One DDC
mode (0x1)
Two DDC
mode (0x2)
Two DDC
mode (0x2)
Four DDC
mode (0x3)
Four DDC
mode (0x3)
NSR mode
(0x7)
VDR mode
(0x8)
Virtual Converter Mapping
Chip Q
Ignore
(Register
0x200[5])
Real (I only)
(0x1)
Complex
(I/Q) (0x0)
Real (I only)
(0x1)
Complex
(I/Q) (0x0)
Real (I only)
(0x1)
Complex
(I/Q) (0x0)
Real or
complex
(0x0)
Real or
complex
(0x0)
0
DDC 0 I
samples
DDC 0 I
samples
DDC 0 I
samples
DDC 0 I
samples
DDC 0 I
samples
DDC 0 I
samples
ADC A
Samples
1
Unused
2
Unused
3
Unused
4
Unused
5
Unused
6
Unused
7
Unused
DDC 0 Q
samples
DDC 1 I
samples
DDC 0 Q
samples
DDC 1 I
samples
DDC 0 Q
samples
ADC B
Samples
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
DDC 1 I
samples
DDC 2 I
samples
DDC 1 I
samples
Unused
DDC 1 Q
samples
DDC 3 I
samples
DDC 1 Q
samples
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
Unused
DDC 2 I
samples
Unused
DDC 2 Q
samples
Unused
DDC 3 I
samples
Unused
DDC 3 Q
samples
Unused
ADC A
Samples
ADC B
Samples
Unused
Unused
Unused
Unused
Unused
Unused
Table 35. JESD204B Output Configurations for N΄ = 16
Number of Virtual
Converters Supported
(Same Value as M)
1
2
4
8
JESD204B Quick
Configuration
(Register 0x570)
0x01
0x40
0x41
0x80
0x81
0x0A
0x49
0x88
0x89
0x13
0x52
0x91
0x1C
0x5B
0x9A
JESD204B
Serial Lane
Rate1
20 × fOUT
10 × fOUT
10 × fOUT
5 × fOUT
5 × fOUT
40 × fOUT
20 × fOUT
10 × fOUT
10 × fOUT
80 × fOUT
40 × fOUT
20 × fOUT
160 × fOUT
80 × fOUT
40 × fOUT
JESD204B Transport Layer Settings2
L
1
2
2
4
4
1
2
4
4
1
2
4
1
2
4
M
1
1
1
1
1
2
2
2
2
4
4
4
8
8
8
F
2
1
2
1
2
4
2
1
2
8
4
2
16
8
4
S
1
1
2
2
4
1
1
1
2
1
1
1
1
1
1
HD
0
1
0
1
0
0
0
1
0
0
0
0
0
0
0
N
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
8 to 16
N΄
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
CS
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
0 to 3
K3
Only valid K values
that are divisible by 4
are supported
1
fOUT is the output sample rate. fOUT = ADC sample rate/chip decimation. The JESD204B serial lane rate must be ≥3.125 Gbps and ≤12.5 Gbps; when the serial lane rate is
≤12.5 Gbps and ≥6.25 Gbps, the low lane rate mode must be disabled (set Bit 4 to 0x0 in Register 0x56E). When the serial lane rate is <6.25 Gbps and ≥3.125 Gbps, the
low lane rate mode must be enabled (set Bit 4 to 0x1 in Register 0x56E).
2
JESD204B transport layer descriptions are as described in the JESD204B Overview section.
3
For F = 1, K = 20, 24, 28, and 32. For F = 2, K = 12, 16, 20, 24, 28, and 32. For F = 4, K = 8, 12, 16, 20, 24, 28, and 32. For F = 8 and F = 16, K = 4, 8, 12, 16, 20, 24, 28, and 32.
Rev. C | Page 69 of 96
AD6674
Data Sheet
Table 36. JESD204B Output Configurations for N΄ = 8
Number of Virtual
Converters Supported
(Same Value as M)
1
2
JESD204B Quick
Configuration
(Register 0x570)
0x00
0x01
0x40
0x41
0x42
0x80
0x81
0x09
0x48
0x49
0x88
0x89
0x8A
JESD204B Transport Layer Settings2
Serial Lane Rate1
10 × fOUT
10 × fOUT
5 × fOUT
5 × fOUT
5 × fOUT
2.5 × fOUT
2.5 × fOUT
20 × fOUT
10 × fOUT
10 × fOUT
5 × fOUT
5 × fOUT
5 × fOUT
L
1
1
2
2
2
4
4
1
2
2
4
4
4
M
1
1
1
1
1
1
1
2
2
2
2
2
2
F
1
2
1
2
4
1
2
2
1
2
1
2
4
S
1
2
2
4
8
4
8
1
1
2
2
4
8
HD
0
0
0
0
0
0
0
0
0
0
0
0
0
N
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
7 to 8
N΄
8
8
8
8
8
8
8
8
8
8
8
8
8
CS
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
0 to 1
K3
Only valid K
values that
are divisible
by 4 are
supported
1
fOUT is the output sample rate. fOUT = ADC sample rate/chip decimation. The JESD204B serial lane rate must be ≥3.125 Gbps and ≤12.5 Gbps; when the serial lane rate is
≤12.5 Gbps and ≥6.25 Gbps, the low lane rate mode must be disabled (set Bit 4 to 0x0 in Register 0x56E). When the serial lane rate is <6.25 Gbps and ≥3.125 Gbps, the
low lane rate mode must be enabled (set Bit 4 to 0x1 in Register 0x56E).
2
JESD204B transport layer descriptions are as described in the JESD204B Overview section.
3
For F = 1, K = 20, 24, 28, and 32. For F = 2, K = 12, 16, 20, 24, 28, and 32. For F = 4, K = 8, 12, 16, 20, 24, 28, and 32. For F = 8 and F = 16, K = 4, 8, 12, 16, 20, 24, 28, and 32.
Example 1: ADC with DDC Option (Two ADCs + Four DDCs)
Example 2: ADC with NSR Option (Two ADCs + NSR)
The chip application mode is four-DDC mode (see Figure 140)
with the following characteristics:
The chip application mode is NSR mode (see Figure 141) with
the following characteristics:







Two 14-bit converters at 1 GSPS
Four DDCs application layer mode with complex outputs
(I/Q)
Chip decimation ratio = 16
DDC decimation ratio = 16 (see Table 15)
The JESD204B output configuration is as follows:


Virtual converters required = 8 (see Table 35)
Output sample rate (fOUT) = 1000/16 = 62.5 MSPS





The JESD204B output configuration is as follows:


Virtual converters required = 2 (see Table 35)
Output sample rate (fOUT) = 500 MSPS
Supported JESD204B output configurations (see Table 35)
include
Supported JESD204B output configurations (see Table 35)
include



Two 14-bit converters at 500 MSPS
NSR blocks enabled for each channel
Chip decimation ratio = 1
N΄ = 16 bits
N = 14 bits
L = 1, M = 8, and F = 16; or L = 2, M = 8, and F = 8 (quick
configuration = 0x1C or 0x5B)
CS = 0 to 1
K = 32
Output serial lane rate = 10 Gbps per lane (L = 1) or
5 Gbps per lane (L = 2)
For L = 1, low lane rate mode disabled
For L = 2, low lane rate mode enabled
Example 1 shows the flexibility in the digital and lane
configurations for the AD6674. The sample rate is 1 GSPS, but
the outputs are all combined into either one or two lanes
depending on the I/O speed capability of the receiving device.








N΄ = 16 bits
N = 9 bits
L = 2, M = 2, and F = 2; L = 4, M = 2, and F = 1 (quick
configuration = 0x49 or 0x88)
CS = 0 to 2
K = 32
Output serial lane rate = 10 Gbps per lane (L = 2) or
5 Gbps per lane (L = 4)
For L = 2, low lane rate mode disabled
For L = 4, low lane rate mode enabled
Example 2 shows the flexibility in the digital and lane
configurations for the AD6674. The sample rate is 500 MSPS,
but the outputs are all combined into either two or four lanes
depending on the I/O speed capability of the receiving device.
Rev. C | Page 70 of 96
Data Sheet
AD6674
ADC A
SAMPLING
AT fS
REAL/I
REAL/Q
DDC 0
I
CONVERTER 0
Q
CONVERTER 1
DDC 1
I
CONVERTER 2
Q
CONVERTER 3
DDC 2
I
CONVERTER 4
Q
CONVERTER 5
DDC 3
I
CONVERTER 6
Q
CONVERTER 7
I/Q
CROSSBAR
MUX
REAL/I
REAL/Q
L JESD204B
LANES UP TO
10Gbps
12400-061
SYSREF±
ADC B
SAMPLING
AT fS
L
JESD204B
LANES
AT UP TO
10Gbps
SYNCHRONIZATION
CONTROL CIRCUITS
Figure 140. Two-ADC + Four-DDC Mode
CMOS
FAST
DETECTION
REAL
REAL
AD6674
14-BIT CORE
AT 500MSPS
NSR
(21% OR 28%
BANDWIDTH)
AD6674
14-BIT CORE
AT 500MSPS
NSR
(21% OR 28%
BANDWIDTH)
CONVERTER 0
AT 500MSPS
CONVERTER 1
AT 500MSPS
2 TO 4
LANES
AT UP TO
10Gbps
FAST
DETECTION
12400-177
REAL
JESD204B TRANSMIT
INTERFACE (JTX)
REAL
CMOS
Figure 141. Two-ADC + NSR Mode
Rev. C | Page 71 of 96
AD6674
Data Sheet
MULTICHIP SYNCHRONIZATION
The AD6674 has a SYSREF± input that allows the user flexible
options for synchronizing the internal blocks. The SYSREF±
input is a source synchronous system reference signal that
enables multichip synchronization. The input clock divider,
DDCs, signal monitor block, and JESD204B link can be
synchronized using the SYSREF± input. For the highest level of
timing accuracy, SYSREF± must meet setup and hold requirements relative to the CLK± input.
The flowchart in Figure 142 describes the internal mechanism
by which multichip synchronization can be achieved in the
AD6674. The AD6674 supports several features that aid users in
meeting the requirements for capturing a SYSREF± signal. The
SYSREF± sample event is defined as either a synchronous low to
high transition or a synchronous high to low transition. Additionally, the AD6674 allows the SYSREF± signal to be sampled
using either the rising edge or falling edge of the CLK± input.
The AD6674 also has the ability to ignore a programmable
number (up to 16) of SYSREF± events. The SYSREF± control
options can be selected using Register 0x120 and Register 0x121.
Rev. C | Page 72 of 96
Data Sheet
AD6674
START
INCREMENT
SYSREF± IGNORE
COUNTER
NO
NO
RESET
SYSREF± IGNORE
COUNTER
SYSREF±
ENABLED?
(0x120)
NO
NO
SYSREF±
ASSERTED?
YES
UPDATE
SETUP/HOLD
DETECTOR STATUS
(0x128)
YES
SYSREF±
IGNORE
COUNTER
EXPIRED?
(0x121)
YES
ALIGN CLOCK
DIVIDER
PHASE TO
SYSREF±
INPUT
CLOCK
DIVIDER
ALIGNMENT
REQUIRED?
YES
YES
NO
SYNCHRONIZATION
MODE?
(0x1FF)
CLOCK
DIVIDER
AUTO ADJUST
ENABLED?
(0x10D)
NO
TIMESTAMP
MODE
SYSREF±
TIMESTAMP
DELAY
(0x123)
INCREMENT
SYSREF±
COUNTER
(0x12A)
CLOCK
DIVIDER
> 1?
(0x10B)
YES
NO
SYSREF±
CONTROL BITS?
(0x559, 0x55A,
0x58F)
YES
SYSREF±
INSERTED
IN JESD204B
CONTROL BITS
NO
RAMP
TEST
MODE
ENABLED?
(0x550)
NORMAL
MODE
YES
SYSREF± RESETS
RAMP TEST
MODE
GENERATOR
BACK TO START
NO
YES
ALIGN PHASE
OF ALL
INTERNAL CLOCKS
(INCLUDING LMFC)
TO SYSREF±
SEND INVALID
8B/10B
CHARACTERS
(ALL 0s)
SYNC~
ASSERTED
NO
SEND K28.5
CHARACTERS
NORMAL
JESD204B
INITIALIZATION
NO
NO
SIGNAL
MONITOR
ALIGNMENT
ENABLED?
(0x26F)
YES
YES
ALIGN SIGNAL
MONITOR
COUNTERS
DDC NCO
ALIGNMENT
ENABLED?
(0x300)
YES
NO
Figure 142. Multichip Synchronization
Rev. C | Page 73 of 96
ALIGN DDC
NCO PHASE
ACCUMULATOR
BACK TO START
12400-509
JESD204B
LMFC
ALIGNMENT
REQUIRED?
AD6674
Data Sheet
SYSREF± SETUP/HOLD WINDOW MONITOR
To assist in ensuring a valid SYSREF± capture, the AD6674 has
a SYSREF± setup and hold window monitor. This feature allows
the system designer to determine the location of the SYSREF±
signals relative to the CLK± signals by reading back the amount
of setup/hold margin on the interface through the memory
map. Figure 143 and Figure 144 show both the setup and hold
status values for different phases of SYSREF±. The setup
detector returns the status of the SYSREF± signal before the
CLK± edge and the hold detector returns the status of the
SYSREF± signal after the CLK± edge. Register 0x128 stores the
status of SYSREF± and lets the user know if the SYSREF± signal
was successfully captured by the ADC.
0xF
0xE
0xD
0xC
0xB
0xA
0x9
REG 0x128[3:0] 0x8
0x7
0x6
0x5
0x4
0x3
0x2
0x1
0x0
CLK±
INPUT
SYSREF±
INPUT
VALID
FLIP-FLOP
HOLD (MIN)
FLIP-FLOP
HOLD (MIN)
Figure 143. SYSREF± Setup Detector
Rev. C | Page 74 of 96
12400-510
FLIP-FLOP
SETUP (MIN)
Data Sheet
AD6674
0xF
0xE
0xD
0xC
0xB
0xA
0x9
REG 0x128[7:4] 0x8
0x7
0x6
0x5
0x4
0x3
0x2
0x1
0x0
CLK±
INPUT
SYSREF±
INPUT
FLIP-FLOP
SETUP (MIN)
FLIP-FLOP
HOLD (MIN)
FLIP-FLOP
HOLD (MIN)
12400-511
VALID
Figure 144. SYSREF± Hold Detector
Table 37 shows the description of the contents of Register 0x128 and how to interpret them.
Table 37. SYSREF± Setup/Hold Monitor, Register 0x128
Register 0x128[7:4] Hold
Status
0x0
0x0 to 0x8
0x8
0x8
0x9 to 0xF
0x0
Register 0x128[3:0] Setup
Status
0x0 to 0x7
0x8
0x9 to 0xF
0x0
0x0
0x0
Description
Possible setup error; the smaller this number, the smaller the setup margin
No setup or hold error (best hold margin)
No setup or hold error (best setup and hold margin)
No setup or hold error (best setup margin)
Possible hold error; the larger this number, the smaller the hold margin
Possible setup or hold error
Rev. C | Page 75 of 96
AD6674
Data Sheet
TEST MODES
ADC TEST MODES
JESD204B BLOCK TEST MODES
The AD6674 has various test options that aid in the system level
implementation. The AD6674 has ADC test modes that are
available in Register 0x550. These test modes are described in
Table 38. When an output test mode is enabled, the analog
section of the ADC is disconnected from the digital back-end
blocks and the test pattern is run through the output formatting
block. Some of the test patterns are subject to output formatting,
and some are not. The PN generators from the PN sequence
tests can be reset by setting Bit 4 or Bit 5 of Register 0x550.
These tests can be performed with or without an analog signal
(if present, the analog signal is ignored), but they do require an
encode clock.
In addition to the ADC test modes, the AD6674 also has flexible
test modes in the JESD204B block. These test modes are listed
in Register 0x573 and Register 0x574. These test patterns can be
inserted at various points along the output data path. These test
insertion points are shown in Figure 124. Table 39 describes the
various test modes available in the JESD204B block. For the
AD6674, a transition from the test modes (Register 0x573 ≠
0x00) to normal mode (0x573 = 0x00) require a SPI soft reset.
This is done by writing 0x81 to Register 0x00 (self cleared).
If the application mode is set to select a DDC mode of operation,
the test modes must be enabled for each DDC enabled. The test
patterns can be enable via Bit 2 and Bit 0 of Register 0x327,
Register 0x347, and Register 0x367, depening on which DDC(s)
are selected. The (I) data uses the test patterns selected for Channel
A and the (Q) data uses the test patterns selected for Channel B.
For the case of DDC3 only, the (I) data uses the test patterns from
Channel A, and the (Q) data does not output test patterns. Bit 0 of
Register 0x387 selects the Channel A test patterns to be used for the
(I) data. For more information, see the AN-877 Application Note,
Interfacing to High Speed ADCs via SPI.
Transport Layer Sample Test Mode
The transport layer samples are implemented in the AD6674 as
defined by Section 5.1.6.3 in the JEDEC JESD204B specification.
These tests are enabled via Register 0x571[5]. The test pattern is
equivalent to the raw samples from the ADC.
Interface Test Modes
The interface test modes are described in Register 0x573, Bits[3:0].
These test modes are also explained in Table 39. The interface
tests can be inserted at various points along the data. See Figure 124
for more information on the test insertion points. Register 0x573,
Bits[5:4], show where these tests are inserted.
Table 38. ADC 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
Expression
Not applicable
00 0000 0000 0000
01 1111 1111 1111
10 0000 0000 0000
10 1010 1010 1010
x23 + x18 + 1
x 9 + x5 + 1
11 1111 1111 1111
Register 0x551 to
Register 0x558
1111
Ramp output
(x) % 214
Default/Seed
Value
Not applicable
Not applicable
Not applicable
Not applicable
Not applicable
0x3AFF
0x0092
Not applicable
Not applicable
Not applicable
Sample (N, N + 1, N + 2, …)
Not applicable
Not applicable
Not applicable
Not applicable
0x1555, 0x2AAA, 0x1555, 0x2AAA, 0x1555
0x3FD7, 0x0002, 0x26E0, 0x0A3D, 0x1CA6
0x125B, 0x3C9A, 0x2660, 0x0c65, 0x0697
0x0000, 0x3FFF, 0x0000, 0x3FFF, 0x0000
For repeat mode: User Pattern 1[15:2], User Pattern 2[15:2],
User Pattern 3[15:2], User Pattern 4[15:2], User Pattern 1[15:2]…
For single mode: User Pattern 1[15:2], User Pattern 2[15:2],
User Pattern 3[15:2], User Pattern 4[15:2], 0x0000…
(x) % 214, (x + 1) % 214, (x + 2) % 214, (x + 3) % 214
Table 39. JESD204B Interface Test Modes
Output Test Mode Bit
Sequence
0000
0001
0010
0011
0100
0101
0110
0111
1000
1110
1111
Pattern Name
Off (default)
Alternating checker board
1/0 word toggle
31-bit PN sequence
23-bit PN sequence
15-bit PN sequence
9-bit PN sequence
7-bit PN sequence
Ramp output
Continuous/repeat user test
Single user test
Expression
Not applicable
0x5555, 0xAAAA, 0x5555…
0x0000, 0xFFFF, 0x0000…
x31 + x28 + 1
x23 + x18 + 1
x15 + x14 + 1
x 9 + x5 + 1
x 7 + x6 + 1
(x) % 216
Register 0x551 to Register 0x558
Register 0x551 to Register 0x558
Rev. C | Page 76 of 96
Default
Not applicable
Not applicable
Not applicable
0x0003AFFF
0x003AFF
0x03AF
0x092
0x07
Ramp size depends on test insertion point
User Pattern 1 to User Pattern 4, then repeat
User Pattern 1 to User Pattern 4, then zeros
Data Sheet
AD6674
Table 40, Table 41, and Table 42 show examples of some of the
test modes when inserted at the JESD204B sample input,
physical layer (PHY) 10-bit input, and scrambler 8-bit input.
UP in the Table 40 to Table 42 represent the user pattern control
bits from the memory map register table (see Table 45).
Data Link Layer Test Modes
The data link layer test modes are implemented in the AD6674
as defined by Section 5.3.3.8.2 in the JEDEC JESD204B specification. These tests are shown in Register 0x574, Bits[2:0]. Test
patterns inserted at this point are useful for verifying the
functionality of the data link layer. When the data link layer
test modes are enabled, disable SYNCINB± by writing 0xC0 to
Register 0x572.
Table 40. JESD204B Sample Input for M = 2, S = 2, N΄ = 16 (Register 0x573[5:4] = 'b00)
Frame
No.
0
0
0
0
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
Converter
No.
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
Sample
No.
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
Alternating
Checkerboard
0x5555
0x5555
0x5555
0x5555
0xAAAA
0xAAAA
0xAAAA
0xAAAA
0x5555
0x5555
0x5555
0x5555
0xAAAA
0xAAAA
0xAAAA
0xAAAA
0x5555
0x5555
0x5555
0x5555
1/0 Word
Toggle
0x0000
0x0000
0x0000
0x0000
0xFFFF
0xFFFF
0xFFFF
0xFFFF
0x0000
0x0000
0x0000
0x0000
0xFFFF
0xFFFF
0xFFFF
0xFFFF
0x0000
0x0000
0x0000
0x0000
Ramp
(x) % 216
(x) % 216
(x) % 216
(x) % 216
(x + 1) % 216
(x + 1) % 216
(x + 1) % 216
(x + 1) % 216
(x + 2) % 216
(x + 2) % 216
(x + 2) % 216
(x + 2) % 216
(x + 3) % 216
(x + 3) % 216
(x + 3) % 216
(x + 3) % 216
(x + 4) % 216
(x + 4) % 216
(x + 4) % 216
(x + 4) % 216
PN9
0x496F
0x496F
0x496F
0x496F
0xC9A9
0xC9A9
0xC9A9
0xC9A9
0x980C
0x980C
0x980C
0x980C
0x651A
0x651A
0x651A
0x651A
0x5FD1
0x5FD1
0x5FD1
0x5FD1
User
Repeat
UP1[15:0]
UP1[15:0]
UP1[15:0]
UP1[15:0]
UP2[15:0]
UP2[15:0]
UP2[15:0]
UP2[15:0]
UP3[15:0]
UP3[15:0]
UP3[15:0]
UP3[15:0]
UP4[15:0]
UP4[15:0]
UP4[15:0]
UP4[15:0]
UP1[15:0]
UP1[15:0]
UP1[15:0]
UP1[15:0]
User
Single
UP1[15:0]
UP1[15:0]
UP1[15:0]
UP1[15:0]
UP2[15:0]
UP2[15:0]
UP2[15:0]
UP2[15:0]
UP3[15:0]
UP3[15:0]
UP3[15:0]
UP3[15:0]
UP4[15:0]
UP4[15:0]
UP4[15:0]
UP4[15:0]
0x0000
0x0000
0x0000
0x0000
User Repeat
UP1[15:6]
UP2[15:6]
UP3[15:6]
UP4[15:6]
UP1[15:6]
UP2[15:6]
UP3[15:6]
UP4[15:6]
UP1[15:6]
UP2[15:6]
UP3[15:6]
UP4[15:6]
User Single
UP1[15:6]
UP2[15:6]
UP3[15:6]
UP4[15:6]
0x000
0x000
0x000
0x000
0x000
0x000
0x000
0x000
PN23
0xFF5C
0xFF5C
0xFF5C
0xFF5C
0x0029
0x0029
0x0029
0x0029
0xB80A
0xB80A
0xB80A
0xB80A
0x3D72
0x3D72
0x3D72
0x3D72
0x9B26
0x9B26
0x9B26
0x9B26
Table 41. Physical Layer 10-Bit Input (Register 0x573[5:4] = 'b01)
10-Bit Symbol No.
0
1
2
3
4
5
6
7
8
9
10
11
Alternating Checkerboard
0x155
0x2AA
0x155
0x2AA
0x155
0x2AA
0x155
0x2AA
0x155
0x2AA
0x155
0x2AA
1/0 Word Toggle
0x000
0x3FF
0x000
0x3FF
0x000
0x3FF
0x000
0x3FF
0x000
0x3FF
0x000
0x3FF
Ramp
(x) % 210
(x + 1) % 210
(x + 2) % 210
(x + 3) % 210
(x + 4) % 210
(x + 5) % 210
(x + 6) % 210
(x + 7) % 210
(x + 8) % 210
(x + 9) % 210
(x + 10) % 210
(x + 11) % 210
Rev. C | Page 77 of 96
PN9
0x125
0x2FC
0x26A
0x198
0x031
0x251
0x297
0x3D1
0x18E
0x2CB
0x0F1
0x3DD
PN23
0x3FD
0x1C0
0x00A
0x1B8
0x028
0x3D7
0x0A6
0x326
0x10F
0x3FD
0x31E
0x008
AD6674
Data Sheet
Table 42. Scrambler 8-Bit Input (Register 0x573[5:4] = 'b10)
8-Bit Octet No.
0
1
2
3
4
5
6
7
8
9
10
11
Alternating Checkerboard
0x55
0xAA
0x55
0xAA
0x55
0xAA
0x55
0xAA
0x55
0xAA
0x55
0xAA
1/0 Word Toggle
0x00
0xFF
0x00
0xFF
0x00
0xFF
0x00
0xFF
0x00
0xFF
0x00
0xFF
Ramp
(x) % 28
(x + 1) % 28
(x + 2) % 28
(x + 3) % 28
(x + 4) % 28
(x + 5) % 28
(x + 6) % 28
(x + 7) % 28
(x + 8) % 28
(x + 9) % 28
(x + 10) % 28
(x + 11) % 28
Rev. C | Page 78 of 96
PN9
0x49
0x6F
0xC9
0xA9
0x98
0x0C
0x65
0x1A
0x5F
0xD1
0x63
0xAC
PN23
0xFF
0x5C
0x00
0x29
0xB8
0x0A
0x3D
0x72
0x9B
0x26
0x43
0xFF
User Repeat
UP1[15:9]
UP2[15:9]
UP3[15:9]
UP4[15:9]
UP1[15:9]
UP2[15:9]
UP3[15:9]
UP4[15:9]
UP1[15:9]
UP2[15:9]
UP3[15:9]
UP4[15:9]
User Single
UP1[15:9]
UP2[15:9]
UP3[15:9]
UP4[15:9]
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x00
Data Sheet
AD6674
SERIAL PORT INTERFACE (SPI)
The AD6674 SPI allows the user to configure the converter for
specific functions or operations through a structured register
space provided inside the ADC. The SPI gives 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 serial port. Memory is organized into bytes
that can be further divided into fields. These fields are documented in the Memory Map section. For detailed operational
information, see the Serial Control Interface Standard.
CONFIGURATION USING THE SPI
Three pins define the SPI of this ADC: the SCLK pin, the SDIO
pin, and the CSB pin (see Table 43). The SCLK (serial clock) pin is
used to synchronize the read and write data presented from/to the
ADC. The SDIO (serial data input/output) pin is a dual-purpose
pin that allows data to be sent and read from the internal ADC
memory map registers. The CSB (chip select bar) pin is an active
low control that enables or disables the read and write cycles.
Table 43. Serial Port Interface Pins
Pin
SCLK
SDIO
CSB
Function
Serial clock. The serial shift clock input, which is used to
synchronize serial interface reads and writes.
Serial data input/output. A dual-purpose pin that
typically serves as an input or an output, depending on
the instruction being sent and the relative position in the
timing frame.
Chip select bar. An active low control that 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. See Figure 4 and
Table 5 for an example of the serial timing and its definitions.
Other modes involving the CSB pin are available. The CSB pin
can be held low indefinitely, which permanently enables the
device; this is called streaming. The CSB can stall high between
bytes to allow additional external timing. When CSB is tied
high, SPI functions are placed in a high impedance mode. This
mode turns on any SPI pin secondary functions.
All data is composed of 8-bit words. The first bit of each
individual byte of serial data indicates whether a read or write
command is issued. This bit allows the SDIO pin to change
direction from an input to an output.
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 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 in LSB first mode. MSB
first is the default on power-up and can be changed via the SPI
port configuration register. For more information about this
and other features, see the Serial Control Interface Standard.
HARDWARE INTERFACE
The pins described in Table 43 comprise the physical interface
between the user programming device and the serial port of the
AD6674. The SCLK pin and the CSB pin 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.
The SPI interface is flexible enough to be controlled by either
FPGAs or microcontrollers. One method for SPI configuration
is described in detail in the AN-812 Application Note,
Microcontroller-Based Serial Port Interface (SPI) Boot Circuit.
Do not activate the SPI port during periods when the full
dynamic performance of the converter is required. Because the
SCLK signal, the CSB signal, and the SDIO signal are typically
asynchronous to the ADC clock, noise from these signals can
degrade converter performance. If the on-board SPI bus is used
for other devices, it may be necessary to provide buffers between
this bus and the AD6674 to prevent these signals from
transitioning at the converter inputs during critical sampling
periods.
SPI ACCESSIBLE FEATURES
Table 44 provides a brief description of the general features that
are accessible via the SPI. These features are described in detail
in the Serial Control Interface Standard. The AD6674 device
specific features are described in the Memory Map section.
Table 44. Features Accessible Using the SPI
Feature Name
Mode
Clock
Test I/O
Output Mode
Serializer/Deserializer (SERDES) Output Setup
Description
Allows the user to set either power-down mode or standby mode
Allows the user to access the clock divider via the SPI
Allows the user to set test modes to have known data on output bits
Allows the user to set up outputs
Allows the user to vary SERDES settings, including swing and emphasis
Rev. C | Page 79 of 96
AD6674
Data Sheet
MEMORY MAP

READING THE MEMORY MAP REGISTER TABLE
Each row in the memory map register table has eight bit locations.
The memory map is roughly divided into seven sections: the
Analog Devices SPI registers, the analog input buffer control
registers, ADC function registers, the DDC function registers,
NSR decimate by 2 and noise shaping requantizer registers,
variable dynamic range registers, and the digital outputs and
test modes registers.
Table 45 (see the Memory Map Register Table section)
documents the default hexadecimal value for each hexadecimal
address shown. The column with the heading Bit 7 (MSB) is the
start of the default hexadecimal value given. For example,
Address 0x561, the output mode register, has a hexadecimal
default value of 0x01. This means that Bit 0 = 1, and the
remaining bits are 0s. This setting is the default output format
value, which is twos complement. For more information on this
function and others, see the Table 45.
Open and Reserved Locations
All address and bit locations that are not included in Table 45
are not currently supported for this device. Write unused bits of
a valid address location with 0s unless the default value is set
otherwise. Writing to these locations is required only when part
of an address location is open (for example, Address 0x561). If
the entire address location is open (for example, Address 0x013),
do not write to this address location.

“Clear a bit” is synonymous with “bit is set to Logic 0” or
“writing Logic 0 for the bit.”
“X” denotes a “don’t care”.
Channel Specific Registers
Some channel setup functions such as buffer input termination
(Register 0x016) can be programmed to a different value for
each channel. In these cases, channel address locations are
internally duplicated for each channel. These registers and bits are
designated in Table 45 as local. These local registers and bits can
be accessed by setting the appropriate Channel A or Channel B
bits in Register 0x008. If both bits are set, the subsequent write
affects the registers of both channels. In a read cycle, set only
Channel A or Channel B to read one of the two registers. If both
bits are set during an SPI read cycle, the device returns the value
for Channel A. Registers and bits designated as global in Table 45
affect the entire device and the channel features for which
independent settings are not allowed between channels. The
settings in Register 0x008 do not affect the global registers and
bits.
SPI Soft Reset
After issuing a soft reset by programming 0x81 to Register 0x000,
the AD6674 requires 5 ms to recover. Therefore, when programming the AD6674 for application setup, ensure that an adequate
delay is programmed into the firmware after asserting the soft
reset and before starting the device setup.
Default Values
Datapath Soft Reset
After the AD6674 is reset, critical registers are loaded with
default values. The default values for the registers are given in
the memory map register table, Table 45.
After programming the desired settings to the SPI registers, issue a
datapath soft reset by programming 0x02 to Register 0x001. This
reset function is implemented upon the next rising edge of the
input clock, after the register is programmed to issue the
datapath soft reset. This reset does not affect the contents of the
memory map registers; it only resets the datapath.
Logic Levels
An explanation of logic level terminology follows:

“Bit is set” is synonymous with “bit is set to Logic 1” or
“writing Logic 1 for the bit.”
Rev. C | Page 80 of 96
Data Sheet
AD6674
MEMORY MAP REGISTER TABLE
All address and bit locations that are not included in Table 45 are not currently supported for this device.
Table 45. Memory Map Registers
Reg.
Addr.
Register
Bit 7
(Hex)
Name
(MSB)
Analog Devices SPI Registers
0x000
INTERFACE_
Soft reset
CONFIG_A
(self
clearing):
clears
memory
map
registers
0x001
INTERFACE_
Single
CONFIG_B
instruction
0x002
0x003
0x004
0x005
0x006
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Bit 1
Bit 0 (LSB)
Default
LSB first
0 = MSB
1 = LSB
Address
ascension
0
0
Address
ascension
LSB first
0 = MSB
1 = LSB
0x00
0
0
0
0
0
Soft reset
(self
clearing):
clears
memory
map
registers
0
0
0
0
0
Notes
0x00
Datapath
soft
reset
(self
clearing):
does not
clear
memory
map
registers
0
00 = normal operation
10 = standby
11 = power-down
011 = high speed ADC
0x03
DEVICE_
CONFIG
(local)
CHIP_TYPE
0
CHIP_ID
(low byte)
CHIP_ID
(high byte)
CHIP_
GRADE
1
1
0
0
1
1
1
1
0xCF
Read
only
0
0
0
0
0
0
0
0
0
0x00
0
0
X
X
X
X
Read
only
Channel A
0x03
0
1
0
0x00
0x01
0x56
0
0x04
0x008
Device
0
index
0x00A
Scratch pad 0
0x00B SPI revision
0
0x00C Vendor ID
0
(low byte)
0x00D Vendor ID
0
(high byte)
Analog Input Buffer Control Registers
0x015
Analog
0
Input (local)
0x016
Input
termination
(local)
0x934
Input
capacitance
(local)
Chip speed grade
1010 = 1000 MSPS
0111 = 750 MSPS
0101 = 500 MSPS
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
1
Channel
B
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
Analog input differential termination
0000 = 400 Ω
0001 = 200 Ω
0010 = 100 Ω
0110 = 50 Ω
0
0
0
Input
disable
0 = normal
operation
1 = input
disabled
1110 =AD6674-1000 and AD6674-750
1100 = AD6674-500
0x1F = 3 pF to GND (default)
0x00 = 1.5 pF to GND
Rev. C | Page 81 of 96
0x00
0x00
0x0C;
0x0E for
AD6674
-1000
and
AD6674
-750
0x1F
Read
only
Read
only
AD6674
Reg.
Addr.
(Hex)
0x018
Register
Name
Buffer
Control 1
(local)
Data Sheet
Bit 7
(MSB)
Bit 6
Bit 5
Bit 4
0000 = 1.0× buffer current
0001 = 1.5× buffer current
0010 = 2.0× buffer current (default for
AD6674-500)
0011 = 2.5× buffer current
0100 = 3.0× buffer current (default for
AD6674-750 and AD6674-1000)
0101 = 3.5× buffer current
…
1111 = 8.5× buffer current
0100 = Setting 1 (default for AD6674-750)
0101 = Setting 2 (default for AD6674-1000)
0110 = Setting 3 (default for AD6674-500)
0111 = Setting 4
(see Table 10 for setting per frequency range)
0
0
0
0
Bit 3
0
Bit 2
0
Bit 1
0
Bit 0 (LSB)
0
Default
0x40;
0x20
for
AD6674
-500
0
0
0
0
0xXX
0x019
Buffer
Control 2
(local)
0x01A
Buffer
Control 3
(local)
0x11A
Buffer
Control 4
(local)
0
0
0x935
Buffer
Control 5
(local)
0
0
0x025
Input fullscale range
(local)
0
0
0
0
0x030
Input fullscale control
(local)
0
0
0
Full-scale control
See Table 10 for recommended settings
for different frequency bands;
default values:
AD6674-1000 = 110
AD6674-750 = 101
AD6674-500 = 001
AD6674-500 = 110 (for <1.82 V)
0
0
0xXX
ADC Function Registers
0x024
V_1P0
0
control
0
0
0
0
0
0
0x00
0
0
0
0
0
0
0
1.0 V
reference
select
0=
internal
1=
external
Diode
selection
0 = no
diode
selected
1=
temperature diode
selected
0x028
Temperature
diode
High
frequency
setting
0 = off
(default)
1 = on
0
0
1000 = Setting 1
1001 = Setting 2 (default for AD6674-750 and
AD6674-1000)
1010 = Setting 3 (default for AD6674-500)
(see Table 10 for setting per frequency range)
0
0
0
0
0
0
0
0
Low
frequency
operation
0 = off
1 = on
(default)
Full-scale adjust
0000 = 1.94 V
1000 = 1.46 V
1001 = 1.58 V
1010 = 1.70 V (default for AD6674-750 and
AD6674-1000)
1011 = 1.82 V
1100 = 2.06 V (default for AD6674-500)
Rev. C | Page 82 of 96
Notes
0x09;
0x0A
for
AD6674
-500
0x00
0x04
0x0A;
0x0C
for
AD6674
-500
0x00
V p-p
differential;
use in
conjunction
with
Reg.
0x030
Used in
conjunction
with
Reg.
0x025
Data Sheet
Reg.
Addr.
(Hex)
0x03F
Register
Name
PDWN/
STBY pin
control
(local)
AD6674
Bit 7
(MSB)
Bit 6
0
0=
PDWN/
STBY
enabled
1=
disabled
PDWN/STBY function
00 = power down
01 = standby
10 = disabled
Bit 5
0
Bit 4
0
Bit 3
0
Bit 2
0
Bit 1
0
Bit 0 (LSB)
0
Default
0x00
Fast Detect A (FD_A)
000 = Fast Detect A output
001 = JESD204B LMFC output
010 = JESD204B internal SYNC~
output
011 = temperature diode
111 = disabled
0
000 = divide by 1
001 = divide by 2
011 = divide by 4
111 = divide by 8
Independently controls Channel A and Channel B
clock divider phase offset
0000 = 0 input clock cycles delayed
0001 = ½ input clock cycles delayed
0010 = 1 input clock cycles delayed
0011 = 1½ input clock cycles delayed
0100 = 2 input clock cycles delayed
0101 = 2½ input clock cycles delayed
…
1111 = 7½ input clock cycles delayed
Clock divider positive
Clock divider negative
skew window
skew window
00 = no positive skew
00 = no negative skew
01 = 1 device clock of
01 = 1 device clock of
positive skew
negative skew
10 = 2 device clocks of
10 = 2 device clocks of
positive skew
negative skew
11 = 3 device clocks of
11 = 3 device clocks of
positive skew
negative skew
0
0
0
clock fine
delay
adjustment
enable
0=
disabled
1=
enabled
0x3F
Clock Fine Delay Adjust[7:0]
twos complement coded control to adjust the fine sample clock skew in ~1.7 ps steps
≤−88 = −151.7 ps skew
−87 = −150.0 ps skew
…
0 = 0 ps skew
…
≥ +87 = +150 ps skew
0
0
0
0
0
0
0 = no
input clock
detected
1 = input
clock
detected
0
SYSREF± mode select
CLK±
0
SYSREF±
SYSREF±
00 = disabled
edge
transition
flag reset
01 = continuous
select
select
0 = normal
10 = N shot
0=
0 = low to
operation
rising
high
1 = flags
1 = high to 1 =
held in
falling
low
reset
0x00
Fast Detect B (FD_B)
000 = Fast Detect B output
001 = JESD204B LMFC output
010 = JESD204B internal SYNC~
output
111 = disabled
0x040
Chip pin
control
0x10B
Clock
divider
0
0
0
0
0x10C
Clock
divider
phase
(local)
0
0
0
0
0x10D
Clock
divider and
SYSREF±
control
0
0
0
0x117
Clock delay
control
Clock
divider
autophase
adjust
0=
disabled
1=
enabled
0
0
0
0
0x118
Clock fine
delay
0x11C
Clock status
0
0x120
SYSREF±
Control 1
0
Rev. C | Page 83 of 96
Notes
Used in
conjunction
with
Reg.
0x040
0x00
0x00
0x00
Clock
dvider
must be
>1
0x00
Enabling
the
clock
fine
delay
adjust
causes a
datapath
soft
reset
Used in
conjunction
with
Reg.
0x117
0x00
0x00
Read
only
AD6674
Reg.
Addr.
(Hex)
0x121
Register
Name
SYSREF±
Control 2
0x123
SYSREF±
timestamp
delay
control
0x128
SYSREF±
Status 1
SYSREF±
and clock
divider
status
0x129
0x12A
0x1FF
SYSREF±
counter
Chip sync
mode
Data Sheet
Bit 7
(MSB)
0
Bit 6
0
0
0
0x00
Chip Q
ignore
0=
normal
(I/Q)
1=
ignore
(I only)
0
0
0x201
Chip
decimation
ratio
0
0
0x228
Customer
offset
Fast detect
(FD) control
(local)
0x249
0x24A
Default
0x00
0
0
0x248
Bit 2
Bit 1
Bit 0 (LSB)
SYSREF± N shot ignore counter select
0000 = next SYSREF± only
0001 = ignore the first SYSREF± transitions
0010 = ignore the first two SYSREF± transitions
…
1111 = ignore the first 16 SYSREF± transitions
0
0
0
FD upper
threshold
LSB (local)
FD upper
threshold
MSB (local)
FD lower
threshold
LSB (local)
FD lower
threshold
MSB (local)
Bit 3
0x00
Chip
application
mode
0x247
Bit 4
0
SYSREF± Timestamp Delay[6:0]
0x00 = no delay
0x01 = 1 clock delay
…
0x7F = 127 clocks delay
SYSREF± hold status
SYSREF± setup status
Refer to Table 37
Refer to Table 37
0
0
0
Clock divider phase when SYSREF± was captured
0000 = in phase
0001 = SYSREF± is ½ cycle delayed from clock
0010 = SYSREF± is 1 cycle delayed from clock
0011 = 1½ input clock cycles delayed
0100 = 2 input clock cycles delayed
0101 = 2½ input clock cycles delayed
…
1111 = 7½ input clock cycles delayed
SYSREF± counter, Bits[7:0], increments when a SYSREF± signal is captured
0x200
0x245
Bit 5
0
0
0
0
0
0
Synchronization mode
00 = normal
01 = timestamp
Chip operating mode
0001 = DDC 0 on
0010 = DDC 0 and DDC 1 on
0011 = DDC 0, DDC 1, DDC 2, and DDC3 on
0111 = NSR enabled (default)
1000 = VDR enabled
Chip decimation ratio select
000 = decimate by 1
001 = decimate by 2
010 = decimate by 4
011 = decimate by 8
100 = decimate by 16
Offset adjust in LSBs from +127 to −128 (twos complement format)
0
0
0
0
0
Force value
Force
of FD_A/
FD_A/
FD_B pins; if
FD_B
force pins is
pins;
true, this
0=
value is
normal
output on
funcFD_x pins
tion;
1 = force
to value
Fast Detect Upper Threshold[7:0]
0
0
Fast Detect Upper Threshold[12:8]
Fast Detect Lower Threshold[7:0]
0
0
0
Fast Detect Lower Threshold[12:8]
Rev. C | Page 84 of 96
Enable fast
detect
output
Notes
Mode
select
(Reg.
0x120,
Bits[2:1])
must be
N shot
Ignored
when
Reg.
0x1FF =
0x00
Read
only
Read
only
Read
only
0x07
0x01;
0x00
for
AD6674
-500
0x00
0x00
0x00
0x00
0x00
0x00
Data Sheet
Reg.
Addr.
(Hex)
0x24B
0x24C
0x26F
0x270
0x271
0x272
0x273
0x274
0x275
0x276
0x277
0x278
0x279
Register
Name
FD dwell
time LSB
(local)
FD dwell
time MSB
(local)
Signal
monitor
synchronization
control
Signal
monitor
control
(local)
Signal
Monitor
Period
Register 0
(local)
Signal
Monitor
Period
Register 1
(local)
Signal
Monitor
Period
Register 2
(local)
Signal
monitor
result
control
(local)
Signal
Monitor
Result
Register 0
(local)
Signal
Monitor
Result
Register 1
(local)
Signal
Monitor
Result
Register 1
(local)
Signal
monitor
period
counter
result
(local)
Signal
monitor
SPORT over
JESD204B
control
(local)
AD6674
Bit 7
(MSB)
Bit 6
Bit 5
Bit 4
Bit 3
Bit 2
Fast Detect Dwell Time[7:0]
Bit 1
Bit 0 (LSB)
Fast Detect Dwell Time[15:8]
0
0
0
0
0
Synchronization mode
00 = disabled
01 = continuous
11 = 1 shot
0
0
0
0
0
0
Peak
detector
0=
disabled
1=
enabled
Signal Monitor Period[7:1]
0
0
0
0
0x00
0
0x80
0x00
Signal Monitor Period[23:16]
0x00
0
0
0
Result
selection
0=
reserved
1 = Peak
detector
Read
only
Signal Monitor Result[15:8]
Readonly
0
0
0
Signal Monitor Result[19:16]
0
0
0
0
Rev. C | Page 85 of 96
Readonly
Readonly
0
00 = reserved
11 = enabled
See the
Signal
Monitor
section
In decimated
output
clock
cycles
In decimated
output
clock
cycles
In decimated
output
clock
cycles
0x01
Signal Monitor Result[7:0]
When 0x0274[0] = 1, Result Bits[19:7] = Peak Detector Absolute Value[12:0]; Result Bits[6:0] = 0
Period Count Result[7:0]
0
0x00
Signal Monitor Period[15:8]
Result
update
1 = update
results
(self clear)
Notes
0x00
0
0
Default
0x00
0x00
Updated
based
on Reg.
0x0274,
Bit 4
Updated
based
on Reg.
0x0274,
Bit 4
Updated
based
on Reg.
0x0274,
Bit 4
Updated
based
on Reg.
0x0274,
Bit 4
AD6674
Reg.
Addr.
(Hex)
0x27A
Register
Name
SPORT over
JESD204B
input
selection
(local)
Data Sheet
Bit 7
(MSB)
0
Bit 6
0
Bit 5
0
Bit 4
0
Bit 3
0
Bit 2
0
Bit 1
Peak
detector
0=
disabled
1=
enabled
Bit 0 (LSB)
0
Digital Downconverter (DDC) Function Registers—see the Digital Downconverter (DDC) section
0
0
0
DDC NCO
0
0
Synchronization mode
0x300
DDC
soft reset
00 = disabled
synchro0 = normal
01 = continuous
nization
operation
11 = one shot
control
1 = reset
Decimation ratio select
IF mode
Complex 0
Gain select
0x310
DDC 0
Mixer
(complex to real
00 = variable IF mode
to real
0 = 0 dB
control
select
disabled)
(mixers and NCO
enable
gain
0 = real
11 = decimate by 2
enabled)
0=
1 = 6 dB
mixer
00 = decimate by 4
01 = 0 Hz IF mode
disabled
gain
1=
01 = decimate by 8
(mixer bypassed, NCO
1=
complex
10 = decimate by 16
disabled)
enabled
mixer
(complex to real
10 = fADC/4 Hz IF mode
enabled)
(fADC/4 downmixing
11 = decimate by 1
mode)
00 = decimate by 2
11 = test mode (mixer
01 = decimate by 4
inputs forced to +FS,
10 = decimate by 8
NCO enabled)
0
I input
0
0
0
0
0
Q input
0x311
DDC 0
select
select
input
0 = Ch. A
0 = Ch. A
selection
1 = Ch. B
1 = Ch. B
0x314
DDC 0
DDC 0 NCO FTW[7:0] twos complement
frequency
LSB
X
X
X
X
DDC 0 NCO FTW[11:8] twos complement
0x315
DDC 0
frequency
MSB
0x320
DDC 0
DDC 0 NCO POW[7:0] twos complement
phase LSB
0x321
DDC 0
X
X
X
X
DDC0 NCO POW[11:8] twos complement
phase MSB
0
I output
0
0
0
0
0
Q output
0x327
DDC 0
test mode
test mode
output test
enable
enable
mode
0=
0 = disabled
selection
disabled
1 = enabled
1=
from Ch. B
enabled
from Ch. A
Decimation ratio select
IF mode
Complex 0
Gain select
0x330
DDC 1
Mixer
(complex to real
00 = variable IF mode
to real
0 = 0 dB
control
select
disabled)
(mixers and NCO
enable
gain
0 = real
11 = decimate by 2
enabled)
0=
1 = 6 dB
mixer
00 = decimate by 4
01 = 0 Hz IF mode
disabled
gain
1=
01 = decimate by 8
(mixer bypassed, NCO
1=
complex
10 = decimate by 16
disabled)
enabled
mixer
(complex to real
10 = fADC/4 Hz IF mode
enabled)
(fADC/4 downmixing
11 = decimate by 1
mode)
00 = decimate by 2
11 = test mode (mixer
01 = decimate by 4
inputs forced to +FS,
10 = decimate by 8
NCO enabled)
0
I input
0x331
DDC 1
0
0
0
0
0
Q input
select
input
select
0 = Ch. A
selection
0 = Ch. A
1 = Ch. B
1 = Ch. B
Rev. C | Page 86 of 96
Default
0x02
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x05
Notes
Data Sheet
Reg.
Addr.
(Hex)
0x334
AD6674
Register
Name
DDC 1
frequency
LSB
DDC 1
frequency
MSB
DDC 1
phase LSB
DDC 1
phase MSB
DDC 1
output test
mode
selection
Bit 7
(MSB)
Bit 6
Bit 5
X
X
X
0x350
0x335
0x340
0x341
Bit 4
Bit 3
Bit 2
DDC 1 NCO FTW[7:0] twos complement
X
DDC 1 NCO POW[7:0] twos complement
X
X
X
0
0
0
0
DDC 2
control
Mixer
select
0 = real
mixer
1=
complex
mixer
Gain select
0 = 0 dB
gain
1 = 6 dB
gain
0x351
DDC 2
input
selection
0
0
IF mode
00 = variable IF mode
(mixers and NCO
enabled)
01 = 0 Hz IF mode
(mixer bypassed, NCO
disabled)
10 = fADC/4 Hz IF mode
(fADC/4 downmixing
mode)
11 = test mode (mixer
inputs forced to +FS,
NCO enabled)
0
0
0x354
DDC 2
frequency
LSB
DDC 2
frequency
MSB
DDC 2
phase LSB
DDC 2
phase MSB
DDC 2
output test
mode
selection
0x355
0x360
0x361
0x367
0x370
DDC 3
control
Bit 0 (LSB)
DDC1 NCO FTW[11:8] twos complement
X
0x347
Bit 1
X
X
Q output
test mode
enable
0 = disabled
1 = enabled
from Ch. B
Complex
to real
enable
0=
disabled
1=
enabled
0
Q input
select
0 = Ch. A
1 = Ch. B
DDC 2 NCO FTW[7:0] twos complement
X
0
X
0
I output
test mode
enable
0=
disabled
1=
enabled
from Ch. A
Decimation ratio select
(complex to real
disabled)
11 = decimate by 2
00 = decimate by 4
01 = decimate by 8
10 = decimate by 16
(complex to real
enabled)
11 = decimate by 1
00 = decimate by 2
01 = decimate by 4
10 = decimate by 8
0
I input
select
0 = Ch. A
1 = Ch. B
X
X
X
0
0
0
0
DDC2 NCO FTW[11:8] twos complement
Mixer
select
0 = real
mixer
1=
complex
mixer
Gain select
0 = 0 dB
gain
1 = 6 dB
gain
IF mode
00 = variable IF mode
(mixers and NCO
enabled)
01 = 0 Hz IF mode
(mixer bypassed, NCO
disabled)
10 = fS/4 Hz IF mode
(fS/4 downmixing
mode)
11 = test mode (mixer
inputs forced to +FS,
NCO enabled)
Q output
test mode
enable
0 = disabled
1 = enabled
from Ch. B
Complex
to real
enable
0=
disabled
1=
enabled
0
Rev. C | Page 87 of 96
0x00
0x00
0x00
0x00
0x00
DDC2 NCO Phase Offset[11:8] twos complement
0
0x00
0x00
DDC 2 NCO Phase Offset[7:0] twos complement
X
0x00
0x00
DDC1 NCO POW[11:8] twos complement
0
Default
0x00
0
I output
test mode
enable
0=
disabled
1=
enabled
from Ch. A
Decimation ratio select
(complex to real
disabled)
11 = decimate by 2
00 = decimate by 4
01 = decimate by 8
10 = decimate by 16
(complex to real
enabled)
11 = decimate by 1
00 = decimate by 2
01 = decimate by 4
10 = decimate by 8
0x00
0x00
0x00
Notes
AD6674
Reg.
Addr.
(Hex)
0x371
0x374
0x375
0x380
0x381
0x387
Register
Name
DDC 3
input
selection
DDC 3
frequency
LSB
DDC 3
frequency
MSB
DDC 3
phase LSB
DDC 3
phase MSB
DDC 3
output test
mode
selection
Data Sheet
Bit 7
(MSB)
0
X
Bit 6
0
X
Bit 5
0
X
Bit 4
0
Bit 2
Q input
select
0 = Ch. A
1 = Ch. B
DDC3 NCO FTW[7:0] twos complement
X
Bit 1
0
Bit 0 (LSB)
I input
select
0 = Ch. A
1 = Ch. B
Default
0x05
DDC3 NCO FTW[11:8] twos complement
0x00
0x00
X
X
X
X
0
0
0
0
0
0
0
I output
test mode
enable
0=
disabled
1=
enabled
from Ch. A
0x00
0
X
X
X
NSR
decimate
by 2
enable
0=
disabled
1=
enabled
0x01;
0x00
for
AD6674
-500
0x420
NSR mode
X
X
0x422
NSR tuning
X
X
X
0x00
NSR mode
X
000 = 21% BW mode
001 = 28% BW mode
NSR tuning word; see the Noise Shaping Requantizer (NSR) section; equations
for the tuning word are dependent on the NSR mode
0x00
0 = dual
VDR BW
real mode
mode
1 = dual
0 = 25%
complex
BW
mode
mode
(Channel A
1 = 43%
= I,
BW
Channel B
mode
= Q)
(only
available
for dual
complex
mode)
VDR center frequency; see the Variable Dynamic
Range (VDR) section for more details on the center
frequency, which is dependent on the VDR mode
0x01
X
X
0
0x434
X
X
X
X
DDC3 NCO POW[11:8] twos complement
X
Variable Dynamic Range (VDR)
0x430
VDR control X
X
Digital Outputs and Test Modes
Rev. C | Page 88 of 96
X
Notes
0x00
DDC3 NCO POW[7:0] twos complement
NSR Decimate by 2 and Noise Shaping Requantizer (NSR)
X
0
0x41E
NSR
Highdecimate
pass/
by 2
low-pass
mode:
0=
enable
LPF
1=
enable
HPF
VDR tuning
Bit 3
0
0x00
0x00
Bit 0 is
ignored
on
AD6674
-750
and
AD6674
-1000
when in
NSR
mode
Data Sheet
Reg.
Addr.
(Hex)
0x550
AD6674
Register
Name
ADC test
modes
(local)
Bit 7
(MSB)
User
pattern
selection
0=
continuous
repeat
1 = single
pattern
0x551
User
Pattern 1
LSB
0x552
Bit 6
0
Bit 5
Reset PN
long gen
0 = long
PN
enable
1 = long
PN reset
Bit 4
Reset PN
short gen
0 = short
PN enable
1 = short
PN reset
Bit 3
Bit 2
Bit 1
Bit 0 (LSB)
Test mode selection
0000 = off (normal operation)
0001 = midscale short
0010 = positive full scale
0011 = negative full scale
0100 = alternating checker board
0101 = PN sequence, long
0110 = PN sequence, short
0111 = 1/0 word toggle
1000 = user pattern test mode (used with
Register 0x550, Bit 7, and User Pattern 1 to
User Patten 4 registers)
1111 = ramp output
0
0
0
Default
0x00
Notes
0
0
0
0
0
0x00
User
Pattern 1
MSB
0
0
0
0
0
0
0
0
0x00
0x553
User
Pattern 2
LSB
0
0
0
0
0
0
0
0
0x00
0x554
User
Pattern 2
MSB
0
0
0
0
0
0
0
0
0x00
0x555
User
Pattern 3
LSB
0
0
0
0
0
0
0
0
0x00
0x556
User
Pattern 3
MSB
0
0
0
0
0
0
0
0
0x00
0x557
User
Pattern 4
LSB
0
0
0
0
0
0
0
0
0x00
0x558
User
Pattern 4
MSB
0
0
0
0
0
0
0
0
0x00
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Used
with
Reg.
0x550,
Reg.
0x573
Rev. C | Page 89 of 96
AD6674
Reg.
Addr.
(Hex)
0x559
Data Sheet
Register
Name
Output
Mode
Control 1
Bit 7
(MSB)
0
0x55A
Output
Mode
Control 2
0x561
Bit 3
0
0
Bit 6
Bit 5
Bit 4
Converter Control Bit 1 selection (only
used when CS (0x58F) = 2 or 3)
000 = tie low (1’b0)
001 = overrange bit
010 = signal monitor bit or
VDR Punish Bit 0
011 = fast detect (FD) bit or
VDR Punish Bit 1
100 = VDR high/low resolution bit
101 = system reference
0
0
0
Output
mode
0
0
0
0
0
0x562
Output
overrange
(OR) clear
Virtual
Converter 7
OR
0 = OR bit
enabled
1 = OR bit
cleared
Virtual
Converter
6 OR
0 = OR bit
enabled
1 = OR bit
cleared
Virtual
Converter
4 OR
0 = OR bit
enabled
1 = OR bit
cleared
0x563
Output
overrange
status
Virtual
Converter 7
OR
0 = no OR
1 = OR
occurred
Virtual
Converter
6 OR
0 = no OR
1 = OR
occurred
Virtual
Converter 5
OR
0 = OR
bit
enabled
1 = OR
bit
cleared
Virtual
Converter 5
OR
0 = no
OR
1 = OR
occurred
0x564
Output
channel
select
0
0
0
0
Virtual
Converter 3
OR
0 = OR
bit
enabled
1 = OR
bit
cleared
Virtual
Converter 3
OR
0 = no
OR
1 = OR
occurre
d
0
0x56E
JESD204B
lane rate
control
0
0
0
0 = serial
lane rate ≥
6.25 Gbps
and ≤
12.5 Gbps
1 = serial
lane rate
must be ≥
3.125 Gbps
and
<6.25 Gbps
Virtual
Converter
4 OR
0 = no OR
1 = OR
occurred
0
0
Rev. C | Page 90 of 96
Bit 2
Bit 1
Bit 0 (LSB)
Converter Control Bit 0 selection (only
used when CS (0x58F) = 3)
000 = tie low (1’b0)
001 = overrange bit
010 = signal monitor bit or
VDR Punish Bit 0
011 = fast detect (FD) bit or VDR
Punish Bit 1
100 = VDR high/low resolution bit
101 = system reference
Converter Control Bit 2 selection (used
when CS (0x58F) = 1, 2, or 3)
000 = tie low (1’b0)
001 = overrange bit
010 = signal monitor bit or
VDR Punish Bit 0
011 = fast detect (FD) bit or VDR
Punish Bit 1
100 = VDR high/low resolution bit
101 = system reference
Data format select
Sample
00 = offset binary
invert
01 = twos complement
0 = normal
1 = sample
invert
Virtual
Virtual
Virtual
Converter 0
ConConverter 2
OR
verter 1
OR
0 = OR bit
OR
0 = or bit
enabled
0 = OR
enabled
1 = OR bit
bit
1 = OR bit
cleared
enabled
cleared
1 = OR
bit
cleared
Virtual
Virtual
Virtual
Converter 0
ConConverter 2
OR
verter 1
OR
0 = no OR
OR
0 = no OR
1 = OR
0 = no
1 = OR
occurred
OR
occurred
1 = OR
occurre
d
0
0
Converter
channel
swap
0 = normal
channel
ordering
1=
channel
swap
enabled
0
0
0
Default
0x00
Notes
0x01
0x01
0x00
0x00
0x00
0x10
Read
only
Data Sheet
Reg.
Addr.
(Hex)
0x56F
0x570
0x571
Register
Name
JESD204B
PLL lock
status
JESD204B
quick
configuration
JESD204B
Link Mode
Control 1
AD6674
Bit 7
(MSB)
PLL lock
0 = not
locked
1=
locked
Standby
mode
0 = all
converter
outputs 0
1 = CGS
(K28.5)
Bit 6
0
Tail bit (T)
PN
0 = disable
1 = enable
T=
N’ − N −
CS
Bit 5
0
Bit 4
0
JESD204B
Link Mode
Control 2
SYNCINB± pin control
00 = normal
10 = ignore SYNCINB±
(force CGS)
11 = ignore SYNCINB±
(force ILAS/user data)
0x573
JESD204B
Link Mode
Control 3
CHKSUM mode
00 = sum of all 8-bit link
configuration registers
01 = sum of individual
link configuration bit
fields
10 = checksum set to
zero
0x574
JESD204B
Link Mode
Control 4
0x578
JESD204B
LMFC offset
JESD204B
DID config
JESD204B
BID config
JESD204B
LID
Config 1
JESD204B
LID
Config 2
JESD204B
LID
Config 3
ILAS delay
0000 = transmit ILAS on first LMFC after SYNCINB±
deasserted
0001 = transmit ILAS on second LMFC after
SYNCINB± deasserted
…
1111 = transmit ILAS on 16th LMFC after SYNCINB±
deasserted
0
0
0
0x581
0x583
0x584
0x585
Bit 2
0
JESD204B quick configuration
Number of lanes (L) = 20x570[7:6]
Number of converters (M) = 20x570[5:3]
Number of octets/frame (F) = 20x570[2:0]
ILAS sequence mode
Lane
Long
00 = ILAS disabled
synchrontrans01 = ILAS enabled
port layer ization
11 = ILAS always on test
0 = disable
test
mode
FACI uses
0=
/K28.7/
disable
1 = enable
1=
FACI uses
enable
/K28.3/
and
/K28.7/
0
8B/10B
SYNCINB± SYNCINB±
bypass
pin invert pin type
0 = normal
0 = active 0 =
1 = bypass
differential
low
1 = active 1 = CMOS
high
0x572
0x580
Bit 3
0
Test insertion point
00 = N’ sample input
01 = 10-bit data at
8B/10B output (for PHY
testing)
10 = 8-bit data at
scrambler input
Bit 1
0
Bit 0 (LSB)
0
Default
0x00
Notes
Read
only
0x88
Refer to
Table 35
and
Table 36
Link
control
0 = active
1 = power
down
0x14
0
8B/10B
bit
invert
0=
normal
1=
invert
abcde
fghij
symbols
JESD204B test mode patterns
0000 = normal operation (test mode disabled)
0001 = alternating checker board
0010 = 1/0 word toggle
0011 = 31-bit PN sequence—x31 + x28 + 1
0100 = 23-bit PN sequence—x23 + x18 + 1
0101 = 15-bit PN sequence—x15 + x14 + 1
0110 = 9-bit PN sequence—x9 + x5 + 1
0111 = 7-bit PN sequence—x7 + x6 + 1
1000 = ramp output
1110 = continuous/repeat user test
1111 = single user test
0
Link layer test mode
000 = normal operation (link layer test
mode disabled)
001 = continuous sequence of /D21.5/
characters
100 = modified RPAT test sequence
101 = JSPAT test sequence
110 = JTSPAT test sequence
LMFC Phase Offset Value[4:0]
0x00
Frame
alignment
character
insertion
(FACI)
0=
enabled
1=
disabled
JESD204B Tx DID Value[7:0]
0
0x00
0x00
0x00
0x00
0
0
0
JESD204B Tx BID Value[3:0]
0
0
0
Lane 0 LID Value[4:0]
0x00
0
0
0
Lane 1 LID Value[4:0]
0x01
0
0
0
Lane 2 LID Value[4:0]
0x01
Rev. C | Page 91 of 96
0x00
AD6674
Reg.
Addr.
(Hex)
0x586
0x58B
Register
Name
JESD204B
LID
Config 4
JESD204B
parameters
(SCR/L)
Data Sheet
Bit 7
(MSB)
0
JESD204B
scrambling
(SCR)
0=
disabled
1=
enabled
Bit 6
0
Bit 5
0
Bit 4
Bit 3
0
0
0
0
0x58C
JESD204B F
config
0x58D
JESD204B K
config
JESD204B
M config
0
0x58F
JESD204B
parameters
(CS/N)
0x590
JESD204B
parameter
(NP)
0x591
JESD204B
parameter
(S)
JESD204B
parameters
(HD and CF)
Number of control bits
(CS) per sample
00 = no control bits
(CS = 0)
01 = 1 control bit (CS =
1); Control Bit 2 only
10 = 2 control bits (CS =
2); Control Bit 2 and
Control Bit 1 only
11 = 3 control bits
(CS = 3); all control bits
(2, 1, 0)
Subclass support
000 = Subclass 0 (no deterministic
latency)
001 = Subclass 1
0
0
1
0x58E
0x592
0x5A0
0x5A1
0x5A2
0x5A3
0x5B0
0x5B2
JESD204B
CHKSUM 0
JESD204B
CHKSUM 1
JESD204B
CHKSUM 2
JESD204B
CHKSUM 3
JESD204B
lane powerdown
JESD204B
lane
SERDOUT0±
assign
Bit 2
Bit 1
Lane 3 LID Value[4:0]
0
Bit 0 (LSB)
JESD204B lanes (L)
00 = 1 lane
01 = 2 lanes
11 = 4 lanes
read only; see
Register 0x570
Number of octets per frame, F = 0x58C[7:0] + 1
HD value
0=
disabled
1=
enabled
1
X
0
0
SERDOUT3±
0 = on
1 = off
X
Number of frames per multi-frame, K = 0x58D[4:0] + 1
Only values where (F × K) mod 4 = 0 are supported
Number of Converters per Link[7:0]
0x00 = link connected to one virtual converter (M = 1)
0x01 = link connected to two virtual converters (M = 2)
0x03 = link connected to four virtual converters (M = 4)
0x07 = link connected to eight virtual converters (M = 8)
0
Converter resolution (N)
0x06 = 7-bit resolution
0x07 = 8-bit resolution
0x08 = 9-bit resolution
0x09 = 10-bit resolution
0x0A = 11-bit resolution
0x0B = 12-bit resolution
0x0C = 13-bit resolution
0x0D = 14-bit resolution
0x0E = 15-bit resolution
0x0F = 16-bit Resolution
0x1F
0x01
Read
only,
see Reg.
0x570
See Reg.
0x570
Read
only
0x0F
Number of bits per sample (N’)
0x7 = 8 bits
0xF = 16 bits
0x2F
Samples per converter frame cycle (S)
S value = 0x591[4:0] +1
0x20
Read
only
Control words per frame clock cycle per link (CF)
CF value = 0x592[4:0]
0x80
Read
only
CHKSUM value for SERDOUT0±[7:0]
0x81
CHKSUM value for SERDOUT1±[7:0]
0x82
CHKSUM value for SERDOUT2±[7:0]
0x82
CHKSUM value for SERDOUT3±[7:0]
0x84
Read
only
Read
only
Read
only
Read
only
0
X
Notes
0x83
0x00
0
1
Default
0x03
SERDOUT2±
0 = on
1 = off
X
1
0
Rev. C | Page 92 of 96
SERDOUT1±
0 = on
1 = off
1
SERDOUT0±
0 = on
1 = off
Physical Lane 0 assignment
000 = Logical Lane 0
001 = Logical Lane 1
010 = Logical Lane 2
011 = Logical Lane 3
0xAA
0x00
Data Sheet
Reg.
Addr.
(Hex)
0x5B3
AD6674
Register
Name
JESD204B
lane
SERDOUT1±
assign
Bit 7
(MSB)
X
Bit 6
X
Bit 5
X
Bit 4
X
Bit 3
0
0x5B5
JESD204B
lane
SERDOUT2±
assign
X
X
X
X
0
0x5B6
JESD204B
lane
SERDOUT3±
assign
X
X
X
X
0
0x5BF
JESD
serializer
drive adjust
0
0
0
0
0x5C1
Deemphasis
select
0
0
0x5C2
Deemphasis
setting for
SERDOUT0±
0
SERDOUT3±
0 = disable
1 = enable
0
0
SERDOUT2±
0 = disable
1 = enable
0
0x5C3
Deemphasis
setting for
SERDOUT1±
0
0
0
0
0x5C4
Deemphasis
setting for
SERDOUT2±
0
0
0
0
0
Rev. C | Page 93 of 96
Bit 2
Bit 1
Bit 0 (LSB)
Physical Lane 1 assignment
000 = Logical Lane 0
001 = Logical Lane 1
010 = Logical Lane 2
011 = Logical Lane 3
Physical Lane 2 assignment
000 = Logical Lane 0
001 = Logical Lane 1
010 = Logical Lane 2
011 = Logical Lane 3
Physical Lane 3 assignment
000 = Logical Lane 0
001 = Logical Lane 1
010 = Logical Lane 2
011 = Logical Lane 3
Swing voltage
0000 = 237.5 mV
0001 = 250 mV
0010 = 262.5 mV
0011 = 275 mV
0100 = 287.5 mV
0101 = 300 mV
0110 = 312.5 mV
0111 = 325 mV
1000 = 337.5 mV
1001 = 350 mV
1010 = 362.5 mV
1011 = 375 mV
1100 = 387.5 mV
1101 = 400 mV
1110 = 412.5 mV
1111 = 425 mV
SERDOUT1±
0
SER0 = disable
DOUT0±
1 = enable
0 = disable
1 = enable
De-emphasis settings
0000 = de-emphasis disabled
1000 = 0.5 dB
1001 = 1.0 dB
1010 = 1.7 dB
1011 = 2.5 dB
1100 = 3.5 dB
1101 = 4.9 dB
1110 = 6.7 dB
1111 = 9.6 dB
De-emphasis settings
0000 = de-emphasis disabled
1000 = 0.5 dB
1001 = 1.0 dB
1010 = 1.7 dB
1011 = 2.5 dB
1100 = 3.5 dB
1101 = 4.9 dB
1110 = 6.7 dB
1111 = 9.6 dB
De-emphasis settings
0000 = de-emphasis disabled
1000 = 0.5 dB
1001 = 1.0 dB
1010 = 1.7 dB
1011 = 2.5 dB
1100 = 3.5 dB
1101 = 4.9 dB
1110 = 6.7 dB
1111 = 9.6 dB
Default
0x11
0x22
0x33
0x05
0x00
0x00
0x00
0x00
Notes
AD6674
Reg.
Addr.
(Hex)
0x5C5
Register
Name
Deemphasis
setting for
SERDOUT3±
Data Sheet
Bit 7
(MSB)
0
Bit 6
0
Bit 5
0
Bit 4
0
Bit 3
Rev. C | Page 94 of 96
Bit 2
Bit 1
Bit 0 (LSB)
De-emphasis settings
0000 = de-emphasis disabled
1000 = 0.5 dB
1001 = 1.0 dB
1010 = 1.7 dB
1011 = 2.5 dB
1100 = 3.5 dB
1101 = 4.9 dB
1110 = 6.7 dB
1111 = 9.6 dB
Default
0x00
Notes
Data Sheet
AD6674
APPLICATIONS INFORMATION
POWER SUPPLY RECOMMENDATIONS
The AD6674 must be powered by the following seven supplies:
AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, AVDD1_SR
= 1.25 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V.
For applications requiring an optimal high power efficiency and
low noise performance, it is recommended that the ADP2164
and ADP2370 switching regulators be used to convert the 3.3 V,
5.0 V, or 12 V input rails to an intermediate rail (1.8 V and
3.8 V). These intermediate rails are then postregulated by very
low noise, low dropout (LDO) regulators (ADP1741, ADM7172,
and ADP125). Figure 145 shows the recommended method. For
more detailed information on the recommended power
solution, refer to the AD6674 evaluation board documentation.
ADP1741
1.8V
AVDD1
1.25V
AVDD1_SR
1.25V
ADP1741
thermal performance of the AD6674. Connect an exposed
continuous copper plane on the PCB to the AD6674 exposed
pad, Pin 0. The copper plane must have several vias to achieve
the lowest possible resistive thermal path for heat dissipation to
flow through the bottom of the PCB. These vias must be solder
filled or plugged. The number of vias and the fill determine the
resultant θJA measured on the board.
To maximize the coverage and adhesion between the ADC and
PCB, partition the continuous copper plane by overlaying a
silkscreen on the PCB into several uniform sections. This
provides several tie points between the ADC and PCB during
the reflow process, whereas using one continuous plane with no
partitions only guarantees one tie point. See Figure 146 for a
PCB layout example. For detailed information on packaging
and the PCB layout of chip scale packages, see the AN-772
Application Note, A Design and Manufacturing Guide for the
Lead Frame Chip Scale Package (LFCSP).
DVDD
1.25V
DRVDD
1.25V
3.6V
ADP125
AVDD3
3.3V
3.3V
ADM7172
OR
ADP1741
AVDD2
2.5V
12400-518
SPIVDD
(1.8V OR 3.3V)
It is not necessary to split all of these power domains in all
cases. The recommended solution shown in Figure 145 provides
the lowest noise, highest efficiency power delivery system for
the AD6674. If only one 1.25 V supply is available, it must be
routed to AVDD1 first and then tapped off and isolated with a
ferrite bead or a filter choke preceded by decoupling capacitors
for AVDD1_SR, DVDD, and DRVDD, in that order. The user
can use several different decoupling capacitors to cover both
high and low frequencies. These must be located close to the
point of entry at the PCB level and close to the devices, with
minimal trace lengths.
EXPOSED PAD THERMAL HEAT SLUG
RECOMMENDATIONS
It is required that the exposed pad on the underside of the ADC
be connected to ground to achieve the best electrical and
12400-180
Figure 145. High Efficiency, Low Noise Power Solution for the AD6674
Figure 146. Recommended PCB Layout of Exposed Pad for the AD6674
AVDD1_SR (PIN 57) AND AGND (PIN 56, PIN 60)
AVDD1_SR (Pin 57) and AGND (Pin 56 and Pin 60) can be
used to provide a separate power supply node to the SYSREF±
circuits of the AD6674. If running in Subclass 1, the AD6674
can support periodic one-shot or gapped signals. To minimize
the coupling of this supply into the AVDD1 supply node,
adequate supply bypassing is needed.
Rev. C | Page 95 of 96
AD6674
Data Sheet
OUTLINE DIMENSIONS
9.10
9.00 SQ
8.90
0.30
0.25
0.18
49
1
0.50
BSC
EXPOSED
PAD
7.70
7.60 SQ
7.50
33
TOP VIEW
0.80
0.75
0.70
0.45
0.40
0.35
16
32
17
BOTTOM VIEW
7.50 REF
0.05 MAX
0.02 NOM
COPLANARITY
0.08
0.203 REF
PKG-004396
SEATING
PLANE
PIN 1
INDICATOR
64
48
0.20 MIN
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
SECTION OF THIS DATA SHEET.
COMPLIANT TO JEDEC STANDARDS MO-220-WMMD
02-12-2014-A
PIN 1
INDICATOR
Figure 147. 64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
9 mm × 9 mm Body, Very Very Thin Quad
(CP-64-15)
Dimensions shown in millimeters
ORDERING GUIDE
Model1
AD6674BCPZ-500
AD6674BCPZRL7-500
AD6674BCPZ-750
AD6674BCPZRL7-750
AD6674BCPZ-1000
AD6674BCPZRL7-1000
AD6674-500EBZ
Temperature Range
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
AD6674-750EBZ
AD6674-1000EBZ
AD6674-LF500EBZ
AD6674-LF750EBZ
AD6674-LF1000EBZ
1
2
Package Description2
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
64-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
Evaluation Board for AD6674-500 (Optimized for Full Analog
Input Bandwidth)
Evaluation Board for AD6674-750 (Optimized for Full Analog
Input Bandwidth)
Evaluation Board for AD6674-1000 (Optimized for Full
Analog Input Bandwidth)
Evaluation Board for AD6674-500 (Optimized for Up to
1 GHz Analog Input Bandwidth)
Evaluation Board for AD6674-750 (Optimized for Up to
1 GHz Input Bandwidth)
Evaluation Board for AD6674-1000 (Optimized for Up to
1 GHz Analog Input Bandwidth)
Z = RoHS Compliant Part.
The AD6674-500EBZ, AD6674-750EBZ, and AD6674-1000EBZ evaluation boards are optimized for the full analog input frequency range.
©2014–2016 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D12400-0-8/16(C)
Rev. C | Page 96 of 96
Package Option
CP-64-15
CP-64-15
CP-64-15
CP-64-15
CP-64-15
CP-64-15
Similar pages