Design Resources Online Documentation Dual, 14-Bit, 1.25 GSPS, 1.2 V/2.5 V, Analog-to-Digital Converter AD9684 FEATURES APPLICATIONS Communications Diversity multiband, multimode digital receivers 3G/4G, TD-SCDMA, W-CDMA, MC-GSM, LTE General-purpose software radios Ultrawideband satellite receiver Instrumentation (spectrum analyzers, network analyzers, integrated RF test solutions) Radar Digital oscilloscopes High speed data acquisition systems DOCSIS CMTS upstream receiver paths HFC digital reverse path receivers AVDD2 (2.5V) AVDD3 (3.3V) DRVDD (1.25V) DVDD (1.25V) SPIVDD (1.8V TO 3.4V) BUFFER VIN+A DIGITAL DOWNCONVERTER FD_B DIGITAL DOWNCONVERTER BUFFER VIN+B VIN–B ADC CORE D0± D1± D2± D3± D4± D5± D6± D7± D8± D9± D10± D11± D12± D13± DCO± STATUS± 14 CONTROL REGISTERS FAST DETECT V_1P0 SIGNAL MONITOR CLOCK GENERATION CLK+ CLK– 16 LVDS/SYNC CONTROL FD_A LVDS OUTPUTS 14 ADC CORE VIN–A FAST DETECT SYNC+ SYNC– SPI CONTROL ÷2 ÷4 AD9684 PDWN/ STBY ÷8 AGND DRGND DGND SDIO SCLK CSB 13015-001 Parallel LVDS (DDR) outputs 1.1 W total power per channel at 500 MSPS (default settings) SFDR = 85 dBFS at 170 MHz fIN (500 MSPS) SNR = 68.6 dBFS at 170 MHz fIN (500 MSPS) ENOB = 10.9 bits at 170 MHz fIN DNL = ±0.5 LSB INL = ±2.5 LSB Noise density = −153 dBFS/Hz at 500 MSPS 1.25 V, 2.50 V, and 3.3 V supply operation No missing codes Internal analog-to-digital converter (ADC) voltage reference Flexible input range and termination impedance 1.46 V p-p to 2.06 V p-p (2.06 V p-p nominal) 400 Ω, 200 Ω, 100 Ω, and 50 Ω differential SYNC± input allows multichip synchronization DDR LVDS (ANSI-644 levels) outputs 2 GHz usable analog input full power bandwidth >96 dB channel isolation/crosstalk Amplitude detect bits for efficient AGC implementation Two integrated wideband digital processors per channel 12-bit numerically controlled oscillator (NCO) 3 cascaded half-band filters Differential clock inputs Serial port control Integer clock divide by 2, 4, or 8 Small signal dither FUNCTIONAL BLOCK DIAGRAM AVDD1 (1.25V) LVDS OUTPUT STAGING Data Sheet Sample & Buy Discussion SIGNAL MONITOR Product Overview Figure 1. GENERAL DESCRIPTION The AD9684 is a dual, 14-bit, 500 MSPS ADC. The device 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 for sampling wide bandwidth analog signals. The AD9684 is optimized for wide input bandwidth, a 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 buffered inputs, supporting a variety of user selectable input ranges. An integrated voltage reference eases design considerations. Each ADC data output is internally connected to an optional decimate by 2 block. The analog input and clock signals are differential inputs. Each ADC data output is internally connected to two digital downconverters (DDCs). Each DDC consists of four cascaded signal processing stages: a 12-bit frequency translator (NCO), and three half-band decimation filters supporting a divide by factor of two, four, and eight. Rev. 0 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 www.analog.com Fax: 781.461.3113 ©2015 Analog Devices, Inc. All rights reserved. Product Overview Online Documentation Design Resources AD9684 Discussion Sample & Buy Data Sheet TABLE OF CONTENTS Features .............................................................................................. 1 DDC I/Q Output Selection ....................................................... 31 Applications ....................................................................................... 1 DDC General Description ........................................................ 31 Functional Block Diagram .............................................................. 1 Frequency Translation ................................................................... 37 General Description ......................................................................... 1 General Description ................................................................... 37 Revision History ............................................................................... 2 DDC NCO Plus Mixer Loss and SFDR ................................... 38 Product Highlights ........................................................................... 3 Numerically Controlled Oscillator .......................................... 38 Specifications..................................................................................... 4 FIR Filters ........................................................................................ 40 DC Specifications ......................................................................... 4 General Description ................................................................... 40 AC Specifications.......................................................................... 5 Half-Band Filters ........................................................................ 41 Digital Specifications ................................................................... 6 DDC Gain Stage ......................................................................... 42 Switching Specifications .............................................................. 7 DDC Complex to Real Conversion Block............................... 42 Timing Specifications .................................................................. 8 DDC Example Configurations ................................................. 43 Absolute Maximum Ratings .......................................................... 16 Digital Outputs ............................................................................... 47 Thermal Characteristics ............................................................ 16 Digital Outputs ........................................................................... 47 ESD Caution ................................................................................ 16 ADC Overrange .......................................................................... 47 Pin Configuration and Function Descriptions ........................... 17 Multichip Synchronization............................................................ 48 Typical Performance Characteristics ........................................... 19 SYNC± Setup and Hold Window Monitor ............................. 49 Equivalent Circuits ......................................................................... 22 Test Modes ....................................................................................... 51 Theory of Operation ...................................................................... 24 ADC Test Modes ........................................................................ 51 ADC Architecture ...................................................................... 24 Serial Port Interface (SPI) .............................................................. 52 Analog Input Considerations.................................................... 24 Configuration Using the SPI ..................................................... 52 Voltage Reference ....................................................................... 26 Hardware Interface ..................................................................... 52 Clock Input Considerations ...................................................... 27 SPI Accessible Features .............................................................. 52 Power-Down/Standby Mode..................................................... 28 Memory Map .................................................................................. 53 Temperature Diode .................................................................... 28 Reading the Memory Map Register Table............................... 53 ADC Overrange and Fast Detect .................................................. 29 Memory Map Register Table ..................................................... 54 ADC Overrange .......................................................................... 29 Applications Information .............................................................. 63 Fast Threshold Detection (FD_A and FD_B) ........................ 29 Power Supply Recommendations............................................. 63 Signal Monitor ................................................................................ 30 Outline Dimensions ....................................................................... 64 Digital Downconverters (DDCs).................................................. 31 Ordering Guide .......................................................................... 64 DDC I/Q Input Selection .......................................................... 31 REVISION HISTORY 4/15—Revision 0: Initial Version Rev. 0 | Page 2 of 64 Product Overview Online Documentation Design Resources Discussion Data Sheet Sample & Buy AD9684 The AD9684 has several functions that simplify the automatic gain control (AGC) function in a communications receiver. The programmable threshold detector allows monitoring of the incoming signal power using the fast detect output bits of the ADC. 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 reduce the system gain to avoid an overrange condition at the ADC input. In addition to the fast detect outputs, the AD9684 also offers signal monitoring capability. The signal monitoring block provides additional information about the signal that the ADC digitized. The dual ADC output data is routed directly to the one external, 14-bit LVDS output port, supporting double data rate (DDR) formatting. An external data clock and status bit are offered for data capture flexibility. The LVDS outputs have several configurations, depending on the acceptable rate of the receiving logic device and the sampling rate of the ADC. Multiple device synchronization is supported through the SYNC± input pins. The AD9684 has flexible power-down options that allow significant power savings when desired. All of these features can be programmed using a 1.8 V to 3.4 V capable 3-wire serial port interface (SPI). The AD9684 is available in a Pb-free, 196-ball ball grid array (BGA) and is specified over the −40°C to +85°C industrial temperature range. This product is protected by a U.S. patent. PRODUCT HIGHLIGHTS 1. 2. 3. 4. 5. 6. 7. Rev. 0 | Page 3 of 64 Wide full power bandwidth supports intermediate frequency (IF) sampling of signals up to 2 GHz. Buffered inputs with programmable input termination ease filter design and implementation. Four integrated wideband decimation filters and 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. SYNC± input allows synchronization of multiple devices. 12 mm × 12 mm, 196-ball BGA_ED. Product Overview Online Documentation Design Resources Sample & Buy Discussion AD9684 Data Sheet SPECIFICATIONS DC SPECIFICATIONS AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified maximum sampling rate (500 MSPS), 1.7 V p-p full-scale differential input, 1.0 V internal reference, AIN = −1.0 dBFS, 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 INPUT-REFERRED NOISE VREF = 1.0 V ANALOG INPUTS Differential Input Voltage Range (Programmable) Common-Mode Voltage (VCM) Differential Input Capacitance 1 Analog Input Full Power Bandwidth POWER SUPPLY AVDD1 AVDD2 AVDD3 DVDD DRVDD SPIVDD IAVDD1 IAVDD2 IAVDD3 IDVDD IDRVDD ISPIVDD POWER CONSUMPTION Total Power Dissipation 2 Power-Down Dissipation Standby 1 2 Temperature Full Full Full Full Full Full Full Full Min 14 −0.3 −6.5 −0.6 −4.5 Typ Max Unit Bits Guaranteed 0 0 0 0 ±0.5 ±2.5 +0.3 +0.3 +6.5 +5.0 +0.7 +5.0 % FSR % FSR % FSR % FSR LSB LSB 25°C 25°C Full ±3 −39 1.0 ppm/°C ppm/°C V 25°C 2.63 LSB rms Full 25°C 25°C 25°C 1.46 2.06 2.05 1.5 2 2.06 V p-p V pF GHz Full Full Full Full Full Full Full Full Full Full Full Full 1.22 2.44 3.2 1.22 1.22 1.22 1.25 2.50 3.3 1.25 1.25 1.8 448 396 103 108 106 2 1.28 2.56 3.4 1.28 1.28 3.4 503 455 124 127 119 6 V V V V V V mA mA mA mA mA mA Full Full Full Differential capacitance is measured between the VIN+x and VIN−x pins (x = A or B). Parallel interleaved LVDS mode. The power dissipation on DRVDD changes with the output data mode used. Rev. 0 | Page 4 of 64 2.2 710 1.0 W mW W Product Overview Online Documentation Design Resources Sample & Buy Discussion Data Sheet AD9684 AC SPECIFICATIONS AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified maximum sampling rate (500 MSPS), 1.7 V p-p full-scale differential input, 1.0 V internal reference, AIN = −1.0 dBFS, default SPI settings, TA = 25°C, unless otherwise noted. Table 2. Parameter 1 ANALOG INPUT FULL SCALE NOISE DENSITY 2 SIGNAL-TO-NOISE RATIO (SNR) 3 fIN = 10 MHz fIN = 170 MHz fIN = 340 MHz fIN = 450 MHz fIN = 765 MHz fIN = 985 MHz fIN = 1950 MHz SIGNAL-TO-NOISE RATIO AND DISTORTION RATIO (SINAD)3 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) 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)3 fIN = 10 MHz fIN = 170 MHz fIN = 340 MHz fIN = 450 MHz fIN = 765 MHz fIN = 985 MHz fIN = 1950 MHz WORST HARMONIC, SECOND OR THIRD3 fIN = 10 MHz fIN = 170 MHz fIN = 340 MHz fIN = 450 MHz fIN = 765 MHz fIN = 985 MHz fIN = 1950 MHz Temperature Full Full 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 Full 25°C 25°C 25°C 25°C 25°C Rev. 0 | Page 5 of 64 Min 67.5 67 10.8 76 Typ 2.06 −153 Max Unit V p-p dBFS/Hz 69.2 68.6 68.4 68.0 64.4 63.8 60.5 dBFS dBFS dBFS dBFS dBFS dBFS dBFS 68.7 68.5 67.6 67.2 63.8 62.5 58.3 dBFS dBFS dBFS dBFS dBFS dBFS dBFS 11.1 10.9 10.8 10.8 10.3 10.1 9.5 Bits Bits Bits Bits Bits Bits Bits 83 85 82 86 81 76 69 dBFS dBFS dBFS dBFS dBFS dBFS dBFS −83 −85 −82 −86 −81 −76 −69 −76 dBFS dBFS dBFS dBFS dBFS dBFS dBFS Product Overview Online Documentation Design Resources Sample & Buy Discussion AD9684 Data Sheet Parameter 1 WORST OTHER, EXCLUDING SECOND OR THIRD HARMONIC3 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), AIN1 AND AIN2 = −7 dBFS fIN1 = 185 MHz, fIN2 = 188 MHz fIN1 = 338 MHz, fIN2 = 341 MHz CROSSTALK 4 FULL POWER BANDWIDTH Temperature Min Typ 25°C Full 25°C 25°C 25°C 25°C 25°C −93 −92 −90 −92 −89 −89 −85 25°C 25°C 25°C 25°C −88 −87 96 2 Max −76 Unit dBFS dBFS dBFS dBFS dBFS dBFS dBFS 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 a low analog input frequency (30 MHz). 3 See Table 9 for the recommended settings for full-scale voltage and buffer current control. 4 Crosstalk is measured at 170 MHz with a −1.0 dBFS analog input on one channel and no input on the adjacent channel. 1 2 DIGITAL SPECIFICATIONS AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified maximum sampling rate (500 MSPS), 1.7 V p-p full-scale differential input, 1.0 V internal reference, AIN = −1.0 dBFS, 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 SYNC INPUTS (SYNC+, SYNC−) 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) LOGIC OUTPUTS (FD_A, FD_B) Logic Compliance Logic 1 Voltage Logic 0 Voltage Input Resistance Temperature Full Full Full Full Full Full Full Full Full Full Full Full Full Full Min 600 Rev. 0 | Page 6 of 64 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 0 Full Full Full Full Full Full Full Typ 0.8 0 mV p-p V kΩ pF CMOS 0.8 × SPIVDD 0.2 × SPIVDD 30 V V kΩ CMOS 0.8 × SPIVDD 0.2 × SPIVDD V V CMOS SPIVDD 0 30 V V kΩ Product Overview Online Documentation Design Resources Sample & Buy Discussion Data Sheet Parameter DIGITAL OUTPUTS (Dx±, 1 DCO±, STATUS±) Logic Compliance Differential Output Voltage Output Common-Mode Voltage (VCM) AC-Coupled Short-Circuit Current (IDSHORT) Differential Return Loss (RLDIFF) 2 Common-Mode Return Loss (RLCM) 2 Differential Termination Impedance 1 2 AD9684 Temperature Min Full Full 25°C 25°C 25°C 25°C Full Typ Max Unit 230 430 mV p-p 0 −100 8 6 80 1.8 +100 V mA dB dB Ω LVDS 100 120 Where x = 0 to 13. Differential and common-mode return loss is measured from 100 MHz to 0.75 MHz × baud rate. SWITCHING SPECIFICATIONS AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, SPIVDD = 1.8 V, specified maximum sampling rate, 1.7 V p-p full-scale differential input, 1.0 V internal reference, AIN = −1.0 dBFS, default SPI settings, TA = 25°C, unless otherwise noted. Table 4. Parameter CLOCK Clock Rate (at CLK+/CLK− Pins) Maximum Sample Rate 1 Minimum Sample Rate 2 Clock Pulse Width High Low LVDS DATA OUTPUT PARAMETERS Data Propagation Delay (tPD) 3 DCO± Propagation Delay (tDCO)3 DCO± to Data Skew Rising Edge Data (tSKEWR)3 Falling Edge Data (tSKEWF)3 STATUS± Propagation Delay (tSTATUS) 4 DCO± to STATUS± Skew (tFRAME)4 Data Propagation Delay (tPD)3 DCO± Propagation Delay (tDCO)3 LATENCY 5 Pipeline Latency Fast Detect Latency HB1 Filter Latency3 HB1 + HB2 Filter Latency3 HB1 + HB2 + HB3 Filter Latency3 HB1 + HB2 + HB3 + HB4 Filter Latency3 Fast Detect Latency Wake-Up Time 6 Standby Power-Down Temperature Min Full Full Full 0.25 500 250 Full Full 1000 1000 Full Full Full Full Full Full Full Full Full Full Full Full Full Full Full 25°C 25°C Rev. 0 | Page 7 of 64 Typ Max Unit 4 GHz MSPS MSPS ps ps 2.225 2.2 −150 850 −150 −25 1.025 2.2 −25 2.225 2.2 ns ns +100 1100 +100 35 28 50 101 217 433 28 1 4 ps ps ns ps ns ns Clock cycles Clock cycles Clock cycles Clock cycles Clock cycles Clock cycles Clock cycles ms ms Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet Parameter APERTURE Aperture Delay (tA) Aperture Uncertainty (Jitter, tj) Out of Range Recovery Time Temperature Min Typ Full Full Full Max Unit 530 55 1 ps fs rms Clock Cycles The maximum sample rate is the clock rate after the divider. The minimum sample rate operates at 300 MSPS. 3 This specification is valid for parallel interleaved, channel multiplexed, and byte mode output modes. 4 This specification is valid for byte mode output mode only. 5 No DDCs used. 6 Wake-up time is defined as the time required to return to normal operation from power-down mode or standby mode. 1 2 TIMING SPECIFICATIONS Table 5. Parameter CLK± to SYNC± TIMING REQUIREMENTS tSU_SR tH_SR SPI TIMING REQUIREMENTS tDS tDH tCLK tS tH tHIGH tLOW tEN_SDIO Description See Figure 2 Device clock to SYNC± setup time Device clock to SYNC± hold time See Figure 3 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 must be in a logic high state Minimum period that SCLK must be in a logic low state Time required for the SDIO pin to switch from an input to an output relative to the SCLK falling edge (not shown in Figure 3) Time required for the SDIO pin to switch from an output to an input relative to the SCLK rising edge (not shown in Figure 3) tDIS_SDIO Min Typ Max 117 −96 Unit ps ps 2 2 40 2 2 10 10 10 ns ns ns ns ns ns ns ns 10 ns Timing Diagrams CLK– CLK+ tH_SR tSU_SR 13015-002 SYNC– SYNC+ Figure 2. SYNC± 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 3. Serial Port Interface Timing Diagram Rev. 0 | Page 8 of 64 D4 D3 D2 D1 D0 DON’T CARE 13015-003 SCLK DON’T CARE Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 APERTURE DELAY N + 35 N VIN±x N + 39 N + 36 N + 40 N–1 N+x N + 37 N+y N + 38 N + 41 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET SYNC+ SYNC– CLK+ CLK– tCLK FIXED DELAY FROM SYNC EVENT TO DCO KNOWN PHASE CONSTANT LATENCY = X CLK CYCLES tDCO tPD DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 90° PHASE ADJUST 1 DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 270° PHASE ADJUST 2 tSKEWF tSKEWR CONVERTER 0 CONVERTER 0 CONVERTER 0 CONVERTER 0 CONVERTER 0 SAMPLE SAMPLE SAMPLE SAMPLE SAMPLE [N] [N + 1] [N + 2] [N + 3] [N + 4] STATUS+ (OVERRANGE/STATUS BIT) STATUS STATUS STATUS STATUS STATUS STATUS STATUS D13± D13 D13 D13 D13 D13 D13 D13 D0± D0 D0 D0 D0 D0 D0 D0 190° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. 2270° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. Figure 4. Parallel Interleaved Mode—One Converter, ≤14-Bit Data Rev. 0 | Page 9 of 64 13015-004 STATUS– Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet APERTURE DELAY N + 36 N VIN±x N + 38 N+x N + 37 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET SYNC+ SYNC– CLK+ CLK– CONSTANT LATENCY = X CLK CYCLES tDCO tCLK tPD DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST tSKEWR STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP STATUS+ (OVERRANGE/STATUS BIT) tSKEWF CONVERTER 0 SAMPLE [N] CONVERTER 1 SAMPLE [N] CONVERTER 0 SAMPLE [N + 1] CONVERTER 1 SAMPLE [N + 1] CONVERTER 0 SAMPLE [N + 2] STATUS STATUS STATUS STATUS STATUS STATUS STATUS STATUS D13± D13 D13 D13 D13 D13 D13 D13 D13 D0± D0 D0 D0 D0 D0 D0 D0 D0 Figure 5. Parallel Interleaved Mode—Two Converters, ≤14-Bit Data, Output Sample Rate < 625 MSPS Rev. 0 | Page 10 of 64 13015-005 STATUS– Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 APERTURE DELAY N + 36 N VIN±x N + 38 N+x N + 37 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET SYNC+ SYNC– CLK+ CLK– CONSTANT LATENCY = X CLK CYCLES tDCO tPD tCLK DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST tSKEWR STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP STATUS+ (OVERRANGE/STAUS BIT) STATUS– STATUS STATUS S[N – y] (ODD BITS) S[N – x] (EVEN BITS) STATUS tSKEWF CONVERTERS SAMPLE [N] CONVERTERS SAMPLE [N] CONVERTERS SAMPLE [N + 1] CONVERTERS SAMPLE [N + 1] CONVERTERS SAMPLE [N + 2] STATUS STATUS STATUS STATUS STATUS S[N] (ODD BITS) S[N + 1] (EVEN BITS) S[N + 1] (ODD BITS) S[N + 2] (EVEN BITS) S[N] S[N – 1] (ODD BITS) (EVEN BITS) CHANNEL A D0±/D1± Figure 6. Channel Multiplexed (Even/Odd) Mode—One Converter, ≤14-Bit Data Rev. 0 | Page 11 of 64 13015-006 CHANNEL A D12±/D13± Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet APERTURE DELAY N + 36 N VIN±x N + 38 N+x N + 37 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET SYNC+ SYNC– CLK+ CLK– tCLK CONSTANT LATENCY = X CLK CYCLES tDCO tPD DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST tSKEWR STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP STATUS+ (OVERRANGE/STATUS BIT) STATUS– STATUS tSKEWF CONVERTERS SAMPLE [N] CONVERTERS SAMPLE [N] CONVERTERS SAMPLE [N + 1] CONVERTERS SAMPLE [N + 1] CONVERTERS SAMPLE [N + 2] STATUS STATUS STATUS STATUS STATUS STATUS STATUS S[N – y] (ODD BITS) S[N – x] (EVEN BITS) S[N] S[N – 1] (ODD BITS) (EVEN BITS) S[N] (ODD BITS) S[N + 1] (EVEN BITS) S[N + 1] (ODD BITS) S[N + 2] (EVEN BITS) S[N – y] (ODD BITS) S[N – x] (EVEN BITS) S[N] S[N – 1] (ODD BITS) (EVEN BITS) S[N] (ODD BITS) S[N + 1] (EVEN BITS) S[N + 1] (ODD BITS) S[N + 2] (EVEN BITS) CHANNEL A D12±/D13± CHANNEL B D12±/D13± CHANNEL B D0±/D1± Figure 7. Channel Multiplexed (Even/Odd) Mode—Two Converters, ≤14-Bit Data, Output Sample Rate < 625 MSPS Rev. 0 | Page 12 of 64 13015-007 CHANNEL A D0±/D1± Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 APERTURE DELAY N+z N VIN±x N–1 N + 39 N + 36 N + 40 N + 42 N+x N + 37 N + 38 N+y N + 41 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET SYNC+ SYNC– CLK+ CLK– tCLK FIXED DELAY FROM SYNC EVENT TO DCO KNOWN PHASE CONSTANT LATENCY = X CLK CYCLES tDCO tPD tSTATUS DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 90° PHASE ADJUST 1 DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 270° PHASE ADJUST 2 tFRAME STATUS– (FRAME CLOCK OUTPUT)3 STATUS+ (OVERRANGE STATUS BIT) FRAME 1 FRAME 0 STAUS+ tSKEWF tSKEWR I0[N] EVEN I0[N] ODD Q 0[N] EVEN Q 0[N] ODD I0[N + 1] EVEN I0[N + 1] Q 0[N + 1] Q 0[N + 1] ODD EVEN ODD PAR 4 PAR STATUS PAR STATUS PAR STATUS PAR STATUS PAR D7± D15 D15 D14 D15 D14 D15 D14 D15 D14 D15 D0± D1 D1 D0 D1 D0 D1 D0 D1 D0 D1 STATUS– 1) ENABLED (ALWAYS ON). 2) DISABLED (ALWAYS OFF). 3) GAPPED PERIODIC (CONDITIONALLY ENABLED BASED ON PSEUDO-RANDOM BIT). 4STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP. Figure 8. LVDS Byte Mode—Two Virtual Converters, One DDC, I/Q Data Decimate by 4 Rev. 0 | Page 13 of 64 13015-008 190° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. 2270° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. 3FRAME CLOCK OUTPUT SUPPORTS 3 MODES OF OPERATION: Rev. 0 | Page 14 of 64 D1 D0± N+y N+z N + 36 N + 37 N + 38 N + 39 N + 40 N + 41 N + 42 tPD tSTATUS D0 D14 STATUS D1 D15 PAR D0 D14 STATUS Q0[N] EVEN tSKEWF I0[N] ODD D1 D15 PAR Q0[N] ODD D0 D14 STATUS I1[N + 1] EVEN FRAME 0 FRAME 0 D1 D15 PAR I1[N] ODD D0 D14 STATUS Q1[N] EVEN FIXED DELAY FROM SYNC EVENT TO DCO KNOWN PHASE I0[N] EVEN 1) ENABLED (ALWAYS ON). 2) DISABLED (ALWAYS OFF). 3) GAPPED PERIODIC (CONDITIONALLY ENABLED BASED ON PSEUDO-RANDOM BIT). 4STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP. D1 D15 PAR tSKEWR tFRAME CONSTANT LATENCY = X CLK CYCLES tDCO D1 D15 PAR Q1[N] ODD D0 D14 STATUS I0[N+1] EVEN D1 D15 PAR I0[N+1] ODD D0 D14 STATUS Q0[N+1] EVEN D1 D15 PAR Q0[N+1] ODD D0 D14 STATUS I1[N + 1] EVEN D1 D15 PAR I1[N+1] ODD FRAME 1 FRAME 1 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET N+x APERTURE DELAY D0 D14 STATUS Q1[N+1] EVEN D1 D15 PAR Q1[N+1] ODD Design Resources 190° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. 2270° PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. 3FRAME CLOCK OUTPUT SUPPORTS 3 MODES OF OPERATION: D15 PAR4 D7± STATUS– STATUS+ (OVERRANGE STATUS BIT) STATUS+ STATUS– (FRAME CLOCK OUTPUT)3 tCLK N–1 N Online Documentation DCO± (DATA CLOCK OUTPUT) 270° PHASE ADJUST 2 DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 90° PHASE ADJUST 1 DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST CLK– CLK+ SYNC– SYNC+ VIN±x Product Overview Discussion Sample & Buy AD9684 Data Sheet Figure 9. LVDS Byte Mode—Four Virtual Converters, Two DDCs, ≤16-Bit Data, I/Q Data Decimate by 8 13015-010 Rev. 0 | Page 15 of 64 N+4 N+5 N+6 N+7 N+8 N+9 D14 D15 D1 D0± D1 D15 PAR D0 D14 STATUS D1 D15 PAR I0[N] ODD D0 D14 STATUS Q0[N] EVEN D1 D15 PAR D0 D14 STATUS Q0[N] I1[N] ODD EVEN PHASE ADJUST IS GENERATED USING THE FALLING EDGE OF CLK±. OF CLK±. D0 STATUS PAR4 I0[N] EVEN N + 10 D0 D14 STATUS Q1[N] EVEN D1 D15 PAR D0 D14 STATUS Q1[N] I2[N] ODD EVEN FRAME 0 D1 D15 PAR I2[N] ODD D0 D14 STATUS Q2[N] EVEN D1 D15 PAR D0 D14 STATUS Q2[N] I3[N] ODD EVEN D1 D15 PAR I0[N] ODD D0 D14 STATUS Q3[N] EVEN D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D1 D15 PAR D0 D14 STATUS D0 D14 D15 D1 STATUS PAR D1 D15 PAR Q3[N] I0[N+1] I0[N+1] Q0[N+1]Q0[N+1] I1[N+1] I1[N+1] Q1[N+1]Q1[N+1] I2[N+1] I2[N+1] Q2[N+1]Q2[N+1] I3[N+1] I3[N+1] Q3[N+1]Q3[N+1] ODD EVEN ODD EVEN ODD EVEN ODD EVEN ODD EVEN ODD EVEN ODD EVEN ODD EVEN ODD FRAME 1 Design Resources 1) ENABLED (ALWAYS ON). 2) DISABLED (ALWAYS OFF). 3) GAPPED PERIODIC (CONDITIONALLY ENABLED BASED ON PSEUDO-RANDOM BIT). 4STATUS BIT SELECTED BY REGISTER 0x559, BITS[2:0] IN THE REGISTER MAP. D1 D15 PAR I1[N] ODD FIXED DELAY FROM SYNC EVENT TO DCO KNOWN PHASE 2270° PHASE ADJUST IS GENERATED USING THE FALLING EDGE 3FRAME CLOCK OUTPUT SUPPORTS 3 MODES OF OPERATION: 190° N+2 N+3 SYNCHRONOUS LOW TO HIGH TRANSITIONS OF THE SYNC SIGNAL CAPTURED ON THE RISING EDGE OF THE CLK SIGNAL CAUSES THE DCO INTERNAL DIVIDER TO BE RESET N+1 APERTURE DELAY CONSTANT LATENCY = X CLK CYCLES N D7± STATUS+ (STATUS BIT) STATUS– STATUS+ (FRAME CLOCK OUTPUT)3 DCO± (DATA CLOCK OUTPUT) 270° PHASE ADJUST2 DCO± (DATA CLOCK OUTPUT) 90° PHASE ADJUST1 DCO± (DATA CLOCK OUTPUT) 180° PHASE ADJUST DCO± (DATA CLOCK OUTPUT) 0° PHASE ADJUST CLK+ (ENCODE CLOCK) SYNC+ (SYSTEM REFERENCE) VIN±x N–1 Online Documentation 13015-011 Product Overview Discussion Data Sheet Figure 10. LVDS Byte Mode—Eight Virtual Converters, Four DDCs, ≤16-Bit Data, I/Q Data Decimate by 16 Sample & Buy AD9684 Product Overview Online Documentation Design Resources Sample & Buy Discussion AD9684 Data Sheet ABSOLUTE MAXIMUM RATINGS THERMAL CHARACTERISTICS Table 6. Parameter Electrical AVDD1 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 VIN±x Maximum Swing PDWN/STBY to AGND Environmental Operating Temperature Range (TCASE) Maximum Junction Temperature Storage Temperature Range (Ambient) Rating Typical θJA, θJB, 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 θJB. The use of appropriate thermal management techniques is recommended to ensure that the maximum junction temperature does not exceed the limits shown in Table 7. 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 4.3 V p-p −0.3 V to SPIVDD + 0.3 V PCB Type JEDEC 2s2p Board −40°C to +85°C 10-Layer PCB 125°C −65°C to +150°C Table 7. Simulated Thermal Data Airflow Velocity (m/sec) 0.0 1.0 2.5 0.0 1.0 2.5 θJB 6.31, 3 5.91, 3 5.71, 3 4.6 4.6 4.6 θJC_TOP 4.71, 5 N/A4 N/A4 4.7 N/A4 N/A4 θJC_BOT 1.21, 5 N/A4 N/A4 1.2 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). 3 Per JEDEC JESD51-8 (still air). 4 N/A means not applicable. 5 Per MIL-STD 883, Method 1012.1. 1 2 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. θJA 17.81, 2 15.61, 2 15.01, 2 13.8 12.7 12.0 ESD CAUTION Rev. 0 | Page 16 of 64 Unit °C/W °C/W °C/W °C/W °C/W °C/W Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 1 2 3 4 5 6 7 8 9 10 11 12 13 14 A AGND AGND AGND AVDD2 AVDD1 AGND CLK+ CLK– AGND AVDD1 AVDD2 AGND AGND AGND A B AVDD3 AGND AGND AVDD2 AVDD1 AGND AGND AGND AGND AVDD1 AVDD2 AGND AGND AVDD3 B C AVDD3 AGND AGND AVDD2 AVDD1 AGND SYNC+ SYNC– AGND AVDD1 AVDD2 AGND AGND AVDD3 C D AGND AGND AGND AVDD2 AVDD1 AGND AVDD1 AGND AGND AVDD1 AVDD2 AGND AGND AGND D E VIN–B AGND AGND AVDD2 AVDD1 AGND AGND AGND AGND AVDD1 AVDD2 AGND AGND VIN–A E F VIN+B AGND AGND AVDD2 AGND AGND AGND AGND AGND AGND AVDD2 AGND AGND VIN+A F G AGND AGND AGND AGND AGND AGND AGND AGND AGND AGND AVDD2 AGND AGND AGND G H AGND AGND AGND CSB AGND AGND AGND AGND AGND V_1P0 AGND AGND AGND AGND H J FD_B AGND AGND SCLK SPIVDD AGND AGND AGND AGND AVDD2 SPIVDD AGND PDWN/STBY FD_A J K DGND DGND AGND SDIO AGND AGND AGND AGND AGND AGND AGND AGND DCO– DCO+ K L DVDD DVDD DGND DGND AGND AGND AGND AGND AGND AGND AGND AGND STATUS– STATUS+ L M D1+ D1– DVDD DVDD DRVDD DRVDD DRVDD DRGND DRGND DRGND DRGND DRGND D13– D13+ M N D2– D3– D4– D5– D6– D0– DRVDD DRGND D7– D8– D9– D10– D11– D12– N P D2+ D3+ D4+ D5+ D6+ D0+ DRVDD DRGND D7+ D8+ D9+ D10+ D11+ D12+ P 1 2 3 4 5 7 8 9 10 11 12 13 14 6 Figure 11. Pin Configuration (Top View) Table 8. Pin Function Descriptions Pin No. Power Supplies A5, A10, B5, B10, C5, C10, D5, D7, D10, E5, E10 A4, A11, B4, B11, C4, C11, D4, D11, E4, E11, F4, F11, G11, J10 B1, B14, C1, C14 L1, L2, M3, M4 M5, M6, M7, N7, P7 J5, J11 K1, K2, L3, L4 M8 to M12, N8, P8 A1, A2, A3, A6, A9, A12, A13, A14, B2, B3, B6, B7, B8, B9, B12, B13, C2, C3, C6, C9, C12, C13, D1, D2, D3, D6, D8, D9, D12, D13, D14, E2, E3, E6 to E9, E12, E13, F2, F3, F5 to F10, F12, F13, G1 to G10, G12, G13, G14, H1, H2, H3, H5 to H9, H11 to H14, J2, J3, J6 to J9, J12, K3, K5 to K12, L5 to L12 Mnemonic Type Description AVDD1 Supply Analog Power Supply (1.25 V Nominal). AVDD2 Supply Analog Power Supply (2.50 V Nominal). AVDD3 DVDD DRVDD SPIVDD DGND DRGND AGND Supply Supply Supply Supply Ground Ground Ground Analog Power Supply (3.3 V Nominal) Digital Power Supply (1.25 V Nominal). Digital Driver Power Supply (1.25 V Nominal). Digital Power Supply for SPI (1.8 V to 3.4 V). Ground Reference for DVDD. Ground Reference for DRVDD. Ground Reference for AVDD. Rev. 0 | Page 17 of 64 13015-012 PIN CONFIGURATION AND FUNCTION DESCRIPTIONS Product Overview Online Documentation Design Resources AD9684 Discussion Sample & Buy Data Sheet Pin No. Analog E14, F14 E1, F1 H10 Mnemonic Type Description VIN−A, VIN+A VIN−B, VIN+B V_1P0 Input Input Input/DNC A7, A8 CMOS Outputs J14, J1 Digital Inputs C7, C8 Data Outputs N6, P6 M1, M2 N1, P1 N2, P2 N3, P3 N4, P4 N5, P5 N9, P9 N10, P10 N11, P11 N12, P12 N13, P13 N14, P14 M13, M14 L13, L14 K13, K14 SPI Controls K4 J4 H4 J13 CLK+, CLK− Input ADC A Analog Input Complement/True. ADC B 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 as 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. Clock Input True/Complement. FD_A, FD_B Output Fast Detect Outputs for Channel A and Channel B. SYNC+, SYNC− Input Active High LVDS SYNC Input—True/Complement. D0−, D0+ D1+, D1− D2−, D2+ D3−, D3+ D4−, D4+ D5−, D5+ D6−, D6+ D7−, D7+ D8−, D8+ D9−, D9+ D10−, D10+ D11−, D11+ D12−, D12+ D13−, D13+ STATUS−, STATUS+ DCO−, DCO+ Output Output Output Output Output Output Output Output Output Output Output Output Output Output Output Output LVDS Lane 0 Output Data—Complement/True. LVDS Lane 1 Output Data—True/Complement. LVDS Lane 2 Output Data—Complement/True. LVDS Lane 3 Output Data—Complement/True. LVDS Lane 4 Output Data—Complement/True. LVDS Lane 5 Output Data—Complement/True. LVDS Lane 6 Output Data—Complement/True. LVDS Lane 7 Output Data—Complement/True. LVDS Lane 8 Output Data—Complement/True. LVDS Lane 9 Output Data—Complement/True. LVDS Lane 10 Output Data—Complement/True. LVDS Lane 11 Output Data—Complement/True. LVDS Lane 12 Output Data—Complement/True. LVDS Lane 13 Output Data—Complement/True. LVDS Status Output Data—Complement/True. LVDS Digital Clock Output Data—Complement/True. SDIO SCLK CSB PDWN/STBY Input/output Input Input Input SPI Serial Data Input/Output. SPI Serial Clock. SPI Chip Select (Active Low). Power-Down Input (Active High). The operation of this pin depends on the SPI mode and can be configured as powerdown or standby. Rev. 0 | Page 18 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 TYPICAL PERFORMANCE CHARACTERISTICS AVDD1 = 1.2 V, AVDD2= 2.5 V, AVDD3 = 3.3 V, DVDD = 1.2 V, DRVDD = 1.2 V, SPIVDD = 1.8 V, sampling rate = 500 MHz, 1.6 V p-p full-scale differential input, AIN = −1.0 dBFS, default SPI settings, TA = 25°C, 256k FFT sample, unless otherwise noted. 0 0 AIN = −1dBFS SNR = 68.9dBFS ENOB = 10.9 BITS SFDR = 83dBFS BUFFER CONTROL 1 = 2.0× –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 13015-014 –140 0 50 75 100 125 150 175 200 225 250 FREQUENCY (MHz) Figure 12. Single Tone FFT with fIN = 10.3 MHz Figure 15. Single Tone FFT with fIN = 450.3 MHz 0 0 AIN = −1dBFS SNR = 68.7dBFS ENOB = 10.9 BITS SFDR = 84dBFS BUFFER CONTROL 1 = 2.0× –20 AIN = −1dBFS SNR = 63.9dBFS ENOB = 10.3 BITS SFDR = 81dBFS BUFFER CONTROL 1 = 5.0× –20 –40 –40 AMPLITUDE (dBFS) AMPLITUDE (dBFS) 25 13015-017 –120 –60 –80 –100 –120 –60 –80 –100 –120 0 25 50 75 100 125 150 175 200 225 250 FREQUENCY (MHz) –140 13015-015 –140 0 25 50 75 100 125 150 175 200 225 250 225 250 FREQUENCY (MHz) Figure 13. Single Tone FFT with fIN = 170.3 MHz 13015-018 AMPLITUDE (dBFS) AIN = −1dBFS SNR = 67.3dBFS ENOB = 10.8 BITS SFDR = 86dBFS BUFFER CONTROL 1 = 4.5× –20 Figure 16. Single Tone FFT with fIN = 765.3 MHz 0 AIN = −1dBFS SNR = 62.8dBFS ENOB = 10.1 BITS SFDR = 76dBFS BUFFER CONTROL 1 = 5.0× 0 AIN = −1dBFS SNR = 67.8dBFS ENOB = 10.8 BITS SFDR = 82dBFS BUFFER CONTROL 1 = 4.5× –40 AMPLITUDE (dBFS) –20 –20 –60 –80 –60 –80 –100 –100 –120 –140 –140 0 25 50 75 100 125 150 175 200 225 FREQUENCY (MHz) 250 Figure 14. Single Tone FFT with fIN = 340.3 MHz Rev. 0 | Page 19 of 64 0 25 50 75 100 125 150 175 200 FREQUENCY (MHz) Figure 17. Single-Tone FFT with fIN = 985.3 MHz 13015-019 –120 13015-016 AMPLITUDE (dBFS) –40 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet 95 0 AIN = −1dBFS SNR = 61.7dBFS ENOB = 9.9 BITS SFDR = 70dBFS BUFFER CONTROL 1 = 8.0× –20 90 SFDR –40 85 SNR/SFDR (dBFS) AMPLITUDE (dBFS) 2.0× 2.0× 3.0× 3.0× 4.0× 4.0× –60 –80 80 75 –100 70 –120 65 0 25 50 75 100 125 150 175 200 225 250 FREQUENCY (MHz) 60 13015-020 –140 0 Figure 18. Single Tone FFT with fIN = 1205.3 MHz 0 AIN = −1dBFS SNR = 60.1dBFS ENOB = 9.7 BITS SFDR = 71dBFS BUFFER CONTROL 1 = 8.0× AIN1 AND AIN2 = –7dBFS SFDR = 88dBFS IMD2 = 95dBFS IMD3 = 88dBFS BUFFER CONTROL 1 = 2.0× –20 AMPLITUDE (dBFS) –60 –80 –100 –120 –40 –60 –80 0 25 50 75 100 125 150 175 200 225 250 FREQUENCY (MHz) –120 0 100 150 200 250 FREQUENCY (MHz) Figure 19. Single Tone FFT with fIN = 1630.3 MHz Figure 22. Two-Tone FFT with fIN1 = 184 MHz and fIN2 = 187 MHz 0 0 AIN = −1dBFS SNR = 59.0dBFS ENOB = 9.5 BITS SFDR = 69dBFS BUFFER CONTROL 1 = 8.0× –20 50 13015-025 –140 13015-021 –100 AIN1 AND AIN2 = –7dBFS SFDR = 87dBFS IMD2 = 94dBFS IMD3 = 87dBFS BUFFER CONTROL 1 = 2.0× –20 AMPLITUDE (dBFS) –40 –60 –80 –100 –40 –60 –80 –100 –140 0 25 50 75 100 125 150 175 200 225 FREQUENCY (MHz) 250 13015-022 –120 Figure 20. Single Tone FFT with fIN = 985.3 MHz –120 0 50 100 150 200 FREQUENCY (MHz) Figure 23. Two-Tone FFT; fIN1 = 338 MHz, fIN2 = 341 MHz Rev. 0 | Page 20 of 64 250 13015-026 AMPLITUDE (dBFS) –40 AMPLITUDE (dBFS) 500 Figure 21. SNR/SFDR vs. Analog Input Frequency (fIN); fIN < 500 MHz; Buffer Control 1 Setting = 2.0×, 3.0×, and 4.0× 0 –20 100 200 300 400 ANALOG INPUT FREQUENCY (MHz) 13015-023 SNR Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 90 0 SFDR –20 85 SNR/SFDR (dBFS) –40 IMD3 (dBc) –60 –80 SFDR (dBFS) 80 75 –100 70 IMD3 (dBFS) –120 –140 –90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6 INPUT AMPLITUDE (dBFS) 65 –40 Figure 24. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN) with fIN1 = 184 MHz and fIN2 = 187 MHz –25 –10 0 15 25 TEMPERATURE (°C) 40 55 70 85 13015-030 SNR 13015-027 SFDR/IMD3 (dBc AND dBFS) SFDR (dBc) Figure 27. SNR/SFDR vs. Temperature, fIN = 170.3 MHz 0 2.30 2.25 –20 2.20 POWER DISSIPATION (W) SFDR/IMD3 (dBc AND dBFS) SFDR (dBc) –40 IMD3 (dBc) –60 –80 SFDR (dBFS) –100 2.15 2.10 2.05 2.00 1.95 1.90 IMD3 (dBFS) Figure 25. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN) with fIN1 = 338 MHz and fIN2 = 341 MHz 100 SFDR (dBc) 80 SFDR (dBFS) 40 SNR (dBc) 20 0 –20 INPUT AMPLITUDE (dBFS) 0 13015-029 –6 –12 –18 –24 –30 –36 –42 –48 –60 –54 –66 –72 –78 –84 –40 –90 SNR/SFDR (dBc AND dBFS) SNR (dBc) 60 Figure 26. SNR/SFDR vs. Input Amplitude, fIN = 170.3 MHz Rev. 0 | Page 21 of 64 500 480 460 440 420 400 380 360 340 SAMPLE RATE (MSPS) Figure 28. Power Dissipation vs. Sample Rate (fS) (Default SPI) 13015-031 INPUT AMPLITUDE (dBFS) 1.80 320 –140 –90 –84 –78 –72 –66 –60 –54 –48 –42 –36 –30 –24 –18 –12 –6 13015-028 1.85 300 –120 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet EQUIVALENT CIRCUITS AVDD3 AVDD3 AVDD3 3pF 1.5pF 200Ω 67Ω 200Ω 28Ω VIN+x VCM BUFFER SPIVDD 200Ω 67Ω 28Ω 10pF 200Ω 400Ω ESD PROTECTED AVDD3 AVDD3 SPIVDD 1kΩ SCLK VIN–x 13015-036 AIN CONTROL (SPI) 3pF 1.5pF 13015-032 30kΩ ESD PROTECTED Figure 29. Analog Inputs Figure 33. SCLK Inputs AVDD1 SPIVDD 25Ω CLK+ ESD PROTECTED 30kΩ 1kΩ CSB AVDD1 20kΩ VCM = 0.85V 13015-033 20kΩ 13015-037 ESD PROTECTED 25Ω Figure 30. Clock Inputs Figure 34. CSB Input AVDD1 SYNC+ SPIVDD 1kΩ ESD PROTECTED SDO 20kΩ SDIO AVDD1 VCM = 0.85V ESD PROTECTED 20kΩ 13015-034 1kΩ Figure 31. SYNC± Inputs Figure 35. SDIO SPIVDD SWING CONTROL (SPI) ESD PROTECTED DRVDD DATA+ OUTPUT DRIVER DATA– DRGND ESD PROTECTED DRVDD Dx± DRGND FD FD_A/FD_B Dx± TEMPERATURE DIODE (FD_A ONLY) FD_x PIN CONTROL (SPI) 13015-035 SYNC– SDI 30kΩ 13015-038 LEVEL TRANSLATOR SPIVDD 1kΩ Figure 32. LVDS Digital Outputs, STATUS±, DCO± Figure 36. FD_A/FD_B Outputs Rev. 0 | Page 22 of 64 13015-039 CLK– Online Documentation Product Overview Design Resources Discussion Sample & Buy Data Sheet AD9684 AVDD2 SPIVDD ESD PROTECTED 1kΩ PDWN/STBY CONTROL (SPI) V_1P0 ESD PROTECTED 13015-040 PDWN/ STBY ESD PROTECTED 30kΩ V_1P0 PIN CONTROL (SPI) Figure 38. V_1P0 Input/Output Figure 37. PDWN/STBY Input Rev. 0 | Page 23 of 64 13015-041 ESD PROTECTED Product Overview Online Documentation Design Resources Sample & Buy Discussion AD9684 Data Sheet THEORY OF OPERATION The AD9684 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 output bits of the ADC. 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 reduce the system gain to avoid an overrange condition at the ADC input. The LVDS outputs can be configured depending on the decimation ratio. Multiple device synchronization is supported through the SYNC± input pins. ADC ARCHITECTURE The architecture of the AD9684 consists of an input buffered pipelined ADC. The input buffer provides 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 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 kickback from the ADC. The buffer is optimized for high linearity, low noise, and low power. The quantized outputs from each stage are combined into a final 14-bit result in the digital correction logic. The pipelined architecture permits the first stage to operate with a new input sample, whereas the remaining stages operate with the preceding samples. Sampling occurs on the rising edge of the clock. ANALOG INPUT CONSIDERATIONS The analog input to the AD9684 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, helps reduce the peak transient current injected from the output stage of the driving source. In addition, low Q inductors or ferrite beads can be placed on each leg 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 For best dynamic performance, the source impedances driving VIN+x and VIN−x must be matched 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 AD9684, the available span is 2.06 V p-p differential. Differential Input Configurations There are several ways to drive the AD9684, either actively or passively. However, optimum performance is achieved by driving the analog input differentially. For applications in which SNR and SFDR are key parameters, differential transformer coupling is the recommended input configuration because the noise performance of most amplifiers is not adequate to achieve the true performance of the AD9684. For low to midrange frequencies, a double balun or double transformer network is recommended for optimum performance of the AD9684 (see Figure 39). For higher frequencies in the second and third Nyquist zones, it is better to remove some of the front-end passive components to ensure wideband operation (see Figure 40). ETC1-11-13/ MABA007159 1:1Z 10Ω 10Ω 0.1µF 25Ω 4pF 0.1µF 25Ω 10Ω 2pF ADC 10Ω 0.1µF 13015-042 The dual ADC cores feature a multistage, differential pipelined architecture with integrated output error correction logic. Each ADC features wide bandwidth inputs that support a variety of user selectable input ranges. An integrated voltage reference eases design considerations. frequencies. Place either a differential capacitor or two singleended 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, see the AN-742 Application Note, the AN-827 Application Note, and the Analog Dialogue article “Transformer-Coupled Front-End for Wideband A/D Converters” (Volume 39, April 2005). In general, the precise values depend on the application. 4pF Figure 39. Differential Transformer-Coupled Configuration for First and Second Nyquist Frequencies 25Ω MARKI BAL-0006 OR BAL-0006SMG 25Ω 25Ω 0.1µF 0.1µF 25Ω 0.1µF ADC 13015-043 The AD9684 has two analog input channels and 14 LVDS output lane pairs. The ADC is designed to sample wide bandwidth analog signals of up to 2 GHz. The AD9684 is optimized for wide input bandwidth, a high sampling rate, excellent linearity, and low power in a small package. Figure 40. Differential Transformer-Coupled Configuration for Second and Third Nyquist Frequencies Rev. 0 | Page 24 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 Input Common Mode 95 Analog Input Controls and SFDR Optimization 4.5× 85 3.0× 75 SFDR (dBFS) The analog inputs of the AD9684 are internally biased to the common mode as shown in Figure 41. The common-mode buffer has a 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. 2.0× 65 1.5× 55 The AD9684 offers flexible controls for the analog inputs, such as input termination and buffer current. All of the available controls are shown in Figure 41. 1.0× 35 50 AVDD3 150 100 200 250 300 350 400 450 500 INPUT FREQUENCY (MHz) 13015-046 45 Figure 43. Buffer Current Sweeps (SFDR vs. Input Frequency and IBUFF), 10 MHz < fIN < 500 MHz VIN+x 3pF 200Ω 67Ω 200Ω 28Ω 90 AVDD3 VCM BUFFER SFDR (dBFS) 200Ω 67Ω 28Ω 200Ω 400Ω 10pF 4.5× 5.0× 6.0× 7.0× 8.0× 85 AVDD3 VIN–x 80 75 3pF 65 500 Figure 41. Analog Input Controls (Should the AIN Using Register 0x018, the buffer currents on each channel can be scaled to optimize the SFDR over various input frequencies and bandwidths of interest. As the input buffer currents are set, the amount of current required by the AVDD3 supply changes. For a complete list of buffer current settings, see Table 29. 550 600 650 700 750 800 850 900 950 1000 INPUT FREQUENCY (MHz) 13015-047 70 13015-044 AIN CONTROL (SPI) REGISTERS (REG 0x008, REG 0x015, REG 0x016, REG 0x018, REG 0x025) Figure 44. Buffer Current Sweeps (SFDR vs. Input Frequency and IBUFF), 500 MHz < fIN < 1000 MHz 80 75 250 70 210 IAVDD3 (mA) 190 170 60 55 150 130 50 110 4.5× 5.0× 6.0× 7.0× 8.0× 8.0× 45 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 90 INPUT FREQUENCY (MHz) 2.5× 3.5× 4.5× 5.5× 6.5× 7.5× 8.5× BUFFER CURRENT SETTING Figure 42. AVDD3 Power (IAVDD3) vs. Buffer Current Control Setting in Register 0x018 13015-045 70 50 1.5× 65 13015-048 SFDR (dBFS) 230 Figure 45. Buffer Current Sweeps (SFDR vs. Input Frequency and IBUFF), 1 GHz < fIN < 2 GHz, Front-End Network Shown in Figure 40 Figure 43, Figure 44, and Figure 45 show how the SFDR can be optimized using the buffer current setting in Register 0x018 for different Nyquist zones. At frequencies greater than 1 GHz, it is better to run the ADC at input amplitudes less than −1 dBFS (−3 dBFS, for example). This greatly improves the linearity of the converted signal without sacrificing SNR performance. Rev. 0 | Page 25 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet 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 47 shows the typical drift characteristics of the internal 1.0 V reference. Table 9 shows the recommended buffer current and full-scale voltage settings for the different analog input frequency ranges. Absolute Maximum Input Swing The absolute maximum input swing allowed at the inputs of the AD9684 is 4.3 V p-p differential. Signals operating near or at this level can cause permanent damage to the ADC. 1.0010 1.0009 1.0008 VOLTAGE REFERENCE V_1P0 VOLTAGE (V) 1.0007 A stable and accurate 1.0 V voltage reference is built into the AD9684. This internal 1.0 V reference sets the full-scale input range of the ADC. For more information on adjusting the input swing, see Table 29. Figure 46 shows the block diagram of the internal 1.0 V reference controls. 1.0006 1.0005 1.0004 1.0003 1.0002 1.0000 VIN–A/ VIN–B 0.9999 0.9998 0 –50 INTERNAL V_1P0 GENERATOR ADC CORE FULL-SCALE VOLTAGE ADJUST 25 90 TEMPERATURE (°C) Figure 47. Typical V_1P0 Drift 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 48 shows how the ADR130 can be used to provide the external 1.0 V reference to the AD9684. The grayed out areas show unused blocks within the AD9684 while using the ADR130 to provide the external reference. INPUT FULL-SCALE RANGE ADJUST SPI REGISTER (REGISTER 0x025) V_1P0 13015-049 V_1P0 PIN CONTROL SPI REGISTER (REGISTER 0x024) Figure 46. Internal Reference Configuration and Controls Register 0x024 enables the user either to 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 reference voltage. For more information on adjusting the fullscale level of the AD9684, see the Memory Map Register Table section. Table 9. SFDR Optimization for Input Frequencies Frequency DC to 250 MHz 250 MHz to 500 MHz 500 MHz to 1 GHz 1 GHz to 2 GHz Buffer Control 1 (Register 0x018) 0x20 (2.0×) 0x70 (4.5×) Input Full-Scale Range (Register 0x025) 0x0C (2.06 V p-p) 0x0C (2.06 V p-p) Input Full-Scale Control (Register 0x030) 0x04 0x04 Input Termination (Register 0x016) 1 0x0C/0x1C/0x6C 0x0C/0x1C/0x6C 0x80 (5.0×) 0x08 (1.46 V p-p) 0x18 0x0C/0x1C/0x6C 0xF0 (8.5×) 0x08 (1.46 V p-p) 0x18 0x0C/0x1C/0x6C The input termination can be changed to accommodate the application with little or no impact to ac performance. INTERNAL V_1P0 GENERATOR ADR130 INPUT 1 NC 2 GND SET 5 3 VIN 0.1µF FULL-SCALE VOLTAGE ADJUST NC 6 VOUT 4 V_1P0 0.1µF FULL-SCALE CONTROL Figure 48. External Reference Using the ADR130 Rev. 0 | Page 26 of 64 13015-054 1 13015-050 1.0001 VIN+A/ VIN+B Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 CLOCK INPUT CONSIDERATIONS Input Clock Divider For optimum performance, drive the AD9684 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 AD9684 contains an input clock divider with the ability to divide the Nyquist input clock by 1, 2, 4, and 8. The divider ratios can be selected using Register 0x10B. This is shown in Figure 52. Figure 49 shows a preferred method for clocking the AD9684. The low jitter clock source is converted from a single-ended signal to a differential signal using an RF transformer. 0.1µF CLK+ 100Ω CLK+ ADC CLK– 0.1µF CLK– ÷2 ÷4 Figure 49. 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 50 and Figure 51. 3.3V 71Ω 33Ω 0.1µF ADC Z0 = 50Ω 0.1µF 13015-052 CLK+ CLK– Input Clock Divider ½ Period Delay Adjustment Figure 50. Differential CML Sample Clock CLK+ LVDS DRIVER CLK+ 100Ω CLK– CLOCK INPUT 50Ω1 50Ω1 Clock Fine Delay Adjustment ADC CLK– 0.1µF 150Ω RESISTORS ARE OPTIONAL. 13015-053 0.1µF The input clock divider inside the AD9684 provides phase delay in increments of ½ the input clock cycle. Program Register 0x10C to enable this delay independently for each channel. 0.1µF 0.1µF CLOCK INPUT Figure 52. Clock Divider Circuit The AD9684 clock divider can be synchronized using the external SYNC± input. A valid SYNC± input 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 13015-055 1:1Z 50Ω 13015-051 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, the appropriate divider ratio must be programmed into the clock divider before applying the clock signal. This ensures that the current transients during device startup are controlled. Figure 51. 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 the 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 device. The AD9684 can be clocked at 2 GHz with the internal clock divider set to 2. The output of the divider offers a 50% duty cycle, high slew rate (fast edge) clock signal to the internal ADC. See the Memory Map section for more details on using this feature. To adjust the AD9684 sampling edge instant, write to Register 0x117 and Register 0x118. Setting Bit 0 of Register 0x117 enables the fine delay 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 adjust takes effect immediately when it is enabled via SPI writes. Enabling the clock fine delay adjustment in Register 0x117 causes a datapath reset. Clock Jitter Considerations High speed, high resolution ADCs are sensitive to the quality of the clock input. The degradation in SNR at a given input frequency (fA) due only to aperture jitter (tJ) can be calculated by 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 53). Rev. 0 | Page 27 of 64 Online Documentation Product Overview Design Resources Discussion Sample & Buy AD9684 POWER-DOWN/STANDBY MODE RMS CLOCK JITTER REQUIREMENT 120 The AD9684 has a PDWN/STBY pin that configures the device in power-down or standby mode. The default operation is the power-down function. The PDWN/STBY pin is a logic high pin. The power-down option can also be set via Register 0x03F and Register 0x040. SNR (dB) 110 100 16 BITS 90 14 BITS 80 12 BITS TEMPERATURE DIODE 70 10 BITS 60 40 30 1 10 100 ANALOG INPUT FREQUENCY (MHz) 1000 13015-056 0.125ps 0.25ps 0.5ps 1.0ps 2.0ps 8 BITS 50 Figure 53. Ideal SNR vs. Analog Input Frequency and Jitter Treat the clock input as an analog signal when aperture jitter may affect the dynamic range of the AD9684. Separate the power supplies for the 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 the clock by the original clock at the last step. For more in-depth information about jitter performance as it relates to ADCs, see the AN-501 Application Note and the AN-756 Application Note. Figure 54 shows the estimated SNR of the AD9684 across the input frequency for different clock induced jitter values. Estimate the SNR using the following equation: − SNR JITTER − SNR ADC SNR (dBFS) = 10log10 10 + 10 10 The AD9684 contains a diode-based temperature sensor for measuring the temperature of the die. This diode can output a voltage and serve as a coarse temperature sensor to monitor the internal die temperature. The temperature diode voltage can be output to the FD_A pin using the SPI. Use Register 0x028, Bit 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, Bits[2:0]. See Table 29 for more information. The voltage response of the temperature diode (with SPIVDD = 1.8 V) is shown in Figure 55. 0.90 0.85 DIODE VOLTAGE (V) 130 Data Sheet 0.80 0.75 0.70 75 0.65 0.60 –55 –45 –35 –25 –15 –5 SNR (dBFS) 65 60 55 100M INPUT FREQUENCY (Hz) 1G 10G 13015-057 10M 15 25 35 45 55 65 75 85 95 105 115 125 Figure 55. Diode Voltage vs. Temperature 50 45 1M 5 TEMPERATURE (°C) 25fs 50fs 75fs 100fs 125fs 150fs 175fs 200fs Figure 54. Estimated SNR Degradation for the AD9684 vs. Input Frequency and Clock Jitter Rev. 0 | Page 28 of 64 13015-058 70 Product Overview Online Documentation Design Resources Discussion Sample & Buy Data Sheet AD9684 ADC OVERRANGE AND FAST DETECT The operation of the upper threshold and lower threshold registers, along with the dwell time registers, is shown in Figure 56. In receiver applications, it is desirable to have a mechanism to reliably determine when the converter is about to be clipped. The standard overrange pin outputs information on the state of the analog input. It is also 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 AD9684 contains fast detect circuitry for individual channels to monitor the threshold and assert the FD_A and FD_B pins. The FD_x indicator is asserted if the input magnitude exceeds the value programmed in the fast detect upper threshold registers, 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 clock cycles (maximum). The approximate upper threshold magnitude is defined by Upper Threshold Magnitude (dBFS) = 20log(Threshold Magnitude/213) ADC OVERRANGE The ADC overrange indicator is asserted when an overrange is detected on the input of the ADC. The overrange indicator can be output on the STATUS± pins (when CSB > 0). The latency of this overrange indicator matches the sample latency. The FD 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, 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 AD9684 also records any overrange condition in any of the four virtual converters. The overrange status of each virtual converter is registered as a sticky bit in Register 0x563. The contents of Register 0x563 can be cleared using Register 0x562, by toggling the bits corresponding to the virtual converter to set and reset the position. Lower Threshold Magnitude (dBFS) = 20log(Threshold Magnitude/213) FAST THRESHOLD DETECTION (FD_A AND FD_B) For example, to set an upper threshold of −6 dBFS, write 0xFFF to Register 0x247 and Register 0x248. To set a lower threshold of −10 dBFS, write 0xA1D to Register 0x249 and Register 0x24A. The fast detect (FD) bit (enabled via the control bits in Register 0x559) is immediately set whenever the absolute value of the input signal exceeds the programmable upper threshold level. The FD bit is cleared only when the absolute value of the input signal drops below the lower threshold level for greater than the programmable dwell time. This feature provides hysteresis and prevents the FD bit from excessively toggling. 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, in Register 0x24B and Register 0x24C. See the Memory Map section (Register 0x040, and Register 0x245 to Register 0x24C in Table 29) for more details. UPPER THRESHOLD DWELL TIME TIMER RESET BY RISE ABOVE LOWER THRESHOLD DWELL TIME FD_A OR FD_B Figure 56. Threshold Settings for FD_A and FD_B Signals Rev. 0 | Page 29 of 64 TIMER COMPLETES BEFORE SIGNAL RISES ABOVE LOWER THRESHOLD 13015-059 MIDSCALE LOWER THRESHOLD Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 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 by reading back the internal values from the SPI port. A global, 24-bit programmable period controls the duration of the measurement. Figure 57 shows the simplified block diagram of the signal monitor block. The peak detector captures the largest signal within the observation period. The detector only observes 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. Derive the peak magnitude using the following equation: Peak Magnitude (dBFS) = 20log(Peak Detector Value/213) The magnitude of the input port signal is monitored over a programmable time period, which is determined by the signal SIGNAL MONITOR PERIOD REGISTER (SMPR) REG 0x271, REG 0x272, REG 0x273 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. 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. 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 magnitude storage register. DOWN COUNTER IS COUNT = 1? LOAD CLEAR FROM INPUT MAGNITUDE STORAGE REGISTER LOAD LOAD SIGNAL MONITOR HOLDING REGISTER COMPARE A>B Figure 57. Signal Monitor Block Rev. 0 | Page 30 of 64 TO STATUS± PINS AND MEMORY MAP 13015-060 FROM MEMORY MAP monitor period register (SMPR). To enable the peak detector function, set Bit 1 of Register 0x270 in the signal monitor control register. The 24-bit SMPR must be programmed before activating this mode. Product Overview Online Documentation Design Resources Discussion Sample & Buy Data Sheet AD9684 DIGITAL DOWNCONVERTERS (DDCs) The AD9684 includes four digital downconverters that provide filtering and reduce the output data rate. This digital processing section includes an NCO, a half-band decimating filter, a finite impulse response (FIR) filter, a gain stage, and a complex to real conversion stage. Each of these processing blocks has a control line that allows the block to be independently enabled and disabled to provide the desired processing function. The DDCs can be configured to output either real data or complex output data. DDC I/Q INPUT SELECTION The AD9684 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 29 for information on how to configure the DDCs. DDC I/Q OUTPUT SELECTION Each DDC channel has two output ports that can be paired to support both real or 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. The I/Q outputs to each DDC channel are controlled by the DDC complex to real enable bit in the DDC control registers (Bit 3 in Register 0x310, Register 0x330, Register 0x350, and Register 0x370). The Chip I only bit in the chip application mode register (Register 0x200, Bit 5) controls the chip output muxing of all the DDC channels. When all DDC channels use real outputs, set this bit high to 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. DDC GENERAL DESCRIPTION The four DDC blocks extract a portion of the full digital spectrum captured by the ADCs. 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 or 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) 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) When real outputs are necessary, this stage converts the complex outputs back to real outputs by performing an fS/4 mixing operation in addition to a filter to remove the complex component of the signal. Figure 58 shows the detailed block diagram of the DDCs implemented in the AD9684. Rev. 0 | Page 31 of 64 Online Documentation Product Overview Design Resources Discussion AD9684 Sample & Buy Data Sheet GAIN = 0dB OR 6dB COMPLEX TO REAL CONVERSION (OPTIONAL) GAIN = 0dB OR 6dB COMPLEX TO REAL CONVERSION (OPTIONAL) COMPLEX TO REAL CONVERSION (OPTIONAL) COMPLEX TO REAL CONVERSION (OPTIONAL) ADC SAMPLING AT fS GAIN = 0dB OR 6dB REAL/I GAIN = 0dB OR 6dB REAL/Q Q HB1 FIR DCM = 2 NCO + MIXER (OPTIONAL) HB2 FIR DCM = BYPASS OR 2 I HB3 FIR DCM = BYPASS OR 2 REAL/I HB4 FIR DCM = BYPASS OR 2 DDC 0 REAL/I CONVERTER 0 Q CONVERTER 1 SYNC± Q CONVERTER 3 REAL/Q Q ADC SAMPLING AT fS HB1 FIR DCM = 2 I HB2 FIR DCM = BYPASS OR 2 REAL/I HB3 FIR DCM = BYPASS OR 2 DDC 2 REAL/I CONVERTER 4 OUTPUT INTERFACE HB1 FIR DCM = 2 REAL/I CONVERTER 2 SYNC± NCO + MIXER (OPTIONAL) REAL/I HB2 FIR DCM = BYPASS OR 2 REAL/Q Q HB3 FIR DCM = BYPASS OR 2 NCO + MIXER (OPTIONAL) HB4 FIR DCM = BYPASS OR 2 I/Q CROSSBAR MUX I HB4 FIR DCM = BYPASS OR 2 DDC 1 REAL/I Q CONVERTER 5 SYNC± SYNC± SYNCHRONIZATION CONTROL CIRCUITS HB1 FIR DCM = 2 REAL/I CONVERTER 6 Q CONVERTER 7 13015-061 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 SYNC± Figure 58. DDC Detailed Block Diagram Figure 59 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 for 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 10 through 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. 0 | Page 32 of 64 Online Documentation Product Overview Design Resources Discussion Sample & Buy Data Sheet AD9684 ADC ADC SAMPLING AT fS REAL REAL INPUT—SAMPLED AT fS BANDWIDTH OF INTEREST IMAGE –fS/2 –fS/3 –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 fS/32 –fS/32 DC fS/16 –fS/16 –fS/8 BANDWIDTH OF INTEREST IMAGE (–6dB LOSS DUE TO NCO + MIXER) BANDWIDTH OF INTEREST (–6dB LOSS DUE TO NCO + MIXER) fS/8 fS/4 fS/3 fS/2 FILTERING STAGE HB4 FIR 4 DIGITAL HALF-BAND FILTERS (HB4 + HB3 + HB2 + HB1) I HALFBAND FILTER Q HALFBAND FILTER HB3 FIR 2 HALFBAND FILTER 2 HALFBAND FILTER HB4 FIR HB2 FIR 2 HALFBAND FILTER 2 HALFBAND FILTER HB3 FIR 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 59. DDC Theory of Operation Example (Real Input, Decimate by 16) Rev. 0 | Page 33 of 64 13015-062 6dB GAIN TO COMPENSATE FOR NCO + MIXER LOSS Product Overview Online Documentation Design Resources AD9684 Discussion Sample & Buy Data Sheet Table 10. DDC Samples When the Chip Decimation Ratio = 1 HB1 FIR (DCM 1 = 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 (DCM1 = 2) (DCM1 = 4) FIR (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 means decimation. Table 11. DDC Samples When the 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 = 4) (DCM 1 = 2) (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. 0 | Page 34 of 64 Product Overview Online Documentation Design Resources Discussion Sample & Buy Data Sheet 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 = 4) (DCM1 = 8) (DCM 1 = 2) 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 AD9684 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 means decimation. Table 12. DDC Samples When the Chip Decimation Ratio = 4 Real (I) Output (Complex to Real Enabled) HB3 FIR + HB2 FIR + HB1 HB4 FIR + HB3 FIR + HB2 FIR FIR (DCM 1 = 4) + HB1 FIR (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 1 Complex (I/Q) Outputs (Complex to Real Disabled) HB2 FIR + HB1 FIR HB3 FIR + HB2 FIR + HB4 FIR + HB3 FIR + HB2 FIR (DCM1 = 4) HB1 FIR (DCM1 = 8) + HB1 FIR (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 means decimation. Table 13. DDC Samples When the Chip Decimation Ratio = 8 Real (I) Output (Complex to Real Enabled) HB4 FIR + HB3 FIR + HB2 FIR + HB1 FIR (DCM 1 = 8) N N+1 N+2 N+3 N+4 N+5 N+6 N+7 1 Complex (I/Q) Outputs (Complex to Real Disabled) HB4 FIR + HB3 FIR + HB2 FIR + HB1 HB3 FIR + HB2 FIR + HB1 FIR(DCM1 = 8) FIR (DCM1 = 16) 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 DCM means decimation. Table 14. DDC Samples When the Chip Decimation Ratio = 16 Real (I) Output (Complex to Real Enabled) HB4 FIR + HB3 FIR + HB2 FIR + HB1 FIR (DCM 1 = 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 means decimation. Rev. 0 | Page 35 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 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 15. Table 15. Chip Decimation Ratio = 4, DDC 0 Decimation = 4 (Complex), and DDC 1 Decimation = 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 Output Port I I0 (N) DDC 0 Output Port Q Q0 (N) Output Port I I1 (N) DDC 1 Output Port Q Not applicable I0 (N + 1) Q0 (N + 1) I1 (N + 1) Not applicable I0 (N + 2) Q0 (N + 2) I1 (N) Not applicable I0 (N + 3) Q0 (N + 3) I1 (N + 1) Not applicable Rev. 0 | Page 36 of 64 Online Documentation Product Overview Design Resources Discussion Sample & Buy Data Sheet AD9684 FREQUENCY TRANSLATION GENERAL DESCRIPTION Variable IF Mode Frequency translation is accomplished using a 12-bit complex NCO with a digital quadrature mixer. The frequency translation translates either a real or complex input signal from an IF to a baseband complex digital output (carrier frequency = 0 Hz). The NCO and the mixers are enabled. The 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 a 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 The input samples are forced to 0.999 × full scale to positive full scale. The NCO is enabled. This test mode allows the NCOs to drive the decimation filters directly. Figure 60 and Figure 61 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 ADC SAMPLING AT fS cos(wt) 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 60. DDC NCO Frequency Tuning Word Selection—Real Inputs Rev. 0 | Page 37 of 64 13015-063 • • • • fS/4 Hz IF Mode Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet NCO FREQUENCY TUNING WORD (FTW) SELECTION 12-BIT NCO FTW = MIXING FREQUENCY/ADC SAMPLE RATE × 4096 I I + I I Q Q 90° PHASE 12-BIT NCO 90° 0° Q Q ADC SAMPLING AT fS Q Q I I – –sin(wt) QUADRATURE ANALOG MIXER + 2 ADCs + QUADRATURE DIGITAL REAL MIXER + NCO COMPLEX INPUT—SAMPLED AT fS QUADRATURE MIXER ADC SAMPLING AT fS 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 13015-064 –fS/2 DC Figure 61. DDC NCO Frequency Tuning Word Selection—Complex Inputs DDC NCO PLUS 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. The NCO introduces an additional 0.05 dB of loss. The total loss of a real input signal mixed down to baseband is 6.05 dB. For this reason, it is recommended to 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 (+fS/2 excluded) are represented using the following frequency words: When mixing a complex input signal down to baseband, the maximum value that each I/Q sample can reach is 1.414 × full scale after it passes through the complex mixer. To avoid an overrange of the I/Q samples and to keep the data bit-widths aligned with real mixing, introduce 3.06 dB of loss (0.707 × fullscale) in the mixer for complex signals. The NCO introduces an additional 0.05 dB of loss. 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 AD9684 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. Calculate the NCO frequency tuning word 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 AD9684 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. 0 | Page 38 of 64 Product Overview Online Documentation Design Resources Discussion Data Sheet Sample & Buy AD9684 For example, if the ADC sampling frequency (fS) is 1250 MSPS and the carrier frequency (fC) is 416.667 MHz, Use the following two methods to synchronize multiple PAWs within the chip: mod( 416.667,1250 ) NCO _ FTW = round 212 = 1365 MHz 1250 • This, in turn, converts to 0x555 in the 12-bit, twos complement representation for NCO_FTW. Calculate the actual carrier frequency based on the following equation: f C _ ACTUAL = NCO _ FTW × f S = 416.56 MHz 212 • A 12-bit POW is available for each NCO to create a known phase relationship between multiple AD9684 chips or individual DDC channels inside one AD9684 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 soft reset bit, accessible through the SPI, or through the assertion of the SYNC± pins. Note that the NCOs must be synchronized either through the SPI or through the SYNC± pins after all writes to the FTW or POW registers are complete. This synchronization 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, described in the Setting Up the NCO FTW and POW section. The phase increment value of each PAW is determined by the FTW. Using the SPI. Use the DDC NCO soft reset bit in the DDC synchronization control register (Register 0x300, Bit 4) to reset all the PAWs in the chip. This is accomplished by toggling the DDC NCO soft reset bit. Note that this method synchronizes DDC channels within the same AD9684 chip only. Using the SYNC± pins. When the SYNC± pins are enabled in the SYNC± control registers (Register 0x120 and Register 0x121), and the DDC synchronization is enabled in Bits[1:0] in the DDC synchronization control register (Register 0x300), any subsequent SYNC± event resets all the PAWs in the chip. Note that this method synchronizes DDC channels within the same AD9684 chip or DDC channels within separate AD9684 chips. Mixer The NCO is accompanied by a mixer, which operates similarly 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. 0 | Page 39 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet FIR FILTERS GENERAL DESCRIPTION Table 16 shows the different bandwidth options by including different half-band filters. In all cases, the DDC filtering stage of the AD9684 provides less than −0.001 dB of pass-band ripple and greater than 100 dB of stop band alias rejection. There are four sets of decimate by 2, low-pass, half-band, FIR filters (labeled HB1 FIR, HB2 FIR, HB3 FIR, and HB4 FIR in Figure 58) 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 17 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 16. DDC Filter Characteristics ADC Sample Rate (MSPS) 1000 1 DDC Decimation Ratio 2 (HB1) 4 (HB1 + HB2) 8 (HB1 + HB2 + HB3) 16 (HB1 + HB2 + HB3 + HB4) Real Output Sample Rate (MSPS) 1000 500 250 125 Complex (I/Q) Output Sample Rate (MSPS) 500 (I) + 500 (Q) 250 (I) + 250 (Q) 125 (I) + 125 (Q) 62.5 (I) + 62.5 (Q) Alias Protected Bandwidth (MHz) 385.0 192.5 96.3 48.1 Ideal SNR Improvement 1 (dB) 1 4 7 10 Pass-Band Ripple (dB) <−0.001 Alias Rejection (dB) >100 The ideal SNR improvement due to oversampling and filtering = 10log(bandwidth/(fS/2)). Table 17. 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) Outputs 1 <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 ratio. Rev. 0 | Page 40 of 64 Alias Protected Bandwidth for Complex (I/Q) Outputs1 <77% × fOUT <77.4% × fOUT <77.8% × fOUT <80% × fOUT 88.8% × fOUT 91.2% × fOUT 96% × fOUT Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 HALF-BAND FILTERS 0 The first decimate by 2, half-band, low-pass FIR filter (HB4) uses an 15-bit, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption. The HB4 filter is used only when complex outputs (decimate by 16) or real outputs (decimate by 8) are enabled; otherwise, the filter is bypassed. Table 18 and Figure 62 show the coefficients and response of the HB4 filter. –40 –60 –80 –100 –120 0 0.1 Table 18. HB4 Filter Coefficients HB4 Coefficient Number C1, C11 C2, C10 C3, C9 C4, C8 C5, C7 C6 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Figure 63. HB3 Filter Response Decimal Coefficient (15-Bit) 99 0 −808 0 4805 8192 HB2 Filter The third decimate by 2, half-band, low-pass FIR filter (HB2) uses a 19-bit, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption. The HB2 filter is only used when complex outputs (decimate by 4, 8, or 16) or real outputs (decimate by 2, 4, or 8) are enabled; otherwise, the filter is bypassed. Table 20 and Figure 64 show the coefficients and response of the HB2 filter. 0 –20 MAGNITUDE (dB) 0.2 NORMALIZED FREQUENCY (× π RAD/SAMPLE) 13015-066 HB4 Filter –20 MAGNITUDE (dB) The AD9684 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. Table 20. HB2 Filter Coefficients –40 –60 –80 –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) 13015-065 –120 Figure 62. HB4 Filter Response HB2 Coefficient Number C1, C19 C2, C18 C3, C17 C4, C16 C5, C15 C6, C14 C7, C13 C8, C12 C9, C11 C10 Decimal Coefficient (19-Bit) 161 0 −1328 0 5814 0 −19272 0 80,160 131,072 HB3 Filter HB3 Coefficient Number C1, C11 C2, C10 C3, C9 C4, C8 C5, C7 C6 –20 Decimal Coefficient (18-Bit) 859 0 −6661 0 38570 65536 Rev. 0 | Page 41 of 64 –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 64. HB2 Filter Response 13015-067 Table 19. HB3 Filter Coefficients 0 MAGNITUDE (dB) The second decimate by 2, half-band, low-pass, FIR filter (HB3) uses an 18-bit, 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, the filter is bypassed. Table 19 and Figure 63 show the coefficients and response of the HB3 filter. Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet HB1 Filter 0 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 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 –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) 13015-068 Table 21. HB1 Filter Coefficients –20 MAGNITUDE (dB) The fourth and final decimate by 2, half-band, low-pass FIR filter (HB1) uses a 21-bit, symmetrical, fixed coefficient filter implementation that is optimized for low power consumption. The HB1 filter is always enabled and cannot be bypassed. Table 21 and Figure 65 show the coefficients and response of the HB1 filter. Figure 65. HB1 Filter Response DDC GAIN STAGE 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 to enable the 6 dB gain to recenter the dynamic range of the signal within the full scale of the output bits. 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 (see Figure 66). DDC COMPLEX TO REAL CONVERSION BLOCK 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 66 shows a simplified block diagram of the complex to real conversion. Rev. 0 | Page 42 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet AD9684 GAIN STAGE HB1 FIR 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 13015-069 Q 0dB OR 6dB HB1 FIR Figure 66. Complex to Real Conversion Block DDC EXAMPLE CONFIGURATIONS Table 22 describes the register settings for multiple DDC example configurations. Table 22. DDC Example Configurations Chip Application Layer One DDC Chip Decimation Ratio 2 DDC Input Type Complex DDC Output Type Complex Bandwidth Per DDC 1 77% × fS Rev. 0 | Page 43 of 64 Number of Virtual Converters Required (M) 2 Register Settings 2 Register 0x200 = 0x01 (one DDC, I/Q selected) Register 0x201 = 0x01 (chip decimate by 2) Register 0x310 = 0x83 (complex mixer, 0 dB gain, variable IF, complex outputs, HB1 filter) Register 0x311 = 0x04 (DDC I input = ADC Channel A, DDC Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Online Documentation Product Overview Design Resources Discussion AD9684 Sample & Buy Data Sheet Chip Application Layer Two DDCs Chip Decimation Ratio 4 DDC Input Type Complex DDC Output Type Complex Bandwidth Per DDC 1 38.5% × fS Number of Virtual Converters Required (M) 4 Two DDCs 4 Complex Real 19.25% × fS 2 Two DDCs 4 Real Real 19.25% × fS 2 Rev. 0 | Page 44 of 64 Register Settings 2 Register 0x200 = 0x02 (two DDCs, I/Q selected) Register 0x201 = 0x02 (chip decimate by 4) Register 0x310, Register 0x330 = 0x80 (complex mixer, 0 dB gain, variable IF, complex outputs, HB2 + HB1 filters) Register 0x311, Register 0x331 = 0x04 (DDC I input = ADC Channel A, DDC Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Register 0x200 = 0x22 (two DDCs, Q ignore selected) Register 0x201 = 0x02 (chip decimate by 4) Register 0x310, Register 0x330 = 0x89 (complex mixer, 0 dB gain, variable IF, real output, HB3 + HB2 + HB1 filters) Register 0x311, Register 0x331 = 0x04 (DDC I input = ADC Channel A, DDC Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Register 0x200 = 0x22 (two DDCs, Q ignore selected) Register 0x201 = 0x02 (chip decimate by 4) Register 0x310, Register 0x330 = 0x49 (real mixer, 6 dB gain, variable IF, real output, HB3 + HB2 + HB1 filters) Register 0x311 = 0x00 (DDC 0 I input = ADC Channel A, DDC 0 Q input = ADC Channel A) Register 0x331 = 0x05 (DDC 1 I input = ADC Channel B, DDC 1 Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Online Documentation Product Overview Design Resources Discussion Data Sheet Sample & Buy AD9684 Chip Application Layer Two DDCs Chip Decimation Ratio 4 DDC Input Type Real DDC Output Type Complex Bandwidth Per DDC 1 38.5% × fS Number of Virtual Converters Required (M) 4 Four DDCs 8 Real Complex 19.25% × fS 8 Rev. 0 | Page 45 of 64 Register Settings 2 Register 0x200 = 0x02 (two DDCs, I/Q selected) Register 0x201 = 0x02 (chip decimate by 4) Register 0x310, Register 0x330 = 0x40 (real mixer, 6 dB gain, variable IF, complex output, HB2 + HB1 filters) Register 0x311 = 0x00 (DDC 0 I input = ADC Channel A, DDC 0 Q input = ADC Channel A) Register 0x331 = 0x05 (DDC 1 I input = ADC Channel B, DDC 1 Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Register 0x200 = 0x03 (four DDCs, I/Q selected) Register 0x201 = 0x03 (chip decimate by 8) Register 0x310, Register 0x330, Register 0x350, Register 0x370 = 0x41 (real mixer, 6 dB gain, variable IF, complex output, HB3 + HB2 + HB1 filters) Register 0x311 = 0x00 (DDC 0 I input = ADC Channel A, DDC 0 Q input = ADC Channel A) Register 0x331 = 0x00 (DDC 1 I input = ADC Channel A, DDC 1 Q input = ADC Channel A) Register 0x351 = 0x05 (DDC 2 I input = ADC Channel B, DDC 2 Q input = ADC Channel B) Register 0x371 = 0x05 (DDC 3 I input = ADC Channel B, DDC 3 Q input = ADC Channel B) Register 0x314, Register 0x315, Register 0x320, Register 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Register 0x354, Register 0x355, Register 0x360, Register 0x361 = FTW and POW set as required by application for DDC 2 Register 0x374, Register 0x375, Register 0x380, Register 0x381 = FTW and POW set as required by application for DDC 3 Product Overview Online Documentation Design Resources Discussion AD9684 Chip Application Layer Four DDCs 1 2 Sample & Buy Data Sheet Chip Decimation Ratio 16 DDC Input Type Real DDC Output Type Complex Bandwidth Per DDC 1 9.625% × fS Number of Virtual Converters Required (M) 8 Register Settings 2 Register 0x200 = 0x03 (four DDCs, I/Q selected) Register 0x201 = 0x04 (chip decimate by 16) Register 0x310, Register 0x330, Register 0x350, Register 0x370 = 0x42 (real mixer, 6 dB gain, variable IF, complex output, HB4 + HB3 + HB2 + HB1 filters) Register 0x311 = 0x00 (DDC 0 I input = ADC Channel A, DDC 0 Q input = ADC Channel A) Register 0x331 = 0x00 (DDC 1 I input = ADC Channel A, DDC 1 Q input = ADC Channel A) Register 0x351 = 0x05 (DDC 2 I input = ADC Channel B, DDC 2 Q input = ADC Channel B) Register 0x371 = 0x05 (DDC 3 I input = ADC Channel B, DDC 3 Q input = ADC Channel B) Register 0x314, 0x315, 0x320, 0x321 = FTW and POW set as required by application for DDC 0 Register 0x334, Register 0x335, Register 0x340, Register 0x341 = FTW and POW set as required by application for DDC 1 Register 0x354, Register 0x355, Register 0x360, Register 0x361 = FTW and POW set as required by application for DDC 2 Register 0x374, Register 0x375, Register 0x380, Register 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 SYNC± pins after all writes to the FTW or POW registers are complete. This ensures the proper operation of the NCO. See the NCO Synchronization section for more information. Rev. 0 | Page 46 of 64 Online Documentation Product Overview Design Resources Discussion Sample & Buy Data Sheet AD9684 DIGITAL OUTPUTS DIGITAL OUTPUTS The AD9684 output drivers are for standard ANSI LVDS, but optionally the drive current can be reduced using Register 0x56A. The reduced drive current for the LVDS outputs potentially reduces the digitally induced noise. As detailed in the AN-877 Application Note, Interfacing to High Speed ADCs via SPI, the data format can be selected for offset binary, twos complement, or gray code when using the SPI control. The AD9684 has a flexible three-state ability for the digital output pins. The three-state mode is enabled when the device is set for power-down mode. As shown in Table 24, the function of the output pins changes based upon the selection of either parallel or byte output mode in Register 0x568. Timing Minimize the length of the output data lines and the corresponding loads to reduce transients within the AD9684. These transients can degrade converter dynamic performance. The lowest typical conversion rate of the AD9684 is 250 MSPS. At clock rates below 250 MSPS, dynamic performance may degrade. Data Clock Output The AD9684 also provides a data clock output (DCO) intended for capturing the data in an external register. The DCO relative to the data output can be adjusted using Register 0x569. ADC OVERRANGE The ADC overrange (OR) indicator is asserted when an overrange is detected on the input of the ADC. The overrange condition is determined at the output of the ADC pipeline and, therefore, is subject to a latency of 28 ADC clocks. An overrange at the input is indicated by the OR bit 28 clock cycles after it occurs. The AD9684 provides latched data with a pipeline delay of 33 input sample clock cycles. Data outputs are available one propagation delay (tPD) after the rising edge of the clock signal. Table 23. LVDS Output Configurations Parallel Output Mode Parallel Interleaved, Two Converters (0x1) Parallel Channel Multiplexed, Two Converters (0x3) Byte Mode, Two Converters (0x5) Byte Mode, Four Converters (0x6) Byte Mode, Eight Converters (0x7) Number of Virtual Converters Supported 2 Virtual Converter Resolution (Max) 14-bit 2 14-bit 2 4 8 16-bit 16-bit 16-bit LVDS Byte Mode Outputs Required DCO + STATUS + D[13:0] DCO + STATUS + D[13:7] = Channel AD[6:0] = Channel B 1 DCO + 1 STATUS + 8 DATA[7:0] 1 DCO + 1 STATUS + 8 DATA[7:0] 1 DCO + 1 STATUS + 8 DATA[7:0] Table 24. Pin Mapping Between LVDS Parallel/Byte Modes Pin Name DCO−, DCO+ STATUS−, STATUS+ D13−, D13+ D12−, D12+ D11−, D11+ D10−, D10+ D9−, D9+ D8−, D8+ D7−, D7+ D6−, D6+ D5−, D5+ D4−, D4+ D3−, D3+ D2−, D2+ D1−, D1+ D0−, D0+ LVDS Parallel Mode Output DCO−, DCO+ OVR−, OVR+ D13−, D13+ D12−, D12+ D11−, D11+ D10−, D10+ D9−, D9+ D8−, D8+ D7−, D7+ D6−, D6+ D5−, D5+ D4−, D4+ D3−, D3+ D2−, D2+ D1−, D1+ D0−, D0+ Rev. 0 | Page 47 of 64 LVDS Byte Mode Output DCO−, DCO+ FCO−, FCO+ STATUS−, STATUS+ DATA7−, DATA7+ DATA6−, DATA6+ DATA5−, DATA5+ DATA4−, DATA4+ DATA3−, DATA3+ DATA2−, DATA2+ DATA1−, DATA1+ DATA0−, DATA0+ Not applicable Not applicable Not applicable Not applicable Not applicable Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet MULTICHIP SYNCHRONIZATION The AD9684 supports several features that aid users in meeting the requirements for capturing SYNC± signals. The SYNC± sample event can be defined as either a synchronous low to high transition or a synchronous high to low transition. Additionally, the AD9684 allows the SYNC± signal to be sampled using either the rising edge or the falling edge of the CLK± input. The AD9684 can also to ignore a programmable number (up to 16) of SYNC± events. The SYNC± control options can be selected using Register 0x120 and Register 0x121. The AD9684 has a SYNC± input that allows the user flexible options for synchronizing the internal blocks. The SYNC± input is a source synchronous system reference signal that enables multichip synchronization. The input clock divider, DDCs, and signal monitor block LVDS output link can be synchronized using the SYNC± input. For the highest level of timing accuracy, SYNC± must meet the setup and hold requirements relative to the CLK± input. The flowchart in Figure 67 shows the internal mechanism by which multichip synchronization can be achieved in the AD9684. START INCREMENT SYNC± IGNORE COUNTER NO NO RESET SYNC± IGNORE COUNTER NO SYNC± ENABLED? REG (0x120) NO SYNC± ASSERTED? YES YES UPDATE SETUP/HOLD DETECTOR STATUS REG (0x128) SYNC± IGNORE COUNTER EXPIRED? REG (0x121) YES ALIGN CLOCK DIVIDER PHASE TO SYNC± INPUT CLOCK DIVIDER ALIGNMENT REQUIRED? YES CLOCK DIVIDER AUTO ADJUST ENABLED? REG (0x10D) YES NO CLOCK ALIGNMENT REQUIRED? YES NO YES ALIGN PHASE OF ALL INTERNAL CLOCKS TO SYNC± YES ALIGN SIGNAL MONITOR COUNTERS INCREMENT SYNC± COUNTER REG (0x12A) CLOCK DIVIDER > 1? REG (0x10B) NO NO NO DDC NCO ALIGNMENT ENABLED? REG (0x300) YES NO Figure 67. Multichip Synchronization Rev. 0 | Page 48 of 64 ALIGN DDC NCO PHASE ACCUMULATOR BACK TO START 13015-082 SIGNAL MONITOR SYNC ENABLED? REG (0x26F) Online Documentation Product Overview Design Resources Discussion Sample & Buy Data Sheet AD9684 SYNC± SETUP AND HOLD WINDOW MONITOR To assist in ensuring a valid SYNC± capture, the AD9684 has a SYNC± setup and hold window monitor. This feature allows the system designer to determine the location of the SYNC± signals relative to the CLK± signals by reading back the amount of setup and hold margin on the interface through the memory map. Figure 68 and Figure 69 show the setup and hold status values for different phases of SYNC±. The setup detector returns the status of the SYNC± signal before the CLK± edge and the hold detector returns the status of the SYNC± signal after the CLK± edge. Register 0x128 stores the status of SYNC± and alerts the user if the SYNC± signal is captured by the ADC. –1 –2 –3 –4 –5 –6 –7 REG 0x128[3:0] –8 7 6 5 4 3 2 1 0 CLK± INPUT VALID SYNC± INPUT FLIP FLOP SETUP (MIN) FLIP FLOP HOLD (MIN) 13015-083 FLIP FLOP HOLD (MIN) Figure 68. SYNC Setup Detector REG 0x128[7:4] –1 –2 –3 –4 –5 –6 –7 –8 7 6 5 4 3 2 1 0 CLK± INPUT SYNC± INPUT FLIP FLOP SETUP (MIN) FLIP FLOP HOLD (MIN) FLIP FLOP HOLD (MIN) Figure 69. SYNC± Hold Detector Rev. 0 | Page 49 of 64 13015-084 VALID Product Overview Online Documentation Design Resources AD9684 Discussion Sample & Buy Data Sheet Table 25 describes the contents of Register 0x128 and how to interpret those contents. Table 25. SYNC± Setup and Hold Monitor Register 0x128 Register 0x128, Bits[7:4], Hold Status 0x0 0x0 to 0x8 0x8 0x8 0x9 to 0xF 0x0 Register 0x128, Bits[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. 0 | Page 50 of 64 Product Overview Online Documentation Design Resources Discussion Sample & Buy Data Sheet AD9684 TEST MODES ADC TEST MODES The AD9684 has various test options that aid in the system level implementation. The AD9684 has ADC test modes that are available in Register 0x550. These test modes are described in Table 26. 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 pseudorandom number (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); however, they do require an encode clock. For more information, see the AN-877 Application Note, Interfacing to High Speed ADCs via SPI. Table 26. ADC Test Modes Output Test Mode Bit Sequence 0000 0001 0010 0011 0100 0101 Expression Not applicable 00 0000 0000 0000 01 1111 1111 1111 10 0000 0000 0000 10 1010 1010 1010 x23 + x18 + 1 Default/Seed Value Not applicable Not applicable Not applicable Not applicable Not applicable 0x3AFF Sample (N, N + 1, N + 2, …) Not applicable Not applicable Not applicable Not applicable 0x1555, 0x2AAA, 0x1555, 0x2AAA, 0x1555 0x3FD7, 0x0002, 0x26E0, 0x0A3D, 0x1CA6 x9 + x 5 + 1 0x0092 0x125B, 0x3C9A, 0x2660, 0x0c65, 0x0697 11 1111 1111 1111 Not applicable 0x0000, 0x3FFF, 0x0000, 0x3FFF, 0x0000 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 Register 0x551 to Register 0x558 Not applicable 1111 Ramp output (x) % 214 Not applicable 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 0110 0111 Rev. 0 | Page 51 of 64 Product Overview Design Resources Online Documentation Discussion AD9684 Sample & Buy Data Sheet SERIAL PORT INTERFACE (SPI) The AD9684 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 (Rev. 1.0). CONFIGURATION USING THE SPI Three pins define the SPI of this ADC: the SCLK pin, the SDIO pin, and the CSB pin (see Table 27). The SCLK (serial clock) pin synchronizes 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 to 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 27. Serial Port Interface Pins Pin SCLK SDIO CSB Function Serial clock. The serial shift clock input, which synchronizes the 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 3 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 pin 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 secondary functions of the SPI pins. 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 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 configuration 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 (Rev. 1.0). HARDWARE INTERFACE The pins described in Table 27 compose the physical interface between the user programming device and the serial port of the AD9684. 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 is flexible enough to be controlled by either field programmable gate arrays (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 AD9684 to prevent these signals from transitioning at the converter inputs during critical sampling periods. SPI ACCESSIBLE FEATURES Table 28 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 (Rev. 1.0). The AD9684 device specific features are described in the Memory Map section. Table 28. Features Accessible Using the SPI Feature Name Mode Clock DDC Test Input/Output Output Mode 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 up the decimation filters for different applications Allows the user to set the test modes to have known data on the output bits Allows the user to set up outputs Rev. 0 | Page 52 of 64 Product Overview Online Documentation Design Resources Discussion Sample & Buy Data Sheet AD9684 MEMORY MAP READING THE MEMORY MAP REGISTER TABLE Logic Levels Each row in the memory map register table has eight bit locations. The memory map is divided into four sections: the Analog Devices, Inc., SPI registers (Register 0x000 to Register 0x00D), the ADC function registers (Register 0x015 to Register 0x278), The DDC function registers (Register 0x300 to Register 0x387), and the digital outputs and test modes registers (Register 0x550 to Register 0x05B). An explanation of logic level terminology follows: Table 29 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 29. Unassigned and Reserved Locations All address and bit locations that are not included in Table 29 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 unassigned (for example, Address 0x561). If the entire address location is open (for example, Address 0x013), do not write to this address location. Default Values After the AD9684 is reset, critical registers are loaded with default values. The default values for the registers are given in Table 29. • • • “Bit is set” is synonymous with “bit is set to Logic 1” or “writing Logic 1 for the bit.” “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” bit. Channel-Specific Registers Some channel setup functions, such as the 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 29 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 29 affect the entire device and the channel features for which independent settings are not allowed between channels. The settings in Register 0x005 do not affect the global registers and bits. SPI Soft Reset After issuing a soft reset by programming 0x81 to Register 0x000, the AD9684 requires 5 ms to recover. Therefore, when programming the AD9684 for application setup, ensure that an adequate delay is programmed into the firmware after asserting the soft reset and before starting the device setup. Rev. 0 | Page 53 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet MEMORY MAP REGISTER TABLE All address locations that are not included in Table 29 are not currently supported for this device and must not be written. Table 29. Memory Map Registers Reg. Addr. Register Bit 7 (Hex) Name (MSB) Analog Devices SPI Registers 0x000 INTERFACE_ Soft reset CONFIG_A (self clearing) 0x001 INTERFACE_ Single CONFIG_B instruction DEVICE_ 0 CONFIG (local) 0x003 CHIP_TYPE 0x004 CHIP_ID (low byte) 0x005 CHIP_ID (high byte) 0x006 CHIP_ GRADE 0x008 Device 0 index 0x00A Scratch 0 pad 0x00B SPI revision 0 0x00C Vendor ID 0 (low byte) 0x00D Vendor ID 0 (high byte) ADC Function Registers 0 0x015 Analog input (local) 0x002 0x016 Input termination (local) 0x018 Input buffer current control (local) 0x024 V_1P0 control 0 0x025 Input fullscale range (local) 0 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 LSB first 0 = MSB 1 = LSB 0 Address ascension 0 0 Address ascension 0 0 0 0 0 0 0 1 1 0 1 0 Bit 1 Bit 0 (LSB) LSB first Soft reset 0 = MSB (self 1 = LSB clearing) Datapath 0 0 soft reset (self clearing) 00 = normal operation 0 10 = standby 11 = power-down 011 = high speed ADC 0 1 1 0 0 Chip speed grade 0101 = 500 MSPS 0 X Default Notes 0x00 0x00 0x00 0x03 0xD3 Read only Read only 0x00 Read only 0x5X Read only 0 1 0 0 0 0 Channel B Channel A 0x03 0 0 0 0 0 0 0 0x00 0 1 0 0 0 1 0 0 0 1 0 1 1 0 0x01 0x56 Read only 0 0 0 0 1 0 0 0x04 Read only 0 0 0 0 0 0 0x00 1 1 0 Input disable 0 = normal operation 1 = input disabled 0 0x0C 0 0 0 0 0x20 0 0 0 1.0 V refer- 0x00 ence select 0 = internal 1= external 0x0C Analog input differential termination 0000 = 400 Ω (default) 0001 = 200 Ω 0010 = 100 Ω 0110 = 50 Ω 0000 = 1.0× buffer current (default) 0001 = 1.5× buffer current 0010 = 2.0× buffer current 0011 = 2.5× buffer current 0100 = 3.0× buffer current 0101 = 3.5× buffer current … 1111 = 8.5× buffer current 0 0 0 0 0 Full-scale adjust 0000 = 1.94 V 1000 = 1.46 V 1001 = 1.58 V 1010 = 1.70 V 1011 = 1.82 V 1100 = 2.06 V (default) 0 Rev. 0 | Page 54 of 64 V p-p differential; use in conjunction with Reg. 0x030 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet Reg. Addr. (Hex) 0x028 AD9684 Register Name Temperature diode (local) Bit 7 (MSB) 0 Bit 6 0 Bit 5 0 Bit 4 0 0x030 Input fullscale control (local) 0 0 0 0x03F PDWN/ STBY pin control (local) 0 0x040 Chip pin control 0= 0 PDWN/ STBY enabled 1= disabled PDWN/STBY function 00 = power down 01 = standby 10 = disabled Full-scale control See Table 9 for recommended settings for different frequency bands Default values: Full scale range ≥ 1.82 V = 001 Full scale range < 1.82 V = 110 0 0 0x10B Clock divider 0 0 0 0 0x10C Clock divider phase (local) 0 0 0 0 0x10D Clock divider and SYNC± control 0 0 0x117 Clock delay control Clock 0 divider automatic phase adjustment 0= disabled 1= enabled 0 0 0 0 Bit 3 0 Fast Detect B (FD_B) 000 = Fast Detect B output 001 = reserved 010 = reserved 111 = disabled Bit 2 0 Bit 1 0 Default 0x00 Notes Used in conjunction with Reg. 0x040 0 Bit 0 (LSB) Diode selection 0 = no diode selected 1= temperature diode selected 0 0 0x04 Input fullscale control (local) 0 0 0 0x00 Used in conjunction with Reg. 0x040 Fast Detect A (FD_A) 000 = Fast Detect A output 001 = reserved 010 = reserved 011 = temperature diode 111 = disabled 000 = divide by 1 0 001 = 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 Clock fine 0 0 0 delay adjust enable 0= disabled 1= enabled Rev. 0 | Page 55 of 64 0x3F 0x00 0x00 0x00 Clock divider must be >1 0x00 Enabling the clock fine delay adjustment causes a datapath soft reset Online Documentation Product Overview Design Resources Discussion Sample & Buy AD9684 Reg. Addr. (Hex) 0x118 Data Sheet Register Name Clock fine delay (local) Bit 7 (MSB) 0x11C Clock status 0 0x120 SYNC± Control 1 0 0x121 SYNC± Control 2 0 0x123 SYNC± timestamp delay control 0x128 SYNC± Status 1 SYNC± and clock divider status 0x129 0x12A 0x1FF Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 (LSB) Clock fine delay adjust, Bits[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 ps skew … 0 = 0 ps skew … ≥+87 = +150 ps skew 0 = no 0 0 0 0 0 0 input clock detected 1 = input clock detected SYNC± mode select CLK± edge SYNC± SYNC± flag 0 0 00 = disabled select transition reset 01 = continuous 0 = rising select 0 = normal 10 = N shot 1 = falling 0 = low to operation high 1 = flags 1 = high to held in reset low SYNC± N shot ignore counter select 0 0 0 0000 = next SYNC± only 0001 = ignore the first SYNC± transitions 0010 = ignore the first two SYNC± transitions … 1111 = ignore the first 16 SYNC± transitions SYNC± timestamp delay, Bits[6:0] 0x00 = no delay 0x01 = 1 clock delay … 0x7F = 127 clocks delay SYNC± hold status, Register 0x128, Bits [7:4] SYNC± setup status, Register 0x128, Bits [3:0] 0 0 SYNC± counter Chip sync mode 0x200 Chip application mode 0 0 0x201 Chip decimation ratio 0 0 0x228 Customer offset Clock divider phase when SYNC± was captured 0000 = in-phase 0001 = SYNC± is ½ cycle delayed from clock 0010 = SYNC± 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 SYNC± counter, Bits[7:0] increments when a SYNC± signal is captured 0 Chip Q ignore 0 = normal (I/Q) 1 = ignore (I only) 0 0 0 0 0 0 Rev. 0 | Page 56 of 64 Notes Used in conjunction with Reg. 0x0117 0x00 Read only 0x00 0x00 0x00 Read only Read only Read only Synchronization mode 0x00 00 = normal 01 = timestamp Chip operating mode 0x00 00 = full bandwidth mode 01 = DDC 0 on 10 = DDC 0 and DDC 1 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 Default 0x00 0x00 0x00 Mode select (Reg. 0x120, Bits[2:1]) must be N shot Ignored when Reg. 0x01FF = 0x00 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet Reg. Addr. (Hex) 0x245 0x247 0x248 0x249 0x24A 0x24B 0x24C 0x26F 0x270 0x271 0x272 0x273 0x274 Register Name Fast detect (FD) control (local) FD upper threshold LSB (local) FD upper threshold MSB (local) FD lower threshold LSB (local) FD lower threshold MSB (local) 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) AD9684 Bit 7 (MSB) 0 0 Bit 6 0 0 Bit 5 0 Bit 4 0 Bit 3 Bit 2 Force FD_A/ Force value of FD_A/ FD_B pins FD_B pins if 0 = normal force pins is function 1 = force to true, this value is value output on FD_x pins Fast detect upper threshold, Bits[7:0] 0 Bit 1 0 Bit 0 (LSB) Enable fast detect output 0 Fast detect upper threshold, Bits[12:8] 0 0x00 0x00 Fast detect dwell time, Bits[15:8] 0x00 0 0 0 0 0 0 0 0 0 0 0 0 0x00 Fast detect dwell time, Bits[7:0] 0 0 0x00 Fast detect lower threshold, Bits[12:8] 0 Notes 0x00 Fast detect lower threshold, Bits[7:0] 0 Default 0x00 Synchronization mode 0x00 00 = disabled 01 = continuous 11 = one-shot Peak detector 0= disabled 1= enabled 0 0x00 Signal monitor period, Bits[7:0] 0x80 In decimated output clock cycles Signal monitor period, Bits[15:8] 0x00 In decimated output clock cycles Signal monitor period, Bits[23:16] 0x00 In decimated output clock cycles Result update 1 = update results (self clear) 0 Rev. 0 | Page 57 of 64 0 0 Result selection 0= reserved 1 = peak detector 0x01 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Reg. Addr. (Hex) 0x275 0x276 0x277 0x278 Register Name Signal Monitor Result Register 0 (local) Signal Monitor Result Register 1 (local) Signal Monitor Result Register 1 (local) Signal monitor period counter result (local) Data Sheet Bit 7 (MSB) Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 (LSB) Default Signal monitor result, Bits[7:0] Read When Register 0x0274, Bit 0 = 1, result Bits[19:7] = peak detector absolute value, Bits [12:0]; result Bits[6:0] = 0 only Signal monitor result, Bits[15:8] 0 0 0 0 Signal monitor result, Bits[19:16] Period count result, Bits[7:0] Digital Downconverter (DDC) Function Registers—See the Digital Downconverters (DDCs) Section DDC NCO 0 0 0 0 0 0x300 DDC synsoft reset chronization 0 = normal control operation 1 = reset IF mode Gain select Complex to 0 Mixer 0x310 DDC 0 00 = variable IF mode 0 = 0 dB real enable control select (mixers and NCO gain 0 = disabled 0 = real enabled) 1 = 6 dB 1 = enabled mixer 01 = 0 Hz IF mode (mixer gain 1= bypassed, NCO disabled) complex 10 = fS/4 Hz IF mode (fS/4 mixer downmixing mode) 11 = test mode (mixer inputs forced to +FS, NCO enabled) 0x311 DDC 0 input selection 0x314 DDC 0 frequency LSB DDC 0 frequency MSB DDC 0 phase LSB DDC 0 phase MSB DDC 0 output test mode selection 0x315 0x320 0x321 0x327 0 0 0 X X X Q input select 0 = Ch. A 1 = Ch. B DDC 0 NCO FTW, Bits[7:0], twos complement 0 0 X DDC 0 NCO POW, Bits[7:0], twos complement X X X X 0 0 0 0 Rev. 0 | Page 58 of 64 Q output test mode enable 0= disabled 1 = enabled from Ch. B Updated based on Reg. 0x274, Bit 4 Read only Updated based on Reg. 0x274, Bit 4 Read only Updated based on Reg. 0x274, Bit 4 0x00 0x00 DDC 0 NCO POW, Bits[11:8], twos complement 0 Read only Synchronization mode (triggered by SYNC±) 00 = disabled 01 = continuous 11 = one-shot Decimation rate select 0x00 (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 I input 0x00 0 select 0 = Ch. A 1 = Ch. B 0x00 DDC 0 NCO FTW, Bits[11:8], twos complement 0 I output test mode enable 0= disabled 1= enabled from Ch. A Notes Updated based on Reg. 0x274, Bit 4 0x00 0x00 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet Reg. Addr. (Hex) 0x330 AD9684 Register Name DDC 1 control Bit 7 (MSB) Mixer select 0 = real mixer 1= complex mixer 0x331 DDC 1 input selection 0 0x334 DDC 1 frequency LSB DDC 1 frequency MSB DDC 1 phase LSB DDC 1 phase MSB DDC 1 output test mode selection 0x335 0x340 0x341 0x347 X Bit 6 Gain select 0 = 0 dB gain 1 = 6 dB gain 0 X Bit 5 Bit 4 Bit 3 IF mode Complex to 00 = variable IF mode real enable (mixers and NCO 0 = disabled enabled) 1 = 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) 0 0 0 Q input select 0 = Ch. A 1 = Ch. B DDC 1 NCO FTW, Bits[7:0], twos complement X X X X X X I input select 0 = Ch. A 1 = Ch. B 0 0 0 0 Gain select 0 = 0 dB gain 1 = 6 dB gain IF mode 00 = variable IF mode (mixers and NCO enabled) 01 = 0Hz 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) 0x351 DDC 2 input selection 0 0 0 0x354 DDC 2 frequency LSB DDC 2 frequency MSB DDC 2 phase LSB DDC 2 phase MSB X X X 0 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, Bits[7:0], twos complement 0 X X X X I output test mode enable 0= disabled 1= enabled from Ch. A Decimation rate 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 I input 0 select 0 = Ch. A 1 = Ch. B 0 DDC 2 NCO FTW, Bits[11:8], twos complement DDC 2 NCO POW, Bits[11:8], twos complement Rev. 0 | Page 59 of 64 0x00 0x00 DDC 2 NCOPOW, Bits[7:0], twos complement X 0x00 0x00 DDC 1 NCO POW, Bits[11:8], twos complement Mixer select 0 = real mixer 1= complex mixer 0x361 0 DDC 1 NCO POW, Bits[7:0], twos complement DDC 2 control 0x360 Bit 1 Bit 0 (LSB) Default Decimation rate select 0x00 (complex to real disabled) 11 = decimate by 2 (complex to real enabled) 11 = decimate by 1 DDC 1 NCO FTW, Bits[11:8], twos complement 0x350 0x355 Bit 2 0 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 Notes Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Reg. Addr. (Hex) 0x367 Data Sheet Register Name DDC 2 output test mode selection Bit 7 (MSB) 0 0x370 DDC 3 control 0x371 DDC 3 input selection 0x374 DDC 3 frequency LSB DDC 3 frequency MSB DDC 3 phase LSB DDC 3 phase MSB DDC 3 output test mode selection 0x375 0x380 0x381 0x387 Bit 6 0 Bit 5 0 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) 0 0 0 X X Bit 4 0 Bit 3 0 Bit 2 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 3 NCO FTW, Bits[7:0] twos complement 0 X 0 X Bit 1 0 Bit 0 (LSB) I output test mode enable 0= disabled 1= enabled from Ch. A Decimation rate 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 I input 0 select 0 = Ch. A 1 = Ch. B DDC 3 NCO POW, Bits[7:0] twos complement X X X 0 0 0 0 0 Reset PN long gen 0 = long PN enable 1 = long PN reset Reset PN short gen 0 = short PN enable 1 = short PN reset Digital Outputs and Test Modes User 0x550 ADC test pattern modes selection (local) 0= continuous repeat 1 = single pattern Q output test mode enable 0 = disabled 1 = enabled from Ch. B 0x551 User Pattern 1 LSB 0 0 0 0 0 0x552 User Pattern 1 MSB 0 0 0 0 0 0 I output test mode enable 0= disabled 1= enabled from Ch. A 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 = one/zero word toggle 1000 = the user pattern test mode (used with Register 0x550, Bit 7 and User Pattern 1 to User Pattern 4 registers) 1111 = ramp output 0 0 0 Rev. 0 | Page 60 of 64 0 0x00 0x05 0x00 0x00 DDC 3 NCO POW, Bits[11:8] twos complement 0 Notes 0x00 DDC 3 NCO FTW, Bits[11:8] twos complement X Default 0x00 0 0 0x00 0x00 0x00 0x00 0x00 Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 Online Documentation Product Overview Design Resources Sample & Buy Discussion Data Sheet Reg. Addr. (Hex) 0x553 AD9684 Register Name User Pattern 2 LSB Bit 7 (MSB) 0 Bit 6 0 Bit 5 0 Bit 4 0 Bit 3 0 Bit 2 0 Bit 1 0 Bit 0 (LSB) 0 Default 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 0x559 Output Mode Control 1 0 0 0x561 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 5 OR 0 = OR bit enabled 1 = OR bit cleared Virtual Converter 4 OR 0 = OR bit enabled 1 = OR bit cleared Virtual Converter 3 OR 0 = OR bit enabled 1 = OR bit cleared 0x563 Output OR status Virtual Converter 6 OR 0 = no OR 1 = OR occurred 0x564 Output channel select Virtual Converter 7 OR 0 = no OR 1 = OR occurred 0 Virtual Converter 5 OR 0 = no OR 1 = OR occurred 0 Virtual Converter 4 OR 0 = no OR 1 = OR occurred 0 Virtual Converter 3 OR 0 = no OR 1 = OR occurred 0 Status bit selection 000 = tie low (1’b0) 001 = overrange bit 010 = signal monitor bit 011 = fast detect (FD) bit 100 = not applicable 101 = system reference Data format select Sample 00 = offset binary invert 01 = twos 0 = normal complement 1 = sample invert Virtual Virtual Virtual ConConvertConverter 2 OR er 0 OR verter 1 0 = OR bit 0 = OR bit OR enabled enabled 0 = OR 1 = OR bit 1 = OR bit bit cleared cleared enabled 1 = OR bit cleared Virtual ConVirtual Virtual verter 2 OR Converter Converter 0 0 = no OR 1 OR OR 1 = OR 0 = no OR 0 = no OR occurred 1 = OR 1 = OR occurred occurred Converter 0 0 channel swap 0 = normal channel ordering 1= channel swap enabled 0 0 0 0 Rev. 0 | Page 61 of 64 Notes Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 Used with Reg. 0x550 and Reg. 0x573 0x00 0x01 0x00 0x00 0x00 Read only Online Documentation Product Overview Design Resources Discussion Sample & Buy AD9684 Reg. Addr. (Hex) 0x568 Data Sheet Register Name LVDS output mode Bit 7 (MSB) 0 Bit 6 0 Bit 5 Bit 4 Frame clock mode (only used when in output data mode is in byte mode) 00 = frame clock always off 01 = frame clock always on 10 = reserved 11 = frame clock conditionally on based on PN23 sequence Bit 3 0 0x569 Digital clock output adjust 0 0 0 0 0 0x56A Output Parallel Driver Adjust 1 0 1 0 0 0x05B Output Parallel Driver Adjust 2 0 0 0 0 0 Bit 1 Bit 0 (LSB) Output data mode 000 = parallel mode (one converter) 001 = parallel interleaved mode (two converters) 010 = parallel channel multiplexed (even/odd) mode (one converter) 011 = parallel channel multiplexed (even/odd) mode (two converters) 100 = byte mode (one converter) 101 = byte mode (two converters) 110 = byte mode (four converters) Others = reserved DCO phase adjustment 0 0x0: 0° 0x1: 90° 0x2: 180° 0x3: 270° LVDS output drive current adjust 0 000 = 2 mA 001 = 2.25 mA 010 = 2.5 mA 011 = 2.75 mA 100 = 3.0 mA 101 = 3.25 mA 110 = 3.5 mA (default) 111 = 3.75 mA Output slew rate 0 control of LVDS driver Interface 00 = 80 ps 01 = 150 ps 10 = 200 ps 11 = 250 ps Rev. 0 | Page 62 of 64 Bit 2 Default 0x00 0x01 0x4C 0x00 Notes Online Documentation Product Overview Design Resources Data Sheet Discussion Sample & Buy AD9684 APPLICATIONS INFORMATION POWER SUPPLY RECOMMENDATIONS The AD9684 must be powered by the following six supplies: AVDD1 = 1.25 V, AVDD2 = 2.5 V, AVDD3 = 3.3 V, DVDD = 1.25 V, DRVDD = 1.25 V, and 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, ADP1740, and ADP125). Figure 70 shows the recommended power supply scheme for the AD9684. LDO ADP125 LDO ADP2164 1.8V BUCK REGULATOR ADP1741 LDO ADP1741 LDO ADP1740 LDO 5V/12V INPUT ADP2370 BUCK REGULATOR 3.8V ADP125 LDO 2.5V: AVDD2 1.8V: SPIVDD 1.25V: AVDD1 1.25V: DVDD 1.25V: DRVDD 3.3V: AVDD3 13015-085 ADP1741 3.3V INPUT It is not necessary to split all of these power domains in all cases. The recommended solution shown in Figure 70 provides the lowest noise, highest efficiency power delivery system for the AD9684. If only one 1.25 V supply is available, route to AVDD1 first and then tap it off and isolate it with a ferrite bead or a filter choke, preceded by decoupling capacitors for SPIVDD, DVDD, and DRVDD, in that order. The user can employ several different decoupling capacitors to cover both high and low frequencies. These capacitors must be located close to the point of entry at the PCB level and close to the devices, with minimal trace lengths. Figure 70. High Efficiency, Low Noise Power Solution for the AD9684 Rev. 0 | Page 63 of 64 Online Documentation Product Overview Design Resources Sample & Buy Discussion AD9684 Data Sheet OUTLINE DIMENSIONS A1 BALL PAD CORNER 14 13 12 11 10 9 8 7 6 5 4 3 2 1 8.20 SQ 11.20 SQ TOP VIEW 1.49 1.38 1.27 A B C D E F G H J K L M N P 10.40 REF SQ 0.80 BOTTOM VIEW 0.80 REF DETAIL A 0.75 REF DETAIL A 1.15 1.05 0.95 0.38 0.33 0.28 0.30 REF PKG-004472 SEATING PLANE 0.50 0.45 0.40 BALL DIAMETER COPLANARITY 0.12 COMPLIANT TO JEDEC STANDARDS MO-275-GGAB-1. 04-24-2015-A A1 BALL PAD CORNER 12.10 12.00 SQ 11.90 Figure 71. 196-Ball Ball Grid Array, Thermally Enhanced [BGA_ED] (BP-196-3) Dimensions shown in millimeters ORDERING GUIDE Model 1 AD9684BBPZ-500 AD9684BBPZRL7-500 AD9684-500EBZ 1 Temperature Range −40°C to +85°C −40°C to +85°C Package Description 196-Ball Ball Grid Array, Thermally Enhanced [BGA_ED] 196-Ball Ball Grid Array, Thermally Enhanced [BGA_ED] Evaluation Board for AD9684-500 Z = RoHS Compliant Part. ©2015 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D13015-0-4/15(0) Rev. 0 | Page 64 of 64 Package Option BP-196-3 BP-196-3