11-Bit/16-Bit, 12 GSPS, RF Digital-to-Analog Converters AD9161/AD9162 Data Sheet FEATURES In baseband mode, wide bandwidth capability combines with high dynamic range to support DOCSIS 3.1 cable infrastructure compliance from the minimum of two carriers to full maximum spectrum of 1.794 GHz. A 2× interpolator filter (FIR85) enables the AD9161/AD9162 to be configured for lower data rates and converter clocking to reduce the overall system power and ease the filtering requirements. In Mix-Mode™ operation, the AD9161/ AD9162 can reconstruct RF carriers in the second and third Nyquist zones up to 7.5 GHz while still maintaining exceptional dynamic range. The output current can be programmed from 8 mA to 38.76 mA. The AD9161/AD9162 data interface consists of up to eight JESD204B serializer/deserializer (SERDES) lanes that are programmable in terms of lane speed and number of lanes to enable application flexibility. DAC update rate up to 12 GSPS (minimum) Direct RF synthesis at 6 GSPS (minimum) DC to 2.5 GHz in baseband 1× bypass mode DC to 6 GHz in 2× nonreturn-to-zero (NRZ) mode 1.5 GHz to 7.5 GHz in Mix-Mode Bypassable interpolation (1× or bypass mode) 2×, 3×, 4×, 6×, 8×, 12×, 16×, 24× Excellent dynamic performance APPLICATIONS Broadband communications systems DOCSIS 3.1 cable modem termination system (CMTS)/ video on demand (VOD)/edge quadrature amplitude modulation (EQAM) Wireless communications infrastructure W-CDMA, LTE, LTE-A, point to point Instrumentation, automatic test equipment (ATE) Radars and jammers A serial peripheral interface (SPI) can configure the AD9161/ AD9162 and monitor the status of all registers. The AD9161/ AD9162 are offered in an 165-ball, 8.0 mm × 8.0 mm, 0.5 mm pitch, CSP_BGA package and in an 169-ball, 11 mm × 11 mm, 0.8 mm pitch, CSP_BGA package, including a leaded ball option for the AD9162. GENERAL DESCRIPTION PRODUCT HIGHLIGHTS The AD9161/AD91621 are high performance, 11-bit/16-bit digital-to-analog converters (DACs) that supports data rates to 6 GSPS. The DAC core is based on a quad-switch architecture coupled with a 2× interpolator filter that enables an effective DAC update rate of up to 12 GSPS in some modes. The high dynamic range and bandwidth makes these DACs ideally suited for the most demanding high speed radio frequency (RF) DAC applications. 1. 2. 3. High dynamic range and signal reconstruction bandwidth supports RF signal synthesis of up to 7.5 GHz. Up to eight lanes JESD204B SERDES interface flexible in terms of number of lanes and lane speed. Bandwidth and dynamic range to meet DOCSIS 3.1 compliance with margin. FUNCTIONAL BLOCK DIAGRAM RESET SDIO SDO CS SCLK IRQ ISET VREF VREF AD9161/AD9162 SPI NRZ RZ MIX SERDIN0± NCO INV SINC SYSREF± HB 2× HB 3× HB 2×, 4×, 8× TO JESD TO DATAPATH TX_ENABLE DAC CORE OUTPUT± CLOCK DISTRIBUTION CLK± 14379-001 HB 2× JESD DATA LATCH SERDIN7± SYNCOUT± Figure 1. Protected by U.S. Patents 6,842,132 and 7,796,971. Rev. 0 1 Document Feedback Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties that may result from its use. Specifications subject to change without notice. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Trademarks and registered trademarks are the property of their respective owners. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781.329.4700 ©2016 Analog Devices, Inc. All rights reserved. Technical Support www.analog.com AD9161/AD9162 Data Sheet TABLE OF CONTENTS Features .............................................................................................. 1 JESD204B Overview .................................................................. 45 Applications ....................................................................................... 1 Physical Layer ............................................................................. 46 General Description ......................................................................... 1 Data Link Layer .......................................................................... 49 Product Highlights ........................................................................... 1 Transport Layer .......................................................................... 57 Functional Block Diagram .............................................................. 1 JESD204B Test Modes ............................................................... 59 Revision History ............................................................................... 2 JESD204B Error Monitoring..................................................... 61 Specifications..................................................................................... 3 Hardware Considerations ......................................................... 63 DC Specifications ......................................................................... 3 Main Digital Datapath ................................................................... 64 DAC Input Clock Overclocking Specifications ........................ 4 Data Format ................................................................................ 64 Power Supply DC Specifications ................................................ 4 Interpolation Filters ................................................................... 64 Serial Port and CMOS Pin Specifications ................................. 7 Digital Modulation ..................................................................... 67 JESD204B Serial Interface Speed Specifications ...................... 8 Inverse Sinc ................................................................................. 69 SYSREF± to DAC Clock Timing Specifications ....................... 8 Downstream Protection ............................................................ 70 Digital Input Data Timing Specifications ................................. 9 Interrupt Request Operation ........................................................ 71 JESD204B Interface Electrical Specifications ........................... 9 Interrupt Service Routine .......................................................... 71 AC Specifications........................................................................ 10 Applications Information .............................................................. 72 Absolute Maximum Ratings .......................................................... 12 Hardware Considerations ......................................................... 72 Thermal Resistance .................................................................... 12 Analog Interface Considerations.................................................. 75 ESD Caution ................................................................................ 12 Analog Modes of Operation ..................................................... 75 Pin Configurations and Function Descriptions ......................... 13 Clock Input.................................................................................. 76 Typical Performance Characteristics ........................................... 17 Shuffle Mode ............................................................................... 77 AD9161 ........................................................................................ 17 DLL............................................................................................... 77 AD9162 ........................................................................................ 27 Voltage Reference ....................................................................... 77 Terminology .................................................................................... 41 Analog Outputs .......................................................................... 78 Theory of Operation ...................................................................... 42 Start-Up Sequence .......................................................................... 80 Serial Port Operation ..................................................................... 43 Register Summary .......................................................................... 82 Data Format ................................................................................ 43 Register Details ............................................................................... 88 Serial Port Pin Descriptions ...................................................... 43 Outline Dimensions ..................................................................... 138 Serial Port Options ..................................................................... 43 Ordering Guide ........................................................................ 139 JESD204B Serial Data Interface .................................................... 45 REVISION HISTORY 5/2016—Revision 0: Initial Version Rev. 0 | Page 2 of 139 Data Sheet AD9161/AD9162 SPECIFICATIONS DC SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, DAC output full-scale current (IOUTFS) = 40 mA, and TA = −40°C to +85°C, unless otherwise noted. Table 1. Parameter RESOLUTION AD9161 DAC Update Rate Minimum Maximum Maximum Adjusted 4 AD9162 DAC Update Rate Minimum Maximum Maximum Adjusted4 ACCURACY Integral Nonlinearity (INL) Differential Nonlinearity (DNL) ANALOG OUTPUTS Gain Error (with Internal Reference) Full-Scale Output Current Minimum Maximum DAC CLOCK INPUT (CLK+, CLK−) Differential Input Power Common-Mode Voltage Input Impedance 5 TEMPERATURE DRIFT Gain Reference Voltage REFERENCE Internal Reference Voltage ANALOG SUPPLY VOLTAGES VDD25_DAC VDD12A 6 VDD12_CLK6 VNEG_N1P2 DIGITAL SUPPLY VOLTAGES DLL_VDD_1P2 DVDD IOVDD 7 SERDES SUPPLY VOLTAGES VDD_1P2 VTT_1P2 Test Conditions/Comments Min 11 Typ Max Unit Bit VDDx 1 = 1.3 V ± 2% 2 VDDx1 = 1.3 V ± 2%2, FIR85 3 2× interpolator enabled VDDx1 = 1.3 V ± 2%2, minimum 2× interpolation 6 12 3 16 6.4 12.8 3.2 1.5 GSPS GSPS GSPS GSPS Bit VDDx1 = 1.3 V ± 2%2 VDDx1 = 1.3 V ± 2%2, FIR853 2× interpolator enabled VDDx1 = 1.3 V ± 2%2 6 12 6 1.5 6.4 12.8 6.4 GSPS GSPS GSPS GSPS ±2.7 ±1.7 LSB LSB −1.7 % RSET = 9.76 kΩ RSET = 9.76 kΩ 7.37 35.8 8 38.76 8.57 41.3 mA mA RLOAD = 90 Ω differential on-chip AC-coupled 3 GSPS input clock −20 0 0.6 90 +10 dBm V Ω Can connect to DVDD Includes VDD12_DCD/DLL Can connect to VDD_1P2 Rev. 0 | Page 3 of 139 105 75 ppm/°C ppm/°C 1.19 V 2.375 1.14 1.14 −1.26 2.5 1.2 1.2 −1.2 2.625 1.326 1.326 −1.14 V V V V 1.14 1.14 1.71 1.2 1.2 2.5 1.326 1.326 3.465 V V V 1.14 1.14 1.2 1.2 1.326 1.326 V V AD9161/AD9162 Parameter DVDD_1P2 PLL_LDO_VDD12 PLL_CLK_VDD12 SYNC_VDD_3P3 BIAS_VDD_1P2 Data Sheet Test Conditions/Comments Can connect to PLL_LDO_VDD12 Can connect to VDD_1P2 Min 1.14 1.14 1.14 3.135 1.14 Typ 1.2 1.2 1.2 3.3 1.2 Max 1.326 1.326 1.326 3.465 1.326 Unit V V V V V VDDx is VDD12_CLK, DVDD, VDD_1P2, DVDD_1P2, and PLL_LDO_VDD12. Any clock speed over 5.1 GSPS requires a maximum junction temperature of 105°C to avoid damage to the device. See Table 11 for details on maximum junction temperature permitted for certain clock speeds. See Table 2 for the complete details on the guaranteed speed performance. 3 FIR85 is the finite impulse response filter with 85 dB digital attenuation that implements 2× NRZ mode. 4 The adjusted DAC update rate is calculated as fDAC divided by the minimum required interpolation factor. For the AD9162, the minimum interpolation factor is 1. Therefore, with fDAC = 6 GSPS, fDAC adjusted = 6 GSPS. For the AD9161, the minimum interpolation is 2×. Therefore, with fDAC = 6 GSPS, fDAC adjusted = 3 GSPS. When FIR85 is enabled, which puts the device into 2× NRZ mode, fDAC = 2 × (DAC clock input frequency), and the minimum interpolation increases to 2× (interpolation value). Thus, for the AD9162, with FIR85 enabled and DAC clock = 6 GSPS, fDAC = 12 GSPS, minimum interpolation = 2×, and the adjusted DAC update rate = 6 GSPS. 5 See the Output Stage Configuration section for more details. 6 For the lowest noise performance, use a separate power supply filter network for the VDD12_CLK and the VDD12A pins. 7 IOVDD can range from 1.8 V to 3.3 V, with ±5% tolerance. 1 2 DAC INPUT CLOCK OVERCLOCKING SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. Maximum guaranteed speed using the temperatures and voltages conditions as shown in Table 2, where VDDx is VDD12_CLK, DVDD, VDD_1P2, DVDD_1P2, and PLL_LDO_VDD12. Any DAC clock speed over 5.1 GSPS requires a maximum junction temperature of 105°C to avoid damage to the device. See Table 11 for details on maximum junction temperature permitted for certain clock speeds. Table 2. Parameter 1 MAXIMUM DAC UPDATE RATE VDDx = 1.2 V ± 5% VDDx = 1.2 V ± 2% VDDx = 1.3 V ± 2% 1 Test Conditions/Comments Min TJMAX = 25°C TJMAX = 85°C TJMAX = 105°C TJMAX = 25°C TJMAX = 85°C TJMAX = 105°C TJMAX = 25°C TJMAX = 85°C TJMAX = 105°C 6.0 5.6 5.4 6.1 5.8 5.6 6.4 6.2 6.0 Typ Max Unit GSPS GSPS GSPS GSPS GSPS GSPS GSPS GSPS GSPS TJMAX is the maximum junction temperature. POWER SUPPLY DC SPECIFICATIONS IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. FIR85 is the finite impulse response with 85 dB digital attenuation. Table 3. Parameter 8 LANES, 2× INTERPOLATION (80%), 3 GSPS Analog Supply Currents VDD25_DAC = 2.5 V VDD12A = 1.2 V VDD12_CLK = 1.2 V VNEG_N1P2 = −1.2 V Digital Supply Currents DVDD = 1.2 V IOVDD 1 = 2.5 V Test Conditions/Comments Numerically controlled oscillator (NCO) on, FIR85 on Includes VDD12_DCD/DLL Rev. 0 | Page 4 of 139 Min Typ Max Unit 100 150 279 −119 93.8 3.7 229 −112 mA µA mA mA 621.3 2.5 971 2.7 mA mA Data Sheet Parameter SERDES Supply Currents VDD_1P2 = 1.2 V DVDD_1P2 = 1.2 V PLL_LDO_VDD12 = 1.2 V SYNC_VDD_3P3 = 3.3 V 8 LANES, 6× INTERPOLATION (80%), 3 GSPS Analog Supply Currents VDD25_DAC = 2.5 V VDD12A = 1.2 V VDD12_CLK = 1.2 V VNEG_N1P2 = −1.2 V Digital Supply Currents DVDD = 1.2 V IOVDD1 = 2.5 V SERDES Supply Currents VDD_1P2 = 1.2 V DVDD_1P2 = 1.2 V PLL_LDO_VDD12 = 1.2 V SYNC_VDD_3P3 = 3.3 V NCO ONLY MODE, 5 GSPS Analog Supply Currents VDD25_DAC = 2.5 V VDD12A = 1.2 V VDD12_CLK = 1.2 V VNEG_N1P2 = −1.2 V Digital Supply Currents DVDD = 1.2 V IOVDD1 = 2.5 V SERDES Supply Currents VDD_1P2 = 1.2 V DVDD_1P2 = 1.2 V PLL_LDO_VDD12 = 1.2 V SYNC_VDD_3P3 = 3.3 V 8 LANES, 4× INTERPOLATION (80%), 5 GSPS Analog Supply Currents VDD25_DAC = 2.5 V VDD12A = 1.2 V VDD12_CLK = 1.2 V AD9161/AD9162 Test Conditions/Comments Min Includes VTT_1P2, BIAS_VDD_1P2 Connected to PLL_CLK_VDD12 IOVDD1 = 2.5 V SERDES Supply Currents VDD_1P2 = 1.2 V DVDD_1P2 = 1.2 V PLL_LDO_VDD12 = 1.2 V SYNC_VDD_3P3 = 3.3 V Max Unit 425.5 62 84.4 9.3 550 86 106 11 mA mA mA mA NCO on, FIR85 on 93.8 3.7 228.7 −120.7 mA µA mA mA Includes VDD12_DCD/DLL 598.4 2.5 mA mA Includes VTT_1P2, BIAS_VDD_1P2 443.4 72.3 81.8 9.4 mA mA mA mA Connected to PLL_CLK_VDD12 93.7 10 340.6 −112 100 150 432 mA µA mA mA Includes VDD12_DCD/DLL 425.5 2.5 753 2.7 mA mA Includes VTT_1P2, BIAS_VDD_1P2 1.4 1.0 0.13 0.32 34 14.1 1.5 0.43 mA mA mA mA 102 80 340.5 408 −120.2 108 150 432.4 mA µA mA mA mA 665.4 706.5 894.6 1090 2.5 1033 2.7 mA mA mA mA mA 411.2 52.1 85.8 9.3 550 73 105 11 mA mA mA mA −119 Connected to PLL_CLK_VDD12 NCO on, FIR85 off (unless otherwise noted) At 6 GSPS VNEG_N1P2 = −1.2 V Digital Supply Currents DVDD = 1.2 V (Includes VDD12_DCD/DLL) DVDD = 1.2 V Typ −127.4 NCO on, FIR85 off NCO off, FIR85 on NCO on, FIR85 on NCO on, FIR85 on, at 6 GSPS Includes VTT_1P2, BIAS_VDD_1P2 Connected to PLL_CLK_VDD12 Rev. 0 | Page 5 of 139 AD9161/AD9162 Parameter 8 LANES, 3× INTERPOLATION (80%), 4.5 GSPS Analog Supply Currents VDD25_DAC = 2.5 V VDD12A = 1.2 V VDD12_CLK = 1.2 V VNEG_N1P2 = −1.2 V Digital Supply Currents DVDD = 1.2 V IOVDD1 = 2.5 V SERDES Supply Currents VDD_1P2 = 1.2 V DVDD_1P2 = 1.2 V PLL_LDO_VDD12 = 1.2 V SYNC_VDD_3P3 = 3.3 V POWER DISSIPATION 3 GSPS NRZ Mode, 2×, FIR85 Enabled, NCO On 2× NRZ Mode, 6×, FIR85 Enabled, NCO On 2× NRZ Mode, 4×, FIR85 Enabled, NCO On 2× NRZ Mode, 1×, FIR85 Enabled, NCO On NRZ Mode, 24×, FIR85 Disabled, NCO On 5 GSPS NCO Mode, FIR85 Disabled, NCO On NRZ Mode, 4×, FIR85 Disabled, NCO On 2× NRZ Mode, 4x, FIR85 Enabled, NCO Off 2× NRZ Mode, 4×, FIR85 Enabled, NCO On NRZ Mode, 8×, FIR85 Disabled, NCO On NRZ Mode, 16×, FIR85 Disabled, NCO On 2× NRZ Mode, 6×, FIR85 Enabled, NCO On NRZ Mode, 3×, FIR85 Disabled, NCO On (4.5 GSPS) 1 Data Sheet Test Conditions/Comments NCO on, FIR85 on Min Typ 94 85 314.3 −112.1 Max 175 Unit mA µA mA mA Includes VDD12_DCD/DLL IOVDD = 2.5 V 948.5 2.5 mA mA Includes VTT_1P2, BIAS_VDD_1P2 432.3 62.3 84.7 9.2 mA mA mA mA 2.1 2.1 2.1 1.94 W W W W 1.3 W Connected to PLL_CLK_VDD12 Using 80%, 2× filter, eight-lane JESD204B Using 80%, 3× filter, eight-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B 1× bypass mode (AD9162 only), eight-lane JESD204B Using 80%, 2× filter, one-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B Using 80%, 2× filter, eight-lane JESD204B Using 80%, 3× filter, eight-lane JESD204B Using 80%, 3× filter, six-lane JESD204B IOVDD can range from 1.8 V to 3.3 V, with ±5% tolerance. Rev. 0 | Page 6 of 139 1.3 2.3 2.35 2.58 2.18 2.09 2.65 2.62 1.83 W W W W W W W W Data Sheet AD9161/AD9162 SERIAL PORT AND CMOS PIN SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. Table 4. Parameter WRITE OPERATION Maximum SCLK Clock Rate SCLK Clock High SCLK Clock Low SDIO to SCLK Setup Time SCLK to SDIO Hold Time CS to SCLK Setup Time SCLK to CS Hold Time READ OPERATION SCLK Clock Rate SCLK Clock High SCLK Clock Low SDIO to SCLK Setup Time SCLK to SDIO Hold Time CS to SCLK Setup Time SCLK to SDIO (or SDO) Data Valid Time CS to SDIO (or SDO) Output Valid to High-Z INPUTS (SDIO, SCLK, CS, RESET, TX_ENABLE) Voltage Input High Low Current Input High Low OUTPUTS (SDIO, SDO) Voltage Output High Low Current Output High Low Symbol fSCLK, 1/tSCLK tPWH tPWL tDS tDH tS tH Test Comments/Conditions See Figure 141 SCLK = 20 MHz SCLK = 20 MHz Min 100 3.5 4 4 1 9 9 Typ Max Unit MHz ns ns ns ns ns ns 2 0.5 1 0.5 See Figure 140 and Figure 141 fSCLK, 1/tSCLK tPWH tPWL tDS tDH tS tDV 20 Not shown in Figure 140 or Figure 141 VIH VIL 1.8 V ≤ IOVDD ≤ 2.5 V 1.8 V ≤ IOVDD ≤ 2.5 V IIH IIL VOH VOL 17 45 MHz ns ns ns ns ns ns ns 0.3 × IOVDD V V 20 20 10 5 10 0.7 × IOVDD 75 −150 1.8 V ≤ IOVDD ≤ 3.3 V 1.8 V ≤ IOVDD ≤ 3.3 V IOH IOL 0.8 × IOVDD 0.2 × IOVDD 4 4 Rev. 0 | Page 7 of 139 µA µA V V mA mA AD9161/AD9162 Data Sheet JESD204B SERIAL INTERFACE SPEED SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. Table 5. Parameter SERIAL INTERFACE SPEED Half Rate Full Rate Oversampling 2× Oversampling Test Conditions/Comments Guaranteed operating range Min Typ 6 3 1.5 0.750 Max Unit 12.5 6.25 3.125 1.5625 Gbps Gbps Gbps Gbps SYSREF± TO DAC CLOCK TIMING SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. Table 6. Parameter1 8 mm × 8 mm BGA Package (AD9162BBCZ) SYSREF± DIFFERENTIAL SWING = 0.4 V Minimum Setup Time, tSYSS Minimum Hold Time, tSYSH SYSREF± DIFFERENTIAL SWING = 0.8 V Minimum Setup Time, tSYSS Minimum Hold Time, tSYSH SYSREF± DIFFERENTIAL SWING = 1.0 V Minimum Setup Time, tSYSS Minimum Hold Time, tSYSH 11 mm × 11 mm BGA Package (AD9161BBCZ, AD9162BBCAZ, AD9162BBCA) SYSREF± DIFFERENTIAL SWING = 1.0 V Minimum Setup Time, tSYSS Minimum Hold Time, tSYSH AC-coupled DC-coupled, common-mode voltage = 0 V DC-coupled, common-mode voltage = 1.25 V AC-coupled DC-coupled, common-mode voltage = 0 V DC-coupled, common-mode voltage = 1.25 V Min Typ Max Unit 163 160 424 318 ps ps 162 169 412 350 ps ps 163 176 376 354 ps ps 65 45 68 19 5 51 117 77 129 63 37 114 ps ps ps ps ps ps The SYSREF± pulse must be at least four DAC clock edges wide plus the setup and hold times in Table 6. For more information, see the Sync Processing Modes Overview section. tSYSS tSYSH SYSREF+ CLK+ MIN 4 DAC CLOCK EDGES Figure 2. SYSREF± to DAC Clock Timing Diagram (Only SYSREF+ and CLK+ Shown) Rev. 0 | Page 8 of 139 14379-002 1 Test Conditions/Comments DC-coupled, common-mode voltage = 1.2 V Data Sheet AD9161/AD9162 DIGITAL INPUT DATA TIMING SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. Table 7. Parameter LATENCY 1 Interface Interpolation Power-Up Time DETERMINISTIC LATENCY Fixed Variable SYSREF± to LOCAL MULTIFRAME CLOCKS (LMFC) DELAY Test Conditions/Comments Min From DAC output off to enabled Typ Max Unit 1 See Table 34 10 PCLK 2 cycle ns 12 2 PCLK2 cycles PCLK2 cycles DAC clock cycles 4 Total latency (or pipeline delay) through the device is calculated as follows: Total Latency = Interface Latency + Fixed Latency + Variable Latency + Pipeline Delay See Table 34 for examples of the pipeline delay per block. 2 PCLK is the internal processing clock for the AD9161/AD9162 and equals the lane rate ÷ 40. 1 JESD204B INTERFACE ELECTRICAL SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −40°C to +85°C, unless otherwise noted. VTT is the termination voltage. Table 8. Parameter JESD204B DATA INPUTS Input Leakage Current Logic High Logic Low Unit Interval Common-Mode Voltage Differential Voltage VTT Source Impedance Differential Impedance Differential Return Loss Common-Mode Return Loss SYSREF± INPUT Differential Impedance DIFFERENTIAL OUTPUTS (SYNCOUT±) 2 Output Differential Voltage Output Offset Voltage 1 2 Symbol Test Conditions/Comments Min TA = 25°C Input level = 1.2 V ± 0.25 V, VTT = 1.2 V Input level = 0 V UI VRCM R_VDIFF ZTT ZRDIFF RLRDIF RLRCM AC-coupled, VTT = VDD_1P2 1 At dc At dc 80 Rev. 0 | Page 9 of 139 100 8 6 1333 +1.85 1050 30 120 110 121 420 1.2 Unit µA −4 80 −0.05 110 350 1.15 As measured on the input side of the ac coupling capacitor. IEEE Standard 1596.3 LVDS compatible. Max 10 165-ball CSP_BGA (AD9162 only) 169-ball CSP_BGA Driving 100 Ω differential load VOD VOS Typ µA ps V mV Ω Ω dB dB Ω Ω 450 1.27 mV V AD9161/AD9162 Data Sheet AC SPECIFICATIONS VDD25_DAC = 2.5 V, VDD12A = VDD12_CLK = 1.2 V, VNEG_N1P2 = −1.2 V, DVDD = 1.2 V, IOVDD = 2.5 V, VDD_1P2 = DVDD_1P2 = PLL_LDO_VDD12 = 1.2 V, SYNC_VDD_3P3 = 3.3 V, IOUTFS = 40 mA, TA = −25°C. Table 9. AD9161 Specifications Parameter SPURIOUS-FREE DYNAMIC RANGE (SFDR) 1 Single Tone, fDAC = 5000 MSPS fOUT = 70 MHz fOUT = 500 MHz fOUT = 1000 MHz fOUT = 2000 MHz fOUT = 4000 MHz Single Tone, fDAC = 5000 MSPS fOUT = 70 MHz fOUT = 500 MHz fOUT = 1000 MHz fOUT = 2000 MHz fOUT = 4000 MHz Data Over Cable Service Interface Specification (DOCSIS) fOUT = 70 MHz fOUT = 70 MHz fOUT = 70 MHz fOUT = 950 MHz fOUT = 950 MHz fOUT = 950 MHz ADJACENT CHANNEL POWER fOUT = 877 MHz fOUT = 877 MHz INTERMODULATION DISTORTION fOUT = 900 MHz fOUT = 900 MHz fOUT = 1800 MHz fOUT = 1800 MHz NOISE SPECTRAL DENSITY (NSD) Single Tone, fDAC = 5000 MSPS fOUT = 550 MHz fOUT = 960 MHz fOUT = 1990 MHz 1 2 Test Conditions/Comments With Marki Microwave BAL-0006SMG 2 Min FIR85 enabled −6 dBFS, shuffle enabled FIR85 enabled fDAC = 3076 MSPS Single carrier Four carriers Eight carriers Single carriers Four carriers Eight carriers fDAC = 5000 MSPS One carrier, first adjacent channel Two carriers, first adjacent channel fDAC = 5000 MSPS, two-tone test 0 dBFS −6 dBFS, shuffle enabled 0 dBFS −6 dBFS, shuffle enabled See the Clock Input section for more details on optimizing SFDR and reducing the image of the fundamental with clock input tuning. The Marki Microwave BAL-0006SMG is used on the AD9162-FMC-EBZ evaluation board. Rev. 0 | Page 10 of 139 Typ Max Unit −82 −75 −65 −70 −55 dBc dBc dBc dBc dBc −75 −75 −70 −75 −50 dBc dBc dBc dBc dBc −70 −68 −65 −70 −68 −64 dBc dBc dBc dBc dBc dBc −76 −75 dBc dBc −75 −80 −71 −75 dBc dBc dBc dBc −157 −155 −155 dBm/Hz dBm/Hz dBm/Hz Data Sheet AD9161/AD9162 Table 10. AD9162 Specifications Parameter SPURIOUS-FREE DYNAMIC RANGE (SFDR) 1 Single Tone, fDAC = 5000 MSPS fOUT = 70 MHz fOUT = 500 MHz fOUT = 1000 MHz fOUT = 2000 MHz fOUT = 4000 MHz Single Tone, fDAC = 5000 MSPS fOUT = 70 MHz fOUT = 500 MHz fOUT = 1000 MHz fOUT = 2000 MHz fOUT = 4000 MHz DOCSIS fOUT = 70 MHz fOUT = 70 MHz fOUT = 70 MHz fOUT = 950 MHz fOUT = 950 MHz fOUT = 950 MHz Wireless Infrastructure fOUT = 960 MHz fOUT = 1990 MHz ADJACENT CHANNEL POWER fOUT = 877 MHz fOUT = 877 MHz fOUT = 1887 MHz fOUT = 1980 MHz INTERMODULATION DISTORTION fOUT = 900 MHz fOUT = 900 MHz fOUT = 1800 MHz fOUT = 1800 MHz NOISE SPECTRAL DENSITY (NSD) Single Tone, fDAC = 5000 MSPS fOUT = 550 MHz fOUT = 960 MHz fOUT = 1990 MHz SINGLE SIDEBAND (SSB) PHASE NOISE AT OFFSET 1 kHz 10 kHz 100 kHz 1 MHz 10 MHz 1 2 Test Conditions/Comments With Marki Microwave BAL-0006SMG 2 Min FIR85 enabled −6 dBFS, shuffle enabled FIR85 enabled fDAC = 3076 MSPS Single carrier Four carriers Eight carriers Single carriers Four carriers Eight carriers fDAC = 5000 MSPS Two-carrier GSM signal at −9 dBFS; across 925 MHz to 960 MHz band Two-carrier GSM signal at −9 dBFS; across 1930 MHz to 1990 MHz band fDAC = 5000 MSPS One carrier, first adjacent channel Two carriers, first adjacent channel One carriers, first adjacent channel Four carriers, first adjacent channel fDAC = 5000 MSPS, two-tone test 0 dBFS −6 dBFS, shuffle enabled 0 dBFS −6 dBFS, shuffle enabled Typ Max Unit −82 −75 −65 −70 −60 dBc dBc dBc dBc dBc −75 −75 −70 −75 −65 dBc dBc dBc dBc dBc −70 −70 −67 −70 −68 −64 dBc dBc dBc dBc dBc dBc −85 dBc −81 dBc −79 −76 −74 −70 dBc dBc dBc dBc −80 −80 −68 −78 dBc dBc dBc dBc −168 −167 −164 dBm/Hz dBm/Hz dBm/Hz −119 −125 −135 −144 −156 dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz fOUT = 3800 MHz, fDAC = 4000 MSPS See the Clock Input section for more details on optimizing SFDR and reducing the image of the fundamental with clock input tuning. The Marki Microwave BAL-0006SMG is used on the AD9162-FMC-EBZ evaluation board. Rev. 0 | Page 11 of 139 AD9161/AD9162 Data Sheet ABSOLUTE MAXIMUM RATINGS THERMAL RESISTANCE Table 11. Parameter ISET, VREF to VBG_NEG SERDINx±, VTT_1P2, SYNCOUT± OUTPUT± to VNEG_N1P2 SYSREF± CLK± to Ground RESET, IRQ, CS, SCLK, SDIO, SDO to Ground Junction Temperature1 fDAC = 6 GSPS fDAC ≤ 5.1 GSPS Ambient Operating Temperature Range (TA) Storage Temperature Range 1 Rating −0.3 V to VDD25_DAC + 0.3 V −0.3 V to SYNC_VDD_3P3 + 0.3 V −0.3 V to VDD25_DAC + 0.3 V GND − 0.5 V to +2.5 V −0.3 V to VDD12_CLK + 0.3 V −0.3 V to IOVDD + 0.3 V Typical θJA and θJC values are specified for a 4-layer JEDEC 2S2P high effective thermal conductivity test board for balled surface-mount packages. θJA is obtained in still air conditions (JESD51-2). Airflow increases heat dissipation, effectively reducing θJA. θJC is obtained with the test case temperature monitored at the bottom of the package. ΨJT is thermal characteristic parameters obtained with θJA in still air test conditions but are not applicable to the CSP_BGA package. Estimate the junction temperature (TJ) using the following equations: 105°C 110°C −40°C to +85°C TJ = TT + (ΨJT × PDISS) −65°C to +150°C Some operating modes of the device may cause the device to approach or exceed the maximum junction temperature during operation at supported ambient temperatures. Removal of heat from the device may require additional measures such as active airflow, heat sinks, or other measures. 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. where: TT is the temperature measured at the top of the package. PDISS is the total device power dissipation. Table 12. Thermal Resistance Package Type 165-Ball CSP_BGA 169-Ball CSP_BGA ESD CAUTION Rev. 0 | Page 12 of 139 θJA 15.4 14.6 θJC 0.04 0.02 Unit °C/W °C/W Data Sheet AD9161/AD9162 PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS A 2 3 4 5 6 VNEG_N1P2 VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC VDD25_DAC 7 8 OUTPUT– OUTPUT+ 9 10 11 12 VDD25_DAC VDD25_DAC VNEG_N1P2 VNEG_N1P2 13 14 15 VSS VSS ISET A VDD12A VDD12A VREF B B VSS VSS VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC VDD25_DAC VDD25_DAC VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC C CLK+ VSS VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC D CLK– VSS VSS VSS VSS VSS D E VSS VSS VSS VSS VSS VDD12_CLK E VDD12_CLK VDD12_CLK VDD12_CLK F F VDD12_CLK VDD12_CLK VDD12_CLK VSS VSS VDD12_DCD/ VDD12_DCD/ DLL DLL VBG_NEG VNEG_N1P2 VDD25_DAC C VSS VSS VSS VSS VDD12_ DCD/DLL VDD12_ DCD/DLL VSS VSS CS G G IRQ VSS VSS H VSS TX_ENABLE VSS VSS VSS VSS VSS VSS VSS SDO VSS H J SERDIN7+ VDD_1P2 RESET VSS VSS VSS VSS VSS SCLK VDD_1P2 SERDIN0+ J K SERDIN7– VDD_1P2 IOVDD DVDD DVDD DVDD DVDD DVDD SDIO VDD_1P2 SERDIN0– K L VSS VSS DVDD_1P2 DVDD_1P2 VSS VSS L M SERDIN6+ VDD_1P2 VTT_1P2 VTT_1P2 VDD_1P2 SERDIN1+ M N SERDIN6– VDD_1P2 VDD_1P2 SERDIN1– N P VSS SYNC_ VDD_3P3 R BIAS_VDD_ 1P2 1 SYSREF+ SYSREF– VSS VSS PLL_CLK_ VDD12 PLL_LDO_ VDD12 VSS SYNCOUT– SYNCOUT+ VDD_1P2 VDD_1P2 DNC VDD_1P2 VDD_1P2 PLL_LDO_ BYPASS VDD_1P2 VDD_1P2 DNC VDD_1P2 VDD_1P2 SYNC_ VDD_3P3 VSS P VSS SERDIN5+ SERDIN5– VSS SERDIN4+ SERDIN4– VSS SERDIN3– SERDIN3+ VSS SERDIN2– SERDIN2+ VSS BIAS_ VDD_1P2 R 2 3 4 5 6 7 8 9 10 11 12 13 14 15 –1.2V ANALOG SUPPLY V 2.5V ANALOG SUPPLY V 1.2V DAC SUPPLY V GROUND 1.2V DAC CLK SUPPLY V SERDES INPUT SERDES 3.3V VCO SUPPLY V SERDES 1.2V SUPPLY V DAC RF SIGNALS SYSREF±/SYNCOUT± CMOS I/O IOVDD REFERENCE DNC = DO NOT CONNECT. 14379-003 1 Figure 3. 165-Ball CSP_BGA Pin Configuration Table 13. 165-Ball CSP_BGA Pin Function Descriptions Pin No. A1, A3, A4, A11, A12, B4, B5, B10, B11, C5, C6, C9, C10, C14 A2, A5, A6, A9, A10, B3, B6, B7, B8, B9, B12, C4, C7, C8, C11, C15 A7 A8 A13, A14, B1, B2, C2, D2, D3, D13, D14, D15, E1, E2, E3, E13, E14, F6, F7, F8, F9, F10, G2, G3, G8, G13, G14, H1, H3, H6, H7, H8, H9, H10, H13, H15, J6, J7, J8, J9, J10, L1, L2, L14, L15, N6, N7, N10, P1, P15, R2, R5, R8, R11, R14 A15 Mnemonic VNEG_N1P2 VDD25_DAC Description −1.2 V Analog Supply Voltage. 2.5 V Analog Supply Voltage. OUTPUT− OUTPUT+ VSS DAC Negative Current Output. DAC Positive Current Output. Supply Return. Connect these pins to ground. ISET B13, B14 B15 VDD12A VREF C1, D1 C12 CLK+, CLK− VBG_NEG E15, F1, F2, F3, F13, F14, F15 G1 VDD12_CLK IRQ Reference Current. Connect this pin to VNEG_N1P2 with a 9.6 kΩ resistor. 1.2 V Analog Supply Voltage. 1.2 V Reference Input/Output. Connect this pin to VSS with a 1 µF capacitor. Positive and Negative DAC Clock Inputs. −1.2 V Reference. Connect this pin to VNEG_N1P2 with a 0.1 µF capacitor. 1.2 V Clock Supply Voltage. Interrupt Request Output (Active Low, Open Drain). Rev. 0 | Page 13 of 139 AD9161/AD9162 Data Sheet Pin No. G6, G7, G9, G10 G15 Mnemonic VDD12_DCD/DLL CS H14 SDO J13 SCLK K13 SDIO J3 RESET H2 TX_ENABLE P5, P11 J2, J14, K2, K14, M2, M14, N2, N14, P3, P4, P6, P7, P9, P10, P12, P13 K3 DNC VDD_1P2 K6, K7, K8, K9, K10 L3, L13 M3, M13 J1, K1 IOVDD N4, N5 DVDD DVDD_1P2 VTT_1P2 SERDIN7+, SERDIN7− SERDIN6+, SERDIN6− SERDIN5+, SERDIN5− SERDIN4+, SERDIN4SERDIN3−, SERDIN3+ SERDIN2−, SERDIN2+ SERDIN1+, SERDIN1− SERDIN0+, SERDIN0− SYSREF+, SYSREF− N8 PLL_CLK_VDD12 N9 N11, N12 PLL_LDO_VDD12 SYNCOUT− , SYNCOUT+ SYNC_VDD_3P3 PLL_LDO_BYPASS BIAS_VDD_1P2 M1, N1 R3, R4 R6, R7 R9, R10 R12, R13 M15, N15 J15, K15 P2, P14 P8 R1, R15 Rev. 0 | Page 14 of 139 Description 1.2 V Digital Supply Voltage. Serial Port Chip Select Bar (Active Low) Input. CMOS levels on this pin are determined with respect to IOVDD. Serial Port Data Output. CMOS levels on this pin are determined with respect to IOVDD. Serial Port Data Clock. CMOS levels on this pin are determined with respect to IOVDD. Serial Port Data Input/Output. CMOS levels on this pin are determined with respect to IOVDD. Reset Bar (Active Low) Input. CMOS levels on this pin are determined with respect to IOVDD. Transmit Enable Input. This pin can be used instead of the DAC output bias power down bits in Register 0x040, Bits [1:0] to enable the DAC output. CMOS levels are determined with respect to IOVDD. Do Not Connect. Do not connect to these pins. 1.2 V SERDES Digital Supply. Supply Voltage for CMOS Input/Output and SPI. Operational for 1.8 V to 3.3 V plus tolerance (see Table 1 for details). 1.2 V Digital Supply Voltage. 1.2 V SERDES Digital Supply Voltage. 1.2 V SERDES VTT Digital Supply Voltage. SERDES Lane 7 Positive and Negative Inputs. SERDES Lane 6 Positive and Negative Inputs. SERDES Lane 5 Positive and Negative Inputs. SERDES Lane 4 Positive and Negative Inputs. SERDES Lane 3 Negative and Positive Inputs. SERDES Lane 2 Negative and Positive Inputs. SERDES Lane 1 Positive and Negative Inputs. SERDES Lane 0 Positive and Negative Inputs. System Reference Positive and Negative Inputs. These pins are self biased for ac coupling. They can be ac-coupled or dc-coupled. 1.2 V SERDES Phase-Locked Loop (PLL) Clock Supply Voltage. 1.2 V SERDES PLL Supply. Negative and Positive LVDS Sync (Active Low) Output Signals. 3.3 V SERDES Sync Supply Voltage. 1.2 V SERDES PLL Supply Voltage Bypass. 1.2 V SERDES Supply Voltage. Data Sheet AD9161/AD9162 1 2 3 4 5 6 7 8 9 10 11 12 13 A VSS VNEG_N1P2 VDD25_DAC VNEG_N1P2 VDD25_DAC OUTPUT– OUTPUT+ VDD25_DAC VNEG_N1P2 VDD25_DAC VSS ISET VREF A B CLK+ VSS VSS VDD25_DAC VNEG_N1P2 VDD25_DAC VDD25_DAC VNEG_N1P2 VDD25_DAC VDD12A VDD12A VDD25_DAC VNEG_N1P2 B C CLK– VSS VSS VSS VDD25_DAC VNEG_N1P2 VNEG_N1P2 VDD25_DAC VBG_NEG VSS VSS VSS VSS C D VSS VDD12_CLK VDD12_CLK VDD12_CLK VDD12_CLK VSS VSS VDD12_CLK VDD12_CLK VDD12_CLK VDD12_CLK VDD12_CLK VDD12_CLK D E VDD12_CLK VSS VSS VSS DVDD DVDD VSS DVDD DVDD VSS VSS VSS VSS E F SYSREF+ SYSREF– VSS VSS VSS VSS VSS VSS VSS VSS VSS CS VSS F G VSS VSS TX_ENABLE IRQ DVDD DVDD DVDD DVDD DVDD SDIO SDO VSS VSS G H SERDIN7+ SERDIN7– VDD_1P2 RESET IOVDD DVDD_1P2 VSS DVDD_1P2 IOVDD SCLK VDD_1P2 SERDIN0– SERDIN0+ H J VSS VSS VDD_1P2 DNC DNC VSS VSS VSS SYNCOUT– SYNCOUT+ VDD_1P2 VSS VSS J K SERDIN6+ SERDIN6– VTT_1P2 SYNC_ VDD_3P3 DNC VSS PLL_CLK_ VDD12 PLL_LDO_ VDD12 DNC SYNC_ VDD_3P3 VTT_1P2 SERDIN1– SERDIN1+ K L VSS VSS VDD_1P2 VDD_1P2 VDD_1P2 VSS DNC VSS VDD_1P2 VDD_1P2 VDD_1P2 VSS VSS L M VSS VSS SERDIN5+ VSS SERDIN4+ VSS PLL_LDO_ BYPASS VSS SERDIN3+ VSS SERDIN2+ VSS VSS M N BIAS_VDD_1P2 VSS SERDIN5– VSS SERDIN4– VSS VSS VSS SERDIN3– VSS SERDIN2– VSS BIAS_ VDD_1P2 N 1 2 3 4 5 6 7 8 9 10 11 12 13 1.2V DAC CLK SUPPLY V SERDES INPUT SERDES 3.3V VCO SUPPLY V SERDES 1.2V SUPPLY V DAC RF SIGNALS SYSREF±/SYNCOUT± CMOS I/O IOVDD DNC = DO NOT CONNECT. REFERENCE 14379-004 –1.2V ANALOG SUPPLY V 2.5V ANALOG SUPPLY V 1.2V DAC SUPPLY V GROUND Figure 4. 169-Ball CSP_BGA Pin Configuration Table 14. 169-Ball CSP_BGA Pin Function Descriptions Pin No. A1, A11, B2, B3, C2, C3, C4, C10, C11, C12, C13, D1, D6, D7, E2, E3, E4, E7, E10, E11, E12, E13, F3, F4, F5, F6, F7, F8, F9, F10, F11, F13, G1, G2, G12, G13, H7, J1, J2, J6, J7, J8, J12, J13, K6, L1, L2, L6, L8, L12, L13, M1, M2, M4, M6, M8, M10, M12, M13, N2, N4, N6, N7, N8, N10, N12 A2, A4, A9, B5, B8, B13, C6, C7 A3, A5, A8, A10, B4, B6, B7, B9, B12, C5, C8 A6 A7 A12 Mnemonic VSS Description Supply Return. Connect these pins to ground. VNEG_N1P2 VDD25_DAC OUTPUT− OUTPUT+ ISET A13 VREF B1, C1 B10, B11 C9 CLK+, CLK− VDD12A VBG_NEG D2, D3, D4, D5, D8, D9, D10, D11, D12, D13, E1 E5, E6, E8, E9, G5, G6, G7, G8, G9 F1, F2 VDD12_CLK DVDD SYSREF+, SYSREF− −1.2 V Analog Supply Voltage. 2.5 V Analog Supply Voltage. DAC Negative Current Output. DAC Positive Current Output. Reference Current. Connect this pin to VNEG_N1P2 with a 9.6 kΩ resistor. 1.2 V Reference Input/Output. Connect this pin to VSS with a 1 µF capacitor. Positive and Negative DAC Clock Inputs. 1.2 V Analog Supply Voltage. −1.2 V Reference. Connect this pin to VNEG_N1P2 with a 0.1 µF capacitor. 1.2 V Clock Supply Voltage. 1.2 V Digital Supply Voltage. System Reference Positive and Negative Inputs. These pins are self biased for ac coupling. They can be ac-coupled or dc-coupled. Rev. 0 | Page 15 of 139 AD9161/AD9162 Data Sheet Pin No. F12 Mnemonic CS G3 TX_ENABLE G4 IRQ G10 SDIO G11 SDO H10 SCLK H3, H11, J3, J11, L3, L4, L5, L9, L10, L11 H4 VDD_1P2 RESET H5, H9 IOVDD H6, H8 H1, H2 DVDD_1P2 SERDIN7+, SERDIN7− SERDIN6+, SERDIN6− SERDIN5+, SERDIN5− SERDIN4+, SERDIN4− SERDIN3+, SERDIN3− SERDIN2+, SERDIN2− SERDIN1−, SERDIN1+ SERDIN0−, SERDIN0+ DNC SYNCOUT− , SYNCOUT+ VTT_1P2 SYNC_VDD_3P3 PLL_CLK_VDD12 PLL_LDO_VDD12 PLL_LDO_BYPASS BIAS_VDD_1P2 K1, K2 M3, N3 M5, N5 M9, N9 M11, N11 K12, K13 H12, H13 J4, J5, K5, K9, L7 J9, J10 K3, K11 K4, K10 K7 K8 M7 N1, N13 Rev. 0 | Page 16 of 139 Description Serial Port Chip Select Bar (Active Low) Input. CMOS levels on this pin are determined with respect to IOVDD. Transmit Enable Input. This pin can be used instead of the DAC output bias power down bits in Register 0x040, Bits [1:0] to enable the DAC output. CMOS levels are determined with respect to IOVDD. Interrupt Request Output (Active Low, Open Drain). Serial Port Data Input/Output. CMOS levels on this pin are determined with respect to IOVDD. Serial Port Data Output. CMOS levels on this pin are determined with respect to IOVDD. Serial Port Data Clock. CMOS levels on this pin are determined with respect to IOVDD. 1.2 V SERDES Digital Supply. Reset Bar (Active Low) Input. CMOS levels on this pin are determined with respect to IOVDD. Supply Voltage for CMOS Input/Output and SPI. Operational for 1.8 V to 3.3 V (see Table 1 for details). 1.2 V SERDES Digital Supply Voltage. SERDES Lane 7 Positive and Negative Inputs. SERDES Lane 6 Positive and Negative Inputs. SERDES Lane 5 Positive and Negative Inputs. SERDES Lane 4 Positive and Negative Inputs. SERDES Lane 3 Positive and Negative Inputs. SERDES Lane 2 Positive and Negative Inputs. SERDES Lane 1 Negative and Positive Inputs. SERDES Lane 0 Negative and Positive Inputs. Do Not Connect. Do not connect to these pins. Negative and Positive LVDS Sync (Active Low) Output Signals. 1.2 V SERDES VTT Digital Supply Voltage. 3.3 V SERDES Sync Supply Voltage. 1.2 V SERDES PLL Clock Supply Voltage. 1.2 V SERDES PLL Supply. 1.2 V SERDES PLL Supply Voltage Bypass. 1.2 V SERDES Supply Voltage. Data Sheet AD9161/AD9162 TYPICAL PERFORMANCE CHARACTERISTICS AD9161 Static Linearity 0.4 0.3 0.3 0.2 0.2 0.1 0.1 DNL (LSB) 0.4 0 –0.1 –0.1 –0.2 –0.2 –0.3 –0.3 500 1000 1500 2000 CODE –0.4 0 0.3 0.2 0.2 0.1 0.1 DNL (LSB) 0.3 0 –0.1 –0.2 –0.2 –0.3 –0.3 1000 1500 2000 CODE –0.4 0 0.3 0.2 0.2 0.1 0.1 DNL (LSB) 0.3 0 –0.1 –0.2 –0.2 –0.3 –0.3 –0.4 1500 CODE 1500 2000 0 –0.1 2000 14379-507 INL (LSB) 0.4 1000 1000 Figure 9. DNL, IOUTFS = 30 mA 0.4 500 500 CODE Figure 6. INL, IOUTFS = 30 mA 0 2000 0 –0.1 14379-506 INL (LSB) 0.4 500 1500 Figure 8. DNL, IOUTFS = 20 mA 0.4 0 1000 CODE Figure 5. INL, IOUTFS = 20 mA –0.4 500 14379-509 0 Figure 7. INL, IOUTFS = 40 mA –0.4 0 500 1000 1500 CODE Figure 10. DNL, IOUTFS = 40 mA Rev. 0 | Page 17 of 139 2000 14379-510 –0.4 14379-508 0 14379-505 INL (LSB) IOUTFS = 40 mA, nominal supplies, TA = 25°C, unless otherwise noted. AD9161/AD9162 Data Sheet AC Performance (NRZ Mode) 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –80 –40 –60 1000 2000 3000 4000 5000 FREQUENCY (MHz) 0 3000 4000 5000 Figure 14. Single-Tone Spectrum at fOUT = 2000 MHz 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –40 –60 –80 0 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-312 –80 0 –40 fDAC 2500MHz fDAC 3000MHz fDAC 5000MHz fDAC 6000MHz –50 2000 3000 4000 5000 FREQUENCY (MHz) Figure 15. Single-Tone Spectrum at fOUT = 2000 MHz (FIR85 Enabled) Figure 12. Single-Tone Spectrum at fOUT = 70 MHz (FIR85 Enabled) –40 1000 14379-315 MAGNITUDE (dBm) 2000 FREQUENCY (MHz) Figure 11. Single-Tone Spectrum at fOUT = 70 MHz fDAC 2500MHz fDAC 3000MHz fDAC 5000MHz fDAC 6000MHz –50 –60 IMD (dBc) –60 –70 –70 –80 –80 –90 –90 –100 0 500 1000 1500 2000 fOUT (MHz) 2500 3000 14379-313 SFDR (dBc) 1000 14379-314 0 14379-311 –80 –100 0 500 1000 1500 2000 fOUT (MHz) Figure 16. IMD vs. fOUT over fDAC Figure 13. SFDR vs. fOUT over fDAC Rev. 0 | Page 18 of 139 2500 3000 14379-416 MAGNITUDE (dBm) IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. Data Sheet AD9161/AD9162 IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –40 –40 DIGITAL SCALE 0dB DIGITAL SCALE –6dB DIGITAL SCALE –12dB DIGITAL SCALE –18dB –50 –50 SFDR (dBc) –60 SFDR (dBc) IOUTFS 20mA IOUTFS 30mA IOUTFS 40mA –70 –80 –60 –70 –80 –90 –90 0 500 1000 1500 2000 2500 fOUT (MHz) –100 14379-317 –100 0 2000 2500 –40 IOUTFS 20mA IOUTFS 30mA IOUTFS 40mA –50 –60 –60 IMD (dBc) IMD (dBc) 1500 Figure 20. SFDR vs. fOUT over DAC IOUTFS DIGITAL SCALE 0dB DIGITAL SCALE –6dB DIGITAL SCALE –12dB DIGITAL SCALE –18dB –50 1000 fOUT (MHz) Figure 17. SFDR vs. fOUT over Digital Full Scale –40 500 14379-322 SHUFFLE FALSE SHUFFLE TRUE –70 –80 –70 –80 –90 –90 0 1000 2000 3000 4000 5000 fOUT (MHz) –100 14379-418 –100 0 1000 2000 3000 4000 5000 fOUT (MHz) Figure 18. IMD vs. fOUT over Digital Full Scale 14379-422 SHUFFLE FALSE SHUFFLE TRUE Figure 21. IMD vs. fOUT over DAC IOUTFS fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz Figure 19. Single-Tone NSD Measured at 70 MHz vs. fOUT over fDAC –160 –165 –170 –175 14379-619 fOUT (MHz) –155 fDAC 2500MHz fDAC 3000MHz fDAC 5000MHz fDAC 6000MHz 0 500 1000 fOUT (MHz) 1500 2000 14379-425 SINGLE-TONE NSD (dBm/Hz) SINGLE-TONE NSD (dBm/Hz) –150 Figure 22. Single-Tone NSD Measured at 10% Offset from fOUT vs. fOUT over fDAC Rev. 0 | Page 19 of 139 AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –150 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –155 W-CDMA NSD (dBm/Hz) –155 –160 –165 –160 –165 500 1000 1500 fOUT (MHz) 2000 2500 3000 Figure 23. W-CDMA NSD Measured at 70 MHz vs. fOUT over fDAC –175 0 500 1000 1500 2000 2500 fOUT (MHz) Figure 25. W-CDMA NSD Measured at 10% Offset from fOUT vs. fOUT over fDAC 14379-336 0 14379-532 –175 14379-225 –170 –170 14379-333 W-CDMA NSD (dBm/Hz) –150 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz Figure 24. Single-Carrier W-CDMA at 877.5 MHz Figure 26. Two-Carrier W-CDMA at 875 MHz Rev. 0 | Page 20 of 139 Data Sheet AD9161/AD9162 AC (Mix-Mode) 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –40 –60 –80 1000 2000 3000 4000 5000 FREQUENCY (MHz) 0 3000 4000 5000 Figure 30. Single-Tone Spectrum at fOUT = 4000 MHz 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –40 –60 –80 0 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-440 –80 Figure 28. Single-Tone Spectrum at fOUT = 2550 MHz (FIR85 Enabled) 0 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-444 MAGNITUDE (dBm) 2000 FREQUENCY (MHz) Figure 27. Single-Tone Spectrum at fOUT = 2550 MHz Figure 31. Single-Tone Spectrum at fOUT = 4000 MHz (FIR85 Enabled) –140 –40 –145 –50 –150 SFDR (dBc) –60 –155 –160 –70 –80 –165 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –170 –175 2000 3000 4000 5000 6000 7000 8000 fOUT (MHz) 9000 10000 –90 14379-424 SINGLE-TONE NSD (dBm/Hz) 1000 14379-344 0 14379-439 –80 Figure 29. Single-Tone NSD vs. fOUT –100 2000 DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB SHUFFLE FALSE SHUFFLE TRUE 3000 4000 5000 6000 7000 fOUT (MHz) Figure 32. SFDR vs. fOUT over Digital Full Scale Rev. 0 | Page 21 of 139 8000 14379-445 MAGNITUDE (dBm) IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –50 –60 IMD (dBc) IMD (dBc) –60 –70 –70 –80 –80 –90 –90 SHUFFLE FALSE SHUFFLE TRUE 3000 4000 5000 6000 7000 8000 fOUT (MHz) –100 2000 14379-446 –100 2000 IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA 3000 –50 –40 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –50 SFDR (dBc) IMD (dBc) –90 –90 2000 3000 4000 5000 6000 7000 8000 9000 14379-447 –80 fOUT (MHz) –100 1000 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA –50 –60 –70 –80 4000 5000 6000 7000 fOUT (MHz) 8000 14379-448 –90 3000 2000 3000 4000 5000 6000 7000 fOUT (MHz) Figure 37. IMD vs. fOUT over fDAC Figure 34. SFDR vs. fOUT over fDAC SFDR (dBc) 8000 –70 –80 –100 2000 7000 –60 –70 –40 6000 Figure 36. IMD vs. fOUT over DAC IOUTFS –60 –100 1000 5000 fOUT (MHz) Figure 33. IMD vs. fOUT over Digital Full Scale –40 4000 14379-449 –50 –40 DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB Figure 35. SFDR vs. fOUT over DAC IOUTFS Rev. 0 | Page 22 of 139 8000 9000 14379-450 –40 Data Sheet AD9161/AD9162 0 0 –10 –10 –20 –20 –30 –40 –50 –60 –30 –40 –50 –60 –70 –70 –80 –80 1000 1500 2000 2500 3000 FREQUENCY (MHz) –90 0 –10 –10 –20 –20 MAGNITUDE (dBc) MAGNITUDE (dBc) 0 –30 –40 –50 –60 –60 –80 3000 FREQUENCY (MHz) –90 14379-361 –90 2500 0 –10 –20 –20 MAGNITUDE (dBc) 0 –30 –40 –50 –60 2000 2500 3000 –60 –80 1500 2500 –50 –70 FREQUENCY (MHz) 2000 –40 –80 1000 1500 –30 –70 3000 14379-363 MAGNITUDE (dBc) 0 500 1000 Figure 42. Four Carriers at 950 MHz Output –10 0 500 FREQUENCY (MHz) Figure 39. Four Carriers at 70 MHz Output –90 3000 –50 –80 2000 2500 –40 –70 1500 2000 –30 –70 1000 1500 Figure 41. Single Carrier at 950 MHz Output 0 500 1000 FREQUENCY (MHz) Figure 38. Single Carrier at 70 MHz Output 0 500 14379-365 500 Figure 40. Eight Carriers at 70 MHz Output –90 0 500 1000 1500 2000 2500 FREQUENCY (MHz) Figure 43. Eight Carriers at 950 MHz Output Rev. 0 | Page 23 of 139 3000 14379-366 0 14379-361 –90 14379-364 MAGNITUDE (dBc) MAGNITUDE (dBc) DOCSIS Performance (NRZ Mode) IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –50 –60 –70 –80 0 200 400 600 800 1000 1200 1400 fOUT (MHz) Figure 44. In-Band Second Harmonic vs. fOUT Performance for One DOCSIS Carrier 0 200 400 600 800 1000 1200 1400 fOUT (MHz) Figure 47. In-Band Third Harmonic vs. fOUT Performance for One DOCSIS Carrier –80 0 200 400 600 800 1000 1200 1400 Figure 45. In-Band Second Harmonic vs. fOUT Performance for Four DOCSIS Carriers –50 –60 –70 –80 –90 0 200 400 600 800 1000 1200 1400 fOUT (MHz) 14379-371 IN-BAND THIRD HARMONIC (dBc) –70 14379-368 IN-BAND SECOND HARMONIC (dBc) –60 fOUT (MHz) Figure 48. In-Band Third Harmonic vs. fOUT Performance for Four DOCSIS Carriers –40 IN-BAND THIRD HARMONIC (dBc) –40 –50 –60 –70 –80 0 200 400 600 800 fOUT (MHz) 1000 1200 1400 Figure 46. In-Band Second Harmonic vs. fOUT Performance for Eight DOCSIS Carriers –50 –60 –70 –80 –90 14379-369 IN-BAND SECOND HARMONIC (dBc) –80 –40 –50 –90 –70 –90 –40 –90 –60 0 200 400 600 800 fOUT (MHz) 1000 1200 1400 14379-372 –90 –50 14379-370 IN-BAND THIRD HARMONIC (dBc) –40 14379-367 IN-BAND SECOND HARMONIC (dBc) –40 Figure 49. In-Band Third Harmonic vs. fOUT Performance for Eight DOCSIS Carriers Rev. 0 | Page 24 of 139 Data Sheet AD9161/AD9162 IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –40 Y-AXIS: FIRST ACP Y-AXIS: SECOND ACP Y-AXIS: THIRD ACP Y-AXIS: FOURTH ACP Y-AXIS: FIFTH ACP –50 ACPR (dBc) –50 –60 –70 –80 –60 –70 0 200 400 600 800 1000 1200 1400 fOUT (MHz) –90 14379-373 –90 0 200 –70 –80 1200 1400 1200 1400 –60 –70 200 400 600 800 1000 1200 1400 –90 14379-374 0 0 200 800 1000 Figure 54. 32-Carrier ACPR vs. fOUT 0 Y-AXIS: FIRST ACP Y-AXIS: SECOND ACP Y-AXIS: THIRD ACP Y-AXIS: FOURTH ACP Y-AXIS: FIFTH ACP –10 –20 MAGNITUDE (dBc) –50 600 fOUT (MHz) Figure 51. Four-Carrier ACPR vs. fOUT –40 400 14379-377 –80 fOUT (MHz) –60 –70 –30 –40 –50 –60 –70 –80 –80 –90 0 200 400 600 800 1000 fOUT (MHz) 1200 1400 14379-375 ACPR (dBc) 1000 Y-AXIS: FIRST ACP Y-AXIS: SECOND ACP Y-AXIS: THIRD ACP Y-AXIS: FOURTH ACP Y-AXIS: FIFTH ACP –50 ACPR (dBc) ACPR (dBc) –40 –60 –90 800 Figure 53. 16-Carrier ACPR vs. fOUT Y-AXIS: FIRST ACP Y-AXIS: SECOND ACP Y-AXIS: THIRD ACP Y-AXIS: FOURTH ACP Y-AXIS: FIFTH ACP –50 600 fOUT (MHz) Figure 50. Single-Carrier Adjacent Channel Power Ratio (ACPR) vs. fOUT –40 400 14379-376 –80 –90 0 500 1000 1500 2000 2500 FREQUENCY (MHz) Figure 55. 194-Carrier, Sinc On, FIR85 Enabled Figure 52. Eight-Carrier ACPR vs. fOUT Rev. 0 | Page 25 of 139 3000 14379-378 ACPR (dBc) –40 Y-AXIS: FIRST ACP Y-AXIS: SECOND ACP Y-AXIS: THIRD ACP Y-AXIS: FOURTH ACP Y-AXIS: FIFTH ACP AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –50 –60 –70 –80 –90 –100 0 200 400 600 800 1000 1200 fGAP (fOUT = fGAP) (MHz) 1400 14379-478 ACLR IN GAP CHANNEL (dBc) –40 Figure 56. ACLR in Gap Channel vs. fGAP Rev. 0 | Page 26 of 139 Data Sheet AD9161/AD9162 AD9162 Static Linearity IOUTFS = 40 mA, nominal supplies, TA = 25°C, unless otherwise noted. 15 4 2 10 5 –2 DNL (LSB) INL (LSB) 0 0 –4 –6 –8 –5 0 10000 20000 30000 40000 50000 60000 CODE –12 14379-005 –10 0 10000 20000 30000 40000 50000 60000 CODE Figure 57. INL, IOUTFS = 20 mA 14379-008 –10 Figure 60. DNL, IOUTFS = 20 mA 15 4 2 10 5 –2 DNL (LSB) INL (LSB) 0 0 –4 –6 –8 –5 0 10000 20000 30000 40000 50000 60000 CODE –12 14379-006 –10 0 10000 20000 30000 40000 50000 60000 CODE Figure 58. INL, IOUTFS = 30 mA 14379-009 –10 Figure 61. DNL, IOUTFS = 30 mA 4 15 2 10 DNL (LSB) 5 0 –2 –4 –6 –8 –5 –10 0 10000 20000 30000 40000 50000 CODE 60000 Figure 59. INL, IOUTFS = 40 mA –12 0 10000 20000 30000 40000 50000 CODE Figure 62. DNL, IOUTFS = 40 mA Rev. 0 | Page 27 of 139 60000 14379-010 –10 14379-007 INL (LSB) 0 AD9161/AD9162 Data Sheet AC Performance (NRZ Mode) 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –40 –60 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-011 0 0 0 0 –20 –20 MAGNITUDE (dBm) 4000 5000 –40 –60 –40 –60 –80 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-012 0 2000 3000 4000 5000 FREQUENCY (MHz) –40 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –50 1000 Figure 67. Single-Tone Spectrum at fOUT = 2000 MHz (FIR85 Enabled) Figure 64. Single-Tone Spectrum at fOUT = 70 MHz (FIR85 Enabled) –40 0 14379-015 MAGNITUDE (dBm) 3000 Figure 66. Single-Tone Spectrum at fOUT = 2000 MHz –80 –50 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –60 IMD (dBc) –60 –70 –70 –80 –80 –90 –90 0 500 1000 1500 2000 fOUT (MHz) 2500 3000 14379-013 SFDR (dBc) 2000 FREQUENCY (MHz) Figure 63. Single-Tone Spectrum at fOUT = 70 MHz –100 1000 14379-014 –80 –80 Figure 65. SFDR vs. fOUT over fDAC –100 0 500 1000 1500 2000 fOUT (MHz) Figure 68. IMD vs. fOUT over fDAC Rev. 0 | Page 28 of 139 2500 3000 14379-016 MAGNITUDE (dBm) IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. Data Sheet AD9161/AD9162 IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –40 SHUFFLE FALSE SHUFFLE TRUE –50 –70 –70 –80 –80 –90 –90 –100 0 500 1000 1500 2000 2500 fOUT (MHz) –100 0 SHUFFLE FALSE SHUFFLE TRUE SFDR (dBc) –70 –80 –90 –90 500 1000 1500 2000 2500 fOUT (MHz) –100 0 –50 1500 2000 2500 IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA –60 IMD (dBc) –70 –70 –80 –80 –90 –90 500 1000 1500 fOUT (MHz) 2000 2500 14379-019 IN-BAND THIRD HARMONIC (dBc) –40 –60 –100 0 1000 Figure 73. SFDR vs. fOUT over DAC IOUTFS SHUFFLE FALSE SHUFFLE TRUE DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB 500 fOUT (MHz) Figure 70. SFDR for In-Band Second Harmonic vs. fOUT over Digital Scale –50 IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA –70 –80 –40 2500 –60 –60 –100 0 2000 –50 14379-018 IN-BAND SECOND HARMONICA (dBc) –50 1500 Figure 72. IMD vs. fOUT over Digital Scale –40 DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB 1000 fOUT (MHz) Figure 69. SFDR vs. fOUT over Digital Scale –40 500 14379-020 IMD (dBc) –60 14379-017 SFDR (dBc) –60 SHUFFLE FALSE SHUFFLE TRUE DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB 14379-021 –50 DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB Figure 71. SFDR for In-Band Third Harmonic vs. fOUT over Digital Scale Rev. 0 | Page 29 of 139 –100 0 500 1000 1500 2000 fOUT (MHz) Figure 74. IMD vs. fOUT over DAC IOUTFS 2500 14379-022 –40 AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –155 –70 –80 1000 1500 2000 2500 fOUT (MHz) –175 400 W-CDMA NSD (dBm/Hz) –155 600 800 1000 1200 1400 1600 1800 2000 fOUT (MHz) Figure 76. Single-Tone NSD Measured at 70 MHz vs. fOUT over fDAC 1600 1800 2000 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –160 –165 –175 400 600 800 1000 1200 1400 1600 fOUT (MHz) –150 fDAC fDAC fDAC fDAC W-CDMA NSD (dBm/Hz) –155 –165 1800 2000 fDAC fDAC fDAC fDAC –160 –165 –170 –170 –175 600 800 1000 1200 1400 fOUT (MHz) 1600 1800 2000 14379-224 SINGLE-TONE NSD (dBm/Hz) 1400 Figure 79. W-CDMA NSD Measured at 10% Offset from fOUT vs. fOUT over fDAC –160 –175 400 1200 –170 –170 14379-024 SINGLE-TONE NSD (dBm/Hz) –150 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –165 –155 1000 Figure 78. W-CDMA NSD Measured at 70 MHz vs. fOUT over fDAC –160 –150 800 fOUT (MHz) Figure 75. SFDR vs. fOUT over Temperature –175 400 600 14379-025 500 14379-023 –100 0 –155 –165 –170 –90 –150 –160 14379-225 SFDR (dBc) –60 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz Figure 77. Single-Tone NSD Measured at 10% Offset from fOUT vs. fOUT over fDAC Rev. 0 | Page 30 of 139 fOUT (MHz) Figure 80. IMD vs. fOUT over Temperature 14379-680 –50 –150 TEMPERATURE = –40°C TEMPERATURE = +25°C TEMPERATURE = +85°C W-CDMA NSD (dBm/Hz) –40 Data Sheet AD9161/AD9162 IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –150 –150 –155 W-CDMA NSD (dBm/Hz) –160 –165 –170 –160 –165 600 800 1000 1200 1400 1600 1800 2000 fOUT (MHz) –175 400 Figure 81. Single-Tone NSD Measured at 70 MHz vs. fOUT over Temperature 1000 1200 1400 1600 1800 2000 Figure 84. W-CDMA NSD Measured at 70 MHz vs. fOUT over Temperature –150 TEMPERATURE = –40°C TEMPERATURE = +25°C TEMPERATURE = +90°C TEMPERATURE = –40°C TEMPERATURE = +25°C TEMPERATURE = +90°C –155 W-CDMA NSD (dBm/Hz) –155 –160 –165 –170 –160 –165 –170 600 800 1000 1200 1400 1600 1800 2000 fOUT (MHz) –175 400 14379-227 SINGLE-TONE NSD (dBm/Hz) 800 fOUT (MHz) –150 –175 400 600 14379-028 –170 800 1000 1200 1400 1600 1800 2000 fOUT (MHz) Figure 85. W-CDMA NSD Measured at 10% Offset from fOUT vs. fOUT over Temperature 14379-029 14379-032 Figure 82. Single-Tone NSD Measured at 10% Offset from fOUT vs. fOUT over Temperature 600 14379-331 –175 400 TEMPERATURE = –40°C TEMPERATURE = +25°C TEMPERATURE = +90°C –155 14379-027 SINGLE-TONE NSD (dBm/Hz) TEMPERATURE = –40°C TEMPERATURE = +25°C TEMPERATURE = +90°C Figure 86. Two-Carrier W-CDMA at 875 MHz Figure 83. Single-Carrier W-CDMA at 877.5 MHz Rev. 0 | Page 31 of 139 AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –60 –65 –70 –70 ACLR (dBc) –65 –75 –80 –85 –85 1000 1200 1400 1600 1800 2000 2200 fOUT (MHz) Figure 87. Single-Carrier, W-CDMA ACLR vs. fOUT (First ACLR, Second ACLR) –65 –90 800 –60 –65 1600 1800 2000 2200 THIRD ACLR FOURTH ACLR FIFTH ACLR ACLR (dBc) –70 –75 –80 –80 –85 –85 1000 1200 1400 1600 1800 2000 2200 fOUT (MHz) Figure 88. Single-Carrier, W-CDMA ACLR vs. fOUT (Third ACLR, Fourth ACLR, Fifth ACLR) –60 1000 1200 1400 1600 1800 SSB PHASE NOISE (dBc/Hz) –120 –140 70MHz 900MHz 1800MHz 3900MHz CLOCK SOURCE –100 –120 –140 10 100 1k 10k 100k 1M 10M 100M OFFSET OVER fOUT (Hz) –180 Figure 89. SSB Phase Noise vs. Offset over fOUT, fDAC = 4000 MSPS (Two Different DAC Clock Sources Used for Best Composite Curve) 10 100 1k 10k 100k 1M 10M 100M OFFSET OVER fOUT (Hz) Figure 92. SSB Phase Noise vs. Offset over fOUT, fDAC = 6000 MSPS Rev. 0 | Page 32 of 139 14379-036 –160 14379-035 –160 2200 Figure 91. Two-Carrier, W-CDMA ACLR vs. fOUT (Third ACLR, Fourth ACLR, Fifth ACLR) –80 –100 2000 fOUT (MHz) –60 70MHz 900MHz 1800MHz 3900MHz CLOCK SOURCE –80 –90 800 14379-031 ACLR (dBc) 1400 Figure 90. Two-Carrier, W-CDMA ACLR vs. fOUT (First ACLR, Second ACLR) THIRD ACLR FOURTH ACLR FIFTH ACLR –75 –180 1200 fOUT (MHz) –70 –90 800 1000 14379-034 –60 SSB HASE NOISE (dBc/Hz) –75 –80 –90 800 FIRST ACLR SECOND ACLR 14379-033 FIRST ACLR SECOND ACLR 14379-030 ACLR (dBc) –60 Data Sheet AD9161/AD9162 AC (Mix-Mode) 0 0 –20 –20 MAGNITUDE (dBm) –40 –60 –40 –60 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-038 0 0 0 0 –20 –20 MAGNITUDE (dBm) 4000 5000 –40 –60 –40 –60 –80 1000 2000 3000 4000 5000 FREQUENCY (MHz) 14379-039 0 Figure 94. Single-Tone Spectrum at fOUT = 2350 MHz (FIR85 Enabled) 0 3000 4000 5000 Figure 97. Single-Tone Spectrum at fOUT = 4000 MHz (FIR85 Enabled) –155 –155 W-CDMA NSD (dBm/Hz) –150 –165 2000 FREQUENCY (MHz) –150 –160 1000 14379-042 MAGNITUDE (dBm) 3000 Figure 96. Single-Tone Spectrum at fOUT = 4000 MHz –80 –160 –165 –170 –170 3000 4000 5000 6000 fOUT (MHz) 7000 14379-040 SINGLE-TONE NSD (dBm/Hz) 2000 FREQUENCY (MHz) Figure 93. Single-Tone Spectrum at fOUT = 2350 MHz –175 1000 14379-041 –80 –80 Figure 95. Single-Tone NSD vs. fOUT –175 3000 4000 5000 6000 fOUT (MHz) Figure 98. W-CDMA NSD vs. fOUT Rev. 0 | Page 33 of 139 7000 14379-599 MAGNITUDE (dBm) IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. –40 –50 –50 –60 –60 SFDR (dBc) –70 –80 3000 4000 SHUFFLE FALSE SHUFFLE TRUE 5000 6000 7000 8000 fOUT (MHz) –100 2000 –50 –40 –50 IMD (dBc) –70 –80 –90 –90 3000 4000 5000 7000 6000 8000 fOUT (MHz) –100 2000 –40 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz –50 3000 4000 5000 6000 7000 8000 9000 fDAC = 2500MHz fDAC = 3000MHz fDAC = 5000MHz fDAC = 6000MHz IMD (dBc) –60 –70 –70 –80 –80 –90 –90 2000 3000 4000 5000 6000 7000 fOUT (MHz) 8000 9000 14379-046 SFDR (dBc) IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA Figure 103. IMD vs. fOUT over DAC IOUTFS –60 –100 1000 8000 fOUT (MHz) Figure 100. IMD vs. fOUT over Digital Scale –50 7000 –70 –80 –40 6000 –60 14379-045 IMD (dBc) –60 –100 2000 5000 Figure 102. SFDR vs. fOUT over DAC IOUTFS SHUFFLE FALSE SHUFFLE TRUE DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB 4000 fOUT (MHz) Figure 99. SFDR vs. fOUT over Digital Scale –40 3000 14379-047 –90 DIGITAL SCALE = 0dB DIGITAL SCALE = –6dB DIGITAL SCALE = –12dB DIGITAL SCALE = –18dB 14379-048 –100 2000 –70 –80 14379-044 –90 IOUTFS = 20mA IOUTFS = 30mA IOUTFS = 40mA 14379-049 SFDR (dBc) –40 –100 1000 Figure 101. SFDR vs. fOUT over fDAC 2000 3000 4000 5000 6000 7000 fOUT (MHz) Figure 104. IMD vs. fOUT over fDAC Rev. 0 | Page 34 of 139 8000 Data Sheet AD9161/AD9162 14379-053 14379-051 IOUTFS = 40 mA, fDAC = 5.0 GSPS, nominal supplies, TA = 25°C, unless otherwise noted. Figure 105. Single-Carrier W-CDMA at 1887.5 MHz –60 FIRST ACLR SECOND ACLR –65 –70 –70 –75 –80 –85 –85 2800 3000 3200 3400 3600 3800 fOUT (MHz) Figure 106. Single-Carrier, W-CDMA ACLR vs. fOUT (First ACLR, Second ACLR) –90 2600 3000 3200 3400 3600 3800 Figure 109. Four-Carrier, W-CDMA ACLR vs. fOUT (First ACLR, Second ACLR) –60 THIRD ACLR FOURTH ACLR FIFTH ACL THIRD ACLR FOURTH ACLR FIFTH ACL –65 –65 –70 ACLR (dBc) –70 –75 –75 –80 –80 3000 3200 fOUT (MHz) 3400 3600 3800 14379-055 2800 Figure 107. Single-Carrier, W-CDMA ACLR vs. fOUT (Third ACLR, Fourth ACLR, Fifth ACLR) –90 2600 2800 3000 3200 fOUT (MHz) 3400 3600 3800 14379-057 –85 –85 –90 2600 2800 fOUT (MHz) –60 ACLR (dBc) –75 –80 –90 2600 FIRST ACLR SECOND ACLR 14379-056 ACLR (dBc) –65 14379-054 ACLR (dBc) –60 Figure 108. Four-Carrier W-CDMA at 1980 MHz Figure 110. Four-Carrier, W-CDMA ACLR vs. fOUT (Third ACLR, Fourth ACLR, Fifth ACLR) Rev. 0 | Page 35 of 139 AD9161/AD9162 Data Sheet 0 –10 –10 –20 –20 –30 –40 –50 –60 –30 –40 –50 –60 –70 –70 –80 –80 –90 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) –90 –10 –20 –20 –30 –30 MAGNITUDE (dBc) –10 –40 –50 –60 –80 –90 2500 2500 3000 –60 –80 2000 2000 –50 –70 1500 1500 –40 –70 3000 FREQUENCY (MHz) –90 14379-059 MAGNITUDE (dBc) 0 1000 1000 Figure 114. Single Carrier at 70 MHz Output (Shuffle On) 0 500 500 FREQUENCY (MHz) Figure 111. Single Carrier at 70 MHz Output 0 0 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) 14379-362 0 14379-361 MAGNITUDE (dBc) 0 14379-058 MAGNITUDE (dBc) DOCSIS Performance (NRZ Mode) IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. Figure 115. Four Carriers at 70 MHz Output (Shuffle On) Figure 112. Four Carriers at 70 MHz Output 0 0 –10 –10 –20 MAGNITUDE (dBc) –40 –50 –60 –30 –40 –50 –60 –70 –70 –80 0 500 1000 1500 2000 2500 FREQUENCY (MHz) 3000 Figure 113. Eight Carriers at 70 MHz Output –90 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) Figure 116. Eight Carriers at 70 MHz Output (Shuffle On) Rev. 0 | Page 36 of 139 14379-363 –80 –90 14379-060 MAGNITUDE (dBc) –20 –30 Data Sheet AD9161/AD9162 0 –10 –10 –20 –20 –30 –40 –50 –60 –30 –40 –50 –60 –70 –70 –80 –80 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) –90 –10 –20 –20 MAGNITUDE (dBc) 0 –30 –40 –50 –60 2500 3000 FREQUENCY (MHz) –90 0 –10 –20 –20 MAGNITUDE (dBc) –10 –30 –40 –50 –60 2500 FREQUENCY (MHz) 3000 3000 –60 –80 2000 2500 –50 –80 1500 2000 –40 –70 1000 1500 –30 –70 14379-062 MAGNITUDE (dBc) 0 500 1000 Figure 121. Eight Carriers at 950 MHz Output 0 0 500 FREQUENCY (MHz) Figure 118. Single Carrier at 950 MHz Output (Shuffle On) –90 3000 –60 –80 2000 2500 –50 –80 1500 2000 –40 –70 1000 1500 –30 –70 14379-364 MAGNITUDE (dBc) 0 500 1000 Figure 120. Four Carriers at 950 MHz Output (Shuffle On) –10 0 500 FREQUENCY (MHz) Figure 117. Single Carrier at 950 MHz Output –90 0 14379-063 0 Figure 119. Four Carriers at 950 MHz Output –90 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) Figure 122. Eight Carriers at 950 MHz Output (Shuffle On) Rev. 0 | Page 37 of 139 14379-366 –90 14379-365 MAGNITUDE (dBc) 0 14379-061 MAGNITUDE (dBc) IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –40 –60 –70 –80 0 200 400 600 800 1000 1200 1400 fOUT (MHz) Figure 123. In-Band Second Harmonic vs. fOUT Performance for One DOCSIS Carrier 0 200 400 600 800 1000 1200 1400 fOUT (MHz) Figure 126. In-Band Third Harmonic vs. fOUT Performance for One DOCSIS Carrier –80 0 200 400 600 800 1000 1200 1400 Figure 124. In-Band Second Harmonic vs. fOUT Performance for Four DOCSIS Carriers –50 –60 –70 –80 –90 0 200 400 600 800 1000 1200 1400 fOUT (MHz) 14379-068 IN-BAND THIRD HARMONIC (dBc) –70 14379-065 IN-BAND SECOND HARMONIC (dBc) –60 fOUT (MHz) Figure 127. In-Band Third Harmonic vs. fOUT Performance for Four DOCSIS Carriers –40 IN-BAND THIRD HARMONIC (dBc) –40 –50 –60 –70 –80 0 200 400 600 800 fOUT (MHz) 1000 1200 1400 Figure 125. In-Band Second Harmonic vs. fOUT Performance for Eight DOCSIS Carriers –50 –60 –70 –80 –90 14379-066 IN-BAND SECOND HARMONIC (dBc) –80 –40 –50 –90 –70 –90 –40 –90 –60 0 200 400 600 800 fOUT (MHz) 1000 1200 1400 14379-069 –90 –50 14379-067 IN-BAND THIRD HARMONIC (dBc) –50 14379-064 IN-BAND SECOND HARMONIC (dBc) –40 Figure 128. In-Band Third Harmonic vs. fOUT Performance for Eight DOCSIS Carriers Rev. 0 | Page 38 of 139 Data Sheet AD9161/AD9162 IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –40 –60 –70 –60 –70 400 600 800 1000 1200 1400 fOUT (MHz) –90 0 200 –70 –80 0 200 400 600 800 1000 1200 1400 fOUT (MHz) 1400 1200 1400 –60 –70 –90 0 –40 400 600 800 1000 Figure 133. 32-Carrier ACPR vs. fOUT 0 Y-AXIS: FIRST ACPR Y-AXIS: SECOND ACPR Y-AXIS: THIRD ACPR Y-AXIS: FOURTH ACPR Y-AXIS: FIFTH ACPR –10 –20 MAGNITUDE (dBc) –50 200 fOUT (MHz) Figure 130. Four-Carrier ACPR vs. fOUT –60 –70 –30 –40 –50 –60 –70 –80 –90 0 200 400 600 800 1000 1200 fOUT (MHz) 1400 –90 0 500 1000 1500 2000 2500 FREQUENCY (MHz) Figure 131. Eight-Carrier ACPR vs. fOUT Figure 134. 194-Carrier, Sinc Enabled, FIR85 Enabled Rev. 0 | Page 39 of 139 3000 14379-075 –80 14379-072 ACPR (dBc) 1200 –80 14379-071 –90 1000 Y-AXIS: FIRST ACPR Y-AXIS: SECOND ACPR Y-AXIS: THIRD ACPR Y-AXIS: FOURTH ACPR Y-AXIS: FIFTH ACPR –50 ACPR (dBc) ACPR (dBc) –40 –60 800 Figure 132. 16-Carrier ACPR vs. fOUT Y-AXIS: FIRST ACPR Y-AXIS: SECOND ACPR Y-AXIS: THIRD ACPR Y-AXIS: FOURTH ACPR Y-AXIS: FIFTH ACPR –50 600 fOUT (MHz) Figure 129. Single-Carrier Adjacent Channel Power Ratio (ACPR) vs. fOUT –40 400 14379-073 200 14379-070 0 14379-074 –80 –80 –90 Y-AXIS: FIRST ACPR Y-AXIS: SECOND ACPR Y-AXIS: THIRD ACPR Y-AXIS: FOURTH ACPR Y-AXIS: FIFTH ACPR –50 ACPR (dBc) –50 ACPR (dBc) –40 Y-AXIS: FIRST ACPR Y-AXIS: SECOND ACPR Y-AXIS: THIRD ACPR Y-AXIS: FOURTH ACPR Y-AXIS: FIFTH ACPR AD9161/AD9162 Data Sheet IOUTFS = 40 mA, fDAC = 3.076 GSPS, nominal supplies, FIR85 enabled, TA = 25°C, unless otherwise noted. –40 –25 –35 –50 ACLR IN GAP CHANNEL (dBc) –55 –65 –75 –85 –95 –105 –60 –70 –80 –90 –125 CENTER 77MHz RES BW 10kHz VBW 1.kHz SPAN 60.0MHz SWEEP 6.041s (1001pts) –100 Figure 135. Gap Channel ACLR at 77 MHz 0 200 400 600 800 1000 1200 fGAP (fOUT = fGAP) (MHz) Figure 136. ACLR in Gap Channel vs. fGAP Rev. 0 | Page 40 of 139 1400 14379-077 –115 14379-076 MAGNITUDE (dBm) –45 Data Sheet AD9161/AD9162 TERMINOLOGY Integral Nonlinearity (INL) INL is the maximum deviation of the actual analog output from the ideal output, determined by a straight line drawn from zero scale to full scale. Differential Nonlinearity (DNL) DNL is the measure of the variation in analog value, normalized to full scale, associated with a 1 LSB change in digital input code. Offset Error Offset error is the deviation of the output current from the ideal of 0 mA. For OUTPUT+, 0 mA output is expected when all inputs are set to 0. For OUTPUT−, 0 mA output is expected when all inputs are set to 1. Gain Error Gain error is the difference between the actual and ideal output span. The actual span is determined by the difference between the output when the input is at its minimum code and the output when the input is at its maximum code. Temperature Drift Temperature drift is specified as the maximum change from the ambient (25°C) value to the value at either TMIN or TMAX. For offset and gain drift, the drift is reported in ppm of full-scale range (FSR) per degree Celsius. For reference drift, the drift is reported in ppm per degree Celsius. Settling Time Settling time is the time required for the output to reach and remain within a specified error band around its final value, measured from the start of the output transition. Spurious-Free Dynamic Range (SFDR) SFDR is the difference, in decibels, between the peak amplitude of the output signal and the peak spurious signal within the dc to Nyquist frequency of the DAC. Typically, energy in this band is rejected by the interpolation filters. This specification, therefore, defines how well the interpolation filters work and the effect of other parasitic coupling paths on the DAC output. Signal-to-Noise Ratio (SNR) SNR is the ratio of the rms value of the measured output signal to the rms sum of all other spectral components below the Nyquist frequency, excluding the first six harmonics and dc. The value for SNR is expressed in decibels. Interpolation Filter If the digital inputs to the DAC are sampled at a multiple rate of the interpolation rate (fDATA), a digital filter can be constructed that has a sharp transition band near fDATA/2. Images that typically appear around the output data rate (fDAC) can be greatly suppressed. Adjacent Channel Leakage Ratio (ACLR) ACLR is the ratio in decibels relative to the carrier (dBc) between the measured power within a channel relative to its adjacent channel. Adjusted DAC Update Rate The adjusted DAC update rate is the DAC update rate divided by the smallest interpolating factor. For clarity on DACs with multiple interpolating factors, the adjusted DAC update rate for each interpolating factor may be given. Physical Lane Physical Lane x refers to SERDINx±. Logical Lane Logical Lane x refers to physical lanes after optionally being remapped by the crossbar block (Register 0x308 to Register 0x30B). Link Lane Link Lane x refers to logical lanes considered in the link. Rev. 0 | Page 41 of 139 AD9161/AD9162 Data Sheet THEORY OF OPERATION The AD9161/AD9162 are 11-bit and 16-bit single RF DACs with a SERDES interface. Figure 1 shows a detailed functional block diagram of the AD9161/AD9162. Eight high speed serial lanes carry data at a maximum speed of 12.5 Gbps, and either a 5 GSPS real input or a 2.5 GSPS complex input data rate to the DAC. Compared to either LVDS or CMOS interfaces, the SERDES interface simplifies pin count, board layout, and input clock requirements to the device. The clock for the input data is derived from the DAC clock, or device clock (required by the JESD204B specification). This device clock is sourced with a high fidelity direct external DAC sampling clock. The performance of the DAC can be optimized by using on-chip adjustments to the clock input accessible through the SPI port. The device can be configured to operate in one-lane, twolane, three-lane, four-lane, six-lane, or eight-lane modes, depending on the required input data rate. The digital datapath of the AD9161/AD9162 offers a bypass (1×) mode (AD9162 only) and several interpolation modes (2×, 3×, 4×, 6×, 8×, 12×, 16×, and 24×) through either an initial halfband (2×) or third-band (3×) filter with programmable 80% or 90% bandwidth, and three subsequent half-band filters (all 90%) with a maximum DAC sample rate of 6 GSPS. An inverse sinc filter is provided to compensate for sinc related roll-off. An additional half-band filter, FIR85, takes advantage of the quadswitch architecture to interpolate on the falling edge of the clock, and effectively double the DAC update rate in 2× NRZ mode. A 48-bit programmable modulus NCO is provided to enable digital frequency shifts of signals with near infinite precision. The NCO can be operated alone in NCO only mode (AD9162 only) or with digital data from the SERDES interface and digital datapath. The 100 MHz speed of the SPI write interface enables rapid updating of the frequency tuning word of the NCO. to 38.76 mA, typically. The differential current outputs are complementary. The DAC uses the patented quad-switch architecture, which enables DAC decoder options to extend the output frequency range into the second and third Nyquist zones with Mix-Mode, return to zero (RZ) mode, and 2× NRZ mode (with FIR85 enabled). Operating as a real mode DAC in 1× bypass (AD9162 only) and NRZ mode, the output signal can range from 0 Hz to 2.5 GHz. Mix-Mode can be used to access 1.5 GHz to around 7.5 GHz. In the interpolation modes, the output can range from 0 Hz to 6 GHz in 2× NRZ mode using the NCO to shift a signal of up to 1.8 GHz instantaneous bandwidth to the desired fOUT. The AD9161/AD9162 are capable of multichip synchronization that can both synchronize multiple DACs and establish a constant and deterministic latency (latency locking) path for the DACs. The latency for each of the DACs remains constant to within several DAC clock cycles from link establishment to link establishment. An external alignment (SYSREF±) signal makes the AD9161/AD9162 Subclass 1 compliant. Several modes of SYSREF± signal handling are available for use in the system. An SPI configures the various functional blocks and monitors their statuses. The various functional blocks and the data interface must be set up in a specific sequence for proper operation (see the Start-Up Sequence section). Simple SPI initialization routines set up the JESD204B link and are included in the evaluation board package. This data sheet describes the various blocks of the AD9161/AD9162 in greater detail. Descriptions of the JESD204B interface, control parameters, and various registers to set up and monitor the device are provided. The recommended start-up routine reliably sets up the data link. The AD9161/AD9162 DAC core provides a fully differential current output with a nominal full-scale current of 38.76 mA. The full-scale output current, IOUTFS, is user adjustable from 8 mA Rev. 0 | Page 42 of 139 Data Sheet AD9161/AD9162 SERIAL PORT OPERATION The serial port is a flexible, synchronous serial communications port that allows easy interfacing with many industry-standard microcontrollers and microprocessors. The serial input/output (I/O) is compatible with most synchronous transfer formats, including both the Motorola SPI and Intel® SSR protocols. The interface allows read/write access to all registers that configure the AD9161/AD9162. MSB first or LSB first transfer formats are supported. The serial port interface can be configured as a 4-wire interface or a 3-wire interface in which the input and output share a single-pin I/O (SDIO). CS F12 The serial clock pin synchronizes data to and from the device and runs the internal state machines. The maximum frequency of SCLK is 100 MHz. All data input is registered on the rising edge of SCLK. All data is driven out on the falling edge of SCLK. SPI PORT 14379-078 SCLK H10 Figure 137. Serial Port Interface Pins (11 mm × 11 mm CSP_BGA) There are two phases to a communication cycle with the AD9161/ AD9162. Phase 1 is the instruction cycle (the writing of an instruction byte into the device), coincident with the first 16 SCLK rising edges. The instruction word provides the serial port controller with information regarding the data transfer cycle, Phase 2 of the communication cycle. The Phase 1 instruction word defines whether the upcoming data transfer is a read or write, along with the starting register address for the following data transfer. A logic high on the CS pin followed by a logic low resets the serial port timing to the initial state of the instruction cycle. From this state, the next 16 rising SCLK edges represent the instruction bits of the current I/O operation. The remaining SCLK edges are for Phase 2 of the communication cycle. Phase 2 is the actual data transfer between the device and the system controller. Phase 2 of the communication cycle is a transfer of one or more data bytes. Eight × N SCLK cycles are needed to transfer N bytes during the transfer cycle. Registers change immediately upon writing to the last bit of each transfer byte, except for the frequency tuning word (FTW) and numerically controlled oscillator (NCO) phase offsets, which change only when the frequency tuning word FTW_LOAD_REQ bit is set. DATA FORMAT The instruction byte contains the information shown in Table 15. Table 15. Serial Port Instruction Word I15 (MSB) R/W SERIAL PORT PIN DESCRIPTIONS Serial Clock (SCLK) SDO G11 SDIO G10 A14 to A0, Bit I14 to Bit I0 of the instruction word, determine the register that is accessed during the data transfer portion of the communication cycle. For multibyte transfers, A[14:0] is the starting address. The remaining register addresses are generated by the device based on the address increment bit. If the address increment bits are set high (Register 0x000, Bit 5 and Bit 2), multibyte SPI writes start on A[14:0] and increment by 1 every eight bits sent/received. If the address increment bits are set to 0, the address decrements by 1 every eight bits. I[14:0] A[14:0] R/W, Bit 15 of the instruction word, determines whether a read or a write data transfer occurs after the instruction word write. Logic 1 indicates a read operation, and Logic 0 indicates a write operation. Chip Select (CS) An active low input starts and gates a communication cycle. CS allows more than one device to be used on the same serial communications lines. The SDIO pin goes to a high impedance state when this input is high. During the communication cycle, the chip select must stay low. Serial Data I/O (SDIO) This pin is a bidirectional data line. In 4-wire mode, this pin acts as the data input and SDO acts as the data output. SERIAL PORT OPTIONS The serial port can support both MSB first and LSB first data formats. This functionality is controlled by the LSB first bit (Register 0x000, Bit 6 and Bit 1). The default is MSB first (LSB bit = 0). When the LSB first bits = 0 (MSB first), the instruction and data bits must be written from MSB to LSB. R/W is followed by A[14:0] as the instruction word, and D[7:0] is the data-word. When the LSB first bits = 1 (LSB first), the opposite is true. A[0:14] is followed by R/W, which is subsequently followed by D[0:7]. The serial port supports a 3-wire or 4-wire interface. When the SDO active bits = 1 (Register 0x000, Bit 4 and Bit 3), a 4-wire interface with a separate input pin (SDIO) and output pin (SDO) is used. When the SDO active bits = 0, the SDO pin is unused and the SDIO pin is used for both the input and the output. Multibyte data transfers can be performed as well by holding the CS pin low for multiple data transfer cycles (eight SCLKs) after the first data transfer word following the instruction cycle. The first eight SCLKs following the instruction cycle read from or write to the register provided in the instruction cycle. For each additional eight SCLK cycles, the address is either incremented or decremented and the read/write occurs on the new register. The direction of the address can be set using ADDRINC or ADDRINC_M (Register 0x000, Bit 5 and Bit 2). When ADDRINC Rev. 0 | Page 43 of 139 AD9161/AD9162 Data Sheet To prevent confusion and to ensure consistency between devices, the chip tests the first nibble following the address phase, ignoring the second nibble. This test is completed independently from the LSB first bits and ensures that there are extra clock cycles following the soft reset bits (Register 0x000, Bit 0 and Bit 7). This test of the first nibble only applies when writing to Register 0x000. SCLK SDIO A0 A1 A2 A12 A13 A14 R/W D0 0 D10 D20 D4N D5N D6N D7N Figure 139. Serial Register Interface Timing, LSB First, Register 0x000, Bit 5 and Bit 2 = 1 CS SCLK DATA TRANSFER CYCLE tDV CS SDIO R/W A14 A13 A3 A2 A1 A0 D7N D6N D5N D30 D20 D10 D00 14379-079 SCLK SDIO DATA TRANSFER CYCLE DATA BIT n DATA BIT n – 1 Figure 140. Timing Diagram for Serial Port Register Read Figure 138. Serial Register Interface Timing, MSB First, Register 0x000, Bit 5 and Bit 2 = 0 tS tH CS tPWH tPWL tDS SDIO tDH INSTRUCTION BIT 15 INSTRUCTION BIT 14 INSTRUCTION BIT 0 Figure 141. Timing Diagram for Serial Port Register Write Rev. 0 | Page 44 of 139 14379-082 SCLK 14379-081 INSTRUCTION CYCLE INSTRUCTION CYCLE CS 14379-080 or ADDRINC_M is 1, the multicycle addresses are incremented. When ADDRINC or ADDRINC_M is 0, the addresses are decremented. A new write cycle can always be initiated by bringing CS high and then low again. Data Sheet AD9161/AD9162 JESD204B SERIAL DATA INTERFACE The various combinations of JESD204B parameters that are supported depend solely on the number of lanes. Thus, a unique set of parameters can be determined by selecting the lane count to be used. In addition, the interpolation rate and number of lanes can be used to define the rest of the configuration needed to set up the AD9161/AD9162. The interpolation rate and the number of lanes are selected in Register 0x110. JESD204B OVERVIEW The AD9161/AD9162 have eight JESD204B data ports that receive data. The eight JESD204B ports can be configured as part of a single JESD204B link that uses a single system reference (SYSREF±) and device clock (CLK±). The JESD204B serial interface hardware consists of three layers: the physical layer, the data link layer, and the transport layer. These sections of the hardware are described in subsequent sections, including information for configuring every aspect of the interface. Figure 142 shows the communication layers implemented in the AD9161/AD9162 serial data interface to recover the clock and deserialize, descramble, and deframe the data before it is sent to the digital signal processing section of the device. The AD9161/AD9162 have a single DAC output; however, for the purposes of the complex signal processing on chip, the converter count is defined as M = 2 whenever interpolation is used. For a particular application, the number of converters to use (M) and the DataRate variable are known. The LaneRate variable and number of lanes (L) can be traded off as follows: DataRate = (DACRate)/(InterpolationFactor) LaneRate = (20 × DataRate × M)/L The physical layer establishes a reliable channel between the transmitter (Tx) and the receiver (Rx), the data link layer is responsible for unpacking the data into octets and descrambling the data. The transport layer receives the descrambled JESD204B frames and converts them to DAC samples. where LaneRate must be between 750 Mbps and 12.5 Gbps. Achieving and recovering synchronization of the lanes is very important. To simplify the interface to the transmitter, the AD9161/AD9162 designate a master synchronization signal for each JESD204B link. The SYNCOUT± pin is used as the master signal for all lanes. If any lane in a link loses synchronization, a resynchronization request is sent to the transmitter via the synchronization signal of the link. The transmitter stops sending data and instead sends synchronization characters to all lanes in that link until resynchronization is achieved. A number of JESD204B parameters (L, F, K, M, N, NP, S, HD) define how the data is packed and tell the device how to turn the serial data into samples. These parameters are defined in detail in the Transport Layer section. The AD9161/AD9162 also have a descrambling option (see the Descrambler section for more information). SYNCOUT± PHYSICAL LAYER SERDIN7± DATA LINK LAYER TRANSPORT LAYER QBD/ DESCRAMBLER FRAME TO SAMPLES I DATA[15:0] DESERIALIZER TO DAC DSP BLOCK Q DATA[15:0] DESERIALIZER 14379-083 SERDIN0± SYSREF± Figure 142. Functional Block Diagram of Serial Link Receiver Table 16. Single-Link JESD204B Operating Modes Parameter L (Lane Count) M (Converter Count) F (Octets per Frame per Lane) S (Samples per Converter per Frame) 1 1 2 4 1 2 2 2 2 1 Rev. 0 | Page 45 of 139 3 3 2 4 3 4 4 2 1 1 Number of Lanes (L) 6 8 6 8 2 1 (real), 2 (complex) 2 1 3 4 (real), 2 (complex) AD9161/AD9162 Data Sheet Table 17. Data Structure per Lane for JESD204B Operating Modes 1 JESD204B Parameters L = 8, M = 1, F = 1, S = 4 L = 8, M = 2, F = 1, S = 2 L = 6, M = 2, F = 2, S = 3 L = 4, M = 2, F = 1, S = 1 L = 3, M = 2, F = 4, S = 3 L = 2, M = 2, F = 2, S = 1 L = 1, M = 2, F = 4, S = 1 1 Lane No. Lane 0 Lane 1 Lane 2 Lane 3 Lane 4 Lane 5 Lane 6 Lane 7 Lane 0 Lane 1 Lane 2 Lane 3 Lane 4 Lane 5 Lane 6 Lane 7 Lane 0 Lane 1 Lane 2 Lane 3 Lane 4 Lane 5 Lane 0 Lane 1 Lane 2 Lane 3 Lane 0 Lane 1 Lane 2 Lane 0 Lane 1 Lane 0 Frame 0 M0S0[15:8] M0S0[7:0] M0S1[15:8] M0S1[7:0] M0S2[15:8] M0S2[7:0] M0S3[15:8] M0S3[7:0] M0S0[15:8] M0S0[7:0] M0S1[15:8] M0S1[7:0] M1S0[15:8] M1S0[7:0] M1S1[15:8] M1S1[7:0] M0S0[15:8] M0S1[15:8] M0S2[15:8] M1S0[15:8] M1S1[15:8] M1S2[15:8] M0S0[15:8] M0S0[7:0] M1S0[15:8] M1S0[7:0] M0S0[15:8] M0S2[15:8] M1S1[15:8] M0S0[15:8] M1S0[15:8] M0S0[15:8] Frame 1 Frame 2 Frame 3 M0S1[15:8] M1S0[15:8] M1S2[15:8] M0S1[7:0] M1S0[7:0] M1S2[7:0] M1S0[15:8] M1S0[7:0] M0S0[7:0] M0S1[7:0] M0S2[7:0] M1S0[7:0] M1S1[7:0] M1S2[7:0] M0S0[7:0] M0S2[7:0] M1S1[7:0] M0S0[7:0] M1S0[7:0] M0S0[7:0] Mx is the converter number and Sy is the sample number. For example, M0S0 means Converter 0, Sample 0. Blank cells are not applicable. PHYSICAL LAYER The physical layer of the JESD204B interface, hereafter referred to as the deserializer, has eight identical channels. Each channel consists of the terminators, an equalizer, a clock and data recovery (CDR) circuit, and the 1:40 demux function (see Figure 143). DESERIALIZER TERMINATION EQUALIZER CDR SPI CONTROL FROM SERDES PLL Figure 143. Deserializer Block Diagram 1:40 14379-084 SERDINx± JESD204B data is input to the AD9161/AD9162 via the SERDINx± 1.2 V differential input pins as per the JESD204B specification. Interface Power-Up and Input Termination Before using the JESD204B interface, it must be powered up by setting Register 0x200, Bit 0 = 0. In addition, each physical lane (PHY) that is not being used (SERDINx±) must be powered down. To do so, set the corresponding Bit x for Physical Lane x in Register 0x201 to 0 if the physical lane is being used, and to 1 if it is not being used. The AD9161/AD9162 autocalibrate the input termination to 50 Ω. Before running the termination calibration, Register 0x2A7 and Register 0x2AE must be written as described in Table 18 to guarantee proper calibration. The termination calibration begins when Register 0x2A7, Bit 0 and Register 0x2AE, Bit 0 transition from low to high. Register 0x2A7 controls autocalibration for Rev. 0 | Page 46 of 139 Data Sheet AD9161/AD9162 Clock Relationships PHY 0, PHY 1, PHY 6, and PHY 7. Register 0x2AE controls autocalibration for PHY 2, PHY 3, PHY 4, and PHY 5. The PHY termination autocalibration routine is as shown in Table 18. The following clocks rates are used throughout the rest of the JESD204B section. The relationship between any of the clocks can be derived from the following equations: DataRate = (DACRate)/(InterpolationFactor) Table 18. PHY Termination Autocalibration Routine Address 0x2A7 Value 0x01 0x2AE 0x01 LaneRate = (20 × DataRate × M)/L Description Autotune PHY 0, PHY 1, PHY 6, and PHY 7 terminations Autotune PHY 2, PHY 3, PHY 4, and PHY 5 terminations ByteRate = LaneRate/10 This relationship comes from 8-bit/10-bit encoding, where each byte is represented by 10 bits. The input termination voltage of the DAC is sourced externally via the VTT_1P2 pins (M3 and M 13 on the 8 mm × 8 mm BGA package, or K3 and K13 on the 11 mm × 11 mm BGA package). Set VTT, the termination voltage, by connecting it to VDD_1P2. It is recommended that the JESD204B inputs be accoupled to the JESD204B transmit device using 100 nF capacitors. PCLK Rate = ByteRate/4 The processing clock is used for a quad-byte decoder. FrameRate = ByteRate/F where F is defined as octets per frame per lane. PCLK Factor = FrameRate/PCLK Rate = 4/F The calibration code of the termination can be read from Bits[3:0] in Register 0x2AC (PHY 0, PHY 1, PHY 6, PHY 7) and Register 0x2B3 (PHY 2, PHY 3, PHY 4, PHY 5). If needed, the termination values can be adjusted or set using several registers. The TERM_BLKx_CTRLREG1 registers (Register 0x2A8 and Register 0x2AF), can override the autocalibrated value. When set to 0xXXX0XXXX, the termination block autocalibrates, which is the normal, default setting. When set to 0xXXX1XXXX, the autocalibration value is overwritten with the value in Bits[3:1] of Register 0x2A8 and Register 0x2AF. Individual offsets from the autocalibration value for each lane can be programmed in Bits[3:0] of Register 0x2BB to Register 0x2C2. The value is a signed magnitude, with Bit 3 as the sign bit. The total range of the termination resistor value is about 94 Ω to 120 Ω, with approximately 3.5% increments across the range (for example, smaller steps at the bottom of the range than at the top). where: M is the JESD204B parameter for converters per link. L is the JESD204B parameter for lanes per link. F is the JESD204B parameter for octets per frame per lane. Receiver Eye Mask The reference clock to the SERDES PLL is always running at a frequency, fREF, that is equal to 1/40 of the lane rate (PCLK Rate). This clock is divided by a DivFactor value (set by SERDES_PLL_ DIV_FACTOR) to deliver a clock to the phase frequency detector (PFD) block that is between 35 MHz and 80 MHz. Table 19 includes the respective SERDES_PLL_DIV_FACTOR register settings for each of the desired PLL_REF_CLK_RATE options available. The AD9161/AD9162 comply with the JESD204B specification regarding the receiver eye mask and is capable of capturing data that complies with this mask. Figure 144 shows the receiver eye mask normalized to the data rate interval with a 600 mV VTT swing. See the JESD204B specification for more information regarding the eye mask and permitted receiver eye opening. LV-OIF-11G-SR RECEIVER EYE MASK The independent SERDES PLL uses integer N techniques to achieve clock synthesis. The entire SERDES PLL is integrated on chip, including the VCO and the loop filter. The SERDES PLL VCO operates over the range of 6 GHz to 12.5 GHz. In the SERDES PLL, a VCO divider block divides the VCO clock by 2 to generate a 3 GHz to 6.25 GHz quadrature clock for the deserializer cores. This clock is the input to the clock and data recovery block that is described in the Clock and Data Recovery section. Table 19. SERDES PLL Divider Settings 525 AMPLITUDE (mV) SERDES PLL Functional Overview of the SERDES PLL Lane Rate (Gbps) 0.750 to 1.5625 1.5 to 3.125 3 to 6.25 6 to 12.5 55 0 –55 0 0.35 0.5 0.65 TIME (UI) 1.00 14379-085 –525 Figure 144. Receiver Eye Mask for 600 mV VTT Swing Rev. 0 | Page 47 of 139 PLL_REF_CLK_RATE, Register 0x084, Bits[5:4] 0b01 = 2× 0b00 = 1× 0b00 = 1× 0b00 = 1× SERDES_PLL_DIV_FAC TOR Register 0x289, Bits[1:0] 0b10 = ÷1 0b10 = ÷1 0b01 = ÷2 0b00 = ÷4 AD9161/AD9162 Data Sheet Register 0x280 controls the synthesizer enable and recalibration. To enable the SERDES PLL, first set the PLL divider register (see Table 19). Then enable the SERDES PLL by writing Register 0x280, Bit 0 = 1. If a recalibration is needed, write Register 0x280, Bit 2 = 0b1 and then reset the bit to 0b0. The rising edge of the bit causes a recalibration to begin. Confirm that the SERDES PLL is working by reading Register 0x281. If Register 0x281, Bit 0 = 1, the SERDES PLL has locked. If Register 0x281, Bit 3 = 1, the SERDES PLL was successfully calibrated. If Register 0x281, Bit 4 or Bit 5 is high, the PLL reaches the lower or upper end of its calibration band and must be recalibrated by writing 0 and then 1 to Register 0x280, Bit 2. After configuring the CDR circuit, reset it and then release the reset by writing 1 and then 0 to Register 0x206, Bit 0. Power-Down Unused PHYs Note that any unused and enabled lanes consume extra power unnecessarily. Each lane that is not being used (SERDINx±) must be powered off by writing a 1 to the corresponding bit of PHY_PD (Register 0x201). Equalization Clock and Data Recovery The deserializer is equipped with a CDR circuit. Instead of recovering the clock from the JESD204B serial lanes, the CDR recovers the clocks from the SERDES PLL. The 3 GHz to 6.25 GHz output from the SERDES PLL, shown in Figure 145, is the input to the CDR. A CDR sampling mode must be selected to generate the lane rate clock inside the device. If the desired lane rate is greater than 6.25 GHz, half rate CDR operation must be used. If the desired lane rate is less than 6.25 GHz, disable half rate operation. If the lane rate is less than 3 GHz, disable full rate and enable 2× oversampling to recover the appropriate lane rate clock. Table 20 gives a breakdown of CDR sampling settings that must be set depending on the LaneRate value. Table 20. CDR Operating Modes To compensate for signal integrity distortions for each PHY channel due to PCB trace length and impedance, the AD9161/AD9162 employ an easy to use, low power equalizer on each JESD204B channel. The AD9161/AD9162 equalizers can compensate for insertion losses far greater than required by the JESD204B specification. The equalizers have two modes of operation that are determined by the EQ_POWER_MODE register setting in Register 0x268, Bits[7:6]. In low power mode (Register 0x268, Bits[7:6] = 2b’01) and operating at the maximum lane rate of 12.5 GBPS, the equalizer can compensate for up to 11.5 dB of insertion loss. In normal mode (Register 0x268, Bits[7:6] = 2b’00), the equalizer can compensate for up to 17.2 dB of insertion loss. This performance is shown in Figure 146 as an overlay to the JESD204B specification for insertion loss. Figure 146 shows the equalization performance at 12.5 Gbps, near the maximum baud rate for the AD9161/AD9162. SPI_DIVISION_RATE, Register 0x230, Bits[2:1] 10b (divide by 4) 01b (divide by 2) 00b (no divide) 00b (no divide) SPI_ENHALFRATE Register 0x230, Bit 5 0 (full rate) 0 (full rate) 0 (full rate) 1 (half rate) MODE HALF RATE FULL RATE, NO DIV FULL RATE, DIV 2 FULL RATE, DIV 4 INTERPOLATION JESD LANES REG 0x110 DAC CLOCK (5GHz) ÷4 PCLK GENERATOR CDR OVERSAMP REG 0x289 PLL REF CLOCK VALID RANGE 35MHz TO 80MHz ÷4, ÷2, OR ÷1 DIVIDE (N) 20 40 80 160 ENABLE HALF RATE DIVISION RATE REG 0x230 SAMPLE CLOCK I, Q TO CDR VALID RANGE 3GHz TO 6.25GHz CP LF PLL_REF_CLK_RATE 1×, 2×, 4× REG 0x084 ÷2 CDR ÷N ÷8 ÷6 TO ÷127, DEFAULT: 10 Figure 145. SERDES PLL Synthesizer Block Diagram Including VCO Divider Block Rev. 0 | Page 48 of 139 JESD LANE CLOCK (SAME RATE AS PCLK) 14379-086 LaneRate (Gbps) 0.750 to 1.5625 1.5 to 3.125 3 to 6.25 6 to 12.5 The CDR circuit synchronizes the phase used to sample the data on each serial lane independently. This independent phase adjustment per serial interface ensures accurate data sampling and eases the implementation of multiple serial interfaces on a PCB. Data Sheet AD9161/AD9162 JESD204B SPEC ALLOWED CHANNEL LOSS EXAMPLE OF JESD204B COMPLIANT CHANNEL 4 INSERTION LOSS (dB) 10 EXAMPLE OF AD9161/AD9162 COMPATIBLE CHANNEL (LOW POWER MODE) AD9161/AD9162 ALLOWED CHANNEL LOSS (LOW POWER MODE) 12 AD9161/AD9162 ALLOWED CHANNEL LOSS (NORMAL MODE) 14 16 EXAMPLE OF AD9161/AD9162 COMPATIBLE CHANNEL (NORMAL MODE) 18 20 22 6.25 14379-087 24 3.125 9.375 FREQUENCY (GHz) Figure 146. Insertion Loss Allowed 0 –5 –15 –20 –25 STRIPLINE = 6" STRIPLINE = 10" STRIPLINE = 15" STRIPLINE = 20" STRIPLINE = 25" STRIPLINE = 30" –35 –40 0 1 2 3 4 5 6 7 8 9 FREQUENCY (GHz) Figure 147. Insertion Loss of 50 Ω Striplines on FR4 10 14379-088 ATTENUATION (dB) –10 –30 –15 –20 –25 6" MICROSTRIP 10" MICROSTRIP 15" MICROSTRIP 20" MICROSTRIP 25" MICROSTRIP 30" MICROSTRIP –30 –35 –40 0 1 2 3 4 5 6 7 8 9 FREQUENCY (GHz) 10 Figure 148. Insertion Loss of 50 Ω Microstrips on FR4 The data link layer of the AD9161/AD9162 JESD204B interface accepts the deserialized data from the PHYs and deframes, and descrambles them so that data octets are presented to the transport layer to be put into DAC samples. The architecture of the data link layer is shown in Figure 149. The data link layer consists of a synchronization FIFO for each lane, a crossbar switch, a deframer, and a descrambler. 6 8 –10 DATA LINK LAYER 0 2 –5 14379-089 Low power mode is recommended if the insertion loss of the JESD204B PCB channels is less than that of the most lossy supported channel for low power mode (shown in Figure 146). If the insertion loss is greater than that, but still less than that of the most lossy supported channel for normal mode (shown in Figure 146), use normal mode. At 12.5 Gbps operation, the equalizer in normal mode consumes about 4 mW more power per lane used than in low power equalizer mode. Note that either mode can be used in conjunction with transmitter preemphasis to ensure functionality and/or optimize for power. 0 ATTENUATION (dB) Figure 147 and Figure 148 are provided as points of reference for hardware designers and show the insertion loss for various lengths of well laid out stripline and microstrip transmission lines, respectively. See the Hardware Considerations section for specific layout recommendations for the JESD204B channel. The AD9161/AD9162 can operate as a single-link high speed JESD204B serial data interface. All eight lanes of the JESD204B interface handle link layer communications such as code group synchronization (CGS), frame alignment, and frame synchronization. The AD9161/AD9162 decode 8-bit/10-bit control characters, allowing marking of the start and end of the frame and alignment between serial lanes. Each AD9161/AD9162 serial interface link can issue a synchronization request by setting its SYNCOUT± signal low. The synchronization protocol follows Section 4.9 of the JESD204B standard. When a stream of four consecutive /K/ symbols is received, the AD9161/AD9162 deactivates the synchronization request by setting the SYNCOUT± signal high at the next internal LMFC rising edge. Then, AD9161/AD9162 wait for the transmitter to issue an initial lane alignment sequence (ILAS). During the ILAS, all lanes are aligned using the /A/ to /R/ character transition as described in the JESD204B Serial Link Establishment section. Elastic buffers hold early arriving lane data until the alignment character of the latest lane arrives. At this point, the buffers for all lanes are released and all lanes are aligned (see Figure 150). Rev. 0 | Page 49 of 139 AD9161/AD9162 Data Sheet DATA LINK LAYER SYNCOUTx± LANE 7 DATA CLOCK SYSREF± CROSSBAR SWITCH SERDIN7± FIFO LANE 0 OCTETS LANE 7 OCTETS SYSTEM CLOCK PHASE DETECT 14379-090 LANE 7 DESERIALIZED AND DESCRAMBLED DATA SERDIN0± FIFO DESCRAMBLE LANE 0 DATA CLOCK QUAD-BYTE DEFRAMER QBD 10-BIT/8-BIT DECODE LANE 0 DESERIALIZED AND DESCRAMBLED DATA PCLK SPI CONTROL Figure 149. Data Link Layer Block Diagram L RECEIVE LANES (EARLIEST ARRIVAL) K K K R D D D D A R Q C L RECEIVE LANES (LATEST ARRIVAL) K K K K K K K R D D C D D A R Q C D D A R D D C D D A R D D 0 CHARACTER ELASTIC BUFFER DELAY OF LATEST ARRIVAL 4 CHARACTER ELASTIC BUFFER DELAY OF EARLIEST ARRIVAL L ALIGNED RECEIVE LANES K K K K K K K R D D D D A R Q C D D A R D D 14379-091 K = K28.5 CODE GROUP SYNCHRONIZATION COMMA CHARACTER A = K28.3 LANE ALIGNMENT SYMBOL F = K28.7 FRAME ALIGNMENT SYMBOL R = K28.0 START OF MULTIFRAME Q = K28.4 START OF LINK CONFIGURATION DATA C = JESD204x LINK CONFIGURATION PARAMETERS D = Dx.y DATA SYMBOL C Figure 150. Lane Alignment During ILAS JESD204B Serial Link Establishment A brief summary of the high speed serial link establishment process for Subclass 1 is provided. See Section 5.3.3 of the JESD204B specifications document for complete details. Step 1: Code Group Synchronization Each receiver must locate /K/ (K28.5) characters in its input data stream. After four consecutive /K/ characters are detected on all link lanes, the receiver block deasserts the SYNCOUT± signal to the transmitter block at the receiver LMFC edge. The transmitter captures the change in the SYNCOUT± signal and at a future transmitter LMFC rising edge starts the ILAS. Step 2: Initial Lane Alignment Sequence The main purposes of this phase are to align all the lanes of the link and to verify the parameters of the link. Before the link is established, write each of the link parameters to the receiver device to designate how data is sent to the receiver block. The ILAS consists of four or more multiframes. The last character of each multiframe is a multiframe alignment character, /A/. The first, third, and fourth multiframes are populated with predetermined data values. Note that Section 8.2 of the JESD204B specifications document describes the data ramp that is expected during ILAS. The AD9161/AD9162 do not require this ramp. The deframer uses the final /A/ of each lane to align the ends of the multiframes within the receiver. The second multiframe contains an /R/ (K.28.0), /Q/ (K.28.4), and then data corresponding to the link parameters. Additional multiframes can be added to the ILAS if needed by the receiver. By default, the AD9161/AD9162 use four multiframes in the ILAS (this can be changed in Register 0x478). If using Subclass 1, exactly four multiframes must be used. After the last /A/ character of the last ILAS, multiframe data begins streaming. The receiver adjusts the position of the /A/ character such that it aligns with the internal LMFC of the receiver at this point. Rev. 0 | Page 50 of 139 Data Sheet AD9161/AD9162 Step 3: Data Streaming Table 21. Crossbar Registers In this phase, data is streamed from the transmitter block to the receiver block. Address 0x308 0x308 0x309 0x309 0x30A 0x30A 0x30B 0x30B Optionally, data can be scrambled. Scrambling does not start until the very first octet following the ILAS. The receiver block processes and monitors the data it receives for errors, including the following: • • • • • Bad running disparity (8-bit/10-bit error) Not in table (8-bit/10-bit error) Unexpected control character Bad ILAS Interlane skew error (through character replacement) If any of these errors exist, they are reported back to the transmitter in one of the following ways (see the JESD204B Error Monitoring section for details): • • • SYNCOUT± signal assertion: resynchronization (SYNCOUT± signal pulled low) is requested at each error for the last two errors. For the first three errors, an optional resynchronization request can be asserted when the error counter reaches a set error threshold. For the first three errors, each multiframe with an error in it causes a small pulse on SYNCOUT±. Errors can optionally trigger an interrupt request (IRQ) event, which can be sent to the transmitter. For more information about the various test modes for verifying the link integrity, see the JESD204B Test Modes section. Lane FIFO The FIFOs in front of the crossbar switch and deframer synchronize the samples sent on the high speed serial data interface with the deframer clock by adjusting the phase of the incoming data. The FIFO absorbs timing variations between the data source and the deframer; this allows up to two PCLK cycles of drift from the transmitter. The FIFO_STATUS_REG_0 register and FIFO_STATUS_REG_1 register (Register 0x30C and Register 0x30D, respectively) can be monitored to identify whether the FIFOs are full or empty. Lane FIFO IRQ An aggregate lane FIFO error bit is also available as an IRQ event. Use Register 0x020, Bit 2 to enable the FIFO error bit, and then use Register 0x024, Bit 2 to read back its status and reset the IRQ signal. See the Interrupt Request Operation section for more information. Crossbar Switch Bits [2:0] [5:3] [2:0] [5:3] [2:0] [5:3] [2:0] [5:3] Logical Lane SRC_LANE0 SRC_LANE1 SRC_LANE2 SRC_LANE3 SRC_LANE4 SRC_LANE5 SRC_LANE6 SRC_LANE7 Write each SRC_LANEy with the number (x) of the desired physical lane (SERDINx±) from which to obtain data. By default, all logical lanes use the corresponding physical lane as their data source. For example, by default, SRC_LANE0 = 0; therefore, Logical Lane 0 obtains data from Physical Lane 0 (SERDIN0±). To use SERDIN4± as the source for Logical Lane 0 instead, the user must write SRC_LANE0 = 4. Lane Inversion Register 0x334 allows inversion of desired logical lanes, which can be used to ease routing of the SERDINx± signals. For each Logical Lane x, set Bit x of Register 0x334 to 1 to invert it. Deframer The AD9161/AD9162 consist of one quad-byte deframer (QBD). The deframer accepts the 8-bit/10-bit encoded data from the deserializer (via the crossbar switch), decodes it, and descrambles it into JESD204B frames before passing it to the transport layer to be converted to DAC samples. The deframer processes four symbols (or octets) per processing clock (PCLK) cycle. The deframer uses the JESD204B parameters that the user has programmed into the register map to identify how the data is packed, and unpacks it. The JESD204B parameters are described in detail in the Transport Layer section; many of the parameters are also needed in the transport layer to convert JESD204B frames into samples. Descrambler The AD9161/AD9162 provide an optional descrambler block using a self synchronous descrambler with the following polynomial: 1 + x14 + x15. Enabling data scrambling reduces spectral peaks that are produced when the same data octets repeat from frame to frame. It also makes the spectrum data independent so that possible frequency selective effects on the electrical interface do not cause data dependent errors. Descrambling of the data is enabled by setting the SCR bit (Register 0x453, Bit 7) to 1. Register 0x308 to Register 0x30B allow arbitrary mapping of physical lanes (SERDINx±) to logical lanes used by the SERDES deframers. Rev. 0 | Page 51 of 139 AD9161/AD9162 Data Sheet Syncing LMFC Signals SYSREF+ 50Ω 50Ω SYSREF– SYSREF± Signal The SYSREF± signal is a differential source synchronous input that synchronizes the LMFC signals in both the transmitter and receiver in a JESD204B Subclass 1 system to achieve deterministic latency. The SYSREF± signal is a rising edge sensitive signal that is sampled by the device clock rising edge. It is best practice that the device clock and SYSREF± signals be generated by the same source, such as the HMC7044 clock generator, so that the phase alignment between the signals is fixed. When designing for optimum deterministic latency operation, consider the timing distribution skew of the SYSREF± signal in a multipoint link system (multichip). The AD9161/AD9162 support a periodic SYSREF± signal. The periodicity can be continuous, strobed, or gapped periodic. The SYSREF± signal can always be dc-coupled (with a commonmode voltage of 0 V to 1.25 V). When dc-coupled, a small amount of common-mode current (<500 µA) is drawn from the SYSREF± pins. See Figure 151 and Figure 152 for the SYSREF± internal circuit. To avoid this common-mode current draw, use a 50% duty cycle periodic SYSREF± signal with ac coupling capacitors. If ac-coupled, the ac coupling capacitors combine with the resistors shown in Figure 151 or Figure 152 to make a high-pass filter with an RC time constant of τ = RC. Select C such that τ > 4/SYSREF± frequency. In addition, the edge rate must be sufficiently fast to meet the SYSREF± vs. DAC clock keep out window (KOW) requirements. It is possible to use ac-coupled mode without meeting the frequency to time constant constraints (τ = RC and τ > 4/SYSREF± frequency) by using SYSREF± hysteresis (Register 0x088 and Register 0x089). However, using hystereis increases the DAC clock KOW (Table 6 does not apply) by an amount depending on the SYSREF± frequency, level of hysteresis, capacitor choice, and edge rate. 100Ω 19kΩ 3kΩ Figure 152. SYSREF± Input Circuit for the 11 mm × 11 mm 169-Ball BGA Sync Processing Modes Overview The AD9161/AD9162 support several LMFC sync processing modes. These modes are one-shot, continuous, and monitor modes. All sync processing modes perform a phase check to confirm that the LMFC is phase aligned to an alignment edge. In Subclass 1, the SYSREF± rising edge acts as the alignment edge; in Subclass 0, an internal processing clock acts as the alignment edge. The SYSREF± signal is sampled by a divide by 4 version of the DAC clock. After SYSREF± is sampled, the phase of the (DAC clock) ÷4 used to sample SYSREF± is stored in Register 0x037, Bits[7:0] and Register 0x038, Bits[3:0] as a thermometer code. This offset can be used by the SERDES data transmitter (for example, FPGA) to align multiple DACs by accounting for this clock offset when transmitting data. The sync modes are described below. See the Sync Procedure section for details on the procedure for syncing the LMFC signals. One-Shot Sync Mode (SYNCMODE = Register 0x03A, Bits[1:0] = 0b10) In one-shot sync mode, a phase check occurs on only the first alignment edge that is received after the sync machine is armed. After the phase is aligned on the first edge, the AD9161/AD9162 transition to monitor mode. Though an LMFC synchronization occurs only once, the SYSREF± signal can still be continuous. In this case, the phase is monitored and reported, but no clock phase adjustment occurs. Continuous Sync Mode (SYNCMODE = Register 0x03A, Bits[1:0] = 0b01) Continuous mode must be used in Subclass 1 only with a periodic SYSREF± signal. In continuous mode, a phase check/alignment occurs on every alignment edge. Monitor Sync Mode (SYNCMODE = Register 0x03A, Bits[1:0]) = 0b00) 200Ω 14379-092 SYSREF– 19kΩ Continuous mode differs from one-shot mode in two ways. First, no SPI cycle is required to arm the device; the alignment edge seen after continuous mode is enabled results in a phase check. Second, a phase check occurs on every alignment edge in continuous mode. 200Ω SYSREF+ 3kΩ 14379-147 The first step in guaranteeing synchronization across links and devices begins with syncing the LMFC signals. In Subclass 0, the LMFC signal is synchronized to an internal processing clock. In Subclass 1, LMFC signals are synchronized to an external SYSREF± signal. Figure 151. SYSREF± Input Circuit for the 8 mm × 8 mm 165-Ball BGA In monitor mode, the user can monitor the phase error in real time. Use this sync mode with a periodic SYSREF± signal. The phase is monitored and reported, but no clock phase adjustment occurs. Rev. 0 | Page 52 of 139 Data Sheet AD9161/AD9162 When an alignment request (SYSREF± edge) occurs, snapshots of the last phase error are placed into readable registers for reference (Register 0x037 and Register 0x038, Bits[3:0]), and the IRQ_SYSREF_JITTER interrupt is set, if appropriate. Sync Procedure The procedure for enabling the sync is as follows: 1. 2. 3. 4. 5. 6. 7. 8. 9. Set up the DAC; the SERDES PLL locks it, and enables the CDR (see the Start-Up Sequence section). Set Register 0x039 (SYSREF± jitter window). A minimum of 4 DAC clock cycles is recommended. See Table 23 for settings. Optionally, read back the SYSREF± count to check whether the SYSREF± pulses are being received. a. Set Register 0x036 = 0. Writing anything to SYSREF_COUNT resets the count. b. Set Register 0x034 = 0. Writing anything to SYNC_LMFC_STAT0 saves the data for readback and registers the count. c. Read SYSREF_COUNT from the value from Register 0x036. Perform a one-shot sync. a. Set Register 0x03A = 0x00. Clear one-shot mode if already enabled. b. Set Register 0x03A = 0x02. Enable one-shot sync mode. The state machine enters monitor mode after a sync occurs. Optionally, read back the sync SYNC_LMFC_STATx registers to verify that sync completed correctly. a. Set Register 0x034 = 0. Register 0x034 must be written to read the value. b. Read Register 0x035 and Register 0x034 to find the value of SYNC_LMFC_STATx. It is recommended to set SYNC_LMFC_STATx to 0 but it can be set to 4, or a LMFC period in DAC clocks – 4, due to jitter. Optionally, read back the sync SYSREF_PHASEx register to identify which phase of the divide by 4 was used to sample SYSREF±. Read Register 0x038 and Register 0x037 as thermometer code. The MSBs of Register 0x037, Bits[7:4], normally show the thermometer code value. Turn the link on (Register 0x300, Bit 0 = 1). Read back Register 0x302 (dynamic link latency). Repeat the reestablishment of the link several times (Step 1 to Step 7) and note the dynamic link latency values. Based on the values, program the LMFC delay (Register 0x304) and the LMFC variable (Register 0x306), and then restart the link. Table 22. Sync Processing Modes Sync Processing Mode No synchronization One shot Continuous Table 23. SYSREF± Jitter Window Tolerance SYSREF± Jitter Window Tolerance (DAC Clock Cycles) ±½ ±4 ±8 ±12 ±16 +20 +24 +28 1 SYSREF_JITTER_WINDOW (Register 0x039, Bits[5:0])1 0x00 0x04 0x08 0x0C 0x10 0x14 0x18 0x1C The two least significant digits are ignored because the SYSREF± signal is sampled with a divide by 4 version of the DAC clock. As a result, the jitter window is set by this divide by 4 clock rather than the DAC clock. It is recommended that at least a four-DAC clock SYSREF± jitter window be chosen. Deterministic Latency JESD204B systems contain various clock domains distributed throughout its system. Data traversing from one clock domain to a different clock domain can lead to ambiguous delays in the JESD204B link. These ambiguities lead to nonrepeatable latencies across the link from power cycle to power cycle with each new link establishment. Section 6 of the JESD204B specification addresses the issue of deterministic latency with mechanisms defined as Subclass 1 and Subclass 2. The AD9161/AD9162 support JESD204B Subclass 0 and Subclass 1 operation, but not Subclass 2. Write the subclass to Register 0x458, Bits[7:5]. Subclass 0 This mode gives deterministic latency to within 32 DAC clock cycles. It does not require any signal on the SYSREF± pins, which can be left disconnected. Subclass 0 still requires that all lanes arrive within the same LMFC cycle and the dual DACs must be synchronized to each other. Subclass 1 This mode gives deterministic latency and allows the link to be synced to within four DAC clock periods. It requires an external SYSREF± signal that is accurately phase aligned to the DAC clock. Deterministic Latency Requirements Several key factors are required for achieving deterministic latency in a JESD204B Subclass 1 system. • • • SYNC_MODE (Register 0x03A, Bits[1:0]) 0b00 0b10 0b01 Rev. 0 | Page 53 of 139 SYSREF± signal distribution skew within the system must be less than the desired uncertainty. SYSREF± setup and hold time requirements must be met for each device in the system. The total latency variation across all lanes, links, and devices must be ≤10 PCLK periods, which includes both variable delays and the variation in fixed delays from lane to lane, link to link, and device to device in the system. AD9161/AD9162 Data Sheet LINK DELAY = DELAYFIXED + DELAYVARIABLE LOGIC DEVICE (JESD204B Tx) CHANNEL JESD204B Rx DSP DAC POWER CYCLE VARIANCE LMFC ILAS DATA ALIGNED DATA AT Rx OUTPUT ILAS DATA FIXED DELAY VARIABLE DELAY 14379-095 DATA AT Tx INPUT Figure 153. JESD204B Link Delay = Fixed Delay + Variable Delay Link Delay The link delay of a JESD204B system is the sum of the fixed and variable delays from the transmitter, channel, and receiver as shown in Figure 153. For proper functioning, all lanes on a link must be read during the same LMFC period. Section 6.1 of the JESD204B specification states that the LMFC period must be larger than the maximum link delay. For the AD9161/AD9162, this is not necessarily the case; instead, the AD9161/AD9162 use a local LMFC for each link (LMFCRx) that can be delayed from the SYSREF± aligned LMFC. Because the LMFC is periodic, this delay can account for any amount of fixed delay. As a result, the LMFC period must only be larger than the variation in the link delays, and the AD9161/AD9162 can achieve proper performance with a smaller total latency. Figure 154 and Figure 155 show a case where the link delay is greater than an LMFC period. Note that it can be accommodated by delaying LMFCRx. POWER CYCLE VARIANCE DATA 14379-093 ILAS EARLY ARRIVING LMFC REFERENCE LATE ARRIVING LMFC REFERENCE Figure 154. Link Delay > LMFC Period Example POWER CYCLE VARIANCE LMFC ALIGNED DATA The RBD described in the JESD204B specification takes values from 1 frame clock cycle to K frame clock cycles, and the RBD of the AD9161/AD9162 takes values from 0 PCLK cycle to 10 PCLK cycles. As a result, up to 10 PCLK cycles of total delay variation can be absorbed. LMFCVar and LMFCDel are both in PCLK cycles. The PCLK factor, or number of frame clock cycles per PCLK cycle, is equal to 4/F. For more information on this relationship, see the Clock Relationships section. Two examples follow that show how to determine LMFCVar and LMFCDel. After they are calculated, write LMFCDel into Register 0x304 for all devices in the system, and write LMFCVar to Register 0x306 for all devices in the system. Link Delay Setup Example, With Known Delays LMFC ALIGNED DATA Setting LMFCDel appropriately ensures that all the corresponding data samples arrive in the same LMFC period. Then LMFCVar is written into the receive buffer delay (RBD) to absorb all link delay variation. This write ensures that all data samples have arrived before reading. By setting these to fixed values across runs and devices, deterministic latency is achieved. ILAS DATA LMFC_DELAY LMFC REFERENCE FOR ALL POWER CYCLES FRAME CLOCK Figure 155. LMFC_DELAY_x to Compensate for Link Delay > LMFC 14379-094 LMFCRX All the known system delays can be used to calculate LMFCVar and LMFCDel. The example shown in Figure 156 is demonstrated in the following steps. Note that this example is in Subclass 1 to achieve deterministic latency, which has a PCLK factor (4/F) of 2 frame clock cycles per PCLK cycle, and uses K = 32 (frames/multiframe). Because PCBFixed << PCLK Period, PCBFixed is negligible in this example and not included in the calculations. 1. 2. The method to select the LMFCDel (Register 0x304) and LMFCVar (Register 0x306) variables is described in the Link Delay Setup Example, With Known Delays section. Rev. 0 | Page 54 of 139 Find the receiver delays using Table 8. RxFixed = 17 PCLK cycles RxVar = 2 PCLK cycles Find the transmitter delays. The equivalent table in the example JESD204B core (implemented on a GTH or GTX gigabit transceiver on a Virtex-6 FPGA) states that the delay is 56 ± 2 byte clock cycles. Data Sheet 4. 5. 6. 7. Because the PCLK Rate = ByteRate/4 as described in the Clock Relationships section, the transmitter delays in PCLK cycles are calculated as follows: TxFixed = 54/4 = 13.5 PCLK cycles TxVar = 4/4 = 1 PCLK cycle Calculate MinDelayLane as follows: MinDelayLane = floor(RxFixed + TxFixed + PCBFixed) = floor(17 + 13.5 + 0) = floor(30.5) MinDelayLane = 30 Calculate MaxDelayLane as follows: MaxDelayLane = ceiling(RxFixed + RxVar + TxFixed + TxVar + PCBFixed)) = ceiling(17 + 2 + 13.5 + 1 + 0) = ceiling(33.5) MaxDelayLane = 34 Calculate LMFCVar as follows: LMFCVar = (MaxDelay + 1) − (MinDelay − 1) = (34 + 1) − (30 − 1) = 35 − 29 LMFCVar = 6 PCLK cycles Calculate LMFCDel as follows: LMFCDel = ((MinDelay − 1) % K = ((30 − 1)) % 32 = (29) % 32 = 29 % 32 LMFCDel = 29 PCLK cycles Write LMFCDel to Register 0x304 for all devices in the system. Write LMFCVar to Register 0x306 for all devices in the system. Link Delay Setup Example, Without Known Delay If the system delays are not known, the AD9161/AD9162 can read back the link latency between LMFCRX for each link and the SYSREF± aligned LMFC. This information is then used to calculate LMFCVar and LMFCDel. Figure 158 shows how DYN_LINK_LATENCY_0 (Register 0x302) provides a readback showing the delay (in PCLK cycles) between LMFCRX and the transition from ILAS to the first data sample. By repeatedly power cycling and taking this measurement, the minimum and maximum delays across power cycles can be determined and used to calculate LMFCVar and LMFCDel. In Figure 158, for Link A, Link B, and Link C, the system containing the AD9161/AD9162 (including the transmitter) is power cycled and configured 20 times. The AD9161/AD9162 are configured as described in the Sync Procedure section. Because the purpose of this exercise is to determine LMFCDel and LMFCVar, the LMFCDel value is programmed to 0 and the DYN_LINK_LATENCY_0 value is read from Register 0x302. The variation in the link latency over the 20 runs is shown in Figure 158 in gray, described as follows: Link A gives readbacks of 6, 7, 0, and 1. Note that the set of recorded delay values rolls over the edge of a multiframe at the boundary of K/ PCLK Factor = 8. Add the number of PCLK cycles per multiframe (PCLKsPerMF) = 8 to the readback values of 0 and 1 because they rolled over the edge of the multiframe. Delay values range from 6 to 9. Link B gives delay values from 5 to 7. Link C gives delay values from 4 to 7. LMFC PCLK FRAME CLOCK DATA AT Tx FRAMER ALIGNED LANE DATA AT Rx DEFRAMER OUTPUT ILAS DATA ILAS Tx VAR DELAY Rx VAR DELAY DATA PCB FIXED DELAY LMFCRX LMFC DELAY = 26 FRAME CLOCK CYCLES TOTAL FIXED LATENCY = 30 PCLK CYCLES Figure 156. LMFC Delay Calculation Example Rev. 0 | Page 55 of 139 TOTAL VARIABLE LATENCY = 4 PCLK CYCLES 14379-096 3. AD9161/AD9162 AD9161/AD9162 Data Sheet 4. The example shown in Figure 158 is demonstrated in the following steps. Note that this example is in Subclass 1 to achieve deterministic latency, which has a PCLK Factor (FrameRate ÷ PCLK Rate) of 2 and uses K = 16; therefore PCLKsPerMF = 8. This example is a hypothetical example used to illustrate the procedure, because K is always 32 on the AD9161/AD9162. 3. SYSREF± LMFCRX ILAS ALIGNED DATA DATA DYN_LINK_LATENCY Figure 157. DYN_LINK_LATENCY_x Illustration LMFC PCLK FRAME CLOCK DYN_LINK_LATENCY_CNT 0 1 2 ALIGNED DATA (LINK A) ALIGNED DATA (LINK B) ALIGNED DATA (LINK C) 3 4 5 6 7 0 1 2 3 ILAS 4 5 6 7 DATA DATA ILAS ILAS DATA LMFCRX DETERMINISTICALLY DELAYED DATA ILAS LMFC_DELAY = 6 (FRAME CLOCK CYCLES) DATA LMFC_VAR = 7 (PCLK CYCLES) Figure 158. Multilink Synchronization Settings, Derived Method Example Rev. 0 | Page 56 of 139 14379-098 2. Calculate the minimum of all delay measurements across all power cycles, links, and devices as follows: MinDelay = min(all Delay values) = 4 Calculate the maximum of all delay measurements across all power cycles, links, and devices as follows: MaxDelay = max(all Delay values) = 9 Calculate the total delay variation (with guard band) across all power cycles, links, and devices as follows: LMFCVar = (MaxDelay + 1) − (MinDelay − 1) = (9 + 1) − (4 − 1) = 10 − 3 = 7 PCLK cycles 14379-097 1. 5. Calculate the minimum delay in PCLK cycles (with guard band) across all power cycles, links, and devices as follows: LMFCDel = ((MinDelay − 1) × PCLK Factor) % K = ((4 − 1)) % 16 = (3) % 16 = 3 % 16 = 3 PCLK cycles Write LMFCDel to Register 0x304 for all devices in the system. Write LMFCVar to Register 0x306 for all devices in the system. Data Sheet AD9161/AD9162 TRANSPORT LAYER TRANSPORT LAYER (QBD) LANE 0 OCTETS DAC A_I0[15:0] DELAY BUFFER 0 F2S_0 DAC A_Q0[15:0] LANE 3 OCTETS PCLK_0 SPI CONTROL LANE 4 OCTETS DAC B_I0[15:0] PCLK_0 TO PCLK_1 FIFO DELAY BUFFER 1 F2S_1 DAC B_Q0[15:0] LANE 7 OCTETS 14379-099 PCLK_1 SPI CONTROL Figure 159. Transport Layer Block Diagram The transport layer receives the descrambled JESD204B frames and converts them to DAC samples based on the programmed JESD204B parameters shown in Table 24. The device parameters are defined in Table 25. Table 24. JESD204B Transport Layer Parameters Parameter F K L M S Description Number of octets per frame per lane: 1, 2, or 4 Number of frames per multiframe: K = 32 Number of lanes per converter device (per link), as follows: 4, or 8 Number of converters per device (per link), as follows: 1 or 2 (2 is used for complex data interface) Number of samples per converter, per frame: 1 or 2 Table 25. JESD204B Device Parameters Parameter CF CS HD N Nʹ (or NP) Description Number of control words per device clock per link. Not supported, must be 0. Number of control bits per conversion sample. Not supported, must be 0. High density user data format. Used when samples must be split across lanes. Set to 1 when F = 1, otherwise 0. Converter resolution = 16. Total number of bits per sample = 16. Certain combinations of these parameters are supported by the AD9161/AD9162. See Table 28 for a list of supported interpolation rates and the number of lanes that is supported for each rate. Table 28 lists the JESD204B parameters for each of the interpolation and number of lanes configuration, and gives an example lane rate for a 5 GHz DAC clock. Table 27 lists JESD204B parameters that have fixed values. A value of yes in Table 26 means the interpolation rate is supported for the number of lanes. A blank cell means it is not supported. Table 26. Interpolation Rates and Number of Lanes Interpolation 1× 2× 3× 4× 6× 8× 12× 16× 24× 1 8 Yes1 Yes Yes Yes Yes Yes Yes Yes Yes 6 4 3 2 1 Yes1 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes1 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes These modes restrict the maximum DAC clock rate to 5 GHz. Table 27. JESD204B Parameters with Fixed Values Parameter K N NP CF HD CS Rev. 0 | Page 57 of 139 Value 32 16 16 0 1 0 AD9161/AD9162 Data Sheet Table 28. JESD204B Parameters for Interpolation Rate and Number of Lanes Interpolation Rate 1 2 2 3 3 4 4 4 4 6 6 6 6 8 8 8 8 8 12 12 12 12 12 16 16 16 16 16 16 24 24 24 24 24 24 1 No. of Lanes 8 6 8 6 8 3 4 6 8 3 4 6 8 2 3 4 6 8 2 3 4 6 8 1 2 3 4 6 8 1 2 3 4 6 8 M 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 F 1 2 1 2 1 4 1 2 1 4 1 2 1 2 4 1 2 1 2 4 1 2 1 4 2 4 1 2 1 4 2 4 1 2 1 S 4 3 2 3 2 3 1 3 2 3 1 3 2 1 3 1 3 2 1 3 1 3 2 1 1 3 1 3 2 1 1 3 1 3 2 PCLK Period (DAC Clocks) 16 12 16 18 24 12 16 24 32 18 24 36 48 16 24 32 48 64 24 36 48 72 96 16 32 48 64 96 128 24 48 72 96 144 192 LMFC Period (DAC Clocks) 128 192 128 288 192 384 128 384 256 576 192 576 384 256 768 256 768 512 384 1152 384 1152 768 512 512 1536 512 1536 1024 768 768 2304 768 2304 1536 Maximum lane rate is 12.5 GHz. These modes must be run with the DAC rate below 3.75 GHz. Rev. 0 | Page 58 of 139 Lane Rate at 5 GHz DAC Clock (GHz) 12.5 16.661 12.5 11.11 8.33 16.661 12.5 8.33 6.25 11.11 8.33 5.55 4.16 12.5 8.33 6.25 4.16 3.12 8.33 5.55 4.16 2.77 2.08 12.5 6.25 4.16 3.12 2.08 1.56 8.33 4.16 2.77 2.08 1.38 1.04 Data Sheet AD9161/AD9162 Configuration Parameters The AD9161/AD9162 modes refer to the link configuration parameters for L, K, M, N, NP, S, and F. Table 29 provides the description and addresses for these settings. Table 29. Configuration Parameters JESD204B Setting L−1 F−1 Description Number of lanes − 1. M−1 Number of ((octets per frame) per lane) − 1. Number of frames per multiframe − 1. Number of converters − 1. N−1 Converter bit resolution − 1. NP − 1 Bit packing per sample − 1. S−1 Number of ((samples per converter) per frame) − 1. High density format. Set to 1 if F = 1. Leave at 0 if F ≠ 1. Device ID. Match the Device ID sent by the transmitter. Bank ID. Match the Bank ID sent by the transmitter. Lane ID for Lane 0. Match the Lane ID sent by the transmitter on Logical Lane 0. JESD204x version. Match the version sent by the transmitter (0x0 = JESD204A, 0x1 = JESD204B). K−1 HD DID BID LID0 JESDV The error counts for failing lanes are reported for one JESD204B lane at a time. The process for performing PRBS testing on the AD9161/AD9162 is as follows: 1. 2. Address Register 0x453, Bits[4:0] Register 0x454, Bits[7:0] Register 0x455, Bits[4:0] Register 0x456, Bits[7:0] Register 0x457, Bits[4:0] Register 0x458, Bits[4:0] Register 0x459, Bits[4:0] Register 0x45A, Bit 7 Register 0x450, Bits[7:0] Register 0x451, Bits[7:0] Register 0x452, Bits[4:0] 3. 4. 5. 6. 7. 8. 9. Register 0x459, Bits[7:5] Data Flow Through the JESD204B Receiver The link configuration parameters determine how the serial bits on the JESD204B receiver interface are deframed and passed on to the DACs as data samples. Deskewing and Enabling Logical Lanes After proper configuration, the logical lanes are automatically deskewed. All logical lanes are enabled or not based on the lane number setting in Register 0x110, Bits[7:4]. The physical lanes are all powered up by default. To disable power to physical lanes that are not being used, set Bit x in Register 0x201 to 1 to disable Physical Lane x, and keep it at 0 to enable it. JESD204B TEST MODES PHY PRBS Testing The JESD204B receiver on the AD9161/AD9162 includes a PRBS pattern checker on the back end of its physical layer. This functionality enables bit error rate (BER) testing of each physical lane of the JESD204B link. The PHY PRBS pattern checker does not require that the JESD204B link be established. It can synchronize with a PRBS7, PRBS15, or PRBS31 data pattern. PRBS pattern verification can be done on multiple lanes at once. Start sending a PRBS7, PRBS15, or PRBS31 pattern from the JESD204B transmitter. Select and write the appropriate PRBS pattern to Register 0x316, Bits[3:2], as shown in Table 30. Enable the PHY test for all lanes being tested by writing to PHY_TEST_EN (Register 0x315). Each bit of Register 0x315 enables the PRBS test for the corresponding lane. For example, writing a 1 to Bit 0 enables the PRBS test for Physical Lane 0. Toggle PHY_TEST_RESET (Register 0x316, Bit 0) from 0 to 1 then back to 0. Set PHY_PRBS_TEST_THRESHOLD_xBITS (Bits[23:0], Register 0x319 to Register 0x317) as desired. Write a 0 and then a 1 to PHY_TEST_START (Register 0x316, Bit 1). The rising edge of PHY_TEST_START starts the test. a. (Optional) In some cases, it may be necessary to repeat Step 4 at this point. Toggle PHY_TEST_RESET (Register 0x316, Bit 0) from 0 to 1, then back to 0. Wait 500 ms. Stop the test by writing PHY_TEST_START (Register 0x316, Bit 1) = 0. Read the PRBS test results. a. Each bit of PHY_PRBS_PASS (Register 0x31D) corresponds to one SERDES lane (0 = fail, 1 = pass). b. The number of PRBS errors seen on each failing lane can be read by writing the lane number to check (0 to 7) in PHY_SRC_ERR_CNT (Register 0x316, Bits[6:4]) and reading the PHY_PRBS_ERR_COUNT (Register 0x31C to Register 0x31A). The maximum error count is 224 − 1. If all bits of Register 0x31C to Register 0x31A are high, the maximum error count on the selected lane is exceeded. Table 30. PHY PRBS Pattern Selection PHY_PRBS_PAT_SEL Setting (Register 0x316, Bits[3:2]) 0b00 (default) 0b01 0b10 PRBS Pattern PRBS7 PRBS15 PRBS31 Transport Layer Testing The JESD204B receiver in the AD9161/AD9162 supports the short transport layer (STPL) test as described in the JESD204B standard. This test can be used to verify the data mapping between the JESD204B transmitter and receiver. To perform this test, this function must be implemented in the logic device and enabled there. Before running the test on the receiver side, the link must be established and running without errors. The STPL test ensures that each sample from each converter is mapped appropriately according to the number of converters (M) and the number of samples per converter (S). As specified in the JESD204B standard, the converter manufacturer specifies Rev. 0 | Page 59 of 139 AD9161/AD9162 Data Sheet what test samples are transmitted. Each sample must have a unique value. For example, if M = 2 and S = 2, four unique samples are transmitted repeatedly until the test is stopped. The expected sample must be programmed into the device and the expected sample is compared to the received sample one sample at a time until all are tested. The process for performing this test on the AD9161/AD9162 is described as follows: 1. 2. 3. 4. 5. 6. Synchronize the JESD204B link. Enable the STPL test at the JESD204B Tx. Depending on JESD204B case, there may be up to two DACs, and each frame may contain up to four DAC samples. Configure the SHORT_TPL_REF_SP_MSB bits (Register 0x32E) and SHORT_TPL_REF_SP_LSB bits (Register 0x32D) to match one of the samples for one converter within one frame. Set SHORT_TPL_SP_SEL (Register 0x32C, Bits[7:4]) to select the sample within one frame for the selected converter according to Table 31. Set SHORT_TPL_TEST_EN (Register 0x32C, Bit 0) to 1. Set SHORT_TPL_TEST_RESET (Register 0x32C, Bit 1) to 1, then back to 0. 7. 8. 9. Wait for the desired time. The desired time is calculated as 1/(sample rate × BER). For example, given a bit error rate of BER = 1 × 10−10 and a sample rate = 1 GSPS, the desired time = 10 sec. Then, set SHORT_TPL_TEST_EN to 0. Read the test result at SHORT_TPL_FAIL (Register 0x32F, Bit 0). Choose another sample for the same or another converter to continue with the test, until all samples for both converters from one frame are verified. (Note that the converter count is M = 2 for all interpolator modes on the AD9161/AD9162 to enable complex signal processing.) Consult Table 31 for a guide to the test sample alignment. Note that the sample order for 1×, eight-lane mode has Sample 1 and Sample 2 swapped. Also, the STPL test for the three-lane and six-lane options is not functional and always fails. Table 31. Short TPL Test Samples Assignment1 JESD204x Mode 1× Eight-Lane (L = 8, M = 1, F = 1, S = 4) Required Samples from JESD204x Tx Send four samples: M0S0, M0S1, M0S2, M0S3, and repeat 2× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 3× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 4× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 6× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 8× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 12× Eight-Lane e (L = 8, M = 2, F = 1, S = 2) 16× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 24× Eight-Lane (L = 8, M = 2, F = 1, S = 2) 2× Six-Lane (L = 6, M = 2, F = 2, S = 3) 3× Six-Lane (L = 6, M = 2, F = 2, S = 3) 4× Six-Lane (L = 6, M = 2, F = 2, S = 3) 6× Six-Lane (L = 6, M = 2, F = 2, S = 3) 8× Six-Lane (L = 6, M = 2, F = 2, S = 3) 12× Six-Lane (L = 6, M = 2, F = 2, S = 3) 16× Six-Lane (L = 6, M = 2, F = 2, S = 3) 24× Six-Lane (L = 6, M = 2, F = 2, S = 3) 4× Six-Lane (L = 3, M = 2, F = 4, S = 3) 6× Three-Lane (L = 3, M = 2, F = 4, S = 3) 8× Three-Lane (L = 3, M = 2, F = 4, S = 3) 12× Three-Lane (L = 3, M = 2, F = 4, S = 3) 16× Three-Lane (L = 3, M = 2, F = 4, S = 3) 24× Three-Lane (L = 3, M = 2, F = 4, S = 3) Send four samples: M0S0, M0S1, M1S0, M1S1, and repeat Send six samples: M0S0, M0S1, M0S2, M1S0, M1S1, M1S2, and repeat Rev. 0 | Page 60 of 139 Samples Assignment SP0: M0S0, SP4: M0S0, SP8: M0S0, SP12: M0S0 SP1: M0S2, SP5: M0S2, SP9: M0S2, SP13: M0S2 SP2: M0S1, SP6: M0S1, SP10: M0S1, SP14: M0S1 SP3: M0S3, SP7: M0S3, SP11: M0S3, SP15: M0S3 SP0: M0S0, SP4: M0S0, SP8: M0S0, SP12: M0S0 SP1: M1S0, SP5: M1S0, SP9: M1S0, SP13: M1S0 SP2: M0S1, SP6: M0S1, SP10: M0S1, SP14: M0S1 SP3: M1S1, SP7: M1S1, SP11: M1S1, SP15: M1S1 Test hardware is not functional; STPL always fails Data Sheet JESD204x Mode 4× Four-Lane (L = 4, M = 2, F = 1, S = 1) 6× Four-Lane (L = 4, M = 2, F = 1, S = 1) 8× Four-Lane (L = 4, M = 2, F = 1, S = 1) 12× Four-Lane (L = 4, M = 2, F = 1, S = 1) 16× Four-Lane (L = 4, M = 2, F = 1, S = 1) 24× Four-Lane (L = 4, M = 2, F = 1, S = 1) 8× Two-Lane (L = 2, M = 2, F = 2, S = 1) 12× Two-Lane (L = 2, M = 2, F = 2, S = 1) 16× Two-Lane (L = 2, M = 2, F = 2, S = 1) 24× Two-Lane (L = 2, M = 2, F = 2, S = 1) 16× One-Lane (L = 1, M = 2, F = 4, S = 1) 24× One-Lane (L = 1, M = 2, F = 4, S = 1) 1 AD9161/AD9162 Required Samples from JESD204x Tx Send two samples: M0S0, M1S0, repeat Samples Assignment SP0: M0S0, SP4: M0S0, SP8: M0S0, SP12: M0S0 SP1: M1S0, SP5: M1S0, SP9: M1S0, SP13: M1S0 SP2: M0S0, SP6: M0S0, SP10: M0S0, SP14: M0S0 SP3: M1S0, SP7: M1S0, SP11: M1S0, SP15: M1S0 Mx is the converter number and Sy is the sample number. For example, M0S0 means Converter 0, Sample 0. SPx is the sample pattern word number. For example, SP0 means Sample Pattern Word 0. Repeated CGS and ILAS Test As per Section 5.3.3.8.2 of the JESD204B specification, the AD9161/AD9162 can check that a constant stream of /K28.5/ characters is being received, or that CGS followed by a constant stream of ILAS is being received. To run a repeated CGS test, send a constant stream of /K28.5/ characters to the AD9161/AD9162 SERDES inputs. Next, set up the device and enable the links. Ensure that the /K28.5/ characters are being received by verifying that SYNCOUT± is deasserted and that CGS has passed for all enabled link lanes by reading Register 0x470. To run the CGS followed by a repeated ILAS sequence test, follow the procedure to set up the links, but before performing the last write (enabling the links), enable the ILAS test mode by writing a 1 to Register 0x477, Bit 7. Then, enable the links. When the device recognizes four CGS characters on each lane, it deasserts the SYNCOUT±. At this point, the transmitter starts sending a repeated ILAS sequence. Read Register 0x473 to verify that initial lane synchronization has passed for all enabled link lanes. values of the bad disparity error (BDE) count register is 1. Reporting of disparity errors that occur at the same character position of an NIT error is disabled. No such disabling is performed for the disparity errors in the characters after an NIT error. Therefore, it is expected behavior that an NIT error may result in a BDE error. A resync is triggered when four NIT errors are injected with Register 0x476, Bit 4 = 1. When this bit is set, the error counter does not distinguish between a concurrent invalid symbol with the wrong running disparity but is in the 8-bit/10-bit decoding table, and a NIT error. Thus, a resync can be triggered when four NIT errors are injected because they are not distinguished from disparity errors. Checking Error Counts The error count can be checked for disparity errors, NIT errors, and unexpected control character errors. The error counts are on a per lane and per error type basis. Each error type and lane has a register dedicated to it. To check the error count, the following steps must be performed: 1. JESD204B ERROR MONITORING Disparity, Not in Table, and Unexpected Control (K) Character Errors As per Section 7.6 of the JESD204B specification, the AD9161/ AD9162 can detect disparity errors, not in table (NIT) errors, and unexpected control character errors, and can optionally issue a sync request and reinitialize the link when errors occur. Note that the disparity error counter counts all characters with invalid disparity, regardless of whether they are in the 8-bit/10-bit decoding table. This is a minor deviation from the JESD204B specification, which only counts disparity errors when they are in the 8-bit/10-bit decoding table. 2. 3. Several other interpretations of the JESD204B specification are noted in this section. When three NIT errors are injected to one lane and QUAL_RDERR (Register 0x476, Bit 4) = 1, the readback Rev. 0 | Page 61 of 139 Choose and enable which errors to monitor by selecting them in Register 0x480, Bits[5:3] to Register 0x487, Bits[5:3]. Unexpected K (UEK) character, BDE, and NIT error monitoring can be selected for each lane by writing a 1 to the appropriate bit, as described in the register map. These bits are enabled by default. The corresponding error counter reset bits are in Register 0x480, Bits[2:0] to Register 0x487, Bits[2:0]. Write a 0 to the corresponding bit to reset that error counter. Registers 0x488, Bits[2:0] to Register 0x48F, Bits[2:0] have the terminal count hold indicator for each error counter. If this flag is enabled, when the terminal error count of 0xFF is reached, the counter ceases counting and holds that value until reset. Otherwise, it wraps to 0x00 and continues counting. Select the desired behavior and program the corresponding register bits per lane. AD9161/AD9162 Data Sheet Table 32. Setting SYNCOUT± Error Pulse Duration Check for Error Count Over Threshold To check for the error count over threshold, follow these steps: 1. 2. Define the error counter threshold. The error counter threshold can be set to a user defined value in Register 0x47C, or left to the default value of 0xFF. When the error threshold is reached, an IRQ is generated or SYNCOUT± is asserted or both, depending on the mask register settings. This one error threshold is used for all three types of errors (UEK, NIT, and BDE). Set the SYNC_ASSERT_MASK bits. The SYNCOUT± assertion behavior is set in Register 0x47D, Bits[2:0]. By default, when any error counter of any lane is equal to the threshold, it asserts SYNCOUT± (Register 0x47D, Bits[2:0] = 0b111). Read the error count reached indicator. Each error counter has a terminal count reached indicator, per lane. This indicator is set to 1 when the terminal count of an error counter for a particular lane has been reached. These status bits are located in Register 0x490, Bits[2:0] to Register 0x497, Bits[2:0]. These registers also indicate whether a particular lane is active by setting Bit 3 = 0b1. Error Counter and IRQ Control For error counter and IRQ control, follow these steps: 1. 2. Enable the interrupts. Enable the JESD204B interrupts. The interrupts for the UEK, NIT, and BDE error counters are in Register 0x4B8, Bits[7:5]. There are other interrupts to monitor when bringing up the link, such as lane deskewing, initial lane sync, good check sum, frame sync, code group sync (Register 0x4B8, Bits[4:0], and configuration mismatch (Register 0x4B9, Bit 0). These bits are off by default but can be enabled by writing 0b1 to the corresponding bit. Read the JESD204B interrupt status. The interrupt status bits are in Register 0x4BA, Bits[7:0] and Register 0x4BB, Bit 0, with the status bit position corresponding to the enable bit position. It is recommended to enable all interrupts that are planned to be used prior to bringing up the JESD204B link. When the link is up, the interrupts can be reset and then used to monitor the link status. F 1 2 4 1 PCLK Factor (Frames/PCLK) 4 2 1 SYNC_ERR_DUR (Register 0x312, Bits[7:4]) Setting1 0 (default) 1 2 These register settings assert the SYNCOUT± signal for two frame clock cycle pulse widths. Unexpected Control Character, NIT, Disparity IRQs For UEK character, NIT, and disparity errors, error count over the threshold events are available as IRQ events. Enable these events by writing to Register 0x4B8, Bits[7:5]. The IRQ event status can be read at Register 0x4BA, Bits[7:5] after the IRQs are enabled. See the Error Counter and IRQ Control section for information on resetting the IRQ. See the Interrupt Request Operation section for more information on IRQs. Errors Requiring Reinitializing A link reinitialization automatically occurs when four invalid disparity characters are received as per Section 7.1 of the JESD204B specification. When a link reinitialization occurs, the resync request is five frames and nine octets long. The user can optionally reinitialize the link when the error count for disparity errors, NIT errors, or UEK character errors reaches a programmable error threshold. The process to enable the reinitialization feature for certain error types is as follows: 1. 2. 3. 4. Monitoring Errors via SYNCOUT± When one or more disparity, NIT, or unexpected control character errors occur, the error is reported on the SYNCOUT± pin as per Section 7.6 of the JESD204B specification. The JESD204B specification states that the SYNCOUT± signal is asserted for exactly two frame periods when an error occurs. For the AD9161/AD9162, the width of theSYNCOUT± pulse can be programmed to ½, 1, or 2 PCLK cycles. The settings to achieve a SYNCOUT± pulse of two frame clock cycles are given in Table 32. Rev. 0 | Page 62 of 139 Choose and enable which errors to monitor by selecting them in Register 0x480, Bits[5:3] to Register 0x487, Bits[5:3]. UEK, BDE, and NIT error monitoring can be selected for each lane by writing a 1 to the appropriate bit, as described in Table 47. These are enabled by default. Enable the sync assertion mask for each type of error by writing to SYNC_ASSERT_MASK (Register 0x47D, Bits[2:0]) according to Table 33. Program the desired error counter threshold into ERRORTHRES (Register 0x47C). For each error type enabled in the SYNC_ASSERT_MASK register, if the error counter on any lane reaches the programmed threshold, SYNCOUT± falls, issuing a sync request. Note that all error counts are reset when a link reinitialization occurs. The IRQ does not reset and must be reset manually. Data Sheet AD9161/AD9162 Table 33. Sync Assertion Mask (SYNC_ASSERT_MASK) CGS, Frame Sync, Checksum, and ILAS IRQs Addr. 0x47D Fail signals for CGS, frame sync, checksum, and ILAS are available as IRQ events. Enable them by writing to Register 0x4B8, Bits[3:0]. The IRQ event status can be read at Register 0x4BA, Bits[3:0] after the IRQs are enabled. Write a 1 to Register 0x4BA, Bit 0 to reset the CGS IRQ. Write a 1 to Register 0x4BA, Bit 1 to reset the frame sync IRQ. Write a 1 to Register 0x4BA, Bit 2 to reset the checksum IRQ. Write a 1 to Register 0x4BA, Bit 3 to reset the ILAS IRQ. Bit No. 2 Bit Name BDE 1 NIT 0 UEK Description Set to 1 to assert SYNCOUT± if the disparity error count reaches the threshold Set to 1 to assert SYNCOUT± if the NIT error count reaches the threshold Set to 1 to assert SYNCOUT± if the UEK character error count reaches the threshold See the Interrupt Request Operation section for more information. Configuration Mismatch IRQ CGS, Frame Sync, Checksum, and ILAS Monitoring Register 0x470 to Register 0x473 can be monitored to verify that each stage of the JESD204B link establishment has occurred. Bit x of CODE_GRP_SYNC (Register 0x470) is high if Link Lane x received at least four K28.5 characters and passed code group synchronization. Bit x of FRAME_SYNC (Register 0x471) is high if Link Lane x completed initial frame synchronization. Bit x of GOOD_CHECKSUM (Register 0x472) is high if the checksum sent over the lane matches the sum of the JESD204B parameters sent over the lane during ILAS for Link Lane x. The parameters can be added either by summing the individual fields in registers or summing the packed register. If Register 0x300, Bit 6 = 0 (default), the calculated checksums are the lower eight bits of the sum of the following fields: DID, BID, LID, SCR, L − 1, F − 1, K − 1, M − 1, N − 1, SUBCLASSV, NP − 1, JESDV, S − 1, and HD. If Register 0x300, Bit 6 = 1, the calculated checksums are the lower eight bits of the sum of Register 0x400 to Register 0x40C and LID. The AD9161/AD9162 have a configuration mismatch flag that is available as an IRQ event. Use Register 0x4B9, Bit 0 to enable the mismatch flag (it is enabled by default), and then use Register 0x4BB, Bit 0 to read back its status and reset the IRQ signal. See the Interrupt Request Operation section for more information. The configuration mismatch event flag is high when the link configuration settings (in Register 0x450 to Register 0x45D) do not match the JESD204B transmitted settings (Register 0x400 to Register 0x40D). This function is different from the good checksum flags in Register 0x472. The good checksum flags ensure that the transmitted checksum matches a calculated checksum based on the transmitted settings. The configuration mismatch event ensures that the transmitted settings match the configured settings. HARDWARE CONSIDERATIONS See the Applications Information section for information on hardware considerations. Bit x of INIT_LANE_SYNC (Register 0x473) is high if Link Lane x passed the initial lane alignment sequence. Rev. 0 | Page 63 of 139 AD9161/AD9162 Data Sheet MAIN DIGITAL DATAPATH HB 2× HB 2× NCO HB 2×, 4×, 8× HB 3× INV SINC 14379-104 JESD Figure 160. Block Diagram of the Main Digital Datapath The block diagram in Figure 160 shows the functionality of the main digital datapath. The digital processing includes an input interpolation block with choice of bypass (1×), 2×, or 3× interpolation, three additional 2× half-band interpolation filters, a final 2× NRZ mode interpolator filter, FIR85, that can be bypassed, and a quadrature modulator that consists of a 48-bit NCO and an inverse sinc block. All of the interpolation filters accept in-phase (I) and quadrature (Q) data streams as a complex data stream. Similarly, the quadrature modulator and inverse sinc function also accept input data as a complex data stream. Thus, any use of the digital datapath functions requires the input data to be a complex data stream. In bypass mode (1× interpolation), the input data stream is expected to be real data. Table 34. Pipeline Delay (Latency) for Various DAC Blocks Mode NCO only 1× (Bypass) 1× (Bypass) 2× 2× 2× 2× 2× 2× 3× 3× 4× 6× 8× 12× 16× 24× 1 2 FIR85 On No No No No No Yes No Yes Yes No No No No No No No No Filter Bandwidth N/A2 N/A2 N/A2 80% 90% 80% 80% 80% 80% 80% 90% 80% 80% 80% 80% 80% 80% Inverse Sinc No No Yes No No No Yes Yes Yes No No No No No No No No NCO Yes No No No No No No No Yes No No No No No No No No Pipeline Delay1 (fDAC Clocks) 48 113 137 155 176 202 185 239 279 168 202 308 332 602 674 1188 1272 The pipeline delay given is a representative number, and may vary by a cycle or two based on the internal handoff timing conditions at startup. N/A means not applicable. The pipeline delay changes based on the digital datapath functions that are selected. See Table 34 for examples of the pipeline delay per block. These delays are in addition to the JESD204B latency. DATA FORMAT The input data format for all modes on the AD9161/AD9162 is 16-bit, twos complement. The digital datapath and the DAC decoder operate in twos complement format. The DAC is a current steering DAC and cannot represent 0—it must either source or sink current. As a result, when the 0 of twos complement is represented in the DAC, it is a +1, and all the positive values thereafter are shifted by +1. This mapping error introduces a ½ LSB shift in the DAC output. The leakage can become apparent when using the NCO to shift a signal that is above or below 0 Hz when synthesized. The NCO frequency is seen as a small spur at the NCO FTW. To avoid the NCO frequency leakage, operate the DAC with a slight digital backoff of one or several codes, and then add 1 to all values in the data stream. These actions remove the NCO frequency leakage but cause a half LSB DC offset. This small dc offset is benign to the DAC and does not affect most applications because the DAC output is ac-coupled through dc blocking capacitors. INTERPOLATION FILTERS The main digital path contains five half-band interpolation filters, plus a final half-band interpolation filter that is used in 2× NRZ mode. The filters are cascaded as shown in Figure 160. The first pair of filters is a 2× (HB2) or 3× (HB3) filter. Each of these filters has two options for bandwidth, 80% or 90%. The 80% filters are lower power than the 90%. The filters default to the lower power 80% bandwidth. To select the filter bandwidth as 90%, program the FILT_BW bit in the DATAPATH_CFG register to 1 (Register 0x111, Bit 4 = 0b1). Following the first pair of filters is a series of 2× half-band filters, each of which halves the usable bandwidth of the previous one. HB4 has 45%, HB5 has 22.5%, and HB6 has 11.25% of the fDATA bandwidth. The final half-band filter, FIR85, is used in the 2× NRZ mode. It is clocked at the 2 × fDAC rate and has a usable bandwidth of 45% of the fDAC rate. The FIR85 filter is a complex filter, and therefore the bandwidth is centered at 0 Hz. The FIR85 filter is used in conjunction with the complex interpolation modes to Rev. 0 | Page 64 of 139 Data Sheet AD9161/AD9162 1× 2× 3× 4× 6× 8× 12× 16× 24× FIR85 Table 35 shows how to select each available interpolation mode, their usable bandwidths, and their maximum data rates. Calculate the available signal bandwidth as the interpolator filter bandwidth, BW, multiplied by fDAC/InterpolationFactor, as follows: –1500 BWSIGNAL = BWFILT × (fDAC/InterpolationFactor) –500 500 1500 2500 FREQUENCY (MHz) 14379-105 The FIR85 filter can be used with the 1× bypass mode, but in that mode, the signal is limited to the lower half of the Nyquist zone, that is, up to fDAC/4. If the NCO is used with FIR85 in 1× bypass mode, images of the NCO upconversion appears because the data interface in 1× bypass is real; therefore, an image reject upconversion cannot be accomplished by the NCO. To synthesize a signal and then upconvert it with the NCO, use the complex data interface and choose an appropriate interpolator prior to FIR85. of each of the filters is shown in Figure 161. The maximum pass band amplitude of all filters is the same; they are different in the illustration to improve understanding. FILTER RESPONSE push the DAC update rate higher and move images further from the desired signal. Figure 161. All Band Responses of Interpolation Filters Filter Performance The interpolation filters interpolate between existing data in such a way that they minimize changes in the incoming data while suppressing the creation of interpolation images. This datapath is shown for each filter in Figure 161. The usable bandwidth (as shown in Table 35) is defined as the frequency band over which the filters have a pass-band ripple of less than ±0.001 dB and an image rejection of greater than 85 dB. A conceptual drawing that shows the relative bandwidth Filter Performance Beyond Specified Bandwidth Some of the interpolation filters are specified to 0.4 × fDATA (with a pass band). The filters can be used slightly beyond this ratio at the expense of increased pass-band ripple and decreased interpolation image rejection. Table 35. Interpolation Modes and Usable Bandwidth Interpolation Mode 1× (Bypass) 2× 3× 4× 6× 8× 12× 16× 24× 2× NRZ (Register 0x111, Bit 0 = 1) INTERP_MODE, Register 0x110, Bits[3:0] 0x00 0x01 0x02 0x03 0x04 0x05 0x06 0x07 0x08 Any combination 3 Available Signal Bandwidth (BW) 1 fDAC/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 BW × fDATA/2 0.45 × fDAC 4 Maximum fDATA (MHz) fDAC 2 fDAC/22 fDAC/3 fDAC/4 fDAC/6 fDAC/8 fDAC/12 fDAC/16 fDAC/24 fDAC (real) or fDAC/2(complex)2 The data rate (fDATA) for all interpolator modes is a complex data rate, meaning each of I data and Q data run at that rate. The data rate for the 1× bypass mode is a real data rate. Available signal bandwidth is the data rate multiplied by the bandwidth of the initial 2× or 3× interpolator filters, which can be set to BW = 80% or BW = 90%. This bandwidth is centered at 0 Hz. 2 The maximum speed for 1× and 2× interpolation is limited by the JESD204B interface, and is 5000 MHz (real) in 1× or 2500 MHz (complex) in 2× interpolation mode. 3 The 2× NRZ filter, FIR85, can be used with any of the interpolator combinations. When used in 1× bypass mode, the desired signal must be placed in the lower half of the first Nyquist zone, as in 0 to DAC clock ÷ 4 MHz. 4 The bandwidth of the FIR85 filter is centered at 0 Hz. 1 Rev. 0 | Page 65 of 139 Data Sheet 20 0 90 –0.2 60 –0.3 50 –0.4 40 –0.5 30 –20 42 45 44 43 –80 –120 –140 –0.6 41 –60 –100 IMAGE REJECTION PASS-BAND RIPPLE 20 40 –40 BANDWIDTH (% fDATA ) 0 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Figure 164. First 2× Half-Band 90% Filter Response 20 0 –20 MAGNITUDE (dB) Most of the filters are specified to 0.45 × fDATA (with pass band). Figure 163 to Figure 170 show the filter response for each of the interpolator filters on the AD9161/AD9162. 0.2 NORMALIZED FREQUENCY (Rad/Sample) Figure 162. Interpolation Filter Performance Beyond Specified Bandwidth for the 80% Filters Figure 162 shows the performance of the interpolation filters beyond 0.4 × fDATA. The ripple increases much slower than the image rejection decreases. This means that if the application can tolerate degraded image rejection from the interpolation filters, more bandwidth can be used. 0.1 14379-159 70 MAGNITUDE (dB) –0.1 MAXIMUM PASS-BAND RIPPLE (dB) 0 80 14379-106 –40 –60 –80 –100 –120 20 –140 –160 0 –20 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 MAGNITUDE (dB) NORMALIZED FREQUENCY (Rad/Sample) 14379-160 0 Figure 165. 3× Third-Band 80% Filter Response –40 20 –60 0 –80 –20 –140 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 NORMALIZED FREQUENCY (Rad/Sample) 1.0 14379-158 –120 MAGNITUDE (dB) –100 –40 –60 –80 –100 Figure 163. First 2× Half-Band 80% Filter Response –120 –140 –160 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 NORMALIZED FREQUENCY (Rad/Sample) Figure 166. 3× Third-Band 90% Filter Response Rev. 0 | Page 66 of 139 1.0 14379-161 MINIMUM INTERPOLATION IMAGE REJECTION (dB) AD9161/AD9162 Data Sheet AD9161/AD9162 20 20 0 0 –20 –20 MAGNITUDE (dB) –60 –80 –100 –60 –80 –100 –120 –120 –140 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 NORMALIZED FREQUENCY (Rad/Sample) –140 14379-162 –160 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 NORMALIZED FREQUENCY (Rad/Sample) Figure 167. Second 2× Half-Band 45% Filter Response 1.0 Figure 170. FIR85 2× Half-Band 45% Filter Response 20 DIGITAL MODULATION 0 The AD9161/AD9162 have digital modulation features to modulate the baseband quadrature signal to the desired DAC output frequency. –20 –40 –120 The AD9161/AD9162 are equipped with several NCO modes. The default NCO is a 48-bit, integer NCO. The A/B ratio of the dual modulus NCO allows the output frequency to be synthesized with very fine precision. NCO mode is selected as shown in Table 36. –140 Table 36. Modulation Mode Selection –60 –80 –100 –160 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 NORMALIZED FREQUENCY (Rad/Sample) 14379-163 MAGNITUDE (dB) –40 14379-165 MAGNITUDE (dB) –40 Figure 168. Third 2× Half-Band 22.5% Filter Response 20 MAGNITUDE (dB) 0 Modulation Mode None 48-Bit Integer NCO 48-Bit Dual Modulus NCO Modulation Type Register 0x111, Register 0x111, Bit 6 Bit 2 0b0 0b0 0b1 0b0 0b1 0b1 –20 48-Bit Dual Modulus NCO –40 This modulation mode uses an NCO, a phase shifter, and a complex modulator to modulate the signal by a programmable carrier signal as shown in Figure 171. This configuration allows output signals to be placed anywhere in the output spectrum with very fine frequency resolution. –60 –80 –100 –140 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 NORMALIZED FREQUENCY (Rad/Sample) Figure 169. Fourth 2× Half-Band 11.25% Filter Response 1.0 14379-164 –120 The NCO produces a quadrature carrier to translate the input signal to a new center frequency. A quadrature carrier is a pair of sinusoidal waveforms of the same frequency, offset 90° from each other. The frequency of the quadrature carrier is set via a FTW. The quadrature carrier is mixed with the I and Q data and then summed into the I and Q datapaths, as shown in Figure 171. Rev. 0 | Page 67 of 139 AD9161/AD9162 Data Sheet Integer NCO Mode The main 48-bit NCO can be used as an integer NCO by using the following formula to create the frequency tuning word (FTW): −fDAC/2 ≤ fCARRIER < +fDAC/2 INTERPOLATION COS(ωn + θ) ω π NCO θ SIN(ωn + θ) FTW[47:0] NCO_PHASE_OFFSET [15:0] 48 OUT_I OUT_Q – where FTW is a 48-bit, twos complement number. + –1 When in 2× NRZ mode (FIR85 enabled with Register 0x111, Bit 0 = 1), the frequency tuning word is calculated as SEL_SIDEBAND 0 1 0 ≤ fCARRIER < fDAC Q DATA FTW = (fCARRIER/fDAC) × 248 where FTW is a 48-bit binary number. Figure 171. NCO Modulator Block Diagram This method of calculation causes fCARRIER values in the second Nyquist zone to appear to move to fDAC − fCARRIER when flipping the FIR85 enable bit and not changing the FTW to account for the change in number format. The intended effect is that a sweep of the NCO from 0 Hz to fDAC − fDAC/248 appears seamless when the FIR85 enable bit is set to Register 0x111, Bit 0 = 0b1 prior to fCARRIER/fDAC = 0.5. As can be seen from examination, the FTWs from 0 to less than fDAC/2 mean the same in either case, but they mean different fCARRIER values from fDAC/2 to fDAC − fDAC/248. This effect must be considered when constructing FTW values and using the 2× NRZ mode. The frequency tuning word is set as shown in Table 37. Table 37. NCO FTW Registers Address 0x114 0x115 0x116 0x117 0x118 0x119 Value FTW[7:0] FTW[15:8] FTW[23:16] FTW[31:24] FTW[39:32] FTW[47:40] INTERPOLATION 14379-108 FTW = (fCARRIER/fDAC) × 2 I DATA Description 8 LSBs of FTW Next 8 bits of FTW Next 8 bits of FTW Next 8 bits of FTW Next 8 bits of FTW 8 MSBs of FTW Modulus NCO Mode The main 48-bit NCO can also be used in a dual modulus mode to create fractional frequencies beyond the 48-bit accuracy. The modulus mode is enabled by programming the MODULUS_EN bit in the DATAPATH_CFG register to 1 (Register 0x111, Bit 2 = 0b1). The frequency ratio for the programmable modulus direct digital synthesis (DDS) is very similar to that of the typical accumulator-based DDS. The only difference is that N is not required to be a power of two for the programmable modulus, but can be an arbitrary integer. In practice, hardware constraints place limits on the range of values for N. As a result, the modulus extends the use of the NCO to applications that require exact rational frequency synthesis. The underlying function of the programmable modulus technique is to alter the accumulator modulus. Implementation of the programmable modulus function within the AD9161/AD9162 is such that the fraction, M/N, is expressible per Equation 1. Note that the form of the equation implies a compound frequency tuning word with X representing the integer part and A/B representing the fractional part. Unlike other registers, the FTW registers are not updated immediately upon writing. Instead, the FTW registers update on the rising edge of FTW_LOAD_REQ (Register 0x113, Bit 0). After an update request, FTW_LOAD_ACK (Register 0x113, Bit 1) must be high to acknowledge that the FTW has updated. The SEL_SIDEBAND bit (Register 0x111, Bit 1 = 0b1) is a convenience bit that can be set to use the lower sideband modulation result, which is equivalent to flipping the sign of the FTW. A f CARRIER M X + B = = 2 48 f DAC N where: X is programmed in Register 0x114 to Register 0x119. A is programmed in Register 0x12A to Register 0x12F. B is programmed in Register 0x124 to Register 0x129. Rev. 0 | Page 68 of 139 (1) Data Sheet AD9161/AD9162 Consider the case in which fDAC = 2500 MHz and the desired value of fCARRIER is 250 MHz. This scenario synthesizes an output frequency that is not a power of two submultiple of the sample rate, namely fCARRIER = (1/10) fDAC, which is not possible with a typical accumulator-based DDS. The frequency ratio, fCARRIER/fDAC, leads directly to M and N, which are determined by reducing the fraction (250,000,000/2,500,000,000) to its lowest terms, that is, M/N = 250,000,000/2,500,000,000 = 1/10 The speed of the SPI port write function is guaranteed, and is a minimum of 100 MHz (see Table 4). Thus, the NCO FTW can be updated in as little as 240 ns with a one register write in automatic update mode. The manner in which the NCO transitions to the new frequency is determined by the frequency change mode selection. The NCO supports several modes of changing frequency: phase continuous or phase noncontinuous. The modes are given in Table 38. Therefore, M = 1 and N = 10. Table 38. NCO Frequency Change Mode After calculation, X = 28147497671065, A = 3, and B = 5. Programming these values into the registers for X, A, and B (X is programmed in Register 0x114 to Register 0x119, B is programmed in Register 0x124 to Register 0x129, and A is programmed in Register 0x12A to Register 0x12F)) causes the NCO to produce an output frequency of exactly 250 MHz given a 2500 MHz sampling clock. For more details, refer to the AN953 Application Note on the Analog Devices, Inc., website. Register 0x800, Bits[7:6] 0b00 0b01 Resetting the NCO can be useful when determining the start time and phase of the NCO. The NCO can be reset by several different methods, including a SPI write, using the TX_ENABLE pin, or by the SYSREF± signal. Due to internal timing variations from device to device, these methods achieve an accuracy of ±6 DAC clock cycles. Program Register 0x800, Bits[7:6] to 0b01 to set the NCO in phase discontinuous switching mode via a write to the SPI port. Then, any time the frequency tuning word is updated, the NCO phase accumulator resets and the NCO begins counting at the new FTW. Changing the Main NCO Frequency In the main 48-bit NCO, the mode of updating the frequency tuning word can be changed from requiring a write to the FTW_LOAD_REQ bit (Register 0x113, Bit 0) to an automatic update mode. In the automatic update mode, the FTW is updated as soon as the chosen FTW word is written. To set the automatic FTW update mode, write the appropriate word to the FTW_REQ_MODE bits (Register 0x113, Bits[6:4]), choosing the particular FTW word that causes the automatic update. For example, if relatively coarse frequency steps are needed, it may be sufficient to write a single word to the MSB byte of the FTW, and therefore the FTW_REQ_MODE bits can be programmed to 110 (Register 0x113, Bits[6:4] = 0b110). Then, each time the most significant byte, FTW5, is written, the NCO FTW is automatically updated. In phase continuous switching, the frequency tuning word of the NCO is updated and the phase accumulator continues to accumulate to the new frequency. In phase noncontinuous mode, the FTW of the NCO is updated and the phase accumulator is reset, making an instantaneous jump to the new frequency. INVERSE SINC The AD9161/AD9162 provide a digital inverse sinc filter to compensate the DAC roll-off over frequency. The filter is enabled by setting the INVSINC_ENABLE bit (Register 0x111, Bit 7) and is disabled by default. The inverse sinc (sinc−1) filter is a seven-tap FIR filter. Figure 172 shows the frequency response of sin(x)/x roll-off, the inverse sinc filter, and the composite response. The composite response has less than ±0.05 dB pass-band ripple up to a frequency of 0.4 × fDACCLK. When 2× NRZ mode is enabled, the inverse sinc filter operates to 0.4 × f2×DACCLK. To provide the necessary peaking at the upper end of the pass band, the inverse sinc filter shown has an intrinsic insertion loss of about 3.8 dB. 1 SIN(x)/x ROLL-OFF SINC–1 FILTER RESPONSE COMPOSITE RESPONSE 0 MAGNITUDE (dB) NCO Reset Description Phase continuous switch Phase noncontinuous switch (reset NCO accumulator) The FTW_REQ_MODE bits can be configured to use any of the FTW words as the automatic update trigger word. This configuration provides convenience when choosing the order in which to program the FTW registers. Rev. 0 | Page 69 of 139 –1 –2 –3 –4 –5 0 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 FREQUENCY (× fDAC ) Figure 172. Responses of Sin(x)/x Roll-Off, the Sinc−1 Filter, and the Composite of the Two 14379-109 Programmable Modulus Example AD9161/AD9162 Data Sheet DOWNSTREAM PROTECTION The AD9161/AD9162 have several features designed to protect the power amplifier (PA) of the system, as well as other downstream blocks. They consist of a control signal from the LMFC sync logic and a transmit enable function. The protection mechanism in each case is the blanking of data that is passed to the DAC decoder. The differences lie in the location in the datapath and slight variations of functionality. The JESD204B serial link has several flags and quality measures to indicate the serial link is up and running error free. If any of these measures flags an issue, a signal from the LMFC sync logic is sent to a mux that stops data from flowing to the DAC decoder and replaces it with 0s. There are several transmit enable features, including a TX_ ENABLE register that can be used to squelch data at several points in the datapath or configure the TX_ENABLE pin to do likewise. controlled function, namely to zero the input to the digital datapath or to zero the output from the digital datapath. In addition, the TX_ENABLE pin can also be configured to ramp down (or up) the full-scale current of the DAC. The ramp down reduces the output power of the DAC by about 20 dB from full scale to the minimum output current. The TX_ENABLE pin can also be programmed to reset the NCO phase accumulator. See Table 39 for a description of the settings available for the TX_ENABLE function. Table 39. TX_ENABLE Settings Register 0x03F Bit 7 Setting 0 1 Bit 6 0 1 Transmit Enable The transmit enable feature can be configured either as a SPI controlled function or a pin controlled function. It can be used for several different purposes. The SPI controlled function has less accurate timing due to its reliance on a microcontroller to program it; therefore, it is typically used as a preventative measure at power-up or when configuring the device. The SPI controlled TX_ENABLE function can be used to zero the input to the digital datapath or to zero the output from the digital datapath, as shown in Figure 173. If the input to the digital datapath is zeroed, any filtering that is selected filters the 0 signal, causing a gradual ramp-down of energy in the digital datapath. If the digital datapath is bypassed, as in 1÷ mode, the data at the input to the DAC immediately drops to zero. The TX_ENABLE pin can be used for more accurate timing when enabling or disabling the DAC output. The effect of the TX_ENABLE pin can be configured by the same TX_ENABLE register (Register 0x03F) as is used for the SPI controlled functions, and it can be made to have the same effects as the SPI Bits[5:4] Bit 3 N/A1 0 1 0 1 Bit 2 Bit 1 0 1 Bit 0 0 1 1 2 Description SPI control: zero data to the DAC SPI control: allow data to pass to the DAC SPI control: zero data at input to the datapath SPI control: allow data to enter the datapath Reserved Use SPI writes to reset the NCO2 Use TX_ENABLE to reset the NCO Use SPI control to zero data to the DAC Use TX_ENABLE pin to zero data to the DAC Use SPI control to zero data at the input to the datapath Use TX_ENABLE pin to zero data at input to the datapath Use SPI registers to control the full-scale current Use TX_ENABLE pin to control the fullscale current N/A means not applicable. Use SPI writes to reset the NCO if resetting the NCO is desired. Register 0x800, Bits[7:6] determine whether the NCO is reset. See Table 38 for more details. DATA TO DAC 0 0 MAIN DIGITAL PATH 0 FROM LMFC SYNC LOGIC FROM REG 0x03F[7] FROM REG 0x03F[6] FROM REG 0x03F[2] FROM REG 0x03F[1] Figure 173. Downstream Protection Block Diagram Rev. 0 | Page 70 of 139 14379-110 TX_ENABLE TX_ENABLE Data Sheet AD9161/AD9162 INTERRUPT REQUEST OPERATION The AD9161/AD9162 provide an interrupt request output signal (IRQ) on Ball G1 (8 mm × 8 mm package) or Ball G4 (11 mm × 11 mm package) that can be used to notify an external host processor of significant device events. On assertion of the interrupt, query the device to determine the precise event that occurred. The IRQ pin is an open-drain, active low output. Pull the IRQ pin high, external to the device. This pin can be tied to the interrupt pins of other devices with open-drain outputs to wire; OR these pins together. Figure 174 shows a simplified block diagram of how the IRQ blocks works. If IRQ_EN is low, the INTERRUPT_SOURCE signal is set to 0. If IRQ_EN is high, any rising edge of EVENT causes the INTERRUPT_SOURCE signal to be set high. If any INTERRUPT_SOURCE signal is high, the IRQ pin is pulled low. INTERRUPT_SOURCE can be reset to 0 by either an IRQ_RESET signal or a DEVICE_RESET signal. Depending on the STATUS_MODE signal, the EVENT_STATUS bit reads back an event signal or INTERRUPT_SOURCE signal. The AD9161/AD9162 have several IRQ register blocks that can monitor up to 75 events (depending on device configuration). Certain details vary by IRQ register block as described in Table 40. Table 41 shows the source registers of the IRQ_EN, IRQ_RESET, and STATUS_MODE signals in Figure 174, as well as the address where EVENT_STATUS is read back. Table 40. IRQ Register Block Details Register Block 0x020, 0x024 Event Reported Per chip 0x4B8 to 0x4BB; 0x470 to 0x473 Per link and lane EVENT_STATUS INTERRUPT_SOURCE if IRQ is enabled; if not, it is the event signal INTERRUPT_SOURCE if IRQ is enabled; if not, 0 INTERRUPT SERVICE ROUTINE Interrupt request management starts by selecting the set of event flags that require host intervention or monitoring. Enable the events that require host action so that the host is notified when they occur. For events requiring host intervention upon IRQ activation, run the following routine to clear an interrupt request: 1. 2. 3. 4. Read the status of the event flag bits that are being monitored. Disable the interrupt by writing 0 to IRQ_EN. Read the event source. Perform any actions that may be required to clear the cause of the event. In many cases, no specific actions may be required. Verify that the event source is functioning as expected. Clear the interrupt by writing 1 to IRQ_RESET. Enable the interrupt by writing 1 to IRQ_EN. 5. 6. 7. 0 1 EVENT_STATUS STATUS_MODE IRQ IRQ_EN EVENT INTERRUPT_SOURCE 0 1 IRQ_EN OTHER INTERRUPT SOURCES IRQ_RESET 14379-111 DEVICE_RESET Figure 174. Simplified Schematic of IRQ Circuitry Table 41. IRQ Register Block Address of IRQ Signal Details Register Block 0x020, 0x024 0x4B8 to 0x4BB 0x470 to 0x473 1 IRQ_EN 0x020; R/W per chip 0x4B8, 0x4B9; W per error type 0x470 to 0x473; W per error type Address of IRQ Signals IRQ_RESET STATUS_MODE 1 0x024; W per chip STATUS_MODE = IRQ_EN 0x4BA, 0x4BB; W per error type N/A, STATUS_MODE = 1 0x470 to 0x473; W per link N/A, STATUS_MODE = 1 N/A means not applicable. Rev. 0 | Page 71 of 139 EVENT_STATUS 0x024; R per chip 0x4BA, 0x4BB; R per chip 0x470 to 0x473; R per link AD9161/AD9162 Data Sheet APPLICATIONS INFORMATION HARDWARE CONSIDERATIONS Power Sequencing Power Supply Recommendations The AD9161/AD9162 require power sequencing to avoid damage to the DAC. A board design with the AD9161/AD9162 must include a power sequencer chip, such as the ADM1184, to ensure that the domains power up in the correct order. The ADM1184 monitors the level of power domains upon power-up. It sends an enable signal to the next grouping of power domains. When all power domains are powered up, a power-good signal is sent to the system controller to indicate all power supplies are powered up. All the AD9161/AD9162 supply domains must remain as noise free as possible for the best operation. Power supply noise has a frequency component that affects performance, and is specified in volts rms terms. An LC filter on the output of the power supply is recommended to attenuate the noise, and must be placed as close to the AD9161/ AD9162 as possible. The VDD12_CLK supply is the most noise sensitive supply on the device, followed by the VDD25_DAC and VNEG_N1P2 supplies, which are the DAC output rails. It is highly recommended that the VDD12_CLK be supplied by itself with an ultralow noise regulator such as the ADM7154 or ADP1761 or better to achieve the best phase noise performance possible. Noisier regulators impose phase noise onto the DAC output. The VDD12A supply can be connected to the digital DVDD supply with a separate filter network. All of the SERDES 1.2 V supplies can be connected to one regulator with separate filter networks. The IOVDD supply can be connected to the VDD25_ DAC supply with a separate filter network, or can be powered from a system controller (for example, a microcontroller), 1.8 V to 3.3 V supply. The power supply sequencing requirement must be met; therefore, a switch or other solution must be used when connected to the IOVDD supply with VDD25_DAC. Take note of the maximum power consumption numbers given in Table 3 to ensure the power supply design can tolerate temperature and IC process variation extremes. The amount of current drawn is dependent on the chosen use cases, and specifications are provided for several use cases to illustrate examples and contributions from individual blocks, and to assist in calculating the maximum required current per supply. Another consideration for the power supply design is peak current handling capability. The AD9161/AD9162 draw more current in the main digital supply when synthesizing a signal with significant amplitude variations, such as a modulated signal, as compared to when in idle mode or synthesizing a dc signal. Therefore, the power supply must be able to supply current quickly to accommodate burst signals such as GSM, TDMA, or other signals that have an on/off time domain response. Because the amount of current variation depends on the signals used, it is best to perform lab testing first to establish ranges. A typical difference can be several hundred milliamperes. The IOVDD, VDD12A, VDD12_CLK, and DVDD domains must be powered up first. Then, the VNEG_N1P2, VDD_1P2, PLL_CLK_VDD12, DVDD_1P2, and SYNC_VDD_3P3 can be powered up. The VDD25_DAC domain must be powered up last. There is no requirement for a power-down sequence. Power and Ground Planes Solid ground planes are recommended to avoid ground loops and to provide a solid, uninterrupted ground reference for the high speed transmission lines that require controlled impedances. It is recommended that power planes be stacked between ground layers for high frequency filtering. Doing so adds extra filtering and isolation between power supply domains in addition to the decoupling capacitors. Do not use segmented power planes as a reference for controlled impedances unless the entire length of the controlled impedance trace traverses across only a single segmented plane. These and additional guidelines for the topology of high speed transmission lines are described in the JESD204B Serial Interface Inputs (SERDIN0± to SERDIN7±) section. For some applications, where highest performance and higher output frequencies are required, the choice of PCB materials significantly impacts results. For example, materials such as polyimide or materials from the Rogers Corporation can be used, for example, to improve tolerance to high temperatures and improve performance. Rogers 4350 material is used for the top three layers in some of the evaluation board designs: between the top signal layer and the ground layer below it, between the ground layer and an internal signal layer, and between that signal layer and another ground layer. JESD204B Serial Interface Inputs (SERDIN0± to SERDIN7±) When considering the layout of the JESD204B serial interface transmission lines, there are many factors to consider to maintain optimal link performance. Among these factors are insertion loss, return loss, signal skew, and the topology of the differential traces. Rev. 0 | Page 72 of 139 Data Sheet AD9161/AD9162 The JESD204B specification limits the amount of insertion loss allowed in the transmission channel (see Figure 146). The AD9161/ AD9162 equalization circuitry allows significantly more loss in the channel than is required by the JESD204B specification. It is still important that the designer of the PCB minimize the amount of insertion loss by adhering to the following guidelines: Keep the differential traces short by placing the AD9161/AD9162 as near the transmitting logic device as possible and routing the trace as directly as possible between the devices. Route the differential pairs on a single plane using a solid ground plane as a reference. It is recommended to route the SERDES lanes on the same layer as the AD9161/AD9162 to avoid vias being used in the SERDES lanes. Use a PCB material with a low dielectric constant (<4) to minimize loss, if possible. When choosing between the stripline and microstrip techniques, keep in mind the following considerations: stripline has less loss (see Figure 147 and Figure 148) and emits less EMI, but requires the use of vias that can add complexity to the task of controlling the impedance; whereas microstrip is easier to implement (if the component placement and density allow routing on the top layer) and eases the task of controlling the impedance. If using the top layer of the PCB is problematic or the advantages of stripline are desirable, follow these recommendations: Minimize the number of vias. If possible, use blind vias to eliminate via stub effects and use microvias to minimize via inductance. If using standard vias, use the maximum via length to minimize the stub size. For example, on an 8-layer board, use Layer 7 for the stripline pair (see Figure 175). For each via pair, place a pair of ground vias adjacent to them to minimize the impedance discontinuity (see Figure 175). LAYER 1 ADD GROUND VIAS y STANDARD VIA LAYER 5 y DIFF+ LAYER 6 GND LAYER 7 LAYER 8 y DIFF– LAYER 3 LAYER 4 GND MINIMIZE STUB EFFECT 14379-100 LAYER 2 Figure 175. Minimizing Stub Effect and Adding Ground Vias for Differential Stripline Traces AD9162. Minimizing the use of vias, or eliminating them all together, reduces one of the primary sources for impedance mismatches on a transmission line (see the Insertion Loss section). Maintain a solid reference beneath (for microstrip) or above and below (for stripline) the differential traces to ensure continuity in the impedance of the transmission line. If the stripline technique is used, follow the guidelines listed in the Insertion Loss section to minimize impedance mismatches and stub effects. Another primary source for impedance mismatch is at either end of the transmission line, where care must be taken to match the impedance of the termination to that of the transmission line. The AD9161/AD9162 handle this internally with a calibrated termination scheme for the receiving end of the line. See the Interface Power-Up and Input Termination section for details on this circuit and the calibration routine. Signal Skew There are many sources for signal skew, but the two sources to consider when laying out a PCB are interconnect skew within a single JESD204B link and skew between multiple JESD204B links. In each case, keeping the channel lengths matched to within 12.5 mm is adequate for operating the JESD204B link at speeds of up to 12.5 Gbps. This amount of channel length match is equivalent to about 85% UI on the AD9161/AD9162 evaluation board. Managing the interconnect skew within a single link is fairly straightforward. Managing multiple links across multiple devices is more complex. However, follow the 12.5 mm guideline for length matching. The AD9161/AD9162 can handle more skew than the 85% UI due to the 6 PCLK buffer in the JESD204B receiver, but matching the channel lengths as close as possible is still recommended. Topology Structure the differential SERDINx± pairs to achieve 50 Ω to ground for each half of the pair. Stripline vs. microstrip tradeoffs are described in the Insertion Loss section. In either case, it is important to keep these transmission lines separated from potential noise sources such as high speed digital signals and noisy supplies. If using stripline differential traces, route them using a coplanar method, with both traces on the same layer. Although this method does not offer more noise immunity than the broadside routing method (traces routed on adjacent layers), it is easier to route and manufacture so that the impedance continuity is maintained. An illustration of broadside vs. coplanar is shown in Figure 176. Tx DIFF A Tx DIFF B Return Loss The JESD204B specification limits the amount of return loss allowed in a converter device and a logic device, but does not specify return loss for the channel. However, every effort must be made to maintain a continuous impedance on the transmission line between the transmitting logic device and the AD9161/ Tx ACTIVE BROADSIDE DIFFERENTIAL Tx LINES Tx DIFF A Tx DIFF B Tx ACTIVE COPLANAR DIFFERENTIAL Tx LINES 14379-101 Insertion Loss Figure 176. Broadside vs. Coplanar Differential Stripline Routing Techniques When considering the trace width vs. copper weight and thickness, the speed of the interface must be considered. At Rev. 0 | Page 73 of 139 AD9161/AD9162 Data Sheet To minimize the impedance mismatch at the pads, select the package size of the capacitor so that the pad size on the PCB matches the trace width as closely as possible. multigigabit speeds, the skin effect of the conducting material confines the current flow to the surface. Maximize the surface area of the conductor by making the trace width made wider to reduce the losses. Additionally, loosely couple differential traces to accommodate the wider trace widths. This coupling helps reduce the crosstalk and minimize the impedance mismatch when the traces must separate to accommodate components, vias, connectors, or other routing obstacles. Tightly coupled vs. loosely coupled differential traces are shown in Figure 177. TIGHTLY COUPLED DIFFERENTIAL Tx LINES Tx DIFF A Tx DIFF B Separate the SYNCOUT± signal from other noisy signals, because noise on the SYNCOUT± might be interpreted as a request for /K/ characters. LOOSELY COUPLED DIFFERENTIAL Tx LINES It is important to keep similar trace lengths for the CLK± and SYSREF± signals from the clock source to each of the devices on either end of the JESD204B links (see Figure 178). If using a clock chip that can tightly control the phase of CLK± and SYSREF±, the trace length matching requirements are greatly reduced. Figure 177. Tightly Coupled vs. Loosely Coupled Differential Traces AC Coupling Capacitors The AD9161/AD9162 require that the JESD204B input signals be ac-coupled to the source. These capacitors must be 100 nF and placed as close as possible to the transmitting logic device. LANE 0 LANE 1 Tx DEVICE Rx DEVICE LANE N – 1 LANE N SYSREF± DEVICE CLOCK SYSREF± CLOCK SOURCE (AD9516-1, ADCLK925) SYSREF± TRACE LENGTH DEVICE CLOCK TRACE LENGTH DEVICE CLOCK SYSREF± TRACE LENGTH DEVICE CLOCK TRACE LENGTH Figure 178. SYSREF± Signal and Device Clock Trace Length Rev. 0 | Page 74 of 139 14379-103 Tx DIFF B The SYNCOUT± and SYSREF± signals on the AD9161/AD9162 are low speed LVDS differential signals. Use controlled impedance traces routed with 100 Ω differential impedance and 50 Ω to ground when routing these signals. As with the SERDIN0± to SERDIN7± data pairs, it is important to keep these signals separated from potential noise sources such as high speed digital signals and noisy supplies. 14379-102 Tx DIFF A SYNCOUT±, SYSREF±, and CLK± Signals Data Sheet AD9161/AD9162 ANALOG INTERFACE CONSIDERATIONS The AD9161/AD9162 use the quad-switch architecture shown in Figure 179. Only one pair of switches is enabled during a halfclock cycle, thus requiring each pair to be clocked on alternative clock edges. A key benefit of the quad-switch architecture is that it masks the code dependent glitches that occur in the conventional two-switch DAC architecture. CLK± CLK IOUTP VG1 DATA INPUT VG1 INPUT DATA D1 D2 VG2 VG3 D3 D4 D5 D6 D7 D8 D3 D10 –D8 D2 FOUR-SWITCH DAC OUTPUT (fS MIX-MODE) VG4 D9 DACCLK_x IOUTN VG2 LATCHES V 3 G When Mix-Mode is used, the output is effectively chopped at the DAC sample rate. This chopping has the effect of reducing the power of the fundamental signal while increasing the power of the images centered around the DAC sample rate, thus improving the dynamic range of these images. D4 –D7 D1 D5 –D9 –D6 –D10 t –D5 D6 –D1 VG4 –D2 D10 –D4 D9 D7 –D3 14379-114 ANALOG MODES OF OPERATION D8 14379-112 Figure 181. Mix-Mode Waveform VSSA Figure 179. Quad-Switch Architecture In two-switch architecture, when a switch transition occurs and D1 and D2 are in different states, a glitch occurs. However, if D1 and D2 happen to be at the same state, the switch transitions and no glitches occur. This code dependent glitching causes an increased amount of distortion in the DAC. In quad-switch architecture (no matter what the codes are), there are always two switches that are transitioning at each half-clock cycle, thus eliminating the code-dependent glitches but, in the process, creating a constant glitch at 2 × fDAC. For this reason, a significant clock spur at 2 × fDAC is evident in the DAC output spectrum. INPUT DATA D1 D2 D3 D4 D5 D6 D7 D8 D9 This ability to change modes provides the user the flexibility to place a carrier anywhere in the first three Nyquist zones, depending on the operating mode selected. Switching between baseband and Mix-Mode reshapes the sinc roll-off inherent at the DAC output. In baseband mode, the sinc null appears at fDACCLK because the same sample latched on the rising clock edge is also latched again on the falling clock edge, thus resulting in the same ubiquitous sinc response of a traditional DAC. In Mix-Mode, the complement sample of the rising edge is latched on the falling edge, therefore pushing the sinc null to 2 × fDACCLK. Figure 182 shows the ideal frequency response of the three modes with the sinc roll-off included. FIRST NYQUIST ZONE D10 SECOND NYQUIST ZONE 0 THIRD NYQUIST ZONE MIX-MODE DACCLK_x –5 RZ MODE D3 D4 D5 t D6 D6 D2 D3 D4 D7 D7 D8 D8 D9 D9 AMPLITUDE (dBFS) FOUR-SWITCH DAC OUTPUT (NORMAL MODE) D1 D2 D10 D10 D5 t 14379-113 TWO-SWITCH DAC OUTPUT D1 –10 –15 NORMAL MODE –20 –25 –30 As a consequence of the quad-switch architecture enabling updates on each half-clock cycle, it is possible to operate that DAC core at 2× the DAC clock rate if new data samples are latched into the DAC core on both the rising and falling edge of the DAC clock. This notion serves as the basis when operating the AD9161/AD9162 in either Mix-Mode or return to zero (RZ) mode. In each case, the DAC core is presented with new data samples on each clock edge: in RZ mode, the rising edge clocks data and the falling edge clocks zero, while in Mix-Mode; the falling edge sample is simply the complement of the rising edge sample value. –35 0FS 0.25FS 0.50FS 0.75FS 1.00FS 1.25FS 1.50FS FREQUENCY (Hz) 14379-115 Figure 180. Two-Switch and Quad-Switch DAC Waveforms Figure 182. Sinc Roll-Off for NRZ, RZ, and Mix-Mode Operation The quad-switch can be configured via SPI (Register 0x152, Bits[1:0]) to operate in either NRZ mode (0b00), RZ mode (0b10), or Mix-Mode (0b01). The AD9161/AD9162 have an additional frequency response characteristic due to the FIR85 filter. This filter samples data on both the rising and falling edges of the DAC clock, in essence doubling the input clock frequency. As a result, the NRZ (normal) mode roll-off in Figure 182 is extended to 2 × fDAC in Figure 182, and follows the Rev. 0 | Page 75 of 139 AD9161/AD9162 Data Sheet Mix-Mode roll-off due to the zero-order hold at 2 × DAC clock (see Figure 183). 0 NRZ MODE 2× NRZ MODE MIX-MODE RZ MODE –3 –6 –9 –15 The clock input has a register that adjusts the phase of the CLK+ and CLK− inputs. This register is located at Address 0x07F. The register has a signed magnitude (1 is negative) value that adds capacitance at 20 fF per step to either the CLK+ or the CLK− input, according to Table 42. The CLK_PHASE_TUNE register can be used to adjust the clock input phase for better DAC image rejection. –18 –21 –24 –27 –30 0 1020 2040 3060 4080 5100 6120 7140 8160 9180 10200 FREQUENCY (MHz) Figure 183. Sinc Roll-Off with 2× NRZ Mode Added, fDAC = 5.1 GSPS CLOCK INPUT The AD9161/AD9162 contain a low jitter, differential clock receiver that is capable of interfacing directly to a differential or single-ended clock source. Because the input is self biased with a nominal impedance of 90 Ω, it is recommended that the clock source be ac-coupled to the CLK± input pins. The nominal differential input is 1 V p-p, but the clock receiver can operate with a span that ranges from 250 mV p-p to 2.0 V p-p. Better phase noise performance is achieved with a higher clock input level. DUTY CYCLE RESTORER CLK+ CROSS CONTROL TO DAC AND DLL CLK– 5kΩ 5kΩ 1.25V 40kΩ 14379-116 16µA Table 42. CLK± Phase Adjust Values Register 0x07F, Bits[5:0] 000000 000001 000010 … 011111 100000 100001 100010 … 111111 Capacitance at CLK+ 0 1 × 20 fF 2 × 20 fF … 31 × 20 fF 0 0 0 … 0 Capacitance at CLK− 0 0 0 … 0 0 1 × 20 fF 2 × 20 fF … 31 × 20 fF The improvement in performance from making these adjustments depends on the accuracy of the balance of the clock input balun and varies from unit to unit. Thus, if a high level of image rejection is required, it is likely that a per unit calibration is necessary. Performing this calibration can yield significant improvements, as much as 20 dB additional rejection of the image due to imbalance. Figure 185 shows the results of tuning clock phase, duty cycle (left at default in this case), and cross control. The improvement to performance, particularly at higher frequencies, can be as much as 20 dB. –20 Figure 184. Clock Input The quality of the clock source, as well as its interface to the AD9161/AD9162 clock input, directly impacts ac performance. Select the phase noise and spur characteristics of the clock source to meet the target application requirements. Phase noise and spurs at a given frequency offset on the clock source are directly translated to the output signal. It can be shown that the phase noise characteristics of a reconstructed output sine wave are related to the clock source by 20 × log10 (fOUT/fCLK) when the DAC clock path contribution is negligible. Figure 187 shows a clock source based on the ADF4355 low phase noise/jitter PLL. The ADF4355 can provide output frequencies from 54 MHz up to 6.8 GHz. The clock control registers exist at Address 0x082 through Address 0x084. CLK_DUTY (Register 0x082) can be used to –30 –40 PHASE 0, CROSS 6 –50 –60 –70 PHASE 28, CROSS 10 –80 –90 0 1000 2000 3000 fOUT (MHz) 4000 5000 6000 14379-221 –36 14379-193 –33 DAC OUTPUT IMAGE POWER (fS – fOUT) (dBc) POWER (dBc) –12 enable duty cycle correction (Bit 7), enable duty cycle offset control (Bit 6), and set the duty cycle offset (Bits[4:0]). The duty cycle offset word is a signed magnitude word, with Bit 4 being the sign bit (1 is negative) and Bits[3:0] the magnitude. The duty cycle adjusts across a range of approximately ±3%. Recommended settings for this register are listed in the Start-Up Sequence section. Figure 185. Performance Improvement from Tuning the Clock Input Rev. 0 | Page 76 of 139 Data Sheet AD9161/AD9162 SHUFFLE MODE The spurious performance of the AD9161/AD9162 can be improved with a feature called shuffle mode. Shuffle mode uses proprietary technology to spread the energy of spurious signals across the DAC output as random noise. Shuffle mode is enabled by programming Register 0x151, Bit 2 = 0b1. Because shuffle is implemented with the MSBs, it is more effective when the DAC is operated with a small amount of digital backoff. The amount of noise rise caused by shuffle mode is directly related to the power in the affected spurious signals. Because the AD9161/AD9162 have good spurious performance without shuffle active, the penalty of shuffle mode to the noise spectral density is typically about 1 dB to 3 dB. Shuffle mode reduces spurious performance related to clock and foldback spurs, but does not affect real harmonics of the DAC output. Examples of the effects of shuffle mode are given in the Typical Performance Characteristics section (see Figure 17, Figure 18, Figure 32, Figure 33, Figure 99, Figure 100, Figure 114, Figure 115, and Figure 116). DLL The CLK± input goes to a high frequency DLL to ensure robust locking of the DAC sample clock to the input clock. The DLL is configured and enabled as part of the recommended start-up sequence. The DLL control registers are located at Register 0x090 through Register 0x09B. The DLL settings are determined during product characterization and are given in the recommended start-up sequence (see the Start-Up Sequence section). It is not normally necessary to change these values, nor is the product characterization data valid on any settings other than the recommended ones. VOLTAGE REFERENCE The AD9161/AD9162 output current is set by a combination of digital control bits and the ISET reference current, as shown in Figure 186. The reference current is obtained by forcing the band gap voltage across an external 9.6 kΩ resistor from ISET (Ball A15 on the 8 mm × 8 mm package and Ball A12 on the 11 mm × 11 mm package) to VNEG_N1P2. The 1.2 V nominal band gap voltage (VREF) generates a 125 µA reference current, ISET, in the 9.6 kΩ resistor, RSET. The maximum full-scale current setting is related to the external resistor by the following equation: IOUTFS = 1.2 V/RSET × 320 (mA) Note the following constraints when configuring the voltage reference circuit: • • • • • The IOUTFS value can be adjusted digitally over an 8 mA to 40 mA range by the ANA_FULL_SCALE_CURRENT[9:0] bits (Register 0x042, Bits[7:0] and Register 0x041, Bits[1:0]). The following equation relates IOUTFS to the ANA_FULL_SCALE_ CURRENT[9:0] bits, which can be set from 0 to 1023. IOUTFS = 32 mA × (ANA_FULL_SCALE_CURRENT[9:0]/1023) + 8 mA Note that the default value of 0x3FF generates 40 mA full scale, and this value is used for most of the characterization presented in this data sheet, unless noted otherwise. AD9161/AD9162 VBG 1.2V VREF ISET 1µF VSS DAC + CURRENT SCALING IOUTFS ISET VNEG_N1P2 14379-119 9.6kΩ ANA_FULL_SCALE_CURRENT [9:0] – Both the 9.6 kΩ resistor and 1 µF bypass capacitor are required for proper operation. Adjusting the DAC output full-scale current, IOUTFS, from its default setting of 40 mA must be performed digitally. The AD9161/AD9162 are not multiplying DACs. Modulation of the reference current, ISET, with an ac signal is not supported. The band gap voltage appearing at the VREF pin must be buffered for use with an external circuitry because it has a high output impedance. An external reference can be used to overdrive the internal reference by connecting it to the VREF pin. Figure 186. Voltage Reference Circuit Rev. 0 | Page 77 of 139 AD9161/AD9162 Data Sheet VOUT AD9161/ AD9162 ADF4355 7.4nH 100pF CLK+ PLL OUTPUT STAGE VCO 100pF CLK– FREF 14379-174 7.4nH VOUT 2GHz TO 6GHz 0dBm Figure 187. Possible Signal Chain for CLK± Input The AD9161/AD9162 provide complementary current outputs, OUTPUT+ and OUTPUT−, that sink current from an external load that is referenced to the 2.5 V VDD25_DAC supply. Figure 188 shows an equivalent output circuit for the DAC. Compared to most current output DACs of this type, the outputs of the AD9161/AD9162 consist of a constant current (IFIXED), and a peak differential ac current, ICS (ICS = ICSP + ICSN). These two currents combine to form the IINTx currents shown in Figure 188. The internal currents, IINTP and IINTN, are sent to the output pin and to an input termination resistance equivalent to 100 Ω pulled to the VDD25_DAC supply (RINT). This termination serves to divide the output current based on the external termination resistors that are pulled to VDD25_DAC. VDD25_DAC IOUTFS = 8mA – 40mA 100Ω ICSP IINTP IFIXED IFIXED IINTN 40 35 30 25 20 15 10 5 0 0 16384 32768 DAC CODE OUTPUT+ OUTPUT– 100Ω VDD25_DAC 49152 65536 Figure 189. Gain Curve for ANA_FULL_SCALE_CURRENT[9:0] = 1023, DAC Offset = 3.8 mA Peak DAC Output Power Capability 14379-120 ICSN 45 14379-121 Equivalent DAC Output and Transfer Function external load, and thus must be ac-coupled through appropriately sized capacitors for the chosen operating frequencies. Figure 189 shows the OUTPUT+ vs. DAC code transfer function when IOUTFS is set to 40 mA. OUTPUT CURRENT (mA) ANALOG OUTPUTS Figure 188. Equivalent DAC Output Circuit The example shown in Figure 188 can be modeled as a pair of dc current sources that source a current of IOUT to each output. This differential ac current source is used to model the signal (that is, a digital code) dependent nature of the DAC output. The polarity and signal dependency of this ac current source are related to the digital code (F) by the following equation: F (code) = (DACCODE – 32,768)/32,768 (2) where: −1 ≤ F (code) < +1. DACCODE = 0 to 65,535 (decimal). To calculate the rms power delivered to the load, consider the following: The current that is measured at the OUTPUT+ and OUTPUT− outputs is as follows: OUTPUT+ = (IFIXED (mA) + (F × IOUTFS)/FMAX(mA)) × (RINT/(RINT + RLOAD) The maximum peak power capability of a differential current output DAC is dependent on its peak differential ac current, IPEAK, and the equivalent load resistance it sees. In the case of a 1:1 balun with 100 Ω differential source termination, the equivalent load that is seen by the DAC ac current source is 50 Ω. If the AD9161/AD9162 are programmed for an IOUTFS = 40 mA, its ideal peak ac current is 20 mA and its maximum power, delivered to the equivalent load, is 10 × (RINT/(RINT + RLOAD) = 8 mW (that is, P = I2R). Because the source and load resistance seen by the 1:1 balun are equal, this power is shared equally. Therefore, the output load receives 4 mW, or 6 dBm maximum power. (3) OUTPUT− = (IFIXED (mA) + ((FMAX − F) × IOUTFS)/FMAX(mA)) ×(RINT/(RINT + RLOAD) The IFIXED value is about 3.8 mA. It is important to note that the AD9161/AD9162 output cannot support dc coupling to the • • • • Peak to rms of the digital waveform Any digital backoff from digital full scale DAC sinc response and nonideal losses in the external network DAC analog roll-off due to switch parasitic capacitance and load impedance For example, a sine wave with no digital backoff ideally measures 6 dBm. If a typical balun loss of 1.2 dB is included, expect to Rev. 0 | Page 78 of 139 Data Sheet AD9161/AD9162 Output Stage Configuration The AD9161/AD9162 are intended to serve high dynamic range applications that require wide signal reconstruction bandwidth (such as a DOCSIS cable modem termination system (CMTS)) and/or high IF/RF signal generation. Optimum ac performance can be realized only when the DAC output is configured for differential (that is, balanced) operation with its output commonmode voltage biased to a stable, low noise 2.5 V nominal analog supply (VDD25_DAC). The output network used to interface to the DAC provides a near 0 Ω dc bias path to VDD25_DAC. Any imbalance in the output impedance over frequency between the OUTPUT+ and OUTPUT− pins degrades the distortion performance (mostly even order) and noise performance. Component selection and layout are critical in realizing the performance potential of the AD9161/AD9162. To assist in matching the AD9161/AD9162 output, an equivalent model of the output was developed, and is shown in Figure 192. This equivalent model includes all effects from the ideal 40 mA current source in the die to the ball of the CSP_ BGA package, including parasitic capacitance, trace inductance and resistance, contact resistance of solder bumps, via inductance, and other effects. 470pH OUTPUT– 248fF 1.14pF 470pH OUTPUT+ 3.59Ω Figure 192. Equivalent Circuit Model of the DAC Output A Smith chart is provided in Figure 193 showing the simulated S11 of the DAC output, using the model in Figure 192. The plot was taken using the circuit in Figure 192, with an ideal balun. For the measured response of the DAC output, see Figure 191. 1.0 2.0 0.5 Most applications that require balanced to unbalanced conversion from 10 MHz to 3 GHz can take advantage of several available transformers that offer impedance ratios of both 2:1 and 1:1. m1 0.2 5.0 S (1, 1) m6 Figure 190 shows the AD9161/AD9162 interfacing to the MiniCircuits TCM1-63AX+ and the TC1-1-43X+ transformers. m2 0 0 m5 m3 m4 MINI-CIRCUITS TCM1-63AX+ TC1-1-43X+ OUTPUT+ 179Ω 40mA 3.59Ω 14379-124 measure 4.8 dBm of actual power in the region where the sinc response of the DAC has negligible influence and analog roll-off has not begun. Increasing the output power is best accomplished by increasing IOUTFS. An example of DAC output characteristics for several balun and board types is shown in Figure 191. –5.0 –0.2 L 50Ω C L 50Ω C 14379-122 VDD25_DAC –0.5 –1.0 FREQUENCY (10MHz TO 6GHz) 5 m1 FREQUENCY = 10MHz S (1, 1) = 0.770/149.556 IMPEDANCE = Z0 × (0.140 + j0.267) m4 FREQUENCY = 2GHz S (1, 1) = 0.583/–148.777 IMPEDANCE = Z0 × (0.282 – j0.259) 0 m2 FREQUENCY = 100MHz S (1, 1) = 0.227/163.083 IMPEDANCE = Z0 × (0.638 + j0.089) m5 FREQUENCY = 4GHz S (1, 1) = 0.794/–170.517 IMPEDANCE = Z0 × (0.116 – j0.082) m3 FREQUENCY = 1GHz S (1, 1) = 0.367/–144.722 IMPEDANCE = Z0 × (0.499 – j0.245) m6 FREQUENCY = 6GHz S (1, 1) = 0.779/168.448 IMPEDANCE = Z0 × (0.125 + j0.100) –5 –10 Figure 193. Simulated Smith Chart Showing the DAC Output Impedance –15 BAL-0006 TC1-1-43X+ TCM1-63AX+ –20 0 1 2 3 4 5 fOUT (GHz) 6 14379-123 OUTPUT POWER (dBm) Figure 190. Recommended Transformer for Wideband Applications with Upper Bandwidths of up to 5 GHz –2.0 14379-125 OUTPUT– Figure 191. Measured DAC Output Response; fDAC = 6 GSPS Rev. 0 | Page 79 of 139 AD9161/AD9162 Data Sheet START-UP SEQUENCE A number of steps is required to program the AD9161/AD9162 to the proper operating state after the device is powered up. This sequence is divided into several steps, and is listed in Table 43, Table 44, and Table 45, along with an explanation of the purpose of each step. Private registers are reserved but must be written for proper operation. Blank cells in Table 43 to Table 45 mean that the value depends on the result as described in the description column. The AD9161/AD9162 are calibrated at the factory as part of the automatic test program. The configure DAC start-up sequence loads the factory calibration coefficients, as well as configures some parameters that optimize the performance of the DAC and the DAC clock DLL (see Table 43). Run this sequence whenever the DAC is powered down or reset. The configure JESD204B sequence configures the SERDES block and then brings up the links (see Table 44). First, run the configure DAC start-up sequence, then run the configure JESD204B sequence. Follow the configure NCO sequence if using the NCO (see Table 45). Note that the NCO can be used in NCO only mode or in conjunction with synthesized data from the SERDES data interface. Only one mode can be used at a time and this mode is selected in the second step in Table 45. The configure DAC start-up sequence is run first, then the configure NCO sequence. Table 43. Configure DAC Start-Up Sequence After Power-Up R/W W W W W W W R Value 0x18 0x52 0xD2 0x02 0x00 0x01 N/A 1 R Register 0x000 0x0D2 0x0D2 0x606 0x607 0x604 0x003, 0x004, 0x005, 0x006 0x604, Bit 1 W W W W W W W R 0x058 0x090 0x080 0x040 0x020 0x09E 0x091 0x092, Bit 0 0x03 0x1E 0x00 0x00 0x0F 0x85 0xE9 0b1 W W 0x0E8 0x152, Bits[1:0] 0x20 1 0b1 Description Configure the device for 4-wire serial port operation (optional: leave at the default of 3-wire SPI) Reset internal calibration registers (private) Clear the reset bit for the internal calibration registers (private) Configure the nonvolatile random access memory (NVRAM) (private) Configure the NVRAM (private) Load the NVRAM. Loads factory calibration factors from the NVRAM. (private) (Optional) read CHIP_TYPE, PROD_ID[15:0], PROD_GRADE, and DEV_REVISION from Register 0x003, Register 0x004, Register 0x005, and Register 0x006 (Optional) read the boot loader pass bit in Register 0x604, Bit 1 = 0b1 to indicate a successful boot load Enable the band gap reference (private) Power up the DAC clock DLL Enable the clock receiver Enable the DAC bias circuits Optional. Enable the interrupts Configure DAC analog parameters (private) Enable the DAC clock DLL Check DLL_STATUS; set Register 0x092, Bit 0 = 1 to indicate the DAC clock DLL is locked to the DAC clock input Enable calibration factors (private) Configure the DAC decode mode (0b00 = NRZ, 0b01 = Mix-Mode, or 0b10 = RZ) N/A means not applicable. Table 44. Configure JESD204B Start-Up Sequence R/W W W W W W W W W W W W W Register 0x300 0x4B8 0x4B9 0x480 0x481 0x482 0x483 0x484 0x485 0x486 0x487 0x110 Value 0x00 0xFF 0x01 0x38 0x38 0x38 0x38 0x38 0x38 0x38 0x38 Description Ensure the SERDES links are disabled before configuring them. Enable JESD204B interrupts. Enable JESD204B interrupts. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Enable SERDES error counters. Configure number of lanes (Bits[7:4]) and interpolation rate (Bits[3:0]). Rev. 0 | Page 80 of 139 Data Sheet AD9161/AD9162 R/W W Register 0x111 W W W W W W W W R 0x230 0x289, Bits[1:0] 0x084, Bits[5:4] 0x200 0x475 0x453, Bit 7 0x458, Bits[7:5] 0x459, Bits[7:5] 0x45D 0x475 0x201, Bits[7:0] 0x2A7 0x2AE 0x29E 0x206 0x206 0x280 0x281, Bit 0 0x01 0x01 0x1F 0x00 0x01 0x03 0b1 W R R R R W W W 0x300 0x470 0x471 0x472 0x473 0x024 0x4BA 0x4BB 0x01 0xFF 0xFF 0xFF 0xFF 0x1F 0xFF 0x01 W W W W W W W W W Value Description Configure the datapath options for Bit 7 (INVSINC_EN), Bit 6 (NCO_EN), Bit 4 (FILT_BW), Bit 2 (MODULUS_EN), Bit 1 (SEL_SIDEBAND), and Bit 0 (FIR85_FILT_EN). See the Register Summary section for details on the options. Set the reserved bits (Bit 5 and Bit 3) to 0b0. Configure the CDR block according to Table 20 for both half rate enable and the divider. Set up the SERDES PLL divider based on the conditions shown in Table 19. Set up the PLL reference clock rate based on the conditions shown in Table 19. 0x00 0x09 0b1 Enable JESD204B block (disable master SERDES power-down). Soft reset the JESD204B quad-byte deframer. (Optional) Enable scrambling on SERDES lanes. Set the subclass type: 0b000 = Subclass 0, 0b001 = Subclass 1. 0b1 Set the JESD204x version to JESD204B. 0x01 Program the calculated checksum value for Lane 0 from values in Register 0x450 to Register 0x45C. Bring the JESD204B quad-byte deframer out of reset. Set any bits to 1 to power down the appropriate physical lane. (Optional) Calibrate SERDES PHY Termination Block 1 (PHY 0, PHY 1, PHY 6, PHY 7). (Optional) Calibrate SERDES PHY Termination Block 2 (PHY 2, PHY 3, PHY 4, PHY 5). Override defaults in the SERDES PLL settings (private). Reset the CDR. Enable the CDR. Enable the SERDES PLL. Read back Register 0x281 until Bit 0 = 1 to indicate the SERDES PLL is locked. Prior to enabling the links, be sure that the JESD204B transmitter is enabled and ready to begin bringing up the link. Enable SERDES links (begin bringing up the link). Read the CGS status for all lanes. Read the frame sync status for all lanes. Read the good checksum status for all lanes. Read the initial lane sync status for all lanes. Clear the interrupts. Clear the SERDES interrupts. Clear the SERDES interrupt. Table 45. Configure NCO Sequence R/W W W Register 0x110 0x111, Bit 6 W W W W W W W W W W W 0x150, Bit 1 0x14E 0x14F 0x113 0x119 0x118 0x117 0x116 0x115 0x114 0x113 Value 0x80 0b1 0x00 0x01 Description (Optional) Perform this write if NCO only mode is desired (AD9162 only). Configure NCO_EN (Bit 6) = 0b1. Configure other datapath options for Bit 7 (INVSINC_EN), Bit 4 (FILT_BW), Bit 2 (MODULUS_EN), Bit 1 (SEL_SIDEBAND), and Bit 0 (FIR85_FILT_EN). See the Register Summary section for details on the options. Set the reserved bits (Bit 5 and Bit 3) to 0b0. Configure the DC_TEST_EN bit: 0b0 = NCO operation with data interface; 0b1 = NCO only mode (AD9162 only). Write the amplitude value for tone amplitude in NCO only mode, Bits [15:8] (AD9162 only). Write the amplitude value for tone amplitude in NCO only mode, Bits [7:0] (AD9162 only). Ensure the frequency tuning word write request is low. Write FTW, Bits[47:40]. Write FTW, Bits[39:32]. Write FTW, Bits[31:24]. Write FTW, Bits[23:16]. Write FTW, Bits[15:8]. Write FTW, Bits[7:0]. Load the FTW to the NCO. Rev. 0 | Page 81 of 139 AD9161/AD9162 Data Sheet REGISTER SUMMARY Table 46. Register Summary Reg. Name Bits Bit 7 Bit 6 Bit 5 Bit 4 0x000 SPI_INTFCONFA [7:0] SOFTRESET_ M LSBFIRST_M ADDRINC_M SDOACTIVE_ SDOACTIVE M Bit 3 0x001 SPI_INTFCONFB [7:0] SINGLEINS CSSTALL 0x002 SPI_DEVCONF [7:0] 0x003 SPI_CHIPTYPE [7:0] CHIP_TYPE 0x00 R 0x004 SPI_PRODIDL [7:0] PROD_ID[7:0] 0x00 R 0x005 SPI_PRODIDH [7:0] 0x006 SPI_CHIPGRADE [7:0] 0x020 IRQ_ENABLE [7:0] RESERVED EN_SYSREF_ JITTER 0x024 IRQ_STATUS [7:0] RESERVED IRQ_SYSREF_ IRQ_DATA_ JITTER READY 0x031 SYNC_LMFC_ DELAY_FRAME [7:0] RESERVED 0x032 SYNC_LMFC_ DELAY0 [7:0] 0x033 SYNC_LMFC_ DELAY1 [7:0] 0x034 SYNC_LMFC_ STAT0 [7:0] 0x035 SYNC_LMFC_ STAT1 [7:0] 0x036 SYSREF_COUNT [7:0] SYSREF_COUNT 0x00 R/W 0x037 SYSREF_PHASE0 [7:0] SYSREF_PHASE[7:0] 0x00 R/W RESERVED DEVSTATUS Bit 2 Bit 1 Bit 0 Reset RW ADDRINC LSBFIRST SOFTRESET 0x00 R/W SOFTRESET1 SOFTRESET0 RESERVED 0x00 R/W CUSTOPMODE SYSOPMODE PROD_ID[15:8] 0x00 R PROD_GRADE DEV_REVISION EN_DATA_ READY 0x00 R EN_LANE_FIFO EN_PRBSQ EN_PRBSI 0x00 R/W IRQ_LANE_ FIFO IRQ_PRBSI 0x00 R/W IRQ_PRBSQ SYNC_LMFC_DELAY_SET_FRM 0x00 R/W SYNC_LMFC_DELAY_SET[7:0] 0x00 R/W RESERVED SYNC_LMFC_DELAY_SET[11:8] SYNC_LMFC_DELAY_STAT[7:0] RESERVED SYSREF_PHASE1 [7:0] 0x039 SYSREF_JITTER_ WINDOW [7:0] RESERVED SYSREF_PHASE[11:8] 0x03A SYNC_CTRL [7:0] 0x03F TX_ENABLE [7:0] 0x040 ANA_DAC_BIAS_ PD [7:0] RESERVED 0x041 ANA_FSC0 [7:0] RESERVED 0x042 ANA_FSC1 [7:0] 0x07F CLK_PHASE_TUNE [7:0] 0x080 CLK_PD [7:0] 0x082 CLK_DUTY [7:0] CLK_DUTY_ EN 0x083 CLK_CRS_CTRL [7:0] CLK_CRS_EN 0x084 PLL_REF_CLK_PD [7:0] 0x088 SYSREF_CTRL0 [7:0] 0x089 SYSREF_CTRL1 [7:0] 0x090 DLL_PD [7:0] 0x091 DLL_CTRL [7:0] 0x092 DLL_STATUS [7:0] 0x093 DLL_GB [7:0] 0x094 DLL_COARSE [7:0] 0x095 DLL_FINE [7:0] 0x096 DLL_PHASE [7:0] RESERVED 0x097 DLL_BW [7:0] RESERVED RESERVED SPI_ DATAPATH_ PRE 0x00 R/W SYNC_MODE RESERVED TXEN_NCO_ TXEN_ DATAPATH_ RESET POST TXEN_DAC_FSC 0xC0 R/W ANA_DAC_ BIAS_PD1 ANA_DAC_BIAS_ PD0 0x03 R/W ANA_FULL_SCALE_CURRENT[1:0] 0x03 R/W RESERVED 0xFF R/W CLK_PHASE_TUNE 0x00 R/W RESERVED DACCLK_PD CLK_DUTY_ BOOST_EN CLK_DUTY_PRG RESERVED RESERVED 0x80 R/W RESERVED RESERVED HYS_ON PLL_REF_CLK_PD SYSREF_RISE HYS_CNTRL[9:8] HYS_CNTRL[7:0] DLL_FINE_ DC_EN DLL_TRACK_ DLL_SEARCH_ DLL_SLOPE ERR ERR DLL_FINE_ XC_EN DLL_COARSE_ DC_EN DLL_COARSE_ DLL_CLK_PD XC_EN DLL_MODE DLL_FAIL RESERVED DLL_LOST DLL_LOCKED 0x00 R/W 0x00 R/W DLL_COARSE 0x00 R/W 0x80 R/W DLL_PHS DLL_FILT_BW 0x1F R/W 0xF0 R/W DLL_FINE Rev. 0 | Page 82 of 139 0x00 R/W DLL_ENABLE DLL_GUARD RESERVED 0x00 R/W 0x00 R/W DLL_SEARCH RESERVED 0x01 R/W 0x80 R/W CLK_CRS_ADJ PLL_REF_CLK_RATE RESERVED 0x00 R/W TXEN_ DATAPATH_ PRE ANA_FULL_SCALE_CURRENT[9:2] CLK_DUTY_ OFFSET_EN 0x00 R/W 0x00 R/W SYSREF_JITTER_WINDOW RESERVED SPI_ DATAPATH_ POST 0x00 R/W 0x00 R/W SYNC_LMFC_DELAY_STAT[11:8] 0x038 0x00 R/W 0x08 R/W DLL_WEIGHT 0x00 R/W Data Sheet AD9161/AD9162 Reg. Name Bits Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 0x098 DLL_READ [7:0] 0x099 DLL_COARSE_RB [7:0] 0x09A DLL_FINE_RB [7:0] 0x09B DLL_PHASE_RB [7:0] 0x09D DIG_CLK_INVERT [7:0] 0x0A0 DLL_CLK_DEBUG [7:0] 0x110 INTERP_MODE [7:0] 0x111 DATAPATH_CFG [7:0] INVSINC_EN 0x113 FTW_UPDATE [7:0] RESERVED 0x114 FTW0 [7:0] FTW[7:0] 0x00 R/W 0x115 FTW1 [7:0] FTW[15:8] 0x00 R/W 0x116 FTW2 [7:0] FTW[23:16] 0x00 R/W 0x117 FTW3 [7:0] FTW[31:24] 0x00 R/W 0x118 FTW4 [7:0] FTW[39:32] 0x00 R/W 0x119 FTW5 [7:0] FTW[47:40] 0x00 R/W 0x11C PHASE_OFFSET0 [7:0] NCO_PHASE_OFFSET[7:0] 0x00 R/W 0x11D PHASE_OFFSET1 [7:0] NCO_PHASE_OFFSET[15:8] 0x00 R/W 0x124 ACC_MODULUS0 [7:0] ACC_MODULUS[7:0] 0x00 R/W 0x125 ACC_MODULUS1 [7:0] ACC_MODULUS[15:8] 0x00 R/W 0x126 ACC_MODULUS2 [7:0] ACC_MODULUS[23:16] 0x00 R/W 0x127 ACC_MODULUS3 [7:0] ACC_MODULUS[31:24] 0x00 R/W 0x128 ACC_MODULUS4 [7:0] ACC_MODULUS[39:32] 0x00 R/W 0x129 ACC_MODULUS5 [7:0] ACC_MODULUS[47:40] 0x00 R/W 0x12A ACC_DELTA0 [7:0] ACC_DELTA[7:0] 0x00 R/W 0x12B ACC_DELTA1 [7:0] ACC_DELTA[15:8] 0x00 R/W 0x12C ACC_DELTA2 [7:0] ACC_DELTA[23:16] 0x00 R/W 0x12D ACC_DELTA3 [7:0] ACC_DELTA[31:24] 0x00 R/W 0x12E ACC_DELTA4 [7:0] ACC_DELTA[39:32] 0x00 R/W 0x12F ACC_DELTA5 [7:0] ACC_DELTA[47:40] 0x14B PRBS [7:0] 0x14C PRBS_ERROR_I [7:0] PRBS_COUNT_I 0x14D PRBS_ERROR_Q [7:0] PRBS_COUNT_Q 0x14E TEST_DC_DATA1 [7:0] DC_TEST_DATA[15:8] 0x14F TEST_DC_DATA0 [7:0] 0x150 DIG_TEST [7:0] 0x151 DECODE_CTRL [7:0] 0x152 DECODE_MODE [7:0] 0x1DF SPI_STRENGTH [7:0] 0x200 MASTER_PD [7:0] 0x201 PHY_PD [7:0] 0x203 GENERIC_PD [7:0] 0x206 CDR_RESET [7:0] 0x230 CDR_OPERATING_ [7:0] MODE_REG_0 0x250 EQ_CONFIG_PHY_ [7:0] 0_1 SPI_EQ_CONFIG1 SPI_EQ_CONFIG0 0x88 R/W 0x251 EQ_CONFIG_PHY_ [7:0] 2_3 SPI_EQ_CONFIG3 SPI_EQ_CONFIG2 0x88 R/W 0x252 EQ_CONFIG_PHY_ [7:0] 4_5 SPI_EQ_CONFIG5 SPI_EQ_CONFIG4 0x88 R/W RESERVED DLL_READ RESERVED DLL_COARSE_RB 0x00 R DLL_PHS_RB DLL_TEST_EN INV_DIG_CLK 0x00 R DIG_CLK_DC_ EN RESERVED DIG_CLK_XC_EN DLL_TEST_DIV JESD_LANES NCO_EN INTERP_MODE RESERVED FILT_BW FTW_REQ_MODE PRBS_GOOD_ PRBS_GOOD_I RESERVED Q 0x00 R/W 0x81 R/W MODULUS_EN SEL_SIDEBAND FIR85_FILT_EN 0x00 R/W RESERVED FTW_LOAD_ SYSREF FTW_LOAD_ ACK 0x00 R/W PRBS_INV_Q PRBS_INV_I FTW_LOAD_REQ 0x00 R/W PRBS_MODE PRBS_RESET PRBS_EN 0x00 R 0x00 R/W 0x00 R/W DC_TEST_EN RESERVED SHUFFLE RESERVED RESERVED RESERVED RESERVED 0x01 R/W 0x00 R/W SPIDRV 0x0F R/W SPI_PD_MASTER SPI_PD_PHY RESERVED Rev. 0 | Page 83 of 139 0x01 R/W 0x00 R/W SPI_SYNC1_PD RESERVED RESERVED 0x00 R/W DECODE_MODE RESERVED RESERVED 0x10 R/W 0x00 R DC_TEST_DATA[7:0] SPI_ ENHALFRATE 0x03 R/W RESERVED RESERVED RESERVED 0x00 R/W 0x00 R DLL_FINE_RB RESERVED RESERVED Reset RW SPI_DIVISION_RATE 0x00 R/W SPI_CDR_RESET 0x01 R/W RESERVED 0x28 R/W AD9161/AD9162 Bits Data Sheet Reg. Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 0x253 EQ_CONFIG_PHY_ [7:0] 6_7 0x268 EQ_BIAS_REG [7:0] 0x280 SYNTH_ENABLE_ CNTRL [7:0] 0x281 PLL_STATUS [7:0] 0x289 REF_CLK_ DIVIDER_LDO [7:0] 0x2A7 TERM_BLK1_ CTRLREG0 [7:0] 0x2A8 TERM_BLK1_ CTRLREG1 [7:0] 0x2AC TERM_BLK1_RD_ REG0 [7:0] 0x2AE TERM_BLK2_ CTRLREG0 [7:0] 0x2AF TERM_BLK2_ CTRLREG1 [7:0] 0x2B3 TERM_BLK2_RD_ REG0 [7:0] 0x2BB TERM_OFFSET_0 [7:0] RESERVED TERM_OFFSET_0 0x00 R/W 0x2BC TERM_OFFSET_1 [7:0] RESERVED TERM_OFFSET_1 0x00 R/W 0x2BD TERM_OFFSET_2 [7:0] RESERVED TERM_OFFSET_2 0x00 R/W 0x2BE TERM_OFFSET_3 [7:0] RESERVED TERM_OFFSET_3 0x00 R/W 0x2BF TERM_OFFSET_4 [7:0] RESERVED TERM_OFFSET_4 0x00 R/W 0x2C0 TERM_OFFSET_5 [7:0] RESERVED TERM_OFFSET_5 0x00 R/W 0x2C1 TERM_OFFSET_6 [7:0] RESERVED TERM_OFFSET_6 0x00 R/W 0x2C2 TERM_OFFSET_7 [7:0] 0x300 GENERAL_JRX_ CTRL_0 [7:0] 0x302 DYN_LINK_ LATENCY_0 [7:0] RESERVED DYN_LINK_LATENCY_0 0x304 LMFC_DELAY_0 [7:0] RESERVED LMFC_DELAY_0 0x00 R/W 0x306 LMFC_VAR_0 [7:0] RESERVED LMFC_VAR_0 0x1F R/W 0x308 XBAR_LN_0_1 [7:0] RESERVED SRC_LANE1 SRC_LANE0 0x08 R/W 0x309 XBAR_LN_2_3 [7:0] RESERVED SRC_LANE3 SRC_LANE2 0x1A R/W 0x30A XBAR_LN_4_5 [7:0] RESERVED SRC_LANE5 SRC_LANE4 0x2C R/W 0x30B XBAR_LN_6_7 [7:0] RESERVED SRC_LANE7 SRC_LANE6 0x3E R/W 0x30C FIFO_STATUS_ REG_0 [7:0] LANE_FIFO_FULL 0x00 R 0x30D FIFO_STATUS_ REG_1 [7:0] LANE_FIFO_EMPTY 0x00 R 0x311 SYNC_GEN_0 [7:0] 0x312 SYNC_GEN_1 [7:0] 0x313 SYNC_GEN_3 [7:0] LMFC_PERIOD 0x00 R 0x315 PHY_PRBS_TEST_ [7:0] EN PHY_TEST_EN 0x00 R/W 0x316 PHY_PRBS_TEST_ [7:0] CTRL 0x317 PHY_PRBS_TEST_ [7:0] THRESHOLD_ LOBITS PHY_PRBS_THRESHOLD_LOBITS 0x00 R/W 0x318 PHY_PRBS_TEST_ [7:0] THRESHOLD_ MIDBITS PHY_PRBS_THRESHOLD_MIDBITS 0x00 R/W SPI_EQ_CONFIG7 Bit 0 SPI_EQ_CONFIG6 EQ_POWER_MODE 0x88 R/W RESERVED RESERVED RESERVED SPI_CP_ OVER_ RANGE_ HIGH_RB 0x62 R/W SPI_RECAL_ SYNTH SPI_CP_ OVER_ RANGE_ LOW_RB SPI_CP_ CAL_VALID_ RB RESERVED RESERVED RESERVED SPI_ENABLE_ SYNTH RESERVED SPI_I_TUNE_R_ CAL_TERMBLK1 SPI_I_SERIALIZER_RTRIM_TERMBLK1 RESERVED SPI_I_TUNE_R_ CAL_TERMBLK2 SPI_I_SERIALIZER_RTRIM_TERMBLK2 RESERVED RESERVED RESERVED PHY_SRC_ERR_CNT EOMF_MASK_0 RESERVED Rev. 0 | Page 84 of 139 PHY_TEST_ START 0x00 R/W 0x00 R EOF_MASK_0 0x00 R/W 0x00 R/W SYNC_SYNCREQ_DUR PHY_PRBS_PAT_SEL 0x00 R/W 0x00 R/W LINK_EN SYNC_ERR_DUR 0x00 R 0x00 R TERM_OFFSET_7 CHECKSUM_ MODE 0x00 R/W 0x00 R/W SPI_O_RCAL_CODE_TERMBLK2 RESERVED RESERVED 0x04 R/W 0x00 R/W SPI_O_RCAL_CODE_TERMBLK1 RESERVED 0x00 R/W SPI_PLL_LOCK_RB 0x00 R SERDES_PLL_DIV_FACTOR RESERVED Reset RW PHY_TEST_RESET 0x00 R/W Data Sheet AD9161/AD9162 Reg. Name Bits Bit 7 0x319 PHY_PRBS_TEST_ [7:0] THRESHOLD_ HIBITS PHY_PRBS_THRESHOLD_HIBITS 0x31A PHY_PRBS_TEST_ [7:0] ERRCNT_LOBITS PHY_PRBS_ERR_CNT_LOBITS 0x00 R 0x31B PHY_PRBS_TEST_ [7:0] ERRCNT_MIDBITS PHY_PRBS_ERR_CNT_MIDBITS 0x00 R 0x31C PHY_PRBS_TEST_ [7:0] ERRCNT_HIBITS PHY_PRBS_ERR_CNT_HIBITS 0x00 R 0x31D PHY_PRBS_TEST_ [7:0] STATUS PHY_PRBS_PASS 0xFF R 0x31E PHY_DATA_ SNAPSHOT_CTRL 0x31F PHY_SNAPSHOT_ [7:0] DATA_BYTE0 PHY_SNAPSHOT_DATA_BYTE0 0x00 R 0x320 PHY_SNAPSHOT_ [7:0] DATA_BYTE1 PHY_SNAPSHOT_DATA_BYTE1 0x00 R 0x321 PHY_SNAPSHOT_ [7:0] DATA_BYTE2 PHY_SNAPSHOT_DATA_BYTE2 0x00 R 0x322 PHY_SNAPSHOT_ [7:0] DATA_BYTE3 PHY_SNAPSHOT_DATA_BYTE3 0x00 R 0x323 PHY_SNAPSHOT_ [7:0] DATA_BYTE4 PHY_SNAPSHOT_DATA_BYTE4 0x00 R 0x32C SHORT_TPL_ TEST_0 [7:0] 0x32D SHORT_TPL_ TEST_1 [7:0] SHORT_TPL_REF_SP_LSB 0x00 R/W 0x32E SHORT_TPL_ TEST_2 [7:0] SHORT_TPL_REF_SP_MSB 0x00 R/W 0x32F SHORT_TPL_ TEST_3 [7:0] 0x334 JESD_BIT_ INVERSE_CTRL [7:0] JESD_BIT_INVERSE 0x400 DID_REG [7:0] DID_RD 0x00 R 0x401 BID_REG [7:0] BID_RD 0x00 R 0x402 LID0_REG [7:0] RESERVED 0x403 SCR_L_REG [7:0] SCR_RD 0x404 F_REG [7:0] 0x405 K_REG [7:0] 0x406 M_REG [7:0] 0x407 CS_N_REG [7:0] 0x408 NP_REG [7:0] SUBCLASSV_RD NP_RD 0x00 R 0x409 S_REG [7:0] JESDV_RD S_RD 0x00 R 0x40A HD_CF_REG [7:0] CF_RD 0x00 R 0x40B RES1_REG [7:0] RES1_RD 0x00 R 0x40C RES2_REG [7:0] RES2_RD 0x00 R 0x40D CHECKSUM0_REG [7:0] LL_FCHK0 0x00 R 0x40E COMPSUM0_REG [7:0] LL_FCMP0 0x00 R 0x412 LID1_REG 0x415 CHECKSUM1_REG [7:0] 0x416 COMPSUM1_REG [7:0] 0x41A LID2_REG 0x41D CHECKSUM2_REG [7:0] 0x41E COMPSUM2_REG [7:0] 0x422 LID3_REG 0x425 CHECKSUM3_REG [7:0] LL_FCHK3 0x00 R 0x426 COMPSUM3_REG [7:0] LL_FCMP3 0x00 R [7:0] [7:0] [7:0] [7:0] Bit 6 Bit 5 Bit 4 RESERVED Bit 3 Bit 2 Bit 1 PHY_GRAB_ MODE SHORT_TPL_M_SEL RESERVED ADJDIR_RD PHADJ_RD LL_LID0 L_RD RESERVED K_RD RESERVED RESERVED LL_LID1 LL_FCHK1 LL_FCHK2 0x00 R 0x00 R 0x00 R 0x00 R LL_FCMP2 0x00 R LL_LID3 Rev. 0 | Page 85 of 139 0x00 R 0x00 R LL_LID2 RESERVED 0x00 R 0x00 R LL_FCMP1 RESERVED 0x00 R 0x00 R N_RD RESERVED 0x00 R 0x00 R M_RD HD_RD SHORT_TPL_TEST_ 0x00 R/W EN 0x00 R/W F_RD CS_RD SHORT_TPL_ TEST_RESET PHY_GRAB_DATA 0x00 R/W SHORT_TPL_FAIL RESERVED Reset RW 0x00 R/W PHY_GRAB_LANE_SEL SHORT_TPL_SP_SEL Bit 0 0x00 R AD9161/AD9162 Data Sheet Reg. Name Bits Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 0x42A LID4_REG [7:0] 0x42D CHECKSUM4_REG [7:0] 0x42E COMPSUM4_REG [7:0] 0x432 LID5_REG 0x435 CHECKSUM5_REG [7:0] LL_FCHK5 0x436 COMPSUM5_REG [7:0] LL_FCMP5 0x43A LID6_REG 0x43D CHECKSUM6_REG [7:0] 0x43E COMPSUM6_REG [7:0] 0x442 LID7_REG 0x445 CHECKSUM7_REG [7:0] LL_FCHK7 0x00 R 0x446 COMPSUM7_REG [7:0] LL_FCMP7 0x00 R 0x450 ILS_DID [7:0] 0x451 ILS_BID [7:0] 0x452 ILS_LID0 [7:0] RESERVED 0x453 ILS_SCR_L [7:0] SCR 0x454 ILS_F [7:0] 0x455 ILS_K [7:0] 0x456 ILS_M [7:0] 0x457 ILS_CS_N [7:0] 0x458 ILS_NP [7:0] 0x459 ILS_S [7:0] 0x45A ILS_HD_CF [7:0] 0x45B ILS_RES1 [7:0] 0x45C ILS_RES2 [7:0] RES2 0x00 R/W 0x45D ILS_CHECKSUM [7:0] FCHK0 0x00 R/W 0x46C LANE_DESKEW [7:0] ILD7 ILS6 ILD5 ILD4 ILD3 ILD2 ILD1 ILD0 0x00 R 0x46D BAD_DISPARITY [7:0] BDE7 BDE6 BDE5 BDE4 BDE3 BDE2 BDE1 BDE0 0x00 R 0x46E NOT_IN_TABLE [7:0] NIT7 NIT6 NIT5 NIT4 NIT3 NIT2 NIT1 NIT0 0x00 R 0x46F UNEXPECTED_ KCHAR [7:0] UEK7 UEK6 UEK5 UEK4 UEK3 UEK2 UEK1 UEK0 0x00 R 0x470 CODE_GRP_SYNC [7:0] CGS7 CGS6 CGS5 CGS4 CGS3 CGS2 CGS1 CGS0 0x00 R 0x471 FRAME_SYNC [7:0] FS7 FS6 FS5 FS4 FS3 FS2 FS1 FS0 0x00 R 0x472 GOOD_ CHECKSUM [7:0] CKS7 CKS6 CKS5 CKS4 CKS3 CKS2 CKS1 CKS0 0x00 R 0x473 INIT_LANE_SYNC [7:0] ILS7 ILS6 ILS5 ILS4 ILS3 ILS2 ILS1 ILS0 0x00 R 0x475 CTRLREG0 [7:0] RX_DIS CHAR_REPL_ DIS SOFTRST FORCESYNCREQ RESERVED REPL_FRM_ENA 0x01 R/W 0x476 CTRLREG1 [7:0] CGS_SEL FCHK_N 0x477 CTRLREG2 [7:0] ILS_MODE RESERVED 0x478 KVAL [7:0] KSYNC 0x01 R/W 0x47C ERRORTHRES [7:0] ETH 0xFF R/W 0x47D SYNC_ASSERT_ MASK [7:0] 0x480 ECNT_CTRL0 [7:0] RESERVED ECNT_ENA0 ECNT_RST0 0x3F R/W 0x481 ECNT_CTRL1 [7:0] RESERVED ECNT_ENA1 ECNT_RST1 0x3F R/W 0x482 ECNT_CTRL2 [7:0] RESERVED ECNT_ENA2 ECNT_RST2 0x3F R/W 0x483 ECNT_CTRL3 [7:0] RESERVED ECNT_ENA3 ECNT_RST3 0x3F R/W 0x484 ECNT_CTRL4 [7:0] RESERVED ECNT_ENA4 ECNT_RST4 0x3F R/W 0x485 ECNT_CTRL5 [7:0] RESERVED ECNT_ENA5 ECNT_RST5 0x3F R/W 0x486 ECNT_CTRL6 [7:0] RESERVED ECNT_ENA6 ECNT_RST6 0x3F R/W 0x487 ECNT_CTRL7 [7:0] RESERVED ECNT_ENA7 ECNT_RST7 0x3F R/W RESERVED Bit 1 Bit 0 LL_LID4 0x00 R LL_FCHK4 0x00 R LL_FCMP4 [7:0] 0x00 R RESERVED [7:0] LL_LID5 0x00 R 0x00 R 0x00 R RESERVED LL_LID6 0x00 R LL_FCHK6 0x00 R LL_FCMP6 [7:0] 0x00 R RESERVED LL_LID7 0x00 R DID 0x00 R/W BID ADJDIR 0x00 R/W PHADJ RESERVED LID0 0x00 R/W L 0x87 R/W F 0x00 R RESERVED K 0x1F R/W M CS 0x01 R RESERVED SUBCLASSV JESDV HD RESERVED N 0x0F R NP 0x0F R/W S 0x01 R/W CF 0x80 R RES1 RESERVED RESERVED QUAL_RDERR DEL_SCR REPDATATEST QUETESTERR AR_ECNTR RESERVED 0x00 R/W NO_ILAS RESERVED SYNC_ASSERT_MASK Rev. 0 | Page 86 of 139 Reset RW 0x14 R/W 0x00 R/W 0x07 R/W Data Sheet AD9161/AD9162 Reg. Name Bits Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 0x488 ECNT_TCH0 [7:0] RESERVED ECNT_TCH0 0x07 R/W 0x489 ECNT_TCH1 [7:0] RESERVED ECNT_TCH1 0x07 R/W 0x48A ECNT_TCH2 [7:0] RESERVED ECNT_TCH2 0x07 R/W 0x48B ECNT_TCH3 [7:0] RESERVED ECNT_TCH3 0x07 R/W 0x48C ECNT_TCH4 [7:0] RESERVED ECNT_TCH4 0x07 R/W 0x48D ECNT_TCH5 [7:0] RESERVED ECNT_TCH5 0x07 R/W 0x48E ECNT_TCH6 [7:0] RESERVED ECNT_TCH6 0x07 R/W 0x48F ECNT_TCH7 [7:0] RESERVED ECNT_TCH7 0x07 R/W 0x490 ECNT_STAT0 [7:0] RESERVED LANE_ENA0 ECNT_TCR0 0x00 R 0x491 ECNT_STAT1 [7:0] RESERVED LANE_ENA1 ECNT_TCR1 0x00 R 0x492 ECNT_STAT2 [7:0] RESERVED LANE_ENA2 ECNT_TCR2 0x00 R 0x493 ECNT_STAT3 [7:0] RESERVED LANE_ENA3 ECNT_TCR3 0x00 R 0x494 ECNT_STAT4 [7:0] RESERVED LANE_ENA4 ECNT_TCR4 0x00 R 0x495 ECNT_STAT5 [7:0] RESERVED LANE_ENA5 ECNT_TCR5 0x00 R 0x496 ECNT_STAT6 [7:0] RESERVED LANE_ENA6 ECNT_TCR6 0x00 R 0x497 ECNT_STAT7 [7:0] RESERVED LANE_ENA7 ECNT_TCR7 0x4B0 LINK_STATUS0 [7:0] BDE0 NIT0 UEK0 ILD0 ILS0 CKS0 FS0 CGS0 0x00 R 0x4B1 LINK_STATUS1 [7:0] BDE1 NIT1 UEK1 ILD1 ILS1 CKS1 FS1 CGS1 0x00 R 0x4B2 LINK_STATUS2 [7:0] BDE2 NIT2 UEK2 ILD2 ILS2 CKS2 FS2 CGS2 0x00 R 0x4B3 LINK_STATUS3 [7:0] BDE3 NIT3 UEK3 ILD3 ILS3 CKS3 FS3 CGS3 0x00 R 0x4B4 LINK_STATUS4 [7:0] BDE4 NIT4 UEK4 ILD4 ILS4 CKS4 FS4 CGS4 0x00 R 0x4B5 LINK_STATUS5 [7:0] BDE5 NIT5 UEK5 ILD5 ILS5 CKS5 FS5 CGS5 0x00 R 0x4B6 LINK_STATUS6 [7:0] BDE6 NIT6 UEK6 ILD6 ILS6 CKS6 FS6 CGS6 0x00 R 0x4B7 LINK_STATUS7 [7:0] BDE7 NIT7 UEK7 ILD7 ILS7 CKS7 FS7 CGS7 0x00 R 0x4B8 JESD_IRQ_ ENABLEA [7:0] EN_BDE EN_NIT EN_UEK EN_ILD EN_ILS EN_CKS EN_FS EN_CGS 0x00 R/W 0x4B9 JESD_IRQ_ ENABLEB [7:0] EN_ILAS 0x00 R/W 0x4BA JESD_IRQ_ STATUSA [7:0] IRQ_CGS 0x00 R/W 0x4BB JESD_IRQ_ STATUSB [7:0] IRQ_ILAS 0x00 R/W 0x800 HOPF_CTRL [7:0] RESERVED IRQ_BDE IRQ_NIT IRQ_UEK IRQ_ILD IRQ_ILS IRQ_CKS RESERVED HOPF_MODE RESERVED Rev. 0 | Page 87 of 139 IRQ_FS Bit 0 Reset RW 0x00 R 0x00 R/W AD9161/AD9162 Data Sheet REGISTER DETAILS Table 47. Register Details Hex. Addr. Name Bits Bit Name Description Reset Access 0x000 SPI_INTFCONFA 7 SOFTRESET_M Soft reset (mirror). Set this to mirror Bit 0. 0x0 R 6 LSBFIRST_M LSB first (mirror). Set this to mirror Bit 1. 0x0 R 5 ADDRINC_M Addr increment (mirror). Set this to mirror Bit 2. 0x0 R 4 SDOACTIVE_M SDO active (mirror). Set this to mirror Bit 3. 0x0 R 3 SDOACTIVE SDO active. Enables 4-wire SPI bus mode. 0x0 R/W 2 ADDRINC 0x0 R/W 1 LSBFIRST LSB first. When set, causes input 0x0 and output data to be oriented as LSB first. If this bit is clear, data is oriented as MSB first. R/W Settings Address increment. When set, causes incrementing streaming addresses; otherwise, descending addresses are generated. 1 Streaming addresses are incremented. 0 Streaming addresses are decremented. 1 Shift LSB in first. 0 Shift MSB in first. 0x001 SPI_INTFCONFB 0 SOFTRESET Soft reset. This bit automatically clears to 0 after performing a reset operation. Setting this bit initiates a reset. This bit is autoclearing after the soft reset is complete. 1 Pulse the soft reset line. 0 Reset the soft reset line. 0x0 R/W 7 SINGLEINS Single instruction. 1 Perform single transfers. 0x0 R/W 6 CSSTALL 0x0 R/W 0x0 R/W 0x0 R/W 0x0 R/W 0 Perform multiple transfers. CS stalling. 0 Disable CS stalling. 1 Enable CS stalling. [5:3] RESERVED Reserved. 2 SOFTRESET1 1 SOFTRESET0 0 RESERVED Soft Reset 1. This bit automatically clears to 0 after performing a reset operation. 1 Pulse the Soft Reset 1 line. 0 Pulse the Soft Reset 1 line. Soft Reset 0. This bit automatically clears to 0 after performing a reset operation. 1 Pulse the Soft Reset 0 line. 0 Pulse the Soft Reset 0 line. 0x002 SPI_DEVCONF Reserved. 0x0 R [7:4] DEVSTATUS Device status. 0x0 R/W [3:2] CUSTOPMODE Customer operating mode. 0x0 R/W Rev. 0 | Page 88 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [1:0] SYSOPMODE Description System operating mode. 0 Normal operation. 1 Low power operation. Reset Access 0x0 R/W 2 Medium power standby. 3 Low power sleep. 0x003 SPI_CHIPTYPE [7:0] CHIP_TYPE Chip type. 0x0 R 0x004 SPI_PRODIDL [7:0] PROD_ID[7:0] Product ID. 0x0 R 0x005 SPI_PRODIDH [7:0] PROD_ID[15:8] Product ID. 0x0 R 0x006 SPI_CHIPGRADE [7:4] PROD_GRADE Product grade. 0x0 R [3:0] DEV_REVISION Device revision. 0x0 R [7:5] RESERVED Reserved. 0x0 R 0x020 IRQ_ENABLE 4 EN_SYSREF_JITTER Enable SYSREF± jitter interrupt. 0 Disable interrupt. 1 Enable interrupt. 0x0 R/W 3 EN_DATA_READY Enable JESD204x receiver ready (JRX_DATA_READY) low interrupt 0 Disable interrupt. 1 Enable interrupt. 0x0 R/W 2 EN_LANE_FIFO 0x0 R/W 0x0 R/W 0x0 R/W Enable lane FIFO overflow/ underflow interrupt. 0 Disable interrupt. 1 Enable interrupt. 1 EN_PRBSQ 0 EN_PRBSI Enable PRBS imaginary error interrupt. 0 Disable interrupt. 1 Enable interrupt. Enable PRBS real error interrupt 0 Disable interrupt. 1 Enable interrupt. 0x024 IRQ_STATUS [7:5] RESERVED Reserved. 0x0 R 4 IRQ_SYSREF_JITTER SYSREF± jitter is too big. Writing 1 clears the status. 0x0 R/W 3 IRQ_DATA_READY JRX_DATA_READY is low. Writing 0x0 1 clears the status. R/W 0 No warning. 1 Warning detected. 2 IRQ_LANE_FIFO 1 IRQ_PRBSQ 0 IRQ_PRBSI Lane FIFO overflow/underflow. Writing 1 clears the status. 0 No warning. 0x0 R/W 0x0 R/W 0x0 R/W 0x0 R 1 Warning detected. PRBS imaginary error. Writing 1 clears the status. 0 No warning. 1 Warning detected. PRBS real error. Writing 1 clears the status. 0 No warning. 1 Warning detected. 0x031 SYNC_LMFC_DELAY_FRAME 0x032 SYNC_LMFC_DELAY0 [7:5] RESERVED Reserved. [4:0] SYNC_LMFC_DELAY_SET_FRM Desired delay from rising edge of 0x0 SYSREF± input to rising edge of LMFC in frames. R/W [7:0] SYNC_LMFC_DELAY_SET[7:0] Desired delay from rising edge of 0x0 SYSREF± input to rising edge of LMFC in DAC clock units. R/W Rev. 0 | Page 89 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name Description Reset Access 0x033 SYNC_LMFC_DELAY1 [7:4] RESERVED Reserved. 0x0 [3:0] SYNC_LMFC_DELAY_SET[11:8] Desired delay from rising edge of 0x0 SYSREF± input to rising edge of LMFC in DAC clock units. R/W 0x034 SYNC_LMFC_STAT0 [7:0] SYNC_LMFC_DELAY_STAT[7:0] Measured delay from rising edge 0x0 of SYSREF± input to rising edge of LMFC in DAC clock units (note: 2 LSBs are always zero). A write to SYNC_LMFC_STATx or SYSREF_PHASEx saves the data for readback. R/W 0x035 SYNC_LMFC_STAT1 [7:4] RESERVED Reserved. R [3:0] SYNC_LMFC_DELAY_STAT[11:8] Measured delay from rising edge 0x0 of SYSREF± input to rising edge of LMFC in DAC clock units (note: 2 LSBs are always zero). A write to SYNC_LMFC_STATx or SYSREF_PHASEx saves the data for readback. R/W 0x036 SYSREF_COUNT [7:0] SYSREF_COUNT Count of SYSREF± signals received. 0x0 A write resets the count. A write to SYNC_LMFC_STATx or SYSREF_PHASEx saves the data for readback. R/W 0x037 SYSREF_PHASE0 [7:0] SYSREF_PHASE[7:0] Phase of measured SYSREF± event. Thermometer encoded. A write to SYNC_LMFC_STATx or SYSREF_PHASEx saves the data for readback. 0x0 R/W 0x038 SYSREF_PHASE1 [7:4] RESERVED Reserved. 0x0 R [3:0] SYSREF_PHASE[11:8] Phase of measured SYSREF± event. Thermometer encoded. A write to SYNC_LMFC_STATx or SYSREF_PHASEx saves the data for readback. 0x0 R/W [7:6] RESERVED Reserved. 0x0 R [5:0] SYSREF_JITTER_WINDOW Amount of jitter allowed on the SYSREF± input. SYSREF± jitter variations bigger than this triggers an interrupt. Units are in DAC clocks. The bottom two bits are ignored. 0x0 R/W [7:2] RESERVED Reserved. 0x0 R 0x039 SYSREF_JITTER_WINDOW 0x03A SYNC_CTRL Settings [1:0] SYNC_MODE 0x0 Synchronization mode. 0x0 00 Do not perform synchronization, monitor SYSREF± to LMFC delay only. 01 Perform continuous synchronization of LMFC on every SYSREF±. R R/W 10 Perform a single synchronization on the next SYSREF±, then switch to monitor mode. 0x03F TX_ENABLE 7 SPI_DATAPATH_POST SPI control of the data at the output of the datapath. 0 Disable or zero the data from the datapath into the DAC. 1 Use the data from the datapath to drive the DAC. Rev. 0 | Page 90 of 139 0x1 R/W Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 6 Settings SPI_DATAPATH_PRE [5:4] RESERVED 3 Description SPI control of the data at the input of the datapath. 0 Disable or zero the data feeding into the datapath. 1 Use the data from the JESD204B lanes to drive into the datapath. Reserved. TXEN_NCO_RESET Reset Access 0x1 R/W 0x0 R Allows TX_ENABLE to control the 0x0 DDS NCO reset. 0 Use the SPI (HOPF_MODE SPI bits) to control the DDS NCO reset. R/W 1 Use the TX_ENABLE pin to control the DDS NCO reset. 2 TXEN_DATAPATH_POST Allows TX_ENABLE to control the 0x0 data at the output of the datapath. R/W 0 Use the SPI (Bit SPI_DATAPATH_ POST) for control. 1 Use the TX_ENABLE pin for control. 1 TXEN_DATAPATH_PRE 0 TXEN_DAC_FSC Allows TX_ENABLE to control the 0x0 data at the input of the datapath. 0 Use the SPI (Bit SPI_DATAPATH_ PRE) for control. 1 Use the TX_ENABLE pin for control. R/W Allows TX_ENABLE to control the 0x0 DAC full-scale current. R/W 0 Use the SPI register ANA_FSC0 and ANA_FSC1 for control. 1 Use the TX_ENABLE pin for control. 0x040 ANA_DAC_BIAS_PD [7:2] RESERVED Reserved. 0x0 R 1 ANA_DAC_BIAS_PD1 Powers down the DAC core bias circuits. A 1 powers down the DAC core bias circuits. 0x1 R/W 0 ANA_DAC_BIAS_PD0 Powers down the DAC core bias circuits. A 1 powers down the DAC core bias circuits. 0x1 R/W [7:2] RESERVED Reserved. 0x0 R [1:0] ANA_FULL_SCALE_CURRENT[1:0] DAC full-scale current. Analog full-scale current adjustment. 0x3 R/W 0x042 ANA_FSC1 [7:0] ANA_FULL_SCALE_CURRENT[9:2] DAC full-scale current. Analog full-scale current adjustment. 0xFF R/W 0x07F CLK_PHASE_TUNE [7:6] RESERVED Reserved. 0x0 R [5:0] CLK_PHASE_TUNE Fine tuning of the clock input phase balance. Adds small capacitors to the CLK+/CLK− inputs, ~ 20 fF per step, signed magnitude. 0x0 R/W 0x041 ANA_FSC0 Rev. 0 | Page 91 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings Description Reset Access Capacitance 0x080 CLK_PD Bits[5:0] At CLK+ At CLK− 000000 000001 000010 0 1 2 0 0 0 … 011111 … 31 … 0 100000 100001 100010 0 0 0 0 1 2 111111 0 31 [7:1] RESERVED Reserved. 0x0 R 0 DAC clock power-down. Powers down the DAC clock circuitry. 0x1 R/W DACCLK_PD 0 Power up. 1 Power down. 0x082 CLK_DUTY 0x083 CLK_CRS_CTRL 0x084 PLL_REF_CLK_PD 7 CLK_DUTY_EN Enable duty cycle control. 0x1 R/W 6 CLK_DUTY_OFFSET_EN Enable duty cycle offset. 0x0 R/W 5 CLK_DUTY_BOOST_EN Enable duty cycle range boost. Extends range to ±5% at cost of 1 dB to 2 dB worse phase noise. 0x0 R/W [4:0] CLK_DUTY_PRG Program the duty cycle offset. 5-bit signed magnitude field, with the MSB as the sign bit and the four LSBs as the magnitude from 0 to 15. A larger magnitude skews duty cycle to a greater amount. Range is ±3%. 0x0 R/W 7 Enable clock cross control adjustment. 0x1 R/W [6:4] RESERVED Reserved. 0x0 R [3:0] CLK_CRS_ADJ Program the clock crossing point. 0x0 R/W [7:6] RESERVED Reserved. 0x0 R 0x0 R/W 0x0 R CLK_CRS_EN [5:4] PLL_REF_CLK_RATE PLL reference clock rate multiplier. 00 Normal rate (1×) PLL reference clock. 01 Double rate (2×) PLL reference clock. 10 Quadruple rate (4×) PLL reference clock. 11 Disable the PLL reference clock. [3:1] RESERVED 0 0x088 SYSREF_CTRL0 Reserved. PLL_REF_CLK_PD PLL reference clock power-down. 0x0 0 Enable the PLL reference clock. 1 Power down the PLL reference clock. R/W [7:4] RESERVED Reserved. 0x0 R 3 HYS_ON SYSREF± hysteresis enable. This bit enables the programmable hysteresis control for the SYSREF± receiver. 0x0 R/W 2 SYSREF_RISE Use SYSREF± rising edge. 0x0 R/W Rev. 0 | Page 92 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings Description Reset Access [1:0] HYS_CNTRL[9:8] Controls the amount of 0x0 hysteresis in the SYSREF± receiver. Each of the 10 bits adds 10 mV of differential hysteresis to the receiver input. R/W 0x089 SYSREF_CTRL1 [7:0] HYS_CNTRL[7:0] Controls the amount of 0x0 hysteresis in the SYSREF± receiver. Each of the 10 bits adds 10 mV of differential hysteresis to the receiver input. R/W 0x090 DLL_PD [7:5] RESERVED Reserved. 0x0 R 4 DLL_FINE_DC_EN Fine delay line duty cycle correction enable. 0x1 R/W 3 DLL_FINE_XC_EN Fine delay line cross control enable. 0x1 R/W 2 DLL_COARSE_DC_EN Coarse delay line duty cycle correction enable. 0x1 R/W 1 DLL_COARSE_XC_EN Coarse delay line cross control enable. 0x1 R/W 0 DLL_CLK_PD Power down DLL and digital clock generator. 0x1 R/W 0x1 R/W 0x1 R/W 0x1 R/W 0x2 R/W 0x0 R/W 0x0 R/W 0 Power up DLL controller. 1 Power down DLL controller. 0x091 DLL_CTRL 7 DLL_TRACK_ERR Track error behavior. 0 Continue on error. 1 Restart on error. 6 DLL_SEARCH_ERR 5 DLL_SLOPE Search error behavior. 0 Stop on error. 1 Retry on error. Desired slope. 0 Negative slope. 1 Positive slope. [4:3] DLL_SEARCH Search direction. 00 Search down from initial point only. 01 Search up from initial point only. 10 Search up and down from initial point. [2:1] DLL_MODE Controller mode. 00 Search then track. 01 Track only. 10 Search only. 0 DLL_ENABLE Controller enable. 0 Disable DLL controller: use static SPI settings. 1 Enable DLL controller: use controller with feedback loop. 0x092 DLL_STATUS 0x093 DLL_GB 0x094 DLL_COARSE 0x095 DLL_FINE [7:3] RESERVED Reserved. 0x0 R 2 DLL_FAIL The DAC clock DLL failed to lock. 0x0 R 1 DLL_LOST The DAC clock DLL has lost lock. 0x0 R/W 0 DLL_LOCKED The DAC clock DLL has achieved lock. 0x0 R [7:4] RESERVED Reserved. 0x0 R [3:0] DLL_GUARD Search guard band. 0x0 R/W [7:6] RESERVED Reserved. 0x0 R [5:0] DLL_COARSE Coarse delay line setpoint. 0x0 R/W [7:0] DLL_FINE Fine delay line setpoint. 0x80 R/W Rev. 0 | Page 93 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name 0x096 DLL_PHASE [7:5] RESERVED Settings [4:0] DLL_PHS Description Reset Access Reserved. 0x0 R 0x8 R/W Desired phase. 0 Minimum allowed phase. 16 Maximum allowed phase. 0x097 DLL_BW [7:5] RESERVED Reserved. 0x0 R [4:2] DLL_FILT_BW Phase measurement filter bandwidth. 0x0 R/W [1:0] DLL_WEIGHT Tracking speed. 0x0 R/W [7:1] RESERVED Reserved. 0x0 R 0 Read request: 0 to 1 transition updates the coarse, fine, and phase readback values. 0x0 R/W [7:6] RESERVED Reserved. 0x0 R [5:0] DLL_COARSE_RB Coarse delay line readback. 0x0 R 0x09A DLL_FINE_RB [7:0] DLL_FINE_RB Fine delay line readback. 0x0 R 0x09B DLL_PHASE_RB [7:5] RESERVED Reserved. 0x0 R [4:0] DLL_PHS_RB Phase readback. 0x0 R [7:3] RESERVED Reserved. 0x0 R 0x0 R/W 0x098 DLL_READ 0x099 DLL_COARSE_RB 0x09D DIG_CLK_INVERT DLL_READ 2 INV_DIG_CLK Invert digital clock from DLL. 0 Normal polarity. 1 DIG_CLK_DC_EN Digital clock duty cycle correction enable. 0x1 R/W 0 DIG_CLK_XC_EN Digital clock cross control enable. 0x1 R/W 7 DLL_TEST_EN 1 Inverted polarity. 0x0A0 DLL_CLK_DEBUG 0x110 INTERP_MODE DLL clock output test enable. 0x0 R/W [6:2] RESERVED Reserved. 0x0 R [1:0] DLL_TEST_DIV DLL clock output divide. 0x0 R/W [7:4] JESD_LANES Number of JESD204B lanes. For 0x8 proper operation of the JESD204B data link, this signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W Interpolation mode. For proper 0x1 operation of the JESD204B data link, this signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0000 1× (bypass) (this mode is not supported in the AD9161). 0001 2×. R/W [3:0] INTERP_MODE 0010 3×. 0011 4×. 0100 6×. 0101 8×. 0110 12×. 0111 16×. 1000 24×. 0x111 DATAPATH_CFG 7 INVSINC_EN Inverse sinc filter enable. 0 Disable inverse sinc filter. 1 Enable inverse sinc filter. Rev. 0 | Page 94 of 139 0x0 R/W Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 6 NCO_EN 5 RESERVED 4 FILT_BW Settings Description Modulation enable. 0 Disable NCO. 1 Enable NCO. Reset Access 0x0 R/W Reserved. 0x0 R Datapath filter bandwidth. 0x0 R/W 0 Filter bandwidth is 80%. 1 Filter bandwidth is 90%. 3 RESERVED Reserved. 0x0 R 2 MODULUS_EN Modulus DDS enable 0x0 R/W 0x0 R/W 0 Disable modulus DDS. 1 Enable modulus DDS. 1 SEL_SIDEBAND Selects upper or lower sideband from modulation result. 0 Use upper sideband. 1 Use lower sideband = spectral flip. 0x113 FTW_UPDATE 0 FIR85_FILT_EN FIR85 filter enable. 0x0 R/W 7 RESERVED Reserved. 0x0 R 0x0 R/W [6:4] FTW_REQ_MODE Frequency tuning word automatic update mode. 000 No automatic requests are generated when the FTW registers are written. 001 Automatically generate FTW_LOAD_REQ after FTW0 is written. 010 Automatically generate FTW_LOAD_REQ after FTW1 is written. 011 Automatically generate FTW_LOAD_REQ after FTW2 is written. 100 Automatically generate FTW_LOAD_REQ after FTW3 is written. 101 Automatically generate FTW_LOAD_REQ after FTW4 is written. 110 Automatically generate FTW_LOAD_REQ after FTW5 is written. 3 RESERVED Reserved. 0x0 R 2 FTW_LOAD_SYSREF FTW load and reset from rising edge of SYSREF±. 0x0 R/W 1 FTW_LOAD_ACK Frequency tuning word update acknowledge. 0 FTW is not loaded. 1 FTW is loaded. 0x0 R 0 FTW_LOAD_REQ Frequency tuning word update request from SPI. 0x0 R/W 0 Clear FTW_LOAD_ACK. 1 0 to 1 transition loads the FTW. 0x114 FTW0 [7:0] FTW[7:0] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W 0x115 FTW1 [7:0] FTW[15:8] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W Rev. 0 | Page 95 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name 0x116 FTW2 [7:0] FTW[23:16] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W 0x117 FTW3 [7:0] FTW[31:24] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W 0x118 FTW4 [7:0] FTW[39:32] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W 0x119 FTW5 [7:0] FTW[47:40] NCO frequency tuning word. This 0x0 is X in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). R/W 0x11C PHASE_OFFSET0 [7:0] NCO_PHASE_OFFSET[7:0] NCO phase offset. 0x0 R/W 0x11D PHASE_OFFSET1 [7:0] NCO_PHASE_OFFSET[15:8] NCO phase offset. 0x0 R/W 0x124 ACC_MODULUS0 [7:0] ACC_MODULUS[7:0] DDS Modulus. This is B in the 0x0 equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. R/W 0x125 ACC_MODULUS1 [7:0] ACC_MODULUS[15:8] DDS Modulus. This is B in the 0x0 equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. R/W 0x126 ACC_MODULUS2 [7:0] ACC_MODULUS[23:16] DDS Modulus. This is B in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248).Note this modulus value is used for all NCO FTWs. 0x0 R/W 0x127 ACC_MODULUS3 [7:0] ACC_MODULUS[31:24] DDS Modulus. This is B in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. 0x0 R/W 0x128 ACC_MODULUS4 [7:0] ACC_MODULUS[39:32] DDS Modulus. This is B in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. 0x0 R/W 0x129 ACC_MODULUS5 [7:0] ACC_MODULUS[47:40] DDS Modulus. This is B in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. 0x0 R/W 0x12A ACC_DELTA0 [7:0] ACC_DELTA[7:0] DDS Delta. This is A in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. Note this delta value is used for all NCO FTWs. 0x0 R/W 0x12B ACC_DELTA1 [7:0] ACC_DELTA[15:8] DDS Delta. This is A in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. Note this delta value is used for all NCO FTWs. 0x0 R/W 0x12C ACC_DELTA2 [7:0] ACC_DELTA[23:16] DDS Delta. This is A in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. Note this delta value is used for all NCO FTWs. 0x0 R/W Settings Rev. 0 | Page 96 of 139 Description Reset Access Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name Description Reset Access 0x12D ACC_DELTA3 [7:0] ACC_DELTA[31:24] DDS Delta. This is 'A' in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this Delta value will be used for all NCO FTWs. 0x0 R/W 0x12E ACC_DELTA4 [7:0] ACC_DELTA[39:32] DDS Delta. This is A in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. Note this delta value is used for all NCO FTWs. 0x0 R/W 0x12F ACC_DELTA5 [7:0] ACC_DELTA[47:40] DDS Delta. This is A in the equation fOUT = fDAC × (M/N) = fDAC × ((X + A/B)/248). Note this modulus value is used for all NCO FTWs. Note this delta value is used for all NCO FTWs. 0x0 R/W 0x14B PRBS 7 PRBS_GOOD_Q 0x0 R 6 PRBS_GOOD_I Good data indicator real channel. 0x0 R Settings Good data indicator imaginary channel. 0 Incorrect sequence detected. 1 Correct PRBS sequence detected. 0 Incorrect sequence detected. 1 Correct PRBS sequence detected. 5 RESERVED Reserved. 0x0 R 4 PRBS_INV_Q Data inversion imaginary channel. 0x1 R/W 0x0 R/W 0x0 R/W 0x0 R/W 0x0 R/W 0 Expect normal data. 1 Expect inverted data. 3 PRBS_INV_I Data inversion real channel. 0 Expect normal data. 2 PRBS_MODE 1 Expect inverted data. Polynomial select. 0 7-bit: x7 + x6 + 1. 1 15-bit: x15 + x14 + 1. 1 PRBS_RESET Reset error counters. 0 Normal operation. 1 Reset counters. 0 PRBS_EN Enable PRBS checker. 0 Disable. 1 Enable. 0x14C PRBS_ERROR_I [7:0] PRBS_COUNT_I Error count value real channel. 0x0 R 0x14D PRBS_ERROR_Q [7:0] PRBS_COUNT_Q Error Count Value Imaginary Channel. 0x0 R 0x14E TEST_DC_DATA1 [7:0] DC_TEST_DATA[15:8] DC test data. DC test mode is not 0x0 supported on the AD9161. R/W 0x14F TEST_DC_DATA0 [7:0] DC_TEST_DATA[7:0] DC test data. DC test mode is not 0x0 supported on the AD9161. R/W 0x150 DIG_TEST [7:2] RESERVED Reserved. 0x0 R 1 DC data test mode enable. DC test mode is not supported on the AD9161. 0x0 R/W DC_TEST_EN 1 DC test mode enable. 0 DC test mode disable. 0 0x151 DECODE_CTRL RESERVED Reserved. 0x0 R/W [7:3] RESERVED Reserved. 0x0 R/W Rev. 0 | Page 97 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 2 Settings SHUFFLE Description Reset Access Shuffle mode. Enables shuffle mode for better spurious performance. 0x0 R/W 0 Disable MSB shuffling (use thermometer encoding). 1 Enable MSB shuffling. 0x152 DECODE_MODE [1:0] RESERVED Reserved. 0x0 R/W [7:2] RESERVED Reserved. 0x0 R 0x0 R/W [1:0] DECODE_MODE Decode mode. 00 Nonreturn-to-zero mode (first Nyquist). 01 Mix-Mode (second Nyquist). 10 Return to zero. 11 Reserved. 0x1DF SPI_STRENGTH 0x200 MASTER_PD 0x201 PHY_PD [7:4] RESERVED Reserved. 0x0 R [3:0] SPIDRV Slew and drive strength for CMOS SPI outputs. Slew = Bits[1:0], drive = Bits[3:2]. 0xF R/W [7:1] RESERVED Reserved. 0x0 R 0 Power down the entire JESD204B 0x1 Rx analog (all eight channels and bias). R/W SPI override to power down the individual PHYs. 0x0 R/W SPI_PD_MASTER [7:0] SPI_PD_PHY Bit 0 controls the SERDIN0± PHY. Bit 1 controls the SERDIN1± PHY. Bit 2 controls the SERDIN2± PHY. Bit 3 controls the SERDIN3± PHY. Bit 4 controls the SERDIN4± PHY. Bit 5 controls the SERDIN5± PHY. Bit 6 controls the SERDIN6± PHY. Bit 7 controls the SERDIN7± PHY. 0x203 GENERIC_PD 0x206 CDR_RESET [7:2] RESERVED Reserved. 0x0 R 1 SPI_SYNC1_PD Power down LVDS buffer for the sync request signal, SYNCOUT. 0x0 R/W 0 RESERVED Reserved. 0x0 R/W [7:1] RESERVED Reserved. 0x0 R 0x1 R/W 0 SPI_CDR_RESET Resets the digital control logic for all PHYs. 0 CDR logic is reset. 1 CDR logic is operational. 0x230 CDR_OPERATING_MODE_REG_0 [7:6] RESERVED Reserved. 0x0 R/W 5 Enables half rate CDR operation, must be enabled for data rates above 6 Gbps. 0x1 R/W 0x0 R/W SPI_ENHALFRATE 0 Disables CDR half rate operation, data rate ≤ 6 Gbps. 1 Enables CDR half rate operation, data rate > 6 Gbps. [4:3] RESERVED Reserved. Rev. 0 | Page 98 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [2:1] SPI_DIVISION_RATE Description Enables oversampling of the input data. 00 No division. Data rate > 3 Gbps. Reset Access 0x0 R/W 0x0 R/W 0x8 R/W 0x8 R/W 0x8 R/W 01 Division by 2. 1.5 Gbps < data rate ≤ 3 Gbps. 10 Division by 4. 750 Mbps < data rate ≤ 1.5 Gbps. 0 0x250 EQ_CONFIG_PHY_0_1 RESERVED Reserved. [7:4] SPI_EQ_CONFIG1 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. [3:0] SPI_EQ_CONFIG0 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. 0x251 EQ_CONFIG_PHY_2_3 [7:4] SPI_EQ_CONFIG3 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. Rev. 0 | Page 99 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings Description Reset Access 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. [3:0] SPI_EQ_CONFIG2 0x8 R/W 0x8 R/W 0x8 R/W 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost Level = 13. 1110 Boost level = 14. 1111 Boost level = 15. 0x252 EQ_CONFIG_PHY_4_5 [7:4] SPI_EQ_CONFIG5 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. [3:0] SPI_EQ_CONFIG4 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. Rev. 0 | Page 100 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings Description Reset Access 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. 0x253 EQ_CONFIG_PHY_6_7 [7:4] SPI_EQ_CONFIG7 0x8 R/W 0x8 R/W 0x1 R/W 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. [3:0] SPI_EQ_CONFIG6 0000 Manual mode (SPI configured values used). 0001 Boost level = 1. 0010 Boost level = 2. 0011 Boost level = 3. 0100 Boost level = 4. 0101 Boost level = 5. 0110 Boost level = 6. 0111 Boost level = 7. 1000 Boost level = 8. 1001 Boost level = 9. 1010 Boost level = 10. 1011 Boost level = 11. 1100 Boost level = 12. 1101 Boost level = 13. 1110 Boost level = 14. 1111 Boost level = 15. 0x268 EQ_BIAS_REG [7:6] EQ_POWER_MODE Controls the equalizer power mode/insertion loss capability. 00 Normal mode. 01 Low power mode. 0x280 SYNTH_ENABLE_CNTRL [5:0] RESERVED Reserved. 0x4 R/W [7:3] RESERVED Reserved. 0x0 R 2 Set this bit high to rerun all of 0x0 the SERDES PLL calibration routines. Set this bit low again to allow additional recalibrations. Rising edge causes the calibration. SPI_RECAL_SYNTH Rev. 0 | Page 101 of 139 R/W AD9161/AD9162 Hex. Addr. Name 0x281 PLL_STATUS Data Sheet Bits Bit Name Description Reset Access 1 RESERVED Settings Reserved. 0x0 R/W 0 SPI_ENABLE_SYNTH Enable the SERDES PLL. Setting this bit turns on all currents and proceeds to calibrate the PLL. Make sure reference clock and division ratios are correct before enabling this bit. 0x0 R/W [7:6] RESERVED Reserved. 0x0 R 5 If set, the SERDES PLL CP output is above valid operating range. 0x0 R SPI_CP_OVER_RANGE_HIGH_RB 0 Charge pump output is within operating range. 1 Charge pump output is above operating range. 4 SPI_CP_OVER_RANGE_LOW_RB If set, the SERDES PLL CP output is below valid operating range. 0 Charge pump output is within operating range. 1 Charge pump output is below operating range. 0x0 R 3 SPI_CP_CAL_VALID_RB This bit tells the user if the 0x0 charge pump calibration has completed and is valid. 0 Charge pump calibration is not valid. 1 Charge pump calibration is valid. R [2:1] RESERVED Reserved. 0x0 R 0 If set, the SERDES synthesizer locked. 0x0 R 0x0 R SPI_PLL_LOCK_RB 0 PLL is not locked. 1 PLL is locked. 0x289 REF_CLK_DIVIDER_LDO [7:2] RESERVED Reserved. [1:0] SERDES_PLL_DIV_FACTOR SERDES PLL reference clock 0x0 division factor. This field controls the division of the SERDES PLL reference clock before it is fed into the SERDES PLL PFD. It must be set so that fREF/DivFactor is between 35 MHz and 80 MHz. 00 Divide by 4 for lane rate between 6 Gbps and 12.5 Gbps. R/W 01 Divide by 2 for lane rate between 3 Gbps and 6 Gbps. 10 Divide by 1 for lane rate between 1.5 Gbps and 3 Gbps. 0x2A7 TERM_BLK1_CTRLREG0 0x2A8 TERM_BLK1_CTRLREG1 [7:1] RESERVED Reserved. 0x0 R 0 Rising edge of this bit starts a termination calibration routine. 0x0 R/W SPI override for termination value for PHY 0, PHY 1, PHY 6, and PHY 7. Value options are as follows: XXX0XXXX Automatically calibrate termination value. XXX1000X Force 000 as termination value. 0x0 R/W SPI_I_TUNE_R_CAL_TERMBLK1 [7:0] SPI_I_SERIALIZER_RTRIM_TERMBLK1 XXX1001X Force 001 as termination value. XXX1010X Force 010 as termination value. XXX1011X Force 011 as termination value. XXX1100X Force 100 as termination value. XXX1101X Force 101 as termination value. Rev. 0 | Page 102 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings Description Reset Access XXX1110X Force 110 as termination value. XXX1111X Force 111 as termination value. XXX1000X Force 000 as termination value. 0x2AC TERM_BLK1_RD_REG0 0x2AE TERM_BLK2_CTRLREG0 0x2AF TERM_BLK2_CTRLREG1 [7:4] RESERVED Reserved. 0x0 R [3:0] SPI_O_RCAL_CODE_TERMBLK1 Readback of calibration code for PHY 0, PHY 1, PHY 6, and PHY 7. 0x0 R [7:1] RESERVED Reserved. 0x0 R 0 SPI_I_TUNE_R_CAL_TERMBLK2 Rising edge of this bit starts a termination calibration routine. 0x0 R/W [7:0] SPI_I_SERIALIZER_RTRIM_TERMBLK2 SPI override for termination value for PHY 2, PHY 3, PHY 4, and PHY 5. Value options are as follows: 0x0 R/W XXX0XXXX Automatically calibrate termination value. XXX1000X Force 000 as termination value. XXX1001X Force 001 as termination value. XXX1010X Force 010 as termination value. XXX1011X Force 011 as termination value. XXX1100X Force 100 as termination value. XXX1101X Force 101 as termination value. XXX1110X Force 110 as termination value. XXX1111X Force 111 as termination value. XXX1000X Force 000 as termination value. 0x2B3 TERM_BLK2_RD_REG0 0x2BB TERM_OFFSET_0 0x2BC TERM_OFFSET_1 0x2BD TERM_OFFSET_2 0x2BE TERM_OFFSET_3 [7:4] RESERVED Reserved. 0x0 R [3:0] SPI_O_RCAL_CODE_TERMBLK2 Readback of calibration code for PHY 2, PHY 3, PHY 4, and PHY 5. 0x0 R [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_0 Add or subtract from the termination calibration value of Physical Lane 0. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_1 Add or subtract from the termination calibration value of Physical Lane 1. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_2 Add or subtract from the termination calibration value of Physical Lane 2. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_3 Add or subtract from the termination calibration value of Physical Lane 3. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W Rev. 0 | Page 103 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name Description Reset Access 0x2BF TERM_OFFSET_4 [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_4 Add or subtract from the termination calibration value of Physical Lane 4. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_5 Add or subtract from the termination calibration value of Physical Lane 5. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_6 Add or subtract from the termination calibration value of Physical Lane 6. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W [7:4] RESERVED Reserved. 0x0 R [3:0] TERM_OFFSET_7 Add or subtract from the termination calibration value of Physical Lane 7. 4-bit signed magnitude value that adds to or subtracts from the termination value. Bit 3 is the sign bit, and Bits[2:0] are the magnitude bits. 0x0 R/W 7 RESERVED Reserved. 0x0 R 6 CHECKSUM_MODE 0x0 R/W 0x0 R 0x2C0 TERM_OFFSET_5 0x2C1 TERM_OFFSET_6 0x2C2 TERM_OFFSET_7 0x300 GENERAL_JRX_CTRL_0 Settings JESD204B link parameter checksum calculation method. 0 Checksum is sum of fields. 1 Checksum is sum of octets. 0x302 DYN_LINK_LATENCY_0 0x304 LMFC_DELAY_0 0x306 LMFC_VAR_0 0x308 XBAR_LN_0_1 [5:1] RESERVED Reserved. 0 This bit brings up the JESD204B 0x0 receiver when all link parameters are programmed and all clocks are ready. R/W [7:5] RESERVED Reserved. 0x0 R [4:0] DYN_LINK_LATENCY_0 Measurement of the JESD204B link delay (in PCLK units). Link 0 dynamic link latency. Latency between current deframer LMFC and the global LMFC. 0x0 R [7:5] RESERVED Reserved. 0x0 R [4:0] LMFC_DELAY_0 Fixed part of the JESD204B link delay (in PCLK units). Delay in frame clock cycles for global LMFC for Link 0. 0x0 R/W [7:5] RESERVED Reserved. 0x0 R [4:0] LMFC_VAR_0 Variable part of the JESD204B link delay (in PCLK units). Location in Rx LMFC where JESD204B words are read out from buffer. This setting must not be more than 10 PCLKs. 0x1F R/W [7:6] RESERVED Reserved. 0x0 R LINK_EN Rev. 0 | Page 104 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [5:3] SRC_LANE1 Description Reset Access Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 1. 0x1 R/W 0x0 R/W 0x0 R 0x3 R/W 0x2 R/W 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. [2:0] SRC_LANE0 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 0. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. 0x309 XBAR_LN_2_3 [7:6] RESERVED Reserved. [5:3] SRC_LANE3 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 3. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. [2:0] SRC_LANE2 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 2. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. Rev. 0 | Page 105 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name 0x30A XBAR_LN_4_5 [7:6] RESERVED Settings [5:3] SRC_LANE5 Description Reset Access Reserved. 0x0 R 0x5 R/W 0x4 R/W Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 5. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. [2:0] SRC_LANE4 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 4. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. 0x30B XBAR_LN_6_7 [7:6] RESERVED Reserved. 0x0 R [5:3] SRC_LANE7 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 7. 0x7 R/W 0x6 R/W 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. [2:0] SRC_LANE6 Select data from SERDIN0±, SERDIN1±, …, or SERDIN7± for Logic Lane 6. 000 Data is from SERDIN0±. 001 Data is from SERDIN1±. 010 Data is from SERDIN2±. 011 Data is from SERDIN3±. 100 Data is from SERDIN4±. 101 Data is from SERDIN5±. 110 Data is from SERDIN6±. 111 Data is from SERDIN7±. 0x30C FIFO_STATUS_REG_0 [7:0] LANE_FIFO_FULL Bit 0 corresponds to FIFO full flag 0x0 for data from SERDIN0±. Bit 1 corresponds to FIFO full flag for data from SERDIN1±. Bit 2 corresponds to FIFO full flag for data from SERDIN2±. Bit 3 corresponds to FIFO full flag for data from SERDIN3±. Bit 4 corresponds to FIFO full flag for data from SERDIN4±. Rev. 0 | Page 106 of 139 R Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings Description Reset Access Bit 5 corresponds to FIFO full flag for data from SERDIN5±. Bit 6 corresponds to FIFO full flag for data from SERDIN6±. Bit 7 corresponds to FIFO full flag for data from SERDIN7±. 0x30D FIFO_STATUS_REG_1 [7:0] LANE_FIFO_EMPTY Bit 0 corresponds to FIFO empty flag for data from SERDIN0±. 0x0 R 0x0 R Bit 1 corresponds to FIFO empty flag for data from SERDIN1±. Bit 2 corresponds to FIFO empty flag for data from SERDIN2±. Bit 3 corresponds to FIFO empty flag for data from SERDIN3±. Bit 4 corresponds to FIFO empty flag for data from SERDIN4±. Bit 5 corresponds to FIFO empty flag for data from SERDIN5±. Bit 6 corresponds to FIFO empty flag for data from SERDIN6±. Bit 7 corresponds to FIFO empty flag for data from SERDIN7±. 0x311 SYNC_GEN_0 [7:3] RESERVED Reserved. 2 Mask EOMF from QBD_0. 0x0 Assert SYNCOUT based on loss of multiframe sync. EOMF_MASK_0 R/W 0 Don not assert SYNCOUT on loss of multiframe. 1 Assert SYNCOUT on loss of multiframe. 1 RESERVED Reserved. 0x0 R/W 0 EOF_MASK_0 Mask EOF from QBD_0. 0x0 Assert SYNCOUT based on loss of frame sync. R/W 0 Do not assert SYNCOUT on loss of frame. 1 Assert SYNCOUT on loss of frame. 0x312 SYNC_GEN_1 [7:4] SYNC_ERR_DUR Duration of SYNCOUT signal low for purpose of sync error report. 0 means half PCLK cycle. Add an additional PCLK = 4 octets for each increment of the value. 0x0 R/W [3:0] SYNC_SYNCREQ_DUR Duration of SYNCOUT signal low 0x0 for purpose of sync request. 0 means 5 frame + 9 octets. Add an additional PCLK = 4 octets for each increment of the value. R/W 0x313 SYNC_GEN_3 [7:0] LMFC_PERIOD LMFC period in PCLK cycle. This is to report the global LMFC period based on PCLK. 0x0 R 0x315 PHY_PRBS_TEST_EN [7:0] PHY_TEST_EN 0x0 R/W 0x316 PHY_PRBS_TEST_CTRL 7 0x0 R Enable PHY BER by ungating the clocks. 1 PHY test enable. 0 PHY test disable. RESERVED Reserved. Rev. 0 | Page 107 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings Description [6:4] PHY_SRC_ERR_CNT Reset Access 0x0 R/W 0x0 R/W Start and stop the PHY PRBS test. 0x0 0 Test not started. R/W 000 Report Lane 0 error count. 001 Report Lane 1 error count. 010 Report Lane 2 error count. 011 Report Lane 3 error count. 100 Report Lane 4 error count. 101 Report Lane 5 error count. 110 Report Lane 6 error count. 111 Report Lane 7 error count. [3:2] PHY_PRBS_PAT_SEL Select PRBS pattern for PHY BER test. 00 PRBS7. 01 PRBS15. 10 PRBS31. 11 Not used. 1 PHY_TEST_START 1 Test started. 0 PHY_TEST_RESET Reset PHY PRBS test state machine and error counters. 0x0 R/W Bits[7:0] of the 24-bit threshold value set the error flag for PHY PRBS test. 0x0 R/W 0x318 PHY_PRBS_TEST_THRESHOLD_MIDBITS [7:0] PHY_PRBS_THRESHOLD_MIDBITS Bits[15:8] of the 24-bit threshold value set the error flag for PHY PRBS test. 0x0 R/W 0x319 PHY_PRBS_TEST_THRESHOLD_HIBITS [7:0] PHY_PRBS_THRESHOLD_HIBITS Bits[23:16] of the 24-bit threshold value set the error flag for PHY PRBS test. 0x0 R/W 0x31A PHY_PRBS_TEST_ERRCNT_LOBITS [7:0] PHY_PRBS_ERR_CNT_LOBITS Bits[7:0] of the 24-bit reported PHY BER test error count from selected lane. 0x0 R 0x31B PHY_PRBS_TEST_ERRCNT_MIDBITS [7:0] PHY_PRBS_ERR_CNT_MIDBITS Bits[15:8] of the 24-bit reported PHY BER test error count from selected lane. 0x0 R 0x31C PHY_PRBS_TEST_ERRCNT_HIBITS [7:0] PHY_PRBS_ERR_CNT_HIBITS Bits[23:16] of the 24-bit reported PHY BER test error count from selected lane. 0x0 R 0x31D PHY_PRBS_TEST_STATUS [7:0] PHY_PRBS_PASS Each bit is for the corresponding lane. Report PHY BER test pass/fail for each lane. 0xFF R 0x31E PHY_DATA_SNAPSHOT_CTRL [7:5] RESERVED Reserved. 0x0 R [4:2] PHY_GRAB_LANE_SEL Select which lane to grab data. 0x0 R/W 0x0 R/W 0 Not reset. 1 Reset. 0x317 PHY_PRBS_TEST_THRESHOLD_LOBITS [7:0] PHY_PRBS_THRESHOLD_LOBITS 000 Grab data from Lane 0. 001 Grab data from Lane 1. 010 Grab data from Lane 2. 011 Grab data from Lane 3. 100 Grab data from Lane 4. 101 Grab data from Lane 5. 110 Grab data from Lane 6. 111 Grab data from Lane 7. 1 PHY_GRAB_MODE Rev. 0 | Page 108 of 139 Use error trigger to grab data. 0 Grab data when PHY_GRAB_DATA is set. 1 Grab data upon bit error. Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 0 Settings PHY_GRAB_DATA Description Reset Access Transition from 0 to 1 causes logic to store current receive data from one lane. 0x0 R/W 0x31F PHY_SNAPSHOT_DATA_BYTE0 [7:0] PHY_SNAPSHOT_DATA_BYTE0 Current data received represents 0x0 PHY_SNAPSHOT_DATA[7:0]. R 0x320 PHY_SNAPSHOT_DATA_BYTE1 [7:0] PHY_SNAPSHOT_DATA_BYTE1 Current data received represents 0x0 PHY_SNAPSHOT_DATA[15:8]. R 0x321 PHY_SNAPSHOT_DATA_BYTE2 [7:0] PHY_SNAPSHOT_DATA_BYTE2 Current data received represents 0x0 PHY_SNAPSHOT_DATA[23:16]. R 0x322 PHY_SNAPSHOT_DATA_BYTE3 [7:0] PHY_SNAPSHOT_DATA_BYTE3 Current data received represents 0x0 PHY_SNAPSHOT_DATA[31:24]. R 0x323 PHY_SNAPSHOT_DATA_BYTE4 [7:0] PHY_SNAPSHOT_DATA_BYTE4 Current data received represents 0x0 PHY_SNAPSHOT_DATA[39:32]. R 0x32C SHORT_TPL_TEST_0 [7:4] SHORT_TPL_SP_SEL Short transport layer sample 0x0 selection. Select which sample to check from a specific DAC. 0000 Sample 0. R/W 0001 Sample 1. 0010 Sample 2. 0011 Sample 3. 0100 Sample 4. 0101 Sample 5. 0110 Sample 6. 0111 Sample 7. 1000 Sample 8. 1001 Sample 9. 1010 Sample 10. 1011 Sample 11. 1100 Sample 12. 1101 Sample 13. 1110 Sample 14. 1111 Sample 15. [3:2] SHORT_TPL_M_SEL Short transport layer test DAC selection. Select which DAC to check. 00 DAC 0. 0x0 R/W 0x0 R/W Short transport layer test enable. 0x0 Enable short transport layer test. R/W 01 DAC 1. 10 DAC 2. 11 DAC 3. 1 SHORT_TPL_TEST_RESET 0 SHORT_TPL_TEST_EN Short transport layer test reset. Resets the result of short transport layer test. 0 Not reset. 1 Reset. 0 Disable. 1 Enable. 0x32D SHORT_TPL_TEST_1 [7:0] SHORT_TPL_REF_SP_LSB Rev. 0 | Page 109 of 139 Short transport layer reference sample LSB. This is the lower eight bits of expected DAC sample. It is used to compare with the received DAC sample at the output of JESD204B Rx. 0x0 R/W AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name 0x32E SHORT_TPL_TEST_2 [7:0] SHORT_TPL_REF_SP_MSB Short transport layer test 0x0 reference sample MSB. This is the upper eight bits of expected DAC sample. It is used to compare with the received sample at JESD204B Rx output. R/W 0x32F SHORT_TPL_TEST_3 [7:1] RESERVED Reserved. 0x0 R 0 Short transport layer test fail. This bit shows if the selected DAC sample matches the reference sample. If they match, the test passes; otherwise, the test fails. 0x0 R Settings SHORT_TPL_FAIL Description Reset Access 0 Test pass. 1 Test fail. 0x334 JESD_BIT_INVERSE_CTRL [7:0] JESD_BIT_INVERSE Each bit of this byte inverses the 0x0 JESD204B deserialized data from one specific JESD204B Rx PHY. The bit order matches the logical lane order. For example, Bit 0 controls Lane 0, Bit 1 controls Lane 1. R/W 0x400 DID_REG [7:0] DID_RD Received ILAS configuration on 0x0 Lane 0. DID is the device ID number. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. R 0x401 BID_REG [7:0] BID_RD Received ILAS configuration on Lane 0. BID is the bank ID, extension to DID. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R 0x402 LID0_REG 7 RESERVED Reserved. 0x0 R 6 ADJDIR_RD Received ILAS configuration on 0x0 Lane 0. ADJDIR is the direction to adjust the DAC LMFC. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. R 5 PHADJ_RD Received ILAS configuration on Lane 0. PHADJ is the phase adjustment request to DAC. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R Received ILAS LID configuration on Lane 0. LID0 is the lane identification for Lane 0. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R Received ILAS configuration on Lane 0. SCR is the Tx scrambling status. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0 Scrambling is disabled. 0x0 R [4:0] LL_LID0 0x403 SCR_L_REG 7 SCR_RD 1 Scrambling is enabled. Rev. 0 | Page 110 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [6:5] RESERVED [4:0] L_RD Description Reset Access Reserved. 0x0 R 0x0 R 0x0 R 0x0 R Received ILAS configuration on 0x0 Lane 0. K is the number of frames per multiframe. Settings of 16 or 32 are valid. On this device, all modes use K = 32 (value in register is K − 1). Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 01111 16 frames per multiframe. R Received ILAS configuration on Lane 0. L is the number of lanes per converter device. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 00000 1 lane per converter device. 00001 2 lanes per converter device. 00011 4 lanes per converter device. 00111 8 lanes per converter device. 0x404 F_REG [7:0] F_RD 0x405 K_REG [7:5] RESERVED Received ILAS configuration on Lane 0. F is the number of octets per frame. Settings of 1, 2, and 4 are valid (value in register is F − 1). Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0 1 octet per frame. 1 2 octets per frame. 11 4 octets per frame. Reserved. [4:0] K_RD 11111 32 frames per multiframe. 0x406 M_REG [7:0] M_RD Received ILAS configuration on Lane 0. M is the number of converters per device. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. M is 1 for real interface and 2 for complex interface (value in register is M − 1). 0x0 R 0x407 CS_N_REG [7:6] CS_RD Received ILAS configuration on Lane 0. CS is the number of control bits per sample. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. CS is always 0 on this device. 0x0 R 5 Reserved. 0x0 R Received ILAS configuration on 0x0 Lane 0. N is the converter resolution. Value in register is N − 1 (for example, 16 bits = 0b01111). R RESERVED [4:0] N_RD 0x408 NP_REG [7:5] SUBCLASSV_RD Received ILAS configuration on Lane 0. SUBCLASSV is the device subclass version. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 000 Subclass 0. 001 Subclass 1. Rev. 0 | Page 111 of 139 0x0 R AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings [4:0] NP_RD 0x409 S_REG [7:5] JESDV_RD Description Reset Access Received ILAS configuration on 0x0 Lane 0. NP is the total number of bits per sample. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. Value in register is NP − 1, for example, 16 bits per sample = 0b01111. R Received ILAS configuration on 0x0 Lane 0. JESDV is the JESD204x version. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 000 JESD204A. R 001 JESD204B. [4:0] S_RD 0x40A HD_CF_REG 7 Received ILAS configuration on Lane 0. S is the number of samples per converter per frame cycle. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. Value in register is S − 1. HD_RD Received ILAS configuration on Lane 0. HD is the high density format. Refer to Section 5.1.3 of JESD204B standard. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0 Low density mode. 1 High density mode. 0x0 R 0x0 R [6:5] RESERVED Reserved. 0x0 R [4:0] CF_RD Received ILAS configuration on 0x0 Lane 0. CF is the number of control words per frame clock period per link. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. CF is always 0 on this device. R 0x40B RES1_REG [7:0] RES1_RD Received ILAS configuration on Lane 0. Reserved Field 1. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R 0x40C RES2_REG [7:0] RES2_RD Received ILAS configuration on Lane 0. Reserved Field 2. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R 0x40D CHECKSUM0_REG [7:0] LL_FCHK0 Received checksum during ILAS on Lane 0. Checksum for Lane 0. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R 0x40E COMPSUM0_REG [7:0] LL_FCMP0 Computed checksum on Lane 0. 0x0 Computed checksum for Lane 0. The JESD204B Rx computes the checksum of the link information received on Lane 0 as specified in Section 8.3 of JESD204B. The computation method is set by the CHECKSUM_MODE bit (Register 0x300, Bit 6) and must match the likewise calculated checksum in Register 0x40D. R Rev. 0 | Page 112 of 139 Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name Description Reset Access 0x412 LID1_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID1 Received ILAS LID configuration 0x0 on Lane 1. Lane identification for Lane 1. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. R 0x415 CHECKSUM1_REG [7:0] LL_FCHK1 Received checksum during ILAS on lane 1. Checksum for Lane 1. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. 0x0 R 0x416 COMPSUM1_REG [7:0] LL_FCMP1 Computed checksum on Lane 1. Computed checksum for Lane 1 (see description for Register 0x40E). 0x0 R 0x41A LID2_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID2 Received ILAS LID configuration 0x0 on Lane 2. Lane identification for Lane 2. R 0x41D CHECKSUM2_REG [7:0] LL_FCHK2 Received checksum during ILAS on Lane 2. Checksum for Lane 2. 0x0 R 0x41E COMPSUM2_REG [7:0] LL_FCMP2 Computed checksum on Lane 2. 0x0 Computed checksum for Lane 2 (see description for Register 0x40E). R 0x422 LID3_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID3 Received ILAS LID configuration 0x0 on Lane 3. Lane identification for Lane 3. R 0x425 CHECKSUM3_REG [7:0] LL_FCHK3 Received checksum during ILAS on Lane 3. Checksum for Lane 3. 0x0 R 0x426 COMPSUM3_REG [7:0] LL_FCMP3 Computed checksum on Lane 3. 0x0 Computed checksum for Lane 3 (see description for Register 0x40E). R 0x42A LID4_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID4 Received ILAS LID configuration 0x0 on Lane 4. Lane identification for Lane 4. R 0x42D CHECKSUM4_REG [7:0] LL_FCHK4 Received checksum during ILAS on Lane 4. Checksum for Lane 4. 0x0 R 0x42E COMPSUM4_REG [7:0] LL_FCMP4 Computed checksum on Lane 4. 0x0 Computed checksum for Lane 4 (see description for Register 0x40E). R 0x432 LID5_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID5 Received ILAS LID configuration 0x0 on Lane 5. Lane identification for Lane 5. R 0x435 CHECKSUM5_REG [7:0] LL_FCHK5 Received checksum during ILAS on lane 5. Checksum for Lane 5. 0x0 R 0x436 COMPSUM5_REG [7:0] LL_FCMP5 Computed checksum on Lane 5. 0x0 Computed checksum for Lane 5 (see description for Register 0x40E). R 0x43A LID6_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID6 Received ILAS LID configuration 0x0 on Lane 6. Lane identification for Lane 6. R 0x43D CHECKSUM6_REG [7:0] LL_FCHK6 Received checksum during ILAS on Lane 6. Checksum for Lane 6. 0x0 R 0x43E COMPSUM6_REG [7:0] LL_FCMP6 Computed checksum on Lane 6. 0x0 Computed checksum for Lane 6 (see description for Register 0x40E). R Settings Rev. 0 | Page 113 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name Description Reset Access 0x442 LID7_REG [7:5] RESERVED Reserved. 0x0 R [4:0] LL_LID7 Received ILAS LID configuration 0x0 on Lane 7. Lane identification for Lane 7. R 0x445 CHECKSUM7_REG [7:0] LL_FCHK7 Received checksum during ILAS on Lane 7. Checksum for Lane 7. 0x0 R 0x446 COMPSUM7_REG [7:0] LL_FCMP7 Computed checksum on Lane 5. 0x0 Computed checksum for Lane 7 (see description for Register 0x40E). R 0x450 ILS_DID [7:0] DID Device (= link) identification 0x0 number. DID is the device ID number. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. Must be set to the value read in Register 0x400. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x451 ILS_BID [7:0] BID Bank ID, extension to DID. This 0x0 signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x452 ILS_LID0 7 RESERVED Reserved. R 6 ADJDIR Direction to adjust DAC LMFC 0x0 (Subclass 2 only). ADJDIR is the direction to adjust DAC LMFC. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 5 PHADJ Phase adjustment to DAC 0x0 (Subclass 2 only). PHADJ is the phase adjustment request to the DAC. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W Lane identification number 0x0 (within link). LID0 is the lane identification for Lane 0. Link information received on Lane 0 as specified in Section 8.3 of JESD204B. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W Settings [4:0] LID0 Rev. 0 | Page 114 of 139 0x0 Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name 0x453 ILS_SCR_L 7 Settings Description Reset Access 0x1 Scramble enable. SCR is the Rx descrambling enable. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. SCR R/W 0 Descrambling is disabled. 1 Descrambling is enabled. [6:5] RESERVED Reserved. 0x0 R [4:0] L 0x7 Number of lanes per converter (minus 1). L is the number of lanes per converter device. Settings of 1, 2, 3, 4, 6, and 8 are valid. Refer to Table 16 and Table 17. R 0x454 ILS_F [7:0] F Number of octets per frame (minus 1). This value of F is not used to soft configure the QBD. Register CTRLREG1 is used to soft-configure the QBD. 0x0 R 0x455 ILS_K [7:5] RESERVED Reserved. 0x0 [4:0] K 0x1F Number of frames per multiframe (minus 1). K is the number of frames per multiframe. On this device, all modes use K = 32 (value in register is K − 1). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R R/W 01111 16 frames per multiframe. 11111 32 frames per multiframe. 0x456 ILS_M [7:0] M Number of converters per device 0x1 (minus 1). M is the number of converters/device. Settings of 1 and 2 are valid. Refer to Table 16 and Table 17. R 0x457 ILS_CS_N [7:6] CS Number of control bits per sample. CS is the number of control bits per sample. Must be set to 0. Control bits are not supported. 0x0 R Reserved. 5 0x458 ILS_NP 0x0 R [4:0] N RESERVED Converter resolution (minus 1). N 0xF is the converter resolution. Must be set to 16 (0x0F). R [7:5] SUBCLASSV 0x0 Device subclass version. SUBCLASSV is the device subclass version. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 000 Subclass 0. 001 Subclass 1. 010 Subclass 2 (not supported.) Rev. 0 | Page 115 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings [4:0] NP 0x459 ILS_S Description Reset Access Total number of bits per sample 0xF (minus 1) NP is the total number of bits per sample. Must be set to 16 (0x0F). Refer to Table 16 and Table 17. [7:5] JESDV JESD204x version. JESDV is the JESD204x version. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 000 JESD204A. 0x0 R R/W 001 JESD204B. [4:0] S 0x45A ILS_HD_CF 7 Number of samples per 0x1 converter per frame cycle (minus 1). S is the number of samples per converter per frame cycle. Settings of 1 and 2 are valid. Refer to Table 16 and Table 17. HD High density format. HD is the high density mode. Refer to Section 5.1.3 of JESD204B standard. 0 Low density mode. R 0x1 R 1 High density mode. [6:5] RESERVED Reserved. 0x0 R [4:0] CF Number of control bits per sample. CF is the number of control words per frame clock period per link. Must be set to 0. Control bits are not supported. 0x0 R 0x45B ILS_RES1 [7:0] RES1 Reserved. Reserved Field 1. This 0x0 signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x45C ILS_RES2 [7:0] RES2 Reserved. Reserved Field 2. This 0x0 signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x45D ILS_CHECKSUM [7:0] FCHK0 Link configuration checksum. 0x0 Checksum for Lane 0. The checksum for the values programmed into Register 0x450 to Register 0x45C must be calculated according to Section 8.3 of the JESD204B specification and written to this register (SUM(Register 0x450 to Register 0x45C) % 256). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x46C LANE_DESKEW 7 ILD7 Interlane deskew status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. Rev. 0 | Page 116 of 139 0x0 R Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 6 Settings ILS6 Description Reset Access Initial lane synchronization status for Lane 6 (ignore this output when NO_ILAS = 1). 0x0 R 0 Synchronization lost. 1 Synchronization achieved. 5 ILD5 Interlane deskew status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 0x0 R 4 ILD4 Interlane deskew status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 0x0 R 3 ILD3 Interlane deskew status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 0x0 R 2 ILD2 0x0 R 1 ILD1 0x0 R 0 ILD0 0x0 R 7 BDE7 Bad disparity error status for Lane 7. 0x0 R 1 Deskew achieved. Interlane deskew status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. Interlane deskew status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. Interlane deskew status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 0x46D BAD_DISPARITY 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 6 BDE6 5 BDE5 Bad disparity error status for Lane 6. 0x0 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Bad disparity errors status for Lane 5. R 0x0 R 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 BDE4 Bad disparity error status for Lane 4. 0x0 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. R 3 BDE3 Bad disparity error status for Lane 3. 0x0 0 Error count < ETH[7:0] value. R 2 BDE2 1 Error count ≥ ETH[7:0] value. Bad disparity error status for Lane 2. 0x0 R 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 1 BDE1 Bad disparity error status for Lane 1. 0x0 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Rev. 0 | Page 117 of 139 R AD9161/AD9162 Hex. Addr. Name 0x46E NOT_IN_TABLE Data Sheet Bits Bit Name Settings Description Reset Access 0 BDE0 Bad disparity error status for Lane 0. 0x0 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. R 7 NIT7 Not in table error status for Lane 7. 0 Error count < ETH[7:0] value. 0x0 R 6 NIT6 0x0 R Not in table errors status for Lane 5. 0x0 R 1 Error count ≥ ETH[7:0] value. Not in table error status for Lane 6. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 5 NIT5 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 NIT4 3 NIT3 2 NIT2 Not in table error status for Lane 4. 0 Error count < ETH[7:0] value. 0x0 R 0x0 R 0x0 R 1 Error count ≥ ETH[7:0] value. Not in table error status for Lane 3. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Not in table error status for Lane 2. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 1 NIT1 Not in table error status for Lane 1. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 0x0 R 0 NIT0 Not in table error status for Lane 0. 0 Error count < ETH[7:0] value. 0x0 R 7 UEK7 Unexpected K character error status for Lane 7. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 0x0 R 6 UEK6 Unexpected K character error status for Lane 6. 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 1 Error count ≥ ETH[7:0] value. 0x46F UNEXPECTED_KCHAR 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 5 UEK5 4 UEK4 3 UEK3 Unexpected K character error status for Lane 5. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character error status for Lane 4. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character error status for Lane 3. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 2 UEK2 Unexpected K character error status for Lane 2. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 1 UEK1 Unexpected K character error status for Lane 1. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Rev. 0 | Page 118 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 0 UEK0 7 CGS7 6 CGS6 Settings Description Unexpected K character error status for Lane 0. 0 Error count < ETH[7:0] value. Reset Access 0x0 R Code group sync status for Lane 7. 0 Synchronization lost. 1 Synchronization achieved. 0x0 R Code group sync status for Lane 6. 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 1 Error count ≥ ETH[7:0] value. 0x470 CODE_GRP_SYNC 0 Synchronization lost. 1 Synchronization achieved. 5 CGS5 4 CGS4 3 CGS3 Code group sync status for Lane 5. 0 Synchronization lost. 1 Synchronization achieved. Code group sync status for Lane 4. 0 Synchronization lost. 1 Synchronization achieved. Code group sync status for Lane 3. 0 Synchronization lost. 1 Synchronization achieved. 2 CGS2 Code group sync status for Lane 2. 0 Synchronization lost. 1 Synchronization achieved. 1 CGS1 Code group sync status for Lane 1. 0 Synchronization lost. 0 CGS0 Code group sync status for Lane 0. 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 7 FS7 Frame sync status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 6 FS6 Frame sync status for Lane 6 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 5 FS5 0x0 R 4 FS4 0x0 R 1 Synchronization achieved. 0x471 FRAME_SYNC 1 Synchronization achieved. Frame sync status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Frame sync status for Lane 4(ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Rev. 0 | Page 119 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 3 Settings FS3 Description Reset Access Frame sync status for Lane 3 (ignore this output when NO_ILAS = 1). 0x0 R 0x0 R 0 Synchronization lost. 1 Synchronization achieved. 2 FS2 Frame sync status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x472 GOOD_CHECKSUM 1 FS1 Frame sync status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 0 FS0 Frame sync status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 7 CKS7 Computed checksum status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 0x0 R 6 CKS6 0x0 R 5 CKS5 0x0 R 4 CKS4 0x0 R 3 CKS3 0x0 R 0x0 R 0x0 R 1 Checksum is correct. Computed checksum status for Lane 6 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. Computed checksum status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. Computed checksum status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. Computed checksum status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 2 CKS2 Computed checksum status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 1 CKS1 Computed checksum status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. Rev. 0 | Page 120 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 0 Settings CKS0 Description Reset Access Computed checksum status for Lane 0 (ignore this output when NO_ILAS = 1). 0x0 R 0x0 R 0 Checksum is incorrect. 1 Checksum is correct. 0x473 INIT_LANE_SYNC 7 ILS7 Initial lane synchronization status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 6 ILS6 Initial lane synchronization status for Lane 6(ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 5 ILS5 Initial lane synchronization status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 4 ILS4 Initial lane synchronization status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 3 ILS3 0x0 R 2 ILS2 0x0 R 1 ILS1 0x0 R 0 ILS0 0x0 R 1 Synchronization achieved. Initial lane synchronization status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Initial lane synchronization status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Initial lane synchronization status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Initial lane synchronization status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x475 CTRLREG0 7 RX_DIS Level input: disable deframer 0x0 receiver when this input = 1. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 1 Disable character replacement of /A/ and /F/ control characters at the end of received frames and multiframes. 0 Enables the substitution. Rev. 0 | Page 121 of 139 R/W AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 6 Settings CHAR_REPL_DIS Description Reset Access When this input = 1, character 0x0 replacement at the end of frame/multiframe is disabled. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W [5:4] RESERVED Reserved. 0x0 R 3 Soft reset. Active high synchronous reset. Resets all hardware to power-on state. 0x0 R/W SOFTRST 1 Disables the deframer reception. 0 Enable deframer logic. 0x476 CTRLREG1 2 FORCESYNCREQ Command from application to 0x0 assert a sync request (SYNCOUT). Active high. R/W 1 RESERVED Reserved. R 0 REPL_FRM_ENA When this level input is set, it 0x1 enables replacement of frames received in error. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W Reserved. R [7:5] RESERVED 4 QUAL_RDERR 0x0 0x0 Error reporting behavior for 0x1 concurrent NIT and RD errors. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0 NIT has no effect on RD error. R/W 1 NIT error masks concurrent RD error. 3 DEL_SCR Alternative descrambler enable. 0x0 (see JESD204B Section 5.2.4) This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 1 Descrambling begins at Octet 2 of user data. 0 Descrambling begins at Octet 0 of user data. This is the common usage. R/W 2 CGS_SEL Determines the QBD behavior 0x1 after code group sync has been achieved. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0 After code group sync is achieved, the QBD asserts SYNCOUT only if there are sufficient disparity errors as per the JESD204B standard. R/W Rev. 0 | Page 122 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings Description Reset Access 1 After code group sync is achieved, if a /K/ is followed by any character other than an /R/ or another /K/, QBD asserts SYNCOUT. 1 NO_ILAS This signal must only be 0x0 programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 1 For single-lane operation, ILAS is omitted. Code group sync is followed by user data. 0 Code group sync is followed by ILAS. For multilane operation, NO_ILAS must always be set to 0. R/W 0 FCHK_N Checksum calculation method. 0x0 This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Register 3), and must not be changed during normal operation. 0 Calculate checksum by summing individual fields (this more closely matches the definition of the checksum field in the JESD204B standard. R/W 1 Calculate checksum by summing the registers containing the packed fields (this setting is provided in case the framer of another vendor performs the calculation with this method). 0x477 CTRLREG2 7 ILS_MODE Data link layer test mode. This 0x0 signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0 Normal mode. R/W 1 Code group sync pattern is followed by a perpetual ILAS sequence. 6 RESERVED Reserved. 5 REPDATATEST Repetitive data test enable, using 0x0 JTSPAT pattern. To enable the test, ILS_MODE must = 0. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 4 QUETESTERR Queue test error mode. This 0x0 signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0 Simultaneous errors on multiple lanes are reported as one error. 1 Detected errors from all lanes are trapped in a counter and sequentially signaled on SYNCOUT. R/W Rev. 0 | Page 123 of 139 0x0 R AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 3 Settings AR_ECNTR Description Reset Access Automatic reset of error counter. 0x0 The error counter that causes assertion of SYNCOUT is automatically reset to 0 when AR_ECNTR = 1. All other counters are unaffected. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W R [2:0] RESERVED Reserved. 0x478 KVAL [7:0] KSYNC Number of 4 × K multiframes 0x1 during ILS. F is the number of octets per frame. Settings of 1, 2, and 4 are valid. Refer to Table 16 and Table 17. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x0 R/W 0x47C ERRORTHRES [7:0] ETH 0xFF Error threshold value. Bad disparity, NIT disparity, and unexpected K character errors are counted and compared to the error threshold value. When the count is equal, either an IRQ is generated or SYNCOUT± is asserted per the mask register settings or both. Function is performed in all lanes. This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. R/W 0x47D SYNC_ASSERT_MASK [7:3] RESERVED Reserved. 0x0 R [2:0] SYNC_ASSERT_MASK SYNCOUT assertion enable mask for BD, NIT, and UEK error conditions. Active high, SYNCOUT assertion enable mask for BD, NIT, and UEK error conditions, respectively. When an error counter, in any lane, has reached the error threshold count, ETH[7:0], and the corresponding SYNC_ASSERT_ MASK bit is set, SYNCOUT is asserted. The mask bits are as follows. Note that the bit sequence is reversed with respect to the other error count controls and the error counters. 0x7 R/W Bit 2 = bad disparity error (BDE). Bit 1 = not in table error (NIT). Bit 0 = unexpected K (UEK) character error. 0x480 ECNT_CTRL0 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA0 Error counter enable for Lane 0. Counters of each lane are addressed as follows: 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). Rev. 0 | Page 124 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [2:0] ECNT_RST0 Description Reset Access Error counters enable for Lane 0, active high. Counters of each lane are addressed as follows: 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x481 ECNT_CTRL1 0x482 ECNT_CTRL2 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA1 Error counters enable for Lane 1, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x7 R/W [2:0] ECNT_RST1 Error counters enable for Lane 1, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x7 R/W [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA2 Error counters enable for Lane 2, active high. Counters of each lane are addressed as follows: 0x7 R/W 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). [2:0] ECNT_RST2 Error counters enable for Lane 2, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x483 ECNT_CTRL3 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA3 Error counters enable for Lane 3, active high. Counters of each lane are addressed as follows: 0x7 R/W 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). [2:0] ECNT_RST3 Error counters enable for Lane 3, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). Rev. 0 | Page 125 of 139 AD9161/AD9162 Data Sheet Hex. Addr. Name Bits Bit Name Description Reset Access 0x484 ECNT_CTRL4 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA4 Error counters enable for Lane 4, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x7 R/W [2:0] ECNT_RST4 Error counters enable for Lane 4, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. 0x7 R/W Settings Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x485 ECNT_CTRL5 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA5 Error counters enable for Lane 5, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. 0x7 R/W 0x7 R/W Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). [2:0] ECNT_RST5 Error counters enable for Lane 5, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x486 ECNT_CTRL6 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA6 Error counters enable for Lane 6, active high. Counters of each lane are addressed as follows: 0x7 R/W 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). [2:0] ECNT_RST6 Error counters enable for Lane 6, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x487 ECNT_CTRL7 [7:6] RESERVED Reserved. 0x0 R [5:3] ECNT_ENA7 Error counters enable for Lane 7, active high. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x7 R/W Rev. 0 | Page 126 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name Settings [2:0] ECNT_RST7 Description Reset Access Reset error counters for Lane 7, active high. Counters of each lane are addressed as follows: 0x7 R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x488 ECNT_TCH0 [7:3] RESERVED Reserved. 0x0 R [2:0] ECNT_TCH0 Terminal count hold enable of error counters for Lane 0. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: 0x7 R/W 0x0 R Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x489 ECNT_TCH1 [7:3] RESERVED Reserved. [2:0] ECNT_TCH1 Terminal count hold enable of 0x7 error counters for Lane 1. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). R/W Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x48A ECNT_TCH2 [7:3] RESERVED Reserved. [2:0] ECNT_TCH2 Terminal count hold enable of 0x7 error counters for Lane 2. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). Rev. 0 | Page 127 of 139 0x0 R R/W AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings Description Reset Access This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x48B ECNT_TCH3 [7:3] RESERVED Reserved. [2:0] ECNT_TCH3 Terminal count hold enable of 0x7 error counters for Lane 3. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: 0x0 R R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x48C ECNT_TCH4 [7:3] RESERVED Reserved. [2:0] ECNT_TCH4 Terminal count hold enable of 0x7 error counters for Lane 4. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: 0x0 R R/W Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x48D ECNT_TCH5 [7:3] RESERVED Reserved. [2:0] ECNT_TCH5 Terminal count hold enable of 0x7 error counters for Lane 5. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. Rev. 0 | Page 128 of 139 0x0 R R/W Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name Description Reset Access 0x48E ECNT_TCH6 [7:3] RESERVED Reserved. 0x0 R [2:0] ECNT_TCH6 Terminal count hold enable of error counters for Lane 6. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. 0x7 R/W Settings Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x48F ECNT_TCH7 [7:3] RESERVED Reserved. 0x0 R [2:0] ECNT_TCH7 Terminal count hold enable of error counters for Lane 7. When set, the designated counter is to hold the terminal count value of 0xFF when it is reached until the counter is reset by the user. Otherwise, the designated counter rolls over. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x7 R/W 0x0 R 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 0. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows. Bit 2 = unexpected K (UEK) character error. R This signal must only be programmed while the QBD is held in soft reset (Register 0x475, Bit 3), and must not be changed during normal operation. 0x490 ECNT_STAT0 [7:4] RESERVED 3 Reserved. LANE_ENA0 This output indicates if lane0 is enabled. 0 Lane0 is held in soft reset. 1 Lane0 is enabled. [2:0] ECNT_TCR0 Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x491 ECNT_STAT1 [7:4] RESERVED Reserved. 0x0 R 3 This output indicates if Lane 1 is enabled. 0x0 R LANE_ENA1 0 Lane 1 is held in soft reset. 1 Lane 1 is enabled. Rev. 0 | Page 129 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name Settings [2:0] ECNT_TCR1 Description Reset Access Terminal count reached indicator 0x0 of error counters for Lane 1. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows. Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). R Bit 0 = bad disparity error (BDE). 0x492 ECNT_STAT2 [7:4] RESERVED 0x0 R 0x0 R [2:0] ECNT_TCR2 Terminal count reached indicator 0x0 of error counters for Lane 2. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows. Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). R [7:4] RESERVED Reserved. 0x0 R 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 3. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows: R 3 Reserved. LANE_ENA2 This output indicates if Lane 2 is enabled. 0 Lane 2 is held in soft reset. 1 Lane 2 is enabled. 0x493 ECNT_STAT3 3 LANE_ENA3 This output indicates if Lane 3 is enabled. 0 Lane 3 is held in soft reset. 1 Lane 3 is enabled. [2:0] ECNT_TCR3 Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x494 ECNT_STAT4 [7:4] RESERVED Reserved. 0x0 R 3 This output indicates if Lane 4 is enabled. 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 4. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows: R LANE_ENA4 0 Lane 4 is held in soft reset. 1 Lane 4 is enabled. [2:0] ECNT_TCR4 Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). Rev. 0 | Page 130 of 139 Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name Description Reset Access 0x495 ECNT_STAT5 [7:4] RESERVED Reserved. 0x0 R 3 This output indicates if Lane 5 is enabled. 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 5. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows: R Settings LANE_ENA5 0 Lane 5 is held in soft reset. 1 Lane 5 is enabled. [2:0] ECNT_TCR5 Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x496 ECNT_STAT6 [7:4] RESERVED Reserved. 0x0 R 3 This output indicates if Lane 6 is enabled. 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 6. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. Bit 1 = not in table error (NIT). R LANE_ENA6 0 Lane 6 is held in soft reset. 1 Lane 6 is enabled. [2:0] ECNT_TCR6 Bit 0 = bad disparity error (BDE). 0x497 ECNT_STAT7 [7:4] RESERVED 3 Reserved. LANE_ENA7 0x0 R 0x0 R Terminal count reached indicator 0x0 of error counters for Lane 7. Set to 1 when the corresponding counter terminal count value of 0xFF has been reached. Counters of each lane are addressed as follows: Bit 2 = unexpected K (UEK) character error. R This output indicates if Lane 7 is enabled. 0 Lane 7 is held in soft reset. 1 Lane 7 is enabled. [2:0] ECNT_TCR7 Bit 1 = not in table error (NIT). Bit 0 = bad disparity error (BDE). 0x4B0 LINK_STATUS0 7 BDE0 6 NIT0 5 UEK0 Bad disparity errors status for Lane 0. 0 Error count < ETH[7:0] value. 0x0 R 0x0 R 0x0 R 1 Error count ≥ ETH[7:0] value. Not in table errors status for Lane 0. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character errors status for Lane 0. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Rev. 0 | Page 131 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 4 Settings ILD0 Description Reset Access Interlane deskew status for Lane 0 (ignore this output when NO_ILAS = 1). 0x0 R 0x0 R 0 Deskew failed. 1 Deskew achieved. 3 ILS0 Initial lane synchronization status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 2 CKS0 Computed checksum status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 0x0 R 1 FS0 Frame sync status for Lane 0 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 0 CGS0 0x0 R 0x0 R 0x0 R 0x0 R 1 Synchronization achieved. Code group sync status for Lane 0. 0 Synchronization lost. 1 Synchronization achieved. 0x4B1 LINK_STATUS1 7 BDE1 Bad Disparity errors status for Lane 1. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 6 NIT1 Not in table errors status for Lane 1. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 5 UEK1 Unexpected K character errors status for Lane 1. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 ILD1 Interlane deskew status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 0x0 R 3 ILS1 Initial lane synchronization status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 2 CKS1 0x0 R 1 FS1 0x0 R 1 Synchronization achieved. Computed checksum status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. Frame sync status for Lane 1 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Rev. 0 | Page 132 of 139 Data Sheet Hex. Addr. Name AD9161/AD9162 Bits Bit Name 0 CGS1 7 BDE2 6 NIT2 Settings Description Code group sync status for Lane 1. 0 Synchronization lost. Reset Access 0x0 R Bad Disparity errors status for Lane 2. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 0x0 R Not in table errors status for Lane 2. 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 1 Synchronization achieved. 0x4B2 LINK_STATUS2 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 5 UEK2 4 ILD2 Unexpected K character errors status for Lane 2. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Interlane deskew status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 3 ILS2 Initial lane synchronization status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 2 CKS2 Computed checksum status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 1 FS2 Frame sync status for Lane 2 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0 CGS2 Code group sync status for Lane 2. 0 Synchronization lost. 1 Synchronization achieved. 0x4B3 LINK_STATUS3 7 BDE3 6 NIT3 5 UEK3 Bad Disparity errors status for Lane 3. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Not in table errors status for Lane 3. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character errors status for Lane 3. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 ILD3 Interlane deskew status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. Rev. 0 | Page 133 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 3 Settings ILS3 Description Reset Access Initial lane synchronization status for Lane 3 (ignore this output when NO_ILAS = 1). 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0 Synchronization lost. 1 Synchronization achieved. 2 CKS3 Computed checksum status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 1 FS3 0 CGS3 Frame sync status for Lane 3 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Code group sync status for Lane 3. 0 Synchronization lost. 1 Synchronization achieved. 0x4B4 LINK_STATUS4 7 BDE4 Bad Disparity errors status for Lane 4. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 6 NIT4 5 UEK4 Not in table errors status for Lane 4. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character errors status for Lane 4. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 ILD4 Interlane deskew status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 3 ILS4 Initial lane synchronization status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 2 CKS4 Computed checksum status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 0x0 R 1 FS4 0x0 R 0 CGS4 0x0 R 1 Checksum is correct. Frame sync status for Lane 4 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Code group sync status for Lane 4. 0 Synchronization lost. 1 Synchronization achieved. Rev. 0 | Page 134 of 139 Data Sheet AD9161/AD9162 Hex. Addr. Name Bits Bit Name 0x4B5 LINK_STATUS5 7 BDE5 6 NIT5 5 UEK5 Settings Description Bad disparity errors status for Lane 5. 0 Error count < ETH[7:0] value. Reset Access 0x0 R 0x0 R 0x0 R 0x0 R 1 Error count ≥ ETH[7:0] value. Not in table errors status for Lane 5. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character errors status for Lane 5. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 ILD5 Interlane deskew status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 3 ILS5 Initial lane synchronization status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0x0 R 2 CKS5 Computed checksum status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 0x0 R 1 FS5 Frame sync status for Lane 5 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 0 CGS5 0x0 R 0x0 R 0x0 R 1 Synchronization achieved. Code group sync status for Lane 5. 0 Synchronization lost. 1 Synchronization achieved. 0x4B6 LINK_STATUS6 7 BDE6 6 NIT6 Bad Disparity errors status for Lane 6. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Not in table errors status for Lane 6. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 5 UEK6 Unexpected K character errors status for Lane 6. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 0x0 R 4 ILD6 Interlane deskew status for Lane 6 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 0x0 R 3 ILS6 Initial lane synchronization status for Lane 6 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 0x0 R 1 Synchronization achieved. Rev. 0 | Page 135 of 139 AD9161/AD9162 Hex. Addr. Name Data Sheet Bits Bit Name 2 Settings CKS6 Description Reset Access Computed checksum status for Lane 6 (ignore this output when NO_ILAS = 1). 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0x0 R 0 Checksum is incorrect. 1 Checksum is correct. 1 FS6 Frame sync status for Lane 6 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 0 CGS6 Code group sync status for Lane 6. 0 Synchronization lost. 1 Synchronization achieved. 0x4B7 LINK_STATUS7 7 BDE7 6 NIT7 5 UEK7 Bad Disparity errors status for Lane 7. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Not in table errors status for Lane 7. 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. Unexpected K character errors status for Lane 7 0 Error count < ETH[7:0] value. 1 Error count ≥ ETH[7:0] value. 4 ILD7 Interlane deskew status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Deskew failed. 1 Deskew achieved. 3 ILS7 Initial lane synchronization status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. 2 CKS7 Computed checksum status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Checksum is incorrect. 1 Checksum is correct. 1 FS7 0 CGS7 Frame sync status for Lane 7 (ignore this output when NO_ILAS = 1). 0 Synchronization lost. 1 Synchronization achieved. Code group sync status for Lane 7. 0 Synchronization lost. 1 Synchronization achieved. 0x4B8 JESD_IRQ_ENABLEA 7 EN_BDE Bad disparity error counter. 0x0 R/W 6 EN_NIT Not in table error counter. 0x0 R/W 5 EN_UEK Unexpected K error counter. 0x0 R/W 4 EN_ILD Interlane deskew. 0x0 R/W 3 EN_ILS Initial lane sync. 0x0 R/W Rev. 0 | Page 136 of 139 Data Sheet Hex. Addr. Name 0x4B9 JESD_IRQ_ENABLEB 0x4BA JESD_IRQ_STATUSA 0x4BB JESD_IRQ_STATUSB 0x800 HOPF_CTRL AD9161/AD9162 Bits Bit Name Settings Description Reset Access 2 EN_CKS Good checksum. This is an 0x0 interrupt that compares two checksums: the checksum that the transmitter sent over the link during the ILAS, and the checksum that the receiver calculated from the ILAS data that the transmitter sent over the link. Note that the checksum IRQ never at any time looks at the checksum that is programmed over the SPI into Register 0x45D. The checksum IRQ only looks at the data sent by the transmitter, and never looks at any data programmed via the SPI. R/W 1 EN_FS Frame sync. 0x0 R/W 0 EN_CGS Code group sync. 0x0 R/W [7:1] RESERVED Reserved. 0x0 R 0 EN_ILAS Configuration mismatch 0x0 (checked for Lane 0 only). The ILAS IRQ compares the two sets of ILAS data that the receiver has: the ILAS data sent over the JESD204B link by the transmitter, and the ILAS data programmed into the receiver via the SPI (Register 0x450 to Register 0x45D). If the data differs, the IRQ is triggered. Note that all of the ILAS data (including the checksum) is compared. R/W 7 IRQ_BDE Bad disparity error counter. 0x0 R/W 6 IRQ_NIT Not in table error counter. 0x0 R/W 5 IRQ_UEK Unexpected K error counter. 0x0 R/W 4 IRQ_ILD Interlane deskew. 0x0 R/W 3 IRQ_ILS Initial lane sync. 0x0 R/W 2 IRQ_CKS Good checksum. 0x0 R/W 1 IRQ_FS Frame sync. 0x0 R/W 0 IRQ_CGS Code group sync. 0x0 R/W [7:1] RESERVED Reserved. 0x0 R 0 Configuration mismatch (checked for Lane 0 only). 0x0 R/W Frequency switch mode. 0x0 00 Phase continuous switch. Changes frequency tuning word, and the phase accumulator continues to accumulate to the new FTW. 01 Phase discontinuous switch. Changes the frequency tuning word and resets the phase accumulator. 10 Reserved. R/W IRQ_ILAS [7:6] HOPF_MODE [5:0] RESERVED Reserved. Rev. 0 | Page 137 of 139 0x0 R AD9161/AD9162 Data Sheet OUTLINE DIMENSIONS 8.05 8.00 SQ 7.95 5.85 BSC A1 BALL CORNER 15 14 13 12 11 10 9 8 7 6 5 4 3 2 A1 BALL CORNER 1 A B C D E 7.00 REF SQ F G H 5.895 BSC J 0.50 BSC K L M N P R 0.50 REF TOP VIEW DETAIL A 0.35 0.30 0.25 DETAIL A 0.24 REF 0.22 NOM 0.15 MIN 0.35 COPLANARITY 0.30 0.08 0.25 BALL DIAMETER PKG-004576 SEATING PLANE 10-28-2014-A 0.86 MAX 0.76 MOM BOTTOM VIEW Figure 194. 165-Ball Chip Scale Package Ball Grid Array [CSP_BGA] (BC-165-1) Dimensions shown in millimeters A1 BALL CORNER 11.05 11.00 SQ 10.95 A1 BALL PAD CORNER 1.285 BSC 13 12 11 10 9 8 7 6 5 4 3 2 1 A 5.935 BSC B C D E F G H J K L M N 9.60 REF SQ 0.80 BSC TOP VIEW 2.405 BSC BOTTOM VIEW 0.70 REF 5.890 BSC DETAIL A *0.95 MAX PKG-004675 SEATING PLANE DETAIL A 0.45 0.40 0.35 BALL DIAMETER 0.31 NOM 0.21 MIN COPLANARITY 0.12 *COMPLIANT TO JEDEC STANDARDS MO-275-FFAC-1 WITH THE EXCEPTION OF PACKAGE HEIGHT AND THICKNESS. Figure 195. 169-Ball Chip Scale Package Ball Grid Array [CSP_BGA] (BC-169-2) Dimensions shown in millimeters Rev. 0 | Page 138 of 139 07-10-2015-A 0.35 0.30 0.25 Data Sheet AD9161/AD9162 ORDERING GUIDE Model 1 AD9161BBCZ AD9161BBCZRL AD9162BBCZ AD9162BBCZRL AD9162BBCAZ AD9162BBCAZRL AD9162BBCA AD9162BBCARL AD9161-FMCC-EBZ AD9162-FMC-EBZ AD9162-FMCB-EBZ AD9162-FMCC-EBZ 1 Temperature Range −40°C to +85°C −40°C to +85°C −40°C to +85°C −40°C to +85°C −40°C to +85°C −40°C to +85°C −40°C to +85°C −40°C to +85°C Package Description 169-Ball Chip Scale Package Ball Grid Array [CSP_BGA] 169-Ball Chip Scale Package Ball Grid Array [CSP_BGA] 165-Ball Chip Scale Package Ball Grid Array [CSP_BGA] 165-Ball Chip Scale Package Ball Grid Array [CSP_BGA] 169 Ball Chip Scale Package Ball Grid Array [CSP_BGA] 169-Ball Chip Scale Package Ball Grid Array [CSP_BGA] 169 Ball Chip Scale Package Ball Grid Array (CSP_BGA) 169-Ball Chip Scale Package Ball Grid Array [CSP_BGA] Evaluation Board Evaluation Board For 8 × 8 mm Package with High Accuracy Balance Balun Evaluation Board For 8 × 8 mm Package with Balun and Match Optimized For Wider Output Bandwidth Evaluation Board for 11 × 11 mm Package with Balun and Match Optimized for Wider Output Bandwidth Z = RoHS Compliant Part. ©2016 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D14379-0-5/16(0) Rev. 0 | Page 139 of 139 Package Option BC-169-2 BC-169-2 BC-165-1 BC-165-1 BC-169-2 BC-169-2 BC-169-2 BC-169-2