LMK04906 Ultra Low Noise Clock Jitter Cleaner/Multiplier with 6 Programmable Outputs 1.0 General Description 3.0 Features The LMK04906 is the industry's highest performance clock jitter attenuator with superior clock jitter cleaning, generation, and distribution with advanced features to meet high performance timing application needs. The LMK04906 accepts 3 clock input ranging from 1 kHz to 750 MHz and generates 6 unique clock output frequencies ranging from 2.26 MHz to 2.6 GHz. The LMK04906 can also buffer a crystal or VCXO to generate a 7th unique clock frequency. The device provides virtually all frequency translation combination required for SONET, Ethernet, Fibre Channel and multi-mode Wireless Base Stations. The LMK04906 input clock frequency and clock multiplication ratio are programmable through a SPI interface. ● Ultra-Low RMS Jitter Performance Device VCO Frequency LMK04906 2370 to 2600 MHz ● 2.0 Target Applications • • • • • • • • • • • • ● 10G/40G/100G OTN line cards SONET/SDH OC-48/STM-16 and OC-192/STM-64line cards GbE/10GbE, 1/2/4/8/10GFC line cards ITU G.709 and custom FEC line cards Synchronous Ethernet Optical modules DSLAM/MSANs Test and measurement Broadcast video Wireless basestations Data converter clocking Microwave ODU and IDUs for Wireless Backhaul ● ● ● ● ● ● ● ● ● ● ● ● — 100 fs RMS jitter (12 kHz to 20 MHz) — 123 fs RMS jitter (100 Hz to 20 MHz) Dual Loop PLLatinum™ PLL Architecture — PLL1 ● Integrated Low-Noise Crystal Oscillator Circuit ● Holdover mode when input clocks are lost — Automatic or manual triggering/recovery PLL2 — ● Normalized [1 Hz] PLL noise floor of -227 dBc/Hz ● Phase detector rate up to 155 MHz ● OSCin frequency-doubler ● Integrated Low-Noise VCO 3 redundant input clocks with LOS — Automatic and manual switch-over modes 50% duty cycle output divides, 1 to 1045 (even and odd) LVPECL, LVDS, or LVCMOS programmable outputs Precision digital delay, fixed or dynamically adjustable 25 ps step analog delay control. 6 differential outputs. Up to 12 single ended. — Up to 5 VCXO/Crystal buffered outputs Clock rates of up to 2600 MHz 0-delay mode Three default clock outputs at power up Multi-mode: Dual PLL, single PLL, and clock distribution Industrial Temperature Range: -40 to 85 °C 3.15 V to 3.45 V operation Package: 64-pin QFN (9.0 x 9.0 x 0.8 mm) 30179139 System Application Diagram PLLatinum™ is a trademark of National Semiconductor Corporation. TRI-STATE® is a registered trademark of National Semiconductor Corporation. PRODUCTION DATA information is current as of publication date. Products conform to specifications per the terms of the Texas Instruments standard warranty. Production processing does not necessarily include testing of all parameters. 301791 SNAS589B Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 30179140 Simplified LMK04906 Block Diagram 4.0 Device Configuration Information NSID Reference Inputs Dedicated Buffered/ Divided OSCin Clock Programmable LVDS/ LVPECL/LVCMOS Outputs (Note 1) VCO LMK04906BISQ 3 1 6 2370 to 2600 MHz Note 1: Up to 2 of these outputs are also able to be driven by the OSCin clock. 2 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 5.0 Common Frequency Plans Standard/Application Output Frequencies Infiniband 100, 200 SATA 75, 150, 300, 600 SAS 37.5, 75, 120, 150 Fast Ethernet 25 1 GbE 125 10 GbE 156.25, 312.5, 625 XAUI 78.125, 156.25, 312.5 VCO Frequency Recommended Crystal Value 2400 MHz 2500 MHz 20 MHz Backplane 227.27... 2G/4G/16G Fiber Channel 106.25, 212.5 10G Fiber Channel 159.375 40/100 GbE 644.53125, 322.265625, 161.1328125 2578.125 MHz 12.5 MHz SONET 19.44, 38.88, 77.76, 155.52, 311.04, 622.08 2488.32 MHz 19.44 MHz A/D Clocking 30.72, 61.44, 122.88, 153.6, 245.76, 491.52, 983.04 2457.6 MHz 19.2 MHz or 12.288 MHz 2550 MHz 6.0 Achievable Frequencies By using the tunable range of the VCO followed by a programmable divider, the LMK04906 can achieve any of the frequencies in the table below Output Divider Value Achieved Frequency (MHz) 1 2370 - 2600 2 1185 - 1300 3 790 - 866.7 4 592.5 - 650 5 474 - 520 6 395.7 - 433 7 338.6 - 371.4 8 296.25 - 325 9 263.3 - 288.9 10 263.3 - 288.9 10 237 - 260 11 to 1045 Any frequency in the range of 2.27 - 236.36 Copyright © 1999-2012, Texas Instruments Incorporated 3 LMK04906 7.0 Functional Block Diagrams and Operating Modes The LMK04906 is a flexible device that can be configured for many different use cases. The following simplified block diagrams help show the user the different use cases of the device. 7.1 Dual PLL Figure 1 illustrates the typical use case of the LMK04906 in dual loop mode. In dual loop mode the reference to PLL1 is either CLKin0, CLKin1, or CLKin2. An external VCXO or tunable crystal will be used to provide feedback for the first PLL and a reference to the second PLL. This first PLL cleans the jitter with the VCXO or low cost tunable crystal by using a narrow loop bandwidth. The VCXO or tunable crystal output may be buffered through the OSCout0 port and optionally on up to 2 of the CLKouts. The VCXO or tunable crystal is used as the reference to PLL2 and may be doubled using the frequency doubler. The internal VCO drives up to six divide/delay blocks which drive 6 clock outputs. Holdover functionality is optionally available when the input reference clock is lost. Holdover works by fixing the tuning voltage of PLL1 to the VCXO or tunable crystal. It is also possible to use an external VCO in place of PLL2's internal VCO. 30179108 FIGURE 1. Simplified Functional Block Diagram for Dual Loop Mode 4 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 7.2 0-Delay Dual PLL Figure 2 illustrates the use case of 0-delay dual loop mode. This configuration is very similar to Section 7.1 Dual PLL except that the feedback to the first PLL is driven by a clock output. This causes the clock outputs to have deterministic phase with the clock input. Since all the clock outputs can be synchronized together, all the clock outputs can be in phase with the clock input signal. The feedback to PLL1 can be connected internally as shown, or externally using FBCLKin (CLKin1) as an input port. It is also possible to use an external VCO in place of PLL2's internal VCO. 30179109 FIGURE 2. Simplified Functional Block Diagram for 0-delay Dual Loop Mode 7.3 Single PLL Figure 3 illustrates the use case of single PLL mode. In single PLL mode only PLL2 is used and PLL1 is powered down. OSCin is used as the reference input. The internal VCO drives up to 6 divide/delay blocks which drive 6 clock outputs. The reference at OSCin can be used to drive the OSCout0 port. OSCin can also optionally drive up to 2 of the clock outputs. It is also possible to use an external VCO in place of PLL2's internal VCO. 30179110 FIGURE 3. Simplified Functional Block Diagram for Single Loop Mode Copyright © 1999-2012, Texas Instruments Incorporated 5 LMK04906 7.4 0-delay Single PLL Figure 4 illustrates the use case of 0-delay single PLL mode. This configuration is very similar to Section 7.3 Single PLL except that the feedback to PLL2 comes from a clock output. This causes the clock outputs to be in phase with the reference input. Since all the clock outputs can be synchronized together, all the clock outputs can be in phase with the clock input signal. The feedback to PLL2 can be performed internally as shown, or externally using FBCLKin (CLKin1) as an input port. It is also possible to use an external VCO in place of PLL2's internal VCO. 30179111 FIGURE 4. Simplified Functional Block Diagram for 0-delay Single Loop Mode 7.5 Clock Distribution Figure 5 illustrates the LMK04906 used for clock distribution. CLKin1 is used to drive up to 6 divide/delay blocks which drive 6 outputs. OSCin can be used to drive the OSCout port. OSCin can also optionally drive up to 2 of the clock outputs. 30179112 FIGURE 5. Simplified Functional Block Diagram for Mode Clock Distribution 6 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 7.6 Detailed LMK04906 Block Diagram Figure 6 illustrates the complete LMK04906 block diagram for the LMK04906 family. 30179101 FIGURE 6. Detailed LMK04906 Block Diagram Copyright © 1999-2012, Texas Instruments Incorporated 7 LMK04906 Table of Contents 1.0 General Description ......................................................................................................................... 1 2.0 Target Applications .......................................................................................................................... 1 3.0 Features ........................................................................................................................................ 1 4.0 Device Configuration Information ....................................................................................................... 2 5.0 Common Frequency Plans ............................................................................................................... 3 6.0 Achievable Frequencies ................................................................................................................... 3 7.0 Functional Block Diagrams and Operating Modes ................................................................................ 4 7.1 Dual PLL ................................................................................................................................. 4 7.2 0-Delay Dual PLL ..................................................................................................................... 5 7.3 Single PLL .............................................................................................................................. 5 7.4 0-delay Single PLL ................................................................................................................... 6 7.5 Clock Distribution ..................................................................................................................... 6 7.6 Detailed LMK04906 Block Diagram ............................................................................................. 7 8.0 Connection Diagram ...................................................................................................................... 12 9.0 Pin Descriptions (Note 2) ................................................................................................................ 13 10.0 Absolute Maximum Ratings ........................................................................................................... 15 11.0 Package Thermal Resistance ........................................................................................................ 15 12.0 Recommended Operating Conditions ............................................................................................ 15 13.0 Electrical Characteristics ............................................................................................................... 16 14.0 Serial MICROWIRE Timing Diagram .............................................................................................. 24 14.1 ADVANCED MICROWIRE TIMING DIAGRAMS ....................................................................... 24 14.1.1 3 Extra Clocks or Double Program ................................................................................ 24 14.1.2 Three Extra Clocks with LEuWire High .......................................................................... 25 14.1.3 Readback .................................................................................................................. 25 15.0 Measurement Definitions .............................................................................................................. 26 15.1 CHARGE PUMP CURRENT SPECIFICATION DEFINITIONS .................................................... 26 15.1.1 Charge Pump Output Current Magnitude Variation Vs. Charge Pump Output Voltage .......... 26 15.1.2 Charge Pump Sink Current Vs. Charge Pump Output Source Current Mismatch ................. 26 15.1.3 Charge Pump Output Current Magnitude Variation Vs. Temperature ................................. 26 15.2 DIFFERENTIAL VOLTAGE MEASUREMENT TERMINOLOGY (Note 27) .................................... 27 16.0 Typical Performance Characteristics .............................................................................................. 28 16.2 CLOCK OUTPUT AC CHARACTERISTICS ............................................................................. 28 17.0 Features ..................................................................................................................................... 29 17.1 SYSTEM ARCHITECTURE ................................................................................................... 29 17.2 PLL1 REDUNDANT REFERENCE INPUTS (CLKin0/CLKin0*, CLKin1/CLKin1*, and CLKin2/ CLKin2*) .................................................................................................................................. 29 17.3 PLL1 TUNABLE CRYSTAL SUPPORT ................................................................................... 29 17.4 VCXO/CRYSTAL BUFFERED OUTPUTS ................................................................................ 29 17.5 FREQUENCY HOLDOVER ................................................................................................... 29 17.6 HITLESS SWITCHING .......................................................................................................... 29 17.7 INTEGRATED LOOP FILTER POLES ..................................................................................... 30 17.8 INTERNAL VCO ................................................................................................................... 30 17.9 EXTERNAL VCO MODE ....................................................................................................... 30 17.10 CLOCK DISTRIBUTION ..................................................................................................... 30 17.10.1 CLKout DIVIDER ...................................................................................................... 30 17.10.2 CLKout DELAY ......................................................................................................... 30 17.10.3 PROGRAMMABLE OUTPUT TYPE ............................................................................ 30 17.10.4 CLOCK OUTPUT SYNCHRONIZATION ...................................................................... 30 17.11 0-DELAY ........................................................................................................................... 31 17.12 DEFAULT STARTUP CLOCKS ............................................................................................ 31 17.13 STATUS PINS ................................................................................................................... 31 17.14 REGISTER READBACK ...................................................................................................... 31 18.0 Functional Description .................................................................................................................. 32 18.1 FUNCTIONAL OVERVIEW .................................................................................................... 32 18.2 MODE SELECTION .............................................................................................................. 32 18.3 INPUTS / OUTPUTS ............................................................................................................. 33 18.3.1 PLL1 Reference Inputs (CLKin0, CLKin1, and CLKin2) .................................................... 33 18.3.2 PLL2 OSCin / OSCin* Port ........................................................................................... 33 18.3.3 CRYSTAL OSCILLATOR ............................................................................................. 33 18.4 INPUT CLOCK SWITCHING .................................................................................................. 33 18.4.1 Input Clock Switching - Manual Mode ............................................................................ 33 18.4.2 Input Clock Switching - Pin Select Mode ........................................................................ 34 18.4.3 Input Clock Switching - Automatic Mode ........................................................................ 34 18.4.4 Input Clock Switching - Automatic Mode with Pin Select ................................................... 35 8 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 18.5 HOLDOVER MODE .............................................................................................................. 18.5.1 Holdover Frequency Accuracy and DAC Performance ..................................................... 18.5.2 Holdover Mode - Automatic Exit of Holdover ................................................................... 18.6 PLLs ................................................................................................................................... 18.6.1 PLL1 ......................................................................................................................... 18.6.2 PLL2 ......................................................................................................................... 18.6.2.1 PLL2 FREQUENCY DOUBLER ......................................................................... 18.6.3 DIGITAL LOCK DETECT ............................................................................................. 18.7 STATUS PINS ..................................................................................................................... 18.7.1 Logic Low .................................................................................................................. 18.7.2 Digital Lock Detect ...................................................................................................... 18.7.3 Holdover Status .......................................................................................................... 18.7.4 DAC ......................................................................................................................... 18.7.5 PLL Divider Outputs .................................................................................................... 18.7.6 CLKinX_LOS ............................................................................................................. 18.7.7 CLKinX Selected ........................................................................................................ 18.7.8 MICROWIRE Readback .............................................................................................. 18.8 VCO ................................................................................................................................... 18.9 CLOCK DISTRIBUTION ........................................................................................................ 18.9.1 Fixed Digital Delay ...................................................................................................... 18.9.1.1 FIXED DIGITAL DELAY - EXAMPLE ................................................................... 18.9.2 Clock Output Synchronization (SYNC) ........................................................................... 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY .............................................. 18.9.2.1.1 Absolute Dynamic Digital Delay ................................................................ 18.9.2.1.1.1 ABSOLUTE DYNAMIC DIGITAL DELAY - EXAMPLE ........................ 18.9.2.1.2 Relative Dynamic Digital Delay ................................................................. 18.9.2.1.2.1 RELATIVE DYNAMIC DIGITAL DELAY - EXAMPLE .......................... 18.9.3 0-Delay Mode ............................................................................................................ 19.0 General Programming Information ................................................................................................. 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY ................. 19.1.1 Example .................................................................................................................... 19.2 RECOMMENDED PROGRAMMING SEQUENCE .................................................................... 19.2.1 Overview ................................................................................................................... 19.3 READBACK ......................................................................................................................... 19.3.1 Readback - Example ................................................................................................... 19.4 REGISTER MAP AND READBACK REGISTER MAP ............................................................... 19.5 DEFAULT DEVICE REGISTER SETTINGS AFTER POWER ON RESET .................................... 19.6 REGISTER R0 TO R5 ........................................................................................................... 19.6.1 CLKoutX_PD, Powerdown CLKoutX Output Path ............................................................ 19.6.2 CLKoutX_OSCin_Sel, Clock output source .................................................................... 19.6.3 CLKoutX_ADLX_SEL[29], CLKoutX_ADLX_SEL[28], Select Analog Delay ......................... 19.6.4 CLKoutX_DDLY, Clock Channel Digital Delay ................................................................ 19.6.5 RESET ...................................................................................................................... 19.6.6 POWERDOWN .......................................................................................................... 19.6.7 CLKoutX_HS, Digital Delay Half Shift ............................................................................ 19.6.8 CLKoutX_DIV, Clock Output Divide ............................................................................... 19.7 REGISTERS R6 TO R8 ......................................................................................................... 19.7.1 CLKoutX_TYPE ......................................................................................................... 19.7.2 CLKoutX_ADLY ......................................................................................................... 19.8 REGISTER R10 ................................................................................................................... 19.8.1 OSCout0_TYPE ......................................................................................................... 19.8.2 EN_OSCout0, OSCout0 Output Enable ......................................................................... 19.8.3 OSCout0_MUX, Clock Output Mux ............................................................................... 19.8.4 PD_OSCin, OSCin Powerdown Control ......................................................................... 19.8.5 OSCout_DIV, Oscillator Output Divide ........................................................................... 19.8.6 VCO_MUX ................................................................................................................ 19.8.7 EN_FEEDBACK_MUX ................................................................................................ 19.8.8 VCO_DIV, VCO Divider ............................................................................................... 19.8.9 FEEDBACK_MUX ...................................................................................................... 19.9 REGISTER R11 ................................................................................................................... 19.9.1 MODE: Device Mode .................................................................................................. 19.9.2 EN_SYNC, Enable Synchronization .............................................................................. 19.9.3 NO_SYNC_CLKoutX .................................................................................................. 19.9.4 SYNC_CLKin2_MUX .................................................................................................. 19.9.5 SYNC_QUAL ............................................................................................................. 19.9.6 SYNC_POL_INV ........................................................................................................ Copyright © 1999-2012, Texas Instruments Incorporated 36 37 37 37 37 37 37 38 38 38 39 39 39 39 39 39 39 39 39 39 40 40 44 45 45 48 48 50 51 51 52 52 52 53 53 53 58 64 64 64 64 65 65 66 66 66 67 67 68 69 69 69 70 70 70 70 70 71 71 72 72 72 73 73 74 74 9 LMK04906 19.9.7 SYNC_EN_AUTO ....................................................................................................... 19.9.8 SYNC_TYPE ............................................................................................................. 19.9.9 EN_PLL2_XTAL ......................................................................................................... 19.10 REGISTER R12 ................................................................................................................. 19.10.1 LD_MUX .................................................................................................................. 19.10.2 LD_TYPE ................................................................................................................ 19.10.3 SYNC_PLLX_DLD .................................................................................................... 19.10.4 EN_TRACK ............................................................................................................. 19.10.5 HOLDOVER_MODE ................................................................................................. 19.11 REGISTER R13 ................................................................................................................. 19.11.1 HOLDOVER_MUX .................................................................................................... 19.11.2 HOLDOVER_TYPE ................................................................................................... 19.11.3 Status_CLKin1_MUX ................................................................................................ 19.11.4 Status_CLKin0_TYPE ............................................................................................... 19.11.5 DISABLE_DLD1_DET ............................................................................................... 19.11.6 Status_CLKin0_MUX ................................................................................................ 19.11.7 CLKin_SELECT_MODE ............................................................................................ 19.11.8 CLKin_Sel_INV ........................................................................................................ 19.11.9 EN_CLKinX ............................................................................................................. 19.12 REGISTER 14 .................................................................................................................... 19.12.1 LOS_TIMEOUT ........................................................................................................ 19.12.2 EN_LOS .................................................................................................................. 19.12.3 Status_CLKin1_TYPE ............................................................................................... 19.12.4 CLKinX_BUF_TYPE, PLL1 CLKinX/CLKinX* Buffer Type ............................................... 19.12.5 DAC_HIGH_TRIP ..................................................................................................... 19.12.6 DAC_LOW_TRIP ...................................................................................................... 19.12.7 EN_VTUNE_RAIL_DET ............................................................................................. 19.13 REGISTER 15 .................................................................................................................... 19.13.1 MAN_DAC ............................................................................................................... 19.13.2 EN_MAN_DAC ......................................................................................................... 19.13.3 HOLDOVER_DLD_CNT ............................................................................................ 19.13.4 FORCE_HOLDOVER ................................................................................................ 19.14 REGISTER 16 .................................................................................................................... 19.14.1 XTAL_LVL ............................................................................................................... 19.15 REGISTER 23 .................................................................................................................... 19.15.1 DAC_CNT ............................................................................................................... 19.16 REGISTER 24 .................................................................................................................... 19.16.1 PLL2_C4_LF, PLL2 Integrated Loop Filter Component .................................................. 19.16.2 PLL2_C3_LF, PLL2 Integrated Loop Filter Component .................................................. 19.16.3 PLL2_R4_LF, PLL2 Integrated Loop Filter Component .................................................. 19.16.4 PLL2_R3_LF, PLL2 Integrated Loop Filter Component .................................................. 19.16.5 PLL1_N_DLY ........................................................................................................... 19.16.6 PLL1_R_DLY ........................................................................................................... 19.16.7 PLL1_WND_SIZE ..................................................................................................... 19.17 REGISTER 25 .................................................................................................................... 19.17.1 DAC_CLK_DIV ......................................................................................................... 19.17.2 PLL1_DLD_CNT ....................................................................................................... 19.18 REGISTER 26 .................................................................................................................... 19.18.1 PLL2_WND_SIZE ..................................................................................................... 19.18.2 EN_PLL2_REF_2X, PLL2 Reference Frequency Doubler ............................................... 19.18.3 PLL2_CP_POL, PLL2 Charge Pump Polarity ................................................................ 19.18.4 PLL2_CP_GAIN, PLL2 Charge Pump Current .............................................................. 19.18.5 PLL2_DLD_CNT ....................................................................................................... 19.18.6 PLL2_CP_TRI, PLL2 Charge Pump TRI-STATE ........................................................... 19.19 REGISTER 27 .................................................................................................................... 19.19.1 PLL1_CP_POL, PLL1 Charge Pump Polarity ................................................................ 19.19.2 PLL1_CP_GAIN, PLL1 Charge Pump Current .............................................................. 19.19.3 CLKinX_PreR_DIV .................................................................................................... 19.19.4 PLL1_R, PLL1 R Divider ............................................................................................ 19.19.5 PLL1_CP_TRI, PLL1 Charge Pump TRI-STATE ........................................................... 19.20 REGISTER 28 .................................................................................................................... 19.20.1 PLL2_R, PLL2 R Divider ............................................................................................ 19.20.2 PLL1_N, PLL1 N Divider ............................................................................................ 19.21 REGISTER 29 .................................................................................................................... 19.21.1 OSCin_FREQ, PLL2 Oscillator Input Frequency Register ............................................... 19.21.2 PLL2_FAST_PDF, High PLL2 Phase Detector Frequency .............................................. 10 74 74 75 75 75 75 75 76 76 77 77 77 77 78 78 78 79 79 79 80 80 80 80 80 81 81 81 82 82 82 82 82 83 83 83 83 83 83 84 84 85 85 85 85 86 86 86 86 86 86 87 87 87 87 87 87 88 88 88 88 88 88 89 90 90 90 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.21.3 PLL2_N_CAL, PLL2 N Calibration Divider .................................................................... 90 19.22 REGISTER 30 .................................................................................................................... 91 19.22.1 PLL2_P, PLL2 N Prescaler Divider .............................................................................. 91 19.22.2 PLL2_N, PLL2 N Divider ............................................................................................ 91 19.23 REGISTER 31 .................................................................................................................... 91 19.23.1 READBACK_LE ....................................................................................................... 91 19.23.2 READBACK_ADDR .................................................................................................. 92 19.23.3 uWire_LOCK ............................................................................................................ 92 20.0 Application Information ................................................................................................................. 93 20.1 FREQUENCY PLANNING WITH THE LMK04906 FAMILY (Note 37) ........................................... 93 20.2 PLL PROGRAMMING ........................................................................................................... 94 20.2.1 Example PLL2 N Divider Programming .......................................................................... 94 20.3 LOOP FILTER ..................................................................................................................... 95 20.3.1 PLL1 ......................................................................................................................... 95 20.3.2 PLL2 ......................................................................................................................... 95 20.4 SYSTEM LEVEL DIAGRAM ................................................................................................... 96 20.5 PIN CONNECTION RECOMMENDATIONS ............................................................................. 98 20.5.1 Vcc Pins and Decoupling ............................................................................................. 98 20.5.1.1 Vcc2, Vcc3, Vcc10, Vcc11, Vcc12, Vcc13 (CLKout Vccs) ....................................... 98 20.5.1.2 Vcc1 (VCO), Vcc4 (Digital), and Vcc9 (PLL2) ........................................................ 98 20.5.1.3 Vcc6 (PLL1 Charge Pump) and Vcc8 (PLL2 Charge Pump) .................................... 98 20.5.1.4 Vcc5 (CLKin), Vcc7 (OSCin & OSCout0) .............................................................. 98 20.5.2 LVPECL Outputs ........................................................................................................ 98 20.5.3 Unused Clock Outputs ................................................................................................ 98 20.5.4 Unused Clock Inputs ................................................................................................... 98 20.5.5 LDO Bypass .............................................................................................................. 98 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY ............................................................... 99 20.6.1 Minimum Lock Time Calculation Example ...................................................................... 99 20.7 CALCULATING DYNAMIC DIGITAL DELAY VALUES FOR ANY DIVIDE ................................... 100 20.7.1 Example ................................................................................................................... 100 20.8 OPTIONAL CRYSTAL OSCILLATOR IMPLEMENTATION (OSCin/OSCin*) ................................ 102 20.9 DRIVING CLKin AND OSCin INPUTS .................................................................................... 106 20.9.1 Driving CLKin Pins with a Differential Source ................................................................. 106 20.9.2 Driving CLKin Pins with a Single-Ended Source ............................................................. 106 20.10 TERMINATION AND USE OF CLOCK OUTPUT (DRIVERS) ................................................... 107 20.10.1 Termination for DC Coupled Differential Operation ....................................................... 107 20.10.2 Termination for AC Coupled Differential Operation ....................................................... 108 20.10.3 Termination for Single-Ended Operation ..................................................................... 109 20.11 POWER SUPPLY .............................................................................................................. 111 20.11.1 Current Consumption / Power Dissipation Calculations ................................................. 111 20.12 THERMAL MANAGEMENT ................................................................................................. 113 21.0 Physical Dimensions ................................................................................................................... 114 22.0 Ordering Information ................................................................................................................... 114 Copyright © 1999-2012, Texas Instruments Incorporated 11 LMK04906 8.0 Connection Diagram 64-Pin QFN Package 30179102 12 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 9.0 Pin Descriptions Pin Number Name(s) (Note 2) I/O Type Description Power Supply for CLKou0 1 Vcc13 PWR 2, 5, 7, 8, 9, 15, 17, 19 22, 47, 51, 55, 56, 60, 61, 64 NC No Connect 3, 4 CLKout0*, CLKout0 O Programmable Clock output 0. 6 SYNC / Status_CLKin2 I/O Programmable CLKout Synchronization input or CLKin2 Status output. 10 Vcc1 PWR Power supply for VCO LDO. 11 LDObyp1 ANLG LDO Bypass, bypassed to ground with 10 µF capacitor. 12 LDObyp2 ANLG LDO Bypass, bypassed to ground with a 0.1 µF capacitor. 13, 14 CLKout1, CLKout1* 16 Vcc2 PWR Power supply for CLKout1. 18 Vcc3 PWR Power supply for CLKout2 20, 21 CLKout2*, CLKout2 O O Programmable Programmable These pins must be left floating. Clock output 1. Clock output 2 23 GND PWR Ground 24 Vcc4 PWR Power supply for digital. Reference Clock Input Port 1 for PLL1. AC or DC Coupled. CLKin1, CLKin1* 25, 26 FBCLKin, FBCLKin* I ANLG Feedback input for external clock feedback input (0delay mode). AC or DC Coupled. External VCO input (External VCO mode). AC or DC Coupled. Fin/Fin* Programmable status pin, default readback output. Programmable to holdover mode indicator. Other options available by programming. 27 Status_Holdover I/O Programmable 28, 29 CLKin0, CLKin0* I ANLG Reference Clock Input Port 0 for PLL1. AC or DC Coupled. 30 Vcc5 PWR Power supply for clock inputs. 31, 32 CLKin2, CLKin2* I ANLG Reference Clock Input Port 2 for PLL1, AC or DC Coupled. 33 Status_LD I/O Programmable 34 CPout1 O ANLG Charge pump 1 output. 35 Vcc6 PWR Power supply for PLL1, charge pump 1. 36, 37 OSCin, OSCin* ANLG Feedback to PLL1, Reference input to PLL2. AC Coupled. 38 Vcc7 PWR Power supply for OSCin port. 39, 40 OSCout0, OSCout0* 41 Vcc8 42 CPout2 43 Vcc9 44 LEuWire I Programmable status pin, default lock detect for PLL1 and PLL2. Other options available by programming. O Programmable PWR Power supply for PLL2, charge pump 2. O ANLG Charge pump 2 output. PWR Power supply for PLL2. I CMOS Copyright © 1999-2012, Texas Instruments Incorporated Buffered output 0 of OSCin port. MICROWIRE Latch Enable Input. 13 LMK04906 Pin Number Name(s) I/O Type 45 CLKuWire I CMOS MICROWIRE Clock Input. 46 DATAuWire I CMOS MICROWIRE Data Input. 48 Vcc10 49, 50 CLKout3, CLKout3* 52 Vcc11 53, 54 CLKout4, CLKout4* 57 Vcc12 58, 59 CLKout5, CLKout5* 62 Status_CLKin0 63 Status_CLKin1 DAP DAP PWR O Programmable PWR O Programmable PWR O I/O I/O Description Power supply for CLKout3. Clock output 3. Power supply for CLKout4. Clock output 4. Power supply for CLKout5. Programmable Clock output 5. Programmable Programmable status pin. Default is input for pin control of PLL1 reference clock selection. CLKin0 LOS status and other options available by programming. Programmable Programmable status pin. Default is input for pin control of PLL1 reference clock selection. CLKin1 LOS status and other options available by programming. GND DIE ATTACH PAD, connect to GND. Note 2: See Application Information section for recommended connections. 14 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 10.0 Absolute Maximum Ratings (Note 3, Note 4, Note 5) If Military/Aerospace specified devices are required, please contact the Texas Instruments Sales Office/ Distributors for availability and specifications. Parameter Ratings Units Supply Voltage (Note 6) Symbol VCC -0.3 to 3.6 V Input Voltage VIN -0.3 to (VCC + 0.3) V Storage Temperature Range TSTG -65 to 150 °C Lead Temperature (solder 4 seconds) TL +260 °C Junction Temperature TJ 150 °C IIN ±5 mA MSL 3 Differential Input Current (CLKinX/X*, OSCin/OSCin*, FBCLKin/FBCLKin*, Fin/Fin*) Moisture Sensitivity Level Note 3: "Absolute Maximum Ratings" indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device is intended to be functional, but do not guarantee specific performance limits. For guaranteed specifications and test conditions, see the Electrical Characteristics. The guaranteed specifications apply only to the test conditions listed. Note 4: This device is a high performance RF integrated circuit with an ESD rating up to 2 kV Human Body Model, up to 150 V Machine Model, and up to 750 V Charged Device Model and is ESD sensitive. Handling and assembly of this device should only be done at ESD-free workstations. Note 5: Stresses in excess of the absolute maximum ratings can cause permanent or latent damage to the device. These are absolute stress ratings only. Functional operation of the device is only implied at these or any other conditions in excess of those given in the operation sections of the data sheet. Exposure to absolute maximum ratings for extended periods can adversely affect device reliability. Note 6: Never to exceed 3.6 V. 11.0 Package Thermal Resistance 64-Lead QFN Parameter Symbol Ratings Units Thermal resistance from junction to ambient on 4-layer JEDEC PCB (Note 7) θJA 19.5 ° C/W Thermal resistance from junction to case (Note 8) θJC 1.5 ° C/W Note 7: Specification assumes 32 thermal vias connect the die attach pad to the embedded copper plane on the 4-layer JEDEC PCB. These vias play a key role in improving the thermal performance of the QFN. Note that the JEDEC PCB is a standard thermal measurement PCB and does not represent best performance a PCB can achieve. It is recommended that the maximum number of vias be used in the board layout. θJA is unique for each PCB. Note 8: Case is defined as the DAP (die attach pad). 12.0 Recommended Operating Conditions Parameter Junction Temperature Ambient Temperature Supply Voltage Symbol Condition Min Typical TJ TA VCC = 3.3 V VCC Copyright © 1999-2012, Texas Instruments Incorporated Max Unit 125 °C -40 25 85 °C 3.15 3.3 3.45 V 15 LMK04906 13.0 Electrical Characteristics (3.15 V ≤ VCC ≤ 3.45 V, -40 °C ≤ TA ≤ 85 °C. Typical values represent most likely parametric norms at VCC = 3.3 V, TA = 25 °C, at the Recommended Operating Conditions at the time of product characterization and are not guaranteed.) Symbol Parameter Conditions Min Typ Max Units 1 3 mA 410 470 mA 500 MHz Current Consumption ICC_PD Power Down Supply Current ICC_CLKS Supply Current with all clocks enabled (Note 10) fCLKin Clock Input Frequency (Note 11) SLEWCLKin Clock Input Slew Rate (Note 25) All clock delays disabled, CLKoutX_DIV = 1045, CLKoutX_TYPE = 1 (LVDS), PLL1 and PLL2 locked. CLKin0/0*, CLKin1/1*, and CLKin2/2* Input Clock Specifications VIDCLKin VSSCLKin VIDCLKin VSSCLKin VCLKin Clock Input Differential Input Voltage (Note 9) Figure 11 Clock Input Single-ended Input Voltage (Note 25) VCLKin0-offset DC offset voltage between CLKin0/ CLKin0* CLKin0* - CLKin0 VCLKin1-offset DC offset voltage between CLKin1/ CLKin1* CLKin1* - CLKin1 VCLKin2-offset DC offset voltage between CLKin2/ CLKin2* CLKin2* - CLKin2 VCLKinX-offset DC offset voltage between CLKinX/ CLKinX* CLKinX* - CLKinX VCLKin- VIH High input voltage VCLKin- VIL Low input voltage 0.001 20% to 80% 0.15 AC coupled CLKinX_BUF_TYPE = 0 (Bipolar) 0.25 1.55 |V| 0.5 3.1 Vpp AC coupled CLKinX_BUF_TYPE = 1 (MOS) 0.25 1.55 |V| 0.5 3.1 Vpp AC coupled to CLKinX; CLKinX* AC coupled to Ground CLKinX_BUF_TYPE = 0 (Bipolar) 0.25 2.4 Vpp AC coupled to CLKinX; CLKinX* AC coupled to Ground CLKinX_BUF_TYPE = 1 (MOS) 0.25 2.4 Vpp Each pin AC coupled CLKin0_BUF_TYPE = 0 (Bipolar) Each pin AC coupled CLKinX_BUF_TYPE = 1 (MOS) DC coupled to CLKinX; CLKinX* AC coupled to Ground CLKinX_BUF_TYPE = 1 (MOS) 0.5 V/ns 20 mV 0 mV 20 mV 55 mV 2.0 VCC V 0.0 0.4 V FBCLKin/FBCLKin* and Fin/Fin* Input Specifications fFBCLKin Clock Input Frequency (Note 25) AC coupled (CLKinX_BUF_TYPE = 0) MODE = 2 or 8; FEEDBACK_MUX = 6 0.001 1000 MHz fFin Clock Input Frequency (Note 25) AC coupled (CLKinX_BUF_TYPE = 0) MODE = 3 or 11 0.001 3100 MHz VFBCLKin/Fin Single Ended Clock Input Voltage (Note 25) AC coupled; (CLKinX_BUF_TYPE = 0) 0.25 2.0 Vpp SLEWFBCLKin/Fin Slew Rate on CLKin (Note 25) AC coupled; 20% to 80%; (CLKinX_BUF_TYPE = 0) 0.15 16 0.5 V/ns Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Symbol Parameter Conditions fPD1 PLL1 Phase Detector Frequency Min Typ Max Units 40 MHz PLL1 Specifications ICPout1SOURCE PLL1 Charge Pump Source Current (Note 13) VCPout1 = VCC/2, PLL1_CP_GAIN = 0 100 VCPout1 = VCC/2, PLL1_CP_GAIN = 1 200 VCPout1 = VCC/2, PLL1_CP_GAIN = 2 400 VCPout1 = VCC/2, PLL1_CP_GAIN = 3 1600 VCPout1=VCC/2, PLL1_CP_GAIN = 0 -100 VCPout1=VCC/2, PLL1_CP_GAIN = 1 -200 VCPout1=VCC/2, PLL1_CP_GAIN = 2 -400 µA ICPout1SINK PLL1 Charge Pump Sink Current (Note 13) VCPout1=VCC/2, PLL1_CP_GAIN = 3 -1600 ICPout1%MIS Charge Pump Sink / Source Mismatch VCPout1 = VCC/2, T = 25 °C 3 ICPout1VTUNE Magnitude of Charge Pump Current Variation vs. Charge Pump Voltage 0.5 V < VCPout1 < VCC - 0.5 V TA = 25 °C 4 % ICPout1%TEMP Charge Pump Current vs. Temperature Variation 4 % ICPout1 TRI Charge Pump TRI-STATE®Leakage Current PN10kHz PLL 1/f Noise at 10 kHz offset. Normalized to 1 GHz Output Frequency PN1Hz Normalized Phase Noise Contribution 0.5 V < VCPout < VCC - 0.5 V µA 10 5 PLL1_CP_GAIN = 400 µA -117 PLL1_CP_GAIN = 1600 µA -118 PLL1_CP_GAIN = 400 µA -221.5 PLL1_CP_GAIN = 1600 µA -223 % nA dBc/Hz dBc/Hz PLL2 Reference Input (OSCin) Specifications fOSCin PLL2 Reference Input (Note 14) SLEWOSCin PLL2 Reference Clock minimum slew rate on OSCin(Note 25) 20% to 80% 0.15 VOSCin Input Voltage for OSCin or OSCin* (Note 25) AC coupled; Single-ended (Unused pin AC coupled to GND) 0.2 2.4 Vpp Differential voltage swing Figure 11 AC coupled 0.2 1.55 |V| 0.4 3.1 Vpp VOSCin-offset DC offset voltage between OSCin/ OSCin* OSCinX* - OSCinX Each pin AC coupled fdoubler_max Doubler input frequency (Note 25) EN_PLL2_REF_2X = 1; OSCin Duty Cycle 40% to 60% VIDOSCin VSSOSCin Copyright © 1999-2012, Texas Instruments Incorporated 500 0.5 MHz V/ns 20 mV 155 MHz 17 LMK04906 Symbol Parameter Conditions Min Typ Max Units fXTAL Crystal Frequency Range (Note 25) RESR < 40 Ω 20.5 MHz PXTAL Crystal Power Dissipation (Note 16) Vectron VXB1 crystal, 20.48 MHz, RESR < 40 Ω XTAL_LVL = 0 100 µW CIN Input Capacitance of LMK04906 OSCin port -40 to +85 °C 6 pF fPD2 Phase Detector Frequency Crystal Oscillator Mode Specifications 6 PLL2 Phase Detector and Charge Pump Specifications ICPoutSOURCE ICPoutSINK 18 PLL2 Charge Pump Source Current (Note 13) PLL2 Charge Pump Sink Current (Note 13) 155 MHz VCPout2=VCC/2, PLL2_CP_GAIN = 0 100 VCPout2=VCC/2, PLL2_CP_GAIN = 1 400 VCPout2=VCC/2, PLL2_CP_GAIN = 2 1600 VCPout2=VCC/2, PLL2_CP_GAIN = 3 3200 VCPout2=VCC/2, PLL2_CP_GAIN = 0 -100 VCPout2=VCC/2, PLL2_CP_GAIN = 1 -400 VCPout2=VCC/2, PLL2_CP_GAIN = 2 -1600 VCPout2=VCC/2, PLL2_CP_GAIN = 3 -3200 3 4 % 4 % ICPout2%MIS Charge Pump Sink/Source Mismatch VCPout2=VCC/2, TA = 25 °C ICPout2VTUNE Magnitude of Charge Pump Current vs. Charge Pump Voltage Variation 0.5 V < VCPout2 < VCC - 0.5 V TA = 25 °C ICPout2%TEMP Charge Pump Current vs. Temperature Variation ICPout2TRI Charge Pump Leakage 0.5 V < VCPout2 < VCC - 0.5 V PLL2_CP_GAIN = 400 µA -118 PN10kHz PLL 1/f Noise at 10 kHz offset (Note 17). Normalized to 1 GHz Output Frequency PLL2_CP_GAIN = 3200 µA -121 PN1Hz Normalized Phase Noise Contribution (Note 18) PLL2_CP_GAIN = 400 µA -222.5 PLL2_CP_GAIN = 3200 µA -227 µA µA 10 10 % nA dBc/Hz dBc/Hz Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Symbol Parameter Conditions fVCO VCO Tuning Range LMK04906 KVCO Fine Tuning Sensitivity (The range displayed in the typical column indicates the lower sensitivity is typical at the lower end of the tuning range, and the higher tuning sensitivity is typical at the higher end of the tuning range). LMK04906 |ΔTCL| Allowable Temperature Drift for Continuous Lock (Note 19, Note 25) After programming R30 for lock, no changes to output configuration are permitted to guarantee continuous lock Min Typ Max Units 2600 MHz Internal VCO Specifications 2370 16 to 21 MHz/V 125 °C CLKout Closed Loop Jitter Specifications using a Commercial Quality VCXO (Note 22) L(f)CLKout JCLKout LVDS/LVPECL/ LVCMOS Offset = 1 kHz LMK04906 Offset = 10 kHz fCLKout = 245.76 MHz Offset = 100 kHz SSB Phase Noise Offset = 800 kHz Measured at Clock Outputs Offset = 10 MHz; LVDS Value is average for all output types Offset = 10 MHz; LVPECL 1600 mVpp (Note 20) Offset = 10 MHz; LVCMOS LMK04906(Note 20) fCLKout = 245.76 MHz Integrated RMS Jitter -122.5 -132.9 -135.2 -143.9 dBc/Hz -156.0 -157.5 -157.1 BW = 12 kHz to 20 MHz 115 BW = 100 Hz to 20 MHz 123 fs rms CLKout Closed Loop Jitter Specifications using the Integrated Low Noise Crystal Oscillator Circuit (Note 23) LMK04906 fCLKout = 245.76 MHz Integrated RMS Jitter Copyright © 1999-2012, Texas Instruments Incorporated BW = 12 kHz to 20 MHz XTAL_LVL = 3 192 BW = 100 Hz to 20 MHz XTAL_LVL = 3 450 19 LMK04906 Symbol Parameter fCLKout-startup Default output clock frequency at device power on (Note 24) Conditions Min Typ Max Units 90 98 110 MHz Default Power On Reset Clock Output Frequency CLKout4, LVDS, LMK04906 Clock Skew and Delay LVDS-to-LVDS, T = 25 °C, FCLK = 800 MHz, RL= 100 Ω AC coupled 30 LVPECL-to-LVPECL, T = 25 °C, FCLK = 800 MHz, RL= 100 Ω emitter resistors = 240 Ω to GND AC coupled 30 Maximum skew between any two LVCMOS outputs, same CLKout or different CLKout (Note 21, Note 25) RL = 50 Ω, CL = 5 pF, T = 25 °C, FCLK = 100 MHz. (Note 21) 100 LVDS or LVPECL to LVCMOS Same device, T = 25 °C, 250 MHz 750 MODE = 2 PLL1_R_DLY = 0; PLL1_N_DLY = 0 1850 MODE = 2 PLL1_R_DLY = 0; PLL1_N_DLY = 0; VCO Frequency = 2949.12 MHz Analog delay select = 0; Feedback clock digital delay = 11; Feedback clock half step = 1; Output clock digital delay = 5; Output clock half step = 0; 0 Maximum CLKoutX to CLKoutY (Note 21, Note 25) |TSKEW| MixedTSKEW td0-DELAY CLKin to CLKoutX delay (Note 21) ps ps ps LVDS Clock Outputs (CLKoutX), CLKoutX_TYPE = 1 fCLKout VOD VSS Differential Output Voltage Figure 12 ΔVOD Change in Magnitude of VOD for complementary output states VOS Output Offset Voltage ΔVOS Change in VOS for complementary output states RL = 100 Ω 1536 T = 25 °C, DC measurement AC coupled to receiver input R = 100 Ω differential termination MHz 250 400 450 |mV| 500 800 900 mVpp 50 mV 1.375 V 35 |mV| -50 1.125 1.25 Output Rise Time 20% to 80%, RL = 100 Ω Output Fall Time 80% to 20%, RL = 100 Ω ISA ISB Output short circuit current - single ended Single-ended output shorted to GND, T = 25 °C -24 24 mA ISAB Output short circuit current differential Complimentary outputs tied together -12 12 mA TR / TF 20 Maximum Frequency (Note 25, Note 26) 200 ps Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Symbol Parameter fCLKout Maximum Frequency (Note 25, Note 26) Conditions Min Typ Max Units LVPECL Clock Outputs (CLKoutX) 20% to 80% Output Rise TR / TF 80% to 20% Output Fall Time 1536 RL = 100 Ω, emitter resistors = 240 Ω to GND CLKoutX_TYPE = 4 or 5 (1600 or 2000 mVpp) MHz 150 ps VCC 1.03 V VCC 1.41 V 700 mVpp LVPECL Clock Outputs (CLKoutX), CLKoutX_TYPE = 2 VOH Output High Voltage VOL Output Low Voltage VOD Output Voltage Figure 12 VSS T = 25 °C, DC measurement Termination = 50 Ω to VCC - 1.4 V 305 380 440 |mV| 610 760 880 mVpp 1200 mVpp LVPECL Clock Outputs (CLKoutX), CLKoutX_TYPE = 3 VOH Output High Voltage VOL Output Low Voltage VOD Output Voltage Figure 12 VSS T = 25 °C, DC measurement Termination = 50 Ω to VCC - 1.7 V VCC 1.07 V VCC 1.69 V 545 625 705 |mV| 1090 1250 1410 mVpp 1600 mVpp LVPECL Clock Outputs (CLKoutX), CLKoutX_TYPE = 4 VOH Output High Voltage VOL Output Low Voltage VOD Output Voltage Figure 12 VSS T = 25 °C, DC Measurement Termination = 50 Ω to VCC - 2.0 V VCC 1.10 V VCC 1.97 V 660 870 965 |mV| 1320 1740 1930 mVpp 2000 mVpp LVPECL (2VPECL) Clock Outputs (CLKoutX), CLKoutX_TYPE = 5 VOH Output High Voltage VOL Output Low Voltage VOD Output Voltage Figure 12 VSS Copyright © 1999-2012, Texas Instruments Incorporated T = 25 °C, DC Measurement Termination = 50 Ω to VCC - 2.3 V VCC 1.13 V VCC 2.20 V 800 1070 1200 |mV| 1600 2140 2400 mVpp 21 LMK04906 Symbol Parameter Conditions Min Typ Max Units fCLKout Maximum Frequency (Note 25, Note 26) 5 pF Load 250 MHz VOH Output High Voltage 1 mA Load VCC 0.1 V VOL Output Low Voltage 1 mA Load IOH Output High Current (Source) VCC = 3.3 V, VO = 1.65 V 28 mA IOL Output Low Current (Sink) VCC = 3.3 V, VO = 1.65 V 28 mA DUTYCLK Output Duty Cycle (Note 25) VCC/2 to VCC/2, FCLK = 100 MHz, T = 25 °C TR Output Rise Time 20% to 80%, RL = 50 Ω, CL = 5 pF 400 ps TF Output Fall Time 80% to 20%, RL = 50 Ω, CL = 5 pF 400 ps LVCMOS Clock Outputs (CLKoutX) 0.1 45 50 55 V % Digital Outputs (Status_CLKinX, Status_LD, Status_Holdover, SYNC) VOH High-Level Output Voltage IOH = -500 µA VOL Low-Level Output Voltage IOL = 500 µA VCC 0.4 V 0.4 V VCC V 0.4 V Digital Inputs (Status_CLKinX, SYNC) VIH High-Level Input Voltage VIL Low-Level Input Voltage IIH IIL High-Level Input Current VIH = VCC Low-Level Input Current VIL = 0 V 1.6 Status_CLKinX_TYPE = 0 (High Impedance) -5 5 Status_CLKinX_TYPE = 1 (Pull-up) -5 5 Status_CLKinX_TYPE = 2 (Pull-down) 10 80 Status_CLKinX_TYPE = 0 (High Impedance) -5 5 Status_CLKinX_TYPE = 1 (Pull-up) -40 -5 Status_CLKinX_TYPE = 2 (Pull-down) -5 5 1.6 VCC V 0.4 V µA µA Digital Inputs (CLKuWire, DATAuWire, LEuWire) 22 VIH High-Level Input Voltage VIL Low-Level Input Voltage IIH High-Level Input Current VIH = VCC 5 25 µA IIL Low-Level Input Current VIL = 0 -5 5 µA Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Symbol Parameter Conditions Min Typ Max Units TECS LE to Clock Set Up Time See MICROWIRE Input Timing 25 ns TDCS Data to Clock Set Up Time See MICROWIRE Input Timing 25 ns TCDH Clock to Data Hold Time See MICROWIRE Input Timing 8 ns TCWH Clock Pulse Width High See MICROWIRE Input Timing 25 ns TCWL Clock Pulse Width Low See MICROWIRE Input Timing 25 ns TCES Clock to LE Set Up Time See MICROWIRE Input Timing 25 ns MICROWIRE Interface Timing TEWH LE Pulse Width See MICROWIRE Input Timing 25 ns TCR Falling Clock to Readback Time See MICROWIRE Readback Timing 25 ns Note 9: See Section 15.2 DIFFERENTIAL VOLTAGE MEASUREMENT TERMINOLOGY (Note 27) for definition of VID and VOD voltages. Note 10: Load conditions for output clocks: LVDS: 100 Ω differential. See applications section Section 20.11.1 Current Consumption / Power Dissipation Calculations for Icc for specific part configuration and how to calculate Icc for a specific design. Note 11: CLKin0, CLKin1, and CLKin2 maximum is guaranteed by characterization, production tested at 200 MHz. Note 12: In order to meet the jitter performance listed in the subsequent sections of this data sheet, the minimum recommended slew rate for all input clocks is 0.5 V/ns. This is especially true for single-ended clocks. Phase noise performance will begin to degrade as the clock input slew rate is reduced. However, the device will function at slew rates down to the minimum listed. When compared to single-ended clocks, differential clocks (LVDS, LVPECL) will be less susceptible to degradation in phase noise performance at lower slew rates due to their common mode noise rejection. However, it is also recommended to use the highest possible slew rate for differential clocks to achieve optimal phase noise performance at the device outputs. Note 13: This parameter is programmable Note 14: FOSCin maximum frequency guaranteed by characterization. Production tested at 200 MHz. Note 15: The EN_PLL2_REF_2X bit (Register 13) enables/disables a frequency doubler mode for the PLL2 OSCin path. Note 16: See Application Section discussion of Crystal Power Dissipation. Section 20.8 OPTIONAL CRYSTAL OSCILLATOR IMPLEMENTATION (OSCin/ OSCin*) Note 17: A specification in modeling PLL in-band phase noise is the 1/f flicker noise, LPLL_flicker(f), which is dominant close to the carrier. Flicker noise has a 10 dB/decade slope. PN10kHz is normalized to a 10 kHz offset and a 1 GHz carrier frequency. PN10kHz = LPLL_flicker(10 kHz) - 20log(Fout / 1 GHz), where LPLL_flicker (f) is the single side band phase noise of only the flicker noise's contribution to total noise, L(f). To measure LPLL_flicker(f) it is important to be on the 10 dB/decade slope close to the carrier. A high compare frequency and a clean crystal are important to isolating this noise source from the total phase noise, L(f). LPLL_flicker(f) can be masked by the reference oscillator performance if a low power or noisy source is used. The total PLL in-band phase noise performance is the sum of LPLL_flicker(f) and LPLL_flat(f). Note 18: A specification modeling PLL in-band phase noise. The normalized phase noise contribution of the PLL, LPLL_flat(f), is defined as: PN1HZ=LPLL_flat(f) 20log(N) - 10log(fPDX). LPLL_flat(f) is the single side band phase noise measured at an offset frequency, f, in a 1 Hz bandwidth and fPDX is the phase detector frequency of the synthesizer. LPLL_flat(f) contributes to the total noise, L(f). Note 19: Maximum Allowable Temperature Drift for Continuous Lock is how far the temperature can drift in either direction from the value it was at the time that the R30 register was last programmed, and still have the part stay in lock. The action of programming the R30 register, even to the same value, activates a frequency calibration routine. This implies the part will work over the entire frequency range, but if the temperature drifts more than the maximum allowable drift for continuous lock, then it will be necessary to reload the R30 register to ensure it stays in lock. Regardless of what temperature the part was initially programmed at, the temperature can never drift outside the frequency range of -40 °C to 85 °C without violating specifications. Note 20: fVCO = 2457.6 MHz, PLL1 parameters: EN_PLL2_REF_2X = 1, PLL2_R = 2, FPD1 = 1.024 MHz, ICP1 = 100 μA, loop bandwidth = 10 Hz. A 122.88 MHz Crystek CVHD-950–122.880. PLL2 parameters: PLL2_R = 1, FPD2 = 122.88 MHz, ICP2 = 3200 μA, C1 = 47 pF, C2 = 3.9 nF, R2 = 620 Ω, PLL2_C3_LF = 0, PLL2_R3_LF = 0, PLL2_C4_LF = 0, PLL2_R4_LF = 0, CLKoutX_DIV = 10, and CLKoutX_ADLY_SEL = 0. Note 21: Equal loading and identical clock output configuration on each clock output is required for specification to be valid. Specification not valid for delay mode. Note 22: VCXO used is a 122.88 MHz Crystek CVHD-950-122.880. Note 23: Crystal used is a 20.48 MHz Vectron VXB1-1150-20M480 and Skyworks varactor diode, SMV-1249-074LF. Note 24: CLKout3 and OSCout0 also oscillate at start-up at the frequency of the VCXO attached to OSCin port. Note 25: Guaranteed by characterization. Note 26: Refer to typical performance charts for output operation performance at higher frequencies than the minimum maximum output frequency. Copyright © 1999-2012, Texas Instruments Incorporated 23 LMK04906 14.0 Serial MICROWIRE Timing Diagram Register programming information on the DATAuWire pin is clocked into a shift register on each rising edge of the CLKuWire signal. On the rising edge of the LEuWire signal, the register is sent from the shift register to the register addressed. A slew rate of at least 30 V/µs is recommended for these signals. After programming is complete the CLKuWire, DATAuWire, and LEuWire signals should be returned to a low state. If the CLKuWire or DATAuWire lines are toggled while the VCO is in lock, as is sometimes the case when these lines are shared with other parts, the phase noise may be degraded during this programming. 30179103 FIGURE 7. MICROWIRE Timing Diagram 14.1 ADVANCED MICROWIRE TIMING DIAGRAMS 14.1.1 3 Extra Clocks or Double Program Figure 8 shows the timing for the programming sequence for loading CLKoutX_DIV > 25 or CLKoutX_DDLY > 12 as described in Section 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY. 30179107 FIGURE 8. MICROWIRE Timing Diagram: Extra CLKuWire Pulses for R0 to R5 24 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 14.1.2 Three Extra Clocks with LEuWire High Figure 9 shows the timing for the programming sequence which allows SYNC_EN_AUTO = 1 when loading CLKoutX_DIV > 25 or CLKoutX_DDLY > 12. When SYNC_EN_AUTO = 1, a SYNC event is automatically generated on the falling edge of LEuWire. See Section 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY. 30179127 FIGURE 9. MICROWIRE Timing Diagram: Extra CLKuWire Pulses for R0 to R5 with LEuWire Asserted 14.1.3 Readback See Section 19.3 READBACK for more information on performing a readback operation. Figure 10 shows timing for LEuWire for both READBACK_LE = 1 and 0. The rising edges of CLKuWire during MICROWIRE readback continue to clock data on DATAuWire into the device during readback. If after the readback, LEuWire transitions from low to high, this data will be latched to the decoded register. The decoded register address consists of the last 5 bits clocked on DATAuWire as shown in the MICROWIRE Timing Diagrams. 30179106 FIGURE 10. MICROWIRE Readback Timing Diagram Copyright © 1999-2012, Texas Instruments Incorporated 25 LMK04906 15.0 Measurement Definitions 15.1 CHARGE PUMP CURRENT SPECIFICATION DEFINITIONS 30179131 I1 = Charge Pump Sink Current at VCPout = VCC - ΔV I2 = Charge Pump Sink Current at VCPout = VCC/2 I3 = Charge Pump Sink Current at VCPout = ΔV I4 = Charge Pump Source Current at VCPout = VCC - ΔV I5 = Charge Pump Source Current at VCPout = VCC/2 I6 = Charge Pump Source Current at VCPout = ΔV ΔV = Voltage offset from the positive and negative supply rails. Defined to be 0.5 V for this device. 15.1.1 Charge Pump Output Current Magnitude Variation Vs. Charge Pump Output Voltage 30179132 15.1.2 Charge Pump Sink Current Vs. Charge Pump Output Source Current Mismatch 30179133 15.1.3 Charge Pump Output Current Magnitude Variation Vs. Temperature 30179134 26 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 15.2 DIFFERENTIAL VOLTAGE MEASUREMENT TERMINOLOGY (Note 27) The differential voltage of a differential signal can be described by two different definitions causing confusion when reading datasheets or communicating with other engineers. This section will address the measurement and description of a differential signal so that the reader will be able to understand and discern between the two different definitions when used. The first definition used to describe a differential signal is the absolute value of the voltage potential between the inverting and noninverting signal. The symbol for this first measurement is typically VID or VOD depending on if an input or output voltage is being described. The second definition used to describe a differential signal is to measure the potential of the non-inverting signal with respect to the inverting signal. The symbol for this second measurement is VSS and is a calculated parameter. Nowhere in the IC does this signal exist with respect to ground, it only exists in reference to its differential pair. VSS can be measured directly by oscilloscopes with floating references, otherwise this value can be calculated as twice the value of VOD as described in the first description. Figure 11 illustrates the two different definitions side-by-side for inputs and Figure 12 illustrates the two different definitions sideby-side for outputs. The VID and VOD definitions show VA and VB DC levels that the non-inverting and inverting signals toggle between with respect to ground. VSS input and output definitions show that if the inverting signal is considered the voltage potential reference, the non-inverting signal voltage potential is now increasing and decreasing above and below the non-inverting reference. Thus the peak-to-peak voltage of the differential signal can be measured. VID and VOD are often defined as volts (V) and VSS is often defined as volts peak-to-peak (VPP). 30179175 FIGURE 11. Two Different Definitions for Differential Input Signals 30179174 FIGURE 12. Two Different Definitions for Differential Output Signals Note 27: Refer to application note AN-912 Common Data Transmission Parameters and their Definitions for more information. Copyright © 1999-2012, Texas Instruments Incorporated 27 LMK04906 16.0 Typical Performance Characteristics 16.2 CLOCK OUTPUT AC CHARACTERISTICS LVDS VOD vs. Frequency LVPECL /w 240 ohm emitter resistors VOD vs. Frequency 500 1200 450 400 VOD (mV) 350 VOD (mV) 2000 mVpp 1600 mVpp 1200 mVpp 700 mVpp 1000 300 250 200 800 600 400 150 100 200 50 0 0 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) 30179141 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) 30179142 LVPECL /w 120 ohm emitter resistors VOD vs. Frequency 1200 VOD (mV) 1000 2000 mVpp 800 600 1600 mVpp 400 200 0 0 500 1000 1500 2000 2500 3000 FREQUENCY (MHz) 30179143 28 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 17.0 Features 17.1 SYSTEM ARCHITECTURE The dual loop PLL architecture of the LMK04906 provides the lowest jitter performance over the widest range of output frequencies and phase noise integration bandwidths. The first stage PLL (PLL1) is driven by an external reference clock and uses an external VCXO or tunable crystal to provide a frequency accurate, low phase noise reference clock for the second stage frequency multiplication PLL (PLL2). PLL1 typically uses a narrow loop bandwidth (10 Hz to 200 Hz) to retain the frequency accuracy of the reference clock input signal while at the same time suppressing the higher offset frequency phase noise that the reference clock may have accumulated along its path or from other circuits. This “cleaned” reference clock provides the reference input to PLL2. The low phase noise reference provided to PLL2 allows PLL2 to operate with a wide loop bandwidth (50 kHz to 200 kHz). The loop bandwidth for PLL2 is chosen to take advantage of the superior high offset frequency phase noise profile of the internal VCO and the good low offset frequency phase noise of the reference VCXO or tunable crystal. Ultra low jitter is achieved by allowing the external VCXO or Crystal’s phase noise to dominate the final output phase noise at low offset frequencies and the internal VCO’s phase noise to dominate the final output phase noise at high offset frequencies. This results in best overall phase noise and jitter performance. The LMK04906 allows subsets of the device to be used to increase the flexibility of device. These different modes are selected using Section 19.9.1 MODE: Device Mode. For instance: • Dual Loop Mode - Typical use case of LMK04906. CLKinX used as reference input to PLL1, OSCin port is connected to VCXO or tunable crystal. • Single Loop Mode - Powers down PLL1. OSCin port is used as reference input. • Clock Distribution Mode - Allows input of CLKin1 to be distributed to output with division, digital delay, and analog delay. See Functional Description for more information on these modes. 17.2 PLL1 REDUNDANT REFERENCE INPUTS (CLKin0/CLKin0*, CLKin1/CLKin1*, and CLKin2/CLKin2*) The LMK04906 has three reference clock inputs for PLL1, CLKin0, CLKin1, and CLKin2. Ref Mux selects CLKin0, CLKin1, or CLKin2. Automatic or manual switching occurs between the inputs. CLKin0, CLKin1, and CLKin2 each have input dividers. The input divider allows different clock input frequencies to be normalized so that the frequency input to the PLL1 R divider remains constant during automatic switching. By programming these dividers such that the frequency presented to the input of the PLL1_R divider is the same prevents the user from needing to reprogram the PLL1 R divider when the input reference is changed to another CLKin port with a different frequency. CLKin1 is shared for use as an external 0-delay feedback (FBCLKin), or for use with an external VCO (Fin). Fast manual switching between reference clocks is possible with a external pins Status_CLKin0, Status_CLKin1, Status_CLKin2. 17.3 PLL1 TUNABLE CRYSTAL SUPPORT The LMK04906 integrates a crystal oscillator on PLL1 for use with an external crystal and varactor diode to perform jitter cleaning. The LMK04906 must be programmed to enable Crystal mode. 17.4 VCXO/CRYSTAL BUFFERED OUTPUTS The LMK04906 provides a dedicated output which is a buffered copy of the PLL2 reference input. This reference input is typically a low noise VCXO or Crystal. When using a VCXO, this output can be used to clock external devices such as microcontrollers, FPGAs, CPLDs, etc. before the LMK04906 is programmed. The OSCout0 buffer output type is programmable to LVDS, LVPECL, or LVCMOS. The dedicated output buffer OSCout0 can output frequency lower than the VCXO or Crystal frequency by programming the OSC Divider. The OSC Divider value range is 1 to 8. Each OSCoutX can individually choose to use the OSC Divider output or to bypass the OSC Divider. Two clock outputs can also be programmed to be driven by OSCin. This allows a total of 2 additional differential outputs to be buffered outputs of OSCin. When programmed in this way, a total of 3 differential outputs can be driven by a buffered copy of OSCin. VCXO/Crystal buffered outputs cannot be synchronized to the VCO clock distribution outputs. The assertion of SYNC will still cause these outputs to become low. Since these outputs will turn off and on asynchronously with respect to the VCO sourced clock outputs during a SYNC, it is possible for glitches to occur on the buffered clock outputs when SYNC is asserted and unasserted. If the NO_SYNC_CLKoutX bits are set these outputs will not be affected by the SYNC event except that the phase relationship will change with the other synchronized clocks unless a buffered clock output is used as a qualification clock during SYNC. 17.5 FREQUENCY HOLDOVER The LMK04906 supports holdover operation to keep the clock outputs on frequency with minimum drift when the reference is lost until a valid reference clock signal is re-established. 17.6 HITLESS SWITCHING The LMK04906 supports hitless switching. Copyright © 1999-2012, Texas Instruments Incorporated 29 LMK04906 17.7 INTEGRATED LOOP FILTER POLES The LMK04906 features programmable 3rd and 4th order loop filter poles for PLL2. These internal resistors and capacitor values may be selected from a fixed range of values to achieve either a 3rd or 4th order loop filter response. The integrated programmable resistors and capacitors compliment external components mounted near the chip. These integrated components can be effectively disabled by programming the integrated resistors and capacitors to their minimum values. 17.8 INTERNAL VCO The output of the internal VCO is routed to a mux which allows the user to select either the direct VCO output or a divided version of the VCO for the Clock Distribution Path. This same selection is also fed back to the PLL2 phase detector through a prescaler and N-divider. The mux selectable VCO divider has a divide range of 2 to 8 with 50% output duty cycle for both even and odd divide values. The primary use of the VCO divider is to achieve divides greater than the clock output divider supports alone. 17.9 EXTERNAL VCO MODE The Fin/Fin* input allows an external VCO to be used with PLL2 of the LMK04906. Using an external VCO reduces the number of available clock inputs by one. 17.10 CLOCK DISTRIBUTION The LMK04906 features a total of 6 outputs driven from the internal or external VCO. All VCO driven outputs have programmable output types. They can be programmed to LVPECL, LVDS, or LVCMOS. When all distribution outputs are configured for LVCMOS or single ended LVPECL a total of 24 outputs are available. If the buffered OSCin output OSCout0 is included in the total number of clock outputs the LMK04906 is able to distribute, then up to 6 differential clocks or up to 12 single ended clocks may be generated with the LMK04906. The following sections discuss specific features of the clock distribution channels that allow the user to control various aspects of the output clocks. 17.10.1 CLKout DIVIDER Each clock output has a single clock output divider. The divider supports a divide range of 1 to 1045 (even and odd) with 50% output duty cycle. When divides of 26 or greater are used, the divider/delay block uses extended mode. The VCO Divider may be used to reduce the divide needed by the clock output divider so that it may operate in normal mode instead of extended mode. This can result in a small current saving if enabling the VCO Divider allows 3 or more clock output divides to change from extended to normal mode. 17.10.2 CLKout DELAY The clock distribution section includes both a fine (analog) and coarse (digital) delay for phase adjustment of the clock outputs. The fine (analog) delay allows a nominal 25 ps step size and range from 0 to 475 ps of total delay. Enabling the analog delay adds a nominal 500 ps of delay in addition to the programmed value. When adjusting analog delay, glitches may occur on the clock outputs being adjusted. Analog delay may not operate at frequencies above the minimum-guaranteed maximum output frequency of 1536 MHz. The coarse (digital) delay allows a group of outputs to be delayed by 4.5 to 12 clock distribution path cycles in normal mode, or from 12.5 to 522 VCO cycles in extended mode. The delay step can be as small as half the period of the clock distribution path by using the CLKoutX_HS bit provided the output divide value is greater than 1. For example 2 GHz VCO frequency without using the VCO divider results in 250 ps coarse tuning steps. The coarse (digital) delay value takes effect on the clock outputs after a SYNC event. There are 3 different ways to use the digital (coarse) delay. 1. Fixed Digital Delay 2. Absolute Dynamic Digital Delay 3. Relative Dynamic Digital Delay These are further discussed in the Functional Description. 17.10.3 PROGRAMMABLE OUTPUT TYPE For increased flexibility all LMK04906 clock outputs (CLKoutX) and OSCout0 can be programmed to an LVDS, LVPECL, or LVCMOS output type. Any LVPECL output type can be programmed to 700, 1200, 1600, or 2000 mVpp amplitude levels. The 2000 mVpp LVPECL output type is a Texas Instruments proprietary configuration that produces a 2000 mVpp differential swing for compatibility with many data converters and is also known as 2VPECL. 17.10.4 CLOCK OUTPUT SYNCHRONIZATION Using the SYNC input causes all active clock outputs to share a rising edge. See Section 18.9.2 Clock Output Synchronization (SYNC) for more information. The SYNC event also causes the digital delay values to take effect. 30 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 17.11 0-DELAY The 0-delay mode synchronizes the input clock phase to the output clock phase. The 0-delay feedback may performed with an internal feedback loop from some of the clock outputs or with an external feedback loop into the FBCLKin port as selected by the FEEDBACK_MUX. Without using 0-delay mode there will be n possible fixed phase relationships from clock input to clock output depending on the clock output divide value. Using an external 0-delay feedback reduces the number of available clock inputs by one. 17.12 DEFAULT STARTUP CLOCKS Before the LMK04906 is programmed, CLKout4 is enabled and operating at a nominal frequency and CLKout3 and OSCout0 are enabled and operating at the OSCin frequency. These clocks can be used to clock external devices such as microcontrollers, FPGAs, CPLDs, etc. before the LMK04906 is programmed. For CLKout3 and OSCout0 to work before the LMK04906 is programmed the device must not be using Crystal mode. 17.13 STATUS PINS The LMK04906 provides status pins which can be monitored for feedback or in some cases used for input depending upon device programming. For example: • The Status_Holdover pin may indicate if the device is in hold-over mode. • The Status_CLKin0 pin may indicate the LOS (loss-of-signal) for CLKin0. • The Status_CLKin0 pin may be an input for selecting the active clock input. • The Status_LD pin may indicate if the device is locked. The status pins can be programmed to a variety of other outputs including analog lock detect, PLL divider outputs, combined PLL lock detect signals, PLL1 Vtune railing, readback, etc. Refer to the MICROWIRE programming section of this datasheet for more information. Default pin programming is captured in Table 16. 17.14 REGISTER READBACK Programmed registers may be read back using the MICROWIRE interface. For readback one of the status pins must be programmed for readback mode. At no time may registers be programed to values other than the valid states defined in the datasheet. Copyright © 1999-2012, Texas Instruments Incorporated 31 LMK04906 18.0 Functional Description 18.1 FUNCTIONAL OVERVIEW In default mode of operation, dual PLL mode with internal VCO, the Phase Frequency Detector in PLL1 compares the active CLKinX reference divided by CLKinX_PreR_DIV and PLL1 R divider with the external VCXO or crystal attached to the PLL2 OSCin port divided by PLL1 N divider. The external loop filter for PLL1 should be narrow to provide an ultra clean reference clock from the external VCXO or crystal to the OSCin/OSCin* pins for PLL2. The Phase Frequency Detector in PLL2 compares the external VCXO or crystal attached to the OCSin port divided by the PLL2 R divider with the output of the internal VCO divided by the PLL2 N divider and N2 pre-scaler and optionally the VCO divider. The bandwidth of the external loop filter for PLL2 should be designed to be wide enough to take advantage of the low in-band phase noise of PLL2 and the low high offset phase noise of the internal VCO. The VCO output is also placed on the distribution path for the clock distribution section. The clock distribution consists of 6 dividers and delays which drive 6 outputs. Each clock output allows the user to select a divide value, a digital delay value, and an analog delay. The 6 dividers drive programmable output buffers. Two outputs allow their input signal to be from the OSCin port directly. When a 0-delay mode is used, a clock output will be passed through the feedback mux to the PLL1 N Divider for synchronization and 0-delay. When an external VCO mode is used, the Fin port will be used to input an external VCO signal. PLL2 Phase comparison will now be with this signal divided by the PLL2 N divider and N2 pre-scaler. The VCO divider may not be used. One less clock input is available when using an external VCO mode. When a single PLL mode is used, PLL1 is powered down. OSCin is used as a reference to PLL2. 18.2 MODE SELECTION The LMK04906 family is capable of operating in several different modes as programmed by Section 19.9.1 MODE: Device Mode. TABLE 1. Device Mode Selection MODE R11[31:27] PLL1 PLL2 PLL2 VCO 0 X X Internal 2 X X Internal 3 X X External 5 X X External 6 X Internal 8 X Internal 11 X External 16 0-delay Clock Dist X X X X X X X X X X X In addition to selecting the device's mode of operation above, some modes require additional configuration. Also there are other features including holdover and dynamic digital delay that can also be enabled. 32 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 TABLE 2. Registers to Further Configure Device Mode of Operation Register Holdover 0-Delay Dynamic Digital Delay HOLDOVER_MODE 2 — — EN_TRACK User — — DAC_CLK_DIV User — — EN_MAN_DAC User — — DISABLE_DLD1_DET User — — EN_VTUNE_RAIL_ DET User — — DAC_HIGH_TRIP User — — DAC_LOW_TRIP User — — FORCE_HOLDOVER 0 — — SYNC_EN_AUTO — — User SYNC_QUAL — — 1 EN_SYNC — — 1 CLKout2_PD — — 0 EN_ FEEDBACK_MUX — 1 1 FEEDBACK_MUX — Feedback Clock Qualifying Clock NO_SYNC_ CLKoutX — — User 18.3 INPUTS / OUTPUTS 18.3.1 PLL1 Reference Inputs (CLKin0, CLKin1, and CLKin2) The reference clock inputs for PLL1 may be selected from either CLKin0, CLKin1, or CLKin2. The user has the capability to manually select one of the inputs or to configure an automatic switching mode of operation. See Section 18.4 INPUT CLOCK SWITCHING for more info. CLKin0, CLKin1, and CLKin2 have dividers which allow the device to switch between reference inputs of different frequencies automatically without needing to reprogram the PLL1 R divider. The CLKin pre-divider values are 1, 2, 4, and 8. CLKin1 input can alternatively be used for external feedback in 0-delay mode (FBCLKin) or for an external VCO input port (Fin). 18.3.2 PLL2 OSCin / OSCin* Port The feedback from the external oscillator being locked with PLL1 drives the OSCin/OSCin* pins. Internally this signal is routed to the PLL1 N Divider and to the reference input for PLL2. This input may be driven with either a single-ended or differential signal and must be AC coupled. If operated in single ended mode, the unused input must be connected to GND with a 0.1 µF capacitor. 18.3.3 CRYSTAL OSCILLATOR The internal circuitry of the OSCin port also supports the optional implementation of a crystal based oscillator circuit. A crystal, a varactor diode, and a small number of other external components may be used to implement the oscillator. The internal oscillator circuit is enabled by setting the EN_PLL2_XTAL bit. See Section 19.9.9 EN_PLL2_XTAL. 18.4 INPUT CLOCK SWITCHING Manual, pin select, and automatic are three different kinds clock input switching modes can be set with the CLKin_SELECT_MODE register. Below is information about how the active input clock is selected and what causes a switching event in the various clock input selection modes. 18.4.1 Input Clock Switching - Manual Mode When CLKin_SELECT_MODE is 0, 1, or 2 then CLKin0, CLKin1, or CLKin2 respectively is always selected as the active input clock. Manual mode will also override the EN_CLKinX bits such that the CLKinX buffer will operate even if CLKinX is is disabled with EN_CLKinX = 0. Entering Holdover If holdover mode is enabled then holdover mode is entered if: • Digital lock detect of PLL1 goes low and DISABLE_DLD1_DET = 0. Exiting Holdover The active clock for automatic exit of holdover mode is the manually selected clock input. Copyright © 1999-2012, Texas Instruments Incorporated 33 LMK04906 18.4.2 Input Clock Switching - Pin Select Mode When CLKin_SELECT_MODE is 3, the pins Status_CLKin0 and Status_CLKin1 select which clock input is active. Clock Switch Event: Pins Changing the state of Status_CLKin0 or Status_CLKin1 pins causes an input clock switch event. Clock Switch Event: PLL1 DLD To prevent PLL1 DLD high to low transition from causing a input clock switch event and causing the device to enter holdover mode, disable the PLL1 DLD detect by setting DISABLE_DLD1_DET = 1. This is the preferred behavior for Pin Select Mode. Configuring Pin Select Mode The Status_CLKin0_TYPE must be programmed to an input value for the Status_CLKin0 pin to function as an input for pin select mode. The Status_CLKin1_TYPE must be programmed to an input value for the Status_CLKin1 pin to function as an input for pin select mode. If the Status_CLKinX_TYPE is set as output, the input value is considered "0." The polarity of Status_CLKin1 and Status_CLKin0 input pins can be inverted with the CLKin_SEL_INV bit. Table 3 defines which input clock is active depending on Status_CLKin0 and Status_CLKin1 state. TABLE 3. Active Clock Input - Pin Select Mode Status_CLKin1 Status_CLKin0 Active Clock 0 0 CLKin0 0 1 CLKin1 1 0 CLKin2 1 1 Holdover The pin select mode will override the EN_CLKinX bits such that the CLKinX buffer will operate even if CLKinX is is disabled with EN_CLKinX = 0. To switch as fast as possible, keep the clock input buffers enabled (EN_CLKinX = 1) that could be switched to. Pin Select Mode and Host When in the pin select mode, the host can monitor conditions of the clocking system which could cause the host to switch the active clock input. The LMK04906 device can also provide indicators on the Status_LD and Status_HOLDOVER like "DAC Rail," "PLL1 DLD", "PLL1 & PLL2 DLD" which the host can use in determining which clock input to use as active clock input. Switch Event without Holdover When an input clock switch event is triggered and holdover mode is disabled, the active clock input immediately switches to the selected clock. When PLL1 is designed with a narrow loop bandwidth, the switching transient is minimized. Switch Event with Holdover When an input clock switch event is triggered and holdover mode is enabled, the device will enter holdover mode and remain in holdover until a holdover exit condition is met as described in Section 18.5 HOLDOVER MODE. Then the device will complete the reference switch to the pin selected clock input. 18.4.3 Input Clock Switching - Automatic Mode When CLKin_SELECT_MODE is 4, the active clock is selected in priority order of enabled clock inputs starting upon an input clock switch event. The priority order of the clocks is CLKin0 → CLKin1 → CLKin2, etc. For a clock input to be eligible to be switched through, it must be enabled using EN_CLKinX. Starting Active Clock Upon programming this mode, the currently active clock remains active if PLL1 lock detect is high. To ensure a particular clock input is the active clock when starting this mode, program CLKin_SELECT_MODE to the manual mode which selects the desired clock input (CLKin0, 1, or 2). Wait for PLL1 to lock PLL1_DLD = 1, then select this mode with CLKin_SELECT_MODE = 4. Clock Switch Event: PLL1 DLD A loss of lock as indicated by PLL1’s DLD signal (PLL1_DLD = 0) will cause an input clock switch event if DISABLE_DLD1_DET = 0. PLL1 DLD must go high (PLL1_DLD = 1) in between input clock switching events. Clock Switch Event: PLL1 Vtune Rail If Vtune_RAIL_DET_EN is set and the PLL1 Vtune voltage crosses the DAC high or low threshold, holdover mode will be entered. Since PLL1_DLD = 0 in holdover a clock input switching event will occur. Clock Switch Event with Holdover If holdover is enabled and an input clock switch event occurs, holdover mode is entered and the active clock is set to the next enabled clock input in priority order. When the new active clock meets the holdover exit conditions, holdover is exited and the active clock will continue to be used as a reference until another PLL1 loss of lock event. PLL1 DLD must go high in between input clock switching events. Clock Switch Event without Holdover If holdover is not enabled and an input clock switch event occurs, the active clock is set to the next enabled clock in priority order. The LMK04906 will keep this new input clock as the active clock until another input clock switching event. PLL1 DLD must go high in between input clock switching events. 34 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 18.4.4 Input Clock Switching - Automatic Mode with Pin Select When CLKin_SELECT_MODE is 6, the active clock is selected using the Status_CLKinX pins upon an input clock switch event according to Table 4. Starting Active Clock Upon programming this mode, the currently active clock remains active if PLL1 lock detect is high. To ensure a particular clock input is the active clock when starting this mode, program CLKin_SELECT_MODE to the manual mode which selects the desired clock input (CLKin0 or 1). Wait for PLL1 to lock PLL1_DLD = 1, then select this mode with CLKin_SELECT_MODE = 6. Clock Switch Event: PLL1 DLD An input clock switch event is generated by a loss of lock as indicated by PLL1's DLD signal (PLL1 DLD = 0). Clock Switch Event: PLL1 Vtune Rail If Vtune_RAIL_DET_EN is set and the PLL1 Vtune voltage crosses the DAC threshold, holdover mode will be entered. Since PLL1_DLD = 0 in holdover, a clock input switching event will occur. Clock Switch Event with Holdover If holdover is enabled and an input clock switch event occurs, holdover mode is entered and the active clock is set to the clock input defined by the Status_CLKinX pins. When the new active clock meets the holdover exit conditions, holdover is exited and the active clock will continue to be used as a reference until another input clock switch event. PLL1 DLD must go high in between input clock switching events. Clock Switch Event without Holdover If holdover is not enabled and an input clock switch event occurs, the active clock is set to the clock input defined by the Status_CLKinX pins. The LMK04906 will keep this new input clock as the active clock until another input clock switching event. PLL1 DLD must go high in between input clock switching events. TABLE 4. Active Clock Input - Auto Pin Mode Status_CLKin1 Status_CLKin0 Active Clock X 1 CLKin0 1 0 CLKin1 0 0 CLKin2 The polarity of Status_CLKin1 and Status_CLKin0 input pins can be inverted with the CLKin_SEL_INV bit. Copyright © 1999-2012, Texas Instruments Incorporated 35 LMK04906 18.5 HOLDOVER MODE Holdover mode causes PLL2 to stay locked on frequency with minimal frequency drift when an input clock reference to PLL1 becomes invalid. While in holdover mode, the PLL1 charge pump is TRI-STATED and a fixed tuning voltage is set on CPout1 to operate PLL1 in open loop. Enable holdover Program Section 19.10.5 HOLDOVER_MODE to enable holdover mode. Holdover mode can be manually enabled by programming the FORCE_HOLDOVER bit. The holdover mode can be set to operate in 2 different sub-modes. • Fixed CPout1 (EN_TRACK = 0 or 1, EN_MAN_DAC = 1). • Tracked CPout1 (EN_TRACK = 1, EN_MAN_DAC = 0). — Not valid when EN_VTUNE_RAIL_DET = 1. Updates to the DAC value for the Tracked CPout1 sub-mode occurs at the rate of the PLL1 phase detector frequency divided by DAC_CLK_DIV. These updates occur any time EN_TRACK = 1. The DAC update rate should be programmed for <= 100 kHz to ensure DAC holdover accuracy. When tracking is enabled the current voltage of DAC can be readback, see Section 19.15.1 DAC_CNT. Entering holdover The holdover mode is entered as described in Section 18.4 INPUT CLOCK SWITCHING. Typically this is because: • FORCE_HOLDOVER bit is set. • PLL1 loses lock according to PLL1_DLD, and — HOLDOVER_MODE = 2 — DISABLE_DLD1_DET = 0 • CPout1 voltage crosses DAC high or low threshold, and — HOLDOVER_MODE = 2 — EN_VTUNE_RAIL_DET = 1 — EN_TRACK = 1 — DAC_HIGH_TRIP = User Value — DAC_LOW_TRIP = User Value — EN_MAN_DAC = 1 — MAN_DAC = User Value During holdover PLL1 is run in open loop mode. • PLL1 charge pump is set to TRI-STATE. • PLL1 DLD will be unasserted. • The HOLDOVER status is asserted • During holdover If PLL2 was locked prior to entry of holdover mode, PLL2 DLD will continue to be asserted. • CPout1 voltage will be set to: — a voltage set in the MAN_DAC register (fixed CPout1). — a voltage determined to be the last valid CPout1 voltage (tracked CPout1). • PLL1 DLD will attempt to lock with the active clock input. The HOLDOVER status signal can be monitored on the Status_HOLDOVER or Status_LD pin by programming the HOLDOVER_MUX or LD_MUX register to "Holdover Status." Exiting holdover Holdover mode can be exited in one of two ways. • Manually, by programming the device from the host. • Automatically, By a clock operating within a specified ppm of the current PLL1 frequency on the active clock input. See Section 18.4 INPUT CLOCK SWITCHING for more detail on which clock input is active. To exit holdover by programming, set HOLDOVER_MODE = Disabled. HOLDOVER_MODE can then be re-enabled by programming HOLDOVER_MODE = Enabled. Care should be taken to ensure that the active clock upon exiting holdover is as expected, otherwise the CLKin_SELECT_MODE register may need to be re-programmed. 36 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 18.5.1 Holdover Frequency Accuracy and DAC Performance When in holdover mode PLL1 will run in open loop and the DAC will set the CPout1 voltage. If Fixed CPout1 mode is used, then the output of the DAC will be a voltage dependant upon the MAN_DAC register. If Tracked CPout1 mode is used, then the output of the DAC will be the voltage at the CPout1 pin before holdover mode was entered. When using Tracked mode and EN_MAN_DAC = 1, during holdover the DAC value is loaded with the programmed value in MAN_DAC, not the tracked value. When in Tracked CPout1 mode the DAC has a worst case tracking error of ±2 LSBs once PLL1 tuning voltage is acquired. The step size is approximately 3.2 mV, therefore the VCXO frequency error during holdover mode caused by the DAC tracking accuracy is ±6.4 mV * Kv. Where Kv is the tuning sensitivity of the VCXO in use. Therefore the accuracy of the system when in holdover mode in ppm is: 30179159 Example: consider a system with a 19.2 MHz clock input, a 153.6 MHz VCXO with a Kv of 17 kHz/V. The accuracy of the system in holdover in ppm is: ±0.71 ppm = ±6.4 mV * 17 kHz/V * 1e6 / 153.6 MHz It is important to account for this frequency error when determining the allowable frequency error window to cause holdover mode to exit. 18.5.2 Holdover Mode - Automatic Exit of Holdover The LMK04906 device can be programmed to automatically exit holdover mode when the accuracy of the frequency on the active clock input achieves a specified accuracy. The programmable variables include PLL1_WND_SIZE and DLD_HOLD_CNT. See Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY to calculate the register values to cause holdover to automatically exit upon reference signal recovery to within a user specified ppm error of the holdover frequency. It is possible for the time to exit holdover to vary because the condition for automatic holdover exit is for the reference and feedback signals to have a time/phase error less than a programmable value. Because it is possible for two clock signals to be very close in frequency but not close in phase, it may take a long time for the phases of the clocks to align themselves within the allowable time/ phase error before holdover exits. 18.6 PLLs 18.6.1 PLL1 PLL1's maximum phase detector frequency (fPD1) is 40 MHz. Since a narrow loop bandwidth should be used for PLL1, the need to operate at high phase detector rate to lower the in-band phase noise becomes unnecessary. The maximum values for the PLL1 R and N dividers is 16,383. Charge pump current ranges from 100 to 1600 µA. PLL1 N divider may be driven by OSCin port at the OSCout0_MUX output (default) or by internal or external feedback as selected by Feedback Mux in 0-delay mode. Low charge pump currents and phase detector frequencies aid design of low loop bandwidth loop filters with reasonably sized components to allow the VCXO or PLL2 to dominate phase noise inside of PLL2 loop bandwidth. High charge pump currents may be used by PLL1 when using VCXOs with leaky tuning voltage inputs to improve system performance. 18.6.2 PLL2 PLL2's maximum phase detector frequency (fPD2) is 155 MHz. Operating at highest possible phase detector rate will ensure low in-band phase noise for PLL2 which in turn produces lower total jitter. The in-band phase noise from the reference input and PLL is proportional to N2. The maximum value for the PLL2 R divider is 4,095. The maximum value for the PLL2 N divider is 262,143. The N2 Prescaler in the total N feedback path can be programmed for values 2 to 8 (all divides even and odd). Charge pump current ranges from 100 to 3200 µA. High charge pump currents help to widen the PLL2 loop bandwidth to optimize PLL2 performance. 18.6.2.1 PLL2 FREQUENCY DOUBLER The PLL2 reference input at the OSCin port may be routed through a frequency doubler before the PLL2 R Divider. The frequency doubler feature allows the phase comparison frequency to be increased when a relatively low frequency oscillator is driving the OSCin port. By doubling the PLL2 phase detector frequency, the in-band PLL2 noise is reduced by about 3 dB. For applications in which the OSCin frequency and PLL2 phase detector frequency are equal, the best PLL2 in-band noise can be achieved when the doubler is enabled (EN_PLL2_REF_2X = 1) and the PLL2 R divide value is 2. Do not use doubler disabled (EN_PLL2_REF_2X = 0) and PLL2 R divide value of 1. When using the doubler take care to use the PLL2 R Divider to reduce the phase detector frequency to the limit of the PLL2 maximum phase detector frequency. Copyright © 1999-2012, Texas Instruments Incorporated 37 LMK04906 18.6.3 DIGITAL LOCK DETECT Both PLL1 and PLL2 support digital lock detect. Digital lock detect compares the phase between the reference path (R) and the feedback path (N) of the PLL. When the time error, which is phase error, between the two signals is less than a specified window size (ε) a lock detect count increments. When the lock detect count reaches a user specified value lock detect is asserted true. Once digital lock detect is true, a single phase comparison outside the specified window will cause digital lock detect to be asserted false. This is illustrated in Figure 13. The incremental lock detect count feature functions as a digital filter to ensure that lock detect isn't asserted for only a brief time when the phases of R and N are within the specified tolerance for only a brief time during initial phase lock. The digital lock detect signal can be monitored on the Status_LD or Status_Holdover pin. The pin may be programmed to output the status of lock detect for PLL1, PLL2, or both PLL1 and PLL2. See Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY for more detailed information on programming the registers to achieve a specified frequency accuracy in ppm with lock detect. The digital lock detect feature can also be used with holdover to automatically exit holdover mode. See Section 18.5 HOLDOVER MODE for more info. 30179128 FIGURE 13. Digital Lock Detect Flowchart 18.7 STATUS PINS The Status_LD, Status_HOLDOVER, Status_CLKin0, Status_CLKin1, and SYNC/Status_CLKin2 pins can be programmed to output a variety of signals for indicating various statuses like digital lock detect, holdover, several DAC indicators, and several PLL divider outputs. 18.7.1 Logic Low This is a vary simple output. In combination with the output _MUX register, this output can be toggled between high and low. Useful to confirm MICROWIRE programming or as a general purpose IO. 38 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 18.7.2 Digital Lock Detect PLL1 DLD, PLL2 DLD, and PLL1 + PLL2 are selectable on certain output pins. See Section 18.6.3 DIGITAL LOCK DETECT for more information. 18.7.3 Holdover Status Indicates if the device is in Holdover mode. See Section 18.5 HOLDOVER MODE for more information. 18.7.4 DAC Various flags for the DAC can be monitored including DAC Locked, DAC Rail, DAC Low, and DAC High. When the PLL1 tuning voltage crosses the low threshold, DAC Low is asserted. When PLL1 tuning voltage crosses the high threshold, DAC High is asserted. When either DAC Low or DAC High is asserted, DAC Rail will also be asserted. DAC Locked is asserted when EN_Track = 1 and DAC is closely tracking the PLL1 tuning voltage. 18.7.5 PLL Divider Outputs The PLL divider outputs are useful for debugging failure to lock issues. It allows the user to measure the frequency the PLL inputs are receiving. The settings of PLL1_R, PLL1_N, PLL2_R, and PLL2_N output pulses at the phase detector rate. The settings of PLL1_R / 2, PLL1_N / 2, PLL2_R / 2, and PLL2_N / 2 output a 50% duty cycle waveform at half the phase detector rate. 18.7.6 CLKinX_LOS The clock input loss of signal indicator is asserted when LOS is enabled (Section 19.12.2 EN_LOS) and the clock no longer detects an input as defined by the time-out threshold, Section 19.12.1 LOS_TIMEOUT. 18.7.7 CLKinX Selected If this clock is the currently selected/active clock, this pin will be asserted. 18.7.8 MICROWIRE Readback The readback data can be output on any pin programmable to readback mode. For more information on readback see Section 19.3 READBACK. 18.8 VCO The integrated VCO uses a frequency calibration routine when register R30 is programmed to lock VCO to target frequency. Register R30 contains the PLL2_N register. During the frequency calibration the PLL2_N_CAL value is used instead of PLL2_N, this allows 0-delay modes to have a separate PLL2 N value for VCO frequency calibration and regular operation. 18.9 CLOCK DISTRIBUTION 18.9.1 Fixed Digital Delay This section discussing Fixed Digital delay and associated registers is fundamental to understanding digital delay and dynamic digital delay. Clock outputs may be delayed or advanced from one another by up to 517.5 clock distribution path periods. By programming a digital delay value from 4.5 to 522 clock distribution path periods, a relative clock output delay from 0 to 517.5 periods is achieved. The CLKoutX_DDLY (5 to 522) and CLKoutX_HS (-0.5 or 0) registers set the digital delay as shown in Table 5. TABLE 5. Possible Digital Delay Values CLKoutX_DDLY CLKoutX_HS Digital Delay 5 1 4.5 5 0 5 6 1 5.5 6 0 6 7 1 6.5 7 0 7 ... ... ... 520 0 520 521 1 520.5 521 0 521 522 1 521.5 522 0 522 Note: Digital delay values only take effect during a SYNC event and if the NO_SYNC_CLKoutX bit is cleared for this clock output. See Section 18.9.2 Clock Output Synchronization (SYNC) for more information. The resolution of digital delay is determined by the frequency of the clock distribution path. The clock distribution path is the output of Mode Mux1 (Figure 6). The best resolution of digital delay is achieved by bypassing the VCO divider. Copyright © 1999-2012, Texas Instruments Incorporated 39 LMK04906 (1) (2) The digital delay between clock outputs can be dynamically adjusted with no or minimum disruption of the output clocks. See Section 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY for more information. 18.9.1.1 FIXED DIGITAL DELAY - EXAMPLE Given a VCO frequency of 2457.6 MHz and no VCO divider, by using digital delay the outputs can be adjusted in 1 / (2 * 2457.6 MHz) = ~203.5 ps steps. To achieve quadrature (90 degree shift) between the 122.88 MHz outputs on CLKout4 and CLKout3 from a VCO frequency of 2457.6 MHz and bypassing the VCO divider, consider the following: 1. The frequency of 122.88 MHz has a period of ~8.14 ns. 2. To delay 90 degrees of a 122.88 MHz clock period requires a ~2.03 ns delay. 3. Given a digital delay step of ~203.5 ps, this requires a digital delay value of 12 steps (2.03 ns / 20.35 ps = 10). 4. Since the 10 steps are half period steps, CLKout3_DDLY is programmed 5 full periods beyond 5 for a total of 10. This result in the following programming: • Clock output dividers to 20. CLKout2_DIV = 20 and CLKout3_DIV = 20. • Set first clock digital delay value. CLKout2_DDLY = 5, CLKout2_HS = 0. • Set second 90 degree shifted clock digital delay value. CLKout3_DDLY = 10, CLKout3_HS = 0. Table 6 shows some of the possible phase delays in degrees achievable in the above example. TABLE 6. Relative phase shift from CLKout2 to CLKout3 CLKout3_DDLY = 5 and CLKout3_HS = 0 CLKout3_DDLY CLKout_HS Relative Digital Delay Degrees of 122.88 MHz 5 1 -0.5 -9° 5 0 0.0 0° 6 1 0.5 9° 6 0 1.0 18° 7 1 1.5 27° 7 0 2.0 36° 8 1 2.5 45° 8 0 3.0 54° 9 1 3.5 63° 9 0 4.0 72° 10 1 4.5 81° 10 0 5.0 90° 11 1 5.5 99° 11 0 6.0 108° 12 1 6.5 117° 12 0 7.0 126° 13 1 7.5 135° 13 0 8.0 144° 14 1 8.5 153° ... ... ... ... Figure 15 illustrates clock outputs programmed with different digital delay values during a SYNC event. Refer to Section 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY for more information on dynamically adjusting digital delay. 18.9.2 Clock Output Synchronization (SYNC) The purpose of the SYNC function is to synchronize the clock outputs with a fixed and known phase relationship between each clock output selected for SYNC. SYNC can also be used to hold the outputs in a low or 0 state. The NO_SYNC_CLKoutX bits can be set to disable synchronization for a clock output. To enable SYNC, EN_SYNC must be set. See Section 19.9.2 EN_SYNC, Enable Synchronization. 40 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 The digital delay value set by CLKoutX_DDLY takes effect only upon a SYNC event. The digital delay due to CLKoutX_HS takes effect immediately upon programming. See Section 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY for more information on dynamically changing digital delay. During a SYNC event, clock outputs driven by the VCO are not synchronized to clock outputs driven by OSCin. OSCout0 is always driven by OSCin. CLKout3 or 4 may be driven by OSCin depending on the CLKoutX_OSCin_Sel bit value. While SYNC is asserted, NO_SYNC_CLKoutX operates normally for CLKout3 and 4 under all circumstances. SYNC operates normally for CLKout3 and 4 when driven by VCO. Effect of SYNC When SYNC is asserted, the outputs to be synchronized are held in a logic low state. When SYNC is unasserted, the clock outputs to be synchronized are activated and will transition to a high state simultaneously with one another except where different digital delay values have been programmed. Refer to Section 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY for SYNC functionality when SYNC_QUAL = 1. TABLE 7. Steady State Clock Output Condition Given Specified Inputs SYNC_TYPE SYNC_POL _INV SYNC Pin Clock Output State 0,1,2 (Input) 0 0 Active 0,1,2 (Input) 0 1 Low 0,1,2 (Input) 1 0 Low 0,1,2 (Input) 1 1 Active 3, 4, 5, 6 (Output) 0 0 or 1 Active 3, 4, 5, 6 (Output) 1 0 or 1 Low Methods of Generating SYNC There are five methods to generate a SYNC event: • Manual: — Asserting the SYNC pin according to the polarity set by SYNC_POL_INV. — Toggling the SYNC_POL_INV bit though MICROWIRE will cause a SYNC to be asserted. • Automatic: — If PLL1_SYNC_DLD or PLL2_SYNC_DLD is set, the SYNC pin will be asserted while DLD (digital lock detect) is false for PLL1 or PLL2 respectively. — Programming Register R30, which contains PLL2_N will generate a SYNC event when using the internal VCO. — Programming Register R0 through R5 when SYNC_EN_AUTO = 1. Note: Due to the speed of the clock distribution path (as fast as ~325 ps period) and the slow slew rate of the SYNC, the exact VCO cycle at which the SYNC is asserted or unasserted by the SYNC is undefined. The timing diagrams show a sharp transition of the SYNC to clarify functionality. Avoiding clock output interruption due to SYNC Any CLKout outputs that have their NO_SYNC_CLKoutX bits set will be unaffected by the SYNC event. It is possible to perform a SYNC operation with the NO_SYNC_CLKoutX bits cleared, then set the NO_SYNC_CLKoutX bits so that the selected clocks will not be affected by a future SYNC. Future SYNC events will not effect these clocks but will still cause the newly synchronized clocks to be re-synchronized using the currently programmed digital delay values. When this happens, the phase relationship between the first group of synchronized clocks and the second group of synchronized clocks will be undefined unless the SYNC pulse is qualified by an output clock. See Section 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY. SYNC Timing When discussing the timing of the SYNC function, one cycle refers to one period of the clock distribution path. Copyright © 1999-2012, Texas Instruments Incorporated 41 LMK04906 30179104 FIGURE 14. Clock Output synchronization using the SYNC pin (Active Low) CLKout0_DIV = 0(valid only for external VCO mode) CLKout2_DIV = 2 CLKout4_DIV = 4 The digital delay for all clock outputs is 5 The digital delay half step for all clock outputs is 0 SYNC_QUAL = 0 (No qualification) Refer to Figure 14 during this discussion on the timing of SYNC. SYNC must be asserted for greater than one clock cycle of the clock distribution path to latch the SYNC event. After SYNC is asserted, the SYNC event is latched on the rising edge of the distribution path clock, at time A. After this event has been latched, the outputs will not reflect the low state for 6 cycles, at time B. Due to the asynchronous nature of SYNC with respect to the output clocks, it is possible that a glitch pulse could be created when the clock output goes low from the SYNC event. This is shown by CLKout4 in Figure 14 and CLKout2 in Figure 15. See Section 18.9.2.1.2 Relative Dynamic Digital Delay for more information on synchronizing relative to an output clock to eliminate or minimize this glitch pulse. After SYNC becomes unasserted the event is latched on the following rising edge of the distribution path clock, time C. The clock outputs will rise at time D, coincident with a rising distribution clock edge that occurs after 6 cycles plus as many more cycles as programmed by the digital delay for that clock output. Therefore, the soonest a clock output will become high is 11 cycles after the SYNC unassertion event registration, time C, when the smallest digital delay value of 5 is set. If CLKoutX_HS = 1 and CLKoutX_DDLY = 5, then the clock output will rise 10.5 cycles after SYNC is unassertion event registration. 42 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 30179105 FIGURE 15. Clock Output synchronization using the SYNC pin (Active Low) CLKout0_DIV = 2, CLKout0_DDLY = 5 CLKout2_DIV = 4, CLKout2_DDLY = 7 CLKout4_DIV = 4, CLKout4_DDLY = 8 CLKout5_DIV = 4, CLKour4_DDLY = 8 CLKout0_HS = 1 CLKout2_HS = 0 CLKout4_HS = 0 CLKout5_HS = 0 SYNC_QUAL = 0 (No qualification) Figure 15 illustrates the timing with different digital delays programmed. • Time A) SYNC assertion event is latched. • Time B) SYNC unassertion latched. • Time C) All outputs toggle and remain low. A glitch pulse can occur at this time as shown by CLKout2. • Time D) After 6 + 4.5 = 10.5 cycles CLKout0 rises. This is the shortest time from SYNC unassertion registration to clock rising edge possible. • Time E) After 6 + 7 = 13 cycles CLKout2 rises. CLKout2 and CLKout4, 5 are programmed for quadrature operation. • Time F) After 6 + 8 = 14 cycles CLKout4 and 5 rise. Copyright © 1999-2012, Texas Instruments Incorporated 43 LMK04906 18.9.2.1 DYNAMICALLY PROGRAMMING DIGITAL DELAY To use dynamic digital delay synchronization qualification set SYNC_QUAL = 1. This causes the SYNC pulse to be qualified by a clock output so that the SYNC event occurs after a specified time from a clock output transition. This allows the relative adjustment of clock output phase in real-time with no or minimum interruption of clock outputs. Hence the term dynamic digital delay. Note that changing the phase of a clock output requires momentarily altering in the rate of change of the clock output phase and therefore by definition results in a frequency distortion of the signal. Without qualifying the SYNC with an output clock, the newly synchronized clocks would have a random and unknown digital delay (or phase) with respect to clock outputs not currently being synchronized. Absolute vs. Relative Dynamic Digital Delay The clock used for qualification of SYNC is selected with the feedback mux (FEEDBACK_MUX). If the clock selected by the feedback mux has its NO_SYNC_CLKoutX = 1, then an absolute dynamic digital delay adjustment will be performed during a SYNC event and the digital delay of the feedback clock will not be adjusted. If the clock selected by the feedback mux has its NO_SYNC_CLKoutX = 0, then a self-referenced or relative dynamic digital delay adjustment will be performed during a SYNC event and the digital delay of the feedback clock will be adjusted. Clocks with NO_SYNC_CLKoutX = 1 always operate without interruption. Dynamic Digital Delay and 0-Delay Mode When using a 0-delay mode absolute dynamic digital delay is recommended. Using relative dynamic digital delay with a 0-delay mode may result in a momentary clock loss on the adjusted clock also being used for 0-delay feedback that may result in PLL1 DLD becoming low. This may result in HOLDOVER mode being activated depending upon device configuration. SYNC and Minimum Step Size The minimum step size adjustment for digital delay is half a clock distribution path cycle. This is achieved by using the CLKoutX_HS bit. The CLKoutX_HS bit change effect is immediate without the need for SYNC. To shift digital delay using CLKoutX_DDLY a SYNC signal must be generated for the change to take effect. Programming Overview To dynamically adjust the digital delay with respect to an existing clock output the device should be programmed as follows: • Set SYNC_QUAL = 1 for clock output qualification. • Set CLKout2_PD = 0. Required for proper operation of SYNC_QUAL = 1. • Set EN_FEEDBACK_MUX = 1 to enable the feedback buffer. • Set FEEDBACK_MUX to the clock output that the newly synchronized clocks will be qualified by. • Set NO_SYNC_CLKoutX = 1 for the output clocks that will continue to operate during the SYNC event. There is no interruption of output on these clocks. — If FEEDBACK_MUX selects a clock output with NO_SYNC_CLKoutX = 1, then absolute dynamic digital delay is performed. — If FEEDBACK_MUX selects a clock output with NO_SYNC_CLKoutX = 0, then self-referenced or relative dynamic digital delay is performed. • The SYNC_EN_AUTO bit may be set to cause a SYNC event to begin when register R0 to R5 is programmed. The auto SYNC feature is a convenience since does not require the application to manually assert SYNC by toggling the SYNC_POL_INV bit or the SYNC pin when changing digital delay. However, under the following condition a special programming sequence is required if SYNC_EN_AUTO = 1: — The CLKoutX_DDLY value being set in the programmed register is 13 or more. • Under the following condition a SYNC_EN_AUTO must = 0: — If the application requires a digital delay resolution of half a clock distribution path cycle in relative dynamic digital delay mode because the HS bit must be fixed per Table 8 for a qualifying clock. Internal Dynamic Digital Delay Timing To dynamically adjust digital delay a SYNC must occur. Once the SYNC is qualified by an output clock, 3 cycles later an internal one shot pulse will occur. The width of the one shot pulse is 3 cycles. This internal one shot pulse will cause the outputs to turn off and then back on with a fixed delay with respect to the falling edge of the qualification clock. This allows for dynamic adjustments of digital delay with respect to an output clock. The qualified SYNC timing is shown in Figure 16 for absolute dynamic digital delay and Figure 17 for relative dynamic digital delay. 44 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Other Timing Requirements When adjusting digital delay dynamically, the falling edge of the qualifying clock selected by the FEEDBACK_MUX must coincide with the falling edge of the clock distribution path. For this requirement to be met, program the CLKoutX_HS value of the qualifying clock output according to Table 8. TABLE 8. Half Step programming requirement of qualifying clock during SYNC event Distribution Path Frequency CLKoutX_DIV value CLKoutX_HS ≥ 1.8 GHz Even Must = 1 during SYNC event. < 1.8 GHz Odd Must = 0 during SYNC event. Even Must = 0 during SYNC event. Odd Must = 1 during SYNC event. 18.9.2.1.1 Absolute Dynamic Digital Delay Absolute dynamic digital delay can be used to program a clock output to a specific phase offset from another clock output. Pros: • Simple direct phase adjustment with respect to another clock output. • CLKoutX_HS will remain constant for qualifying clock. — Can easily use auto sync feature (SYNC_EN_AUTO = 1) when digital delay adjustment requires half step digital delay requirements. • Can be used with 0-delay mode. Cons: • For some phase adjustments there may be a glitch pulse due to SYNC assertion. — For example see CLKout4 in Figure 14 and CLKout2 in Figure 15. 18.9.2.1.1.1 ABSOLUTE DYNAMIC DIGITAL DELAY - EXAMPLE To illustrate the absolute dynamic digital delay adjust procedure, consider the following example. System Requirements: • VCO Frequency = 2457.6 MHz • CLKout0 = 819.2 MHz (CLKout0_DIV = 3) • CLKout2 = 307.2 MHz (CLKout2_DIV = 8) • CLKout4 = 245.76 MHz (CLKout4_DIV = 10) • For all clock outputs during initial programming: — CLKoutX_DDLY = 5 — CLKoutX_HS = 1 — NO_SYNC_CLKoutX = 0 The application requires the 307.2 MHz clock to be stepped in 22.5 degree steps (~203.4 ps), which is the minimum step resolution allowable by the clock distribution path requiring use of the half step bit (CLKoutX_HS). That is 1 / 2457.6 MHz / 2 = ~203.4 ps. During the stepping of the 307.2 MHz clock the 819.2 MHz and 245.76 MHz clock must not be interrupted. Step 1: The device is programmed from register R0 to R30 with values that result in the device being locked and operating as desired, see the system requirements above. The phase of all the output clocks are aligned because all the digital delay and half step values were the same when the SYNC was generated by programming register R30. The timing of this is as shown in Figure 14. Step 2: Now the registers will be programmed to prepare for changing digital delay (or phase) dynamically. TABLE 9. Register Setup for Absolute Dynamic Digital Delay Example Register Purpose SYNC_QUAL = 1 Use a clock output for qualifying the SYNC pulse for dynamically adjusting digital delay. EN_SYNC = 1 (default) Required for SYNC functionality. CLKout3_PD = 0 Required when SYNC_QUAL = 1. CLKout3 outputs may be powered down or in use. EN_FEEDBACK_MUX = 1 Enable the feedback mux for SYNC operation for dynamically adjusting digital delay. FEEDBACK_MUX = 2 (CLKout4) Use the fixed 245.76 MHz clock as the SYNC qualification clock. NO_SYNC_CLKout0 = 1 This clock output (819.2 MHz) won't be affected by SYNC. It will always operate without interruption. Copyright © 1999-2012, Texas Instruments Incorporated 45 LMK04906 Register Purpose NO_SYNC_CLKout4 = 1 This clock output (245.76 MHz) won't be affected by SYNC. It will always operate without interruption. This clock will also be the qualifying clock in this example. CLKout4_HS = 1 Since CLKout4 is the qualifying clock and CLKoutX_DIV is even, the half step bit must be set to 1. See Table 8. SYNC_EN_AUTO = 1 Automatic generation of SYNC is allowed for this case. After the registers in Table 9 have been programmed, the application may now dynamically adjust the digital delay of CLKout2 (307.2 MHz). Step 3: Adjust digital delay of CLKout2. Refer to Table 10 for the programming values to set a specified phase offset from the absolute reference clock. Table 10 is dependant upon the qualifying clock divide value of 12, refer to Section 20.7 CALCULATING DYNAMIC DIGITAL DELAY VALUES FOR ANY DIVIDE for information on creating tables for any divide value. TABLE 10. Programming for Absolute Digital Delay Adjustment Degrees of Adjustment from initial 307.2MHz phase +/-0 or +/-360 degrees 22.5 degrees Programming CLKout2_DDLY = 14; CLKout2_HS = 1 -337.5 degrees CLKout2_DDLY = 14; CLKout2_HS = 0 45 degrees -315 degrees CLKout2_DDLY = 15; CLKout2_HS = 1 67.5 degrees -292.54 degrees CLKout2_DDLY = 5; CLKout2_HS = 0 90 degrees -270 degrees CLKout2_DDLY = 6; CLKout2_HS = 1 112.5 degrees -247.5 degrees CLKout2_DDLY = 6; CLKout2_HS = 0 135 degrees -25 degrees CLKout2_DDLY = 7; CLKout2_HS = 1 157.5 degrees -202.5 degrees CLKout2_DDLY = 7; CLKout2_HS = 0 180 degrees -180 degrees CLKout2_DDLY = 8; CLKout2_HS = 1 247.5 degrees -112.5 degrees CLKout2_DDLY = 8; CLKout2_HS = 0 270 degrees -90 degrees CLKout2_DDLY = 9; CLKout2_HS = 1 292.5 degrees -67.5 degrees CLKout2_DDLY = 9; CLKout2_HS = 0 315 degrees -45 degrees CLKout2_DDLY = 10; CLKout2_HS = 1 337.5 degrees -22.5 degrees CLKout2_DDLY = 10; CLKout2_HS = 0 After setting the new digital delay values, the act of programming R1 will start a SYNC automatically because SYNC_EN_AUTO = 1. If the user elects to reduce the number of SYNCs because they are not required when only CLKout2_HS is set, then SYNC_EN_AUTO is = 0 and the SYNC may now be generated by toggling the SYNC pin or by toggling the SYNC_POL_INV bit. Because of the internal one shot pulse, no strict timing of the SYNC pin or SYNC_POL_INV bit is required. After the SYNC event, the clock output will adjust according to Table 10. See Figure 16 for a detailed view of the timing diagram. The timing diagram critical points are: • Time A) SYNC assertion event is latched. • Time B) First qualifying falling clock output edge. • Time C) Second qualifying falling clock output edge. • Time D) Internal one shot pulse begins. 5 cycles later clock outputs will be forced low • Time E) Internal one shot pulse ends. 5 cycles + digital delay cycles later the synced clock outputs rise. • Time F) Clock outputs are forced low. (CLKout2 is already low). • Time G) Beginning of digital delay cycles. • Time H) For CLKout2_DDLY = 14; the clock output rises now. 46 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 30179152 FIGURE 16. Absolute Dynamic Digital Delay Programming Example (SYNC_QUAL = 1, Qualify with clock output) Copyright © 1999-2012, Texas Instruments Incorporated 47 LMK04906 18.9.2.1.2 Relative Dynamic Digital Delay Relative dynamic digital delay can be used to program a clock output to a specific phase offset from another clock output. Pros: • Simple direct phase adjustment with respect to same clock output. • The clock output will always behave the same during digital delay adjustment transient. For some divide values there will be no glitch pulse. Cons: • For some clock divide values there may be a glitch pulse due to SYNC assertion. • Adjustments of digital delay requiring the half step bit (CLKoutX_HS) for finer digital delay adjust is complicated. • Use with 0-delay mode may result in PLL1 DLD becoming low and HOLDOVER mode becoming activated. — DISABLE_DLD1_DET can be set to prevent HOLDOVER from becoming activated due to PLL1 DLD becoming low. 18.9.2.1.2.1 RELATIVE DYNAMIC DIGITAL DELAY - EXAMPLE To illustrate the relative dynamic digital delay adjust procedure, consider the following example. System Requirements: • VCO Frequency = 2457.6 MHz • CLKout0 = 819.2 MHz (CLKout0_DIV = 3) • CLKout2 = 491.52 MHz (CLKout2_DIV = 5) • CLKout4 = 491.52 MHz (CLKout4_DIV = 5) • For all clock outputs during initial programming: — CLKoutX_DDLY = 5 — CLKoutX_HS = 0 — NO_SYNC_CLKoutX = 0 The application requires the 491.52 MHz clock to be stepped in 22.5degree steps (~203.4 ps), which is the minimum step resolution allowable by the clock distribution path. That is 1 / 2457.62 MHz / 2 = ~203.4 ps. During the stepping of the 491.52 MHz clocks the 819.2 MHz clock must not be interrupted. Step 1: The device is programmed from register R0 to R30 with values that result in the device being locked and operating as desired, see the system requirements above. The phase of all the output clocks are aligned because all the digital delay and half step values were the same when the SYNC was generated by programming register R30. The timing of this is as shown in Figure 14. Step 2: Now the registers will be programmed to prepare for changing digital delay (or phase) dynamically. Register Purpose SYNC_QUAL = 1 Use clock output for qualifying the SYNC pulse for dynamically adjusting digital delay. EN_SYNC = 1 (default) Required for SYNC functionality. CLKout3_PD = 0 Required when SYNC_QUAL = 1. CLKout3 outputs may be powered down or in use. EN_FEEDBACK_MUX = 1 Enable the feedback mux for SYNC operation for dynamically adjusting digital delay. FEEDBACK_MUX = 1 (CLKout2) Use the clock itself as the SYNC qualification clock. NO_SYNC_CLKout0 = 1 This clock output (819.2 MHz) won't be affected by SYNC. It will always operate without interruption. NO_SYNC_CLKout4 = 1 CLKout4’s phase is not to change with respect to CLKout0. SYNC_EN_AUTO = 0 (default) Automatic generation of SYNC is not allowed because of the half step requirement in relative dynamic digital delay mode. SYNC must be generated manually by toggling the SYNC_POL_INV bit or the SYNC pin. After the above registers have been programmed, the application may now dynamically adjust the digital delay of the 491.52 MHz clocks. Step 3: Adjust digital delay of CLKout2 by one step which is 22.5 degrees or ~203.4 ps. Refer to Table 11 for the programming sequence to step one half clock distribution period forward or backwards. Refer to Section 20.7 CALCULATING DYNAMIC DIGITAL DELAY VALUES FOR ANY DIVIDE for more information on how to calculate digital delay and half step values for other cases. To fulfill the qualifying clock output half step requirement in Table 8 when dynamically adjusting digital delay, the CLKoutX_HS bit must be cleared for clocks with even divides. So before any dynamic digital delay adjustment, CLKoutX_HS must be clear because the clock divide value is even. To achieve the final required digital delay adjustment, the CLKoutX_HS bit may set after SYNC. 48 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 TABLE 11. Programming sequence for one step adjust Step direction and current HS state Programming Sequence Adjust clock output one step forward. CLKout2_HS is 0. 1. CLKout2_HS = 1. Adjust clock output one step forward. CLKout2_HS is 1. 1. CLKout2_DDLY = 11. 2. Perform SYNC event. 3. CLKout2_HS = 0. Adjust clock output one step backward. CLKout2_HS is 0. 1. CLKout2_HS = 1. 2. CLKout2_DDLY =11. 3. Perform SYNC event. Adjust clock output one step backward. CLKout2_HS is 1. 1. CLKout2_HS = 0. After programing the updated CLKout2_DDLY and CLKout2_HS values, perform a SYNC event. The SYNC may be generated by toggling the SYNC pin or by toggling the SYNC_POL_INV bit. Because of the internal one shot pulse, no strict timing of the SYNC pin or SYNC_POL_INV bit is required. After the SYNC event, the clock output will be at the specified phase. See Figure 17 for a detailed view of the timing diagram. The timing diagram critical points are: • Time A) SYNC assertion event is latched. • Time B) First qualifying falling clock output edge. • Time C) Second qualifying falling clock output edge. • Time D) Internal one shot pulse begins. 5 cycles later clock outputs will be forced low. • Time E) Internal one shot pulse ends. 5 cycles + digital delay cycles later the synced clock outputs rise. • Time F) Clock outputs are forced low. (CLKouts are already low). • Time G) Beginning of digital delay cycles. • Time H) For CLKout2_DDLY = 11; the clock output rises now. 30179155 FIGURE 17. Relative Dynamic Digital Delay Programming Example, 2nd adjust. (SYNC_QUAL = 1, Qualify with clock output) Starting condition is after half step is removed (CLKout2_HS = 0). Copyright © 1999-2012, Texas Instruments Incorporated 49 LMK04906 18.9.3 0-Delay Mode When 0-delay mode is enabled the clock output selected by the Feedback Mux is connected to the PLL1 N counter to ensure a fixed phase relationship between the selected CLKin and the fed back CLKout. When all the clock outputs are synced together, all the clock outputs will share the same fixed phase relationship between the selected CLKin and the fed back CLKout. The feedback can be internal or external using FBCLKin port. When 0-delay mode is enabled the lowest frequency clock output is fed back to the Feedback Mux to ensure a repeatable fixed CLKin to CLKout phase relationship between all clock outputs. If a clock output that is not the lowest frequency output is selected for feedback, then clocks with lower frequencies will have an unknown phase relationship with respect the other clocks and clock input. There will be a number of possible phase relationships equal to Feedback_Clock_Frequency / Lower_Clock_Frequency that may occur. The Feedback Mux can select a clock output of some of the clocks for internal feedback or the FBCLKin port for external 0-delay feedback. To use 0-delay mode, the bit EN_FEEDBACK_MUX must be set (=1) to power up the feedback mux. See Section 20.2 PLL PROGRAMMING for more information on programming PLL1_N for 0-delay mode. When using an external VCO mode, internal 0-delay feedback must be used since the FBCLKin port is shared with the Fin input. Table 12 outlines several registers to program for 0-delay mode. TABLE 12. Programming 0-Delay Mode 50 Register Purpose MODE = 2 or 5 Select one of the 0-delay modes for device. EN_FEEDBACK_MUX = 1 Enable feedback mux. FEEDBACK_MUX = Application Specific Select CLKout or FBCLKin for 0-delay feedback. CLKoutX_DIV The divide value of the clock selected by FEEDBACK_MUX is important for PLL2 N value calculation PLL1_N PLL1_N value used with CLKoutX_DIV in loop. Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.0 General Programming Information LMK04906 devices are programmed using 32-bit registers. Each register consists of a 5-bit address field and 27-bit data field. The address field is formed by bits 0 through 4 (LSBs) and the data field is formed by bits 5 through 31 (MSBs). The contents of each register is clocked in MSB first (bit 31), and the LSB (bit 0) last. During programming, the LEuWire signal should be held low. The serial data is clocked in on the rising edge of the CLKuWire signal. After the LSB (bit 0) is clocked in the LEuWire signal should be toggled low-to-high-to-low to latch the contents into the register selected in the address field. It is recommended to program registers in numeric order, for example R0 to R16, and R24 to R31 to achieve proper device operation. Figure 7 illustrates the serial data timing sequence. To achieve proper frequency calibration, the OSCin port must be driven with a valid signal before programming register R30. Changes to PLL2 R divider or the OSCin port frequency require register R30 to be reloaded in order to activate the frequency calibration process. 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY In some cases when programming register R0 to R5 to change the CLKoutX_DIV divide value or CLKoutX_DDLY delay value, 3 additional CLKuWire cycles must occur after loading the register for the newly programmed divide or delay value to take effect. These special cases include: • When CLKoutX_DIV is > 25. • When CLKoutX_DDLY is > 12. Note, loading the digital delay value only prepares for a future SYNC event. Also, since SYNC_EN_AUTO bit = 1 automatically generates a SYNC on the falling edge of LE when R0 to R5 is programmed, further programming considerations must be made when SYNC_EN_AUTO = 1. These special programming cases requiring the additional three clock cycles may be properly programmed by one of the following methods shown in Table 13. TABLE 13. R0 to R5 Special Case SYNC _EN_ AUTO Programming Method 0 or 1 No Additional Clocks Required (Normal) CLKoutX_DIV > 25 or CLKoutX_DDLY > 12 0 Three Extra CLKuWire Clocks (Or program another register) CLKoutX_DIV > 25 or CLKoutX_DDLY > 12 1 Three Extra CLKuWire Clocks while LEuWire is High CLKoutX_DIV & CLKoutX_DDLY CLKoutX_DIV ≤ 25 and CLKoutX_DDLY ≤ 12 Method: No Additional Clocks Required (Normal) No special consideration to CLKuWire is required when changing divide value to ≤ 25, digital delay value to ≤ 12, or when the digital delay and divide value do not change. See MICROWIRE timing Figure 7. Copyright © 1999-2012, Texas Instruments Incorporated 51 LMK04906 Method: Three Extra CLKuWire Clocks Three extra clocks must be provided before CLKoutX_DIV > 25 or CLKoutX_DDLY > 12 take effect. See MICROWIRE timing Figure 8. Also, by programming another register the three clock requirement can be satisfied. Method: Three Extra CLKuWire Clocks with LEuWire Asserted When SYNC_EN_AUTO = 1 the falling edge of LEuWire will generate a SYNC event. CLKoutX_DIV and CLKoutX_DDLY values must be updated before the SYNC event occurs. So 3 CLKuWire rising edges must occur before LEuWire goes low. See MICROWIRE timing Figure 9. Initial Programming Sequence During the recommended programming sequence the device is programmed in order from R0 to R31, so it is expected at least one additional register will be programmed after programming the last CLKoutX_DIV or CLKoutX_DDLY value in R0 to R5. This will result in the extra needed CLKuWire rising edges, so this special note is of little concern. If programming R0 to R5 to change CLKout frequency or digital delay or dynamic digital delay at a later time in the application, care must be taken to provide these extra CLKuWire cycles to properly load the new divide and/or delay values. 19.1.1 Example In this example, all registers have been programmed, the PLLs are locked. An LMK04906 has been generating a clock output frequency of 61.44 MHz on CLKout4 using a VCO frequency of 2457.6 MHz and a divide value of 40. SYNC_EN_AUTO = 0. At a later time the application requires a 30.72 MHz output on CLKout4. By reprogramming register R4 with CLKout4_DIV = 80 twice, the divide value of 80 is set for clock output 4 which results in an output frequency of 30.72 MHz (2457.6 MHz / 80 = 30.72 MHz) on CLKout4. In this example the required 3 CLKuWire cycles were achieved by reprogramming the R4 register with the same value twice. 19.2 RECOMMENDED PROGRAMMING SEQUENCE Registers are programmed in numeric order with R0 being the first and R31 being the last register programmed. The recommended programming sequence involves programming R0 with the reset bit (b17) set to 1 to ensure the device is in a default state. If R0 is programmed again, the reset bit must be cleared to 0 during the programming of R0. 19.2.1 Overview • Program R0 with RESET bit = 1. This ensures that the device is configured with default settings. When RESET = 1, all other R0 bits are ignored. — If R0 is programmed again during the initial configuration of the device, the RESET bit must be cleared. • R0 through R5: CLKouts. — Program as necessary to configure the clock outputs, CLKout0 to CLKout5 as desired. These registers configure clock output controls such as powerdown, digital delay and divider value, analog delay select, and clock source select. • R6 through R8: CLKouts. — Program as necessary to configure the clock outputs, CLKout0 to CLKout5 as desired. These registers configure the output format for each clock outputs and the analog delay for the clock outputs. • R9: Required programming — Program this register as shown in the register map for proper operation. • R10: OSCouts, VCO divider, and 0-delay. — Enable and configure clock outputs OSCout0/1. — Set and select VCO divider (VCO bypass is recommended). — Set 0-delay feedback source if used. • R11: Part mode, SYNC, and XTAL. — Program to configure the mode of the part, to configure SYNC functionality and pin, and to enable crystal mode. • R12: Pins, SYNC, and holdover mode. — Status_LD pin, more SYNC options to generate a SYNC upon PLL1 and/or PLL2 lock detect. — Enable clock features such as holdover. • R13: Pins, holdover mode, and CLKins. — Status_HOLDOVER, Status_CLKin0, and Status_CLKin1 pin controls. — Enable clock inputs for use in specific part modes. • R14: Pins, LOS, CLKins, and DAC. — Status_CLKin1 pin control. — Loss of signal detection, CLKin type, DAC rail detect enable and high and low trip points. • R15: DAC and holdover mode. — Program to enable and set the manual DAC value. — HOLDOVER mode options. • R16: Crystal amplitude. — Increasing XTAL_LVL can improve tunable crystal phase noise performance. • R24: PLL1 and PLL2. 52 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 • • • • • • • — PLL1 N and R delay and PLL1 digital lock delay value. — PLL2 integrated loop filter. R25: DAC and PLL1. — Program to configure DAC update clock divider and PLL1 digital lock detect count. R26: PLL2. — Program to configure PLL2 options. R27: CLKins and PLL1. — Clock input pre-dividers. — Program to configure PLL1 options. R28: PLL1 and PLL2. — Program to configure PLL2 R and PLL1 N. R29: OSCin and PLL2. — Program to configure oscillator input frequency, PLL2 fast phase detector frequency mode, and PLL2 N calibration value. R30: PLL2. — Program to configure PLL2 prescaler and PLL2 N value. R31: uWire lock. — Program to set the uWire_LOCK bit. 19.3 READBACK At no time should the MICROWIRE registers be programmed to any value other than what is specified in the datasheet. For debug of the MICROWIRE interface, it is recommended to simply program an output pin mux to active low and then toggle the output type register between output and inverting output while observing the output pin for a low to high transition. For example, to verify MICROWIRE programming, set the LD_MUX = 0 (Low) and then toggle the LD_TYPE register between 3 (Output, pushpull) and 4 (Output inverted, push-pull). The result will be that the Status_LD pin will toggle from low to high. Readback from the MICROWIRE programming registers is available. The MICROWIRE readback function can be enabled on the Status_LD, Status_HOLDOVER, Status_CLKin0, Status_CLKin1, or SYNC pin by programming the corresponding MUX register to “uWire Readback” and the corresponding TYPE register to "Output (push-pull)." Power on reset defaults the Status_HOLDOVER pin to “uWire Readback.” Figure 10 illustrates the serial data timing sequence for a readback operation for both cases of READBACK_LE = 0 (POR default) and READBACK_LE = 1. To perform a readback operation first set the register to be read back by programming the READBACK_ADDR register. Then after any MICROWIRE write operation, with the LEuWire pin held low continue to clock the CLKuWire pin. On every rising edge of the CLKuWire pin a new data bit is clocked onto the any pins programmed for uWire Readback. If the READBACK_LE bit is set, the LEuWire pin should be left high after LEuWire rising edge while continuing to clock the CLKuWire pin. It is allowable to perform a register read back in the same MICROWIRE operation which set the READBACK_ADDR register value. Data is clocked out MSB first. After 27 clocks all the data values will have been read and the read operation is complete. If READBACK_LE = 1, the LEuWire line may now be lowered. It is allowable for the CLKuWire pin to be clocked additional cycles, but the data on the readback pin will be invalid. CLKuWire must be low before the falling edge of LEuWire. 19.3.1 Readback - Example To readback register R3 perform the following steps: • Write R31 with READBACK_ADDR = 3; READBACK_LE = 0. DATAuWire and CLKuWire are toggled as shown in Figure 7 with new data being clocked in on rising edges of CLKuWire • Toggle LEuWire high and then low as shown in Figure 7 and Figure 10. LEuWire is returned low because READBACK_LE = 0. • Toggle CLKuWire high and then low 27 times to read back all 27 bits of register R3. Data is read MSB first. Data is valid on falling edge of CLKuWire. • Read operation is complete. 19.4 REGISTER MAP AND READBACK REGISTER MAP Table 14 provides the register map for device programming. Normally any register can be read from the same data address it is written to. However, READBACK_LE has adifferent readback address. Also, the DAC_CNT register is a read only register. Table 15 shows the address for READBACK_LE and DAC_CNT. Bits marked as reserved are undefined upon readback. Observe that only the DATA bits are readback during a readback which can result in an offset of 5 bits between the two register tables. Copyright © 1999-2012, Texas Instruments Incorporated 53 54 0 0 0 CLKout 0_PD CLKout 1_PD CLKout 2_PD Register R0 R1 R2 CLKout3_ C0LKout4_ OSCin_Sel OSCin_Sel CLKout CLKout 3_PD 4_PD 0 1 0 1 CLKout3_ CLKout4_ CLKout5 ADLY_SEL ADLY_SEL ADLY_SEL R9 0 CLKout 5_PD R8 0 1 0 1 0 1 0 CLKout5_TYPE [27:24] 1 CLKout2_TYPE [23:20] CLKout3_TYPE [27:24] 0 CLKout0_TYPE [23:20] CLKout1_TYPE [27:24] CLKout5_DDLY [27:18] CLKout4_DDLY [27:18] CLKout3_DDLY [27:18] 0 1 0 1 CLKout4_TYPE [19:16] 0 0 0 0 0 0 1 0 1 CLKout5_ADLY [15:11] CLKout3_ADLY [15:11] CLKout1_ADLY [15:11] 0 1 0 0 0 0 8 7 6 1 0 1 CLKout4_ADLY [9:5] CLKout2_ADLY [9:5] CLKout0_ADLY [9:5] POWERDOWN Data [26:0] RESET 0 9 CLKout 5_HS R7 10 CLKout5_DIV [15:5] 11 CLKout 4_HS 0 0 12 CLKout4_DIV [15:5] 13 CLKout3_DiV [15:5] 14 CLKout2_DIV [15:5] 15 CLKout 3_HS 0 18 CLKout 2_HS CLKout2_DDLY [27:18] 19 CLKout 1_HS R6 0 0 0 R5 CLKout2_ ADLY_SEL R4 0 CLKout0_ ADLY_SEL 0 20 CLKout1_DIV [15:5] 21 CLKout0_DIV [15:5] 22 16 23 CLKout 0_HS 17 24 CLKout1_DDLY [27:18] 25 CLKout0_DDLY [27:18] 26 CLKout1_ ADLY_SEL 29 27 28 0 R3 30 31 TABLE 14. Register Map 0 5 0 0 0 0 0 0 0 0 0 0 4 2 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 1 0 0 1 1 0 0 Address [4:0] 3 1 0 1 0 1 0 1 0 1 0 0 LMK04906 Copyright © 1999-2012, Texas Instruments Incorporated 25 24 0 22 EN_OSCout0 23 0 Copyright © 1999-2012, Texas Instruments Incorporated LOS_ TIMEOUT [31:30] XTAL_ LVL 0 EN_LOS 0 0 1 0 CLKin2_BUF_TYPE 1 0 1 0 1 0 CLKin0_BUF_TYPE EN_MAN_DAC 0 0 0 0 1 0 0 0 1 0 0 0 0 1 0 1 0 1 1 1 1 0 0 0 1 0 0 2 1 0 1 1 1 1 0 0 0 1 1 0 0 1 1 Address [4:0] 3 EN_TRACK CLKin_Sel_INV 1 0 0 4 EN_CLKin1 DAC_LOW_TRIP [11:6] CLKin _Select _MODE [11:8] 0 0 5 0 6 FEEDBACK _MUX [7:5] 7 HOLDOVER _MODE [7:6] 0 0 8 FORCE_ HOLDOVER 0 0 0 HOLDOVER_DLD_CNT [19:6] 0 Status_ CLKin0 _MUX [14:12] 0 9 VCO_DIV [10:8] 10 EN_VTUNE_ RAIL_DET 0 0 DAC_HIGH_TRIP [19:14] 0 0 0 1 SYNC _TYPE [14:12] 11 12 13 14 EN_CLKin0 R16 R15 CLKin1_BUF_TYPE MAN_DAC [31:22] 0 0 1 Status_ CLKin0 _TYPE [18:16] 0 SYNC _CLKin2 _ MUX [19:18] 0 0 SYNC_QUAL 1 0 15 OSCout_DIV [18:16] 16 SYNC_POL_INV DISABLE_ DLD1_DET 0 Status_ CLKin1 _TYPE [26:24] 17 SYNC_EN_AUTO Status_ CLKin1 _MUX [22:20] 18 Data [26:0] 19 EN_PLL2_XTAL EN_CLKin2 R14 0 EN_SYNC R13 NO_SYNC_CLKout5 HOLDOVER _TYPE [26:24] 0 (Note 28) SYNC_PLL1 NO_SYNC_CLKout2 _DLD SYNC_PLL2 NO_SYNC_CLKout3 _DLD HOLDOVER_MUX [31:27] NO_SYNC_CLKout4 0 20 21 NO_SYNC_CLKout1 LD_TYPE [26:24] 0 OSCout0_TYPE [27:24] 26 OSCout0_MUX NO_SYNC_CLKout0 LD_MUX [31:27] 1 0 27 PD_OSCin R12 0 28 29 VCO_MUX MODE [31:27] 0 Register R10 30 EN_ FEEDBACK_MUX R11 31 0 1 0 1 0 1 0 0 LMK04906 55 Register 56 29 25 0 0 0 0 0 0 0 0 21 20 0 0 1 0 18 17 READBACK_ADDR [20:16] 13 12 PLL1_N_DLY [14:12] 14 0 11 0 0 0 PLL2_N [22:5] 0 PLL2_N_CAL [22:5] PLL1_N [19:6] PLL1_R [19:6] 0 PLL2_DLD_CNT [19:6] 0 15 0 PLL2_R3_LF [18:16] 16 PLL1_DLD_CNT [19:6] 0 Data [26:0] 19 0 PLL2_R4_LF [22:20] 22 CLKin0_ PreR_DIV 0 PLL2_P 0 0 R31 0 PLL2_ FAST_PDF 0 1 0 23 CLKin1_ PreR_DIV 0 0 R30 0 1 CLKin2_ PreR_DI V 1 OSCin_FREQ [26:24] 0 PLL2_CP _GAIN [27:26] 24 0 PLL1_CP_POL R29 0 26 PLL2_C3_LF [27:24] PLL1_CP_GAIN PLL2_R 0 27 9 8 0 0 0 PLL1_R_DLY [10:8] 10 6 0 0 PLL1_ WND_ SIZE 7 0 0 5 0 PLL2_CP_TRI PLL1_CP_TRI R28 0 PLL2_ CP_POL EN_PLL2_ REF_2X 0 0 R27 28 DAC_CLK_DIV [31:22] PLL2_C4_LF [31:28] 30 PLL2_ WND_SIZE [31:30] R26 R25 R24 31 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 0 0 0 0 1 1 0 0 1 1 0 0 Address [4:0] 3 1 4 1 0 1 0 1 0 1 0 0 LMK04906 uWire_LOCK READBACK _LE Copyright © 1999-2012, Texas Instruments Incorporated Register RD 23 RD 20 RD 19 0 1 RD 13 1 0 RD 12 Data [26:0] RD 14 RESERVED [26:10] RD 15 RD 16 RD 17 RD 18 DAC_CNT [23:14] LD_TYPE [21:19] RD 21 SYNC_PLL2_DLD RESERVED [26:24] RD 22 SYNC_PLL1_DLD Copyright © 1999-2012, Texas Instruments Incorporated RD R31 RD 24 LD_MUX [26:22] RD 25 READBACK_LE RD R23 RD R12 RD 26 TABLE 15. Readback Register Map 0 RD 11 0 RD 10 0 RD 9 0 RD 7 0 RD 6 0 RD 5 RESERVED [13:0] 0 RD 8 0 RD 4 RD 3 RD 2 RD 1 1 RD 0 LMK04906 uWire_LOCK HOLDOVER_MODE [2:1] EN_TRACK 57 LMK04906 Note 28: Although the value of 0 is written here, during readback the value of READBACK_LE will be read at this location. See Section 19.4 REGISTER MAP AND READBACK REGISTER MAP. 19.5 DEFAULT DEVICE REGISTER SETTINGS AFTER POWER ON RESET Table 16 illustrates the default register settings programmed in silicon for the LMK04906 after power on or asserting the reset bit. Capital X and Y represent numeric values. Clock Output Control Group TABLE 16. Default Device Register Settings after Power On/Reset Field Name Default Value (decimal) Default State CLKout0_PD 1 PD CLKout1_PD 1 PD CLKout2_PD 1 PD CLKout3_PD 0 Normal CLKout4_PD 0 Normal CLKout5_PD 1 PD CLKout3_OSCin_Sel 1 OSCin CLKout4_OSCin_Sel 0 VCO CLKoutX_ADLY_SEL 0 None CLKoutX_DDLY 0 5 RESET 0 Not in reset POWERDOWN 0 Disabled (device is active) Register Bit Location (MSB:LSB) R0 R1 Powerdown control for analog and digital delay, divider, and both output buffers R2 R3 31 R4 R5 Selects the clock source for a clock output from internal VCO or external OSCin R3 30 R4 30 Add analog delay for clock output R0 to R5 Digital delay value 28, 29 R0 to R5 27:18 [10] Performs power on reset for device R0 17 Device power down control R1 17 R0 to R5 16 CLKoutX_HS 0 No shift CLKout0_DIV 25 Divide-by-25 R0 CLKout1_DIV 25 Divide-by-25 R1 CLKout2_DIV 25 Divide-by-25 CLKout3_DIV 1 Divide-by-1 CLKout4_DIV 25 Divide-by-25 R4 CLKout5_DIV 25 Divide-by-25 R5 CLKout1_TYPE 0 Powerdown R6 8 LVCMOS (Norm/ Norm) R7 CLKout3_TYPE 58 Field Description Half shift for digital delay Divide for clock outputs R2 R3 15:5 [11] 27:24 [4] CLKout5_TYPE 0 CLKout0_TYPE 0 Individual clock output format. Select from LVDS/LVPECL/ Powerdown LVCMOS. Powerdown CLKout2_TYPE 0 Powerdown R7 CLKout4_TYPE 1 LVDS R8 CLKoutX_ADLY 0 No delay Analog delay setting for clock output OSCout0_TYPE 1 LVDS OSCout0 default clock output R10 27:24 [4] EN_OSCout0 1 Enabled Enable OSCout0 output buffer R10 22 OSCout0_MUX 0 Bypass Divider Select OSCout divider for OSCout0 or bypass R10 20 PD_OSCin 0 OSCin powered Allows OSCin to be powered down. For use in clock distribution mode. R10 19 OSCout_DIV 0 R10 18:16 [3] Divide-by-8 OSCout divider value R8 R6 23:20 [4] 19:16 [4] R6 to R8 15:11, 9:5 [5] Copyright © 1999-2012, Texas Instruments Incorporated Clock Synchronization Mode Group LMK04906 Field Name Default Value (decimal) Default State VCO_MUX 0 VCO EN_FEEDBACK_MUX 0 Disabled VCO_DIV 2 Register Bit Location (MSB:LSB) Select VCO or VCO Divider output R10 12 Feedback MUX is powered down. R10 11 Field Description Divide-by-2 VCO Divide value R10 10:8 [3] R10 7:5 [3] FEEDBACK_MUX 0 CLKout0 Selects CLKout to feedback into the PLL1 N divider MODE 0 Internal VCO Device mode R11 31:27 [5] EN_SYNC 1 Enabled Enables synchronization circuitry. R11 26 NO_SYNC_CLKout5 0 Will sync R11 25 NO_SYNC_CLKout4 1 Will not sync R11 24 NO_SYNC_CLKout3 1 R11 23 NO_SYNC_CLKout2 0 Will not sync Disable individual clock output Will sync from becoming synchronized. R11 22 NO_SYNC_CLKout1 0 Will sync R11 21 NO_SYNC_CLKout0 0 Will sync R11 20 Mux controlling SYNC pin when set to output R11 19:18 [2] Allows SYNC operations to be qualified by a clock output. R11 17 Sets the polarity of the SYNC pin when input R11 16 SYNC_CLKin2_MUX 0 Logic Low SYNC_QUAL 0 Not qualified SYNC_POL_INV 1 Logic Low SYNC_EN_AUTO 0 Manual SYNC is not started by programming a register R0 to R5. R11 15 SYNC_TYPE 1 Input /w Pull-up SYNC IO pin type R11 14:12 [3] Copyright © 1999-2012, Texas Instruments Incorporated 59 CLKin Control Other Mode Control Group LMK04906 60 Default Value (decimal) Default State Field Description EN_PLL2_XTAL 0 Disabled LD_MUX 3 PLL1 & 2 DLD LD_TYPE 3 SYNC_PLL2_DLD 0 Normal SYNC_PLL1_DLD 0 Normal EN_TRACK 1 HOLDOVER_MODE Register Bit Location (MSB:LSB) Enable Crystal oscillator for OSCin R11 5 Lock detect mux selection when output R12 31:27 [5] R12 26:24 [3] Force synchronization mode until PLL2 locks R12 23 Force synchronization mode until PLL1 locks R12 22 Enable Tracking DAC tracking of the PLL1 tuning voltage R12 8 2 Enable Holdover Causes holdover to activate when lock is lost R12 7:6 [2] HOLDOVER_MUX 7 uWire Readback Holdover mux selection R13 31:27 [5] HOLDOVER_TYPE 3 Output HOLDOVER IO pin type (Push-Pull) R13 26:24 [3] Status_CLKin1_MUX 0 Logic Low Status_CLKin1 pin MUX selection R13 22:20 [3] Status_CLKin0_TYPE 2 Input /w PullStatus_CLKin0 IO pin type down R13 18:16 [3] DISABLE_DLD1_DET 0 Disables PLL1 DLD falling edge Not Disabled from causing HOLDOVER mode to be entered R13 15 Status_CLKin0_MUX 0 Logic Low Status_CLKin0 pin MUX selection R13 14:12 [3] CLKin_SELECT_MODE 3 Manual Select Mode to use in determining reference CLKin for PLL1 R13 11:9 [3] CLKin_Sel_INV 0 Active High Invert Status 0 and 1 pin polarity for input R13 8 EN_CLKin2 1 Usable Set CLKin2 to be usable R13 7 EN_CLKin1 1 Usable Set CLKin1 to be usable R13 6 EN_CLKin0 1 Usable Set CLKin0 to be usable R13 5 LOS_TIMEOUT 0 R14 31:30 [2] EN_LOS 1 Loss of Signal Detect at CLKin R14 28 Input /w PullStatus_CLKin1 pin IO pin type down R14 26:24 [3] Field Name Output LD IO pin type (Push-Pull) 1200 ns, 420 Time until no activity on CLKin kHz asserts LOS Enabled Status_CLKin1_TYPE 2 CLKin2_BUF_TYPE 0 Bipolar CLKin2 Buffer Type R14 22 CLKin1_BUF_TYPE 0 Bipolar CLKin1 Buffer Type R14 21 CLKin0_BUF_TYPE 0 Bipolar CLKin0 Buffer Type R14 20 Copyright © 1999-2012, Texas Instruments Incorporated Group LMK04906 Default Value (decimal) Register Bit Location (MSB:LSB) 0 Voltage from Vcc at which ~50 mV from holdover mode is entered if Vcc EN_VTUNE_RAIL_DAC is enabled. R14 19:14 [6] DAC_LOW_TRIP 0 Voltage from GND at which ~50 mV from holdover mode is entered if GND EN_VTUNE_RAIL_DAC is enabled. R14 11:6 [6] EN_VTUNE_RAIL_DET 0 Field Name DAC Control DAC_HIGH_TRIP MAN_DAC EN_MAN_DAC HOLDOVER_DLD_CNT Field Description Disabled Enable PLL1 unlock state when DAC trip points are achieved R14 5 512 3V/2 Writing to this register will set the value for DAC when in manual override. Readback from this register is DAC value. R15 31:22 [10] 0 Disabled Set manual DAC override R15 20 R15 19:6 [14] R15 5 R16 31:30 [2] 512 FORCE_HOLDOVER 0 XTAL_LVL 0 Copyright © 1999-2012, Texas Instruments Incorporated Default State Lock must be valid n many clocks 512 counts of PLL1 PDF before holdover mode is exited. Holdover not Forces holdover mode. forced 1.65 Vpp Sets drive power level of Crystal 61 PLL Control Group LMK04906 Field Name Default Value (decimal) Default State PLL2_C4_LF 0 10 pF PLL2_C3_LF 0 PLL2_R4_LF Register Bit Location (MSB:LSB) PLL2 integrated capacitor C4 value R24 31:28 [4] 10 pF PLL2 integrated capacitor C3 value R24 27:24 [4] 0 200 Ω PLL2 integrated resistor R4 value R24 22:20 [3] PLL2_R3_LF 0 200 Ω PLL2 integrated resistor R3 value R24 18:16 [3] PLL1_N_DLY 0 No delay Delay in PLL1 feedback path to decrease lag from input to output R24 14:12 [3] PLL1_R_DLY 0 No delay Delay in PLL1 reference path to increase lag from input to output R24 10:8 [3] PLL1_WND_SIZE 3 40 ns Window size used for digital lock detect for PLL1 R24 7:6 [2] DAC_CLK_DIV 4 R25 31:22 [10] DAC update clock divisor. Divide-by-4 Divides PLL1 phase detector frequency. PLL1_DLD_CNT 1024 1024 cycles Lock must be valid n many cycles before LD is asserted R25 19:6 [14] PLL2_WND_SIZE 0 Reserved (Note 29) Window size used for digital lock detect for PLL2 R26 31:30 [2] EN_PLL2_REF_2X 0 Disabled, 1x Doubles reference frequency of PLL2. R26 29 PLL2_CP_POL 0 Negative Polarity of PLL2 Charge Pump R26 28 PLL2_CP_GAIN 3 3.2 mA PLL2 Charge Pump Gain R26 27:26 [2] R26 19:6 [14] PLL2_DLD_CNT 8192 Number of PDF cycles which phase error must be within DLD 8192 Counts window before LD state is asserted. PLL2_CP_TRI 0 Active PLL2 Charge Pump Active R26 5 PLL1_CP_POL 1 Positive Polarity of PLL1 Charge Pump R27 28 PLL1_CP_GAIN 0 100 uA PLL1 Charge Pump Gain R27 27:26 [2] R27 25:24 [2] CLKin2_PreR_DIV 0 CLKin2 Pre-R divide value (1, 2, Divide-by-1 4, or 8) CLKin1_PreR_DIV 0 Divide-by-1 CLKin1 Pre-R divide value (1, 2, 4, or 8) R27 23:22 [2] CLKin0_PreR_DIV 0 Divide-by-1 CLKin0 Pre-R divide value (1, 2, 4, or 8) R27 21:20 [2] PLL1_R 96 Divide-by-96 PLL1 R Divider (1 to 16383) PLL1_CP_TRI 0 PLL2_R 4 R27 19:6 [14] PLL1 Charge Pump Active R27 5 Divide-by-4 PLL2 R Divider (1 to 4095) R28 31:20 [12] PLL1 N Divider (1 to 16383) R28 19:6 [14] OSCin frequency range R29 26:24 [3] R29 23 Must be programmed to PLL2_N value. R29 22:5 [18] Active 192 Divideby-192 OSCin_FREQ 7 448 to 511 MHz PLL2_FAST_PDF 1 PLL2 PDF > When set, PLL2 PDF of greater 100 MHz than 100 MHz may be used PLL2_N_CAL 48 Divide-by-48 PLL1_N 62 Field Description PLL2_P 2 Divide-by-2 PLL2 N Divider Prescaler (2 to 8) R30 26:24 [3] PLL2_N 48 Divide-by-48 PLL2 N Divider (1 to 262143) R30 22:5 [18] Copyright © 1999-2012, Texas Instruments Incorporated Group LMK04906 Default Value (decimal) Default State READBACK_LE 0 LEuWire Low for Readback READBACK_ADDR 31 uWire_LOCK 0 Field Name Field Description State LEuWire pin must be in for readback Register 31 Register to read back Writable The values of registers R0 to R30 are lockable Register Bit Location (MSB:LSB) R31 21 R31 20:16 [5] R31 5 Note 29: This register must be reprogrammed to a value of 2 (3.7 ns) during user programming. Copyright © 1999-2012, Texas Instruments Incorporated 63 LMK04906 19.6 REGISTER R0 TO R5 Registers R0 through R5 control the 6 clock outputs CLKout0 to CLKout5. Register R0 controls CLKout0, Register R1 controls CLKout1, and so on. All functions of the bits in these six registers are identical except the different registers control different clock outputs. The X in CLKoutX_PD, CLKoutX_ADLY_SEL, CLKoutX_DDLY, CLKoutX_HS, CLKoutX_DIV denote the actual clock output which may be from 0 to 5. The RESET bit is only in register R0. The POWERDOWN bit is only in register R1. The CLKoutX_OSCin_Sel bit is only in registers R3 and R4. 19.6.1 CLKoutX_PD, Powerdown CLKoutX Output Path This bit powers down the clock output as specified by CLKoutX. This includes the divider, digital delay, analog delay, and output buffers. CLKoutX_PD R0-R5[31] State 0 Power up clock output 1 Power down clock output 19.6.2 CLKoutX_OSCin_Sel, Clock output source This bit sets the source for the clock CLKoutX. The selected source will be either from a VCO via Mode Mux1 or from the OSCin buffer. This bit is valid only for registers R3 and R4, clock outputs CLKout3 and CLKout4 respectively. All other clock outputs are driven by a VCO via Mode Mux1. CLKoutX_OSCin_Sel R3-R4[30] Clock output source 0 VCO 1 OSCin 19.6.3 CLKoutX_ADLX_SEL[29], CLKoutX_ADLX_SEL[28], Select Analog Delay These bits individually select the analog delay block (Section 19.7.2 CLKoutX_ADLY) for use with CLKoutX. If a clock output does not use analog delay, the analog delay block is powered down. CLKoutX_ADLX_SEL[29], CLKoutX_ADLX_SEL[28] 64 R0-R5[28],[29] State 0 Analog delay powered down 1 Analog delay on CLKoutX Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.6.4 CLKoutX_DDLY, Clock Channel Digital Delay CLKoutX_DDLY and CLKoutX_HS sets the digital delay used for CLKoutX. This value only takes effect during a SYNC event and if the NO_SYNC_CLKoutX bit is cleared for this clock output. See Section 18.9.2 Clock Output Synchronization (SYNC). Programming CLKoutX_DDLY can require special attention. See section Section 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY for more details. Using a CLKoutX_DDLY value of 13 or greater will cause the clock output to operate in extended mode regardless of the clock ouptut's divide value or the half step value. One clock cycle is equal to the period of the clock distribution path. The period of the clock distribution path is equal to VCO Divider value divided by the frequency of the VCO. If the VCO divider is disabled or an external VCO is used, the VCO divide value is treated as 1. tclock distribution path = VCO divide value / fVCO CLKoutX_DDLY, 10 bits R0-R5[27:18] Delay 0 (0x00) 5 clock cycles 1 (0x01) 5 clock cycles 2 (0x02) 5 clock cycles 3 (0x03) 5 clock cycles 4 (0x04) 5 clock cycles 5 (0x05) 5 clock cycles 6 (0x06) 6 clock cycles 7 (0x07) 7 clock cycles ... ... 12 (0x0C) 12 clock cycles 13 (0x0D) 13 clock cycles ... ... 520 (0x208) 520 clock cycles 521 (0x209) 521 clock cycles 522 (0x20A) 522 clock cycles Power Mode Normal Mode Extended Mode 19.6.5 RESET The RESET bit is located in register R0 only. Setting this bit will cause the silicon default values to be loaded. When programming register R0 with the RESET bit set, all other programmed values are ignored. After resetting the device, the register R0 must be programmed again (with RESET = 0) to set non-default values in register R0. The reset occurs on the falling edge of the LEuWire pin which loaded R0 with RESET = 1. The RESET bit is automatically cleared upon writing any other register. For instance, when R0 is written to again with default values. RESET R0[17] State 0 Normal operation 1 Reset (automatically cleared) Copyright © 1999-2012, Texas Instruments Incorporated 65 LMK04906 19.6.6 POWERDOWN The POWERDOWN bit is located in register R1 only. Setting the bit causes the device to enter powerdown mode. Normal operation is resumed by clearing this bit with MICROWIRE. POWERDOWN R1[17] State 0 Normal operation 1 Powerdown 19.6.7 CLKoutX_HS, Digital Delay Half Shift This bit subtracts a half clock cycle of the clock distribution path period to the digital delay of CLKoutX. CLKoutX_HS is used together with CLKoutX_DDLY to set the digital delay value. When changing CLKoutX_HS, the digital delay immediately takes effect without a SYNC event. CLKoutX_HS R0-R5[16] State 0 Normal 1 Subtract half of a clock distribution path period from the total digital delay 19.6.8 CLKoutX_DIV, Clock Output Divide CLKoutX_DIV sets the divide value for the clock output. The divide may be even or odd. Both even and odd divides output a 50% duty cycle clock. Using a divide value of 26 or greater will cause the clock output to operate in extended mode regardless of the clock output's digital delay value. Programming CLKoutX_DIV can require special attention. See section Section 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY for more details. CLKoutX_DIV, 11 bits R0-R5[15:5] Divide Value 0 (0x00) Reserved 1 (0x01) 1 (Note 30) 2 (0x02) 2 (Note 31) 3 (0x03) 3 4 (0x04) 4 (Note 31) 5 (0x05) 5 (Note 31) 6 (0x06) 6 ... ... 24 (0x18) 24 25 (0x19) 25 26 (0x1A) 26 27 (0x1B) 27 ... ... 1044 (0x414) 1044 1045 (0x415) 1045 Power Mode Normal Mode Extended Mode Note 30: CLKoutX_HS must = 0 for divide by 1. Note 31: After programming PLL2_N value, a SYNC must occur on channels using this divide value. Programming PLL2_N does generate a SYNC event automatically which satisfies this requirement, but NO_SYNC_CLKoutX must be set to 0 for these clock outputs. 66 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.7 REGISTERS R6 TO R8 Registers R6 to R8 set the clock output types and analog delays. 19.7.1 CLKoutX_TYPE The clock output types of the LMK04906 are individually programmable. The CLKoutX_TYPE registers set the output type of an individual clock output to LVDS, LVPECL, LVCMOS, or powers down the output buffer. Note that LVPECL supports four different amplitude levels and LVCMOS supports single LVCMOS outputs, inverted, and normal polarity of each output pin for maximum flexibility. The programming addresses table shows at what register and address the specified clock output CLKoutX_TYPE register is located. The CLKoutX_TYPE table shows the programming definition for these registers. CLKoutX_TYPE Programming Addresses CLKoutX Programming Address CLKout0 R6[23:20] CLKout1 R6[27:24] CLKout2 R723:20] CLKout3 R7[27:24] CLKout4 R8[19:16] CLKout5 R8[27:24] CLKoutX_TYPE, 4 bits R6-R8[31:28, 27:24, 23:20] Definition 0 (0x00) Power down 1 (0x01) LVDS 2 (0x02) LVPECL (700 mVpp) 3 (0x03) LVPECL (1200 mVpp) 4 (0x04) LVPECL (1600 mVpp) 5 (0x05) LVPECL (2000 mVpp) 6 (0x06) LVCMOS (Norm/Inv) 7 (0x07) LVCMOS (Inv/Norm) 8 (0x08) LVCMOS (Norm/Norm) 9 (0x09) LVCMOS (Inv/Inv) 10 (0x0A) LVCMOS (Low/Norm) 11 (0x0A) LVCMOS (Low/Inv) 12 (0x0C) LVCMOS (Norm/Low) 13 (0x0D) LVCMOS (Inv/Low) 14 (0x0E) LVCMOS (Low/Low) Copyright © 1999-2012, Texas Instruments Incorporated 67 LMK04906 19.7.2 CLKoutX_ADLY These registers control the analog delay of the clock output CLKoutX. Adding analog delay to the output will increase the noise floor of the output. For this analog delay to be active for a clock output, it must be selected with CLKoutX_ADL_SEL. If neither clock output in a clock output selects the analog delay, then the analog delay block is powered down. In addition to the programmed delay, a fixed 500 ps of delay will be added by engaging the delay block. The programming addresses table shows at what register and address the specified clock output CLKoutX_ADLY register is located. The CLKoutX_ADLY table shows the programming definition for these registers. CLKoutX_ADLY Programming Addresses CLKoutX_ADLY Programming Address CLKout0_ADLY R6[9:5] CLKout1_ADLY R6[15:11] CLKout2_ADLY R7[9:5] CLKout3_ADLY R7[15:11] CLKout4_ADLY R8[9:5] CLKout5_ADLY R8[15:11] CLKoutX_ADLY, 5 bits 68 R6-R8[15:11, 9:5] Definition 0 (0x00) 500 ps + No delay 1 (0x01) 500 ps + 25 ps 2 (0x02) 500 ps + 50 ps 3 (0x03) 500 ps + 75 ps 4 (0x04) 500 ps + 100 ps 5 (0x05) 500 ps + 125 ps 6 (0x06) 500 ps + 150 ps 7 (0x07) 500 ps + 175 ps 8 (0x08) 500 ps + 200 ps 9 (0x09) 500 ps + 225 ps 10 (0x0A) 500 ps + 250 ps 11 (0x0B) 500 ps + 275 ps 12 (0x0C) 500 ps + 300 ps 13 (0x0D) 500 ps + 325 ps 14 (0x0E) 500 ps + 350 ps 15 (0x0F) 500 ps + 375 ps 16 (0x10) 500 ps + 400 ps 17 (0x11) 500 ps + 425 ps 18 (0x12) 500 ps + 450 ps 19 (0x13) 500 ps + 475 ps 20 (0x14) 500 ps + 500 ps 21 (0x15) 500 ps + 525 ps 22 (0x16) 500 ps + 550 ps 23 (0x17) 500 ps + 575 ps Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.8 REGISTER R10 19.8.1 OSCout0_TYPE The OSCout0 clock output has a programmable output type. The OSCout0_TYPE register sets the output type to LVDS, LVPECL, LVCMOS, or powers down the output buffer. Note that LVPECL supports four different amplitude levels and LVCMOS supports dual and single LVCMOS outputs with inverted, and normal polarity of each output pin for maximum flexibility. To turn on the output, the OSCout0_TYPE must be set to a non-power down setting and enabled with Section 19.8.2 EN_OSCout0, OSCout0 Output Enable. OSCout0_TYPE, 4 bits R10[27:24] Definition 0 (0x00) Powerdown 1 (0x01) LVDS 2 (0x02) LVPECL (700 mVpp) 3 (0x03) LVPECL (1200 mVpp) 4 (0x04) LVPECL (1600 mVpp) 5 (0x05) LVPECL (2000 mVpp) 6 (0x06) LVCMOS (Norm/Inv) 7 (0x07) LVCMOS (Inv/Norm) 8 (0x08) LVCMOS (Norm/Norm) 9 (0x09) LVCMOS (Inv/Inv) 10 (0x0A) LVCMOS (Low/Norm) 11 (0x0B) LVCMOS (Low/Inv) 12 (0x0C) LVCMOS (Norm/Low) 13 (0x0D) LVCMOS (Inv/Low) 14 (0x0E) LVCMOS (Low/Low) 19.8.2 EN_OSCout0, OSCout0 Output Enable EN_OSCout0 is used to enable an oscillator buffered output. EN_OSCout0 R10[22] Output State 0 OSCout0 Disabled 1 OSCout0 Enabled OSCout0 note: In addition to enabling the output with EN_OSCout0. The OSCout0_TYPE must be programmed to a non-power down value for the output buffer to power up. Copyright © 1999-2012, Texas Instruments Incorporated 69 LMK04906 19.8.3 OSCout0_MUX, Clock Output Mux Sets OSCout0 buffer to output a divided or bypassed OSCin signal. The divisor is set by Section 19.8.5 OSCout_DIV, Oscillator Output Divide. OSCout0_MUX R10[20] Mux Output 0 Bypass divider 1 Divided 19.8.4 PD_OSCin, OSCin Powerdown Control Except in clock distribution mode, the OSCin buffer must always be powered up. In clock distribution mode, the OSCin buffer must be powered down if not used. PD_OSCin R10[19] OSCin Buffer 0 Normal Operation 1 Powerdown 19.8.5 OSCout_DIV, Oscillator Output Divide The OSCout divider can be programmed from 2 to 8. Divide by 1 is achieved by bypassing the divider with Section 19.8.3 OSCout0_MUX, Clock Output Mux. Note that OSCout_DIV will be in the PLL1 N feedback path if OSCout0_MUX selects divided as an output. When OSCout_DIV is in the PLL1 N feedback path, the OSCout_DIV divide value must be accounted for when programming PLL1 N. See Section 20.2 PLL PROGRAMMING for more information on programming PLL1 to lock. OSCout_DIV, 3 bits R10[18:16] Divide 0 (0x00) 8 1 (0x01) 2 2 (0x02) 2 3 (0x03) 3 4 (0x04) 4 5 (0x05) 5 6 (0x06) 6 7 (0x07) 7 19.8.6 VCO_MUX When the internal VCO is used, the VCO divider can be selected to divide the VCO output frequency to reduce the frequency on the clock distribution path. It is recommended to use the VCO directly unless: • Very low output frequencies are required. • If using the VCO divider results in three or more clock output divider/delays changing from extended to normal power mode, a small power savings may be achieved by using the VCO divider. A consequence of using the VCO divider is a small degradation in phase noise. VCO_MUX R10[12] Divide 0 VCO selected 1 VCO divider selected 19.8.7 EN_FEEDBACK_MUX When using 0-delay or dynamic digital delay (SYNC_QUAL = 1), EN_FEEDBACK_MUX must be set to 1 to power up the feedback mux. EN_FEEDBACK_MUX 70 R10[11] Divide 0 Feedback mux powered down 1 Feedback mux enabled Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.8.8 VCO_DIV, VCO Divider Divide value of the VCO Divider. See Section 20.2 PLL PROGRAMMING for more information on programming PLL2 to lock. VCO_DIV, 3 bits R10[10:8] Divide 0 (0x00) 8 1 (0x01) 2 2 (0x02) 2 3 (0x03) 3 4 (0x04) 4 5 (0x05) 5 6 (0x06) 6 7 (0x07) 7 19.8.9 FEEDBACK_MUX When in 0-delay mode, the feedback mux selects the clock output to be fed back into the PLL1 N Divider. FEEDBACK_MUX, 3 bits R10[7:5] Divide 0 (0x00) Reserved 1 (0x01) CLKout1 2 (0x02) Reserved 3 (0x03) CLKout3 4 (0x04) CLKout4 5 (0x05) CLKout5 6 (0x06) FBCLKin/FBCLKin* Copyright © 1999-2012, Texas Instruments Incorporated 71 LMK04906 19.9 REGISTER R11 19.9.1 MODE: Device Mode MODE determines how the LMK04906 operates from a high level. Different blocks of the device can be powered up and down for specific application requirements from a dual loop architecture to clock distribution. The LMK04906 can operate in: • Dual PLL mode with the internal VCO or an external VCO. • Single PLL mode uses PLL2 and powers down PLL1. OSCin is used for PLL reference input. • Clock Distribution mode allows use of CLKin1 to distribute to clock outputs CLKout0 through CLKout11, and OSCin to distribute to OSCout0, and optionally CLKout3 through CLKout9. For the PLL modes, 0-delay can be used have deterministic phase with the input clock. For the PLL modes it is also possible to use an external VCO. MODE, 5 bits R11[31:27] Value 0 (0x00) Dual PLL, Internal VCO 1 (0x01) Reserved 2 (0x02) Dual PLL, Internal VCO, 0-Delay 3 (0x03) Dual PLL, External VCO (Fin) 4 (0x04) Reserved 5 (0x05) Dual PLL, External VCO (Fin), 0-Delay 6 (0x06) PLL2, Internal VCO 7 (0x07) Reserved 8 (0x08) PLL2, Internal VCO, 0–Delay 9 (0x09) Reserved 10 (0x0A) Reserved 11 (0x0B) PLL2, External VCO (Fin) 12 (0x0C) Reserved 13 (0x0D) Reserved 14 (0x0E) Reserved 15 (0x0F) Reserved 16 (0x10) Clock Distribution 19.9.2 EN_SYNC, Enable Synchronization The EN_SYNC bit (default on) must be enabled for synchronization to work. Synchronization is required for dynamic digital delay. The synchronization enable may be turned off once the clocks are operating to save current. If EN_SYNC is set after it has been cleared (a transition from 0 to 1), a SYNC is generated that can disrupt the active clock outputs. Setting the NO_SYNC_CLKoutX bits will prevent this SYNC pulse from affecting the output clocks. Setting the EN_SYNC bit is not a valid method for synchronizing the clock outputs. See the section for more information on synchronization. EN_SYNC 72 R11[26] Definition 0 Synchronization disabled 1 Synchronization enabled Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.9.3 NO_SYNC_CLKoutX The NO_SYNC_CLKoutX bits prevent individual clock outputs from becoming synchronized during a SYNC event. A reason to prevent individual clock output from becoming synchronized is that during synchronization, the clock output is in a fixed low state or can have a glitch pulse. By disabling SYNC on a clock output, it will continue to operate normally during a SYNC event. Digital delay requires a SYNC operation to take effect. If NO_SYNC_CLKoutX is set before a SYNC event, the digital delay value will be unused. Setting the NO_SYNC_CLKoutX bit has no effect on clocks already synchronized together. NO_SYNC_CLKoutX Programming Addresses NO_SYNC_CLKoutX Programming Address CLKout0 R11:20 CLKout1 R11:21 CLKout2 R11:22 CLKout3 R11:23 CLKout4 R11:24 CLKout5 R11:25 NO_SYNC_CLKoutX R11[25, 24, 23, 22, 21, 20] Definition 0 CLKoutX will synchronize 1 CLKoutX will not synchronize 19.9.4 SYNC_CLKin2_MUX Mux controlling SYNC/Status_CLKin2 pin. All the outputs logic is active high when SYNC_TYPE = 3 (Output). All the outputs logic is active low when SYNC_TYPE = 4 (Output Inverted). For example, when SYNC_MUX = 0 (Logic Low) and SYNC_TYPE = 3 (Output) then SYNC outputs a logic low. When SYNC_MUX = 0 (Logic Low) and SYNC_TYPE = 4 (Output Inverted) then SYNC outputs a logic high. SYNC_MUX, 2 bits R11[19:18] Sync pin output 0 (0x00) Logic Low 1 (0x01) CLKin2 LOS 2 (0x02) CLKin2 Selected 3 (0x03) uWire Readback Copyright © 1999-2012, Texas Instruments Incorporated 73 LMK04906 19.9.5 SYNC_QUAL When SYNC_QUAL is set, clock outputs will be synchronized to an existing clock output selected by FEEDBACK_MUX. By using the NO_SYNC_CLKoutX bits, selected clock outputs will not be interrupted during the SYNC event. Qualifying the SYNC by an output clock means that the pulse which turns the clock outputs off and on will have a fixed time relationship to the qualifying output clock. SYNC_QUAL = 1 requires CLKout2_PD = 0 for proper operation. CLKout2_TYPE may be set to Powerdown mode. See Section 18.9.2 Clock Output Synchronization (SYNC) for more information. SYNC_QUAL R11[17] Mode 0 No qualification 1 Qualification by clock output from feedback mux (Must set CLKout2_PD = 0) 19.9.6 SYNC_POL_INV Sets the polarity of the SYNC pin when input. When SYNC is asserted the clock outputs will transition to a low state. See Section 18.9.2 Clock Output Synchronization (SYNC) for more information on SYNC. A SYNC event can be generated by toggling this bit through the MICROWIRE interface. SYNC_POL_INV R11[16] Polarity 0 SYNC is active high 1 SYNC is active low 19.9.7 SYNC_EN_AUTO When set, causes a SYNC event to occur when programming register R0 to R5 to adjust digital delay values. The SYNC event will coincide with the LEuWire pin falling edge. Refer to Section 19.1 SPECIAL PROGRAMMING CASE FOR R0 to R5 for CLKoutX_DIV & CLKoutX_DDLY for more information on possible special programming considerations when SYNC_EN_AUTO = 1. SYNC_EN_AUTO R11[15] Mode 0 Manual SYNC 1 SYNC Internally Generated 19.9.8 SYNC_TYPE Sets the IO type of the SYNC pin. SYNC_TYPE, 3 bits R11[14:12] Polarity 0 (0x00) Input 1 (0x01) Input /w pull-up resistor 2 (0x02) Input /w pull-down resistor 3 (0x03) Output (push-pull) 4 (0x04) Output inverted (push-pull) 5 (0x05) Output (open source) 6 (0x06) Output (open drain) When in output mode the SYNC input is forced to 0 regardless of the SYNC_MUX setting. A synchronization can then be activated by uWire by programming the SYNC_POL_INV register to active low to assert SYNC. SYNC can then be released by programming SYNC_POL_INV to active high. Using this uWire programming method to create a SYNC event saves the need for an IO pin from another device. 74 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.9.9 EN_PLL2_XTAL If an external crystal is being used to implement a discrete VCXO, the internal feedback amplifier must be enabled with this bit in order to complete the oscillator circuit. EN_PLL2_XTAL R11[5] Oscillator Amplifier State 0 Disabled 1 Enabled 19.10 REGISTER R12 19.10.1 LD_MUX LD_MUX sets the output value of the LD pin. All the outputs logic is active high when LD_TYPE = 3 (Output). All the outputs logic is active low when LD_TYPE = 4 (Output Inverted). For example, when LD_MUX = 0 (Logic Low) and LD_TYPE = 3 (Output) then Status_LD outputs a logic low. When LD_MUX = 0 (Logic Low) and LD_TYPE = 4 (Output Inverted) then Status_LD outputs a logic high. LD_MUX, 5 bits R12[31:27] Divide 0 (0x00) Logic Low 1 (0x01) PLL1 DLD 2 (0x02) PLL2 DLD 3 (0x03) PLL1 & PLL2 DLD 4 (0x04) Holdover Status 5 (0x05) DAC Locked 6 (0x06) Reserved 7 (0x07) uWire Readback 8 (0x08) DAC Rail 9 (0x09) DAC Low 10 (0x0A) DAC High 11 (0x0B) PLL1_N 12 (0x0C) PLL1_N/2 13 (0x0D) PLL2 N 14 (0x0E) PLL2 N/2 15 (0x0F) PLL1_R 16 (0x10) PLL1_R/2 17 (0x11) PLL2 R (Note 32) 18 (0x12) PLL2 R/2 (Note 32) Note 32: Only valid when HOLDOVER_MUX is not set to 2 (PLL2_DLD) or 3 (PLL1 & PLL2 DLD). 19.10.2 LD_TYPE Sets the IO type of the LD pin. LD_TYPE, 3 bits R12[26:24] Polarity 0 (0x00) Reserved 1 (0x01) Reserved 2 (0x02) Reserved 3 (0x03) Output (push-pull) 4 (0x04) Output inverted (push-pull) 5 (0x05) Output (open source) 6 (0x06) Output (open drain) 19.10.3 SYNC_PLLX_DLD By setting SYNC_PLLX_DLD a SYNC mode will be engaged (asserted SYNC) until PLL1 and/or PLL2 locks. SYNC_QUAL must be 0 to use this functionality. Copyright © 1999-2012, Texas Instruments Incorporated 75 LMK04906 SYNC_PLL2_DLD R12[23] Sync Mode Forced 0 No 1 Yes SYNC_PLL1_DLD R12[22] Sync Mode Forced 0 No 1 Yes 19.10.4 EN_TRACK Enable the DAC to track the PLL1 tuning voltage. For optional use in in holdover mode. Tracking can be used to monitor PLL1 voltage by readback of DAC_CNT register in any mode. EN_TRACK R12[8] DAC Tracking 0 Disabled 1 Enabled 19.10.5 HOLDOVER_MODE Enable the holdover mode. HOLDOVER_MODE, 2 bits 76 R12[7:6] Holdover Mode 0 Reserved 1 Disabled 2 Enabled 3 Reserved Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.11 REGISTER R13 19.11.1 HOLDOVER_MUX HOLDOVER_MUX sets the output value of the Status_Holdover pin. The outputs are active high when HOLDOVER_TYPE = 3 (Output). The outputs are active low when HOLDOVER_TYPE = 4 (Output Inverted). HOLDOVER_MUX, 5 bits R13[31:27] Divide 0 (0x00) Logic Low 1 (0x01) PLL1 DLD 2 (0x02) PLL2 DLD 3 (0x03) PLL1 & PLL2 DLD 4 (0x04) Holdover Status 5 (0x05) DAC Locked 6 (0x06) Reserved 7 (0x07) uWire Readback 8 (0x08) DAC Rail 9 (0x09) DAC Low 10 (0x0A) DAC High 11 (0x0B) PLL1 N 12 (0x0C) PLL1 N/2 13 (0x0D) PLL2 N 14 (0x0E) PLL2 N/2 15 (0x0F) PLL1 R 16 (0x10) PLL1 R/2 17 (0x11) PLL2 R (Note 33) 18 (0x12) PLL2 R/2 (Note 33) Note 33: Only valid when LD_MUX is not set to 2 (PLL2_DLD) or 3 (PLL1 & PLL2 DLD). 19.11.2 HOLDOVER_TYPE Sets the IO mode of the Status_Holdover pin. HOLDOVER_TYPE, 3 bits R13[26:24] Polarity 0 (0x00) Reserved 1 (0x01) Reserved 2 (0x02) Reserved 3 (0x03) Output (push-pull) 4 (0x04) Output inverted (push-pull) 5 (0x05) Output (open source) 6 (0x06) Output (open drain) 19.11.3 Status_CLKin1_MUX Status_CLKin1_MUX sets the output value of the Status_CLKin1 pin. If Section 19.12.3 Status_CLKin1_TYPE is set to an input type, this register has no effect. This MUX register only sets the output signal. The outputs are active high when Status_CLKin1_TYPE = 3 (Output). The outputs are active low when Status_CLKin1_TYPE = 4 (Output Inverted). Status_CLKin1_MUX, 3 bits R13[22:20] Divide 0 (0x00) Logic Low 1 (0x01) CLKin1 LOS 2 (0x02) CLKin1 Selected 3 (0x03) DAC Locked 4 (0x04) DAC Low Copyright © 1999-2012, Texas Instruments Incorporated 77 LMK04906 R13[22:20] Divide 5 (0x05) DAC High 6 (0x06) uWire Readback 19.11.4 Status_CLKin0_TYPE Status_CLKin0_TYPE sets the IO type of the Status_CLKin0 pin. Status_CLKin0_TYPE, 3 bits R13[18:16] Polarity 0 (0x00) Input 1 (0x01) Input /w pull-up resistor 2 (0x02) Input /w pull-down resistor 3 (0x03) Output (push-pull) 4 (0x04) Output inverted (push-pull) 5 (0x05) Output (open source) 6 (0x06) Output (open drain) 19.11.5 DISABLE_DLD1_DET DISABLE_DLD1_DET disables the HOLDOVER mode from being activated when PLL1 lock detect signal transitions from high to low. When using Pin Select Mode as the input clock switch mode, this bit should normally be set. DISABLE_DLD1_DET R13[15] Holdover DLD1 Detect 0 PLL1 DLD causes clock switch event 1 PLL1 DLD does not cause clock switch event 19.11.6 Status_CLKin0_MUX CLKin0_MUX sets the output value of the Status_CLKin0 pin. If Section 19.11.4 Status_CLKin0_TYPE is set to an input type, this register has no effect. This MUX register only sets the output signal. The outputs logic is active high when Status_CLKin0_TYPE = 3 (Output). The outputs logic is active low when Status_CLKin0_TYPE = 4 (Output Inverted). Status_CLKin0_MUX, 3 bits 78 R13[14:12] Divide 0 (0x00) Logic Low 1 (0x01) CLKin0 LOS 2 (0x02) CLKin0 Selected 3 (0x03) DAC Locked 4 (0x04) DAC Low 5 (0x05) DAC High 6 (0x06) uWire Readback Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.11.7 CLKin_SELECT_MODE CLKin_SELECT_MODE sets the mode used in determining reference CLKin for PLL1. CLKin_SELECT_MODE, 3 bits R13[11:9] Mode 0 (0x00) CLKin0 Manual 1 (0x01) CLKin1 Manual 2 (0x02) CLKin2 Manual 3 (0x03) Pin Select Mode 4 (0x04) Auto Mode 5 (0x05) Reserved 6 (0x06) Auto mode & next clock pin select 7 (0x07) Reserved 19.11.8 CLKin_Sel_INV CLKin_Sel_INV sets the input polarity of Status_CLKin0 and Status_CLKin1 pins. CLKin_Sel_INV R13[8] Input 0 Active High 1 Active Low 19.11.9 EN_CLKinX Each clock input can individually be enabled to be used during auto-switching CLKin_SELECT_MODE. Clock input switching priority is always CLKin0 → CLKin1 → CLKin2 → CLKin0. EN_CLKin2 R13[7] Input 0 No 1 Yes EN_CLKin1 R13[6] Valid 0 No 1 Yes EN_CLKin0 R13[5] Valid 0 No 1 Yes Copyright © 1999-2012, Texas Instruments Incorporated 79 LMK04906 19.12 REGISTER 14 19.12.1 LOS_TIMEOUT This bit controls the amount of time in which no activity on a CLKin causes LOS (Loss-of-Signal) to be asserted. LOS_TIMEOUT, 2 bits R14[31:30] Timeout 0 (0x00) 1200 ns, 420 kHz 1 (0x01) 206 ns, 2.5 MHz 2 (0x02) 52.9 ns, 10 MHz 3 (0x03) 23.7 ns, 22 MHz 19.12.2 EN_LOS Enables the LOS (Loss-of-Signal) timeout control. EN_LOS R14[28] LOS 0 Disabled 1 Enabled 19.12.3 Status_CLKin1_TYPE Sets the IO type of the Status_CLKin1 pin. Status_CLKin1_TYPE, 3 bits R14[26:24] Polarity 0 (0x00) Input 1 (0x01) Input /w pull-up resistor 2 (0x02) Input /w pull-down resistor 3 (0x03) Output (push-pull) 4 (0x04) Output inverted (push-pull) 5 (0x05) Output (open source) 6 (0x06) Output (open drain) 19.12.4 CLKinX_BUF_TYPE, PLL1 CLKinX/CLKinX* Buffer Type There are two input buffer types for the PLL1 reference clock inputs: either bipolar or CMOS. Bipolar is recommended for differential inputs such as LVDS and LVPECL. CMOS is recommended for DC coupled single ended inputs. When using bipolar, CLKinX and CLKinX* input pins must be AC coupled when using a differential or single ended input. When using CMOS, CLKinX and CLKinX* input pins may be AC or DC coupled with a differential input. When using CMOS in single ended mode, the unused clock input pin (CLKinX or CLKinX*) must be AC grounded. The used clock input pin (CLKinX* or CLKinX) may be AC or DC coupled to the signal source. The programming addresses table shows at what register and address the specified CLKinX_BUF_TYPE bit is located. The CLKinX_BUF_TYPE table shows the programming definition for these registers. CLKinX_BUF_TYPE Programming Addresses CLKinX_BUF_TYPE Programming Address CLKin2_BUF_TYPE R14[22] CLKin1_BUF_TYPE R14[21] CLKin0_BUF_TYPE R14[20] CLKinX_BUF_TYPE 80 R14[22, 21, 20] CLKinX Buffer Type 0 Bipolar 1 CMOS Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.12.5 DAC_HIGH_TRIP Voltage from Vcc at which holdover mode is entered if EN_VTUNE_RAIL_DAC is enabled. Will also set flags which can be monitored out Status_LD/Status_Holdover pins. Step size is ~51 mV DAC_HIGH_TRIP, 6 bits R14[19:14] Trip voltage from Vcc (V) 0 (0x00) 1 × Vcc / 64 1 (0x01) 2 × Vcc / 64 2 (0x02) 3 × Vcc / 64 3 (0x03) 4 × Vcc / 64 4 (0x04) 5 × Vcc / 64 ... ... 61 (0x3D) 62 × Vcc / 64 62 (0x3E) 63 × Vcc / 64 63 (0x3F) 64 × Vcc / 64 19.12.6 DAC_LOW_TRIP Voltage from GND at which holdover mode is entered if EN_VTUNE_RAIL_DAC is enabled. Will also set flags which can be monitored out Status_LD/Status_Holdover pins. Step size is ~51 mV DAC_LOW_TRIP, 6 bits R14[11:6] Trip voltage from GND (V) 0 (0x00) 1 × Vcc / 64 1 (0x01) 2 × Vcc / 64 2 (0x02) 3 × Vcc / 64 3 (0x03) 4 × Vcc / 64 4 (0x04) 5 × Vcc / 64 ... ... 61 (0x3D) 62 × Vcc / 64 62 (0x3E) 63 × Vcc / 64 63 (0x3F) 64 × Vcc / 64 19.12.7 EN_VTUNE_RAIL_DET Enables the DAC Vtune rail detection. When the DAC achieves a specified Vtune, if this bit is enabled, the current clock input is considered invalid and an input clock switch event is generated. EN_VTUNE_RAIL_DET R14[5] State 0 Disabled 1 Enabled Copyright © 1999-2012, Texas Instruments Incorporated 81 LMK04906 19.13 REGISTER 15 19.13.1 MAN_DAC Sets the DAC value when in manual DAC mode in ~3.2 mV steps. MAN_DAC, 10 bits R15[31:22] DAC Voltage 0 (0x00) 0 × Vcc / 1023 1 (0x01) 1 × Vcc / 1023 2 (0x02) 2 × Vcc / 1023 ... ... 1023 (0x3FF) 1023 × Vcc / 1023 19.13.2 EN_MAN_DAC This bit enables the manual DAC mode. EN_MAN_DAC R15[20] DAC Mode 0 Automatic 1 Manual 19.13.3 HOLDOVER_DLD_CNT Lock must be valid for this many clocks of PLL1 PDF before holdover mode is exited. HOLDOVER_DLD_CNT, 14 bits R15[19:6] Exit Counts 0 (0x00) Reserved 1 (0x01) 1 2 (0x02) 2 ... ... 16,383 (0x3FFF) 16,383 19.13.4 FORCE_HOLDOVER This bit forces the holdover mode. When holdover is forced, if in fixed CPout1 mode, then the DAC will set the programmed MAN_DAC value. If in tracked CPout1 mode, then the DAC will set the current tracked DAC value. Setting FORCE_HOLDOVER does not constitute a clock input switch event unless DISABLE_DLD1_DET = 0, since in holdover mode, PLL1_DLD = 0 this will trigger the clock input switch event. FORCE_HOLDOVER 82 R15[5] Holdover 0 Disabled 1 Enabled Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.14 REGISTER 16 19.14.1 XTAL_LVL Sets the peak amplitude on the tunable crystal. Increasing this value can improve the crystal oscillator phase noise performance at the cost of increased current and higher crystal power dissipation levels. XTAL_LVL, 2 bits R15[31:22] Peak Amplitude(Note 34) 0 (0x00) 1.65 Vpp 1 (0x01) 1.75 Vpp 2 (0x02) 1.90 Vpp 3 (0x03) 2.05 Vpp Note 34: At crystal frequency of 20.48 MHz 19.15 REGISTER 23 This register must not be programmed, it is a readback only register. 19.15.1 DAC_CNT The DAC_CNT register is 10 bits in size and located at readback bit position [23:14]. When using tracking mode for holdover, the DAC value can be readback at this address. 19.16 REGISTER 24 19.16.1 PLL2_C4_LF, PLL2 Integrated Loop Filter Component Internal loop filter components are available for PLL2, enabling either 3rd or 4th order loop filters without requiring external components. Internal loop filter capacitor C4 can be set according to the following table. PLL2_C4_LF, 4 bits R24[31:28] Loop Filter Capacitance (pF) 0 (0x00) 10 pF 1 (0x01) 15 pF 2 (0x02) 29 pF 3 (0x03) 34 pF 4 (0x04) 47 pF 5 (0x05) 52 pF 6 (0x06) 66 pF 7 (0x07) 71 pF 8 (0x08) 103 pF 9 (0x09) 108 pF 10 (0x0A) 122 pF 11 (0x0B) 126 pF 12 (0x0C) 141 pF 13 (0x0D) 146 pF 14 (0x0E) Reserved 15 (0x0F) Reserved Copyright © 1999-2012, Texas Instruments Incorporated 83 LMK04906 19.16.2 PLL2_C3_LF, PLL2 Integrated Loop Filter Component Internal loop filter components are available for PLL2, enabling either 3rd or 4th order loop filters without requiring external components. Internal loop filter capacitor C3 can be set according to the following table. PLL2_C3_LF, 4 bits R24[27:24] Loop Filter Capacitance (pF) 0 (0x00) 10 pF 1 (0x01) 11 pF 2 (0x02) 15 pF 3 (0x03) 16 pF 4 (0x04) 19 pF 5 (0x05) 20 pF 6 (0x06) 24 pF 7 (0x07) 25 pF 8 (0x08) 29 pF 9 (0x09) 30 pF 10 (0x0A) 33 pF 11 (0x0B) 34 pF 12 (0x0C) 38 pF 13 (0x0D) 39 pF 14 (0x0E) Reserved 15 (0x0F) Reserved 19.16.3 PLL2_R4_LF, PLL2 Integrated Loop Filter Component Internal loop filter components are available for PLL2, enabling either 3rd or 4th order loop filters without requiring external components. Internal loop filter resistor R4 can be set according to the following table. PLL2_R4_LF, 3 bits 84 R24[22:20] Resistance 0 (0x00) 200 Ω 1 (0x01) 1 kΩ 2 (0x02) 2 kΩ 3 (0x03) 4 kΩ 4 (0x04) 16 kΩ 5 (0x05) Reserved 6 (0x06) Reserved 7 (0x07) Reserved Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.16.4 PLL2_R3_LF, PLL2 Integrated Loop Filter Component Internal loop filter components are available for PLL2, enabling either 3rd or 4th order loop filters without requiring external components. Internal loop filter resistor R3 can be set according to the following table. PLL2_R3_LF, 3 bits R24[18:16] Resistance 0 (0x00) 200 Ω 1 (0x01) 1 kΩ 2 (0x02) 2 kΩ 3 (0x03) 4 kΩ 4 (0x04) 16 kΩ 5 (0x05) Reserved 6 (0x06) Reserved 7 (0x07) Reserved 19.16.5 PLL1_N_DLY Increasing delay of PLL1_N_DLY will cause the outputs to lead from CLKinX. For use in 0-delay mode. PLL1_N_DLY, 3 bits R24[14:12] Definition 0 (0x00) 0 ps 1 (0x01) 205 ps 2 (0x02) 410 ps 3 (0x03) 615 ps 4 (0x04) 820 ps 5 (0x05) 1025 ps 6 (0x06) 1230 ps 7 (0x07) 1435 ps 19.16.6 PLL1_R_DLY Increasing delay of PLL1_R_DLY will cause the outputs to lag from CLKinX. For use in 0-delay mode. PLL1_R_DLY, 3 bits R24[10:8] Definition 0 (0x00) 0 ps 1 (0x01) 205 ps 2 (0x02) 410 ps 3 (0x03) 615 ps 4 (0x04) 820 ps 5 (0x05) 1025 ps 6 (0x06) 1230 ps 7 (0x07) 1435 ps 19.16.7 PLL1_WND_SIZE PLL1_WND_SIZE sets the window size used for digital lock detect for PLL1. If the phase error between the reference and feedback of PLL1 is less than specified time, then the PLL1 lock counter increments. Refer to Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY for more information. PLL1_WND_SIZE, 2 bits R24[7:6] Definition 0 5.5 ns 1 10 ns 2 18.6 ns 3 40 ns Copyright © 1999-2012, Texas Instruments Incorporated 85 LMK04906 19.17 REGISTER 25 19.17.1 DAC_CLK_DIV The DAC update clock frequency is the PLL1 phase detector frequency divided by this divisor. DAC_CLK_DIV, 10 bits R25[31:22] Divide 0 (0x00) Reserved 1 (0x01) 1 2 (0x02) 2 3 (0x03) 3 ... ... 1,022 (0x3FE) 1022 1,023 (0x3FF) 1023 19.17.2 PLL1_DLD_CNT The reference and feedback of PLL1 must be within the window of phase error as specified by PLL1_WND_SIZE for this many phase detector cycles before PLL1 digital lock detect is asserted. Refer to Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY for more information. PLL1_DLD_CNT, 14 bits R25[19:6] Divide 0 Reserved 1 1 2 2 3 3 ... ... 16,382 (0x3FFE) 16,382 16,383 (0x3FFF) 16,383 19.18 REGISTER 26 19.18.1 PLL2_WND_SIZE PLL2_WND_SIZE sets the window size used for digital lock detect for PLL2. If the phase error between the reference and feedback of PLL2 is less than specified time, then the PLL2 lock counter increments. This value must be programmed to 2 (3.7 ns). Refer to Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY for more information. PLL2_WND_SIZE, 2 bits R26[31:30] Definition 0 Reserved 1 Reserved 2 3.7 ns 3 Reserved 19.18.2 EN_PLL2_REF_2X, PLL2 Reference Frequency Doubler Enabling the PLL2 reference frequency doubler allows for higher phase detector frequencies on PLL2 than would normally be allowed with the given VCXO or Crystal frequency. Higher phase detector frequencies reduces the PLL N values which makes the design of wider loop bandwidth filters possible. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. EN_PLL2_REF_2X R26[29] Description 0 Reference frequency normal (Note 35) 1 Reference frequency doubled (2x) Note 35: When the doubler is not enabled, PLL2_R should not be programmed to 1. 86 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.18.3 PLL2_CP_POL, PLL2 Charge Pump Polarity PLL2_CP_POL sets the charge pump polarity for PLL2. The internal VCO requires the negative charge pump polarity to be selected. Many VCOs use positive slope. A positive slope VCO increases output frequency with increasing voltage. A negative slope VCO decreases output frequency with increasing voltage. PLL2_CP_POL R26[28] Description 0 Negative Slope VCO/VCXO 1 Positive Slope VCO/VCXO 19.18.4 PLL2_CP_GAIN, PLL2 Charge Pump Current This bit programs the PLL2 charge pump output current level. The table below also illustrates the impact of the PLL2 TRI-STATE bit in conjunction with PLL2_CP_GAIN. PLL2_CP_GAIN, 2 bits R26[27:26] PLL2_CP_TRI R27[5] Charge Pump Current (µA) X 1 Hi-Z 0 (0x00) 0 100 1 (0x01) 0 400 2 (0x02) 0 1600 3 (0x03) 0 3200 19.18.5 PLL2_DLD_CNT The reference and feedback of PLL2 must be within the window of phase error as specified by PLL2_WND_SIZE for PLL2_DLD_CNT cycles before PLL2 digital lock detect is asserted. Refer to Section 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY for more information PLL2_DLD_CNT, 14 bits R26[19:6] Divide 0 (0x00) Reserved 1 (0x01) 1 2 (0x02) 2 3 (0x03) 3 ... ... 16,382 (0x3FFE) 16,382 16,383 (0x3FFF) 16,383 19.18.6 PLL2_CP_TRI, PLL2 Charge Pump TRI-STATE This bit allows for the PLL2 charge pump output pin, CPout2, to be placed into TRI-STATE. PLL2_CP_TRI R26[5] Description 0 PLL2 CPout2 is active 1 PLL2 CPout2 is at TRI-STATE 19.19 REGISTER 27 19.19.1 PLL1_CP_POL, PLL1 Charge Pump Polarity PLL1_CP_POL sets the charge pump polarity for PLL1. Many VCXOs use positive slope. A positive slope VCXO increases output frequency with increasing voltage. A negative slope VCXO decreases output frequency with increasing voltage. PLL1_CP_POL R27[28] Description 0 Negative Slope VCO/VCXO 1 Positive Slope VCO/VCXO Copyright © 1999-2012, Texas Instruments Incorporated 87 LMK04906 19.19.2 PLL1_CP_GAIN, PLL1 Charge Pump Current This bit programs the PLL1 charge pump output current level. The table below also illustrates the impact of the PLL1 TRI-STATE bit in conjunction with PLL1_CP_GAIN. PLL1_CP_GAIN, 2 bits R26[27:26] PLL1_CP_TRI R27[5] Charge Pump Current (µA) X 1 Hi-Z 0 (0x00) 0 100 1 (0x01) 0 200 2 (0x02) 0 400 3 (0x03) 0 1600 19.19.3 CLKinX_PreR_DIV The pre-R dividers before the PLL1 R divider can be programmed such that when the active clock input is switched, the frequency at the input of the PLL1 R divider will be the same. This allows PLL1 to stay in lock without needing to re-program the PLL1 R register when different clock input frequencies are used. This is especially useful in the auto CLKin switching modes. CLKinX_PreR_DIV Programming Addresses CLKinX_PreR_DIV Programming Address CLKin2_PreR_DIV R27[25:24] CLKin1_PreR_DIV R27[23:22] CLKin0_PreR_DIV R27[21:20] CLKinX_PreR_DIV, 2 bits R27[25:24, 23:22, 21:20] Divide 0 (0x00) 1 1 (0x01) 2 2 (0x02) 4 3 (0x03) 8 19.19.4 PLL1_R, PLL1 R Divider The reference path into the PLL1 phase detector includes the PLL1 R divider. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. The valid values for PLL1_R are shown in the table below. PLL1_R, 14 bits R27[19:6] Divide 0 (0x00) Reserved 1 (0x01) 1 2 (0x02) 2 3 (0x03) 3 ... ... 16,382 (0x3FFE) 16,382 16,383 (0x3FFF) 16,383 19.19.5 PLL1_CP_TRI, PLL1 Charge Pump TRI-STATE This bit allows for the PLL1 charge pump output pin, CPout1, to be placed into TRI-STATE. PLL1_CP_TRI R27[5] Description 0 PLL1 CPout1 is active 1 PLL1 CPout1 is at TRI-STATE 19.20 REGISTER 28 19.20.1 PLL2_R, PLL2 R Divider The reference path into the PLL2 phase detector includes the PLL2 R divider. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. 88 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 The valid values for PLL2_R are shown in the table below. PLL2_R, 12 bits R28[31:20] Divide 0 (0x00) Not Valid 1 (0x01) 1 (Note 36) 2 (0x02) 2 3 (0x03) 3 ... ... 4,094 (0xFFE) 4,094 4,095 (0xFFF) 4,095 Note 36: When using PLL2_R divide value of 1, the PLL2 reference doubler should be used (EN_PLL2_REF_2X = 1). 19.20.2 PLL1_N, PLL1 N Divider The feedback path into the PLL1 phase detector includes the PLL1 N divider. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. The valid values for PLL1_N are shown in the table below. PLL1_N, 14 bits R28[19:6] Divide 0 (0x00) Not Valid 1 (0x01) 1 2 (0x02) 2 ... ... 4,095 (0xFFF) 4,095 Copyright © 1999-2012, Texas Instruments Incorporated 89 LMK04906 19.21 REGISTER 29 19.21.1 OSCin_FREQ, PLL2 Oscillator Input Frequency Register The frequency of the PLL2 reference input to the PLL2 Phase Detector (OSCin/OSCin* port) must be programmed in order to support proper operation of the frequency calibration routine which locks the internal VCO to the target frequency. OSCin_FREQ, 3 bits R29[26:24] OSCin Frequency 0 (0x00) 0 to 63 MHz 1 (0x01) >63 MHz to 127 MHz 2 (0x02) >127 MHz to 255 MHz 3 (0x03) Reserved 4 (0x04) >255 MHz to 400 MHz 19.21.2 PLL2_FAST_PDF, High PLL2 Phase Detector Frequency When PLL2 phase detector frequency is greater than 100 MHz, set the PLL2_FAST_PDF to ensure proper operation of device. PLL2_FAST_PDF R29[23] PLL2 PDF 0 Less than or equal to 100 MHz 1 Greater than 100 MHz 19.21.3 PLL2_N_CAL, PLL2 N Calibration Divider During the frequency calibration routine, the PLL uses the divide value of the PLL2_N_CAL register instead of the divide value of the PLL2_N register to lock the VCO to the target frequency. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. PLL2_N_CAL, 18 bits 90 R30[22:5] Divide 0 (0x00) Not Valid 1 (0x01) 1 2 (0x02) 2 ... ... 262,143 (0x3FFFF) 262,143 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 19.22 REGISTER 30 If an internal VCO mode is used, programming Register 30 triggers the frequency calibration routine. This calibration routine will also generate a SYNC event. See Section 18.9.2 Clock Output Synchronization (SYNC) for more details on a SYNC. 19.22.1 PLL2_P, PLL2 N Prescaler Divider The PLL2 N Prescaler divides the output of the VCO as selected by Mode_MUX1 and is connected to the PLL2 N divider. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. PLL2_P, 3 bits R30[26:24] Divide Value 0 (0x00) 8 1 (0x01) 2 2 (0x02) 2 3 (0x03) 3 4 (0x04) 4 5 (0x05) 5 6 (0x06) 6 7 (0x07) 7 19.22.2 PLL2_N, PLL2 N Divider The feeback path into the PLL2 phase detector includes the PLL2 N divider. Each time register 30 is updated via the MICROWIRE interface, a frequency calibration routine runs to lock the VCO to the target frequency. During this calibration PLL2_N is substituted with PLL2_N_CAL. Refer to Section 20.2 PLL PROGRAMMING for more information on how to program the PLL dividers to lock the PLL. The valid values for PLL2_N are shown in the table below. PLL2_N, 18 bits R30[22:5] Divide 0 (0x00) Not Valid 1 (0x01) 1 2 (0x02) 2 ... 262,143 (0x3FFFF) 262,143 19.23 REGISTER 31 19.23.1 READBACK_LE Sets the required state of the LEuWire pin when performing register readback. Refer to Section 19.3 READBACK READBACK_LE R31[21] Register 0 (0x00) LE must be low for readback 1 (0x01) LE must be high for readback Copyright © 1999-2012, Texas Instruments Incorporated 91 LMK04906 19.23.2 READBACK_ADDR Sets the address of the register to read back when performing readback. When reading register 12, the READBACK_ADDR will be read back at R12[20:16]. When reading back from R31 bits 6 to 31 should be ignored. Only uWire_LOCK is valid. Refer to Section 17.14 REGISTER READBACK for more information on readback. READBACK_ADDR, 5 bits R31[20:16] Register 0 (0x00) R0 1 (0x01) R1 2 (0x02) R2 3 (0x03) R3 4 (0x04) R4 5 (0x05) R5 6 (0x06) R6 7 (0x07) R7 8 (0x08) R8 9 (0x09) Reserved 10 (0x0A) R10 11 (0x0B) R11 12 (0x0C) R12 13 (0x0D) R13 14 (0x0E) R14 15 (0x0F) R15 16 (0x10) Reserved 17 (0x11) Reserved ... ... 22 (0x16) Reserved 23 (0x17) Reserved 24 (0x18) R24 25 (0x19) R25 26 (0x1A) R26 27 (0x1B) R27 28 (0x1C) R28 29 (0x1D) R29 30 (0x1E) R30 31 (0x1F) R31 19.23.3 uWire_LOCK Setting uWire_LOCK will prevent any changes to uWire registers R0 to R30. Only by clearing the uWire_LOCK bit in R31 can the uWire registers be unlocked and written to once more. It is not necessary to lock the registers to perform a readback operation. uWire_LOCK 92 R31[5] State 0 Registers unlocked 1 Registers locked, Write-protect Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 20.0 Application Information 20.1 FREQUENCY PLANNING WITH THE LMK04906 FAMILY (Note 37) Calculating the value of the output dividers for use with the LMK04906 family is simple due to the architecture of the LMK04906. That is, the VCO divider may be bypassed and the clock output dividers allow for even and odd output divide values from 2 to 1045. For most applications it is recommended to bypass the VCO divider. The procedure for determining the needed LMK04906 device and clock output divider values for a set of clock output frequencies is straightforward. 1. Calculate the least common multiple (LCM) of the clock output frequencies. 2. Determine which VCO ranges will support the target clock output frequencies given the LCM. 3. Determine the clock output divide values based on VCO frequency. 4. Determine the PLL2 reference frequency doubler mode and PLL2_P, PLL2_N, and PLL2_R divider values given the OSCin VCXO or crystal frequency and VCO frequency. For example, given the following target output frequencies: 200 MHz, 120 MHz, and 25 MHz with a VCXO frequency of 40 MHz: First determine the LCM of the three frequencies. LCM(200 MHz, 120 MHz, 25 MHz) = 600 MHz. The LCM frequency is the lowest frequency for which all of the target output frequencies are integer divisors of the LCM. Note, if there is one frequency which causes the LCM to be very large, greater than 3 GHz for example, determine if there is a single frequency requirement which causes this. It may be possible to select the VCXO/crystal frequency to satisfy this frequency requirement through OSCout or CLKout3/4 driven by OSCin. In this way it is possible to get non-integer related frequencies at the outputs. Second, since the LCM is not in a VCO frequency range supported by the LMK04906, multiply the LCM frequency by an integer which causes it to fall into a valid VCO frequency range of an LMK04906 device. In this case 600 MHz * 4 = 2400 MHz which is valid for the LMK04906. Third, continuing the example by using a VCO frequency of 2400 MHz and the LMK04906, the CLKout dividers can be calculated by simply dividing the VCO frequency by the output frequency. To output 200 MHz, 120 MHz, and 25 MHz the output dividers will be 12, 20, and 96 respectively. • 2400 MHz / 200 MHz = 12 • 2400 MHz / 120 MHz = 20 • 2400 MHz / 25 MHz = 96 Fourth, PLL2 must be locked to its input reference. Refer to Section 20.2 PLL PROGRAMMING for more information on this topic. By programming the clock output dividers and the PLL2 dividers the VCO can lock to the frequency of 2400 MHz and the clock outputs dividers will each divide the VCO frequency down to the target output frequencies of 200 MHz, 120 MHz, and 25 MHz. Note 37: Refer to application note AN-1865 Frequency Synthesis and Planning for PLL Architectures for more information on this topic and LCM calculations. Copyright © 1999-2012, Texas Instruments Incorporated 93 LMK04906 20.2 PLL PROGRAMMING To lock a PLL the divided reference and divided feedback from VCO or VCXO must result in the same phase detector frequency. The tables below illustrate how the divides are structured for the reference path (R) and feedback path (N) depending on the MODE of the device. PLL1 Phase Detector Frequency — Reference Path (R) MODE (R) PLL1 PDF = All CLKinX Frequency / CLKinX_PreR_DIV / PLL1_R PLL1 Phase Detector Frequency — Feedback Path (N) MODE Internal VCO Dual PLL Internal VCO /w 0-delay VCO_MUX OSCout0 — Bypass PLL1 PDF (N) = VCXO Frequency / PLL1_N — Divided VCXO Frequency / OSCin_DIV / PLL1_N Bypass — VCO Frequency / CLKoutX_DIV / PLL1_N (Note 38) Divided — VCO Frequency / VCO_DIV / CLKoutX_DIV / PLL1_N (Note 38) Note 38: The actual CLKoutX_DIV used is selected by Section 19.8.9 FEEDBACK_MUX. PLL2 Phase Detector Frequency — Reference Path (R) EN_PLL2_REF_2X PLL2 PDF (R) = Disabled OSCin Frequency / PLL2_R (Note 39) Enabled OSCin Frequency * 2 / PLL2_R (Note 39) Note 39: For applications in which the OSCin frequency and PLL2 phase detector frequency are equal, the best PLL2 in-band noise can be achieved when the doubler is enabled (EN_PLL2_REF_2X = 1) and the PLL2 R divide value is 2. Do not use doubler disabled (EN_PLL2_REF_2X = 0) and PLL2 R divide value of 1. PLL2 Phase Detector Frequency — Feedback Path (N) MODE VCO_MUX PLL2 PDF (N) = Dual PLL Dual PLL /w 0-delay VCO VCO Frequency / PLL2_P / PLL2_N Single PLL Dual PLL Dual PLL /w 0-delay VCO Divider VCO Frequency / VCO_DIV / PLL2_P / PLL2_N Single PLL Dual PLL External VCO Dual PLL External VCO /w 0-delay Single PLL /w 0-delay — VCO VCO Frequency / VCO_DIV / PLL2_P / PLL2_N VCO Frequency / CLKoutX_DIV / PLL2_N VCO Divider VCO Frequency / VCO_DIV / CLKoutX_DIV / PLL2_N PLL2 Phase Detector Frequency — Feedback Path (N) during VCO Frequency Calibration MODE All Internal VCO Modes VCO_MUX PLL2 PDF (N_CAL) = VCO VCO Frequency / PLL2_P / PLL2_N_CAL VCO Divider VCO Frequency / VCO_DIV / PLL2_P / PLL2_N_CAL 20.2.1 Example PLL2 N Divider Programming To program PLL2 to lock an LMK04906 using Dual PLL mode to a VCO frequency of 2400 MHz using a 40 MHz VCXO reference, first determine the total PLL2 N divide value. This is VCO Frequency / PLL2 phase detector frequency. This example assumes the PLL2 reference frequency doubler is enabled and a PLL2 R divider value of 2 (Note 39) which results in PLL2 R divide value of 1 which results in PLL2 phase detector frequency the same as PLL2 reference frequency (40 MHz). 2400 MHz / 40 MHz = 60, so the total PLL2 N divide value is 60. The dividers in the PLL2 N feedback path for Dual PLL mode include PLL2_P and PLL2_N. PLL2_P can be programmed from 2 to 8 even and odd. PLL2_N can be programmed from 1 to 263,143 even and odd. Since the total PLL2 N divide value of 60 contains 94 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 the factors 2, 2, 3, and 5, it would be allowable to program PLL2_P to 2, 3 or 5. It is simplest to use the smallest divide, so PLL2_P = 2, and PLL2_N = 30 which results in a Total PLL2 N = 60. For this example and in most cases, PLL2_N_CAL will have the same value as PLL2_N. However when using Single PLL mode with 0-delay, the values will differ. When using an external VCO, PLL2_N_CAL value is unused. 20.3 LOOP FILTER Each PLL of the LMK04906 requires a dedicated loop filter. 20.3.1 PLL1 The loop filter for PLL1 must be connected to the CPout1 pin. Figure 18shows a simple 2-pole loop filter. The output of the filter drives an external VCXO module or discrete implementation of a VCXO using a crystal resonator and external varactor diode. Higher order loop filters may be implemented using additional external R and C components. It is recommended the loop filter for PLL1 result in a total closed loop bandwidth in the range of 10 Hz to 200 Hz. The design of the loop filter is application specific and highly dependent on parameters such as the phase noise of the reference clock, VCXO phase noise, and phase detector frequency for PLL1. TI's Clock Conditioner Owner’s Manual covers this topic in detail and TI's Clock Design Tool can be used to simulate loop filter designs for both PLLs. These resources may be found: http://www.ti.com/lsds/ti/analog/clocksandtimers/ clocks_and_timers.page. 20.3.2 PLL2 As shown in Figure 18, the charge pump for PLL2 is directly connected to the optional internal loop filter components, which are normally used only if either a third or fourth pole is needed. The first and second poles are implemented with external components. The loop must be designed to be stable over the entire application-specific tuning range of the VCO. The designer should note the range of KVCO listed in the table of Electrical Characteristics and how this value can change over the expected range of VCO tuning frequencies. Because loop bandwidth is directly proportional to KVCO, the designer should model and simulate the loop at the expected extremes of the desired tuning range, using the appropriate values for KVCO. When designing with the integrated loop filter of the LMK04906 family, considerations for minimum resistor thermal noise often lead one to the decision to design for the minimum value for integrated resistors, R3 and R4. Both the integrated loop filter resistors (R3 and R4) and capacitors (C3 and C4) also restrict the maximum loop bandwidth. However, these integrated components do have the advantage that they are closer to the VCO and can therefore filter out some noise and spurs better than external components. For this reason, a common strategy is to minimize the internal loop filter resistors and then design for the largest internal capacitor values that permit a wide enough loop bandwidth. In situations where spur requirements are very stringent and there is margin on phase noise, a feasible strategy would be to design a loop filter with integrated resistor values larger than their minimum value. 30179171 FIGURE 18. PLL1 and PLL2 Loop Filters Copyright © 1999-2012, Texas Instruments Incorporated 95 LMK04906 20.4 SYSTEM LEVEL DIAGRAM Figure 19 and Figure 20 show an LMK04906 family device with external circuitry for clocking and for power supply to serve as a guideline for good practices when designing with the LMK04906 family. Refer to Section 20.5 PIN CONNECTION RECOMMENDATIONS for more details on the pin connections and bypassing recommendations. Also refer to the evaluation board. PCB design will also play a role in device performance. 30179178 FIGURE 19. Example Application – System Schematic Except for Power Figure 19 shows the primary reference clock input is at CLKin0/0*. A secondary reference clock is driving CLKin1/1*. Both clocks are depicted as AC coupled differential drivers. The VCXO attached to the OSCin/OSCin* port is configured as an AC coupled single-ended driver. Any of the input ports (CLKin0/0*, CLKin1/1*, CLKin2/2*or OSCin/OSCin*) may be configured as either differential or single-ended. These options are discussed later in the data sheet. See for more information on PLL1 and PLL2 loop filters. The clock outputs are all AC coupled with 0.1 µF capacitors. Some clock outputs are depicted as LVPECL with 240 Ω emitter resistors and some clock outputs as LVDS. However, the output format of the clock outputs will vary by user programming, so the user should use the appropriate source termination for each clock output. Later sections of this data sheet illustrate alternative methods for AC coupling, DC coupling and terminating the clock outputs. 96 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 PCB design will influence crosstalk performance. Tightly coupled clock traces will have less crosstalk than loosely coupled clock traces. Also proximity to other clocks traces will influence crosstalk. 30179177 FIGURE 20. Example Application – Power System Schematic Figure 20 shows an example decouping and bypassing scheme for the LMK04906. Components drawn in dotted lines are optional. Two power planes are used in this design, one for the clock outputs and one for other PLL circuits. PCB design will influence impedance to the supply. Vias and traces will increase the impedance to the power supply. Ensure good direct return current paths. Copyright © 1999-2012, Texas Instruments Incorporated 97 LMK04906 20.5 PIN CONNECTION RECOMMENDATIONS 20.5.1 Vcc Pins and Decoupling All Vcc pins must always be connected. Integrated capacitance on the LMK04906 makes external high frequency decoupling capacitors (≤ 1 nF) unnecessary. The internal capacitance is more effective at filtering high frequency noise than off device bypass capacitance because there is no bond wire inductance between the LMK04906 circuit and the bypass capacitor. 20.5.1.1 Vcc2, Vcc3, Vcc10, Vcc11, Vcc12, Vcc13 (CLKout Vccs) Each of these pins has an internal 200 pF of capacitance. Ferrite beads may be used to reduce crosstalk between different clock output frequencies on the same LMK04906 device. Ferrite beads placed between the power supply and a clock Vcc pin will reduce noise between the Vcc pin and the power supply. When several output clocks share the same frequency a single ferrite bead can be used between the power supply and each same frequency CLKout Vcc pin. When using ferrite beads on CLKout Vcc pins, care must be taken to ensure the power supply can source the needed switching current. • In most cases a ferrite bead may be placed and the internal capacitance is sufficient. • If a ferrite bead is used with a low frequency output (typically ≤ 10 MHz) and a high current switching clock output format such as non-complementary LVCMOS or high swing LVPECL is used, then... — the ferrite bead can be removed to the lower impedance to the main power supply and bypass capacitors, or — localized capacitance can be placed between the ferrite bead and Vcc pin to support the switching current. Note that decoupling capacitors used between the ferrite bead and a CLKout Vcc pin can permit high frequency switching noise to couple through the capacitors into the ground plane and onto other CLKout Vcc pins with decoupling capacitors. This can degrade crosstalk performance. 20.5.1.2 Vcc1 (VCO), Vcc4 (Digital), and Vcc9 (PLL2) Each of these pins has internal bypass capacitance. Ferrite beads should not be used between these pins and the power supply/large bypass capacitors because these Vcc pins don’t produce much noise or a ferrite bead can cause phase noise disturbances and resonances. The typical application diagram in Figure 20 shows all these Vccs connected to together to Vcc without a ferrite bead. 20.5.1.3 Vcc6 (PLL1 Charge Pump) and Vcc8 (PLL2 Charge Pump) Each of these pins has an internal bypass capacitor. Use of a ferrite bead between the power supply/large bypass capacitors and PLL1 is optional. PLL1 charge pump can be connected directly to Vcc along with Vcc1, Vcc4, and Vcc9. Depending on the application, a 0.1 uF capacitor may be placed close to PLL1 charge pump Vcc pin. A ferrite bead should be placed between the power supply/large bypass capacitors and Vcc8. Most applications have high PLL2 phase detector frequencies and (> 50 MHz) such that the internal bypassing is sufficient and a ferrite bead can be used to isolate this switching noise from other circuits. For lower phase detector frequencies a ferrite bead is optional and depending on application a 0.1 uF capacitor may be added on Vcc8. 20.5.1.4 Vcc5 (CLKin), Vcc7 (OSCin & OSCout0) Each of these pins has an internal 100 pF of capacitance. No ferrite bead should be placed between the power supply/large bypass capacitors and Vcc5 or Vcc7. These pins are unique since they supply an output clock and other circuitry. Vcc5 supplies CLKin. Vcc7 supplies OSCin, OSCout0, and PLL2 circuitry. 20.5.2 LVPECL Outputs When using an LVPECL output it is not recommended to place a capacitor to ground on the output as might be done when using a capacitor input LC lowpass filter. The capacitor will appear as a short to the LVPECL output drivers which are able to supply large amounts of switching current. The effect of the LVPECL sourcing large switching currents can result in: 1. Large switching currents through the Vcc pin of the LVPECL power supply resulting in more Vcc noise and possible Vcc spikes. 2. Large switching currents injected into the ground plane through the capacitor which could couple onto other Vcc pins with bypass capacitors to ground resulting in more Vcc noise and possible Vcc spikes. 20.5.3 Unused Clock Outputs Leave unused clock outputs floating and powered down. 20.5.4 Unused Clock Inputs Unused clock inputs can be left floating. 20.5.5 LDO Bypass The LDObyp1 and LDObyp2 pins should be connected to GND through external capacitors, as shown in the diagram. 98 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 20.6 DIGITAL LOCK DETECT FREQUENCY ACCURACY The digital lock detect circuit is used to determine PLL1 locked, PLL2 locked, and holdover exit events. A window size and lock count register are programmed to set a ppm frequency accuracy of reference to feedback signals of the PLL for each event to occur. When a PLL digital lock event occurs the PLL's digital lock detect is asserted true. When the holdover exit event occurs, the device will exit holdover mode. TABLE 17. Event PLL Window size Lock count PLL1 Locked PLL1 PLL1_WND_SIZE PLL1_DLD_CNT PLL2 Locked PLL2 PLL2_WND_SIZE PLL2_DLD_CNT Holdover PLL1 exit PLL1_WND_SIZE HOLDOVER_DLD_CNT For a digital lock detect event to occur there must be a “lock count” number of phase detector cycles of PLLX during which the time/phase error of the PLLX_R reference and PLLX_N feedback signal edges are within the user programmable "window size." Since there must be at least "lock count" phase detector events before a lock event occurs, a minimum digital lock event time can be calculated as "lock count" / fPDX where X = 1 for PLL1 or 2 for PLL2. By using Equation 3, values for a "lock count" and "window size" can be chosen to set the frequency accuracy required by the system in ppm before the digital lock detect event occurs: (3) The effect of the "lock count" value is that it shortens the effective lock window size by dividing the "window size" by "lock count". If at any time the PLLX_R reference and PLLX_N feedback signals are outside the time window set by "window size", then the “lock count” value is reset to 0. 20.6.1 Minimum Lock Time Calculation Example To calculate the minimum PLL2 digital lock time given a PLL2 phase detector frequency of 40 MHz and PLL2_DLD_CNT = 10,000. Then the minimum lock time of PLL2 will be 10,000 / 40 MHz = 250 µs. Copyright © 1999-2012, Texas Instruments Incorporated 99 LMK04906 20.7 CALCULATING DYNAMIC DIGITAL DELAY VALUES FOR ANY DIVIDE This section explains how to calculate the dynamic digital delay for any divide value. Dynamic digital delay allows the time offset between two or more clock outputs to be adjusted with no or minimal interruption of clock outputs. Since the clock outputs are operating at a known frequency, the time offset can also be expressed as a phase shift. When dynamically adjusting the digital delay of clock outputs with different frequencies the phase shift should be expressed in terms of the higher frequency clock. The step size of the smallest time adjustment possible is equal to half the period of the Clock Distribution Path, which is the VCO frequency (Equation 1) or the VCO frequency divided by the VCO divider (Equation 2) if not bypassed. The smallest degree phase adjustment with respect to a clock frequency will be 360 * the smallest time adjustment * the clock frequency. The total number of phase offsets that the LMK04906 family is able to achieve using dynamic digital delay is equal 1 / (higher clock frequency * the smallest phase adjustment). Equation 4 calculates the digital delay value that must be programmed for a synchronizing clock to achieve a 0 time/phase offset from the qualifying clock. Once this digital delay value is known, it is possible to calculate the digital delay value for any phase offset. The qualifying clock for dynamic digital delay is selected by the FEEDBACK_MUX. When dynamic digital delay is engaged with same clock output used for the qualifying clock and the new synchronized clock, it is termed relative dynamic digital delay since causing another SYNC event with the same digital delay value will offset the clock by the same phase once again. The important part of relative dynamic digital delay is that the CLKoutX_HS must be programmed correctly when the SYNC event occurs (Table 8). This can result in needing to program the device twice. Once to set the new CLKoutX_DDLY with CLKoutX_HS as required for the SYNC event, and again to set the CLKoutX_HS to its desired value. Digital delay values are programmed using the CLKoutX_DDLY and CLKoutX_HS registers as shown in Equation 5. For example, to achieve a digital delay of 13.5, program CLKoutX_DDLY = 14 and CLKoutX_HS = 1. (4) Equation 4 uses the ceiling operator. To find the ceiling of a fractional number round up. An integer remains the same value. Digital delay = CLKoutX_DDLY - (0.5 * CLKoutX_HS) (5) Note: since the digital delay value for 0 time/phase offset is a function of the qualifying clock's divide value, the resulting digital delay value can be used for any clock output operating at any frequency to achieve a 0 time/phase offset from the qualifying clock. Therefore the calculated time shift table will also be the same as in Table 18 20.7.1 Example Consider a system with: • A VCO frequency of 2000 MHz. • The VCO divider is bypassed, therefore the clock distribution path frequency is 2000 MHz. • CLKout0_DIV = 10 resulting in a 200 MHz frequency on CLKout0. • CLKout2_DIV = 20 resulting in a 100 MHz frequency on CLKout2. For this system the minimum time adjustment is 0.25 ns, which is 0.5 / (2000 MHz). Since the higher frequency is 200 MHz, phase adjustments will be calculated with respect to the 200 MHz frequency. The 0.25 ns minimum time adjustment results in a minimum phase adjustment of 18 degrees, which is 360 degrees / 200 MHz * 0.25 ns. To calculate the digital delay value to achieve a 0 time/phase shift of CLKout2 when CLKout0 is the qualifying clock. Solve Equation 4 using the divide value of 10. To solve the equation 16/10 = 1.6, the ceiling of 1.6 is 2. Then to finish solving the equation solve (2 + 0.5) * 10 - 11.5 = 13.5. A digital delay value of 13.5 is programmed by setting CLKout2_DDLY = 14 and CLKout2_HS = 1. To calculate the digital delay value to achieve a 0 time/phase shift of CLKout0 when CLKout2 is the qualifying clock, solve Equation 4 using the divide value of CLKout2, which is 20. This results in a digital delay of 18.5 which is programmed as CLKout0_DDLY = 19 and CLKout0_HS = 1. Once the 0 time/phase shift digital delay programming value is known a table can be constructed with the digital delay value to be programmed for any time/phase offset by decrementing or incrementing the digital delay value by 0.5 for the minimum time/phase adjustment. A complete filled out table for use of CLKout0 as the qualifying clock is shown in Table 18. It was created by entering a digital delay of 13.5 for 0 degree phase shift, then decrementing the digital delay down to the minimum value of 4.5. Since this did not result in all the possible phase shifts, the digital delay was then incremented from 13.5 to 14.0 to complete all possible phase shifts. 100 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 TABLE 18. Example Digital Delay Calculation Digital delay Calculated time shift (ns) Relative time shift to 200 MHz (ns) Phase shift of 200 MHz (degrees) 4.5 -4.5 0.5 36 5 -4.25 0.75 54 5.5 -4.0 1.0 72 6 -3.75 1.25 90 6.5 -3.5 1.5 108 7 -3.25 1.75 126 7.5 -3.0 2.0 144 8 -2.75 2.25 162 8.5 -2.5 2.5 180 9 -2.25 2.75 198 9.5 -2.0 3.0 216 10 -1.75 3.25 234 10.5 -1.5 3.5 252 11 -1.25 3.75 270 11.5 -1.0 4.0 288 12 -0.75 4.25 306 12.5 -0.5 4.5 324 342 13 -0.25 4.75 13.5 0 0 0 14 0.25 0.25 18 14.5 0.5 0.5 36 Observe that the digital delay value of 4.5 and 14.5 will achieve the same relative time shift/phase delay. However programming a digital delay of 14.5 will result in a clock off time for the synchronizing clock to achieve the same phase time shift/phase delay. Digital delay value is programmed as CLKoutX_DDLY - (0.5 * CLKoutX_HS). So to achieve a digital delay of 13.5, program CLKoutX_DDLY = 14 and CLKoutX_HS = 1. To achieve a digital delay of 14, program CLKoutX_DDLY = 14 and CLKoutX_HS = 0. Copyright © 1999-2012, Texas Instruments Incorporated 101 LMK04906 20.8 OPTIONAL CRYSTAL OSCILLATOR IMPLEMENTATION (OSCin/OSCin*) The LMK04906 family features supporting circuitry for a discretely implemented oscillator driving the OSCin port pins. Figure 21 illustrates a reference design circuit for a crystal oscillator: 30179163 FIGURE 21. Reference Design Circuit for Crystal Oscillator Option This circuit topology represents a parallel resonant mode oscillator design. When selecting a crystal for parallel resonance, the total load capacitance, CL, must be specified. The load capacitance is the sum of the tuning capacitance (CTUNE), the capacitance seen looking into the OSCin port (CIN), and stray capacitance due to PCB parasitics (CSTRAY), and is given by Equation 6. (6) CTUNE is provided by the varactor diode shown in Figure 21, Skyworks model SMV1249-074LF. A dual diode package with common cathode provides the variable capacitance for tuning. The single diode capacitance ranges from approximately 31 pF at 0.3 V to 3.4 pF at 3 V. The capacitance range of the dual package (anode to anode) is approximately 15.5 pF at 3 V to 1.7 pF at 0.3 V. The desired value of VTUNE applied to the diode should be VCC/2, or 1.65 V for VCC = 3.3 V. The typical performance curve from the data sheet for the SMV1249-074LF indicates that the capacitance at this voltage is approximately 6 pF (12 pF / 2). The nominal input capacitance (CIN) of the LMK04906 family OSCin pins is 6 pF. The stray capacitance (CSTRAY) of the PCB should be minimized by arranging the oscillator circuit layout to achieve trace lengths as short as possible and as narrow as possible trace width (50 Ω characteristic impedance is not required). As an example, assume that CSTRAY is 4 pF. The total load capacitance is nominally: 102 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 (7) Consequently the load capacitance specification for the crystal in this case should be nominally 14 pF. The 2.2 nF capacitors shown in the circuit are coupling capacitors that block the DC tuning voltage applied by the 4.7 kΩ and 10 kΩ resistors. The value of these coupling capacitors should be large, relative to the value of CTUNE (CC1 = CC2 >> CTUNE), so that CTUNE becomes the dominant capacitance. For a specific value of CL, the corresponding resonant frequency (FL) of the parallel resonant mode circuit is: (8) FS = Series resonant frequency C1 = Motional capacitance of the crystal CL = Load capacitance C0 = Shunt capacitance of the crystal, specified on the crystal datasheet The normalized tuning range of the circuit is closely approximated by: (9) CL1, C L2 = The endpoints of the circuit’s load capacitance range, assuming a variable capacitance element is one component of the load. FCL1, FCL2 = parallel resonant frequencies at the extremes of the circuit’s load capacitance range. A common range for the pullability ratio, C0/C1, is 250 to 280. The ratio of the load capacitance to the shunt capacitance is ~(n * 1000), n < 10. Hence, picking a crystal with a smaller pullability ratio supports a wider tuning range because this allows the scale factors related to the load capacitance to dominate. Examples of the phase noise and jitter performance of the LMK04906 with a crystal oscillator are shown in Table 19. This table illustrates the clock output phase noise when a 20.48 MHz crystal is paired with PLL1. TABLE 19. Example RMS Jitter and Clock Output Phase Noise for LMK04906 with a 20.48 MHz Crystal Driving OSCin (T = 25 °C, VCC = 3.3 V) (Note 40) RMS Jitter (ps) Integration Bandwidth 100 Hz – 20 MHz 10 kHz – 20 MHz Clock Output Type PLL2 PDF = 20.48 MHz (EN_PLL2_REF2X = 0, XTAL_LVL = 3) PLL2 PDF = 40.96 MHz (EN_PLL2_REF2X = 1, XTAL_LVL = 3) fCLK = 245.76 MHz fCLK = 122.88 MHz fCLK = 245.76 MHz LVCMOS 374 412 382 LVDS 419 421 372 LVPECL 1.6 Vpp 460 448 440 LVCMOS 226 195 190 LVDS 231 205 194 LVPECL 1.6 Vpp 226 191 188 Phase Noise (dBc/Hz) Offset 100 Hz 1 kHz Clock Output Type PLL2 PDF = 20.48 MHz (EN_PLL2_REF2X = 0, XTAL_LVL = 3) PLL2 PDF = 40.96 MHz (EN_PLL2_REF2X = 1, XTAL_LVL = 3) fCLK = 245.76 MHz fCLK = 122.88 MHz fCLK = 245.76 MHz LVCMOS -87 -93 -87 -86 LVDS -86 -91 LVPECL 1.6 Vpp -86 -92 -85 LVCMOS -115 -121 -115 LVDS -115 -123 -116 LVPECL 1.6 Vpp -114 -122 -116 Copyright © 1999-2012, Texas Instruments Incorporated 103 LMK04906 10 kHz 100 kHz 1 MHz 40 MHz LVCMOS -117 -128 -122 LVDS -117 -128 -122 LVPECL 1.6 Vpp -117 -128 -122 LVCMOS -130 -135 -129 LVDS -130 -135 -129 LVPECL 1.6 Vpp -129 -135 -129 LVCMOS -150 -154 -148 LVDS -149 -153 -148 LVPECL 1.6 Vpp -150 -154 -148 LVCMOS -159 -162 -159 LVDS -157 -159 -157 LVPECL 1.6 Vpp -159 -161 -159 Note 40: Performance data and crystal specifications contained in this section are based on Vectron model VXB1-1150-20M480, 20.48 MHz. PLL1 has a narrow loop bandwidth, PLL2 loop parameters are: C1 = 150 pF, C2 = 120 nF, R2 = 470 Ω, Charge Pump current = 3.2 mA, Phase detector frequency = 20.48 MHz or 40.96 MHz, VCO frequency = 2949.12 MHz. Loop filter was optimized for 40.96 MHz phase detector performance. Example crystal specifications are presented in Table 20. TABLE 20. Example Crystal Specifications Parameter Value Nominal Frequency (MHz) 20.48 Frequency Stability, T = 25 °C ± 10 ppm Operating temperature range -40 °C to +85 °C Frequency Stability, -40 °C to +85 °C ± 15 ppm Load Capacitance 14 pF Shunt Capacitance (C0) 5 pF Maximum Motional Capacitance (C1) 20 fF ± 30% Equivalent Series Resistance 25 Ω Maximum Drive level 2 mWatts Maximum C0/C1 ratio 225 typical, 250 Maximum See Figure 22 for a representative tuning curve. 180 140 100 PPM 60 20 -20 -60 -100 -140 -180 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 VTUNE (V) 30179193 FIGURE 22. Example Tuning Curve, 20.48 MHz Crystal The tuning curve achieved in the user's application may differ from the curve shown above due to differences in PCB layout and component selection. 104 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 This data is measured on the bench with the crystal integrated with the LMK04906 family. Using a voltmeter to monitor the VTUNE node for the crystal, the PLL1 reference clock input frequency is swept in frequency and the resulting tuning voltage generated by PLL1 is measured at each frequency. At each value of the reference clock frequency, the lock state of PLL1 should be monitored to ensure that the tuning voltage applied to the crystal is valid. The curve shows over the tuning voltage range of 0.3 VDC to 3.0 VDC, the frequency range is -140 to +91 ppm; or equivalently, a tuning range of -2850 Hz to +1850 Hz. The measured tuning voltage at the nominal crystal frequency (20.48 MHz) is 1.7 V. Using the diode data sheet tuning characteristics, this voltage results in a tuning capacitance of approximately 6.5 pF. The tuning curve data can be used to calculate the gain of the oscillator (KVCO). The data used in the calculations is taken from the most linear portion of the curve, a region centered on the crossover point at the nominal frequency (20.48 MHz). For a well designed circuit, this is the most likely operating range. In this case, the tuning range used for the calculations is ± 1000 Hz (± 0.001 MHz), or ± 81.4 ppm. The simplest method is to calculate the ratio: (10) ΔF2 and ΔF1 are in units of MHz. Using data from the curve this becomes: (11) A second method uses the tuning data in units of ppm: (12) Copyright © 1999-2012, Texas Instruments Incorporated 105 LMK04906 FNOM is the nominal frequency of the crystal and is in units of MHz. Using the data, this becomes: (13) In order to ensure startup of the oscillator circuit, the equivalent series resistance (ESR) of the selected crystal should conform to the specifications listed in the table of Electrical Characteristics. It is also important to select a crystal with adequate power dissipation capability, or drive level. If the drive level supplied by the oscillator exceeds the maximum specified by the crystal manufacturer, the crystal will undergo excessive aging and possibly become damaged. Drive level is directly proportional to resonant frequency, capacitive load seen by the crystal, voltage and equivalent series resistance (ESR). For more complete coverage of crystal oscillator design, see Application Note AN-1939 at http:// www.ti.com/lsds/ti/analog/clocksandtimers/ clocks_and_timers.page or http://www.ti.com/ general/docs/techdocs.tsp?silold=1. 20.9 DRIVING CLKin AND OSCin INPUTS 20.9.1 Driving CLKin Pins with a Differential Source Both CLKin ports can be driven by differential signals. It is recommended that the input mode be set to bipolar (CLKinX_BUF_TYPE = 0) when using differential reference clocks. The LMK04906 family internally biases the input pins so the differential interface should be AC coupled. The recommended circuits for driving the CLKin pins with either LVDS or LVPECL are shown in Figure 23 and Figure 24. 30179181 FIGURE 23. CLKinX/X* Termination for an LVDS Reference Clock Source 30179187 FIGURE 24. CLKinX/X* Termination for an LVPECL Reference Clock Source Finally, a reference clock source that produces a differential sine wave output can drive the CLKin pins using the following circuit. Note: the signal level must conform to the requirements for the CLKin pins listed in the Electrical Characteristics table. 30179124 FIGURE 25. CLKinX/X* Termination for a Differential Sinewave Reference Clock Source 20.9.2 Driving CLKin Pins with a Single-Ended Source The CLKin pins of the LMK04906 family can be driven using a single-ended reference clock source, for example, either a sine wave source or an LVCMOS/LVTTL source. Either AC coupling or DC coupling may be used. In the case of the sine wave source that is expecting a 50 Ω load, it is recommended that AC coupling be used as shown in the circuit below with a 50 Ω termination. Note: The signal level must conform to the requirements for the CLKin pins listed in the Electrical Characteristics table. CLKinX_BUF_TYPE in Register 11 is recommended to be set to bipolar mode (CLKinX_BUF_TYPE = 0). 106 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 30179122 FIGURE 26. CLKinX/X* Single-ended Termination If the CLKin pins are being driven with a single-ended LVCMOS/LVTTL source, either DC coupling or AC coupling may be used. If DC coupling is used, the CLKinX_BUF_TYPE should be set to MOS buffer mode (CLKinX_BUF_TYPE = 1) and the voltage swing of the source must meet the specifications for DC coupled, MOS-mode clock inputs given in the table of Electrical Characteristics. If AC coupling is used, the CLKinX_BUF_TYPE should be set to the bipolar buffer mode (CLKinX_BUF_TYPE = 0). The voltage swing at the input pins must meet the specifications for AC coupled, bipolar mode clock inputs given in the table of Electrical Characteristics. In this case, some attenuation of the clock input level may be required. A simple resistive divider circuit before the AC coupling capacitor is sufficient. 30179185 FIGURE 27. DC Coupled LVCMOS/LVTTL Reference Clock 20.10 TERMINATION AND USE OF CLOCK OUTPUT (DRIVERS) When terminating clock drivers keep in mind these guidelines for optimum phase noise and jitter performance: • Transmission line theory should be followed for good impedance matching to prevent reflections. • Clock drivers should be presented with the proper loads. For example: — LVDS drivers are current drivers and require a closed current loop. — LVPECL drivers are open emitters and require a DC path to ground. • Receivers should be presented with a signal biased to their specified DC bias level (common mode voltage) for proper operation. Some receivers have self-biasing inputs that automatically bias to the proper voltage level. In this case, the signal should normally be AC coupled. It is possible to drive a non-LVPECL or non-LVDS receiver with an LVDS or LVPECL driver as long as the above guidelines are followed. Check the datasheet of the receiver or input being driven to determine the best termination and coupling method to be sure that the receiver is biased at its optimum DC voltage (common mode voltage). For example, when driving the OSCin/OSCin* input of the LMK04906 family, OSCin/OSCin* should be AC coupled because OSCin/OSCin* biases the signal to the proper DC level (See Figure 19) This is only slightly different from the AC coupled cases described in Section 20.9.2 Driving CLKin Pins with a Single-Ended Source because the DC blocking capacitors are placed between the termination and the OSCin/OSCin* pins, but the concept remains the same. The receiver (OSCin/OSCin*) sets the input to the optimum DC bias voltage (common mode voltage), not the driver. 20.10.1 Termination for DC Coupled Differential Operation For DC coupled operation of an LVDS driver, terminate with 100 Ω as close as possible to the LVDS receiver as shown in Figure 28. 30179120 FIGURE 28. Differential LVDS Operation, DC Coupling, No Biasing of the Receiver Copyright © 1999-2012, Texas Instruments Incorporated 107 LMK04906 For DC coupled operation of an LVPECL driver, terminate with 50 Ω to VCC - 2 V as shown in Figure 29. Alternatively terminate with a Thevenin equivalent circuit (120 Ω resistor connected to VCC and an 82 Ω resistor connected to ground with the driver connected to the junction of the 120 Ω and 82 Ω resistors) as shown in Figure 30 for VCC = 3.3 V. 30179118 FIGURE 29. Differential LVPECL Operation, DC Coupling 30179121 FIGURE 30. Differential LVPECL Operation, DC Coupling, Thevenin Equivalent 20.10.2 Termination for AC Coupled Differential Operation AC coupling allows for shifting the DC bias level (common mode voltage) when driving different receiver standards. Since AC coupling prevents the driver from providing a DC bias voltage at the receiver it is important to ensure the receiver is biased to its ideal DC level. When driving non-biased LVDS receivers with an LVDS driver, the signal may be AC coupled by adding DC blocking capacitors, however the proper DC bias point needs to be established at the receiver. One way to do this is with the termination circuitry in Figure 31. 30179119 FIGURE 31. Differential LVDS Operation, AC Coupling, External Biasing at the Receiver Some LVDS receivers may have internal biasing on the inputs. In this case, the circuit shown in Figure 31 is modified by replacing the 50 Ω terminations to Vbias with a single 100 Ω resistor across the input pins of the receiver, as shown in Figure 32. When using AC coupling with LVDS outputs, there may be a startup delay observed in the clock output due to capacitor charging. The previous figures employ a 0.1 µF capacitor. This value may need to be adjusted to meet the startup requirements for a particular application. 108 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 30179182 FIGURE 32. LVDS Termination for a Self-Biased Receiver LVPECL drivers require a DC path to ground. When AC coupling an LVPECL signal use 120 Ω emitter resistors close to the LVPECL driver to provide a DC path to ground as shown in Figure 33. For proper receiver operation, the signal should be biased to the DC bias level (common mode voltage) specified by the receiver. The typical DC bias voltage for LVPECL receivers is 2 V. A Thevenin equivalent circuit (82 Ω resistor connected to VCC and a 120 Ω resistor connected to ground with the driver connected to the junction of the 82 Ω and 120 Ω resistors) is a valid termination as shown in Figure 33 for VCC = 3.3 V. Note this Thevenin circuit is different from the DC coupled example in Figure 30. 30179117 FIGURE 33. Differential LVPECL Operation, AC Coupling, Thevenin Equivalent, External Biasing at the Receiver 20.10.3 Termination for Single-Ended Operation A balun can be used with either LVDS or LVPECL drivers to convert the balanced, differential signal into an unbalanced, singleended signal. It is possible to use an LVPECL driver as one or two separate 800 mVpp signals. When using only one LVPECL driver of a CLKoutX/ CLKoutX* pair, be sure to properly terminated the unused driver. When DC coupling one of the LMK04906 family clock LVPECL drivers, the termination should be 50 Ω to VCC - 2 V as shown in Figure 34. The Thevenin equivalent circuit is also a valid termination as shown in Figure 35 for Vcc = 3.3 V. 30179115 FIGURE 34. Single-Ended LVPECL Operation, DC Coupling 30179116 FIGURE 35. Single-Ended LVPECL Operation, DC Coupling, Thevenin Equivalent Copyright © 1999-2012, Texas Instruments Incorporated 109 LMK04906 When AC coupling an LVPECL driver use a 120 Ω emitter resistor to provide a DC path to ground and ensure a 50 Ω termination with the proper DC bias level for the receiver. The typical DC bias voltage for LVPECL receivers is 2 V (See Section 20.9.2 Driving CLKin Pins with a Single-Ended Source). If the companion driver is not used it should be terminated with either a proper AC or DC termination. This latter example of AC coupling a single-ended LVPECL signal can be used to measure single-ended LVPECL performance using a spectrum analyzer or phase noise analyzer. When using most RF test equipment no DC bias point (0 VDC) is required for safe and proper operation. The internal 50 Ω termination of the test equipment correctly terminates the LVPECL driver being measured as shown in Figure 36. 30179114 FIGURE 36. Single-Ended LVPECL Operation, AC Coupling 110 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 20.11 POWER SUPPLY 20.11.1 Current Consumption / Power Dissipation Calculations From Table 21 the current consumption can be calculated for any configuration. For example, the current for the entire device with 1 LVDS (CLKout0) and 1 LVPECL 1.6 Vpp /w 240 ohm emitter resistors (CLKout1) output active with a clock output divide = 1, and no other features enabled can be calculated by adding up the following blocks: core current, clock buffer, one LVDS output buffer current, and one LVPECL output buffer current. There will also be one LVPECL output drawing emitter current, which means some of the power from the current draw of the device is dissipated in the external emitter resistors which doesn't add to the power dissipation budget for the device but is important for LDO ICC calculations. For total current consumption of the device, add up the significant functional blocks. In this example, 228.1 mA = • 140 mA (core current) • 17.3 mA (base clock distribution) • 25.5 mA (CLKout0 & 1 divider) • 14.3 mA (LVDS buffer) • 31 mA (LVPECL 1.6 Vpp buffer /w 240 ohm emitter resistors) Once total current consumption has been calculated, power dissipated by the device can be calculated. The power dissipation of the device is equation to the total current entering the device multiplied by the voltage at the device minus the power dissipated in any emitter resistors connected to any of the LVPECL outputs. If no emitter resistors are connected to the LVPECL outputs, this power will be 0 watts. Continuing the above example which has 228.1 mA total Icc and one output with 240 ohm emitter resitors. Total IC power = 717.7 mW = 3.3 V * 228.1 mA - 35 mW. TABLE 21. Typical Current Consumption for Selected Functional Blocks (TA = 25 °C, VCC = 3.3 V) Block Condition Typical ICC (mA) Power dissipated in device (mW) Power dissipated externally (Note 41) (mW) Core and Functional Blocks Core MODE = 0: Dual Loop, Internal VCO PLL1 and PLL2 locked 140 462 - MODE = 2: Dual Loop, Internal VCO, 0-Delay PLL1 and PLL2 locked; Includes EN_FEEDBACK_MUX = 1 155 512 - MODE = 3: Dual Loop, External VCO PLL1 and PLL2 locked 127 419 - MODE = 5: Dual Loop, External VCO, 0-Delay PLL1 and PLL2 locked; Includes EN_FEEDBACK_MUX = 1 142 469 - MODE = 6: Single Loop (PLL2), Internal VCO PLL2 locked 116 383 - MODE = 11: Single Loop (PLL2), External VCO PLL2 locked 103 340 - PD_OSCin = 0 42 139 - PD_OSCin = 1 34.5 114 - 2 6.6 - 17.3 57.1 - MODE = 16: Clock Distribution EN_TRACK Tracking is enabled (EN_TRACK = 1) Base Clock Distribution At least 1 CLKoutX_PD = 0 CLKout Outputs Each CLKout Output 2.8 9.2 - When a clock output is enabled, this contributes the divider/delay block 25.5 84.1 - Clock Divider/ Digital Delay Divider / digital delay in extended mode 29.6 97.7 - VCO Divider VCO Divider current 7.7 25.4 - HOLDOVER mode When in holdover mode 2.2 7.2 - Feedback Mux Feedback mux must be enabled for 0-delay modes and digital delay mode (SYNC_QUAL = 1) 4.9 16.1 - SYNC Asserted While SYNC is asserted, this extra current is drawn 1.7 5.6 - EN_SYNC = 1 Required for SYNC functionality. May be turned off once SYNC is complete to save power. 6 19.8 - Copyright © 1999-2012, Texas Instruments Incorporated 111 LMK04906 Block Condition SYNC_QUAL = 1 Delay enabled, delay > 7 (CLKout_MUX = 2, 3) Crystal Mode OSCin Doubler Enabling the Crystal Oscillator Analog Delay Power dissipated externally (Note 41) (mW) 8.7 28.7 - XTAL_LVL = 0 1.8 5.9 - XTAL_LVL = 1 2.7 9 - XTAL_LVL = 2 3.6 12 - XTAL_LVL = 3 4.5 15 - 2.8 9.2 - CLKoutX_ANLG_DLY = 0 to 3 3.4 11.2 - CLKoutX_ANLG_DLY = 4 to 7 3.8 12.5 - CLKoutX_ANLG_DLY = 8 to 11 4.2 13.9 - CLKoutX_ANLG_DLY = 12 to 15 4.7 15.5 - CLKoutX_ANLG_DLY = 16 to 23 5.2 17.2 - 2.8 9.2 - EN_PLL2_REF_2X = 1 Analog Delay Value Typical ICC (mA) Power dissipated in device (mW) Clock Output Has Analog Delay Selected. Example: CLKout0_ADLY_SEL = 1 Clock Output Buffers LVDS LVPECL LVCMOS 100 ohm differential termination 14.3 47.2 - LVPECL 2.0 Vpp, AC coupled using 240 ohm emitter resistors 32 70.6 35 LVPECL 1.6 Vpp, AC coupled using 240 ohm emitter resistors 31 67.3 35 LVPECL 1.6 Vpp, AC coupled using 120 ohm emitter resistors 46 91.8 60 LVPECL 1.2 Vpp, AC coupled using 240 ohm emitter resistors 30 59 40 LVPECL 0.7 Vpp, AC coupled using 240 ohm emitter resistors 29 55.7 40 LVCMOS Pair (CLKoutX_TYPE = 6 to 9) CL = 5 pF 3 MHz 24 79.2 - 30 MHz 26.5 87.5 - 36.5 120.5 - LVCMOS Single (CLKoutX_TYPE = 10 to 13) CL = 5 pF 3 MHz 15 49.5 - 30 MHz 16 52.8 - 21.5 71 - 150 MHz 150 MHz Note 41: Power is dissipated externally in LVPECL emitter resistors. The externally dissipated power is calculated as twice the DC voltage level of one LVPECL clock output pin squared over the emitter resistance. That is to say power dissipated in emitter resistors = 2 * Vem2 / Rem. Note 42: Assuming θJA = 15 °C/W, the total power dissipated on chip must be less than (125 °C – 85 °C) / 16 °C/W = 2.5 W to guarantee a junction temperature is less than 125 °C. Note 43: Worst case power dissipation can be estimated by multiplying typical power dissipation with a factor of 1.15. 112 Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 20.12 THERMAL MANAGEMENT Power consumption of the LMK04906 can be high enough to require attention to thermal management. For reliability and performance reasons the die temperature should be limited to a maximum of 125 °C. That is, as an estimate, TA (ambient temperature) plus device power consumption times θJA should not exceed 125 °C. The package of the device has an exposed pad that provides the primary heat removal path as well as excellent electrical grounding to a printed circuit board. To maximize the removal of heat from the package a thermal land pattern including multiple vias to a ground plane must be incorporated on the PCB within the footprint of the package. The exposed pad must be soldered down to ensure adequate heat conduction out of the package. A recommended land and via pattern is shown in Figure 37. More information on soldering QFN, formerly referred to as LLP, packages can be obtained: http:// www.national.com/analog/packaging/. A recommended footprint including recommended solder mask and solder paste layers can be found at: http://www.national.com/ analog/packaging/gerber for the SQA64 package. 30179173 FIGURE 37. Recommended Land and Via Pattern To minimize junction temperature it is recommended that a simple heat sink be built into the PCB (if the ground plane layer is not exposed). This is done by including a copper area of about 2 square inches on the opposite side of the PCB from the device. This copper area may be plated or solder coated to prevent corrosion but should not have conformal coating (if possible), which could provide thermal insulation. The vias shown in Figure 37 should connect these top and bottom copper layers and to the ground layer. These vias act as “heat pipes” to carry the thermal energy away from the device side of the board to where it can be more effectively dissipated. Copyright © 1999-2012, Texas Instruments Incorporated 113 LMK04906 21.0 Physical Dimensions inches (millimeters) unless otherwise noted 22.0 Ordering Information Order Number Ref Inputs Buffered OSCin Outputs Programmable Outputs VCO Packaging 3 1 6 2.5 GHz 1000 Unit Tape and Reel LMK04906BISQE LMK04906BISQ LMK04906BISQX 114 Package Marking 250 Unit Tape and Reel K04906 2500 Unit Tape and Reel Copyright © 1999-2012, Texas Instruments Incorporated LMK04906 Notes Copyright © 1999-2012, Texas Instruments Incorporated 115 Notes Copyright © 1999-2012, Texas Instruments Incorporated IMPORTANT NOTICE Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other changes to its semiconductor products and services per JESD46C and to discontinue any product or service per JESD48B. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and complete. All semiconductor products (also referred to herein as “components”) are sold subject to TI’s terms and conditions of sale supplied at the time of order acknowledgment. TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TI’s terms and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily performed. TI assumes no liability for applications assistance or the design of Buyers’ products. Buyers are responsible for their products and applications using TI components. To minimize the risks associated with Buyers’ products and applications, Buyers should provide adequate design and operating safeguards. TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the third party, or a license from TI under the patents or other intellectual property of TI. Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional restrictions. Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice. TI is not responsible or liable for any such statements. Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use of any TI components in safety-critical applications. In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TI’s goal is to help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and requirements. Nonetheless, such components are subject to these terms. No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties have executed a special agreement specifically governing such use. Only those TI components which TI has specifically designated as military grade or “enhanced plastic” are designed and intended for use in military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and regulatory requirements in connection with such use. TI has specifically designated certain components which meet ISO/TS16949 requirements, mainly for automotive use. Components which have not been so designated are neither designed nor intended for automotive use; and TI will not be responsible for any failure of such components to meet such requirements. Products Applications Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers DLP® Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps DSP dsp.ti.com Energy and Lighting www.ti.com/energy Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial Interface interface.ti.com Medical www.ti.com/medical Logic logic.ti.com Security www.ti.com/security Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video RFID www.ti-rfid.com OMAP Mobile Processors www.ti.com/omap TI E2E Community e2e.ti.com Wireless Connectivity www.ti.com/wirelessconnectivity Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265 Copyright © 2012, Texas Instruments Incorporated