LMX2541SQX2060E/NOPB

Product Folder Sample & Buy Support & Community Tools & Software Technical Documents LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 LMX2541 Ultra-Low Noise PLLatinum Frequency Synthesizer With Integrated VCO 1 Features 3 Description • The LMX2541 device is an ultra low-noise frequency synthesizer which integrates a high-performance delta-sigma fractional N PLL, a VCO with fully integrated tank circuit, and an optional frequency divider. The PLL offers an unprecedented normalized noise floor of –225 dBc/Hz and can be operated with up to 104 MHz of phase-detector rate (comparison frequency) in both integer and fractional modes. The PLL can also be configured to work with an external VCO. 1 • • • • • • • • • • • • • • • • • • Multiple Frequency Options Available (See Device Comparison Table) Frequencies From 31.6 MHz to 4000 MHz Very Low RMS Noise and Spurs –225 dBc/Hz Normalized PLL Phase Noise Integrated RMS Noise (100 Hz to 20 MHz) – 2 mRad (100 Hz to 20 MHz) at 2.1 GHz – 3.5 mRad (100 Hz to 20 MHz) at 3.5 GHz Ultra Low-Noise Integrated VCO External VCO Option (Internal VCO Bypassed) VCO Frequency Divider 1 to 63 (All Values) Programmable Output Power Up to 104-MHz Phase Detector Frequency Integrated Low-Noise LDOs Programmable Charge Pump Output Partially Integrated Loop Filter Digital Frequency Shift Keying (FSK) Modulation Pin Integrated Reference Crystal Oscillator Circuit Hardware and Software Power Down FastLock Mode and VCO-Based Cycle Slip Reduction Analog and Digital Lock Detect 1.6-V Logic Compatibility The LMX2541 integrates several low-noise, highprecision LDOs and output driver matching network to provide higher supply noise immunity and more consistent performance, while reducing the number of external components. When combined with a highquality reference oscillator, the LMX2541 generates a very stable, ultra low-noise signal. The LMX2541 is offered in a family of 6 devices with varying VCO frequency range from 1990 MHz up to 4 GHz. Using a flexible divider, the LMX2541 can generate frequencies as low as 31.6 MHz. The LMX2541 is a monolithic integrated circuit, fabricated in a proprietary BiCMOS process. Device programming is facilitated using a three-wire MICROWIRE interface that can operate down to 1.6 volts. Supply voltage ranges from 3.15 V to 3.45 V. The LMX2541 is available in a 36-pin 6-mm × 6-mm × 0.8-mm WQFN package. Device Information(1) 2 • • • • • Applications PART NUMBER Wireless Infrastructure (UMTS, LTE, WiMax) Broadband Wireless Wireless Meter Reading Test and Measurement FM Mobile Radio space PACKAGE VCO FREQUENCY (MHz) LMX2541SQ2060E WQFN (36) 1990 - 2240 LMX2541SQ2380E WQFN (36) 2200 - 2530 LMX2541SQ2690E WQFN (36) 2490 - 2865 LMX2541SQ3030E WQFN (36) 2810 - 3230 LMX2541SQ3320E WQFN (36) 3130 - 3600 LMX2541SQ3740E WQFN (36) 3480 - 4000 (1) For all available packages, see the orderable addendum at the end of the data sheet. System Block Diagram Clean Reference Clock Dist. Dirty Input Clock PLL Ultra-Clean Local Oscillator PLL LMK04000 LMX2541 (Clock Jitter Cleaner) (RF Synthesizer) 1 An IMPORTANT NOTICE at the end of this data sheet addresses availability, warranty, changes, use in safety-critical applications, intellectual property matters and other important disclaimers. PRODUCTION DATA. LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Table of Contents 1 2 3 4 5 6 7 Features .................................................................. Applications ........................................................... Description ............................................................. Revision History..................................................... Device Comparison Table..................................... Pin Configuration and Functions ......................... Specifications......................................................... 7.1 7.2 7.3 7.4 7.5 7.6 7.7 1 1 1 2 3 4 6 Absolute Maximum Ratings ...................................... 6 ESD Ratings.............................................................. 6 Recommended Operating Conditions....................... 6 Thermal Information .................................................. 6 Electrical Characteristics........................................... 7 Timing Requirements .............................................. 11 Typical Characteristics ............................................ 12 8 Parameter Measurement Information ................ 23 9 Detailed Description ............................................ 25 8.1 Bench Test Setups.................................................. 23 9.1 Overview ................................................................. 25 9.2 9.3 9.4 9.5 9.6 Functional Block Diagrams ..................................... Feature Description................................................. Device Functional Modes........................................ Programming........................................................... Register Maps ......................................................... 25 26 36 37 38 10 Application and Implementation........................ 51 10.1 Application Information.......................................... 51 10.2 Typical Application ............................................... 59 11 Power Supply Recommendations ..................... 61 12 Layout................................................................... 62 12.1 Layout Guidelines ................................................. 62 12.2 Layout Example .................................................... 63 13 Device and Documentation Support ................. 64 13.1 13.2 13.3 13.4 Device Support...................................................... Trademarks ........................................................... Electrostatic Discharge Caution ............................ Glossary ................................................................ 64 64 64 64 14 Mechanical, Packaging, and Orderable Information ........................................................... 64 4 Revision History Changes from Revision I (February 2013) to Revision J • Added ESD Ratings table, Feature Description section, Device Functional Modes, Application and Implementation section, Power Supply Recommendations section, Layout section, Device and Documentation Support section, and Mechanical, Packaging, and Orderable Information section .................................................................................................. 1 Changes from Revision H (February 2013) to Revision I • 2 Page Page Changed layout of National Data Sheet to TI format ........................................................................................................... 36 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 5 Device Comparison Table VCO _DIV LMX2541SQ2060E LMX2541SQ2380E LMX2541SQ2690E LMX2541SQ3030E LMX2541SQ3320E LMX2541SQ3740E Fmin Fmax Fmin Fmax Fmin Fmax Fmin Fmax Fmin Fmax Fmin Fmax 1 1990.0 2240.0 2200.0 2530.0 2490.0 2865.0 2810.0 3230.0 3130.0 3600.0 3480.0 4000.0 2 995.0 1120.0 1100.0 1265.0 1245.0 1432.5 1405.0 1615.0 1565.0 1800.0 1740.0 2000.0 3 663.3 746.7 733.3 843.3 830.0 955.0 936.7 1076.7 1043.3 1200.0 1160.0 1333.3 4 497.5 560.0 550.0 632.5 622.5 716.3 702.5 807.5 782.5 900.0 870.0 1000.0 5 398.0 448.0 440.0 506.0 498.0 573.0 562.0 646.0 626.0 720.0 696.0 800.0 6 331.7 373.3 366.7 421.7 415.0 477.5 468.3 538.3 521.7 600.0 580.0 666.7 7 284.3 320.0 314.3 361.4 355.7 409.3 401.4 461.4 447.1 514.3 497.1 571.4 8 248.8 280.0 275.0 316.3 311.3 358.1 351.3 403.8 391.3 450.0 435.0 500.0 ... ... ... ... ... ... ... ... ... ... ... ... ... 63 31.6 35.6 34.9 40.2 39.5 45.5 44.6 51.3 49.7 57.1 55.2 63.5 All devices have continuous frequency coverage below a divide value of 8 (7 for most devices) down to their minimum frequency achievable with divide by 63. The numbers in bold show the upper end of this minimum continuous frequency range. For instance, the LMX2541SQ3740E option offers continuous frequency coverage from 55.2 MHz to 571.4 MHz and LMX2541SQ2060E offers continuous frequency coverage from 31.6 MHz to 280 MHz. If using the part in External VCO mode, all parts have roughly the same performance and any option will do. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 3 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 6 Pin Configuration and Functions RFout NC LE CLK DATA VccDiv Bypass VccBias VccDig NJK Package 36-Pin WQFN Top View 36 35 34 33 32 31 30 29 28 GND 1 27 GND VregRFout 2 26 VregFRAC VccRFout 3 25 VccFRAC L1 4 24 RFoutEN 23 VccOSCin 0 GND 21 OSCin VregVCO 8 20 Ftest/LD VrefVCO 9 19 VccPLL2 11 12 13 14 15 16 17 18 VccCP2 10 FLout 7 CPout VccVCO Vtune OSCin* VccCP1 22 VccPLL1 6 ExtVCOin L2 CE 5 GND Lmid Pin Functions PIN 4 TYPE DESCRIPTION NAME NO. Bypass 30 Bypass Put a cap to the VccBias pin. CE 11 CMOS Chip Enable. The device needs to be programmed for this pin to properly power down the device. CLK 33 High-Z Input CPout 16 Output DATA 32 High-Z Input MICROWIRE serial data input. High impedance CMOS input. ExtVCOin 12 RF Input Optional input for use with an external VCO. This pin should be AC coupled if used or left open if not used. FLout 17 Output Fastlock output. Ftest/LD 20 Output Software controllable multiplexed CMOS output. Can be used to monitor PLL lock condition. GND 0 GND GND 1 GND GND 10 GND GND 27 GND L1 4 NC Do not connect this pin. L2 6 NC Do not connect this pin. LE 34 High-Z Input Lmid 5 NC Do not connect this pin. NC 35 NC No connect. OSCin 21 High-Z Input MICROWIRE clock input. High impedance CMOS input. This pin is used for the digital FSK modulation feature. Charge pump output. The DAP pad must be grounded. MICROWIRE Latch Enable input. High impedance CMOS input. Oscillator input signal. If not being used with an external crystal, this input should be AC coupled. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Pin Functions (continued) PIN TYPE DESCRIPTION NAME NO. OSCin* 22 High-Z Input RFout 36 RF Output RFoutEN 24 Input VccBias 29 Supply Supply for Bias circuitry that is for the whole chip. VccCP1 14 Supply Power supply for PLL charge pump. VccCP2 18 Supply Power supply for PLL charge pump. VccDig 28 Supply Supply for digital circuitry, such the MICROWIRE. VccDiv 31 Supply Supply for the output divider VccFRAC 25 Supply (LDO Input) VccOSCin 23 Supply Supply for the OSCin buffer. VccPLL1 13 Supply Power supply for PLL. VccPLL2 19 Supply Power supply for PLL. VccRFout 3 Supply (LDO Input) Supply for the RF output buffer. VccVCO 7 Supply (LDO Input) Supply for the VCO. VregFRAC 26 LDO Output Regulated power supply used for the fractional delta-sigma circuitry. VregRFout 2 LDO Output LDO Output for RF output buffer. Complementary oscillator input signal. Can also be used with an external crystal. If not being used with an external crystal, this input should be AC coupled. RF output. Must be AC coupled if used. Software programmable output enable pin. Power Supply for the PLL fractional circuitry. VrefVCO 9 LDO Bypass LDO Bypass VregVCO 8 LDO Output LDO Output for VCO Vtune 15 High-Z Input Tuning voltage input to the VCO. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 5 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 7 Specifications 7.1 Absolute Maximum Ratings over operating free-air temperature range (unless otherwise noted) (1) (2) MIN MAX UNIT Vcc Power Supply Voltage –0.3 3.6 V VIN Input Voltage to pins other than Vcc Pins –0.3 (Vcc + 0.3) V TL Lead Temperature (solder 4 sec.) 260 °C Tstg Storage Temperature 150 °C (1) (2) (3) (3) –65 Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings only, which do not imply functional operation of the device at these or any other conditions beyond those indicated under Recommended Operating Conditions. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. If Military/Aerospace specified devices are required, please contact the Texas Instruments Sales Office/Distributors for availability and specifications. Never to exceed 3.6 V. 7.2 ESD Ratings VALUE V(ESD) (1) (2) Electrostatic discharge Human body model (HBM), per ANSI/ESDA/JEDEC JS-001 (1) ±2500 Charged-device model (CDM), per JEDEC specification JESD22C101 (2) ±1750 Machine model (MM) ±400 UNIT V JEDEC document JEP155 states that 500-V HBM allows safe manufacturing with a standard ESD control process. JEDEC document JEP157 states that 250-V CDM allows safe manufacturing with a standard ESD control process. 7.3 Recommended Operating Conditions MIN NOM MAX Vcc Power Supply Voltage (All Vcc Pins) 3.15 3.3 3.45 UNIT V TA Ambient Temperature –40 85 °C 7.4 Thermal Information THERMAL METRIC (1) LMX2541 SQ2060E LMX2541 SQ2060E LMX2541 SQ2060E 9 Thermal Vias (2) 13 Thermal Vias (3) 16 Thermal Vias (4) RθJA Junction-to-ambient thermal resistance 31.7 30.3 29.8 ψJT Junction-to-top characterization parameter 7.3 7.3 7.3 (1) (2) (3) (4) 6 UNIT °C/W For more information about traditional and new thermal metrics, see the IC Package Thermal Metrics application report, SPRA953. Recommended for Most Reliable Solderability. Compromise Between Solderability, Heat Dissipation, and Fractional Spurs. Recommended for Optimal Heat Dissipation and Fractional Spurs. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 7.5 Electrical Characteristics (3.15 V ≤ VCC ≤ 3.45 V, -40°C ≤ TA ≤ 85 °C; except as specified. Typical values are at Vcc = 3.3 V, 25 C.) PARAMETER TEST CONDITIONS MIN TYP MAX VCO_DIV>1 170 204 VCO_DIV=1 130 156 UNIT CURRENT CONSUMPTION Default Power Mode ICC Entire Chip Supply Current with all blocks enabled IPLL Current for External VCO Mode RFoutEN = LOW 72 94 mA IDIV Current for Divider Only Mode VCO_DIV >1 Default Power Mode 84 110 mA ICCPD Power Down Current CE = 0 V, Device Initialized 100 250 µA 300 µA (1) (1) mA OSCILLATOR (NORMAL MODE OPERATION WITH XO=0) IIHOSC in Oscillator Input High Current for OSCin and OSCin* VIH = 2.75 V IILOSCi n Oscillator Input Low Current for OSCin and OSCin* pins VIL = 0 –100 OSC_2X = 1 fOSCin OSCin Frequency Range (2) dvOSCin Slew Rate vOSCin Oscillator Sensitivity OSC_2X = 0 µA 5 52 MODE = 0 5 700 MODE = 1 5 900 (2) MHz 150 dvOSCin ≥ 150 V/µs V/µs Single-Ended 0.2 2.0 Differential 0.4 3.1 5 20 MHz 100 Ω Vpp OSCILLATOR (CRYSTAL MODE WITH XO=1) fXTAL Crystal Frequency Range VIH = 2.75 V ESRXTA Crystal Equivalent Series Resistance L This a requirement for the crystal, not a characteristic of the LMX2541. PXTAL Power Dissipation in Crystal This requirement is for the crystal, not a characteristic of the LMX2541. COSCin Input Capacitance of OSCin 200 µW 6 pF PLL fPD Phase Detector Frequency ICPout Charge Pump Output Current Magnitude 104 CPG = 1X 100 CPG = 2X 200 CPG = 3X 300 ... µA ... CPG=32X ICPoutTR CP TRI-STATE Current I 0.4 V < VCPout < Vcc - 0.4 ICPoutM M Charge Pump Sink vs. Source Mismatch ICPoutV 3200 1 5 VCPout = Vcc / 2 TA = 25°C 3% 10% Charge Pump Current vs. CP Voltage Variation 0.4 V < VCPout < Vcc - 0.4 TA = 25°C 4% ICPoutT CP Current vs. Temperature Variation VCPout = Vcc / 2 8% Normalized PLL 1/f Noise LNPLL_flicker(10 kHz) CPG = 1X LN(f) CPG = 32X –124.5 Normalized PLL Noise Floor LNPLL_flat(1 Hz) CPG = 1X –220.8 CPG = 32X –225.4 (3) (1) (2) (3) MHz –116 nA dBc/Hz dBc/Hz The LMX2541 RFout power level is programmable with the program words of VCOGAIN, OUTTERM, and DIVGAIN. Changing these words can change the output power of the VCO as well as the current consumption of the output buffer. For the purpose of consistency in electrical specifications, "Default Power Mode" is defined to be the settings of VCOGAIN = OUTTERM = DIVGAIN = 12. Not tested in production. Specified by characterization. OSCin is tested only to 400 MHz. See Application and Implementation for more details on these parameters. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 7 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Electrical Characteristics (continued) (3.15 V ≤ VCC ≤ 3.45 V, -40°C ≤ TA ≤ 85 °C; except as specified. Typical values are at Vcc = 3.3 V, 25 C.) PARAMETER fExtVCOin PLL Input Frequency pExtVCOi n PLL Input Sensitivity ( (2) applies to Max Limit Only) TEST CONDITIONS MIN TYP MAX RFout Buffer Enabled and VCO_DIV > 1 400 4000 RFout Buffer Disabled and VCO_DIV = 1 400 6000 fExtVCOin ≤ 4 GHz –15 10 fExtVCOin > 4 GHz –5 10 2060E 1990 2240 2380E 2200 2530 2690E 2490 2865 3030E 2810 3230 3320E 3130 3600 3740E 3480 4000 UNIT MHz dBm VCO SPECIFICATIONS fVCO Internal VCO Frequency Range Maximum Allowable Temperature Drift for Continuous Lock ΔTCL pRFout RF Output Power ΔPRFout Change in Output Power KVtune Fine Tuning Sensitivity HSRFout DERFout Second Harmonic (5) Duty Cycle Error (5) PSHVCO VCO Frequency Pushing PULVCO VCO Frequency Pulling (4) (5) 8 Mode = Full Chip Mode This is the frequency before the VCO divider. (2) (4) , 125 Maximum Frequency Default Power Mode VCO_DIV=1 °C 2060E 3.5 2380E 2.8 2690E 1.6 3030E 1.2 3320E 0.2 3740E –0.3 Fixed Temperature with 100 MHz frequency change at the output 0.3 Fixed frequency with a change over the entire temperature range 0.4 The lower number in the range applies when the VCO is at its lowest frequency and the higher number applies when the VCO is at its highest frequency. A linear approximation can be used for frequencies between these two cases. 2060E 13 - 23 2380E 16 - 30 2690E 17 - 32 3030E 20 - 37 3320E 21 - 37 3740E 24 - 42 VCO_DIV = 2 –20 50 Ω Load VCO_DIV = 3 –20 Default Power Mode VCO_DIV = 2 3% 50 Ω Load VCO_DIV = 3 3% VCO_DIV = 1 ±800 VCO_DIV > 1 ±60 (1) CVregVCO = 4.7 µF, Open Loop VSWR 1.7 to 1 (6 dB Pad) dBm dB Default Power Mode (1) MHz 600 MHz/V dBc kHz/V kHz 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 R0 register was last programmed, and still have the device stay in lock. The action of programming the R0 register, even to the same value, activates a frequency calibration routine. This implies that the device 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 R0 register to ensure that it stays in lock. Regardless of what temperature the device was initially programmed at, the temperature can never drift outside the frequency range of -40°C ≤TA≤ 85°C without violating specifications. The duty cycle error (DE) and second harmonic (HS) are theoretically related by the equation HS = 10·log| 2π·DE | - 6 dB. A square wave with 3% duty cycle theoretically has a second harmonic of -20 dBc. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Electrical Characteristics (continued) (3.15 V ≤ VCC ≤ 3.45 V, -40°C ≤ TA ≤ 85 °C; except as specified. Typical values are at Vcc = 3.3 V, 25 C.) PARAMETER σΦ RMS Phase Error TEST CONDITIONS Integration Bandwidth = 100 Hz to 20 MHz Middle VCO Frequency 100 MHz Wenzel Crystal Reference Integer Mode Optimized Loop Bandwidth MIN TYP 2060E 1.6 2380E 1.8 2690E 2.1 3030E 2.1 3320E 2.3 3740E 2.6 MAX UNIT mRad VCO PHASE NOISE (6) fRFout = Min VCO Frequency L(f)Fout Phase Noise 2060E fRFout = Max VCO Frequency fRFout = Min VCO Frequency L(f)Fout Phase Noise 2380E fRFout = Max VCO Frequency fRFout = Min VCO Frequency L(f)Fout Phase Noise 2690E fRFout = Max VCO Frequency (6) 10-kHz Offset –89.7 100-kHz Offset –113.7 1-MHz Offset –134.9 10-MHz Offset –155.4 20-MHz Offset –160.3 10-kHz Offset –86.5 100-kHz Offset –111.4 1-MHz Offset –132.8 10-MHz Offset –153.4 20-MHz Offset –158.5 10-kHz Offset –87.9 100-kHz Offset –112.7 1-MHz Offset –133.8 10-MHz Offset –154.2 20-MHz Offset –159.5 10-kHz Offset –83.4 100-kHz Offset –109.1 1-MHz Offset –130.8 10-MHz Offset –151.8 20-MHz Offset –157.5 10-kHz Offset –86.9 100-kHz Offset –111.8 1-MHz Offset –133.3 10-MHz Offset –154.2 20-MHz Offset –159.4 10-kHz Offset –82.3 100-kHz Offset –108.4 1-MHz Offset –130.3 10-MHz Offset –151.1 20-MHz Offset –156.7 dBc/Hz dBc/Hz dBc/Hz The VCO phase noise is measured assuming that the loop bandwidth is sufficiently narrow that the VCO noise dominates. The phase noise is measured with AC_TEMP_COMP = 5 and the device is reloaded at each test frequency. The typical performance characteristics section shows how the VCO phase noise varies over temperature and frequency. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 9 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Electrical Characteristics (continued) (3.15 V ≤ VCC ≤ 3.45 V, -40°C ≤ TA ≤ 85 °C; except as specified. Typical values are at Vcc = 3.3 V, 25 C.) PARAMETER TEST CONDITIONS fRFout = Min VCO Frequency L(f)Fout Phase Noise 3030E fRFout = Max VCO Frequency fRFout = Min VCO Frequency L(f)Fout Phase Noise 3320E fRFout = Max VCO Frequency fRFout = Min VCO Frequency L(f)Fout Phase Noise 3740E fRFout = Max VCO Frequency MIN TYP 10-kHz Offset –86.1 100-kHz Offset –110.5 1-MHz Offset –132.0 10-MHz Offset –152.2 20-MHz Offset –157.1 10-kHz Offset –82.2 100-kHz Offset –107.7 1-MHz Offset –129.4 10-MHz Offset –150.5 20-MHz Offset –156.1 10-kHz Offset –84.1 100-kHz Offset –109.1 1-MHz Offset –130.7 10-MHz Offset –151.6 20-MHz Offset –156.9 10-kHz Offset –82.0 100-kHz Offset –107.0 1-MHz Offset –128.5 10-MHz Offset –149.6 20-MHz Offset –155.2 10-kHz Offset –83.9 100-kHz Offset –108.3 1-MHz Offset –129.9 10-MHz offset –150.6 20-MHz Offset –156.5 10-kHz Offset –81.6 100-kHz Offset –106.5 1-MHz Offset –127.7 10-MHz Offset –148.6 20-MHz Offset –154.2 MAX UNIT dBc/Hz dBc/Hz dBc/Hz DIGITAL INTERFACE (DATA, CLK, LE, CE, Ftest/LD, FLout,RFoutEN) VIH High-Level Input Voltage VIL Low-Level Input Voltage IIH High-Level Input Current VIH = 1.75, XO = 0 IIL Low-Level Input Current VIL = 0 V , XO = 0 VOH High-Level Output Voltage IOH = 500 µA VOL Low-Level Output Voltage IOL = -500 µA ILeak Leakage Current Ftest/LD and FLout Pins Only 10 1.6 Submit Documentation Feedback Vcc V 0.4 V –5 5 µA –5 5 µA 2 V 0 –5 0.4 V 5 µA Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 7.6 Timing Requirements MIN NOM MAX UNIT tCE Clock to Enable Low Time See Figure 1 25 ns tCS Data to Clock Set Up Time See Figure 1 25 ns tCH Data to Clock Hold Time See Figure 1 20 ns tCWH Clock Pulse Width High See Figure 1 25 ns tCWL Clock Pulse Width Low See Figure 1 25 ns tCES Enable to Clock Set Up Time See Figure 1 25 ns tEWH Enable Pulse Width High See Figure 1 25 ns MSB DATA D19 LSB D18 D17 D16 D15 D0 C3 C2 C1 C0 CLK tCES tCS tCH tCWH tCWL tES LE tEWH Figure 1. Serial Data Timing Diagram Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 11 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 7.7 Typical Characteristics PLL NORMALIZED NOISE FLOOR (dBc/Hz) PLL NORMALIZED NOISE FLOOR (dBc/Hz) 7.7.1 Not Ensured Characteristics -206 -210 -214 -218 -222 -226 1 10 10 2 3 10 10 4 -216 -218 -220 -222 -224 -226 1X 2X SLEW RATE (V/Ps) -105 16X 32X -115 PLL NORMALIZED 1/f NOISE (dBc/Hz) PLL NORMALIZED 1/f NOISE (dBc/Hz) 8X Figure 3. PLL Normalized Noise Floor vs Charge Pump Gain (Slew Rate = 2000 V/μs) Figure 2. PLL Normalized Noise Floor vs OSCin Slew Rate (KPD = 32X) -109 -113 -117 -121 -125 1 10 4X CHARGE PUMP GAIN -117 -119 -121 -123 -125 10 2 10 3 4 1X 10 2X 4X 8X 16X 32X SLEW RATE (V/Ps) CHARGE PUMP GAIN Figure 4. PLL Normalized 1/f Noise vs OSCin Slew Rate (KPD = 32X) Figure 5. PLL Normalized 1/f Noise vs Charge Pump Gain (Slew Rate = 2000 V/μs) -90 Modeled Noise Measured Noise 1/f Noise Flat Noise -93 Phase Noise (dBc/Hz) -96 -99 -102 -105 -108 -111 -114 -117 -120 1x102 1x103 1x104 Offset (Hz) 1x105 1x106 D001 See Phase Noise Measurement Test Setup Figure 6. PLL 1/f and Noise Floor Measurement of the LMX2541SQ3740E 12 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Not Ensured Characteristics (continued) -80 -145 -88 -147.5 Phase Noise (dBc/Hz) Phase Noise (dBc/Hz) -96 -104 -112 -120 -128 -136 -150 -152.5 -155 -157.5 -144 -160 -152 -160 1x103 1x104 1x105 1x106 Offset (Hz) 1x107 -162.5 5x101 1x108 1x102 D001 Engaging the divider reduces the phase noise by 20 × log(VCO_DIV) except at far offsets where it adds noise. Figure 7. Divider Noise Floor vs Divider Value (fVCO = 3700 MHz, Various Values for VCO_DIV) 2x102 5x102 Offset (Hz) 1x103 2x103 D001 The above plot shows how this noise floor changes as a function of the frequency of the RFout pin. Figure 8. Divider Noise Floor vs Frequency Table 1. Relative VCO Phase Noise Over Temperature Drift (AC_TEMP_COMP = 24, Vcc = 3.3 V) (1) TEMPERATURE (1) PHASE NOISE CHANGE IN CELSIUS FOR VARIOUS OFFSETS LOCK CURRENT 10 kHz 100 kHz 1 MHz 10 MHz 20 MHz –40 –40 +0.4 –2.0 –1.6 –1.8 –1.6 –40 25 +0.3 +0.5 +0.5 +0.5 +0.4 –40 85 +0.9 +2.0 +2.4 +2.5 +2.3 25 –40 +0.2 –2.2 –1.7 –2.0 –1.8 25 25 This is the default condition to which these other numbers are normalized to. 25 85 +0.6 +1.5 +2.0 +2.0 +1.9 85 –40 +0.2 –2.2 –1.7 –1.9 –1.8 85 25 +0.2 +0.2 +0.3 +0.2 +0.2 85 85 +0.6 +1.8 +2.2 +2.3 +2.1 The table shows the typical degradation for VCO phase noise when the VCO is locked at one temperature and the temperature is allowed to drift to another temperature. A negative value indicates a phase noise improvement. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 13 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 7.7.2 Output Power in Bypass Mode The following plots show the trends in output power as a function of temperature, voltage, and frequency. For states where VCOGAIN and OUTTERM are not 12, the table below shows how the output power is modified based on these programmable settings. The measurement of the output power is sensitive to the test circuit. All the numbers in the electrical specifications and typical performance curves were obtained from a characterization setup that accommodate temperature testing and changing of parts. In a more optimized setup the measured RF output power is typically on the order of 1.5 to 2.4 dB higher. 6 5 OUTTERM=12 4 Vcc = 3.3V OUTTERM=15 2 POWER (dBm) POWER (dBm) 3 Vcc = 3.45V 2 1 0 -2 OUTTERM=9 OUTTERM=6 -6 Vcc = 3.15V -1 OUTTERM=3 -2 1750 2250 2750 3250 3750 -10 1750 4250 2250 Figure 9. Output Power vs Voltage (VCO_DIV = 1, VCOGAIN = 12, OUTTERM = 12, TA = 25°C) 3750 4250 6 TA = 25o C 4 VCOGAIN=12 TA = -40o C VCOGAIN=15 2 POWER (dBm) 3 POWER (dBm) 3250 Figure 10. Output Power vs OUTTERM and FREQUENCY (VCO_DIV = 1, TA = 25 °C, Vcc = 3.3 V, VCOGAIN = 12) 5 2 1 TA = 85o C 0 VCOGAIN=9 -2 VCOGAIN=6 -6 VCOGAIN=3 -1 -2 1750 2750 FREQUENCY (MHz) FREQUENCY (MHz) 2250 2750 3250 3750 -10 1750 4250 FREQUENCY (MHz) 2250 2750 3250 3750 4250 FREQUENCY (MHz) Figure 11. Output Power vs Temperature (VCO_DIV = 1, VCOGAIN = 12, OUTTERM = 12, Vcc = 3.3 V) Figure 12. Output Power vs VCOGAIN and FREQUENCY (VCO_DIV = 1, TA = 25 °C, Vcc = 3.3 V, OUTTERM = 12) Table 2. Change in Output Power in Bypass Mode as a Function of VCOGAIN and OUTTERM VCOGAIN OUTTERM 14 3 6 9 12 15 3 –9.7 –8.4 –7.9 –7.8 –7.9 6 –6.6 –4.5 –3.6 –3.4 –3.6 9 –5.7 –3.1 –1.7 –1.3 –1.3 12 –5.4 –2.5 –0.8 +0.0 +0.1 15 –5.3 –2.2 –0.3 +0.8 +1.1 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 7.7.3 Output Power in Divided Mode The measurement of the output power is sensitive to the test circuit. All the numbers in the electrical specifications and typical performance curves were obtained from a characterization setup that accommodate temperature testing and changing of parts. In a more optimized setup the measured RF output power is typically on the order of 1.5 to 2.4 dB higher. 10 9 Vcc = 3.3 V 8 6 4 Vcc = 3.45 V POWER (dBm) POWER (dBm) OUTTERM=12 6 7 5 4 3 2 1 OUTTERM=6 2 0 OUTTERM=9 -2 -4 -6 Vcc = 3.15 V OUTTERM=3 -8 0 -1 OUTTERM=15 8 0 500 1000 1500 -10 2500 2000 0 500 Figure 13. Output Power vs Voltage (VCO_DIV > 1, DIVGAIN = 12, OUTTERM = 12, TA = 25°C) 9 10 8 7 6 POWER (dBm) POWER (dBm) 4 TA = 25o C 3 2 1 0 500 1000 2500 DIVGAIN=15 DIVGAIN=12 2 0 DIVGAIN=6 -2 -4 DIVGAIN=9 -6 TA = 85o C 0 -1 2000 4 TA = -40o C 5 1500 Figure 14. Output Power vs OUTTERM and FREQUENCY (VCO_DIV > 1, TA = 25 °C, Vcc = 3.3 V, DIVGAIN = 12) 8 6 1000 FREQUENCY (MHz) FREQUENCY (MHz) -8 1500 2000 -10 2500 DIVGAIN=3 0 500 1000 1500 2000 2500 FREQUENCY (MHz) FREQUENCY (MHz) Figure 15. Output Power vs Temperature ( VCO_DIV > 1, DIVGAIN = OUTTERM = 12, Vcc = 3.3 V) Figure 16. Output Power vs DIVGAIN and FREQUENCY(VCO_DIV > 1, TA = 25 °C, Vcc = 3.3 V, OUTTERM = 12) Table 3. Change in Output Power in Divided Mode as a Function of DIVGAIN and OUTTERM (1) DIVGAIN OUTTERM (1) 3 6 9 12 15 3 –10.2 –6.1 –5.7 –5.5 –5.5 6 –9.8 –4.4 –2.4 –2.1 –2.0 9 –9.8 –4.3 –1.5 –0.7 –0.5 12 –9.9 –4.3 –1.4 +0.0 +0.2 15 –9.9 –4.4 –1.4 +0.3 +0.7 The table shows the RELATIVE output power to the case of VCOGAIN = OUTTERM = 12. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 15 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 7.7.4 RFout Output Impedance The impedance of the RFout pin varies as a function of frequency, VCO_DIV, OUTTERM, VCOGAIN, DIVGAIN, and frequency. When in bypass mode (VCO_DIV = 1), the DIVGAIN word has no impact on the output impedance. When in divided mode (VCO_DIV>1), the VCOGAIN has no impact on the output impedance. This graphic shows how the input impedance varies as a function of frequency for both the bypass and divided cases. 4 2 Marker 1: 1 GHz 1 VCO_DIV =1 Marker 2: 2 GHz 3 2 1 VCO_DIV > 1 Marker 3: 3 GHz Marker 4: 4 GHz Start 30 MHz Stop 5 GHz Figure 17. RFout Output Impedance Table 4. RFout Output Impedance vs. VCOGAIN (Bypass Mode) (1) Freq. (MHz) (1) 16 VCOGAIN=3 VCOGAIN=6 VCOGAIN=9 VCOGAIN=12 VCOGAIN=15 REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY 50 3.8 2.1 5.5 1.9 7.3 1.8 9.5 1.7 10.1 1.7 100 4.8 4.1 6.1 3.9 7.8 3.7 9.8 3.6 10.3 3.6 200 5.4 5.7 6.8 6.0 8.7 6.3 10.9 6.5 11.4 6.6 400 5.5 9.4 7.5 10.0 9.8 10.6 12.4 11.0 13.1 11.0 600 5.8 15.1 8.1 15.4 10.7 15.7 13.7 15.7 14.5 15.6 800 7.0 20.7 9.6 20.8 12.6 20.8 15.8 20.3 16.7 20.1 1000 9.2 26.3 12.1 26.1 15.4 25.6 19.0 24.6 19.8 24.1 1200 10.7 28.6 13.4 27.9 16.3 26.9 19.3 25.5 20.0 25.0 1400 12.2 30.9 14.7 29.7 17.1 28.2 19.7 26.4 20.2 25.9 1600 13.7 33.2 15.9 31.5 18.0 29.5 20.1 27.4 20.5 26.8 1800 15.2 35.5 17.2 33.3 18.8 30.8 20.5 28.3 20.7 27.7 2000 14.5 39.5 16.4 37.4 17.9 35.0 19.6 32.5 19.8 31.9 2200 15.6 42.9 17.4 40.7 18.7 38.2 20.3 35.6 20.4 35.0 2400 14.2 47.6 16.0 45.3 17.4 42.8 19.0 40.1 19.2 39.4 2600 12.2 51.3 14.1 48.7 15.6 46.5 17.2 43.5 17.3 42.5 2800 11.5 57.9 13.7 55.3 15.3 52.4 17.0 49.0 17.1 48.3 3000 10.6 67.1 13.1 64.0 14.8 60.5 16.3 56.5 16.4 55.7 3200 13.1 77.3 15.7 73.2 17.3 69.0 18.4 64.2 18.4 63.3 3400 17.6 88.1 20.0 82.8 21.1 77.4 21.7 71.8 21.5 70.8 3600 29.0 96.0 30.6 90.2 30.9 83.6 30.2 76.7 29.8 75.6 3800 38.2 99.4 38.0 94.4 36.4 87.3 34.1 80.5 33.4 79.4 This is for the VCO divider in bypass mode (VCO_DIV=1) and the RFout pin powered up. OUTTERM was set to 12. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 RFout Output Impedance (continued) Table 4. RFout Output Impedance vs. VCOGAIN (Bypass Mode)(1) (continued) Freq. (MHz) VCOGAIN=3 VCOGAIN=6 REAL IMAGINA RY REAL 4000 43.5 106.0 4200 48.0 119.3 4400 62.4 4600 87.0 4800 5000 VCOGAIN=9 IMAGINA RY REAL 41.6 99.0 45.9 109.8 137.9 56.4 149.4 76.0 128.1 153.7 168.1 134.7 VCOGAIN=12 VCOGAIN=15 IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY 38.9 92.0 35.5 85.1 34.8 83.7 43.1 101.9 37.2 94.2 36.0 93.0 126.6 49.8 117.6 42.3 109.5 40.8 108.3 138.1 65.4 129.5 54.3 122.2 52.3 121.2 109.7 145.6 93.0 140.1 76.7 135.9 74.0 135.5 145.4 135.5 124.9 138.0 105.4 141.1 102.4 141.9 Table 5. RFout Output Impedance vs. OUTTERM (Bypass Mode) (1) Freq. (MHz) (1) OUTTERM=3 OUTTERM=6 OUTTERM=9 OUTTERM=12 TERM=15 REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY REAL IMAGINA RY 50 27.9 1.6 16.2 1.9 12.3 1.8 9.5 1.7 7.8 1.7 100 28.5 2.8 16.7 3.6 12.7 3.6 9.8 3.6 8.0 3.5 200 29.2 3.8 18.1 5.9 14.0 6.3 10.9 6.5 9.0 6.6 400 28.8 5.7 19.2 9.5 15.3 10.3 12.4 11.0 10.6 11.2 600 28.8 8.8 20.4 13.7 16.5 14.9 13.7 15.7 11.9 16.0 800 29.1 11.7 22.5 17.5 18.7 19.2 15.8 20.3 14.0 20.8 1000 28.6 13.4 22.8 19.2 19.3 21.2 16.5 22.5 14.6 23.1 1200 28.0 15.0 23.1 20.9 19.8 23.2 17.1 24.7 15.2 25.4 1400 27.5 16.7 23.3 22.7 20.4 25.2 17.7 26.9 15.8 27.7 1600 27.0 18.4 23.6 24.4 20.9 27.2 18.4 29.0 16.5 30.0 1800 26.4 20.1 23.9 26.1 21.4 29.2 19.0 31.2 17.1 32.3 2000 25.9 21.8 24.1 27.9 22.0 31.1 19.6 33.4 17.7 34.6 2200 25.3 23.5 24.4 29.6 22.5 33.1 20.3 35.6 18.3 36.9 2400 23.1 26.9 22.9 33.2 21.3 37.1 19.0 40.1 17.0 41.8 2600 20.1 29.3 20.5 35.4 19.3 39.6 17.2 42.9 15.1 44.9 2800 18.5 34.2 19.6 40.4 18.8 45.0 17.0 49.0 14.8 51.6 3000 16.6 40.6 18.1 46.9 17.8 51.9 16.3 56.5 14.3 59.7 3200 16.5 47.0 18.9 53.4 19.3 58.9 18.4 64.2 16.7 68.2 3400 17.1 53.8 20.4 60.1 21.8 65.8 21.7 71.8 20.4 76.6 3600 20.8 59.4 25.4 65.0 28.3 70.5 30.2 76.8 30.3 82.5 3800 22.0 64.9 27.3 69.7 31.1 74.6 34.1 80.5 35.4 86.1 4000 23.0 70.0 28.1 74.9 32.1 80.0 35.5 86.4 37.6 92.0 4200 23.7 77.9 28.6 82.8 32.8 87.7 37.0 94.2 39.9 100.9 4400 23.7 93.2 30.1 98.0 35.4 102.9 42.3 109.4 47.8 116.6 4600 27.3 107.4 36.6 112.0 44.8 116.3 54.3 122.2 62.6 128.9 4800 40.1 126.6 52.2 129.8 63.3 132.3 76.7 135.9 89.3 140.5 5000 61.4 142.8 76.2 143.3 89.5 142.3 105.5 141.0 121.0 140.5 The VCO divider was bypassed (VCO_DIV = 1) and the RFout pin was enabled. The VCOGAIN word was set to 12. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 17 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Table 6. RFout Output Impedance vs. DIVGAIN (Divided Mode) (1) Freq. (MHz) (1) DIVGAIN=3 DIVGAIN=6 DIVGAIN=9 DIVGAIN=12 DIVGAIN=15 REAL IMAGINAR Y REAL IMAGINAR Y REAL IMAGINAR Y REAL IMAGINAR Y REAL IMAGINAR Y 50 3.2 2.2 3.6 2.1 5.8 2.0 13.9 1.9 22.3 1.6 100 4.5 4.1 4.6 4.0 6.6 3.8 14.7 3.2 23.2 2.3 200 5.7 5.3 6.4 5.7 7.0 5.9 15.0 4.7 23.0 2.7 400 5.0 9.2 5.6 9.4 7.7 9.5 15.6 7.7 22.8 4.4 600 5.2 14.6 5.7 14.6 7.8 14.6 15.9 12.1 22.2 7.9 800 6.0 20.2 6.5 20.2 8.7 20.2 16.9 16.5 22.3 11.4 1000 7.9 25.7 8.4 25.7 10.7 25.5 18.7 20.5 22.9 14.6 1200 11.0 29.9 11.6 30.0 13.9 29.5 21.4 23.1 24.3 16.8 1400 13.2 32.3 13.9 32.3 16.1 31.7 22.5 24.3 23.9 18.2 1600 14.2 34.4 15.0 34.3 17.1 33.5 22.5 25.8 23.1 20.1 1800 13.9 37.2 14.6 37.0 16.7 36.2 21.6 28.2 21.7 22.9 2000 13.5 41.1 14.3 40.9 16.4 39.9 20.9 31.4 20.6 26.4 2200 14.8 45.1 15.6 44.7 17.8 43.6 21.7 34.5 20.9 29.7 2400 14.1 49.4 14.9 49.0 17.1 47.7 20.4 38.1 19.3 33.5 2600 12.4 52.1 13.2 51.6 15.5 50.1 18.2 40.4 16.8 36.2 2800 11.8 59.3 12.5 58.7 15.0 56.8 17.0 46.2 15.3 42.0 3000 10.7 68.3 11.5 67.6 14.0 65.2 15.2 53.4 13.0 49.2 3200 13.1 78.6 14.0 77.6 16.7 74.5 16.5 61.1 13.8 56.9 3400 18.1 89.6 18.9 88.4 21.6 84.4 19.4 69.2 16.0 65.1 3600 29.2 98.6 29.8 96.9 31.9 91.6 26.1 75.4 21.7 71.6 3800 36.0 105.8 36.5 103.9 37.8 97.5 28.9 81.1 24.0 77.8 4000 43.6 101.4 43.7 99.5 43.7 92.9 32.3 78.8 27.1 76.3 4200 40.6 122.9 40.8 120.3 40.6 111.8 26.6 94.7 20.7 91.8 4400 63.6 143.0 62.9 139.6 59.9 128.6 37.8 111.4 30.0 109.2 4600 90.9 155.3 88.8 151.4 81.1 139.6 49.9 125.8 40.3 124.9 4800 135.8 159.1 131.2 155.7 116.3 145.5 73.7 142.1 61.7 144.0 5000 179.4 135.1 173.2 133.9 153.3 131.4 107.1 147.7 94.5 155.2 This was done with RFout buffer powered up and with OUTTERM=12. VCO_DIV was set to 50. Table 7. RFout Output Impedance vs. OUTTERM (Divided Mode) (1) Freq.(MHz) (1) 18 OUTTERM=3 REAL IMAGINA RY OUTTERM=6 REAL IMAGINA RY OUTTERM=9 OUTTERM=12 REAL IMAGINA RY REAL IMAGINA RY OUTTERM=15 REAL IMAGINA RY 50 44.1 –0.3 31.8 1.0 21.2 1.7 14.0 1.9 9.3 2.0 100 44.9 –2.2 32.8 0.7 22.1 2.5 14.8 3.2 10.0 3.5 200 43.2 –7.2 33.2 –1.2 23.3 2.8 16.1 4.7 11.3 5.6 400 33.2 –8.1 28.5 –1.5 21.9 4.5 15.7 7.7 11.2 9.1 600 28.0 –3.8 25.7 1.8 21.4 8.0 15.9 12.1 11.4 13.9 800 25.1 1.1 24.0 5.6 21.7 11.5 16.9 16.5 12.5 19.0 1000 23.7 5.8 23.3 9.6 22.4 14.7 18.7 20.5 14.6 23.8 1200 23.5 9.3 23.7 12.4 23.8 16.7 21.4 23.1 17.7 27.2 1400 22.6 12.3 22.9 14.8 23.5 18.1 22.5 24.3 19.5 28.9 1600 21.5 15.3 21.8 17.4 22.6 20.0 22.5 25.8 20.2 30.5 1800 20.2 18.8 20.5 20.7 21.3 22.8 21.6 28.2 19.7 33.0 This was done in divided mode (VCO_DIV=50) with VCOGAIN=12. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Table 7. RFout Output Impedance vs. OUTTERM (Divided Mode)(1) (continued) Freq.(MHz) OUTTERM=3 OUTTERM=6 OUTTERM=9 OUTTERM=12 OUTTERM=15 2000 19.1 22.9 19.4 24.5 20.1 26.3 20.9 31.4 19.3 36.4 2200 19.4 26.4 19.7 28.0 20.5 29.6 21.7 34.5 20.6 39.8 2400 17.9 30.4 18.2 32.0 18.9 33.4 20.4 38.1 19.8 43.6 2600 15.7 33.3 15.9 34.9 16.5 36.1 18.2 40.4 17.9 45.7 2800 14.5 39.0 14.5 40.7 15.1 42.0 17.0 46.2 17.2 51.9 3000 12.7 46.1 12.6 47.9 12.9 49.2 15.2 53.4 15.8 59.5 3200 13.5 53.5 13.3 55.5 13.8 56.9 16.5 61.1 18.0 67.8 3400 15.5 61.3 15.4 63.5 15.9 65.0 19.4 69.2 22.0 76.5 3600 20.9 67.5 21.1 70.0 21.7 71.5 26.1 75.4 30.5 82.8 3800 22.7 73.3 23.1 76.0 23.9 77.6 28.9 81.1 34.7 88.2 4000 25.4 71.7 26.2 74.5 27.1 76.1 32.3 78.8 39.0 84.7 4200 19.0 86.1 19.8 89.5 20.7 91.4 26.6 94.7 34.6 101.8 4400 26.6 102.0 28.3 106.3 29.9 108.7 37.8 111.4 49.4 118.0 4600 34.9 116.4 37.8 121.5 40.1 124.1 49.8 125.9 65.3 130.6 4800 52.1 134.8 57.4 140.3 61.1 143.1 73.7 141.9 93.8 141.8 5000 78.5 147.4 87.4 152.0 93.3 154.0 107.2 148.0 129.0 138.6 7.7.4.1 OSCin and Fin Sensitivity This chart shows the typical sensitivity for a sine wave. Note that at lower frequencies, there is a constant slope that suggests that the part fails when the slew rate falls below 27 V/us. The electrical specifications call for a minimum of 150 V/us to ensure margin. Also, as some of the other performance graphs show, the OSCin slew rate has an impact on fractional spurs and phase noise as well. It is recommended to design to the electrical specifications, not the typical performance plots. Variation over voltage and temperature is typically very small and on the order than less ±1 dB. Figure 18. OSCin Sensitivity for Single-Ended SINE Wave Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 19 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 20 10 TA = 25oC Ensured Operating Range pEXTVCOin (dBm) 0 TA = 85oC -10 -20 TA = -40oC -30 -40 1000 0 2000 4000 3000 6000 5000 7000 8000 fExtVCOin (MHz) This plot shows the ExtVCOin sensitivity which applies only when the device is being used in External VCO mode. Variation over voltage is typically very small and on the order of less than ± 1 dB. Figure 19. SINE wave ExtVCOin Sensitivity Marker 1: 50 MHz Marker 2: 100 MHz Marker 3: 500 MHz XO = 1 1 4 2 3 Marker 4: 1000 MHz 2 XO = 0 Start 50 MHz Stop 1000 MHz 3 4 Figure 20. OSCin Input Impedance Table 8. OSCin Frequency FREQUENCY (MHz) 20 OSCin (NORMAL MODE) OSCin (XO MODE) OSCin# (NORMAL MODE) REAL IMAGINARY REAL IMAGINARY REAL IMAGINARY 1 3945.3 2261.6 9452.3 2182.1 3975.5 2287.0 5 4846.0 –189.6 2397.9 –916.7 4890.1 –150.1 10 4253.4 –1850.1 428.2 –1105.7 4297.4 –1886.7 20 2295.3 –2366.9 248.4 –591.8 2288.6 –2383.8 30 1290.0 –2087.0 187.1 –410.1 1304.3 –2079.1 40 847.9 –1716.1 163.5 –313.3 855.5 –1718.0 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Table 8. OSCin Frequency (continued) FREQUENCY (MHz) OSCin (NORMAL MODE) OSCin (XO MODE) OSCin# (NORMAL MODE) 50 581.3 –1464.9 147.9 –257.1 590.7 –1471.6 60 439.2 –1254.1 138.3 –219.0 449.4 –1264.2 70 337.9 –1105.7 131.1 –192.0 349.0 –1115.4 80 269.4 –983.6 127.0 –171.8 276.3 –989.1 90 223.4 –869.9 119.7 –158.0 231.9 –876.2 100 179.2 –776.8 114.5 –143.9 186.9 –783.9 200 52.4 –379.8 93.9 –85.1 54.3 –382.5 300 31.2 –247.0 80.9 –68.9 31.9 –247.4 400 23.5 –181.7 72.3 –58.1 23.8 –180.5 500 20.4 –140.5 65.1 –49.4 20.4 –138.4 600 18.4 –110.2 58.1 –42.1 18.2 –107.6 700 17.0 –88.0 51.9 –35.6 16.7 –85.3 800 15.8 –71.2 47.4 –29.5 15.7 –68.4 900 15.2 –57.6 43.6 –23.4 14.7 –56.3 1000 15.1 –45.2 40.9 –17.2 14.3 –44.7 Marker 1: 100 MHz 7 8 Marker 2: 1 GHz Marker 3: 2 GHz 6 Marker 4: 3 GHz Marker 5: 4 GHz 1 5 Marker 6: 5 GHz Marker 7: 6 GHz 2 Marker 8: 7 GHz 3 4 Start 100 MHz Stop 7000 MHz Figure 21. ExtVCOin Input Impedance Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 21 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Table 9. ExtVCOin Frequency 22 FREQUENCY REAL IMAGINARY 100 627.9 –1532.3 200 193.8 –852.6 400 56.4 –434.5 600 31.3 –287.4 800 23.2 –212.9 1000 17.8 –167.0 1200 15.4 –134.9 1400 14.0 –111.4 1600 12.8 –93.7 1800 11.8 –79.5 2000 11.2 –67.5 2200 10.7 –57.4 2400 10.2 –48.6 2600 10.5 –42.0 2800 9.1 –35.5 3000 7.8 –29.0 3200 7.2 –23.4 3400 6.6 –18.3 3600 5.9 –13.3 3800 5.3 –8.5 4000 5.0 –3.7 4200 4.5 –1.4 4400 4.0 0.9 4600 3.5 3.1 4800 2.6 7.7 5000 1.7 12.1 6000 0.9 26.7 7000 2.3 51.9 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 8 Parameter Measurement Information 8.1 Bench Test Setups DC Blocking Capacitor 10 MHz SMA Cable OSCin Signal Generator Device Under Test Semiconductor Parameter Analyzer SMA Cable CPout Pin 3.3 V Evaluation Board Power Supply Figure 22. Charge Pump Currents Test Setup The charge pump is tested in external VCO mode (MODE=1), although it is no external VCO hooked up. The CPout pin should be disconnected from the any external VCO tuning pin, external loop filter, and also the Vtune pin on the device. A signal is then applied to the OSCin pin to ensure that the R counter is oscillating. This signal does not have to be clean and the frequency is very critical. These currents at the CPout pin are typically measured with a semiconductor parameter analyzer. 8.1.1 Charge Pump Current Measurements In order to test the TRI-STATE current, the CPT bit is set to one and the current is measured. Aside from having no other sources of leakage attached to this pin, it is also important that the board be well cleaned before doing this test. The temperature and voltage at the charge pump can then be varied and the resulting leakage current is then recorded. Typically, the leakage currents are worst at higher temperatures and higher charge pump voltages. In order to test the source and sink currents, the CPT bit is set to active mode and the frequency is programmed to something much higher than can be achieved in order to force the charge pump to rail. The reason why this is necessary is that the duty cycle of the charge pump is not 100% unless it is forced against one of the rails. If the charge pump polarity bit (CPP) is set to positive, then the charge pump source current is measured. To measure the sink current, the CPT bit is set to negative. The part is then programmed and the charge pump will rail in one direction. The semiconductor parameter analyzer measures the current at a particular charge pump voltage. The phase detector polarity bit, CPP, can be toggled to test between the negative and positive charge pump gains. In order to test leakage, set the TRI-STATE bit, CPT, to 1 so that this can be measured. For the most accurate measurements, it is desirable that the CPout and Vtune pin are not shorted together for these measurements. Once these currents are measured, then the datasheet parameters can be calculated. A summary of these charge pump tests is given in the table below. MEASUREMENT PLL_R PLL_N CPG CPT Leakage Current X Source Current 1 Sink Current 1 4000 CPP X X 1 (TRI-STATE) X 4000 0 - 31 0 (Active) 1 (Positive) 0 - 31 0 (Active) 0 (Negative) Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 23 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 8.1.2 Charge Pump Current Definitions Figure 23. Charge Pump Current Definitions 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 0.4 volts for this part. Figure 24. Variation of Charge Pump Current Magnitude vs. Charge Pump Voltage Figure 25. Variation of Charge Pump Current Magnitude vs. Temperature Figure 26. Charge Pump Sink vs. Source Current Mismatch 24 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9 Detailed Description 9.1 Overview The LMX2541 is a low-noise synthesizer that can be used with the internal VCO or with an external VCO. The functional description gives more details on this. 9.2 Functional Block Diagrams Clean Reference Clock Dist. Dirty Input Clock PLL Ultra-Clean Local Oscillator PLL LMK04000 LMX2541 (Clock Jitter Cleaner) (RF Synthesizer) Figure 27. System Block Diagram Vtune VregVCO VrefVCO VCO VREG Charge Pump CPout Fast Lock FLout N Divider 4/5 Prescaler ExtVCOin 2-63 Divider Modulus Control I Comp Ftest/LD RFout VREG RFout MUX FRAC VREG RFoutEN 2X OSCin DATA MUX OSCin* VregFRAC R Divider Serial Interface Control CLK LE CE Figure 28. Functional Block Diagram Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 25 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.3 Feature Description The LMX2541 is a low power, high performance frequency synthesizer system which includes a PLL, Partially Integrated Loop Filter, VCO, VCO Divider, and Programmable Output Buffer. There are three basic modes that the device can be configured in: Full Chip Mode, External VCO Mode, and Divider Only Mode. Full chip mode is intended to be used with the internal VCO and PLL. There is also the option of External VCO mode, which allows the user to connect their own external VCO. Finally, there is Divider only, which is just the VCO divider and output buffer. The active blocks for these modes are described in Table 10: Table 10. Active Blocks AVAILABLE BLOCKS MODE PLL LOOP FILTER VCO VCO DIVIDER OUTPUT BUFFER Full Chip Yes Yes Yes Yes Yes External VCO Yes No No Yes Yes Divider Only No No No Yes Yes 9.3.1 PLL Reference Oscillator Input Pins There are three basic ways that the OSCin/OSCin* pins may be configured as shown in the table below: MODE DESCRIPTION XO BIT Crystal Device is used with a crystal oscillator 1 Single Ended Device is driven with a single-ended source, such as a TCXO. 0 Use this mode when driving with a differential signal, such as an LVDS signal. 0 Differential In addition to the way that the OSCin/OSCin* pins are driven, there are also bits that effect the frequency that the chip uses. The OSC_FREQ word needs to be programmed correctly, or the VCO may have issues locking to the proper frequency, because the VCO frequency calibration is based on this word. Table 11. Word Name and Function WORD NAME FUNCTION OSC_FREQ This needs to be set correctly if the internal VCO is used for proper calibration. OSC2X This allows the oscillator frequency to be doubled. The R divider is bypassed in this case. Higher slew rates tend to yield the best fractional spurs and phase noise, so a square wave signal is best for OSCin. Single ended mode and differential mode have similar results if a square wave is used to drive the OSCin pin. If using a sine wave, higher frequencies tend to work better due to their higher slew rates. 9.3.2 PLL R Divider The R divider divides the OSCin frequency down to the phase detector frequency. If the doubler is enabled, then the R divider is bypassed. 9.3.3 PLL Phase Detector and Charge Pump The phase detector compares the outputs of the R and N dividers and generates a correction current corresponding to the phase error. This charge pump current is software programmable to 32 different levels. The phase detector frequency, fPD, can be calculated as follows: fPD = fOSCin / R 26 (1) Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.3.4 PLL N Divider and Fractional Circuitry The N divider in the LMX2541 includes fractional compensation and can achieve any fractional denominator (PLL_DEN) from 1 to 4,194,303. The integer portion, PLL_N, is the whole part of the N divider value and the fractional portion, PLL_NUM / PLL_DEN, is the remaining fraction. PLL_N, PLL_NUM, and PLL_DEN are software programmable. So in general, the total N divider value, N, is determined by: N = PLL_N + PLL_NUM / PLL_DEN (2) The order of the delta-sigma modulator is programmable from integer mode to fourth order. There are also several dithering modes that are also programmable. In order to make the fractional spurs consistent, the modulator is reset any time that the R0 register is programmed. 9.3.5 Partially Integrated Loop Filter The LMX2541 integrates the third pole (formed by R3_LF and C3_LF) and fourth pole (formed by R4_LF and C4_LF) of the loop filter. The values for these integrated components can be programmed independently through the MICROWIRE interface. The larger the values of these components, the stronger the attenuation of the internal loop filter. The maximum attenuation can be achieved by setting the internal resistors and capacitors to their maximum value and the minimum attenuation can be attained by setting all of these to their minimum setting. This partially integrated loop filter can only be used in full chip mode. Charge Pump R4_LF CPout R3_LF Vtune C2_LF C1_LF R2_LF Figure 29. Partially Integrated Loop Filter 9.3.6 Low Noise, Fully Integrated VCO The LMX2541 includes a fully integrated VCO, including the inductors. The VCO (Voltage Controlled Oscillator) takes the voltage from the loop filter and converts this into a frequency. The VCO frequency is related to the other frequencies and divider values as follows: fVCO = fPD × N = fOSCin × N / R (3) In order to the reduce the VCO tuning gain and therefore improve the VCO phase noise performance, the VCO frequency range is divided into many different frequency bands. This creates the need for frequency calibration in order to determine the correct frequency band given a desired output frequency. The frequency calibration routine is activated any time that the R0 register is programmed. It is important that the OSC_FREQ word is set correctly to have this work correctly. The VCO also has an internal amplitude calibration algorithm to optimize the phase noise which is also activated any time the R0 register is programmed. The optimum internal settings for this are temperature dependent. If the temperature is allowed to drift too much without being re-calibrated, some minor phase noise degradation could result. For applications where this is an issue, the AC_TEMP_COMP word can be used to sacrifice phase noise at room temperature in order to improve the VCO phase noise over all temperatures. The maximum allowable drift for continuous lock, ΔTCL, is stated in the electrical specifications. For this part, a number of +125 C means the part will never lose lock if the part is operated under recommended operating conditions. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 27 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.3.7 Programmable VCO Divider The VCO divider can be programmed to any value from 2 to 63 as well as bypass mode if device is in full chip mode. In external VCO mode or divider mode, all values except bypass mode can be used for the VCO divider. The VCO divider is not in the feedback path between the VCO and the PLL and therefore has no impact on the PLL loop dynamics. After this programmable divider is changed, it may be beneficial to reprogram the R0 register to recallibrate the VCO . The frequency at the RFout pin is related to the VCO frequency and divider value, VCO_DIV, as follows: fRFout = fVCO / VCO_DIV (4) When this divider is enabled, there will be some far-out phase noise contribution to the VCO noise. Also, it may be beneficial for VCO phase noise to reprogram the R0 register to recalibrate the VCO if the VCO_DIV value is changed from bypass to divided, or vice-versa. The duty cycle for this divider is always 50%, even for odd divide values. Because of the architecture of this divider that allows it to work to high frequencies and always have a 50% duty cycle, there are a few extra considerations: • In divider only mode, there must be five clock cycles on the ExtVCOin pin after the divide value is programmed in order to cause the divide value to properly changed. It is fine to use more than 5 clock cycles for this purpose. • For a divide of 4 or 5 ONLY, the R4 register needs to be programmed one more time after the device is fully programmed in order synchronize the divider. Failure to do so will cause the VCO divider to divide by the wrong value. Furthermore, if the VCO signal ever goes away, as is the case when the part is powered down, it is necessary to reprogram the R4 register again to re-synchronize the divider. Furthermore, if the R0 register is ever programmed in full chip mode, it is also necessary to reprogram the R4 register. 9.3.8 Programmable RF Output Buffer The output power at the RFout pin can be programmed to various levels as well as on and off states. The output state of this pin is controlled by the RFoutEN pin as well as the RFOUT word. The RF output buffer can be disabled while still keeping the PLL in lock. In addition to this, the actual output power level of this pin can be adjusted using the VCOGAIN, DIVGAIN, and OUTTERM programming words. The reader should note that VCOGAIN controls the gain of the VCO buffer, not the tuning constant in of the VCO. 9.3.9 Powerdown Modes The LMX2541 can be powered up and down using the CE pin or the POWERDOWN bit. When the device is powered down, the programming and VCO calibration information is retained, so it is not necessary to reprogram the device when the device comes out of the powered down state (The one exception is when the VCO_DIV value is 4 or 5, which has already been discussed.). The following table shows how to use the bit and pin. CE PIN POWERDOWN BIT DEVICE STATE Low Don't Care Powered Down High 0 Powered Up 1 Powered Down The device can be programmed in the powerdown state. However, the VCO frequency needs to be changed when the device is powered up because the VCO calibration does not run in the powerdown state. Also, the special programming for VCO_DIV = 4 or 5 has to be done when the part is powered up. In order for the CE pin to properly power the device down when it is held low, the all registers in the device need to have been programmed at least one time. 9.3.10 Fastlock The LMX2541 includes the Fastlock feature that can be used to improve the lock times. When the frequency is changed, a timeout counter is used to engage the fastlock for a programmable amount of time. During the time that the device is in Fastlock, the FLout pin changes from high impedance to low, thus switching in the external resistor R2pLF with R2_LF as well as changing the internal loop filter values for R3_LF and R4_LF. 28 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 R4_LF R3_LF Vtune Charge Pump CPout C2_LF Fastlock Control FLout C1_LF R2pLF R2_LF Figure 30. Fastlock Schematic Table 12 shows the charge pump gain, loop filter resistors, and FLout pin change between normal operation and Fastlock. Table 12. Normal Operation vs Fastlock PARAMETER NORMAL OPERATION FASTLOCK Charge Pump Gain CPG FL_CPG Loop Filter Resistor R3_LF R3_LF FL_R3_LF Loop Filter Resistor R4_LF R4_LF FL_R4_LF FLout Pin High Impedance Low Once the loop filter values and charge pump gain are known for normal mode operation, they can be determined for fastlock operation as well. In normal operation, one cannot use the highest charge pump gain and still use fastlock because there will be no larger current to switch in. If the resistors and the charge pump current are done simultaneously, then the phase margin can be preserved while increasing the loop bandwidth by a factor of K as shown in the following table: Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 29 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com PARAMETER SYMBOL CALCULATION Charge pump gain in Fastlock FL_CPG Typically choose to be the largest value. Loop Bandwidth Multiplier K K= sqrt (FL_CPG/CPG) Internal Loop Filter Resistor FL_R3_LF FL_R3_LF = R3_LF / K Internal Loop Filter Resistor FL_R4_LF FL_R4_LF = R4_LF / K External Loop Filter Resistor R2pLF R2pLF = R2_LF / (K - 1) 9.3.11 Lock Detect The Ftest/LD pin of the LMX2541 can be configured to output a signal that gives an indication for the PLL being locked. There are two styles of lock detect; analog and digital. The analog lock detect signal is more of a legacy feature and consists a series of narrow pulses that correspond to when the charge pump comes on. These pulses can be integrated with an external RC filter to create generate a lock detect signal. Analog lock detect can be configured in a push-pull output or an open drain output. The analog open drain lock detect signal can be integrated with a similar RC filter and requires an additional pullup resistor. This pullup resistor can be much larger than the resistor in the RC filter in order to make unbalanced time constants for improved sensitivity. The digital lock detect function can also be selected for the Ftest/LD pin to give a logic level indication of lock or unlock. The digital lock detect circuitry works by comparing the difference between the phase of the inputs to the phase detector with a RC generated delay of ε. To indicate a locked state (Lock = HIGH) the phase error must be less than ε for 5 consecutive phase detector cycles. Once in lock (Lock = HIGH), the RC delay is changed to δ. To indicate an out of lock state (Lock = LOW), the phase error must become greater than δ. The values of ε and δ are programmable with the DLOCK word. 30 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 START LD = LOW (Not Locked) Phase Error < H NO YES Phase Error < H NO YES Phase Error < H NO YES Phase Error < H NO YES Phase Error < H NO YES LD = HIGH (Locked) NO Phase Error > G YES Figure 31. Lock Detect 9.3.12 Current Consumption The current consumption of the LMX2541 has many factors that influence it. Determining the current consumption for the entire device involves knowing which blocks are powered up and adding their currents together. The current in the electrical specifications gives some typical cases, but there could be some variation over factors such as the phase detector frequency. Also, the output buffer current can be impacted by the software controllable settings. By subtracting or adding combinations of the currents for the RFout buffer and VCO divider, the current consumption for the device can be estimated for any usable configuration. The currents for the buffer and VCO divider are as follows: BLOCK CURRENT (mA) RF Output Buffer ~ 40 (See Programmable Output Power with On/Off) VCO Divider 32 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 31 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.3.13 Fractional Spurs 9.3.13.1 Primary Fractional Spurs The primary fractional spurs occur at multiples of the channel spacing and can change based on the fraction. For instance, if the phase detector frequency is 10 MHz, and the channel spacing is 100 kHz, then this could be achieved using a fraction of 1/100. The fractional spurs would be at offsets that are multiples 100 kHz. 9.3.13.2 Sub-Fractional Spurs Sub-fractional spurs occur at sub-multiples of the channel spacing, Fch. For instance, in the above example, there could be a sub-fractional spur at 50 kHz. The occurrence of these spurs is dependent on the modulator order. Integer mode and the first order modulator never have sub-fractional spurs. If the fractional denominator can be chosen to avoid factors of 2 or 3, then there will also be no sub-fractional spurs. Sub-fractional spurs get worse for higher order modulators. Dithering tends to reduce sub-fractional spurs at the expense of increasing PLL phase noise. Table 13 provides guidance on predicting sub-fractional spur offset frequencies. Table 13. Sub-Fractional Spur Offset Frequencies vs. Modulator Order and Fractional Denominator Factors ORDER FRACTIONAL DENOMINATOR FACTORS NO FACTOR of 2 or 3 FACTOR of 2 but not 3 FACTOR of 3 but not 2 FACTOR of 2 and 3 None Integer Mode None None None 1st Order Modulator None None None None 2nd Order Modulator None Fch/2 None Fch/2 3rd Order Modulator None Fch/2 Fch/3 Fch/6 4th Order Modulator None Fch/4 Fch/3 Fch/12 9.3.14 Impact of VCO_DIV on Fractional Spurs Because the fractional and sub-fractional spur levels do not depend on output frequency, there is a big benefit to division. In general, every factor of 2 gives a 6 dB improvement to fractional spurs. Also, because the spur offset frequency is not divided, the channel spacing at the VCO can be also increased to improve the spurs. However, if the on-chip VCO is used, crosstalk can cause spurs at a frequency of fRFout mod fPD. Consider the following example of a 50 MHz phase detector frequency and VCO_DIV = 2. If the VCO is at 3000.1 MHz and divided by 2 to get 1500.05 MHz, there will be a spur at an offset of 50 kHz (1500.05 MHz mod 50 MHz). However, if the VCO frequency is at 3050.1 MHz, the output will be at 1525.05 MHz, but the spur will be at a much farther offset that can easily be filtered by the loop filter of 25.05 MHz (1525.05 MHz mod 50 MHz). 9.3.15 PLL Phase Noise 9.3.15.1 Figure 6, LMX2541SQ3740E Raw Phase Noise Measurement Plot Description The above plot demonstrates the PLL phase noise of the LMX2541SQ3700E operating at 3700 MHz output frequency, phase detector frequency of 100 MHz, and charge pump gain of 32X. The loop bandwidth was made as wide as possible to fully expose the PLL phase noise and reference source was a 100 MHz Wenzel crystal. This measurement was done in integer mode. To better understand the impact of using fractional mode, consult the applications section. The measured noise is the sum of the PLL 1/f noise and noise floor. At offsets below 1 kHz, the PLL 1/f noise dominates and changes at a rate of 10 dB/decade. The noise at 1 kHz is dominated by this 1/f noise and has a value of -103 dBc/Hz. In the 100 - 200 kHz offset range, the noise is -113.7 dBc/Hz and is dominated by the PLL noise floor. It can be shown that if the effects of the loop filter peaking and the 1/f noise are subtracted away from this measurement, it would be about 0.6 dB better. If the phase detector frequency is changed with the VCO frequency held constant, the PLL noise floor will change, but the 1/f noise will remain the same. If the VCO frequency is changed, both the 1/f noise and PLL noise floor change at a rate of 20 dB/decade. 32 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.3.15.2 Figure 37, LMX2541SQ2690 System Phase Noise Plot Description For this plot, a third order modulator with dithering disabled was used with a fractional denominator of 500000. The charge pump gain was 32X and the loop filter components were C1 = 2.2 nF, C2 = 22 nF, R2 = 470 Ω. The internal loop filter components were C3_LF = 20 pF, C4_LF = 100 pF, R3_LF = 1 kΩ, R4_LF = 200 Ω. The VCO frequency is 2720.1 MHz. The OSCin signal was a 500 MHz differential LVPECL output of the LMK04033. 9.3.15.3 Phase Noise of PLL Disregarding the impact of reference oscillator noise, loop filter resistor thermal noise, and loop filter shaping, the phase noise of the PLL can be decomposed into three components: flicker noise, flat noise, and fractional noise. These noise sources add in an RMS sense to produce the total PLL noise. In other words: LPLL(f) = 10·log(10(LPLL_flat(f) / 10 ) + 10(LPLL_flicker (f) / 10 )+ 10(LPLL_fractional(f) / 10 ) (5) Table 14. Potential Influencing Factors POTENTIAL INFLUENCING FACTORS SYMBOL f fVCO fPD KPD FRAC LPLL_flat(f) No Yes Yes Yes No LPLL_flicker(f) Yes Yes No Yes No LPLL_fractional(f) Yes No Yes No Yes The preceding table shows which factors of offset frequency (f), VCO frequency (fVCO), phase detector frequency (fPD), charge pump gain (KPD), and the fractional settings (FRAC) can potentially influence each phase noise component. The fractional settings include the fraction, modulator order, and dithering. For the flat noise and flicker noise, it is possible to normalize each of these noise sources into a single index. By normalizing these noise sources to an index, it makes it possible to calculate the flicker and flat noise for an arbitrary condition. These indices are reported in the electrical characteristics section and in the typical performance curves. Table 15. Noise Component NOISE COMPONENT INDEX RELATIONSHIP LPLL_flat(f) LNPLL_flat (1 Hz) LPLL_flat(f) = LNPLL_flat(1 Hz) + 20·log(N) + 10·log(fPD) LPLL_flicker(f) LNPLL_flicker (10 kHz) LPLL_flicker(f) = LNPLL_flicker(10 kHz) - 10·log(10 kHz / f) + 20·log( fVCO / 1 GHz ) The flat noise is dependent on the PLL N divider value (N) and the phase detector frequency (fPD) and the 1 Hz Normalized phase noise ( LNPLL_flat(1 Hz) ). The 1 Hz normalized phase noise can also depend on the charge pump gain as well. In order to make an accurate measurement of just the flat noise component, the offset frequency must be chosen sufficiently smaller then the loop bandwidth of the PLL, and yet large enough to avoid a substantial noise contribution from the reference and PLL flicker noise. This becomes easier to measure for lower phase detector frequencies. The flicker noise, also known as 1/f noise, can be normalized to 1 GHz carrier frequency and 10 kHz offset, LNPLL_flicker(10 kHz). Flicker noise can dominate at low offsets from the carrier and has a 10 dB/decade slope and improves with higher charge pump currents and at higher offset frequencies . To accurately measure the flicker noise it is important to use a high phase detector frequency and a clean crystal to make it such that this measurement is on the 10 dB/decade slope close to the carrier. LPLL_flicker(f) can be masked by the reference oscillator performance if a low power or noisy source is used. An alternative way to interpret the flicker noise is the 1/f noise corner, fcorner. This would be the offset frequency where the flat noise and flicker noise are equal. This corner frequency changes as a function of the phase detector frequency and can be related to the flat and flicker noise indices as shown below. fcorner = 10( (LNPLL_flicker(10 kHz) - LNPLL_flat(1 Hz) - 140) / 10 ) × fPD (6) Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 33 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Based on the values for LNPLL_flicker(10 kHz) and LNPLL_flat(1Hz) as reported in the electrical specifications, the corner frequency can be calculated. For example, one of the plots in the typical performance characteristics shows the phase noise with a 100 MHz phase detector frequency and 32X charge pump gain. In this case, this corner frequency works out to be 0.000123 × 100 MHz = 12.3 kHz. KPD LNPLL_flicker(10 kHz) LNPLL_flat(1 Hz) fcorner 1X -116.0 dBc/Hz -220.8 dBc/Hz 0.000302 × fPD 32X -124.5 dBc/Hz -225.4 dBc/Hz 0.000123 × fPD For integer mode or a first order modulator, there is no fractional noise (disregarding fractional spurs). For higher order modulators, the fractional engine may or may not add significant phase noise depending on the fraction and choice of dithering. 9.3.16 Impact of Modulator Order, Dithering, and Larger Equivalent Fractions on Spurs and Phase Noise To achieve a fractional N value, an integer N divider is modulated between different values. This gives rise to three main degrees of freedom with the LMX2541 delta-sigma engine: the modulator order, dithering, and the way that the fractional portion is expressed. The first degree of freedom, the modulator order, can be selected as zero (integer mode), one, two, three, or four. One simple technique to better understand the impact of the deltasigma fractional engine on noise and spurs is to tune the VCO to an integer channel and observe the impact of changing the modulator order from integer mode to a higher order. A higher fractional modulator order in theory yields lower primary fractional spurs. However, this can also give rise to sub-fractional spurs in some applications. The second degree of freedom is dithering. Dithering seeks to improve the sub-fractional spurs by randomizing the sequence of N divider values. In theory, a perfectly randomized sequence would eliminate all sub-fractional spurs, but add phase noise by spreading the energy that would otherwise be contained in the spurs. The third degree of freedom is the way that the fraction is expressed. For example, 1/10 can be expressed as a larger equivalent fraction of 100000/1000000. Using larger equivalent fractions tends to increase randomization similar to dithering. In general, the very low phase noise of the LMX2541 exposes the modulator noise when dithering and large fractions are used, so use these with caution. The avid reader is highly encouraged to read application note 1879 for more details on fractional spurs. The following table summarizes the relationships between spur types, phase noise, modulator order, dithering and fractional expression. ACTION INCREASE MODULATOR ORDER INCREASE DITHERING USING LARGER EQUIVALENT FRACTIONS Phase Noise WORSE (But only for larger fractions or more dithering) WORSE WORSE Primary Fractional Spur BETTER NO IMPACT NO IMPACT Sub-Fractional Spurs WORSE (Creates more sub -fractional spurs) BETTER BETTER NOISE / SPUR TYPE 9.3.17 Modulator Order In general, the fractional mode of the PLL enables the use of a higher phase detector frequency relative to the channel spacing, which enables the in-band noise of the PLL to be lower. The choice of modulator order to be used in fractional mode is based on how much higher fPD can be made relative to the channel spacing and the acceptable spur levels. The LMX2541 has a programmable modulator order which allows the user to make a trade-off between PLL noise and primary and sub-fractional spur performance. The following table provides some general guidelines for choosing modulator order: Note that the spurs due to crosstalk will not be impacted by modulator order. 34 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 ORDER Integer Mode 1st Order Modulator GUIDELINES FOR USE • • Use if fPD can be made very high without using a fractional N value. Use if it is not desired to make fPD higher using a fractional N value. This could be the case if the loop bandwidth is very narrow and smaller loop filter capacitors are desired. • Use 1st order if fPD can be increased by at least a factor of four over the integer case and fractional spur frequencies and levels are acceptable. If the channel spacing is 5 MHz or greater, the 1st order modulator may provide better spur performance than integer mode. • • 2nd Order Modulator 3rd Order Modulator 4th Order Modulator • • If the spurs of the 1st order modulator are unacceptable, use a higher order modulator. If the spurious components are due to crosstalk they will not be improved by increasing modulator order. In this case , use the lowest order modulator that gives acceptable performance. Use if the spurs of the 1st order modulator are unacceptable. In general, use the lowest order modulator unless a higher order modulator yields an improvement in primary fractional spurs. If the spurious components are due to crosstalk, they will not be improved by increasing the modulator order. 9.3.18 Programmable Output Power with On/Off The power level of the RFout pin is programmable, including on/off controls. The RFoutEN pin and RFOUT word can be used to turn the RFout pin on and off while still keeping the VCO running and in lock. In addition to on/off states, the power level can also be programmed in various steps using the VCOGAIN, DIVGAIN, and OUTTERM programming words. There are tables in the Typical Characteristics section that discuss the impact of these words on the output power. In addition to impacting the output power, these words also impact the current consumption of the device. This data was obtained as an average over all frequencies. In general, it is desirable to find the combination of programming words that gives the lowest current consumption for a given output power level. All numbers reported are relative to the case of VCOGAIN = OUTTERM = 12. According to this data, using a VCOGAIN or OUTTERM value of 12 or greater yields only a small increase in output power, but a large increase in current consumption. Table 16. Change in Current Consumption in Bypass Mode as a Function of VCOGAIN and OUTTERM VCOGAIN OUTTERM 3 6 9 12 15 3 -26.0 -22.3 -18.6 -15.1 -11.8 6 -18.5 -15.5 -12.6 -9.7 -6.9 9 -11.1 -9.0 -6.9 -4.7 -2.5 12 -3.8 -2.6 -1.4 +0.0 +1.5 15 +3.3 +3.7 +4.0 +4.5 +5.3 Table 17. Change in Current Consumption in Divided Mode as a Function of DIVGAIN and OUTTERM DIVGAIN OUTTERM 3 6 9 12 15 3 -24.4 -21.7 -18.7 -15.9 -13.3 6 -16.2 -14.6 -12.6 -10.1 -8.0 9 -8.3 -7.6 -6.8 -5.0 -3.2 12 -0.5 -0.7 -0.7 +0.0 +1.3 15 +7.1 +6.0 +5.2 +4.9 +5.6 9.3.19 Loop Filter Loop filter design can be rather complicated, but there are design tools and references available at www.ti.com. The loop bandwidth can impact the size of loop filter capacitors and also how the phase noise is filtered. For optimal integrated phase noise, choose the bandwidth to be about 20% wider than the frequency where the inband PLL phase noise (as described in PLL Phase Noise) and open loop VCO noise cross. This optimal loop bandwidth may need adjustment depending on the application requirements. Reduction of spurs can be achieved Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 35 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com by reducing the loop bandwidth. On the other hand, a wider loop bandwidth may be required for faster lock time. Note that using the integrated loop filter components can lead to a significant restriction on the loop bandwidth and should be used with care. 2 kΩ for R3_LF and R4_LF is a good starting point. If the integrated loop filter restricts the loop bandwidth, then first try to relieve this restriction by reducing the integrated loop filter resistors and then reduce the capacitors only if necessary. 9.3.20 Internal VCO Digital Calibration Time When the LMX2541 is used in full chip mode, the integrated VCO can impact the lock time of the system. This digital calibration chooses the closest VCO frequency band, which typically gets the device within a frequency error 10 MHz or less of the final settling frequency, although this final frequency error can change slightly between the different options of the LMX2541. Once this digital calibration is finished, this remaining frequency error must settle out, and this remaining lock time is dictated by the loop bandwidth. Based on measured data, this digital calibration time can be approximated by the following formula: LockTime = A + B/CLK + C·ΔF + D·( ΔF/CLK ) (7) SYMBOL1 VALUE UNITS Locktime Varies µs A 30 μs B 3800 None C 0.1 us/MHz D 2 µs ΔF Varies MHz CLK fOSCin / 2 for 0 ≤ OSC_FREQ ≤ 63 fOSCin / 4 for 64 ≤ OSC_FREQ ≤ 127 fOSCin / 8 for 128 ≤ OSC_FREQ None For example, consider the LMX2541SQ3320E changing from 3600 to 3400 with an OSCin frequency of 100 MHz. In this case, ΔF = 200 (direction of frequency change does not matter), fOSCin = 100 MHz, and OSC_FREQ=100. The calibration circuitry is run at a clock speed of CLK = 100 MHz / 4 = 25 MHz. When this values are substituted in the formula, the resulting lock time is 218 μs. After this time, the VCO will be within about 10 MHz of the final frequency and this final frequency error will settle out in an analog fashion. This final frequency error can be slightly different depending on which option of the LMX2541 is being used. 9.4 Device Functional Modes 9.4.1 External VCO Mode The LMX2541 also has provisions to be driven with an external VCO as well. In this mode, the user has the option of using the RFout pin output, although if this pin is used, the VCO input frequency is restricted to 4 GHz. If not used, the RFout pin should be left open. The VCO input is connected to the ExtVCOin pin. Because the internal VCO is not being used, the part option that is being used does not have a large impact on phase noise or spur performance. It is also possible to switch between both Full Chip mode and External VCO mode. 9.4.2 Digital FSK Mode The LMX2541 supports 2-level digital frequency shift keying (FSK) modulation. The bit rate is limited by the loop bandwidth of the PLL loop. As a general rule of thumb, it is desirable to have the loop bandwidth at least twice the bit rate. This is achieved by changing the N counter rapidly between two states. The fractional numerator and denominator are restricted to a length of 12 bits. The 12 LSB’s of the numerator and denominator set the center frequency, Fcenter, and the 10 MSB’s of the numerator set the frequency deviation, Fdev. The LMX2541 has the ability to switch between two different numerator values based on the voltage at the DATA pin. When DATA is low, the output frequency will be Fcenter – Fdev and when the DATA pin is high the output frequency will be Fcenter + Fdev. A limitation of the FSK mode is the frequency deviation cannot cause the N counter to cross integer boundaries. When using FSK mode, the FDM bit needs to be set to zero. 36 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.5 Programming There are several other considerations for programming: • The DATA is clocked into a shift register on each rising edge of the CLK signal. On the rising edge of the LE signal, the data is sent from the shift registers to an actual counter. • A slew rate of at least 30 V/μs is recommended for the CLK, DATA, and LE signals. • After the programming is complete, the CLK, DATA, and LE signals should be returned to a low state. • When using the part in Full Chip Mode with the Integrated VCO, LE should be kept high no more than 1 us after the programming of the R0 register. Failure to do so may interfere with the digital VCO calibration. • If the CLK and DATA lines are toggled while the in VCO is in lock , as is sometimes the case when these lines are shared with other parts, the phase noise may be degraded during the time of this programming. 9.5.1 General Programming Information The LMX2541 is programmed using several 32-bit registers used to control the LMX2541 operation. A 32-bit shift register is used as a temporary register to indirectly program the on-chip registers. The shift register consists of a data field and an address field. The last 4 register bits, CTRL[3:0] form the address field, which is used to decode the internal register address. The remaining 28 bits form the data field DATA[27:0]. While LE is low, serial data is clocked into the shift register upon the rising edge of clock (data is programmed MSB first). When LE goes high, data is transferred from the data field into the selected register bank. For initial device programming the register programming sequence must be done in the order as shown in the register map. The action of programming register R7 and bringing LE low resets all the registers to default values, including hidden registers. The programming of register R0 is also special for the device when operating in full chip mode because the action of programming either one of these registers activates the VCO calibration. In addition to changing the values of various words, the programming of certain registers triggers certain events as described in the table below: PROGRAMMING EVENT Action of programming register R7 and bring LE low EVENT TRIGGERED Resets all registers, including hidden ones, to a default state CONFIGURATIONS WHERE IT HAS AN IMPACT SIGNIFICANCE All This needs to be the first programming step for all configurations. If register R7 is ever programmed again, all programming information will be reset to the default state. The VCO calibration tunes the VCO to the correct frequency band and optimizes the phase noise. It is necessary whenever the internal VCO frequency is changed. Also, if the temperature drifts considerably, then this calibration can better optimize the phase noise. Action of programming register R0 and bringing LE low Activates the VCO calibration Only in Full Chip Mode Action of programming register R4 Synchronizes the VCO Divider Only when the RFout pin ins enabled and Consult the Feature Description for more details. the VCO divider is set to 4 or 5 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 37 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.6 Register Maps The following table lists the registers as well as the order that they should be programmed. Register 7 is programmed first and the action of programming register R7 resets all the registers after the LE pin is pulled to a low state. Register R0 is programmed last because it activates the VCO calibration. The one exception to this is when the VCO_DIV value is 4 or 5. Consult the programming section on VCO_DIV for more details. Table 18. Register Map 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 DATA[27:0] 0 C0 0 1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 R13 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 R12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 R9 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 1 0 0 1 R8 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 1 1 0 0 1 1 1 1 0 0 0 0 1 1 0 0 1 0 1 0 1 0 0 0 0 1 1 0 0 1 0 PLL_R[11:0] 0 0 0 1 PLL_N[11:0] 0 0 0 0 R5 1 R4 0 0 0 1 0 0 0 0 FL_CPG[4:0] C4_LF[3:0] 0 FS K 0 0 DLOCK[2:0] R2 0 0 0 0 R1 0 0 0 0 0 CP T 0 0 1 1 1 FL_RF_LF[2:0] FL_R3_LF[2:0] R4_LF[2:0] R3_LF[2:0] C3_LF[3:0] R3 DITH[1:0] ORDER[2:0] FD M OS C _2X 1 VCOGAIN[3:0] OUTTERM[3:0] VCO_DIV_OPT [2:0] AC_TEMP_COMP[4:0] RFOUT[1 :0] DIVGAIN[3:0] FL_TOC[13:0] VCO_DIV[5:0] CP P MUX[3:0] OSC_FREQ[7:0] CPG[4:0] DEN[21:0] PLL_NUM[21:16] R0 38 0 1 1 C1 0 0 0 2 C2 R7 R6 0 3 C3 PLL_N[17:12] PLL_NUM[15:0] Submit Documentation Feedback XO PO WE MODE[1: R 0] DO WN Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1 Register R7 Although Register 7 has no elective bits to program, it is very important to program this register because the action of doing so with the bit sequence shown in the register map resets all the registers, including hidden registers with test bits that are not disclosed. Register 7 should always be programmed first, because it will reset all other programming information. The register reset occurs only after the LE signal has transitioned from low to high and back to low again. 9.6.1.1 Register R13 This register needs to be programmed only in the event that the RFout pin is being used and VCO_DIV = 1. 9.6.1.1.1 VCO_DIV_OPT[2:0] This word optimizes the RFout power level based on the VCO_DIV and VCO_GAIN words. 9.6.1.2 CONDITION VCO_DIV_OPT COMMENTS RFout Pin Disabled OR VCO_DIV>1 OR VCO_GAIN12 4 Register R12 This register needs to be programmed as shown in the register map in the event that the internal VCO is being used. When using external VCO mode, this register does not need to be programmed. 9.6.1.3 Register R9 Program this register as shown in the register map. 9.6.1.4 Register R8 9.6.1.4.1 AC_TEMP_COMP[4:0] This word optimizes the VCO phase noise for possible temperature drift. When the VCO frequency is changed, the internal tuning algorithm optimizes the phase noise for the current temperature. In fixed frequency applications, temperature drift may lead to sub-optimal phase noise over time. In dynamic frequency applications, the re-tuning of the VCO frequency overcomes this problem because the phase noise is re-optimized each time the VCO frequency is changed. The AC_TEMP_COMP word can be used to optimize the VCO phase noise for temperature drift for these different scenarios. The following table indicates which values of this word should be used for each scenario. AC_TEMP_COMP APPLICATION TYPE 5 Dynamic Frequency 24 Fixed Frequency All Other States Invalid Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 39 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.6.1.5 Register R6 Register R6 has words that impact the output power of the RFout pin. 9.6.1.5.1 RFOUT[1:0] - RFout enable pin This word works in combination with the RFoutEN Pin to control the state of the RFout pin. RFOUT RFoutEN PIN RFout PIN STATE 0 Don't Care Disabled 2 Don't Care Enabled Low Disabled High Enabled 1 or 3 9.6.1.5.2 DIVGAIN[3:0], VCOGAIN[3:0], and OUTTERM[3:0] - Power Controls for RFout These three words may be programmed in a value from 0 to 15 and work in conjunction to control the output power level of the RFout pin. Increasing any of these values increases the output power at the expense of higher current consumption of the buffer. Although there may be more than one way to get the same output power, some combinations may have lower current. The typical performance characteristics show these trade-offs. The default setting for all these bits is 12. The value of VCO_DIV determines which two of these three words have an impact. VCO_DIV BITS that IMPACT POWER 1 (Bypass) OUTTERM, VCOGAIN >1 (Not Bypass) OUTTERM, DIVGAIN 9.6.1.6 Register R5 This register controls the fastlock mode which enables a wider loop bandwidth when the device is changing frequencies. 9.6.1.6.1 FL_TOC[13:0] -- Time Out Counter for FastLock When the value of this word is 3 or less, FastLock time out counter is disabled, and the FLout pin can be used for general purpose I/O. When this value is 4 or greater, the time out counter is engaged for the amount of phase detector cycles shown in the table below. TOC VALUE FLout PIN STATE FASTLOCK ENGAGEMENT TIME 0 High Impedance Disabled 1 Low Always Engaged 2 Low Disabled 3 High Disabled 4 Low Engaged for 4 × 2 Phase Detector Cycles . . . Low Engaged for 16383 × 2 Phase Detector Cycles 16383 When this count is active, the FLout Pin is grounded, the FastLock current is engaged, and the resistors R3 and R4 are also potentially changed. The table below summarizes the bits that control various values in and out of FastLock. 40 CHARGE PUMP CURRENT FastLock STATE FLout Not Engaged High Impedance CPG R3_LF R4_LF Engaged Grounded FL_CPG FL_R3_LF FL_R4_LF Submit Documentation Feedback R3_LF VALUE R4_LF VALUE Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1.6.2 FL_R3_LF[2:0] -- Value for Internal Loop Filter Resistor R3 During Fastlock FL_R3_LF VALUE R3 RESISTOR DURING FASTLOCK (kΩ) 0 Low ( 200 Ω ) 1 1 2 2 3 4 4 16 5-7 Reserved 9.6.1.6.3 FL_R4_LF[2:0] -- Value for Internal Loop Filter Resistor R4 During Fastlock FL_R4_LF VALUE R3 RESISTOR DURING FASTLOCK (kΩ) 0 Low ( 200 Ω ) 1 1 2 2 3 4 4 16 5-7 Reserved 9.6.1.6.4 FL_CPG[4:0] -- Charge Pump Current for Fastlock When FastLock is enabled, this is the charge pump current that is used for faster lock time. FL_CPG FASTLOCK CHARGE PUMP STATE TYPICAL FASTLOCK CHARGE PUMP CURRENT at 3.3 VOLTS (µA) 0 1X 100 1 2X 200 2 3X 300 3 4X 400 ... ... ... 31 32X 3200 9.6.1.7 Register R4 This register controls miscellaneous functions of the device. The action of programming the R4 register also synchronizes the VCO divider, which is necessary when VCO_DIV = 4 or 5. 9.6.1.7.1 OSC_FREQ [7:0] -- OSCin Frequency for VCO Calibration Clocking This word is used for the VCO frequency calibration. This word should be set to the OSCin frequency rounded to the nearest MHz. OSC_FREQ OSCin FREQUENCY 0 Illegal State 1 1 MHz 2 2 MHz ... ... 255 255 MHz and higher Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 41 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.6.1.7.2 VCO_DIV[5:0] - VCO Divider The output of the VCO is divided by the value of VCO_DIV, which can range from 1 (Bypass Mode) to 63 and all values in between with the limitation that the VCO divider can only be set to bypass mode when the device is operating in full chip mode. When the VCO divider is set to 4 or 5 ONLY, there is one extra programming step required to synchronize the VCO divider. Consult the Feature Description for more details. VCO_DIV VCO OUTPUT DIVIDE 0 n/a Illegal State 1 Bypass Mode This state only available for MODE=Full Chip Mode 2 Divide by 2 3 Divide by 3 4 Divide by 4 5 Divide by 5 6 Divide by 6 ... ... 62 Divide by 62 63 Divide by 63 COMMENTS Extra programming is required for divide by 4 and divide by 5 only. Refer to the Feature Description for more details. 9.6.1.7.3 R3_LF[2:0] -- Value for Internal Loop Filter Resistor R3 This word controls the state of the internal loop filter resistor R3_LF when the device is in Full Chip Mode and Fastlock is not active. R3_LF VALUE R3 RESISTOR DURING FASTLOCK (kΩ) 0 Low ( 200 Ω ) 1 1 2 2 3 4 4 16 5-7 Reserved 9.6.1.7.4 R4_LF[2:0] -- Value for Internal Loop Filter Resistor R4 This word controls the state of the internal loop filter resistor R4_LF when the device is in Full Chip Mode and Fastlock is not active. 42 R4_LF VALUE R3 RESISTOR DURING FASTLOCK (kΩ) 0 Low ( 200 Ω ) 1 1 2 2 3 4 4 16 5-7 Reserved Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1.7.5 C3_LF[3:0] -- Value for C3 in the Internal Loop Filter This word controls the state of the internal loop filter resistor C3_LF when the device is Full Chip Mode. C3_LF C3 (pF) 0 0 1 1 2 5 3 6 4 10 5 11 6 15 7 16 8 20 9 21 10 25 11 26 12 30 13 31 14 35 15 36 9.6.1.7.6 C4_LF[3:0] -- Value for C4 in the Internal Loop Filter This word controls the state of the internal loop filter resistor C4_LF when the device is Full Chip Mode. C4_LF C4 (pF) 0 0 1 5 2 20 3 25 4 40 5 45 6 60 7 65 8 100 9 105 10 120 11 125 12 140 13 145 14 160 15 165 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 43 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.6.1.8 Register R3 This register controls miscellaneous features of the device. 9.6.1.8.1 MODE[1:0] -- Operational Mode The LMX2541 can be run in several operational modes as listed in the table below: MODE NAME DIVIDER PLL VCO 0 Full Enabled Enabled Enabled 1 External VCO Enabled Enabled Disabled 2 Divider Only Enabled Disabled Disabled 3 Test (Reserved) Enabled Enabled Enabled 9.6.1.8.2 Powerdown -- Powerdown Bit Enabling this bit powers down the entire device, although register and VCO calibration information is retained. 9.6.1.8.3 XO - Crystal Oscillator Mode Select When this bit is enabled, a crystal with appropriate load capacitors can be attached between the OSCin and OSCin* pins in order to form a crystal oscillator. 9.6.1.8.4 CPG[4:0] -- Charge Pump Current This word programs the charge pump current gain. The current is programmable between 100 uA and 3.2 mA in 100 uA steps. 44 CPG CHARGE PUMP STATE TYPICAL CHARGE PUMP CURRENT (µA) 0 1X 100 1 2X 200 2 3X 300 ... 3 4X ... ... 31 32X Submit Documentation Feedback 3200 Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1.8.5 MUX[3:0] -- Multiplexed Output for Ftest/LD Pin The MUX[3:0] word is used to program the output of the Ftest/LD Pin. This pin can be used for a general purpose I/O pin, a lock detect pin, and for diagnostic purposes. When programmed to the digital lock detect state, the output of the Ftest/LD pin will be high when the device is in lock, and low otherwise. The output voltage level of the Ftest/LD is not equal to the supply voltage of the device, but rather is given by VOH and VOL in the electrical characteristics specification. Because the Ftest/LD pin is close to the OSCin pin, the state of this pin can have an impact on the performance of the device. If any of the diagnostic modes (8-13) are used, the OSCin sensitivity can be severely degraded, so these should only be used for diagnostic purposes. The fractional spurs can also be impacted a little by the MUX programming word. The Push-Pull digital lock detect modes, like mode 3, tend to have the best fractional spurs, so these states are recommended, even if the digital lock detect function is not needed. MUX OUTPUT TYPE FUNCTION 0 High Impedance Disabled 1 Push-Pull Logical High State 2 Push-Pull Logical Low State 3 Push-Pull Digital Lock Detect 4 Push-Pull Inverse Digital Lock Detect 5 Open Drain Digital Lock Detect 6 Open Drain Analog Lock Detect 7 Push-Pull Analog Lock Detect 8 Push-Pull N Divider 9 Push-Pull N Divider / 2 10 Push-Pull R Divider 11 Push-Pull R Divider / 2 12 Push-Pull PFD Up 13 Push-Pull PFD Down 14-15 N/A Reserved COMMENTS GENERAL PURPOSE I/O MODES LOCK DETECT MODES Consult Feature Description for more details State 3 is recommended for optimal spurious performance. DIAGNOSTIC MODES These allow the user to view the outputs of the N divider, R divider, and phase frequency detector (PFD) and are intended only for diagnostic purposes. Typically, the output is narrow pulses, but when the output is divided by 2, there is a 50% duty cycle. The use of these modes (including R Divider) can degrade the OSCin sensitivity. 9.6.1.8.6 CPP - Charge Pump Polarity This bit sets the polarity of the phase detector. CPP CHARGE PUMP POLARITY TYPICAL APPLICATIONS 0 Negative Full Chip Mode External VCO Mode with an inverting active loop filter. 1 Positive External VCO Mode with a passive loop filter. 9.6.1.8.7 OSC2X-- OSCin Frequency Doubler Enabling this bit doubles the OSCin frequency. This is useful in achieving a higher phase detector frequency to improve PLL phase noise, push out noise from the delta-sigma modulator, and sometimes reduce fractional spurs . Note that when this bit is enabled, the R divider is bypassed. OSC_2X STATE 0 Normal 1 OSCin frequency is doubled Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 45 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 9.6.1.8.8 FDM - Extended Fractional Denominator Mode Enable Enabling this bit allows the fractional numerator and denominator to be expanded from 10 bits to 22 bits. In 10-bit mode, only the first 10 bits of the fractional numerator and denominator are considered. When using FSK mode, this bit has to be disabled. FDM FRACTIONAL MODE 0 10-bit 1 (Default) 22-bit 9.6.1.8.9 ORDER[2:0] -- Delta-Sigma Modulator Order This word determines the order of the delta-sigma modulator in the PLL. In general, higher order fractional modulators tend to reduce the primary fractional spurs that occur at increments of the channel spacing, but can also create spurs that are at a fraction of the channel spacing. The optimal choice of modulator order is very application specific, however, a third order modulator is a good starting point. The first order modulator has no analog compensation or dithering DELTA-SIGMA MODULATOR MODE 0 Disabled Integer 1 First Order 2 Second Order 3 Third Order 4 Fourth Order 5-7 Illegal States ORDER COMMENTS Allows larger N Counter This has no analog compensation or dithering Fractional Traditional Delta-Sigma Operation n/a n/a 9.6.1.8.10 DITH[1:0] -- Dithering Dithering randomizes the delta-sigma modulator output. This reduces sub-fractional spurs at the expense of adding phase noise. In general, it is recommended to keep the dithering strength at None or Weak for most applications. Dithering should never be used when the device is used in integer mode or a first order modulator. When using dithering with the other delta-sigma modulator orders, it is beneficial to disable it in the case where the fractional numerator is zero, because it can actually create sub-fractional spurs. DITH DITHERING STRENGTH 0 Weak 1 Medium 2 Strong 3 Disabled 9.6.1.8.11 CPT - Charge Pump TRI-STATE When this bit is enabled, the charge pump is at TRI-STATE. The TRI-STATE mode could be useful for open loop modulation applications or as diagnostic tool for measuring the VCO noise, but is generally not used. 46 CPT CHARGE PUMP 0 Normal Operation 1 TRI-STATE Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1.8.12 DLOCK[2:0] - Controls for Digital Lock Detect This word controls operation of the digital lock detect function through selection of the window sizes (ε and δ). In order to indicate the PLL is locked, there must be 5 consecutive phase detector output cycles in which the time offset between the R and N counter outputs is less than ε. This will cause the Ftest/LD pin output to go high. Once lock is indicated, it will remain in this state until the time offset between the R and N counter outputs exceeds δ. For this device, ε and δ are the same. If the OSCin signal goes away, the digital lock detect circuit will reliably indicate an unlocked condition. Consult the Feature Description for more details. A larger window size makes the lock detect circuit less sensitive, but may be necessary in some situations to reduce chattering. DLOCK WINDOW SIZE (ε and δ) 0 (Default) 3.5 1 5.5 2 7.5 3 9.5 4 11.5 5 13.5 6 -7 Reserved There are restrictions when using digital lock detect, based on the phase detector frequency (fPD), Modulator Order (ORDER), and VCO frequency (fVCO). The first restriction involves a minimum window size (εmin), the second one involves a maximum window size (εmax), and the third involves further restrictions on the maximum phase detector frequency that are implied by the window size that is selected. The first restriction involves the minimum window size (εmin). This minimum window size cannot be greater than the maximum programmable value of 13.5 ns for valid operation of the digital lock detect. Possible remedies for this solution would be reducing the delta-sigma order, using a higher VCO frequency and using a larger VCO_DIV value, or using analog lock detect. 13.5 ns ≥ εmin = 2ORDER-1 / fVCO (8) The second restriction is the maximum window size (εmax). If the calculated maximum window size is less than the minimum programmable window size of 3.5 ns, then this indicates that the digital lock detect cannot be used in this condition. Possible remedies for this could be to decrease the phase detector frequency, use analog lock detect, decrease the delta-sigma order, or decrease the VCO frequency. 3.5 ns ≤ εmax = 1/fPD - εmin - 2 ns (9) The third restriction comes from rearranging the equation for εmax. fPD ≤ 1 / ( εmin + εmax + 2 ns ) (10) In addition to this restriction on the maximum phase detector rate, recall that there are also restrictions on the maximum phase detector rate implied by the electrical specifications ( fPD≤ 104 MHz ) and by the minimum continuous N divider value (fPD ≤ fVCO / NMin). Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 47 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 fVCO www.ti.com εmin ORDER (MHz) MAXIMUM POSSIBLE PHASE DETECTOR FREQUENCY (MHz) (ns) ε = 3.5 ns ε = 5.5 ns ε = 7.5 ns ε = 9.5 ns ε = 11.5 ns ε = 13.5 ns Min (104, fVCO / 12) Min (104, fVCO / 12) Min (87.0, fVCO / 12) Min (74.1, fVCO / 12) Min (64.5, fVCO / 12) All 0 0.0 Min (104, fVCO / 12) 400 4 20.0 FAIL FAIL FAIL FAIL FAIL FAIL 400 3 10.0 FAIL FAIL FAIL FAIL 26.7 26.7 400 2 5.0 FAIL 30.8 30.8 30.8 30.8 30.8 500 4 16.0 FAIL FAIL FAIL FAIL FAIL FAIL 500 3 8.0 FAIL FAIL FAIL 33.3 33.3 33.3 500 2 4.0 FAIL 38.5 38.5 38.5 38.5 38.5 600 4 13.3 FAIL FAIL FAIL FAIL FAIL 31.6 600 3 6.7 FAIL FAIL 40.0 40.0 40.0 40.0 600 2 3.3 46.2 46.2 46.2 46.2 46.2 46.2 1200 4 6.7 FAIL FAIL 63.2 63.2 63.2 63.2 1200 3 3.3 80.0 80.0 80.0 80.0 74.1 64.5 1200 2 1.7 92.3 92.3 92.3 87.0 74.1 64.5 1530 4 5.2 FAIL 80.3 80.3 80.3 74.1 64.5 1530 3 2.6 102.0 102.0 102.0 87.0 74.1 64.5 1530 2 1.3 104.0 104.0 104.0 87.0 74.1 64.5 1800 4 4.4 FAIL 91.8 91.8 87.0 74.1 64.5 1800 3 2.2 104.0 104.0 104.0 87.0 74.1 64.5 1800 2 1.1 104.0 104.0 104.0 87.0 74.1 64.5 2000 4 4.0 FAIL 100.0 100.0 87.0 74.1 64.5 2000 3 2.0 104.0 104.0 104.0 87.0 74.1 64.5 2000 2 1.0 104.0 104.0 104.0 87.0 74.1 64.5 3000 4 2.7 104.0 104.0 104.0 87.0 74.1 64.5 3000 3 1.3 104.0 104.0 104.0 87.0 74.1 64.5 3000 2 0.7 104.0 104.0 104.0 87.0 74.1 64.5 4000 4 2.0 104.0 104.0 104.0 87.0 74.1 64.5 4000 3 1.0 104.0 104.0 104.0 87.0 74.1 64.5 4000 2 0.5 104.0 104.0 104.0 87.0 74.1 64.5 In the previous table, consider the case of operating in integer mode with ORDER=0. For this case, lock detect can theoretically work for all VCO frequencies provided that the phase detector frequency does not violate the maximum possible value. For instance, it would be an invalid condition to operate in integer mode with a VCO frequency of 900 MHz and a phase detector frequency of 100 MHz because 100 MHz exceeds the limit of 900 MHz/12 = 75 MHz. If the phase detector was lowered to 75 MHz to meet this restriction, then this condition would be valid provided that the window size was programmed to be 9.5 ns or less. Consider another example of a 400 MHz VCO frequency with a fourth order modulator. Because the minimum window size of 20 ns is above the maximum programmable value of 13.5 ns, digital lock detect cannot be used in this configuration. If the modulator order was reduced to 2nd order, then it would function provided that the phase detector frequency was less 30.8 MHz. 9.6.1.8.13 FSK - Frequency Shift Keying This bit enables a binary FSK modulation mode using the PLL N counter. Consult the applications section for more details. 48 FSK FSK MODE 0 Disabled 1 Enabled Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 9.6.1.9 Register R2 This word contains all the bits of the fractional denominator. These bits apply if the device is being used fractional mode. 9.6.1.9.1 PLL_DEN[21:0] -- Fractional Denominator These bits determine the fractional denominator. PLL_DEN[21:0] Fractional Denominator 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ... . . . . . . . . . . . . . . . . . . . . . 0 . 4194303 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 9.6.1.10 Registers R1 and R0 Both registers R1 and R0 contain information for the PLL R counter, N counter, and fractional numerator. The action of programming register R0, even to the same value, runs the VCO calibration when the device has the internal VCO operating. There are some programming words that are split across these two registers. 9.6.1.10.1 PLL_R[11:0] -- PLL R Divider Value The R divider divides the OSCin signal. Note that if the doubler is enabled, the R divider is bypassed. PLL_R[11:0] 0 Illegal State 1 0 0 0 0 0 0 0 0 0 0 0 1 2 0 0 0 0 0 0 0 0 0 0 1 0 3 0 0 0 0 0 0 0 0 0 0 1 1 ... ... ... ... ... ... ... ... ... ... ... ... ... 4095 1 1 1 1 1 1 1 1 1 1 1 1 9.6.1.10.2 PLL_N[17:0] PLL N Divider Value When using integer mode, the PLL N divider value is split up into two different locations. In fractional mode, only the 12 LSB bits of the N counter are used. Based on the order of the modulator, the range is shown in the table below. PLL_N[17:12] PLL_N[11:0] 1), OUTTERM, VCOGAIN, and DIVGAIN bits. It is not very sensitive to the actual frequency option of the part. For instance, the LMX2541SQ2060E and the LMX2541SQ2380E both should have similar output power at 2.2 GHz. Note that this same test setup can also be used to measure harmonics. 10.1.3 Phase Noise Measurement Test Setup The basic setup technique for all noise tests is to measure the noise at the output of the RFout pin in Internal VCO Mode (MODE=0) with a phase noise analyzer. For all measurements, the internal loop filter components (LF_R3, LF_R4, LF_C3, and LF_C4) should be set to their minimum values. There are some special considerations depending on what kind of noise is being measured. 10.1.3.1 PLL Phase Noise Measurement To • • • • get an accurate measurement of the PLL phase noise, one needs to ensure four things. The PLL loop bandwidth is sufficiently wide so that the VCO noise does not degrade the measurement The measurement is not corrupted by peaking in the loop filter response. The reference source is sufficiently clean so that this does not degrade the measurement. A distinction is made between the PLL flat noise and the PLL 1/f noise If the PLL loop bandwidth is made as wide as possible, then this helps keep the peaking of the loop filter response and the VCO noise from degrading the measurement. For the ultimate accuracy, this loop filter response can be factored into the measurement. As for the cleanliness of the reference source, the best sources tend to be those that are fixed, such as a 100 MHz Wenzel oscillator. Signal generators tend to be noisy, but if that is all that is available, then there are a few things that can help compensate for this. One technique is to use a higher frequency and divide this down to a lower frequency. For instance, a 500 MHz signal divided down to 20 MHz typically has much better phase noise than a direct 20 MHz signal, if it comes from a signal generator. Another technique is to measure the noise of the reference source and then multiply it up and then subtract it from the measurement. For instance, if the signal source was 500 MHz and the output frequency was 4 GHz, this signal source noise would be multiplied up by a factor of 20·log(4 GHz / 500 MHz) = 18 dB. Once that is done, the 1/f noise and the flat noise can be measured. 52 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 10.1.3.1.1 PLL Phase Noise Measurement - 1/f Noise The 1/f noise dominates closer to the carrier. Special care should be taken to ensure that this is not the noise of the reference source. The noise contribution of the reference source at the RFout pin can be calculating by measuring what is coming into the OSCin pin and then adding a correction factor of 20·log( fRFout / fOSCin) . A characteristic of this noise is that it follows a 10 dB/decade slope. If the slope of the measured noise looks more than 10 dB/decade, it is likely to be the reference source, not the LMX2541 device. Another characteristic of the 1/f noise is that it is independent of phase detector frequency. So to fully expose the 1/f noise, raise the phase detector frequency as high as possible, because this lowers the flat noise, but not the 1/f noise. 10.1.3.1.2 PLL Phase Noise Measurement - Flat Noise The PLL flat noise is measured at an offset that is not too close to the PLL 1/f noise, but also well inside the loop bandwidth. Many phase noise profiles have a point where the PLL noise flattens to a minimum value between the carrier and the loop bandwidth. This is where the flat noise should be measured. To measure the 1 Hz normalized phase noise, it is often easier to measure this with a lower phase detector frequency so that this flat noise is higher and easier to measure. 10.1.3.2 VCO Phase Noise Measurement In order to measure the VCO phase noise, the loop filter resistors should be set to their minimum value to reduce their noise contribution. The loop bandwidth should also be made as narrow as possible. Because the loop bandwidth is very narrow, the cleanliness of the OSCin signal is therefore not as important. The phase noise is measured outside the loop bandwidth of the system. An alternative way that might not be as accurate, but is much easier to do is to lock the device to a frequency and then set the CPT bit to 1 to disable the charge pump. The VCO will drift a little, the averaging on the equipment should be reset after this bit is changed and one cannot take to long to take this measurement. Test equipment that tracks the signal source is better if using this open loop technique. 10.1.3.3 Divider Phase Noise Measurement The basic method for measuring the divider noise is to drive the divider with a noise source of known value and then subtract away this noise. The divider noise floor tends to be flat, whereas the VCO phase noise decreases with offset frequency, so this measurement is made at as far of an offset that is possible. When using Internal VCO Mode (MODE=0), the raw VCO phase noise with VCO_DIV=1 can be measured. Then the VCO divider can be programmed to get close to the desired frequency. For example, the VCO frequency can be set to 4 GHz and the phase noise measured. This phase noise data can be saved or downloaded. Suppose then that one was interested in the divider noise at 400 MHz. The VCO divider could be set to 10 and then 20 dB is subtracted from the VCO phase noise to figure its contribution at 400 MHz. Provided that the actual phase noise measured at 400 MHz with VCO_DIV is above this, then one can assume that this is the noise of the divider. An alternative way to measure this is to drive the OSCin pin and use Divider Only (MODE=2) to measure the phase noise. This gives direct control of the frequency, but one should be sure that the noise being measured is the device and not the frequency source. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 53 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 10.1.4 Input and Output Impedance Test Setup Calibrate for Open, Short, Load Here Frequency Pin Network Analyzer Device Under Test Evaluation Board Power Supply A network analyzer can be used to measure the input impedance of the OSCin and ExtVCOin pin as well as the output impedance of the RFout pin. The general technique is to connect the desired pin with no DC blocking capacitor to a network analyzer and measure the impedance directly. The part needs to be programmed to ensure that it is in a known state. There are some special considerations that should be taken for different measurements of the three different impedances. 10.1.4.1 OSCin Input Impedance Measurement For this pin, the provided calibration standards are typically good enough for a decent measurement. A singleended measurement at the OSCin or OSCin* pins can be made For a differential measurement, this needs to be treated by the instrument as a two port network. 10.1.4.2 ExtVCOin Input Impedance Measurement Because this pin goes higher in frequency, it is often difficult get a good measurement at higher frequencies because of the effects of the board and SMA connector. One technique that can be used is instead of using the provided calibration standards that come with the equipment, solder resistors directly to the board in order to calibrate out the effects of the board as well. A 0 ohm resistor functions as a short, no resistor functions as an open, and two parallel 100 ohm resistors serve as a load. These should be soldered as close to the part as possible. Once this calibration is done, the measurement can be done as normal. 10.1.4.3 RFout Output Impedance Measurement Although output and input impedance are not the same thing, they can be measured in a similar way. Because this pin is a higher frequency, it is better to use the same method for calibration as used for the ExtVCOin pin. The other consideration for the RFout pin is that there are many different settings that impact this input impedance. When in bypass mode (VCO_DIV=1), the VCOGAIN and OUTTERM words can change the impedance. When in divided mode (VCO_DIV>1), the DIVGAIN and OUTTERM words can impact the impedance. 54 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 10.1.5 ExtVCOin (NOT OSCin) Input Sensitivity Test Setup SMA Cable Signal Generator Matching Network ExtVCOin DC Blocking Capacitor SMA Cable Frequency Counter Device Under Test Ftest/LD Pin Evaluation Board Power Supply In order to measure the ExtVCOin Input sensitivity, the part is put in External VCO mode and a signal is applied to the ExtVCOin pin. A matching network, which is typically a 3 dB pad, is used and this loss is added to the measured numbers as well as any potential cable losses (on the order of 1 dB). A signal is applied at a known frequency and power and the output of the N counter is monitored using the Ftest/LD pin and setting it to look at the N counter output divided by 2. Typically, the divide by 2 function is better because if it is not used, the duty cycle from the Ftest/LD pin is not 50% and this can sometimes confuse frequency counters. The part is set in fractional mode with a large fraction of 502 + 2097150/4194301 to ensure that the fractional circuitry gets fully tested. Accounting for the extra divide by 2 from the Ftest/LD pin, the divided output frequency should be the input frequency divided by 1005 to a 1 ppm tolerance. 10.1.6 OSCin Input Sensitivity Test Setup SMA Cable OSCin Signal Generator Device Under Test SMA Cable RFout Pin Frequency Counter Evaluation Board Power Supply 10.1.6.1 Input Sensitivity Test Procedure There are two things that are important to consider when measuring the OSCin sensitivity. • The action of setting the Ftest/LD pin to monitor the R divider output degrades the OSCin sensitivity. • The internal VCO frequency calibration is based on the OSCin signal Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 55 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com Because of these considerations, the OSCin sensitivity needs to be measured in a closed loop test in such a way that the internal frequency calibration is not distorting the measurement. To do this, a known frequency and power level are set at the OSCin pin and the power level is changed until the PLL becomes more than 1 ppm off frequency. The PLL_R divider is varied to maintain a phase detector frequency of 1 MHz to ensure that the PLL loop does not become unstable. The frequency counter needs to be synchronized in frequency to the signal generator. It is better to use a narrower loop bandwidth for this test because the phase noise of the PLL might degrade when the OSCin power level gets to close to the sensitivity limits. Typically, a 0.1 uF capacitor is used as a DC block for the signal at the OSCin pin. The sensitivity at the OSCin pin is measured with a single-ended input. This test can be run in internal VCO mode (MODE=0) or external VCO mode (MODE=1). When doing the test in internal VCO mode, the part needs to be initially locked and then the R counter is programmed to adjust for the OSCin frequency. However, in internal VCO mode, the PLL_N counter cannot be programmed, because the action of programming this counter activates the internal VCO frequecy calibration, which can interfere with the test. 10.1.6.2 OSCin Slew Rate Tests There are two methods that can be used to test the OSCin slew rate. One method is to use test equipment that actually allows the user to vary the slew rate directly, but this type of equipment typically does not give the user enough range of adjustability. Another method is to calculate the slew rate based on the slope of a sine wave of known frequency and amplitude. For this method, the slew rate can be calculated from the frequency and peak to peak amplitude of the OSCin signal as follows: SlewOSCin = 2 × π × fOSCin × VppOSCin 10.1.7 Typical Connections LMK04000 Output Figure 32. Full Chip Mode, Differential OSCin Microcontroller +3.3 V 0.1 PF 0.1 PF Ferrite 100 : +3.3 V VccCP1 VrefVCO +3.3V LMX2541 VregVCO 4.7 PF 4.7: VccCP2 VccVCO 1 PF 10: Ferrite Ferrite Ferrite Ferrite +3.3V +3.3V +3.3V +3.3V VregRFout Vtune R2_LF R2pLF C1_LF FLout C2_LF 330: +3.3V VccDiv 18: 100 pF RFout VccRFout CPout VccDig ExtVCOin +3.3V 0.1 PF 0.1 PF 0.1 PF VccFRAC Ferrite 0.1 PF VccPLL2 4.7: Ferrite 0.1 PF Bypass VccPLL1 0.1 PF VregFRAC +3.3V VccBias VccOSCin OSCin* OSCin 1 PF Ftest/LD CE LE DATA CLK RFoutEN +3.3V 330: To Circuit 56 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 Figure 33. External VCO Mode, Single-Ended OSCin, RFout Pin not Used 51: 51: +3.3V Ferrite 0.1 PF 0.1 PF Microcontroller +3.3V Ferrite +3.3V VccPLL2 4.7: 0.1 PF Bypass VccPLL1 VccFRAC Ferrite 0.1 PF VregFRAC +3.3V VccBias VccOSCin OSCin* OSCin 1 PF Ftest/LD CE LE DATA CLK RFoutEN +3.3V Ferrite VccCP1 VrefVCO VregVCO Ferrite +3.3V 1 PF 4.7: VccCP2 VccVCO 0.1 PF +3.3V LMX2541 0.1 PF +3.3V 0.1 PF Ferrite +3.3V 18: +3.3V VccDiv R3_LF C3_LF 18: R4_LF C4_LF To Circuit Ferrite C1_LF CPout 18: R2pLF R2_LF C2_LF FLout 100 pF RFout VccRFout Vtune VccDig 0.1 PF VregRFout ExtVCOin +3.3V For both of the able connection diagrams, L1, L2, and Lmid should be left open, but the pads should be placed on these pins for optimal solderability. The GND pins should have separate vias to ground and the GND DAP also needs to be grounded with 9 vias. The VccVCO, VccRFout, and VccDiv pins can be shorted to the power plane, but need to be connected. For the other Vcc pins, ferrite beads and bypass capacitors may be added in order to improve spurious performance. VregVCO and VrefVCO need to be connected even if the internal VCO is not being used. The VregRFout pin only needs to be connected if the RFout pin is being used. When a block is not used, it is always still necessary to connect the corresponding Vcc pin, but the bypassing is not necessary, as shown in the above diagram for the external VCO mode. 10.1.7.1 OSCin/OSCin* Connections For single-ended operation, the signal is driven into the OSCin pin. The OSCin* pin is terminated the same as the OSCin pin. This is a typical case if the device is driven by a TCXO. For both single-ended and differential operation, the input is AC coupled because the OSCin/OSCin* pins self-bias to an optimal DC operating point. Better performance for both phase noise and fractional spurs is obtained for signals with a higher slew rate, such as a square wave. This is especially important for lower frequency signals, because slower frequency sine waves have lower slew rates. Fractional spurs are typically about four dB better when running in differential mode as opposed to single-ended mode. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 57 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 50: Figure 34. Single-Ended Operation LMX2541 OSCin 0.1 PF 0.1 PF 50: OSCin* For differential operation, as is the case when using an LVDS or LVPECL driver, a 100 Ω resistor is placed across the OSCin/OSCin* traces Figure 35. Differential Operation LMX2541 0.1 PF 100: OSCin OSCin* 0.1 PF A third way to configure the device is in crystal mode (XO = 1). For this, the crystal is placed across the OSCin/OSCin* pins. Crystals are specified for a specific load capacitance, CLoad. The load capacitors shown in the figure each have a value of CLoad/2. Figure 36. Crystal Mode Operation LMX2541 OSCin OSCin* 58 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 10.2 Typical Application 10.2.1 Design Requirements Consider generating 4000.25 MHz from a 100-MHz XO input frequency. For this design example, use the parameters listed in Table 19 as the user-input parameters. Table 19. Design Parameters PARAMETER VALUE Fout 4000.25 MHz This parameter was given. Fosc 100 MHz This parameter was given. Fpd 25 MHz Although the phase detector could be chosen to be 100 MHz, it is being chosen lower to reduce the integer boundary spur at 250 kHz offset. Loop Bandwidth 53 kHz A wider bandwidth gives better in-band phase noise, but then the integer boundary spur is higher. This gives a reasonable trade-off. Maximum charge pump gain is best for the lowest PLL phase noise. Kpd 3.2 mA C1_LF 2.2 nF C2_LF 22 nF R2_LF 470 Ω C3_LF 20 pF C4_LF 100 pF R3_LF 1 kΩ R4_LF 200 Ω REASON FOR CHOOSING The external loop filter components can be found using design software from TI. The internal loop filter was chosen with the resistors to a lower setting to allow a wider bandwidth. However, it is best to put the most capacitance on C4 as it gives more filtering. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 59 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 10.2.2 Detailed Design Procedure Use the WEBENCH® Clock Architect to calculate the values of C2_LF and R2_LF. Use C4 for the largest capacitance because it offers more filtering. 10.2.3 Application Curves For Figure 38 and Figure 39, a third order modulator with dithering disabled was used with a fractional denominator of 500000. The charge pump gain was 32X and the loop filter components were C1 = 2.2 nF, C2 = 22 nF, R2 = 470 Ω. The internal loop filter components were C3_LF = 20 pF, C4_LF = 100 pF, R3_LF = 1 kΩ, R4_LF = 200 Ω. -70 -90 -100 -110 -120 -130 -140 -90 Fpd = 50 MHz , 5.8 mRad Fpd =100 MHz, 5.1 mRad Input Signal, 5.1 mRad -100 -110 -120 -130 -140 -150 -150 -160 1x102 1x103 1x104 1x105 1x106 1x107 1x108 Offset (Hz) D001 See Phase Noise Measurement Test Setup. The VCO Frequency is 2720.1 MHz. The OSCin signal was a 500 MHz differential LVPECL output of the LMK4033. Figure 37. LMX2541SQ2690 System Phase Noise 60 -80 Phase Noise (dBc/Hz) Phase Noise (dBc/Hz) -80 -70 Fpd=50MHz, 4.7 mRad Fpd=100MHz, 4.4 mRad Input Signal, 4.2 mRad -160 1x102 1x103 1x104 1x105 1x106 1x107 1x108 Offset (Hz) D001 The OSCin signal was a 500 MHz differential LVPECL output of the LMK04033. Figure 38. LMX2541SQ3320E System Phase Noise Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 3 -70 -90 -100 -110 -120 -130 2.4 CHANGE in VCO Noise (dB) Phase Noise (dBc/Hz) -80 Fpd=50 MHZ, 3.3 mRad Fpd=100 MHz, 2.8 mRad Input Signal, 3.0 mRad 1.8 1.2 0.6 0 -0.6 -1.2 Offset=10 kHz Offset=100 kHz Offset=1 MHz Offset=20 MHz -1.8 -2.4 -3 -40 -140 -150 -160 1x102 1x103 1x104 1x105 1x106 1x107 1x108 Offset (Hz) D001 The OSCin signal was a 500 MHz differential LVPECL output of the LMK04033. Figure 39. LMX2541SQ3740E System Phase Noise Figure 41. Phase Noise at 4000.25 MHz -20 0 20 40 60 Temperature (C) 80 100 D001 The above plot shows how much the VCO phase noise typically change over temperature relative to room temperature. The typical values for represent an average over all frequencies and part options and therefore there are some small variations over part options and frequencies that are not shown. VCO phase noise numbers room temperature are reported in the electrical specifications. A negative value indicates a phase noise improvement. Figure 40. VCO Phase Noise Degradation vs. Temperature and Offset (VCO Relocked at Each Temperature Vcc = 3.3 V, AC_TEMP_COMP = 5) Figure 42. Integer Boundary Spur Measurement 11 Power Supply Recommendations The LMX2541 has an internal low-noise regulator for the internal VCO, which makes it more immune to supply noise than many devices with an integrated VCO. That being said, the VCO is extremely sensitive to noise so it is recommended to use an LDO or clean power supply for this device; a switching power supply would likely generate unwanted spurs on the output. Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 61 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 12 Layout 12.1 Layout Guidelines 12.1.1 Configuring the LMX2541 for Optimal Performance 1. Determine the Channel Spacing (fCH) – For a system that has a VCO that tunes over several frequencies, the channel spacing is the tuning increment. In the case that the VCO frequency is fixed, this channel spacing is the greatest number that divides both the VCO frequency and the OSCin frequency. 2. Determine OSCin Frequency (fOSCin) – If the OSCin frequency is not already determined, then there are several considerations. A higher frequency is generally, but not always, preferable. One reason for this is that it has a higher slew rate if it is a sine wave. Another reason is that the clock for the VCO frequency calibration is based on the OSCin frequency and in general will run faster for higher OSCin frequencies. – Although a higher OSCin frequency is desirable, there are also reasons to use a lower frequency. If the OSCin frequency is strategically chosen, the worst case fractional spur channels might fall out of band. Also, if the OSCin frequency can be chosen such that the fractional denominator can avoid factors of 2 and/or 3, the sub-fractional spurs can be reduced. 3. Determine the Phase Detector Frequency (fPD) , Charge Pump Gain (KPD) and Fractional Denominator (FDEN) – In general, choose the highest phase detector frequency and charge pump gain, unless it leads to loop filter capacitor values that are unrealistically large for a given loop bandwidth. In this case, reducing either the phase detector frequency or the charge pump gain can yield more feasible capacitor values. Other reasons for not using the highest charge pump gain is to allow some adjustment margin to compensate for changes in the VCO gain or allow the use of Fastlock. – For choosing the fractional denominator, start with FDEN = fPD/fCH. As discussed previously, there might be reasons to choose larger equivalent fractions. 4. Design the Loop Filter 5. Determine the Modulator Order 6. Determine Dithering and Potential Larger Equivalent Fractional Value 62 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 LMX2541 www.ti.com SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 12.2 Layout Example Figure 43. Layout Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 63 LMX2541 SNOSB31J – JULY 2009 – REVISED DECEMBER 2014 www.ti.com 13 Device and Documentation Support 13.1 Device Support For the Clock Architect tool, go to http://www.ti.com/lsds/ti/analog/webench/clock-architect.page 13.2 Trademarks All trademarks are the property of their respective owners. 13.3 Electrostatic Discharge Caution These devices have limited built-in ESD protection. The leads should be shorted together or the device placed in conductive foam during storage or handling to prevent electrostatic damage to the MOS gates. 13.4 Glossary SLYZ022 — TI Glossary. This glossary lists and explains terms, acronyms, and definitions. 14 Mechanical, Packaging, and Orderable Information The following pages include mechanical, packaging, and orderable information. This information is the most current data available for the designated devices. This data is subject to change without notice and revision of this document. For browser-based versions of this data sheet, refer to the left-hand navigation. 64 Submit Documentation Feedback Copyright © 2009–2014, Texas Instruments Incorporated Product Folder Links: LMX2541 PACKAGE OPTION ADDENDUM www.ti.com 10-Dec-2020 PACKAGING INFORMATION Orderable Device Status (1) Package Type Package Pins Package Drawing Qty Eco Plan (2) Lead finish/ Ball material MSL Peak Temp Op Temp (°C) Device Marking (3) (4/5) (6) LMX2541SQ2060E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412060E LMX2541SQ2380E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412380E LMX2541SQ2690E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412690E LMX2541SQ3030E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413030E LMX2541SQ3320E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413320E LMX2541SQ3740E/NOPB ACTIVE WQFN NJK 36 1000 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413740E LMX2541SQE2060E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412060E LMX2541SQE2380E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412380E LMX2541SQE2690E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 412690E LMX2541SQE3030E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413030E LMX2541SQE3320E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413320E LMX2541SQE3740E/NOPB ACTIVE WQFN NJK 36 250 RoHS & Green SN Level-3-260C-168 HR -40 to 85 413740E (1) The marketing status values are defined as follows: ACTIVE: Product device recommended for new designs. LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect. NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in a new design. PREVIEW: Device has been announced but is not in production. Samples may or may not be available. OBSOLETE: TI has discontinued the production of the device. (2) RoHS: TI defines "RoHS" to mean semiconductor products that are compliant with the current EU RoHS requirements for all 10 RoHS substances, including the requirement that RoHS substance do not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered at high temperatures, "RoHS" products are suitable for use in specified lead-free processes. TI may reference these types of products as "Pb-Free". RoHS Exempt: TI defines "RoHS Exempt" to mean products that contain lead but are compliant with EU RoHS pursuant to a specific EU RoHS exemption. Green: TI defines "Green" to mean the content of Chlorine (Cl) and Bromine (Br) based flame retardants meet JS709B low halogen requirements of