700 MHz to 2700 MHz Rx Mixer with Integrated IF DGA, Fractional-N PLL, and VCO ADRF6620 Data Sheet FUNCTIONAL BLOCK DIAGRAM LOIN+ LOIN- PFD CHARGE CP PUMP + /1, /2, /4, /8 VTUNE LDO 2.5 V SERIAL PORT INTERFACE LDO VCO LDO 3.3V DECL4 DECL1 LOCK_DET VPTAT /2 CS SCLK SDIO MUXOUT LOIN+ LOIN- VTUNE CP DECL2 FRAC N = INT + MOD 11489-001 REFIN /8 /4 /2 x1 x2 RFSW1 APPLICATIONS IFOUT1- IFOUT1+ IFOUT2- IFOUT2+ RFIN0 RFIN1 RFIN2 RFIN3 RFSW0 Integrated fractional-N phase-locked loop (PLL) RF input frequency range: 700 MHz to 2700 MHz Internal local oscillator (LO) frequency range: 350 MHz to 2850 MHz Input P1dB: 17 dBm Output IP3: 45 dBm Single-pole four-throw (SP4T) RF input switch Digital step attenuator (DSA) range: 0 dB to 15 dB Integrated RF tunable balun allowing single-ended 50 input Multicore integrated voltage controlled oscillator (VCO) Digitally programmable variable gain amplifier (DGA) -3 dB bandwidth: >600 MHz Balanced 150 IF output impedance Programmable via 3-wire serial port interface (SPI) Single 5 V supply MXOUT+ MXOUT- IFIN+ IFIN- FEATURES Figure 1. Wireless receivers Digital predistortion (DPD) receivers GENERAL DESCRIPTION The ADRF6620 is a highly integrated active mixer and synthesizer that is ideally suited for wireless receiver subsystems. The feature rich device consists of a high linearity broadband active mixer; an integrated fractional-N PLL; low phase noise, multicore VCO; and IF DGA. In addition, the ADRF6620 integrates a 4:1 RF switch, an on-chip tunable RF balun, programmable RF attenuator, and low dropout (LDO) regulators. This highly integrated device fits within a small 7 mm x 7 mm footprint. The high isolation 4:1 RF switch and on-chip tunable RF balun enable the ADRF6620 to support four single-ended 50 terminated RF inputs. A programmable attenuator ensures optimal RF input drive to the high linearity mixer core. The integrated DSA has an attenuation range of 0 dB to 15 dB with a step size of 1 dB. Rev. 0 The ADRF6620 offers two alternatives for generating the differential LO input signal: externally, via a high frequency, low phase noise LO signal, or internally, via the on-chip fractional-N PLL synthesizer. The integrated synthesizer enables continuous LO coverage from 350 MHz to 2850 MHz. The PLL reference input can support a wide frequency range because the divide and multiply blocks can be used to increase or decrease the reference frequency to the desired value before it is passed to the phase frequency detector (PFD). The integrated high linearity DGA provides an additional gain range from 3 dB to 15 dB in steps of 0.5 dB for maximum flexibility in driving an analog-to-digital converter (ADC). The ADRF6620 is fabricated using an advanced silicon-germanium BiCMOS process. It is available in a 48-lead, RoHS-compliant, 7 mm x 7 mm LFCSP package with an exposed pad. Performance is specified over the -40C to +85C temperature range. Document Feedback Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties that may result from its use. Specifications subject to change without notice. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Trademarks and registered trademarks are the property of their respective owners. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781.329.4700 (c)2013 Analog Devices, Inc. All rights reserved. Technical Support www.analog.com ADRF6620 Data Sheet TABLE OF CONTENTS Features .............................................................................................. 1 Serial Port Interface (SPI) ......................................................... 27 Applications ....................................................................................... 1 Basic Connections ...................................................................... 28 Functional Block Diagram .............................................................. 1 RF Input Balun Insertion Loss Optimization ......................... 30 General Description ......................................................................... 1 IP3 and Noise Figure Optimization ......................................... 31 Revision History ............................................................................... 2 Interstage Filtering Requirements ............................................ 35 Specifications..................................................................................... 3 IF DGA vs. Load......................................................................... 38 RF Input to IF DGA Output System Specifications ................. 3 ADC Interfacing ......................................................................... 39 Synthesizer/PLL Specifications ................................................... 4 Power Modes ............................................................................... 40 RF Input to Mixer Output Specifications .................................. 6 Layout .......................................................................................... 40 IF DGA Specifications ................................................................. 7 Register Map ................................................................................... 41 Digital Logic Specifications ......................................................... 8 Register Address Descriptions ...................................................... 42 Absolute Maximum Ratings ............................................................ 9 Register 0x00, Reset: 0x00000, Name: SOFT_RESET ........... 42 Thermal Resistance ...................................................................... 9 Register 0x01, Reset: 0x8B7F, Name: Enables ........................ 42 ESD Caution .................................................................................. 9 Register 0x02, Reset: 0x0058, Name: INT_DIV ..................... 43 Pin Configuration and Function Descriptions ........................... 10 Register 0x03, Reset: 0x0250, Name: FRAC_DIV ................. 43 Typical Performance Characteristics ........................................... 11 Register 0x04, Reset: 0x0600, Name: MOD_DIV .................. 43 RF Input to DGA Output System Performance ..................... 11 Register 0x20, Reset: 0x0C26, Name: CP_CTL ...................... 44 Phase-Locked Loop (PLL)......................................................... 13 Register 0x21, Reset: 0x0003, Name: PFD_CTL .................... 45 RF Input to Mixer Output Performance ................................. 17 Register 0x22, Reset: 0x000A, Name: FLO_CTL ................... 46 IF DGA ........................................................................................ 20 Register 0x23, Reset: 0x0000, Name: DGA_CTL................... 47 Spurious Performance................................................................ 22 Register 0x30, Reset: 0x00000, Name: BALUN_CTL ............ 48 Theory of Operation ...................................................................... 24 Register 0x31, Reset: 0x08EF, Name: MIXER_CTL .............. 48 RF Input Switches ....................................................................... 24 Register 0x40, Reset: 0x0010, Name: PFD_CTL2 .................. 49 Tunable Balun ............................................................................. 25 Register 0x42, Reset: 0x000E, Name: DITH_CTL1 ............... 50 RF Digital Step Attenuator (DSA) ............................................ 25 Register 0x43, Reset: 0x0001, Name: DITH_CTL2 ............... 50 Active Mixer ................................................................................ 25 Outline Dimensions ....................................................................... 51 Digitally Programmable Variable Gain Amplifier (DGA) .... 25 Ordering Guide .......................................................................... 51 LO Generation Block ................................................................. 26 REVISION HISTORY 7/13--Revision 0: Initial Version Rev. 0 | Page 2 of 52 Data Sheet ADRF6620 SPECIFICATIONS VCCx = 5 V, TA = 25C, unless otherwise noted. Table 1. Parameter LO INPUT Internal LO Frequency Range External LO Frequency Range LO Input Level LO Input Impedance RF INPUT Input Frequency Input Return Loss Input Impedance RF DIGITAL STEP ATTENUATOR Attenuation Range POWER SUPPLY Power Consumption Test Conditions/Comments Min LO_DIV_A = 00 350 350 -6 Typ 0 50 700 Max Unit 2850 3200 +6 MHz MHz dBm 2700 MHz dB 15 5.25 dB V 12 50 Step size = 1 dB 0 4.75 LO output buffer disabled External LO + IF DGA enabled Internal LO + IF DGA enabled Only IF DGA enabled 5.0 1.3 1.7 0.6 6 Power-Down Current W W W mA RF INPUT TO IF DGA OUTPUT SYSTEM SPECIFICATIONS VCCx = 5 V, TA = 25C, high-side LO injection, fIF = 200 MHz, internal LO frequency, IF DGA output load = 150 , and 2 V p-p differential output with third-order low-pass filter, unless otherwise noted. For mixer settings for maximum linearity, see Table 16. All losses from input and output traces and baluns are de-embedded from results Table 2. RF Switch + Balun + RF Attenuator + Mixer + IF DGA Parameter DYNAMIC PERFORMANCE AT fRF = 900 MHz Voltage Conversion Gain Output P1dB Output IP3 Output IP2 Noise Figure DYNAMIC PERFORMANCE AT fRF = 1900 MHz Voltage Conversion Gain Output P1dB Output IP3 Output IP2 Noise Figure DYNAMIC PERFORMANCE AT fRF = 2100 MHz Voltage Conversion Gain Output P1dB Output IP3 Output IP2 Noise Figure DYNAMIC PERFORMANCE AT fRF = 2700 MHz Voltage Conversion Gain Output P1dB Output IP3 Output IP2 Noise Figure Test Conditions/Comments fIF = 200 MHz 1 V p-p each output tone, 1 MHz tone spacing 1 V p-p each output tone, 1 MHz tone spacing Noise figure optimized fIF = 200 MHz 1 V p-p each output tone, 1 MHz tone spacing 1 V p-p each output tone, 1 MHz tone spacing Noise figure optimized fIF = 200 MHz 1 V p-p each output tone, 1 MHz tone spacing 1 V p-p each output tone, 1 MHz tone spacing Noise figure optimized fIF = 200 MHz 1 V p-p each output tone, 1 MHz tone spacing 1 V p-p each output tone, 1 MHz tone spacing Noise figure optimized Rev. 0 | Page 3 of 52 Min Typ Max Unit 12 18 43 78 16 dB dBm dBm dBm dB 11 18 45 75 18.5 10.5 18 45 66 19 dB dBm dBm dBm dB dB dBm dBm dBm dBm dB 9 18 44 74 21 dB dBm dBm dBm dB ADRF6620 Data Sheet SYNTHESIZER/PLL SPECIFICATIONS VCCx = 5 V, TA = 25C, fREF = 153.6 MHz, fREF power = 4 dBm, fPFD = 38.4 MHz, and loop filter bandwidth = 120 kHz, unless otherwise noted. Table 3. Parameter PLL REFERENCE PLL Reference Frequency PLL Reference Level PFD FREQUENCY INTERNAL VCO RANGE OPEN-LOOP VCO PHASE NOISE fVCO2 = 3.4 GHz fVCO1 = 4.6 GHz fVCO0 = 5.5 GHz SYNTHESIZER SPECIFICATIONS fLO = 1.710 GHz, fVCO2 = 3.420 GHz fPFD Spurs Closed-Loop Phase Noise Integrated Phase Noise Figure of Merit (FOM) 1 Test Conditions/Comments For PLL lock condition VTUNE = 2 V, LO_DIV_A = 00 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset VCO sensitivity (KV) 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset VCO sensitivity (KV) 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset VCO sensitivity (KV) Measured at LO output, LO_DIV_A = 01 fREF = 153.6 MHz, fPFD = 38.4 MHz, 120 kHz loop filter fPFD x 1 fPFD x 2 fPFD x 3 fPFD x 4 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset 10 kHz to 40 MHz integration bandwidth Rev. 0 | Page 4 of 52 Min 12 -15 24 2800 Typ +4 Max Unit 464 +14 58 5700 MHz dBm MHz MHz -39 -81 -103 -123 -125 -143 -147 -155 88 -39 -74 -101 -123 -125 -143 -147 -156 89 -39 -69 -99 -121 -124 -142 -146 -155 72 dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz MHz/V dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz MHz/V dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz MHz/V -83 -89 -90 -93 -97 -110 -107 -128 -132 -144 -152 -158 0.21 -222 dBc dBc dBc dBc dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz rms dBc/Hz Data Sheet Parameter fLO = 2.305 GHz, fVCO1 = 4.610 GHz fPFD Spurs Closed-Loop Phase Noise Integrated Phase Noise Figure of Merit1 fLO = 2.75 GHz, fVCO2 = 5.5 GHz fPFD Spurs Closed-Loop Phase Noise Integrated Phase Noise Figure of Merit1 1 ADRF6620 Test Conditions/Comments Min Typ Max Unit fPFD x 1 fPFD x 2 fPFD x 3 fPFD x 4 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset 10 kHz to 40 MHz integration bandwidth -84 -87 -91 -92 -93 105 -103 -116 -130 -144 -152 -156 0.3 -222 dBc dBc dBc dBc dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz rms dBc/Hz fPFD x 1 fPFD x 2 fPFD x 3 fPFD x 4 1 kHz offset 10 kHz offset 100 kHz offset 800 kHz offset 1 MHz offset 6 MHz offset 10 MHz offset 40 MHz offset 10 kHz to 40 MHz integration bandwidth -82 -88 -93 -96 -93 -101 -99 -122 -128 -144 -151 -154 0.38 -222 dBc dBc dBc dBc dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz dBc/Hz rms dBc/Hz Figure of merit (FOM) is computed as phase noise (dBc/Hz) - 10 log 10(fPFD) - 20 log 10(fLO/fPFD). The FOM was measured across the full LO range, with fREF = 160 MHz and fREF power = 4 dBm (500 V/s slew rate) with a 40 MHz fPFD. The FOM was computed at 50 kHz offset. Rev. 0 | Page 5 of 52 ADRF6620 Data Sheet RF INPUT TO MIXER OUTPUT SPECIFICATIONS VCCx = 5 V, TA = 25C, high-side LO injection, fIF = 200 MHz, external LO frequency, and RF attenuation = 0 dB, unless otherwise noted. Mixer settings configured for maximum linearity (see Table 16). All losses from input and output traces and baluns are de-embedded from results. Table 4. RF Switch + Balun + RF Attenuator + Mixer Parameter VOLTAGE GAIN MIXER OUTPUT IMPEDANCE DYNAMIC PERFORMANCE AT fRF= 900 MHz Voltage Conversion Gain Input P1dB Input IP3 Input IP2 Noise Figure LO to RF Leakage RF to LO Leakage LO to IF Leakage RF to IF Leakage Isolation1 DYNAMIC PERFORMANCE AT fRF =1900 MHz Voltage Conversion Gain Input P1dB Input IP3 Input IP2 Noise Figure LO to RF Leakage RF to LO Leakage LO to IF Leakage RF to IF Leakage Isolation1 DYNAMIC PERFORMANCE AT fRF = 2100 MHz Voltage Conversion Gain Input P1dB Input IP3 Input IP2 Noise Figure LO to RF Leakage RF to LO Leakage LO to IF Leakage RF to IF Leakage Isolation1 DYNAMIC PERFORMANCE AT fRF = 2700 MHz Voltage Conversion Gain Input P1dB Input IP3 Input IP2 Noise Figure LO to RF Leakage RF to LO Leakage LO to IF Leakage RF to IF Leakage Isolation1 1 Test Conditions/Comments Differential 255 load Differential (see Figure 87) -5 dBm each input tone, 1 MHz tone spacing -5 dBm each input tone, 1 MHz tone spacing With respect to 0 dBm RF input power Isolation between RFIN0 and RFIN3 -5 dBm each input tone, 1 MHz tone spacing -5 dBm each input tone, 1 MHz tone spacing With respect to 0 dBm RF input power Isolation between RFIN0 and RFIN3 -5 dBm each input tone, 1 MHz tone spacing -5 dBm each input tone, 1 MHz tone spacing With respect to 0 dBm RF input power Isolation between RFIN0 and RFIN3 -5 dBm each input tone, 1 MHz tone spacing -5 dBm each input tone, 1 MHz tone spacing With respect to 0 dBm RF input power Isolation between RFIN0 and RFIN3 Min Typ -4 255 Max Unit dB -2 17 40 65 15 -70 -60 -32 -45 -52 dB dBm dBm dBm dB dBm dBc dBm dBc dBc -3 17 40 62 17 -60 -50 -35 -43 -47 dB dBm dBm dBm dB dBm dBc dBm dBc dBc -3.5 18 40 54.5 18 -60 -40 -35 -40 -45 dB dBm dBm dBm dB dBm dBc dBm dBc dBc -4.7 19 40 56 21 -60 -45 -40 -42 -41 dB dBm dBm dBm dB dBm dBc dBm dBc dBc Isolation between RF inputs. An input signal was applied to RFIN0 while RFIN1 to RFIN3 were terminated with 50 . The IF signal amplitude was measured at the mixer output. The internal switch was then configured for RFIN3, and the feedthrough was measured as a delta from the fundamental. Rev. 0 | Page 6 of 52 Data Sheet ADRF6620 IF DGA SPECIFICATIONS VCCx = 5 V, TA = 25C, RS = RL = 150 differential, fIF = 200 MHz, 2 V p-p differential output,unless otherwise noted. All losses from input and output traces and baluns are de-embedded from results. Table 5. Parameter BANDWIDTH -1 dB Bandwidth -3 dB Bandwidth SLEW RATE INPUT STAGE Input P1dB Input Impedance Common-Mode Input Voltage Common-Mode Rejection Ratio (CMRR) GAIN Power/Voltage Gain, Step Size = 0.5 dB Gain Flatness Gain Conformance Error Gain Temperature Sensitivity Gain Step Response OUTPUT STAGE Output P1dB Output Impedance NOISE/HARMONIC PERFORMANCE at 200 MHz Output IP3 Output IP2 HD2 HD3 Noise Figure Test Conditions/Comments Min Typ Max Unit VOUT = 2 V p-p VOUT = 2 V p-p 500 700 5.5 MHz MHz V/ns At minimum gain 17 150 1.5 50 dBm V dB 3 50 MHz < fC < 200 MHz See Figure 88 1 V p-p each output tone, 1 MHz tone spacing 1 V p-p each output tone, 1 MHz tone spacing VOUT = 2 V p-p VOUT = 2 V p-p Rev. 0 | Page 7 of 52 0.2 0.1 0.008 15 15 dB dB dB dB/C ns 18 150 dBm 45 63 -87 -84 10 dBm dBm dBc dBc dB ADRF6620 Data Sheet DIGITAL LOGIC SPECIFICATIONS Table 6. Parameter SERIAL PORT INTERFACE TIMING Input Voltage High Input Voltage Low Output Voltage High Output Voltage Low Serial Clock Period Setup Time Between Data and Rising Edge of SCLK Hold Time Between Data and Rising Edge of SCLK Setup Time Between Falling Edge of CS and SCLK Hold Time Between Rising Edge of CS and SCLK Minimum Period SCLK Can Be in Logic High State Minimum Period SCLK Can Be in Logic Low State Maximum Time Delay Between Falling Edge of SCLK and Output Data Valid for a Read Operation Maximum Time Delay Between CS Deactivation and SDIO Bus Return to High Impedance Symbol Test Conditions/Comments VIH VIL VOH VOL tSCLK tDS tDH tS tH tHIGH tLOW tACCESS Min Typ Max 1.4 231 V V V V ns ns ns ns ns ns ns ns 5 ns 0.70 IOH = -100 A IOL = +100 A 2.3 0.2 38 8 8 10 10 10 10 tZ Unit Timing Diagram tHIGH tDS tS tH tSCLK tACCESS tLOW tDH CS DON'T CARE SDIO DON'T CARE DON'T CARE tZ A6 A5 A4 A3 A2 A1 A0 R/W D15 D14 D13 Figure 2. Serial Port Interface Timing Rev. 0 | Page 8 of 52 D3 D2 D1 D0 DON'T CARE 11489-002 SCLK Data Sheet ADRF6620 ABSOLUTE MAXIMUM RATINGS THERMAL RESISTANCE Table 7. Parameter VCCx RFSW0, RFSW1 RFIN0, RFIN1, RFIN2, RFIN3 LOIN-, LOIN+ REFIN IFIN-, IFIN+ CS, SCLK, SDIO VTUNE Operating Temperature Range Storage Temperature Range Maximum Junction Temperature Table 8. Thermal Resistance Rating -0.5 V to +5.5 V -0.3 V to +3.6 V 20 dBm 16 dBm -0.3 V to +3.6 V -1.2 V to +3.6 V -0.3 V to +3.6 V -0.3 V to +3.6 V -40C to +85C -65C to +150C 150C Package Type 48-Lead LFCSP ESD CAUTION Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only; functional operation of the device at these or any other conditions above those indicated in the operational section of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. Rev. 0 | Page 9 of 52 JC 1.62 Unit C/W ADRF6620 Data Sheet 48 47 46 45 44 43 42 41 40 39 38 37 GND VTUNE DECL4 LOIN+ LOIN- MUXOUT SDIO SCLK CS RFSW1 RFSW0 DECL3 PIN CONFIGURATION AND FUNCTION DESCRIPTIONS VCC1 1 DECL1 2 ADRF6620 TOP VIEW (Not to Scale) 36 35 34 33 32 31 30 29 28 27 26 25 GND RFIN0 GND GND RFIN1 GND GND RFIN2 GND GND RFIN3 GND NOTES 1. THE EXPOSED PAD MUST BE CONNECTED TO A GROUND PLANE WITH LOW THERMAL IMPEDANCE. 11489-003 VCC3 VCC4 IFIN- IFIN+ GND MXOUT+ MXOUT- GND LOOUT+ LOOUT- GND VCC5 13 14 15 16 17 18 19 20 21 22 23 24 CP 3 GND 4 GND 5 REFIN 6 DECL2 7 IFOUT1+ 8 IFOUT1- 9 IFOUT2+ 10 IFOUT2- 11 VCC2 12 PIN 1 INDICATOR Figure 3. Pin Configuration Table 9. Pin Function Descriptions 1 Pin No. 1, 12, 13, 14, 24 2, 7, 37, 46 3 4, 5, 17, 20, 23, 25, 27, 28, 30, 31, 33, 34, 36, 48 6 8 to 11 15, 16 18, 19 21, 22 26, 29, 32, 35 Mnemonic VCC1, VCC2, VCC3, VCC4, VCC5 DECL1, DECL2, DECL3, DECL4 CP GND Description 5 V Power Supplies. Decouple all power supply pins to ground, using 100 pF and 0.1 F capacitors. Place the decoupling capacitors near the pins. Decouple all DECLx pins to ground, using 100 pF, 0.1 F, and 10 F capacitors. Place the decoupling capacitors near the pins. Synthesizer Charge Pump Output. Connect this pin to the VTUNE pin through the loop filter. Ground. REFIN IFOUT1+, IFOUT1-, IFOUT2+, IFOUT2- Synthesizer Reference Frequency Input. IF DGA Outputs. Connect the positive pins such that IFOUT1+ and IFOUT2+ are tied together. Similarly, connect the negative pins such that IFOUT1- and IFOUT2- are tied together. Refer to the Layout section for a recommended layout that minimizes parasitic capacitance and optimizes performance. Differential IF DGA Inputs. AC couple the mixer outputs to the IF DGA inputs. Differential Mixer Outputs. AC couple the mixer outputs to the IF DGA inputs. Differential LO Outputs. The differential output impedance is 50 . RF Inputs. These single-ended RF inputs have a 50 input impedance and must be ac-coupled. External Pin Control of RF Input Switches. For logic high, connect these pins to 2.5 V logic. SPI Chip Select, Active Low. 3.3 V tolerant logic levels. SPI Clock. 3.3 V tolerant logic levels. SPI Data Input or Output. 3.3 V tolerant logic levels. Multiplexer Output. This output pin provides the PLL reference signal or the PLL lock detect signal. Differential Local Oscillator Inputs. The differential input impedance is 50 . VCO Tuning Voltage. Connect this pin to the CP pin through the loop filter. Exposed Pad. The exposed pad must be connected to a ground plane with low thermal impedance. 38, 39 40 41 42 43 IFIN-, IFIN+ MXOUT+, MXOUT- LOOUT+, LOOUT- RFIN3, RFIN2, RFIN1, RFIN0 RFSW0, RFSW1 CS SCLK SDIO MUXOUT 44, 45 47 49 LOIN-, LOIN+ VTUNE EPAD 1 For more connection information about these pins, see Table 14. Rev. 0 | Page 10 of 52 Data Sheet ADRF6620 TYPICAL PERFORMANCE CHARACTERISTICS RF INPUT TO DGA OUTPUT SYSTEM PERFORMANCE VCCx = 5 V, TA = 25C, RFDSA_SEL = 00 (0 dB), RFSW_SEL = 00 (RFIN0), BAL_CIN and BAL_COUT optimized for maximum gain; MIXER_BIAS, MIXER_RDAC, and MIXER_CDAC optimized for highest linearity, DGA at maximum gain; third-order low-pass filter between the mixer output and IF DGA input; high-side LO, internal LO frequency, IF frequency = 200 MHz, unless otherwise noted. All losses from input and output traces and baluns are de-embedded from results. 15 15 14 14 13 TA = -40C RF FREQUENCY = 1900MHz GAIN (dB) 11 10 RF FREQUENCY = 2100MHz 10 TA = +25C TA = +85C 9 RF FREQUENCY = 2700MHz 9 8 7 6 5 8 4 7 3 2 6 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) 0 50 11489-004 1000 250 300 350 400 450 500 22 20 TA = +85C TA = +25C 18 18 TA = -40C 16 OP1dB (dBm) 14 12 10 8 14 12 10 8 6 6 4 4 2 2 1000 1400 1800 2200 RF FREQUENCY (MHz) Figure 5. OP1dB vs. RF Frequency 2600 3000 0 50 11489-005 0 600 200 Figure 6. Gain vs. IF Frequency; LO Sweep with Fixed RF, IF Roll-Off 22 16 150 IF FREQUENCY (MHz) Figure 4. Gain vs. RF Frequency; IF Frequency = 200 MHz 20 100 11489-007 1 5 600 RF RF RF RF FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz 100 150 200 250 300 350 IF FREQUENCY (MHz) 400 450 500 11489-008 GAIN (dB) RF FREQUENCY = 900MHz 12 11 12 OP1dB (dBm) 13 Figure 7. OP1dB vs. IF Frequency; LO Sweep with Fixed RF, IF Roll-Off Rev. 0 | Page 11 of 52 ADRF6620 Data Sheet 95 85 75 75 OIP2 (dBm), OIP3 (dBm) 85 65 TA = -40C TA = +25C 55 TA = +85C OIP3 (dBm) 45 35 OIP3 (dBm) 45 35 15 15 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) 5 50 RF RF RF RF FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz 100 150 200 250 300 350 400 450 500 IF FREQUENCY (MHz) Figure 11. OIP2/OIP3 vs. IF Frequency; LO Sweep with Fixed RF, IF Roll-Off; Measured on 1 V p-p on Each Tone at DGA Output Figure 8. OIP2/OIP3 vs. RF Frequency; Measured on 1 V p-p on Each Tone at DGA Output 15 95 14 LO FREQUENCY = 1100MHz 13 85 LO FREQUENCY = 2100MHz 12 OIP2 (dBm) 75 OIP2 (dBm), OIP3 (dBm) 11 10 GAIN (dB) 55 25 5 600 OIP2 (dBm) 65 25 11489-006 OIP2 (dBm), OIP3 (dBm) OIP2 (dBm) 11489-009 95 LO FREQUENCY = 2300MHz 9 8 7 6 5 4 65 LO FREQUENCY = 1100MHz LO FREQUENCY = 2300MHz 55 45 OIP3 (dBm) 35 LO FREQUENCY = 2100MHz 25 3 2 15 100 150 200 250 300 350 400 450 500 IF FREQUENCY (MHz) 5 50 11489-110 0 50 Figure 9. Gain vs. IF Frequency; RF Sweep with Fixed LO; IF and RF Roll-Off; Measured on 1 V p-p on Each Tone at DGA Output 200 250 300 350 400 450 500 Figure 12. OIP2/OIP3 vs. IF Frequency; RF Sweep with Fixed LO; IF and RF Roll-Off; Measured on 1 V p-p on Each Tone at DGA Output 500 450 85 OIP2 (dBm) TA = +85C 400 SUPPLY CURRENT (mA) 75 65 55 OIP3 (dBm) 45 35 25 350 300 TA = +25C TA = -40C 250 200 150 100 15 0 1 FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz 2 3 4 5 6 7 50 8 RFDSA 9 10 11 12 13 14 15 0 600 11489-111 RF RF RF RF 1000 1400 1800 2200 2600 RF FREQUENCY (MHz) Figure 10. OIP2/OIP3 vs. RFDSA; Measured on 1 V p-p on Each Tone at DGA Output Rev. 0 | Page 12 of 52 Figure 13. Supply Current vs. RF Frequency 3000 11489-113 OIP2 (dBm), OIP3 (dBm) 150 IF FREQUENCY (MHz) 95 5 100 11489-112 1 Data Sheet ADRF6620 PHASE-LOCKED LOOP (PLL) VCCx = 5 V, TA = 25C, 120 kHz loop filter, fREF = 153.6 MHz, PLL reference amplitude = 4 dBm, fPFD = 38.4 MHz, measured at LO output, unless otherwise noted. 10M 100M Figure 14. VCO2 Open-Loop VCO Phase Noise vs. Offset Frequency; fVCO2 = 3.4 GHz, LO_DIV_A = 00, VTUNE = 2 V PHASE NOISE (dBc/Hz) PHASE NOISE (dBc/Hz) 100k 1M 10M 100M OFFSET FREQUENCY (Hz) Figure 15. VCO1 Open-Loop Phase Noise vs. Offset Frequency; fVCO1 = 4.6 GHz, LO_DIV_A = 00, VTUNE = 2 V PHASE NOISE (dBc/Hz) PHASE NOISE (dBc/Hz) 100k 1M 10M 100M OFFSET FREQUENCY (Hz) Figure 16. VCO0 Open-Loop Phase Noise vs. Offset Frequency; fVCO0 = 5.5 GHz, LO_DIV_A = 00, VTUNE = 2 V 100M -60 -65 -70 -75 -80 -85 -90 -95 -100 -105 -110 -115 -120 -125 -130 -135 -140 -145 -150 -155 -160 1k LO_DIV_A = 00 LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 10k 100k 1M 10M 100M Figure 18. VCO1 Closed-Loop Phase Noise for Various LO_DIV_A Dividers vs. Offset Frequency; fVCO1 = 4.6 GHz 11489-012 10k 10M OFFSET FREQUENCY (Hz) 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 -110 -120 -130 -140 -150 -160 1k 1M Figure 17. VCO2 Closed-Loop Phase Noise for Various LO_DIV_A Dividers vs. Offset Frequency; fVCO2 = 3.4 GHz 11489-011 10k 100k OFFSET FREQUENCY (Hz) 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 -110 -120 -130 -140 -150 -160 1k 10k 11489-014 1M LO_DIV_A = 00 LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 -60 -65 -70 -75 -80 -85 -90 -95 -100 -105 -110 -115 -120 -125 -130 -135 -140 -145 -150 -155 -160 1k LO_DIV_A = 00 LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 10k 100k 1M OFFSET FREQUENCY (Hz) 10M 100M 11489-015 100k OFFSET FREQUENCY (Hz) -60 -65 -70 -75 -80 -85 -90 -95 -100 -105 -110 -115 -120 -125 -130 -135 -140 -145 -150 -155 -160 1k 11489-013 PHASE NOISE (dBc/Hz) 10k 11489-010 PHASE NOISE (dBc/Hz) 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 -110 -120 -130 -140 -150 -160 1k Figure 19. VCO0 Closed-Loop Phase Noise for Various LO_DIV_A Dividers vs. Offset Frequency; fVCO0 = 5.532 GHz Rev. 0 | Page 13 of 52 ADRF6620 Data Sheet 3.0 200 TA = -40C TA = +25C TA = +85C 2.6 2.4 210 VTUNE (V) FOM (dBc/Hz/Hz) 205 TA = -40C TA = +25C TA = +85C 2.8 215 220 2.2 2.0 1.8 1.6 1.4 225 2000 2200 2400 2600 2800 1.0 2800 LO FREQUENCY (MHz) 3200 PHASE NOISE (dBc/Hz) 10kHz OFFSET 100kHz OFFSET 800kHz OFFSET 5600 -120 1MHz OFFSET -125 -130 -135 -140 10MHz OFFSET -150 6MHz OFFSET 40MHz OFFSET -155 5379 5779 -160 2579 PHASE NOISE (dBc/Hz) 50kHz OFFSET 400kHz OFFSET 1MHz OFFSET 10MHz OFFSET 2784 3379 3779 4179 4579 4979 5379 5779 Figure 24. Open-Loop Phase Noise vs. VCO Frequency; LO_DIV_A = 00 1kHz OFFSET 2584 2979 VCO FREQUENCY (MHz) 11489-018 PHASE NOISE (dBc/Hz) 5200 -115 -145 11489-017 PHASE NOISE (dBc/Hz) -110 1kHz OFFSET 1984 2184 2384 LO FREQUENCY (MHz) 4800 TA = -40C TA = +25C TA = +85C -105 Figure 21. Open-Loop Phase Noise vs. VCO Frequency; LO_DIV_A = 00 -85 TA = -40C -90 TA = +25C -95 TA = +85C -100 -105 -110 -115 -120 -125 -130 -135 -140 -145 -150 -155 -160 -165 1384 1584 1784 4400 Figure 23. VTUNE vs. VCO Frequency -100 3779 4179 4579 4979 VCO FREQUENCY (MHz) 4000 VCO FREQUENCY (MHz) Figure 20. PLL Figure of Merit (FOM) vs. LO Frequency 0 TA = -40C -10 TA = +25C -20 TA = +85C -30 -40 -50 -60 -70 -80 -90 -100 -110 -120 -130 -140 -150 -160 2579 2979 3379 3600 11489-020 1800 -85 TA = -40C -90 TA = +25C -95 TA = +85C -100 100kHz OFFSET -105 -110 -115 -120 800kHz OFFSET -125 -130 -135 -140 6MHz OFFSET -145 -150 40MHz OFFSET -155 -160 -165 1384 1584 1784 1984 2184 2384 2584 2784 LO FREQUENCY (MHz) Figure 22. 120 kHz Bandwidth Loop Phase Noise, LO_DIV_A = 01; Offset = 1 kHz, 50 kHz, 400 kHz, 1 MHz, and 10 MHz Figure 25. 120 kHz Bandwidth Loop Phase Noise, LO_DIV_A = 01; Offset = 100 kHz, 800 kHz, 6 MHz, and 40 MHz Rev. 0 | Page 14 of 52 11489-021 1600 11489-016 230 1400 11489-019 1.2 Data Sheet 0.7 0.6 LO_DIV_A = 11 LO_DIV_A = 10 LO_DIV_A = 01 0.5 0.4 0.3 0.7 0.6 0.5 0.4 0.3 0.2 0.2 0.1 0.1 3168 3568 3968 4368 4768 5168 5568 VCO FREQUENCY (MHz) Figure 26. 10 kHz to 40 MHz Integrated Phase Noise vs. VCO Frequency; LO_DIV_A = 01, 10, and 11, Including Spurs, for Various LO Divider Ratios -85 -90 -95 -100 -105 3568 3968 4368 4768 5168 5568 -75 4368 4768 5168 5568 LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 TA = -40C TA = +25C TA = +85C -80 -85 -90 -95 -100 -105 -110 2768 3168 3568 3968 4368 4768 5168 5568 VCO FREQUENCY (MHz) Figure 30. fPFD Spurs vs. VCO Frequency; 2x PFD Offset; Measured at LO Output -70 -70 -80 -85 -90 -95 -100 -105 3168 3568 3968 4368 4768 5168 VCO FREQUENCY (MHz) 5568 Figure 28. fPFD Spurs vs. VCO Frequency; 3x PFD Offset; Measured at LO Output -75 TA = -40C TA = +25C TA = +85C LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 -80 -85 -90 -95 -100 -105 -110 -115 -120 2768 3168 3568 3968 4368 4768 5168 VCO FREQUENCY (MHz) Figure 31. fPFD Spurs vs. VCO Frequency; 4x PFD Offset; Measured at LO Output Rev. 0 | Page 15 of 52 5568 11489-032 REFERENCE SPURS (dBc), 4x PFD OFFSET LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 TA = -40C TA = +25C TA = +85C 11489-029 REFERENCE SPURS (dBc), 3x PFD OFFSET 3968 Figure 29. 10 kHz to 40 MHz Integrated Phase Noise vs. VCO Frequency; LO_DIV_A = 01, 10, and 11, Excluding Spurs, for Various LO Divider Ratios Figure 27. fPFD Spurs vs. VCO Frequency; 1x PFD Offset; Measured at LO Output -110 2768 3568 VCO FREQUENCY (MHz) 11489-028 3168 VCO FREQUENCY (MHz) -75 3168 11489-031 LO_DIV_A = 01 LO_DIV_A = 10 LO_DIV_A = 11 TA = -40C TA = +25C TA = +85C -80 -110 2768 0 2768 REFERENCE SPURS (dBc), 2x PFD OFFSET -75 LO_DIV_A = 11 LO_DIV_A = 10 LO_DIV_A = 01 -70 -70 REFERENCE SPURS (dBc), 1x PFD OFFSET 0.8 11489-128 0.8 0 2768 TA = -40C TA = +25C TA = +85C 0.9 11489-126 INTEGRATED PHASE NOISE, WITH SPUR ( rms) 0.9 1.0 TA = -40C TA = +25C TA = +85C INTEGRATED PHASE NOISE, WITHOUT SPUR ( rms) 1.0 ADRF6620 ADRF6620 Data Sheet 10 300 TA = -40C TA = +25C TA = +85C 290 8 LO AMPLITUDE (dBm) 230 2 0 LO_DRV_LVL = 01 -2 -4 220 -6 210 -8 200 350 850 1350 1850 2350 2850 LO FREQUENCY (MHz) -10 350 REFERENCE SPURS (dBc), 1x PFD OFFSET -70 -10 -20 -30 -40 -50 -60 -70 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) 2863.2 2858.2 2853.2 2848.2 2843.2 2838.2 2833.2 2828.2 75 100 125 150 TIME (s) 175 200 225 250 11489-137 2823.2 50 2850 -74 -76 -78 -80 -82 -84 -86 -88 -90 -92 -94 -96 -98 1584 1784 1984 2184 2384 2584 2784 Figure 36. fPFD Spurs, LO_DIV_A = 01, 1x PFD Offset; Measured on LO Output and DGA Output 2868.2 25 2350 LO FREQUENCY (MHz) Figure 33. RF to LO Output Feedthrough, LO_DRV_LVL = 00 0 1850 LO OUTPUT DGA OUTPUT -72 -100 1384 11489-136 -80 600 1350 Figure 35. LO Amplitude vs. LO Frequency; LO_DRV_LVL = 00, 01, 10, and 11 0 RF TO LO FEEDTHROUGH (dBc) 850 LO FREQUENCY (MHz) Figure 32. Supply Current vs. LO Frequency; LO_DRV_LVL = 00, 01, 10, and 11 LO FREQUENCY (MHz) LO_DRV_LVL = 00 TA = -40C TA = +25C TA = +85C 11489-135 250 LO_DRV_LVL = 00 LO_DRV_LVL = 01 240 LO_DRV_LVL = 10 LO_DRV_LVL = 11 4 Figure 34. LO Frequency Settling Time, Loop Filter Bandwidth = 120 kHz Rev. 0 | Page 16 of 52 11489-023 260 11489-132 SUPPLY CURRENT (mA) 270 2818.2 LO_DRV_LVL = 11 LO_DRV_LVL = 10 6 280 Data Sheet ADRF6620 RF INPUT TO MIXER OUTPUT PERFORMANCE VCCx = 5 V, TA = 25C, RL = 250 , external LO, PLO = 0 dBm, RFDSA_SEL = 00 (0 dB), RFSW_SEL = 00 (RFIN0), BAL_CIN and BAL_COUT optimized, MIXER_BIAS, MIXER_RDAC, and MIXER_CDAC optimized for highest linearity, DGA and LO output disabled, unless otherwise noted. All losses from input and output traces and baluns are de-embedded from results. 0 0 -1 -1 RF RF RF RF -40C -2 FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz -2 +85C -4 -5 -6 -6 -7 -7 1400 1800 2200 2600 3000 -8 0 20 18 18 16 16 14 14 IP1dB (dBm) 22 20 12 10 8 6 4 4 TA = -40C TA = +25C TA = +85C 2600 3000 600 700 800 900 1000 FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz 0 11489-035 2200 RF FREQUENCY (MHz) 0 100 200 300 400 500 600 700 800 900 1000 IF FREQUENCY (MHz) Figure 41. Mixer IP1dB vs. IF Frequency; LO Sweep with Fixed RF, IF Roll-Off Figure 38. Mixer IP1dB vs. RF Frequency 100 100 TA = -40C TA = +25C TA = +85C 90 80 80 IIP2 (dBm), IIP3 (dBm) IIP2 (dBm) 70 60 50 IIP3 (dBm) 40 30 70 60 50 30 10 10 1800 2200 2600 3000 RF FREQUENCY (MHz) 11489-036 20 1400 IIP3 (dBm) 40 20 1000 IIP2 (dBm) RF RF RF RF 0 0 100 200 300 400 500 FREQUENCY = 900MHz FREQUENCY = 1900MHz FREQUENCY = 2100MHz FREQUENCY = 2700MHz 600 700 800 900 1000 IF FREQUENCY (MHz) Figure 42. Mixer IIP2/IIP3 vs. IF Frequency; PIN = -5 dBm/Tone, 1 MHz Spacing, LO Sweep with Fixed RF, IF Roll-Off Figure 39. Mixer IIP2/IIP3 vs. RF Frequency; PIN = -5 dBm/Tone, 1 MHz Spacing Rev. 0 | Page 17 of 52 11489-039 90 0 600 RF RF RF RF 2 1800 500 10 6 1400 400 12 8 1000 300 IF FREQUENCY (MHz) 22 0 600 200 Figure 40. Mixer Gain vs. IF Frequency; LO Sweep with Fixed RF, IF Roll-Off Figure 37. Mixer Gain vs. RF Frequency 2 100 11489-038 1000 RF FREQUENCY (MHz) IP1dB (dBm) -4 -5 -8 600 IIP2 (dBm), IIP3 (dBm) -3 11489-037 GAIN (dB) -3 11489-034 GAIN (dB) +25C ADRF6620 0 -1 Data Sheet 100 RFSW_SEL = 00 RFSW_SEL = 01 RFSW_SEL = 10 RFSW_SEL = 11 90 RFSW_SEL = 00 RFSW_SEL = 01 RFSW_SEL = 10 RFSW_SEL = 11 IIP2 (dBm) 80 IIP2 (dBm), IIP3 (dBm) GAIN (dB) -2 -3 -4 -5 70 60 IIP3 (dBm) 50 40 30 -6 20 -7 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) Figure 43. Mixer Gain vs. RF Frequency; RFSW_SEL = 00, 01, 10, and 11 0 ISOLATION RFSW_SEL = 00 TO 11 ISOLATION RFSW_SEL = 00 TO 01 ISOLATION RFSW_SEL = 00 TO 10 -5 -10 ISOLATION (dBc) ISOLATION (dBc) -30 -35 -40 -45 -50 -20 -25 -30 -35 -40 -65 -55 -70 -60 1400 1800 2200 2600 3000 -65 600 11489-142 1000 RF FREQUENCY (MHz) 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) Figure 44. Mixer Input to Mixer Output Isolation vs. RF Frequency; RFSW_SEL = 00 Driven Figure 47. Mixer Input to Mixer Output Isolation vs. RF Frequency; RFSW_SEL = 11 Driven 0 ISOLATION RFSW_SEL = 01 TO 11 ISOLATION RFSW_SEL = 01 TO 00 ISOLATION RFSW_SEL = 01 TO 10 -5 -10 -15 -20 -20 ISOLATION (dBc) -15 -25 -30 -35 -40 -45 -30 -35 -40 -45 -50 -55 -55 -60 -60 -65 ISOLATION RFSW_SEL = 10 TO 11 ISOLATION RFSW_SEL = 10 TO 00 ISOLATION RFSW_SEL = 10 TO 01 -25 -50 -65 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) 11489-141 ISOLATION (dBc) ISOLATION RFSW_SEL = 11 TO 11 ISOLATION RFSW_SEL = 11 TO 00 ISOLATION RFSW_SEL = 11 TO 01 -50 -60 -70 600 3000 -45 -55 -10 2600 -15 -25 0 2200 Figure 46. Mixer IIP2/IIP3 vs. RF Frequency; RFSW_SEL = 00, 01, 10, and 11 -20 -5 1800 RF FREQUENCY (MHz) -15 -75 600 1400 11489-145 -10 1000 Figure 45. Mixer Input to Mixer Output Isolation vs. RF Frequency; RFSW_SEL = 01 Driven -70 600 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) Figure 48. Mixer Input to Mixer Output Isolation vs. RF Frequency; RFSW_SEL = 10 Driven Rev. 0 | Page 18 of 52 11489-144 0 -5 0 600 11489-143 10 11489-140 -8 600 LO TO IF FEEDTHROUGH (dBm) Data Sheet ADRF6620 0 300 -5 275 INTERNAL LO 250 -10 225 -15 200 -20 EXTERNAL LO ICC (mA) 175 150 -25 125 -30 100 -35 75 50 -45 25 1200 1600 2000 2400 2800 3200 LO FREQUENCY (MHz) 0 600 11489-146 -50 800 TA = -40C TA = +25C TA = +85C 1000 1400 1800 2200 2600 3000 RF FREQUENCY (MHz) 11489-149 -40 Figure 52. ICC vs. RF Frequency; DGA and LO Output Disabled Figure 49. LO to IF Feedthrough at Mixer Output Without Filtering 24 0 23 -5 21 SSB NOISE FIGURE (dB) RF TO IF FEEDTHROUGH (dBc) 22 -10 -15 -20 -25 -30 -35 -40 20 OPTIMIZED FOR HIGH LINEARITY 19 18 17 16 15 NOISE FIGURE OPTIMIZED 14 13 -45 12 -50 1600 2000 2400 2800 3200 RF FREQUENCY (MHz) Figure 50. RF to IF Feedthrough at Mixer Output Without Filtering; Mixer Input Power = 0 dBm 0 LO TO RF FEEDTHROUGH (dBm) -20 -30 -40 -50 EXTERNAL LO -70 INTERNAL LO -80 -90 600 850 1100 1350 1600 1850 2100 2350 2600 2850 LO FREQUENCY (MHz) 11489-148 -100 -110 350 1000 1400 1800 2200 2600 RF FREQUENCY (MHz) Figure 53. SSB Noise Figure vs. RF Frequency (see Table 16) -10 -60 10 600 Figure 51. LO to RF Feedthrough; PLO = 0 dBm Rev. 0 | Page 19 of 52 11489-150 1200 11489-147 11 -55 800 ADRF6620 Data Sheet IF DGA GAIN = 7dB GAIN = 3dB TA = +85C 100 150 200 TA = +25C 250 300 TA = -40C 350 400 450 500 IF FREQUENCY (MHz) 0.1 0 -0.1 -0.2 -0.3 -0.4 3 4 20 18 18 16 16 14 14 12 12 OP1dB (dB) 5 6 7 8 9 10 11 12 13 14 15 -0.5 10 8 10 8 6 4 TA = +85C TA = +25C TA = -40C 100 150 200 250 300 350 400 450 500 IF FREQUENCY (MHz) TA = +85C TA = +25C TA = -40C 2 0 11489-152 2 3 4 5 6 7 8 9 10 11 12 13 14 15 GAIN (dB) Figure 55. DGA OP1dB vs. Frequency and Temperature; Maximum Gain 11489-155 4 Figure 58. DGA OP1dB vs. Gain Setting and Temperature 70 65 60 OIP2 (dBm) OIP2 (dBm) OIP2 (dBm), OIP3 (dBm) 55 OIP3 (dBm) 50 45 40 OIP3 (dBm) 35 30 25 20 15 10 TA = +85C TA = +25C TA = -40C 150 200 250 300 350 400 450 500 IF FREQUENCY (MHz) 11489-153 100 TA = +85C TA = +25C TA = -40C 5 Figure 56. DGA OIP2/OIP3 vs. IF Frequency and Temperature; Maximum Gain 0 3 4 5 6 7 8 9 10 11 12 13 14 15 GAIN (dB) Figure 59. DGA OIP2/OIP3 vs. Gain Setting and Temperature Rev. 0 | Page 20 of 52 11489-156 OP1dB (dB) 0.2 GAIN (dB) 6 OIP2 (dBm), OIP3 (dBm) 0.3 20 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 50 0.4 Figure 57. DGA Gain and Gain Step Error vs. Gain Setting and Temperature Figure 54. DGA Gain vs. IF Frequency and Temperature 0 50 0.5 TA = -40C TA = +25C TA = +85C GAIN STEP ERROR (dB) GAIN = 11dB 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 11489-259 GAIN = 15dB GAIN (dB) 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 50 11489-151 GAIN (dB) VCCx = 5 V, TA = 25C, RS = RL = 150 , IF = 200 MHz, 2 V p-p differential output, unless otherwise noted. All losses from input and output traces and baluns are de-embedded from results. Data Sheet ADRF6620 -60 -20 -70 -20 -80 -30 -80 -30 -50 -110 -60 -120 -70 -80 -130 -80 -90 -140 -90 -50 -110 -60 -120 -70 -130 -140 200 250 300 350 400 450 IF FREQUENCY (MHz) -150 3 8 9 10 11 12 13 -100 15 14 OIP2 (dBm) 65 60 -90 -40 -100 -50 -110 -60 -120 -70 -130 -80 -140 -90 55 OIP2 (dBm), OIP3 (dBm) -30 HD3 (dBc) -20 -80 1 2 3 4 5 6 7 8 45 40 OIP3 (dBm) 35 30 25 20 GAIN = 15dB GAIN = 11dB GAIN = 7dB GAIN = 3dB 10 5 0 -7 11489-158 -7 -6 -5 -4 -3 -2 -1 0 50 15 -100 -150 9 10 POUT (dBm) -6 -5 -4 -3 -5 -1 0 1 2 3 4 5 POUT (dBm) Figure 61. DGA HD2/HD3 vs. Output Power (POUT) and Gain Setting Figure 64. DGA OIP2/OIP3 vs. Output Power (POUT) and Gain Setting 0 0 TA = +85C TA = +25C TA = -40C TA = +85C TA = +25C TA = -40C -10 -20 -40 -50 IMD2 (dBc) -60 -70 -80 -30 -40 -50 -70 -80 IMD3 (dBc) -90 -90 -100 -100 50 100 150 IMD2 (dBc) -60 200 250 300 350 400 450 500 IF FREQUENCY (MHz) Figure 62. DGA IMD2/IMD3 vs. IF Frequency and Temperature; Maximum Gain IMD3 (dBc) 3 4 5 6 7 8 9 10 11 12 13 GAIN (dB) Figure 65. DGA IMD2/IMD3 vs. Gain Setting Rev. 0 | Page 21 of 52 14 15 11489-162 IMD2 (dBc), IMD3 (dBc) -20 -30 11489-159 IMD2 (dBc), IMD3 (dBc) 7 70 -10 -70 -10 6 Figure 63. DGA HD2/HD3 vs. Gain Setting and Temperature 0 GAIN = 15dB GAIN = 11dB GAIN = 7dB GAIN = 3dB -60 5 GAIN (dB) Figure 60. DGA HD2/HD3 vs. IF Frequency and Temperature; Maximum Gain -50 4 11489-161 150 11489-157 100 -100 500 HD2 (dBc) -40 -40 HD3 (dBc) -90 -100 -90 -100 50 -10 HD3 (dBc) -10 -70 -150 HD2 (dBc) 0 TA = +85C TA = +25C TA = -40C 11489-160 TA = +85C TA = +25C TA = -40C -60 HD2 (dBc) -50 0 -50 ADRF6620 Data Sheet SPURIOUS PERFORMANCE (N x fRF) - (M x fLO) spur measurements were made using the standard evaluation board. Mixer spurious products were measured in decibels (dB) relative to the carrier (dBc) from the IF output power level. Data is shown for all spurious components greater than -115 dBc and frequencies of less than 3 GHz. 915 MHz Performance VCCx = 5 V, TA = 25C, RF power = 0 dBm, internal LO, fRF = 914 MHz, fLO = 1114 MHz M 0 N 0 1 2 3 4 5 6 -43 -72 -102 1 -34 0 -60 -73 -102 2 -35 -52 -72 -103 <-115 <-115 3 4 5 6 -16 -67 -78 <-115 -105 <-115 -74 <-115 <-115 <-115 <-115 -80 <-115 <-115 <-115 <-115 <-115 1910 MHz Performance VCCx = 5 V, TA = 25C, RF power = 0 dBm, internal LO, fRF = 1910 MHz, fLO = 2110 MHz. M 0 N 0 1 2 3 4 5 6 1 -38.208 -0.001 -59.208 -40.462 2 -50.9 -69.655 -106.741 3 4 -62.35 -74.322 <-115 <-115 -106.429 <-115 <-115 5 6 <-115 -110.954 <-115 <-115 2140 MHz Performance VCCx = 5 V, TA = 25C, RF power = 0 dBm, internal LO, fRF = 2140 MHz, fLO = 2340 MHz. M 0 N 0 1 2 3 4 5 6 -36 1 -40 0 -58 2 -45 -67 <-115 3 -59 -74 <-115 Rev. 0 | Page 22 of 52 4 <-115 <-115 <-115 5 6 <-115 <-115 <-115 <-115 <-115 Data Sheet ADRF6620 2700 MHz Performance VCCx = 5 V, TA = 25C, RF power = 0 dBm, internal LO, fRF = 2700 MHz, fLO = 2500 MHz. M 0 N 0 1 2 3 4 5 6 -40.126 1 -38.613 -0.001 -58.299 2 -43.84 -67.06 3 -62.116 -73.603 Rev. 0 | Page 23 of 52 4 <-115 <-115 5 <-115 <-115 6 <-115 <-115 ADRF6620 Data Sheet THEORY OF OPERATION The ADRF6620 integrates the essential elements of a multichannel loopback receiver that is typically used in digital predistortion systems. The main features of the ADRF6620 include a single-pole four throw (SP4T) RF input switch with tunable balun, variable attenuation, a wideband active mixer, and digitally programmable variable gain amplifier (DGA). In addition, the ADRF6620 integrates a local oscillator (LO) generation block consisting of a synthesizer and a multicore voltage controlled oscillator (VCO) with an octave range and low phase noise. The synthesizer uses a fractional-N phase-locked loop (PLL) to enable continuous LO coverage from 350 MHz to 2850 MHz. Putting all the building blocks of the ADRF6620 together, the signal path through the device starts at the RF input, where one of four single-ended RF inputs is selected by the input mux and converted to a differential signal via a tunable balun. The differential RF signal is attenuated to an optimal input level via the digital step attenuator with 15 dB of attenuation range in steps of 1 dB. The RF signal is then mixed via a Gilbert cell mixer with the LO signal down to an IF frequency. The 255 terminated differential output of the mixer is brought off chip to a pair of inductors and passed through an IF filter. The output of the IF filter is ac-coupled off chip and fed to an on-chip digital attenuator and IF DGA. The output of the IF DGA is then passed to an off-chip analog-to-digital converter (ADC). 100 ns. When serial port control is used, the switch time is 100 ns, plus the latency of the SPI programming. The RFSW_MUX bit (Register 0x23, Bit 11) selects whether the RF input switch is controlled via the external pins or the SPI port. By default at power-up, the device is configured for serial control. Writing to the RFSW_SEL bits (Register 0x23, Bits[10:9]) allows selection of one of the four RF inputs. Alternatively, by setting the RFSW_MUX bit high, the RFSW0 and RFSW1 pins can be used to select the RF input. Table 10 summarizes the different control options for the RF inputs. To maintain good channel-to-channel isolation, ensure that unused RF inputs are properly terminated. The RFINx ports are internally terminated with 50 resistors and have a dc bias level of 2.5 V. To avoid disrupting the dc level, the recommended termination is a dc blocking capacitor to GND. Figure 66 shows the recommended configuration when only RFIN0 is used, and the other RF input ports are properly terminated. RFIN0 35 50 RFIN1 32 0.1F 50 RFIN2 29 0.1F The ADRF6620 integrates a SP4T switch where one of four RF inputs is selected. The desired RF input can be selected using either pin control or register writes via the SPI. Compared to the serial write approach, pin control allows faster control over the switch. When the RFSW0 pin (Pin 38) and the RFSW1 pin (Pin 39) are used, the RF switches can switch at speeds of up to 50 RFIN3 26 0.1F 50 11489-168 RF INPUT SWITCHES Figure 66. Terminating Unused RF Input Ports Table 10. RF Input Selection Table RFSW_MUX (Register Address 0x23[11]) Bit 11 0 0 0 0 1 1 1 1 1 SPI Control, RFSW_SEL (Register Address 0x23[10:9]) Bit 10 Bit 9 0 0 0 1 1 0 1 1 X1 X1 X1 X1 X1 X1 X1 X1 X = don't care. Rev. 0 | Page 24 of 52 Pin Control RFSW1, Pin 39 RFSW0, Pin 38 X1 X1 1 X X1 1 X X1 X1 X1 0 0 0 1 1 0 1 1 RF Input RFIN0 RFIN1 RFIN2 RFIN3 RFIN0 RFIN1 RFIN2 RFIN3 Data Sheet ADRF6620 TUNABLE BALUN +5V 9 IFIN+ RS RIN gm AMP ROUT 8 11 10 LOGIC IFOUT1- IFOUT1+ IFOUT2- IFOUT2+ REF RL Figure 68. Simplified IF DGA Schematic An independent internal voltage reference circuit sets the dc voltage level at the input of the amplifier to approximately 1.5 V. This reference is not accessible and cannot be adjusted. 11489-040 BAL_COUT REG 0x30[7:5] 15 ATTENUATOR IFIN- RFINx BAL_CIN REG 0x30[3:1] 16 11489-041 The ADRF6620 integrates a programmable balun operating over a frequency range from 700 MHz to 2700 MHz. The tunable balun offers the benefit of ease of drivability from a single-ended 50 RF input, and the single-ended-to-differential conversion of the balun optimizes common-mode rejection. Figure 67. Integrated Tunable Balun The RF balun is tuned by switching parallel capacitances on the primary and secondary sides by writing to Register 0x30. The added capacitance, in parallel with the inductive windings of the balun, changes the resonant frequency of the inductive capacitive (LC) tank. Therefore, selecting the proper combination of BAL_ CIN (Register 0x30, Bits[3:1]) and BAL_COUT (Register 0x30, Bits[7:5]) sets the desired frequency and minimizes the insertion loss of the balun. Under most circumstances, the input and output can be tuned together; however, sometimes for matching reasons, it may be advantageous to tune them separately. See the RF Input Balun Insertion Loss Optimization section for the recommended BAL_CIN and BAL_COUT settings. The IF DGA consumes 35 mA through the VCC2 pin (Pin 12) and 75 mA through the two output choke inductors. The IF DGA can be powered down by disabling the IF_AMP_EN bit (Register 0x01, Bit 11). In its power-down state, the IF DGA current reduces to 6 mA. The dc bias level at the input remains at approximately 1.5 V when the DGA is disabled. At minimum attenuation, the gain of the IF DGA is 15 dB when driving a 150 load. The source and load resistance of the amplifier is set to 150 in a matched condition. If the load or the source resistance is not equal to 150 , the following equations can be used to determine the resulting gain and input/output resistances. RF DIGITAL STEP ATTENUATOR (DSA) Voltage Gain = AV = 0.044 x (1000||RL) The RF DSA follows the tunable balun. The attenuation range is 0 dB to 15 dB with a step size of 1 dB. DSA attenuation is set using the RFDSA_SEL bits (Register 0x23, Bits[8:5]). RIN = (1000 + RL)/(1 + 0.044 x RL) ACTIVE MIXER The double balanced mixer uses high performance SiGe NPN transistors. This mixer is based on the Gilbert cell design of four cross-connected transistors. The mixer output has a 255 differential output resistance. Bias the mixer outputs using either a pair of supply referenced RF chokes or an output transformer with the center tap connected to the positive supply. DIGITALLY PROGRAMMABLE VARIABLE GAIN AMPLIFIER (DGA) The ADRF6620 integrates a differential IF DGA consisting of a 150 digitally controlled passive attenuator followed by a highly linear transconductance amplifier with feedback. The attenuation range is 12 dB, and the transconductor amplifier has a fixed gain of 15 dB. Therefore, at minimum attenuation, the gain of the IF DGA is 15 dB; at maximum attenuation, the gain is 3 dB. The attenuation is controlled by addressing the IF_ATTN bits in Register 0x23, Bits[4:0]. The attenuation step size is 0.5 dB. S21 (Gain) = 2 x RIN/(RIN + RS) x AV ROUT = (1000 + RS)/(1 + 0.044 x RS) The dc current to the outputs of each amplifier is supplied through two external choke inductors. The inductance of the chokes and the resistance of the load, in parallel with the output resistance of the device, add a low frequency pole to the response. The parasitic capacitance of the chokes adds to the output capacitance of the part. This total capacitance, in parallel with the load and output resistance, sets the high frequency pole of the device. In general, the larger the inductance of the choke, the higher the parasitic capacitance. Therefore, this trade-off must be considered when the value and type of the choke are selected. For each polarity, the amplifier has two output pins that are oriented in an alternating fashion: IFOUT1+ (Pin 8), IFOUT1- (Pin 9), IFOUT2+ (Pin 10), and IFOUT2- (Pin 11). When designing the board, minimize the parasitic capacitance caused by routing the corresponding outputs together. See the Layout section for the recommended printed circuit board (PCB) layout. Rev. 0 | Page 25 of 52 ADRF6620 Data Sheet LO GENERATION BLOCK Internal LO Mode The ADRF6620 offers two modes for sourcing the LO signal to the mixer. The first mode uses the on-chip PLL and VCO. This mode of operation provides a high quality LO that meets the performance requirements of most applications. Using the onchip synthesizer and VCO removes the burden of generating and distributing a high frequency LO signal. The ADRF6620 includes an on-chip VCO and PLL for LO synthesis. The PLL, shown in Figure 69, consists of a reference input, phase and frequency detector (PFD), charge pump, and a programmable integer divider with prescaler. The reference path takes in a reference clock and divides it down by a factor of 1, 2, 4, or 8 or multiplies it by a factor of 2 before passing it to the PFD. The PFD compares this signal to the divided down signal from the VCO. Depending on the PFD polarity selected, the PFD sends an up/down signal to the charge pump if the VCO signal is slow/fast compared to the reference frequency. The charge pump sends a current pulse to the off-chip loop filter to increase or decrease the tuning voltage (VTUNE). The second mode bypasses the integrated LO generation block and allows the LO to be supplied externally. This second mode can provide a very high quality signal directly to the mixer core. Sourcing the LO signal externally may be necessary in demanding applications that require the lowest possible phase noise performance. The ADRF6620 integrates three VCO cores that cover an octave range from 2.8 GHz to 5.7 GHz. Table 11 summarizes the frequency range for each VCO. The desired VCO can be selected by addressing the VCO_SEL bits (Register 0x22, Bits[2:0]). External LO Mode External or internal LO mode can be selected via the VCO_SEL bits (Register 0x22, Bits[2:0]). To configure for external LO mode, set Register 0x22, Bits[2:0] to 011 and apply the differential LO signals to Pin 44 (LOIN-) and Pin 45 (LOIN+). The external LO frequency range is 350 MHz to 3.2 GHz. The ADRF6620 offers the flexibility of using a higher LO frequency signal and dividing it down before it drives the mixer. The LO divider can be found in the LO_DIV_A bits (Register 0x22, Bits[4:3]), where options include /1, /2, /4, or /8. VCO_SEL (Register 0x22, Bits[2:0]) 000 001 010 011 The external LO input pins present a broadband differential 50 input impedance. The LOIN+ and LOIN- input pins must be ac-coupled. When not in use, LOIN+ and LOIN- can be left unconnected. The N-divider divides down the differential VCO signal to the PFD frequency. The N-divider can be configured for fractional mode or integer mode by addressing the DIV_MODE bit (Register 0x02, Bit 11). The default configuration is set for fractional mode. Table 11. VCO Range Frequency Range (GHz) 5.2 to 5.7 4.1 to 5.2 2.8 to 4.1 External LO VCO_SEL REG 0x22[2:0] REFSEL REG 0x21[2:0] LOIN+ /8 /4 /2 PFD_POLARITY REG 0x21[3] PFD + CP CHARGE PUMP LO_DIV_A REG 0x22[4:3] LOIN- /1, /2, /4, /8 VTUNE LOOUT+ TO MIXER LOOUT- TO MIXER LPF x1 x2 CP_CTRL REG 0x20[13:0] N = INT + FRAC MOD DIV_MODE: REG 0x02[11] INT_DIV: REG 0x02[10:0] FRAC_DIV: REG 0x03[10:0] MOD_DIV: REG 0x04[10:0] Figure 69. LO Generation Block Diagram Rev. 0 | Page 26 of 52 /2 11489-042 REFIN EXTERNAL LOOP FILTER Data Sheet ADRF6620 The following equations can be used to determine the N value and PLL frequency: f PFD = f LO = Additional LO Controls f VCO 2x N N = INT + FRAC MOD f PFD x 2 x N LO_DIVIDER where: fPFD is the phase frequency detector frequency. fVCO is the voltage controlled oscillator frequency. N is the fractional divide ratio (INT + FRAC/MOD) INT is the integer divide ratio programmed in Register 0x02. FRAC is the fractional divider programmed in Register 0x03. MOD is the modulus divide ratio programmed in Register 0x04. fLO is the LO frequency going to the mixer core when the loop is locked. LO_DIVIDER is the final divider block that divides the VCO frequency down by 1, 2, 4, or 8 before it reaches the mixer (see Table 12). This control is located in the LO_DIV_A bits (Register 0x22, Bits[4:3]). Table 12. LO Divider LO_DIV_A (Register 0x22, Bits[4:3]) 00 01 10 11 captures these effects. In general, higher bandwidth loops tend to lock more quickly than lower bandwidth loops. To access the LO signal going to the mixer core through the LOOUT+ and LOOUT- pins (Pin 21 and Pin 22), enable the LO_DRV_EN bit in Register 0x01, Bit 7. This setting offers direct monitoring of the LO signal to the mixer for debug purposes; or the LO signal can be used to daisy-chain many devices synchronously. One ADRF6620 can serve as the master where the LO signal is sourced, and the subsequent slave devices share the same LO signal from the master. This flexibility substantially eases the LO requirements of a system with multiple LOs. The LO output drive level is controlled by the LO_DRV_LVL bits (Register 0x22, Bits[8:7]). Table 13 shows the available drive levels. Table 13. LO Drive Level LO_DRV_LVL (Register 0x22, Bits[8:7]) 00 01 10 11 Amplitude (dBm) -4 0.5 3 4.5 SERIAL PORT INTERFACE (SPI) The SPI port of the ADRF6620 allows the user to configure the device through a structured register space provided inside the chip. Registers are accessed via the serial port interface and can be written to or read from via the serial port interface. LO_DIVIDER 1 2 4 8 The lock detect signal is available as one of the selectable outputs through the MUXOUT pin; a logic high indicates that the loop is locked. The MUXOUT pin is controlled by the REF_MUX_SEL bits (Register 0x21, Bits[6:4]); the PLL lock detect signal is the default configuration. To ensure that the PLL locks to the desired frequency, follow the proper write sequence of the PLL registers. The PLL registers must be configured accordingly to achieve the desired frequency, and the last writes must be to Register 0x02 (INT_DIV), Register 0x03 (FRAC_DIV), or Register 0x04 (MOD_DIV). When one of these registers is programmed, an internal VCO calibration is initiated, which is the last step in locking the PLL. The time it takes to lock the PLL after the last register is written can be broken down into two parts: VCO band calibration and loop settling. After the last register is written, the PLL automatically performs a VCO band calibration to choose the correct VCO band. This calibration takes approximately 5120 PFD cycles. For a 40 MHz fPFD, this corresponds to 128 s. After calibration is complete, the feedback action of the PLL causes the VCO to eventually lock to the correct frequency. The speed with which this locking occurs depends on the nonlinear cycle-slipping behavior, as well as the small-signal settling of the loop. For an accurate estimation of the lock time, download the ADIsimPLL tool, which correctly The serial port interface consists of three control lines: SCLK, SDIO, and CS. SCLK (serial clock) is the serial shift clock. The SCLK signal clocks data on its rising edge. SDIO (serial data input/output) is an input or output depending on the instruction being sent and the relative position in the timing frame. CS (chip select bar) is an active low control that gates the read and write cycles. The falling edge of CS, in conjunction with the rising edge of SCLK, determines the start of the frame. All SCLK and SDIO activity is ignored when CS is high. Table 6 and Figure 2 show the serial timing and its definitions. The ADRF6620 protocol consists of seven register address bits, followed by a read/write indicator and 16 data bits. Both the address and data fields are organized from MSB to LSB. On a write cycle, up to 16 bits of serial write data are shifted in, MSB to LSB. If the rising edge of CS occurs before the LSB of the serial data is latched, only the bits that were clocked in are written to the device. If more than 16 data bits are shifted in, the 16 most recent bits are written to the device. The ADRF6620 input logic level for the write cycle supports a logic level as low as 1.8 V. On a read cycle, up to 16 bits of serial read data are shifted out, MSB to LSB. Data shifted out beyond 16 bits is undefined. It is not necessary for readback content at a given register address to correspond with the write data of the same address. The output logic level for a read cycle is 2.5 V. Rev. 0 | Page 27 of 52 ADRF6620 Data Sheet BASIC CONNECTIONS +5V DNP 0.1F (0402) 4 RFSW1 RFSW0 RFIN0 RFIN2 RFIN3 100pF (0402) REF_IN RFIN3 REFIN 11 10 29 21 26 6 49.9 (0402) MUXOUT MUXOUT 0 (0402) 43 EXPOSED PADDLE LOIN+ LOIN- /8 /4 /2 x1 x2 PFD CHARGE PUMP + CP 22 /1, /2, /4, /8 VTUNE FRAC N = INT + MOD /2 LOCK_DET VPTAT 44 LDO VCO LDO LO LDO 2.5V 46 37 7 LDO 3.3V 47 3 24 14 13 VCC5 1 12 100pF (0402) 100pF (0402) 100pF (0402) 100pF (0402) 100pF (0402) 0.1F (0402) 0.1F (0402) 0.1F (0402) 0.1F (0402) 0.1F (0402) 0.1F (0402) 10F (0603) 3 4 IFOUT2+ LOOUT+ 100pF (0402) LOOUT- LOIN+ DNP TC1-1-43A+ 3 4 1 LOOUT 6 100pF (0402) TC1-1-43A+ 3 4 LOIN- 1 VTUNE 100pF (0402) CP 22pF (0402) LOIN 6 10k (0402) 3k (0402) 10k (0402) 6.8pF (0402) OPEN (0402) VTUNE_TP OPEN 22pF (0402) 2.7nF (0603) 100pF (0402) 0.1F (0402) 100pF (0402) 0.1F (0402) 0.1F (0402) 10F (0603) 10F (0603) 100pF (0402) 11489-043 10F (0603) 6 2 100pF 0 (0402) (0402) IFOUT2- 2 100pF (0402) +5V RED LOIN 100pF (0402) 45 TCM3-1T+ 0.1F (0402) 1 DNP IFOUT1+ DECL1 100pF (0402) 32 DECL2 RFIN2 8 RFIN1 1H 100pF 0 (0402) (0402) 9 IFOUT1- DECL3 RFIN1 100pF (0402) 1H 35 DECL4 100pF (0402) +5V 42 41 40 SPI INTERFACE VCC1 RFIN0 18 19 15 16 ADRF6620 38 VCC2 S1 17 20 23 25 27 28 30 31 33 34 36 48 VCC3 100pF (0402) 5 CSB SCLK SDIO 39 VCC4 10k S2 (0402) SDIO 10k (0402) 3.3V MXOUT- MXOUT+ GND 3.3V CS 39nH (0402) IFIN+ 0.1F (0402) IFIN- DNP 39nH (0402) SCLK 470nH (0603) 470nH (0603) Figure 70. Basic Connection Diagram Table 14. Basic Connections Pin No. 5 V Power 1 12 13 14 24 PLL/VCO 3 Mnemonic Description Basic Connection VCC1 VCC2 VCC3 VCC4 VCC5 LO, VCO, mixer power supply IF DGA power supply Factory calibration pin Factory calibration pin RF front-end power supply Decouple all power supply pins to ground using 100 pF and 0.1 F capacitors. Place the decoupling capacitors close to the pins. CP Synthesizer charge pump output Synthesizer reference frequency input Connect this pin to the VTUNE pin through the loop filter. 6 REFIN 21, 22 LOOUT+, LOOUT- Differential LO outputs The nominal input level of this pin is 1 V p-p. The input range is 12 MHz to 464 MHz. This pin is internally biased and must be accoupled and terminated externally with a 50 resistor. Place the ac coupling capacitor between the pin and the resistor. When driven from an 50 RF signal generator, the recommended input level is 4 dBm. The differential output impedance of these pins is 50 . The pins Rev. 0 | Page 28 of 52 Data Sheet Pin No. ADRF6620 Mnemonic Description 44, 45 LOIN-, LOIN+ Differential LO inputs 43 MUXOUT PLL multiplex output 47 VTUNE VCO tuning voltage RFIN3, RFIN2 RFIN1, RFIN0 RF inputs RFSW0, RFSW1 Pin control of the RF inputs IFOUT1+, IFOUT1-, IFOUT2+, IFOUT2- IF DGA outputs IFIN-, IFIN+ IF DGA inputs MXOUT+, MXOUT- Differential mixer outputs The output stage of the mixer is an open collector configuration that requires a dc bias of 5 V. Use bias choke inductors to achieve this configuration. Carefully choose the bias choke inductors such that they can handle a maximum current of 50 mA on each side. The differential output impedance of the mixer is 255 . CS SCLK SDIO SPI chip select SPI clock SPI data input and output Active low. 3.3 V logic levels. 3.3 V tolerant logic levels. 3.3 V tolerant logic levels. DECL1 DECL2 DECL3 DECL4 3.3 V LDO decoupling 2.5 V LDO decoupling LO LDO decoupling VCO LDO decoupling Decouple all DECLx pins to ground using 100 pF, 0.1 F, and 10 F capacitors. Place the decoupling capacitors close to the pin. GND Ground Connect these pins to the GND of the PCB. Exposed pad (EPAD) The exposed thermal pad is on the bottom of the package. The exposed pad must be soldered to ground. RF Inputs 26, 29, 32, 35 38, 39 IF DGA 8, 9, 10, 11 15, 16 Mixer Outputs 18, 19 Serial Port Interface 40 41 42 LDO Decoupling 2 7 37 46 GND 4, 5, 17, 20, 23, 25, 27, 28, 30, 31, 33, 34, 36, 48 49 (EPAD) Basic Connection are internally biased to 2.5 V and must be ac-coupled. The differential input impedance of these pins is 50 . The pins are internally biased to 2.5 V and must be ac-coupled. This output pin provides the PLL reference signal or the PLL lock detect signal. This pin is driven by the output of the loop filter; its nominal input voltage range is 1.5 V to 2.5 V. The single-ended RF inputs have a 50 input impedance and are internally biased to 2.5 V. These pins must be ac-coupled. Terminate unused RF inputs with a dc blocking capacitor to GND to improve isolation. Refer to the Layout section for the recommended PCB layout for optimized channel-to-channel isolation. See Table 10 for the pin settings for RF input pin control. For logic high, connect these pins to 2.5 V logic. The differential IF DGA outputs have two output pins for each polarity. They are oriented in alternating fashion: IFOUT1+ (Pin 8), IFOUT1- (Pin 9), IFOUT2+ (Pin 10), and IFOUT2- (Pin 11). Connect the positive pins such that IFOUT1+ and IFOUT2+ are tied together. Similarly, connect the negative pins such that IFOUT1- and IFOUT2- are tied together. Refer to the Layout section for a recommended layout that minimizes parasitic capacitance and optimizes on performance. The output stage of the IF DAG is an open-collector configuration that requires a dc bias of 5 V. Use bias choke inductors to achieve this configuration. Choose the bias choke inductors such that they can handle a maximum current of 50 mA on each side. By design, the IF DGA is optimized for linearity when the source and load are terminated with 150 . AC couple the mixer outputs to the IF DGA inputs. See the Interstage Filtering Requirements section for the recommended filter designs. Rev. 0 | Page 29 of 52 ADRF6620 Data Sheet RF INPUT BALUN INSERTION LOSS OPTIMIZATION As shown in Figure 71 to Figure 74, the gain of the ADRF6620 mixer has been characterized for every combination of BAL_CIN and BAL_COUT (Register 0x30). As shown, a range of BAL_CIN and BAL_COUT values can be used to optimize the gain of the ADRF6620. The optimized values do not change with temperature. After the values are chosen, the absolute gain changes over temperature; however, the signature of the BAL_CIN and BAL_COUT values is fixed. 0 At lower input frequencies, more capacitance is needed. This increase is achieved by programming higher codes into BAL_CIN and BAL_COUT. At high frequencies, less capacitance is required; therefore, lower BAL_CIN and BAL_COUT codes are appropriate. Table 16 provides a list of recomended BAL_CIN and BAL_COUT codes for popular radio frequencies. Use Figure 71 to Figure 74 and Table 16 only as guides; do not interpret them in the absolute sense because every application and PCB design varies. Additional fine-tuning may be necessary to achieve the maximum gain. 0 -40C +25C +85C -1 -2 -3 -2 -4 GAIN (dB) GAIN (dB) -40C +25C +85C -1 -3 -5 -6 -4 -7 -8 -5 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 BAL_CIN BAL_COUT 0 1 2 3 4 5 6 7 BAL_CIN/BAL_COUT 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 BAL_CIN BAL_COUT 0 1 2 3 4 5 6 7 BAL_CIN/BAL_COUT Figure 71. Gain vs. BAL_CIN and BAL_COUT at RF = 900 MHz 0 11489-045 -6 11489-044 -9 -10 Figure 73. Gain vs. BAL_CIN and BAL_COUT at RF = 1900 MHz 0 -40C +25C +85C -2 -40C +25C +85C -2 -4 GAIN (dB) -6 -6 -8 -8 -10 -12 -14 -10 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 BAL_CIN BAL_COUT 0 1 2 3 4 5 6 7 BAL_CIN/BAL_COUT -18 Figure 72. Gain vs. BAL_CIN and BAL_COUT at RF = 2100 MHz 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 BAL_CIN BAL_COUT 0 1 2 3 4 5 6 7 BAL_CIN/BAL_COUT Figure 74. Gain vs. BAL_CIN and BAL_COUT at RF = 2700 MHz Rev. 0 | Page 30 of 52 11489-047 -16 -12 11489-046 GAIN (dB) -4 Data Sheet ADRF6620 IP3 AND NOISE FIGURE OPTIMIZATION The ADRF6620 can be configured for either improved performance or reduced power consumption. In applications where performance is critical, the ADRF6620 offers IP3 or noise figure optimization. However, if power consumption is the priority, the mixer bias current can be reduced to save on the overall power at the expense of degraded performance. Whatever the application specific needs are, the ADRF6620 offers configurability that balances performance and power consumption. Adjustments to the mixer bias setting have the most impact on performance and power. For this reason, mixer bias should be the first adjustment. The active mixer core of the ADRF6620 is a linearized transconductor. With increased bias current, the transconductor becomes more linear, resulting in higher IP3. The improved IP3, however, is at the expense of degraded noise figure and increased power consumption (see Figure 75). For a 1-bit change of the mixer bias (MIXER_BIAS, Register 0x31, Bits[11:9]), the current increases by 7.71 mA. Inevitably, there is a limit on how much the bias current can increase before the improvement in linearity no longer justifies the increase in power and noise. The mixer core reaches a saturation point where further increases in bias current do not translate to improved performance. When that point is reached, it is best to decrease the bias current to a level where the desired performance is achieved. Depending on the system specifications of the customer, a balance between linearity, noise figure, and power can be attained. In addition to bias optimization, the ADRF6620 also has configurable distortion cancellation circuitry. The linearized transconductor input of the ADRF6620 is made up of a main path and a secondary path. Through adjustments of the amplitude and phase of the secondary path, the distortion generated by the main path can be canceled, resulting in improved IPd3 performance. The amplitude and phase adjustments are located in the following serial interface bits: MIXER_RDAC (Register 0x31, Bits[8:5]) and MIXER_CDAC (Register 0x31, Bits[4:0]). 220 RF FREQ: 900MHz 1900MHz 210 2100MHz 2600MHz 205 215 200 190 185 7.71 mA 180 1 175 170 165 160 155 150 0 1 2 3 4 5 6 7 MIXER BIAS 11489-057 ICC (mA) 195 Figure 75. Change in Current Consumption vs. MIXER_BIAS Rev. 0 | Page 31 of 52 ADRF6620 Data Sheet Figure 76 to Figure 83 show the IIP3 and noise figure sweeps for all MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS combinations. The IIP3 vs. MIXER_RDAC and MIXER_CDAC figures show both a surface and a contour plot in one figure. The contour plot is located directly underneath the surface plot. The best approach for reading the figure is to localize the peaks on the surface plot, which indicate maximum IIP3, and to follow the same color pattern to the contour plot to determine the optimized MIXER_RDAC and MIXER_CDAC values. The overall shape of the IIP3 plot does not vary with the MIXER_BIAS setting; therefore, only MIXER_BIAS = 011 is displayed. The data shows that MIXER_BIAS has the largest impact on performance. As previously mentioned and evident in the data, IIP3 improves with increased MIXER_BIAS, and noise figure is optimized at the lowest bias setting. Taking a more detailed look at the data, the different MIXER_RDAC and MIXER_CDAC combinations can result in a ~5 dB to +10 dB change in IIP3, but the noise figure changes by only ~0.5 dB. These trends become very important in deciding the trade-offs between IP3, noise figure, and power consumption. The total current consumption of the ADRF6620 does not change with MIXER_RDAC and MIXER_ CDAC and varies only with the mixer bias settings (see Figure 75). 19.5 40 19.0 18.5 NOISE FIGURE (dB) 30 25 20 0 18.0 17.5 17.0 16.5 16.0 10 15 5 0 10 11489-093 15.5 C A D _C ER IX M 5 15 MIXER_RDAC 15.0 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 Figure 76. IIP3 vs. MIXER_RDAC, MIXER_CDAC; MIXER_BIAS = 011 at RF Frequency = 900 MHz 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MIXER_RDAC MIXER_CDAC 11489-062 IIP3 (dBm) 35 MIXER BIAS 900-0 900-6 900-2 900-7 900-4 Figure 78. Noise Figure vs. MIXER_RDAC, MIXER_CDAC, and Various MIXER_BIAS Values at RF Frequency = 900 MHz 40 22.0 21.5 21.0 20.5 NOISE FIGURE (dB) 30 25 20 15 20.0 19.5 19.0 18.5 18.0 17.5 17.0 AC RD R_ XE 16.0 5 15.5 0 0 5 10 15 MIXER_CDAC 11489-094 MI 16.5 10 15.0 Figure 77. IIP3 vs. MIXER_RDAC, MIXER_CDAC; MIXER_BIAS = 011 at RF Frequency = 1900 MHz MIXER BIAS 1900-0 1900-6 1900-2 1900-7 1900-4 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MIXER_RDAC MIXER_CDAC Figure 79. Noise Figure vs. MIXER_RDAC, MIXER_CDAC, and Various MIXER_BIAS Values at RF Frequency = 1900 MHz Rev. 0 | Page 32 of 52 11489-063 IIP3 (dBm) 35 ADRF6620 45 NOISE FIGURE (dB) 35 30 25 20 15 10 C A RD R_ 5 5 10 MIXER_CDAC 15 0 MI XE 11489-060 15 0 0 Figure 80. IIP3 vs. MIXER_RDAC, MIXER_CDAC; MIXER_BIAS = 011 at RF Frequency = 2100 MHz NOISE FIGURE (dB) 35 30 25 20 AC RD _ ER IX M 10 5 10 MIXER_CDAC 15 0 11489-061 IIP3 (dBm) 40 15 0 2 3 4 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 5 6 7 8 9 10 11 12 13 14 15 MIXER_RDAC MIXER_CDAC Figure 82. Noise Figure vs. MIXER_RDAC, MIXER_CDAC, and Various MIXER_BIAS Values at RF Frequency = 2100 MHz 45 20 1 26.5 26.0 25.5 25.0 24.5 24.0 23.5 23.0 22.5 22.0 21.5 21.0 20.5 20.0 19.5 19.0 18.5 18.0 17.5 17.0 MIXER BIAS 16.5 2600-0 2600-6 16.0 2600-2 2600-7 15.5 2600-4 15.0 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 0 1 2 3 4 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 5 6 7 8 9 10 11 12 13 14 15 MIXER_RDAC MIXER_CDAC Figure 83. Noise Figure vs. MIXER_RDAC, MIXER_CDAC, and Various MIXER_BIAS Values at RF Frequency = 2700 MHz Figure 81. IIP3 vs. MIXER_RDAC, MIXER_CDAC; MIXER_BIAS = 011 at RF Frequency = 2700 MHz Rev. 0 | Page 33 of 52 11489-065 IIP3 (dBm) 40 23.5 23.0 22.5 22.0 21.5 21.0 20.5 20.0 19.5 19.0 18.5 18.0 17.5 17.0 16.5 MIXER BIAS 16.0 2100-6 2100-0 2100-7 2100-2 15.5 2100-4 15.0 0 7 15 0 7 15 0 7 15 0 7 15 0 7 15 0 11489-064 Data Sheet ADRF6620 Data Sheet specific MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS settings are shown in Figure 84. 50 IIP3 45 40 35 30 25 20 15 10 NOISE FIGURE 5 0 0.6 1.1 IIP3: OPT IIP3 IIP3: OPT NOISE FIGURE IIP3: IIP3 AND NOISE FIGURE BALANCE NF: OPT IIP3 NF: OPT NOISE FIGURE NF: IIP3 AND NOISE FIGURE BALANCE 1.6 2.1 2.6 RF FREQUENCY (GHz) 11489-066 IIP3 (dBm)/NOISE FIGURE (dB) As an example, the MIXER_RDAC, MIXER_CDAC, and MIXER_ BIAS settings of the ADRF6620 were carefully selected, based on three individual goals that resulted in three sets of MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS values. The first goal was for optimized IIP3. To achieve the most optimal IIP3 performance, the MIXER_BIAS was set to a higher current setting, and MIXER_ RDAC and MIXER_CDAC were selected at the peaks. This configuration allowed for the most optimal IIP3 performance. However, it also consumed the most power, and the noise figure was degraded. The second goal was to achieve a balance among IIP3, the noise figure, and power consumption. Finally, the third goal was for an optimized noise figure. This configuration resulted in the lowest power consumption while IIP3 was not optimized. Table 15 summarizes the test conditions; Table 16 shows the corresponding MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS values. The resulting IIP3 and noise figure performance for the Figure 84. Example IIP3 and Noise Figure Optimization Table 15. Mixer Optimization Summary Parameter Optimized IIP3 Noise Figure, IIP3, and Power Consumption Balance Optimized Noise Figure Test Conditions/Comments MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS were configured for optimized IIP3 performance. MIXER_BIAS was limited to 0, 1, or 2 decimal for improved noise figure while allowing IIP3 to degrade. MIXER_RDAC and MIXER_CDAC were chosen for optimized IIP3 because MIXER_RDAC and MIXER_CDAC have a larger impact on IIP3 than on noise figure. MIXER_BIAS was set to 0 decimal for the best noise figure. MIXER_RDAC and MIXER_CDAC were chosen for optimized IIP3 because they have a larger impact on IIP3 than on noise figure. Table 16. Recommended BAL_CIN, BAL_COUT, MIXER_RDAC, MIXER_CDAC, and MIXER_BIAS Settings (in Decimal) RF Frequency (MHz) 600 700 800 900 940 1000 1100 1200 1300 1400 1500 1600 1700 1800 1840 1900 2000 2100 2140 2200 2300 2400 2500 2600 2700 2800 2900 3000 BAL_CIN 7 7 5 3 3 2 1 1 0 0 0 0 0 0 0 0 0 1 1 2 2 1 1 1 1 1 1 0 BAL_COUT 7 7 5 4 3 3 2 2 2 2 2 2 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 Optimized IIP3 RDAC CDAC BIAS 6 10 4 5 14 4 3 13 3 0 15 0 5 12 4 5 11 4 5 10 4 5 9 4 8 8 4 6 7 4 6 7 4 8 7 4 6 6 4 9 6 4 9 6 5 9 6 5 7 5 5 9 5 5 9 5 4 7 4 4 7 4 4 7 4 4 7 4 4 7 4 4 7 4 4 7 4 4 7 4 4 7 4 4 IIP3 and Noise Figure Balance RDAC CDAC BIAS 4 15 2 4 15 2 3 14 2 3 13 2 5 11 2 4 10 2 3 10 1 3 9 1 3 9 1 4 8 1 5 7 2 5 7 2 5 6 2 5 6 2 5 6 2 6 5 2 3 6 0 5 5 1 5 5 1 5 5 1 5 5 1 5 5 1 5 5 1 5 5 1 5 5 1 4 15 2 4 15 2 3 14 2 Rev. 0 | Page 34 of 52 Optimized Noise Figure RDAC CDAC BIAS 4 15 0 4 15 0 2 15 0 2 14 0 2 13 0 3 11 0 2 11 0 2 10 0 2 10 0 2 9 0 3 8 0 2 8 0 4 7 0 4 7 0 3 7 0 3 7 0 3 6 0 3 6 0 3 6 0 3 6 0 3 6 0 3 6 0 3 6 0 3 6 0 3 6 0 4 15 0 4 15 0 2 15 0 Data Sheet ADRF6620 INTERSTAGE FILTERING REQUIREMENTS MXOUT+ L1 RF 19 MXOUT- 15 C2 C1 L4 LO 16 L3 MXOUT+ IFIN+ 11489-048 10 IFOUT2+ 8 270 7 250 PARALLEL RESISTANCE 6 230 5 4 210 3 Figure 85. Low-Pass IF Filter When designing the low-pass filter, it is important to consider the output impedance of the mixer and the input impedance of the IF DGA. The output impedance of the mixer has both a real and reactive component, and its equivalent model is shown in Figure 86. Correspondingly, Figure 87 shows the impedance vs. frequency for the mixer output. 190 PARALLEL CAPACITANCE 2 170 1 0 100 200 300 400 500 600 700 FREQUENCY (MHz) 800 900 150 1000 11489-050 0 PARALLEL RESISTANCE () 290 9 Figure 87. Mixer Output Impedance vs. Frequency Likewise, Figure 88 shows the impedance vs. frequency for the IF DGA. The four-port S parameter files for the IF DGA and mixer are available on analog.com and can serve as a useful tool to accurately capture the input and output impedance when designing the interstage filter. As a first-order approximation at low frequencies, the mixer output has a fixed impedance of approximately 255 , and the input impedance of the IF DAG is approximately 150 . Therefore, design the low-pass filter to have an input impedance of 255 and an output impedance of 150 . IFIN- 11 IFOUT2- 11489-049 10 16 500 OUTPUT CAPACITANCE INPUT CAPACITANCE 450 OUTPUT RESISTANCE INPUT RESISTANCE 400 14 350 12 300 18 IFOUT1+ 1.1pF MXOUT- 20 0.1F + Figure 86. Equivalent Model of the Mixer Output Impedance 0.1F 9 IFOUT1- 8 90 82.5 PARALLEL CAPACITANCE (pF) 18 L2 + 10 Rev. 0 | Page 35 of 52 250 PARALLEL RESISTANCE 8 200 6 150 4 PARALLEL RESISTANCE () +5V 2.5pF 100 PARALLEL CAPACITANCE 50 2 0 0 100 200 300 400 500 600 700 800 900 0 1000 FREQUENCY (MHz) Figure 88. IF DGA Input/Output Impedance vs. Frequency 11489-051 The low-pass filter resides between the mixer outputs and the IF DGA inputs, as shown in Figure 85. The signal flow starts with the differential outputs of the mixer being dc biased to positive supply (5 V) via a pair of pull-up inductors, L1 and L2. The inductor value is determined by the low frequency cutoff of the signal band of interest. Next, the third-order low-pass filter attenuates the high frequency sum term. The combination of the pull-up inductors and the low-pass filter results in a bandpass filter profile. The outputs of the filter are then ac-coupled through series capacitors and routed to the on-chip IF DGA via the IFIN+ and IFIN- pins. 82.5 PARALLEL CAPACITANCE (pF) Filtering at the mixer output may be necessary for improved linearity performance. For applications where the frequency plan requires low RF frequency inputs and IF outputs, the resulting sum term at the mixer outputs, fRF + fLO, may fall within the band of interest. The unwanted sum term may cause the IF DGA to operate in its nonlinear region because of the unnecessary presence of additional signal power. As a result, the linearity performance degrades where OIP3 and OIP2 decrease substantially. For this reason, a low-pass filter is necessary to attenuate the unwanted signal while maintaining the integrity of the wanted signal within the band of interest. In addition, the low-pass filter serves to suppress the LO feedthrough. Because of the absence of blockers in a typical DPD receive application, a lower order filter, such as a third-order Chebyshev, is typically adequate. ADRF6620 Data Sheet Most important, the low-pass interstage filter must attenuate the sum term (fRF + fLO) and LO feedthrough to prevent unnecessary overdrive of the DGA. The level of attenuation that is required to achieve optimal OIP3 performance is shown in Figure 89, where OIP3 vs. (fRF + fLO) amplitude is plotted. To maintain performance, attenuate the amplitude of the sum term to at least -16 dBm (see Figure 89). Beyond this point, the OIP3 degrades decibel per decibel for increased amplitudes. Table 17. Example Filter Design Parameter RS RL Pass-Band Edge Attenuation at Pass-Band Edge Stop-Band Edge Attenuation at Stop-Band Edge Filter Type 46 Using filter equations from a textbook or filter design software, a third-order Chebyshev filter can be designed to satisfy all the specifications in Table 17, as shown in Figure 91. The mixer output capacitance of 1.1 pF can be absorbed into the filter, resulting in a reduction in C1 from 2 pF to 0.8 pF. In addition, depending on the PCB board stack-up, C2 can be further reduced, or eliminated, because the capacitance of the PCB board can be used as the third pole of the filter. The components used in the simulation were the Coilcraft 0805CS inductors and Murata GRM15 series capacitors. Figure 90 shows the filter profile that satisfies all the filter specifications in Table 17. 44 42 38 36 34 -18 -16 -14 -12 -10 -8 -6 AMPLITUDE (dBm) -4 11489-052 32 0 Figure 89. OIP3 vs. (fRF + fLO) Amplitude -5 The ADRF6620 is optimized for use in digital predistortion (DPD) receivers. An example filter design for DPD is shown in Figure 91. Table 17 lists the interstage filter design targets. In most DPD systems for cellular transmission, the pass band is between 50 MHz and 500 MHz. For this reason, the pull-up inductors have a low frequency cutoff of 50 MHz, and the passband edge of the interstage low-pass filter is 500 MHz. This results in a band-pass filter profile with a maximally flat pass band from 50 MHz to 500 MHz. The stop-band attenuation at 1400 MHz is 20 dB, which typically provides the necessary attenuation of the mixer sum term with some margin. AMPLITUDE (dBm) -10 -15 -20 -25 -30 -35 -40 0 200 400 600 800 1000 1200 1400 1600 1800 2000 FREQUENCY (MHz) 11489-054 OIP3 (dBm) 40 30 -20 Value 255 150 500 MHz 0.5 dB 1400 MHz 20 dB Third-order Chebyshev Figure 90. Third-Order Chebyshev Filter Profile +5V L1 470nH RL L2 470nH MXOUT+ + 2.5pF 90 + L3 24nH 82.5 + C1 + 0.8pF 1.1pF 82.5 MXOUT- C2 1pF 150 + L4 24nH 0.1F + MIXER OUTPUT IMPEDANCE EQUIVALENT MODEL THIRD-ORDER CHEBYSHEV FILTER Figure 91. Low-Pass Interstage Filter Design Rev. 0 | Page 36 of 52 IFIN+ 0.1F DC BLOCKING CAPS IFIN- IDEAL IF AMP INPUT IMPEDANCE 11489-053 RS Data Sheet ADRF6620 12 11 L3 = L4 = 47nH, C1 = C2 = OPEN L3 = L4 = 39nH, C1 = C2 = OPEN Because the capacitance of the ADRF6620 evaluation board closely approximates the C1 and C2 capacitors, they can be removed from the design. However, this may not be the case for every PCB design with different stack-ups. Figure 93 compares the OIP3 and OIP2 performance of the ADRF6620 with and without filtering at the mixer output. 85 OIP2 (dBm)/OIP3 (dBm) Maintaining the same third-order Chebyshev filter design shown in Figure 91, the component values can be tuned to optimize performance with some trade-offs. To achieve maximally flat passband response, the trade-off is signal bandwidth (see Figure 92). The L3 and L4 inductors are replaced with 47 nH, and the capacitors are not populated. This configuration results in the flattest pass-band ripple; however, the signal bandwidth starts to roll off at 300 MHz. A narrower bandwidth translates to more attenuation of the mixer sum and LO leakage, which is a desirable effect if the wider signal bandwidth is not a requirement. Use the results shown in Figure 92 only as a guide, and design the interstage filter to the specific PCB board conditions. The plots in Figure 92 were measured using the ADRF6620 evaluation board. 75 OIP2 WITH FILTER 65 OIP2 WITH NO FILTER 55 OIP3 WITH FILTER 45 35 OIP3 WITH NO FILTER 10 100 150 200 250 300 350 IF FREQUENCY (MHz) 450 500 Figure 93. OIP2/OIP3 Performance With and Without Filtering at the DGA Output; RF Frequency = 900 MHz; High-Side LO Injection, LO Sweep 7 6 5 4 50 400 11489-056 15 50 L3 = L4 = 24nH, C1 = 0.8pF, C2 = 1pF 8 100 150 200 250 300 350 400 450 IF FREQUENCY (MHz) 500 11489-055 GAIN (dB) 25 9 Figure 92. Interstage Filter Design Trade-Offs Rev. 0 | Page 37 of 52 ADRF6620 Data Sheet IF DGA VS. LOAD 20 By design, the IF DGA is optimized for performance in a matched condition where the source and load resistances are both 150 . If the load or the source resistance is not equal to 150 (see the Digitally Programmable Variable Gain Amplifier (DGA) section), use the following equations to determine the resulting gain and input/output resistances: 18 RL = 73 10 8 6 S21 (Gain) = 2 x RIN/(RIN + RS) x AV 2 ROUT = (1000 + RS)/(1 + 0.044 x RS) 0 In a configuration where the mixer outputs of the ADRF6620 are routed to the IF DGA inputs, the matched condition is no longer satisfied because the source impedance, as seen by the IF DGA, is the 255 output impedance of the mixer outputs. As a result, the gain and output resistance of the amplifier vary from the expected 15 dB (see Figure 94). 0 100 200 300 400 500 600 700 800 900 1000 FREQUENCY (MHz) Figure 95. IF DGA Gain vs. Frequency for Different Loads -20 -30 -40 IF DGA IMD3 (dBc) 11489-067 RL Figure 94. Mixer Loading of the IF DGA The ideal load is 150 for the matched condition; however, this may not be the most readily available load impedance. As a result, load vs. performance trade-offs must be considered. In the matched condition, the IF DGA is optimized for linearity; therefore, the third-order intermodulation product degrades with load. Table 18 shows some common output loads, and Figure 95, Figure 96, and Figure 97 show the effects of loading on gain, IMD2, and IMD3. -50 RL = 50 -60 -70 RL = 73 -80 -90 RL = 500 -110 40 120 200 RL = 150 280 360 440 520 FREQUENCY (MHz) 600 680 11489-069 -100 Figure 96. IF DGA IMD3 vs. Frequency for Different Loads -20 150 LOAD 50 LOAD 500 LOAD 73 LOAD -30 As the equations in this section indicate, the manner in which the IF DGA is loaded affects the input resistance, RIN, of the amplifier. RIN, in turn, determines the load resistance of the interstage filter between the mixer outputs and the IF DGA inputs. The interstage filter has a source impedance of 255 from the mixer outputs and a load impedance of RIN for the particular RL load (see Table 18). As a result of the impedance mismatch, the insertion loss of the interstage filter must be included in the level planning calculations. IF DGA IMD2 (dBc) -40 -50 -60 -70 -80 -90 FREQUENCY (MHz) 11489-070 680 640 600 560 520 480 440 400 360 320 280 240 200 160 80 -110 120 -100 40 ROUT RL = 50 11489-068 4 RIN RL = 150 12 RIN = (1000 + RL)/(1 + 0.044 x RL) 255 RL = 500 14 IF DGA GAIN (dB) Voltage Gain = AV = 0.044 x (1000||RL) 16 Figure 97. IF DGA IMD2 vs. Frequency for Different Loads Table 18. Common Output Loads RS () 255 255 255 255 RIN () 65 151 255 328 AV (Linear) 14.7 5.7 3 2.1 AV (dB) 23.3 15.2 9.5 6.4 S21 (Linear) 6 4.3 3 2.4 Rev. 0 | Page 38 of 52 S21 (dB) 15.5 12.6 9.5 7.5 ROUT () 102.7 102.7 102.7 102.7 RL () 500 150 73 50 Data Sheet ADRF6620 The parallel combination of the 176 with the 1 k of the ADC input impedance results in an equivalent 150 differential output load as seen by the IF DGA of the ADRF6620. In addition, the input capacitance of the AD9434 can be used as the fourth pole of the antialiasing filter. The final schematic design is shown in Figure 99. The antialiasing filter is maximally flat, with a passband bandwidth of 500 MHz. Table 19 shows the component values for the antialiasing filter design for DPD. Figure 98 shows the simulated antialiasing filter design. ADC INTERFACING The integrated IF DGA of the ADRF6620 provides variable and sufficient drive capability for both buffered and unbuffered ADCs. It also provides isolation between the sampling edges of the ADC and the mixer core. As result, only an antialiasing filter is required when interfacing with an ADC. The ADRF6620 is optimized for use in cellular base station digital predistortion (DPD) systems. Predistortion is used to improve the linearity of transmitter power amplifiers (PA). Because the input signal to the DPD path is the known transmitted signal, the hardware specifications are not typically as stringent as the main receive path. The signal-to-noise ratio (SNR) of the ADC is not paramount, due to the autocorrelation with the known transmitted signal. For this reason, lower resolution ADCs are usually adequate, and 11-bit to 14-bit resolution typically suffices. A more critical consideration is the analog bandwidth of the converter. Traditional DPD systems require 3x to 5x the transmit bandwidth. Therefore, for a 100 MHz Tx bandwidth, the DPD bandwidth must be at least 500 MHz for fifth-order correction. Table 19. Component Values for 500 MHz Antialiasing Filter Design Parameter L1 = L2 C1 L3 = L4 C2 L5 = L6 L7 = L8 C3 L9 = L10 The AD9434 complements the ADRF6620 very well in a DPD design. The AD9434 is a 12-bit, 370 MSPS/500 MSPS buffered ADC. Its full power analog bandwidth is 1 GHz, making it wide enough for fifth-order correction with substantial margin. The sampling rate of the AD9434 is insufficient in satisfying the sampling theorem; however, this may be acceptable in DPD applications where undersampling is often permissible. Because the receive signal in the DPD path is the known transmitted signal, the desired signal and its aliases are clearly distinguished. -10 AMPLITUDE (dB) -15 -20 -25 -30 -35 -40 -50 0 200 400 600 800 1000 1200 1400 1600 1800 2000 FREQUENCY (MHz) 11489-100 -45 Figure 98. Simulated Antialiasing Filter Design +5V L6 0.1F 0.1F + L3 L7 L9 C2 + C1 88 + + 0.1F + L4 0.1F ADRF6620 IF AMP Figure 99. ADRF6620 Interface to the AD9434 Rev. 0 | Page 39 of 52 L8 AD9434 88 C3 L10 1k 1.3pF 11489-071 L5 + L2 + + 255 Manufacturer Coilcraft Murata Coilcraft Murata Coilcraft Coilcraft Murata Coilcraft 0 +5V L1 Type 0805CS GRM15 0805CS GRM15 0805LS 0805CS GRM15 0805CS -5 The antialiasing filter resides between the ADRF6620 and the AD9434. Because aliasing is common practice in a DPD receive chain, the antialiasing filter requirements can be relaxed. A second-order or third-order filter is sufficient in reducing the high frequency noise from folding back into the band of interest. When designing the antialiasing filter, it is important to consider the output impedance of the IF DGA of the ADRF6620 and the input impedance of the AD9434. The differential resistance of the AD9434 is 1 k, and the parallel capacitance is 1.3 pF. For the matched load condition, where the IF DGA is optimized for gain and linearity, load the IF DGA with 150 . To do this, place a 176 resistor in parallel with the input of the ADC. ADRF6620 MIXER OUTPUT Value 470 nH DNP 39 nH DNP 1 H 15 nH 2.7 pF 27 nH ADRF6620 Data Sheet POWER MODES The ADRF6620 has many building blocks, and these blocks can be independently powered off by writing to Register 0x01 (see Table 23). External LO Mode In external LO mode, the internal PLL and VCO are disabled, which reduces the current consumption by approximately 100 mA. Table 20 lists the register settings that are required to configure external LO mode. traces as far away from each other as possible (and at an angle, if possible) to prevent cross coupling. The input impedance of the RF inputs is 50 , and the traces leading to the pin must also have a 50 characteristic impedance. Terminate unused RF inputs with a dc blocking capacitor to ground. Table 20. Serial Port Configuration for External LO Mode State On On Off Off On Off Off On On On On External LO RFIN 0 Register 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x01 = 0x8B53 0x22, Bits[2:0] = 011 GND GND RFIN 1 GND GND RFIN 2 GND GND RFIN 3 GND 11489-072 Bit Name LDO_3P3_EN VCO_LDO_EN CP_EN DIV_EN VCO_EN REF_BUF_EN LO_DRV_EN LO_PATH_EN MIX_EN IF_AMP_EN LO_LDO_EN VCO_SEL GND Figure 100. Recommended Layout for the RF Inputs IF DGA Disable Mode In applications where the IF DGA is not used, it can be powered down. Power-down is achieved by disabling the IF_AMP_EN bit (Register 0x01, Bit 11 = 0). By disabling the amplifier, the current consumption of the ADRF6620 decreases by approximately 25 mA, along with a 35 mA to 50 mA current savings through each bias inductor at the output of the amplifier. When the IF DGA is disabled, its input and output impedance is high-Z. For this reason, the input and output pins can be left open. If the preference is not to leave the nodes open, the alternative option is to terminate the pins to ground via a 1 k resistor. The IF DGA outputs on the ADRF6620 have two output pins for each polarity, and they are oriented in an alternating fashion, as follows: IFOUT1+ (Pin 8), IFOUT1- (Pin 9), IFOUT2+ (Pin 10), and IFOUT2- (Pin 11). When designing the board, minimize the parasitic capacitance due to the routing that connects the corresponding outputs together. A good practice is to avoid any ground or power plane under this routing region and under the chokes to minimize the parasitic capacitance. Figure 101 shows the recommended layout. The IF DGA output pins with the same polarity are tied together on the bottom of the board with the blue traces and vias. LAYOUT Rev. 0 | Page 40 of 52 IFOUT1+ IFOUT1- IFOUT2+ IFOUT2- 11489-073 Careful layout of the ADRF6620 is necessary for optimizing performance and minimizing stray parasitics. Because the ADRF6620 supports four RF inputs, the layout of the RF section is critical in achieving isolation between each channel. Figure 100 shows the recommended layout for the RF inputs. Each RF input, RFIN0 to RFIN3, is isolated between ground pins, and the best layout approach is to keep the traces short and direct. To achieve this layout, connect the pins directly to the center ground pad of the exposed pad of the ADRF6620. This approach minimizes the trace inductance and promotes better isolation between the channels. In additional, for improved isolation, do not route the RFIN0 to RFIN3 traces in parallel to each other; instead, spread the traces immediately after each one leaves the pins. Keep the Figure 101. Recommended Layout for the IF DGA Outputs (Green traces are routings on top of the board, and blue traces are routings on the bottom of the board.) Data Sheet ADRF6620 REGISTER MAP Table 21. Register Map Summary Table Reg Name 0x00 SOFT_RESET 0x01 Enables 0x02 INT_DIV 0x03 FRAC_DIV 0x04 MOD_DIV 0x20 CP_CTL 0x21 PFD_CTL 0x22 FLO_CTL 0x23 DGA_CTL 0x30 BALUN_CTL 0x31 MIXER_CTL 0x40 PFD_CTL2 0x42 DITH_CTL1 0x43 DITH_CTL2 Bits [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] [15:8] [7:0] Bit 15 Bit 7 Bit 14 Bit 6 Bit 13 Bit 5 Bit 12 Bit 4 Bit 11 Bit 10 Bit 9 Bit 8 Bit 3 Bit 2 Bit 1 Bit 0 RESERVED RESERVED SOFT_RESET LO_LDO_EN RESERVED RESERVED RESERVED IF_AMP_EN RESERVED MIX_EN LO_PATH_EN LO_DRV_EN RESERVED REF_BUF_EN VCO_EN DIV_EN CP_EN VCO_LDO_EN LDO_3P3_EN RESERVED DIV_MODE INT_DIV[10:8] INT_DIV[7:0] RESERVED FRAC_DIV[10:8] FRAC_DIV[7:0] RESERVED MOD_DIV[10:8] MOD_DIV[7:0] RESERVED RESERVED CSCALE RESERVED RESERVED BLEED_DIR BLEED RESERVED RESERVED REF_MUX_SEL PFD_POLARITY REFSEL RESERVED LO_DRV_LVL[1] LO_DRV_LVL[0] RESERVED LO_DIV_A VCO_SEL RESERVED RFSW_MUX RFSW_SEL RFDSA_SEL[3] RFDSA_SEL[2:0] IF_ATTN RESERVED BAL_COUT RESERVED BAL_CIN RESERVED RESERVED MIXER_BIAS MIXER_RDAC[3] MIXER_RDAC[2:0] RESERVED MIXER_CDAC RESERVED RESERVED ABLDLY CPCTRL CLKEDGE RESERVED RESERVED DITH_EN DITH_MAG DITH_VAL DITH_VAL[15:8] DITH_VAL[7:0] Rev. 0 | Page 41 of 52 Reset RW 0x00000 W 0x8B7F RW 0x0058 RW 0x0250 RW 0x0600 RW 0x0C26 RW 0x0003 RW 0x000A RW 0x0000 RW 0x00000 RW 0x08EF RW 0x0010 RW 0x000E RW 0x0001 RW ADRF6620 Data Sheet REGISTER ADDRESS DESCRIPTIONS REGISTER 0x00, RESET: 0x00000, NAME: SOFT_RESET Table 22. Bit Descriptions for SOFT_RESET Bit 0 Bit Name SOFT_RESET Settings Description Soft reset Reset 0x0000 Access W REGISTER 0x01, RESET: 0x8B7F, NAME: ENABLES Table 23. Bit Descriptions for Enables Bits 15 11 9 8 7 5 4 3 2 1 0 Bit Name LO_LDO_EN IF_AMP_EN MIX_EN LO_PATH_EN LO_DRV_EN REF_BUF_EN VCO_EN DIV_EN CP_EN VCO_LDO_EN LDO_3P3_EN Settings Description Power up LO LDO IF DGA enable Mixer enable External LO path enable LO driver enable Reference buffer enable Power up VCOs Power up dividers Power up charge pump Power up VCO LDO Power up 3.3 V LDO Rev. 0 | Page 42 of 52 Reset 0x1 0x1 0x1 0x1 0x0 0x1 0x1 0x1 0x1 0x1 0x1 Access RW RW RW RW RW RW RW RW RW RW RW Data Sheet ADRF6620 REGISTER 0x02, RESET: 0x0058, NAME: INT_DIV Table 24. Bit Descriptions for INT_DIV Bits 11 Bit Name DIV_MODE Settings Description Reset 0x0 Access RW 0x58 RW Description Set divider FRAC value Reset 0x250 Access RW Description Set divider MOD value Reset 0x600 Access RW 0 1 [10:0] INT_DIV Fractional Integer Set divider INT value REGISTER 0x03, RESET: 0x0250, NAME: FRAC_DIV Table 25. Bit Descriptions for FRAC_DIV Bits [10:0] Bit Name FRAC_DIV Settings REGISTER 0x04, RESET: 0x0600, NAME: MOD_DIV Table 26. Bit Descriptions for MOD_DIV Bits [10:0] Bit Name MOD_DIV Settings Rev. 0 | Page 43 of 52 ADRF6620 Data Sheet REGISTER 0x20, RESET: 0x0C26, NAME: CP_CTL Table 27. Bit Descriptions for CP_CTL Bits [13:10] Bit Name CSCALE Settings 0001 0011 0111 1111 5 BLEED_DIR 0 1 [4:0] BLEED 00000 00001 ... ... ... 11110 11111 Description Charge pump current 250 A 500 A 750 A 1000 A Charge pump bleed direction Sink Source Charge pump bleed 0 A 15.625 A N x 15.625 A 468.75 A 484.375 A Rev. 0 | Page 44 of 52 Reset 0x3 Access RW 0x1 RW 0x06 RW Data Sheet ADRF6620 REGISTER 0x21, RESET: 0x0003, NAME: PFD_CTL Table 28. Bit Descriptions for PFD_CTL Bits [6:4] Bit Name REF_MUX_SEL Settings 000 001 010 011 100 101 110 111 3 PFD_POLARITY 0 1 [2:0] REFSEL 000 001 010 011 100 Description Set REF output divide ratio/VPTAT/LOCK_DET LOCK_DET VPTAT REFCLK REFCLK/2 REFCLK x 2 RESERVED REFCLK/4 RESERVED Set PFD polarity Positive KV VCO Negative KV VCO Set REF input divide ratio x2 x1 DIV2 DIV4 DIV8 Rev. 0 | Page 45 of 52 Reset 0x0 Access RW 0x0 RW 0x3 RW ADRF6620 Data Sheet REGISTER 0x22, RESET: 0x000A, NAME: FLO_CTL Table 29. Bit Descriptions for FLO_CTL Bits [8:7] Bit Name LO_DRV_LVL Settings 00 01 10 11 [4:3] LO_DIV_A 00 01 10 11 [2:0] VCO_SEL 000 001 010 011 100 101 110 111 Description LO amplitude -4 dBm 0.5 dBm +3 dBm +4.5 dBm LO_DIV_A DIV1 DIV2 DIV4 DIV8 Select VCO core/external LO 5.2 GHz to 5.7 GHz 4.1 GHz to 5.2 GHz 2.8 GHz to 4.1 GHz EXT LO VCO_PWRDWN VCO_PWRDWN VCO_PWRDWN VCO_PWRDWN Rev. 0 | Page 46 of 52 Reset 0x0 Access RW 0x1 RW 0x2 RW Data Sheet ADRF6620 REGISTER 0x23, RESET: 0x0000, NAME: DGA_CTL Table 30. Bit Descriptions for DGA_CTL Bits 11 Bit Name RFSW_MUX Settings 0 1 [10:9] RFSW_SEL 00 01 10 11 [8:5] RFDSA_SEL 0000 0001 ... 1110 1111 [4:0] IF_ATTN 00000 00001 ... 10111 11000 Description Set switch control. Serial CNTRL Pin CNTRL Set RF input. RFIN0 RFIN1 RFIN2 RFIN3 Set RFDSA attenuation. Range: 0 dB to 15 dB in steps of 1 dB. 0 dB 1 dB 14 dB 15 dB IF Attenuation. Range: 3 dB to 15 dB in steps of 0.5 dB. 3 dB 3.5 dB 14.5 dB 15 dB Rev. 0 | Page 47 of 52 Reset 0x0 Access RW 0x0 RW 0x0 RW 0x0 RW ADRF6620 Data Sheet REGISTER 0x30, RESET: 0x00000, NAME: BALUN_CTL Table 31. Bit Descriptions for BALUN_CTL Bits [7:5] Bit Name BAL_COUT Settings 000 ... 111 [3:1] BAL_CIN 000 ... 111 Description Set balun output capacitance Minimum capacitance ... Maximum capacitance Set balun input capacitance Minimum capacitance ... Maximum capacitance Reset 0x0 Access RW 0x0 RW Reset 0x4 Access RW 0x7 0xF RW RW REGISTER 0x31, RESET: 0x08EF, NAME: MIXER_CTL Table 32. Bit Descriptions for MIXER_CTL Bits [11:9] Bit Name MIXER_BIAS Settings 000 ... 111 [8:5] [3:0] MIXER_RDAC MIXER_CDAC Description Set mixer bias value Minimum Maximum Set mixer RDAC value Set mixer CDAC value Rev. 0 | Page 48 of 52 Data Sheet ADRF6620 REGISTER 0x40, RESET: 0x0010, NAME: PFD_CTL2 Table 33. Bit Descriptions for PFD_CTL2 Bits [6:5] Bit Name ABLDLY Settings 00 01 10 11 [4:2] CPCTRL 000 001 010 011 100 [1:0] CLKEDGE 00 01 10 11 Description Set antibacklash delay 0 ns 0.5 ns 0.75 ns 0.9 ns Set charge pump control. Both on Pump down Pump up Tristate PFD Set PFD edge sensitivity Div and REF down edge Div down edge, REF up edge Div up edge, REF down edge Div and REF up edge Rev. 0 | Page 49 of 52 Reset 0x0 Access RW 0x4 RW 0x0 RW ADRF6620 Data Sheet REGISTER 0x42, RESET: 0x000E, NAME: DITH_CTL1 Table 34. Bit Descriptions for DITH_CTL1 Bits 3 Bit Name DITH_EN Settings 0 1 [2:1] 0 DITH_MAG DITH_VAL Description Set dither enable Disable Enable Set dither magnitude Set dither value Reset 0x1 Access RW 0x3 0x0 RW RW Description Set dither value Reset 0x1 Access RW REGISTER 0x43, RESET: 0x0001, NAME: DITH_CTL2 Table 35. Bit Descriptions for DITH_CTL2 Bits [15:0] Bit Name DITH_VAL Settings Rev. 0 | Page 50 of 52 Data Sheet ADRF6620 OUTLINE DIMENSIONS 0.30 0.23 0.18 PIN 1 INDICATOR 37 36 48 1 0.50 BSC TOP VIEW 0.80 0.75 0.70 0.45 0.40 0.35 EXPOSED PAD 24 SEATING PLANE 5.65 5.50 SQ 5.35 13 BOTTOM VIEW 0.05 MAX 0.02 NOM COPLANARITY 0.08 0.20 REF PIN 1 INDICATOR 0.20 MIN 5.50 REF FOR PROPER CONNECTION OF THE EXPOSED PAD, REFER TO THE PIN CONFIGURATION AND FUNCTION DESCRIPTIONS SECTION OF THIS DATA SHEET. COMPLIANT TO JEDEC STANDARDS MO-220-WKKD. 06-06-2012-B 7.10 7.00 SQ 6.90 Figure 102. 48-Lead Lead Frame Chip Scale Package [LFCSP_WQ] 7 mm x 7 mm Body, Very Very Thin Quad (CP-48-9) Dimensions shown in millimeters ORDERING GUIDE Model 1 ADRF6620ACPZ-R7 ADRF6620-EVALZ 1 Temperature Range -40C to +85C Package Description 48-Lead Lead Frame Chip Scale Package [LFCSP_WQ] Evaluation Board Z = RoHS Compliant Part. Rev. 0 | Page 51 of 52 Package Option CP-48-9 ADRF6620 Data Sheet NOTES (c)2013 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D11489-0-7/13(0) Rev. 0 | Page 52 of 52