400 MHz to 6 GHz Quadrature Demodulator ADL5380 FEATURES FUNCTIONAL BLOCK DIAGRAM ENBL ADJ ADL5380 IHI BIAS ILO LOIP RFIN V2I QUADRATURE PHASE SPLITTER RFIP LOIN QHI QLO 07585-001 Operating RF and LO frequency: 400 MHz to 6 GHz Input IP3 30 dBm @ 900 MHz 28 dBm @1900 MHz Input IP2: >65 dBm @ 900 MHz Input P1dB (IP1dB): 11.6 dBm @ 900 MHz Noise figure (NF) 10.9 dB @ 900 MHz 11.7 dB @ 1900 MHz Voltage conversion gain: ~7 dB Quadrature demodulation accuracy @ 900 MHz Phase accuracy: ~0.2 Amplitude balance: ~0.07 dB Demodulation bandwidth: ~390 MHz Baseband I/Q drive: 2 V p-p into 200 Single 5 V supply Figure 1. APPLICATIONS Cellular W-CDMA/GSM/LTE Microwave point-to-(multi)point radios Broadband wireless and WiMAX GENERAL DESCRIPTION The ADL5380 is a broadband quadrature I-Q demodulator that covers an RF/IF input frequency range from 400 MHz to 6 GHz. With a NF = 10.9 dB, IP1dB = 11.6 dBm, and IIP3 = 29.7 dBm @ 900 MHz, the ADL5380 demodulator offers outstanding dynamic range suitable for the demanding infrastructure direct-conversion requirements. The differential RF inputs provide a well-behaved broadband input impedance of 50 and are best driven from a 1:1 balun for optimum performance. Excellent demodulation accuracy is achieved with amplitude and phase balances of ~0.07 dB and ~0.2, respectively. The demodulated in-phase (I) and quadrature (Q) differential outputs are fully buffered and provide a voltage conversion gain of ~7 dB. The buffered baseband outputs are capable of driving a 2 V p-p differential signal into 200 . The fully balanced design minimizes effects from second-order distortion. The leakage from the LO port to the RF port is <-50 dBm. Differential dc offsets at the I and Q outputs are typically <20 mV. Both of these factors contribute to the excellent IIP2 specification, which is >65 dBm. The ADL5380 operates off a single 4.75 V to 5.25 V supply. The supply current is adjustable by placing an external resistor from the ADJ pin to either the positive supply, VS, (to increase supply current and improve IIP3) or to ground (which decreases supply current at the expense of IIP3). The ADL5380 is fabricated using the Analog Devices, Inc., advanced silicon-germanium bipolar process and is available in a 24-lead exposed paddle LFCSP. Rev. 0 Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties that may result from its use. Specifications subject to change without notice. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Trademarks and registered trademarks are the property of their respective owners. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781.329.4700 www.analog.com Fax: 781.461.3113 (c)2009 Analog Devices, Inc. All rights reserved. ADL5380 TABLE OF CONTENTS Features .............................................................................................. 1 V-to-I Converter ......................................................................... 22 Applications ....................................................................................... 1 Mixers .......................................................................................... 22 Functional Block Diagram .............................................................. 1 Emitter Follower Buffers ........................................................... 22 General Description ......................................................................... 1 Bias Circuit .................................................................................. 22 Revision History ............................................................................... 2 Applications Information .............................................................. 23 Specifications..................................................................................... 3 Basic Connections ...................................................................... 23 Absolute Maximum Ratings............................................................ 5 Power Supply............................................................................... 23 ESD Caution .................................................................................. 5 Local Oscillator (LO) Input ...................................................... 23 Pin Configuration and Function Descriptions ............................. 6 RF Input ....................................................................................... 24 Typical Performance Characteristics ............................................. 7 Baseband Outputs ...................................................................... 24 Low Band Operation .................................................................... 7 Error Vector Magnitude (EVM) Performance ........................... 24 Midband Operation ................................................................... 11 Low IF Image Rejection............................................................. 25 High Band Operation ................................................................ 14 Example Baseband Interface ..................................................... 26 Distributions for fLO = 900 MHz ............................................... 17 Characterization Setups ................................................................. 30 Distributions for fLO = 1900 MHz............................................. 18 Evaluation Board ............................................................................ 32 Distributions for fLO = 2700 MHz............................................. 19 Thermal Grounding and Evaluation Board Layout............... 34 Distributions for fLO = 3600 MHz............................................. 20 Outline Dimensions ....................................................................... 35 Distributions for fLO = 5800 MHz............................................. 21 Ordering Guide .......................................................................... 35 Circuit Description ......................................................................... 22 LO Interface................................................................................. 22 REVISION HISTORY 7/09--Revision 0: Initial Version Rev. 0 | Page 2 of 36 ADL5380 SPECIFICATIONS VS = 5 V, TA = 25C, fLO = 900 MHz, fIF = 4.5 MHz, PLO = 0 dBm, ZO = 50 , unless otherwise noted. Baseband outputs differentially loaded with 450 . Loss of the balun used to drive the RF port was de-embedded from these measurements. Table 1. Parameter OPERATING CONDITIONS LO and RF Frequency Range LO INPUT Input Return Loss LO Input Level I/Q BASEBAND OUTPUTS Voltage Conversion Gain Demodulation Bandwidth Quadrature Phase Error I/Q Amplitude Imbalance Output DC Offset (Differential) Output Common Mode 0.1 dB Gain Flatness Output Swing Peak Output Current POWER SUPPLIES Voltage Current ENABLE FUNCTION Off Isolation Turn-On Settling Time Turn-Off Settling Time ENBL High Level (Logic 1) ENBL Low Level (Logic 0) DYNAMIC PERFORMANCE at RF = 900 MHz Conversion Gain Input P1dB RF Input Return Loss Second-Order Input Intercept (IIP2) Third-Order Input Intercept (IIP3) LO to RF RF to LO IQ Magnitude Imbalance IQ Phase Imbalance Noise Figure Noise Figure Under Blocking Conditions Condition Min Typ 0.4 LOIP, LOIN LO driven differentially through a balun at 900 MHz -6 QHI, QLO, IHI, ILO 450 differential load on I and Q outputs at 900 MHz 200 differential load on I and Q outputs at 900 MHz 1 V p-p signal, 3 dB bandwidth At 900 MHz 0 dBm LO input at 900 MHz Dependent on ADJ pin setting VADJ ~ 4 V (set by 1.5 k from ADJ pin to VS) VADJ ~ 4.8 V (set by 200 from ADJ pin to VS) VADJ ~ 2.4 V (ADJ pin open) Differential 200 load Each pin VS = VCC1, VCC2, VCC3 -10 0 Unit 6 GHz +6 dB dBm 6.9 5.9 390 0.2 0.07 10 dB dB MHz Degrees dB mV VS - 2.5 VS - 2.8 VS - 1.2 37 2 12 V V V MHz V p-p mA 4.75 1.5 k from ADJ pin to VS; ENBL pin low 1.5 k from ADJ pin to VS; ENBL pin high Pin ENBL Max 5.25 245 145 -70 45 950 ENBL high to low ENBL low to high 2.5 1.7 V mA mA dB ns ns V V VADJ ~ 4 V (set by 1.5 k from ADJ pin to VS) RFIP, RFIN driven differentially through a balun -5 dBm each input tone -5 dBm each input tone RFIN, RFIP terminated in 50 LOIN, LOIP terminated in 50 With a -5 dBm input interferer 5 MHz away Rev. 0 | Page 3 of 36 6.9 11.6 -19 68 29.7 -52 -67 0.07 0.2 10.9 13.1 dB dBm dB dBm dBm dBm dBc dB Degrees dB dB ADL5380 Parameter DYNAMIC PERFORMANCE at RF = 1900 MHz Conversion Gain Input P1dB RF Input Return Loss Second-Order Input Intercept (IIP2) Third-Order Input Intercept (IIP3) LO to RF RF to LO IQ Magnitude Imbalance IQ Phase Imbalance Noise Figure Noise Figure Under Blocking Conditions DYNAMIC PERFORMANCE at RF = 2700 MHz Conversion Gain Input P1dB RF Input Return Loss Second-Order Input Intercept (IIP2) Third-Order Input Intercept (IIP3) LO to RF RF to LO IQ Magnitude Imbalance IQ Phase Imbalance Noise Figure DYNAMIC PERFORMANCE at RF = 3600 MHz Conversion Gain Input P1dB RF Input Return Loss Second-Order Input Intercept (IIP2) Third-Order Input Intercept (IIP3) LO to RF RF to LO IQ Magnitude Imbalance IQ Phase Imbalance Noise Figure Noise Figure Under Blocking Conditions DYNAMIC PERFORMANCE at RF = 5800 MHz Conversion Gain Input P1dB RF Input Return Loss Second-Order Input Intercept (IIP2) Third-Order Input Intercept (IIP3) LO to RF RF to LO IQ Magnitude Imbalance IQ Phase Imbalance Noise Figure Noise Figure Under Blocking Conditions Condition VADJ ~ 4 V (set by 1.5 k from ADJ pin to VS) RFIP, RFIN driven differentially through a balun -5 dBm each input tone -5 dBm each input tone RFIN, RFIP terminated in 50 LOIN, LOIP terminated in 50 With a -5 dBm input interferer 5 MHz away VADJ ~ 4 V (set by 1.5 k from ADJ pin to VS) RFIP, RFIN driven differentially through a balun -5 dBm each input tone -5 dBm each input tone RFIN, RFIP terminated in 50 LOIN, LOIP terminated in 50 Min Typ Max Unit 6.8 11.6 -13 61 27.8 -49 -77 0.07 0.25 11.7 14 dB dBm dB dBm dBm dBm dBc dB Degrees dB dB 7.4 11 -10 54 28 -49 -73 0.07 0.5 12.3 dB dBm dB dBm dBm dBm dBc dB Degrees dB 6.3 9.6 -11 48 21 -46 -72 0.14 1.1 14.2 16.2 dB dBm dB dBm dBm dBm dBc dB Degrees dB dB 5.8 8.2 -7.5 44 20.6 -47 -62 0.07 -1.25 15.5 18.9 dB dBm dB dBm dBm dBm dBc dB Degrees dB dB VADJ ~ 4.8 V (set by200 from ADJ pin to VS) RFIP, RFIN driven differentially through a balun -5 dBm each input tone -5 dBm each input tone RFIN, RFIP terminated in 50 LOIN, LOIP terminated in 50 With a -5 dBm input interferer 5 MHz away VADJ ~ 2.4 V (ADJ pin left open) RFIP, RFIN driven differentially through a balun -5 dBm each input tone -5 dBm each input tone RFIN, RFIP terminated in 50 LOIN, LOIP terminated in 50 With a -5 dBm input interferer 5 MHz away Rev. 0 | Page 4 of 36 ADL5380 ABSOLUTE MAXIMUM RATINGS Table 2. Parameter Supply Voltage: VCC1, VCC2, VCC3 LO Input Power RF Input Power Internal Maximum Power Dissipation JA 1 Maximum Junction Temperature Operating Temperature Range Storage Temperature Range 1 Rating 5.5 V 13 dBm (re: 50 ) 15 dBm (re: 50 ) 1370 mW 53C/W 150C -40C to +85C -65C to +125C 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. ESD CAUTION Per JDEC standard JESD 51-2. For information on optimizing thermal impedance, see the Thermal Grounding and Evaluation Board Layout section. Rev. 0 | Page 5 of 36 ADL5380 24 23 22 21 20 19 VCC3 GND3 RFIP RFIN GND3 ADJ PIN CONFIGURATION AND FUNCTION DESCRIPTIONS PIN 1 INDICATOR ADL5380 TOP VIEW (Not to Scale) 18 17 16 15 14 13 GND3 GND2 QHI QLO GND2 VCC2 07585-002 1 2 3 4 5 6 ENBL 7 GND4 8 LOIP 9 LOIN 10 GND4 11 NC 12 GND3 GND1 IHI ILO GND1 VCC1 NOTES 1. NC = NO CONNECT. 2. THE EXPOSED PAD SHOULD BE CONNECTED TO A LOW IMPEDANCE THERMAL AND ELECTRICAL GROUND PLANE. Figure 2. Pin Configuration Table 3. Pin Function Descriptions Pin No. 1, 2, 5, 8, 11, 14, 17, 18, 20, 23 3, 4, 15, 16 Mnemonic GND1, GND2, GND3, GND4 Description Ground Connect. IHI, ILO, QLO, QHI 6, 13, 24 VCC1, VCC2, VCC3 7 ENBL 9, 10 LOIP, LOIN 12 19 NC ADJ 21, 22 RFIN, RFIP I Channel and Q Channel Mixer Baseband Outputs. These outputs have a 50 differential output impedance (25 per pin). Each output pair can swing 2 V p-p (differential) into a load of 200 . The output 3 dB bandwidth is ~400 MHz. Supply. Positive supply for LO, IF, biasing, and baseband sections. Decouple these pins to the board ground using the appropriate-sized capacitors. Enable Control. When pulled low, the part is fully enabled; when pulled high, the part is partially powered down and the output is disabled. Local Oscillator Input. Pins must be ac-coupled. A differential drive through a balun is necessary to achieve optimal performance. Recommended balun is the Mini-Circuits TC1-1-13 for lower frequencies, the Johanson Technology 3600 balun for midband frequencies, and the Johanson Technology 5400 balun for high band frequencies. Balun choice depends on the desired frequency range of operation. Do not connect this pin. A resistor to VS that optimizes third-order intercept. For operation <3 GHz, RADJ = 1.5 k. For operation from 3 GHz to 4 GHz, RADJ = 200 . For operation >5 GHz, RADJ = open. See the Circuit Description section for more details. RF Input. A single-ended 50 signal can be applied differentially to the RF inputs through a 1:1 balun. Recommended balun is the Mini-Circuits TC1-1-13 for lower frequencies, the Johanson Technology 3600 balun for midband frequencies, and the Johanson Technology 5400 balun for high band frequencies. Balun choice depends on the desired frequency range of operation. Exposed Paddle. Connect to a low impedance thermal and electrical ground plane. EP Rev. 0 | Page 6 of 36 ADL5380 TYPICAL PERFORMANCE CHARACTERISTICS VS = 5 V, TA = 25C, LO drive level = 0 dBm, RF input balun loss is de-embedded, unless otherwise noted. LOW BAND OPERATION RF = 400 MHz to 3 GHz; Mini-Circuits TC1-1-13 balun on LO and RF inputs, 1.5 k from the ADJ pin to VS. 1.0 18 TA = -40C TA = +25C TA = +85C 14 0.8 0.6 INPUT P1dB GAIN MISMATCH (dB) GAIN (dB), IP1dB (dBm) 16 12 10 GAIN 8 0.4 0.2 0 -0.2 -0.4 6 -0.6 3000 07585-005 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 LO FREQUENCY (MHz) LO FREQUENCY (MHz) Figure 3. Conversion Gain and Input 1 dB Compression Point (IP1dB) vs. LO Frequency 80 800 400 3000 -1.0 07585-003 2600 2800 2200 2400 1800 2000 1600 1400 1200 800 1000 600 2 400 TA = -40C TA = +25C TA = +85C -0.8 600 4 Figure 5. IQ Gain Mismatch vs. LO Frequency 2 I CHANNEL Q CHANNEL 1 70 BASEBAND RESPONSE (dB) 0 INPUT IP2 50 INPUT IP3 (I AND Q CHANNELS) 30 -2 -3 -4 -5 -6 3000 -8 07585-004 2600 2800 2200 2400 1800 2000 1600 1400 -7 1200 800 1000 10 TA = -40C TA = +25C TA = +85C 600 20 -1 10 100 BASEBAND FREQUENCY (MHz) LO FREQUENCY (MHz) Figure 6. Normalized IQ Baseband Frequency Response Figure 4. Input Third-Order Intercept (IIP3) and Input Second-Order Intercept Point (IIP2) vs. LO Frequency Rev. 0 | Page 7 of 36 1000 07585-006 40 400 IIP3, IIP2 (dBm) 60 ADL5380 16 NOISE FIGURE (dB) 15 14 13 12 11 10 30 3000 260 240 20 SUPPLY CURRENT 15 200 NOISE FIGURE 5 0 1.0 2.0 2.5 3.0 3.5 4.0 160 4.5 Figure 10. IIP3, Noise Figure, and Supply Current vs. VADJ, fLO = 900 MHz 4 QUADRATURE PHASE ERROR (Degrees) 1.5 180 VADJ (V) Figure 7. Noise Figure vs. LO Frequency 25 23 3 21 NOISE FIGURE (dB) 2 1 0 -1 TA = -40C TA = +25C TA = +85C -2 17 15 13 1920MHz 11 920MHz -3 5 -30 07585-008 3000 2800 GAIN (dB), IP1dB (dBm), NOISE FIGURE (dB) 65 14 55 NOISE FIGURE 8 50 45 GAIN 6 40 4 35 IIP3 2 -5 -4 -3 -2 30 -1 0 1 2 LO LEVEL (dBm) 3 4 5 6 25 07585-009 10 IIP3, IIP2 ( dBm) 60 IP1dB 60 18 70 IIP2, Q CHANNEL 12 5 IIP2, Q CHANNEL 16 55 IIP2, I CHANNEL 14 50 NOISE FIGURE 45 12 IP1dB 10 40 GAIN 8 35 IIP3, IIP2 (dBm) 75 IIP2, I CHANNEL 16 0 Figure 11. Noise Figure vs. Input Blocker Level, fLO = 900 MHz, fLO = 1900 MHz (RF Blocker 5 MHz Offset) 20 18 -20 -15 -10 -5 RF BLOCKER INPUT POWER (dBm) 30 6 IIP3 25 4 2 -6 20 -5 -4 -3 -2 -1 0 1 2 LO LEVEL (dBm) 3 4 5 6 Figure 12. Conversion Gain, IP1dB, Noise Figure, IIP3, and IIP2 vs. LO Level, fLO = 2700 MHz Figure 9. Conversion Gain, IP1dB, Noise Figure, IIP3, and IIP2 vs. LO Level, fLO = 900 MHz Rev. 0 | Page 8 of 36 07585-012 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 LO FREQUENCY (MHz) -25 07585-011 7 Figure 8. IQ Quadrature Phase Error vs. LO Frequency 0 -6 19 9 -4 GAIN (dB), IP1dB (dBm), NOISE FIGURE (dB) 220 10 07585-007 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 LO FREQUENCY (MHz) 280 INPUT IP3 25 9 8 300 TA = -40C TA = +25C TA = +85C SUPPLY CURRENT (mA) 17 IIP3 (dBm) AND NOISE FIGURE (dB) TA = -40C TA = +25C TA = +85C 07585-010 35 18 ADL5380 35 0 TA = -40C TA = +25C TA = +85C IIP3 (dBm) AND NOISE FIGURE (dB) 30 -5 RETURN LOSS (dB) 25 INPUT IP3 20 15 10 NOISE FIGURE -10 -15 -20 1.0 1.5 2.0 2.5 3.0 VADJ (V) 3.5 4.0 -25 07585-013 0 4.5 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 RF FREQUENCY (GHz) Figure 16. RF Port Return Loss vs. RF Frequency Measured on Characterization Board Through TC1-1-13 Balun 80 -20 70 -30 60 -40 LEAKAGE (dBm) 50 900MHz: GAIN 900MHz: IP1dB 900MHz: IIP2, I CHANNEL 900MHz: IIP2, Q CHANNEL 2700MHz: GAIN 2700MHz: IP1dB 2700MHz: IIP2, I CHANNEL 2700MHz: IIP2, Q CHANNEL 40 30 20 -60 -70 -80 10 1 2 3 4 -100 VADJ (V) 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 LO FREQUENCY (GHz) Figure 14. Conversion Gain, IP1dB, and IIP2 vs. VADJ, fLO = 900 MHz, fLO = 2700 MHz IP1dB, IIP3 (dBm) 30 90 -20 85 -30 80 IIP3 25 75 IIP2 20 70 15 65 10 60 IP1dB 5 0 4.5 6.5 8.5 10.5 12.5 14.5 16.5 BASEBAND FREQUENCY (MHz) -40 LEAKAGE (dBc) 35 I CHANNEL Q CHANNEL 55 18.5 50 -50 -60 -70 -80 -90 07585-015 TA = -40C TA = +25C TA = +85C Figure 17. LO-to-RF Leakage vs. LO Frequency IIP2, I AND Q CHANNELS (dBm) 40 07585-017 -90 07585-014 0 -50 Figure 15. IP1dB, IIP3, and IIP2 vs. Baseband Frequency -100 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 RF FREQUENCY (GHz) Figure 18. RF-to-LO Leakage vs. RF Frequency Rev. 0 | Page 9 of 36 07585-018 GAIN (dB), IP1dB (dBm), IIP2 I AND Q CHANNELS (dBm) Figure 13. IIP3 and Noise Figure vs. VADJ, fLO = 2700 MHz 07585-016 5 ADL5380 0 -2 -6 -8 -10 -12 -14 -16 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 LO FREQUENCY (GHz) 07585-019 RETURN LOSS (dB) -4 Figure 19. LO Port Return Loss vs. LO Frequency Measured on Characterization Board Through TC1-1-13 Balun Rev. 0 | Page 10 of 36 ADL5380 MIDBAND OPERATION RF = 3 GHz to 4 GHz; Johanson Technology 3600BL14M050T balun on LO and RF inputs, 200 from VADJ to VS. 12 IP1dB 11 10 9 8 GAIN 7 6 5 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 LO FREQUENCY (GHz) 3.8 3.9 4.0 18 Figure 20. Conversion Gain and Input 1 dB Compression Point (IP1dB) vs. LO Frequency 16 14 40 IP1dB 10 35 8 30 GAIN 6 25 4 20 IIP3 2 15 -5 -4 -3 -2 -1 0 1 LO LEVEL (dBm) 2 3 4 5 6 10 Figure 23. Conversion Gain, IP1dB, Noise Figure, IIP3, and IIP2 vs. LO Level, fLO = 3600 MHz 17 I CHANNEL Q CHANNEL 16 NOISE FIGURE (dB) INPUT IP2 60 50 40 INPUT IP3 I AND Q CHANNELS 15 14 13 12 TA = -40C TA = +25C TA = +85C 11 10 20 3.1 3.2 3.3 3.4 3.5 3.6 3.7 LO FREQUENCY (GHz) 3.8 3.9 4.0 8 07585-021 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 LO FREQUENCY (GHz) 3.8 3.9 4.0 3.9 4.0 Figure 24. Noise Figure vs. LO Frequency Figure 21. Input Third-Order Intercept (IIP3) and Input Second-Order Intercept Point (IIP2) vs. LO Frequency 4 1.0 0.6 QUADRATURE PHASE ERROR (Degrees) TA = -40C TA = +25C TA = +85C 0.8 0.4 0.2 0 -0.2 -0.4 -0.6 -1.0 3.0 3.2 3.4 3.6 LO FREQUENCY (GHz) 3.8 4.0 07585-022 -0.8 TA = -40C TA = +25C TA = +85C 3 2 1 0 -1 -2 -3 -4 3.0 3.1 3.2 3.4 3.6 3.3 3.5 3.7 LO FREQUENCY (GHz) 3.8 Figure 25. IQ Quadrature Phase Error vs. LO Frequency Figure 22. IQ Gain Mismatch vs. LO Frequency Rev. 0 | Page 11 of 36 07585-025 3.0 07585-024 9 10 GAIN MISMATCH (dB) IIP3, IIP2 (dBm) 45 NOISE FIGURE 18 TA = -40C TA = +25C TA = +85C 30 50 12 80 70 55 IIP2, Q CHANNEL IIP2, I CHANNEL 0 -6 07585-020 4 60 IIP3, IIP2 (dBm) TA = -40C TA = +25C TA = +85C 13 GAIN (dB), IP1dB (dBm) GAIN (dB), IP1dB (dBm), NOISE FIGURE (dB) 20 07585-023 14 ADL5380 INPUT IP3 20 15 -20 280 -30 260 -40 240 10 NOISE FIGURE 5 220 LEAKAGE (dBm) TA = -40C TA = +25C TA = +85C 25 300 CURRENT (mA) IIP3 (dBm) AND NOISE FIGURE (dB) 30 -50 -60 200 -70 180 -80 1.5 2.0 2.5 3.0 3.5 4.0 4.5 07585-026 0 1.0 VADJ (V) 3.1 3.3 3.4 3.5 3.6 3.7 LO FREQUENCY (GHz) 3.8 3.9 4.0 3.9 4.0 3.9 4.0 Figure 29. LO-to-RF Leakage vs. LO Frequency Figure 26. IIP3, Noise Figure, and Supply Current vs. VADJ, fLO = 3600 MHz 25 -20 23 -30 -40 21 LEAKAGE (dBc) 19 17 15 -50 -60 -70 -80 13 -90 -25 -20 -15 -10 -5 0 5 RF POWEL LEVEL (dBm) -100 07585-027 11 -30 3.1 3.2 3.3 3.4 3.5 3.6 3.7 RF FREQUENCY (GHz) 3.8 07585-030 NOISE FIGURE (dB) 3.2 07585-029 SUPPLY CURRENT Figure 30. RF-to-LO Leakage vs. RF Frequency Figure 27. Noise Figure vs. Input Blocker Level, fLO = 3600 MHz (RF Blocker 5 MHz Offset) 0 80 70 RETURN LOSS (dB) GAIN (dB), IP1dB (dBm), IIP2 I AND Q CHANNELS (dBm) -2 60 50 40 3600MHz: GAIN 3600MHz: IP1dB 3600MHz: IIP2, I CHANNEL 3600MHz: IIP2, Q CHANNEL 30 20 10 -4 -6 -8 -10 1 2 3 4 V ADJ (V) -12 3.1 3.2 3.3 3.4 3.5 3.6 3.7 RF FREQUENCY (GHz) 3.8 07585-031 -10 07585-028 0 Figure 31. RF Port Return Loss vs. RF Frequency Measured on Characterization Board Through Johanson Technology 3600 Balun Figure 28. Conversion Gain, IP1dB, and IIP2 vs. VADJ, fLO = 3600 MHz Rev. 0 | Page 12 of 36 ADL5380 0 RETURN LOSS (dB) -5 -10 -15 -20 -30 3.1 3.2 3.3 3.4 3.5 3.6 3.7 LO FREQUENCY (GHz) 3.8 3.9 4.0 07585-032 -25 Figure 32. LO Port Return Loss vs. LO Frequency Measured on Characterization Board Through Johanson Technology 3600 Balun Rev. 0 | Page 13 of 36 ADL5380 HIGH BAND OPERATION RF = 5 GHz to 6 GHz; Johanson Technology 5400BL15B050E balun on LO and RF inputs, the ADJ pin is open. 8 7 GAIN 6 5 TA = -40C TA = +25C TA = +85C 4 3 5.1 5.2 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 6.0 Figure 33. Conversion Gain and Input 1 dB Compression Point (IP1dB) vs. LO Frequency 16 50 45 14 IIP2, I CHANNEL 12 10 40 35 IP1dB 30 8 GAIN 6 4 25 20 IIP3 2 15 0 -6 -5 -4 -3 -2 -1 0 1 LO LEVEL (dBm) 2 3 4 5 6 10 Figure 36. Conversion Gain, IP1dB, Noise Figure, IIP3, and IIP2 vs. LO Level, fLO = 5800 MHz 20 80 TA = -40C TA = +25C TA = +85C 70 18 17 NOISE FIGURE (dB) INPUT IP2 50 40 INPUT IP3 (I AND Q CHANNELS) 30 TA = -40C TA = -25C TA = +85C 19 I CHANNEL Q CHANNEL 60 IIP3, IIP2 (dBm) 55 NOISE FIGURE IIP2, Q CHANNEL IIP3, IIP2 (dBm) INPUT P1dB 9 18 07585-036 GAIN (dB), IP1dB (dBm), NOISE FIGURE (dB) 10 07585-033 GAIN (dB), INPUT P1dB (dBm) 11 2 60 20 12 16 15 14 13 12 11 10 20 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 6.0 8 07585-034 5.2 5.0 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 6.0 4 1.0 TA = -40C TA = +25C TA = +85C 0.6 3 IQ PHASE MISMATCH (Degrees) 0.8 0.4 0.2 0 -0.2 -0.4 -0.6 TA = -40C TA = +25C TA = +85C 2 1 0 -1 -2 5.1 5.2 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 6.0 Figure 35. IQ Gain Mismatch vs. LO Frequency -4 5.1 5.2 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 Figure 38. IQ Quadrature Phase Error vs. LO Frequency Rev. 0 | Page 14 of 36 6.0 07585-038 -3 -0.8 07585-035 IQ AMPLITUDE MISMATCH (dB) 5.2 Figure 37. Noise Figure vs. LO Frequency Figure 34. Input Third-Order Intercept (IIP3) and Input Second-Order Intercept Point (IIP2) vs. LO Frequency -1.0 5.1 07585-037 9 10 5.1 ADL5380 300 TA = -40C TA = +25C TA = +85C -40 260 NOISE FIGURE 15 240 10 220 200 2.0 2.5 3.0 3.5 -70 -90 180 1.5 -60 -80 SUPPLY CURRENT 5 0 1.0 -50 4.0 4.5 -10 0 5.1 VADJ (V) 5.2 5.3 5.4 5.5 5.6 5.7 LO FREQUENC Y (GHz) 5.8 5.9 6.0 5.9 6.0 07585-042 20 -30 280 LEAKAGE (dBm) INPUT IP3 CURRENT (mA) 25 -20 07585-039 Figure 42. LO-to-RF Leakage vs. LO Frequency Figure 39. IIP3, Noise Figure, and Supply Current vs. VADJ, fLO = 5800 MHz -20 25 -30 -40 LEAKAGE (dBc) NOISE FIGURE (dB) 20 15 10 -50 -60 -70 -80 5 -25 -20 -15 -10 RF POWER LEVEL (dBm) -5 -100 07585-040 0 -30 5.1 Figure 40. Noise Figure vs. Input Blocker Level, fLO = 5800 MHz (RF Blocker 5 MHz Offset) 5.3 5.4 5.5 5.6 5.7 RF FREQUENCY (MHz) 5.8 Figure 43. RF-to-LO Leakage vs. RF Frequency 60 0 -2 50 -4 RETURN LOSS (dB) 40 5800MHz: GAIN 5800MHz: IP1dB 5800MHz: IIP2, I CHANNEL 5800MHz: IIP2, Q CHANNEL 30 20 -6 -8 -10 -12 10 -14 0 1 2 3 4 VADJ (V) 07585-041 GAIN (dB), IP1dB (dBm), IIP2 I AND Q CHANNEL (dBm) 5.2 07585-043 -90 Figure 41. Conversion Gain, IP1dB, and IIP2 vs. RBIAS, fLO = 5800 MHz -16 5.1 5.2 5.3 5.4 5.5 5.6 5.7 RF FREQUENCY (GHz) 5.8 5.9 6.0 07585-044 IIP3 (dBm) AND NOISE FIGURE (dB) 30 Figure 44. RF Port Return Loss vs. RF Frequency Measured on Characterization Board Through Johanson Technology 5400 Balun Rev. 0 | Page 15 of 36 ADL5380 -0 -2 RETURN LOSS (dB) -4 -6 -8 -10 -12 -16 5.1 5.2 5.3 5.4 5.5 5.6 5.7 LO FREQUENCY (GHz) 5.8 5.9 6.0 07585-045 -14 Figure 45. LO Port Return Loss vs. LO Frequency Measured on Characterization Board Through Johanson Technology 5400 Balun Rev. 0 | Page 16 of 36 ADL5380 100 90 90 DISTRIBUTION PERCENTAGE (%) 100 80 70 60 50 40 30 TA = -40C TA = +25C TA = +85C 10 28 29 30 31 32 INPUT IP3 (dBm) 70 50 40 30 20 10 33 34 0 45 50 60 65 70 INPUT IP2 (dBm) 75 80 85 100 100 IP1dB GAIN 90 80 70 60 50 40 30 20 TA = -40C TA = +25C TA = +85C 0 4 5 6 7 8 9 10 11 GAIN (dB), IP1dB (dBm) 70 60 50 40 30 TA = -40C TA = +25C TA = +85C 20 10 12 13 14 0 9.5 07585-047 10 80 10.0 100 90 90 DISTRIBUTION PERCENTAGE (%) 100 80 70 60 50 40 30 TA = -40C TA = +25C TA = +85C 0 -0.3 -0.2 -0.1 0 0.1 GAIN MISMATCH (dB) 12.5 TA = -40C TA = +25C TA = +85C 80 70 60 50 40 30 20 10 0.2 0.3 07585-048 10 12.0 Figure 50. Noise Figure Distributions Figure 47. Gain and IP1dB Distributions 20 10.5 11.0 11.5 NOISE FIGURE (dB) 07585-050 DISTRIBUTION PERCENTAGE (%) 90 DISTRIBUTION PERCENTAGE (%) 55 Figure 49. IIP2 Distributions for I Channel and Q Channel Figure 46. IIP3 Distributions DISTRIBUTION PERCENTAGE (%) I CHANNEL Q CHANNEL 60 0 -1.0 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 QUADRATURE PHASE ERROR (Degrees) 0.8 Figure 51. IQ Quadrature Phase Error Distributions Figure 48. IQ Gain Mismatch Distributions Rev. 0 | Page 17 of 36 1.0 07585-051 0 TA = -40C TA = +25C TA = +85C 80 07585-049 20 07585-046 DISTRIBUTION PERCENTAGE (%) DISTRIBUTIONS FOR fLO = 900 MHz ADL5380 100 90 90 DISTRIBUTION PERCENTAGE (%) 100 80 70 60 50 40 30 TA = -40C TA = +25C TA = +85C 10 24 25 80 60 50 40 30 20 10 26 27 28 29 INPUT IP3 (dBm) 30 31 32 0 45 Figure 52. IIP3 Distributions 50 60 65 INPUT IP2 (dBm) 70 75 80 100 TA = -40C TA = +25C TA = +85C 90 IP1dB GAIN 70 60 50 40 30 20 10 80 70 60 50 40 30 20 TA = -40C TA = +25C TA = +85C 10 4 5 6 7 8 9 10 11 12 13 14 GAIN (dB), IP1dB (dBm) 0 10.5 07585-053 0 11.0 12.0 12.5 13.0 13.5 NOISE FIGURE (dB) Figure 53. Gain and IP1dB Distributions Figure 56. Noise Figure Distributions 100 100 90 TA = -40C TA = +25C TA = +85C 90 DISTRIBUTION PERCENTAGE (%) TA = -40C TA = +25C TA = +85C 80 70 60 50 40 30 20 80 70 60 50 40 30 20 10 -0.2 -0.1 0 0.1 GAIN MISMATCH (dB) 0.2 0.3 07585-054 10 0 -0.3 11.5 07585-056 80 DISTRIBUTION PERCENTAGE (%) 90 DISTRIBUTION PERCENTAGE (%) 55 Figure 55. IIP2 Distributions for I Channel and Q Channel 100 DISTRIBUTION PERCENTAGE (%) I CHANNEL Q CHANNEL 70 0 -1.0 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 QUADRATURE PHASE ERROR (Degrees) 0.8 Figure 57. IQ Quadrature Phase Error Distributions Figure 54. IQ Gain Mismatch Distributions Rev. 0 | Page 18 of 36 1.0 07585-057 0 TA = -40C TA = +25C TA = +85C 07585-055 20 07585-052 DISTRIBUTION PERCENTAGE (%) DISTRIBUTIONS FOR fLO = 1900 MHz ADL5380 100 100 90 90 DISTRIBUTION PERCENTAGE (%) 80 70 60 50 40 TA = -40C TA = +25C TA = +85C 30 20 10 20 22 24 26 28 30 INPUT IP3 (dBm) 32 34 36 60 50 40 30 20 0 35 40 50 55 60 INPUT IP2 (dBm) 65 70 75 100 90 IP1dB GAIN 80 TA = -40C TA = +25C TA = +85C 90 DISTRIBUTION PERCENTAGE (%) TA = -40C TA = +25C TA = +85C 70 60 50 40 30 20 10 80 70 60 50 40 30 20 5 6 7 8 9 10 11 GAIN (dB), IP1dB (dBm) 12 13 14 0 10.5 07585-059 4 11.0 90 90 DISTRIBUTION PERCENTAGE (%) 100 80 70 60 50 40 30 0 -0.3 TA = -40C TA = +25C TA = +85C -0.2 -0.1 0 0.1 GAIN MISMATCH (dB) 13.5 14.0 80 2.0 TA = -40C TA = +25C TA = +85C 70 60 50 40 30 20 10 0.2 0.3 07585-060 10 12.0 12.5 13.0 NOISE FIGURE (dB) Figure 62. Noise Figure Distributions 100 20 11.5 07585-062 10 Figure 59. Gain and IP1dB Distributions DISTRIBUTION PERCENTAGE (%) 45 Figure 61. IIP2 Distributions for I Channel and Q Channel 100 DISTRIBUTION PERCENTAGE (%) 70 07585-061 18 Figure 58. IIP3 Distributions 0 80 10 07585-058 0 I CHANNEL Q CHANNEL TA = -40C TA = +25C TA = +85C 07585-063 DISTRIBUTION PERCENTAGE (%) DISTRIBUTIONS FOR fLO = 2700 MHz Figure 60. IQ Gain Mismatch Distributions 0 -2.0 -1.5 -1.0 -0.5 0 0.5 1.0 QUADRATURE PHASE ERROR (Degrees) 1.5 Figure 63. IQ Quadrature Phase Error Distributions Rev. 0 | Page 19 of 36 ADL5380 100 100 90 90 DISTRIBUTION PERCENTAGE (%) 80 70 60 TA = -40C TA = +25C TA = +85C 50 40 30 20 80 70 60 50 40 30 20 15 17 19 21 23 25 27 INPUT IP3 (dBm) 29 31 33 0 07585-064 0 35 Figure 64. IIP3 Distributions 45 50 55 INPUT IP2 (dBm) 60 65 70 Figure 67. IIP2 Distributions for I Channel and Q Channel 100 100 IP1dB GAIN 80 TA = -40C TA = +25C TA = +85C 70 60 TA = -40C TA = +25C TA = +85C 90 DISTRIBUTION PERCENTAGE (%) 90 50 40 30 20 10 80 70 60 50 40 30 20 10 4 5 6 7 8 9 10 11 GAIN (dB), IP1dB (dBm) 12 13 14 0 12.5 07585-065 0 Figure 65. Gain and IP1dB Distributions 13.5 14.0 14.5 15.0 NOISE FIGURE (dB) 15.5 16.0 2.5 Figure 68. Noise Figure Distributions 100 100 90 DISTRIBUTION PERCENTAGE (%) 70 TA = -40C TA = +25C TA = +85C 90 TA = -40C TA = +25C TA = +85C 80 60 50 40 30 20 10 80 70 60 50 40 30 20 10 -0.2 -0.1 0 0.1 GAIN MISMATCH (dB) 0.2 0.3 07585-066 0 -0.3 13.0 07585-068 DISTRIBUTION PERCENTAGE (%) 40 07585-067 10 10 DISTRIBUTION PERCENTAGE (%) I CHANNEL Q CHANNEL TA = -40C TA = +25C TA = +85C 07585-069 DISTRIBUTION PERCENTAGE (%) DISTRIBUTIONS FOR fLO = 3600 MHz Figure 66. IQ Gain Mismatch Distributions 0 -0.5 0 0.5 1.0 1.5 2.0 QUADRATURE PHASE ERROR (Degrees) Figure 69. IQ Quadrature Phase Error Distributions Rev. 0 | Page 20 of 36 ADL5380 DISTRIBUTIONS FOR fLO = 5800 MHz 100 100 TA = -40C TA = +25C TA = +85C 80 90 DISTRIBUTION PERCENTAGE (%) 70 60 50 40 30 20 10 70 60 50 40 30 20 21 22 INPUT IP3 (dBm) 23 24 0 30 40 45 50 55 INPUT IP2 (dBm) 60 65 70 Figure 73. IIP2 Distributions for I Channel and Q Channel 100 100 TA = -40C TA = +25C TA = +85C 80 90 DISTRIBUTION PERCENTAGE (%) 90 IP1dB GAIN 70 60 50 40 30 20 10 TA = -40C TA = +25C TA = +85C 80 70 60 50 40 30 20 3 4 5 6 7 GAIN (dB), IP1dB (dBm) 8 9 10 0 13.0 07585-071 2 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 NOISE FIGURE (dB) Figure 74. Noise Figure Distributions 100 100 90 90 DISTRIBUTION PERCENTAGE (%) TA = -40C TA = +25C TA = +85C 80 70 60 50 40 30 20 80 TA = -40C TA = +25C TA = +85C 70 60 50 40 30 20 10 -0.2 -0.1 0 0.1 GAIN MISMATCH (dB) 0.2 0.3 07585-072 10 0 -0.3 13.5 07585-074 10 Figure 71. Gain and IP1dB Distributions DISTRIBUTION PERCENTAGE (%) 35 07585-073 20 07585-070 19 Figure 70. IIP3 Distributions DISTRIBUTION PERCENTAGE (%) 80 10 0 18 0 I CHANNEL Q CHANNEL TA = -40C TA = +25C TA = +85C 0 -3 -2 -1 0 1 2 QUADRATURE PHASE ERROR (Degrees) Figure 75. IQ Quadrature Phase Error Distributions Figure 72. IQ Gain Mismatch Distributions Rev. 0 | Page 21 of 36 3 07585-075 DISTRIBUTION PERCENTAGE (%) 90 ADL5380 CIRCUIT DESCRIPTION The ADL5380 can be divided into five sections: the local oscillator (LO) interface, the RF voltage-to-current (V-to-I) converter, the mixers, the differential emitter follower outputs, and the bias circuit. A detailed block diagram of the device is shown in Figure 76. ENBL ADJ ADL5380 IHI BIAS ILO LOIP RFIN QUADRATURE PHASE SPLITTER V2I RFIP RADJ 200 to VS 600 to VS 1.54 k to VS 3.8 k to VS 10 k to VS Open 9 k to GND 3.5 k to GND 1.5 k to GND ~VADJ (V) 4.8 4.5 4 3.5 3 2.5 2 1.5 1 ~ Baseband CommonMode Output (V) 2.2 2.3 2.5 2.7 3 3.2 3.4 3.6 3.8 MIXERS LOIN The ADL5380 has two double-balanced mixers: one for the inphase channel (I channel) and one for the quadrature channel (Q channel). These mixers are based on the Gilbert cell design of four cross-connected transistors. The output currents from the two mixers are summed together in the resistive loads that then feed into the subsequent emitter follower buffers. 07585-076 QHI QLO Table 4. ADJ Pin Resistor Values and Approximate ADJ Pin Voltages Figure 76. Block Diagram The LO interface generates two LO signals at 90 of phase difference to drive two mixers in quadrature. RF signals are converted into currents by the V-to-I converters that feed into the two mixers. The differential I and Q outputs of the mixers are buffered via emitter followers. Reference currents to each section are generated by the bias circuit. A detailed description of each section follows. LO INTERFACE The LO interface consists of a polyphase quadrature splitter followed by a limiting amplifier. The LO input impedance is set by the polyphase, which splits the LO signal into two differential signals in quadrature. The LO input impedance is nominally 50 . Each quadrature LO signal then passes through a limiting amplifier that provides the mixer with a limited drive signal. For optimal performance, the LO inputs must be driven differentially. V-TO-I CONVERTER EMITTER FOLLOWER BUFFERS The output emitter followers drive the differential I and Q signals off chip. The output impedance is set by on-chip 25 series resistors that yield a 50 differential output impedance for each baseband port. The fixed output impedance forms a voltage divider with the load impedance that reduces the effective gain. For example, a 500 differential load has 1 dB lower effective gain than a high (10 k) differential load impedance. BIAS CIRCUIT A band gap reference circuit generates the reference currents used by different sections. The bias circuit can be enabled and partially disabled using ENBL (Pin 7). If ENBL is grounded or left open, the part is fully enabled. Pulling ENBL high shuts off certain sections of the bias circuitry, reducing the standing power to about half of its fully enabled consumption and disabling the outputs. The differential RF input signal is applied to a V-to-I converter that converts the differential input voltage to output currents. The V-to-I converter provides a differential 50 input impedance. The V-to-I bias current can be adjusted up or down using the ADJ pin (Pin 19). Adjusting the current up improves IIP3 and IP1dB but degrades SSB NF. Adjusting the current down improves SSB NF but degrades IIP3 and IP1dB. The current adjustment can be made by connecting a resistor from the ADJ pin (Pin 19) to VS to increase the bias current or to ground to decrease the bias current. Table 4 approximately dictates the relationship between the resistor used (RADJ), the resulting ADJ pin voltage, and the resulting baseband common-mode output voltage. Rev. 0 | Page 22 of 36 ADL5380 APPLICATIONS INFORMATION BASIC CONNECTIONS LOCAL OSCILLATOR (LO) INPUT Figure 78 shows the basic connections schematic for the ADL5380. For optimum performance, drive the LO port differentially through a balun. The recommended balun for each performance level includes the following: POWER SUPPLY The nominal voltage supply for the ADL5380 is 5 V and is applied to the VCC1, VCC2, and VCC3 pins. Connect ground to the GND1, GND2, GND3, and GND4 pins. Solder the exposed paddle on the underside of the package to a low thermal and electrical impedance ground plane. If the ground plane spans multiple layers on the circuit board, these layers should be stitched together with nine vias under the exposed paddle. The AN-772 Application Note discusses the thermal and electrical grounding of the LFCSP in detail. Decouple each of the supply pins using two capacitors; recommended capacitor values are 100 pF and 0.1 F. * Up to 3 GHz is the Mini-Circuits TC1-1-13. * From 3 GHz to 4 GHz is the Johanson Technology 3600BL14M050. * From 4.9 GHz to 6 GHz is the Johanson Technology 5400BL15B050. AC couple the LO inputs to the device with 100 pF capacitors. The LO port is designed for a broadband 50 match from 400 MHz to 6 GHz. The LO return loss can be seen in Figure 19. Figure 77 shows the LO input configuration. LO INPUT LOIP 10 LOIN 07585-077 BALUN 9 100pF 100pF Figure 77. Differential LO Drive The recommended LO drive level is between -6 dBm and +6 dBm. The applied LO frequency range is between 400 MHz and 6 GHz. RFIN BALUN 100pF 100pF RADJ VS VS 22 21 20 RFIN GND3 GND3 GND3 18 2 GND1 GND2 17 3 IHI QHI 16 ADL5380 4 ILO 7 GND4 6 VCC1 LOIN VCC2 13 8 9 10 11 12 100pF QLO VS 100pF 0.1F 100pF BALUN LO_SE Figure 78. Basic Connections Schematic Rev. 0 | Page 23 of 36 07585-078 100pF LOIP 0.1F GND2 14 GND4 VS QHI QLO 15 5 GND1 ENBL ILO NC IHI 19 ADJ 23 RFIP 1 24 GND3 100pF VCC3 0.1F ADL5380 RF INPUT IHI AC couple the RF inputs to the device with 100 pF capacitors. Figure 79 shows the RF input configuration. RFIN 22 RFIP 100pF BALUN 100pF 07585-079 RF INPUT Figure 79. RF Input The differential RF port return loss is characterized, as shown in Figure 80. -8 15 QLO 07585-081 4 Figure 81. Baseband Output Configuration -10 EVM is a measure used to quantify the performance of a digital radio transmitter or receiver. A signal received by a receiver has all constellation points at their ideal locations; however, various imperfections in the implementation (such as magnitude imbalance, noise floor, and phase imbalance) cause the actual constellation points to deviate from their ideal locations. In general, a demodulator exhibits three distinct EVM limitations vs. received input signal power. At strong signal levels, the distortion components falling in-band due to nonlinearities in the device cause strong degradation to EVM as signal levels increase. At medium signal levels, where the demodulator behaves in a linear manner and the signal is well above any notable noise contributions, the EVM has a tendency to reach an optimum level determined dominantly by the quadrature accuracy of the demodulator and the precision of the test equipment. As signal levels decrease, such that noise is a major contribution, the EVM performance vs. the signal level exhibits a decibel-fordecibel degradation with decreasing signal level. At lower signal levels, where noise proves to be the dominant limitation, the decibel EVM proves to be directly proportional to the SNR. The ADL5380 shows excellent EVM performance for various modulation schemes. Figure 82 shows the EVM performance of the ADL5380 with a 16 QAM, 200 kHz low IF. -12 -14 -16 0 -18 -5 -20 -10 -22 -15 -26 -20 EVM (dB) -24 -28 -30 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 RF FREQUENCY (GHz) 4.5 5.0 5.5 6.0 07585-080 DIFFERENTIAL RETURN LOSS RF PORT (dB) ILO ERROR VECTOR MAGNITUDE (EVM) PERFORMANCE Up to 3 GHz is the Mini-Circuits TC1-1-13. From 3 GHz to 4 GHz is the Johanson Technology 3600BL14M050. From 4.9 GHz to 6 GHz is the Johanson Technology 5400BL15B050. 21 QHI Figure 80. Differential RF Port Return Loss -25 -30 -35 -40 BASEBAND OUTPUTS -45 The baseband outputs QHI, QLO, IHI, and ILO are fixed impedance ports. Each baseband pair has a 50 differential output impedance. The outputs can be presented with differential loads as low as 200 (with some degradation in gain) or high impedance differential loads (500 or greater impedance yields the same excellent linearity) that is typical of an ADC. The TCM9-1 9:1 balun converts the differential IF output to a single-ended output. When loaded with 50 , this balun presents a 450 load to the device. The typical maximum linear voltage swing for these outputs is 2 V p-p differential. The output 3 dB bandwidth is 390 MHz. Figure 81 shows the baseband output configuration. -50 -90 Rev. 0 | Page 24 of 36 -70 -50 -30 RF INPUT POWER (dBm) -10 Figure 82. EVM, RF = 900 MHz, IF = 200 kHz vs. RF Input Power for a 16 QAM 160ksym/s Signal 10 07585-082 * 16 ADL5380 The RF inputs have a differential input impedance of approximately 50 . For optimum performance, drive the RF port differentially through a balun. The recommended balun for each performance level includes the following: * * 3 ADL5380 Figure 84 exhibits multiple W-CDMA low-IF EVM performance curves over a wide RF input power range into the ADL5380. In the case of zero-IF, the noise contribution by the vector signal analyzer becomes predominant at lower power levels, making it difficult to measure SNR accurately. -10 -15 -20 EVM (dB) Figure 83 shows the zero-IF EVM performance of a 10 MHz IEEE 802.16e WiMAX signal through the ADL5380. The differential dc offsets on the ADL5380 are in the order of a few millivolts. However, ac coupling the baseband outputs with 10 F capacitors eliminates dc offsets and enhances EVM performance. With a 10 MHz BW signal, 10 F ac coupling capacitors with the 500 differential load results in a high-pass corner frequency of ~64 Hz, which absorbs an insignificant amount of modulated signal energy from the baseband signal. By using ac coupling capacitors at the baseband outputs, the dc offset effects, which can limit dynamic range at low input power levels, can be eliminated. 0 -25 -30 0Hz IF -35 -10 2.5MHz LOW-IF 5MHz LOW-IF -40 -45 -80 -30 -60 -50 -40 -30 -20 RF INPUT POWER (dBm) -10 0 10 Figure 84. EVM, RF = 1900 MHz, IF = 0 Hz, IF = 2.5 MHz, IF = 5 MHz, and IF = 7.5 MHz vs. RF Input Power for a W-CDMA Signal (AC-Coupled Baseband Outputs) 5.8GHz -40 7.5MHz LOW-IF -70 07585-084 EVM (dB) -20 3.5GHz LOW IF IMAGE REJECTION -50 -60 -75 -65 -55 -45 -35 -25 -15 -5 RF INPUT POWER (dBm) 5 07585-083 2.6GHz Figure 83. EVM, RF = 2.6 GHz, RF = 3.5 GHz, and RF = 5.8 GHz, IF = 0 Hz vs. RF Input Power for a 16 QAM 10 MHz Bandwidth Mobile WiMAX Signal (AC-Coupled Baseband Outputs) The image rejection ratio is the ratio of the intermediate frequency (IF) signal level produced by the desired input frequency to that produced by the image frequency. The image rejection ratio is expressed in decibels. Appropriate image rejection is critical because the image power can be much higher than that of the desired signal, thereby plaguing the down-conversion process. Figure 85 illustrates the image problem. If the upper sideband (lower sideband) is the desired band, a 90 shift to the Q channel (I channel) cancels the image at the lower sideband (upper sideband). Phase and gain balance between I and Q channels are critical for high levels of image rejection. COSLOt 0 IF IF -IF 0 +IF -90 0 +IF 0 +IF +90 LO USB 0 -IF 0 +IF 07585-085 LSB SINLOt Figure 85. Illustration of the Image Problem Rev. 0 | Page 25 of 36 ADL5380 Figure 86 and Figure 87 show the excellent image rejection capabilities of the ADL5380 for low IF applications, such as W-CDMA. The ADL5380 exhibits image rejection greater than 45 dB over a broad frequency range. 60 2.5MHz LOW IF 5MHz LOW IF 7MHz LOW IF 40 30 20 0 400 07585-103 10 800 1200 1600 2000 2400 2800 RF FREQUENCY (MHz) 3200 3600 4000 Figure 86. Low Band and Midband Image Rejection vs. RF Frequency for a W-CDMA Signal, IF = 2.5 MHz, 5 MHz, and 7.5 MHz As an example, a second-order Butterworth, low-pass filter design is shown in Figure 88 where the differential load impedance is 500 and the source impedance of the ADL5380 is 50 . The normalized series inductor value for the 10-to-1, load-to-source impedance ratio is 0.074 H, and the normalized shunt capacitor is 14.814 F. For a 10.9 MHz cutoff frequency, the single-ended equivalent circuit consists of a 0.54 H series inductor followed by a 433 pF shunt capacitor. 60 IMAGE REJECTION (dB) 50 40 30 The order and type of filter network depends on the desired high frequency rejection required, pass-band ripple, and group delay. Filter design tables provide outlines for various filter types and orders, illustrating the normalized inductor and capacitor values for a 1 Hz cutoff frequency and 1 load. After scaling the normalized prototype element values by the actual desired cut-off frequency and load impedance, the series reactance elements are halved to realize the final balanced filter network component values. 2.5MHz LOW IF 5MHz LOW IF 7MHz LOW IF The balanced configuration is realized as the 0.54 H inductor is split in half to realize the network shown in Figure 88. 20 10 07585-104 0 5000 RS = 50 5200 5400 5600 RF FREQUENCY (MHz) 5800 NORMALIZED SINGLE-ENDED CONFIGURATION VS 6000 CN 14.814F RS = 0.1 RL Figure 87. High Band Image Rejection vs. RF Frequency for a W-CDMA Signal, IF = 2.5 MHz, 5 MHz, and 7.5 MHz RS = 50 EXAMPLE BASEBAND INTERFACE In most direct-conversion receiver designs, it is desirable to select a wanted carrier within a specified band. The desired channel can be demodulated by tuning the LO to the appropriate carrier frequency. If the desired RF band contains multiple carriers of interest, the adjacent carriers are also down converted to a lower IF frequency. These adjacent carriers can be problematic if they are large relative to the wanted carrier because they can overdrive the baseband signal detection circuitry. As a result, it is often necessary to insert a filter to provide sufficient rejection of the adjacent carriers. LN = 0.074H fC = 1Hz 0.54H DENORMALIZED SINGLE-ENDED EQUIVALENT VS RS = 25 2 RS = 25 2 433pF RL= 500 fC = 10.9MHz 0.27H BALANCED CONFIGURATION VS RL= 500 433pF RL 2 = 250 RL = 250 2 0.27H Figure 88. Second-Order Butterworth, Low-Pass Filter Design Example Rev. 0 | Page 26 of 36 07585-087 IMAGE REJECTION (dB) 50 It is necessary to consider the overall source and load impedance presented by the ADL5380 and ADC input when designing the filter network. The differential baseband output impedance of the ADL5380 is 50 . The ADL5380 is designed to drive a high impedance ADC input. It may be desirable to terminate the ADC input down to lower impedance by using a terminating resistor, such as 500 . The terminating resistor helps to better define the input impedance at the ADC input at the cost of a slightly reduced gain (see the Circuit Description section for details on the emitter-follower output loading effects). ADL5380 Figure 89 and Figure 90 show the measured frequency response and group delay of the filter. 900 800 700 DELAY (ns) A complete design example is shown in Figure 91. A sixth-order Butterworth differential filter having a 1.9 MHz corner frequency interfaces the output of the ADL5380 to that of an ADC input. The 500 load resistor defines the input impedance of the ADC. The filter adheres to typical direct conversion W-CDMA applications where, 1.92 MHz away from the carrier IF frequency, 1 dB of rejection is desired, and, 2.7 MHz away from the carrier IF frequency, 10 dB of rejection is desired. 600 500 400 300 100 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Figure 90. Sixth-Order Baseband Filter Group Delay 0 -5 -10 -15 -20 0 FREQUENCY (MHz) 0 0.5 1.0 1.5 2.0 2.5 3.0 FREQUENCY (MHz) 3.5 07585-088 MAGNITUDE RESPONSE (dB) 5 Figure 89. Sixth-Order Baseband Filter Response Rev. 0 | Page 27 of 36 1.8 07585-089 200 10 ADL5380 RFIN BALUN 100pF 100pF VS VS 23 22 21 20 GND3 RFIP RFIN GND3 1 GND3 19 ADJ 24 VCC3 100pF 0.1F GND3 18 2 GND1 GND2 17 3 IHI QHI 16 ADL5380 4 ILO NC 7 GND4 VCC2 13 LOIN 6 VCC1 100pF LOIP GND2 14 GND4 0.1F 5 GND1 ENBL VS QLO 15 8 9 10 11 12 100pF VS 100pF 0.1F 100pF BALUN LO_SE 27H 27H 10H 270pF 100pF 68pF CAC 10F CAC 10F 27H 27H 27H 27H 10H 10H 500 CAC 10F 270pF 100pF 68pF 27H 27H 10H 500 ADC INPUT ADC INPUT Figure 91. Sixth-Order Low-Pass Butterworth, Baseband Filter Schematic Rev. 0 | Page 28 of 36 07585-090 CAC 10F ADL5380 1.5H 5 0 -5 -10 -15 -20 -25 -30 1.5H 0 5 10 15 20 25 30 35 40 FREQUENCY (MHz) Figure 93. Fourth-Order Low-Pass LTE Filter Magnitude Response Figure 92. Fourth-Order Low-Pass LTE Filter Schematic 60 GROUP DELAY (ns) 50 40 30 20 10 0 0 5 10 15 20 25 FREQUENCY (MHz) 30 35 40 07585-093 2.2H 07585-091 -40 07585-092 -35 200 22pF 100pF 50 2.2H Figure 93 and Figure 94 illustrate the magnitude response and group delay response of the fourth-order filter, respectively. FREQUENCY RESPONSE (dB) As the load impedance of the filter increases, the filter design becomes more challenging in terms of meeting the required rejection and pass band specifications. In the previous W-CDMA example, the 500 load impedance resulted in the design of a sixth-order filter that has relatively large inductor values and small capacitor values. If the load impedance is 200 , the filter design becomes much more manageable. Figure 92 shows a fourth-order filter designed for a 10 MHz wide LTE signal. As shown in Figure 92, the resultant inductor and capacitor values become much more practical with a 200 load. Figure 94. Fourth-Order Low-Pass LTE Filter Group Delay Response Rev. 0 | Page 29 of 36 ADL5380 CHARACTERIZATION SETUPS The two setups shown in Figure 95 and Figure 96 were used for making NF measurements. Figure 95 shows the setup for measuring NF with no blocker signal applied while Figure 96 was used to measure NF in the presence of a blocker. For both setups, the noise was measured at a baseband frequency of 10 MHz. For the case where a blocker was applied, the output blocker was at a 15 MHz baseband frequency. Note that great care must be taken when measuring NF in the presence of a blocker. The RF blocker generator must be filtered to prevent its noise (which increases with increasing generator output power) from swamping the noise contribution of the ADL5380. At least 30 dB of attention at the RF and image frequencies is desired. For example, assume a 915 MHz signal applied to the LO inputs of the ADL5380. To obtain a 15 MHz output blocker signal, the RF blocker generator is set to 930 MHz and the filters tuned such that there is at least 30 dB of attenuation from the generator at both the desired RF frequency (925 MHz) and the image RF frequency (905 MHz). Finally, the blocker must be removed from the output (by the 10 MHz low-pass filter) to prevent the blocker from swamping the analyzer. Figure 95 to Figure 97 show the general characterization bench setups used extensively for the ADL5380. The setup shown in Figure 97 was used to do the bulk of the testing and used sinusoidal signals on both the LO and RF inputs. An automated Agilent VEE program was used to control the equipment over the IEEE bus. This setup was used to measure gain, IP1dB, IIP2, IIP3, I/Q gain match, and quadrature error. The ADL5380 characterization board had a 9-to-1 impedance transformer on each of the differential baseband ports to do the differential-tosingle-ended conversion, which presented a 450 differential load to each baseband port, when interfaced with 50 test equipment. For all measurements of the ADL5380, the loss of the RF input balun was de-embedded. Due to the wideband nature of the ADL5380, three different board configurations had to be used to characterize the product. For low band characterization (400 MHz to 3 GHz), the Mini-Circuits TC1-1-13 balun was used on the RF and LO inputs to create differential signals at the device pins. For midband characterization (3 GHz to 4 GHz), the Johanson Technology 3600BL14M050T was used, and for high band characterization (5 GHz to 6 GHz), the Johanson Technology 5400BL15B050E balun was used. SNS OUTPUT RF ADL5380 VPOS CHAR BOARD I LO INPUT 6dB PAD HP 6235A POWER SUPPLY R1 50 AGILENT N8974A NOISE FIGURE ANALYZER LOW-PASS FILTER IEEE GND Q FROM SNS PORT CONTROL AGILENT 8665B SIGNAL GENERATOR PC CONTROLLER Figure 95. General Noise Figure Measurement Setup Rev. 0 | Page 30 of 36 07585-095 IEEE ADL5380 BAND-PASS TUNABLE FILTER BAND-REJECT TUNABLE FILTER 6dB PAD R&S SMT03 SIGNAL GENERATOR RF GND ADL5380 6dB PAD VPOS CHAR BOARD LOW-PASS FILTER I LO 6dB PAD HP 6235A POWER SUPPLY R&S FSEA30 SPECTRUM ANALYZER R1 50 Q HP 87405 LOW NOISE PREAMP 07585-096 BAND-PASS CAVITY FILTER AGILENT 8665B SIGNAL GENERATOR Figure 96. Measurement Setup for Noise Figure in the Presence of a Blocker 3dB PAD RF AMPLIFIER 3dB PAD IN RF OUT 3dB PAD IEEE VP GND 3dB PAD AGILENT 11636A R&S SMT06 6dB PAD IEEE RF SWITCH MATRIX VPOS CHAR BOARD LO I 6dB PAD IEEE 6dB PAD AGILENT E3631 POWER SUPPLY RF INPUT AGILENT E8257D SIGNAL GENERATOR IEEE PC CONTROLLER IEEE R&S FSEA30 SPECTRUM ANALYZER Figure 97. General Characterization Setup Rev. 0 | Page 31 of 36 HP 8508A VECTOR VOLTMETER 07585-097 IEEE Q 6dB PAD ADL5380 IEEE RF GND INPUT CHANNELS A AND B R&S SMT06 ADL5380 EVALUATION BOARD The Johanson Technology 5400BL15K050 shares the same footprint and can be used for operation between 4900 MHz to 5800 MHz. The ADL5380 evaluation board is available. There are two versions of the board, optimized for performance for separate frequency ranges. For operation <3 GHz, an FR4 material-based board with the TC1-1-13 balun footprint is available. For operation between 3 GHz to 6 GHz, a Rogers(R) material-based RO3003 board with the Johanson Technology 3600BL14M050 balun (optimal for operation between 3 GHz and 4 GHz) footprint is available. The board can be used for single-ended or differential baseband analysis. The default configuration of the board is for single-ended baseband analysis. RFx T3x C5x C12x R19x R23x R5x IPx 23 22 21 20 19 GND3 RFIP RFIN GND3 ADJ C8x C11x 24 VCC3 VPOS 1 GND3 GND3 18 2 GND1 GND2 17 3 IHI R16x R17x T4x C16x R7x VPOS 4 ILO 5 GND1 QHI R3x R6x ADL5380 QLO 15 R15x R9x QNx R12x VCC2 13 VCC1 NC C7x GND4 6 C6x LOIN C9x R10x LOIP VPOS T2x C15x R13x GND4 R4x R18x GND2 14 ENBL INx QPx R14x 16 7 8 9 10 11 12 R2x VPOS C10x R1x R11x VPOS C2x C3x C4x C1x P1x VPOS T1x LONx LOPx LO_SE 2. FOR OPERATION BETWEEN 4.9GHZ TO 6GHZ, THE JOHANSON TECHNOLOGY 5400BL15K050 BALUN, WHICH SHARES A SIMILAR FOOTPRINT AS THE 4GHZ BALUN, CAN BE USED. Figure 98. Evaluation Board Schematic Rev. 0 | Page 32 of 36 07585-098 NOTES 1. X = B, FOR LOW FREQUENCY OPERATION UP TO 3GHz, TC1-1-13 BALUN ON RF AND LO PORTS. X = A, FOR FREQUENCY OPERATION FROM 3GHz TO 4GHz, JOHANSON TECHNOLOGY 3600BL14M050 BALUN ON RF AND LO PORTS. ADL5380 Table 5. Evaluation Board Configuration Options Component VPOSx, GNDx R10x, R12x, R19x C6x to C11x Description Power Supply and Ground Vector Pins. Power Supply Decoupling. Shorts or power supply decoupling resistors. Default Condition Not applicable R10x, R12x, R19x = 0 (0603) The capacitors provide the required dc coupling up to 6 GHz. P1x, R11x, R9x, R1x R23x Device Enable. When connected to VS, the device is active. C6x, C7x, C8x = 100 pF (0402), C9x, C10x, C11x = 0.1 F (0603) P1x, R9x = DNI, R1x = DNI, R11x = 0 R23B = 1.5 k (0603), R23A = 200 (0603) C1x, C4x = DNI, C2x, C3x, C5x, C12x = 100 pF (0402) R2x to R7x = open, R13x to R18x = 0 (0402) C1x to C5x, C12x R2x to R7x, R13x to R18x T2x, T4x C15x, C16x T1x T3x Adjust Pin. The resistor value here sets the bias voltage at this pin and optimizes third-order distortion. AC Coupling Capacitors. These capacitors provide the required ac coupling from 400 MHz to 4 GHz. Single-Ended Baseband Output Path. This is the default configuration of the evaluation board. R13x to R18x are populated for appropriate balun interface. R2x to R5x are not populated. Baseband outputs are taken from QHI and IHI. The user can reconfigure the board to use full differential baseband outputs. R2x to R5x provide a means to bypass the 9:1 TCM9-1 transformer to allow for differential baseband outputs. Access the differential baseband signals by populating R2x to R5x with 0 and not populating R13x to R18x. This way the transformer does not need to be removed. The baseband outputs are taken from the SMAs of QHI, QLO, IHI, and ILO. R6x and R7x are provisions for applying a specific differential load across the baseband outputs IF Output Interface. TCM9-1 converts a differential high impedance IF output to a single-ended output. When loaded with 50 , this balun presents a 450 load to the device. The center tap can be decoupled through a capacitor to ground. Decoupling Capacitors. C15x and C16x are the decoupling capacitors used to reject noise on the center tap of the TCM9-1. LO Input Interface. A 1:1 RF balun that converts the single-ended RF input to differential signal is used. RF Input Interface. A 1:1 RF balun that converts the single-ended RF input to differential signal is used. Rev. 0 | Page 33 of 36 T2x, T4x = TCM9-1, 9:1 (Mini-Circuits) C15x, C16x = 0.1 F (0402) T1B = TC1-1-13, 1:1 (Mini-Circuits) for operation <3 GHz, T1A = Johanson Technology 3600BL14M050 for operation from 3 GHz to 4 GHz, Johanson Technology 5400BL15K050 for operation from 4900 MHz to 5800 MHz T3B = TC1-1-13, 1:1 (Mini-Circuits) for operation <3 GHz, T3A = Johanson Technology 3600BL14M050 for operation from 3 GHz to 4 GHz, Johanson Technology 5400BL15K050 for operation from 4900 MHz to 5800 MHz 07585-101 07585-099 ADL5380 Figure 99. Low Band Evaluation Board Top Layer Figure 100. Midband/High Band Evaluation Board Top Layer Silkscreen 07585-102 07585-100 Figure 101. Low Band Evaluation Board Bottom Layer Figure 102. Midband/High Band Evaluation Board Bottom Layer Silkscreen 12 mil. THERMAL GROUNDING AND EVALUATION BOARD LAYOUT The package for the ADL5380 features an exposed paddle on the underside that should be well soldered to a low thermal and electrical impedance ground plane. This paddle is typically soldered to an exposed opening in the solder mask on the evaluation board. Figure 103 illustrates the dimensions used in the layout of the ADL5380 footprint on the ADL5380 evaluation board (1 mil = 0.0254 mm). 25 mil. 23 mil. 82 mil. Notice the use of nine via holes on the exposed paddle. These ground vias should be connected to all other ground layers on the evaluation board to maximize heat dissipation from the device package. 12 mil. 98.4 mil. 133.8 mil. 07585-105 19.7 mil. Figure 103. Dimensions for Evaluation Board Layout for the ADL5380 Package Under these conditions, the thermal impedance of the ADL5380 was measured to be approximately 30C/W in still air. Rev. 0 | Page 34 of 36 ADL5380 OUTLINE DIMENSIONS 0.60 MAX 4.00 BSC SQ TOP VIEW 0.50 BSC 3.75 BSC SQ 0.50 0.40 0.30 1.00 0.85 0.80 12 MAX 0.80 MAX 0.65 TYP 0.30 0.23 0.18 SEATING PLANE PIN 1 INDICATOR 24 1 19 18 2.65 2.50 SQ 2.35 EXPOSED PAD (BOTTOMVIEW) 13 12 7 6 0.23 MIN 2.50 REF 0.05 MAX 0.02 NOM 0.20 REF COPLANARITY 0.08 FOR PROPER CONNECTION OF THE EXPOSED PAD, REFER TO THE PIN CONFIGURATION AND FUNCTION DESCRIPTIONS SECTION OF THIS DATA SHEET. 082908-A PIN 1 INDICATOR 0.60 MAX COMPLIANT TO JEDEC STANDARDS MO-220-VGGD-8 Figure 104. 24-Lead Lead Frame Chip Scale Package [LFCSP_VQ] 4 mm x 4 mm Body, Very Thin Quad (CP-24-3) Dimensions shown in millimeters ORDERING GUIDE Model ADL5380ACPZ-R7 1 ADL5380ACPZ-WP1 ADL5380-29A-EVALZ1 ADL5380-30A-EVALZ1 1 Temperature Range -40C to +85C -40C to +85C Package Description 24-Lead LFCSP_VQ 24-Lead LFCSP_VQ Mid Band (3 GHz to 4 GHz) Evaluation Board Low Band (400 MHz to 3 GHz) Evaluation Board Z = RoHS Compliant Part. Rev. 0 | Page 35 of 36 Package Option CP-24-3 CP-24-3 Ordering Quantity 1,500, 7" Tape and Reel 64, Waffle Pack 1 1 ADL5380 NOTES (c)2009 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D07585-0-7/09(0) Rev. 0 | Page 36 of 36