 |
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
| Datasheet File OCR Text: |
| Low Cost DC-150 MHz Variable Gain Amplifier AD8330* FEATURES Fully Differential Signal Path May also be Used with Single-Sided Signals Inputs from 0.3 mV to 1 V rms, Rail-to-Rail Outputs Differential RIN = 1 k ; ROUT (Each Output) 75 Automatic Offset Compensation (Optional) Linear-in-dB and Linear-in-Magnitude Gain Modes 0 dB to 50 dB, for 0 V < VDBS < 1.5 V (30 mV/dB) Inverted Gain Mode: 50 dB to 0 dB at -30 mV/dB 0.03 to 10 Nominal Gain for 15 mV < VMAG < 5 V Constant Bandwidth: 150 MHz at All Gains Low Noise: 5 nV/Hz Typical at Maximum Gain Low Distortion: -62 dBc Typ Low Power: 20 mA Typ at VS of 2.7 V - 6 V Available in Space-Saving 3 3 LFCSP Package APPLICATIONS Pre-ADC Signal Conditioning 75 Cable Driving Adjust AGC Amplifiers GENERAL DESCRIPTION FUNCTIONAL BLOCK DIAGRAM 16 ENBL 1 VPSI 15 OFST 14 VPOS CM AND OFFSET CONTROL 13 CNTR VPSO 12 BIAS AND VREF INHI 2 VGA CORE INLO 3 OUTPUT STAGES OPHI 11 OPLO 10 MODE 4 VDBS 5 GAIN INTERFACE CMGN 6 7 OUTPUT CONTROL CMOP 9 COMM VMAG 8 90 70 The AD8330 is a wideband variable gain amplifier for use in applications requiring a fully differential signal path, low noise, welldefined gain, and moderately low distortion, from dc to 150 MHz. The input pins can also be driven from a single ended source. The peak differential input is 2 V, allowing sine wave operation at 1 V rms with generous headroom. The output pins can optionally drive single-sided loads and each swing essentially rail-to-rail. The differential output resistance is 150 . The output swing is a linear function of the voltage applied to the VMAG pin, which internally defaults to 0.5 V, to provide a peak output of 2 V. This may be raised to 10 V p-p, limited by the supply voltage. The basic gain function is linear-in-dB, controlled by the voltage applied to pin VDBS. The gain ranges from 0 dB to 50 dB for control voltages between 0 V and 1.5 V--a slope of 30 mV dB. The gain linearity is typically within 0.1 dB. By changing the logic level on pin MODE, the gain will decrease over the same range, with opposite slope. A second gain control port is provided at pin VMAG and allows the user to vary the numeric gain from a factor of 0.03 to 10. All the parameters of the AD8330 have low sensitivities to temperature and supply voltages. Using VMAG, the basic 0 dB to 50 dB range can be repositioned to any value from 20 dB higher (that is, 20 dB to 70 dB) to at least 30 dB lower (that is, -30 dB to +20 dB) to suit the application, providing an unprecedented gain range of over 100 dB. A unique aspect of the AD8330 is that its bandwidth and pulse response are essentially constant for all gains, not only over the basic 50 dB linear-in-dB *Protected by U.S.Patent No. 5,969,657; other patents pending. 50 GAIN - dB 30 10 -10 -30 -50 100k 1M 10M FREQUENCY - Hz 100M 300M Figure 1. AC Response over the Extended Gain Range range, but also when using the linear-in-magnitude function. The exceptional stability of the HF response over the gain range is of particular value in those VGA applications where it is essential to maintain accurate gain law-conformance at high frequencies. An external capacitor at pin OFST sets the high-pass corner of an offset reduction loop, whose frequency may be as low as 5 Hz. When this pin is grounded, the signal path becomes dc-coupled. When used to drive an ADC, an external common-mode control voltage at pin CNTR can be driven to within 0.5 V of either ground or VS to accommodate a wide variety of requirements. By default, the two outputs are positioned at the midpoint of the supply, VS/2. Other features, such as two levels of power-down (fully off and a hibernate mode), further extend the practical value of this exceptionally versatile VGA. The AD8330 is available in 16-lead LFCSP and 16-lead QSOP packages and is specified for operation from -40C to +85C. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781/329-4700 www.analog.com Fax: 781/326-8703 (c) 2004 Analog Devices, Inc. All rights reserved. REV. B 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. 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. V, T 12 pF on OPHI and OPLO, R = 0.75 V, otherwise AD8330-SPECIFICATIONS V(V ==50/C, V= 25 =C,0CV,=Differential Operation, unless 0/C, V = noted.)V S A L L DBS MAG OFST MODE = HI, Parameter INPUT INTERFACE Full-Scale Input Input Resistance Input Capacitance Voltage Noise Spectral Density Common-Mode Voltage Level Input Offset Drift Permissible CM Range1 Common-Mode AC Rejection OUTPUT INTERFACE Small Signal -3 dB Bandwidth Peak Slew Rate Peak-to-Peak Output Swing Common-Mode Voltage Voltage Noise Spectral Density Differential Output Impedance HD22 HD32 OUTPUT OFFSET CONTROL AC-Coupled Offset High-Pass Corner Frequency COMMON-MODE CONTROL Usable Voltage Range Input Resistance DECIBEL GAIN CONTROL Normal Voltage Range Elevated Range Gain Scaling Gain Linearity Error Absolute Gain Error Bias Current Incremental Resistance Gain Settling Time to 0.5 dB Error Mode Up/Down Mode Up Logic Level Mode Down Logic Level LINEAR GAIN INTERFACE Peak Output Scaling, Gain vs. VMAG Gain Multiplication Factor vs. VMAG Usable Input Range Default Voltage Incremental Resistance Bandwidth Conditions Pins INHI, INLO VDBS = 0 V, Differential Drive VDBS = 1.5 V Pin-to-Pin Either Pin to COMM f = 1 MHz, VDBS = 1.5 V; Inputs ac-shorted Pin OFST Connected to COMM Min 1.4 4.5 800 Typ 2 6.3 1k 4 5 3.0 1 2 Max Unit V mV pF nV/Hz V mV rms V/C V dB dB MHz V/s V V V nV/Hz dBc dBc mV rms kHz 1.2k 0 f = 1 MHz, 0.1 V rms f = 50 MHz Pins OPHI, OPLO 0 V < VDBS < 1.5 V VDBS = 0 VMAG 2 V (Peaks are Supply Limited) Pin CNTR O/C f = 1 MHz, VDBS = 0 Pin-to-Pin VOUT = 1 V p-p, f = 10 MHz, RL = 1 k VOUT = 1 V p-p, f = 10 MHz, RL = 1 k Pin OFST CHPF on Pin OFST (0 V< VDBS < 1.5 V) CHPF = 3.3 nF, from OFST to CNTR (Scales as 1/CHPF) Pin CNTR 0.5 From Pin CNTR to VS/2 Pins VDBS, CMGN, MODE CMGN Connected to COMM CMGN O/C (VCMGN Rises to 0.2 V) Mode HIGH or LOW 0.3 V VDBS 1.2 V VDBS = 0 Flows out of pin VDBS VDBS Stepped from 0.05 V-1.45 V or 1.45 V-0.05 V Pin MODE Gain Increases with VDBS, MODE = O/C Gain Decreases with VDBS Pins VMAG, CMGN See Circuit Description Section Gain is Nominal when VMAG = 0.5 V VMAG O/C For VMAG 0.1 V 4 -60 -55 150 1500 2 4.5 2.5 62 150 -62 -53 10 100 VS 1.8 4 2.4 120 2.2 2.6 180 4.5 V k V V mV/dB dB dB nA M ns 27 -0.35 -2 0 to 1.5 0.2 to 1.7 30 33 0.1 +0.35 0.5 +2 100 100 250 1.5 0.5 3.8 0 0.48 4.0 2 0.5 4 150 4.2 5 0.52 V V V/V V V k MHz -2- REV. B AD8330 Parameter CHIP ENABLE Logic Voltage for Full Shutdown Logic Voltage for Hibernate Mode Logic Voltage for Full Operation Current in Full Shutdown Current in Hibernate Mode Minimum Time Delay3 POWER SUPPLY Supply Voltage Quiescent Current Conditions Pin ENBL Output Pins Remain at CNTR 1.3 2.3 1.5 20 1.5 1.7 Pins VPSI, VPOS, VPSO, COMM, CMOP 2.7 VDBS = 0.75 V 20 6 27 V mA 0.5 1.7 100 V V V A mA s Min Typ Max Unit NOTES 1 The use of an input common-mode voltage significantly different than the internally set value is not recommended due to its effect on noise performance. See Figure 13. 2 See Typical Performance Characteristics for more detailed information on distortion in a variety of operating conditions. 3 For minimum sized coupling capacitors. REV. B -3- AD8330 ABSOLUTE MAXIMUM RATINGS 1 Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 V Power Dissipation RQ Package2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.62 W CP Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.67 W Input Voltage at Any Pin . . . . . . . . . . . . . . . . . . . VS + 200 mV Storage Temperature . . . . . . . . . . . . . . . . . . . -65C to +105C JA Operating Temperature Range . . . . . . . . . . . . -40C to +85C Lead Temperature (Soldering 60 sec) . . . . . . . . . . . . . . 300C NOTES 1 Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only and 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. 2 Four-Layer JEDEC Board (252P). RQ-16 Package . . . . . . . . . . . . . . . . . . . . . . . . . 105.4C/W CP-16 Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60C/W JC RQ-16 Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39C/W ORDERING GUIDE Model AD8330ACP-R2 AD8330ACP-REEL AD8330ACP-REEL7 AD8330ACPZ-R2* AD8330ACPZ-REEL* AD8330ACPZ-REEL7* AD8330ARQ AD8330ARQ-REEL AD8330ARQ-REEL7 AD8330ARQZ* AD8330ARQZ-REEL* AD8330ARQZ-REEL7* AD8330-EVAL *Z = Pb-free part. Temperature Range -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C -40C to +85C Package Description LFCSP LFCSP LFCSP LFCSP LFCSP LFCSP QSOP QSOP QSOP QSOP QSOP QSOP Evaluation Board Package Option CP-16-3 CP-16-3 CP-16-3 CP-16-3 CP-16-3 CP-16-3 RQ-16 RQ-16 RQ-16 RQ-16 RQ-16 RQ-16 Branding JFA JFA JFA JFA JFA JFA CAUTION ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on the human body and test equipment and can discharge without detection. Although the AD8330 features proprietary ESD protection circuitry, permanent damage may occur on devices subjected to high energy electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance degradation or loss of functionality. -4- REV. B AD8330 PIN CONFIGURATIONS 16-Lead LFCSP VPOS CNTR ENBL OFST 16-Lead QSOP OFST 1 ENBL 2 VPSI 3 16 15 VPOS CNTR VPSO 16 15 14 13 AD8330 14 VSPI INHI INLO MODE 1 2 3 4 12 VPSO OPHI OPLO CMOP TOP VIEW 13 OPHI (Not to Scale) 12 OPLO INLO 5 MODE 6 VDBS 7 CMGN 8 11 10 9 INHI 4 AD8330 TOP VIEW (Not to Scale) 11 10 9 CMOP VMAG COMM 5 6 7 8 COMM CMGN VMAG VDBS PIN FUNCTION DESCRIPTIONS 16-Lead LFCSP 16-Lead QSOP Pin No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Mnemonic VPSI INHI INLO MODE VDBS CMGN COMM VMAG CMOP OPLO OPHI VPSO CNTR VPOS OFST ENBL Description Positive Supply for Input Stages Differential Signal Input, Positive Polarity Differential Signal Input, Negative Polarity Logic Input: Selects Gain Slope. High = Gain Up versus VDBS Input for Linear-in-dB Gain Control Voltage, VDBS Common Baseline for Gain Control Interfaces Ground for Input and Gain Control Bias Circuitry Input for Gain/Amplitude Control, VMAG Ground for Output Stages Differential Signal Output, Negative Polarity Differential Signal Output, Positive Polarity Positive Supply for Output Stages Common-Mode Output Voltage Control Positive Supply for Inner Stages Used in Offset Control Modes Power Enable, Active High Pin No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Mnemonic OFST ENBL VPSI INHI INLO MODE VDBS CMGN COMM VMAG CMOP OPLO OPHI VPSO CNTR VPOS Description Used in Offset Control Modes Power Enable, Active High Positive Supply for Input Stages Differential Signal Input, Positive Polarity Differential Signal Input, Negative Polarity Logic Input: Selects Gain Slope. High = Gain Up versus VDBS Input for Linear-in-dB Gain Control Voltage, VDBS Common Baseline for Gain Control Interfaces Ground for Input and Gain Control Bias Circuitry Input for Gain/Amplitude Control, VMAG Ground for Output Stages Differential Signal Output, Negative Polarity Differential Signal Output, Positive Polarity Positive Supply for Output Stages Common-Mode Output Voltage Control Positive Supply for Inner Stages REV. B -5- AD8330-Typical Performance Characteristics VS = 5 V, TA = 25 C, CL = 12 pF, VDBS = 0.75 V, VMODE = High (or O/C) VMAG = O/C, RL = O/C, VOFST = 0, Differential Operation, unless otherwise stated. 50 45 LO MODE 40 HI MODE 2.0 NORMALIZED @ V DBS = 0.75V 1.5 1.0 GAIN ERROR - dB 35 100MHz 0.5 10, 50MHz 0 -0.5 1MHz -1.0 100MHz -1.5 -2.0 1MHz 50MHz GAIN - dB 30 25 20 15 10 5 0 0 0.25 0.50 0.75 VDBS - V 1.00 1.25 1.50 10MHz 0 0.2 0.4 0.6 0.8 1.0 VDBS - V 1.2 1.4 1.6 TPC 1. Gain vs. VDBS TPC 4. Gain Error vs. VDBS at Various Frequencies 10 9 20 15 10 2340 UNITS MODE = LO GAIN MULTIPLICATION FACTOR 8 7 % OF UNITS 6 5 4 3 2 1 0 0 1 2 VMAG - V 3 4 5 5 0 -30.6 -30.5 -30.4 -30.3 -30.2 -30.1 -30.0 -29.9 -29.8 -29.7 -29.6 -29.5 -29.4 -29.3 -29.2 -29.1 -29.0 20 MODE = HI 15 10 5 0 29.1 29.2 29.3 29.4 29.5 29.6 29.7 29.8 29.9 30.0 30.1 30.2 30.3 30.4 30.5 30.6 GAIN SCALING - mV/dB TPC 2. Linear Gain Multiplication Factor vs. VMAG TPC 5. Gain Slope Histogram 1.0 0.8 0.6 GAIN ERROR - dB 60 50 40 30 20 VDBS = 1.5V 1.2V 0.9V 0.6V 0.3V 0V 0.4 T = -40 C GAIN - dB 0.2 0 -0.2 10 0 -10 -20 -30 -40 T = +85 C -0.4 T = +25 C -0.6 -0.8 -1.0 0 0.2 0.4 0.6 0.8 VDBS - V 1.0 1.2 1.4 1.6 -50 100k 1M 10M FREQUENCY - Hz 100M 500M TPC 3. Gain Linearity Error Normalized at 25C vs. VDBS, at Three Temperatures, f = 1 MHz TPC 6. Frequency Response in 10 dB Steps for Various Values of VDBS -6- REV. B AD8330 50 40 30 20 4.8V 1.52V 20 .48V 25 1048 UNITS ENABLE MODE % OF UNITS .15V .048V .015V GAIN - dB 15 10 0 -10 -20 -30 -40 100k 10 5 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 1M 10M FREQUENCY - Hz 100M 500M DIFFERENTIAL OFFSET - mV TPC 7. Frequency Response for Various Values of VMAG TPC 10. Differential Input Offset Histogram 10 VDBS = 0.1V 8 OUTPUT BALANCE - dB GROUP DELAY - ns 10 0 -10 -20 -30 -40 -50 -60 -70 -80 6 4 2 0 100k 1M 10M FREQUENCY - Hz 100M 300M -90 100k 1M 10M FREQUENCY - Hz -0.1 100M TPC 8. Group Delay vs. Frequency TPC 11. Output Balance Error vs. Frequency for a Representative Part 0 200 190 -1 180 OFFSET VOLTAGE - mV T = -40 C -3 T = +25 C OUTPUT RESISTANCE - -2 170 160 150 140 130 120 -4 -5 -6 T = +85 C -7 0 0.2 0.4 0.6 0.8 VDBS - V 1.0 1.2 1.4 1.6 110 100 100k 1M 10M FREQUENCY - Hz 100M 300M TPC 9. Differential Output Offset vs. VDBS for Three Temperatures, for a Representative Part TPC 12. Output Impedance vs. Frequency REV. B -7- 1.0 0 AD8330 90 VDBS = 1.5V 80 70 60 VDBS = .75V OFST: ENABLED DISABLED 5000 6000 VDBS = 1.5V f = 1MHz CMRR - dB 50 40 30 20 10 0 -10 50k 100k VDBS = 0V NOISE - nV/ Hz 1M 10M FREQUENCY - Hz 100M 4000 3000 2000 1000 0 0 0.5 1.0 VMAG - V 1.5 2.0 2.5 TPC 13. CMRR vs. Frequency TPC 16. Output Referred Noise vs. VMAG 1500 f = 1MHz 1200 T = +85 C T = +25 C 80 70 60 VMAG = 0.5V f = 1MHz T = +85 C NOISE - nV/ Hz NOISE - nV/ Hz 900 50 40 T = +25 C 30 20 T = -40 C T = -40 C 600 300 10 0 0 0.2 0.4 0.6 0.8 1.0 VOLTAGE - V 1.2 1.4 1.6 0 0 0.2 0.4 0.6 0.8 VDBS - V 1.0 1.2 1.4 1.6 TPC 14. Output Referred Noise vs. VDBS for Three Temperatures TPC 17. Input Referred Noise vs. VDBS for Three Temperatures 700 f = 1MHz 600 180 f = 1MHz 160 140 500 NOISE - nV/ Hz NOISE - nV/ Hz 120 100 VMAG = 0.125V 400 VMAG = 0.5V 80 60 40 300 200 100 20 VMAG = 2V 0 0 0.5 1.0 VMAG - V 1.5 2.0 2.5 0 0 0.2 0.4 0.6 0.8 VDBS - V 1.0 1.2 1.4 1.6 TPC 15. Output Referred Noise vs. VMAG TPC 18. Input Referred Noise vs. VDBS for Three Values of VMAG -8- REV. B AD8330 7 VDBS = 1.5V 6 0 f = 10MHz -10 -20 DISTORTION - dBc -30 -40 HD3, RL = 1k -50 -60 -70 -80 1M 10M FREQUENCY - Hz 100M 0 0.3 0.6 0.9 VOUT - V p-p 1.2 1.5 HD2, RL = 1k 5 NOISE - nV/ Hz 4 3 2 1 0 100k TPC 19. Input Referred Noise vs. Frequency TPC 22. Harmonic Distortion vs. VOUT-DIFFERENTIAL VMAG = 0.5 V 0 VDBS = 0.75V VOUT = 1V p-p -10 RL = 1k -20 DISTORTION - dBc 0 f = 10MHz -10 -20 DISTORTION - dBc -30 -40 HD3, RL = 1k -50 -60 -70 *OUTPUT AMPLITUDE HARD LIMITED -80 1M 10M FREQUENCY - Hz 100M 0 1 2 3 VOUT - V p-p 4 5 HD2, RL = 1k HD2 AND HD3, RL = 150 * -30 -40 -50 HD 2 -60 -70 -80 100k HD 3 TPC 20. Harmonic Distortion vs. Frequency TPC 23. Harmonic Distortion vs. VOUT-DIFFERENTIAL VMAG = 2.0 V 0 VDBS = 0.75V VOUT = 1V p-p -10 RL = 1k -20 DISTORTION - dBc -30 -40 -50 -60 HD 2 -70 -80 0 10 20 30 CLOAD - pF 40 50 HD 3 DISTORTION - dBc 0 f = 10MHz VOUT - 1V p-p -10 RL - 1k -20 -30 HD3 -40 -50 HD2 -60 -70 0 0.2 0.4 0.6 0.8 1.0 VDBS - V 1.2 1.4 1.6 TPC 21. Harmonic Distortion vs. CLOAD TPC 24. Harmonic Distortion vs. VDBS REV. B -9- AD8330 10 f = 10MHz 0 INPUT VOLTAGE - dBVrms 13 f = 10MHz 15 28 23 20 33 OIPC - dBVrms -10 3 P1dB 10 f = 50MHz 5 23 -20 -7 18 -30 -17 -40 -27 0 13 -50 0 0.2 0.4 0.6 0.8 1.0 VDBS - V 1.2 1.4 -37 1.6 -5 0 0.2 0.4 0.6 0.8 1.0 VDBS - V 1.2 1.4 8 1.6 TPC 25. Input Voltage 1 dBV vs. VDBS TPC 28. Output IP3 vs. VDBS 20 OUTPUT V1dB COMPRESSION - dBVrms 33 f = 10MHz 25 f = 10MHz 38 10 23 20 33 15 28 f = 50MHz OIP3 - dBVrms 0 13 P1dB 10 23 -10 3 5 18 -20 -7 0 -30 -17 13 -5 8 -40 0 1 2 3 VMAG - V 4 5 -27 6 -10 0 0.2 0.4 0.6 0.8 VMAG - V 1.0 1.2 1.4 3 1.6 TPC 26. Output Voltage 1 dB vs. VMAG TPC 29. Output IP3 vs. VMAG 0 VDBS = 0.75V -10 VOUT = 1V p-p 1.5 1.0 -20 -30 0.5 IMD3 - dBc -40 -50 -60 -70 VOUT - V VDBS = 0V 0 -0.5 -1.0 -80 -90 1M -1.5 -50 10M FREQUENCY - Hz 100M -25 0 25 TIME - ns 50 75 100 TPC 27. IM3 Distortion vs. Frequency TPC 30. Full-Scale Transient Response, VDBS = 0 V -10- REV. B OUTPUT IP3 - dBm OUTPUT OIP3 - dBm AD8330 1.5 1V 1.0 0.5 VOUT - V VDBS = 0.75V 0 -0.5 -1.0 1V -1.5 -50 -25 0 25 TIME - ns 50 75 100 400ns TPC 31. Full-Scale Transient Response, VDBS = 0.75 V, f = 1 MHz, VOUT = 2 V p-p 1.5 TPC 34. VDBS Interface Response Top: VDBS, Bottom: VOUT 2V 1.0 0.5 VOUT - V VDBS = 1.5V 0 -0.5 -1.0 1mV -1.5 -50 -25 0 25 TIME - ns 50 75 100 400ns TPC 32. Full-Scale Transient Response, VDBS = 1.5 V, f = 1 MHz, VOUT = 2 V p-p TPC 35. VMAG Interface Response Top: VMAG, Bottom: VOUT 500mV CL = 12pF 1V VMAG = 5V VMAG = 0.5V CL = 54pF CL = 24pF VMAG = 0.05V 12.5ns 100mV 12.5ns TPC 33. Transient Response vs. for Various Load Capacitances, G = 25 dB TPC 36. Transient Response vs. VMAG REV. B -11- AD8330 26 2.00V OUTPUT 24 SUPPLY CURRENT - mA 22 +85 C 20 +25 C 18 -40 C INPUT 16 50mV 25ns 14 0 0.2 0.4 0.6 0.8 1.0 VDBS - V 1.2 1.4 1.6 TPC 37. Overdrive Response, VDBS = 1.5 V, VMAG = 0.5 V, 18.5 dB Overdrive TPC 40. Supply Current vs. VDBS at Three Temperatures 2V 3.125V 2.5V 1.875V 3.125V 2.5V 1.875V 1V 400ns 100ns TPC 38. ENBL Interface Response. Top: VENBL; Bottom: VOUT, f = 10 MHz TPC 41. CNTR Transient Response Top: Input to CNTR; Bottom, VOUT Single Ended -10 -20 -30 VPSI -40 -50 VPSO -60 VPOS -70 -80 -90 -100 -110 1M 10M FREQUENCY - Hz 100M 200M VDBS = 0.75V PSRR - dB TPC 39. PSRR vs. Frequency -12- REV. B AD8330 CIRCUIT DESCRIPTION Many monolithic variable gain amplifiers use techniques that share common principles that are broadly classified as translinear, a term referring to circuit cells whose functions depend directly on the very predictable properties of bipolar junction transistors, notably the linear dependence of their transconductance on collector current. Since the discovery of these cells in 1967, and their commercial exploitation in products developed during the early 1970s, accurate wide bandwidth analog multipliers, dividers, and variable gain amplifiers have invariably employed translinear principles. While these techniques are well understood, the realization of a high performance variable gain amplifier (VGA) requires special technologies and attention to many subtle details in its design. The AD8330 is fabricated on a proprietary silicon-on-insulator, complementary bipolar IC process and draws on decades of experience in developing many leading-edge products using translinear principles to provide an unprecedented level of versatility. Figure 2 shows a basic representative cell comprising just four transistors. This, or a very closely related form, is at the heart of most translinear multipliers, dividers, and VGAs. The key concepts are as follows: First, the ratio of the currents in the left-hand and right-hand pairs of transistors are identical; this is represented by the modulation factor, x, which may have values between -1 and +1. Second, the input signal is arranged to modulate the fixed tail current ID to cause the variable value of x introduced in the left-hand pair to be replicated in the right-hand pair, and thus generate the output by modulating its nominally fixed tail current IN. Third, the current-gain of this cell is very exactly G = IN/ID over many decades of variable bias current. In practice, the realization of the full potential of this circuit involves many other factors, but these three elementary ideas remain essential. By varying IN, the overall function is that of a two-quadrant analog multiplier, exhibiting a linear relationship to both the signal modulation factor x and this numerator current. On the other hand, by varying ID, a two-quadrant analog divider is realized, having a hyperbolic gain function with respect to the input factor x, controlled by this denominator current. The AD8330 exploits both modes of operation. However, since a hyperbolic gain function is generally of less value than one in which the decibel gain is a linear function of a control input, a special interface is included to provide either increasing or decreasing exponential control of ID. INPUT IS xlD G = IN/ID OUTPUT IS xlN LOOP AMPLIFIER (1-x) ID 2 (1-x) ID 2 + Q1 Q2 Q3 - Q4 (1-x) IN 2 (1+x) IN 2 ENBL VPSI OFST BIAS AND VREF VPOS CM MODE AND OFFSET CONTROL CNTR VPSO INHI AD8330 OUTPUT STAGES OPHI INLO VGA CORE OPLO MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG Figure 3. Block Schematic of the AD8330 Overall Structure Figure 3 shows a block schematic of the AD8330 in which the key sections are located. More detailed discussions of its structure and features are provided later; this figure provides a general overview of its capabilities. The VGA core contains a much elaborated version of the cell shown in Figure 2. The current called ID is controlled exponentially (linear in decibels) through the decibel gain interface at the pin VDBS and its local common CMGN. The gain span (that is, the decibel difference between maximum and minimum values) provided by this control function is slightly more than 50 dB. The absolute gain from input to output is a function of source and load impedance and also depends on the voltage on a second gain-control pin, VMAG, which is explained below. Normal Operating Conditions To minimize confusion, these normal operating conditions are defined as follows: the input pins are voltage driven (the source impedance is assumed to be zero); the output pins are open circuited (the load impedance is assumed to be infinite); pin VMAG is unconnected, which sets up the output bias current (IN in the four-transistor gain cell) to its nominal value; pin CMGN is grounded; and MODE is either tied to a logic high or left unconnected, to set the UP gain mode. The effects of other operating conditions can then be considered separately. Throughout this data sheet, the end-to-end voltage gain for the normal operating conditions will be referred to as the Basic Gain. Under these conditions, it runs from 0 dB when VDBS = 0 (where this voltage is more exactly measured with reference to pin CMGN, which may not necessarily be tied to ground) up to 50 dB for VDBS = 1.5 V. The gain does not "fold-over" when the VDBS pin is driven below ground or above its nominal fullscale value. The input is accepted at the differential port INHI/INLO. These pins are internally biased to roughly the midpoint of the supply VS (it is actually ~2.75 V for VS = 5 V, VDBS = 0, and 1.5 V for VS = 3 V), but the AD8330 is able to accept a forced commonmode value, from zero to VS, with certain limitations. This interface provides good common-mode rejection up to high frequencies (see TPC 13) and thus can be driven in either a single-sided or differential manner. However, operation using a differential drive is preferable, and this is assumed in the specifications, unless otherwise stated. -13- ID BIAS CURRENT DENOMINATOR NUMERATOR BIAS CURRENT IN Figure 2. The Basic Core of the AD8330 REV. B AD8330 The pin-to-pin input resistance is specified as 950 20%. The driving-point impedance of the signal source may range from zero up to values considerably in excess of this resistance, with a corresponding variation in noise figure (see Figure 10). In most cases, the input will be coupled via two capacitors, chosen to provide adequate low frequency transmission. This results in the minimum input noise, which is increased when some other common-mode voltage is forced onto these pins, as explained later. The short circuit, inputreferred noise at maximum gain is approximately 5 nV/Hz. The output pins OPHI/OPLO operate at a common-mode voltage at the midpoint of the supply, VS/2, within a few millivolts. This ensures that an analog-to-digital converter (ADC) attached to these outputs operates within the often narrow range permitted by their design. When a common-mode voltage other than VS/2 is required at this interface, it can easily be forced by applying an externally provided voltage to the output centering pin, CNTR. This voltage may run from zero to the full supply, though it must be noted that the use of such extreme values would leave only a small range for the differential output signal swing. The differential impedance measured between OPHI and OPLO is 150 20%. It follows that both the gain and the full-scale voltage swing will depend on the load impedance; both are nominally halved when this is also 150 . A fixedimpedance output interface, rather than an op amp style voltage-mode output, is preferable in high speed applications since the effects of complex reactive loads on the gain and phase can be better controlled. The top end of the AD8330's ac response is optimally flat for a 12 pF load on each pin, but this is not critical and the system will remain stable for any value of load capacitance including zero. Another useful feature of this VGA in connection with the driving of an ADC is that the peak output magnitude can be precisely controlled by the voltage on pin VMAG. Usually, this voltage is internally preset to 500 mV, and the peak differential, unloaded output swing is 2 V 3%. However, any voltage from zero to at least 5 V can be applied to this pin to alter the peak output in an exactly proportional way. Since either output pin can swing rail-to-rail, which in practice means down to at least 0.35 V and to within the same voltage below the supply, the peak-to-peak output between these pins can be as high as 10 V using VS = 6 V. VPSI CM MODE FEEDBACK TRANSIMPEDANCE OUTPUT STAGE INHI 500 V=0 INLO 500 V=0 ROUT = 150 OPLO O/P CM-MODE NORMALLY AT VP/2 CNTR LINEAR-IN-dB INTERFACE MODE VDBS VDBS COMM 12.65 A-4mA OR 4mA-12.65 A VMAG 5k COMM OPHI VPSO Linear-in-dB Gain Control (V DBS) A gain control law that is linear in decibels is frequently claimed for VGAs based more loosely on these principles. However, closer inspection reveals that their conformance to this ideal gain function is poor, usually only an approximation over part of the gain range. Furthermore, the calibration (so many decibels per volt) is invariably left unspecified, and the resulting gain often varies wildly with temperature. All Analog Devices VGAs featuring a linear-in-dB gain law, such as the X-AMPTM family, provide exact, constant gain scaling over the fully specified gain range, and the deviation from the ideal response is within a small fraction of a dB. For the AD8330, the scaling of both its gain interfaces is substantially independent of process, supply voltage, or temperature. The Basic Gain, GB, is simply GB (dB) = VDBS 30 mV (1) where VDBS is in volts. Alternatively, this can be expressed as a numerical gain magnitude GBN = 10 V DBS 0.6 V (2) As discussed later, the gain may be increased or decreased by changing the voltage VMAG applied to the VMAG pin. The internally set default value of 500 mV is derived from the same band gap reference that determines the decibel scaling. The tolerance on this voltage, and mismatches in certain on-chip resistors, cause small gain errors (see Specifications). While not all applications of VGAs demand accurate gain calibration, there are many situations in which it will be a valuable asset, for example, in reducing design tolerances. Figure 4 shows the core circuit in somewhat more detail. The range and scaling of VDBS is independent of the supply voltage, and the gain-control pin, VDBS, presents a high incremental input resistance (~100 M) with a low bias current (~100 nA), making the AD8330 easy to drive from a variety of gain-control sources. Inversion of the Gain Slope The AD8330 supports many new features that further extend the versatility of this VGA in wide bandwidth, gain control systems. For example, the logic pin MODE allows the slope of the gain function to be inverted, so that the basic gain starts at +50 dB for a gain voltage VDBS of zero and runs down to 0 dB when this voltage is at its maximum specified value of 1.5 V. The basic forms of these two gain control modes are shown in Figure 5. 50 MODE PIN LOW, GAIN DECREASES WITH VDBS MODE PIN HIGH, GAIN INCREASES WITH VDBS 40 30 dB 20 MAGNITUDE INTERFACE 100 A VMAG 10 0 0 0.25 0.50 0.75 V 1.0 1.25 1.50 Figure 5. The Two Gain Directions of the AD8330 Figure 4. Schematic of Key Components -14- REV. B AD8330 Gain Magnitude Control (V MAG) 0.10 0.05 0 -0.05 -0.10 1.2 1.0 0.8 0.6 0.4 0.2 0 2.5 2.0 1.5 1.0 0.5 0 -0.5 -1.0 -1.5 -2.0 -400 VMAG VIN In addition to the basic linear-in-dB control, two more gain control features are provided. The voltage applied to pin VMAG provides accurate linear-in-magnitude gain control with a very rapid response. The bandwidth of this interface is >100 MHz. When this pin is unconnected, VMAG assumes its default value of 500 mV (see Figure 4) to set up the basic 0 dB-50 dB range. But any voltage from ~15 mV to 5 V may be applied to either lower the gain by up to 30 dB or to raise it by 20 dB. The combined gain span is thus 100 dB, that is, the 50 dB Basic Gain span provided by VDBS plus a 60 dB linear-in-magnitude span provided by VMAG. The latter modifies the basic numerical gain GBN to generate a total gain, expressed here in magnitude terms V GT = GBN MAG (3) 0.5 V Using this to calculate the output voltage VOUT -300 -200 -100 0 100 200 300 TIME - ns VOUT = 2G BN VIN VMAG (4) Figure 7. Using VMAG in Modulation Mode from which it is apparent that the AD8330 implements a linear, two-quadrant multiplier with a bipolar VIN and a unipolar VMAG. Since the AD8330 is a dc-coupled system (the management of dc offsets at high gains is discussed later), it may be used in many applications where a wideband two-quadrant multiplier function is required, from dc up to about 100 MHz from either input (VIN or VMAG). As VMAG is varied, so also is the peak output magnitude, up to a point where this is limited by the absolute output limit imposed by the supply voltage. In the absence of the latter effect, the peak output into an open circuited load is just VOUT_PK = 2VMAG Another gain-related feature allows both of the gain control ranges to be accurately raised by 200 mV. To enable this offset, open circuit Pin 6 (CMGN) and add a 0.1 F capacitor to ground. In use, the nominal range for VDBS now extends from 0.2 V to 1.7 V and VMAG from 0.2 V to 5.2 V. These specifications apply for any supply voltage. This allows the use of DACs whose output range does not include ground as sources for the gain control function(s). Note that the 200 mV that appears on this pin will affect the response to an externally applied VMAG, but when pin VMAG is unconnected, the internally set default value of 0.5 V still applies. Furthermore, the pin CMGN can, if desired, be driven by a user supplied voltage to reposition the baseline for VDBS (or for an externally applied VMAG) to any other voltage up to 500 mV. In all cases, the gain scaling, its law conformance, and temperature stability are unaffected. Two Classes of Variable Gain Amplifiers (5) while for a load resistance of RL directly across OPHI and OPLO, it is 2 VMAG R L VOUT_PK = (6) (R L + 150) These capabilities are illustrated in Figure 6, where VS = 6 V, RL = O/C, VDBS = 0 V, VIN was swept from -2.5 V dc to +2.5 V dc and VMAG was set to 0.25 V, 0.5 V, 1 V, and 2 V. Except for the last value of VMAG, the peak output follows Equation 5; this exceeds the supply-limited value when VMAG = 2 V and the peak output is 5.65 V, that is, 6 V- 0.35 V. Figure 7 demonstrates the high speed multiplication capability. The signal input is a 100 MHz, 0.1 V sine wave, VDBS is set to 0.6 V, and VMAG is a square wave at 5 MHz alternating from 0.25 V to 1 V. The output is ideally a sine wave switching in amplitude between 0.5 V and 2 V. 8 6 4 2 VOUT - V It may be noted at this point that there are two broad classes of VGA. The first type is designed to cope with a very wide range of input amplitudes and, by virtue of its gain control function, compress this range down to an essentially constant output. This is the function needed in an AGC system. Such a VGA is called an IVGA, referring to a structure optimized to address a wide range of input amplitudes. By contrast, an OVGA is optimized to deliver a wide range of output values while operating with an essentially constant input amplitude. This is the function that might be needed, for example, in providing a variable drive to a power amplifier. It will be apparent from the foregoing that the AD8330 is both an IVGA and an OVGA in one package. This is an unusual and possibly confusing degree of versatility for a VGA; therefore, these two distinct control functions are discussed at separate points throughout this data sheet to explaining the operation and applications of this product. It is nevertheless useful to briefly demonstrate the capabilities of these features when used together. VMAG = 2V 1V 0.5V 0.25V 0 -2 -4 -6 -8 -3 -2 -1 0 VMAG - V 1 2 3 Figure 6. Effect of VMAG on Gain and Peak Output REV. B -15- AD8330 Amplitude/Phase Response 80 60 40 20 0 -20 -40 -60 100k The ac response of the AD8330 is remarkably consistent not only over the full 50 dB of its basic gain range, but also with changes of gain due to alteration of VMAG, as demonstrated in Figure 8. This is an overlay of two sets of results: first with a very low VMAG of 16 mV, which reduces the overall gain by 30 dB [20 log10 (500 mV/16 mV)]; second, with VMAG = 5 V, which increases the gain by 20 dB = 20 log10 (5 V/0.5 V). 90 70 GAIN - dB GAIN - dB NOISE - nV/ Hz 10k 1k 100 10 1 .001 .01 .1 VGAIN - V 1 10 50 30 10 -10 -30 -50 100k 0 1M 10M 100M 300M GAIN - Degrees -50 -100 -150 -200 -250 -300 -350 100k 1M 10M FREQUENCY - Hz G = +70dB Figure 9. Gain Control Function and Input Referred Noise Spectral Density over a 120 dB Range Noise, Input Capacity, and Dynamic Range G = -20dB 100M 300M Figure 8. AC Performance over a 100 dB Gain Range Obtained by Using Two Values of VMAG This 50 dB step change in gain produces the two sets of gain curves, having a total gain span of 100 dB. It is apparent that the amplitude and phase response are essentially independent of the gain over this wide range, an aspect of the AD8330's performance potential unprecedented in any prior VGA. It is unusual for an application to require such a wide range of gains, of course; and as a practical matter, the peak output voltage for VMAG = 16 mV is reduced by the factor 16/500, compared to its nominal value of 2 V, to only 64 mV. As already noted, most applications of VGAs require that they operate in a mode that is predominantly of either an IVGA or OVGA style, rather than mixed modes. With this limitation in mind, and simply in order to illustrate the unusual possibilities afforded by the AD8330, it is noted that with appropriate drive to VDBS and VMAG in tandem, the gain span is a remarkable 120 dB, extending from -50 dB to +70 dB, as shown in Figure 9 for operation at 1 MHz and 100 MHz. In this case, VDBS and VMAG are driven from a common control voltage, VGAIN, which is varied from 1.2 mV to 5 V, with 30% (1.5/5) of VGAIN applied to VDBS, and 100% applied to VMAG. The gain varies in a linear-in-dB manner with VDBS, while the response from VMAG is linear-in-magnitude. Consequently, the overall numerical gain is the product of these two functions is The design of variable-gain amplifiers invariably incurs some compromises in noise performance. However, the structure of the AD8330 is such that this penalty is minimal. Examination of the simplified schematic (Figure 4) shows that the input voltage is converted to current-mode form by the two 500 resistors at pins INHI and INLO, whose combined Johnson noise contributes 4.08 nV/Hz. The total input noise at full gain, when driven from a low impedance source, is typically 5 nV/Hz after accounting for the voltage and current noise contributions of the loop amplifier. For a 200 kHz channel bandwidth, this amounts to 2.24 V rms. The peak input at full gain is 6.4 mV, or +4.5 mV rms for a sine wave signal. The signal-to-noise ratio at full input, that is, the dynamic range, for these conditions is thus 20 log10(4.5 mV/2.24 V), or 66 dB. The value of VMAG has essentially no effect on the input-referred noise, but it is assumed to be 0.5 V. Below midgain (25 dB, VDBS = 0.75 V), noise in the output section dominates, and the total input noise is 11 nV/Hz, or 4.9 V rms in a 200 kHz bandwidth, while the peak input is 78 mV rms. Thus, the dynamic range has increased to 84 dB. At minimum gain, the input noise is up to 120 nV/Hz, or 53.7 mV rms in the assumed 200 kHz bandwidth, while the input capacity is 2 V, or +1.414 V rms (sine), a dynamic range of 88.4 dB. In calculating the dynamic range for other channels' bandwidths, f, subtract 10 log10(f /200 kHz) from these illustrative values. A system operating with a 2 MHz bandwidth, for example, will exhibit dynamic range values that are uniformly 10 dB lower; used in an audio application with a 20 kHz bandwidth, they will be 10 dB higher. Noise figure is a misleading metric for amplifiers that are not impedance matched at their input, which is the special condition resulting only when both the voltage and current components of a signal, that is, the signal power, are used at the input port. When a source of impedance RS is terminated using a resistor of RS (a condition that is not to be confused with matching), only one of these components is used, either the current (as in the AD8330) or the voltage. Then, even if the amplifier is perfect, the noise figure cannot be better than 3 dB. The 1 k internal termination resistance would result in a minimum noise figure of 3 dB for an RS of 1 k if the amplifier were noise-free. However, this is not the case and the minimum noise figure will occur at a slightly different value of RS (see Figure 10 and Using the AD8330 section). REV. B GAIN = VGAIN / 0.5 V x 0.3 x 10 VGAIN 0.6 V (7) In rare cases where such a wide gain range might be of value, the calibration will still be accurate and temperature stable. -16- AD8330 15 14 13 12 NOISE FIGURE 3.2 T = +85 C 3.1 T = +25 C DC VOLTAGE AT INHI, INLO - V 3.0 T = -40 C 2.9 11 10 9 8 7 6 5 10 2.8 2.7 2.6 100 RS - 1000 10k 0 0.2 0.4 0.6 0.8 1.0 VDBS V 1.2 1.4 1.6 Figure 10. Noise Figure for Source Resistance of 50 to 5 k, at f = 10 MHz (lower) and 100 MHz (Simulation) 144 140 Figure 12. Common-Mode Voltage at Input Pins vs. VDBS, for VS = 5 V, T = -35C, +25C, and +85C 26 24 INPUT REFERRED NOISE - nV/ Hz VDBS = 0.5V DYNAMIC RANGE - dB/ Hz CONSTANT 1V rms OUTPUT, BOTH CASES 136 X-AMP WITH 40dB OF GAIN AND AN INPUT NSD OF nV/ Hz 22 20 18 16 14 12 10 8 SIMULATION 6 4 VDBS = 1.5V VDBS = 0.75V VDBS = 0.6V 132 128 124 120 116 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 VDBS - V 0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 COMMON-MODE VOLTAGE AT INHI, INLO - V 4.4 4.8 Figure 11. Dynamic Range in dB/ Hz vs. VDBS (VMAG = 0.5 V, 1 V rms output) Compared with a Representative X-AMP (Simulation) Dynamic Range Figure 13. Input Noise vs. Common-Mode Input Voltage for VDBS = 0.5 V, 0.6 V, 0.75 V, and 1.5 V Output Noise and Peak Swing The ratio of peak output swing, expressed in rms terms, to the output-referred noise spectral density provides a measure of dynamic range, in dB/Hz. For a certain class of variable gain amplifiers, exemplified by Analog Devices' X-AMP family, the dynamic range is essentially independent of the gain setting, because the peak output swing and noise are both constant. The AD8330 provides a different dynamic range profile, since there is no longer a constant relationship between these two parameters. Figure 11 compares the dynamic range of the AD8330 to a representative X-AMP. Input Common-Mode Range and Rejection Ratio The output noise of the AD8330 is the input noise multiplied by the overall gain, which includes any optional change to the voltage VMAG applied to pin VMAG. The peak output swing is also proportional to this voltage, which, at low gains and high values of VMAG, will affect the output noise. The scaling for VDBS = 0 is as follows: VOUT_PK = 4 VMAG (8) (9) VNOISE_OUT = (85 + 70 VMAG ) nV / Hz The inputs INHI and INLO should be ac-coupled in most applications, to achieve the stated noise performance. When direct coupling is used, care must be taken in setting the dc voltage level at these inputs, in general, and particularly when minimizing noise is critical. This objective is complicated by the fact that the commonmode level varies with the basic gain voltage VDBS. Figure 12 shows this relationship for a supply voltage of 5 V, for temperatures of -35C, +25C, and +85C. Figure 13 shows the input noise spectral density (RS = 0) versus the input common-mode voltage, for VDBS = 0.5, 0.6 V, 0.75 V, and 1.5 V. It is apparent that there is a broad range over which the noise is unaffected by this dc level. The input CMRR is excellent (see TPC 13). REV. B For example, using a reduced value of VMAG = 0.25 V, which lowers all gain values by 6 dB, the peak output swing is 1 V (differentially) and the output noise spectral density evaluates to 102.5 nV/Hz. The peak output swing is no different at full gain, but the noise is now VNOISE_OUT = (0.1 + 0.32VMAG ) V / Hz (10) for RS = 0 and VDBS = 1.5 V, assuming an input noise of 5 nV/Hz. The output noise for very small values of VMAG (at or below 15 mV) is not precise, partly because the small input offset associated with this interface has a large effect on the gain. -17- AD8330 Offset Compensation The AD8330 includes an offset compensation feature, which is operational in the default condition (no connection to pin OFST). This loop introduces a high-pass filter function into the signal path, whose -3 dB corner frequency is at f HPF = (2 R INT C HP ) 1 (11) where CHP is the external capacitance added from OFST to CNTR, and RINT is an internal resistance of approximately 480 , having a maximum uncertainty of about 20%. This evaluates to fHPF 330 CHP from CNTR to ground can lower this voltage, or one to the supply will raise it. On the other hand, this pin may be driven by an external voltage source to set the common-mode level, to satisfy the needs of a following ADC, for example. Any value from 0.5 V above ground to 0.5 V below the supply is permissible. Of course, when using an extreme common-mode level, the available output swing will be limited, and it is recommended that a value equal or close to the default of VCNTR = VS/2 be used. There may be a few millivolts of offset between the applied voltage and the actual common-mode level at the output pins. The input common-mode voltage VCMI at pins INHI, INLO is slaved to the output, but with a shifted value of (CHP in F) (12) VCMI = 0.757 VCNTR + 1.12 V (14) A small amount of peaking at this corner when using small capacitor values can be avoided by adding a series resistor. Useful combinations are CHP = 3 nF, RHP = 180 , f = 100 kHz; CHP = 33 nF, RHP = 10 , f = 10 kHz; CHP = 0.33 F, RHP = 0 , f = 1 kHz; CHP = 3.3 F, RHP = 0 , f = 100 Hz. The offset compensation feature can be disabled simply by grounding the OFST pin. This provides a dc-coupled signal path, with no other effects on the overall ac response. Input offsets must be externally nulled in this mode of operation, as shown in Figure 15. Effects of Loading on Gain and AC Response for VDBS = 0.75 and T = 25C. Thus, the default value for VCMI when VS = 5 V is 3.01 V (see Figure 12). USING THE AD8330 There are very few precautions that need to be observed in applying the AD8330 to a wide variety of circumstances. A selection of specific applications is presented later; a few general aspects of utilization are discussed here. As in all high frequency circuits, careful observation of the ground nodes associated with each function is important. Three positive supply pins are provided. VPSI supports the input circuitry, which may often be operating at a relatively high sensitivity; VPOS, which supports general bias sources, needs no decoupling; VPSO is used to bias the output stage, where decoupling may be useful in maintaining a glitch-free output. Figure 14 shows the general case, where VPSI and VPSO are each provided with their own decoupling network, but this may not be needed in all cases. Because of the differential nature of the signal path, power supply decoupling is in general much less critical than in a single-sided amplifier, and where the minimization of board-level components is especially crucial, it may be found that these pins need no decoupling at all. On the other hand, when the signal source is VS 2.7V-6V RD1 CHPF CD2 RD2 The differential output impedance RO is 150 and the frequency response of the output stage is optimized for operation with a certain load capacitance on each output pin, OPHI and OPLO, to ground, in combination with a load resistance RL directly across these pins. In the absence of these capacitances, there will be a small amount of peaking at the top extremity of the ac response. Suitable combinations are: RL = , CL = 12 pF; RL = 150 , CL = 25 pF; RL = 75 , CL = 40 pF; RL = 50 , CL = 50 pF. The gain calibration is specified for an open-circuited load, such as the high input resistance of an ADC. When resistively loaded, all gain values are nominally lowered as follows: G LOADED = GUNLOADED R L (150 + R L ) (13) Thus when RL = 150 , the gain is reduced by 6 dB; for RL = 75 , the reduction is 9.5 dB; and for RL = 50 , it is 12 dB. Gain Errors Due to On-Chip Resistor Tolerances CD1 ENBL VPSI OFST BIAS AND V-REF VPOS CM MODE AND OFFSET CONTROL CNTR VPSO CD3 In all cases where external resistors are used, keep in mind that all on-chip resistances, including the RO and the input resistance, RI, are subject to variances of up to 20%, which will need to be accounted for when calculating the gain with input and output loading. This sensitivity can be avoided by adjusting the source and load resistances to bear an inverse relationship as follows: If RS = RI then make RL = RO /; or, if RL = RO then make RS = RI /. The simplest case is when RS = 1 k and RL = 150 . Here the gain is 12 dB lower than the basic value. The reduction of peak swing at the load can be corrected by using VMAG = 1 V, which also restores 6 dB of gain; using VMAG = 2 V restores the full basic gain while also doubling the peak available output swing. Output (Input) Common-Mode Control INHI INPUT, 0V TO 2V MAX INLO VGA CORE OUTPUT STAGES OPHI OUTPUT, 2V MAX OPLO NC MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG NC GROUND The output voltages are nominally positioned at the midpoint of the supply, VS/2, over the range 2.7 V < VS < 6 V, and this voltage appears at pin CNTR, which is not normally expected to be loaded (the source resistance is ~4 k). However, some circumstances may require a small change in this voltage, and a resistor BASIC GAIN BIAS VDBS: 0V TO 1.5V Figure 14. Power Supply Decoupling and Basic Connections -18- REV. B AD8330 single sided, extra attention may be needed to the decoupling on pin VPSI, while if the output is loaded on only one of its two output pins, this may require care in decoupling the VPSO pin. The general common COMM and the output stage common CMOP are usually grounded as shown in the figure; however, the Applications section shows how a negative supply can optionally be used. The AD8330 is enabled by taking the ENBL pin to a logical high (or, in all cases, the supply). The "UP" gain mode is enabled either by leaving the MODE pin unconnected or taken to a logical HI; when the opposite gain direction is needed, this pin should be grounded or driven to a logical low. The low-pass corner of the offset loop is determined by the capacitor CHPF; this is preferably tied to the CNTR pin, which in turn should be decoupled to ground. The gain-interface common pin CMGN is grounded, and the output magnitude control pin VMAG is left unconnected, or may optionally be connected to a 500 mV source for basic gain calibration. Connections to the input and output pins are not shown in this figure because of the many options that are available. When the AD8330 is used to drive an ADC, pins OPHI and OPLO may be connected directly to the differential inputs of a suitable converter, such as an AD9214. If an adjustment is needed to this commonmode level, it can be introduced by applying that voltage to CNTR, or, more simply, by using a resistor from this pin to either ground or the supply (see applications). This pin can also supply the common-mode voltage to an ADC that supports such a feature. When the loads to be driven introduce a dc resistive path to ground, coupling capacitors must be used; these should be of sufficient value to pass the lowest frequency components of the signal without excessive attenuation. Keep in mind that the voltage swing on such loads will alternate both above and below ground, requiring that the subsequent component must be able to cope with negative signal excursions. Gain and Swing Adjustments When Loaded each input pin, their minimum value can be readily found from this expression CIN_CPL = 320 F fHPF (15) where fHPF is the -3dB frequency expressed in hertz. Thus, for an fHPF of 10 kHz, 33 nF capacitors would be used. It may occasionally be possible to avoid the use of coupling capacitors, when the dc level of the driving source is within a certain range, as shown in Figure 13. This range extends from 3.5 V to 4.5 V when using a 5 V supply, and at high basic gains, where the effect of an incorrect dc level would be most troublesome, causing an increase in noise level due to internal aspects of the input stage. For example, suppose the driver IC is an LNA having an output topology in which its load resistors are taken to the supply, and the output is buffered by emitter followers. This presents a source for the AD8330 that could readily be directly coupled. DC-Coupled Signal Path In many cases, where the VGA is not required to provide its lowest noise, the full common-mode input range of zero to VS may be used without problems, avoiding the need for any ac coupling means. However, such direct coupling at both the input and output will not automatically result in a fully dc-coupled signal path. The internal offset compensation loop must also be disengaged, by connecting the OFST pin to ground. Keep in mind that at the maximum basic gain of 50 dB ( 316), every millivolt of offset at the input, arising from whatever source, causes an output offset of 316 mV, which is an appreciable fraction of the peak output swing. Since the offset correction loop is placed after the front-end variable gain sections of the AD8330, the most effective way of dealing with such offsets is at the input pins, as shown in Figure 15. For example, assume, for illustrative purposes, that the resistances associated with each side of the source in a certain application are 50 . If this source has a very low (op amp) output impedance, the extra resistors should be inserted, with a negligible noise penalty and an attenuation of only 0.83 dB. The resistor values shown provide a trim range of about 2 mV. VS 2.7V-6V RD1 CD2 RD2 The output can also be coupled to a load via a transformer, in which case it may be possible to achieve a higher load power by impedance transformation. For example, using a 2:1 turns ratio, a 50 final load will present a 200 load on the output. The gain loss (relative to the basic value with no termination) will be 20 log10{(200+150)/200} or 4.86 dB, which can be restored by raising the voltage on the VMAG pin by a factor of 104.86/20 or 1.75, from its basic value of 0.5 V to 0.875 V. This also restores the peak swing at the 200 level to 2 V, or 1 V into the 50 final load. Whenever a stable supply voltage is available, the additional voltage may be provided simply by adding a resistor from this pin to the supply. The calculation is based on knowing that the internal bias is delivered via a 5 k source; since an additional 0.375 V is needed, the current in this external resistor must be 0.375 V/5 k = 75 A. Thus, using a 5 V supply, a resistor of 5 V-0.875 V/75 A = 55 k would be used. Based on this example, the corrections for other load conditions should be easy to calculate. If the effects on gain and peak output swing due to supply variations cannot be tolerated, VMAG must be driven by an accurate voltage. Input Coupling CD1 ENBL VPSI OFST BIAS AND V-REF VPOS CM MODE AND OFFSET CONTROL CNTR VPSO CD3 INHI RS ASSUMED 50k TO BE 50 ON EACH SIDE INLO 75k VGA CORE OUTPUT STAGES OPHI OUTPUT, 2V MAX OPLO MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG NC GROUND The dc common-mode voltage at the input pins varies with the supply, the basic gain bias and temperature (see Figure 12); for this reason, many applications will need to use coupling capacitors from the source, which should be large enough to support the lowest frequencies to be transmitted. Using one capacitor at REV. B -19- BASIC GAIN BIAS VDBS: 0V TO 1.5V Figure 15. Input Offset Nulling in a DC-Coupled System AD8330 90 VDBS = 1.5V 80 70 60 VDBS = .75V OFST: ENABLED DISABLED CMRR - dB 50 40 30 20 10 0 -10 50k 100k VDBS = 0V Adding a dummy 75 to OPLO results in Line 3: the gain is a further 2.5 dB lower, at about 14 dB. The CM artifacts are no longer present but there is now a small amount of peaking. If objectionable, this may be eliminated by raising both of the capacitors on the output pins to 25 pF, as shown in Line 4 of Figure 17. The gain reduction incurred both by using only one output and by the additional effect of loading can be overcome by taking advantage of the VMAG feature, provided primarily for just such circumstances. Thus, to restore the basic gain in the first case (Line 1), a 1 V source should be applied to this pin; to restore the gain in the second case, this voltage should be raised by a factor of 1.5, to 1.5 V. In cases 3 and 4, a further factor of 1.33 is needed to make up the 2.5 dB loss, that is, VMAG should be raised to 2 V. With the restoration of gain, the peak output swing at the load is likewise restored to 2 V. Pulse Operation 1M 10M FREQUENCY - Hz 100M Figure 16. Input CMRR vs. Frequency for Various Values of VDBS Using Single-Sided Sources and Loads Where the source provides a single-sided output, either INHI or INLO may be used for the input, with of course a polarity change when using INLO. The unused pin must be connected either through a capacitor to ground, or a dc bias point that corresponds closely to the dc level on the active signal pin. The input CMRR over the full frequency range is illustrated in Figure 16. In some cases, an additional element such as a SAW filter (having a singlesided balanced configuration) or a flux-coupled transformer may be interposed. Where this element must be terminated in the correct impedance, other than 1 k, it will be necessary to add either shunt or series resistors at this interface. 30 LINE 1 When using the AD8330 in applications where its transient response is of greater interest and the outputs are conveyed to their load via coaxial cables, the added capacitances may be slightly different in value, and may be placed either at the sending or load end of the cables, or divided between these nodes. Figure 18 shows an illustrative example in which dual 1 meter 75 cables are driven through dc-blocking capacitors and independently terminated at ground level. Because of the considerable variation between applications, only general recommendations can be made with regard to minimizing pulse overshoot and droop. The former can be optimized by adding small load capacitances, if necessary; the latter require the use of sufficiently large capacitors C1. VS 2.7V-6V CD2 RD2 20 GAIN - dB LINE 3 10 0 LINE 4 -10 -20 -30 0 ENBL VPSI OFST BIAS AND V-REF VPOS CM MODE AND OFFSET CONTROL CNTR VPSO LINE 2 CD3 LINE 2 RL1 C1 INHI VGA CORE OUTPUT STAGES C1 INLO OPLO CL2 RL2 OPHI CL1 PHASE - Degrees -100 -200 -300 -400 -500 -600 1M 10M FREQUENCY - Hz 100M LINE 4 LINE 1 LINE 3 500M MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG NC Figure 17. AC Gain and Phase for Various Loading Conditions When driving a single-sided load, either OPHI or OPLO can be used. These outputs are very symmetric, so the only effect of this choice is to select the desired polarity. However, when the frequency range of interest extends to the upper limits of the AD8330, a dummy resistor of the same value should be attached to the unused output. Figure 17 illustrates the ac gain and phase response for various loads and VDBS = 0.75 V. Line 1 shows the unloaded (CL = 12 pF) case for reference; the gain is 6 dB lower (20 dB) using just the single-sided output. Adding a 75 load just from OPHI to an ac ground results in Line 2. The gain is now a factor of 1.5 or 3.54 dB lower, but artifacts of the output common-mode control loop now appear in both the magnitude and phase response. -20- Figure 18. Driving Dual Cables with Grounded Loads REV. B AD8330 1.2 1.0 0.8 0.6 0.4 0.2 0 -0.2 0.2 0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 1.2 1.0 0.8 0.6 0.4 0.2 0 -0.2 0.2 0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 Table I. Preserving Absolute Gain RS() 10 15 20 30 50 75 100 150 200 300 500 750 1k 1.5k 2k RL() 15k 10k 7.5k 5.0k 3.0k 2.0k 1.5k 1.0k 750 500 300 200 150 100 75 Uncorrected Loss Factor dB 0.980 0.971 0.961 0.943 0.907 0.865 0.826 0.756 0.694 0.592 0.444 0.327 0.250 0.160 0.111 0.17 0.26 0.34 0.51 0.85 1.26 1.66 2.43 3.17 4.56 7.04 9.72 12.0 15.9 19.1 VMAG Required to Correct Loss 0.510 0.515 0.520 0.530 0.551 0.578 0.605 0.661 0.720 0.845 1.125 1.531 2.000 3.125 4.500 0 5ns 10ns 15ns 20ns 25ns Figure 19. Typical Pulse Responses for Figure 18 Figure 19 shows typical results for VDBS = 0.24 V, a square wave input amplitude of 450 mV (the actual combination is not important) and a rise time of 2 ns. VMAG raised to 2.0 V is used. In the upper waveforms the load capacitors are both zero, and a small amount of overshoot is visible; with 40 pF the response is cleaner. A shunt capacitance of 20 pF from OPHI to OPLO will have a similar effect. Coupling capacitors for this demonstration are sufficiently large to prevent any visible droop over this time scale. The outputs at the load side will eventually assume a mean value of zero, with negative and positive excursions depending on the duty cycle. The bandwidth from pin VMAG to these outputs is somewhat higher than from the normal input pins. Thus when this pin is used to rapidly modulate the primary signal, some further experimentation with response optimization may be required. In general, the AD8330 is very tolerant of a wide range of loading conditions. Preserving Absolute Gain Calculation of Noise Figure The AD8330 noise is a consequence of its intrinsic voltage noise spectral density (ENSD) and the current noise spectral density (INSD). Their combined effect generates a net input noise, VNOISE_IN, which is a function of the device's input resistance, RI, nominally 1 k, and the differential source resistance, RS, as follows: VNOISE_IN = {E 2 NSD + I NSD (R I + R S ) 2 2 } (16) Although the AD8330 is not laser trimmed, its absolute gain calibration, based mainly on ratios, is very good. Full details can be found in the Specifications section and in the typical performance curves. Nevertheless, having finite input and output impedances, the gain is necessarily dependent on the source and load conditions. The loss incurred when either of these is finite causes an error in the absolute gain, which may also be uncertain due to the approximately 20% tolerance in the absolute value of the input and output impedances. Often, such losses and uncertainties can be tolerated and accommodated by a correction to the gain control bias. On the other hand, the error in the loss can be essentially nulled by using appropriate modifications to either the source impedance (RS) or the load impedance (RL), or both, in some cases by padding them with series or shunt components. The formulation for this correction technique was described previously. However, to simplify its use, Table I shows spot values for combinations of RS and RL resulting in an overall loss that will not be dependent on sample-to-sample variations in on-chip resistances. Furthermore, this fixed and predictable loss can be corrected by an adjustment to VMAG, as indicated in Table I. Note that we assume purely resistive source and input impedances as a concession to simplicity. A more thorough treatment of noise mechanisms, for the case where the source is reactive, is beyond the scope of these brief notes. Also note that VNOISE_IN is the voltage noise spectral density appearing across the differential input pins, INHI, INLO. In preparing for the calculation of noise figure, VSIG is defined as the open-circuit signal voltage across the source and VIN as the differential input to the AD8330. The relationship is simply VR VIN = SIG I (17) (R I + R S ) At maximum gain, ENSD is 4.1 nV/Hz, and INSD is 3 pA/Hz. Thus, the short-circuit voltage noise is VNOISE_IN = 4.1 nV / Hz ( ) + (3 pA / 2 Hz ) (1 k + 0) 2 2 = 5.08 nV / Hz (18) Next, examine the net noise when RS = RI = 1 k, often incorrectly called the "matching" condition, rather than "source impedance termination," which is the actual situation in this case. Repeating the procedure VNOISE_IN = 4.1 nV / Hz ( ) + (3 pA / 2 Hz ) (1 k + 1 k) 2 2 7.3 nV / Hz (19) REV. B -21- AD8330 The noise figure is just the decibel representation of the noise factor, NFAC, which is commonly defined as follows: P1 dB and V1 dB N FAC = Signal-to-noise ratio at input Signal-to-noise ratio at output (20) However, this is equivalent to Signal-to-noise ratio at the source N FAC = Signal-to-noise ratio at the input pins (21) Let VNSD be the voltage noise spectral density kTRS due to the source resistance. Then we have N FAC = VIN / VNOISE _ IN RS / ( RI + RS ) VSIG RI / ( RI + RS ) / VNSD In addition to the nonlinearities that arise within the core of the AD8330, at moderate output levels, a further metric that is more commonly stated for RF components that deliver appreciable power to a load is the 1 dB compression point. This is defined in a very specific manner: it is that point at which, with increasing output level, the power delivered to the load eventually falls to a value that is 1 dB lower than it would be for a perfectly linear system. (While this metric is sometimes called the 1 dB gain compression point, it is important to note that this is not the output level at which the incremental gain has fallen by 1 dB). As was shown in Figure 6, the output of the AD8330 limits quite abruptly, and the gain drops sharply above the clipping level. The output power, on the other hand, using an external resistive load, RL, continues to increase. In the most extreme case, the waveform changes from the sinusoidal form of the test signal, with an amplitude just below the clipping level, say, VCLIP, to a squarewave of precisely the same amplitude. The change in power over this range is from (VCLIP/2)2/RL to (VCLIP)2/RL, that is, a factor of 2, or 3 dB in power terms. It can be shown that for an ideal limiting amplifier, the 1 dB compression point occurs for an overdrive factor of 2 dB. For example, if the AD8330 is driving a 150 load and VMAG has been set to 2 V, the peak output is nominally 4 V (as noted above, the actual value when loaded may differ due to the mismatch between on-chip and external resistors), or 2.83 V rms for a sine wave output, which corresponds to a power of 53.3 mW, that is, 17.3 dBm in 150 . Thus, the P1dB level, at 2 dB above clipping, is 19.3 dBm. While not involving power transfer, it is sometimes useful to state the V1dB, which is the output voltage (unloaded or loaded) that is 2 dB above clipping for a sine waveform. In the above example, this voltage is still 2.83 V rms, which can be expressed as 9.04 dBV (0 dBV corresponds to a 1 V sine wave). Thus the V1dB is at 11.04 dBV. APPLICATIONS { { } } = RI VNOISE _ IN RS VNSD (22) using Equation 17. Thus, using the result Equation 19 for a source resistance of 1 k, having a noise-spectral density of 4.08 nV/Hz, we have N FAC = (1 k)(7.3 nV / (1 k)(4.08 nV / Hz ) = 1.79 Hz ) (23) Finally, converting this to decibels using N FIG = 10 log10 (N FAC ) (24) we find the noise figure in this case to be 5.06 dB, which is somewhat lower than the value shown in Figure 10 for this operating condition. Noise as a Function of V DBS The chief consequence of lowering the basic gain using VDBS is that the current noise spectral density INSD increases with the square root of the basic gain magnitude, GBN: INSD = 3 pA / Hz GBN (25) Thus, at the maximum basic gain of 316, INSD has risen to 53.3 pA/Hz, and if the noise figure using the procedures just explained is recalculated, we find it has risen to 17.2 dB. Distortion Considerations Continuously variable gain amplifiers invariably employ nonlinear circuit elements; consequently it is common for their distortion to be higher than well-designed fixed gain amplifiers. The translinear multiplier principles used in the AD8330 in principle yield extremely low distortion, a result of the fundamental linearization technique that is an inherent aspect of these circuits. In practice, however, the effect of device mismatches and junction resistances in the core cell, and other mechanisms in its supporting circuitry inevitably cause distortion, further aggravated by other effects in the later output stages. Some of these effects are very consistent from one sample to the next, while those due to mismatches (causing predominantly even-order distortion components) will be quite variable. Where the highest linearity (and also lowest noise) is demanded, consider using one of the X-AMP products such as the AD603 (single-channel), AD604 (dual-channel), or AD8332 (wideband dual-channel with ultralow noise LNAs). The AD8330's versatility, very constant ac response over a wide range of gains, large signal dynamic range, output swing, singlesupply operation, and low power consumption will commend this VGA to a diverse variety of applications. Only a few can be described here, including the most basic uses and some unusual ones. ADC Driving The AD8330 is well-suited to driving a high speed converter. There are now many available, but to illustrate the general features we will use one of the least expensive, the AD9214, which is available in three grades for operation at 65 MHz, 80 MHz, and 105 MHz; the AD9214BRS-80 is a good complement to the general capabilities of this VGA. -22- REV. B AD8330 VS, 3.3V 0.1 F 8k CHPF 0.1 F 3.3 OVERRANGE 3.3 0.1 F VPOS CNTR VPSO 10 ENBL VPSI OFST BIAS AND V-REF OR AVDD PWRDN DFS/GAIN DrVDD CM MODE AND OFFSET CONTROL 0.1 F INHI INPUT, 2V MAX INLO VGA CORE OUTPUT STAGES OPLO OPHI AIN AD9217BRS-80 AIN REFSENSE NC MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG NC ANALOG GROUND REF AGND CLK 0.1 F CLOCK DGND D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 GAIN BIAS, VDBS, 0V-1.5V DIGITAL GROUND Figure 20. Driving an Analog-to-Digital Converter (Preliminary) Figure 20 shows the connections. A 3.3 V supply is used for both parts. The ADC requires that its input pins be positioned at one third of the supply, or 1.1 V. Since the default output level of the VGA is one half the supply or 1.65 V, a small correction is introduced by the 8 k resistor from CNTR to ground. The ADC specifications require that the common-mode input be within 0.2 V of the nominal 1.1 V; variations of up to 20% in the AD8330's on-chip resistors will change this voltage by only 70 mV. With the connections shown, the AD9214 is able to receive an input of 2 V p-p; the peak output of the AD8330 can be reduced if desired by adding a resistor from VMAG to ground. An overrange condition is signaled by a HI state on pin OR of the AD9214. DFS/GAIN is unconnected in this example; this produces an offset-binary output. To provide a twos complement output, it should be connected to the REF pin. For ADCs running at sampling rates substantially below the bandwidth of the AD8330, an intervening noise filter is recommended to limit the noise bandwidth. A one-pole filter can easily be created with a single differential capacitor between OPHI and OPLO outputs. For a corner frequency of fc, the capacitor should have a value of CFILT = 1 / 942 fC For example a 10 MHz corner requires about 100 pF. Simple AGC Amplifier DATA OUTPUTS VS, 2.7V-6V 33nF 10 ENBL VPSI OFST BIAS AND V-REF VPOS CM MODE AND OFFSET CONTROL CNTR VPSO 4.7 0.1 F 0.1 F INHI INPUT, 5mV TO 1V rms INLO VGA CORE OUTPUT STAGES OPLO OPHI SEE TEXT OUTPUT, ~1V rms MODE VDBS GAIN INTERFACE CMGN OUTPUT CONTROL CMOP COMM VMAG NC Q2 Q1 0.1 F C1 0.1 F R1 100k 0.1 F Figure 21. Simple AGC Amplifier (Preliminary) (26) Figure 21 illustrates the use of the inverted gain mode and the offset gain range (0.2 V < VDBS < 1.7 V) in supporting a low cost AGC loop. Q1 is used as a detector. When OPHI is sufficiently higher than CNTR, due to the signal swing it conducts and charges C1. This raises VDBS and rapidly lowers the gain. Note that MODE is grounded (see Figure 5). The minimum voltage needed across R1 to set up the full gain is now 0.2 V, since CMGN is dc open circuited (this does not alter VMAG), while the maximum is 1.7 V. When the loop is settled, the average current in Q1 is VDBS/R1, which varies from 2 A at maximum gain (VDBS = 0.2 V) to 17 A at minimum gain (VDBS = 1.7 V). This change in Q1's current causes an increase of ~0.25 dB over the full gain range in the differential output of nominally 0.75 dBV at midrange (3.08 V p-p), corresponding to a 200:1 compression ratio. This is plotted in Figure 22 for a representative 100 kHz input. REV. B -23- AD8330 1.0 Wide Range True RMS Voltmeter 0.9 LEVELED OUTPUT - dBV 0.8 The AD8362 is an rms responding detector providing a dynamic range of 60 dB from low frequencies to 2.7 GHz. This may be increased to 110 dB using an AD8330 as a preconditioner, provided the noise bandwidth is limited by an interstage low-pass or band-pass filter. The VGA also provides an input port that is easier to drive than the 200 input of the AD8362. Figure 24 shows the general scheme. Both the AD8330 and AD8362 provide linear-in-decibel control interfaces. Thus when the output of the latter is used to control the gain of the former, this functional form is unaffected. The overall scaling is 33 mV/dB. Figure 25 shows the time domain response using a loop filter capacitor of 10 nF, for inputs ranging from 10 V to 1 V rms, that is, a 100 dB measurement range. 1.75 1.50 1.25 1.00 0.75 GAIN ERROR - dB 0.7 0.6 0.5 -50 -40 -30 -20 INPUT TO AD8330 - dBV -10 0 Figure 22. AGC Output vs. Input Amplitude (Simulation) The upper panel in Figure 23 shows time-domain output for fourteen 3 dB steps in input amplitude from 5.4 mV to 1.7 V. The waveforms in Figure 22 show the AGC voltage VDBS. This simple detector exhibits a temperature variation in the differential output amplitude of about 4 mV/C. It provides a fast attack time (an increase in the input is quickly leveled to the nominal output, due to the high peak currents in Q1) and a slow release time (a decrease in the input is not restored as quickly). The voltage at the VDBS pin may be used as an RSSI output, scaled 30 mV/dB. Note that the attack time can be halved by adding a second transistor as shown in the box (Figure 21). For operation at lower frequencies, the AGC hold capacitor must be increased. VDBS 0.50 0.25 0 3 2 1 0 -1 -2 -3 -4 0 10 20 30 40 50 60 70 80 90 OUTPUT 100 110 120 130 140 150 TIME - s Figure 23. Time Domain Waveforms (Simulation) (See Text) 5V 3.3 3.3 3.3 0.1 F 0.1 F 0.1 F 1 AD8362 COMM ACOM 16 0.1 F 10 F 2 3.6V ENBL VPS1 OFST VPOS CNTR VPSO CHPF VREF 15 3 DECL VTGT 14 INHI INPUT INLO OPHI 4 INHI VPOS 13 AD8330 OPLO CFLT 18nF 3.6V 5 INLO VOUT 12 VOUT MODE VDBS CMGN COMM CMOP VMAG 0.1 F 6 DECL VSET 11 7 PWDN ACOM 10 10 F 6.04k 8 COMM CLPF 9 4.02k Figure 24. Wide Range True RMS Voltmeter (Preliminary) -24- REV. B AD8330 4 3 OUTPUT - V 2 1 0 0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 TIME - ms 3.6 4.0 4.4 4.8 Figure 25. Time Domain Response of RMS Voltmeter (Simulation) REV. B -25- AD8330 OUTLINE DIMENSIONS 16-Lead Shrink Small Outline Package [QSOP] (RQ-16) Dimensions shown in inches 0.193 BSC 16 9 0.154 BSC 1 8 0.236 BSC PIN 1 0.065 0.049 0.069 0.053 0.010 0.025 0.004 BSC COPLANARITY 0.004 0.012 0.008 SEATING PLANE 0.010 0.006 8 0 0.050 0.016 COMPLIANT TO JEDEC STANDARDS MO-137AB 16-Lead Lead Frame Chip Scale Package [LFCSP] 3 mm 3 mm Body (CP-16-3) Dimensions shown in millimeters 0.50 0.40 0.30 PIN 1 INDICATOR 1.65 1.50 SQ* 1.35 3.00 BSC SQ 0.45 PIN 1 INDICATOR TOP VIEW 0.60 MAX 13 16 12 (BOTTOM VIEW) 1 EXPOSED PAD 2.75 BSC SQ 0.50 BSC 9 8 4 5 0.25 MIN NOTE: THE EXPOSED PAD IS NOT CONNECTED INTERNALLY. FOR INCREASED RELIABILITY OF THE SOLDER JOINTS AND MAXIMUM THERMAL CAPABILITY, IT IS RECOMMENDED THAT IT BE SOLDERED TO THE GROUND PLANE. 12 MAX 1.00 0.85 0.80 SEATING PLANE 0.30 0.23 0.18 0.80 MAX 0.65 TYP 0.05 MAX 0.02 NOM 0.20 REF 1.50 REF *COMPLIANT TO JEDEC STANDARDS MO-220-VEED-2 EXCEPT FOR EXPOSED PAD DIMENSION -26- REV. B AD8330 Revision History Location 10/04--Data Sheet changed from REV. A to REV. B. Page Changes to ABSOLUTE MAXIMUM RATINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Changes to ORDERING GUIDE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Change to TPC 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Note added to CP-16 Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4/03--Data Sheet changed from REV. 0 to REV. A. Update OUTLINE DIMENSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 REV. B -27- -28- C03217-0-10/04(B) |
Price & Availability of AD8330-EVAL
 |
|
|
|
|
All Rights Reserved © IC-ON-LINE 2003 - 2022 |
| [Add Bookmark] [Contact Us] [Link exchange] [Privacy policy] |
|
Mirror Sites : [www.datasheet.hk]
[www.maxim4u.com] [www.ic-on-line.cn]
[www.ic-on-line.com] [www.ic-on-line.net]
[www.alldatasheet.com.cn]
[www.gdcy.com]
[www.gdcy.net] |