CURRENT FEEDBACK AMPLIFIERS/DRIVERS High Output Drive 26 dBm Differential Line Drive for ADSL Transmitters 40 V p-p Differential Output Voltage, RL = 50 ⍀ @ 1 MHz 500 mA Continuous Current, R L = 5 ⍀ 1 A Peak Current, 1% Duty Cycle, RL = 15 ⍀ for DMT Low Distortion –68 dB @ 1 MHz THD, RL = 100 ⍀, V O = 40 V p-p High Speed 120 MHz Bandwidth (–3 dB) 1500 V/s Differential Slew Rate, VO = 10 V p-p, G = +5 70 ns Settling Time to 0.1% VOLTAGE FEEDBACK AMPLIFIERS/RECEIVERS High Input Performance 4 nV/√Hz Voltage Noise 15 mV Max Input Offset Voltage Low Distortion –68 dB @ 1 MHz THD, VO = 10 V p-p, RL = 200 ⍀ High Speed 100 MHz Bandwidth (–3 dB) 180 V/s Slew Rate High Output Drive 70 mA Output Current Drive APPLICATIONS ADSL, VDSL and HDSL Line Interface Driver and Receiver CRT Convergence and Astigmatism Adjustment Coil and Transformer Drivers Composite Audio Amplifiers PRODUCT DESCRIPTION RECEIVER B FUNCTIONAL BLOCK DIAGRAM AD816 TAB IS +VS DRIVER A & B FEATURES Flexible Configuration Two Low Noise Voltage Feedback Amplifiers with High Current Drive, Ideal for ADSL Receivers or Drivers for Low Impedance Loads such as CRT Coils Two High Current Drive Amplifiers, Ideal for an ADSL Differential Driver or Single Ended Drivers for Low Impedance Loads such as CRT Coils Thermal Overload Protection 15 9 +VS 8 –VS 7 6 5 4 3 NC OUT2 RECEIVER –IN2 RECEIVER +IN2 RECEIVER +IN2 DRIVER –IN2 DRIVER OUT2 DRIVER +VS –VS OUT1 DRIVER –IN1 DRIVER +IN1 DRIVER +IN1 RECEIVER 2 1 –IN1 RECEIVER OUT1 RECEIVER 14 13 12 11 B A RECEIVER A a 500 mA Differential Driver and Dual Low Noise (VF) Amplifiers AD816* 10 NC = NO CONNECT The two high output drive amplifiers are capable of supplying a minimum of 500 mA continuous output current and up to 1A peak output current, and when configured differentially, 40 V p-p differential output swing can be achieved on ± 15 V supplies into a load of 50 Ω. The drivers have 120 MHz of bandwidth and 1,500 V/µs of differential slew rate while featuring total harmonic distortion of –68 dB at 1 MHz into a 100 Ω load, specifications required for high frequency telecommunication subscriber line drivers. The low noise voltage feedback amplifiers are fully independent and can be configured differentially for use as receiver amplifiers within a subscriber line hybrid interface or individually for signal conditioning or filtering. The low noise of 4 nV/√Hz and distortion of –68 dB at 1 MHz enable low level signals to be resolved and amplified in the presence of large common-mode voltages. 100 MHz of bandwidth and 180 V/µs of slew rate combined with a load drive capability of 70 mA enable these amplifiers to drive passive filters and low inductance coils. The AD816 has thermal overload protection for system reliability and is available in low thermal resistance power packages. The AD816 operates over the industrial temperature range (–40°C to +85°C). The AD816 consists of two high current drive and two low noise amplifiers. These can be configured differentially for driving low impedance loads and receiving signals over twisted pair cable or could be used independently for single ended driving application such as correction circuits within high resolution CRT Monitors. 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 which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781/329-4700 World Wide Web Site: http://www.analog.com Fax: 781/326-8703 © Analog Devices, Inc., 1999 AD816–SPECIFICATIONS DRIVER AMPLIFIERS (@ T = +25ⴗC, V = ⴞ15 V dc, R = 1 k⍀ and R A Model DYNAMIC PERFORMANCE Small Signal Bandwidth (–3 dB) Bandwidth (0.1 dB) Differential Slew Rate Settling Time to 0.1% NOISE/HARMONIC PERFORMANCE Total Harmonic Distortion (Differential) Input Voltage Noise Input Current Noise (+I IN) Input Current Noise (–I IN) Differential Gain Error Differential Phase Error DC PERFORMANCE Input Offset Voltage S F LOAD = 50 ⍀ unless otherwise noted) Conditions AD816A Typ Max VS Min ± 15 100 120 MHz G = +2, RF = 499 Ω, VIN = 0.125 V rms, RL = 100 Ω G = +2, RF = 499 Ω, VIN = 0.125 V rms, RL = 100 Ω G = +2, RF = 499 Ω, VIN = 0.125 V rms, RL = 100 Ω VOUT = 10 V p-p, G = +5, RL = 100 Ω 10 V Step, G = +2 ±5 90 110 MHz ± 15 ± 15 ± 15 1400 10 1500 70 MHz V/µs ns f = 1 MHz, RLOAD = 100 Ω, VOUT = 40 V p-p f = 10 kHz, G = +2 (Single Ended) f = 10 kHz, G = +2 f = 10 kHz, G = +2 NTSC, G = +2, RLOAD = 25 Ω NTSC, G = +2, RLOAD = 25 Ω ± 15 ± 5, ± 15 ± 5, ± 15 ± 5, ± 15 ± 15 ± 15 –68 1.85 1.8 19 0.05 0.45 dBc nV/√Hz pA/√Hz pA/√Hz % Degrees ±5 ± 15 5 10 ± 5, ± 15 40 0.5 ± 5, ± 15 5 20 ± 5, ± 15 2 ± 5, ± 15 10 TMIN to TMAX Input Offset Voltage Drift Differential Offset Voltage TMIN to TMAX Differential Offset Voltage Drift –Input Bias Current TMIN to TMAX +Input Bias Current TMIN to TMAX Differential Input Bias Current Open-Loop Transresistance INPUT CHARACTERISTICS Differential Input Resistance TMIN to TMAX VOUT = ± 10 V, RL = 1 kΩ TMIN to TMAX ± 5, ± 15 0.7 0.6 ± 15 +Input –Input ± 15 ± 15 ±5 ± 5, ± 15 ± 5, ± 15 Differential Input Capacitance Input Common-Mode Voltage Range Common-Mode Rejection Ratio Differential Common-Mode Rejection Ratio OUTPUT CHARACTERISTICS Voltage Swing Units TMIN to TMAX TMIN to TMAX Single Ended, R LOAD = 25 Ω Continuous Output Current Differential, R LOAD = 50 Ω TMIN to TMAX RLOAD = 5 Ω Peak Output Current Short Circuit Current 10 µs Pulse, 1% Duty Cycle, RL = 15 Ω Note 1 ± 15 ±5 ± 15 ± 15 ± 15 ±5 ± 15 ± 15 56 80 23 2.2 46 45 500 200 2 12 15 25 2 5 60 100 5 5 50 50 mV mV mV µV/°C mV mV µV/°C µA µA µA µA µA µA MΩ MΩ 7 15 1.4 13.5 3.5 60 100 MΩ Ω pF ±V ±V dB dB 24.5 3.6 49 V p-p V p-p V p-p V p-p mA mA A A 750 100 1.0 1.0 NOTES 1 See Power Considerations section. Specifications subject to change without notice. –2– REV. B AD816 RECEIVER AMPLIFIERS (@ T = +25ⴗC, V A Model DYNAMIC PERFORMANCE Small Signal Bandwidth (–3 dB) Bandwidth (0.1 dB) Slew Rate Settling Time to 0.1% NOISE/HARMONIC PERFORMANCE Total Harmonic Distortion Input Voltage Noise Current Noise Differential Gain Error Differential Phase Error S = ⴞ15 V dc, RF = 1 k⍀ and RLOAD = 500 ⍀ unless otherwise noted) VS G = +2, RL = 100 Ω G = +2, RL = 100 Ω G = +2 G = +2 VOUT = 4 V p-p VOUT = 10 V p-p Step, G = +2 ± 15 ±5 ± 15 ±5 ± 15 ± 15 100 80 30 40 180 45 MHz MHz MHz MHz V/µs ns f = 1 MHz, RLOAD = 200 Ω f = 10 kHz f = 10 kHz NTSC, G = +2, RLOAD = 150 Ω ± 15 ± 5, ± 15 ± 5, ± 15 ± 15 ±5 ± 15 ±5 –68 4 2 0.04 0.05 0.03 0.06 dBc nV/√Hz pA/√Hz % % Degrees Degrees ± 5, ± 15 7.5 ± 5, ± 15 20 5 NTSC, G = +2, RLOAD = 150 Ω DC PERFORMANCE Input Offset Voltage Min AD816A Typ Max Conditions TMIN to TMAX Offset Voltage Drift Input Bias Current TMIN to TMAX Input Offset Current Offset Current Drift Open-Loop Gain ± 5, ± 15 VOUT = ± 7.5 V, RLOAD = 150 Ω TMIN to TMAX INPUT CHARACTERISTICS Input Resistance Input Capacitance Input Common-Mode Voltage Range Common-Mode Rejection Ratio OUTPUT CHARACTERISTICS Output Voltage Swing Output Current Short Circuit Current VCM = ± 5 V Single Ended, RLOAD = 150 Ω TMIN to TMAX Single Ended, RLOAD = 150 Ω TMIN to TMAX RL = 150 Ω ± 15 ± 15 3 1 ± 15 ± 15 ±5 ±5 ± 15 +13 –12 +3.8 –2.7 82 ± 15 ± 15 ±5 ±5 ± 15 ± 15 25.2 25.2 6.2 6.0 65 0.5 1 6 0.08 0.1 0.1 0.1 15 15 7 15 2 Units mV mV µV/°C µA µA µA nA/°C V/mV V/mV 300 1.5 +14.3 –13.4 +4.3 –3.4 110 kΩ pF V V V V dB 25.5 V p-p V p-p V p-p V p-p mA mA 6.4 70 105 Specifications subject to change without notice. (@ TA = +25ⴗC, VS = ⴞ15 V dc, RF = 1 k⍀ and RLOAD = 50 ⍀ (Driver), RLOAD = 500 ⍀ (Receiver) COMMON CHARACTERISTICS unless otherwise noted) Model Conditions VS MATCHING CHARACTERISTICS Crosstalk: Driver to Driver Drivers to Receivers Receiver to Receiver f = 1 MHz, VIN = 200 mV rms, RLOAD = 100 Ω ± 15 f = 1 MHz, VIN = 200 mV rms, RLOAD = 100 Ω ± 15 f = 1 MHz, VIN = 200 mV rms, RLOAD = 500 Ω ± 15 POWER SUPPLY Operating Range Quiescent Current Driver Supply Rejection Ratio Receiver Supply Rejection Ratio ± 15 ± 15 ± 15, ± 5 ± 15, ± 5 TMIN to TMAX TMIN to TMAX TMIN to TMAX Specifications subject to change without notice. REV. B –3– Min AD816A Typ Max –67 –64 –81 ±5 46 –49 –69 –66 –75 Units dB dB dB ± 18 56 59 V mA mA dB dB AD816 ABSOLUTE MAXIMUM RATINGS 1 MAXIMUM POWER DISSIPATION Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . ± 18 V Total Internal Power Dissipation2 Plastic (Y, YS and VR) . . 3.05 W (Observe Derating Curves) Input Voltage (Common Mode) . . . . . . . . . . . . . . . . . . . . ± VS Differential Input Voltage . . . . . . . . . . . . . . . . . . . . . . . . ± 6 V Output Short Circuit Duration . . . . . . . . . . . . . . . . . . . . . . Observe Power Derating Curves Storage Temperature Range Y, YS, VR Package . . . . . . . . . . . . . . . . . . –65°C to +125°C Operating Temperature Range AD816A . . . . . . . . . . . . . . . . . . . . . . . . . . . –40°C to +85°C Lead Temperature Range (Soldering, 10 sec) . . . . . . . +300°C The maximum power that can be safely dissipated by the AD816 is limited by the associated rise in junction temperature. The maximum safe junction temperature for the plastic encapsulated parts is determined by the glass transition temperature of the plastic, about 150°C. Exceeding this limit temporarily may cause a shift in parametric performance due to a change in the stresses exerted on the die by the package. Exceeding a junction temperature of 175°C for an extended period can result in device failure. The AD816 has thermal shutdown protection, which guarantees that the maximum junction temperature of the die remains below a safe level. However, shorting the output to ground or either power supply for an indeterminate period will result in device failure. To ensure proper operation, it is important to observe the derating curves and refer to the section on power considerations. NOTES 1 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. 2 Specification is for device in free air: 15-Lead Through Hole and Surface Mount: θJA = 41°C/W. It must also be noted that in high (noninverting) gain configurations (with low values of gain resistor), a high level of input overdrive can result in a large input error current, which may result in a significant power dissipation in the input stage. This power must be included when computing the junction temperature rise due to total internal power. PIN CONFIGURATION Y-15 VR-15, YS-15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 TOP VIEW OUT1 RECEIVER –IN1 RECEIVER +IN1 RECEIVER +IN1 DRIVER –IN1 DRIVER OUT1 DRIVER –VS +VS OUT2 DRIVER –IN2 DRIVER +IN2 DRIVER +IN2 RECEIVER –IN2 RECEIVER OUT2 RECEIVER NC OUT1 RECEIVER –IN1 RECEIVER +IN1 RECEIVER +IN1 DRIVER –IN1 DRIVER OUT1 DRIVER –VS +VS OUT2 DRIVER –IN2 DRIVER +IN2 DRIVER +IN2 RECEIVER –IN2 RECEIVER OUT2 RECEIVER NC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 TOP VIEW MAXIMUM POWER DISSIPATION – Watts 14 TJ = 1508C 13 θJA = 168C/W SOLDERED DOWN TO COPPER HEAT SINK AREA (STILL AIR = 0FT/MIN) 12 11 10 9 AD816 AVR, AY 8 7 6 5 4 θJA = 418C/W (STILL AIR = 0FT/MIN) NO HEAT SINK AD816 AVR, AY 3 2 1 0 –50 –40 –30 –20 –10 0 10 20 30 40 50 60 AMBIENT TEMPERATURE – 8C 70 80 90 Figure 1. Plot of Maximum Power Dissipation vs. Temperature (Copper Heat Sink Area = 2 in.2) ORDERING GUIDE Model Temperature Range Package Description Package Option AD816AY AD816AYS AD816AVR –40°C to +85°C –40°C to +85°C –40°C to +85°C 15-Lead Through-Hole SIP with Staggered Leads and 90° Lead Form 15-Lead Through-Hole SIP with Staggered Leads and Straight Lead Form 15-Lead Surface Mount DDPAK Y-15 YS-15 VR-15 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 AD816 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– WARNING! ESD SENSITIVE DEVICE REV. B Typical Driver Performance Characteristics–AD816 60 60 DIFFERENTIAL OUTPUT VOLTAGE – Volts p-p VS = 615V 25 50 20 40 15 30 10 VS = 65V 20 10 5 –IB, VS = 615V 50 INPUT BIAS CURRENT – mA SINGLE-ENDED OUTPUT VOLTAGE – Volts p-p 30 1 10 INPUT VOLTAGE NOISE 100 1k FREQUENCY – Hz 1 100k 10k Figure 3. Driver Input Current and Voltage Noise vs. Frequency –50 –70 PSRR – dB 50V 100V RL = 50V (DIFFERENTIAL) 400V RL = 200V (DIFFERENTIAL) –90 –100 –110 100 1k 10k 100k FREQUENCY – Hz 1M 10M Figure 6. Driver Total Harmonic Distortion vs. Frequency 80 VS = 615V G = +2 RL = 100V –30 –40 –50 100 –80 COMMON-MODE REJECTION – dB –20 80 VS = 615V G = +10 VOUT = 40V p-p –60 0 –10 0 20 40 60 JUNCTION TEMPERATURE – 8C –40 TOTAL HARMONIC DISTORTION – dBc NONINVERTING INPUT CURRENT NOISE –20 Figure 5. Driver Input Bias Current vs. Temperature CURRENT NOISE – pA/ Hz VOLTAGE NOISE – nV/ Hz 10 20 +IB, VS = 65V, 615V 100 10 –IB, VS = 65V 0 –40 Figure 2. Driver Output Voltage Swing vs. Load Resistance INVERTING INPUT CURRENT NOISE 30 10 0 0 10 100 1k 10k LOAD RESISTANCE – (Differential – V) (Single-Ended – V/2) 100 40 –PSRR +PSRR –60 –70 –80 70 VS = 615V 60 50 1kV 40 1kV 30 20 VOUT VIN 1kV 1kV –90 –100 0.01 0.1 1 10 FREQUENCY – MHz 100 Figure 4. Driver Power Supply Rejection vs. Frequency REV. B 10 10k 300 100k 1M FREQUENCY – Hz 10M 100M Figure 7. Driver Common-Mode Rejection vs. Frequency –5– AD816–Typical Driver Performance Characteristics 1400 G = +5 RL = 100V 2400 60 TA = +258C DIFFERENTIAL SR 1000 80 2000 1600 800 +SR 600 1200 400 800 200 400 VS = 65V 40 RTI OFFSET – mV 1200 2800 DIFFERENTIAL SLEW RATE – V/ms SINGLE-ENDED SLEW RATE – V/ms (PER AMPLIFIER) –SR 0 5 10 15 OUTPUT STEP SIZE – V p-p VS = 610V 0 VIN f = 0.1Hz 100V –20 SINGLE DRIVER 49.9V –40 0 0 VS = 615V 20 –60 –2.0 –1.6 –1.2 20 Figure 8. Driver Slew Rate vs. Output Step Size RL= 5V 1kV 1kV 0 –0.8 –0.4 0.4 0.8 LOAD CURRENT – Amps VOUT 1.2 1.6 2.0 Figure 11. Driver Thermal Nonlinearity vs. Output Current Drive 15 DIFFERENTIAL OUTPUT VOLTAGE – V p-p VS = 610V TA = +258C 10 VS = 615V RTI OFFSET – mV VS = 65V 5 0 VIN f = 0.1Hz 100V –5 SINGLE DRIVER 49.9V –10 1kV 1kV –15 –20 VOUT RL= 25V TA = +258C VS = 615V 40 RL = 100V 30 RL = 50V 20 RL = 25V 10 RL = 1V 0 –16 –12 –8 –4 0 4 VOUT – Volts 8 12 16 0 20 100 90 10 0% 5V 4 10 6 8 FREQUENCY – MHz 12 14 Figure 12. Driver Large Signal Frequency Response CLOSED-LOOP OUTPUT RESISTANCE – V Figure 9. Driver Gain Nonlinearity vs. Output Voltage 2 1ms 100 10 VS = 615V 1 0.1 0.01 30k Figure 10. Driver 40 V p-p Differential Sine Wave; RL = 50 Ω, f = 100 kHz VS = 65V 100k 300k 1M 3M 10M FREQUENCY – Hz 30M 100M 300M Figure 13. Driver Closed-Loop Output Resistance vs. Frequency –6– REV. B Typical Driver Characteristics–AD816 2 0.010 0.005 0.000 –0.005 PHASE –0.010 –0.015 GAIN –0.020 –0.025 –0.030 1 2 3 4 5 6 7 8 9 2 BACK TERMINATED LOADS (75V) 3 4 5 6 7 8 9 0.12 0.10 0.08 0.06 G = +2 0.04 RF = 1kV 0.02 NTSC 0.00 –0.02 –0.04 10 11 –3 VIN = 0.5Vrms –6 –9 –3 VIN = 0.25Vrms –12 –6 –9 –15 VIN = 125mVrms –18 –12 –21 –15 VIN = 62.5mVrms –24 –27 100k 1M –18 10M FREQUENCY – Hz 100M VIN = 200mVrms –10 INPUT 100V DRIVER A OUTPUT OUTPUT –30 499V 499V 0 50V 100V 100V 499V –40 –50 DRIVER A = INPUT DRIVER B = OUTPUT –60 –70 DRIVER B = INPUT DRIVER A = OUTPUT –80 –90 –100 10k 100k 1M 10M FREQUENCY – Hz 100M 300M –3 VIN = 0.5Vrms –6 –9 VIN = 0.25Vrms –12 –15 VIN = 125mVrms –21 –24 –27 100k VIN = 62.5mVrms 1M 10M FREQUENCY – Hz 100M 0 –3 –4 RF = 604V –5 –0.1 RF = 750V –0.2 –6 –0.3 –7 –0.4 100k 2 1 1M 10M FREQUENCY – Hz 100M –8 300M VIN = 200mVrms G +2 RL = 100V RS = 100V RF = 499V 0 –1 RF = 604V –2 –3 RF = 750V –4 –5 –6 –7 100k 300M Figure 16. Driver Small and Large Signal Frequency Response, G = +1 REV. B 0.1 NORMALIZED FREQUENCY RESPONSE – dB OUTPUT/INPUT LEVEL – dBV 0 –2 RF = 750V 3 G = +1 RF = 499V RL = 100V RS = 100V VIN = 1.0Vrms –1 RF = 604V Figure 18. Driver Frequency Response and Flatness, G = +5 Figure 15. Driver Output-to-Output Crosstalk vs. Frequency 3 VIN = 50mVrms G +5 RL = 100V RS = 100V 499V NORMALIZED FLATNESS – dB CROSSTALK – dB 50V 1 RF = 499V DRIVER 100V INPUT B NORMALIZED FREQUENCY RESPONSE – dB 2 0 –18 –21 300M Figure 17. Driver Small and Large Signal Frequency Response, G = +2 Figure 14. Driver Differential Gain and Differential Phase (Per Amplifier) –20 OUTPUT LEVEL – dBV 1 6 G = +2 RF = 499V 3 RL = 100V RS = 100V 0 0 INPUT LEVEL – dBV GAIN DIFF PHASE – Degrees PHASE 0.5 0.4 0.3 G = +2 RF = 1kV 0.2 0.1 NTSC 0.0 –0.1 –0.2 –0.3 10 11 DIFF PHASE – Degrees DIFF GAIN – % DIFF GAIN – % 6 BACK TERMINATED LOADS (25V) 0.04 0.03 0.02 0.01 0.00 –0.01 –0.02 –0.03 –0.04 1M 10M FREQUENCY – Hz 100M 300M Figure 19. Driver Frequency Response vs. RF, G = +2 –7– AD816–Typical Driver Performance Characteristics 499V 1kV 0.1mF 0.1mF 499V 1kV 8 VIN PULSE GENERATOR 100V 100V DRIVER A/B 0.1mF 7 TR/TF = 250ps 8 AD816 AD816 55V 10mF +15V 10mF +15V VIN RL = 100V PULSE GENERATOR 10mF TR/TF = 500ps –15V DRIVER A/B 7 0.1mF RL = 100V 10mF 50V –15V Figure 20. Test Circuit Gain = –1 Figure 24. Driver Test Circuit, Gain = +2 Figure 21. Driver 500 mV Step Response, G = –1 Figure 25. 10 V Step Response, G = +2 Figure 22. Driver 4 V Step Response, G = –1 Figure 26. Driver 400 mV Step Response, G = +2 RF 10mF +15V 0.1mF RG 8 AD816 VIN PULSE GENERATOR TR/TF = 250ps 100V DRIVER A/B 0.1mF 7 50V RL = 100V 10mF –15V Figure 27. Driver 20 V Step Response, G = +5 Figure 23. Test Circuit, Gain = 1 + RF/RG –8– REV. B Typical Receiver Performance Characteristics–AD816 –40 HARMONIC DISTORTION – dB 40 30 20 10 0 –50 –60 –70 –80 –90 3 10 100 1k 10k 100k FREQUENCY – Hz 1M –100 100 10M 10k 100k FREQUENCY – Hz 1M 10M 9 3 5 VIN = 1.0Vrms 1kV 4 VIN 1kV –3 VOUT 2 G = +2 RF = 1kV CF = 2.2pF RL = 100V RS = 0V 0 VIN = 0.5Vrms 100V 50V VS = 615V INPUT LEVEL – dBV 3 1k Figure 31. Receiver Harmonic Distortion vs. Frequency Figure 28. Receiver Input Voltage Noise Spectral Density GAIN – dB G = +5 VOUT = 14V p-p RF = 4kV RL = 1kV 1 0 –1 –2 VS = 65V –3 –6 3 0 –3 –9 –12 6 VIN = 0.25Vrms –6 –9 –15 VIN = 0.125Vrms –12 –18 –15 –21 OUTPUT LEVEL (RTO) – dBV INPUT VOLTAGE NOISE – nV/ Hz 50 VIN = 0.0625Vrms –4 –24 –5 100k –27 100k 1M 10M FREQUENCY – Hz 100M 300M Figure 29. Receiver Closed-Loop Gain vs. Frequency, Gain = –1 –18 1M 10M FREQUENCY – Hz 100M –21 300M Figure 32. Receiver Small and Large Signal Frequency Response, Gain = +2 100 100 90 80 80 PSR – dB CMR – dB 70 60 1kV VIN 40 1kV POSITIVE SUPPLY 60 NEGATIVE SUPPLY 50 40 VOUT 1kV 30 1kV 20 0 1k 10k 100k FREQUENCY – Hz 1M 10 100 10M Figure 30. Receiver Common-Mode Rejection vs. Frequency REV. B 1k 10k 100k 1M FREQUENCY – Hz 10M 100M Figure 33. Receiver Power Supply Rejection vs. Frequency –9– AD816–Typical Receiver Performance Characteristics 2.2pF 1kV 1kV 0.1mF 1kV 0.1mF VIN 1kV AD816 VIN VOUT 0.1mF AD816 50V VOUT REC A/B TR/ TF = 250ps RL 7 PULSE GENERATOR 8 PULSE GENERATOR 8 REC A/B 10mF +15V 10mF +15V 0.1mF RL = 500V 7 10mF 10mF 50V –15V –15V TR/ TF = 500ps Figure 38. Test Circuit, Gain = –1 Figure 34. Test Circuit, Gain = +2 50ns 5V Figure 39. Receiver 10 V Step Response, G = –1 Figure 35. Receiver 10 V Step Response, G = +2 50ns Figure 40. Receiver 400 mV Step Response, G = –1 Figure 36. Receiver 400 mV Step Response, G = +2 0 0 REC A INPUT –10 OUTPUT OUTPUT REC B CROSSTALK – dB 50V –30 –40 100V INPUT –20 1kV 100V 100V 1kV 1kV 2.2pF –70 2.2pF RECEIVER B : INPUT RECEIVER A : OUTPUT 2.2pF 2.2pF 499V OUTPUT OUTPUT 1kV 1kV REC B 50V 100V INPUT 50V DRV B 100V 100V INPUT DRIVER A: INPUT RECEIVER A: OUTPUT –60 –70 1kV 499V –40 –50 50V 1kV 100V 499V VIN = 200mVrms INPUT DRIVER B: INPUT RECEIVER A: OUTPUT DRIVER A: INPUT RECEIVER B: OUTPUT –80 –80 RECEIVER A = INPUT RECEIVER B = OUTPUT –90 –100 0.01 499V –30 1kV –50 –60 REC A OUTPUT OUTPUT 50V 50V CROSSTALK – dB –20 INPUT 100V DRV A VIN = 200mVrms –10 0.1 1 10 FREQUENCY – MHz –90 100 –100 0.01 300 DRIVER B: INPUT RECEIVER A: OUTPUT 0.1 1 10 FREQUENCY – MHz 100 300 Figure 41. Driver-to-Receiver Crosstalk vs. Frequency Figure 37. Receiver Output-to-Output Crosstalk vs. Frequency –10– REV. B AD816 THEORY OF OPERATION (DRIVER) Table I. Driver Resistor Values The AD816 driver is a dual current feedback amplifier with high (500 mA) output current capability. Being a current feedback amplifier, the AD816 driver’s open-loop behavior is expressed as transimpedance, ∆VO/∆I–IN, or TZ. The open-loop transimpedance behaves just as the open-loop voltage gain of a voltage feedback amplifier, that is, it has a large dc value and decreases at roughly 6 dB/octave in frequency. Since RIN is proportional to 1/gM, the equivalent voltage gain is just TZ × gM, where the gM in question is the transconductance of the input stage. Figure 42 shows the driver connected as a follower with gain. Basic analysis yields the following results: RF RG RIN = 1/gM ≈ 25 Ω G = 1+ RF RG VOUT ∞ 499 499 125 110 R R VOUT =VIO 1+ F ± I BN RN 1+ F ± I BI RF RG RG VIN Figure 42. Current-Feedback Amplifier Operation RF Recognizing that G × RIN << RF for low gains, it can be seen to the first order that bandwidth for this amplifier is independent of gain (G). I BI RG Considering that additional poles contribute excess phase at high frequencies, there is a minimum feedback resistance below which peaking or oscillation may result. This fact is used to determine the optimum feedback resistance, RF. In practice parasitic capacitance at the inverting input terminal will also add phase in the feedback loop so that picking an optimum value for RF can be difficult. Achieving and maintaining gain flatness of better than 0.1 dB at frequencies above 10 MHz requires careful consideration of several issues. Choice of Feedback and Gain Resistors The fine scale gain flatness will, to some extent, vary with feedback resistance. It is therefore recommended that once optimum resistor values have been determined, 1% tolerance values should be used if it is desired to maintain flatness over a wide range of production lots. Table I shows optimum values for several useful gain configurations. These should be used as a starting point in any application. REV. B 604 499 499 499 1k There are three major noise and offset terms to consider in a current feedback amplifier. For offset errors refer to the equation below. For noise error the terms are root-sum-squared to give a net output error. In the circuit below (Figure 43), they are input offset (VIO) which appears at the output multiplied by the noise gain of the circuit (1 + RF/RG), noninverting input current (IBN × RN) also multiplied by the noise gain, and the inverting input current, which when divided between RF and RG and subsequently multiplied by the noise gain always appear at the output as IBI × RF. The input voltage noise of the AD816 is less than 4 nV/√Hz. At low gains, however, the inverting input current noise times RF is the dominant noise source. Careful layout and device matching contribute to better offset and drift. The typical performance curves in conjunction with the equations below can be used to predict the performance of the AD816 in any application. where: RIN RG (⍀) DRIVER DC ERRORS AND NOISE T Z (S ) VO =G× VIN T Z (S ) + G × RIN + RF RN G = +1 –1 +2 +5 +10 RF (⍀) RN VIO I BN AD816 DRIVERS VOUT Figure 43. Driver Output Offset Voltage THEORY OF OPERATION (RECEIVER) Each AD816 receiver is a wide band high performance operational amplifier. It also provides a constant slew rate, bandwidth and settling time over its entire specified temperature range. The AD816 receiver consists of a degenerated NPN differential pair driving matched PNPs in a folded-cascode gain stage. The output buffer stage employs emitter followers in a class AB amplifier which deliver the necessary current to the load while maintaining low levels of distortion. A protection resistor in series with the noninverting input is required in circuits where the input to the receiver could be subject to transients on continuous overload voltages exceeding the ± 6 V maximum differential limit. The resistor provides protection for the input transistors, by limiting their maximum base current. –11– AD816 relationship between junction temperature (TJ) and various components of θJA. PRINTED CIRCUIT BOARD LAYOUT CONSIDERATIONS As to be expected for a wideband amplifier, PC board parasitics can affect the overall closed-loop performance. Of concern are stray capacitances at the output and the inverting input nodes. If a ground plane is to be used on the same side of the board as the signal traces, a space (5 mm min) should be left around the signal lines to minimize coupling. TJ = TA + PIN θJA DIE MOUNT) θ B (DIE MOUNT TA Adequate power supply bypassing can be critical when optimizing the performance of a high frequency circuit. Inductance in the power supply leads can form resonant circuits that produce peaking in the amplifier’s response. In addition, if large current transients must be delivered to the load, then bypass capacitors (typically greater than 1 µF) will be required to provide the best settling time and lowest distortion. A parallel combination of 10.0 µF and 0.1 µF is recommended. Under some low frequency applications, a bypass capacitance of greater than 10 µF may be necessary. Due to the large load currents delivered by the AD816, special consideration must be given to careful bypassing. The ground returns on both supply bypass capacitors as well as signal common must be “star” connected as shown in Figure 44. +VS RECEIVER A +IN IN DRIVER A RF RG (OPTIONAL) θ A (JUNCTION TO TJ POWER SUPPLY BYPASSING +OUT Equation 1 OUT RF RF RG TO CASE) θ A + θ B = θ JC CASE θ CA θ JC TJ θ JA PIN TA WHERE: PIN = DEVICE POWER DISSIPATION TA = AMBIENT TEMPERATURE TJ = JUNCTION TEMPERATURE θ JC = THERMAL RESISTANCE – JUNCTION TO CASE θ CA = THERMAL RESISTANCE – CASE TO AMBIENT Figure 45. A Breakdown of Various Package Thermal Resistances Figure 46 gives the relationship between output voltage swing into various loads and the power dissipated by the AD816 (PIN). This data is given for both sine wave and square wave (worst case) conditions. It should be noted that these graphs are for mostly resistive (phase < ±10°) loads. When the power dissipation requirements are known, Equation 1 and the graph on Figure 47 can be used to choose an appropriate heat sinking configuration. RF –OUT RG OUT DRIVER B IN –IN 4 RECEIVER B f = 1kHz VS = 615V RL = 50V SQUARE WAVE –VS PIN – Watts SINE WAVE Figure 44. Signal Ground Connected in “Star” Configuration 3 RL = 100V 2 POWER CONSIDERATIONS RL = 200V The 500 mA drive capability of the AD816 driver enables it to drive a 50 Ω load at 40 V p-p when it is configured as a differential driver. This implies a power dissipation, PIN, of nearly 5 watts. To ensure reliability, the junction temperature of the AD816 should be maintained at less than 175°C. For this reason, the AD816 will require some form of heat sinking in most applications. The thermal diagram of Figure 45 gives the basic 1 10 20 30 VOUT – Volts p-p 40 Figure 46. Total Power Dissipation vs Differential Driver Output Voltage –12– REV. B AD816 Normally, the AD816 will be soldered directly to a copper pad. Figure 47 plots θJA against size of copper pad. This data pertains to copper pads on both sides of G10 epoxy glass board connected together with a grid of feedthroughs on 5 mm centers. This data shows that loads of 100 ohms or greater will usually not require any more than this. This is a feature of the AD816’s 15-lead power SIP package. An important component of θJA is the thermal resistance of the package to heatsink. The data given is for a direct soldered connection of package to copper pad. The use of heatsink grease either with or without an insulating washer will increase this number. Several options now exist for dry thermal connections. These are available from Bergquist as part # SP600-90. Consult with the manufacturer of these products for details of their application. The AD816 is equipped with a thermal shutdown circuit. This circuit ensures that the temperature of the AD816 die remains below a safe level. In normal operation, the circuit shuts down the AD816 at approximately 180°C and allows the circuit to turn back on at approximately 140°C. This built-in hysteresis means that a sustained thermal overload will cycle between power-on and power-off conditions. The thermal cycling typically occurs at a rate of 1 ms to several seconds, depending on the power dissipation and the thermal time constants of the package and heat sinking. Figures 48 and 49 illustrate the thermal shutdown operation after driving OUT1 to the + rail, and OUT2 to the – rail, and then short-circuiting to ground each output of the AD816. The AD816 will not be damaged by momentary operation in this state, but the overload condition should be removed. COPPER HEAT SINK AREA (TOP AND BOTTOM) – in2 1 2 3 35 30 θJA – 8C/W AD816AVR, AY (θJC = 28C/W) 25 20 Figure 48. OUT2 Shorted to Ground Through a 2 Ω Resistor, Square Wave Is OUT1, RF = 1 k Ω, RG = 222 Ω 15 10 0 0.5k 1k 1.5k 2k 2.5k COPPER HEAT SINK AREA (TOP AND BOTTOM) – mm2 Figure 47. Power Package Thermal Resistance vs. Heat Sink Area Other Power Considerations There are additional power considerations applicable to the AD816. First, as with many current feedback amplifiers, there is an increase in supply current when delivering a large peak-to-peak voltage to a resistive load at high frequencies. This behavior is affected by the load present at the amplifier’s output. Figure 12 summarizes the full power response capabilities of the AD816 driver. These curves apply to the differential driver applications (right-hand side of Figure 52). In Figure 12, maximum continuous peak-to-peak output voltage is plotted vs. frequency for various resistive loads. Exceeding this value on a continuous basis can damage the AD816. REV. B Figure 49. OUT1 Shorted to Ground Through a 2 Ω Resistor, Square Wave Is OUT2, RF = 1 k Ω, RG = 222 Ω –13– AD816 APPLICATIONS ADSL Transceiver The AD816 is designed for the primary purpose of providing an integrated solution for the transmit and receive functions of an ADSL modem. ADSL or Asymmetrical Digital Subscriber Line is a means for delivering up to 6 Mbps from a telephone central office (CO) into a home over the conventional telephone twisted pair (local loop) and a few hundred kbps simultaneously in the opposite direction. The transmit/receive block is commonly referred to as a hybrid, which is an old telephone term, and the function was originally performed with passive circuitry in early phone systems. The hybrid’s function is to deliver maximum transmit power down the line, while providing the receive circuitry with a maximum receive signal and a minimized (self) transmit signal. As the line gets longer, this separation becomes much more difficult, because the transmit signal must be larger to reach the other end with acceptable SNR, while the receive signal is more attenuated by the longer line. The figure of merit for the performance of the hybrid is commonly called trans-hybrid loss and is a measure of how much the transmit signal that appears in the receive circuit has been attenuated relative to the amplitude of the transmit signal itself. It is measured in dBs and is a function of frequency. In addition to the passive circuits that have been used over time, active circuit techniques can enhance the hybrid’s performance. Figure 50 shows one of the various hybrid circuits that uses the AD816 in an ADSL application. The high power op amps serve as the transmitter, while the low noise amplifiers serve as the receiver. The power amplifiers of the AD816 (D1 and D2) are arranged in a differential configuration that receives its inputs from the differential outputs of a D/A converter. The outputs differentially drive the transformer primary with a turns ratio of 1:2. The line on the secondary side of the transformer has an impedance of 120 Ω. Thus one quarter of this resistance (30 Ω) is required for back termination on the primary side due to the impedance scaling by the square of the turns ratio. This resistance is divided in half (15 Ω) and put on each side of the drive buffers for symmetry (R101 and R201). The receive section (R1 and R2) is configured as a pair of difference amplifiers that together produce a differential output that consists of the receive signal in addition to the transmit signal attenuated by the trans-hybrid loss. The circuit is highly symmetrical, so a single-ended explanation can be easily generalized to understand the differential operation. D1 output terminals (Pin 6 of the AD816) drives the top of the primary of T1 through R101. A voltage divider is formed +15V 0.1mF V+ 5 10mF 8 4 R101 15V 6 D1 T1 XFRMR 1 2 C601 0.1mF TELEPHONE 4 715V 7 806V 715V 5 R201 15V 10 V– 11 D2 7 9 10 9 6 8 AD816 0.1mF R102 196V R202 196V 10mF –15V R103 196V R203 196V C201 8.2mF R204 1.18kV L201 12mH C101 8.2mF R104 1.18kV C602 0.1mF R106 348V R105 162V TWISTED PAIR 3 2 1 AD816 RCV OUT+ R1 R107 1kV L101 12mH R206 348V R205 162V R108 2.37kV 12 13 AD816 14 RCV OUT– R2 R207 1kV R208 2.37kV Figure 50. AD816 as an ADSL Transceiver –14– REV. B AD816 by R101 and all the downstream circuitry comprised of T1, the transmission line and its termination. For an ideal transformer, transmission line and termination, this will appear to be 15 Ω, and thus the signal appearing at Pins 1 and 2 of T1 will be the output of D1 divided by two in the ideal case. This signal is applied to the input of R1 (Receive 1 of the AD816) (Pin 3) via R105. Dual Composite Amplifier In some ADSL systems (DMT), there is a need to transmit higher crest factor signals. Typically this is done by increasing the turns ratio of T1 to as much as 4:1. In this case, R101 and R201 would be 3.75 Ω, and the peak current of the AD816 (1 A) would be the drive limit of the transmitter. The circuit in Figure 51 shows an example of such a circuit. It uses receiver amp R1 for the low noise first stage and driver D1 for the high output current second stage. Both local and overall feedback are used to get the desired response. A composite amplifier uses two different op amps together in a circuit to yield an overall performance that has some of the advantages of each op amp. In the case of the AD816, two composite amplifiers can be constructed that offer the low noise of the receiver amps in addition to the high current output of the driver amps. R1 is configured as a difference amplifier. The negative side (Pin 2) is driven by another signal that is a divided down version of the output of D1. This circuit is formed by R102 as one side of the voltage divider along with R103, C101, R104 and L101 as the other half of the divider. If the frequency dependent impedance part of this circuit matches the transformer, transmission line and termination impedance, then the signals applied to both sides of the difference-amp-configured R1 will be the same, and the transmit signal will be totally subtracted out by the circuit. In a real-world situation, it is not practical (or even possible) to subtract out all of the transmit signal (100% trans-hybrid loss), but only provide a first order cancellation which goes a long way toward reducing the dynamic range of the RCVOUT signal. The overall performance of this circuit depends on the ability to build a lumped element network that matches the impedance of the transmission line over the frequency range required for ADSL (≈ 20 kHz to 1.1 MHz). The circuits formed by D2 and R2 of the AD816 are totally symmetric with those formed by D1 and R1 and work in the same fashion. All the components in the D1, R1 circuits that are numbered with 100 range numbers are numbered with 200 range numbers in the D2, R2 circuits. 2 R1 4 3 VIN D1 6 VOUT 5 Figure 51. AD816 Composite Amplifier Creating Differential Signals If only a single-ended signal is available to drive the AD816 and a differential output signal is desired, a circuit can be used to perform the single-ended to differential conversion. The circuit shown in Figure 52 performs this function. It uses the AD816 with the gain of one receiver set at +1 and the gain of the other at –1. The 1 kΩ resistor across the input terminals of the follower makes the noise gain (NG = 2) equal to the inverter’s. The two receiver outputs then differentially drive the inputs to the AD816 driver with no common-mode signal to first order. The receive signal from the telephone line creates a differential signal across the primary of T1. There is, however, a two to one reduction in amplitude due to turns ratio of T1. This differential signal is applied to the + inputs (Pins 3 and 12) of R1 and R2. The receive amplifiers buffer this signal and present a differential output at Pins 1 and 14. There is no significant receive signal applied to the negative inputs of R1 and R2 due to the attenuating effects of R101 and R201 and the low output impedances of D1 and D2. +15V +15V 0.1mF 100V 4 3 1kV 8 AD816 1 5 6 RF 499V 1kV RG 100V RL 1kV 1kV 6 RF 499V 10 RECEIVER #2 AD816 4 7 DRIVER #2 0.1mF AD816 11 100V 9 7 0.1mF –15V –15V Figure 52. Differential Driver with Single-Ended Differential Converter REV. B –15– 10mF 8 DRIVER #1 AD816 2 5 Thus, the overall circuit provides first order cancellation of the transmit signal and differential buffering of the receive signal. 0.1mF RECEIVER #1 10mF AD816 OUTLINE DIMENSIONS Dimensions shown in inches and (mm). 0.600 (15.24) BSC 0.079 (2.006) DIA 2 PLACES 8° 0° 0.137 (3.479) TYP 1 15 0.798 (20.27) 0.778 (19.76) 0.182 (4.62) 0.172 (4.37) 0.182 (4.62) 0.172 (4.37) 0.031 (0.79) SEATING 0.024 (0.60) PLANE 0.666 ±0.006 (16.916 ±0.152) LONG LEAD 0.024 (0.61) 0.014 (0.36) 0.209 ±0.010 (5.308 ±0.254) SEATING PLANE 0.050 0.031 (0.79) (1.27) 0.024 (0.60) BSC 0.700 (17.78) BSC 15-Lead Through-Hole SIP with Staggered Leads and Straight Lead Form (YS-15) 0.063 (1.60) 0.057 (1.45) 0.394 (10.007) 0.137 (3.48) TYP 0.042 (1.07) TYP 1 15 0.516 (13.106) 0.110 0.152 (3.86) (2.79) BSC 0.148 (3.76) 0.694 (17.63) 0.684 (17.37) 0.426 (10.82) 0.416 (10.57) 0.100 (2.54) BSC 0.671 ±0.006 (17.043 ±0.152) SHORT 0.080 (2.03) LEAD 0.065 (1.65) 2 PLACES 0.079 (2.006) DIA 2 PLACES 0.042 (1.066) TYP PIN 1 0.024 (0.61) 0.014 (0.36) 0.798 (20.27) 0.778 (19.76) 0.063 (1.60) 0.057 (1.45) 0.394 (10.007) 0.080 (2.03) 0.065 (1.65) 2 PLACES PIN 1 0.700 (17.78) BSC 0.079 (2.007) DIA 2 PLACES 0.601 ±0.010 (15.265 0.710 (18.03) ±0.254) 0.690 (17.53) LONG LEAD 0.627 ±0.010 (15.926 ±0.254) SHORT LEAD 0.024 (0.61) 0.014 (0.36) 0.798 (20.27) 0.778 (19.76) 0.176 (4.47) 0.150 (3.81) 0.169 0.200 (4.29) (5.08) BSC BSC PRINTED IN U.S.A. PIN 1 0.080 (2.03) 0.065 (1.65) 2 PLACES 0.152 (3.86) 0.148 (3.76) 0.691 ±0.010 (17.551 ±0.254) 0.766 ±0.010 (19.456 ±0.254) 0.791 ±0.010 (20.091 ±0.254) 15 0.694 (17.63) 0.684 (17.37) 0.426 (10.82) 0.416 (10.57) 1 0.146 (3.70) 0.138 (3.50) 0.088 (2.24) 0.068 (1.72) 0.694 (17.63) 0.684 (17.37) 0.137 (3.479) 0.516 TYP (13.106) 0.042 (1.066) TYP 0.110 (2.79) BSC 0.516 (13.106) 0.063 (1.60) 0.057 (1.45) 0.394 (10.007) 0.426 (10.82) 0.416 (10.57) 0.110 (2.79) 0.152 (3.86) BSC 0.148 (3.76) C2191b–0–12/99 (rev. B) 15-Lead Through-Hole SIP with Staggered Leads and 90ⴗ Lead Form (Y-15) 15-Lead Surface Mount DDPAK (VR-15) 0.182 (4.62) 0.172 (4.37) SEATING PLANE 0.050 (1.27) BSC 0.031 (0.79) 0.024 (0.60) –16– REV. B