Features • • • • • • • • • Dual ADC with 8-bit Resolution 1 Gsps Sampling Rate per Channel, 2 Gsps in Interlaced Mode Single or 1:2 Demultiplexed Output LVDS Output Format (100Ω) 500 mVpp Analog Input (Differential Only) Differential or Single-ended 50Ω PECL/LVDS Compatible Clock Inputs Power Supply: 3.3V (Analog), 3.3V (Digital), 2.25V (Output) LQFP144 Package Temperature Range: – 0°C < TA < 70°C (Commercial Grade) – -40°C < TA < 85°C (Industrial Grade) • 3-wire Serial Interface – 16-bit Data, 3-bit Address – 1:2 or 1:1 Output Demultiplexer Ratio Selection – Full or Partial Standby Mode – Analog Gain (±1.5 dB) Digital Control – Input Clock Selection – Analog Input Switch Selection – Binary or Gray Logical Outputs – Synchronous Data Ready Reset – Data Ready Delay Adjustable on Both Channels – Interlacing Functions: Offset and Gain (Channel to Channel) Calibration Digital Fine SDA (Fine Sampling Delay Adjust) on One Channel – Internal Static or Dynamic Built-In Test (BIT) Dual 8-bit 1 Gsps ADC AT84AD001B Smart ADC™ Performance • • • • • • • • • Low Power Consumption: 0.7W Per Channel Power Consumption in Standby Mode: 120 mW 1.5 GHz Full Power Input Bandwidth (-3 dB) SNR = 42 dB Typ (6.8 ENOB), THD = -51 dBc, SFDR = -54 dBc at Fs = 1 Gsps Fin = 500 MHz 2-tone IMD3: -54 dBc (499 MHz, 501 MHz) at 1 Gsps DNL = 0.25 LSB, INL = 0.5 LSB Channel to Channel Input Offset Error: 0.5 LSB Max (After Calibration) Gain Matching (Channel to Channel): 0.5 LSB Max (After Calibration) Low Bit Error Rate (10-13) at 1 Gsps Application • • • • Instrumentation Satellite Receivers Direct RF Down Conversion WLAN 2153C–BDC–04/04 1 Description The AT84AD001B is a monolithic dual 8-bit analog-to-digital converter, offering low 1.4W power consumption and excellent digitizing accuracy. It integrates dual on-chip track/holds that provide an enhanced dynamic performance with a sampling rate of up to 1 Gsps and an input frequency bandwidth of over 1.5 GHz. The dual concept, the integrated demultiplexer and the easy interleaving mode make this device user-friendly for all dual channel applications, such as direct RF conversion or data acquisition. The smart function of the 3-wire serial interface eliminates the need for external components, which are usually necessary for gain and offset tuning and setting of other parameters, leading to space and power reduction as well as system flexibility. Functional Description The AT84AD001B is a dual 8-bit 1 Gsps ADC based on advanced high-speed BiCMOS technology. Each ADC includes a front-end analog multiplexer followed by a Sample and Hold (S/H), and an 8-bit flash-like architecture core analog-to-digital converter. The output data is followed by a switchable 1:1 or 1:2 demultiplexer and LVDS output buffers (100Ω). Two over-range bits are provided for adjustment of the external gain control on each channel. A 3-wire serial interface (3-bit address and 16-bit data) is included to provide several adjustments: • Analog input range adjustment (±1.5 dB) with 8-bit data control using a 3-wire bus interface (steps of 0.18 dB) • Analog input switch: both ADCs can convert the same analog input signal I or Q • Gray or binary encoder output. Output format: DMUX 1:1 or 1:2 with control of the output frequency on the data ready output signal • Partial or full standby on channel I or channel Q • Clock selection: • – Two independent clocks: CLKI and CLKQ – One master clock (CLKI) with the same phase for channel I and channel Q – One master clock but with two phases (CLKI for channel I and CLKIB for channel Q) ISA: Internal Settling Adjustment on channel I and channel Q • FiSDA: Fine Sampling Delay Adjustment on channel Q • Adjustable Data Ready Output Delay on both channels • Test mode: decimation mode (by 16), Built-In Test. A calibration phase is provided to set the two DC offsets of channel I and channel Q close to code 127.5 and calibrate the two gains to achieve a maximum difference of 0.5 LSB. The offset and gain error can also be set externally via the 3-wire serial interface. The AD84AD001B operates in fully differential mode from the analog inputs up to the digital outputs. The AD84AD001B features a full-power input bandwidth of 1.5 GHz. 2 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 1. Simplified Block Diagram CLKI Divider 2 to16 Clock Buffer DDRB DoirI + Vini S/H Vinib - 8bit ADC I DRDA I DMUX 1:2 or 1:1 I 8 Gain control I Calibration Gain/offset ISA I LVDS Clock Buffer LVDS Buffer I 2 16 DOAI DOAIN 16 DOBI DOBIN 2 Data BIT 3-wire Serial Interface 3WSI Gain control Q Calibration Gain/offset ISA Q & FiSDA + S/H Vinqb - Clock Ldn DMUX control Mode 2 DoirQ Vinq DOIRI DOIRIN DMUX control Input switch INPUT MUX CLKIO 8bit ADC Q 8 DMUX 1: 2 or 1: 1 Q LVDS buffe r Q DOIRQ DOIRQN 16 DOAQ DOAQN 16 DOBQ DOBQN CLKQ Clock Buffer DDRB Divider 2 to 16 DRDA Q LVDS Clock Buffer 2 CLKQO 3 2153C–BDC–04/04 Typical Applications Figure 2. Satellite Receiver Application Satellite Low Noise Converter (Connected to the Dish) Bandpass Amplifier Dish Satellite Tuner Low Pass Filter Bandpass Amplifier 11..12 GHz Tunable Band Filter IF Band Filter AGC 1..2 GHz Synthesizer 1.5 … 2.5 GHz Local oscillator I I I Local Oscillator Control Functions: AT84AD001B Clock and Carrier 90 Q Recovery... Q 0 Q Clock Q Quadrature Demodulation 4 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 3. Dual Channel Digital Oscilloscope Application DAC Gain A Channel A A Analog switch Channel B ADC B DAC Offset FISO RAM DAC Offset Display µP ADC A DAC Gain Channel Mode Selection Clock selection Timing circuit DACs Smart dual ADC DACs Table 1. Absolute Maximum Ratings Parameter Symbol Value Unit Analog positive supply voltage VCCA 3.6 V Digital positive supply voltage VCCD 3.6 V Output supply voltage VCCO 3.6 V VCCA to VCCD ± 0.8 V VCCO 1.6 V Analog input voltage VINI or VINIB VINQ or VINQB 1/-1 V Digital input voltage VD -0.3 to VCCD + 0.3 V Clock input voltage VCLK or VCLKB -0.3 to VCCD + 0.3 V VCLK - VCLKB -2 to 2 V Maximum junction temperature TJ 125 °C Storage temperature Tstg -65 to 150 °C Tleads 300 °C Maximum difference between VCCA and VCCD Minimum VCCO Maximum difference between VCLK and VCLKB Lead temperature (soldering 10s) Note: Absolute maximum ratings are limiting values (referenced to GND = 0V), to be applied individually, while other parameters are within specified operating conditions. Long exposure to maximum ratings may affect device reliability. 5 2153C–BDC–04/04 Table 2. Recommended Conditions of Use Parameter Symbol Comments Recommended Value Unit Analog supply voltage VCCA 3.3 V Digital supply voltage VCCD 3.3 V Output supply voltage VCCO 2.25 V VINi -VIniB or VINQ -VINQB 500 mVpp Vinclk 600 mVpp ISA -50 ps 0 < TA < 70 -40 < TA < 85 °C Differential analog input voltage (full-scale) Differential clock input level Internal Settling Adjustment (ISA) with a 3-wire serial interface for channel I and channel Q Operating temperature range TAmbient Commercial grade Industrial grade Electrical Operating Characteristics Unless otherwise specified: • VCCA = 3.3V; VCCD = 3.3V; VCCO = 2.25V • VINI - VINB or VINQ - VINQB = 500 mVpp full-scale differential input • LVDS digital outputs (100Ω) • TA (typical) = 25° C • Full temperature range: 0° C < TA < 70° C (commercial grade) or -40° C < TA < 85° C (industrial grade) Table 3. Electrical Operating Characteristics in Nominal Conditions Parameter Symbol Min Resolution Typ Max 8 Unit Bits Power Requirements Positive supply voltage - Analog - Digital Output digital (LVDS) and serial interface VCCA VCCD VCCO Supply current (typical conditions) - Analog - Digital - Output Supply current (1:2 DMUX mode) - Analog - Digital - Output 6 3.15 3.15 2.0 3.3 3.3 2.25 3.45 3.45 2.5 V V V ICCA ICCD ICCO 150 230 100 180 275 120 mA mA mA ICCA ICCD ICCO 150 260 175 180 310 210 mA mA AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 3. Electrical Operating Characteristics in Nominal Conditions (Continued) Parameter Typ Max ICCA ICCD ICCO 150 290 180 180 350 215 ICCA ICCD ICCO 80 160 55 95 190 65 mA mA mA Supply current (1 channel only, 1:2 DMUX mode) - Analog - Digital - Output ICCA ICCD ICCO 80 170 90 95 205 110 mA mA mA Supply current (full standby mode) - Analog - Digital - Output ICCA ICCD ICCO 12 24 3 17 34 5 mA mA mA Nominal dissipation (1 clock, 1:1 DMUX mode, 2 channels) PD 1.4 1.7 W Nominal dissipation (full standby mode) stbpd 120 Supply current (2 input clocks, 1:2 DMUX mode) - Analog - Digital - Output Supply current (1 channel only, 1:1 DMUX mode) - Analog - Digital - Output Symbol Min Unit mA mW Analog Inputs Full-scale differential analog input voltage VINi - VIniB or VINQ - VINQB Analog input capacitance I and Q CIN Full power input bandwidth (-3 dB) FPBW mV 450 500 550 mV 2 Gain flatness (-0.5 dB) pF 1.5 GHz 500 MHz Clock Input Logic compatibility for clock inputs and DDRB Reset (pins 124,125,126,127,128,129) PECL/LVDS clock inputs voltages (VCLKI/IN or VCLKQ/QN) Differential logical level PECL/ECL/LVDS VIL - VIH Clock input power level 600 -9 0 Clock input capacitance mV 6 2 dBm pF Digital Outputs Logic compatibility for digital outputs (depending on the value of VCCO) Differential output voltage swings (assuming VCCO = 2.25V) LVDS VOD 220 270 350 mV 7 2153C–BDC–04/04 Table 3. Electrical Operating Characteristics in Nominal Conditions (Continued) Parameter Symbol Min Typ Max Unit Output levels (assuming VCCO = 2.25V) 100Ω differentially terminated Logic 0 voltage Logic 1 voltage VOL VOH 1.0 1.25 1.1 1.35 1.2 1.45 V V Output offset voltage (assuming VCCO = 2.25V) 100Ω differentially terminated VOS 1125 1250 1325 mV Output impedance RO 50 W Output current (shorted output) 12 mA Output current (grounded output) 30 mA Output level drift with temperature 1.3 mV/°C Digital Input (Serial Interface) Maximum clock frequency (input clk) Fclk 50 MHz Input logical level 0 (clk, mode, data, ldn) -0.4 0 0.4 V Input logical level 1 (clk, mode, data, ldn) VCCO - 0.4 VCCO - 0.4 VCCO + 0.4 V Output logical level 0 (cal) -0.4 0 0.4 V Output logical level 1 (cal) VCCO - 0.4 VCCO VCCO + 0.4 V 15 pF Maximum output load (cal) Note: The gain setting is 0 dB, one clock input, no standby mode [full power mode], 1:1 DMUX, calibration off. Table 4. Electrical Operating Characteristics Parameter Symbol Min Typ Max Unit DC Accuracy No missing code Guaranteed over specified temperature range Differential non-linearity DNL 0.25 0.6 LSB Integral non-linearity INL 0.5 1 LSB Gain error (single channel I or Q) with calibration -0.5 0 0.5 LSB Input offset matching (single channel I or Q) with calibration -0.5 0 0.5 LSB 0.062 0.064 Gain error drift against temperature Gain error drift against VCCA Mean output offset code with calibration 127 LSB/°C LSB/mV 127.5 128 LSB BER 10-13 10-10 Error/ sample TS 170 Transient Performance Bit Error Rate Fs = 1 Gsps Fin = 250 MHz ADC settling time channel I or Q (between 10% - 90% of output response) VIni -ViniB = 500 mVpp Note: 8 ps Gain setting is 0 dB, two clock inputs, no standby mode [full power mode], 1:2 DMUX, calibration on. AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 5. AC Performances Parameter Symbol Min Typ Max Unit 42 44 dBc 40 42 dBc 41 dBc 7 7.2 Bits 6.5 6.8 Bits 6.2 Bits 48 54 dBc 45 51 dBc 42 dBc 50 56 dBc 48 54 dBc 43 dBc -54 dBc ±0.5 dB AC Performance Signal-to-noise Ratio Fs = 1 Gsps Fin = 20 MHz Fs = 1 Gsps Fin = 500 MHz Fs = 1 Gsps Fin = 1 GHz SNR Effective Number of Bits Fs = 1 Gsps Fin = 20 MHz Fs = 1 Gsps Fin = 500 MHz Fs = 1 Gsps Fin = 1 GHz ENOB Total Harmonic Distortion (First 9 Harmonics) Fs = 1 Gsps Fin = 20 MHz Fs = 1 Gsps Fin = 500 MHz Fs = 1 Gsps Fin = 1 GHz |THD| Spurious Free Dynamic Range Fs = 1 Gsps Fin = 20 MHz Fs = 1 Gsps Fin = 500 MHz Fs = 1 Gsps Fin = 1 GHz |SFDR| Two-tone Inter-modulation Distortion (Single Channel) FIN1 = 499 MHz , FIN2 = 501 MHz at Fs = 1 Gsps IMD Band flatness from DC up to 600 MHz Phase matching using auto-calibration and FiSDA in interlace mode (channel I and Q) Fin = 250 MHz Fs = 1 Gsps dϕ Crosstalk channel I versus channel Q Fin = 250 MHz, Fs = 1 Gsps(2) Cr Notes: -0.7 0 0.7 -55 ° dB 1. Differential input [-1 dBFS analog input level], gain setting is 0 dB, two input clock signals, no standby mode, 1:1 DMUX, ISA = -50 ps. 2. Measured on the AT84AD001TD-EB Evaluation Board. 9 2153C–BDC–04/04 Table 6. AC Performances in Interlace Mode Parameter Symbol Min Typ Max Unit Maximum equivalent clock frequency Fint = 2 x Fs Where Fs = external clock frequency Fint 2 Minimum clock frequency Fint 20 Msps Differential non-linearity in interlace mode intDNL 0.25 LSB Integral non-linearity in interlace mode intINL 0.5 LSB 42 dBc 40 dBc 7.1 Bits 6.8 Bits 52 dBc 49 dBc 54 dBc 52 dBc -54 dBc Interlace Mode Gsps Signal-to-noise Ratio in Interlace Mode Fint = 2 Gsps Fin = 20 MHz iSNR Fint = 2 Gsps Fin = 250 MHz Effective Number of Bits in Interlace Mode Fint = 2 Gsps Fin = 20 MHz iENOB Fint = 2 Gsps Fin = 250 MHz Total Harmonic Distortion in Interlace Mode Fint = 2 Gsps Fin = 20 MHz |iTHD| Fint = 2 Gsps Fin = 250 MHz Spurious Free Dynamic Range in Interlace Mode Fint = 2 Gsps Fin = 20 MHz Fint = 2 Gsps Fin = 250 MHz |iSFDR| Two-tone Inter-modulation Distortion (Single Channel) in Interlace Mode FIN1 = 249 MHz , FIN2 = 251 MHz at Fint = 2 Gsps Note: 10 iIMD One analog input on both cores, clock I samples the analog input on the rising and falling edges. The calibration phase is necessary. The gain setting is 0 dB, one input clock I, no standby mode, 1:1 DMUX, FiSDA adjustment. AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 7. Switching Performances Parameter Symbol Min Typ Max Unit Switching Performance and Characteristics - See “Timing Diagrams” on page 12. Maximum operating clock frequency Maximum operating clock frequency in BIT and decimation modes FS 1 Gsps FS (BIT, DEC) 750 Minimum clock frequency (no transparent mode) Minimum clock frequency (with transparent mode) FS Msps 10 Msps 1 Ksps Minimum clock pulse width [high] (No transparent mode) TC1 0.4 0.5 50 ns Minimum clock pulse width [low] (No transparent mode) TC2 0.4 0.5 50 ns Aperture delay: nominal mode with ISA & FiSDA TA 1 ns Aperture uncertainty Jitter 0.4 ps (rms) Data output delay between input clock and data TDO 3.8 ns Data Ready Output Delay TDR 3 ns TRDR 2 ns TD2 1/2 Fs +Tdrda ps Tdrda range -560 to 420 ps Data Ready Reset to Data Ready Data Output Delay with Data Ready Data Ready (CLKO) Delay Adjust (140 ps steps) Output skew 50 100 ps Output rise/fall time for DATA (20% - 80%) TR/TF 300 350 500 ps Output rise/fall time for DATA READY (20% - 80%) TR/TF 300 350 500 ps 3 (port B) 3.5 (port A, 1:1 DMUX mode) 4 (port A, 1:2 DMUX mode) Data pipeline delay (nominal mode) TPD Data pipeline delay (nominal mode) in S/H transparent mode DDRB recommended pulse width Clock cycles 2.5 (port B) 3 (port A, 1:1 DMUX mode) 3.5 (port A, 1:2 DMUX mode) 1 ns 11 2153C–BDC–04/04 Timing Diagrams Figure 4. Timing Diagram, ADC I or ADC Q, 1:2 DMUX Mode, Clock I for ADC I, Clock Q for ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 1 1 X X 1 X 0 0 TA N+3 N+1 VIN N+2 N CLKI or CLKQ Pipeline delay = 4 clock cycles DOIA[0:7] or DOQA[0:7] N-4 TDO Pipeline delay = 3 clock cycles DOIB[0:7] or DOQB[0:7] N N - 2 N-3 TDO N-1 N +1 Programmable delay TD2 CLKOI or CLKOQ (= CLKI/2) CLKOI or CLKOQ (= CLKI/4) Figure 5. 1:1 DMUX Mode, Clock I = ADC I, Clock Q = ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 1 1 X X 0 X 0 0 TA N+3 N+1 VIN N+2 N CLKI or CLKQ Pipeline delay = 3.5 clock cycles DOIA[0:7] or DOQA[0:7] N -3 N -2 TDO N - 1 N N+1 CLKOI or CLKOQ DOIB[0:7] and DOQB[0:7] are high impedance 12 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 6. 1:2 DMUX Mode, Clock I = ADC I, Clock I = ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 1 0 X X 1 X 0 0 TA N+3 N+2 N+1 VIN N CLKI TDO Pipeline delay = 4 clock cycles DOIA[0:7] NI - 4 NI NI - 2 Pipeline delay = 3 clock cycles TDO DOIB[0:7] NI - 3 NI - 1 NI +1 DOQA[0:7] NQ - 4 NQ - 2 NQ DOQB[0:7] NQ - 3 NQ - 1 NQ +1 TD2 CLKOI (= CLKI/2) CLKOI (= CLKI/4) CLKOQ is high impedance 13 2153C–BDC–04/04 Figure 7. 1:1 DMUX Mode, Clock I = ADC I, Clock I = ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 1 0 X X 0 X 0 0 TA N+3 N+1 VIN N+2 N CLKI TDO Pipeline delay = 3.5 clock cycles DOIA[0:7] N -3 N -2 N - 1 N N+1 DOQA[0:7] N -3 N -2 N - 1 N N+1 CLKOI DOIB[0:7] and DOQB[0:7] are high impedance CLKOQ is high impedance 14 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 8. 1:2 DMUX Mode, Clock I = ADC I, Clock IN = ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 0 X X X 1 X 0 0 N+6 N+4 N+2 TA N+5 VIN N N+1 N+3 CLKI CLKIN TDO Pipeline delay = 4 clock cycles DOQA[0:7] N -8 Pipeline delay = 3 clock cycles DOQB[0:7] N N- 4 N -6 TDO N -2 Pipeline delay = 3.5 clock cycles N+2 TDO DOIA[0:7] N -7 N -3 N+1 DOIB[0:7] N -5 N -1 N+3 TD2 CLKOI (= CLKI/2) CLKOI (= CLKI/4) CLKOQ is high impedance 15 2153C–BDC–04/04 Figure 9. 1:1 DMUX Mode, Clock I = ADC I, Clock IN = ADC Q Address: D7 D6 D5 D4 D3 D2 D1 D0 0 X X X 0 X 0 0 N+6 N+4 N+2 TA N+5 VIN N N+3 N+1 CLKI CLKIN TDO Pipeline delay = 3.5 clock cycles DOQA[0:7] N -6 N -4 N -5 N+2 N +1 N+3 TDO Pipeline delay = 3 clock cycles DOIA[0:7] N N - 2 N -3 N - 1 CLKOI (= CLKI/2) DOIB[0:7] and DOQB[0:7] are high impedance CLKOQ is high impedance Figure 10. 1:1 DMUX Mode, Decimation Mode Test (1:16 Factor) Address: D7 D6 D5 D4 D3 D2 D1 D0 1 0 X X 0 X 0 0 N - 16 VIN N N + 16 N + 32 16 clock cycles CLKI DOIA[0:7] N - 16 N N + 16 N + 32 N + 48 DOQA[0:7] N - 16 N N + 16 N + 32 N + 48 CLKOI DOIB[0:7] and DOQB[0:7] are high impedance CLKOQ is high impedance Notes: 16 1. The maximum clock input frequency in decimation mode is 750 Msps. 2. Frequency(CLKOI) = Frequency(Data) = Frequency(CLKI)/16. AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 11. Data Ready Reset 500 ps CLKI or CLKQ 500 ps 1 ns min DDRB FORBIDDEN FORBIDDEN ALLOWED ALLOWED Figure 12. Data Ready Reset 1:1 DMUX Mode TA N VIN N+1 Clock in Reset CLKI or CLKQ Pipeline Delay + TDO DOIA[0:7] or DOQA[0:7] N TDR CLKOI or CLKOQ TDR 2 ns DDRB 1 ns min Note: The Data Ready Reset is taken into account only 2 ns after it is asserted. The output clock first completes its cycle (if the reset occurs when it is high, it goes low only when its half cycle is complete; if the reset occurs when it is low, it remains low) and then only, remains in reset state (frozen to a low level in 1:1 DMUX mode). The next falling edge of the input clock after reset makes the output clock return to normal mode (after TDR). 17 2153C–BDC–04/04 Figure 13. Data Ready Reset 1:2 DMUX Mode TA N VIN N+1 Clock in Reset CLKI or CLKQ Pipeline Delay + TDO DOIA[0:7] or DOQA[0:7] N DOIB[0:7] or DOQB[0:7] N+1 TDR TDR CLKOI or CLKOQ (= CLKI/2) TDR + 2 cycles CLKOI or CLKOQ (= CLKI/4) TDR + 2 cycles 2 ns DDRB 1 ns min Notes: 1. In 1:2 DMUX, Fs/2 mode: The Data Ready Reset is taken into account only 2 ns after it is asserted. The output clock first completes its cycle (if the reset occurs when it is low, it goes high only when its half cycle is complete; if the reset occurs when it is high, it remains high) and then only, remains in reset state (frozen to a high level in 1:2 DMUX Fs/2 mode). The next rising edge of the input clock after reset makes the output clock return to normal mode (after TDR). 2. In 1:2 DMUX, Fs/4 mode: The Data Ready Reset is taken into account only 2 ns after it is asserted. The output clock first completes its cycle (if the reset occurs when it is high, it goes low only when its half cycle is complete; if the reset occurs when it is low, it remains low) and then only, remains in reset state (frozen to a low level in 1:2 DMUX Fs/4 mode). The next rising edge of the input clock after reset makes the output clock return to normal mode (after TDR). 18 AT84AD001B 2153C–BDC–04/04 AT84AD001B Functions Description Table 8. Description of Functions Name Function VCCA Positive analog power supply VCCD Positive digital power supply VCCO Positive output power supply GNDA Analog ground GNDD Digital ground GNDO Output ground VINI, VINIB Differential analog inputs I VINQ, VINQB Differential analog inputs Q CLKOI, CLKOIN, CLKOQ, CLKOQN Differential output data ready I and Q VCCA = 3.3V VCCD = 3.3V VCCO = 2.25V VINI 32 D0AI0 D0AI0N D0BI0 D0BI0N DOAI7 DOAI7N DOBI7 DOBI7N 32 D0AQ0 D0AQ0 DOAQ7 DOAQ7 VINIB CLKI, CLKIN, CLKQ, CLKQN Differential clock inputs I and Q DDRB, DDRBN Synchronous data ready reset I and Q Mode Bit selection for 3-wire bus or nominal setting Clk Input clock for 3-wire bus interface Data Input data for 3-wire bus Ldn Beginning and end of register line for 3-wire bus interface <D0AI0:DOAI7> <D0AI0N:DOAI7N> <D0BI0:DOBI7> <D0BI0N:DOBI7N> Differential output data port channel I <D0AQ0:DOAQ7> <D0AQ0N:DOAQ7N> <D0BQ0:DOBQ7> <D0BQ0N:DOBQ7N> Differential output data port channel Q VINQ DOBQ0 DOQBQ7 DOBQ0N DOQBQ7N VINQB AT84AD001B CLKI 4 DOIRI, DOIRIN DOIRQ, DOIRQN CLKQ 4 CLOCKOI, CLOCKOIB CLOCKOQ, CLOCKOQB CLKQB 2 VtestI VtestQ CLKIB Vdiode GNDA GNDD GNDO mode clk data ldn DOIRI, DOIRIN DOIRQ, DOIRQN Differential output IN range data I and Q VtestQ Test voltage output for ADC Q (to be left open) VtestI Test voltage output for ADC I (to be left open) Cal Output bit status internal calibration Vdiode Test diode voltage for Tj measurement 19 2153C–BDC–04/04 Digital Output Coding (Nominal Settings) Table 9. Digital Output Coding (Nominal Setting) Differential Analog Input Voltage Level Digital Output I or Q (Binary Coding) Out-of-range Bit > 250 mV > Positive full-scale + 1/2 LSB 11111111 1 250 mV 248 mV Positive full-scale + 1/2 LSB Positive full-scale - 1/2 LSB 11111111 11111110 0 0 1 mV -1 mV Bipolar zero + 1/2 LSB Bipolar zero - 1/2 LSB 10000000 01111111 0 0 -248 mV -250 mV Negative full-scale + 1/2 LSB Negative full-scale - 1/2 LSB 00000001 00000000 0 0 < -250 mV < Negative full-scale - 1/2 LSB 00000000 1 Pin Description Table 10. AT84AD001B LQFP 144 Pin Description Symbol Pin number Function GNDA, GNDD, GNDO 10, 12, 22, 24, 36, 38, 40, 42, 44, 46, 51, 54, 59, 61, 63, 65, 67, 69, 85, 87, 97, 99, 109, 111, 130, 142, 144 Ground pins. To be connected to external ground plane VCCA 41, 43, 45, 60, 62, 64 Analog positive supply: 3.3V typical VCCD 9, 21, 37, 39, 66, 68, 88, 100, 112, 123, 141 3.3V digital supply VCCO 11, 23, 86, 98, 110, 143 2.25V output and 3-wire serial interface supply VINI 57, 58 In-phase (+) analog input signal of the sample & hold differential preamplifier channel I VINIB 55, 56 Inverted phase (-) of analog input signal (VINI) VINQ 47, 48 In-phase (+) analog input signal of the sample & hold differential preamplifier channel Q VINQB 49, 50 Inverted phase (-) of analog input signal (VINQ) CLKI 124 In-phase (+) clock input signal CLKIN 125 Inverted phase (-) clock input signal (CLKI) CLKQ 129 In-phase (+) clock input signal 20 AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 10. AT84AD001B LQFP 144 Pin Description (Continued) Symbol Pin number Function CLKQN 128 Inverted phase (-) clock input signal (CLKQ) DDRB 126 Synchronous data ready reset I and Q DDRBN 127 Inverted phase (-) of input signal (DDRB) DOAI0, DOAI1, DOAI2, DOAI3, DOAI4, DOAI5, DOAI6, DOAI7 117, 113, 105, 101, 93, 89, 81, 77 In-phase (+) digital outputs first phase demultiplexer (channel I) DOAI0 is the LSB. D0AI7 is the MSB DOAI0N, DOAI1N, DOAI2N, DOAI3N, DOAI4N, DOAI5N, DOAI6N, DOAI7N, 118, 114, 106, 102, 94, 90, 82, 78 Inverted phase (-) digital outputs first phase demultiplexer (channel I) DOAI0N is the LSB. D0AI7N is the MSB DOBI0, DOBI1, DOBI2, DOBI3, DOBI4, DOBI5, DOBI6, DOBI7 119, 115, 107, 103, 95, 91, 83, 79 In-phase (+) digital outputs second phase demultiplexer (channel I) DOBI0 is the LSB. D0BI7 is the MSB DOBI0N, DOBI1N, DOBI2N, DOBI3N, DOBI4N, DOBI5N, DOBI6N, DOBI7N 120, 116, 108, 104, 96, 92, 84, 80 Inverted phase (-) digital outputs second phase demultiplexer (channel I) DOBI0N is the LSB. D0BI7N is the MSB DOAQ0, DOAQ1, DOAQ2, DOAQ3, DOAQ4, DOAQ5, DOAQ6, DOAQ7 136, 140, 4, 8, 16, 20, 28, 32 In-phase (+) digital outputs first phase demultiplexer (channel Q) DOAI0 is the LSB. D0AQ7 is the MSB DOAQ0N, DOAQ1N, DOAQ2N, DOAQ3N, DOAQ4N, DOAQ5N, DOAQ6N, DOAQ7N 135, 139, 3, 7, 15, 19, 27, 31 Inverted phase (-) digital outputs first phase demultiplexer (channel Q) DOAI0N is the LSB. D0AQ7N is the MSB DOBQ0, DOBQ1, DOBQ2, DOBQ3, DOBQ4, DOBQ5, DOBQ6, DOBQ7 134, 138, 2, 6, 14, 18, 26, 30 In-phase (+) digital outputs second phase demultiplexer (channel Q) DOBQ0 is the LSB. D0BQ7 is the MSB DOBQ0N, DOBQ1N, DOBQ2N, DOBQ3N, DOBQ4N, DOBQ5N, DOBQ6N, DOBQ7N 133, 137, 1 ,5, 13, 17, 25, 29 Inverted phase (-) digital outputs second phase demultiplexer (channel Q) DOBQ0N is the LSB. D0BQ7N is the MSB DOIRI 75 In-phase (+) out-of-range bit input (I phase) combined demultiplexer out-of-range is high on the leading edge of code 0 and code 256 DOIRIN 76 Inverted phase of output signal DOIRI DOIRQ 34 In-phase (+) out-of-range bit input (Q phase) combined demultiplexer out-of-range is high on the leading edge of code 0 and code 256 DOIRQN 33 Inverted phase of output signal DOIRQ MODE 74 Bit selection for 3-wire bus interface or nominal setting CLK 73 Input clock for 3-wire bus interface DATA 72 Input data for 3-wire bus LND 71 Beginning and end of register line for 3- wire bus interface CLKOI 121 Output clock in-phase (+) channel I 21 2153C–BDC–04/04 Table 10. AT84AD001B LQFP 144 Pin Description (Continued) Symbol Pin number Function CLKOIN 122 Inverted phase (-) output clock channel I CLKOQ 132 Output clock in-phase (+) channel Q, 1/2 input clock frequency CLKOQN 131 Inverted phase (-) output clock channel Q VtestQ, VtestI 52, 53 Pins for internal test (to be left open) Cal 70 Calibration output bit status Vdiode 35 Positive node of diode used for die junction temperature measurements Figure 14. AT84AD001B Pinout (Top View) LQFP 144 20 by 20 by 1.4 mm Atmel - Dual 8-bit 22 AT84AD001B 2153C–BDC–04/04 AT84AD001B Typical Characterization Results Nominal conditions (unless otherwise specified): Typical Full Power Input Bandwidth • VCCA = 3.3V; VCCD = 3.3V; VCCO = 2.25V • VINI - VINB or VINQ to VINQB = 500 mVpp full-scale differential input • LVDS digital outputs (100Ω) • TA (typical) = 25° C • Full temperature range: 0°C < TA < 70°C (commercial grade) or -40°C < TA < 85° C (industrial grade) • Fs = 500 Msps • Pclock = 0 dBm • Pin = -1 dBFS • Gain flatness (±0.5 dB) from DC to > 500 MHz • Full power input bandwidth at -3 dB > 1.5 GHz Figure 15. Full Power Input Bandwidth 0 -1 -2 -3 dB Bandwidth -3 dBFS -4 -5 -6 -7 -8 -9 -10 -11 100 300 500 700 900 1100 1300 1500 1700 1900 2100 2300 2500 2700 2900 Fin (MHz) 23 2153C–BDC–04/04 Typical Crosstalk Figure 16. Crosstalk (Fs = 500 Msps) 80 70 60 dBc 50 40 30 20 10 0 0 100 200 300 400 500 600 700 800 900 1000 Fin (MHz) Note: Typical DC, INL and DNL Patterns Measured on the AT84AD001TD-EB Evaluation Board. 1:2 DMUX mode, Fs/4 DR type Figure 17. Typical INL (Fs = 50 Msps, Fin = 1 MHz, Saturated Input) 0,6 0,4 INL (Lsb) 0,2 0 -0,2 -0,4 -0,6 1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226 241 256 Codes 24 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 18. Typical DNL (Fs = 50 Msps, Fin = 1 MHz, Saturated Input) 0,3 0,2 DNL (Lsb) 0,1 0 -0,1 -0,2 -0,3 1 16 31 46 61 76 91 106 121 136 151 166 181 196 211 226 241 256 Codes Typical Step Response Figure 19. Step Response 250 Codes 200 150 100 50 0 2.4E-12 1.3E-09 2.5E-09 3.8E-09 5.0E-09 6.3E-09 7.5E-09 8.8E-09 Time (s) Channel IA • Fs = 1 Gsps • Pclock = 0 dBm • Fin = 100 MHz • Pin = -1 dBFS Channel QA 25 2153C–BDC–04/04 Figure 20. Step Response (Zoom) 250 200 Codes 90% 150 Tr = 160 ps 100 50 10% 0 4.9E-09 6.1E-09 Channel IA • Fs = 1 Gsps • Pclock = 0 dBm • Fin = 500 MHz • Pin = -1 dBFS 7.4E-09 Time (s) Channel QA Figure 21. Step Response 250 Codes 200 150 100 50 0 4.9E-13 2.5E-10 5.0E-10 7.5E-10 1.0E-09 1.3E-09 1.5E-09 1.8E-09 Time (s) Channel IA 26 Channel QA AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 22. Step Response (Zoom) 250 90% Codes 200 150 Tr = 170 ps 100 50 10% 0 9.8E-10 1.2E-09 1.5E-09 Channel IA Typical Dynamic Performances Versus Sampling Frequency Time (s) Channel QA Figure 23. ENOB Versus Sampling Frequency in Nyquist Conditions (Fin = Fs/2) 7.6 7.4 ENOB (Bit) 7.2 7.0 6.8 6.6 6.4 6.2 6.0 100 200 300 400 500 600 700 800 900 1000 1100 Fs (Msps) Figure 24. SFDR Versus Sampling Frequency in Nyquist Conditions (Fin = Fs/2) -50 SFDR (dBc) -53 -56 -59 -62 -65 100 300 500 700 900 1100 Fs (Msps) 27 2153C–BDC–04/04 Figure 25. THD Versus Sampling Frequency in Nyquist Conditions (Fin = Fs/2) -48 -50 THD (dBc) -52 -54 -56 -58 -60 100 300 500 700 900 1100 Fs (Msps) Figure 26. SNR Versus Sampling Frequency in Nyquist Conditions (Fin = Fs/2) 45 SNR (dBc) 44 43 42 41 40 100 300 500 700 900 1100 Fs (Msps) Typical Dynamic Performances Versus Input Frequency Figure 27. ENOB Versus Input Frequency (Fs = 1 Gsps) 8.0 7.5 ENOB (Bit) 7.0 6.5 6.0 5.5 5.0 0 200 400 600 800 1000 Fin (MHz) 28 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 28. SFDR Versus Input Frequency (Fs = 1 Gsps) -35 -40 SFDR (dBc) -45 -50 -55 -60 -65 0 200 400 600 800 1000 800 1000 800 1000 Fin (MHz) Figure 29. THD Versus Input Frequency (Fs = 1 Gsps) -35 -40 THD (dBc) -45 -50 -55 -60 -65 0 200 400 600 Fin (MHz) Figure 30. SNR Versus Input Frequency (Fs = 1 Gsps) 50 48 46 SNR (dBc) 44 42 40 38 36 34 32 30 0 200 400 600 Fin (MHz) 29 2153C–BDC–04/04 Typical Reconstructed Signals and Signal Spectrum Figure 31. Fs = 1 Gsps and Fin = 20 MHz (1:2 DMUX, Fs/2 DR Type, FiSDA = -15 ps, ISA = -50 ps) 20 250 Ch IA Ch QA 0 200 150 -40 dBc Codes -20 -60 100 -80 50 Ch IA Ch QA -100 -120 0 1 513 1025 1537 2049 2561 3073 0 3585 31 62 93 125 156 187 218 249 Fout/2 F (Msps) Samples Figure 32. Fs = 1 Gsps and Fin = 500 MHz (1:2 DMUX, Fs/2 DR Type, FiSDA = -15 ps, ISA = -50 ps) 20 250 Ch IA Ch QA 0 200 -40 dBc Codes -20 150 -60 100 -80 50 Ch IA Ch QA 0 -100 -120 1 513 1025 1537 2049 2561 3073 3585 0 31 62 93 Samples 125 156 187 218 249 Fout/2 F (Msps) Figure 33. Fs = 1 Gsps and Fin = 1 GHz (1:2 DMUX, Fs/2 DR Type, FiSDA = -15 ps, ISA = -50 ps) 20 250 Ch IA Ch QA 0 200 150 dBc Codes -20 -40 -60 100 -80 50 -100 Ch IA Ch QA -120 0 1 513 1025 1537 2049 Samples Note: 30 2561 3073 3585 0 31 62 93 125 F (Msps) 156 187 218 249 Fout/2 The spectra are given with respect to the output clock frequency observed by the acquisition system (Figures 31 to 33). AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 34. Fs = 1 Gsps and Fin = 20 MHz (Interleaving Mode Fint = 2 Gsps, Fs/4 DR Type, FiSDA = -15 ps, ISA = -50 ps) 20 250 0 200 dBc Codes -20 150 -40 -60 100 -80 50 -100 -120 0 1 2048 4095 6142 8189 0 10236 12283 14330 16377 125 250 375 500 624 749 874 Fs (MHz) Samples 999 Fs/2 Figure 35. Fs = 1 Gsps and Fin = 250 MHz (Interleaving Mode Fint = 2 Gsps, Fs/4 DR Type, FiSDA = -15 ps, ISA = -50 ps) 20 250 0 200 150 dBc Codes -20 -40 -60 100 -80 50 -100 0 -120 1 2048 4095 6142 8189 Samples 10236 12283 14330 16377 0 125 250 375 500 Fs (MHz) 624 749 874 999 Fs/2 31 2153C–BDC–04/04 Typical Performance Sensitivity Versus Power Supplies and Temperature Figure 36. ENOB Versus VCCA = VCCD (Fs = 1 Gsps, Fin = 500 MHz, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) 7.4 7.2 ENOB (Bit) 7.0 6.8 6.6 6.4 6.2 6.0 3.1 3.15 3.2 3.25 3.3 3.35 3.4 3.45 3.5 Vcca = Vccd (V) Figure 37. SFDR Versus VCCA = VCCD (Fs = 1 Gsps, Fin = 500 MHz, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) -40 SFDR (dBc) -45 -50 -55 -60 3.1 3.15 3.2 3.25 3.3 3.35 3.4 3.45 3.5 Vcca = Vccd (V) 32 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 38. THD Versus VCCA = VCCD (Fs = 1 Gsps, Fin = 500 MHz, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) -40 THD (dBc) -45 -50 -55 -60 3.1 3.15 3.2 3.25 3.3 3.35 3.4 3.45 3.5 Vcca = Vccd (V) Figure 39. SNR Versus VCCA = VCCD (Fs = 1 Gsps, Fin = 500 MHz, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) 45.0 SNR (dBc) 44.0 43.0 42.0 41.0 40.0 3.1 3.15 3.2 3.25 3.3 3.35 3.4 3.45 3.5 Vcca = Vccd (V) 33 2153C–BDC–04/04 Figure 40. ENOB Versus Junction Temperature (Fs = 1 Gsps, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) 8.0 7.5 1 Gsps 20 MHz ENOB (Bit) 7.0 1 Gsps 502 MHz 6.5 6.0 1 Gsps 998 MHz 5.5 5.0 -50 -25 0 25 Tj (˚C) 50 75 100 Figure 41. SFDR Versus Junction Temperature (Fs = 1 Gsps, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) -35 1 Gsps 998 MHz -40 SFDR (dBc) -45 1 Gsps 502 MHz -50 -55 1 Gsps 20 MHz -60 -65 -50 34 -25 0 25 Tj (˚C) 50 75 100 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 42. THD Versus Junction Temperature (Fs = 1 Gsps, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) -35 1 Gsps 998 MHz THD (dBc) -40 -45 1 Gsps 502 MHz -50 1 Gsps 20 MHz -55 -60 -50 -25 0 25 50 75 100 Tj (˚C) Figure 43. SNR Versus Junction Temperature (Fs = 1 Gsps, 1:2 DMUX, Fs/4 DR Type, ISA = -50 ps) 45.0 SNR (dBc) 44.0 1 Gsps 20 MHz 43.0 1 Gsps 502 MHz 42.0 41.0 1 Gsps 998 MHz 40.0 -50 -25 0 25 Tj (˚C) 50 75 100 35 2153C–BDC–04/04 Test and Control Features 3-wire Serial Interface Control Setting Table 11. 3-wire Serial Interface Control Settings Mode Characteristics Mode = 1 (2.25V) 3-wire serial bus interface activated Mode = 0 (0V) 3-wire serial bus interface deactivated Nominal setting: Dual channel I and Q activated One clock I 0 dB gain DMUX mode 1:1 DRDA I & Q = 0 ps ISA I & Q = 0 ps FiSDA Q = 0 ps Binary output Decimation test mode OFF Calibration setting OFF Data Ready = Fs /2 36 AT84AD001B 2153C–BDC–04/04 AT84AD001B 3-wire Serial Interface and Data Description The 3-wire bus is activated with the control bit mode set to 1. The length of the word is 19 bits: 16 for the data and 3 for the address. The maximum clock frequency is 50 MHz. Table 12. 3-wire Serial Interface Address Setting Description Address Setting 000 Standby Gray/binary mode 1:1 or 1:2 DMUX mode Analog input MUX Clock selection Auto-calibration Decimation test mode Data Ready Delay Adjust 001 Analog gain adjustment Data7 to Data0: gain channel I Data15 to Data8: gain channel Q Code 00000000: -1.5 dB Code 10000000: 0 dB Code 11111111: 1.5 dB Steps: 0.011 dB 010 Offset compensation Data7 to Data0: offset channel I Data15 to Data8: offset channel Q Data7 and Data15: sign bits Code 11111111b: 31.75 LSB Code 10000000b: 0 LSB Code 00000000b: 0 LSB Code 01111111b: -31.75 LSB Steps: 0.25 LSB Maximum correction: ±31.75 LSB 011 100 Gain compensation Data6 to Data0: channel I/Q (Q is matched to I) Code 11111111b: -0.315 dB Code 10000000b: 0 dB Code 0000000b: 0 dB Code 0111111b: 0.315 dB Steps: 0.005 dB Data6: sign bit Internal Settling Adjustment (ISA) Data2 to Data0: channel I Data5 to Data3: channel Q Data15 to Data6: 1000010000 37 2153C–BDC–04/04 Table 12. 3-wire Serial Interface Address Setting Description (Continued) Address 101 110 111 Notes: 38 Setting Testability Data3 to Data0 = 0000 Mode S/H transparent OFF: Data4 = 0 ON: Data4 = 1 Data7 = 0 Data8 = 0 Built-In Test (BIT) Data0 = 0 BIT Inactive Data0 = 1 BIT Active Data1 = 0 Static BIT Data1 = 1 Dynamic BIT If Data1 = 1, then Ports BI & BQ = Rising Ramp Ports AI & AQ = Decreasing Ramp If Data1 = 0, then Data2 to Data9 = Static Data for BIT Ports BI & BQ = Data2 to Data9 Ports AI & AQ = NOT (Data2 to Data9) Data Ready Delay Adjust (DRDA) Data2 to Data0: clock I Data5 to Data3: clock Q Steps: 140 ps 000: -560 ps 100: 0 ps 111: 420 ps Fine Sampling Delay Adjustment (FiSDA) on channel Q Data10 to Data6: channel Q Steps: 5 ps Data4: sign bit Code 11111: -75 ps Code 10000: 0 ps Code 00000: 0 ps Code 01111: 75 ps 1. The Internal Settling Adjustment could change independently of the two analog sampling times (TA channels I and Q) of the sample/hold (with a fixed digital sampling time) with steps of ±50 ps: Nominal mode will be given by Data2…Data0 = 100 or Data5…Data3 = 100. Data5…Data3 = 000 or Data2…Data0 = 000: sampling time is -200 ps compared to nominal. Data2…Data0 = 111 or Data5…Data3 = 111: sampling time is 150 ps compared to nominal. We recommend setting the ISA to -50 ps to optimize the ADC’s dynamic performances. 2. The Fine Sampling Delay Adjustment enables you to change the sampling time (steps of ±5 ps) on channel Q more precisely, particularly in the interleaved mode. 3. A Built-In Test (BIT) function is available to rapidly test the device’s I/O by either applying a defined static pattern to the dual ADC or by generating a dynamic ramp at the output of the dual ADC. This function is controlled via the 3-wire bus interface at the address 110. The maximum clock frequency in dynamic BIT mode is 750 Msps. Please refer to “Built-In Test (BIT)” on page 43 for more information about this function. 4. The decimation mode enables you to lower the output bit rate (including the output clock rate) by a factor of 16, while the internal clock frequency remains unchanged. The maximum clock frequency in decimation mode is 750 Msps. 5. The “S/H transparent” mode (address 101, Data4) enables bypassing of the ADC’s track/hold. This function optimizes the ADC’s performances at very low input frequencies (Fin < 50 MHz). 6. In the Gray mode, when the input signal is overflow (that is, the differential analog input is greater than 250 mV), the output data must be corrected using the output DOIR: If DOIR = 1: Data7 unchanged Data6 = 0, Data5 = 0, Data4 = 0, Data3 = 0, Data2 = 0, Data1 = 0, Data0 = 0. In 1:2 DMUX mode, only one out-of-range bit is provided for both A and B ports. AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 13. 3-wire Serial Interface Data Setting Description Setting for Address: 000 D15 D14 D13 D12 D11 D10 D9(1) D8 D7 D6 D5 D4 D3 D2 D1 D0 X X X X X X 0 X X X X X X X 1 1 X X X X X X 0 X X X X X X X 0 1 X X X X X X 0 X X X X X X X 1 0 No standby mode X X X X X X 0 X X X X X X X 0 0 Binary output mode X X X X X X 0 X X X X X X 1 X X Gray output mode X X X X X X 0 X X X X X X 0 X X DMUX 1:2 mode X X X X X X 0 X X X X X 1 X X X DMUX 1:1 mode X X X X X X 0 X X X X X 0 X X X Analog selection mode Input I →ADC I Input Q →ADC Q X X X X X X 0 X X X 1 1 X X X X Analog selection mode Input I →ADC I Input I →ADC Q X X X X X X 0 X X X 1 0 X X X X Analog selection mode Input Q →ADC I Input Q →ADC Q X X X X X X 0 X X X 0 X X X X X Clock Selection mode CLKI →ADC I CLKQ →ADC Q X X X X X X 0 X 1 1 X X X X X X Clock selection mode CLKI →ADC I CLKI →ADC Q X X X X X X 0 X 1 0 X X X X X X Clock selection mode CLKI →ADC I CLKIN →ADC Q X X X X X X 0 X 0 X X X X X X X Decimation OFF mode X X X X X X 0 0 X X X X X X X X Decimation ON mode X X X X X X 0 1 X X X X X X X X Keep last calibration calculated value(4) No calibration phase X X X X 0 1 0 X X X X X X X X X No calibration phase(5) No calibration value X X X X 0 0 0 X X X X X X X X X Start a new calibration phase X X X X 1 1 0 X X X X X X X X X Full standby mode Standby channel I (2) Standby channel Q (3) 39 2153C–BDC–04/04 Table 13. 3-wire Serial Interface Data Setting Description (Continued) Setting for Address: 000 D15 D14 D13 D12 D11 D10 D9(1) D8 D7 D6 D5 D4 D3 D2 D1 D0 Control wait bit calibration(6) X X a b X X 0 X X X X X X X X X In 1:2 DMUX FDataReady I & Q = Fs/2 X 0 X X X X 0 X X X X X X X X X In 1:2 DMUX FDataReady I & Q = Fs/4 X 1 X X X X 0 X X X X X X X X X Notes: 1. 2. 3. 4. D9 must be set to “0” Mode standby channel I: use analog input I Vini, Vinib and Clocki. Mode standby channel Q: use analog input Q Vinq, Vinqb and Clockq. Keep last calibration calculated value - no calibration phase: D11 = 0 and D10 = 1. No new calibration is required. The values taken into account for the gain and offset are either from the last calibration phase or are default values (reset values). 5. No calibration phase - no calibration value: D11 = 0 and D10 = 0. No new calibration phase is required. The gain and offset compensation functions can be accessed externally by writing in the registers at address 010 for the offset compensation and at address 011 for the gain compensation. 6. The control wait bit gives the possibility to change the internal setting for the auto-calibration phase: For high clock rates (> 500 Msps) use a = b = 1. For clock rates > 250 Msps and < 500 Msps use a = 1 and b = 0. For clock rates > 125 Msps and < 250 Msps use a = 0 and b = 1. For low clock rates < 125 Msps use a = 0 and b = 0. 3-wire Serial Interface Timing Description The 3-wire serial interface is a synchronous write-only serial interface made of three wires: • sclk: serial clock input • sldn: serial load enable input • sdata: serial data input The 3-wire serial interface gives write-only access to as many as 8 different internal registers of up to 16 bits each. The input format is always fixed with 3 bits of register address followed by 16 bits of data. The data and address are entered with the Most Significant Bit (MSB) first. The write procedure is fully synchronous with the rising clock edge of “sclk” and described in the write chronogram (Figure 44 on page 41). 40 • “sldn” and “sdata” are sampled on each rising clock edge of “sclk” (clock cycle). • “sldn” must be set to 1 when no write procedure is performed. • A minimum of one rising clock edge (clock cycle) with “sldn” at 1 is required for a correct start of the write procedure. • A write starts on the first clock cycle with “sldn” at 0. “sldn” must stay at 0 during the complete write procedure. • During the first 3 clock cycles with “sldn” at 0, 3 bits of the register address from MSB (a[2]) to LSB (a[0]) are entered. • During the next 16 clock cycles with “sldn” at 0, 16 bits of data from MSB (d[15]) to LSB (d[0]) are entered. • An additional clock cycle with “sldn” at 0 is required for parallel transfer of the serial data d[15:0] into the addressed register with address a[2:0]. This yields 20 clock cycles with “sldn” at 0 for a normal write procedure. AT84AD001B 2153C–BDC–04/04 AT84AD001B • A minimum of one clock cycle with “sldn” returned at 1 is requested to close the write procedure and make the interface ready for a new write procedure. Any clock cycle where “sldn” is at 1 before the write procedure is completed interrupts this procedure and no further data transfer to the internal registers is performed. • Additional clock cycles with “sldn” at 0 after the parallel data transfer to the register (done at the 20th consecutive clock cycle with “sldn” at 0) do not affect the write procedure and are ignored. It is possible to have only one clock cycle with “sldn” at 1 between two following write procedures. • 16 bits of data must always be entered even if the internal addressed register has less than 16 bits. Unused bits (usually MSBs) are ignored. Bit signification and bit positions for the internal registers are detailed in Table 12 on page 37. To reset the registers, the Pin mode can be used as a reset pin for chip initialization, even when the 3-wire serial interface is used. Figure 44. Write Chronogram Mode 1 2 3 4 a[1] a[0] d[15] 5 13 14 15 16 17 18 19 d[1] d[0] 20 sclk sldn sdata Internal register value a[2] d[8] d[7] d[6] d[5] d[4] d[3] d[2] New d Reset setting Reset Write procedure Figure 45. Timing Definition Twlmode Mode Tsclk Twsclk Tdmode Tdmode sclk Tssldn Thsldn Tssdata Thsdata sldn sdata 41 2153C–BDC–04/04 Table 14. Timing Description Value Name Parameter Unit Min Typ Max Tsclk Sclk period 20 ns Twsclk High or low time of sclk 5 ns Tssldn Setup time of sldn before rising edge of sclk 4 ns Thsldn Hold time of sldn after rising edge of sclk 2 ns Tssdata Setup time of sdata before rising edge of sclk 4 ns Thsdata Hold time of sdata after rising edge of sclk 2 ns Twlmode Minimum low pulse width of mode 5 ns Tdmode Minimum delay between an edge of mode and the rising edge of sclk 10 ns Calibration Description The AT84AD001B offers the possibility of reducing offset and gain matching between the two ADC cores. An internal digital calibration may start right after the 3-wire serial interface has been loaded (using data D12 of the 3-wire serial interface with address 000). The beginning of calibration disables the two ADCs and a standard data acquisition is performed. The output bit CAL goes to a high level during the entire calibration phase. When this bit returns to a low level, the two ADCs are calibrated with offset and gain and can be used again for a standard data acquisition. If only one channel is selected (I or Q) the offset calibration duration is divided by two and no gain calibration between the two channels is necessary. Figure 46. Internal Timing Calibration 3-wire Serial Interface LDN CAL Tcal The Tcal duration is a multiple of the clock frequency ClockI (master clock). Even if a dual clock scheme is used during calibration, ClockQ will not be used. The control wait bits (D13 and D14) give the possibility of changing the calibration’s setting depending on the clock’s frequency: 42 • For high clock rates (> 500 Msps) use a = b = 1, Tcal = 10112 clock I periods. • For clock rates > 250 Msps and < 500 Msps use a = 1, b = 0, Tcal = 6016 clock I periods. • For clock rates > 125 Msps and < 250 Msps use a = 0, b = 1 ,Tcal = 3968 clock I periods. • For low clock rates (< 125 Msps) use a = 0, b = 0 , Tcal = 2944 clock I periods. AT84AD001B 2153C–BDC–04/04 AT84AD001B The calibration phase is necessary when using the AT84AD001B in interlace mode, where one analog input is sampled at both ADC cores on the common input clock’s rising and falling edges. This operation is equivalent to converting the analog signal at twice the clock frequency Table 15. Matching Between Channels Value Parameter Min Gain error (single channel I or Q) without calibration Gain error (single channel I or Q) with calibration -0.5 0 -0.5 0 Unit LSB 0.5 0 Mean offset code without calibration (single channel I or Q) Mean offset code with calibration (single channel I or Q) Max 0 Offset error (single channel I or Q) without calibration Offset error (single channel I or Q) with calibration Typ LSB LSB 0.5 LSB 127.5 127 127.5 128 During the ADC’s auto-calibration phase, the dual ADC is set with the following: • Decimation mode ON • 1:1 DMUX mode • Binary mode Any external action applied to any signal of the ADC’s registers is inhibited during the calibration phase. Gain and Offset Compensation Functions It is also possible for the user to have external access to the ADC’s gain and offset compensation functions: • Offset compensation between I and Q channels (at address 010) • Gain compensation between I and Q channels (at address 011) To obtain manual access to these two functions, which are used to set the offset to middle code 127.5 and to match the gain of channel Q with that of channel I (if only one channel is used, the gain compensation does not apply), it is necessary to set the ADC to “manual” mode by writing 0 at bits D11 and D10 of address 000. Built-In Test (BIT) A Built-In Test (BIT) function is available to allow rapid testing of the device’s I/O by either applying a defined static pattern to the ADC or by generating a dynamic ramp at the ADC’s output. The dynamic ramp can be used with a clock frequency of up to 750 Msps. This function is controlled via the 3-wire bus interface at address 101. • The BIT is active when Data0 = 1 at address 110. • The BIT is inactive when Data0 = 0 at address 110. • The Data1 bit allows choosing between static mode (Data1 = 0) and dynamic mode (Data1 = 1). When the static BIT is selected (Data1 = 0), it is possible to write any 8-bit pattern by defining the Data9 to Data2 bits. Port B then outputs an 8-bit pattern equal to Data9 ... Data2, and Port A outputs an 8-bit pattern equal to NOT (Data9 ... Data2). 43 2153C–BDC–04/04 Example: Address = 110 Data = D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 X X X X X X 0 1 0 1 0 1 0 1 0 1 One should then obtain 01010101 on Port B and 10101010 on Port A. When the dynamic mode is chosen (Data1 = 1) port B outputs a rising ramp while Port A outputs a decreasing one. Note: Decimation Mode The decimation mode is provided to enable rapid testing of the ADC at a maximum clock frequency of 750 Msps. In decimation mode, one data out of 16 is output, thus leading to a maximum output rate of 46.875 Msps. Note: Die Junction Temperature Monitoring Function In dynamic mode, use the DRDA function to align the edges of CLKO with the middle of the data. Frequency (CLKO) = frequency (Data) = Frequency (CLKI)/16. A die junction temperature measurement setting is included on the board for junction temperature monitoring. The measurement method forces a 1 mA current into a diode-mounted transistor. Caution should be given to respecting the polarity of the current. In any case, one should make sure the maximum voltage compliance of the current source is limited to a maximum of 1V or use a resistor serial-mounted with the current source to avoid damaging the transistor device (this may occur if the current source is reverse-connected). The measurement setup is illustrated in Figure 47. Figure 47. Die Junction Temperature Monitoring Setup VDiode (Pin 35) 1 mA GNDD (Pin 36) 44 Protection Diodes AT84AD001B 2153C–BDC–04/04 AT84AD001B The VBE diode’s forward voltage in relation to the junction temperature (in steady-state conditions) is shown in Figure 48. Figure 48. Diode Characteristics Versus TJ 860 840 820 Diode Voltage (mV) 800 780 760 740 720 700 680 660 640 620 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 Junction Temperature (˚C) VtestI, VtestQ VtestI and VtestQ pins are for internal test use only. These two signals must be left open. Equivalent Input/Output Schematics Figure 49. Simplified Input Clock Model VCCD CLK 100Ω 50Ω VCCD/2 50Ω 100Ω CLKB GNDD 45 2153C–BDC–04/04 Figure 50. Simplified Data Ready Reset Buffer Model VCCD DDRB 100Ω 50Ω VCCD/2 50Ω 100Ω DDRBN GNDD Figure 51. Analog Input Model Vcca Vcca DC Coupling (Common Mode = Ground = 0V) 50Ω Vinl Reverse Termination Sel Input I ESD GND VinI VinI Double Pad GND – 0.4V MAX ESD GND 50Ω VinQ Reverse Termination GND GND VinQ VinQ Double Pad 46 Sel Input Q AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 52. Data Output Buffer Model VCCO DOAIO, DOAI7 DOBIO, DOBI7 DOAION, DOAI7N DOBION, DOBI7N GNDO Definitions of Terms Table 16. Definitions of Terms Abbreviation Definition Description BER Bit Error Rate The probability to exceed a specified error threshold for a sample at a maximum specified sampling rate. An error code is a code that differs by more than ±4 LSB from the correct code DNL Differential Non-Linearity The differential non-linearity for an output code i is the difference between the measured step size of code i and the ideal LSB step size. DNL (i) is expressed in LSBs. DNL is the maximum value of all DNL (i). A DNL error specification of less than 1 LSB guarantees that there are no missing output codes and that the transfer function is monotonic ENOB Effective Number of Bits FPBW Full Power Input Bandwidth The analog input frequency at which the fundamental component in the digitally reconstructed output waveform has fallen by 3 dB with respect to its low frequency value (determined by FFT analysis) for input at full-scale -1 dB (-1 dBFS) IMD Inter-Modulation Distortion The two tones intermodulation distortion (IMD) rejection is the ratio of either of the two input tones to the worst third order intermodulation products INL Integral Non-Linearity The integral non-linearity for an output code i is the difference between the measured input voltage at which the transition occurs and the ideal value of this transition. INL (i) is expressed in LSBs and is the maximum value of all |INL (i)| JITTER Aperture uncertainty The sample-to-sample variation in aperture delay. The voltage error due to jitters depends on the slew rate of the signal at the sampling point Noise Power Ratio The NPR is measured to characterize the ADC’s performance in response to broad bandwidth signals. When applying a notch-filtered broadband white noise signal as the input to the ADC under test, the Noise Power Ratio is defined as the ratio of the average out-ofnotch to the average in-notch power spectral density magnitudes for the FFT spectrum of the ADC output sample test NPR A SINAD – 1.76 + 20 log ----------Fs/2 ENOB = ----------------------------------------------------------------------------6.02 Where A is the actual input amplitude and Fs is the full scale range of the ADC under test 47 2153C–BDC–04/04 Table 16. Definitions of Terms (Continued) Abbreviation Definition Description ORT Overvoltage Recovery Time The time to recover a 0.2% accuracy at the output, after a 150% full-scale step applied on the input is reduced to midscale PSRR Power Supply Rejection Ratio The ratio of input offset variation to a change in power supply voltage SFDR Spurious Free Dynamic Range The ratio expressed in dB of the RMS signal amplitude, set at 1 dB below full-scale, to the RMS value of the highest spectral component (peak spurious spectral component). The peak spurious component may or may not be a harmonic. It may be reported in dB (related to the converter -1 dB full-scale) or in dBc (related to the input signal level) SINAD Signal to Noise and Distortion Ratio The ratio expressed in dB of the RMS signal amplitude, set to 1 dB below full-scale (-1 dBFS) to the RMS sum of all other spectral components including the harmonics, except DC SNR Signal to Noise Ratio The ratio expressed in dB of the RMS signal amplitude, set to 1 dB below full-scale, to the RMS sum of all other spectral components excluding the first 9 harmonics SSBW Small Signal Input Bandwidth The analog input frequency at which the fundamental component in the digitally reconstructed output waveform has fallen by 3 dB with respect to its low frequency value (determined by FFT analysis) for input at full-scale -10 dB (-10 dBFS) TA Aperture delay The delay between the rising edge of the differential clock inputs (CLK, CLKB) [zero crossing point] and the time at which VIN and VINB are sampled TC Encoding Clock period TC1 = minimum clock pulse width (high) TC = TC1 + TC2 TC2 = minimum clock pulse width (low) TD1 Time Delay from Data Transition to Data Ready The general expression is TD1 = TC1 + TDR - TDO with TC = TC1 + TC2 = 1 encoding clock period TD2 Time Delay from Data Ready to Data The general expression is TD2 = TC2 + TDR - TDO with TC = TC1 + TC2 = 1 encoding clock period TDO Digital Data Output Delay The delay from the rising edge of the differential clock inputs (CLK, CLKB) [zero crossing point] to the next point of change in the differential output data (zero crossing) with a specified load TDR Data Ready Output Delay The delay from the falling edge of the differential clock inputs (CLK, CLKB) [zero crossing point] to the next point of change in the differential output data (zero crossing) with a specified load TF Fall Time The time delay for the output data signals to fall from 20% to 80% of delta between the low and high levels THD Total Harmonic Distortion The ratio expressed in dB of the RMS sum of the first 9 harmonic components to the RMS input signal amplitude, set at 1 dB below full-scale. It may be reported in dB (related to the converter -1 dB full-scale) or in dBc (related to the input signal level ) TPD Pipeline Delay The number of clock cycles between the sampling edge of an input data and the associated output data made available (not taking into account the TDO) TR Rise Time 48 The time delay for the output data signals to rise from 20% to 80% of delta between the low and high levels AT84AD001B 2153C–BDC–04/04 AT84AD001B Table 16. Definitions of Terms (Continued) Abbreviation Definition Description TRDR Data Ready Reset Delay The delay between the falling edge of the Data Ready output asynchronous reset signal (DDRB) and the reset to digital zero transition of the Data Ready output signal (DR) TS Settling Time The time delay to rise from 10% to 90% of the converter output when a full-scale step function is applied to the differential analog input VSWR Voltage Standing Wave Ratio The VSWR corresponds to the ADC input insertion loss due to input power reflection. For example, a VSWR of 1.2 corresponds to a 20 dB return loss (99% power transmitted and 1% reflected) 49 2153C–BDC–04/04 Using the AT84AD001B Dual 8-bit 1 Gsps ADC Decoupling, Bypassing and Grounding of Power Supplies The following figures show the recommended bypassing, decoupling and grounding schemes for the dual 8-bit 1 Gsps ADC power supplies. Figure 53. VCCD and VCCA Bypassing and Grounding Scheme L PC Board 3.3V VCCD L 1µF VCCA 100 pF PC Board GND C C Figure 54. VCCO Bypassing and Grounding Scheme L VCCO PC Board 2.25V 1µF 100 pF PC Board GND C Note: L and C values must be chosen in accordance with the operation frequency of the application. Figure 55. Power Supplies Decoupling Scheme VCCA VCCA 100 pF 10 nF GNDA GNDA GNDO VCCD VCCO VCCO 100 pF 10 nF GNDO 100 pF 10 nF GNDD Note: 50 The bypassing capacitors (1 µF and 100 pF) should be placed as close as possible to the board connectors, whereas the decoupling capacitors (100 pF and 10 nF) should be placed as close as possible to the device. AT84AD001B 2153C–BDC–04/04 AT84AD001B Analog Input Implementation The analog inputs of the dual ADC have been designed with a double pad implementation as illustrated in Figure 56. The reverse pad for each input should be tied to ground via a 50Ω resistor. The analog inputs must be used in differential mode only. Figure 56. Termination Method for the ADC Analog Inputs in DC Coupling Mode 50Ω VinI 50Ω Source VinI Channel I GND GND VinIB 50Ω VinIB Dual ADC 50Ω VinQ 50Ω Source VinQ Channel Q GND VinQB GND 50Ω VinQB 51 2153C–BDC–04/04 Figure 57. Termination Method for the ADC Analog Inputs in AC Coupling Mode 50Ω VinI 50Ω Source VinI Channel I GND VinIB GND 50Ω VinIB Dual ADC 50Ω VinQ 50Ω Source VinQ Channel Q GND VinQB GND 50Ω VinQB Clock Implementation The ADC features two different clocks (I or Q) that must be implemented as shown in Figure 58. Each path must be AC coupled with a 100 nF capacitor. Figure 58. Differential Termination Method for Clock I or Clock Q ADC Package CLK 50Ω VCCD/2 100 nF CLKB Note: 52 50Ω Differential Buffer 100 nF When only clock I is used, it is not necessary to add the capacitors on the CLKQ and CLKQN signal paths; they may be left floating. AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 59. Single-ended Termination Method for Clock I or Clock Q VCCD AC coupling capacitor 50Ω Source R1 CLK 50Ω 50Ω AC coupling capacitor R2 CLKB 50Ω VCCD/2 Output Termination in 1:1 Ratio When using the integrated DMUX in 1:1 ratio, the valid port is port A. Port B remains unused. Port A functions in LVDS mode and the corresponding outputs (DOAI or DOAQ) have to be 100Ω differentially terminated as shown in Figure 60 on page 54. The pins corresponding to Port B (DOBI or DOBQ pins) must be left floating (in high impedance state). Figure 60 shows the example of a 1:1 ratio of the integrated DMUX for channel I (the same applies to channel Q). 53 2153C–BDC–04/04 Figure 60. Example of Termination for Channel I Used in DMUX 1:1 Ratio (Port B Unused) DOBI0 / DOBI0N DOBI1 / DOBI1N DOBI2 / DOBI2N Port B DOBI3 / DOBI3N Floating (High Z) DOBI4 / DOBI4N DOBI5 / DOBI5N DOBI6 / DOBI6N DOBI7 / DOBI7N Dual ADC Package DOAI0 / DOAI0N DOAI1 / DOAI1N VCCO DOAI2 / DOAI2N DOAI3 / DOAI3N Port A DOAI4 / DOAI4N DOAI5 / DOAI5N DOAI0 Z0 = 50Ω DOAI0N Z0 = 50Ω LVDS In DOAI6 / DOAI6N DOAI7 / DOAI7N 100Ω LVDS In Note: If the outputs are to be used in single-ended mode, it is recommended that the true and false signals be terminated with a 50Ω resistor. Using the Dual ADC With Figure 61 on page 55 illustrates the configuration of the dual ADC (1:2 DMUX mode, independent I and Q clocks) driving an LVDS system (ASIC/FPGA) with potential addiand ASIC/FPGA Load tional DMUXes used to halve the speed of the dual ADC outputs. 54 AT84AD001B 2153C–BDC–04/04 AT84AD001B Figure 61. Dual ADC and ASIC/FPGA Load Block Diagram Data rate = FsI/2 Port A DEMUX 8 :16 Channel I Data rate = FsQ/2 Data rate = FsQ/4 CLKI/CLKIN @ FsI Dual 8-bit 1 Gsps ADC Port A Channel Q DMUX 8 :16 ASIC / FPGA Port B Channel I DMUX 8 :16 CLKQ/CLKQN @ FsQ Port B DMUX 8 :16 Channel Q Note: The demultiplexers may be internal to the ASIC/FPGA system. 55 2153C–BDC–04/04 Thermal Characteristics Simplified Thermal Model for LQFP 144 20 x 20 x 1.4 mm The following model has been extracted from the ANSYS FEM simulations. Assumptions: no air, no convection and no board. Figure 62. Simplified Thermal Model for LQFP Package Silicon Junction 355 µm silicon die 25 mm 2 λ = 0.95W/cm/˚C 0.6˚C/watt 40 µm Epoxy/Ag glue λ = 0.02 W/cm/˚C 1.4˚C/watt Copper paddle λ = 2.5W/cm/˚C Package top Resin λ = 0.007W/cm/˚C 0.1˚C/watt 6.1˚C/watt 1.5˚C/watt 5.5˚C/watt Leads tip Aluminium paddle Resin Copper alloy leadframe λ = 0.007W/cm/˚C λ = 25W/cm/˚C Aluminium paddle λ = 0.75W/cm/˚C 0.1˚C/watt Resin bottom λ = 0.007W/cm/˚C 4.3˚C/watt Package bottom Assumptions: Die 5.0 x 5.0 = 25 mm 2 40 µm thick Epoxy/Ag glue 8.3˚C/watt 100 µm air gap λ = 0.00027W/cm/˚C 11.4˚C/watt Package bottom connected to: 100 µm thermal grease gap diamater 12 mm λ = 0.01W/cm/˚C (user dependent) Top of user board 1.5˚C/watt Note: The above are typical values with an assumption of uniform power dissipation over 2.5 x 2.5 mm2 of the top surface of the die. Thermal Resistance from Junction to Bottom of Leads Assumptions: no air, no convection and no board. Thermal Resistance from Junction to Top of Case Assumptions: no air, no convection and no board. Thermal Resistance from Junction to Bottom of Case Assumptions: no air, no convection and no board. Thermal Resistance from Junction to Bottom of Air Gap The thermal resistance from the junction to the bottom of the air gap (bottom of package) is 17.9° C/W typical. 56 The thermal resistance from the junction to the bottom of the leads is 15.2° C/W typical. The thermal resistance from the junction to the top of the case is 8.3° C/W typical. The thermal resistance from the junction to the bottom of the case is 6.4° C/W typical. AT84AD001B 2153C–BDC–04/04 AT84AD001B Thermal Resistance from Junction to Ambient The thermal resistance from the junction to ambient is 25.2° C/W typical. Note: In order to keep the ambient temperature of the die within the specified limits of the device grade (that is TA max = 70°C in commercial grade and 85°C in industrial grade) and the die junction temperature below the maximum allowed junction temperature of 105°C, it is necessary to operate the dual ADC in air flow conditions (1m/s recommended). In still air conditions, the junction temperature is indeed greater than the maximum allowed TJ. - TJ = 25.2°C/W x 1.4W + TA = 35.28 + 70 = 105.28°C for commercial grade devices - TJ = 25.2°C/W x 1.4W + TA = 35.28 + 85 = 125.28°C for industrial grade devices Thermal Resistance from Junction to Board The thermal resistance from the junction to the board is 13° C/W typical. 57 2153C–BDC–04/04 Ordering Information Part Number Package Temperature Range Screening Comments AT84XAD001BTD LQFP 144 Ambient Prototype Prototype version Please contact your local Atmel sales office AT84AD001BCTD LQFP 144 C grade 0°C < TA < 70°C Standard AT84AD001BITD LQFP 144 I grade -40°C < TA < 85°C Standard AT84AD001TD-EB LQFP 144 Ambient Prototype 58 Evaluation Kit AT84AD001B 2153C–BDC–04/04 AT84AD001B Packaging Information Figure 63. Type of Package N Dims. A A1 A2 D D1 E E1 L e b ddd ccc o 1 B E1 A E Notes: D D1 Body +2.00 mm footprint Tols. Leads 144L max. 1.60 0.05 min./0.15 max. +/- 0.05 1.40 +/-0.20 22.00 +/-0.10 20.00 +/-0.20 22.00 +/-0.10 20.00 +0.15/-0.10 0.60 basic 0.50 +/-0.05 0.22 0.08 max. 0.08 o 0 o- 5 1. All dimensions are in millimeters 2. Dimensions shown are nominal with tolerances as indicated 3. L/F: eftec 64T copper or equivalent 4. Foot length: "L" is measured at gauge plane at 0.25 mm above the seating plane D 12 o TYP. A2 e A1 A o 12 TYP. 0.20 RAD max. 0.20 RAD nom. 6o A C Stand off A1 0.25 Seating plane C Lead coplanarity 0 0.17 max b L Note: + o -4 ddd e c A-B e De ccc c Thermally enhanced package: LQFP 144, 20 x 20 x 1.4 mm. 59 2153C–BDC–04/04 Atmel Corporation 2325 Orchard Parkway San Jose, CA 95131, USA Tel: 1(408) 441-0311 Fax: 1(408) 487-2600 Regional Headquarters Europe Atmel Sarl Route des Arsenaux 41 Case Postale 80 CH-1705 Fribourg Switzerland Tel: (41) 26-426-5555 Fax: (41) 26-426-5500 Asia Room 1219 Chinachem Golden Plaza 77 Mody Road Tsimshatsui East Kowloon Hong Kong Tel: (852) 2721-9778 Fax: (852) 2722-1369 Japan 9F, Tonetsu Shinkawa Bldg. 1-24-8 Shinkawa Chuo-ku, Tokyo 104-0033 Japan Tel: (81) 3-3523-3551 Fax: (81) 3-3523-7581 Atmel Operations Memory 2325 Orchard Parkway San Jose, CA 95131, USA Tel: 1(408) 441-0311 Fax: 1(408) 436-4314 RF/Automotive Theresienstrasse 2 Postfach 3535 74025 Heilbronn, Germany Tel: (49) 71-31-67-0 Fax: (49) 71-31-67-2340 Microcontrollers 2325 Orchard Parkway San Jose, CA 95131, USA Tel: 1(408) 441-0311 Fax: 1(408) 436-4314 La Chantrerie BP 70602 44306 Nantes Cedex 3, France Tel: (33) 2-40-18-18-18 Fax: (33) 2-40-18-19-60 ASIC/ASSP/Smart Cards 1150 East Cheyenne Mtn. Blvd. Colorado Springs, CO 80906, USA Tel: 1(719) 576-3300 Fax: 1(719) 540-1759 Biometrics/Imaging/Hi-Rel MPU/ High Speed Converters/RF Datacom Avenue de Rochepleine BP 123 38521 Saint-Egreve Cedex, France Tel: (33) 4-76-58-30-00 Fax: (33) 4-76-58-34-80 Zone Industrielle 13106 Rousset Cedex, France Tel: (33) 4-42-53-60-00 Fax: (33) 4-42-53-60-01 1150 East Cheyenne Mtn. Blvd. Colorado Springs, CO 80906, USA Tel: 1(719) 576-3300 Fax: 1(719) 540-1759 Scottish Enterprise Technology Park Maxwell Building East Kilbride G75 0QR, Scotland Tel: (44) 1355-803-000 Fax: (44) 1355-242-743 Literature Requests www.atmel.com/literature Disclaimer: Atmel Corporation makes no warranty for the use of its products, other than those expressly contained in the Company’s standard warranty which is detailed in Atmel’s Terms and Conditions located on the Company’s web site. The Company assumes no responsibility for any errors which may appear in this document, reserves the right to change devices or specifications detailed herein at any time without notice, and does not make any commitment to update the information contained herein. No licenses to patents or other intellectual property of Atmel are granted by the Company in connection with the sale of Atmel products, expressly or by implication. Atmel’s products are not authorized for use as critical components in life support devices or systems. © Atmel Corporation 2004. All rights reserved. Atmel ® and combinations thereof are the registered trademarks and Smart ADC ™ is the trademark of Atmel Corporation or its subsidiaries. Other terms and product names may be the trademarks of others. Printed on recycled paper. 2153C–BDC–04/04 0M