HI5703 S E M I C O N D U C T O R 10-Bit, 40 MSPS A/D Converter December 1997 Features Description * * * * * * * * * The HI5703 is a monolithic, 10-bit, analog-to-digital converter fabricated in Harris's BiCMOS process. It is designed for high speed applications where wide bandwidth and low power consumption are essential. Its 40 MSPS speed is made possible by a fully differential pipeline architecture with an internal sample and hold. Sampling Rate . . . . . . . . . . . . . . . . . . . . . . . . . 40 MSPS 8.55 Bits Guaranteed at fIN = 10MHz Low Power Wide Full Power Input Bandwidth . . . . . . . . . . 250MHz On Chip Sample and Hold Fully Differential or Single-Ended Analog Input Single Supply Voltage . . . . . . . . . . . . . . . . . . . . . . . . +5V TTL Compatible Interface 3.3V Digital Outputs Available The HI5703 has excellent dynamic performance while consuming only 400mW power at 40 MSPS. Data output latches are provided which present valid data to the output bus with a latency of 7 clock cycles. It is pin-to-pin compatible with the HI5702. Applications For lower power consumption or internal reference, please refer to the HI5746 or HI5767. * Professional Video Digitizing * Medical Imaging Ordering Information * Digital Communication Systems * High Speed Data Acquisition * Additional Reference Documents - AN9534 Using the HI5703 Evaluation Board - AN9413 Driving the Analog Input of the HI5702 - AN9214 Using Harris High Speed A/D Converters Pinout PART NUMBER TEMP. RANGE (oC) HI5703KCB 0 to 70 28 Ld SOIC (W) HI5703EVAL 25 Evaluation Board M28.3 Typical Application Schematic HI5703 HI5703 (SOIC) TOP VIEW 3.25V VREF+ (7) VREF - (8) 2.0V DVCC1 1 28 D0 (LSB) (28) D0 D0 DGND 2 27 D1 (27) D1 D1 DVCC1 3 26 D2 (26) D2 D2 DGND 4 25 D3 (25) D3 D3 AVCC 5 AGND 6 24 D4 (24) D4 D4 DGND (2) (20) D5 D5 DGND (21) (19) D6 D6 VREF + 7 22 CLK DGND (4) (18) D7 D7 VREF - 8 21 DGND (17) D8 D8 (MSB) (16) D9 D9 20 D5 VIN- 10 19 D6 VDC 11 18 D7 AGND 12 17 D8 AVCC 13 OE 14 AGND (12) AGND (6) 23 DVCC2 VIN+ 9 16 D9 VIN + VIN + (9) VIN - (1) DVCC1 VDC (11) (3) DVCC1 VIN - (10) (23) DVCC2 15 DFS CLOCK CLK (22) (13) AVCC DFS (15) (5) AVCC OE (14) CAUTION: These devices are sensitive to electrostatic discharge. Users should follow proper IC Handling Procedures. Copyright PKG. NO. PACKAGE (c) Harris Corporation 1997 1 AGND DGND BNC 10F AND 0.1F CAPS ARE PLACED AS CLOSE TO PART AS POSSIBLE 0.1F + 10F 0.1F + 10F +5V +5V File Number 3950.6 HI5703 Functional Block Diagram CLOCK BIAS VDC CLK VINVIN+ S/H STAGE 1 DFS 2-BIT FLASH 2-BIT DAC OE + - X2 D9 (MSB) D8 D7 D6 DIGITAL DELAY AND DIGITAL ERROR CORRECTION STAGE 9 D5 D4 D3 2-BIT FLASH 2-BIT DAC D2 D1 + D0 (LSB) - X2 STAGE 10 1-BIT FLASH AVCC AGND DVCC1 DVCC2 DGND 2 VREF + VREF - HI5703 Absolute Maximum Ratings TA = 25oC Thermal Information Thermal Resistance (Typical, Note 1) JA (oC/W) SOIC Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Maximum Junction Temperature . . . . . . . . . . . . . . . . . . . . . . . 150oC Maximum Storage Temperature Range . . . . . . . . . -65oC to 150oC Maximum Lead Temperature (Soldering 10s) . . . . . . . . . . . . . 300oC (SOIC - Lead Tips Only) Supply Voltage, AVCC or DVCC to AGND or DGND. . . . . . . . . . +6V DGND to AGND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.3V Digital I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DGND to DVCC Analog I/O Pins. . . . . . . . . . . . . . . . . . . . . . . . . . . . . AGND to AVCC Operating Conditions Temperature Range, HI5703KCB . . . . . . . . . . . . . . . . . .0oC to 70oC CAUTION: Stresses above those listed in "Absolute Maximum Ratings" may cause permanent damage to the device. This is a stress only rating and operation of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied. NOTE: 1. JA is measured with the component mounted on an evaluation PC board in free air. Electrical Specifications AVCC = DVCC1 = DVCC2 = +5.0V; VREF+ = 3.25V; VREF - = 2.0V; fS = 36 MSPS at 50% Duty Cycle; CL = 20pF; TA = 25oC; Differential Analog Input; Unless Otherwise Specified PARAMETER TEST CONDITION MIN TYP MAX UNITS ACCURACY Resolution 10 - - Bits Integral Linearity Error, INL fIN = DC - 1 2.0 LSB Differential Linearity Error, DNL (Guaranteed No Missing Codes) fIN = DC - 0.5 1 LSB Offset Error, VOS fIN = DC - 4 - LSB Full Scale Error, FSE fIN = DC - 1 - LSB DYNAMIC CHARACTERISTICS Minimum Conversion Rate No Missing Codes - 0.5 1 MSPS Maximum Conversion Rate No Missing Codes 40 - - MSPS Effective Number of Bits, ENOB fIN = 1MHz - 9.2 - Bits fIN = 5MHz - 9.2 - Bits fIN = 10MHz 8.55 8.9 - Bits fIN = 1MHz - 57 - dB Signal to Noise and Distortion Ratio, SINAD RMS Signal = -------------------------------------------------------------RMS Noise + Distortion fIN = 5MHz - 57 - dB fIN = 10MHz 53.2 55 - dB Signal to Noise Ratio, SNR RMS Signal = --------------------------------RMS Noise fIN = 1MHz - 58 - dB fIN = 5MHz - 58 - dB fIN = 10MHz 53.2 57 - dB Total Harmonic Distortion, THD fIN = 1MHz - -64 - dBc fIN = 5MHz - -63 - dBc fIN = 10MHz - -60 - dBc fIN = 1MHz - -75 - dBc fIN = 5MHz - -75 - dBc fIN = 10MHz - -73 - dBc fIN = 1MHz - -66 - dBc fIN = 5MHz - -64 - dBc fIN = 10MHz - -63 - dBc fIN = 1MHz - 66 - dBc fIN = 5MHz - 64 - dBc fIN = 10MHz 2nd Harmonic Distortion 3rd Harmonic Distortion Spurious Free Dynamic Range, SFDR 54 63 - dBc Intermodulation Distortion, IMD f1 = 1MHz, f2 = 1.02MHz - -59 - dBc Differential Gain Error fS = 17.72MHz, 6 Step, Mod Ramp - 0.5 - % Differential Phase Error fS = 17.72MHz, 6 Step, Mod Ramp - 0.1 - Degree - 1 - Cycle 0.2V Overdrive - 1 - Cycle Transient Response Over-Voltage Recovery 3 HI5703 Electrical Specifications AVCC = DVCC1 = DVCC2 = +5.0V; VREF+ = 3.25V; VREF - = 2.0V; fS = 36 MSPS at 50% Duty Cycle; CL = 20pF; TA = 25oC; Differential Analog Input; Unless Otherwise Specified (Continued) PARAMETER TEST CONDITION MIN TYP MAX UNITS Maximum Peak-to-Peak Differential Analog Input Range (VIN+ - VIN-) - 1.25 - V Maximum Peak-to-Peak Single-Ended Analog Input Range - 2.5 - V - 1 - M - 7 - pF -10 - +10 A - 0.5 - A 0.625 - 4.375 V - 250 - MHz 300 400 500 ANALOG INPUT Analog Input Resistance, RIN (Note 3) Analog Input Capacitance, CIN Analog Input Bias Current, IB+ or IB- (Note 3) Differential Analog Input Bias Current IB DIFF = (IB+ - IB-) Analog Input Common Mode Voltage Range (VIN+ + VIN-) / 2 Differential Mode (Note 1) Full Power Input Bandwidth (FPBW) REFERENCE INPUT Total Reference Resistance, RL Reference Current 2.5 3.125 4.2 mA Positive Reference Voltage Input, VREF+ (Note 2) - 3.25 3.3 V Negative Reference Voltage Input, VREF - (Note 2) 1.95 2.0 - V Reference Common Mode Voltage (VREF+ + VREF -) / 2 (Note 2) 2.575 2.625 2.675 V DC Bias Voltage Output, VDC - 2.8 - V Max Output Current - - 1 mA 2.0 - - V DC BIAS VOLTAGE DIGITAL INPUTS Input Logic High Voltage, VIH Input Logic Low Voltage, VIL - - 0.8 V Input Logic High Current, IIH VIH = 5V - - 10.0 A Input Logic Low Current, IIL VIL = 0V - - 10.0 A - 7 - pF - - mA mA Input Capacitance, CIN DIGITAL OUTPUTS Output Logic Sink Current, IOL VO = 0.4V; DVCC2 = 5V 1.6 Output Logic Source Current, IOH VO = 2.4V; DVCC2 = 5V -0.2 - - Output Three-State Leakage Current, IOZ VO = 0/5V; DVCC2 = 5V - 1 10 A Output Logic Sink Current, IOL VO = 0.4V; DVCC2 = 3.3V 1.6 - - mA Output Logic Source Current, IOH VO = 2.4V; DVCC2 = 3.3V -0.2 - - mA Output Three-State Leakage Current, IOZ VO = 0/3.3V; DVCC2 = 3.3V - 1 10 A - 5 - pF Aperture Delay, tAP - 5 - ns Aperture Jitter, tAJ - 5 - ps Data Output Delay, tOD - 7 - ns 5 7 18 ns Data Output Hold, tH - 4 - ns Data Output Enable Time, tEN - 7 - ns Output Capacitance, COUT TIMING CHARACTERISTICS AVCC = DVCC1 = 5V 10%, DVCC2 = 3.3V 5%, 0oC TA 70oC Data Output Enable Time, tDIS - 7 - ns Clock Pulse Width (Low) 40 MSPS Clock 11.875 12.5 13.125 ns Clock Pulse Width (High) 40 MSPS Clock 11.875 12.5 13.125 ns 4 HI5703 Electrical Specifications AVCC = DVCC1 = DVCC2 = +5.0V; VREF+ = 3.25V; VREF - = 2.0V; fS = 36 MSPS at 50% Duty Cycle; CL = 20pF; TA = 25oC; Differential Analog Input; Unless Otherwise Specified (Continued) PARAMETER TEST CONDITION MIN TYP MAX UNITS Data Latency, tLAT For a Valid Sample (Note 2) - - 7 Cycles Power-Up Initialization Data Invalid Time (Note 2) - - 20 Cycles Analog Supply Voltage, AVCC 4.75 5.0 5.25 V Digital Supply Voltage, DVCC1 4.75 5.0 5.25 V At 3.30V 3.135 3.3 3.465 V At 5.0V 4.75 5.0 5.25 V POWER SUPPLY CHARACTERISTICS Digital Output Supply Voltage, DVCC2 Total Supply Current, ICC VIN+ - VIN- = +1.25V and DFS = "0" - 80 - mA Analog Supply Current, AICC VIN+ - VIN- = +1.25V and DFS = "0" - 48 - mA Digital Supply Current, DICC1 VIN+ - VIN- = +1.25V and DFS = "0" - 30 - mA Digital Output Supply Current, DICC2 VIN+ - VIN- = +1.25V and DFS = "0" - 2 - mA Power Dissipation VIN+ - VIN- = +1.25V and DFS = "0" - 400 - mW Offset Error Sensitivity, VOS AVCC or DVCC = 5V 5% - 1.5 - LSB Full Scale Error Sensitivity, FSE AVCC or DVCC = 5V 5% - 0.2 - LSB SN + 5 HN + 5 SN + 6 HN + 6 SN + 7 NOTES: 2. Parameter guaranteed by design or characterization and not production tested. 3. With the clock low and DC input. Timing Waveforms ANALOG INPUT CLOCK INPUT SN - 1 HN - 1 SN HN SN + 1 HN + 1 SN + 2 HN + 7 SN + 8 HN + 8 INPUT S/H 1ST STAGE 2ND STAGE B1, N - 1 B2, N - 2 10TH STAGE DATA OUTPUT DN - 7 B1, N B1, N + 1 B1, N + 4 B1, N + 5 B1, N + 6 B1, N + 7 B2, N - 1 B2, N B2, N + 4 B2, N + 5 B2, N + 6 B10, N - 5 B10, N - 4 B10, N B10, N + 1 B10, N + 2 DN - 6 DN - 2 DN - 1 tLAT NOTES: 4. SN : N-th sampling period. 5. HN : N-th holding period. 6. BM , N : M-th stage digital output corresponding to N-th sampled input. 7. DN : Final data output corresponding to N-th sampled input. FIGURE 1. HI5703 INTERNAL CIRCUIT TIMING 5 DN B10, N + 3 DN + 1 HI5703 Timing Waveforms (Continued) ANALOG INPUT tAP tAJ CLOCK INPUT 1.5V 1.5V tOD tH 2.0V DATA OUTPUT DATA N DATA N - 1 0.8V FIGURE 2. INPUT-TO-OUTPUT TIMING Typical Performance Curves fS = 40 MSPS TEMPERATURE = 25oC -40 fS = 40 MSPS TEMPERATURE = 25oC -45 THD 8.0 -50 dBc EFFECTIVE NUMBER OF BITS (ENOB) -35 9.0 -55 -60 7.0 -65 SFDR -70 6.0 1 2 4 6 8 10 20 40 -75 60 80 100 1 2 4 6 8 10 20 40 60 80 100 INPUT FREQUENCY (MHz) INPUT FREQUENCY (MHz) NOTE: SFDR depicted here does not include any harmonic distortion. FIGURE 3. EFFECTIVE NUMBER OF BITS (ENOB) vs INPUT FREQUENCY FIGURE 4. TOTAL HARMONIC DISTORTION (THD) AND SPURIOUS FREE DYNAMIC RANGE (SFDR) vs INPUT FREQUENCY 450 65 430 SNR 60 POWER DISSIPATION (mW) -THD dB 55 SINAD 50 fS = 40 MSPS TEMPERATURE = 25oC 45 40 VIN + - VIN- = +1.25V AND DFS = "0" TEMPERATURE = 25oC 410 390 370 350 330 310 290 270 250 35 1 2 4 6 8 10 20 INPUT FREQUENCY (MHz) 40 0 60 80 100 FIGURE 5. SINAD, SNR, AND -THD vs INPUT FREQUENCY 5 10 15 20 25 30 fS (MSPS) 35 40 45 50 FIGURE 6. POWER DISSIPATION vs SAMPLE FREQUENCY 6 HI5703 Typical Performance Curves (Continued) 10.0 40 MSPS 8.5 OUTPUT DELAY TIME (ns) EFFECTIVE NUMBER OF BITS (ENOB) 9.0 8.0 45 MSPS 7.5 7.0 6.5 fIN = 10MHz 6.0 50 MSPS 9.0 8.0 7.0 tOD 6.0 5.5 5.0 -40 -20 0 20 40 60 80 5.0 -40 -30 -20 -10 100 0 TEMPERATURE (oC) FIGURE 7. EFFECTIVE NUMBER OF BITS (ENOB) vs TEMPERATURE AND SAMPLE FREQUENCY 20 30 40 50 60 70 80 85 FIGURE 8. OUTPUT DELAY TIME (TOD) vs TEMPERATURE 9.0 EFFECTIVE NUMBER OF BITS (ENOB) 450 fS = 40 MSPS 440 POWER DISSIPATION (mW) 10 TEMPERATURE (oC) VIN+ - VIN- = 1.25V, DFS = 0 430 420 410 400 390 380 -40 fS = 40 MSPS fIN = 10MHz TEMPERATURE = 25oC 8.5 8.0 7.5 -20 0 20 40 60 80 85 45 46 47 48 TEMPERATURE (oC) FIGURE 9. POWER DISSIPATION vs TEMPERATURE 49 50 51 52 DUTY CYCLE (%) 53 54 55 FIGURE 10. EFFECTIVE NUMBER OF BITS (ENOB) vs DUTY CYCLE (TH/TTOTAL) 9.0 INTERMODULATION DISTORTION (dBc) EFFECTIVE NUMBER OF BITS (ENOB) 9.5 12.5ns 8.5 50% 8.0 fIN = fS/ 4 TEMPERATURE = 25oC 7.5 7.0 0 10 20 30 40 50 58.0 IMD 1MHz 56.0 54.0 IMD 10MHz -40 -30 -20 -10 fS (MSPS) FIGURE 11. EFFECTIVE NUMBER OF BITS (ENOB) vs SAMPLE FREQUENCY 0 10 20 30 40 TEMPERATURE 50 60 70 FIGURE 12. INTERMODULATION DISTORTION (IMD) vs TEMPERATURE 7 80 85 HI5703 Typical Performance Curves (Continued) 0dB fIN = 10MHz -10dB fS = 40 MSPS OUTPUT LEVEL (dB) -20dB -30dB -40dB -50dB -60dB -70dB -80dB -90dB -100dB 0 200 400 600 800 1000 1200 1400 1600 1800 2047 FREQUENCY BIN FIGURE 13. 4096 POINT FFT SPECTRAL PLOT 0dB fIN = 1MHz -10dB fS = 40 MSPS OUTPUT LEVEL (dB) -20dB -30dB -40dB -50dB -60dB -70dB -80dB -90dB -100dB 0 200 400 600 800 1000 1200 1400 FREQUENCY BIN FIGURE 14. 4096 POINT FFT SPECTRAL PLOT 8 1600 1800 2047 HI5703 fully-differential output for the converter core. During the sampling phase, the VIN pins see only the on-resistance of a switch and CS . The relatively small values of these components result in a typical full power input bandwidth of 250MHz for the converter. TABLE 1. PIN DESCRIPTION PIN # NAME DESCRIPTION 1 DVCC1 Digital Supply (+5.0V) 2 DGND Digital Ground 3 DVCC1 Digital Supply (+5.0V) 4 DGND Digital Ground 5 AVCC Analog Supply (+5.0V) 6 AGND Analog Ground 7 VREF+ Positive Reference Voltage Input 8 VREF - Negative Reference Voltage Input 9 VIN+ Positive Analog Input 10 VIN- Negative Analog Input 11 VDC DC Bias Voltage Output 12 AGND Analog Ground 13 AVCC Analog Supply (+5.0V) 14 OE Digital Output Enable Control Input 15 DFS Data Format Select Input 16 D9 Data Bit 9 Output (MSB) 17 D8 Data Bit 8 Output 18 D7 Data Bit 7 Output 19 D6 Data Bit 6 Output 20 D5 Data Bit 5 Output 21 DGND 22 CLK 23 DVCC2 24 D4 Data Bit 4 Output 25 D3 Data Bit 3 Output 26 D2 Data Bit 2 Output 27 D1 Data Bit 1 Output 28 D0 Data Bit 0 Output (LSB) 1 VIN+ 1 1 1 CS 2 VIN- CH -+ VOUT+ +- VOUT- CS 1 CH 1 FIGURE 15. ANALOG INPUT SAMPLE-AND-HOLD As illustrated in the functional block diagram and the timing diagram in Figure 1, nine identical pipeline subconverter stages, each containing a two-bit flash converter and a twobit multiplying digital-to-analog converter, follow the S/H circuit with the tenth stage being a one bit flash converter. Each converter stage in the pipeline will be sampling in one phase and amplifying in the other clock phase. Each individual subconverter clock signal is offset by 180 degrees from the previous stage clock signal resulting in alternate stages in the pipeline performing the same operation. Digital Ground Sample Clock Input The two-bit digital output of each stage is fed to a digital delay line controlled by the internal clock. The purpose of the delay line is to align the digital output data to the corresponding sampled analog input signal. This delayed data is fed to the digital error correction circuit which corrects the error in the output data with the information contained in the redundant bits to form the final ten bit output for the converter. Digital Output Supply (+3.3V to +5V) Because of the pipeline nature of this converter, the data on the bus is output at the 7th cycle of the clock after the analog sample is taken. This delay is specified as the data latency. After the data latency time, the data representing each succeeding sample is output at the following clock pulse. The output data is synchronized to the external clock by a double buffered latching technique. Detailed Description Theory of Operation The HI5703 is a 10-bit fully differential sampling pipeline A/D converter with digital error correction. Figure 15 depicts the circuit for the front end differential-in-differential-out sampleand-hold (S/H). The switches are controlled by an internal clock which is a non-overlapping two phase signal, 1 and 2 , derived from the master clock. During the sampling phase, 1 , the input signal is applied to the sampling capacitors, CS . At the same time the holding capacitors, CH , are discharged to analog ground. At the falling edge of 1 the input signal is sampled on the bottom plates of the sampling capacitors. In the next clock phase, 2 , the two bottom plates of the sampling capacitors are connected together and the holding capacitors are switched to the op-amp output nodes. The charge then redistributes between CS and CH completing one sample-and-hold cycle. The output is a fully-differential, sampled-data representation of the analog input. The circuit not only performs the sample-andhold function but will also convert a single-ended input to a The digital output bits are available in offset binary or two's complement format, the format being set by the Data Format Select (DFS) input. Reference Voltage Inputs, VREF - and VREF+ The HI5703 requires two reference voltages connected to the VREF pins. The HI5703 is tested with VREF - equal to 2V and VREF+ equal to 3.25V for a fully differential input voltage range of 1.25V. VREF+ and VREF - can differ from the above voltages as long as the reference common mode voltage, ((VREF+ + VREF -)/2), does not exceed 2.625V 50mV and the limits on VREF+ and VREF - are not exceeded. In order to minimize overall converter noise it is recommended that adequate high frequency decoupling be provided at the reference voltage input pins, VREF+ and VREF -. 9 HI5703 connected from VIN+ to VIN- will help filter any high frequency noise on the inputs, also improving performance. Values around 20pF are sufficient and can be used on AC coupled inputs as well. Note, however, that the value of capacitor C chosen must take into account the highest frequency component of the analog input signal. Analog Input, Differential Connection The analog input to the HI5703 is a differential input that can be configured in various ways depending on the signal source and the required level of performance. A fully differential connection (Figure 16 and Figure 17) will give the best performance for the converter. Analog Input, Single-Ended Connection The configuration shown in Figure 18 may be used with a single ended AC coupled input. VIN+ VIN R HI5703 VDC VIN+ VIN R R -VIN HI5703 VDC VIN- VIN- FIGURE 16. AC COUPLED DIFFERENTIAL INPUT FIGURE 18. AC COUPLED SINGLE ENDED INPUT Since the HI5703 is powered by a single +5V analog supply, the analog input is limited to be between ground and +5V. For the differential input connection this implies the analog input common mode voltage can range from 0.625V to 4.375V. The performance of the ADC does not change significantly with the value of the analog input common mode voltage. Again, assume the difference between VREF+, typically 3.25V, and VREF-, typically 2V, is 1.25V. If VIN is a 2.5VP-P sinewave, then VIN+ is a 2.5VP-P sinewave riding on a positive voltage equal to VDC. The converter will be at positive fullscale when VIN+ is at VDC + 1.25V and will be at negative fullscale when VIN+ is equal to VDC - 1.25V. Sufficient headroom must be provided such that the input voltage never goes above +5V or below AGND. In this case, VDC could range between 1.25V and 3.75V without a significant change in ADC performance. The simplest way to produce VDC is to use the VDC output of the HI5703. A DC voltage source, VDC , equal to 2.8V (typical), is made available to the user to help simplify circuit design when using an AC coupled differential input. This low output impedance voltage source is not designed to be a reference but makes an excellent bias source and stays within the analog input common mode voltage range over temperature. It has a temperature coefficient of approximately +200ppm/oC. For the AC coupled differential input (Figure 16) assume the difference between VREF+, typically 3.25V, and VREF, typically 2V, is 1.25V. Fullscale is achieved when the VIN+ and VIN- inputs are 1.25VP-P , with VIN- being 180 degrees out of phase with VIN+. The converter will be at positive fullscale when the VIN+ input is at VDC + 0.625V and VIN- is at VDC - 0.625V (VIN+ - VIN- = 1.25V). Conversely, the converter will be at negative full scale when the VIN+ input is equal to VDC - 0.625V and VIN- is at VDC + 0.625V (VIN+ - VIN- = -1.25V). The analog input can be DC coupled (Figure 17) as long as the inputs are within the analog input common mode voltage range (0.625V VDC 4.375V). C HI5703 VDC -VIN VDC VIN+ R C VDC HI5703 VIN- The resistor, R, in Figure 19 is not absolutely necessary but may be used as a load setting resistor. A capacitor, C, connected from VIN+ to VIN- will help filter any high frequency noise on the inputs, also improving performance. Values around 20pF are sufficient and can be used on AC coupled inputs as well. Note, however, that the value of capacitor C chosen must take into account the highest frequency component of the analog input signal. VIN+ R VIN VDC FIGURE 19. DC COUPLED SINGLE ENDED INPUT VIN VDC The single ended analog input can be DC coupled (Figure 19) as long as the input is within the analog input common mode voltage range. R VIN- A single ended source may give better overall system performance if it is first converted to differential before driving the HI5703. Refer to the application notes AN9534, "Using the HI5703 Evaluation Board", and AN9413, "Driving the Analog Input of the HI5702". Application note AN9413 FIGURE 17. DC COUPLED DIFFERENTIAL INPUT The resistors, R, in Figure 17 are not absolutely necessary but may be used as load setting resistors. A capacitor, C, 10 HI5703 The part should be mounted on a board that provides separate low impedance connections for the analog and digital supplies and grounds. For best performance, the supplies to the HI5703 should be driven by clean, linear regulated supplies. The board should also have good high frequency decoupling capacitors mounted as close as possible to the converter. If the part is powered off a single supply then the analog supply and ground pins should be isolated by ferrite beads from the digital supply and ground pins. applies to the HI5703 as well as the HI5702 and describes several different ways of driving the analog differential inputs. Digital Output Control and Clock Requirements The HI5703 provides a standard high-speed interface to external TTL logic families. In order to ensure rated performance of the HI5703, the duty cycle of the clock should be held at 50% 5%. It must also have low jitter and operate at standard TTL levels. Refer to the application notes "Using Harris High Speed A/D Converters" (AN9214) for additional considerations when using high speed converters. Performance of the HI5703 will only be guaranteed at conversion rates above 1 MSPS. This ensures proper performance of the internal dynamic circuits. Similarly, when power is first applied to the converter, a maximum of 20 cycles at a sample rate above 1 MSPS will have to be performed before valid data is available. Static Performance Definitions Offset Error (VOS) The midscale code transition should occur at a level 1/4 LSB above half-scale. Offset is defined as the deviation of the actual code transition from this point. A Data Format Select (DFS) pin is provided which will determine the format of the digital data outputs. When at logic low, the data will be output in offset binary format. When at logic high, the data will be output in two's complement format. Refer to Table 2 for further information. Full-Scale Error (FSE) The last code transition should occur for an analog input that is 3/4 LSB below positive Fullscale (+FS) with the offset error removed. Full-scale error is defined as the deviation of the actual code transition from this point. The output enable pin, OE, when pulled high will three-state the digital outputs to a high impedance state. Set the OE input to logic low for normal operation. OE INPUT DIGITAL DATA OUTPUTS 0 Active 1 High Impedance Differential Linearity Error (DNL) DNL is the worst case deviation of a code width from the ideal value of 1 LSB. Integral Linearity Error (INL) INL is the worst case deviation of a code center from a best fit straight line calculated from the measured data. Supply and Ground Considerations The HI5703 has separate analog and digital supply and ground pins to keep digital noise out of the analog signal path. The digital data outputs also have a separate supply pin, DVCC2 , which can be powered from a 3.3V to 5.0V supply. This allows the outputs to interface with 3.3V logic if so desired. Power Supply Sensitivity Each of the power supplies are moved plus and minus 5% and the shift in the offset and full scale error (in LSBs) is noted. TABLE 2. A/D CODE TABLE OFFSET BINARY OUTPUT CODE (DFS LOW) CODE CENTER DESCRIPTION TWO'S COMPLEMENT OUTPUT CODE (DFS HIGH) M L M L S S S S DIFFERENTIAL B B B INPUT VOLTAGE B (VIN+ - VIN-) D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 +Full Scale (+FS) 1/ LSB 4 1.24939V 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 +FS - 11/4 LSB 1.24695V 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 0 1.83mV 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 +3/4 LSB -1/4 LSB -FS + 13/4 LSB -Full Scale (-FS) + 3/4 LSB -0.610mV 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -1.24573V 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 -1.24817V 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 NOTE: 8. The voltages listed above represent the ideal center of each output code shown as a function of the reference voltage. 9. VREF+ = 3.25V and VREF - = 2.0V. 11 HI5703 Dynamic Performance Definitions Fast Fourier Transform (FFT) techniques are used to evaluate the dynamic performance of the HI5703. A low distortion sine wave is applied to the input, it is coherently sampled, and the output is stored in RAM. The data is then transformed into the frequency domain with an FFT and analyzed to evaluate the dynamic performance of the A/D. The sine wave input to the part is -0.5dB down from full scale for all these tests. the input by 200mV, and measuring the number of cycles it takes for the output code to settle within 10-bit accuracy. Full Power Input Bandwidth (FPBW) Full power input bandwidth is the analog input frequency at which the amplitude of the digitally reconstructed output has decreased 3dB below the amplitude of the input sine wave. The input sine wave has an amplitude which swings from -FS to +FS. The bandwidth given is measured at the specified sampling frequency. SNR and SINAD are quoted in dB. The distortion numbers are quoted in dBc (decibels with respect to carrier) and DO NOT include any correction factors for normalizing to full scale. Video Definitions Effective Number Of Bits (ENOB) Differential Gain and Differential Phase are two commonly found video specifications for characterizing the distortion of a chrominance signal as it is offset through the input voltage range of an ADC. The effective number of bits (ENOB) is calculated from the SINAD data by: ENOB = (SINAD - 1.76 + VCORR) / 6.02 where: VCORR = 0.5 dB Differential Gain (DG) VCORR adjusts the ENOB for the amount the input is below fullscale. Differential Gain is the peak difference in chrominance amplitude (in percent) relative to the reference burst. Signal To Noise and Distortion Ratio (SINAD) Differential Phase (DP) SINAD is the ratio of the measured RMS signal to RMS sum of all the other spectral components below the Nyquist frequency, fS /2, excluding DC. Differential Phase is the peak difference in chrominance phase (in degrees) relative to the reference burst. Signal To Noise Ratio (SNR) Timing Definitions SNR is the ratio of the measured RMS signal to RMS noise at a specified input and sampling frequency. The noise is the RMS sum of all of the spectral components below fS /2 excluding the fundamental, the first five harmonics and DC. Refer to Figure 1 and Figure 2 for these definitions. Aperture Delay (tAP) Aperture delay is the time delay between the external sample command (the falling edge of the clock) and the time at which the signal is actually sampled. This delay is due to internal clock path propagation delays. Total Harmonic Distortion (THD) THD is the ratio of the RMS sum of the first 5 harmonic components to the RMS value of the fundamental input signal. Aperture Jitter (tAJ) 2nd and 3rd Harmonic Distortion Aperture jitter is the RMS variation in the aperture delay due to variation of internal clock path delays. This is the ratio of the RMS value of the applicable harmonic component to the RMS value of the fundamental input signal. Data Hold Time (tH) Spurious Free Dynamic Range (SFDR) Data hold time is the time to where the previous data (N - 1) is no longer valid. SFDR is the ratio of the fundamental RMS amplitude to the RMS amplitude of the next largest spectral component in the spectrum below fS /2. Data Output Delay Time (tOD) Data output delay time is the time to where the new data (N) is valid. Intermodulation Distortion (IMD) Nonlinearities in the signal path will tend to generate intermodulation products when two tones, f1 and f2 , are present at the inputs. The ratio of the measured signal to the distortion terms is calculated. The terms included in the calculation are (f1+f2), (f1-f2), (2f1), (2f2), (2f1+f2), (2f1-f2), (f1+2f2), (f1-2f2). The ADC is tested with each tone 6dB below full scale. Data Latency (tLAT) After the analog sample is taken, the digital data is output on the bus at the 7th cycle of the clock. This is due to the pipeline nature of the converter where the data has to ripple through the stages. This delay is specified as the data latency. After the data latency time, the data representing each succeeding sample is output at the following clock pulse. The digital data lags the analog input sample by 7 cycles. Transient Response Transient response is measured by providing a full scale transition to the analog input of the ADC and measuring the number of cycles it takes for the output code to settle within 10-bit accuracy. Power-Up initialization This time is defined as the maximum number of clock cycles that are required to initialize the converter at power-up. The requirement arises from the need to initialize the dynamic circuits within the converter. Over-Voltage Recovery Over-Voltage Recovery is measured by providing a full scale transition to the analog input of the ADC which overdrives 12 HI5703 AMP HFA1100 HFA1105 HFA1106 HFA1135 HFA1145 HFA1245 A/D HI5702 HI5703 DSP/P D/A HSP9501 HSP48410 HSP48908 HSP48212 HSP43891 HSP43168 HSP43216 HFA1100: HFA1105: HFA1106: HFA1135: HFA1145: HFA1245: HI5702: HI5703: HSP9501: HSP48410: 850MHz Video Op Amp 300MHz Video Op Amp 250MHz Video Op Amp with Bandwidth Limit Control 350MHz Video Op Amp with Output Limiting 300MHz Video Op Amp with Output Disable Dual 350MHz Video Op Amp with Output Disable 10-Bit, 40 MSPS, A/D Converter 10-Bit, 40 MSPS, Low Power A/D Converter Programmable Data Buffer Histogrammer/Accumulating Buffer, 10-Bit Pixel Resolution HSP48908: 2-D Convolver, 3 x 3 Kernal Convolution, 8-Bit HI5780 HI1171 CA3338 HSP48212: HSP43891: HSP43168: HSP43216: HI5780: HI1171: CA3338: HA5020: HA2842: HFA1115: HFA1212: HFA1412: AMP HA5020 HA2842 HFA1115 HFA1212 HFA1412 Digital Video Mixer Digital Filter, 30MHz, 9-Bit Dual FIR Filter, 10-Bit, 33MHz/45MHz Digital Half Band Filter 10-Bit, 80 MSPS, Video D/A Converter 8-Bit, 40 MSPS, Video D/A Converter 8-Bit, 50 MSPS, Video D/A Converter 100MHz Video Op Amp High Output Current, Video Op Amp 225MHz Programmable Gain Video Buffer with Output Limiting 350MHz, Dual Programmable Gain Video Buffer 350MHz, Quad Programmable Gain Video Buffer In addition, CMOS Logic Families in HC/HCT, AC/ACT, FCT and CD4000 are available. FIGURE 20. 10-BIT VIDEO IMAGING COMPONENTS AMP HFA1100 HFA1110 HFA3101 HFA3102 HFA3600 A/D HI5702 HI5703 DSP/P HSP43168 HSP43216 HSP43220 HSP43891 HSP50016 HSP50110 HSP50210 D/A HI5721 HI5780 HI20201 HI20203 AMP HFA1112 HFA1113 HFA1100: 850MHz Op Amp HFA1110: 750MHz Unity Gain Video Buffer HFA3101: Gilbert Cell Transistor Array HFA3102: Dual Long-Tailed Pair Transistor Array HFA3600: Low Noise Amplifier/Mixer HI5702: 10-Bit, 40 MSPS, A/D Converter HI5703: 10-Bit, 40 MSPS, Low Power A/D Converter HSP43168: Dual FIR Filter, 10-Bit, 33MHz/45MHz HSP43216: Digital Half Band Filter HSP43220: Decimating Digital Filter HSP43891: Digital Filter, 30MHz, 9-Bit HSP50016: Digital Down Converter HSP50110: Digital Quadrature Tuner HSP50210: Digital Costas Loop HI5721: 10-Bit, 100 MSPS, Communications D/A Converter HI5780: 10-Bit, 80 MSPS, D/A Converter HI20201: 10-Bit, 160 MSPS, High Speed D/A Converter HI20203: 8-Bit, 160 MSPS, High Speed D/A Converter HFA1112: 850MHz Programmable Gain Video Buffer HFA1113: 850MHz Programmable Gain Video Buffer with Output Limiting In addition, CMOS Logic Families in HC/HCT, AC/ACT, FCT and CD4000 are available. FIGURE 21. 10-BIT COMMUNICATIONS COMPONENTS 13