ADC10321 10-Bit, 20MSPS, 98mW A/D Converter with Internal Sample and Hold General Description The ADC10321 is a low power, high performance CMOS analog-to-digital converter that digitizes signals to 10 bits resolution at sampling rates up to 25Msps while consuming a typical 98mW from a single 5V supply. Reference force and sense pins allow the user to connect an external reference buffer amplifier to ensure optimal accuracy. No missing codes is guaranteed over the full operating temperature range. The unique two stage architecture achieves 9.2 Effective Bits with a 10MHz input signal and a 20MHz clock frequency. Output formatting is straight binary coding. To ease interfacing to 3V systems, the digital I/O power pins of the ADC10321 can be tied to a 3V power source, making the outputs 3V compatible. When not converting, power consumption can be reduced by pulling the PD (Power Down) pin high, placing the converter into a low power standby state, where it typically consumes less than 4mW. The ADC10321's speed, resolution and single supply operation makes it well suited for a variety of applications in video, imaging, communications, multimedia and high speed data acquisition. Low power, single supply operation ideally suit the ADC10321 for high speed portable applications, and its speed and resolution are ideal for charge coupled device (CCD) input systems. The ADC10321 comes in a space saving 32-pin TQFP and operates over the industrial (-40C TA +85C) temperature range. n Guaranteed No Missing Codes n Tri-State Outputs n TTL/CMOS or 3V Logic Input/Output Compatible Key Specifications n n n n n n n n Resolution Conversion Rate ENOB@ 10MHz Input DNL Conversion Latency PSRR Power Consumption Low Power Standby Mode 10 Bits 20 Msps 9.2 Bits (typ) 0.35 LSB (typ) 2 Clock Cycles 56dB 98mW (typ) < 4mW (typ) Applications n n n n n n n Digital Video Communications Document Scanners Medical Imaging Electro-Optics Plain Paper Copiers CCD Imaging Features n Internal Sample-and-Hold n Single +5V Operation n Low Power Standby Mode (c) 2003 National Semiconductor Corporation DS100897 www.national.com ADC10321 10-Bit, 20MSPS, 98mW A/D Converter with Internal Sample and Hold January 2003 ADC10321 Connection Diagram 10089701 Ordering Information Commercial (-40C TA +85C) NS Package ADC10321CIVT TQFP Block Diagram 10089702 www.national.com 2 ADC10321 Pin Descriptions and Equivalent Circuits Pin No. Symbol Description Equivalent Circuit Analog I/O 30 VIN Analog Input signal to be converted. Conversion range is VREF+ S to VREF- S. 31 VREF+ F Analog input that goes to the high side of the reference ladder of the ADC. This voltage should force VREF+ S to be in the range of 2.3V to 4.0V. 32 VREF+ S Analog output used to sense the voltage near the top of the ADC reference ladder. 2 VREF- F Analog input that goes to the low side of the reference ladder of the ADC. This voltage should force VREF- S to be in the range of 1.3V to 3.0V. 1 VREF- S Analog output used to sense the voltage near the bottom of the ADC reference ladder. 9 CLK Converter digital clock input. VIN is sampled on the falling edge of CLK input. 8 PD Power Down input. When this pin is high, the converter is in the Power Down mode and the data output pins are in a high impedance state. 26 OE Output Enable pin. When this pin and the PD pin are low, the output data pins are active. When this pin or the PD pin is high, the output data pins are in a high impedance state. 14 thru 19 and 22 thru 25 D0 -D9 3, 7, 28 Digital Output pins providing the 10 bit conversion results. D0 is the LSB, D9 is the MSB. Valid data is present just after the falling edge of the CLK input. Positive analog supply pins. These pins should be connected to a clean, quiet voltage source of +5V. VA and VD should have a common supply and be separately bypassed with 10F to 50F capacitors in parallel with 0.1F capacitors. VA 3 www.national.com ADC10321 Pin Descriptions and Equivalent Circuits Pin No. Symbol 5, 10 (Continued) Description Equivalent Circuit Positive digital supply pins. These pins should be connected to a clean, quiet voltage source of +5V. VA and VD should have a common supply and be separately bypassed with 10F to 50F capacitors in parallel with 0.1F capacitors. VD 12, 21 VD I/O Positive supply pins for the digital output drivers. These pins should be connected to a clean, quiet voltage source of +3V to +5V and be separately bypassed with 10F capacitors. 4, 27, 29 AGND The ground return for the analog supply. AGND and DGND should be connected together close to the ADC10321 package. 6, 11 DGND The ground return for the digital supply. AGND and DGND should be connected together close to the ADC10321 pacjage. 13, 20 DGND I/O www.national.com The ground return of the digital output drivers. 4 Machine Model (Notes 1, 200V 2) Soldering Temp., Infrared, 10 sec. (Note 6) If Military/Aerospace specified devices are required, please contact the National Semiconductor Sales Office/ Distributors for availability and specifications. Storage Temperature Positive Supply Voltage (V = VA = VD) Voltage on Any I/O Pin Operating Ratings(Notes 1, 2) 6.5V Operating Temperature -0.3V to (VA or VD) +0.3V) Package Input Current (Note 3) Package Dissipation at TA = 25C +4.5V to +5.5V VD I/O Supply Voltage See (Note 4) ESD Susceptibility (Note 5) Human Body Model -40C TA +85C VA, VD Supply Voltage 25mA 50mA Input Current at Any Pin (Note 3) 235C -65C to +150C 1500V +2.7V to 5.5V VIN Voltage Range 1.3V to (VA-1.0V) VREF + Voltage Range 2.3V to (VA-1.0V) VREF- Voltage Range 1.3V to 3.0V PD, CLK, OE Voltage -0.3V to + 5.5V Converter Electrical Characteristics The following specifications apply for VA = +5.0VDC, VD = 5.0VDC, VD I/O = 5.0VDC, VREF+ = +3.5VDC, VREF- = +1.5VDC, CL = 20pF, fCLK = 20MHz, RS = 25. Boldface limits apply for TA = TMIN to TMAX: all other limits TA = 25C(Note 7) Symbol Parameter Conditions Typical (Note 8) Limits (Note 9) 0.45 0.35 1.0 0.85 LSB(max) 10 Bits Units Static Converter Characteristics INL Integral Non-Linearity DNL Differential-Non Linearity Resolution with No Missing Codes LSB(max) Zero Scale Offset Error -6 mV(max) Full-Scale Error -6 mV(max) Dynamic Converter Characteristics ENOB Effective Number of Bits fIN = 1.0MHz fIN = 4.43MHz fIN = 10MHz 9.5 9.5 9.2 9.0 Bits Bits(min) Bits S/(N+D) Signal-to-Noise Plus Distortion Ratio fIN = 1.0MHz fIN = 4.43MHz fIN = 10MHz 59 59 57 56 dB dB(min) dB SNR Signal-to-Noise Ratio fIN = 1.0MHz fIN = 4.43MHz fIN = 10MHz 60 60 58 58 dB dB(min) dB THD Total Harmonic Distortion fIN = 1.0MHz fIN = 4.43MHz fIN = 10MHz -71 -70 -66 -59 dB dB(min) dB SFDR Spurious Free Dynamic Range fIN = 1.0MHz fIN = 4.43MHz fIN = 10MHz 74 72 68 60 dB dB dB DG Differential Gain Error fIN = 4.43MHz, fCLK = 17.72MHz 0.5 %(max) DP Differential Phase Error fIN = 4.43MHz, fCLK = 17.72MHz 0.5 deg(max) Overrange Output Code VIN > VREF+ Underrange Output Code VIN < VREF- BW Full Power Bandwidth PSRR Power Supply Rejection Ratio 1023 0 Change in Full Scale with 4.5V to 5.5V Supply Change 150 MHz 56 dB Reference and Analog Input Characteristics VIN 1.3 4.0 Analog Input Range 5 V(min) V(max) www.national.com ADC10321 Absolute Maximum Ratings ADC10321 Converter Electrical Characteristics (Continued) The following specifications apply for VA = +5.0VDC, VD = 5.0VDC, VD I/O = 5.0VDC, VREF+ = +3.5VDC, VREF- = +1.5VDC, CL = 20pF, fCLK = 20MHz, RS = 25. Boldface limits apply for TA = TMIN to TMAX: all other limits TA = 25C(Note 7) Symbol Parameter Typical (Note 8) Conditions CIN Analog VIN Input Capacitance 5 IIN Input Leakage Current 10 RREF Reference Ladder Resistance VREF+ VREF- (VREF+) - (VREF -) Limits (Note 9) Units pF A 1000 850 1150 (min) (max) Positive Reference Voltage 3.5 4.0 V(max) Negative Reference Voltage 1.5 1.3 V(min) 2.0 1.0 2.7 V(min) V(max) Total Reference Voltage DC and Logic Electrical Characteristics The following specifications apply for VA = +5.0VDC, VD = +5.0VDC, VD I/O = 5.0VDC, VREF+ = +3.5VDC, VREF- = +1.5VDC, CL = 20 pF, fCLK = 20MHz, RS = 25. Boldface limits apply for TA = TMIN to TMAX: all other limits TA = 25C(Note 7) Symbol Parameter Conditions Typical (Note 8) Limits (Note 9) Units CLK, OE, PD, Digital Input Characteristics VIH Logical "1" Input Voltage VD = 5.5V 2.0 V(min) VIL Logical "0" Input Voltage VD = 4.5V 1.0 V(max) IIH Logical "1" Input Current VIH = VD 10 A IIL Logical "0" Input Current VIL = DGND -10 A D00 - D13 Digital Output Characteristics VOH Logical "1" Output Voltage VD I/O = + 4.5V, IOUT = -0.5mA VD I/O = + 2.7V, IOUT = -0.5mA 4.0 2.4 V(min) V(min) VOL Logical "0" Output Voltage VD I/O = + 4.5V, IOUT = -1.6mA VD I/O = + 2.7V, IOUT = -1.6mA 0.4 0.4 V(max) V(max) IOZ TRI-STATE Output Current VOUT = DGND VOUT = VD -10 10 A A Output Short Circuit Current VD I/O = 3V mA VD I/O = 5V 12 25 IOS mA Power Supply Characteristics IA Analog Supply Current PD = LOW, Ref not included PD = HIGH, Ref not included 14.5 0.5 16 mA(max) ID + IDI/O Digital Supply Current PD = LOW, Ref not included PD = HIGH, Ref not included 5 0.2 6 mA(max) PD Power Consumption 98 110 mW (max) AC Electrical Characteristics The following specifications apply for VA = +5.0VDC, VD I/O = 5.0VDC, VREF+ = +3.5VDC, VREF- = +1.5VDC, fCLK = 20MHz, trc = tfc = 5ns, RS = 25. CL (data bus loading) = 20 pF, Boldface limits apply for TA = TMIN to TMAX: all other limits TA = 25C(Note 7) Symbol Parameter Conditions Typical (Note 8) Limits (Note 9) Units (Limits) 20 MHz(min) fCLK1 Maximum Clock Frequency 25 fCLK2 Minimum Clock Frequency 1 tCH Clock High Time 23 ns(min tCL Clock Low Time 23 ns(min) 45 55 %(min) %(max) Duty Cycle www.national.com 50 6 MHz(max) (Continued) The following specifications apply for VA = +5.0VDC, VD I/O = 5.0VDC, VREF+ = +3.5VDC, VREF- = +1.5VDC, fCLK = 20MHz, trc = tfc = 5ns, RS = 25. CL (data bus loading) = 20 pF, Boldface limits apply for TA = TMIN to TMAX: all other limits TA = 25C(Note 7) Symbol Parameter Conditions Typical (Note 8) Pipeliine Delay (Latency) trc, tfc Clock Input Rise and Fall Time tr, tf Output Rise and Fall Times 10 tOD Fall of CLK to data valid 20 tOH Output Data Hold Time tDIS Rising edge of OE to valid data tEN Falling edge of OE to valid data tVALID Data valid time tAD tAJ tWU Limits (Note 9) Units (Limits) 2.0 Clock Cycles 5 ns(max) 25 ns(max) ns 12 ns From output High, 2K to Ground 25 ns From output Low, 2K to VD I/O 18 ns 1K to VCC 25 ns 40 ns Apeture Delay 4 ns Aperture Jitter < 30 ps Full Scale Step Response tr = 10ns 1 conversion Overrange Recovery Time VIN step from (VREF+ +100mV) to (VREF-) 1 conversion 700 ns PD low to 1/2 LSB accurate conversion (Wake-Up time) Note 1: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device is functional, but do not guarantee specific performance limits. For guaranteed specifications and test conditions, see the Electrical Characteristics. The guaranteed specifications apply only for the test conditions listed. Some performance characteristics may degrade when the device is not operated under the listed test conditions. Note 2: All voltages are measured with respect to GND = AGND = DGND = 0V, unless otherwise specified. Note 3: When the input voltage at any pin exceeds the power supplies ( VIN < AGND or VIN > VA or VD), the current at that pin should be limited to 25mA. The 50mA maximum package input current rating limits the number of pins that can safely exceed the power supplies with an input current of 25mA to two. Note 4: The absolute maximum junction temperature (TJmax) for this device is 150C. The maximum allowable power dissipation is dictated by TJmax, the junction-to-ambient thermal resistance (JA), and the ambient temperature (TA), and can be calculated using the formula PDMAX = (TJmax - TA)/JA. In the 32-pin TQFP, JA is 69C/W, so PDMAX = 1,811 mW at 25C and 942mW at the maximum operating ambient temperature of 85C. Note that the power dissipation of this device under normal operation will typically be about 110mW (98mW quiescent power + 2mW reference ladder power +10mW due to 10 TTL load on each digital output). The values for maximum power dissipation listed above will be reached only when the ADC10321 is operated in a severe fault condition (e.g. when input or output pins are driven beyond the power supply voltages, or the power supply polarity is reversed). Obviously, such conditions should always be avoided. Note 5: Human body model is 100 pF capacitor discharged through a 1.5k resistor. Machine model is 220 pF discharged through ZERO . Note 6: The 235C reflow temperature refers to infared reflow. For Vapor Phase Reflow (VPR), the following conditions apply: Maintain the temperature at the top of the package body above 183C for a minimum 60 seconds. The temperature measured on the package body must not exceed 220C. Only one excursion above 183C is allowed per reflow cycle. Note 7: The inputs are protected as shown below. Input voltage magnitudes up to 500mV beyond the supply rails will not damage this device. However, errors in the A/D conversion can occur if the input goes above VA or below AGND by more than 300 mV. 7 www.national.com ADC10321 AC Electrical Characteristics ADC10321 AC Electrical Characteristics (Continued) 10089712 10089724 10089711 Note 8: Typical figures are at TA = TJ = 25C, and represent most likely parametric norms. Note 9: Tested limits are guaranteed to Nationsl's AOQL (Average Outgoing Quality Level). Note 10: When the input signal is between VREF+ and (VA + 300mV), the output code will be 3FFh, or all 1s. When the input signal is between -300 mV and VREF-, the output code will be 000h, or all 0s. Typical Performance Characteristics VA = VD = VDI/O = 5V, fCLK = 20MHz, unless otherwise specified. Typical INL 10089726 10089725 INL vs Clock Duty Cycle Typical DNL 10089738 DNL vs VA DNL vs Clock Duty Cycle 10089740 8 10089727 DNL vs fCLK 10089739 10089729 www.national.com INL vs VA INL vs fCLK 10089728 SINAD & ENOB vs Temperature and fIN 10089730 SINAD & ENOB vs VA SINAD & ENOB vs fCLK and fIN 10089732 10089731 IA + ID vs. Temperature 10089737 Spectral Response at 20 MSPs 10089735 9 www.national.com ADC10321 Typical Performance Characteristics VA = VD = VDI/O = 5V, fCLK = 20MHz, unless otherwise specified. (Continued) ADC10321 OUTPUT DELAY is the time delay after the fall of the input clock before the data update is present at the output pins. OUTPUT HOLD TIME is the length of time that the output data is valid after the fall of the input clock. OVER RANGE RECOVERY TIME is the time required after VIN goes from AGND to VREF+ or VIN goes from VA to VREF- for the converter to recover and make a conversion with its rated accuracy. Specification Definitions APERTURE JITTER is the variation in aperture delay from sample to sample. Aperture jitter shows up as input noise. APERTURE DELAY See Sampling Delay. DIFFERENTIAL GAIN ERROR is the percentage difference between the output amplitudes of a given amplitude small signal, high frequency sine wave input at two different dc input levels. DIFFERENTIAL PHASE ERROR is the difference in the output phase of a small signal sine wave input at two different dc input levels. DIFFERENTIAL NON-LINEARITY (DNL) is the measure of the maximum deviation from the ideal step size of 1 LSB. EFFECTIVE NUMBER OF BITS (ENOB, or EFFECTIVE BITS) is another method of specifying Signal-to-Noise and Distortion Ratio (S/N+D or SINAD). ENOB is defined as (SINAD -1.76) / 6.02. PIPELINE DELAY (LATENCY) is the number of clock cycles between initiation of conversion and when that data is presented to the output driver stage. Data for any given sample is available by the Pipeline Delay plus the Output Delay after that sample is taken. New data is available at every clock cycle, but the data lags the conversion by the pipeline delay. PSRR (POWER SUPPLY REJECTION RATIO) is the ratio of the change in dc power supply voltage to the resulting change in Full Scale Error, expressed in dB. SAMPLING (APERTURE) DELAY or APERTURE TIME is that time required after the fall of the clock input for the sampling switch to open. The sample is effectively taken this amount of time after the fall of the clock input. SIGNAL TO NOISE RATIO (SNR) is the ratio, expressed in dB, of the rms value of the input signal to the rms value of the sum of all other spectral components below one-half the sampling frequency, not including harmonics or dc. FULL POWER BANDWIDTH is a measure of the frequency at which the reconstructed output fundamental drops 3 dB below its 1MHz value for a full scale input. The test is performed with fIN equal to 100 kHz plus integral multiples of fCLK. The input frequency at which the output is -3 dB relative to the1MHz input signal is the full power bandwidth. FULL SCALE (FS) INPUT RANGE of the ADC is the input range of voltages over which the ADC will digitize that input. For VREF+ = 3.50V and VREF- = 1.50V, FS = (VREF+) - (VREF-) = 2.00V. FULL SCALE OFFSET ERROR is a measure of how far the last code transition is from the ideal 112 LSB below VREF+ and is defined as V1023 -1.5 LSB - VREF+ , where V1023 is the voltage at which the transitions from code 1022 to 1023 occurs. FULL SCALE STEP RESPONSE is defined as the time required after VIN goes from VREF- to VREF+, or VREF+ to VREF-, and settles sufficiently for the converter to recover and make a conversion with its rated accuracy. SIGNAL TO NOISE PLUS DISTORTION (S/(N+D) or SINAD) is the ratio, expressed in dB, of the RMS value of the input signal to the RMS value of all of the other spectral components below half the clock frequency, including harmonics but excluding dc. SPURIOUS FREE DYNAMIC RANGE (SFDR) is the difference, expressed in dB or dBc, between the RMS values of the input signal and the peak spurious signal, where a spurious signal is any signal present in the output spectrum that is not present at the input. TOTAL HARMONIC DISTORTION (THD) is the ratio, expressed in dB, of the rms total of the first six harmonic components, to the rms value of the input signal. ZERO SCALE OFFSET ERROR is the difference between the ideal input voltage (12 LSB) and the actual input voltage that just causes a transition from an output code of zero to an output code of one. INTEGRAL NON-LINEARITY (INL) is a measure of the deviation of each individual code from a line drawn from negative full scale (12 LSB below the first code transition) through positive full scale (112 LSB above the last code transition). The deviation of any given code from this straight line is measured from the center of that code value. www.national.com 10 ADC10321 Timing Diagram 10089715 FIGURE 1. ADC10321 Timing Diagram 10089717 10089716 FIGURE 2. AC Test Circuit FIGURE 3. tEN, tDIS Test Circuit 11 www.national.com ADC10321 bandwidth, low distortion and minimal Differential Gain and Differential Phase. The LMH6702 performs best with a feedback resistor of about 100 ohms. Care should be taken to keep digital noise out of the analog input circuitry to maintain highest noise performance. Functional Description The ADC10321 maintains excellent dynamic performance for input signals up to half the clock frequency. The use of an internal sample-and-hold amplifier (SHA) enables sustained dynamic performance for signals of input frequency beyond the clock rate, lowers the converter's input capacitance and reduces the number of external components required. The analog signal at VIN that is within the voltage range set by VREF+ S and VREF- S are digitized to ten bits at up to 25 MSPS. Input voltages below VREF- S will cause the output word to consist of all zeroes. Input voltages above VREF+ S will cause the output word to consist of all ones. VREF+ S has a range of 2.3 to 4.0 Volts, while VREF- S has a range of 1.3 to 3.0 Volts. VREF+ S should always be at least 1.0 Volt more positive than VREF- S. 2.0 REFERENCE INPUTS Note: Throughout this data sheet reference is made to VREF+ and to VREF-. These refer to the internal voltage across the reference ladder and are, nominally, VREF+ S and VREF- S, respectively. Figure 4 shows a simple reference biasing scheme with minimal components. While this circuit might suffice for some applications, it does suffer from thermal drift because the external resistor at pin 2 will have a different temperature coefficient than the on-chip resistors. Also, the on-chip resistors, while well matched to each other, will have a large tolerance compared with any external resistors, causing the value of VREF- to be quite variable. No d.c. current should be allowed to flow through pin 1 or 32 or linearity errors will result near the zero scale and full scale ends of the signal excursion. The sense pins were designed to be used with high impedance opamp inputs for high accuracy biasing. The circuit of Figure 5 is an improvement over the circuit of Figure 4 in that both ends of the reference ladder are defined with reference voltages. This reduces problems of high reference variability and thermal drift, but requires two reference sources. In addition to the usual reference inputs, the ADC10321 has two sense outputs for precision control of the ladder voltages. These sense outputs (VREF+ S and VREF- S) compensate for errors due to IR drops between the source of the reference voltages and the ends of the reference ladder itself. Data is acquired at the falling edge of the clock and the digital equivalent of that data is available at the digital outputs 2.0 clock cycles plus tOD later. The ADC10321 will convert as long as the clock signal is present at pin 9 and the PD pin is low. The Output Enable pin (OE), when low, enables the output pins. The digital outputs are in the high impedance state when the OE pin is low or the PD pin is high. Applications Information 1.0 THE ANALOG INPUT The analog input of the ADC10321 is a switch (transmission gate) followed by a switched capacitor amplifier. The capacitance seen at the input changes with the clock level, appearing as about 3pF when the clock is low, and about 5pF when the clock is high. This small change in capacitance can be reasonably assumed to be a fixed capacitance. Care should be taken to avoid driving the input beyond the supply rails, even momentarily, as during power-up. The LMH6702 has been found to be a good device to drive the ADC10321 because of its low voltage capability, wide www.national.com With the addition of two op-amps, the voltages at the top and bottom of the reference ladder can be forced to the exact value desired, as shown in Figure 6. 12 ADC10321 Applications Information (Continued) 10089718 FIGURE 4. Simple, low component count reference biasing 13 www.national.com ADC10321 Applications Information (Continued) 10089719 FIGURE 5. Better low component count reference biasing Voltages at the reference sense pins (VREF+ S and VREF- S) should be within the range specified in the Operating Ratings table (2.3V to 4.0V for VREF+ and 1.3V to 3.0V for VREF-). Any device used to drive the reference pins should be able to source sufficient current into the VREF+ F pin and sink sufficient current from the VREF- F pin when the ladder is at its minimum value of 850 Ohms. The reference voltage at the top of the ladder (VREF+) may take on values as low as 1.0V above the voltage at the bottom of the ladder (VREF-) and as high as (VA - 1.0V) Volts. The voltage at the bottom of the ladder (VREF-) may take on values as low as 1.3 Volts and as high as 3.0V. However, to minimize noise effects and ensure accurate conversions, the total reference voltage range (VREF+ VREF-) should be a minimum of 2.0V and a maximum of 2.7V. The VREF+ F and VREF- F pins should each be bypassed to AGND with 10F tantalum or electrolytic and 0.1F ceramic capacitors. The circuit of Figure 6 may be used if it is desired to obtain precise reference voltages. The LMC6082 in this circuit was chosen for its low offset voltage, low voltage rail-to-rail capability and low cost. Since the current flowing through the sense lines (those lines associated with VREF+ S and VREF- S) is essentially zero, there is negligible voltage drop across any resistance in series with these sense pins and the voltage at the inverting input of the op-amp accurately represents the voltage at the top (or bottom) of the ladder. The op-amp drives the force input, forcing the voltage at the ends of the ladder to equal the voltage at the op-amp's non-inverting input, plus any offset voltage. For this reason, op-amps with low VOS, such as the LMC6081 and LMC6082, should be used for this application. www.national.com 14 ADC10321 Applications Information (Continued) 10089720 FIGURE 6. Setting precision reference voltages 3.0 POWER SUPPLY CONSIDERATIONS A/D converters draw sufficient transient current to corrupt their own power supplies if not adequately bypassed. A 10F to 50F tantalum or aluminum electrolytic capacitor should be placed within an inch (2.5 centimeters) of the A/D power pins, with a 0.1F ceramic chip capacitor placed as close as possible to each of the converter's power supply pins. Leadless chip capacitors are preferred because they have low lead inductance. While a single voltage source should be used for the analog and digital supplies of the ADC10321, this supply should not be the supply that is used for other digital circuitry on the board. As is the case with all high speed converters, the ADC10321 should be assumed to have little high frequency power supply rejection. A clean analog power source should be used. No pin should ever have a voltage on it that is in excess of the supply voltages or below ground, not even on a transient basis. This can be a problem upon application of power to a circuit. Be sure that the supplies to circuits driving the CLK, PD, OE, analog input and reference pins do not come up any faster than does the voltage at the ADC10321 power pins. 4.0 THE ADC10321 CLOCK Although the ADC10321 is tested and its performance is guaranteed with a 20MHz clock, it typically will function with clock frequencies from 1MHz to 25MHz. Performance is best if the clock rise and fall times are 5ns or less. If the CLK signal is interrupted, or its frequency is too low, the charge on internal capacitors can dissipate to the point where the accuracy of the output data will degrade. This is what limits the minimum sample rate. The duty cycle of the clock signal can affect the performance of the A/D Converter. Because achieving a precise duty cycle is difficult, this device is designed to maintain performance over a range of duty cycles. While it is specified and performance is guaranteed with a 50% clock duty cycle, performance is typically maintained over a clock duty cycle range of 45% to 55%. 15 www.national.com ADC10321 Applications Information 5.0 LAYOUT AND GROUNDING (Continued) Proper routing of all signals and proper ground techniques are essential to ensure accurate conversion. Separate analog and digital ground planes are required to meet data sheet limits. The analog ground plane should be low impedance and free of noise form other parts of the system. The clock line should be series terminated at the source end in the characteristic impedance of that line. Use a series resistor right after the source such that the source impedance plus that series resistor equals the characteristic impedance of the clock line. This resistor should be as close to the source as possible, but in no case should it be further away than Each bypass capacitor should be located as close to the appropriate converter pin as possible and connected to the pin and the appropriate ground plane with short traces. The analog input should be isolated from noisy signal traces to avoid coupling of spurious signals into the input. Any external component (e.g., a filter capacitor) connected between the converter's input and ground should be connected to a very clean point in the analog ground return. Figure 7 gives an example of a suitable layout, including power supply routing, ground plane separation, and bypass capacitor placement. All analog circuitry (input amplifiers, filters, reference components, etc.) should be placed on or over the analog ground plane. All digital circuitry and I/O lines should be over the digital ground plane. where tr is the rise time of the clock signal and tPR is the propagation rate down the board. For a Board of FR-4 material, tPR is typically about 150 ps/inch. To maintain a consistent impedance along the clock line, use stripline or microstrip techniques (see Application Note AN1113) and avoid the use of through-holes in the line. It might also be necessary to terminate the ADC end of the clock line with a series RC to ground such that the resistor value equals the characteristic impedance of the clock line and the capacitor value is Digital and analog signal lines should never run parallel to each other in close proximity with each other. They should only cross each other when absolutely necessary, and then only at 90 angles. Violating this rule can result in digital noise getting into the input, which degrades accuracy and dynamic performance (THD, SNR, SINAD). where tPR is again the propagation rate down the clock line, L is the length of the line in inches and ZO is the characteristic impedance of the clock line. 10089721 FIGURE 7. An acceptable layout pattern www.national.com 16 operation. It is not uncommon for high speed digital circuits (e.g., 74F and 74AC devices) to exhibit undershoot that goes more than a volt below ground. A resistor of 50 to 100 in series with the offending digital input will usually eliminate the problem. Care should be taken not to overdrive the inputs of the ADC10321 (or any device) with a device that is powered from supplies outside the range of the ADC10321 supply. Such practice may lead to conversion inaccuracies and even to device damage. (Continued) 6.0 DYNAMIC PERFORMANCE The ADC10321 is ac tested and its dynamic performance is guaranteed. To meet the published specifications, the clock source driving the CLK input must be free of jitter. For best ac performance, isolating the ADC clock from any digital circuitry should be done with adequate buffers, as with a clock tree. See Figure 8 Meeting dynamic specifications is also dependent upon keeping digital noise out of the input, as mentioned in Sections 1.0 and 5.0. Attempting to drive a high capacitance digital data bus. The more capacitance the output drivers has to charge for each conversion, the more instantaneous digital current is required from VD and DGND. These large charging current spikes can couple into the analog section, degrading dynamic performance. Adequate bypassing and maintaining separate analog and digital ground planes will reduce this problem on the board. Buffering the digital data outputs (with an 74F541, for example) may be necessary if the data bus to be driven is heavily loaded. Dynamic performance can also be improved by adding series resistors of 47 at each digital output. Driving the VREF+ F pin or the VREF- F pin with devices that can not source or sink the current required by the ladder. As mentioned in section 2.0, be careful to see that any driving devices can source sufficient current into the VREF+ F pin and sink sufficient current from the VREF- F pin. If these pins are not driven with devices than can handle the required current, they will not be held stable and the converter output will exhibit excessive noise. Using a clock source with excessive jitter. This will cause the sampling interval to vary, causing excessive output noise and a reduction in SNR performance. Simple gates with RC timing is generally inadequate. Using the same voltage source for VD and other digital logic. As mentioned in Section 3.0, VD should use the same power source used by VA, but should be decoupled from VA. 10089722 FIGURE 8. Isolating the ADC clock from digital circuitry 7.0 COMMON APPLICATION PITFALLS Driving the inputs (analog or digital) beyond the power supply rails. For proper operation, all inputs should not go more than 300mV beyond the supply pins. Exceeding these limits on even a transient basis can cause faulty or erratic 17 www.national.com ADC10321 Applications Information ADC10321 10-Bit, 20MSPS, 98mW A/D Converter with Internal Sample and Hold Physical Dimensions inches (millimeters) unless otherwise noted 32-Lead TQFP Package Ordering Number ADC10321CIVT NS Package Number VBE32A LIFE SUPPORT POLICY NATIONAL'S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and whose failure to perform when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury to the user. National Semiconductor Americas Customer Support Center Email: new.feedback@nsc.com Tel: 1-800-272-9959 www.national.com National Semiconductor Europe Customer Support Center Fax: +49 (0) 180-530 85 86 Email: europe.support@nsc.com Deutsch Tel: +49 (0) 69 9508 6208 English Tel: +44 (0) 870 24 0 2171 Francais Tel: +33 (0) 1 41 91 8790 2. A critical component is any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness. National Semiconductor Asia Pacific Customer Support Center Fax: 65-6250 4466 Email: ap.support@nsc.com Tel: 65-6254 4466 National Semiconductor Japan Customer Support Center Fax: 81-3-5639-7507 Email: nsj.crc@jksmtp.nsc.com Tel: 81-3-5639-7560 National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.