a 5 V Integrated High Speed ADC/Quad DAC System AD7339 FEATURES 8-Bit A/D Converter Two 8-Bit D/A Converters Two 8-Bit Serial D/A Converters Single +5 V Supply Operation On-Chip Reference Power-Down Mode 52-Lead PQFP Package FUNCTIONAL BLOCK DIAGRAM DVDD1 AGND1 DGND1 ADCPDB AIN D0 ADC T/H D7 ADCCLK DA0 DA7 DAC A REGISTER PARALLEL DAC A DACA DAC B REGISTER PARALLEL DAC B DACB DACCLK DB0 DB7 DACPDB SDATA SCLK SERIAL CONTROL LOGIC DAC 0 REGISTER SERIAL DAC 0 SDAC0F DAC 1 REGISTER SERIAL DAC 1 SDAC1F SDAC0S SDAC1S LATCH VREF SDACPDB 2.5V REFERENCE GENERAL DESCRIPTION The AD7339 is a composite IC that contains both DAC and ADC functions. The device includes an 8-bit parallel A-to-D converter. Two 8-bit parallel DACs are also included as are two serial control DACs. These serial DACs are 8-bit DACs. The AD7339, which operates with a single 5 V power supply, has a bandgap reference on board with a nominal value of 2.5 V. To reduce the power consumption of the part, each section, except the reference, can be individually powered down when not in use. VREFA AVDD AD7339 DVDD2 DVDD3 VREFB DGND2 DGND3 AGND2 AGND3 The AD7339 is available in a 52-lead PQFP package. REV. 0 Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781/329-4700 World Wide Web Site: http://www.analog.com Fax: 781/326-8703 (c) Analog Devices, Inc., 1997 = DVDD = +5 V 6 10%, AGND = DGND = 0 V, T = T AD7339-SPECIFICATIONS1 (AVDD wise noted) A Parameter B Version Units ADC Resolution Differential Nonlinearity Integral Nonlinearity Zero Input Offset Error Signal Range 8 1 1 3 1 Bits LSB max LSB max LSB V max 1.024 2.048 42.7 6.8 48 4.7 x 1011 5 Offset Binary MHz MSPS dB min Bits min dB min 8 1 1 VBIAS VSWING 14/25 x VREFA/B VREFA/B 2.304 40 5 50 46 0.2 Bits LSB max LSB max Full Power Input Bandwidth Conversion Rate Signal to (Noise + Distortion) Effective No. of Bits (ENOB) Intermodulation Distortion Error Rate Input Capacitance Coding PARALLEL DACS Resolution Differential Nonlinearity Integral Nonlinearity Output Signal Range VSWING VBIAS Update Rate Bipolar Zero Offset Error Gain Error Output Harmonic Content in Band 0 MHz to 1.152 MHz Gain Matching Between DACs Crosstalk To B Channel from A Channel To A Channel from B Channel To VREFB from A Channel To VREFA from B Channel Load Resistance Load Capacitance Full-Scale Settling Time Coding SERIAL DACS Resolution Differential Nonlinearity Integral Nonlinearity Output Range 00H FFH Update Rate Load Resistance Load Capacitance ISINK ISOURCE Full-Scale Settling Time Coding MIN to TMAX, unless other- Test Conditions/Comments ADCCLK = 2.048 MHz 8 Bits Monotonic The input must be biased about 1.4 V. Therefore, ac coupling with a 1 nF capacitor is needed if the bias voltage does not equal 1.4 V. The input should be driven with a maximum source impedance of 50 . See Terminology pF max 00H to FFH with 80H = 0 V DACCLK = 2.304 MHz V nom V nom MHz max mV max % typ dB min dB min dB 55 55 55 55 1.8 50 4 Offset Binary dB min dB min dB min dB min k min pF max s typ 8 1 1.5 Bits LSB LSB 0.2 AVDD - 0.247 SCLK/10 20 100 1 100 2.5 Straight Binary V max V min kHz max k max pF max mA typ A typ s typ 8 Bits Monotonic VREFA/B means VREFA for DACA and VREFB for DACB. Factory Trim. Does Not Include Gain Error For a Full-Scale Digital Sine Wave in Band 0 kHz to 76.8 kHz For a Full-Scale Digital Sine Wave in Band 0 kHz to 128 kHz For Amplitudes Which Equal Full Scale -10 dB 1.8 k Load Between DACA and VREFA, and Between DACB and VREFB A Channel has a full-scale output of frequency 128 kHz. B Channel has a full-scale output of frequency 128 kHz. A Channel has a full-scale output of frequency 128 kHz. B Channel has a full-scale output of frequency 128 kHz. Connected Between DACA/B and VREFA/B 00H to FFH with 80H = Bias Voltage SCLK is a gated 256 kHz clock. -2- 8 Bits Monotonic With Respect to Full Scale See Figure 1 When AVDD > 5.247 V, the analog output will equal 2 VREF. REV. 0 AD7339 Parameter B Version Units REFERENCE VREF Voltage VREFA/VREFB Voltage Load Capacitance 2.5 2% 2.5 5% 0.1 1 1 V min/max V min/max F max Each reference output must have a load capacitance of 100 pF minimum for compensation purposes. mA max mA max LOGIC INPUTS VINH, Input High Voltage VINL, Input Low Voltage IINH, Input Leakage Current CIN, Input Capacitance DVDD - 0.8 0.8 10 15 V min V max A max pF max LOGIC OUTPUTS VOH, Output High Voltage VOL, Output Low Voltage COUT, Output Capacitance DVDD - 0.4 0.4 15 V min V max pF max 4.5/5.5 45 4.5 5 V min/max mA max Active Mode mA max +25C. No Load on VREF mA max -40C to +85C. No Load on VREF ISINK ISOURCE POWER SUPPLIES AVDD, DVDD IDD Power-Down Current Test Conditions/Comments |IOUT| 1 mA |IOUT| 2 mA NOTES 1 Operating temperature range is as follows: B Version; -40C to +85C. Specifications subject to change without notice. 2VREF OUTPUT VOLTAGE - Volts +5.5V POWER SUPPLY 4.753 +5V POWER SUPPLY 4.253 +4.5V POWER SUPPLY 0.2 0 217 243 ANALOG OUTPUT VOLTAGE 255 Figure 1. Analog Output Voltage from Serial DACs vs. Power Supply REV. 0 -3- AD7339 TIMING CHARACTERISTICS Parameter ADC t1 t2 t3 t4 t5 PARALLEL DACS t6 t7 t8 t9 t10 t11 t12 SERIAL DACS t13 t14 t15 t16 t17 t18 t19 (AVDD = +5 V 6 10%; AGND = DGND = 0 V; TA = TMlN to TMAX, unless otherwise noted) Limit at TA = -408C to +858C Units 480 210 210 100 200 ns min ns min ns min ns min ns min 430 200 200 130 50 150 250 ns min ns min ns min ns min ns min ns max ns max 3.9 1.94 1.94 950 950 480 100 s min s min s min ns min ns min ns min s max Description See Figure 3. ADCCLK Period ADCCLK Width Low ADCCLK Width High Data Valid After Falling Edge of ADCCLK Data Valid Before Subsequent Falling Edge of ADCCLK See Figure 4. DACCLK Period DACCLK Width Low DACCLK Width High Data Setup Before DACCLK Rising Edge Time Data Hold After DACCLK Rising Edge Time Propagation Delay Settling Time (from 10% to 90%) See Figure 5. SCLK Period SCLK Width Low SCLK Width High Data Setup Before SCLK Rising Edge Latch Enable Setup Time After SCLK Falling Edge LATCH Pulsewidth Conversion Delay 2mA TO OUTPUT PIN IOL +2.1V CL 15pF 1mA IOH Figure 2. Load Circuit for Timing Specifications -4- REV. 0 AD7339 SAMPLE N-2 SAMPLE N-1 SAMPLE N SAMPLE N+1 SAMPLE N+2 t2 ADCCLK t3 t4 t1 N-3 D0 - D7 N-2 t5 N-1 N N+1 Figure 3. ADC Timing t6 t7 DACCLK t8 t9 DA0 - DA7 DB0 - DB7 DATA t10 DATA t11 DACA DACB 10% 90% t12 Figure 4. Parallel DACs Timing t 14 SCLK t 15 t 13 SDATA D1 t 16 D0 D9(MSB) D8 t 17 LATCH t 18 t 19 SDAC0S SDAC1S 10% Figure 5. Serial DACs Timing REV. 0 -5- D7 AD7339 ABSOLUTE MAXIMUM RATINGS 1 NOTES 1 Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only; functional operation of the device at these or any other conditions above those listed in the operational sections of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. 2 Transient currents of up to 100 mA will not cause SCR latchup. (TA = +25C unless otherwise noted) AVDD, DVDD to GND . . . . . . . . . . . . . . . . . -0.3 V to +7 V AGND to DGND . . . . . . . . . . . . . . . . . . . . . -0.3 V to +0.3 V Digital I/O Voltage to DGND . . . . . . -0.3 V to VDD + 0.3 V ADC Analog Input Voltage . . . . . . . . . . . . . . . . . . . . . . . 2 V Input/Output Current at any Pin Except Supplies2 . . . 20 mA Operating Temperature Range Industrial (B Version) . . . . . . . . . . . . . . . . . -40C to +85C Storage Temperature Range . . . . . . . . . . . . -65C to +150C Maximum Junction Temperature . . . . . . . . . . . . . . . . +150C PQFP, JA Thermal Impedance . . . . . . . . . . . . . . . . . 90C/W Lead Temperature, Soldering Vapor Phase (60 sec) . . . . . . . . . . . . . . . . . . . . . . . . +215C Infrared (15 sec) . . . . . . . . . . . . . . . . . . . . . . . . . . . . +220C ORDERING GUIDE Model Temperature Range Package Description Package Option AD7339BS -40C to +85C Plastic Quad Flatpack (PQFP) S-52 CAUTION ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on the human body and test equipment and can discharge without detection. Although this device features proprietary ESD protection circuitry, permanent damage may occur on devices subjected to high energy electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance degradation or loss of functionality. -6- WARNING! ESD SENSITIVE DEVICE REV. 0 AD7339 SDAC1S DACB VREFB DACA DA0 VREFA DA1 DA2 DA3 DA4 DA5 DA6 DA7 PIN CONFIGURATION 52 51 50 49 48 47 46 45 44 43 42 41 40 DACCLK 1 39 SDAC1F 38 SDAC0S DB0 3 37 SDAC0F DB1 4 36 AGND1 DB2 5 35 VREF PIN 1 IDENTIFIER DVDD1 2 DB3 6 AD7339 34 AGND3 DB4 7 TOP VIEW (Not to Scale) 33 AVDD 32 AGND2 DB6 9 31 AIN DB7 10 30 ADCPDB DACPDB 11 29 DVDD3 DGND1 12 28 DGND3 SDACPDB 13 27 ADCCLK DB5 8 D0 D1 D2 D3 DVDD2 DGND2 D4 D5 D6 D7 SDATA SCLK LATCH 14 15 16 17 18 19 20 21 22 23 24 25 26 PIN FUNCTION DESCRIPTIONS Pin Number Mnemonic Function Power Supply 33 2 12 36 22 29 21 28 32 34 ADCs AVDD DVDD1 DGND1 AGND1 DVDD2 DVDD3 DGND2 DGND3 AGND2 AGND3 Analog power supply connection. Digital power supply for the parallel DACs. Digital ground connection for the parallel DACs. Analog ground connection for the parallel DACs. Digital power supply for the ADC. Digital power supply for the ADC. Digital ground connection for the ADC. Digital ground connection for the ADC. Analog ground connection for the ADC. Analog ground connection for the reference. 31 AIN Analog input to the ADC. The analog input must be appropriately ac coupled. The AD7339 can accept an analog input of 1 V maximum. 27 ADCCLK ADC Input Clock, CMOS Logic Input. The analog input is sampled on the rising edge of ADCCLK. ADCCLK is nominally set to 2.048 MHz. 26-23, 20-17 D0-D7 30 ADCPDB Digital Output from the ADC. The 8-bit digital word from the ADC is in offset binary. The digital output uses CMOS logic. Digital Input. When ADCPDB is low, the ADC is powered down. While in this mode, ADCCLK should be tied low. The ADC is powered up by taking ADCPDB high. Parallel DACs 45-52 DA0-DA7 3-10 DB0-DB7 1 DACCLK 43, 41 DACA, DACB REV. 0 Digital input to the parallel A DAC. The digital input uses CMOS logic and the word is presented to the DAC in offset binary format. Digital input to the parallel B DAC. The digital input uses CMOS logic and the word is presented to the DAC in offset binary format. Input clock to the parallel DACs. The digital words in the A and B DAC registers are loaded into the DACs on the rising edge of DACCLK. DACCLK has a nominal frequency of 2.304 MHz and uses CMOS logic. Analog outputs from the A and B DACs. Both DACs have an analog output of VREFA/ VREFB X volts where VREFA = VREFB = 2.5 V nominal and X = 1.4 V. -7- AD7339 Pin Number Mnemonic Function 11 DACPDB Digital Input. The parallel DACs, VREFA and VREFB, can be powered down using pin DACPDB. When DACPDB is low, both of the parallel DACs and the VREFA/VREFB outputs are placed in a standby mode, drawing a minimal current. The reference, which is available on the VREF pin, is not powered down. Serial DACs 16 SDATA 14 SCLK 15 LATCH 38 SDAC0S 37 SDAC0F 40 SDAC1S Serial Input Data. Serial data is latched into the AD7339 registers on the rising edge of SCLK. The digital data uses CMOS logic. Data is loaded into the latches in 10-bit bursts (MSB first), the 2 MSBs of the word indicating the DAC to which the digital word is being loaded while the 8 LSBs contain the digital word being loaded into the DAC. The serial DACs use offset binary. Serial Input Clock. Data is latched into the registers on the rising edge of SCLK, which is nominally set to 256 kHz. SCLK is a gated clock--the clock should be active only when data is being loaded into the latches. The clock should idle low between conversions. Latch Enable Input. LATCH is used to load the digital data from the latch into the DAC and begin conversion. Both DACs are loaded with the digital data in their respective latches. LATCH is pulsed high to load the DACs, the DACs being loaded on the rising edge of LATCH. Analog Output from Serial DAC0. The analog output from this DAC will have a value of 0.2 V to AVDD - 0.247 V. Feedback Analog Input. By connecting a resistor between SDAC0F and SDAC0S, the gain of the DAC0 buffer can be altered and the magnitude of the analog output adjusted accordingly. Analog Output from Serial DAC1. The analog output from this DAC will have a value of 0.2 V to AVDD - 0.247 V. 39 SDAC1F 13 SDACPDB Reference 35 VREF 44, 42 VREFA/VREFB Feedback Analog Input. By connecting a resistor between SDAC1F and SDAC1S, the gain of the DAC1 buffer can be altered and the magnitude of the analog output adjusted accordingly. Digital Input. The serial DACs are powered down using SDACPDB. When this pin is tied low, the serial DACs are placed in standby mode. The onboard bandgap reference is available on the VREF pin. The reference has a value of 2.5 V nominal. A bypass capacitor of 0.1 F is required between VREF and AGND. This output cannot be powered down. A buffered version of the reference is available on VREFA/VREFB. The analog outputs from the parallel DACs are biased about the reference voltage. DACA is biased about VREFA while DACB is biased about VREFB. VREFA and VREFB can be used with DACA and DACB to provide differential analog inputs to the circuitry connected to the DACs. These outputs are powered down using DACPDB. These outputs should be decoupled using a capacitance of 100 pF minimum. -8- REV. 0 AD7339 Parallel DACs FUNCTIONAL DESCRIPTION A-to-D Converter The A/D conversion circuitry consists of a track-and-hold amplifier followed by a flash A-to-D converter. Figure 6 shows the architecture of the ADC. AD7339 AIN T/H COMPARATOR NETWORK HOLD DECODE LOGIC OUTPUT REGISTER OUTPUT DRIVERS D7 D6 D5 D4 D3 D2 D1 D0 The circuitry for each parallel DAC consists of a current source DAC followed by a buffer that converts the current to a voltage. Figure 8 shows the functional block diagram for the parallel DACs. The loading of both the A and B DAC is controlled by the DACCLK signal, which is nominally set to 2.304 MHz. The digital input to each DAC is latched in on the rising edge of the DACCLK signal so that both DACs simultaneously perform the D-to-A conversion. AD7339 DA0 DA7 REFERENCE DAC A DAC B REGISTER DAC B DACA DACCLK TIMING AND CONTROL LOGIC RESISTOR LADDER DAC A REGISTER DB0 DB7 ADCCLK ADCPDB DACB REFERENCE Figure 6. ADC Architecture VREFA Track-and-Hold Amplifier CONTROL LOGIC The track-and-hold amplifier on the analog input of the AD7339's ADC allows the ADC to accurately convert input frequencies to 8-bit accuracy. The input bandwidth of the track-and-hold amplifier is much greater than the Nyquist rate of the ADC. The operation of the track-and-hold is essentially transparent to the user. The track-and-hold amplifier goes from its tracking mode to its hold mode on the rising edge of ADCCLK. Analog Input The ADC accepts an analog input of 2 V p-p. The analog input is biased about 1.4 V internally. If the signal applied to the ADC is biased about 1.4 V, then dc coupling can be used. AC coupling is needed if the analog input is biased about any voltage other than 1.4 V. A capacitor of 1 nF is suitable for ac coupling. Figure 7 shows the ideal input/output transfer function for the ADC. The designed code transitions occur midway between successive integer LSB values (1/2 LSB, 3/2 LSB, 5/2 LSB . . .) with 1 LSB = FS/256 = 2 V/256 = 7.8 mV. ADC OUTPUT CODE 11111111 AD7339 ADC 11111110 10000010 10000001 -1V 10000000 +1V - 1LSB 01111111 01111110 DACPDB Figure 8. Parallel DACs Functional Block Diagram The analog output from each DAC is biased about the reference voltage VREFA (DAC A) or VREFB (DAC B). The analog output is 1.4 V about the reference voltage. Since the analog outputs are biased about the reference voltage, the reference outputs can be used with the analog outputs to form a differential signal for the circuitry that follows the DACs. The AD7339 includes a calibration feature that reduces the offset between the DAC output bias voltage and the VREFA/ VREFB voltage. A 4-bit offset nulling feature is used to factory trim the offset. The device also has a 4-bit offset register that is user controlled; i.e., the user can disable the factory trimmed offset and use the 4-bit register instead. This allows the user to calibrate out the system offset; however, the user is also responsible for calibrating out the AD7339 offset. The 4-bit offset register is accessed via the serial interface that is used by DAC 0 and DAC 1. Table III gives the addresses for accessing these registers. D5 of the 10-bit data word enables the user to write to the 4-bit offset register. When this bit is set to 0, the factory trimmed value is used as the offset value, while the user programmed value is used when D5 equals 1. When the offset is user controlled, D4 is used to inform the AD7339 to reduce or increase the DAC output voltage. When D4 equals 0, the DAC output is reduced, while the DAC output is increased when D4 equals 1. When user trimming is being used, the 4-bit word to be loaded into the register is contained in the 4 LSBs of the 10-bit word being written to the serial port. 00000001 00000000 0V ANALOG INPUT VOLTAGE - AIN Figure 7. ADC Transfer Function REV. 0 VREFB -9- AD7339 The 8-bit word is loaded into the DAC from the register using LATCH. Data is loaded into the DACs on the falling edge of LATCH. When the D-to-A conversion is performed, the analog output is altered accordingly. The analog output will remain valid until the next falling edge of LATCH, at which stage the next digital word in the register is converted. LATCH is normally low, the input being pulsed to load the DACs, the DACs being loaded on the falling edge of LATCH. The 4-bit offset nulling feature has a LSB size of 7.6 mV; thereby, allowing the user to vary the DAC output by 115 mV. Table I. Writing to the Parallel DACs Offset Registers D9 D8 D7 D6 D5 Address X X Factory/ User Offset D4 D3 D2 D1 D0 Decr/ Incr Data Word The analog output is available on the SDAC0S/SDAC1S pin. Each DAC has an analog output of 0.2 V to AVDD - 0.247 V, an input of 00H generating an analog output of 0.2 V while a digital input of FFH produces an analog output of AVDD - 0.247 V, i.e., the serial DACs use straight binary coding. The analog output is generated by the on board reference. Therefore, when AVDD is greater than 5.247 V, VOUT = 2 VREF when the digital word equals all 1s. However, when AVDD is less than 5.247 V, the output is limited to 0.247 V below AVDD as the amplifier clips the output. The DACs use offset binary coding with 1 LSB = FS/256 = 2.8/256 = 10.94 mV. Table II shows the ideal input code to output voltage relationship. Table II. Ideal Input/Output Code Table DAC Latch Contents MSB LSB 00000000 00000001 01111111 10000000 10000001 11111110 11111111 Analog Output, VOUT* -1.4 V -1.38906 V -0.01094 V 0V +0.01094 V +1.37812 V +1.38906 V The output from the current source is converted to a voltage using an operational amplifier. The amplifier is configured to gain the signal by two; however, the gain of the amplifier can be adjusted by tying a resistor between SDAC0F/SDAC1F and SDAC0S/SDAC1S. The resistors on board the AD7339 have a value of 20 k. Power-Down *These are the nominal output voltages with V OUT = 1.4 V. Serial DACs The AD7339 has two serial DACs on board. The serial DACs have an architecture similar to the parallel DACs. The 8-bit digital word to each DAC is serially loaded. The serial DACs have a common serial port. To distinguish between the two DACs, 10-bit bursts are transferred to the DACs, the two MSBs identifying the DAC to which the 8-bit word is to be loaded. Table III shows the truth table for the two MSBs. The serial word is loaded into the serial register using SDATA and SCLK. SCLK is a gated clock of nominal value 256 kHz, which should be active only when the 10-bit word is being loaded into the register; i.e., SCLK should consist of 10 pulses. If SCLK is continuous, or if it consists of more than 10 pulses, the data shifted into the serial register will be shifted out of the serial register so the register will not contain valid data. When the serial register is not being written to, SCLK should idle low. The serial data bits are read into the serial register on the rising edge of SCLK, the two MSBs of the word identifying the DAC to which the word is being written, and the eight LSBs of the 10-bit word containing the 8-bit word to be converted, the 8-bit word being transferred MSB first. SDATA idles low. Table III. Serial DACs Truth Table D9 D8 DAC to be Written to 0 0 1 1 0 1 0 1 DAC A Offset Register Is Loaded DAC 1 Register Is Loaded DAC 0 Register Is Loaded DAC B Offset Register Is Loaded Each section of the AD7339 can be individually powered down. The ADC, parallel DACs and serial DACs have individual power-down pins, which allows each section to be powered down when it is not being used, thus minimizing the current consumption of the AD7339. Pin ADCPDB is used to place the ADC in sleep mode. When this pin is taken low, the ADC is powered down. For normal operation, ADCPDB is high. When the parallel DACs are not being used, they can be placed in power-down mode using DACPDB. When DACPDB is low, both DACs are powered down. The reference outputs VREFA and VREFB are also powered down. During power-down, the analog outputs DACA and DACB, as well as the reference outputs, are pulled down to ground. When the DACs are powered up, the analog outputs settle to the bias voltage VREFA/VREFB. The serial DACs are powered down using SDACPDB. When this pin is tied low, the serial DACs are placed in sleep mode. When a converter is powered up, 100 s are required for the analog and digital circuitry to settle. Conversions can commence when the circuitry has settled. The reference on board the AD7339 is permanently powered up. While the outputs VREFA and VREFB can be powered down, the reference voltage, which is available on pin VREF, is always available. -10- REV. 0 AD7339 TERMINOLOGY Differential Nonlinearity The difference between the measured and the ideal 1 LSB change between any two adjacent codes in the ADC or DACs. A specified Differential Nonlinearity of 1 LSB max over the operating temperature range ensures monotonicity. Integral Nonlinearity The maximum deviation of any code from a straight line passing through the endpoints of the transfer function. The endpoints of the transfer function are zero scale, a point 0.5 LSB below the first code transition (000 . . . 00 to 000 . . . 01) and full scale, a point 0.5 LSB above the last code transition (111 . . . 10 to 111 . . . 11). The error is expressed in LSBs. Signal to (Noise + Distortion) Signal to (Noise + Distortion) is measured signal-to-noise at the output of the ADC. The signal is the rms magnitude of the fundamental. Noise is the rms sum of all nonfundamental signals up to half the sampling frequency (FS/2) excluding dc. Signal to (Noise + Distortion) is dependent on the number of quantization levels used in the digitization process; the more levels, the smaller the quantization noise. The theoretical signal to (Noise + Distortion) ratio for a sine wave signal is given by Signal to (Noise + Distortion) = (6.02N + 1.76) dB where N is the number of bits. Thus for an ideal 8-bit converter, Signal to (Noise + Distortion) is 49.92 dB. products, of order (m + n), at sum and difference frequencies of mfa nfb, where m, n = 0, 1, 2, 3, . . .. Intermodulation terms are those for which m or n is not equal to zero. For the AD7339, the Intermodulation Distortion is the level to which the second and third intermodulation terms are suppressed below a full scale output signal level, the second order terms being (fa + fb) and (fa - fb) while the third order terms are (2fa + fb), (2fa - fb), (fa + 2fb) and (fa - 2fb). Error Rate The rate at which A-to-D conversion errors occur. DACS Bipolar Zero Offset Error The deviation between the measured output voltage and the bias voltage (VREFA or VREFB, depending on which DAC is being tested) when the DAC is loaded with code 100 . . . 00 after gain error has been adjusted out. Gain Error A measure of the output error between an ideal DAC and the actual device output with all 1s loaded after offset error has been adjusted out. Update Rate The rate at which the DACs can be loaded with new data. The parallel DACs have an update rate of 2.304 MHz while the serial DACs have an update rate of 256/10 kHz maximum. Gain Matching Between DACs ADC Effective Number of Bits (ENOB) Signal to (Noise + Distortion) is expressed in dBs; rewriting the Signal to (Noise + Distortion) formula, it is possible to get a measure of performance expressed in effective number of bits. The effective number of bits for a device can be calculated directly from its measured Signal to (Noise + Distortion) value. ENOB = (SNR - 1.76)/6.02 where SNR is the Signal to (Noise + Distortion). Zero Input Offset Error This is the offset error in the ADC when the analog input is zero. Ideally, the digital output should equal 100 . . . 00. The offset error is the deviation from the ideal output code. The offset error is expressed in LSBs. Intermodulation Distortion With inputs consisting of sine waves at two frequencies, fa and fb, any active device with nonlinearities will create distortion REV. 0 The matching between the analog output amplitudes of the parallel DACs when the same digital word is written to each DAC. Crosstalk The ratio of the amplitude of a full-scale signal appearing on one channel to the amplitude of the same signal which couples onto another channel. Crosstalk is expressed in dBs. Output Harmonic Content When the digital word is converted to analog form, harmonics will also be generated. The Output Harmonic Content specifies the amount by which these harmonics are attenuated relative to the fundamental frequency. With the parallel DACs, a full sine wave of frequency 0 kHz to 128 kHz is input. The resulting analog output is evaluated and the amount by which the harmonics in the frequency band 0 MHz to 1.1152 MHz are attenuated is measured relative to the magnitude of the fundamental output signal. -11- crossover of digital and analog signals. Traces at opposite sides of the board should run at right angles to each other. This will reduce the effects of feedthrough through the board. A microstrip technique is by far the best, but is not always possible with a double-sided board. In this technique, the component side of the board is dedicated to ground planes while signals are placed on the other side. Good decoupling is important. The analog and digital supplies to the AD7339 are independent and separately pinned out to minimize coupling between analog and digital sections of the device. All analog and digital supplies should be decoupled to AGND and DGND respectively using 0.1 F ceramic capacitors in parallel with 10 F tantalum capacitors. To achieve the best from the decoupling capacitors, they should be placed as close as possible to the device, ideally right up against the device. In systems where a common supply is used to drive both the AVDD and DVDD of the AD7339, it is recommended that the system's AVDD supply be used. This supply should have the recommended analog supply decoupling between the AVDD pins of the AD7339 and AGND and the recommended digital supply decoupling capacitors between the DVDD pins and DGND. GROUNDING AND LAYOUT The printed circuit board that houses the AD7339 should be designed so that the analog and digital sections are separated and confined to certain areas of the board. This facilitates the use of ground planes that can be easily separated. A minimum etch technique is generally best for ground planes as it gives the best shielding. Digital and analog ground planes should be joined in only one place. If the AD7339 is the only device requiring an AGND-toDGND connection, the ground planes should be connected at the AGND and DGND pins of the AD7339. If the AD7339 is in a system where multiple devices require AGND-to-DGND connections, the connection should be made at one point only, a star point that should be established as close as possible to the AD7339. Avoid running digital lines under the device as these will couple noise onto the die. The analog ground plane should be allowed to run under the AD7339 to avoid noise coupling. The power supply lines to the AD7339 should use as large a track as possible to provide low impedance paths and reduce the effects of glitches on the power supply line. Fast switching signals such as clocks should be shielded with digital ground to avoid radiating noise to other sections of the board. Avoid C3049-8-10/97 AD7339 OUTLINE DIMENSIONS Dimensions shown in inches and (mm). 52-Lead Plastic Quad Flatpack (PQFP) (S-52) 0.557 (14.15) 0.537 (13.65) 0.398 (10.11) 0.390 (9.91) 0.094 (2.39) 0.084 (2.13) 0.037 (0.95) 0.026 (0.65) 52 1 40 39 0.398 (10.11) 0.390 (9.91) 0.557 (14.15) 0.537 (13.65) PIN 1 TOP VIEW (PINS DOWN) 0.012 (0.30) 0.006 (0.15) 13 14 0.008 (0.20) 0.006 (0.15) 0.082 (2.09) 0.078 (1.97) 0.0256 (0.65) BSC -12- 27 26 PRINTED IN U.S.A. SEATING PLANE 0.014 (0.35) 0.010 (0.25) REV. 0