Preliminary Technical Data Complete Quad, 16-Bit, High Accuracy, Serial Input, ±5V DACs AD5765 FEATURES GENERAL DESCRIPTION Complete quad, 16-bit digital-to-analog converters (DACs) Programmable output range: ±4.096 V, ±4.201 V, or ±4.311 V ±1 LSB maximum INL error, ±1 LSB maximum DNL error Low noise: 60 nV/√Hz Settling time: 10 µs maximum Integrated reference buffers On-chip die temperature sensor Output control during power-up/brownout Programmable short-circuit protection Simultaneous updating via LDAC Asynchronous CLR to zero code Digital offset and gain adjust Logic output control pins DSP-/microcontroller-compatible serial interface Temperature range: −40°C to +105°C iCMOS® process technology1 The AD5765 is a quad, 16-bit, serial input, bipolar voltage output digital-to-analog converter that operates from supply voltages of ±4.75 V up to ±5.25 V. Nominal full-scale output range is ±4.096 V. The AD5765 provides integrated output amplifiers, reference buffers and proprietary power-up/powerdown control circuitry. The part also features a digital I/O port, which is programmed via the serial interface. The parts incorporate digital offset and gain adjust registers per channel. APPLICATIONS Industrial automation Open-/closed-loop servo control Process control Data acquisition systems Automatic test equipment Automotive test and measurement High accuracy instrumentation The AD5765 is a high performance converter that offers guaranteed monotonicity, integral nonlinearity (INL) of ±1 LSB, low noise, and 10 µs settling time. During power-up (when the supply voltages are changing), the outputs are clamped to 0 V via a low impedance path. The AD5765 uses a serial interface that operates at clock rates of up to 30 MHz and is compatible with DSP and microcontroller interface standards. Double buffering allows the simultaneous updating of all DACs. The input coding is programmable to either twos complement or offset binary formats. The asynchronous clear function clears all DAC registers to either bipolar zero or zero scale depending on the coding used. The AD5765 is ideal for both closed-loop servo control and open-loop control applications. The AD5765 is available in a 32-lead TQFP, and offers guaranteed specifications over the −40°C to +105°C industrial temperature range. See Figure 1, the functional block diagram. Table 1. Related Devices Part No. AD5764 AD5763 1 Description Complete quad, 16-bit, high accuracy, serial input, ±10V output DAC Complete dual, 16-bit, high accuracy, serial input, ±5V DAC For analog systems designers within industrial/instrumentation equipment OEMs who need high performance ICs at higher voltage levels, iCMOS is a technology platform that enables the development of analog ICs capable of 30 V and operating at ±15 V supplies, allowing dramatic reductions in power consumption and package size, and increased ac and dc performance. Rev. PrA 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 that may result from its use. Specifications subject to change without notice. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Trademarks and registered trademarks are the property of their respective owners. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781.329.4700 www.analog.com Fax: 781.461.3113 ©2007 Analog Devices, Inc. All rights reserved. AD5765 Preliminary Technical Data TABLE OF CONTENTS Features .............................................................................................. 1 Function Register ....................................................................... 21 Applications....................................................................................... 1 Data Register............................................................................... 21 General Description ......................................................................... 1 Coarse Gain Register ................................................................. 21 Revision History ............................................................................... 2 Fine Gain Register...................................................................... 22 Functional Block Diagram .............................................................. 3 Offset Register ............................................................................ 22 Specifications..................................................................................... 4 Offset and Gain Adjustment Worked Example...................... 23 AC Performance Characteristics .................................................... 6 AD5765 Features ............................................................................ 24 Timing Characteristics..................................................................... 7 Analog Output Control ............................................................. 24 Absolute Maximum Ratings.......................................................... 10 Digital Offset and Gain Control............................................... 24 ESD Caution................................................................................ 10 Programmable Short-Circuit Protection ................................ 24 Pin Configuration and Function Descriptions........................... 11 Digital I/O Port........................................................................... 24 Typical Performance Characteristics ........................................... 13 Local Ground Offset Adjust...................................................... 24 Terminology .................................................................................... 17 Applications Information .............................................................. 25 Theory of Operation ...................................................................... 18 Typical Operating Circuit ......................................................... 25 DAC Architecture....................................................................... 18 Layout Guidelines........................................................................... 26 Reference Buffers........................................................................ 18 Galvanically Isolated Interface ................................................. 26 Serial Interface ............................................................................ 18 Microprocessor Interfacing....................................................... 26 Simultaneous Updating via LDAC ........................................... 19 Evaluation Board ........................................................................ 27 Transfer Function ....................................................................... 20 Outline Dimensions ....................................................................... 28 Asynchronous Clear (CLR)....................................................... 20 Ordering Guide .......................................................................... 28 REVISION HISTORY Preliminary Version: PrA December 11, 2007 Rev. PrA | Page 2 of 31 Preliminary Technical Data AD5765 FUNCTIONAL BLOCK DIAGRAM PGND AVDD AVSS AVDD AVSS REFGND DVCC DGND REFERENCE BUFFERS AD5765 16 SDIN SCLK SYNC INPUT SHIFT REGISTER AND CONTROL LOGIC INPUT REG A DAC REG A RSTOUT REFAB RSTIN VOLTAGE MONITOR AND CONTROL ISCC 16 G1 DAC A VOUTA G2 GAIN REG A AGNDA OFFSET REG A SDO INPUT REG B DAC REG B 16 G1 DAC B VOUTB G2 GAIN REG B AGNDB OFFSET REG B D0 D1 INPUT REG C DAC REG C 16 G1 DAC C VOUTC G2 GAIN REG C AGNDC OFFSET REG C BIN/2sCOMP INPUT REG D DAC REG D 16 G1 DAC D VOUTD G2 CLR GAIN REG D AGNDD OFFSET REG D LDAC Figure 1. Rev. PrA | Page 3 of 31 REFERENCE BUFFERS TEMP SENSOR REFCD TEMP AD5765 Preliminary Technical Data SPECIFICATIONS AVDD = 4.75 V to 5.25 V, AVSS = −4.75 V to −5.25 V, AGNDX = DGND = REFGND = PGND = 0 V; REFAB = REFCD = 2.048 V; DVCC = 2.7 V to 5.25 V, RLOAD = 5 kΩ, CLOAD = 200 pF. All specifications TMIN to TMAX, unless otherwise noted. Table 2. B Grade1 C Grade1 Unit 16 ±2 ±1 ±2 16 ±1 ±1 ±2 Bits LSB max LSB max mV max Bipolar Zero TC2 Zero-Scale Error ±2 ±2 ±2 ±2 ppm FSR/°C max mV max Zero-Scale TC2 Gain Error ±2 ±0.02 ±2 ±0.02 ppm FSR/°C max % FSR max ±2 0.5 ±2 0.5 ppm FSR/°C max LSB max 2.048 1 ±10 1 to 2.1 2.048 1 ±10 1 to 2.1 V nominal MΩ min µA max V min to V max ±1% for specified performance Typically 100 MΩ Typically ±30 nA ±4.311 ±4.42 ±13 ±15 10 ±1 ±4.311 ±4.42 ±13 ±15 10 ±1 V min/V max V min/V max ppm FSR/500 hours typ ppm FSR/1000 hours typ mA typ mA max REFIN = 2.048V REFIN = 2.1V 200 1000 0.3 200 1000 0.3 pF max pF max Ω max Parameter ACCURACY Resolution Relative Accuracy (INL) Differential Nonlinearity Bipolar Zero Error Gain TC2 DC Crosstalk2 REFERENCE INPUT2 Reference Input Voltage DC Input Impedance Input Current Reference Range OUTPUT CHARACTERISTICS2 Output Voltage Range3 Output Voltage Drift vs. Time Short Circuit Current Load Current Capacitive Load Stability RL = ∞ RL = 10 kΩ DC Output Impedance DIGITAL INPUTS2 VIH, Input High Voltage VIL, Input Low Voltage Input Current Pin Capacitance Test Conditions/Comments Outputs unloaded Guaranteed monotonic At 25°C; error at other temperatures obtained using bipolar zero TC At 25°C; error at other temperatures obtained using zero scale TC At 25°C; error at other temperatures obtained using gain TC RISCC = 6 kΩ, see Figure 29 For specified performance DVCC = 2.7 V to 5.25 V, JEDEC compliant 2 0.8 ±1 10 2 0.8 ±1 10 V min V max µA max pF max Rev. PrA | Page 4 of 31 Per pin Per pin Preliminary Technical Data Parameter DIGITAL OUTPUTS (D0, D1, SDO)2 Output Low Voltage Output High Voltage Output Low Voltage Output High Voltage High Impedance Leakage Current High Impedance Output Capacitance POWER REQUIREMENTS AVDD/AVSS DVCC Power Supply Sensitivity2 ∆VOUT/∆ΑVDD AIDD AISS DICC Power Dissipation AD5765 B Grade1 C Grade1 Unit Test Conditions/Comments 0.4 DVCC − 1 0.4 0.4 DVCC − 1 0.4 V max V min V max DVCC − 0.5 DVCC − 0.5 V min ±1 5 ±1 5 µA max pF typ DVCC = 5 V ± 5%, sinking 200 µA DVCC = 5 V ± 5%, sourcing 200 µA DVCC = 2.7 V to 3.6 V, sinking 200 µA DVCC = 2.7 V to 3.6 V, sourcing 200 µA SDO only SDO only 4.75 to 5.25 2.7 to 5.25 4.75 to 5.25 2.7 to 5.25 V min/V max V min/V max −85 1.75 1.38 1.2 138 −85 1.75 1.38 1.2 138 dB typ mA/channel max mA/channel max mA max mW typ 1 Temperature range: -40°C to +105°C; typical at +25°C. Guaranteed by design and characterization; not production tested. 3 Output amplifier headroom requirement is 0.5 V minimum. 2 Rev. PrA | Page 5 of 31 Outputs unloaded Outputs unloaded VIH = DVCC, VIL = DGND, 750 µA typ ±5 V operation output unloaded AD5765 Preliminary Technical Data AC PERFORMANCE CHARACTERISTICS AVDD = 4.75 V to 5.25 V, AVSS = −4.75 V to −5.25 V, AGNDX = DGND = REFGND = PGND = 0 V; REFAB = REFCD = 2.048 V; DVCC = 2.7 V to 5.25 V, RLOAD = 5 kΩ, CLOAD = 200 pF. All specifications TMIN to TMAX, unless otherwise noted. Table 3. Parameter DYNAMIC PERFORMANCE1 Output Voltage Settling Time Slew Rate Digital-to-Analog Glitch Energy Glitch Impulse Peak Amplitude Channel-to-Channel Isolation DAC-to-DAC Crosstalk Digital Crosstalk Digital Feedthrough Output Noise (0.1 Hz to 10 Hz) Output Noise (0.1 Hz to 100 kHz) 1/f Corner Frequency Output Noise Spectral Density Complete System Output Noise Spectral Density2 1 2 A Grade B Grade C Grade Unit Test Conditions/Comments 8 10 2 5 8 25 80 8 2 2 8 10 2 5 8 25 80 8 2 2 8 10 2 5 8 25 80 8 2 2 µs typ µs max µs typ V/µs typ nV-sec typ mV max dB typ nV-sec typ nV-sec typ nV-sec typ Full-scale step to ±1 LSB 0.1 45 1 60 80 0.1 45 1 60 80 0.1 45 1 60 80 LSB p-p typ µV rms max kHz typ nV/√Hz typ nV/√Hz typ Guaranteed by design and characterization; not production tested. Includes noise contributions from integrated reference buffers, 16-bit DAC and output amplifier. Rev. PrA | Page 6 of 31 512 LSB step settling Effect of input bus activity on DAC outputs Measured at 10 kHz Measured at 10 kHz Preliminary Technical Data AD5765 TIMING CHARACTERISTICS AVDD = 4.75 V to 5.25 V, AVSS = −4.75 V to −5.25 V, AGNDX = DGND = REFGND = PGND = 0 V; REFAB = REFCD = 2.048 V; DVCC = 2.7 V to 5.25 V, RLOAD = 5 kΩ, CLOAD = 200 pF. All specifications TMIN to TMAX, unless otherwise noted. Table 4. Parameter1, 2, 3 t1 t2 t3 t4 t54 t6 t7 t8 t9 t10 t11 t12 t13 t14 t155, 6 t16 t17 t18 Limit at TMIN, TMAX 33 13 13 13 13 40 2 5 1.4 400 10 500 10 10 2 25 13 2 170 Unit ns min ns min ns min ns min ns min ns min ns min ns min µs min ns min ns min ns max µs max ns min µs max ns max ns min µs min ns min Description SCLK cycle time SCLK high time SCLK low time SYNC falling edge to SCLK falling edge setup time 24th SCLK falling edge to SYNC rising edge Minimum SYNC high time Data setup time Data hold time SYNC rising edge to LDAC falling edge (all DACs updated) SYNC rising edge to LDAC falling edge (single DAC updated) LDAC pulse width low LDAC falling edge to DAC output response time DAC output settling time CLR pulse width low CLR pulse activation time SCLK rising edge to SDO valid SYNC rising edge to SCLK falling edge SYNC rising edge to DAC output response time (LDAC = 0) LDAC falling edge to SYNC rising edge 1 Guaranteed by design and characterization; not production tested. All input signals are specified with tR = tF = 5 ns (10% to 90% of DVCC) and timed from a voltage level of 1.2 V. See Figure 2, Figure 3, and Figure 4. 4 Standalone mode only. 5 Measured with the load circuit of Figure 5. 6 Daisy-chain mode only. 2 3 Rev. PrA | Page 7 of 31 AD5765 Preliminary Technical Data t1 SCLK 1 2 24 t3 t6 t2 t4 t5 SYNC t8 t7 SDIN DB0 DB23 t10 t9 LDAC t10 t18 t12 t11 VOUTX LDAC = 0 t12 t17 VOUTX t13 CLR t14 05303-002 VOUTX Figure 2. Serial Interface Timing Diagram t1 SCLK 24 t3 t6 48 t2 t5 t16 t4 SYNC t7 SDIN t8 DB23 DB0 INPUT WORD FOR DAC N DB23 DB0 t15 INPUT WORD FOR DAC N–1 DB23 SDO UNDEFINED DB0 INPUT WORD FOR DAC N t9 t10 LDAC Figure 3. Daisy-Chain Timing Diagram Rev. PrA | Page 8 of 31 Preliminary Technical Data AD5765 SCLK 24 48 SYNC DB23 DB0 DB23 DB0 NOP CONDITION INPUT WORD SPECIFIES REGISTER TO BE READ DB23 SDO UNDEFINED DB0 SELECTED REGISTER DATA CLOCKED OUT Figure 4. Readback Timing Diagram 200µA VOH (MIN) OR VOL (MAX) CL 50pF 200µA IOH Figure 5. Load Circuit for SDO Timing Diagram Rev. PrA | Page 9 of 31 05303-005 TO OUTPUT PIN IOL 05303-004 SDIN AD5765 Preliminary Technical Data ABSOLUTE MAXIMUM RATINGS TA = 25°C, unless otherwise noted. Transient currents of up to 100 mA do not cause SCR latch-up. Table 5. Parameter AVDD to AGNDX, DGND AVSS to AGNDX, DGND DVCC to DGND Digital Inputs to DGND Digital Outputs to DGND REFIN to AGNDX, PGND VOUTA, VOUTB, VOUTC, VOUTD to AGNDX AGNDX to DGND Operating Temperature Range (TA) Industrial Storage Temperature Range Junction Temperature (TJ max) Power Dissipation 32-Lead TQFP θJA Thermal Impedance θJC Thermal Impedance Lead Temperature Soldering Rating −0.3 V to +17 V +0.3 V to −17 V −0.3 V to +7 V −0.3 V to DVCC + 0.3 V or 7 V (whichever is less) −0.3 V to DVCC + 0.3 V −0.3 V to AVDD + 0.3V AVSS to AVDD 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. ESD CAUTION −0.3 V to +0.3 V −40°C to +105°C −65°C to +150°C 150°C (TJ max – TA)/ θJA 65°C/W 12°C/W JEDEC Industry Standard J-STD-020 Rev. PrA | Page 10 of 31 Preliminary Technical Data AD5765 REFAB REFCD NC REFGND TEMP AVSS AVDD BIN/2sCOMP PIN CONFIGURATION AND FUNCTION DESCRIPTIONS 32 31 30 29 28 27 26 25 24 AGNDA 23 VOUTA 22 VOUTB 21 AGNDB 20 AGNDC LDAC 6 19 VOUTC D0 7 18 VOUTD D1 8 17 AGNDD SYNC 1 PIN 1 SCLK 2 SDIN 3 AD5765 SDO 4 TOP VIEW (Not to Scale) CLR 5 ISCC AVSS PGND AVDD DVCC DGND RSTIN 10 11 12 13 14 15 16 RSTOUT 9 NC = NO CONNECT Figure 6. Pin Configuration Table 6. Pin Function Descriptions Pin No. 1 Mnemonic SYNC 2 SCLK 3 4 51 6 SDIN SDO CLR1 LDAC 7, 8 D0, D1 9 RSTOUT 10 RSTIN 11 12 13, 31 14 15, 30 16 DGND DVCC AVDD PGND AVSS ISCC 17 18 AGNDD VOUTD 19 VOUTC Description Active Low Input. This is the frame synchronization signal for the serial interface. While SYNC is low, data is transferred in on the falling edge of SCLK. Serial Clock Input. Data is clocked into the shift register on the falling edge of SCLK. This operates at clock speeds up to 30 MHz. Serial Data Input. Data must be valid on the falling edge of SCLK. Serial Data Output. Used to clock data from the serial register in daisy-chain or readback mode. Negative Edge Triggered Input. Asserting this pin sets the DAC registers to 0x0000. Load DAC. Logic input. This is used to update the DAC registers and consequently the analog outputs. When tied permanently low, the addressed DAC register is updated on the rising edge of SYNC. If LDAC is held high during the write cycle, the DAC input register is updated but the output update is held off until the falling edge of LDAC. In this mode, all analog outputs can be updated simultaneously on the falling edge of LDAC. The LDAC pin must not be left unconnected. D0 and D1 form a digital I/O port. The user can set up these pins as inputs or outputs that are configurable and readable over the serial interface. When configured as inputs, these pins have weak internal pull-ups to DVCC. When programmed as outputs, D0 and D1 are referenced by DVCC and DGND. Reset Logic Output. This is the output from the on-chip voltage monitor used in the reset circuit. If desired, it can be used to control other system components. Reset Logic Input. This input allows external access to the internal reset logic. Applying a Logic 0 to this input clamps the DAC outputs to 0 V. In normal operation, RSTIN should be tied to Logic 1. Register values remain unchanged. Digital Ground Pin. Digital Supply Pin. Voltage ranges from 2.7 V to 5.25 V. Positive Analog Supply Pins. Voltage ranges from 4.75 V to 5.25 V. Ground Reference Point for Analog Circuitry. Negative Analog Supply Pins. Voltage ranges from –4.75 V to –5.25 V. This pin is used in association with an optional external resistor to AGND to program the short-circuit current of the output amplifiers. Refer to the AD5765 Features section on page 25 for further details. Ground Reference Pin for DAC D Output Amplifier. Analog Output Voltage of DAC D. Buffered output with a nominal full-scale output range of ±4.096 V. The output amplifier is capable of directly driving a 5 kΩ, 200 pF load. Analog Output Voltage of DAC C. Buffered output with a nominal full-scale output range of ±4.096 V. The output amplifier is capable of directly driving a 5 kΩ, 200 pF load. Rev. PrA | Page 11 of 31 AD5765 Preliminary Technical Data Pin No. 20 21 22 Mnemonic AGNDC AGNDB VOUTB 23 VOUTA 24 25 AGNDA REFAB 26 REFCD 27 28 29 NC REFGND TEMP 32 BIN/2sCOMP 1 Description Ground Reference Pin for DAC C Output Amplifier. Ground Reference Pin for DAC B Output Amplifier. Analog Output Voltage of DAC B. Buffered output with a nominal full-scale output range of ±4.096 V. The output amplifier is capable of directly driving a 5 kΩ, 200 pF load. Analog Output Voltage of DAC A. Buffered output with a nominal full-scale output range of ±4.096 V. The output amplifier is capable of directly driving a 5 kΩ, 200 pF load. Ground Reference Pin for DAC A Output Amplifier. External Reference Voltage Input for Channel A and Channel B. Reference input range is 1 V to 2.1 V; programs the full-scale output voltage. REFIN = 2.048 V for specified performance. External Reference Voltage Input for Channel C and Channel D. Reference input range is 1 V to 2.1 V; programs the full-scale output voltage. REFIN = 2.048 V for specified performance. No Connect. Reference Ground Return for the Reference Generator and Buffers. This pin provides an output voltage proportional to temperature. The output voltage is 1.4V typical at 25°C die temperature; variation with temperature is 5mV/°C. Determines the DAC Coding. This pin should be hardwired to either DVCC or DGND. When hardwired to DVCC, input coding is offset binary. When hardwired to DGND, input coding is twos complement (see Table 7). Internal pull-up device on this logic input. Therefore, it can be left floating and defaults to a logic high condition. Rev. PrA | Page 12 of 31 Preliminary Technical Data AD5765 TYPICAL PERFORMANCE CHARACTERISTICS Figure 7. Integral Nonlinearity Error vs. Code Figure 10. Differential Nonlinearity Error vs. Temperature Figure 8. Differential Nonlinearity Error vs. Code Figure 11. Integral Nonlinearity Error vs. Supply voltage Figure 9. Integral Nonlinearity Error vs. Temperature Figure 12. Differential Nonlinearity Error vs. Supply Voltage Rev. PrA | Page 13 of 31 AD5765 Preliminary Technical Data Figure 13.Integral Nonlinearity Error vs. Reference voltage Figure 16. AIDD/AISS vs. AVDD/AVSS Figure 14. Differential Nonlinearity Error vs Reference Voltage Figure 17. Zero-Scale Error vs. Temperature Figure 15. Total Unadjusted Error vs. Reference Voltage Figure 18. Bipolar Zero Error vs. Temperature Rev. PrA | Page 14 of 31 Preliminary Technical Data AD5765 Figure 19. Gain Error vs. Temperature Figure 22. Source and Sink Capability of Output Amplifier with Negative Full-Scale Loaded Figure 20.DICC vs. Logic Input Voltage Figure 23. Positive Full-Scale Step Figure 21. Source and Sink Capability of Output Amplifier with Positive FullScale Loaded Rev. PrA | Page 15 of 31 Figure 24. Negative Full-Scale Step AD5765 Preliminary Technical Data Figure 25. Settling Time vs. Load Capacitance Figure 28. VOUT vs. AVDD/AVSS on Power-up Figure 26. Major Code Transition Glitch Energy Figure 29. Short-Circuit Current vs. RISCC Figure 27. Peak-to-Peak Noise (100 kHz Bandwidth) Figure 30. TEMP Output Voltage vs. Temperature Rev. PrA | Page 16 of 31 Preliminary Technical Data AD5765 TERMINOLOGY Relative Accuracy or Integral Nonlinearity (INL) For the DAC, relative accuracy or integral nonlinearity (INL) is a measure of the maximum deviation, in LSBs, from a straight line passing through the endpoints of the DAC transfer function. A typical INL vs. code plot can be seen in Figure 7. Differential Nonlinearity (DNL) Differential nonlinearity is the difference between the measured change and the ideal 1 LSB change between any two adjacent codes. A specified differential nonlinearity of ±1 LSB maximum ensures monotonicity. This DAC is guaranteed monotonic. A typical DNL vs. code plot can be seen in Figure 8. Monotonicity A DAC is monotonic if the output either increases or remains constant for increasing digital input code. The AD5765 is monotonic over its full operating temperature range. Bipolar Zero Error Bipolar zero error is the deviation of the analog output from the ideal half-scale output of 0 V when the DAC register is loaded with 0x8000 (offset binary coding) or 0x0000 (twos complement coding). A plot of bipolar zero error vs. temperature can be seen in Figure 18. Bipolar Zero Temperature Coefficient Bipolar zero TC is the measure of the change in the bipolar zero error with a change in temperature. It is expressed in ppm FSR/°C. Full-Scale Error Full-scale error is a measure of the output error when full-scale code is loaded to the DAC register. Ideally the output voltage should be 2 × VREF − 1 LSB. Full-scale error is expressed in percentage of full-scale range. Negative Full-Scale Error/Zero Scale Error Negative full-scale error is the error in the DAC output voltage when 0x0000 (offset binary coding) or 0x8000 (twos complement coding) is loaded to the DAC register. Ideally, the output voltage should be −2 × VREF. A plot of zero-scale error vs. temperature can be seen in Figure 17. Output Voltage Settling Time Output voltage settling time is the amount of time it takes for the output to settle to a specified level for a full-scale input change. Slew Rate The slew rate of a device is a limitation in the rate of change of the output voltage. The output slewing speed of a voltageoutput D/A converter is usually limited by the slew rate of the amplifier used at its output. Slew rate is measured from 10% to 90% of the output signal and is given in V/µs. Gain Error Gain error is a measure of the span error of the DAC. It is the deviation in slope of the DAC transfer characteristic from the ideal, expressed as a percentage of the full-scale range. A plot of gain error vs. temperature can be seen in Figure 19. Total Unadjusted Error Total unadjusted error (TUE) is a measure of the output error considering all the various errors. A plot of total unadjusted error vs. reference can be seen in Figure 15. Zero-Scale Error TC Zero-scale error TC is a measure of the change in zero-scale error with a change in temperature. Zero-scale error TC is expressed in ppm FSR/°C. Gain Error TC Gain error TC is a measure of the change in gain error with changes in temperature. Gain Error TC is expressed in (ppm of FSR)/°C. Digital-to-Analog Glitch Energy Digital-to-analog glitch impulse is the impulse injected into the analog output when the input code in the DAC register changes state. It is normally specified as the area of the glitch in nV-secs and is measured when the digital input code is changed by 1 LSB at the major carry transition (0x7FFF to 0x8000) (see Figure 26). Digital Feedthrough Digital feedthrough is a measure of the impulse injected into the analog output of the DAC from the digital inputs of the DAC but is measured when the DAC output is not updated. It is specified in nV-secs and measured with a full-scale code change on the data bus, that is, from all 0s to all 1s, and vice versa. Power Supply Sensitivity Power supply sensitivity indicates how the output of the DAC is affected by changes in the power supply voltage. DC Crosstalk DC crosstalk is the dc change in the output level of one DAC in response to a change in the output of another DAC. It is measured with a full-scale output change on one DAC while monitoring another DAC, and is expressed in LSBs. DAC-to-DAC Crosstalk DAC-to-DAC crosstalk is the glitch impulse transferred to the output of one DAC due to a digital code change and subsequent output change of another DAC. This includes both digital and analog crosstalk. It is measured by loading one of the DACs with a full-scale code change (all 0s to all 1s and vice versa) with LDAC low and monitoring the output of another DAC. The energy of the glitch is expressed in nV-sec. Channel-to-Channel Isolation Channel-to-channel isolation is the ratio of the amplitude of the signal at the output of one DAC to a sine wave on the reference input of another DAC. It is measured in dB. Digital Crosstalk Digital crosstalk is a measure of the impulse injected into the analog output of one DAC from the digital inputs of another DAC but is measured when the DAC output is not updated. It is specified in nV-secs and measured with a full-scale code change on the data bus, that is, from all 0s to all 1s, and vice versa. Rev. PrA | Page 17 of 31 AD5765 Preliminary Technical Data THEORY OF OPERATION The AD5765 is a quad, 16-bit, serial input, bipolar voltage output DAC and operates from supply voltages of ±4.75 V to ±5.25 V and has a buffered output voltage of up to ±4.311 V. Data is written to the AD5765 in a 24-bit word format, via a 3-wire serial interface. The device also offers an SDO pin, which is available for daisychaining or readback. SERIAL INTERFACE The AD5765 incorporates a power-on reset circuit, which ensures that the DAC registers power up loaded with 0x0000. The AD5765 features a digital I/O port that can be programmed via the serial interface, on-chip reference buffers and per channel digital gain, and offset registers. The input shift register is 24 bits wide. Data is loaded into the device MSB first as a 24-bit word under the control of a serial clock input, SCLK. The input register consists of a read/write bit, three register select bits, three DAC address bits and 16 data bits as shown in Table 8. The timing diagram for this operation is shown in Figure 2. The AD5765 is controlled over a versatile 3-wire serial interface that operates at clock rates of up to 30 MHz and is compatible with SPI®, QSPI™, MICROWIRE™, and DSP standards. Input Shift Register DAC ARCHITECTURE The DAC architecture of the AD5765 consists of a 16-bit current mode segmented R-2R DAC. The simplified circuit diagram for the DAC section is shown in Figure 31. The four MSBs of the 16-bit data word are decoded to drive 15 switches, E1 to E15. Each of these switches connects one of the 15 matched resistors to either AGNDX or IOUT. The remaining 12 bits of the data-word drive switches S0 to S11 of the 12-bit R-2R ladder network. R VREF 2R 2R 2R R 2R R 2R 2R Upon power-up, the DAC registers are loaded with zero code (0x0000) and the outputs are clamped to 0 V via a low impedance path. The outputs can be updated with the zero code value at this time by asserting either LDAC or CLR. The corresponding output voltage depends on the state of the BIN/2sCOMP pin. If the BIN/2sCOMP pin is tied to DGND, then the data coding is twos complement and the outputs update to 0 V. If the BIN/2sCOMP pin is tied to DVCC, then the data coding is offset binary and the outputs update to negative full-scale. To have the outputs power-up with zero code loaded to the outputs, the CLR pin should be held low during power-up. 2R Standalone Operation E15 R/8 E14 E1 S11 S10 S0 IOUT 4 MSBs DECODED INTO 15 EQUAL SEGMENTS VOUTX 05303-060 AGNDX 12-BIT, R-2R LADDER Figure 31. DAC Ladder Structure REFERENCE BUFFERS The AD5765 operates with an external reference. The reference inputs (REFAB and REFCD) have an input range up to 2.1 V. This input voltage is then used to provide a buffered positive and negative reference for the DAC cores. The positive reference is given by +VREF = 2VREF The negative reference to the DAC cores is given by −VREF = −2VREF These positive and negative reference voltages (along with the gain register values) define the output ranges of the DACs. The serial interface works with both a continuous and noncontinuous serial clock. A continuous SCLK source can only be used if SYNC is held low for the correct number of clock cycles. In gated clock mode, a burst clock containing the exact number of clock cycles must be used and SYNC must be taken high after the final clock to latch the data. The first falling edge of SYNC starts the write cycle. Exactly 24 falling clock edges must be applied to SCLK before SYNC is brought back high again. If SYNC is brought high before the 24th falling SCLK edge, then the data written is invalid. If more than 24 falling SCLK edges are applied before SYNC is brought high, then the input data is also invalid. The input register addressed is updated on the rising edge of SYNC. In order for another serial transfer to take place, SYNC must be brought low again. After the end of the serial data transfer, data is automatically transferred from the input shift register to the addressed register. When the data has been transferred into the chosen register of the addressed DAC, all DAC registers and outputs can be updated by taking LDAC low. Rev. PrA | Page 18 of 31 Preliminary Technical Data SDO DISABLE bit; this bit is cleared by default. Readback mode is invoked by setting the R/W bit = 1 in the serial input register write. With R/W = 1, Bit A2 to Bit A0, in association with Bit REG2, Bit REG1, and Bit REG0, select the register to be read. The remaining data bits in the write sequence are don’t care. During the next SPI write, the data appearing on the SDO output contain the data from the previously addressed register. For a read of a single register, the NOP command can be used in clocking out the data from the selected register on SDO. The readback diagram in Figure 4 shows the readback sequence. For example, to read back the fine gain register of Channel A on the AD5765, the following sequence should be implemented: AD57651 MOSI SDIN SCK SCLK PC7 SYNC PC6 LDAC MISO SDO SDIN AD57651 SCLK SYNC 1. Write 0xA0XXXX to the AD5765 input register. This configures the AD5765 for read mode with the fine gain register of Channel A selected. Note that all the data bits, DB15 to DB0, are don’t cares. 2. Follow this with a second write, an NOP condition, 0x00XXXX. During this write, the data from the fine gain register is clocked out on the SDO line, that is, data clocked out contain the data from the fine gain register in Bit DB5 to Bit DB0. LDAC SDO SDIN AD57651 SCLK SYNC LDAC SIMULTANEOUS UPDATING VIA LDAC 1ADDITIONAL PINS OMITTED FOR CLARITY 05303-061 SDO Figure 32. Daisy-Chaining the AD5765 Daisy-Chain Operation For systems that contain several devices, the SDO pin can be used to daisy-chain several devices together. This daisy-chain mode can be useful in system diagnostics and in reducing the number of serial interface lines. The first falling edge of SYNC starts the write cycle. The SCLK is continuously applied to the input shift register when SYNC is low. If more than 24 clock pulses are applied, the data ripples out of the shift register and appears on the SDO line. This data is clocked out on the rising edge of SCLK and is valid on the falling edge. By connecting the SDO of the first device to the SDIN input of the next device in the chain, a multidevice interface is constructed. Each device in the system requires 24 clock pulses. Therefore, the total number of clock cycles must equal 24N, where N is the total number of AD5765 devices in the chain. When the serial transfer to all devices is complete, SYNC is taken high. This latches the input data in each device in the daisy chain and prevents any further data from being clocked into the input shift register. The serial clock can be a continuous or a gated clock. Depending on the status of both SYNC and LDAC, and after data has been transferred into the input register of the DACs, there are two ways in which the DAC registers and DAC outputs can be updated. Individual DAC Updating In this mode, LDAC is held low while data is being clocked into the input shift register. The addressed DAC output is updated on the rising edge of SYNC. Simultaneous Updating of All DACs In this mode, LDAC is held high while data is being clocked into the input shift register. All DAC outputs are updated by taking LDAC low any time after SYNC has been taken high. The update now occurs on the falling edge of LDAC. A continuous SCLK source can only be used if SYNC is held low for the correct number of clock cycles. In gated clock mode, a burst clock containing the exact number of clock cycles must be used and SYNC must be taken high after the final clock to latch the data. Readback Operation Before a readback operation is initiated, the SDO pin must be enabled by writing to the function register and clearing the Rev. PrA | Page 19 of 31 OUTPUT I/V AMPLIFIER VREFIN LDAC 16-BIT DAC VOUT DAC REGISTER INPUT REGISTER SCLK SYNC SDIN INTERFACE LOGIC SDO 05303-062 68HC11 1 AD5765 Figure 33. Simplified Serial Interface of Input Loading Circuitry for One DAC Channel AD5765 Preliminary Technical Data TRANSFER FUNCTION The output voltage expression for the AD5765 is given by ⎡ D ⎤ VOUT = −2 × VREFIN + 4 × VREFIN ⎢ ⎣ 65536 ⎥⎦ Table 7 shows the ideal input code to output voltage relationship for the AD5765 for both offset binary and twos complement data coding. where: D is the decimal equivalent of the code loaded to the DAC. VREFIN is the reference voltage applied at the REFAB/REFCD pins. Table 7. Ideal Output Voltage to Input Code Relationship Digital Input Analog Output Offset Binary Data Coding MSB 1111 1000 1000 0111 0000 1111 0000 0000 1111 0000 1111 0000 0000 1111 0000 LSB 1111 0001 0000 1111 0000 VOUTX 2VREF × (32767/32768) 2VREF × (1/32768) 0V −2VREF × (1/32768) −2VREF × (32767/32768) ASYNCHRONOUS CLEAR (CLR) CLR is a negative edge triggered clear that allows the outputs to be cleared to either 0 V (twos complement coding) or negative full scale (offset binary coding). It is necessary to maintain CLR low for a minimum amount of time (see Figure 2) for the operation to complete. When the CLR signal is returned high, the output remains at the cleared value until a new value is programmed. If at power-on, CLR is at 0 V, then all DAC outputs are updated with the clear value. A clear can also be initiated through software by writing the command 0x04XXXX to the AD5765. Twos Complement Data Coding MSB 0111 0000 0000 1111 1000 1111 0000 0000 1111 0000 1111 0000 0000 1111 0000 LSB 1111 0001 0000 1111 0000 VOUTX 2VREF × (32767/32768) 2VREF × (1/32768) 0V −2VREF × (1/32768) −2VREF × (32767/32768) Table 8. AD5765 Input Register Format MSB LSB DB23 DB22 DB21 DB20 DB19 DB18 DB17 DB16 R/W 0 REG2 REG1 REG0 A2 A1 A0 DB15 DB14 DB13 DB12 DB11 DB10 DB9 DB8 DB7 DB6 DB5 DB4 DB3 DB2 DB1 DATA Table 9. Input Register Bit Functions Bit R/W Description Indicates a read from or a write to the addressed register. REG2, REG1, REG0 Used in association with the address bits to determine if a read or write operation is to the data register, offset register, gain register, or function register. REG2 REG1 REG0 Function 0 0 0 Function Register 0 1 0 Data Register 0 1 1 Coarse Gain Register 1 0 0 Fine Gain Register 1 0 1 Offset Register These bits are used to decode the DAC channels. A2 A1 A0 Channel Address 0 0 0 DAC A 0 0 1 DAC B 0 1 0 DAC C 0 1 1 DAC D 1 0 0 ALL DACs Data Bits. A2, A1, A0 D15:D0 Rev. PrA | Page 20 of 31 DB0 Preliminary Technical Data AD5765 FUNCTION REGISTER The function register is addressed by setting the three REG bits to 000. The values written to the address bits and the data bits determine the function addressed. The functions available via the function register are outlined in Table 10 and Table 11. Table 10. Function Register Options REG2 0 0 REG1 0 0 REG0 0 0 A2 0 0 A1 0 0 A0 0 1 0 0 0 0 0 0 1 1 0 0 0 1 DB15:DB6 DB5 Don’t Care LocalGroundOffset Adjust DB4 DB3 DB2 NOP, Data = Don’t Care D1 D1 D0 Direction Value Direction DB1 DB0 D0 Value SDO Disable CLR, Data = Don’t Care LOAD, Data = Don’t Care Table 11. Explanation of Function Register Options Option NOP Local-GroundOffset Adjust D0/D1 Direction Description No operation instruction used in readback operations. Set by the user to enable local-ground-offset adjust function. Cleared by the user to disable local-ground-offset adjust function (default). Refer to Features section for further details. Set by the user to enable D0/D1 as outputs. Cleared by the user to enable D0/D1 as inputs (default). Refer to the Features section for further details. I/O Port Status Bits. Logic values written to these locations determine the logic outputs on the D0 and D1 pins when configured as outputs. These bits indicate the status of the D0 and D1 pins when the I/O port is active as an input. When enabled as inputs, these bits are don’t cares during a write operation. Set by the user to disable the SDO output. Cleared by the user to enable the SDO output (default). Addressing this function resets the DAC outputs to 0 V in twos complement mode and negative full scale in binary mode. Addressing this function updates the DAC registers and consequently the analog outputs. D0/D1 Value SDO Disable CLR LOAD DATA REGISTER The data register is addressed by setting the three REG bits to 010. The DAC address bits select with which DAC channel the data transfer is to take place (see Table 9). The data bits are in positions DB15 to DB0 as shown in Table 12. Table 12. Programming the AD5765 Data Register REG2 REG1 REG0 0 1 0 A2 A1 A0 DB15 DB14 DB13 DB12 DB11 DB10 DAC Address DB9 DB8 DB7 DB6 DB5 DB4 DB3 DB2 DB1 DB0 16-Bit DAC Data COARSE GAIN REGISTER The coarse gain register is addressed by setting the three REG bits to 011. The DAC address bits select with which DAC channel the data transfer is to take place (see Table 9). The coarse gain register is a 2-bit register and allows the user to select the output range of each DAC as shown in Table 13 and Table 14. Table 13. Programming the AD5765 Coarse Gain Register REG2 0 REG1 1 REG0 1 A2 A1 A0 DAC Address Table 14. Output Range Selection Output Range ±4.096 V (default) ±4.201 V ±4.331 V CG1 0 0 1 CG0 0 1 0 Rev. PrA | Page 21 of 31 DB15 …. DB2 Don’t Care DB1 CG1 DB0 CG0 AD5765 Preliminary Technical Data FINE GAIN REGISTER The fine gain register is addressed by setting the three REG bits to 100. The DAC address bits select with which DAC channel the data transfer is to take place (see Table 9). The fine gain register is a 6-bit register and allows the user to adjust the gain of each DAC channel by −32 LSBs to +31 LSBs in 1 LSB increments as shown in Table 15 and Table 16. The adjustment is made to both the positive full-scale and negative full-scale points simultaneously, each point being adjusted by ½ of one step. The fine gain register coding is twos complement. Table 15. Programming AD5765 Fine Gain Register REG2 1 REG1 0 REG0 0 A2 A1 A0 DAC Address DB15:DB6 Don’t Care DB5 FG5 DB4 FG4 DB3 FG3 DB2 FG2 DB1 FG1 DB0 FG0 Table 16. AD5765 Fine Gain Register Options Gain Adjustment +31 LSBs +30 LSBs No Adjustment (default) −31 LSBs −32 LSBs FG5 0 0 0 1 1 FG4 1 1 0 0 0 FG3 1 1 0 0 0 FG2 1 1 0 0 0 FG1 1 1 0 0 0 FG0 1 0 0 1 0 OFFSET REGISTER The offset register is addressed by setting the three REG bits to 101. The DAC address bits select with which DAC channel the data transfer is to take place (see Table 9). The AD5765 offset register is an 8-bit register and allows the user to adjust the offset of each channel by −16 LSBs to +15.875 LSBs in increments of ⅛ LSB as shown in Table 17 and Table 18. The offset register coding is twos complement. Table 17. Programming the AD5765 Offset Register REG2 1 REG1 0 REG0 1 A2 A1 A0 DAC Address DB15:DB8 Don’t Care DB7 OF7 DB6 OF6 DB5 OF5 DB4 OF4 DB3 OF3 DB2 OF2 DB1 OF1 Table 18. AD5765 Offset Register options Offset Adjustment +15.875 LSBs +15.75 LSBs No Adjustment (default) −15.875 LSBs −16 LSBs OF7 0 0 0 1 1 OF6 1 1 0 0 0 OF5 1 1 0 0 0 Rev. PrA | Page 22 of 31 OF4 1 1 0 0 0 OF3 1 1 0 0 0 OF2 1 1 0 0 0 OF1 1 1 0 0 0 OF0 1 0 0 1 0 DB0 OF0 Preliminary Technical Data AD5765 OFFSET AND GAIN ADJUSTMENT WORKED EXAMPLE Convert this to a negative twos complement number by inverting all bits and adding 1: 11011000. Using the information provided in the previous section, the following worked example demonstrates how the AD5765 functions can be used to eliminate both offset and gain errors. As the AD5765 is factory calibrated, offset and gain errors should be negligible. However, errors can be introduced by the system that the AD5765 is operating within, for example, a voltage reference value that is not equal to 2.048 V introduces a gain error. An output range of ±4.096 V and twos complement data coding is assumed. 11011000 is the value that should be programmed to the offset register. Removing Offset Error The AD5765 can eliminate an offset error in the range of −2 mV to +1.98 mV with a step size of ⅛ of a 16-bit LSB. Note that this twos complement conversion is not necessary in the case of a positive offset adjustment. The value to be programmed to the offset register is simply the binary representation of the adjustment value. Removing Gain Error The AD5765 can eliminate a gain error at negative full-scale output in the range of −2 mV to +1.94 mV with a step size of ½ of a 16-bit LSB. Calculate the step size of the gain adjustment. Gain Adjust Step Size = Calculate the step size of the offset adjustment. Offset Adjust Step Size = 8.192 = 15.625 µV 216 × 8 Measure the offset error by programming 0x0000 to the data register and measuring the resulting output voltage, for this example the measured value is 614 µV. Measure the gain error by programming 0x8000 to the data register and measuring the resulting output voltage. The gain error is the difference between this value and −4.096 V; for this example, the gain error is −0.8 mV. Calculate how many gain adjustment steps this value represents. Calculate the number of offset adjustment steps that this value represents. Number of Steps = Measured Offset Value 614 µV = = 40 Steps Offset Step Size 15.625 µV The offset error measured is positive, therefore, a negative adjustment of 40 steps is required. The offset register is 8 bits wide and the coding is twos complement. The required offset register value can be calculated as follows: 8.192 = 62.5 µV 216 × 2 Number of Steps = Measured Gain Value 0.8 mV = = 13 Steps Gain Step Size 62.5 µV The gain error measured is negative (in terms of magnitude); therefore, a positive adjustment of 13 steps is required. The gain register is 6 bits wide and the coding is twos complement, the required gain register value can be determined as follows: Convert adjustment value to binary: 001101. The value to be programmed to the gain register is simply this binary number. Convert adjustment value to binary: 00101000. Rev. PrA | Page 23 of 31 AD5765 AD5765 FEATURES ANALOG OUTPUT CONTROL In many industrial process control applications, it is vital that the output voltage be controlled during power-up and during brownout conditions. When the supply voltages are changing, the output pins are clamped to 0 V via a low impedance path. To prevent the output amp being shorted to 0 V during this time, transmission gate G1 is also opened (see Figure 34). These conditions are maintained until the power supplies stabilize and a valid word is written to the DAC register. At this time, G2 opens and G1 closes. Both transmission gates are also externally controllable via the reset logic (RSTIN) control input. For instance, if RSTIN is driven from a battery supervisor chip, the RSTIN input is driven low to open G1 and close G2 on powerdown or during a brownout. Conversely, the on-chip voltage detector output (RSTOUT) is also available to the user to control other parts of the system. The basic transmission gate functionality is shown in Figure 34. RSTOUT DIGITAL I/O PORT The AD5765 contains a 2-bit digital I/O port (D1 and D0). These bits can be configured as inputs or outputs independently, and can be driven or have their values read back via the serial interface. The I/O port signals are referenced to DVCC and DGND. When configured as outputs, they can be used as control signals to multiplexers or can be used to control calibration circuitry elsewhere in the system. When configured as inputs, the logic signals from limit switches, for example, can be applied to D0 and D1 and can be read back via the digital interface. DIE TEMPERATURE SENSOR RSTIN The on-chip die temperature sensor provides a voltage output that is linearly proportional to the centigrade temperature scale. Its nominal output voltage is 1.4 V at +25°C die temperature, varying at 5 mV/°C, giving a typical output range of 1.175 V to 1.9 V over the full temperature range. Its low output impedance, and linear output simplify interfacing to temperature control circuitry and A/D converters. The temperature sensor is provided as more of a convenience rather than a precise feature; it is intended for indicating a die temperature change for recalibration purposes. VOLTAGE MONITOR AND CONTROL G1 VOUTA AGNDA 05303-063 G2 Figure 34. Analog Output Control Circuitry DIGITAL OFFSET AND GAIN CONTROL LOCAL GROUND OFFSET ADJUST The AD5765 incorporates a digital offset adjust function with a ±16 LSB adjust range and 0.125 LSB resolution. The gain register allows the user to adjust the AD5765 full-scale output range. The full-scale output can be programmed to achieve fullscale ranges of ±4.096 V, ±4.201 V, and ±4.311 V. A fine gain trim is also provided. PROGRAMMABLE SHORT-CIRCUIT PROTECTION The short-circuit current of the output amplifiers can be programmed by inserting an external resistor between the ISCC pin and PGND. The programmable range for the current is 500 µA to 10 mA, corresponding to a resistor range of 120 kΩ to 6 kΩ. The resistor value is calculated as follows: R≈ If the ISCC pin is left unconnected, the short-circuit current limit defaults to 5 mA. It should be noted that limiting the short circuit current to a small value can affect the slew rate of the output when driving into a capacitive load, therefore, the value of short-circuit current programmed should take into account the size of the capacitive load being driven. The AD5765 incorporates a local-ground-offset adjust feature which, when enabled in the function register, adjusts the DAC outputs for voltage differences between the individual DAC ground pins and the REFGND pin ensuring that the DAC output voltages are always with respect to the local DAC ground pin. For instance, if pin AGNDA is at +5 mV with respect to the REFGND pin and VOUTA is measured with respect to AGNDA, then a −5 mV error results, enabling the localground-offset adjust feature adjusts VOUTA by +5 mV, eliminating the error. 60 I SC Rev. PrA | Page 24 of 31 Preliminary Technical Data AD5765 APPLICATIONS INFORMATION TYPICAL OPERATING CIRCUIT Precision Voltage Reference Selection Figure 35 shows the typical operating circuit for the AD5765. The only external components needed for this precision 16-bit DAC are a reference voltage source, decoupling capacitors on the supply pins and reference inputs, and an optional shortcircuit current setting resistor. Because the device incorporates reference buffers, it eliminates the need for an external bipolar reference and associated buffers. This leads to an overall savings in both cost and board space. To achieve the optimum performance from the AD5765 over its full operating temperature range, a precision voltage reference must be used. Thought should be given to the selection of a precision voltage reference. The AD5765 has two reference inputs, REFAB and REFCD. The voltages applied to the reference inputs are used to provide a buffered positive and negative reference for the DAC cores. Therefore, any error in the voltage reference is reflected in the outputs of the device. In Figure 35, AVDD is connected to +5 V and AVSS is connected to −5 V. In Figure 35, AGNDA is connected to REFGND. There are four possible sources of error to consider when choosing a voltage reference for high accuracy applications: initial accuracy, temperature coefficient of the output voltage, long term drift, and output voltage noise. +5V 10µF ADR420 100nF 2 VOUT 6 GND VIN Initial accuracy error on the output voltage of an external reference could lead to a full-scale error in the DAC. Therefore, to minimize these errors, a reference with low initial accuracy error specification is preferred. Choosing a reference with an output trim adjustment, such as the ADR430, allows a system designer to trim system errors out by setting the reference voltage to a voltage other than the nominal. The trim adjustment can also be used at temperature to trim out any error. 4 +5V –5V 10µF 10µF 100nF 100nF 100nF BIN/2sCOMP 1 SYNC SCLK 2 SCLK SDIN 3 SDIN SDO 4 SDO REFAB NC REFCD AVSS TEMP AVDD REFGND SYNC BIN/2sCOMP 32 31 30 29 28 27 26 25 +5V AGNDA 24 VOUTA 23 VOUTA VOUTB 22 VOUTB AGNDB 21 AD5765 AGNDC 20 LDAC VOUTC 19 VOUTC D0 7 D0 VOUTD 18 VOUTD D1 8 D1 AGNDD 17 10µF ISCC AVSS PGND 100nF NC = NO CONNECT 10µF +5V +5V The temperature coefficient of a reference output voltage affects INL, DNL, and TUE. A reference with a tight temperature coefficient specification should be chosen to reduce the dependence of the DAC output voltage on ambient conditions. In high accuracy applications, which have a relatively low noise budget, reference output voltage noise needs to be considered. Choosing a reference with as low an output noise voltage as practical for the system resolution required is important. Precision voltage references such as the ADR420 (XFET® design) produce low output noise in the 0.1 Hz to 10 Hz region. However, as the circuit bandwidth increases, filtering the output of the reference may be required to minimize the output noise. 10µF 100nF 100nF RSTIN AVDD 10 11 12 13 14 15 16 RSTOUT DVCC 9 DGND RSTIN CLR 6 RSTOUT 5 LDAC Long-term drift is a measure of how much the reference output voltage drifts over time. A reference with a tight long-term drift specification ensures that the overall solution remains relatively stable over its entire lifetime. –5V Figure 35. Typical Operating Circuit Table 19. Some Precision References Recommended for Use with the AD5765 Part No. ADR430 ADR420 ADR390 Initial Accuracy (mV Max) ±1 ±1 ±4 Long-Term Drift (ppm Typ) 40 50 50 Temp Drift (ppm/°C Max) 3 3 9 Rev. PrA | Page 25 of 31 0.1 Hz to 10 Hz Noise (µV p-p Typ) 3.5 1.75 5 AD5765 Preliminary Technical Data LAYOUT GUIDELINES The power supply lines of the AD5765 should use as large a trace 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 parts of the board, and should never be run near the reference inputs. A ground line routed between the SDIN and SCLK lines helps reduce crosstalk between them (not required on a multilayer board, which has a separate ground plane, however, it is helpful to separate the lines). It is essential to minimize noise on the reference inputs, because it couples through to the DAC output. Avoid crossover of digital and analog signals. Traces on opposite sides of the board should run at right angles to each other. This reduces the effects of feedthrough on the board. A microstrip technique is recommended, but not always possible with a double-sided board. In this technique, the component side of the board is dedicated to ground plane, and signal traces are placed on the solder side. GALVANICALLY ISOLATED INTERFACE In many process control applications, it is necessary to provide an isolation barrier between the controller and the unit being controlled to protect and isolate the controlling circuitry from any hazardous common-mode voltages that might occur. Isocouplers provide voltage isolation in excess of 2.5 kV. The serial loading structure of the AD5765 makes it ideal for isolated interfaces because the number of interface lines is kept to a minimum. Figure 36 shows a 4-channel isolated interface to the AD5765 using an ADuM1400. For more information, go to www.analog.com. µCONTROLLER SERIAL CLOCK OUT SERIAL DATA OUT SYNC OUT CONTROL OUT ADuM14001 VIA VIB VIC VID ENCODE DECODE ENCODE DECODE ENCODE DECODE ENCODE DECODE VOA VOB VOC VOD TO SCLK TO SDIN TO SYNC TO LDAC 05303-065 In any circuit where accuracy is important, careful consideration of the power supply and ground return layout helps to ensure the rated performance. The printed circuit board on which the AD5765 is mounted should be designed so that the analog and digital sections are separated and confined to certain areas of the board. If the AD5765 is in a system where multiple devices require an AGND-to-DGND connection, the connection should be made at one point only. The star ground point should be established as close as possible to the device. The AD5765 should have ample supply bypassing of 10 µF in parallel with 0.1 µF on each supply located as close to the package as possible, ideally right up against the device. The 10 µF capacitors are the tantalum bead type. The 0.1 µF capacitor should have low effective series resistance (ESR) and low effective series inductance (ESI) such as the common ceramic types, which provide a low impedance path to ground at high frequencies to handle transient currents due to internal logic switching. 1ADDITIONAL PINS OMITTED FOR CLARITY Figure 36. Isolated Interface MICROPROCESSOR INTERFACING Microprocessor interfacing to the AD5765 is via a serial bus that uses a standard protocol that is compatible with microcontrollers and DSP processors. The communications channel is a 3-wire (minimum) interface consisting of a clock signal, a data signal, and a synchronization signal. The AD5765 requires a 24-bit data-word with data valid on the falling edge of SCLK. For all the interfaces, the DAC output update can be done automatically when all the data is clocked in, or it can be done under the control of LDAC. The contents of the DAC register can be read using the readback function. AD5765 to MC68HC11 Interface Figure 37 shows an example of a serial interface between the AD5765 and the MC68HC11 microcontroller. The serial peripheral interface (SPI) on the MC68HC11 is configured for master mode (MSTR = 1), clock polarity bit (CPOL = 0), and the clock phase bit (CPHA = 1). The SPI is configured by writing to the SPI control register (SPCR) (see the MC68HC11 User Manual). SCK of the MC68HC11 drives the SCLK of theAD5765, the MOSI output drives the serial data line (DIN) of the AD5744/AD5765, and the MISO input is driven from SDO. The SYNC is driven from one of the port lines, in this case, PC7. When data is being transmitted to the AD5765, the SYNC line (PC7) is taken low and data is transmitted MSB first. Data appearing on the MOSI output is valid on the falling edge of SCK. Eight falling clock edges occur in the transmit cycle, so, in order to load the required 24-bit word, PC7 is not brought high until the third 8-bit word has been transferred to the DAC input shift register. AD57651 MC68HC111 MISO SDO MOSI SDIN SCK SCLK PC7 SYNC 1ADDITIONAL PINS OMITTED FOR CLARITY Figure 37. AD5765 to MC68HC11 Interface Rev. PrA | Page 26 of 31 Preliminary Technical Data AD5765 LDAC is controlled by the PC6 port output. The DAC can be updated after each 3-byte transfer by bringing LDAC low. This example does not show other serial lines for the DAC. For example, if CLR were used, it could be controlled by port output PC5. output is updated using the LDAC pin via the DSP. Alternatively, the LDAC input can be tied permanently low, and then the update takes place automatically when TFS is taken high. AD5765 to 8XC51 Interface The AD5765 requires a clock synchronized to the serial data. For this reason, the 8XC51 must be operated in Mode 0. In this mode, serial data enters and exits through RxD, and a shift clock is output on TxD. P3.3 and P3.4 are bit programmable pins on the serial port and are used to drive SYNC and LDAC, respectively. The 8CX51 provides the LSB of its SBUF register as the first bit in the data stream. The user must ensure that the data in the SBUF register is arranged correctly, because the DAC expects the MSB first. When data is to be transmitted to the DAC, P3.3 is taken low. Data on RxD is clocked out of the microcontroller on the rising edge of TxD and is valid on the falling edge. As a result, no glue logic is required between this DAC and the microcontroller interface. 8XC511 AD57651 RxD SDIN TxD SCLK P3.3 SYNC P3.4 LDAC AD57651 ADSP2101/ ADSP21031 DR SDO DT SDIN SCLK SCLK TFS SYNC RFS FO LDAC 1ADDITIONAL PINS OMITTED FOR CLARITY Figure 39. AD5765 to ADSP2101/ADSP2103 Interface AD5765 to PIC16C6x/7x Interface The PIC16C6x/7x synchronous serial port (SSP) is configured as an SPI master with the clock polarity bit set to 0. This is done by writing to the synchronous serial port control register (SSPCON). See the PIC16/17 Microcontroller User Manual. In this example, I/O port RA1 is being used to pulse SYNC and enable the serial port of the AD5765. This microcontroller transfers only eight bits of data during each serial transfer operation; therefore, three consecutive write operations are needed. Figure 40 shows the connection diagram. AD57651 PIC16C6x/7x1 1ADDITIONAL PINS OMITTED FOR CLARITY SDI/RC4 SDO SDO/RC5 SDIN SCLK/RC3 SCLK RA1 SYNC Figure 38. AD5765 to 8XC51 Interface The 8XC51 transmits data in 8-bit bytes with only eight falling clock edges occurring in the transmit cycle. Because the DAC expects a 24-bit word, SYNC (P3.3) must be left low after the first eight bits are transferred. After the third byte has been transferred, the P3.3 line is taken high. The DAC can be updated using LDAC via P3.4 of the 8XC51. 1ADDITIONAL PINS OMITTED FOR CLARITY Figure 40. AD5765 to PIC16C6x/7x Interface EVALUATION BOARD AD5765 to ADSP2101/ADSP2103 Interface An interface between the AD5765 and the ADSP2101/ ADSP2103 is shown in Figure 39. The ADSP2101/ADSP2103 should be set up to operate in the SPORT transmit alternate framing mode. The ADSP2101/ADSP2103 are programmed through the SPORT control register and should be configured as follows: internal clock operation, active low framing, and 24-bit word length. Transmission is initiated by writing a word to the Tx register after the SPORT has been enabled. As the data is clocked out of the DSP on the rising edge of SCLK, no glue logic is required to interface the DSP to the DAC. In the interface shown, the DAC The AD5765 comes with a full evaluation board to aid designers in evaluating the high performance of the part with a minimum of effort. All that is required with the evaluation board is a power supply and a PC. The AD5765 evaluation kit includes a populated, tested AD5765 printed circuit board. The evaluation board interfaces to the USB port of the PC. Software is available with the evaluation board, which allows the user to easily program the AD5765. The software runs on any PC that has Microsoft® Windows® 2000/NT/XP installed. The EVAL-AD5765EBZ data sheet is available, which gives full details on operating the evaluation board. Rev. PrA | Page 27 of 31 AD5765 Preliminary Technical Data OUTLINE DIMENSIONS 0.75 0.60 0.45 1.20 MAX 9.00 BSC SQ 25 32 24 1 PIN 1 7.00 BSC SQ TOP VIEW 0.15 0.05 (PINS DOWN) 0° MIN 1.05 1.00 0.95 0.20 0.09 7° 3.5° 0° 0.08 MAX COPLANARITY SEATING PLANE 17 8 9 VIEW A VIEW A 16 0.80 BSC LEAD PITCH ROTATED 90° CCW 0.45 0.37 0.30 COMPLIANT TO JEDEC STANDARDS MS-026ABA Figure 41. 32-Lead Thin Plastic Quad Flat Package [TQFP] (SU-32-2) Dimensions shown in millimeters ORDERING GUIDE Model AD5765BSUZ AD5765CSUZ INL ± 2 LSB ± 1 LSB Temperature Range −40°C to +105°C −40°C to +105°C Rev. PrA | Page 28 of 31 Package Description 32-lead TQFP 32-lead TQFP Package Option SU-32-2 SU-32-2 Preliminary Technical Data AD5765 NOTES Rev. PrA | Page 29 of 31 AD5765 NOTES Rev. PrA | Page 30 of 31 Preliminary Technical Data AD5765 NOTES ©2007 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. PR07249-0-12/07(PrA) Rev. PrA | Page 31 of 31