12-Bit Monitor and Control System with Multichannel ADC, DACs, Temperature Sensor, and Current Sense AD7294-2 Data Sheet FEATURES APPLICATIONS 12-bit SAR ADC with 3 s conversion time 4 uncommitted analog inputs Differential/single-ended VREF, 2 x VREF input ranges 2 high-side current sense inputs 5 V to 59.4 V operating range 0.75% maximum gain error 200 mV input range 2 external diode temperature sensor inputs -55C to +150C measurement range 2C accuracy Series resistance cancellation 1 internal temperature sensor: 2C accuracy Built-in monitoring features Minimum/maximum recorder for each channel Programmable alert thresholds Programmable hysteresis Four 12-bit, monotonic, 15 V DACs 5 V span, 0 V to 10 V offset 8 s settling time 10 mA sink and source capability Power-on reset (POR) to 0 V Internal 2.5 V reference 2-wire, fast mode I2C interface Temperature range: -40C to +105C Package type: 64-lead TQFP Pin compatible with the AD7294 Cellular base stations GSM, EDGE, UMTS, CDMA, TD-SCDMA, W-CDMA, WiMAX Point-to-multipoint and other RF transmission systems 12 V, 24 V, 48 V automotive applications Industrial controls Rev. 0 GENERAL DESCRIPTION The AD7294-2 contains all the functions that are required for general-purpose monitoring and control of current, voltage, and temperature, integrated into a single-chip solution. The part includes low voltage (200 mV) analog input sense amplifiers for current monitoring across shunt resistors, temperature sense inputs, and four uncommitted analog input channels multiplexed into a SAR analog-to-digital converter (ADC) with a 3 s conversion time. A high accuracy internal reference is provided to drive both the digital-to-analog converters (DACs) and the ADC. Four 12-bit DACs provide the outputs for voltage control. The AD7294-2 also includes limit registers for alarm functions. The part is designed for high voltage compliance: 59.4 V on the current sense inputs and up to a 15 V DAC output voltage. The AD7294-2 is a highly integrated solution that offers all the functionality necessary for precise control of the power amplifier in cellular base station applications. In these types of applications, the DACs provide 12-bit resolution to control the bias currents of the power transistors. Thermal diode-based temperature sensors are incorporated to compensate for temperature effects. The ADC monitors the high-side current and temperature. This functionality is provided in a 64-lead TQFP, which operates over a temperature range of -40C to +105C. Document Feedback 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 (c)2013 Analog Devices, Inc. All rights reserved. Technical Support www.analog.com AD7294-2* PRODUCT PAGE QUICK LINKS Last Content Update: 02/23/2017 COMPARABLE PARTS DESIGN RESOURCES View a parametric search of comparable parts. * AD7294-2 Material Declaration * PCN-PDN Information EVALUATION KITS * Quality And Reliability * AD7294-2 Evaluation Board * Symbols and Footprints DOCUMENTATION DISCUSSIONS Data Sheet View all AD7294-2 EngineerZone Discussions. * AD7294-2: 12-Bit Monitor and Control System with Multichannel ADC, DACs, Temperature Sensor, and Current Sense Data Sheet SAMPLE AND BUY User Guides * UG-605: Evaluating the AD7294-2 12-Bit Monitor and Control System Visit the product page to see pricing options. TECHNICAL SUPPORT Submit a technical question or find your regional support number. DOCUMENT FEEDBACK Submit feedback for this data sheet. This page is dynamically generated by Analog Devices, Inc., and inserted into this data sheet. A dynamic change to the content on this page will not trigger a change to either the revision number or the content of the product data sheet. This dynamic page may be frequently modified. AD7294-2 Data Sheet TABLE OF CONTENTS Features .............................................................................................. 1 Command Register .................................................................... 26 Applications ....................................................................................... 1 ADC Result Register .................................................................. 26 General Description ......................................................................... 1 TSENSE1 and TSENSE2 Result Registers ......................................... 27 Revision History ............................................................................... 2 TSENSEINT Result Register .......................................................... 27 Functional Block Diagram .............................................................. 3 DACA, DACB, DACC, and DACD Value Registers....................... 27 Specifications..................................................................................... 4 Alert Status Register A, Alert Status Register B, and Alert Status Register C ......................................................................... 28 DAC Specifications....................................................................... 4 Channel Sequence Register ....................................................... 28 ADC Specifications ...................................................................... 5 Configuration Register .............................................................. 29 General Specifications ................................................................. 7 Power-Down Register ................................................................ 30 Timing Characteristics ................................................................ 8 DATALOW and DATAHIGH Registers .......................................... 30 Absolute Maximum Ratings ............................................................ 9 Hysteresis Registers .................................................................... 30 Thermal Resistance ...................................................................... 9 ESD Caution .................................................................................. 9 Pin Configuration and Function Descriptions ........................... 10 Typical Performance Characteristics ........................................... 12 Terminology .................................................................................... 17 DAC Terminology ...................................................................... 17 ADC Terminology ...................................................................... 17 Theory of Operation ...................................................................... 18 ADC Overview ........................................................................... 18 ADC Transfer Functions ........................................................... 18 Analog Inputs .............................................................................. 19 Current Sensor ............................................................................ 20 Analog Comparator Loop ......................................................... 22 Temperature Sensor ................................................................... 22 DAC Operation ........................................................................... 23 ADC and DAC Reference.......................................................... 24 VDRIVE Feature .............................................................................. 24 Register Settings .............................................................................. 25 Address Pointer Register ........................................................... 25 Remote Channel TSENSE1 and TSENSE2 Offset Registers ........... 31 I C Interface .................................................................................... 32 2 General I2C Timing .................................................................... 32 Serial Bus Address Byte ............................................................. 32 Interface Protocol ....................................................................... 33 Modes of Operation ....................................................................... 36 Command Mode ........................................................................ 36 Autocycle Mode .......................................................................... 37 Alerts and Limits Theory .............................................................. 38 ALERT_FLAG Bit ....................................................................... 38 Alert Status Registers ................................................................. 38 DATALOW and DATAHIGH Monitoring Features ...................... 38 Hysteresis..................................................................................... 39 Applications Information .............................................................. 40 Base Station Power Amplifier Monitor and Control ............. 40 Gain Control of Power Amplifier............................................. 41 Outline Dimensions ....................................................................... 42 Ordering Guide .......................................................................... 42 REVISION HISTORY 6/13--Revision 0: Initial Version Rev. 0 | Page 2 of 44 Data Sheet AD7294-2 FUNCTIONAL BLOCK DIAGRAM RSENSE VPP1, VPP2 RS1(+) RS1(-) RS2(+) HIGH-SIDE CURRENT SENSE ISENSE 1 OVERRANGE VIN0 VIN1 VIN2 VIN3 D1(+) D2(+) T1 AGND1 TO DAC OUTV+ AB, AVDD AGND7 DAC OUTV+ CD, 2.5V REF HIGH-SIDE CURRENT SENSE 12-BIT DAC 100k 200k VOUTA VREF 10.41 100k MUX 12-BIT ADC 12-BIT DAC 200k 100k OFFSET IN A 200k VOUTB LIMIT REGISTERS 100k 12-BIT DAC T2 200k 100k TEMP SENSOR D2(-) D1(-) VOUTC 100k AD7294-2 CONTROL LOGIC OFFSET IN B 200k 12-BIT DAC 200k 100k 200k OFFSET IN C VOUTD I2C INTERFACE PROTOCOL DGND (x3) RESET SDA SCL AS2 AS1 AS0 DCAP ALERT/ BUSY Figure 1. Rev. 0 | Page 3 of 44 100k 200k OFFSET IN D 10936-001 ISENSE 2 OVERRANGE TO LOAD REFOUT/ REFOUT/ REFIN DAC RS2(-) REFIN ADC AD7294-2 Data Sheet SPECIFICATIONS DAC SPECIFICATIONS AVDD = 4.5 V to 5.5 V, AGND1 to AGND7 = DGND = 0 V, internal 2.5 V reference; VDRIVE = 2.7 V to 5.5 V; TA = -40C to +105C, unless otherwise noted. DAC OUTV+ AB and DAC OUTV+ CD = 4.5 V to 16.5 V; OFFSET IN x is floating; therefore, the DAC output span = 0 V to 5 V. Table 1. Parameter ACCURACY 1 Resolution Relative Accuracy (INL) Differential Nonlinearity (DNL) Zero-Code Error Full-Scale Error Offset Error Offset Error Temperature Coefficient Gain Error Gain Error Drift DAC OUTPUT CHARACTERISTICS Output Voltage Span Output Voltage Offset Offset Input Pin Range DC Input Impedance 2 Output Voltage Settling Time2 Slew Rate2 Short-Circuit Current2 Load Current2 Capacitive Load Stability2 DC Output Impedance2 REFERENCE Reference Output Voltage Reference Input Voltage Range Input Current Input Capacitance2 VREF Output Impedance2 Reference Temperature Coefficient 1 2 Min Typ Max Unit 1 0.3 2.5 3 1 6 10 4 2 Bits LSB LSB mV mV mV mV ppm/C 12 5 0.025 5 0 0 0 0.155 % FSR ppm/C 2 x VREF 10 V V 5 V k s 75 8 1.1 40 10 V/s mA mA nF 10 1 2.49 0 2.5 400 20 5 10 2.51 AVDD - 2 480 25 V V A pF ppm/C Linearity calculated using a reduced code range: Code 10 to Code 4095. Guaranteed by design and characterization; not production tested. Rev. 0 | Page 4 of 44 Test Conditions/Comments Guaranteed monotonic DAC OUTV+ = 5.5 V Measured in the linear region, TA = -40C to +105C Measured in the linear region, TA = 25C 0 V to 5 V for a 2.5 V reference The output voltage span can be positioned in the 0 V to 15 V range; if the OFFSET IN x pin is left floating, the offset = 2/3 x VREF, giving an output of 0 V to 2 x VREF VOUT = 3 x VOFFSET - 2 x VREF + VDAC 100 k to VREF, and 200 k to AGND; see Figure 45 1/4 to 3/4 change within 1/2 LSB, measured from last SCL edge Full-scale current shorted to ground Source and/or sink within 200 mV of supply RL = 0.2% maximum at 25C, AVDD = 5 V VREF = 2.5 V A buffer is required if the reference output is used to drive external loads Data Sheet AD7294-2 ADC SPECIFICATIONS AVDD = 4.5 V to 5.5 V, AGND1 to AGND7 = DGND = 0 V, internal or external 2.5 V reference; VDRIVE = 2.7 V to 5.5 V; VPPx = AVDD to 59.4 V; TA = -40C to +105C, unless otherwise noted. Table 2. Parameter DC ACCURACY Resolution Integral Nonlinearity (INL) 1 Min Differential Nonlinearity (DNL)1 Single-Ended Mode Offset Error Offset Error Match Gain Error Gain Error Match Differential Mode Positive Gain Error Positive Gain Error Match Zero Code Error Zero Code Error Match Negative Gain Error Negative Gain Error Match CONVERSION RATE Conversion Time 2 Autocycle Update Rate2 Throughput Rate ANALOG INPUT 3 Single-Ended Input Range Typ Max Unit 12 0.5 0.5 0.5 1 1.5 0.99 Bits LSB LSB LSB 1 0.4 0.5 0.4 7 4.5 LSB LSB LSB LSB LSB LSB 3 50 s s kSPS 0 V to VREF mode 0 V to 2 x VREF mode 0 V to VREF mode 0 V to 2 x VREF mode 0 V to VREF mode 0 V to 2 x VREF mode 73 72 dB dB 72.5 dB fIN = 10 kHz sine wave; differential mode fIN = 10 kHz sine wave; single-ended and pseudo differential modes fIN = 10 kHz sine wave; differential mode 71.5 dB Total Harmonic Distortion (THD)1 -81 -79 dB dB Spurious-Free Dynamic Range (SFDR)1 -81 -79 dB dB Channel-to-Channel Isolation2 -90 dB Fully Differential Input Range: VIN+ - VIN- Input Capacitance2 DC Input Leakage Current DYNAMIC PERFORMANCE Signal-to-Noise Ratio (SNR)1 Signal-to-Noise and Distortion Ratio (SINAD)1 VREF 2 x VREF VREF 2 x VREF +VREF +2 x VREF fSCL = 400 kHz V V V V V V pF A Pseudo Differential Input Range: VIN+ - VIN- 4 0 0 0 0 -VREF -2 x VREF Differential mode Single-ended or pseudo differential mode Differential, single-ended, and pseudo differential modes LSB LSB LSB LSB 1 0.5 3 0.5 1 0.5 22.22 Test Conditions/Comments 30 1 Rev. 0 | Page 5 of 44 fIN = 10 kHz sine wave; single-ended and pseudo differential modes fIN = 10 kHz sine wave; differential mode fIN = 10 kHz sine wave; single-ended and pseudo differential modes fIN = 10 kHz sine wave; differential mode fIN = 10 kHz sine wave; single-ended and pseudo differential modes fIN = 0.5 Hz to 100 kHz AD7294-2 Parameter TEMPERATURE SENSOR--INTERNAL Operating Range Accuracy Resolution Update Rate TEMPERATURE SENSOR--EXTERNAL Operating Range Accuracy Resolution Low Level Output Current Source2 Medium Level Output Current Source2 High Level Output Current Source2 Maximum Series Resistance (RS) for External Diode2 Maximum Parallel Capacitance (CP) for External Diode2 CURRENT SENSE VPP Supply Range Gain Error Differential Input RS(+)/RS(-) Input Bias Current CMRR/PSRR2 Offset Error Offset Drift Amplifier Peak-to-Peak Noise2 VPP Supply Current REFERENCE Reference Output Voltage Reference Input Voltage Range2 DC Leakage Current VREF Output Impedance2 Input Capacitance2 Reference Temperature Coefficient Data Sheet Min Typ -40 Max Unit Test Conditions/Comments +105 2 2.5 C C C C ms Internal temperature sensor, TA = -30C to +90C Internal temperature sensor, TA = -40C to +105C LSB size +150 2 10 C C C A A A k 1 nF 59.4 0.75 V % FSR mV A dB V V/C V mA 0.25 5 -55 0.25 8 32 128 External transistor is 2N3906 Limited by external diode TA = TDIODE = -40C to +105C LSB size For <0.5C additional error, CP = 0 (see Figure 29) RS = 0 (see Figure 28) VPP = AVDD to 59.4 V AVDD 200 25 110 50 3 400 0.18 2.49 0.1 2.5 32 780 0.25 2.51 4.1 2 5 20 10 25 V V A pF ppm/C See the Terminology section for more information. Guaranteed by design and characterization; not production tested. VIN+ and VIN- must remain within AGND/AVDD. (The analog input pins are VIN3 to VIN0.) 4 VIN- = 0 V for specified performance. For full input range on VIN-, see Figure 36. 1 2 3 Rev. 0 | Page 6 of 44 2.5 V reference 2.5 V reference Inputs shorted to VPP Referred to input Per channel, VPP1 = VPP2 = 59.4 V 0.2% maximum at 25C A buffer is required if the reference output is used to drive external loads Data Sheet AD7294-2 GENERAL SPECIFICATIONS AVDD = 4.5 V to 5.5 V, AGND1 to AGND7 = DGND = 0 V, internal or external 2.5 V reference; VDRIVE = 2.7 V to 5.5 V; VPPx = AVDD to 59.4 V; DAC OUTV+ AB and DAC OUTV+ CD = 4.5 V to 16.5 V; OFFSET IN x is floating; therefore, DAC output span = 0 V to 5 V; TA = -40C to +105C, unless otherwise noted. Table 3. Parameter LOGIC INPUTS Input High Voltage Input Low Voltage Input Leakage Current Input Hysteresis 1 Input Capacitance1 Glitch Rejection1 Symbol Min VIH VIL IIN VHYST CIN 0.7 VDRIVE Typ 0.3 VDRIVE 1 0.05 VDRIVE 8 50 Maximum External Capacitance of I2C Address Pins When Floating1 LOGIC OUTPUTS SDA, ALERT Output Low Voltage Floating-State Leakage Current1 Floating-State Output Capacitance1 ISENSE OVERRANGE Output High Voltage Output Low Voltage Overrange Setpoint1 POWER REQUIREMENTS VPP1, VPP2 AVDD DAC OUTV+ xx VDRIVE IDD DAC OUTV+ xx, IDD 30 VOL V V A V pF ns SDA, SCL only SDA, SCL only pF VDRIVE - 0.2 0.2 V V mV 5.3 0.6 59.4 5.5 16.5 5.5 7.5 1.2 V V V V mA mA 70 110 mW 4.4 5.5 mA 35 60 70 A mW VFS x 1.2 AVDD 4.5 4.5 2.7 DAC OUTV+ x, IDD Power Dissipation Test Conditions/Comments V V A pF VOH VOL VFS Unit 0.4 0.6 1 8 Power Dissipation Power-Down IDD 1 Max Guaranteed by design and characterization; not production tested. Rev. 0 | Page 7 of 44 Input filtering suppresses noise spikes of less than 50 ns Tristate input SDA and ALERT/BUSY are open-drain outputs ISINK = 3 mA ISINK = 6 mA ISENSEx OVERRANGE are push-pull outputs ISOURCE = 200 A for push-pull outputs ISINK = 200 A for push-pull outputs VFS = VREF ADC/12.5 AVDD + VDRIVE; DAC outputs unloaded At midscale output voltage, DAC outputs unloaded AVDD and VDRIVE; ADC, DACs, and temperature sensor powered down AD7294-2 Data Sheet TIMING CHARACTERISTICS I2C Serial Interface AVDD = 4.5 V to 5.5 V, AGND1 to AGND7 = DGND = 0 V, internal or external 2.5 V reference; VDRIVE = 2.7 V to 5.5 V; VPPx = AVDD to 59.4 V; DAC OUTV+ AB and DAC OUTV+ CD = 4.5 V to 16.5 V; OFFSET IN x is floating, therefore, DAC output span = 0 V to 5 V; TA = -40C to +105C, unless otherwise noted. Table 4. Parameter 1 fSCL t1 t2 t3 t4 t5 t6 Limit at TMIN, TMAX 400 2.5 0.6 1.3 0.6 100 0.9 0 0.6 0.6 1.3 300 0 300 20 x (VDRIVE/5.5 V) 0 300 400 t7 t8 t9 t10 2 t112 Cb 3 Unit kHz max s min s min s min s min ns min s max s min s min s min s min ns max ns min ns max ns min ns min ns max pF max Symbol tHIGH tLOW tHD,STA tSU,DAT tHD,DAT tHD,DAT tSU,STA tSU,STO tBUF tR tR tF tF tF tF Description SCL clock frequency SCL cycle time SCL high time SCL low time Start/repeated start condition hold time Data setup time Data hold time Data hold time Setup time for repeated start Stop condition setup time Bus free time between a stop and a start condition Rise time of SCL and SDA when receiving Rise time of SCL and SDA when receiving (CMOS compatible) Fall time of SDA when transmitting Fall time of SCL and SDA when transmitting Fall time of SDA when receiving (CMOS compatible) Fall time of SCL and SDA when receiving Capacitive load for each bus line See Figure 2. tR and tF are measured between 0.3 VDD and 0.7 VDD. 3 Cb is the total capacitance in pF of one bus line. 1 2 Timing and Circuit Diagrams SDA t3 t9 t10 t11 t4 SCL t2 t1 t5 START CONDITION REPEATED START CONDITION STOP CONDITION Figure 2. I2C-Compatible Serial Interface Timing Diagram 200A IOL VOH (MIN) OR VOL (MAX) TO OUTPUT PIN CL 50pF 200A t8 t7 IOH Figure 3. Load Circuit for Digital Output Rev. 0 | Page 8 of 44 10936-002 t6 10936-003 t4 Data Sheet AD7294-2 ABSOLUTE MAXIMUM RATINGS TA = 25C, unless otherwise noted.1 Table 5. Parameter VPPx to AGND AVDD to AGND DAC OUTV+ AB to AGND DAC OUTV+ CD to AGND VDRIVE to OPGND Digital Inputs to OPGND RESET to OPGND SDA/SCL to OPGND Digital Outputs to OPGND RS(+)/RS(-) to VPPx REFOUT/REFIN ADC to AGND REFOUT/REFIN DAC to AGND OPGND to AGND OPGND to DGND AGND to DGND VOUTx to AGND Analog Inputs to AGND Operating Temperature Range Storage Temperature Range Junction Temperature (TJ MAX) ESD, Human Body Model Reflow Soldering Peak Temperature 1 Rating -0.3 V to +65 V -0.3 V to +7 V -0.3 V to +17 V -0.3 V to +17 V -0.3 V to +7 V -0.3 V to VDRIVE + 0.3 V -0.3 V to +7 V -0.3 V to +7 V -0.3 V to VDRIVE + 0.3 V VPPx - 0.3 V to VPPx + 0.3 V -0.3 V to AVDD + 0.3 V -0.3 V to AVDD + 0.3 V -0.3 V to +0.3 V -0.3 V to +0.3 V -0.3 V to +0.3 V -0.3 V to DAC OUTV+ xx + 0.3 V -0.3 V to AVDD + 0.3 V -40C to +105C -65C to +150C 150C 1 kV 260C 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 indicated in the operational section of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. To conform with IPC-2221 industrial standards, it is advisable to use conformal coating on the high voltage pins. THERMAL RESISTANCE Table 6. Thermal Resistance Package Type 64-Lead TQFP ESD CAUTION Transient currents of up to 100 mA do not cause SCR latch-up. Rev. 0 | Page 9 of 44 JA 54 JC 16 Unit C/W AD7294-2 Data Sheet VIN3 VIN2 VIN1 REF OUT/REFIN ADC VIN0 DCAP AV DD AGND6 AGND7 NC NC RS1(+) RS1(-) VPP1 VPP2 NC PIN CONFIGURATION AND FUNCTION DESCRIPTIONS 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 NC 1 RS2(-) 2 RS2(+) 3 46 ISENSE 1 OVERRANGE NC 4 45 ISENSE 2 OVERRANGE NC 5 44 RESET AGND1 48 DGND PIN 1 INDICATOR 47 DGND 43 DGND 6 AGND2 7 NC 8 D2(-) 9 42 VDRIVE 41 OPGND AD7294-2 TQFP TOP VIEW (Not to Scale) D2(+) 10 40 SCL 39 SDA D1(+) 11 38 AS0 D1(-) 12 AGND3 13 37 AS1 36 AS2 35 ALERT/BUSY FACTORY TEST 14 34 AGND5 REF OUT/REFIN DAC 15 33 NC NC 16 10936-005 NC VOUTD OFFSET IN D DAC OUT GND CD VOUTC DAC OUTV+ CD NC OFFSET IN C AGND4 OFFSET IN B VOUTB DAC OUTV+ AB DAC OUT GND AB VOUTA NC OFFSET IN A 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 NOTES 1. NC = NO INTERNAL CONNECTION. Figure 4. Pin Configuration Table 7. Pin Function Descriptions Pin No. 2, 61 3, 60 1, 4, 5, 8, 16, 17, 25, 32, 33, 57 59, 64 14 Mnemonic RS2(-), RS1(-) RS2(+), RS1(+) NC Description Connection for External Shunt Resistor. Connection for External Shunt Resistor. This pin has no internal connection. FACTORY TEST 56 AVDD 6, 7, 13, 24, 34, 55, 58 AGND1 to AGND7 9, 12 D2(-), D1(-) 10, 11 D2(+), D1(+) 15 REFOUT/REFIN DAC Factory Test Pin. To maintain pin compatibility with the AD7294, this pin can tolerate being connected to voltages of up to 5.5 V. Analog Supply Pin. The operating range is 4.5 V to 5.5 V. This pin provides the supply voltage for all the analog circuitry on the AD7294-2. This supply should be decoupled to AGND with one 10 F tantalum capacitor and a 0.1 F ceramic capacitor. Analog Ground. Ground reference point for all analog circuitry on the AD7294-2. Refer all analog input signals and any external reference signal to this AGND voltage. Connect all seven of these AGND pins to the AGND plane of the system. Note that AGND5 is a DAC ground reference point and should be used as a star ground for circuitry being driven by the DAC outputs. Ideally, the AGND and DGND voltages should be at the same potential and must not be more than 0.3 V apart, even on a transient basis. Temperature Sensor Analog Inputs. These pins are connected to the external temperature sensing transistor. See Figure 43 and Figure 44. Temperature Sensor Analog Inputs. These pins are connected to the external temperature sensing transistor. See Figure 43 and Figure 44. DAC Reference Output/Input Pin. The REFOUT/REFIN DAC pin is common to all four DAC channels. On power-up, the default configuration of this pin is as an external reference (REFIN). Enable the internal reference by writing to the power-down register; see Table 27. Decoupling capacitors (220 nF recommended) are connected to this pin to decouple the reference buffer. If the output is buffered, the on-chip reference can be taken from this pin and applied externally to the rest of a system. A maximum external reference voltage of AVDD - 2 V can be supplied to the REFOUT portion of the REFOUT/REFIN DAC pin. Rev. 0 | Page 10 of 44 Data Sheet Pin No. 18, 23, 26, 31 Mnemonic OFFSET IN A to OFFSET IN D 19, 22, 27, 30 VOUTA to VOUTD 20, 29 35 DAC OUT GND AB, DAC OUT GND CD DAC OUTV+ AB, DAC OUTV+ CD ALERT/BUSY 38, 37, 36 AS0, AS1, AS2 39 40 SDA SCL 41 42 OPGND VDRIVE 43, 47, 48 44 DGND 46, 45 ISENSE1 OVERRANGE, ISENSE2 OVERRANGE VIN3 to VIN0 21, 28 49, 50, 51, 52 53 RESET REFOUT/REFIN ADC 54 DCAP 62, 63 VPP1, VPP2 AD7294-2 Description DAC Analog Offset Input Pins. These pins set the desired output range for each DAC channel. The DACs have an output voltage span of 5 V, which can be shifted from 0 V to 5 V to a maximum output voltage of 10 V to 15 V by supplying an offset voltage to these pins. These pins can be left floating, in which case decouple them to AGND with a 100 nF capacitor. Buffered Analog DAC Outputs for Channel A to Channel D. Each DAC analog output is driven from an output amplifier that can be offset using the OFFSET IN x pin. The DAC has a maximum output voltage span of 5 V that can be level shifted to a maximum output voltage level of 15 V. Each output is capable of sourcing and sinking 10 mA and driving a 10 nF load. Analog Ground. Analog ground pins for the DAC output amplifiers on VOUTA and VOUTB, and VOUTC and VOUTD, respectively. Analog Supply. Analog supply pins for the DAC output amplifiers on VOUTA and VOUTB, and VOUTC and VOUTD, respectively. The operating range is 4.5 V to 16.5 V. Digital Output. Selectable as an alert or busy output function in the configuration register. This is an open-drain output. An external pull-up resistor is required. When configured as an alert, this pin acts as an out-of-range indicator and becomes active when the conversion result violates the DATAHIGH or DATALOW register values. See the Alert Status Registers section. When configured as a busy output, this pin becomes active when a conversion is in progress. Digital Logic Inputs. Together, the logic state of these inputs selects a unique I2C address for the AD7294-2. See Table 34 for more information. Digital Input/Output. Serial bus bidirectional data; external pull-up resistor required. Serial I2C Bus Clock. The data transfer rate in I2C mode is compatible with both 100 kHz and 400 kHz operating modes. Open-drain input; external pull-up resistor required. Dedicated Ground Pin for I2C Interface. Logic Power Supply. The voltage supplied at this pin determines at what voltage the interface operates. Decouple this pin to DGND. The voltage range on this pin is 2.7 V to 5.5 V; it may be different from the voltage level at AVDD but should never exceed it by more than 0.3 V. To set the input and output thresholds, connect this pin to the supply to which the I2C bus is pulled. Digital Ground. This pin is the ground for all digital circuitry. Reset. Taking RESET low performs a reset of the I2C interface logic. The logic input threshold of the pin is set by VDRIVE (Pin 42). To maintain pin compatibility with the AD7294, this pin can tolerate being connected to voltages of up to 5.5 V. Fault Comparator Outputs. These pins connect to the high-side current sense amplifiers. Uncommitted ADC Analog Inputs. These pins are programmable as four single-ended channels or two true differential analog input channel pairs. See Table 2 and Table 10 for more information. ADC Reference Output/Input Pin. The REFOUT/REFIN ADC pin provides the reference source for the ADC. On power-up, the default configuration of this pin is as an external reference (REFIN). Enable the internal reference by writing to the power-down register (see Table 27). Connect decoupling capacitors (220 nF recommended) to this pin to decouple the reference buffer. If the output is buffered, the on-chip reference can be taken from this pin and applied externally to the rest of a system. A maximum external reference voltage of 2.5 V can be supplied to the REFOUT portion of this pin. External Decoupling Capacitor Input for Internal Temperature Sensor. Decouple this pin to AGND using a 0.1 F capacitor. In normal operation, the voltage is typically 1.25 V. Current Sensor Supply Pins. Power supply pins for the high-side current sense amplifiers. The operating range is from AVDD to 59.4 V. Decouple these supplies to AGND. Refer to the Current Sense Filtering section for more information about using these pins. Rev. 0 | Page 11 of 44 AD7294-2 Data Sheet TYPICAL PERFORMANCE CHARACTERISTICS 0 0 -20 AMPLITUDE (dB) -40 -60 -80 -80 -100 -120 -120 -140 0 2000 4000 6000 8000 10000 FREQUENCY (kHz) -140 0 2000 4000 6000 8000 10000 FREQUENCY (kHz) Figure 5. Signal-to-Noise Ratio, Single-Ended Input, VREF Range Figure 8. Signal-to-Noise Ratio, Differential Input, 2 x VREF Range 1.00 0 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE fSAMPLE = 20kSPS fSIGNAL = 10303Hz SINGLE-ENDED -20 AVDD = 5V, VDRIVE = 5V VREF RANGE 0.75 0.50 INL (LSB) -40 AMPLITUDE (dB) -60 -100 10936-105 AMPLITUDE (dB) -40 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE fSAMPLE = 20kSPS fSIGNAL = 10303Hz DIFFERENTIAL -20 10936-108 AVDD = 5V, VDRIVE = 5V VREF RANGE fSAMPLE = 20kSPS fSIGNAL = 10303Hz SINGLE-ENDED -60 -80 0.25 0 -0.25 -100 -0.50 -120 0 2000 4000 6000 8000 10000 FREQUENCY (kHz) -1.00 10936-106 -140 0 1000 1500 2000 2500 3000 3500 4095 CODE Figure 6. Signal-to-Noise Ratio, Single-Ended Input, 2 x VREF Range Figure 9. ADC INL, Single-Ended Input, VREF Range 1.00 0 AVDD = 5V, VDRIVE = 5V VREF RANGE fSAMPLE = 20kSPS fSIGNAL = 10303Hz DIFFERENTIAL -20 AVDD = 5V, VDRIVE = 5V VREF RANGE 0.75 0.50 DNL (LSB) -40 -60 -80 0.25 0 -0.25 -100 -0.50 -120 -140 0 2000 4000 6000 8000 10000 FREQUENCY (kHz) -1.00 0 500 1000 1500 2000 2500 3000 3500 CODE Figure 10. ADC DNL, Single-Ended Input, VREF Range Figure 7. Signal-to-Noise Ratio, Differential Input, VREF Range Rev. 0 | Page 12 of 44 4095 10936-110 -0.75 10936-107 AMPLITUDE (dB) 500 10936-109 -0.75 Data Sheet AD7294-2 1.00 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE 0.50 0.25 0.25 INL (LSB) 0.50 0 0 -0.25 -0.25 -0.50 -0.50 -0.75 -0.75 -1.00 0 500 1000 1500 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE 0.75 2000 2500 3000 3500 4095 CODE -1.00 10936-111 0 1500 2000 2500 3000 3500 4095 Figure 14. ADC INL, Differential Input, VREF Range 1.00 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE 0.75 0.50 0.25 0.25 DNL (LSB) 0.50 0 0 -0.25 -0.25 -0.50 -0.50 -0.75 -0.75 500 1000 1500 2000 2500 3000 3500 4095 CODE -1.00 10936-112 -1.00 0 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE 0.75 0 500 1000 1500 2000 2500 3000 3500 4095 CODE Figure 12. ADC DNL, Single-Ended Input, 2 x VREF Range 10936-114 1.00 DNL (LSB) 1000 CODE Figure 11. ADC INL, Single-Ended Input, 2 x VREF Range Figure 15. ADC DNL, Differential Input, 2 x VREF Range 1.00 1.00 AVDD = 5V, VDRIVE = 5V VREF RANGE 0.75 0.50 0.25 0.25 DNL (LSB) 0.50 0 0 -0.25 -0.25 -0.50 -0.50 -0.75 -0.75 -1.00 0 500 1000 1500 AVDD = 5V, VDRIVE = 5V 2 x VREF RANGE 0.75 2000 2500 3000 3500 CODE 4095 10936-113 INL (LSB) 500 Figure 13. ADC INL, Differential Input, VREF Range -1.00 0 500 1000 1500 2000 2500 3000 3500 CODE Figure 16. ADC DNL, Differential Input, 2 x VREF Range Rev. 0 | Page 13 of 44 4095 10936-116 INL (LSB) 0.75 10936-115 1.00 AD7294-2 Data Sheet 1.5 0.6 1.0 0.4 DNL ERROR (LSB) MAX INL INL (LSB) 0.5 0 MIN INL -0.5 -1.0 0.2 0 -0.2 -0.4 AVDD = 5V VDRIVE = 5V 10936-120 3584 3840 3328 3072 2816 2560 2304 1792 2048 1280 1536 768 1024 -0.6 512 6 5 0 4 2 3 REFERENCE VOLTAGE (V) 1 256 0 10936-117 -1.5 CODE Figure 20. DAC DNL Figure 17. ADC INL vs. Reference Voltage 1.0 0.8 0.6 MAX DNL DNL (LSB) 0.4 0.2 1 0 AVDD = 5V VDRIVE = 5V -0.2 -0.4 10936-121 MIN DNL -0.8 0 2 3 4 REFERENCE VOLTAGE (V) 1 6 5 10936-118 -0.6 CH1 20V A CH4 1.02V Figure 21. 0.1 Hz to 10 Hz DAC Output Noise, Input Code = 0x800 Figure 18. ADC DNL vs. Reference Voltage 2.0 5.0 4.5 DAC OUTPUT VOLTAGE (V) 1.5 1.0 0.5 0 -0.5 -1.0 4.0 3.5 3.0 2.5 0nF 1nF 10nF 2.0 1.5 1.0 -1.5 3840 3584 3072 3328 0 10936-119 CODE 2816 2560 2304 2048 1792 1536 1280 768 1024 512 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 TIME (s) Figure 22. Settling Time for a 1/4 to 3/4 Output Voltage Step Figure 19. DAC INL Rev. 0 | Page 14 of 44 10936-122 0.5 -2.0 256 INL ERROR (LSB) M1.00s Data Sheet AD7294-2 100 3.760 CHANGE IN OUTPUT VOLTAGE (mV) DAC OUTPUT VOLTAGE (V) 80 3.758 0nF 1nF 10nF 3.756 3.754 3.752 60 40 20 0 -20 -40 -60 -100 -50 10936-123 3.750 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 TIME (s) Figure 23. Zoomed-In Settling Time for a 1/4 to 3/4 Output Voltage Step -40 -30 -20 0 10 20 -10 LOAD CURRENT (mA) 30 40 10936-126 -80 50 Figure 26. DAC Output Voltage vs. Load Current, Input Code = 0x800 1.0 55 0.9 50 TEMPERATURE READING (C) 0.8 0.6 0.5 0.4 0.3 0.2 40 35 30 0 5 10 15 20 25 30 35 40 SINK CURRENT (mA) 20 -0.1 10936-124 0 -1 -0.2 -2 -0.3 -3 ERROR (C) 0 -0.1 -0.4 -0.5 -0.6 -8 -9 -1.0 -10 25 30 35 40 10936-125 -0.9 20 0.5 0.6 0.7 0.8 0.9 1.0 -6 -0.8 15 0.4 -5 -7 SOURCE CURRENT (mA) 0.3 -4 -0.7 10 0.2 Figure 27. Response of Temperature Sensor to a Step Function 0 5 0.1 TIME (Seconds) Figure 24. DAC Sinking Current at Input Code = 0x000, (VOUT = 0 V) 0 0 10936-127 25 0.1 VOUT (V) 45 Figure 25. DAC Sourcing Current at Input Code = 0xFFF, (VOUT = AVDD) Rev. 0 | Page 15 of 44 0 0.5 1.0 1.5 2.0 CAPACITANCE FROM Dx(+) TO Dx(-) (nF) 2.5 10936-128 VOUT (V) 0.7 Figure 28. Temperature Error vs. Capacitance from Dx(+) to Dx(-) Data Sheet 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 -50 -60 PSRR (dB) -70 -90 2 4 6 8 10 12 14 16 SERIES RESISTANCE (k) -120 Figure 29. Temperature Error vs. Series Resistance 1k 0 -10 -15 -20 -25 -30 -35 -40 100k 1M FREQUENCY (Hz) 10M 100M 10936-130 -45 10k 1M 10M Figure 31. ISENSE Power Supply Rejection Ratio (PSRR) vs. Supply Ripple Frequency Without VPP Supply Decoupling Capacitors for a 500 mV Ripple -5 1k 100k FREQUENCY (Hz) 5 -50 100 10k 10936-131 -110 0 AMPLITUDE (dB) -80 -100 10936-129 ERROR (C) AD7294-2 Figure 30. Frequency Response of the High-Side Current Sensor Rev. 0 | Page 16 of 44 Data Sheet AD7294-2 TERMINOLOGY DAC TERMINOLOGY Relative Accuracy A measure of the maximum deviation, in LSBs, from a straight line passing through the endpoints of the DAC transfer function. Differential Nonlinearity (DNL) 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 by design. Zero Code Error A measure of the output error when zero code (0x0000) is loaded to the DAC register. Ideally, the output should be 0 V. The zero code error is always positive in the AD7294-2 because the output of the DAC cannot go below 0 V. Zero code error is expressed in mV. Full-Scale Error A measure of the output error when full-scale code (0xFFFF) is loaded to the DAC register. Ideally, the output should be VDD - 1 LSB. Full-scale error is expressed in mV. Gain Error A measure of the span error of the DAC. It is the deviation in slope of the DAC transfer characteristic from ideal, expressed as a percent of the full-scale range (% FSR). Gain Error Drift A measure of the change in gain error with changes in temperature. It is expressed in (ppm of full-scale range)/C. noise. The theoretical signal-to-noise-and-distortion ratio for an ideal N-bit converter with a sine wave input is given by Signal-to-Noise-and-Distortion = (6.02 N + 1.76) dB Thus, the SINAD is 74 dB for an ideal 12-bit converter. Total Harmonic Distortion (THD) The ratio of the rms sum of harmonics to the fundamental. For the AD7294-2, THD is defined as V2 + V3 + V4 + V5 + V6 2 THD (dB) = 20 log 2 2 2 2 V1 where V1 is the rms amplitude of the fundamental and V2, V3, V4, V5, and V6 are the rms amplitudes of the second through sixth harmonics. Integral Nonlinearity (INL) The maximum deviation from a straight line passing through the endpoints of the ADC transfer function. The endpoints are zero scale, a point 1 LSB below the first code transition, and full scale, a point 1 LSB above the last code transition. Differential Nonlinearity (DNL) The difference between the measured change and the ideal 1 LSB change between any two adjacent codes in the ADC. Offset Error The deviation of the first code transition (00...000) to (00...001) from the ideal--that is, AGND + 1 LSB. Offset Error Match The difference in offset error between any two channels. ADC TERMINOLOGY Signal-to-Noise-and-Distortion Ratio (SINAD) The measured ratio of signal-to-noise and distortion at the output of the ADC. The signal is the rms amplitude of the fundamental. Noise is the sum of all nonfundamental signals up to half the sampling frequency (fS/2), excluding dc. The ratio is dependent on the number of quantization levels in the digitization process; the more levels, the smaller the quantization Gain Error The deviation of the last code transition (111 ... 110 to 111 ... 111) from the ideal (that is, REFIN - 1 LSB) after the offset error has been adjusted out. Gain Error Match The difference in gain error between any two channels. Rev. 0 | Page 17 of 44 AD7294-2 Data Sheet THEORY OF OPERATION 1LSB = 2 x VREF /4096 011...111 011...110 The various monitored and uncommitted input signals are multiplexed into the ADC. The AD7294-2 has four uncommitted analog input channels, VIN0 to VIN3. These four channels allow singleended, differential, and pseudo differential mode measurements of various system signals. 100...001 100...000 -VREF + 1LSB VREF - 1LSB +VREF - 1LSB ANALOG INPUT Figure 33. Differential Transfer Characteristics with VREF VREF Input Range For VIN0 to VIN3 in single-ended mode, the output code is straight binary, where VIN = 0 V, DOUT = x000, VIN = VREF - 1 LSB, and DOUT = 0xFFF 111...110 In differential mode, the code is twos complement, where VIN+ - VIN- = 0 V, and DOUT = 0x00 VIN+ - VIN- = VREF - 1 LSB, and DOUT = 00x7FF VIN+ - VIN- = -VREF, and DOUT = 0x800 111...000 Channel 5 and Channel 6 (current sensor inputs) are twos complement, where 111...111 ADC CODE 111...111 100...010 ADC TRANSFER FUNCTIONS The designed code transitions occur at successive integer LSB values (1 LSB, 2 LSB, and so on). In single-ended mode, the LSB size is VREF/4096 when the 0 V to VREF range is used, and 2 x VREF/4096 when the 0 V to 2 x VREF range is used. The ideal transfer characteristic for the ADC, when outputting straight binary coding, is shown in Figure 32. 000...001 000...000 1LSB = VREF /4096 10936-017 The AD7294-2 provides the user with a 9-channel multiplexer, an on-chip track-and-hold, and a successive approximation ADC based on four capacitive DACs. The analog input range for the part can be selected as a 0 V to VREF input or a 2 x VREF input, configured with either single-ended or differential analog inputs. An on-chip 2.5 V reference can be disabled when an external reference is preferred. If the internal ADC reference is to be used elsewhere in a system, the output must first be buffered. In differential mode, the LSB size is 2 x VREF/4096 when the 0 V to VREF range is used, and 4 x VREF/4096 when the 0 V to 2 x VREF range is used. The ideal transfer characteristic for the ADC, when outputting twos complement coding, is shown in Figure 33 (with the 2 x VREF range). ADC CODE ADC OVERVIEW VIN+ - VIN- = 0 mV, and DOUT = 0x000 011...111 VIN+ - VIN- = VREF/12.5 - 1 LSB, and DOUT = 0x7FF VIN+ - VIN- = -VREF/12.5, and DOUT = 0x800 000...010 000...001 Channel 7 to Channel 9 (temperature sensor inputs) are twos complement with the LSB equal to 0.25C, where 000...000 0V 1LSB VREF - 1LSB NOTE 1. VREF IS EITHER VREF OR 2 x VREF . Figure 32. Single-Ended Transfer Characteristics 10936-016 ANALOG INPUT TIN = 0C, and DOUT = 0x000 TIN = +255.75C, and DOUT = 0x7FF TIN = -256C, and DOUT = 0x800 Rev. 0 | Page 18 of 44 Data Sheet AD7294-2 ANALOG INPUTS The AD7294-2 has four analog inputs, VIN3 to VIN0. Depending on the configuration register setup, they can be configured as two single-ended inputs, two pseudo differential channels, or two fully differential channels (see the Register Settings section). Single-Ended Mode The AD7294-2 can have four single-ended analog input channels. In applications where the signal source has high impedance, it is recommended that the analog input be buffered before it is applied to the ADC. The analog input range can be programmed to either of the following modes: 0 V to VREF or 0 V to 2 x VREF. In 2 x VREF mode, the input is effectively divided by 2 before the conversion takes place. Note that the voltage, with respect to GND on the ADC analog input pins, cannot exceed AVDD. If the analog input signal to be sampled is bipolar, the internal reference of the ADC can be used to externally bias up this signal so that it is correctly formatted for the ADC. Figure 34 shows a typical connection diagram when operating the ADC in single-ended mode. 0V 0V R VIN VIN0 3R -1.25V AD7294-21 VIN3 REFOUT ADC (VIN+ + VIN-)/2 The common-mode voltage is, therefore, the voltage on which the two inputs are centered. The result is that the span of each input is VCM VREF/2. This common-mode voltage must be set up externally, and its range varies with the reference value, VREF. As the value of VREF increases, the common-mode range decreases. When driving the inputs with an amplifier, the actual common-mode range is determined by the output voltage swing of the amplifier. The common-mode voltage must be within this common-mode range to guarantee the functionality of the AD7294-2. Driving Differential Inputs 10936-018 0.47F PINS OMITTED FOR CLARITY. The common-mode voltage is the average of the two signals. If the 2 x VREF range is used, the input signal amplitude extends from -2 x VREF (VIN+ = 0 V, VIN- = VREF) to +2 x VREF (VIN- = 0 V, VIN+ = VREF). R 1ADDITIONAL Assuming that the 0 V to VREF range is selected, the amplitude of the differential signal is, therefore, -VREF to +VREF peak-topeak (2 x VREF), regardless of the common-mode voltage (VCM). When a conversion takes place, the common-mode voltage is rejected, resulting in a virtually noise-free signal of amplitude -VREF to +VREF, corresponding to the digital output codes of -2048 to +2047 in twos complement format. +2.5V R +1.25V Simultaneously drive VIN0 and VIN1 by two signals, each of amplitude VREF (or 2 x VREF, depending on the range chosen), that are 180 out of phase. Figure 34. Single-Ended Mode Connection Diagram Differential Mode The AD7294-2 can have two differential analog input pairs. Differential signals have some benefits over single-ended signals, including noise immunity based on the commonmode rejection of the device and improvements in distortion performance. Figure 35 defines the fully differential analog input of the AD7294-2. The differential modes that are available on VIN0 to VIN3 (see Table 10) require that VIN+ and VIN- be driven simultaneously with two equal signals that are 180 out of phase. The commonmode voltage on which the analog input is centered must be set up externally. The common-mode range is determined by VREF, the power supply, and the particular amplifier used to drive the analog inputs. Differential modes of operation with either an ac or dc input provide the best THD performance over a wide frequency range. Because not all applications have a signal that is preconditioned for differential operation, there is often a need to perform a single-ended-to-differential conversion. Using an Op Amp Pair VREF p-p An op amp pair can be used to directly couple a differential signal to one of the analog input pairs of the AD7294-2. The circuit configuration that is illustrated in Figure 38 shows how a dual op amp can be used to convert a single-ended bipolar signal into a differential unipolar input signal. AD7294-21 VREF p-p VIN- 1ADDITIONAL PINS OMITTED FOR CLARITY. 10936-019 COMMON-MODE VOLTAGE VIN+ Figure 35. Differential Input Definition The amplitude of the differential signal is the difference between the signals applied to VIN+ and VIN- in each differential pair (VIN+ - VIN-). The resulting converted data is stored in twos complement format in the result register. The voltage applied to Point A sets up the common-mode voltage. As shown in Figure 38, Point A connects to the reference, but any value in the common-mode range can be the input at Point A to set up the common-mode voltage. The AD8022 is a suitable dual op amp that can be used in this configuration to provide differential drive to the AD7294-2. Rev. 0 | Page 19 of 44 AD7294-2 Data Sheet Care is required when choosing the op amp because the selection depends on the required power supply and system performance objectives. The driver circuit in Figure 38 is optimized for dc coupling applications that require best distortion performance. 2.5 AVDD = 5V VDRIVE = 5V 2.0 1.5 VIN- (V) The differential op amp driver circuit shown in Figure 38 is configured to convert and level shift a single-ended, ground referenced (bipolar) signal to a differential signal that is centered at the VREF level of the ADC. 1.0 0.5 Pseudo Differential Mode -0.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 VREF (V) 10936-137 0 The four uncommitted analog input channels can be configured as two pseudo differential pairs. Two uncommitted inputs, VIN0 and VIN1, are a pseudo differential pair, as are VIN2 and VIN3. In this mode, VIN+ is connected to the signal source, which can have a maximum amplitude of VREF (or 2 x VREF, depending on the range that is chosen) to make use of the full dynamic range of the part. A dc input is applied to VIN-. The voltage applied to this input provides an offset from ground or a pseudo ground for the VIN+ input. The channel specified as VIN+ is determined by the ADC channel allocation. The differential mode must be selected to operate in the pseudo differential mode. The resulting converted pseudo differential data is stored in twos complement format in the result register. Figure 36. VIN- Input Range vs. VREF in Pseudo Differential Mode VREF p-p AD7294-21 VIN+ DC INPUT VOLTAGE VIN- REFOUT/REFIN ADC For VIN0, the governing equation for the pseudo differential mode is VOUT = 2(VIN+ - VIN-) - VREF_ADC 1ADDITIONAL PINS OMITTED FOR CLARITY. 10936-026 0.47F Figure 37. Pseudo Differential Mode Connection Diagram where VIN+ is the single-ended signal and VIN- is a dc voltage. CURRENT SENSOR The benefit of pseudo differential inputs is that they separate the analog input signal ground from the ADC ground, allowing dc common-mode voltages to be cancelled. Two bidirectional high-side current sense amplifiers are provided that can accurately amplify differential current shunt voltages in the presence of high common-mode voltages from AVDD up to 59.4 V. Each amplifier can accept a 200 mV differential input. Both current sense amplifiers have a fixed gain of 12.5 and use an internal 2.5 V reference. Figure 36 shows the typical voltage range for VIN- while in pseudo differential mode, and Figure 37 shows a connection diagram for pseudo differential mode. An analog comparator is also provided with each amplifier for fault detection. The threshold is defined as 1.2 x Full-Scale Voltage Range 2 x VREF p-p GND 440 3.75V 2.5V 1.25V 220 V+ 27 VIN+ V- 220k 220 220 V+ A V- 3.75V 2.5V 1.25V 27 AD7294-21 VIN- REFOUT ADC 10k 0.47F 1ADDITIONAL PINS OMITTED FOR CLARITY. 10936-023 20k Figure 38. Dual Op Amp Circuit to Convert a Single-Ended Bipolar Signal into a Differential Unipolar Signal Rev. 0 | Page 20 of 44 Data Sheet AD7294-2 When this limit is reached, the output is latched onto a dedicated pin. This output remains high until the latch is cleared by writing to the appropriate register. RSENSE ILOAD AV DD TO 59.4V VPPx RSx(+) RSx(-) AD7294-2 R1 40k R2 40k A1 Q2 A2 R3 100k R4 100k VOUT TO MUX 10936-029 Q1 Figure 39. High-Side Current Sense The AD7294-2 current sense comprises two main blocks: a differential and an instrumentation amplifier. A load current flowing through the external shunt resistor produces a voltage at the input terminals of the AD7294-2. Resistors R1 and R2 connect the input terminals to the differential amplifier (A1). A1 nulls the voltage that appears across its own input terminals by adjusting the current through R1 and R2 with Transistors Q1 and Q2. Common-mode feedback maintains the sum of these currents at approximately 50 A. When the input signal to the AD7294-2 is zero, the currents in R1 and R2 are equal. When the differential signal is nonzero, the current increases through one of the resistors and decreases in the other. The current difference is proportional to the size and polarity of the input signal. The differential currents through Q1 and Q2 are converted into a differential voltage by R3 and R4. A2 is configured as an instrumentation amplifier, buffering this voltage and providing additional gain. Therefore, for an input voltage of 200 mV at the pins, an output span of 2.5 V is generated. The current sensors on the AD7294-2 are designed to remove any flicker noise and offset that are present in the sensed signal. This is achieved by using a chopping technique that is transparent to the user. The VSENSE signal is first converted by the AD7294-2, the analog inputs to the amplifiers are then swapped, and the differential voltage is once again converted by the AD7294-2. be required to use the full input range of the ADC, thus achieving maximum SNR performance. When the sense current is known, the voltage range of the AD7294-2 current sensor (200 mV) is divided by the maximum sense current to yield a suitable shunt value. If the power dissipation in the shunt resistor is too large, the size of the shunt resistor can be reduced; in this case, less of the ADC input range is used. Using less of the ADC input range produces conversion results that are more susceptible to noise and offset errors because offset errors are fixed and are, thus, more significant when smaller input ranges are used. RSENSE must be able to dissipate the I2R losses. If the power dissipation rating of the resistor is exceeded, its value may drift or the resistor may be damaged, resulting in an open circuit. This open circuit can cause a differential voltage across the terminals of the AD7294-2 in excess of the absolute maximum ratings. Additional protection is afforded to the current sensors on the AD7294-2 by the recommended current limiting resistors, RF1 and RF2, as shown in Figure 40. The AD7294-2 can handle a maximum continuous current of 30 mA; thus, an RF2 of 1 k provides adequate protection for the AD7294-2. If ISENSE has a large high frequency component, take care to choose a resistor with low inductance. Low inductance metal film resistors are best suited for these applications. Current Sense Filtering In some applications, it may be desirable to use external filtering to reduce the input bandwidth of the amplifier (see Figure 40). The -3 dB differential bandwidth of this filter is equal to BWDM = 1/(4 x x RC) Note that the maximum series resistance on the RS(+) and RS(-) inputs (see Figure 39) is limited to a maximum of 1 k due to back-to-back ESD protection diodes from RS(+) and RS(-) to VPPx. Also, note that if RF1 and RF2 are in series with R1 and R2 (see Figure 39), the gain of the amplifier is affected. Any mismatch between RF1 and RF2 can introduce an offset error. VPP RSENSE RF1 ILOAD RF2 10nF The two conversion results enable the digital removal of any offset or noise. Switches on the amplifier inputs enable this chopping technique to be implemented. The process typically requires 6 s, in total, to return a final result. VPPx RSx(+) CF RSx(-) AD7294-2 The resistor values used in conjunction with the current sense amplifiers are determined by the specific application requirements in terms of voltage, current, and power. Small resistors minimize power dissipation, have low inductance to prevent any induced voltage spikes, and provide good tolerance, which reduces current variations. The final values chosen are a compromise between low power dissipation and good accuracy. Low value resistors have less power dissipated in them, but higher value resistors may 10936-098 Choosing RSENSE Figure 40. Current Sense Filtering (RSx Can Be Either RS1 or RS2) For certain RF applications, the optimum value for RF1 and RF2 is 1 k, whereas CF can range from 1 F to 10 F. There is an additional decoupling capacitor for the VPPx supply. Its value is application dependent, but for initial evaluation, values in the range of 1 nF to 100 nF are recommended. Rev. 0 | Page 21 of 44 AD7294-2 Data Sheet Kelvin Sense Resistor Connection VDD SENSE RESISTOR CURRENT FLOW FROM SUPPLY 16 x I 4xI I I-BIAS MUX LPF T1 T2 LIMIT REGISTERS TEMP SENSOR D1- D2- TO ADC fC = 65kHz D2+ D1+ MUX BIAS DIODE REMOTE SENSING TRANSISTORS AD7294-2 CURRENT FLOW TO LOAD DCAP ALERT 10936-032 When using a low value sense resistor for high current measurement, the problem of parasitic series resistance can arise. The lead resistance can be a substantial fraction of the rated resistance, making the total resistance a function of lead length. Avoid this problem by using a Kelvin sense connection. This type of connection separates the current path through the resistor and the voltage drop across the resistor. Figure 41 shows the correct way to connect the sense resistor between the RS(+) and RS(-) pins of the AD7294-2. Figure 42. Internal and Remote Temperature Sensors KELVIN SENSE TRACES RSx(+) RSx(-) 10936-031 AD7294-2 Figure 41. Kelvin Sense Connections (RSx Can Be Either RS1 or RS2) ANALOG COMPARATOR LOOP The AD7294-2 contains two setpoint comparators that are used for independent analog control. This circuitry enables users to quickly detect if the sensed voltage across the shunt resistor has increased above the preset (VREF x 1.2)/12.5. If this increase occurs, the ISENSEx OVERRANGE pin is set to a high logic level, enabling appropriate action to be taken to prevent any damage to the external circuitry. The setpoint threshold level is fixed internally in the AD7294-2, and the current sense amplifier saturates above this level. The comparator is also triggered if a voltage of less than AVDD is applied to the RSENSE resistor or the VPPx pin. TEMPERATURE SENSOR Each input integrates, in turn, over a period of several hundred microseconds (s). This integration takes place continuously in the background, leaving the user free to perform conversions on the other channels. When the integration is complete, a signal passes to the control logic to initiate a conversion automatically. If the ADC is in command mode, the temperature conversion is performed as soon as the next conversion is completed. In autocycle mode, the conversion is inserted into an appropriate place in the current sequence (see the Register Settings section for further details. If the ADC is idle, the conversion takes place immediately. Three registers store the result of the last conversion on each temperature channel; these can be read at any time. In addition, in command mode, one or both of the two external channel registers can be read out as part of the output sequence. Remote Sensing Diode The AD7294-2 is designed to work with discrete transistors, 2N3904 and 2N3906. If an alternative transistor is used, the AD7294-2 operates as specified, provided that the conditions explained in the following sections are adhered to. Ideality Factor The AD7294-2 contains one local and two remote temperature sensors. The temperature sensors continuously monitor the three temperature inputs, and new readings are automatically available every 5 ms. The ideality factor of the transistor, nf, is a measure of the deviation of the thermal diode from ideal behavior. The AD7294-2 is trimmed for an nf value of 1.008. Use the following equation to calculate the error introduced at a Temperature T (C) when using a transistor whose nf does not equal 1.008: The on-chip, band gap temperature sensor measures the temperature of the system. Diodes are used in conjunction with the two remote temperature sensors to monitor the temperature of other critical board components. To factor in this error, the user can write the T value to the offset register. The AD7294-2 automatically adds it to, or subtracts it from, the temperature measurement. The temperature sensor module on the AD7294-2 is based on the three-current principle (see Figure 42), where three currents are passed through a diode and the forward voltage drop is measured at each diode, allowing the temperature to be calculated free of errors caused by series resistance. T = (nf - 1.008) x (273.15 K + T) Base Emitter Voltage The AD7294-2 operates as specified if the base emitter voltage is greater than 0.25 V at 8 A at the highest operating temperature, and less than 0.95 V at 128 A for the lowest operating temperature. Rev. 0 | Page 22 of 44 Data Sheet AD7294-2 hFE Variation Use a transistor with little variation in hFE (~50 to 150). Little variation in hFE indicates tight control of the VBE characteristics. to the amplifier. This architecture is inherently monotonic, voltage out, and low glitch. It is also linear because all of the resistors are of equal value. For RF applications, the use of high Q capacitors, functioning as a filter, protects the integrity of the measurement. Connect these capacitors, such as Johanson Technology 10 pF, high Q capacitors, Reference Code 500R07S100JV4T, between the base and the emitter, as close to the external device as possible. However, large capacitances affect the accuracy of the temperature measurement; thus, the recommended maximum capacitor value is 100 pF. In most cases, a capacitor is not required; the selection of any capacitor is dependent on the noise frequency level. R R TO OUTPUT AMPLIFIERS R AD7294-2 D+ R 10pF R 10936-028 D- 10936-099 2N3904 NPN Figure 43. Measuring Temperature Using an NPN Transistor Figure 45. Resistor String Structure Output Amplifiers AD7294-2 The purpose of Op Amp A1 is to buffer the DAC output range from 0 V to VREF. A second amplifier, Op Amp A2, is configured such that when an offset is applied to OFFSET IN x, its output voltage is 3x the offset voltage minus 2x the DAC voltage. D+ D- 10936-100 2N3906 PNP 10pF Figure 44. Measuring Temperature Using a PNP Transistor VOUT = 3 x VOFFSET - 2 x VDAC Series Resistance Cancellation The DAC word is digitally inverted on chip such that The AD7294-2 is designed to automatically cancel out the effect of parasitic, base, and collector resistance on the temperature reading. This feature provides a more accurate result, without the need for any user characterization of the parasitic resistance. The AD7294-2 can compensate for up to 10 k in a process that is transparent to the user. DAC OPERATION The AD7294-2 contains four 12-bit DACs that provide digital control with 12 bits of resolution and a 2.5 V internal reference. The DAC core is a thin film 12-bit string DAC with a 5 V output span and an output buffer that can drive the high voltage output stage. The DAC has a span of 0 V to 5 V with a 2.5 V reference input. The output range of the DAC, which is controlled by the offset input, can be positioned from 0 V to 15 V. Resistor String The resistor string structure is shown in Figure 45. It consists of a string of 2n resistors, each of Value R. The code loaded to the DAC register determines at which node on the string the voltage is tapped off to be fed into the output amplifier. The voltage is tapped off by closing one of the switches connecting the string VOUT = 3 x VOFFSET + 2 x (VDAC - VREF) D and VDAC = VREF x n 2 where: VDAC is the output of the DAC before digital inversion. D is the decimal equivalent of the binary code that is loaded to the DAC register. n is the bit resolution of the DAC. An example of the offset function is given in Table 8. Table 8. Offset Voltage Function Example Offset Voltage (V) 1.67 3.33 5.00 VOUT with 0x000 (V) 0 5 10 VOUT with 0xFFF (LSB) 5V-1 10 V - 1 15 V - 1 The user has the option of leaving the offset pin open, in which case the voltage on the noninverting input of Op Amp A2 is set by the resistor divider, giving Rev. 0 | Page 23 of 44 VOUT = 2 x VDAC AD7294-2 Data Sheet This configuration generates the 5 V output span from a 2.5 V reference. Digitally inverting the DAC allows the circuit to operate as a generic DAC when no offset is applied. If the offset pin is not driven, it is best practice to place a 100 nF capacitor between the pin and ground to improve both the settling time and the noise performance of the DAC. Note that a significant amount of power can be dissipated in the DAC outputs. ADC AND DAC REFERENCE The AD7294-2 has two independent, internal, high performance 2.5 V references, one for the ADC and the other for the four on-chip DACs. If the application requires an external reference, it can be applied to the REFOUT/REFIN DAC pin and/or to the REFOUT/REFIN ADC pin. Buffer the internal reference before it is used by external circuitry. Decouple both the REFOUT/REFIN DAC pin and the REFOUT/REFIN ADC pin to AGND, using a 220 nF capacitor. On power-up, the AD7294-2 is configured for use with an external reference. To enable the internal references, write a zero to both the D4 and D5 bits in the power-down register (see the Register Settings section). Both the ADC and DAC references require a minimum of 60 s to power up and settle to 12-bit performance when a 220 nF decoupling capacitor is used. The AD7294-2 can also operate with an external reference. Suitable reference sources for the AD7294-2 include the AD780, AD1582, ADR431, REF193, and ADR391. In addition, choosing a reference with an output trim adjustment, such as the ADR441, allows a system designer to trim system errors by setting a reference voltage to a voltage other than the nominal. Long-term drift is a measure of how much the reference drifts over time. A reference with a low long-term drift specification ensures that the overall solution remains stable during its entire lifetime. If an external reference is used, select a low temperature coefficient specification to reduce the temperature dependence of the system output voltage on ambient conditions. VDRIVE FEATURE The AD7294-2 also has a VDRIVE feature to control the voltage at which the I2C interface operates. The VDRIVE pin is connected to the supply to which the I2C bus is pulled. This pin sets the input and output threshold levels for the digital logic pins and the ISENSEx OVERRANGE pins. The VDRIVE feature allows the AD7294-2 to easily interface to both 3 V and 5 V processors. For example, if the AD7294-2 is operated with a VDD of 5 V, the VDRIVE pin can be powered from a 3 V supply, allowing a large dynamic range with low voltage digital processors. Thus, the AD7294-2 can be used with the 2 x VREF input range with a VDD of 5 V, yet it remains capable of interfacing to 3 V digital parts. Decouple the VDRIVE pin to DGND with a 100 nF capacitor and a 1 F capacitor. Rev. 0 | Page 24 of 44 Data Sheet AD7294-2 REGISTER SETTINGS The AD7294-2 contains internal registers (see Figure 46) that store conversion results, high and low conversion limits, and information to configure and control the device. COMMAND REGISTER Address 0x00 0x01 DAC REGISTERS TSENSE RESULT REGISTERS x 3 0x02 DATA ALERT STATUS REGISTERS x 3 CHANNEL SEQUENCE REGISTER 0x03 CONFIGURATION REGISTER 0x04 POWER-DOWN REGISTER DATA HIGH/ DATA LOW REGISTERS x 18 HYSTERESIS REGISTERS x 9 SDA SCL 10936-039 TSENSE OFFSET REGISTERS x 2 SERIAL BUS INTERFACE The address pointer register is an 8-bit register in which the six LSBs are used as pointer bits to store an address that points to one of the AD7294-2 data registers (see Table 9). Table 9. Register Addresses ADC RESULT REGISTER ADDRESS POINTER REGISTER ADDRESS POINTER REGISTER Figure 46. Register Structure Each data register has an address to which the address pointer register points when communicating with it. The command register is the only register that is a write only register; the remainder of the addresses have both read and write access. 0x05 0x06 0x07 0x08 0x09 0x0A 0x0B 0x0C 0x0D 0x0E 0x0F 0x10 0x11 0x12 0x13 0x14 0x15 0x16 0x17 0x18 0x19 0x1A 0x1B 0x1C 0x1D 0x1E 0x1F 0x20 0x21 0x22 0x23 0x24 0x25 0x26 0x27 0x40 0x41 Rev. 0 | Page 25 of 44 Register Name Command register ADC result register DACA value TSENSE1 result DACB value TSENSE2 result DACC value TSENSEINT result DACD value Alert Status Register A Alert Status Register B Alert Status Register C Channel sequence register Configuration register Power-down register DATALOW Register VIN0 DATAHIGH Register VIN0 Hysteresis Register VIN0 DATALOW Register VIN1 DATAHIGH Register VIN1 Hysteresis Register VIN1 DATALOW Register VIN2 DATAHIGH Register VIN2 Hysteresis Register VIN2 DATALOW Register VIN3 DATAHIGH Register VIN3 Hysteresis Register VIN3 DATALOW Register ISENSE1 DATAHIGH Register ISENSE1 Hysteresis Register ISENSE1 DATALOW Register ISENSE2 DATAHIGH Register ISENSE2 Hysteresis Register ISENSE2 DATALOW Register TSENSE1 DATAHIGH Register TSENSE1 Hysteresis Register TSENSE1 DATALOW Register TSENSE2 DATAHIGH Register TSENSE2 Hysteresis Register TSENSE2 DATALOW Register TSENSEINT DATAHIGH Register TSENSEINT Hysteresis Register TSENSEINT TSENSE1 offset register TSENSE2 offset register Factory test mode Factory test mode Access W R W R W R W R W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W N/A N/A AD7294-2 Data Sheet COMMAND REGISTER Bit D15 is reserved as an ALERT_FLAG bit. Table 12 lists the contents of the first byte that is read from the AD7294-2 results register; Table 13 lists the contents of the second byte read. A write to the command register (Address 0x00) puts the part into command mode. In command mode, the part cycles through the selected channels from LSB (Bit D0) to MSB (Bit D7) on each subsequent read (see Table 11). A channel is selected for conversion if a 1 is written to the desired bit in the command register. On power-up, all bits in the command register are set to 0. If the external TSENSE channels are selected in the command register byte, it is not actually requesting a conversion. The result of the last automatic conversion is output as part of the sequence (see the Modes of Operation section). ADC Channel Allocation The three channel address bits indicate which channel the result in the result register represents. Table 10 describes the channel ID bits (S.E. indicates single-ended, and DIFF indicates differential). Table 10. ADC Channel Allocation Function VIN0 (S.E.) or VIN0 - VIN1 (DIFF) VIN1 (S.E.) or VIN1 - VIN0 (DIFF) VIN2 (S.E.) or VIN2 - VIN3 (DIFF) VIN3 (S.E.) or VIN3 - VIN2 (DIFF) ISENSE1 ISENSE2 TSENSE1 TSENSE2 If a command mode conversion is required while the autocycle mode is active, it is necessary to disable the autocycle mode before proceeding to the command mode (see the Autocycle Mode section for more details). ADC RESULT REGISTER The ADC result register (Address 0x01) is a 16-bit, read only register. The conversion results for the four uncommitted ADC inputs and the two ISENSE channels are stored in the result register for reading. Bit D14 to Bit D12 are the channel allocation bits, each of which identifies the ADC channel that corresponds to the subsequent result (see the ADC Channel Allocation section). Bit D11 to Bit D0 contain the most recent ADC result. CHID2 0 Channel ID CHID1 0 CHID0 0 0 0 1 0 1 0 0 1 1 1 1 1 1 0 0 1 1 0 1 0 1 Table 11. Command Register1 Bits Channel Name 1 MSB D7 Read out last result from TSENSE2 D6 Read out last result from TSENSE1 D5 ISENSE2 D4 ISENSE1 D3 VIN3 (S.E.) or VIN3 - VIN2 (DIFF) D2 VIN2 (S.E.) or VIN2 - VIN3 (DIFF) D1 VIN1 (S.E.) or VIN1 - VIN0 (DIFF) LSB D0 VIN0 (S.E.) or VIN0 - VIN1 (DIFF) S.E. indicates single-ended, and DIFF indicates differential. Table 12. ADC Result Register (First Read) MSB D15 ALERT_FLAG D14 CHID2 D13 CHID1 D12 CHID0 D11 B11 D10 B10 D9 B9 LSB D8 B8 D4 B4 D3 B3 D2 B2 D1 B1 LSB D0 B0 Table 13. ADC Result Register (Second Read) MSB D7 B7 D6 B6 D5 B5 Rev. 0 | Page 26 of 44 Data Sheet AD7294-2 TSENSE1 AND TSENSE2 RESULT REGISTERS and are set to zero. Conversions take place approximately every 5 ms. The temperature data format in Table 18 applies to the internal temperature sensor data. The TSENSE1 result register (Address 0x02) and TSENSE2 result register (Address 0x03) are 16-bit, read only registers. The MSB, Bit D15, is the ALERT_FLAG bit; Bits[D14:D12] contain the three ADC channel allocation bits. Bit D11 is reserved for flagging diode open circuits. Bits[D10:D0] store the temperature reading from the ADC in an 11-bit, twos complement format (see Table 14 and Table 15). Conversions take place approximately every 5 ms. The temperature reading from the ADC is stored in an 11-bit twos complement format, D10 to D0, to accommodate both positive and negative temperature measurements. The temperature data format is provided in Table 18. Table 14. TSENSEx Result Register (First Read) DACA, DACB, DACC, AND DACD VALUE REGISTERS MSB D15 ALERT_FLAG D14 CHID2 D13 CHID1 D12 CHID0 D11 B11 D10 B10 D9 B9 LSB D8 B8 Table 15. TSENSEx Result Register (Second Read) MSB D7 B7 D6 B6 D5 B5 D4 B4 D3 B3 D2 B2 D1 B1 LSB D0 B0 TSENSEINT RESULT REGISTER Temperature Value Format Writing to Address 0x01 to Address 0x04 sets the DACA, DACB, DACC, and DACD output voltage codes, respectively. Bits[D11:D0] in the write result register are the data bits sent to DACA. Note that Bits[D15:D12] are ignored. Table 16. DAC Register (First Write)1 MSB D15 X 1 The TSENSEINT result register (Address 0x04) is a 16-bit, read only register used to store the ADC data generated from the internal temperature sensor. Similar to the TSENSE1 and TSENSE2 result registers, this register stores the temperature readings from the ADC in an 11-bit, twos complement format (D10 to D0) and uses the MSB as a general alert flag. Bits[D14:D11] are not used D14 X D13 X D12 X D11 B11 D10 B10 LSB D8 B8 D9 B9 X is don't care. Table 17. DAC Register (Second Write) MSB D7 B7 D6 B6 D5 B5 D4 B4 D3 B3 D2 B2 D1 B1 LSB D0 B0 Table 18. TSENSEINT Data Format Input Value (C) D10 (MSB) -256 D9 +128 D8 +64 D7 +32 D6 +16 Rev. 0 | Page 27 of 44 D5 +8 D4 +4 D3 +2 D2 +1 D1 +0.5 D0 (LSB) +0.25 AD7294-2 Data Sheet ALERT STATUS REGISTER A, ALERT STATUS REGISTER B, AND ALERT STATUS REGISTER C Alert Status Register A (Address 0x05), Alert Status Register B (Address 0x06), and Alert Status Register C (Address 0x07) are 8-bit, read/write registers that provide information about an alert event. If a conversion results in activation of the ALERT/BUSY pin or the ALERT_FLAG bit in the ADC result register or the TSENSEx result registers, the alert status registers can be read to gain more information. To clear the full content of any alert status registers, write a code of 0xFF (all 1s) to the relevant register. Alternatively, write to the respective alert bit in the selected alert status register to clear the alert that is associated with that bit. The entire contents of all the alert status registers can be cleared by writing a 1 to Bit D1 and Bit D2 in the configuration register, as shown in Table 24. However, this operation then enables the ALERT/BUSY pin for subsequent conversions. See the Alerts and Limits Theory section for more information. CHANNEL SEQUENCE REGISTER The channel sequence register (Address 0x08) is an 8-bit, read/write register that allows the user to sequence the ADC conversions to be performed in autocycle mode. Table 22 shows the contents of the channel sequence register. See the Modes of Operation section for more information. Table 19. Alert Status Register A Alert Bit Function D7 VIN3 high alert D6 VIN3 low alert D5 VIN2 high alert D4 VIN2 low alert D3 VIN1 high alert D2 VIN1 low alert D1 VIN0 high alert D0 VIN0 low alert D5 ISENSE2 overrange D4 ISENSE1 overrange D3 ISENSE2 high alert D2 ISENSE2 low alert D1 ISENSE1 high alert D0 ISENSE1 low alert D5 TSENSEINT high alert D4 TSENSEINT low alert D3 TSENSE2 high alert D2 TSENSE2 low alert D1 TSENSE1 high alert D0 TSENSE1 low alert D5 ISENSE2 D4 ISENSE1 D3 VIN3 D2 VIN2 D1 VIN1 D0 VIN0 Table 20. Alert Status Register B Alert Bit Function D7 Reserved D6 Reserved Table 21. Alert Status Register C Alert Bit Function D7 Open diode flag D6 Overtemp alert Table 22. Channel Sequence Register Channel Bit Function D7 Reserved D6 Reserved Rev. 0 | Page 28 of 44 Data Sheet AD7294-2 CONFIGURATION REGISTER The configuration register (Address 0x09) is a 16-bit, read/write register that sets the operating mode of the AD7294-2. The bit functions of the configuration register are outlined in Table 23 and Table 24. On power-up, the configuration register is reset to 0x0000. Sample Delay and Bit Trial Delay It is recommended that no I2C bus activity occur when a conversion is taking place; however, this may not be possible, for example, when operating in autocycle mode. Bit D14 and Bit D13 in the configuration register are used to delay critical sample intervals and bit trials from occurring while there is activity on the I2C bus. On power-up, Bit D14 (noise-delayed sampling), Bit D13 (noise-delayed bit trials), and Bit D3 (I2C filters) are enabled (set to 0). This configuration is appropriate for low frequency applications because the bit trials are prevented from occurring when there is activity on the I2C bus, thereby ensuring good dc linearity performance. For high frequency input signals, it may be desirable to have a known sampling point; thus, the noise-delayed sampling can be disabled by writing 1 to Bit D14 in the configuration register. This ensures that the sampling instance is fixed relative to SDA, resulting in improved SNR performance. If noise-delay samplings extend longer than 1 s, the current conversion terminates. This termination can occur if there are edges on SDA that are outside the I2C specification. When noisedelayed sampling is enabled, the rise and fall times must meet the I2C-specified standard. When Bit D13 is enabled, the conversion time may vary. The default configuration for Bit D3 (enabled) is recommended for normal operation because it ensures that the I2C requirements for tOf (minimum)and tSP are met. The I2C filters reject glitches of less than 50 ns. If this function is disabled, the conversion results are more susceptible to noise from the I2C bus. Table 23. Configuration Register Bit Function Descriptions, Bits[D15:D8] Channel Bit Function D15 Reserved D14 Noise-delayed sampling. Use to delay critical sample intervals from occurring when there is activity on the I2C bus. Enabled = 0 Disabled = 1 Setting D13 Noise-delayed bit trials. Use to delay critical bit trials from taking place when there is activity on the I2C bus. D12 Autocycle mode D11 Pseudo differential mode for VIN2/VIN3 D10 Pseudo differential mode for VIN0/VIN1 D9 Differential mode for VIN2/VIN3 D8 Differential mode for VIN0/VIN1 Enabled = 0 Disabled = 1 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Table 24. Configuration Register Bit Function Descriptions, Bits[D7:D0] Channel Bit Function Setting D7 2 x VREF range for VIN3 D6 2 x VREF range for VIN2 D5 2 x VREF range for VIN1 D4 2 x VREF range for VIN0 D3 I2C filters D2 ALERT pin D1 BUSY pin (D2 = 0), clear alerts (D2 = 1) Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 1 Disabled = 0 Enabled = 0 Disabled = 1 Enabled D2 = 1 D1 = 0 Disabled D2 = 0 Enabled D1 = 1 + D0 = 0 Disabled D1 = 0 D0 Select ALERT pin polarity (active high/ active low) Active high = 1 Active low = 0 Table 25. ALERT/BUSY Pin Function Descriptions D2 0 0 1 1 D1 0 1 0 1 ALERT/BUSY Pin Function Descriptions Pin does not provide any interrupt signal. Configures pin as a busy output. Configures pin as an alert output. Resets the ALERT/BUSY output pin, the ALERT_FLAG bit in the conversion result register, and the entire alert status register (if any is active). 11 is written to Bits[D2:D1] in the configuration register to reset the ALERT/BUSY pin, the ALERT_FLAG bit, and the alert status register. Following this write, the content of the configuration register read 10 for Bit D2 and Bit D1, respectively, when read back. Table 26. ADC Input Mode Examples D11 0 0 0 D10 0 0 1 D9 0 0 0 D8 0 1 1 Description All channels single-ended Differential mode on VIN0/VIN1 Pseudo differential mode on VIN0/VIN1 Rev. 0 | Page 29 of 44 AD7294-2 Data Sheet POWER-DOWN REGISTER The power-down register (Address 0x0A) is an 8-bit, read/write register that powers down various sections of the AD7294-2 device. On power-up, the default value for the power-down register is 0x70. The content of the power-down register is listed in Table 27. Table 27. Power-Down Register Bit Descriptions Bit D7 D6 D5 D4 D3 D2 D1 D0 Description Powers down the ADC and DAC reference buffers and the temperature sensor. Sets the DAC outputs to high impedance and disables the ISENSE1 and ISENSE2 alerts. Reserved. Powers down the ADC reference buffer. To allow for an external reference, set to 1 at power-up. Powers down the DAC reference buffer. To allow for an external reference, set to 1 at power-up. Powers down the temperature sensor. Disables the ISENSE1 alerts. Disables the ISENSE2 alerts. Sets the DAC outputs to high impedance. DATALOW AND DATAHIGH REGISTERS VIN0 to VIN3 Channels The DATALOW and DATAHIGH registers (Address 0x0B and Address 0x0C, VIN0; Address 0x0E and Address 0x0F, VIN1; Address 0x11 and Address 0x12, VIN2; Address 0x14 and Address 0x15, VIN3), one pair for each VINx channel, are 16-bit, read/write registers (see Table 29 and Table 30). General alert is flagged by the MSB, Bit D15. Bits[D14:D12] are not used in the register and are set to 0. The remaining 12 bits set the low and high limits for the relevant channel. For single-ended mode, the default values for VIN0 to VIN3 are 0x000 (DATALOW) and 0xFFF (DATAHIGH) in binary format. For differential mode, the default values for VIN0 to VIN3 are 0x800 (DATALOW) and 0x7FF (DATAHIGH) in twos complement format. Note that, if the part is configured in either single-ended or differential mode and the mode is changed, the limits in the DATALOW and DATAHIGH registers must be reprogrammed. TSENSE1, TSENSE2, and TSENSEINT Channels Channel 7 to Channel 9 (Address 0x1D and Address 0x1E, TSENSE1; Address 0x20 and Address 0x21, TSENSE2; and Address 0x23 and Address 0x24, TSENSEINT) default to 0x400 (DATALOW) and 0x3FF (DATAHIGH) as the limits because they are in 11-bit, twos complement format. Table 28. Default Values for DATALOW and DATAHIGH Registers ADC Channel VIN0 VIN1 VIN2 VIN3 ISENSE1 ISENSE2 TSENSE1 TSENSE2 TSENSEINT Single-Ended DATALOW DATAHIGH 0x000 0xFFF 0x000 0xFFF 0x000 0xFFF 0x000 0xFFF N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A Differential DATALOW DATAHIGH 0x800 0x7FF 0x800 0x7FF 0x800 0x7FF 0x800 0x7FF 0x800 0x7FF 0x800 0x7FF 0x400 0x3FF 0x400 0x3FF 0x400 0x3FF Table 29. DATALOW, DATAHIGH Register (First Read/Write) MSB D15 ALERT_FLAG D14 0 D13 0 D12 0 D11 B11 D10 B10 D9 B9 LSB D8 B8 Table 30. DATALOW, DATAHIGH Register (Second Read/Write) MSB D7 B7 D6 B6 D5 B5 D4 B4 D3 B3 D2 B2 D1 B1 LSB D0 B0 HYSTERESIS REGISTERS Each hysteresis register (Address 0x0D, VIN0; Address 0x10, VIN1; Address 0x13, VIN2; Address 0x16, VIN3; Address 0x19, ISENSE1; Address 0x1C, ISENSE2; Address 0x1F, TSENSE1; Address 0x22, TSENSE2; Address 0x25, TSENSEINT) is a 16-bit, read/write register in which only the 12 LSBs of the register are used. The MSB signals the alert event. If 0xFFF is written to the hysteresis register, the hysteresis register enters the minimum/maximum mode (see the Alerts and Limits Theory section). Table 31. Hysteresis Register (First Read/Write) MSB D15 ALERT_FLAG D14 0 D13 0 D12 0 D11 B11 D10 B10 D9 B9 LSB D8 B8 Table 32. Hysteresis Register (Second Read/Write) MSB D7 B7 Rev. 0 | Page 30 of 44 D6 B6 D5 B5 D4 B4 D3 B3 D2 B2 D1 B1 LSB D0 B0 Data Sheet AD7294-2 REMOTE CHANNEL TSENSE1 AND TSENSE2 OFFSET REGISTERS The TSENSE1 offset register (Address 0x26) and TSENSE2 offset register (Address 0x27) allow the user to add or subtract an offset to the temperature. These 8-bit, read/write registers store data in a twos complement format. The data is subtracted from the temperature readings taken by the TSENSE1 and TSENSE2 temperature sensors. The offset is implemented before the values are stored in the TSENSE1 and TSENSE2 result registers. The offset registers can be used to compensate for transistors with different ideality factors because the TSENSEx results are based on the 2N3906 transistor ideality factor. Different transistors with different ideality factors result in different offsets within the region of interest, which can be compensated for by using this register. Table 33. TSENSE1, TSENSE2 Offset Register Data Format Input Value (C) Rev. 0 | Page 31 of 44 MSB D7 -32 D6 +16 D5 +8 D4 +4 D3 +2 D2 +1 D1 +0.5 LSB D0 +0.25 AD7294-2 Data Sheet I2C INTERFACE GENERAL I2C TIMING SERIAL BUS ADDRESS BYTE Figure 47 shows the timing for general read and write operations using an I2C-compliant interface. The I2C bus uses open-drain drivers; therefore, when no device is driving the bus, both SCL and SDA are high. This is known as idle state. When the bus is idle, the master initiates a data transfer by establishing a start condition, defined as a high-to-low transition on the serial data line (SDA) while the serial clock line (SCL) remains high. This indicates that a data stream follows. The master device is responsible for generating the clock. The first byte the user writes to the device is the slave address byte. Similar to all I2C-compatible devices, the AD7294-2 has a 7-bit serial address. The five LSBs are user-programmable via the three, three-state input pins, as shown in Table 34. Data is sent over the serial bus in groups of nine bits: eight bits of data from the transmitter followed by an acknowledge bit (ACK) from the receiver. Data transitions on the SDA line must occur during the low period of the clock signal and remain stable during the high period. The receiver should pull the SDA line low during the acknowledge bit to signal that the preceding byte has been received correctly. If this is not the case, cancel the transaction. Table 34. Slave Address Control Using Three-State Input Pins1 Table 34 shows that the ASx pins are sometimes left floating. Note that, in such cases, the stray capacitance on the pin must be less than 30 pF to allow correct detection of the floating state; therefore, any PCB trace must be kept as short as possible. AS2 L L L L L L L L L H H H H H H H H H NC NC NC NC NC NC NC NC NC The first byte that the master sends must consist of a 7-bit slave address, followed by a data direction bit. Each device on the bus has a unique slave address; therefore, the first byte sets up communication with a single slave device for the duration of the transaction. The transaction can be used either to write to a slave device (data direction bit = 0) or to read data from it (data direction bit = 1). In the case of a read transaction, it is often necessary first to write to the slave device (in a separate write transaction) to tell it from which register to read. Reading and writing cannot be combined in one transaction. When the transaction is complete, the master can keep control of the bus, initiating a new transaction by generating another start bit (high-to-low transition on SDA while SCL is high). This is known as a repeated start (Sr). Alternatively, the bus can be relinquished by releasing the SCL line followed by releasing the SDA line. This low-to-high transition on SDA while SCL is high is known as a stop bit (P), and it leaves the I2C bus in its idle state (that is, no current is consumed by the bus). The example in Figure 47 shows a simple write transaction with an AD7294-2 as the slave device. In this example, the AD7294-2 register pointer is being readied for a future read transaction. 1 AS1 L L L H H H NC NC NC L L L H H H NC NC NC L L L H H H NC NC NC AS0 L H NC L H NC L H NC L H NC L H NC L H NC L H NC L H NC L H NC Slave Address (A6 to A0) 0x61 0x62 0x63 0x64 0x65 0x66 0x67 0x68 0x69 0x6A 0x6B 0x6C 0x6D 0x6E 0x6F 0x70 0x71 0x72 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7A 0x7B H = tie the pin to VDRIVE, L = tie the pin to DGND, NC = pin left floating. SCL START BY MASTER A6 A5 A4 A3 A2 A1 A0 SLAVE ADDRESS BYTE R/W P7 P6 P5 ACK. BY AD7294-2 USER PROGRAMMABLE 5 LSBs Figure 47. General I2C Timing Rev. 0 | Page 32 of 44 P4 P3 P2 REGISTER ADDRESS P1 P0 ACK. BY AD7294-2 STOP BY MASTER 10936-040 SDA Data Sheet AD7294-2 INTERFACE PROTOCOL To write data to the register, use the following command sequence: The AD7294-2 uses the following I2C protocols. 1. 2. Writing a Single Byte of Data to an 8-Bit Register The alert status registers (Address 0x05 to Address 0x07), the power-down register (Address 0x0A), the channel sequence register (Address 0x08), the temperature offset registers (Address 0x26 and Address 0x27), and the command register (Address 0x00) are 8-bit registers. Therefore, only one byte of data can be written to each. In this operation, the master device sends a byte of data to the slave device (see Figure 48). 1 The master device asserts a start condition on SDA. The master sends the 7-bit slave address followed by a zero for the direction bit, indicating a write operation. The addressed slave device asserts an acknowledge on SDA. The master sends a register address. The slave asserts an acknowledge on SDA. The master sends a data byte. The slave asserts an acknowledge on SDA. The master asserts a stop condition to end the transaction. 3. 4. 5. 6. 7. 8. 9 1 9 SCL A6 A5 A4 A3 A2 A1 A0 P7 R/W START BY MASTER P6 P5 P4 P3 P2 P1 P0 ACK. BY AD7294-2 FRAME 1 SLAVE ADDRESS BYTE ACK. BY AD7294-2 FRAME 2 ADDRESS POINTER REGISTER BYTE 1 9 SCL (CONTINUED) SDA (CONTINUED) D7 D6 D5 D4 D3 D2 D1 D0 ACK. BY STOP BY AD7294-2 MASTER FRAME 3 DATA BYTE S SLAVE ADDRESS 0 A REG POINTER FROM MASTER TO SLAVE FROM SLAVE TO MASTER A DATA S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE Figure 48. Single Byte Write Sequence Rev. 0 | Page 33 of 44 A P 10936-061 SDA AD7294-2 Data Sheet Writing Two Bytes of Data to a 16-Bit Register Writing to Multiple Registers The limit and hysteresis registers (Address 0x0B to Address 0x25), the result registers (Address 0x01 to Address 0x04), and the configuration register (Address 0x09) are 16-bit registers. Therefore, two bytes of data are required to write a value to any one of these registers (see Figure 49). Use the following sequence to write the two bytes of data to one of these registers: Use the following sequence to write to multiple address registers: 1. The master device asserts a start condition on SDA. 2. The master sends the 7-bit slave address followed by the write bit (low). 3. The addressed slave device (the AD7294-2) asserts an acknowledge on SDA. 4. The master sends a register address; for example, the Alert Status Register A register address. 5. The slave asserts an acknowledge on SDA. 6. The master sends the data byte. 7. The slave asserts an acknowledge on SDA. 8. The master sends a second register address; for example, the configuration register. 9. The slave asserts an acknowledge on SDA. 10. The master sends the first data byte to the second register address. 11. The slave asserts an acknowledge on SDA. 12. The master sends the second data byte. 13. The slave asserts an acknowledge on SDA. 14. The master asserts a stop condition on SDA to end the transaction. The previous examples describe writing to two registers only (Alert Status Register A and the configuration register). However, the AD7294-2 can read from multiple registers following one write operation, as shown in Figure 50. 1. 2. The master device asserts a start condition on SDA. The master sends the 7-bit slave address, followed by the write bit (low). 3. The addressed slave device asserts an acknowledge on SDA. 4. The master sends a register address. 5. The slave asserts an acknowledge on SDA. 6. The master sends the first data byte (most significant). 7. The slave asserts an acknowledge on SDA. 8. The master sends the second data byte (least significant). 9. The slave asserts an acknowledge on SDA. 10. The master asserts a stop condition on SDA to end the transaction. SLAVE ADDRESS 0 FROM MASTER TO SLAVE FROM SLAVE TO MASTER A REG POINTER A DATA[15:8] A DATA[7:0] S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE A P 10936-059 S Figure 49. Writing Two Bytes of Data to a 16-Bit Register ... SLAVE ADDRESS DATA[15:8] A FROM MASTER TO SLAVE FROM SLAVE TO MASTER 0 A POINT TO PD REG (ADDR 0X0A) DATA[7:0] A A DATA[7:0] A POINT TO CONFIG REG (ADDR 0x09) A ... P S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE 10936-054 S Figure 50. Writing to Multiple Registers Rev. 0 | Page 34 of 44 Data Sheet AD7294-2 Reading Data from an 8-Bit Register Reading Two Bytes of Data from a 16-Bit Register Reading the contents from any of the 8-bit registers is a singlebyte read operation, as shown in Figure 51. In this protocol, the first part of the transaction writes to the register pointer. When the register address has been set up, any number of reads can be performed from that particular register address without having to write to the address pointer register again. When the required number of reads is completed, the master should not acknowledge the final byte. This tells the slave to stop transmitting, allowing a stop condition to be asserted by the master. Additional reads from this register can be performed in a future transaction without the need to rewrite to the register pointer. In this example, the master device reads three lots of two-byte data from a slave device (see Figure 52). However, any number of lots consisting of two bytes can be read. This protocol assumes that the particular register address has been set up by a singlebyte write operation to the address pointer register (see the previous read example). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. If a read from a different address is required, the relevant register address must be written to the address pointer register. Again, any number of reads from this register can then be performed. In the next example, the master device receives two bytes from a slave device, as follows: 4. 5. 6. 7. 8. S ... 0 SLAVE ADDRESS A DATA[7:0] FROM MASTER TO SLAVE FROM SLAVE TO MASTER A REG POINTER 16. A SR SLAVE ADDRESS 1 A DATA[7:0] ... A P S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE Figure 51. Reading Two Single Bytes of Data from a Selected Register SLAVE ADDRESS S ... DATA[15:8] 1 A FROM MASTER TO SLAVE FROM SLAVE TO MASTER A DATA[7:0] A DATA[15:8] A DATA[7:0] A DATA[15:8] A DATA[7:0] A ... P S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE 10936-060 3. The master device asserts a start condition on SDA. The master sends the 7-bit slave address followed by the read bit (high). The addressed slave device asserts an acknowledge on SDA. The master receives a data byte. The master asserts an acknowledge on SDA. The master receives another 8-bit data byte. The master asserts a no acknowledge on SDA to inform the slave that the data transfer is complete. The master asserts a stop condition on SDA to end the transaction. 10936-055 1. 2. The master device asserts a start condition on SDA. The master sends the 7-bit slave address followed by the read bit (high). The addressed slave device asserts an acknowledge on SDA. The master receives the data byte. The master asserts an acknowledge on SDA. The master receives a second data byte. The master asserts an acknowledge on SDA. The master receives a data byte. The master asserts an acknowledge on SDA. The master receives a second data byte. The master asserts an acknowledge on SDA. The master receives a data byte. The master asserts an acknowledge on SDA. The master receives a second data byte. The master asserts a no acknowledge on SDA to notify the slave that the data transfer is complete. The master asserts a stop condition on SDA to end the transaction. Figure 52. Reading Three Lots of Two Bytes of Data from the Conversion Result Register Rev. 0 | Page 35 of 44 AD7294-2 Data Sheet MODES OF OPERATION The conversion sequence is as follows: There are two different methods of initiating a conversion on the AD7294-2: command mode and autocycle mode. 1. 2. COMMAND MODE In command mode, the AD7294-2 ADC converts on demand on either a single channel or a sequence of channels. To enter this mode, the required combination of channels is written into the command register (Address 0x00). The first conversion takes place at the end of this write operation, in time for the result to be read out in the next read operation. While this result is being read out, the next conversion in the sequence takes place, and so on. The master device asserts a start condition on SDA. The master sends the 7-bit slave address followed by the write bit (low). The addressed slave device (AD7294-2) asserts an acknowledge on SDA. The master sends the command register address (0x00). The slave asserts an acknowledge on SDA. The master sends Data Byte 0x13, which selects the VIN0, VIN1, and ISENSE1 channels. The slave asserts an acknowledge on SDA. The master sends the ADC result register address (0x01). The slave asserts an acknowledge on SDA. The master sends the 7-bit slave address followed by the write bit (high). The slave (AD7294-2) asserts an acknowledge on SDA. The master receives a data byte, which contains the ALERT_FLAG bit, the channel ID bits, and the four MSBs of the conversion result for the VIN0 channel. The master then asserts an acknowledge on SDA. The master receives the second data byte, which contains the eight LSBs of the converted result for the VIN0 channel. The master then asserts an acknowledge on SDA. Step 10 and Step 11 are repeated for the VIN1 channel and the ISENSE1 channel. When the master receives the results from all the selected channels, the slave again converts and outputs the result for the first channel in the selected sequence. Step 10 to Step 12 are repeated. The master asserts a no acknowledge on SDA and a stop condition on SDA to end the conversion and exit command mode. 3. 4. 5. 6. 7. To exit the command mode, the master does not acknowledge the final byte of data. This stops the AD7294-2 from transmitting, allowing the master to assert a stop condition on the bus. When switching to read mode, therefore, it is important that, after writing to the command register, a repeated start (Sr) signal be used rather than a stop (P) followed by a start (S). Otherwise, the device exits command mode after the first conversion. 8. 9. 10. After writing to the command register, the register pointer is returned to its previous value. If a new pointer value is required (typically for the ADC result register, Address 0x01), it can be written immediately following the command byte. This extra write operation does not affect the conversion sequence because the second conversion is triggered only at the start of the first read operation. 11. 12. 13. The maximum throughput that can be achieved using this mode with a 400 kHz I2C clock is (400 kHz/18) = 22.2 kSPS. Figure 53 shows the command mode converting on a sequence of channels including VIN0, VIN1, and ISENSE1. 14. If no read occurs in a 5 ms period, the AD7294-2 automatically exits command mode. To change the conversion sequence, write a new sequence to the command register. * 0 SLAVE ADDRESS A POINT TO COMMAND REG (ADDR 0x00) COMMAND = 0x13 A * ... POINT TO ADC RESULT REG (ADDR 0x01) ... VIN0[7:0] A SR 1 SLAVE ADDRESS A ALERT? CH ID (000) * ... * ALERT? ... A VIN0[7:0] A ALERT? VIN1[11:8] CH ID (001) VIN1[7:0] A A VIN0[11:8] ... * CH ID (100) A ISENSE 1[11:8] ........ A ISENSE 1[7:0] ISENSE 1[7:0] A A P ALERT? CH ID (000) FROM MASTER TO SLAVE FROM SLAVE TO MASTER * = POSITION OF A CONVERSION START Figure 53. Command Mode Operation Rev. 0 | Page 36 of 44 ... A VIN0[11:8] A ... S = START CONDITION SR = REPEATED START P = STOP CONDITION A = ACKNOWLEDGE A = NO ACKNOWLEDGE 10936-056 S Data Sheet AD7294-2 AUTOCYCLE MODE The AD7294-2 can be configured to convert continuously on a programmable sequence of channels, making it the ideal mode of operation for system monitoring. These conversions occur in the background approximately every 50 s and are transparent to the master. Typically, this mode is used to automatically monitor a selection of channels with either the limit registers programmed to signal an out-of-range condition via the alert function or the minimum/maximum recorders tracking the variation over time of a particular channel. Reads and writes can be performed at any time (the ADC result register, Address 0x01, contains the most recent conversion result). On power-up, the autocycle mode is disabled. To enable it, write to Bit D12 in the configuration register (Address 0x09) and select the desired channels for conversion in the channel sequence register (Address 0x08). If a command mode conversion is required while the autocycle mode is active, it is necessary to disable the autocycle mode before proceeding to the command mode. This is achieved either by clearing Bit D12 of the configuration register or by writing 0x00 to the channel sequence register. When the command mode conversion is complete, the user must exit command mode by issuing a stop condition before reenabling autocycle mode. When switching from autocycle mode to command mode, the temperature sensor must be given sufficient time to settle and complete a new temperature integration cycle. Therefore, temperature sensor conversions performed within the first 500 ms after switching from autocycle mode to command mode may trigger false temperature high and low alarms. It is recommended that the temperature sensor alarms be disabled for the first 500 ms after mode switching by writing 0x400 to the DATALOW register TSENSEx and 0x3FF to the DATAHIGH register TSENSEx. Reconfigure the temperature sensor alerts to the desired alarm level when the 500 ms period has elapsed. Alternatively, ignore any temperature alerts triggered during the first 500 ms after mode switching. Rev. 0 | Page 37 of 44 AD7294-2 Data Sheet ALERTS AND LIMITS THEORY ALERT_FLAG BIT The ALERT_FLAG bit indicates whether the conversion result being read or any other channel result has violated the limit registers associated with it. If an alert occurs and the ALERT_ FLAG bit is set, the master can read the alert status register to obtain more information on where the alert occurred. ALERT STATUS REGISTERS The alert status registers are 8-bit, read/write registers that provide information on an alert event. If a conversion results in activation of the ALERT/BUSY pin or the ALERT_FLAG bit in the ADC result register or TSENSE result registers, the alert status register can be read to get more information (see Figure 54 for the alert register structure). D7 D6 D5 D4 D3 D2 D1 D0 RESERVED RESERVED ISENSE 2 OVERRANGE* ISENSE 1 OVERRANGE* ISENSE 2 HIGH ALERT ISENSE 2 LOW ALERT ISENSE 1 HIGH ALERT ISENSE 1 LOW ALERT D7 D6 D5 D4 D3 D2 D1 D0 OPEN DIODE FLAG* OVERTEMP ALERT* TSENSE INT HIGH ALERT TSENSE INT LOW ALERT TSENSE 2 HIGH ALERT TSENSE 2 LOW ALERT TSENSE 1 HIGH ALERT TSENSE 1 LOW ALERT D7 D6 D5 D4 D3 D2 D1 D0 ALERT STATUS REGISTER A ALERT STATUS REGISTER B OR ALERT FLAG ALERT/BUSY CONFIGURATION REGISTER D2 = 1, D1 = 0 To clear the full contents of any alert register, write a code of 0xFF (all 1s) to the relevant registers. Alternatively, the user can write to the respective alert bit in the selected alert register to clear the alert that is associated with that bit. The entire contents of all the alert status registers can be cleared by writing a 1 to Bit D2 and Bit D1 in the configuration register, as shown in Table 24. However, this operation then enables the ALERT/BUSY pin for subsequent conversions. If the result moves outside the lower or upper limit set by the user, the AD7294-2 signals an alert using hardware (via the ALERT/BUSY pin), software (via the ALERT_FLAG bit), or both, depending on the configuration. ALERT STATUS REGISTER C * THESE BITS ARE ALWAYS ACTIVE. ALL OTHER BITS CAN BE PROGRAMMED TO BE ACTIVE OR NOT, AS REQUIRED. The AD7294-2 internal circuitry can generate an alert if either the D1() or the D2() input pins for the external temperature sensor are open circuit. The most significant bit (MSB) of Alert Status Register C (see Table 21) alerts the user when an open diode flag occurs on the external temperature sensors. If the internal temperature sensor detects an AD7294-2 die temperature of greater than 150C, the overtemperature alert bit (Bit D6 in Alert Status Register C) is set, and the DAC outputs are set to a high impedance state. The remaining six bits in Address 0x06 store alert event data for TSENSEINT, TSENSE2, and TSENSE1 with two status bits per channel, one corresponding to each of the DATAHIGH and DATALOW limits. DATALOW AND DATAHIGH MONITORING FEATURES 10936-057 VIN3 HIGH ALERT VIN3 LOW ALERT VIN2 HIGH ALERT VIN2 LOW ALERT VIN1 HIGH ALERT VIN1 LOW ALERT VIN0 HIGH ALERT VIN0 LOW ALERT Bit D4 and Bit D5 represent the ISENSE1 OVERRANGE and ISENSE2 OVERRANGE values of VREF/10.41. During power-up, it is possible for the fault outputs to be triggered, depending on which supply comes up first. It is recommended that these bits be cleared on power-up as part of the initialization routine by writing a 1 to both D4 and D5. Figure 54. Alert Status Register Structure Alert Status Register A (see Table 19) consists of four channels with two status bits per channel, one bit corresponding to each of the DATALOW and DATAHIGH limits. This register stores the alert event data for VIN3 to VIN0, which are the standard voltage inputs. When the contents of this register are read, any bit with a status of 1 indicates a violation of its associated limit; that is, it identifies the channel and whether the violation occurred on the upper or lower limit. If a second alert event occurs on another channel before the contents of the alert register are read, the bit corresponding to the second alert event is also set. Alert Status Register B (see Table 20) consists of three channels, also with two status bits per channel, representing the specified DATALOW and DATAHIGH limits. Bits[D3:D0] correspond to the low and high limit alerts for the current sense inputs. The DATALOW register stores the lower limit that activates the ALERT/BUSY output pin and/or the ALERT_FLAG bit in the conversion result register. If the conversion result is less than the value in the DATALOW register, an alert occurs. The DATAHIGH register stores the upper limit that activates the ALERT/BUSY output pin and/or the ALERT_FLAG bit in the conversion result register. If the conversion result is greater than the value in the DATAHIGH register, an alert occurs. An alert associated with either the DATALOW or DATAHIGH register is cleared automatically when the monitored signal is back in range; that is, the conversion result is between the limits. The contents of the alert register are updated after each conversion. A conversion is performed every 50 s in autocycle mode; as a result, the contents of the alert register may change every 50 s. If the ALERT pin signals an alert event and the content of the alert register is not read before the next conversion is complete, the contents of the register may be changed if the signal that is being monitored returns between the specified limits. In such circumstances, the ALERT pin no longer signals the occurrence of an alert event. Rev. 0 | Page 38 of 44 Data Sheet AD7294-2 The hysteresis register can be used to avoid flicker on the ALERT/ BUSY pin. If the hysteresis function is enabled, the conversion result must return to a value of at least N LSBs above the DATALOW or N LSBs below the DATAHIGH register value for the ALERT/BUSY output pin and ALERT_FLAG bit to be reset. The value of N is taken from the 12-bit hysteresis register associated with that channel. By setting the hysteresis register to a code that is close to the maximum output code for the ADC (for example, 0x77D), the DATALOW or DATAHIGH alerts are not cleared automatically by the AD7294-2. Bit D11 of DATALOW or DATAHIGH Register TSENSEx is the diode open circuit flag. If this bit is set to 0, it indicates the presence of an open circuit between the Dx(+) and Dx(-) pins. An alert that is triggered on either ISENSE OVERRANGE pin remains until it is cleared by a write to the alert status register. The contents of the DATALOW and DATAHIGH registers are reset to their default values on power-up (see Table 28). HYSTERESIS The hysteresis value determines the reset point for the ALERT/ BUSY pin and/or ALERT_FLAG bit if a violation of the limits occurs. The hysteresis register stores the hysteresis value, N, when using the limit registers. Each pair of limit registers has a dedicated hysteresis register. For example, if a hysteresis value of 8 LSBs is required on the upper and lower limits of VIN0, the 16-bit word, 0000 0000 0000 1000, should be written to Hysteresis Register VIN0 (Register 0x0D, see Table 9). On power-up, the hysteresis registers contain a value of 8 LSBs for nontemperature result registers and 8C, or 32 LSBs, for the TSENSE registers. If a different hysteresis value is required, that value must be written to the hysteresis register for the channel in question. The advantage of having hysteresis registers associated with each of the limit registers is that hysteresis prevents chatter on the alert bits associated with each ADC channel. Figure 55 shows the limit checking operation. Using the Limit Registers to Store Minimum/Maximum Conversion Results If 0xFFF is written to the hysteresis register for a particular channel, the DATALOW and DATAHIGH registers for that channel no longer act as limit registers, as previously described, but, instead, act as storage registers for the maximum and minimum conversion results. This function is useful when an alert signal is not required in an application, but it is still required to monitor the minimum and maximum conversion values over time. Note that on powerup, the contents of the DATAHIGH register for each channel are set to the maximum code, whereas the contents of the DATALOW registers are set to the minimum code by default. HIGH LIMIT HIGH LIMIT - HYSTERESIS INPUT SIGNAL LOW LIMIT + HYSTERESIS LOW LIMIT TIME Figure 55. Limit Checking Rev. 0 | Page 39 of 44 10936-067 ALERT SIGNAL AD7294-2 Data Sheet APPLICATIONS INFORMATION The circuit in Figure 56 is a typical system connection diagram for the AD7294-2. The device monitors and controls the overall performance of two final stage amplifiers. The gain control and phase adjustment of the driver stage are incorporated in the application and are carried out by the two available uncommitted outputs of the AD7294-2. Both high-side current senses measure the amount of current on the respective final stage amplifiers. The comparator outputs, the ISENSE1 OVERRANGE and ISENSE2 OVERRANGE pins, are the controlling signals for the switches on the RF inputs of the LDMOS power FETs. If the high-side current sense reads a value above a specified limit compared with the setpoint, the RF IN signal is switched off by the comparator. The AD7294-2 contains all the functions that are required for general-purpose monitoring and control of current, voltage, and temperature. With its 59.4 V maximum common-mode range, the device is useful in industrial and automotive applications where current sensing in the presence of a high common-mode voltage is required. For example, the part is ideally suited for monitoring and controlling a power amplifier in a cellular base station. BASE STATION POWER AMPLIFIER MONITOR AND CONTROL The AD7294-2 is used in a power amplifier signal chain to achieve the optimal bias condition for the LDMOS transistor. The main factors influencing the bias conditions are temperature, supply voltage, gate voltage drift, and general processing parameters. The overall performance of a power amplifier configuration is determined by the inherent trade-offs required in efficiency, gain, and linearity. The high level of integration offered by the AD7294-2 allows the use of a single chip to dynamically control the drain bias current to maintain a constant value over temperature and time, thus significantly improving the overall performance of the power amplifier. The AD7294-2 incorporates the functionality of eight discrete components, bringing considerable board area savings over alternative solutions. By measuring the transmitted power (Tx) and the received power (Rx), the device can dynamically change the drivers and PA signal to optimize performance. This application requires a logarithmic detector/controller, such as the Analog Devices AD8317 or AD8362. RSENSE VDD RSENSE RF CHOKE RF CHOKE RS1(+) ISENSE 2 OVERRANGE ISENSE 1 OVERRANGE REF Rx POWER Rx POWER MONITOR REF RS2(-) HIGH SIDE CURRENT SENSE AD7294-2* 12-BIT DAC SETPOINT 240mV RF OUT VOUTA FILTER RF OUT VIN0 VIN1 LDMOS RF IN MUX 12-BIT ADC VIN2 VIN3 D1+ D2+ T1 RF IN T2 TEMP SENSOR D2- 12-BIT DAC VOUTB 12-BIT DAC VOUTC GAIN CONTROL VOUTD GAIN CONTROL 12-BIT DAC FILTER LDMOS D1- *ADDITIONAL PINS OMITTED FOR CLARITY. Figure 56. Typical HPA Monitor and Control Application Rev. 0 | Page 40 of 44 10936-036 Tx POWER MONITOR Tx POWER HIGH SIDE CURRENT SENSE RS2(+) COMPARATORS AND REGISTERS RF CUTOFF RS1(-) Data Sheet AD7294-2 The VOUT pin of the AD8362 is applied to the gain control terminal of the power amplifier. For this output power control loop to be stable, a ground referenced capacitor must be connected to the CLPF pin of the AD8362. This capacitor integrates the error signal (which is actually a current) that is present when the loop is not balanced. In a system where a variable gain amplifier (VGA) or variable voltage attenuator (VVA) feeds the power amplifier, only one AD8362 is required. In such a case, the gain on one of the parts (VVA, PA) is fixed, and VOUTx feeds the control input of the other. GAIN CONTROL OF POWER AMPLIFIER In gain control mode, a setpoint voltage that is proportional in decibels (dB) to the desired output power is applied to a power detector such as the AD8362. A sample of the output power from the power amplifier (PA), through a directional coupler and attenuator (or by other means), is fed to the input of the AD8362. The VOUT pin is connected to the gain control terminal of the PA (see Figure 57). Based on the defined relationship between VOUT and the RF input signal, the AD8362 adjusts the voltage on VOUT (VOUT is now an error amplifier output) until the level at the RF input corresponds to the applied VSET. The AD7294-2 completes a feedback loop that tracks the output of the AD8362 and adjusts the VSET input of the AD8362 accordingly. ENVELOPE OF TRANSMITTED SIGNAL POWER AMPLIFIER RF IN DIRECTIONAL COUPLER ATTENUATOR C5 1nF T2 AD7294-2 C7 0.1nF AD8362 INHI VOUT INLO VSET VIN 1:4 CLPF Figure 57. Setpoint Controller Operation Rev. 0 | Page 41 of 44 VOUT 10936-037 C6 0.1nF AD7294-2 Data Sheet OUTLINE DIMENSIONS 9.20 9.00 SQ 8.80 1.20 MAX 0.75 0.60 0.45 64 49 48 1 PIN 1 7.20 7.00 SQ 6.80 TOP VIEW 0.15 0.05 0.20 0.09 7 3.5 0 0.08 MAX COPLANARITY SEATING PLANE VIEW A (PINS DOWN) 33 32 16 17 VIEW A 0.40 BSC LEAD PITCH ROTATED 90 CCW 0.23 0.18 0.13 COMPLIANT TO JEDEC STANDARDS MS-026-ABD 012108-A 0 MIN 1.05 1.00 0.95 Figure 58. 64-Lead Thin Plastic Quad Flat Package [TQFP] (SU-64-1) Dimensions shown in millimeters ORDERING GUIDE Model 1 AD7294-2BSUZ AD7294-2BSUZ-RL 1 Temperature Range -40C to +105C -40C to +105C Package Description 64-Lead Thin Plastic Quad Flat Package [TQFP] 64-Lead Thin Plastic Quad Flat Package [TQFP] Z = RoHS Compliant Part. Rev. 0 | Page 42 of 44 Package Option SU-64-1 SU-64-1 Data Sheet AD7294-2 NOTES Rev. 0 | Page 43 of 44 AD7294-2 Data Sheet NOTES I2C refers to a communications protocol originally developed by Philips Semiconductors (now NXP Semiconductors). (c)2013 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D10936-0-6/13(0) Rev. 0 | Page 44 of 44