19-3291; Rev 0; 5/04 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference Features ♦ 1.8Msps Sampling Rate ♦ Only 50mW (typ) Power Dissipation ♦ Only 1µA (max) Shutdown Current ♦ High-Speed, SPI-Compatible, 3-Wire Serial Interface ♦ 61dB S/(N + D) at 525kHz Input Frequency ♦ Internal True-Differential Track/Hold (T/H) ♦ Internal 4.096V Reference ♦ No Pipeline Delays ♦ Small 12-Pin TQFN Package Ordering Information PART TEMP RANGE PINPACKAGE INPUT The MAX1076/MAX1078 come in a 12-pin TQFN package, and are available in the commercial (0°C to +70°C) and extended (-40°C to +85°C) temperature ranges. MAX1076CTC-T 0°C to +70°C 12 TQFN-12 Unipolar MAX1076ETC-T -40°C to +85°C 12 TQFN-12 Unipolar MAX1078CTC-T 0°C to +70°C 12 TQFN-12 Bipolar Applications MAX1078ETC-T -40°C to +85°C 12 TQFN-12 Bipolar Data Acquisition Communications Bill Validation Portable Instruments Motor Control Typical Operating Circuit Pin Configuration TOP VIEW AIN+ N.C. SCLK 12 11 10 0.01µF 10µF AIN- 1 REF 2 RGND MAX1076 MAX1078 9 CNVST 8 DOUT 7 3 +1.8V TO VDD +4.75V TO +5.25V 0.01µF VDD DIFFERENTIAL + INPUT VOLTAGE - AIN+ 10µF VL DOUT AIN- MAX1076 CNVST MAX1078 VL µC/DSP SCLK 4 VDD 5 6 N.C. GND REF 4.7µF 0.01µF RGND GND TQFN SPI/QSPI are trademarks of Motorola, Inc. MICROWIRE is a trademark of National Semiconductor Corp. ________________________________________________________________ Maxim Integrated Products For pricing, delivery, and ordering information, please contact Maxim/Dallas Direct! at 1-888-629-4642, or visit Maxim’s website at www.maxim-ic.com. 1 MAX1076/MAX1078 General Description The MAX1076/MAX1078 are low-power, high-speed, serial-output, 10-bit, analog-to-digital converters (ADCs) that operate at up to 1.8Msps and have an internal reference. These devices feature true-differential inputs, offering better noise immunity, distortion improvements, and a wider dynamic range over single-ended inputs. A standard SPI™/QSPI™/MICROWIRE™ interface provides the clock necessary for conversion. These devices easily interface with standard digital signal processor (DSP) synchronous serial interfaces. The MAX1076/MAX1078 operate from a single +4.75V to +5.25V supply voltage. The MAX1076/MAX1078 include a 4.096V internal reference. The MAX1076 has a unipolar analog input, while the MAX1078 has a bipolar analog input. These devices feature a partial power-down mode and a full power-down mode for use between conversions, which lower the supply current to 2mA (typ) and 1µA (max), respectively. Also featured is a separate power-supply input (VL), which allows direct interfacing to +1.8V to VDD digital logic. The fast conversion speed, low-power dissipation, excellent AC performance, and DC accuracy (±0.5 LSB INL) make the MAX1076/MAX1078 ideal for industrial process control, motor control, and base-station applications. MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference ABSOLUTE MAXIMUM RATINGS VDD to GND ..............................................................-0.3V to +6V VL to GND ................-0.3V to the lower of (VDD + 0.3V) and +6V Digital Inputs to GND .................-0.3V to the lower of (VDD + 0.3V) and +6V Digital Output to GND ....................-0.3V to the lower of (VL + 0.3V) and +6V Analog Inputs and REF to GND..........-0.3V to the lower of (VDD + 0.3V) and +6V RGND to GND .......................................................-0.3V to +0.3V Maximum Current into Any Pin............................................50mA Continuous Power Dissipation (TA = +70°C) 12-Pin TQFN (derate 16.9mW/°C above +70°C) ......1349mW Operating Temperature Ranges MAX107_ CTC ...................................................0°C to +70°C MAX107_ ETC.................................................-40°C to +85°C Junction Temperature ......................................................+150°C Storage Temperature Range .............................-60°C to +150°C Lead Temperature (soldering, 10s) .................................+300°C Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. ELECTRICAL CHARACTERISTICS (VDD = +5V ±5%, VL = VDD, fSCLK = 28.8MHz, 50% duty cycle, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS DC ACCURACY Resolution 10 Bits Relative Accuracy INL (Note 1) ±0.5 LSB Differential Nonlinearity DNL (Note 2) ±0.5 LSB ±2 LSB Offset Error Offset-Error Temperature Coefficient ±1 Gain Error Offset nulled ppm/°C ±2 Gain Temperature Coefficient ±2 LSB ppm/°C DYNAMIC SPECIFICATIONS (fIN = 525kHz sine wave, VIN = VREF, unless otherwise noted.) Signal-to-Noise Plus Distortion Total Harmonic Distortion SINAD Spurious-Free Dynamic Range SFDR Intermodulation Distortion THD IMD 60 Up to the 5th harmonic 61 dB -80 -74 dB -80 -74 dB fIN1 = 250kHz, fIN2 = 300kHz -78 dB Full-Power Bandwidth -3dB point, small-signal method 20 MHz Full-Linear Bandwidth S/(N + D) > 56dB, single ended 2 MHz CONVERSION RATE Minimum Conversion Time tCONV (Note 3) Maximum Throughput Rate Minimum Throughput Rate Track-and-Hold Acquisition Time (Note 4) tACQ (Note 5) Aperture Delay Aperture Jitter External Clock Frequency 2 0.556 1.8 (Note 6) µs Msps 10 ksps 104 ns 5 ns 30 fSCLK _______________________________________________________________________________________ ps 28.8 MHz 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference (VDD = +5V ±5%, VL = VDD, fSCLK = 28.8MHz, 50% duty cycle, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS ANALOG INPUTS (AIN+, AIN-) Differential Input Voltage Range VIN AIN+ - AIN-, MAX1076 0 VREF AIN+ - AIN-, MAX1078 -VREF / 2 +VREF / 2 0 VDD V ±1 µA Absolute Input Voltage Range DC Leakage Current V Input Capacitance Per input pin 16 pF Input Current (Average) Time averaged at maximum throughput rate 75 µA REFERENCE OUTPUT (REF) REF Output Voltage Range Static, TA = +25°C 4.086 Voltage Temperature Coefficient 4.096 4.106 ±50 Load Regulation Line Regulation ISOURCE = 0 to 2mA 0.3 ISINK = 0 to 200µA 0.5 VDD = 4.75V to 5.25V, static 0.5 V ppm/°C mV/mA mV/V DIGITAL INPUTS (SCLK, CNVST) Input-Voltage Low VIL Input-Voltage High VIH Input Leakage Current 0.3 x VL V ±10 µA pF 0.7 x VL IIL V 0.05 DIGITAL OUTPUT (DOUT) Output Load Capacitance For stated timing performance 30 Output-Voltage Low COUT VOL ISINK = 5mA, VL ≥ 1.8V 0.4 Output-Voltage High VOH ISOURCE = 1mA, VL ≥ 1.8V Output Leakage Current IOL Output high impedance VL - 0.5V V V ±0.2 ±10 µA 5.25 V VDD V POWER REQUIREMENTS Analog Supply Voltage VDD Digital Supply Voltage VL Analog Supply Current, Normal Mode IDD Analog Supply Current, Partial Power-Down Mode IDD Analog Supply Current, Full Power-Down Mode IDD 4.75 1.8 Static, fSCLK = 28.8MHz 8 11 Static, no SCLK 5 7 Operational, 1.8Msps 10 13 fSCLK = 28.8MHz 2 No SCLK 2 fSCLK = 28.8MHz No SCLK PSR 1 µA 1 1 2.5 Static, fSCLK = 28.8MHz 0.4 1 Partial/full power-down mode, fSCLK = 28.8MHz 0.2 0.5 0.1 ±0.2 1 µA ±3.0 mV Static, no SCLK (all modes) Positive-Supply Rejection mA 0.3 Operational, full-scale input at 1.8Msps Digital Supply Current (Note 7) mA VDD = 5V ±5%, full-scale input mA _______________________________________________________________________________________ 3 MAX1076/MAX1078 ELECTRICAL CHARACTERISTICS (continued) MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference TIMING CHARACTERISTICS (VDD = +5V ±5%, VL = VDD, fSCLK = 28.8MHz, 50% duty cycle, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) PARAMETER SYMBOL CONDITIONS MIN SCLK Pulse-Width High tCH VL = 1.8V to VDD 15.6 SCLK Pulse-Width Low tCL VL = 1.8V to VDD 15.6 SCLK Rise to DOUT Transition tDOUT TYP MAX UNITS ns ns CL = 30pF, VL = 4.75V to VDD 14 CL = 30pF, VL = 2.7V to VDD 17 CL = 30pF, VL = 1.8V to VDD ns 24 DOUT Remains Valid After SCLK tDHOLD VL = 1.8V to VDD 4 ns CNVST Fall to SCLK Fall tSETUP VL = 1.8V to VDD 10 ns tCSW VL = 1.8V to VDD 20 CNVST Pulse Width ns Power-Up Time; Full Power-Down tPWR-UP 2 ms Restart Time; Partial Power-Down tRCV 16 Cycles Note 1: Relative accuracy is the deviation of the analog value at any code from its theoretical value after the gain error and the offset error have been nulled. Note 2: No missing codes over temperature. Note 3: Conversion time is defined as the number of clock cycles (16) multiplied by the clock period. Note 4: At sample rates below 10ksps, the input full-linear bandwidth is reduced to 5kHz. Note 5: The listed value of three SCLK cycles is given for full-speed continuous conversions. Acquisition time begins on the 14th rising edge of SCLK and terminates on the next falling edge of CNST. The IC idles in acquisition mode between conversions. Note 6: Undersampling at the maximum signal bandwidth requires the minimum jitter spec for SINAD performance. Note 7: Digital supply current is measured with the VIH level equal to VL, and the VIL level equal to GND. VL CNVST tCSW tSETUP tCL tCH SCLK DOUT tDHOLD tDOUT 6kΩ DOUT DOUT 6kΩ 4 GND GND a) HIGH-Z TO VOH, VOL TO VOH, AND VOH TO HIGH-Z Figure 1. Detailed Serial-Interface Timing CL CL b) HIGH-Z TO VOL, VOH TO VOL, AND VOL TO HIGH-Z Figure 2. Load Circuits for Enable/Disable Times _______________________________________________________________________________________ 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference -0.1 512 756 1024 0 -0.1 -0.2 -0.2 256 0 -512 -256 0 256 0 512 256 512 756 DIGITAL OUTPUT CODE DIGITAL OUTPUT CODE DIGITAL OUTPUT CODE DIFFERENTIAL NONLINEARITY vs. DIGITAL OUTPUT CODE (MAX1078) OFFSET ERROR vs. TEMPERATURE (MAX1076) OFFSET ERROR vs. TEMPERATURE (MAX1078) 0 -0.1 0 -0.25 -0.2 -256 0 256 512 0 -0.50 -40 -15 10 35 60 85 -40 -15 10 35 TEMPERATURE (°C) TEMPERATURE (°C) GAIN ERROR vs. TEMPERATURE (MAX1076) GAIN ERROR vs. TEMPERATURE (MAX1078) DYNAMIC PERFORMANCE vs. INPUT FREQUENCY (MAX1076) 0.25 0 -0.25 0.25 0 -0.25 -0.50 -0.50 -0.75 -0.75 -1.00 10 35 TEMPERATURE (°C) 60 85 SNR 61.5 61.4 61.3 SINAD 61.2 61.1 -1.00 -15 MAX1076/78 toc09 0.50 GAIN ERROR (LSB) 0.50 0.75 61.6 DYNAMIC PERFORMANCE (dB) 0.75 MAX1076/78 toc08 MAX1076/78 toc07 1.00 85 60 DIGITAL OUTPUT CODE 1.00 -40 0.25 -0.25 -0.50 -512 MAX1076/78 toc06 0.25 OFFSET ERROR (LSB) OFFSET ERROR (LSB) 0.1 1024 0.50 MAX1076/78 toc05 0.50 MAX1076/78 toc04 0.2 INL (LSB) 0 -0.1 -0.2 GAIN ERROR (LSB) 0.1 INL (LSB) 0 0.2 MAX1076/78 toc02 0.1 INL (LSB) 0.1 INL (LSB) 0.2 MAX1076/78 toc01 0.2 DIFFERENTIAL NONLINEARITY vs. DIGITAL OUTPUT CODE (MAX1076) INTEGRAL NONLINEARITY vs. DIGITAL OUTPUT CODE (MAX1078) MAX1076/78 toc03 INTEGRAL NONLINEARITY vs. DIGITAL OUTPUT CODE (MAX1076) 61.0 -40 -15 10 35 TEMPERATURE (°C) 60 85 100 200 300 400 500 ANALOG INPUT FREQUENCY (kHz) _______________________________________________________________________________________ 5 MAX1076/MAX1078 Typical Operating Characteristics (VDD = +5V, VL = VDD, fSCLK = 28.8MHz, fSAMPLE = 1.8Msps, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) Typical Operating Characteristics (continued) (VDD = +5V, VL = VDD, fSCLK = 28.8MHz, fSAMPLE = 1.8Msps, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) -84 SNR 61.50 -86 -88 MAX1078 88 SFDR (dB) THD (dB) 61.75 MAX1076 -82 90 MAX1076/78 toc11 -80 MAX1076/78 toc10 DYNAMIC PERFORMANCE (dB) SFDR vs. INPUT FREQUENCY THD vs. INPUT FREQUENCY 62.00 86 -90 61.25 84 -92 SINAD MAX1078 MAX1076 -94 61.00 300 400 500 300 100 500 400 300 400 ANALOG INPUT FREQUENCY (kHz) FFT PLOT (MAX1076) FFT PLOT (MAX1078) TOTAL HARMONIC DISTORTION vs. SOURCE IMPEDANCE -80 -20 -40 -80 -100 -120 -120 -60 fIN = 500kHz -60 -100 150 300 450 600 750 fIN = 100kHz -90 -100 0 900 -70 -80 -140 -140 MAX1076/78 toc15 fIN = 500kHz SINAD = 61.4dB SNR = 61.5dB THD = -93.8dB SFDR = 84.5dB 500 -50 MAX1076/78 toc14 0 THD (dB) -60 0 200 ANALOG INPUT FREQUENCY (kHz) AMPLITUDE (dB) 150 300 450 600 750 900 10 1000 100 ANALOG FREQUENCY (kHz) SOURCE IMPEDANCE (Ω) TWO-TONE IMD PLOT (MAX1076) TWO-TONE IMD PLOT (MAX1078) VDD/VL FULL POWER-DOWN SUPPLY CURRENT vs. TEMPERATURE -20 -40 -60 -80 -100 fSAMPLE = 2Msps fIN1 = 250.039kHz fIN2 = 300.059kHz IMD = 82.1dB -20 AMPLITUDE (dB) fIN2 fIN1 0 -40 fIN2 fIN1 -60 -80 -100 -120 1.0 VDD/VL SUPPLY CURRENT (µA) fSAMPLE = 2Msps fIN1 = 250.039kHz fIN2 = 300.059kHz IMD = -81.9dB MAX1076/78 toc16 0 MAX1076/78 toc17 ANALOG FREQUENCY (kHz) MAX1076/78 toc18 -40 200 ANALOG FREQUENCY (kHz) fIN = 500kHz SINAD = 61.5dB SNR = 61.4dB THD = -88.5dB SFDR = 87.0dB -20 100 MAX1076/78 toc13 0 AMPLITUDE (dB) 82 -96 200 100 0.8 VDD, NO SCLK 0.6 VDD, fSCLK = 28.8MHz 0.4 VL, NO SCLK 0.2 -120 -140 0 -140 0 200 400 600 800 ANALOG FREQUENCY (kHz) 6 MAX1076/78 toc12 DYNAMIC PERFORMANCE vs. INPUT FREQUENCY (MAX1078) AMPLITUDE (dB) MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference 1000 0 200 400 600 800 ANALOG FREQUENCY (kHz) 1000 -40 -15 10 35 TEMPERATURE (°C) _______________________________________________________________________________________ 60 85 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference VL = 5V, fSCLK = 28.8MHz 100 VL = 3V, fSCLK = 28.8MHz 50 6 PARTIAL POWER-DOWN 3 -15 10 35 60 MAX1076/78 toc21 9 6 3 0 -15 -40 85 10 35 60 85 500 0 1000 1500 TEMPERATURE (°C) TEMPERATURE (°C) fSAMPLE (kHz) VL SUPPLY CURRENT vs. TEMPERATURE VL SUPPLY CURRENT vs. CONVERSION RATE REFERENCE VOLTAGE vs. TEMPERATURE CONVERSION, fSCLK = 28.8MHz 0.6 FULL/PARTIAL POWER-DOWN, fSCLK = 28.8MHz 0.2 0.8 VL = 5V 0.6 VL = 3V 0.4 VL = 1.8V 2000 MAX1076/78 toc24 VL SUPPLY CURRENT (mA) 0.8 4.12 REFERENCE VOLTAGE (V) 1.0 MAX1076/78 toc22 1.0 MAX1076/78 toc23 -40 4.10 4.08 0.2 0 -15 10 35 60 0 85 4.06 0 500 TEMPERATURE (°C) 1000 1500 2000 -40 -15 fSAMPLE (kHz) 60 85 4.08 4.07 4.12 MAX1076/78 toc26 4.09 35 REFERENCE VOLTAGE vs. LOAD CURRENT (SINK) REFERENCE VOLTAGE (V) 4.10 10 TEMPERATURE (°C) REFERENCE VOLTAGE vs. LOAD CURRENT (SOURCE) MAX1076/78 toc25 -40 REFERENCE VOLTAGE (V) VL SUPPLY CURRENT (mA) CONVERSION, fSCLK = 28.8MHz 0 0 0.4 9 12 VDD SUPPLY CURRENT (mA) VDD SUPPLY CURRENT (mA) 150 MAX1076/78 toc20 12 MAX1076/78 toc19 VL SUPPLY CURRENT (µA) 200 VDD SUPPLY CURRENT vs. CONVERSION RATE VDD SUPPLY CURRENT vs. TEMPERATURE VL PARTIAL/FULL POWER-DOWN SUPPLY CURRENT vs. TEMPERATURE 4.11 4.10 4.09 4.08 4.06 0 2 4 6 LOAD CURRENT (mA) 8 10 0 100 200 300 400 500 LOAD CURRENT (µA) _______________________________________________________________________________________ 7 MAX1076/MAX1078 Typical Operating Characteristics (continued) (VDD = +5V, VL = VDD, fSCLK = 28.8MHz, fSAMPLE = 1.8Msps, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.) 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference MAX1076/MAX1078 Pin Description PIN NAME FUNCTION 1 AIN- Negative Analog Input 2 REF Reference Voltage Output. Internal 4.096V reference output. Bypass REF with a 0.01µF capacitor and a 4.7µF capacitor to RGND. 3 RGND Reference Ground. Connect RGND to GND. VDD Positive Analog Supply Voltage (+4.75V to +5.25V). Bypass VDD with a 0.01µF capacitor and a 10µF capacitor to GND. 5, 11 N.C. No Connection 6 GND Ground. GND is internally connected to EP. 7 VL 8 DOUT Serial Data Output. Data is clocked out on the rising edge of SCLK. 9 CNVST Convert Start. Forcing CNVST high prepares the part for a conversion. Conversion begins on the falling edge of CNVST. The sampling instant is defined by the falling edge of CNVST. 10 SCLK Serial Clock Input. Clocks data out of the serial interface. SCLK also sets the conversion speed. 12 AIN+ Positive Analog Input — EP 4 Positive Logic Supply Voltage (1.8V to VDD). Bypass VL with a 0.01µF capacitor and a 10µF capacitor to GND. Exposed Paddle. EP is internally connected to GND. Detailed Description The MAX1076/MAX1078 use an input T/H and successive-approximation register (SAR) circuitry to convert an analog input signal to a digital 10-bit output. The serial interface requires only three digital lines (SCLK, CNVST, and DOUT) and provides easy interfacing to microprocessors (µPs) and DSPs. Figure 3 shows the simplified internal structure for the MAX1076/MAX1078. True-Differential Analog Input T/H The equivalent circuit of Figure 4 shows the input architecture of the MAX1076/MAX1078, which is composed of a T/H, a comparator, and a switched-capacitor digital-toanalog converter (DAC). The T/H enters its tracking mode on the 14th SCLK rising edge of the previous conversion. Upon power-up, the T/H enters its tracking mode immediately. The positive input capacitor is connected to AIN+. The negative input capacitor is connected to AIN-. The T/H enters its hold mode on the falling edge of CNVST and the difference between the sampled positive and negative input voltages is converted. The time required for the T/H to acquire an input signal is determined by how quickly its input capacitance is charged. If the input signal’s source impedance is high, the acquisition time lengthens. The acquisition time, tACQ, is the minimum 8 time needed for the signal to be acquired. It is calculated by the following equation: tACQ ≥ 8 × (RS + RIN) × 16pF where RIN = 200Ω, and RS is the source impedance of the input signal. Note: tACQ is never less than 104ns, and any source impedance below 12Ω does not significantly affect the ADC’s AC performance. Input Bandwidth The ADC’s input-tracking circuitry has a 20MHz smallsignal bandwidth, making it possible to digitize highspeed transient events and measure periodic signals with bandwidths exceeding the ADC’s sampling rate by using undersampling techniques. To avoid high-frequency signals being aliased into the frequency band of interest, anti-alias filtering is recommended. Analog Input Protection Internal protection diodes that clamp the analog input to VDD and GND allow the analog input pins to swing from GND - 0.3V to VDD + 0.3V without damage. Both inputs must not exceed VDD or be lower than GND for accurate conversions. _______________________________________________________________________________________ 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference VL CIN+ REF MAX1076/MAX1078 VDD CAPACITIVE DAC RIN+ AIN+ REF 4.096V AIN+ 10-BIT SAR ADC T/H AIN- OUTPUT BUFFER DOUT VAZ COMP CONTROL LOGIC AIN- RGND CONTROL LOGIC AND TIMING CNVST CIN- RIN- CIN+ RIN+ ACQUISITION MODE SCLK MAX1076 MAX1078 GND CAPACITIVE DAC AIN+ Figure 3. Functional Diagram VAZ Serial Interface Initialization After Power-Up and Starting a Conversion Upon initial power-up, the MAX1076/MAX1078 require a complete conversion cycle to initialize the internal calibration. Following this initial conversion, the part is ready for normal operation. This initialization is only required after a hardware power-up sequence and is not required after exiting partial or full power-down mode. To start a conversion, pull CNVST low. At CNVST’s falling edge, the T/H enters its hold mode and a conversion is initiated. SCLK runs the conversion and the data can then be shifted out serially on DOUT. Timing and Control Conversion-start and data-read operations are controlled by the CNVST and SCLK digital inputs. Figures 1 and 5 show timing diagrams, which outline the serialinterface operation. A CNVST falling edge initiates a conversion sequence: the T/H stage holds the input voltage, the ADC begins to convert, and DOUT changes from high impedance to logic low. SCLK is used to drive the conversion process, and it shifts data out as each bit of the conversion is determined. SCLK begins shifting out the data after the 4th rising edge of SCLK. DOUT transitions t DOUT after each SCLK’s rising edge and remains valid 4ns (tDHOLD) after the next rising edge. The 4th rising clock edge produces the MSB of the conversion at DOUT, and the MSB remains valid 4ns after the 5th rising edge. Since there are 10 data bits, 2 sub-bits (S1 and S0), and 3 leading zeros, at least 16 rising clock edges are need- COMP CONTROL LOGIC AINCIN- RIN- HOLD/CONVERSION MODE Figure 4. Equivalent Input Circuit ed to shift out these bits. For continuous operation, pull CNVST high between the 14th and the 16th SCLK rising edges. If CNVST stays low after the falling edge of the 16th SCLK cycle, the DOUT line goes to a highimpedance state on either CNVST’s rising edge or the next SCLK’s rising edge. Partial Power-Down and Full Power-Down Modes Power consumption can be reduced significantly by placing the MAX1076/MAX1078 in either partial powerdown mode or full power-down mode. Partial powerdown mode is ideal for infrequent data sampling and fast wake-up time applications. Pull CNVST high after the 3rd SCLK rising edge and before the 14th SCLK rising edge to enter and stay in partial power-down mode (see Figure 6). This reduces the supply current to 2mA. While in partial power-down mode, the reference remains enabled to allow valid conversions once the IC is returned to normal mode. Drive CNVST low and allow at least 14 SCLK cycles to elapse before driving CNVST high to exit partial power-down mode. Full power-down mode is ideal for infrequent data sampling and very low supply-current applications. The MAX1076/MAX1078 have to be in partial power-down mode in order to enter full power-down mode. Perform the SCLK/CNVST sequence described above to enter _______________________________________________________________________________________ 9 MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference CNVST tSETUP tACQUIRE POWER-MODE SELECTION WINDOW 1 SCLK 2 3 4 HIGH IMPEDANCE 8 D9 DOUT D8 D7 CONTINUOUS-CONVERSION 16 SELECTION WINDOW 14 D6 D5 D4 D3 D2 D1 D0 S1 S0 Figure 5. Interface-Timing Sequence CNVST MUST GO HIGH AFTER THE 3RD BUT BEFORE THE 14TH SCLK RISING EDGE CNVST ONE 8-BIT TRANSFER SCLK DOUT GOES HIGH IMPEDANCE ONCE CNVST GOES HIGH 1ST SCLK RISING EDGE DOUT 0 0 0 MODE D9 D8 D7 D6 D5 NORMAL PPD REF ENABLED (4.096V) Figure 6. SPI Interface—Partial Power-Down Mode EXECUTE PARTIAL POWER-DOWN TWICE CNVST FIRST 8-BIT TRANSFER SECOND 8-BIT TRANSFER SCLK 1ST SCLK RISING EDGE DOUT MODE 0 0 0 DOUT ENTERS TRI-STATE ONCE CNVST GOES HIGH 1ST SCLK RISING EDGE D9 D8 D7 D6 NORMAL REF 0 D5 PPD 0 0 0 0 0 0 RECOVERY 0 FPD ENABLED (4.096V) DISABLED Figure 7. SPI Interface—Full Power-Down Mode partial power-down mode. Then repeat the same sequence to enter full power-down mode (see Figure 7). Drive CNVST low, and allow at least 14 SCLK cycles to elapse before driving CNVST high to exit full powerdown mode. While in full power-down mode, the reference is disabled to minimize power consumption. Be sure to allow at least 2ms recovery time after exiting full power-down mode for the reference to settle. In 10 partial/full power-down mode, maintain a logic low or a logic high on SCLK to minimize power consumption. Transfer Function Figure 8 shows the unipolar transfer function for the MAX1076. Figure 9 shows the bipolar transfer function for the MAX1078. The MAX1076 output is straight binary, while the MAX1078 output is two’s complement. ______________________________________________________________________________________ 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference Internal Reference The MAX1076/MAX1078 have an on-chip voltage reference trimmed to 4.096V. The internal reference output is connected to REF and also drives the internal capacitive DAC. The output can be used as a reference voltage source for other components and can source up to 2mA. Bypass REF with a 0.01µF capacitor and a 4.7µF capacitor to RGND. OUTPUT CODE 111...110 111...101 000...011 How to Start a Conversion 000...010 Connection to Standard Interfaces The MAX1076/MAX1078 serial interface is fully compatible with SPI/QSPI and MICROWIRE (see Figure 11). If a serial interface is available, set the CPU’s serial interface in master mode so the CPU generates the serial clock. Choose a clock frequency up to 28.8MHz. SPI and MICROWIRE When using SPI or MICROWIRE, the MAX1076/MAX1078 are compatible with all four modes programmed with the CPHA and CPOL bits in the SPI or MICROWIRE control register. Conversion begins with a CNVST falling edge. DOUT goes low, indicating a conversion is in progress. Two consecutive 1-byte reads are required to get the full 10 bits from the ADC. DOUT transitions on SCLK rising edges. DOUT is guaranteed to be valid tDOUT later and FULL-SCALE TRANSITION 111...111 The internal reference is continuously powered up during both normal and partial power-down modes. In full power-down mode, the internal reference is disabled. Be sure to allow at least 2ms recovery time after hardware power-up or exiting full power-down mode for the reference to reach its intended value. An analog-to-digital conversion is initiated by CNVST and clocked by SCLK, and the resulting data is clocked out on DOUT by SCLK. With SCLK idling high or low, a falling edge on CNVST begins a conversion. This causes the analog input stage to transition from track to hold mode, and DOUT to transition from high impedance to being actively driven low. A total of 16 SCLK cycles are required to complete a normal conversion. If CNVST is low during the 16th falling SCLK edge, DOUT returns to high impedance on the next rising edge of CNVST or SCLK, enabling the serial interface to be shared by multiple devices. If CNVST returns high after the 14th, but before the 16th SCLK rising edge, DOUT remains active so continuous conversions can be sustained. The highest throughput is achieved when performing continuous conversions. Figure 10 illustrates a conversion using a typical serial interface. MAX1076/MAX1078 Applications Information FS = VREF ZS = 0 V 1 LSB = REF 1024 000...001 000...000 0 1 2 3 FS DIFFERENTIAL INPUT VOLTAGE (LSB) FS - 3/2 LSB Figure 8. Unipolar Transfer Function (MAX1076 Only) OUTPUT CODE FULL-SCALE TRANSITION V FS = REF 2 ZS = 0 -VREF - FS = 2 V 1 LSB = REF 1024 011...111 011...110 000...010 000...001 000...000 111...111 111...110 111...101 100...001 100...000 -FS 0 DIFFERENTIAL INPUT VOLTAGE (LSB) FS FS - 3/2 LSB Figure 9. Bipolar Transfer Function (MAX1078 Only) ______________________________________________________________________________________ 11 MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference CNVST SCLK 1 14 16 1 DOUT 0 0 0 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 S1 S0 0 Figure 10. Continuous Conversion with Burst/Continuous Clock I/O SCK MISO +3V TO +5V CNVST SCLK DOUT MAX1076 MAX1078 SS A) SPI CS SCK MISO +3V TO +5V CNVST SCLK DOUT MAX1076 MAX1078 SS B) QSPI I/O SK SI CNVST SCLK DOUT MAX1076 MAX1078 C) MICROWIRE Figure 11. Common Serial-Interface Connections to the MAX1076/MAX1078 12 ______________________________________________________________________________________ 0 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference MAX1076/MAX1078 CNVST 8 1 9 16 SCLK DOUT HIGH-Z D9 D8 D4 D5 D6 D7 D3 D2 D1 S1 D0 HIGH-Z S0 Figure 12. SPI/MICROWIRE Serial-Interface Timing—Single Conversion (CPOL = CPHA = 0), (CPOL = CPHA = 1) CNVST SCLK 14 1 0 DOUT 0 0 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 16 S1 S0 1 0 0 Figure 13. SPI/MICROWIRE Serial-Interface Timing—Continuous Conversion (CPOL = CPHA = 0), (CPOL = CPHA = 1) CNVST DOUT 16 2 SCLK HIGH-Z HIGH-Z D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 S1 S0 Figure 14. QSPI Serial-Interface Timing—Single Conversion (CPOL = 1, CPHA = 1) remains valid until tDHOLD after the following SCLK rising edge. When using CPOL = 0 and CPHA = 0 or CPOL = 1 and CPHA = 1, the data is clocked into the µP on the following rising edge. When using CPOL = 0 and CPHA = 1 or CPOL = 1 and CPHA = 0, the data is clocked into the µP on the next falling edge. See Figure 11 for connections and Figures 12 and 13 for timing. See the Timing Characteristics section to determine the best mode to use. QSPI Unlike SPI, which requires two 1-byte reads to acquire the 10 bits of data from the ADC, QSPI allows the minimum number of clock cycles necessary to clock in the data. The MAX1076/MAX1078 require 16 clock cycles from the µP to clock out the 10 bits of data. Figure 14 shows a transfer using CPOL = 1 and CPHA = 1. The conversion result contains three zeros, followed by the 10 data bits, 2 sub-bits, and a trailing zero with the data in MSB-first format. DSP Interface to the TMS320C54_ The MAX1076/MAX1078 can be directly connected to the TMS320C54_ family of DSPs from Texas Instruments, Inc. Set the DSP to generate its own clocks or use external clock signals. Use either the standard or buffered serial port. Figure 15 shows the simplest interface between the MAX1076/MAX1078 and ______________________________________________________________________________________ 13 MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference VL MAX1076 SCLK MAX1078 CNVST DVDD CLKX TMS320C54_ CLKR DVDD CLKR TMS320C54_ CNVST FSR DOUT DR FSX FSR DOUT VL MAX1076 MAX1078 SCLK DR CLOCK CONVERT Figure 15. Interfacing to the TMS320C54_ Internal Clocks Figure 16. Interfacing to the TMS320C54_ External Clocks the TMS320C54_, where the transmit serial clock (CLKX) drives the receive serial clock (CLKR) and SCLK, and the transmit frame sync (FSX) drives the receive frame sync (FSR) and CNVST. For continuous conversion, set the serial port to transmit a clock, and pulse the frame sync signal for a clock period before data transmission. The serial-port configuration (SPC) register should be set up with internal frame sync (TXM = 1), CLKX driven by an on-chip clock source (MCM = 1), burst mode (FSM = 1), and 16-bit word length (FO = 0). This setup allows continuous conversions provided that the data-transmit register (DXR) and the data-receive register (DRR) are serviced before the next conversion. Alternatively, autobuffering can be enabled when using the buffered serial port to execute conversions and read the data without CPU intervention. Connect the VL pin to the TMS320C54_ supply voltage when the MAX1076/MAX1078 are operating with an analog supply voltage higher than the DSP supply voltage. The word length can be set to 8 bits with FO = 1 to implement the power-down modes. The CNVST pin must idle high to remain in either power-down state. Another method of connecting the MAX1076/MAX1078 to the TMS320C54_ is to generate the clock signals external to either device. This connection is shown in Figure 16 where serial clock (CLOCK) drives the CLKR and SCLK and the convert signal (CONVERT) drives the FSR and CNVST. MCM = 0, CLKX is taken from the CLKX pin FSM = 1, burst mode The serial port must be set up to accept an external receive-clock and external receive-frame sync. The SPC register should be written as follows: TXM = 0, external frame sync 14 FO = 0, data transmitted/received as 16-bit words This setup allows continuous conversion, provided that the DRR is serviced before the next conversion. Alternatively, autobuffering can be enabled when using the buffered serial port to read the data without CPU intervention. Connect the VL pin to the TMS320C54_ supply voltage when the MAX1076/MAX1078 are operating with an analog supply voltage higher than the DSP supply voltage. The MAX1076/MAX1078 can also be connected to the TMS320C54_ by using the data transmit (DX) pin to drive CNVST and the CLKX generated internally to drive SCLK. A pullup resistor is required on the CNVST signal to keep it high when DX goes high impedance and 0001hex should be written to the DXR continuously for continuous conversions. The power-down modes may be entered by writing 00FFhex to the DXR (see Figures 17 and 18). DSP Interface to the ADSP21_ _ _ The MAX1076/MAX1078 can be directly connected to the ADSP21_ _ _ family of DSPs from Analog Devices, Inc. Figure 19 shows the direct connection of the MAX1076/MAX1078 to the ADSP21_ _ _. There are two modes of operation that can be programmed to interface with the MAX1076/MAX1078. For continuous conversions, idle CNVST low and pulse it high for one clock cycle during the LSB of the previous transmitted word. The ADSP21_ _ _ STCTL and SRCTL registers should be configured for early framing (LAFR = 0) and for an active-high frame (LTFS = 0, LRFS = 0) signal. In this mode, the data-independent frame-sync bit (DITFS = 1) ______________________________________________________________________________________ 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference MAX1076/MAX1078 CNVST SCLK 1 S0 DOUT 0 1 0 0 0 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 S1 S0 0 0 Figure 17. DSP Interface—Continuous Conversion CNVST SCLK DOUT 1 1 0 0 0 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 S1 S0 0 0 0 Figure 18. DSP Interface—Single-Conversion, Continuous/Burst Clock can be selected to eliminate the need for writing to the transmit-data register more than once. For single conversions, idle CNVST high and pulse it low for the entire conversion. The ADSP21_ _ _ STCTL and SRCTL registers should be configured for late framing (LAFR = 1) and for an active-low frame (LTFS = 1, LRFS = 1) signal. This is also the best way to enter the power-down modes by setting the word length to 8 bits (SLEN = 1001). Connect the VL pin to the ADSP21_ _ _ supply voltage when the MAX1076/MAX1078 are operating with a supply voltage higher than the DSP supply voltage (see Figures 17 and 18). Layout, Grounding, and Bypassing For best performance, use PC boards. Wire-wrap boards are not recommended. Board layout should ensure that digital and analog signal lines are separated from each other. Do not run analog and digital (especially clock) lines parallel to one another, or digital lines underneath the ADC package. Figure 20 shows the recommended system ground connections. Establish a single-point analog ground (star ground point) at GND, separate from the logic ground. Connect all other analog grounds and DGND to this star ground point for further noise reduction. The ground return to the power supply for this ground should be low impedance and as short as possible for noise-free operation. High-frequency noise in the VDD power supply can affect the ADC’s high-speed comparator. Bypass this supply to the single-point analog ground with 0.01µF and 10µF bypass capacitors. Minimize capacitor lead lengths for best supply-noise rejection. Definitions Integral Nonlinearity Integral nonlinearity (INL) is the deviation of the values on an actual transfer function from a straight line. This straight line can be either a best-straight-line fit or a line drawn between the end points of the transfer function, once offset and gain errors have been nullified. The static linearity parameters for the MAX1076/MAX1078 are measured using the end-points method. Differential Nonlinearity Differential nonlinearity (DNL) is the difference between an actual step width and the ideal value of 1 LSB. A DNL error specification of 1 LSB or less guarantees no missing codes and a monotonic transfer function. Aperture Jitter Aperture jitter (tAJ) is the sample-to-sample variation in the time between the samples. ______________________________________________________________________________________ 15 MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference VL MAX1076 SCLK MAX1078 CNVST VDDINT TCLK SUPPLIES ADSP21_ _ _ RCLK VDD VL TFS 10µF RFS DOUT GND 10µF DR 0.1µF 0.1µF Figure 19. Interfacing to the ADSP21_ _ _ VDD GND RGND VL DGND VL Aperture Delay Aperture delay (tAD) is the time defined between the falling edge of CNVST and the instant when an actual sample is taken. Signal-to-Noise Ratio For a waveform perfectly reconstructed from digital samples, signal-to-noise ratio (SNR) is the ratio of full-scale analog input (RMS value) to the RMS quantization error (residual error). The theoretical minimum analog-to-digital noise is caused by quantization error, and results directly from the ADC’s resolution (N bits): DIGITAL CIRCUITRY MAX1076 MAX1078 Figure 20. Power-Supply Grounding Condition Total Harmonic Distortion Total harmonic distortion (THD) is the ratio of the RMS sum of the first five harmonics of the input signal to the fundamental itself. This is expressed as: SNR = (6.02 x N + 1.76)dB In reality, there are other noise sources besides quantization noise, including thermal noise, reference noise, clock jitter, etc. Therefore, SNR is computed by taking the ratio of the RMS signal to the RMS noise, which includes all spectral components minus the fundamental, the first five harmonics, and the DC offset. Signal-to-Noise Plus Distortion Signal-to-noise plus distortion (SINAD) is the ratio of the fundamental input frequency’s RMS amplitude to the RMS equivalent of all other ADC output signals: SINAD(dB) = 20 x log (SignalRMS / NoiseRMS) Effective Number of Bits Effective number of bits (ENOB) indicates the global accuracy of an ADC at a specific input frequency and sampling rate. An ideal ADC’s error consists of quantization noise only. With an input range equal to the full-scale range of the ADC, calculate the ENOB as follows: ENOB = 16 V22 + V32 + V42 + V52 THD = 20 x log V 1 where V1 is the fundamental amplitude, and V2 through V5 are the amplitudes of the 2nd- through 5th-order harmonics. Spurious-Free Dynamic Range Spurious-free dynamic range (SFDR) is the ratio of the RMS amplitude of the fundamental (maximum signal component) to the RMS value of the next largest distortion component. Full-Power Bandwidth Full-power bandwidth is the frequency at which the input signal amplitude attenuates by 3dB for a full-scale input. (SINAD − 1.76) 6.02 ______________________________________________________________________________________ 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference Intermodulation Distortion Any device with nonlinearities creates distortion products when two sine waves at two different frequencies (f1 and f2) are input into the device. Intermodulation distortion (IMD) is the total power of the IM2 to IM5 intermodulation products to the Nyquist frequency relative to the total input power of the two input tones, f1 and f2. The individual input tone levels are at -7dBFS. The intermodulation products are as follows: • 2nd-order intermodulation products (IM2): f1 + f2, f2 - f1 • 3rd-order intermodulation products (IM3): 2f1 - f2, 2f2 - f1, 2f1 + f2, 2f2 + f1 • 4th-order intermodulation products (IM4): 3f1 - f2, 3f2 - f1, 3f1 + f2, 3f2 + f1 • 5th-order intermodulation products (IM5): 3f1 - 2f2, 3f2 - 2f1, 3f1 + 2f2, 3f2 + 2f1 Chip Information TRANSISTOR COUNT: 13,016 PROCESS: BiCMOS ______________________________________________________________________________________ 17 MAX1076/MAX1078 Full-Linear Bandwidth Full-linear bandwidth is the frequency at which the signal-to-noise plus distortion (SINAD) is equal to 56dB. Package Information (The package drawing(s) in this data sheet may not reflect the most current specifications. For the latest package outline information go to www.maxim-ic.com/packages.) 24L QFN THIN.EPS MAX1076/MAX1078 1.8Msps, Single-Supply, Low-Power, TrueDifferential, 10-Bit ADCs with Internal Reference PACKAGE OUTLINE 12, 16, 20, 24L THIN QFN, 4x4x0.8mm 21-0139 C 1 2 PACKAGE OUTLINE 12, 16, 20, 24L THIN QFN, 4x4x0.8mm 21-0139 C 2 2 Maxim cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim product. No circuit patent licenses are implied. Maxim reserves the right to change the circuitry and specifications without notice at any time. 18 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600 © 2004 Maxim Integrated Products Printed USA is a registered trademark of Maxim Integrated Products.