19-5134; Rev 1; 9/10 KIT ATION EVALU E L B AVAILA Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package The MAX11646/MAX11647 low-power, 10-bit, 1-/2channel analog-to-digital converters (ADCs) feature internal track/hold (T/H), voltage reference, a clock, and an I 2 C-compatible 2-wire serial interface. These devices operate from a single supply of 2.7V to 3.6V (MAX11647) or 4.5V to 5.5V (MAX11646) and require only 6A at a 1ksps sample rate. AutoShutdownTM powers down the devices between conversions, reducing supply current to less than 1A at lower throughput rates. The MAX11646/MAX11647 each measure two single-ended or one differential input. The fully differential analog inputs are software configurable for unipolar or bipolar and single-ended or differential operation. The full-scale analog input range is determined by the internal reference or by an externally applied reference voltage ranging from 1V to VDD. The MAX11647 features a 2.048V internal reference and the MAX11646 features a 4.096V internal reference. The MAX11646/MAX11647 are available in an ultra-tiny 1.9mm x 2.2mm WLP package and an 8-pin MAX (R) package. These ADCs are guaranteed over the extended temperature range (-40C to +85C). For pin-compatible 12-bit parts, refer to the MAX11644/MAX11645 data sheet. Features Ultra-Tiny 1.9mm x 2.2mm Wafer Level Package High-Speed I2C-Compatible Serial Interface 400kHz Fast Mode 1.7MHz High-Speed Mode Single Supply 2.7V to 3.6V (MAX11647) 4.5V to 5.5V (MAX11646) Internal Reference 2.048V (MAX11647) 4.096V (MAX11646) External Reference: 1V to VDD Internal Clock 2-Channel Single-Ended or 1-Channel Fully Differential Internal FIFO with Channel-Scan Mode Low Power 670A at 94.4ksps 230A at 40ksps 60A at 10ksps 6A at 1ksps 0.5A in Power-Down Mode Software-Configurable Unipolar/Bipolar Applications Ordering Information Handheld Portable Applications Medical Instruments Battery-Powered Test Equipment Power-Supply Monitoring Solar-Powered Remote Systems Received-Signal-Strength Indicators PART TEMP RANGE PINI2C SLAVE PACKAGE ADDRESS MAX11646EUA+ -40C to +85C 8 MAX 0110110 MAX11647EUA+ -40C to +85C 8 MAX 0110110 MAX11647EWC+ -40C to +85C 12 WLP 0110110 +Denotes a lead(Pb)-free/RoHs-compliant package. System Supervision Typical Operating Circuit and Selector Guide appear at end of data sheet. AutoShutdown is a trademark of Maxim Integrated Products, Inc. MAX is a registered trademark of Maxim Integrated Products, Inc. ________________________________________________________________ Maxim Integrated Products 1 For pricing, delivery, and ordering information, please contact Maxim Direct at 1-888-629-4642, or visit Maxim's website at www.maxim-ic.com. MAX11646/MAX11647 General Description MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package ABSOLUTE MAXIMUM RATINGS Operating Temperature Range ...........................-40C to +85C Junction Temperature ......................................................+150C Storage Temperature Range .............................-60C to +150C Lead Temperature (soldering, 10s) MAX only ....................................................................+300C Soldering Temperature (reflow) .......................................+260C VDD to GND ..............................................................-0.3V to +6V AIN0, AIN1, REF to GND ............-0.3V to the lower of (VDD + 0.3V) and 6V SDA, SCL to GND.....................................................-0.3V to +6V Maximum Current Into Any Pin .........................................50mA Continuous Power Dissipation (TA = +70C) 8-Pin MAX (derate 4.5mW/C above +70C) .............362mW 12-Pin WLP (derate 16.1mW/C above +70C) .........1288mW 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 = 2.7V to 3.6V (MAX11647), VDD = 4.5V to 5.5V (MAX11646), VREF = 2.048V (MAX11647), VREF = 4.096V (MAX11646), fSCL = 1.7MHz, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25C. See Tables 1-5 for programming notation.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS DC ACCURACY (Note 2) Resolution 10 Bits Relative Accuracy INL (Note 3) 1 LSB Differential Nonlinearity DNL No missing codes over temperature 1 LSB 1 LSB Offset Error Offset-Error Temperature Coefficient Relative to FSR Gain Error (Note 4) Gain-Temperature Coefficient Relative to FSR 0.3 ppm/C 1 LSB 0.3 ppm/C Channel-to-Channel Offset Matching 0.1 LSB Channel-to-Channel Gain Matching 0.1 LSB DYNAMIC PERFORMANCE (fIN(SINE-WAVE) = 10kHz, VIN(P-P) = VREF, fSAMPLE = 94.4ksps) Signal-to-Noise and Distortion SINAD 60 dB Up to the fifth harmonic -70 dB 70 dB Full-Power Bandwidth SINAD > 57dB 3.0 MHz Full-Linear Bandwidth -3dB point 5.0 MHz Total Harmonic Distortion THD Spurious-Free Dynamic Range SFDR CONVERSION RATE Conversion Time (Note 5) Throughput Rate Track/Hold Acquisition Time 2 tCONV f SAMPLE Internal clock External clock 6.8 10.6 Internal clock, SCAN[1:0] = 01 53 External clock s ksps 94.4 800 _______________________________________________________________________________________ ns Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package (VDD = 2.7V to 3.6V (MAX11647), VDD = 4.5V to 5.5V (MAX11646), VREF = 2.048V (MAX11647), VREF = 4.096V (MAX11646), fSCL = 1.7MHz, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25C. See Tables 1-5 for programming notation.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN Internal Clock Frequency Aperture Delay (Note 6) TYP MAX 2.8 tAD External clock, fast mode 60 External clock, high-speed mode 30 UNITS MHz ns ANALOG INPUT (AIN0/AIN1) Input Voltage Range, SingleEnded and Differential (Note 7) Input Multiplexer Leakage Current Input Capacitance Unipolar 0 VREF Bipolar 0 VREF/2 On/off-leakage current, VAIN_ = 0V or VDD 0.01 CIN 1 22 V A pF INTERNAL REFERENCE (Note 8) Reference Voltage VREF Reference-Voltage Temperature Coefficient TA = +25C MAX11647 1.968 2.048 2.128 MAX11646 3.939 4.096 4.256 TCVREF 25 REF Short-Circuit Current ppm/C 2 REF Source Impedance V 1.5 mA k EXTERNAL REFERENCE REF Input Voltage Range VREF (Note 9) REF Input Current IREF f SAMPLE = 94.4ksps 1 VDD V 40 A DIGITAL INPUTS/OUTPUTS (SCL, SDA) Input High Voltage 0.7 x VDD VIH Input Low Voltage V 0.3 x VDD VIL Input Hysteresis 0.1 x VDD VHYST Input Current I IN Input Capacitance CIN Output Low Voltage VOL V V 10 VIN = 0V to VDD 15 I SINK = 3mA A pF 0.4 V POWER REQUIREMENTS Supply Voltage Supply Current VDD IDD MAX11647 2.7 3.6 MAX11646 4.5 5.5 f SAMPLE = 94.4ksps external clock Internal reference 900 1150 f SAMPLE = 40ksps internal clock External reference 670 900 Internal reference 530 External reference 230 f SAMPLE = 10ksps internal clock Internal reference 380 External reference 60 f SAMPLE =1ksps internal clock Internal reference 330 External reference Shutdown (internal reference off) V A 6 0.5 10 _______________________________________________________________________________________ 3 MAX11646/MAX11647 ELECTRICAL CHARACTERISTICS (continued) MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package ELECTRICAL CHARACTERISTICS (continued) (VDD = 2.7V to 3.6V (MAX11647), VDD = 4.5V to 5.5V (MAX11646), VREF = 2.048V (MAX11647), VREF = 4.096V (MAX11646), fSCL = 1.7MHz, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25C. See Tables 1-5 for programming notation.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS 0.01 0.5 LSB/V POWER REQUIREMENTS Power-Supply Rejection Ratio PSRR Full-scale input (Note 10) TIMING CHARACTERISTICS (Figure 1) (VDD = 2.7V to 3.6V (MAX11647), VDD = 4.5V to 5.5V (MAX11646), VREF = 2.048V (MAX11647), VREF = 4.096V (MAX11646), fSCL = 1.7MHz, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25C. See Tables 1-5 for programming notation.) (Note 1) PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS 400 kHz TIMING CHARACTERISTICS FOR FAST MODE Serial-Clock Frequency f SCL Bus Free Time Between a STOP (P) and a START (S) Condition tBUF 1.3 s Hold Time for a START (S) Condition tHD:STA 0.6 s Low Period of the SCL Clock tLOW 1.3 s High Period of the SCL Clock tHIGH 0.6 s Setup Time for a REPEATED START Condition (Sr) t SU:STA 0.6 s Data Hold Time tHD:DAT Data Setup Time t SU:DAT (Note 11) 0 900 100 ns ns Rise Time of Both SDA and SCL Signals, Receiving tR Measured from 0.3VDD to 0.7VDD 20 + 0.1CB 300 ns Fall Time of SDA Transmitting tF Measured from 0.3VDD to 0.7VDD (Note 12) 20 + 0.1CB 300 ns Setup Time for a STOP (P) Condition 0.6 t SU:STO s Capacitive Load for Each Bus CB 400 pF Pulse Width of Spike t SP 50 ns 1.7 MHz TIMING CHARACTERISTICS FOR HIGH-SPEED MODE (CB = 400pF, Note 13) Serial-Clock Frequency Hold Time, REPEATED START Condition (Sr) f SCLH (Note 14) tHD:STA 160 ns Low Period of the SCL Clock tLOW 320 ns High Period of the SCL Clock tHIGH 120 ns Setup Time for a REPEATED START Condition (Sr) t SU: STA 160 ns Data Hold Time tHD:DAT Data Setup Time t SU:DAT 4 (Note 11) 0 10 _______________________________________________________________________________________ 150 ns ns Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package (VDD = 2.7V to 3.6V (MAX11647), VDD = 4.5V to 5.5V (MAX11646), VREF = 2.048V (MAX11647), VREF = 4.096V (MAX11646), fSCL = 1.7MHz, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25C. See Tables 1-5 for programming notation.) (Note 1) PARAMETER SYMBOL CONDITIONS MAX UNITS 20 80 ns Measured from 0.3VDD to 0.7VDD 20 160 ns Rise Time of SCL Signal (Current Source Enabled) tRCL Measured from 0.3VDD to 0.7VDD Rise Time of SCL Signal After Acknowledge Bit tRCL1 MIN TYP Fall Time of SCL Signal tFCL Measured from 0.3VDD to 0.7VDD 20 80 ns Rise Time of SDA Signal tRDA Measured from 0.3VDD to 0.7VDD 20 160 ns Fall Time of SDA Signal tFDA Measured from 0.3VDD to 0.7VDD (Note 12) 20 160 ns Setup Time for a STOP (P) Condition t SU: STO Capacitive Load for Each Bus Line CB Pulse Width of Spike t SP 160 (Notes 11 and 14) 0 ns 400 pF 10 ns All WLP devices are 100% production tested at TA = +25C. Specifications over temperature limits are guaranteed by design and characterization. Note 2: For DC accuracy, the MAX11646 is tested at VDD = 5V and the MAX11647 is tested at VDD = 3V, with an external reference for both ADCs. All devices are configured for unipolar, single-ended inputs. Note 3: Relative accuracy is the deviation of the analog value at any code from its theoretical value after the full-scale range and offsets have been calibrated. Note 4: Offset nulled. Note 5: Conversion time is defined as the number of clock cycles needed for conversion multiplied by the clock period. Conversion time does not include acquisition time. SCL is the conversion clock in the external clock mode. Note 6: A filter on the SDA and SCL inputs suppresses noise spikes and delays the sampling instant. Note 7: The absolute input voltage range for the analog inputs (AIN0/AIN1) is from GND to VDD. Note 8: When the internal reference is configured to be available at REF (SEL[2:1] = 11), decouple REF to GND with a 0.1F capacitor and a 2k series resistor (see the Typical Operating Circuit). Note 9: ADC performance is limited by the converter's noise floor, typically 300VP-P. Note 10: Measured as follows for the MAX11647: Note 1: 2N VFS (3 . 6V) - VFS (2 . 7V) x V REF (3 . 6V - 2 . 7V) and for the MAX11646, where N is the number of bits: 2N VFS (5 . 5V) - VFS (4 . 5V) x V REF (5 . 5V - 4 . 5V) Note 11: A master device must provide a data hold time for SDA (referred to VIL of SCL) to bridge the undefined region of SCL's falling edge (see Figure 1). Note 12: The minimum value is specified at TA = +25C. Note 13: CB = total capacitance of one bus line in pF. Note 14: fSCL must meet the minimum clock low time plus the rise/fall times. _______________________________________________________________________________________ 5 MAX11646/MAX11647 TIMING CHARACTERISTICS (Figure 1) (continued) Typical Operating Characteristics (VDD = 3.3V (MAX11647), VDD = 5V (MAX11646), fSCL = 1.7MHz, external clock, fSAMPLE = 94.4ksps, single-ended, unipolar, TA = +25C, unless otherwise noted.) INTEGRAL NONLINEARITY vs. DIGITAL OUTPUT CODE 0.4 0.3 -0.1 AMPLITUDE (dBc) INL (LSB) 0.1 0 -0.1 -0.2 -140 200 400 600 800 400 600 800 0 1000 20k 30k 40k FREQUENCY (Hz) SUPPLY CURRENT vs. TEMPERATURE SHUTDOWN SUPPLY CURRENT vs. SUPPLY VOLTAGE SHUTDOWN SUPPLY CURRENT vs. TEMPERATURE MAX11647 EXTERNAL REFERENCE MAX11646 0.5 IDD (A) 0.4 0.3 0.2 450 EXTERNAL REFERENCE 0.50 0.45 0.35 0.30 0.25 0.20 MAX11647 0.15 0.10 0.1 MAX11647 MAX11646 0.40 SUPPLY CURRENT (A) SETUP BYTE EXT REF: 10111011 INT REF: 11011011 INTERNAL REFERENCE SDA = SCL = VDD 50k MAX11646 toc06 MAX11646 toc04 MAX11646 INTERNAL REFERENCE 0.6 350 0.05 0 300 -40 -25 -10 5 20 35 50 65 80 0 3.2 2.7 3.7 TEMPERATURE (C) 4.2 4.7 5.2 A) INTERNAL REFERENCE ALWAYS ON B) EXTERNAL REFERENCE 700 B 600 500 400 1.0008 35 NORMALIZED TO REFERENCE VALUE TA = +25C 1.0006 MAX11646 1.0004 1.0002 1.0000 0.9998 300 0.9996 200 0.9994 100 0.9992 0 20 1.0010 VREF NORMALIZED AVERAGE IDD (A) 800 A 5 MAX11647 0.9990 0 20 40 60 CONVERSION RATE (ksps) 80 100 50 TEMPERATURE (C) INTERNAL REFERENCE VOLTAGE vs. TEMPERATURE MAX11646 toc07 1000 900 -40 -25 -10 SUPPLY VOLTAGE (V) AVERAGE SUPPLY CURRENT vs. CONVERSION RATE (EXTERNAL CLOCK) 6 10k DIGITAL OUTPUT CODE 700 400 200 DIGITAL OUTPUT CODE 800 500 0 1000 MAX11646 toc05 0 550 -160 -0.5 -0.3 600 -100 -120 -0.4 650 -80 -0.3 -0.2 750 -60 MAX11646 toc08 DNL (LSB) 0 fSAMPLE = 94.4ksps fIN = 10kHz -20 -40 0.2 0.1 0 MAX11646 toc02 0.2 FFT PLOT 0.5 MAX11646 toc01 0.3 MAX11646 toc03 DIFFERENTIAL NONLINEARITY vs. DIGITAL OUTPUT CODE SUPPLY CURRENT (A) MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package -40 -25 -10 5 20 35 50 65 TEMPERATURE (C) _______________________________________________________________________________________ 80 65 80 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package NORMALIZED REFERENCE VOLTAGE vs. SUPPLY VOLTAGE 1.00004 -0.1 -0.2 1.00000 0.99998 0.99996 0.99994 0.99992 MAX11646 toc10 MAX11646, NORMALIZED TO REFERENCE VALUE AT VDD = 5V 1.00002 MAX11647, NORMALIZED TO REFERENCE VALUE AT VDD = 3.3V -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 0.99990 -40 -25 -10 2.7 3.0 3.3 3.6 3.9 4.2 4.5 4.8 5.1 5.4 5 20 35 50 65 80 TEMPERATURE (C) VDD (V) OFFSET ERROR vs. SUPPLY VOLTAGE GAIN ERROR vs. TEMPERATURE -0.1 -0.2 MAX11646 toc12 1.0 MAX11646 toc11 0 0.9 0.8 GAIN ERROR (LSB) -0.3 -0.4 -0.5 -0.6 0.7 0.6 0.5 0.4 -0.7 0.3 -0.8 0.2 -0.9 0.1 -1.0 0 2.7 3.0 3.3 3.6 3.9 4.2 4.5 4.8 5.1 5.4 -40 -25 -10 VDD (V) 5 20 35 50 65 80 TEMPERATURE (C) GAIN ERROR vs. SUPPLY VOLTAGE MAX11646 toc13 1.0 0.9 0.8 GAIN ERROR (LSB) OFFSET ERROR (LSB) VREF NORMALIZED 1.00006 OFFSET ERROR (LSB) 1.00008 OFFSET ERROR vs. TEMPERATURE 0 MAX11646 toc09 1.00010 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 2.7 3.2 3.7 4.2 4.7 5.2 VDD (V) _______________________________________________________________________________________ 7 MAX11646/MAX11647 Typical Operating Characteristics (continued) (VDD = 3.3V (MAX11647), VDD = 5V (MAX11646), fSCL = 1.7MHz, external clock, fSAMPLE = 94.4ksps, single-ended, unipolar, TA = +25C, unless otherwise noted.) Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package MAX11646/MAX11647 Pin Configuration TOP VIEW (BUMPS ON BOTTOM) 1 3 4 MAX11647 TOP VIEW + AIN0 1 AIN1 2 N.C. 2 3 MAX11646 MAX11647 REF 4 8 VDD 7 GND 6 SDA 5 SCL MAX A AIN0 AIN1 GND REF B GND GND GND GND C VDD GND SDA SCL WLP Pin Description PIN 8 NAME MAX WLP 1,2 A1, A2 AIN0, AIN1 3 -- N.C. FUNCTION Analog Inputs No connection. Not internally connected. 4 A4 REF Reference Input/Output. Selected in the setup register (see Tables 1 and 6). 5 C4 SCL Clock Input 6 C3 SDA Data Input/Output 7 A3, B1-B4, C2 GND Ground 8 C1 VDD Positive Supply. Bypass to GND with a 0.1F capacitor. _______________________________________________________________________________________ Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package tR tF MAX11646/MAX11647 A. F/S-MODE 2-WIRE SERIAL-INTERFACE TIMING t SDA tSU:DAT tHD:DAT tLOW tHD:STA tBUF tSU:STA tSU:STO SCL tHD:STA tHIGH tR tF S A Sr P B. HS-MODE 2-WIRE SERIAL-INTERFACE TIMING S tRDA tFDA SDA tSU:DAT tHD:DAT tLOW tBUF tHD:STA tSU:STO tSU:STA SCL tHD:STA tHIGH tRCL tFCL tRCL1 S Sr HS MODE A P S F/S MODE Figure 1. 2-Wire Serial-Interface Timing _______________________________________________________________________________________ 9 MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package SDA SCL INPUT SHIFT REGISTER VDD CONTROL LOGIC SETUP REGISTER GND INTERNAL OSCILLATOR CONFIGURATION REGISTER AIN0 AIN1 10-BIT ADC T/H ANALOG INPUT MUX REF OUTPUT SHIFT REGISTER AND RAM REF REFERENCE 4.096V (MAX11646) 2.048V (MAX11647) MAX11646 MAX11647 Figure 2. Functional Diagram Power Supply VDD IOL VOUT SDA 400pF IOH Figure 3. Load Circuit Detailed Description The MAX11646/MAX11647 ADCs use successiveapproximation conversion techniques and fully differential input T/H circuitry to capture and convert an analog signal to a serial 10-bit digital output. The MAX11646/MAX11647 measure either two singleended inputs or one differential input. These devices feature a high-speed, 2-wire serial interface supporting data rates up to 1.7MHz. Figure 2 shows the simplified internal structure for the MAX11646/MAX11647. 10 The MAX11646/MAX11647 operate from a single supply and consume 670A (typ) at sampling rates up to 94.4ksps. The MAX11647 features a 2.048V internal reference and the MAX11646 features a 4.096V internal reference. These devices can be configured for use with an external reference from 1V to VDD. Analog Input and Track/Hold The MAX11646/MAX11647 analog input architecture contains an analog input multiplexer (mux), a fully differential T/H capacitor, T/H switches, a comparator, and a fully differential switched capacitive digital-toanalog converter (DAC) (Figure 4). In single-ended mode, the analog-input multiplexer connects CT/H between the analog input selected by CS0 (see the Configuration/Setup Bytes (Write Cycle) section) and GND (Table 3). In differential mode, the analog input multiplexer connects CT/H to the + and - analog inputs selected by CS0 (Table 4). During the acquisition interval, the T/H switches are in the track position and CT/H charges to the analog input signal. At the end of the acquisition interval, the T/H switches move to the hold position retaining the charge on CT/H as a stable sample of the input signal. ______________________________________________________________________________________ Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package The time required for the T/H circuitry to acquire an input signal is a function of the input sample capacitance. If the analog input source impedance is high, the acquisition time constant lengthens and more time must be allowed between conversions. The acquisition time (tACQ) is the minimum time needed for the signal to be acquired. It is calculated by: Sufficiently low source impedance is required to ensure an accurate sample. A source impedance of up to 1.5k does not significantly degrade sampling accuracy. To minimize sampling errors with higher source impedances, connect a 100pF capacitor from the analog input to GND. This input capacitor forms an RC filter with the source impedance limiting the analog-input bandwidth. For larger source impedances, use a buffer amplifier to maintain analog-input signal integrity and bandwidth. When operating in internal clock mode, the T/H circuitry enters its tracking mode on the eighth rising clock edge of the address byte (see the Slave Address section). The T/H circuitry enters hold mode on the falling clock edge of the acknowledge bit of the address byte (the ninth clock pulse). A conversion or a series of conversions is then internally clocked and the MAX11646/MAX11647 hold SCL low. With external clock mode, the T/H circuitry enters track mode after a valid address on the rising edge of the clock during the read (R/W = 1) bit. Hold mode is then entered on the rising edge of the second clock pulse during the shifting out of the first byte of the result. The conversion is performed during the next 10 clock cycles. tACQ 9 (RSOURCE + RIN) CIN where RSOURCE is the analog input source impedance, RIN = 2.5k, and CIN = 22pF. tACQ is 1.5/fSCL for internal clock mode and tACQ = 2/fSCL for external clock mode. Analog Input Bandwidth The MAX11646/MAX11647 feature input-tracking circuitry with a 5MHz small-signal bandwidth. The 5MHz input bandwidth makes it possible to digitize highspeed transient events and measure periodic signals with bandwidths exceeding the ADC's sampling rate by using under sampling techniques. To avoid high-frequency signals being aliased into the frequency band of interest, anti-alias filtering is recommended. Analog Input Range and Protection Internal protection diodes clamp the analog input to VDD and GND. These diodes allow the analog inputs to swing from (VGND - 0.3V) to (VDD + 0.3V) without causing damage to the device. For accurate conversions the inputs must not go more than 50mV below GND or above VDD. HOLD ANALOG INPUT MUX REF CT/H AIN0 TRACK CAPACITIVE DAC TRACK VDD/2 HOLD AIN1 HOLD TRACK HOLD TRACK TRACK GND CAPACITIVE DAC CT/H HOLD REF MAX11646 MAX11647 Figure 4. Equivalent Input Circuit ______________________________________________________________________________________ 11 MAX11646/MAX11647 During the conversion interval, the switched capacitive DAC adjusts to restore the comparator input voltage to 0V within the limits of 10-bit resolution. This action requires 10 conversion clock cycles and is equivalent to transferring a charge of 11pF (VIN+ - VIN-) from CT/H to the binary-weighted capacitive DAC, forming a digital representation of the analog input signal. MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Single-Ended/Differential Input The SGL/DIF of the configuration byte configures the MAX11646/MAX11647 analog input circuitry for singleended or differential inputs (Table 2). In single-ended mode (SGL/DIF = 1), the digital conversion results are the difference between the analog input selected by CS0 and GND (Table 3). In differential mode (SGL/ DIF = 0), the digital conversion results are the difference between the + and the - analog inputs selected by CS0 (Table 4). Unipolar/Bipolar When operating in differential mode, the BIP/UNI bit of the setup byte (Table 1) selects unipolar or bipolar operation. Unipolar mode sets the differential input range from 0 to VREF. A negative differential analog input in unipolar mode causes the digital output code to be zero. Selecting bipolar mode sets the differential input range to VREF/2. The digital output code is binary in unipolar mode and two's complement in bipolar mode. See the Transfer Functions section. In single-ended mode, the MAX11646/MAX11647 always operate in unipolar mode irrespective of BIP/UNI. The analog inputs are internally referenced to GND with a full-scale input range from 0 to VREF. 2-Wire Digital Interface The MAX11646/MAX11647 feature a 2-wire interface consisting of a serial-data line (SDA) and serial-clock line (SCL). SDA and SCL facilitate bidirectional communication between the MAX11646/MAX11647 and the master at rates up to 1.7MHz. The MAX11646/MAX11647 are slaves that transfer and receive data. The master (typically a microcontroller) initiates data transfer on the bus and generates the SCL signal to permit that transfer. SDA and SCL must be pulled high. This is typically done with pullup resistors (750 or greater) (see the Typical Operating Circuit). Series resistors (RS) are optional. They protect the input architecture of the MAX11646/ MAX11647 from high voltage spikes on the bus lines, minimize crosstalk, and undershoot of the bus signals. Bit Transfer One data bit is transferred during each SCL clock cycle. A minimum of 18 clock cycles are required to transfer the data in or out of the MAX11646/ MAX11647. The data on SDA must remain stable during the high period of the SCL clock pulse. Changes in SDA while SCL is stable are considered control signals (see the START and STOP Conditions section). Both SDA and SCL remain high when the bus is not busy. 12 START and STOP Conditions The master initiates a transmission with a START condition (S), a high-to-low transition on SDA while SCL is high. The master terminates a transmission with a STOP condition (P), a low-to-high transition on SDA while SCL is high (Figure 5). A repeated START condition (Sr) can be used in place of a STOP condition to leave the bus active and the mode unchanged (see the HS Mode section). Acknowledge Bits Data transfers are acknowledged with an acknowledge bit (A) or a not-acknowledge bit (A). Both the master and the MAX11646/MAX11647 (slave) generate acknowledge bits. To generate an acknowledge, the receiving device must pull SDA low before the rising edge of the acknowledge-related clock pulse (ninth pulse) and keep it low during the high period of the clock pulse (Figure 6). To generate a not-acknowledge, the receiver allows SDA to be pulled high before the rising edge of the acknowledge-related clock pulse and leaves SDA high during the high period of the clock pulse. Monitoring the acknowledge bits allows for detection of unsuccessful data transfers. An unsuccessful data transfer happens if a receiving device is busy or if a system fault has occurred. In the event of an unsuccessful data transfer, the bus master should reattempt communication at a later time. Sr S P SDA SCL Figure 5. START and STOP Conditions S NOT ACKNOWLEDGE SDA ACKNOWLEDGE SCL 1 2 Figure 6. Acknowledge Bits ______________________________________________________________________________________ 8 9 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package SLAVE ADDRESS 0110110 SLAVE ADDRESS S 0 1 1 0 1 1 0 R/W A SDA 1 SCL 2 3 4 5 6 7 8 9 Figure 7. Slave Address Byte Slave Address A bus master initiates communication with a slave device by issuing a START condition followed by a slave address. When idle, the MAX11646/MAX11647 continuously wait for a START condition followed by their slave address. When the MAX11646/MAX11647 recognize their slave address, they are ready to accept or send data. The slave address has been factory programmed and is always 0110110 for the MAX11646/MAX11647 (Figure 7). The least significant bit (LSB) of the address byte (R/W) determines whether the master is writing to or reading from the MAX11646/MAX11647 (R/W = 0 selects a write condition, R/W = 1 selects a read condition). After receiving the address, the MAX11646/MAX11647 (slave) issue an acknowledge by pulling SDA low for one clock cycle. high-speed mode (HS mode) to achieve conversion rates up to 94.4ksps. Figure 1 shows the bus timing for the MAX11646/MAX11647's 2-wire interface. HS Mode At power-up, the MAX11646/MAX11647 bus timing is set for F/S mode. The bus master selects HS mode by addressing all devices on the bus with the HS-mode master code 0000 1XXX (X = don't care). After successfully receiving the HS-mode master code, the MAX11646/MAX11647 issue a not-acknowledge, allowing SDA to be pulled high for one clock cycle (Figure 8). After the not-acknowledge, the MAX11646/ MAX11647 are in HS mode. The bus master must then send a repeated START followed by a slave address to initiate HS-mode communication. If the master generates a STOP condition the MAX11646/MAX11647 return to F/S mode. Bus Timing At power-up, the MAX11646/MAX11647 bus timing is set for fast mode (F/S mode), allowing conversion rates up to 22.2ksps. The MAX11646/MAX11647 must operate in HS-MODE MASTER CODE S 0 0 0 0 1 X X X A Sr SDA SCL F/S MODE HS MODE Figure 8. F/S-Mode to HS-Mode Transfer ______________________________________________________________________________________ 13 MAX11646/MAX11647 DEVICE MAX11646/MAX11647 MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Configuration/Setup Bytes (Write Cycle) A write cycle begins with the bus master issuing a START condition followed by 7 address bits (Figure 7) and a write bit (R/W = 0). If the address byte is successfully received, the MAX11646/MAX11647 (slave) issue an acknowledge. The master then writes to the slave. The slave recognizes the received byte as the setup byte (Table 1) if the most significant bit (MSB) is 1. If the MSB is 0, the slave recognizes that byte as the configu- ration byte (Table 2). The master can write either 1 or 2 bytes to the slave in any order (setup byte then configuration byte, configuration byte then setup byte, setup byte or configuration byte only; see Figure 9). If the slave receives a byte successfully, it issues an acknowledge. The master ends the write cycle by issuing a STOP condition or a repeated START condition. When operating in HS mode, a STOP condition returns the bus into F/S mode (see the HS Mode section). MASTER TO SLAVE SLAVE TO MASTER A. 1-BYTE WRITE CYCLE 1 7 1 1 S SLAVE ADDRESS W A 1 8 1 NUMBER OF BITS SETUP OR A P or Sr CONFIGURATION BYTE MSB DETERMINES WHETHER SETUP OR CONFIGURATION BYTE B. 2-BYTE WRITE CYCLE 7 1 S 1 1 1 8 SETUP OR W A CONFIGURATION BYTE SLAVE ADDRESS A 8 1 1 NUMBER OF BITS SETUP OR A P or Sr CONFIGURATION BYTE MSB DETERMINES WHETHER SETUP OR CONFIGURATION BYTE Figure 9. Write Cycle Table 1. Setup Byte Format 14 BIT 7 (MSB) BIT 6 BIT 5 BIT 4 BIT 3 BIT 2 BIT 1 BIT 0 (LSB) REG SEL2 SEL1 SEL0 CLK BIP/UNI RST X BIT NAME 7 REG 6 SEL2 5 SEL1 4 SEL0 3 CLK 2 BIP/UNI 1 RST 0 X DESCRIPTION Register bit. 1 = setup byte, 0 = configuration byte (see Table 2). Three bits select the reference voltage (Table 6). Default to 000 at power-up. 1 = external clock, 0 = internal clock. Defaulted to 0 at power-up. 1 = bipolar, 0 = unipolar. Defaulted to 0 at power-up (see the Unipolar/Bipolar section). 1 = no action, 0 = resets the configuration register to default. Setup register remains unchanged. Don't-care bit. This bit can be set to 1 or 0. ______________________________________________________________________________________ Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package MAX11646/MAX11647 Table 2. Configuration Byte Format BIT 7 (MSB) BIT 6 BIT 5 BIT 4 BIT 3 BIT 2 BIT 1 BIT 0 (LSB) REG SCAN1 SCAN0 X X X CS0 SGL/DIF BIT NAME 7 REG 6 SCAN1 5 SCAN0 4 X 3 X 2 X 1 CS0 0 SGL/DIF DESCRIPTION Register bit. 1= setup byte (see Table 1), 0 = configuration byte. Scan-select bits. Two bits select the scanning configuration (Table 5). Defaults to 00 at power-up. Channel-select bit. CS0 selects which analog input channels are to be used for conversion (Tables 3 and 4). Defaults to 0000 at power-up. 1 = single-ended, 0 = differential (Tables 3 and 4). Defaults to 1 at power-up. See the SingleEnded/Differential Input section. X = Don't care. Table 3. Channel Selection in SingleEnded Mode (SGL/DIF = 1) CS0 AIN0 0 + 1 AIN1 + Table 4. Channel Selection in Differential Mode (SGL/DIF = 0) GND CS0 AIN0 AIN1 - 0 + - 1 - + - X = Don't care. X = Don't care. Data Byte (Read Cycle) A read cycle must be initiated to obtain conversion results. Read cycles begin with the bus master issuing a START condition followed by 7 address bits and a read bit (R/W = 1). If the address byte is successfully received, the MAX11646/MAX11647 (slave) issue an acknowledge. The master then reads from the slave. The result is transmitted in 2 bytes; first 6 bits of the first byte are high, then MSB through LSB are consecutively clocked out. After the master has received the byte(s), it can issue an acknowledge if it wants to continue reading or a not-acknowledge if it no longer wishes to read. If the MAX11646/MAX11647 receive a not- acknowledge, they release SDA, allowing the master to generate a STOP or a repeated START condition. See the Clock Modes and Scan Mode sections for detailed information on how data is obtained and converted. Clock Modes The clock mode determines the conversion clock and the data acquisition and conversion time. The clock mode also affects the scan mode. The state of the setup byte's CLK bit determines the clock mode (Table 1). At power-up, the MAX11646/MAX11647 are defaulted to internal clock mode (CLK = 0). ______________________________________________________________________________________ 15 MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Internal Clock When configured for internal clock mode (CLK = 0), the MAX11646/MAX11647 use their internal oscillator as the conversion clock. In internal clock mode, the MAX11646/MAX11647 begin tracking the analog input after a valid address on the eighth rising edge of the clock. On the falling edge of the ninth clock, the analog signal is acquired and the conversion begins. While converting the analog input signal, the MAX11646/ MAX11647 hold SCL low (clock stretching). After the conversion completes, the results are stored in internal memory. If the scan mode is set for multiple conversions, they all happen in succession with each additional result stored in memory. The MAX11646/ MAX11647 contain two 10-bit blocks of memory. Once all conversions are complete, the MAX11646/MAX11647 release SCL, allowing it to be pulled high. The master can now clock the results out of the memory in the same order the scan conversion has been done at a clock rate of up to 1.7MHz. SCL is stretched for a maximum of 7.6s per channel (see Figure 10). The device memory contains all of the conversion results when the MAX11646/MAX11647 release SCL. The converted results are read back in a first-in/first-out (FIFO) sequence. The memory contents can be read continuously. If reading continues past the result stored in memory, the pointer wraps around and point to the first result. Note that only the current conversion results are read from memory. The device must be addressed with a read command to obtain new conversion results. The internal clock mode's clock stretching quiets the SCL bus signal, reducing the system noise during conversion. Using the internal clock also frees the bus master (typically a microcontroller) from the burden of running the conversion clock, allowing it to perform other tasks that do not need to use the bus. MASTER TO SLAVE SLAVE TO MASTER A. SINGLE CONVERSION WITH INTERNAL CLOCK 1 7 1 1 S SLAVE ADDRESS R A 8 CLOCK STRETCH RESULT 2 MSBs 8 A RESULT 8 LSBs 1 1 NUMBER OF BITS A P or Sr tACQ tCONV B. SCAN MODE CONVERSIONS WITH INTERNAL CLOCK 1 7 1 1 S SLAVE ADDRESS R A 8 CLOCK STRETCH tACQ1 tCONV1 tACQ2 tCONV2 CLOCK STRETCH 1 8 1 RESULT 1 ( 2MSBs) A RESULT 1 (8 LSBs) A 8 1 8 tACQN tCONVN Figure 10. Internal Clock Mode Read Cycles 16 1 1 RESULT N (8MSBs) A RESULT N (8LSBs) A P or Sr ______________________________________________________________________________________ NUMBER OF BITS Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package The MAX11646/MAX11647 must operate in external clock mode for conversion rates from 40ksps to 94.4ksps. Below 40ksps internal clock mode is recommended due to much smaller power consumption. Scan Mode SCAN0 and SCAN1 of the configuration byte set the scan mode configuration. Table 5 shows the scanning configurations. The scanned results are written to memory in the same order as the conversion. Read the results from memory in the order they were converted. Each result needs a 2-byte transmission, the first byte begins with six empty bits during which SDA is left high. Each byte has to be acknowledged by the master or the memory transmission is terminated. It is not possible to read the memory independently of conversion. MASTER TO SLAVE SLAVE TO MASTER A. SINGLE CONVERSION WITH EXTERNAL CLOCK 1 7 1 1 8 1 8 1 1 S SLAVE ADDRESS R A RESULT (2 MSBs) A RESULT (8 LSBs) A P OR Sr NUMBER OF BITS tACQ tCONV B. SCAN MODE CONVERSIONS WITH EXTERNAL CLOCK 1 7 1 1 S SLAVE ADDRESS R A 8 RESULT 1 (2 MSBs) 1 8 1 8 1 8 A RESULT 2 (8 LSBs) A RESULT N (2 MSBs) A RESULT N (8 LSBs) tACQ2 tACQN tACQ1 tCONV1 1 1 NUMBER OF BITS A P OR Sr tCONVN Figure 11. External Clock Mode Read Cycle Table 5. Scanning Configuration SCAN1 SCAN0 0 0 Scans up from AIN0 to the input selected by CS0. SCANNING CONFIGURATION 0 1 Converts the input selected by CS0 eight times (see Tables 3 and 4).* 1 0 Reserved. Do not use. 1 1 Converts input selected by CS0.* *When operating in external clock mode, there is no difference between SCAN[1:0] = 01 and SCAN[1:0] = 11, and converting occurs perpetually until not acknowledge occurs. ______________________________________________________________________________________ 17 MAX11646/MAX11647 External Clock When configured for external clock mode (CLK = 1), the MAX11646/MAX11647 use the SCL as the conversion clock. In external clock mode, the MAX11646/ MAX11647 begin tracking the analog input on the ninth rising clock edge of a valid slave address byte. Two SCL clock cycles later the analog signal is acquired and the conversion begins. Unlike internal clock mode, converted data is available immediately after the first four empty high bits. The device continuously converts input channels dictated by the scan mode until given a not acknowledge. There is no need to re-address the device with a read command to obtain new conversion results (see Figure 11). The conversion must complete in 1ms or droop on the track-and-hold capacitor degrades conversion results. Use internal clock mode if the SCL clock period exceeds 60s. MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Applications Information Power-On Reset The configuration and setup registers (Tables 1 and 2) default to a single-ended, unipolar, single-channel conversion on AIN0 using the internal clock with VDD as the reference. The memory contents are unknown after power-up. Automatic Shutdown Automatic shutdown occurs between conversions when the MAX11646/MAX11647 are idle. All analog circuits participate in automatic shutdown except the internal reference due to its prohibitively long wake-up time. When operating in external clock mode, a STOP, notacknowledge, or repeated START condition must be issued to place the devices in idle mode and benefit from automatic shutdown. A STOP condition is not necessary in internal clock mode to benefit from automatic shutdown because power-down occurs once all conversion results are written to memory (Figure 10). When using an external reference or VDD as a reference, all analog circuitry is inactive in shutdown and supply current is less than 0.5A (typ). The digital conversion results obtained in internal clock mode are maintained in memory during shutdown and are available for access through the serial interface at any time prior to a STOP or a repeated START condition. When idle, the MAX11646/MAX11647 continuously wait for a START condition followed by their slave address (see the Slave Address section). Upon reading a valid address byte the MAX11646/MAX11647 power up. The internal reference requires 10ms to wake up, so when using the internal reference it should be powered up 10ms prior to conversion or powered continuously. Wake-up is invisible when using an external reference or VDD as the reference. Automatic shutdown results in dramatic power savings, particularly at slow conversion rates and with internal clock. For example, at a conversion rate of 10ksps, the average supply current for the MAX11647 is 60A (typ) and drops to 6A (typ) at 1ksps. At 0.1ksps the average supply current is just 1A, or a minuscule 3W of power consumption (see Average Supply Current vs. Conversion Rate (External Clock) in the Typical Operating Characteristics). Reference Voltage SEL[2:0] of the setup byte (Table 1) control the reference and the REF configuration (Table 6). Internal Reference The internal reference is 4.096V for the MAX11646 and 2.048V for the MAX11647. When REF is configured to be an internal reference output (SEL[2:1] = 11), decouple REF to GND with a 0.1F capacitor and a 2k series resistor (see the Typical Operating Circuit). Once powered up, the reference always remains on until reconfigured. The internal reference requires 10ms to wake up and is accessed using SEL0 (Table 6). When in shutdown, the internal reference output is in a high-impedance state. The reference should not be used to supply current for external circuitry. The internal reference does not require an external bypass capacitor and works best when left unconnected (SEL1 = 0). External Reference The external reference can range from 1V to VDD. For maximum conversion accuracy, the reference must be able to deliver up to 40A and have an output impedance of 500 or less. If the reference has a higher output impedance or is noisy, bypass it to GND as close as possible to REF with a 0.1F capacitor. Table 6. Reference Voltage and REF Format SEL2 SEL1 SEL0 REFERENCE VOLTAGE REF INTERNAL REFERENCE STATE 0 0 X 0 1 X VDD Not connected Always off External reference Reference input 1 0 0 Internal reference Always off Not connected* Always off 1 0 1 1 1 0 Internal reference Not connected* Always on Internal reference Reference output 1 1 1 Always off Internal reference Reference output Always on X = Don't care. *Preferred configuration for internal reference. 18 ______________________________________________________________________________________ Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Layout, Grounding, and Bypassing Only use PCBs. Wire-wrap configurations are not recommended since the layout should ensure proper separation of analog and digital traces. Do not run analog and digital lines parallel to each other, and do not lay out digital signal paths underneath the ADC package. Use separate analog and digital PCB ground sections with only one star point (Figure 14) connecting the two ground systems (analog and digital). For lowest noise operation, ensure the ground return to the star ground's power supply is low impedance and as short as possible. Route digital signals far away from sensitive analog and reference inputs. OUTPUT CODE FULL-SCALE TRANSITION 11 . . . 111 MAX11646 MAX11647 11 . . . 110 11 . . . 101 FS = VREF ZS = GND V 1 LSB = REF 1024 00 . . . 011 High-frequency noise in the power supply (VDD) could influence the proper operation of the ADC's fast comparator. Bypass VDD to the star ground with a network of two parallel capacitors, 0.1F and 4.7F, located as close as possible to the MAX11646/MAX11647 powersupply pin. Minimize capacitor lead length for best supply noise rejection, and add an attenuation resistor (5) in series with the power supply if it is extremely noisy. Definitions 00 . . . 010 00 . . . 001 Integral Nonlinearity 00 . . . 000 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 MAX11646/ MAX11647's INL is measured using the endpoint. 0 1 2 3 FS FS - 3/2 LSB INPUT VOLTAGE (LSB) Figure 12. Unipolar Transfer Function OUTPUT CODE 011 . . . 111 V FS = REF 2 011 . . . 110 ZS = 0 000 . . . 010 000 . . . 001 MAX11646 MAX11647 SUPPLIES 3V OR 5V -VREF 2 V 1 LSB = REF 1024 VLOGIC = 3V/5V GND -FS = 000 . . . 000 4.7F R* = 5 111 . . . 111 111 . . . 110 0.1F 111 . . . 101 VDD GND 3V/5V DGND 100 . . . 001 100 . . . 000 0 - FS MAX11646 MAX11647 +FS - 1 LSB DIGITAL CIRCUITRY INPUT VOLTAGE (LSB) *VCOM VREF/2 *VIN = (AIN+) - (AIN-) Figure 13. Bipolar Transfer Function *OPTIONAL Figure 14. Power-Supply Grounding Connection ______________________________________________________________________________________ 19 MAX11646/MAX11647 Transfer Functions Output data coding for the MAX11646/MAX11647 is binary in unipolar mode and two's complement in bipolar mode with 1 LSB = (VREF/2N) where N is the number of bits (10). Code transitions occur halfway between successive-integer LSB values. Figures 12 and 13 show the input/output (I/O) transfer functions for unipolar and bipolar operations, respectively. MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Differential Nonlinearity Effective Number of Bits Differential nonlinearity (DNL) is the difference between an actual step width and the ideal value of 1 LSB. A DNL error specification of less than 1 LSB guarantees no missing codes and a monotonic transfer function. 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 ADC's full-scale range, calculate the ENOB as follows: Aperture Jitter Aperture jitter (tAJ) is the sample-to-sample variation in the time between the samples. Aperture Delay Aperture delay (tAD) is the time between the falling edge of the sampling clock and the instant when an actual sample is taken. Signal-to-Noise Ratio For a waveform perfectly reconstructed from digital samples, the theoretical maximum SNR is the ratio of the fullscale analog input (RMS value) to the RMS quantization error (residual error). The ideal, theoretical minimum analog-to-digital noise is caused by quantization error only and results directly from the ADC's resolution (N Bits): SNRMAX[dB] = 6.02dB N + 1.76dB In reality, there are other noise sources besides quantization noise: thermal noise, reference noise, clock jitter, etc. 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. SignalRMS SINAD(dB) = 20 x log NoiseRMS + THDRMS ENOB = (SINAD - 1.76)/6.02 Total Harmonic Distortion Total harmonic distortion (THD) is the ratio of the RMS sum of the input signal's first five harmonics to the fundamental itself. This is expressed as: THD = 20 x log V 2 +V 2 +V 2 +V 2 3 4 5 2 V1 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 RMS amplitude of the fundamental (maximum signal component) to the RMS value of the next largest distortion component. Signal-to-Noise Plus Distortion Signal-to-noise plus distortion (SINAD) is the ratio of the fundamental input frequency's RMS amplitude to RMS equivalent of all other ADC output signals. SINAD (dB) = 20 log (SignalRMS/NoiseRMS) 20 ______________________________________________________________________________________ Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Selector Guide 3.3V or 5V PART 0.1F VDD RS* ANALOG INPUTS AIN0 AIN1 MAX11646 2 SingleEnded/1 Differential 4.096 4.5 to 5.5 1 MAX11647 2 SingleEnded/1 Differential 2.048 2.7 to 3.6 1 SDA MAX11646 MAX11647 SCL RS* INTERNAL SUPPLY INPUT INL REFERENCE VOLTAGE CHANNELS (LSB) (V) (V) RC NETWORK* 2k CREF 0.1F REF GND Chip Information 5V 5V C RP SDA SCL *OPTIONAL RP PROCESS: BiCMOS Package Information For the latest package outline information and land patterns, go to www.maxim-ic.com/packages. Note that a "+", "#", or "-" in the package code indicates RoHS status only. Package drawings may show a different suffix character, but the drawing pertains to the package regardless of RoHS status. PACKAGE TYPE 8 MAX PACKAGE CODE U8CN+1 OUTLINE NO. 21-0036 LAND PATTERN NO. 90-0092 12 WLP W121C2+1 21-0009 Refer to Application Note 1891 ______________________________________________________________________________________ 21 MAX11646/MAX11647 Typical Operating Circuit MAX11646/MAX11647 Low-Power, 1-/2-Channel, I2C, 10-Bit ADCs in Ultra-Tiny 1.9mm x 2.2mm Package Revision History REVISION NUMBER REVISION DATE 0 1/10 Initial release 1 9/10 Added the WLP package to the Ordering Information, Absolute Maximum Ratings, Pin Configuration, Pin Description, and Package Information sections DESCRIPTION PAGES CHANGED -- 1, 2, 8, 20 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. 22 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600 (c) 2010 Maxim Integrated Products Maxim is a registered trademark of Maxim Integrated Products, Inc.