MAX1458CAE

19-1373; Rev 0; 5/98 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner General Description The MAX1458 highly integrated analog-sensor signal processor is optimized for piezoresistive sensor calibration and compensation without any external components. It includes a programmable current source for sensor excitation, a 3-bit programmable-gain amplifier (PGA), a 128-bit internal EEPROM, and four 12-bit DACs. Achieving a total error factor within 1% of the sensor’s repeatability errors, the MAX1458 compensates offset, offset temperature coefficient, full-span output (FSO), FSO temperature coefficient (FSOTC), and FSO nonlinearity of silicon piezoresistive sensors. The MAX1458 calibrates and compensates first-order temperature errors by adjusting the offset and span of the input signal via digital-to-analog converters (DACs), thereby eliminating quantization noise. Built-in testability features on the MAX1458 result in the integration of three traditional sensor-manufacturing operations into one automated process: • Pretest: Data acquisition of sensor performance under the control of a host test computer. • Calibration and compensation: Computation and storage (in an internal EEPROM) of calibration and compensation coefficients computed by the test computer and downloaded to the MAX1458. • Final test operation: Verification of transducer calibration and compensation without removal from the pretest socket. Although optimized for use with piezoresistive sensors, the MAX1458 may also be used with other resistive sensors (i.e., accelerometers and strain gauges) with some additional external components. PART MAX1458CAE MAX1458C/D MAX1458AAE Features o Medium Accuracy (±1%), Single-Chip Sensor Signal Conditioning o Sensor Errors Trimmed Using Correction Coefficients Stored in Internal EEPROM— Eliminates the Need for Laser Trimming and Potentiometers o Compensates Offset, Offset-TC, FSO, FSOTC, FSO Linearity o Programmable Current Source (0.1mA to 2.0mA) for Sensor Excitation o Fast Signal-Path Settling Time (1MΩ. Rail-to-rail input range. Sensor Excitation Current Output. This current source drives the bridge. Positive Sensor Input. Input impedance >1MΩ. Rail-to-rail input range. Positive Power-Supply Input. Connect a 0.1µF capacitor from VDD to VSS. Voltage Limit Input. This pin sets the maximum voltage at OUT. If left unconnected, the output voltage will be limited to 4.6V (VDD = 5V). Connect to VDD for maximum output swing. The acceptable range is 4.5V ≤ VLIMIT ≤ VDD. 7 8 9 10 12 13 14 15 16 WE VSS ISRC OUT INM BDRIVE INP VDD LIMIT _______________Detailed Description The MAX1458 provides an analog amplification path for the sensor signal. Calibration and temperature compensation are achieved by varying the offset and gain of a programmable-gain amplifier (PGA) and by varying the sensor bridge current. The PGA uses a switchedcapacitor CMOS technology, with an input-referred coarse offset trimming range of approximately ±63mV (9mV steps). An additional output-referred fine offset trim is provided by the Offset DAC (approximately 2.8mV steps). The PGA provides eight gain values from +41V/V to +230V/V. The bridge current source is programmable from 0.1mA to 2mA. The MAX1458 uses four 12-bit DACs and one 3-bit DAC, with calibration coefficients stored by the user in 4 an internal 128-bit EEPROM. This memory contains the following information as 12-bit-wide words: • Configuration register • Offset calibration coefficient • Offset temperature error compensation coefficient • FSO (full-span output) calibration coefficient • FSO temperature error compensation coefficient • 24 user-defined bits for customer programming of manufacturing data (e.g., serial number and date) Figure 1 shows a typical pressure-sensor output and defines the offset, full-scale, and full-span output values as a function of voltage. _______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner FSOTC Compensation Silicon piezoresistive transducers (PRTs) exhibit a large positive input resistance tempco (TCR) so that, while under constant current excitation, the bridge voltage (V BDRIVE ) increases with temperature. This dependence of VBDRIVE on the sensor temperature can be used to compensate the sensor temperature errors. PRTs also have a large negative full-span output sensitivity tempco (TCS) so that, with constant voltage excitation, full-span output (FSO) will decrease with temperature, causing a full-span output temperature coefficient (FSOTC) error. However, if the bridge voltage can be made to increase with temperature at the same rate that TCS decreases with temperature, the FSO will remain constant. FSOTC compensation is accomplished by resistor RFTC and the FSOTC DAC, which modulate the excitation reference current at ISRC as a function of temperature (Figure 3). FSO DAC sets V ISRC and remains constant with temperature while the voltage at FSOTC varies with temperature. FSOTC is the buffered output of the FSOTC DAC. The reference DAC voltage is VBDRIVE, which is temperature dependent. The FSOTC DAC alters the tempco of the current source. When the tempco of the bridge voltage is equal in magnitude and opposite in polarity to the TCS, the FSOTC errors are compensated and FSO will be constant with temperature. VBDRIVE that must be added to the output summing junction to correct the error. Use the Offset TC DAC to adjust the amount of BDRIVE voltage that is added to the output summing junction (Figure 2). MAX1458 Analog Signal Path The fully differential analog signal path consists of four stages: • Front-end summing junction for coarse offset correction • 3-bit PGA with eight selectable gains ranging from 41 through 230 • Three-input-channel summing junction • Differential to single-ended output buffer (Figure 2) OFFSET TC Compensation Compensating offset TC errors involves first measuring the uncompensated offset TC error, then determining the percentage of the temperature-dependent voltage Coarse Offset Correction The sensor output is first fed into a differential summing junction (INM (negative input) and INP (positive input)) with a CMRR > 90dB, an input impedance of approximately 1MΩ, and a common-mode input voltage range from VSS to VDD. At this summing junction, a coarse offset-correction voltage is added, and the resultant voltage is fed into the PGA. The 3-bit (plus sign) input-referred Offset DAC (IRO DAC) generates the coarse offset-correction voltage. The DAC voltage reference is 1.25% of VDD; thus, a VDD of 5V results in a front-end offset-correction voltage ranging from -63mV to +63mV, in 9mV steps (Table 1). To add an offset to the input signal, set the IRO sign bit high; to subtract an offset from the input signal, set the IRO sign bit low. The IRO DAC bits (C2, C1, C0, and IRO sign bit) are programmed in the configuration register (see Internal EEPROM section). 1.25% VDD 4.5 IRO DAC BDRIVE OFFTC DAC A2 A1 A0 SOTC A = 2.3 ± LIMIT OUT VOLTAGE (V) FULL-SPAN OUTPUT (FSO) INP FULL-SCALE (FS) 0.5 OFFSET PMIN PMAX PRESSURE VDD Offset DAC SOFF INM Σ PGA Σ A = 2.3 ± A=1 Figure 1. Typical Pressure-Sensor Output Figure 2. Signal-Path Block Diagram 5 _______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Table 1. Input-Referred Offset DAC Correction Values IRO DAC OFFSET CORRECTION % of VDD (%) +1.25 +1.08 +0.90 +0.72 +0.54 +0.36 +0.18 0 0 -0.18 -0.36 -0.54 -0.72 -0.90 -1.08 -1.25 OFFSET CORRECTION AT VDD = 5V (mV) +63 +54 +45 +36 +27 +18 +9 0 0 -9 -18 -27 -36 -45 -54 -63 Table 2. PGA Gain Settings and IRO DAC Step Size PGA VALUE 0 1 2 3 4 5 6 7 A2 A1 A0 PGA GAIN (V/V) 41 68 95 122 149 176 203 230 OUTPUTREFERRED IRO DAC STEP SIZE (VDD = 5V) (V) 0.369 0.612 0.855 1.098 1.341 1.584 1.827 2.070 VALUE +7 +6 +5 +4 +3 +2 +1 +0 -0 -1 -2 -3 -4 -5 -6 -7 SIGN 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 C2 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 C1 1 1 0 0 1 1 0 0 0 0 1 1 0 0 1 1 C0 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1 0 1 0 1 0 1 0 1 Output Buffer OUT can drive 0.1µF of capacitance. If CS is brought low, OUT becomes high impedance (resulting in typical output impedance of 1MΩ). The output is current limited and can be shorted to either VDD or VSS indefinitely. The maximum output voltage can be limited using the LIMIT pin. Output limiting can be performed for sensor diagnostic purposes. Connect LIMIT to VDD to disable the voltage-limiting feature. Bridge Drive Programmable-Gain Amplifier The programmable-gain amplifier (PGA), which is used to set the coarse FSO, uses a switched-capacitor CMOS technology and contains eight selectable gain levels from 41 to 230, in increments of 27 (Table 2). The output of the PGA is fed to the output summing junction. The three PGA gain bits A2, A1, and A0 are stored in the configuration register. Output Summing Junction The third stage in the analog signal path consists of a summing junction for the PGA output, offset correction, and the offset TC correction. Both the offset and the offset TC correction voltages are gained by a factor of 2.3 before being fed into the summing junction, increasing the offset and offset TC correction range. The offset sign bit and offset TC sign bit are stored in the configuration register. The offset sign bit determines if the offset correction voltage is added to (sign bit is high) or subtracted from (sign bit is low) the PGA output. Negative offset TC errors require a logic high for the offset TC sign bit. Alternately, positive offset TC errors dictate a logic low for the offset TC sign bit. The output of the summing junction is fed to the output buffer. 6 Fine FSO correction is accomplished by varying the sensor excitation current with the 12-bit FSO DAC (Figure 3). Sensor bridge excitation is performed by a programmable current source capable of delivering up to 2mA. The reference current at ISRC is established by resistor RISRC and by the voltage at node ISRC (controlled by the FSO DAC). The reference current flowing through this pin is multiplied by a current mirror (AA ≅ 14) and then made available at BDRIVE for sensor excitation. Modulation of this current with respect to temperature can be used to correct FSOTC errors, while modulation with respect to the output voltage (VOUT) can be used to correct FSO linearity errors. Digital-to-Analog Converters The four 12-bit, sigma-delta DACs typically settle in less than 100ms. The four DACs have a corresponding memory register in EEPROM for storage of correction coefficients. Use the FSO DAC for fine FSO adjustments. The FSO DAC takes its reference from VDD and controls VISRC which, in conjunction with RISRC, sets the baseline sensor excitation current. The Offset DAC also takes its reference from VDD and provides a 1.22mV resolution with _______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 VDD FSO DAC VDD I = IISRC ISRC AA ≈ 14IISRC = IBDRIVE BDRIVE FSOTC DAC FSOTC RFTC RISRC EXTERNAL SENSOR Figure 3. Bridge Excitation Circuit a VDD of 5V. The output of the Offset DAC is fed into the output summing junction where it is gained by approximately 2.3, which increases the resulting output-referred offset correction resolution to 2.8mV. Both the Offset TC and FSOTC DACs take their reference from BDRIVE, a temperature-dependent voltage. A nominal VBDRIVE of 2.5V results in a step size of 0.6mV. The Offset TC DAC output is fed into the output summing junction where it is gained by approximately 2.3, thereby increasing the Offset TC correction range. The buffered FSOTC DAC output is available at FSOTC and is connected to ISRC via RFTC to correct FSOTC errors. Table 3. Configuration Register EEPROM ADDRESS (hex) 00h 01h 02h 03h 04h 05h 06h 07h 08h 09h 0Ah 0Bh DESCRIPTION Offset TC Sign Bit, SOTC Offset Sign Bit, SOFF PGA Gain (MSB), A2 PGA Gain, A1 PGA Gain (LSB), A0 Reserved “0” Reserved “0” Internal Resistor (RFTC and RISRC) Selection Input-Referred Offset (IRO) Sign Bit Input-Referred Offset (MSB) Input-Referred Offset Input-Referred Offset (LSB) Internal Resistors The MAX1458 contains three internal resistors (RISRC, RFTC, and RTEMP) optimized for common silicon PRTs. RISRC (in conjunction with the FSO DAC) programs the nominal sensor excitation current. RFTC (in conjunction with the FSOTC DAC) compensates the FSOTC errors. Both RISRC and RFTC have a nominal value of 75kΩ. If external resistors are used, RISRC and RFTC can be disabled by resetting the appropriate bit (address 07h reset to zero) in the configuration register (Table 3). R TEMP is a high-tempco resistor with a TC of +4600ppm/°C and a nominal resistance of 100kΩ at +25°C. This resistor can be used with certain sensor types that require an external temperature sensor. Internal EEPROM The MAX1458 has a 128-bit internal EEPROM arranged as eight 16-bit words. The four uppermost bits for each register are reserved. The internal EEPROM is used to store the following (also shown in the memory map in Table 4): 7 _______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Table 4. EEPROM Memory Map EE Address Contents EE Address Contents EE Address Contents EE Address Contents EE Address Contents 0F 1 1F 1 2F 1 3F 1 0E 0 1E 0 2E 0 3E 0 0D 0 1D 0 2D 1 3D 1 0C 0 1C 1 2C 0 3C 1 1B MSB 2B MSB 3B MSB 3A 39 38 37 2A 29 28 27 1A 19 18 17 0B 0A 09 08 07 06 05 04 03 02 01 00 Configuration 16 15 14 13 12 11 10 LSB 24 23 22 21 20 LSB 34 33 32 31 30 LSB Offset 26 25 Offset TC 36 35 FSO 4F 1 4E 1 4D 0 4C 0 4B MSB 4A 49 48 47 46 45 44 43 42 41 40 LSB FSOTC 5F 5E 0 5D 0 5C 0 5B 0 5A 0 59 0 58 0 57 0 56 0 55 0 54 0 53 0 52 0 51 0 50 0 Reserved* 0 EE Address Contents EE Address Contents 6F 0 6E 0 6D 0 6C 0 6B 6A 69 68 67 66 65 64 63 62 61 60 User defined bits 7F 0 7E 0 7D 0 7C 0 7B 7A 79 78 77 76 75 74 73 72 71 70 User defined bits = Reserved Bits Note: The MAX1458 processes the Reserved Bits in the EEPROM. If these bits are not properly programmed, the configuration and DAC registers will not be updated correctly. * The contents of the Reserved EE Address 50–5F must all be reset to zero. • Configuration register (Table 3) • 12-bit calibration coefficients for the Offset and FSO DACs • 12-bit compensation coefficients for the Offset TC and FSOTC DACs • Two general-purpose registers available to the user for storing process information such as serial number, batch date, and check sums Program the EEPROM one bit at a time. The bits have addresses from 0 to 127 (7F hex). Configuration Register The configuration register (Table 3) determines the PGA gain, the polarity of the offset and offset TC coefficients, and the coarse offset correction (IRO DAC). It also enables/disables internal resistors (R FTC and RISRC). DAC Registers The Offset, Offset TC, FSO, and FSOTC registers store the coefficients used by their respective calibration/ compensation DACs. 8 _______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner Detailed Description of the Digital Lines Chip-Select (CS) and Write-Enable (WE) CS is used to enable OUT, control serial communication, and force an update of the configuration and DAC registers. • A low on CS disables serial communication and places OUT in a high-impedance state. • A transition from low to high on CS forces an update of the configuration and DAC registers from the EEPROM when the “U” bit is zero. • A transition from high to low on CS terminates programming mode. • A logic high on CS enables OUT and serial communication (see Communication Protocol section). WE controls the refresh rate for the internal configuration and DAC registers from the EEPROM and enables the erase/write operations. If communication has been initiated (see Communication Protocol section), internal register refresh is disabled. • A low on WE disables the erase/write operations and also disables register refreshing from the EEPROM. • A high on WE selects a refresh rate of approximately 400 times per second and enables EEPROM erase/write operations. • It is recommended that WE be connected to V SS after the MAX1458 EEPROM has been programmed. Data Input/Output (DIO) The DIO line is an input/output pin used to issue commands to the MAX1458 (input mode) or read the EEPROM contents (output mode). In input mode (the default mode), data on DIO is latched on each rising edge of SCLK. Therefore, data on DIO must be stable at the rising edge of SCLK and should transition on the falling edge of SCLK. DIO will switch to output mode after receiving a “READ EEPROM” command, and will return the data bit addressed by the digital value in the “READ EEPROM” command. After a low-to-high transition or CS, DIO returns to input mode and is ready to accept more commands. MAX1458 Communication Protocol To initiate communication, the first six bits on DIO after CS transitions from low to high m ust be 1010U0 (defined as the INIT SEQUENCE). The MAX1458 will then begin accepting 16-bit control words (Figure 4). If the INIT SEQUENCE is not detected, all subsequent data on DIO is ignored until CS again transitions from low to high and the correct INIT SEQUENCE is received. The “U” bit of the INIT SEQUENCE controls the updating of the DACs and configuration register from the internal EEPROM. If this bit is low (U = 0), all four internal DACs and the configuration register will be updated from the EEPROM on the next rising edge of CS (this is also the default on power-up). If the “U” bit is high, the DACs and configuration register will not be updated from the internal EEPROM; they will retain their current value on any subsequent CS rising edge. The MAX1458 continues to accept control words until CS is brought low. SCLK (Serial Clock) SCLK must be driven externally and is used to input commands to the MAX1458 and read EEPROM contents. Input data on DIO is latched on the rising edge of SCLK. Noise on SCLK may disrupt communication. In noisy environments, place a capacitor (0.01µF) between SCLK and VSS. CS tMIN = 200µs SCLK 16 CLK CYCLES 16 CLK CYCLES n x 16 CLK CYCLES DIO X 1 0 1 0 U 0 D0 D1 CM3 D0 D1 CM3 CONTROL WORDS BEGIN PROGRAMMING SEQUENCE CONTROL WORD CONTROL WORD Figure 4. Communication Sequence _______________________________________________________________________________________ 9 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Control Words After receiving the INIT SEQUENCE on DIO, the MAX1458 begins latching in 16-bit control words, LSB first (Figure 5). The first 12 bits (D0–D11) represent the data field. The last four bits of the control word (the MSBs, CM0–CM3) are the command field. The MAX1458 supports the commands listed in Table 5. Table 5. MAX1458 Commands FUNCTION ERASE EEPROM BEGIN EEPROM WRITE at Address READ EEPROM at Address Maxim Reserved END EEPROM WRITE at Address WRITE Data to Configuration Register WRITE Offset DAC WRITE Offset TC DAC WRITE FSO DAC WRITE FSOTC DAC No Operation HEX CM3 CM2 CM1 CM0 CODE 1h 2h 3h 4h 5h 8h 9h Ah Bh Ch 0h 6h, 7h, Dh, Eh, Fh 0 0 0 0 0 1 1 1 1 1 0 0 0 1 1 1 0 0 0 1 1 0 0 0 0 1 0 1 1 1 1 1 0 1 1 0 0 0 0 1 1 0 0 1 1 0 1 1 1 0 1 0 1 0 1 0 1 0 0 0 1 1 0 1 ERASE EEPROM Command When an ERASE EEPROM command is issued, all of the memory locations in the EEPROM are reset to a logic “0.” The data field of the 16-bit word is ignored. Important: An internal charge pump develops voltages greater than 20V for EEPROM programming operations. The EEPROM control logic requires 50ms to erase the EEPROM. After sending a WRITE or ERASE command, failure to wait 50ms before issuing another command may result in data being accidentally written to the EEPROM. The maximum number of ERASE EEPROM cycles should not exceed 100. BEGIN EEPROM WRITE Command The BEGIN EEPROM WRITE command stores a logic high at the memory location specified by the lower seven bits of the data field (A0–A6). The higher bits of the data field (A7–A11) are ignored (Figure 6). Note that to write to the internal EEPROM, WE and CS must be Load Register SCLK DATA LSB DIO D0 D1 D2 D3 D4 D5 D6 D7 D8 D9 MSB LSB COMMAND MSB D10 D11 CM0 CM2 CM2 CM3 16-BIT CONFIGURATION WORD LSB MSB LSB MSB Figure 5. Control-Word Timing Diagram CS WE tMIN = 200µs SCLK 16 CLK CYCLES 16 CLK CYCLES n x 16 CLK CYCLES DIO X 1 0 1 0 U 0 A0 A1 CM3 BEGIN EEPROM WRITE TWRITE A0 A1 CM3 END EEPROM WRITE tWAIT D0 D1 CM3 INIT SEQUENCE n COMMAND WORDS Figure 6. Timing Diagram for WRITE EEPROM Operation 10 ______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner high. In addition, the EEPROM should only be written to at TA = +25°C and VDD = 5V. Writing to the internal EEPROM is a time-consuming process and should only be required once. All calibration/compensation coefficients are determined by writing directly to the DAC and configuration registers. Use the following procedure to write these calibration/compensation coefficients to the EEPROM: 1) Issue an ERASE EEPROM command. 2) Wait 50ms (tWRITE). 3) Issue on END EEPROM WRITE command at address 00h. 4) Wait 1ms (tWAIT). 5) Issue a BEGIN EEPROM WRITE command (Figure 7) at the address of the bit to be set. 6) Wait 50ms. 7) Issue an END EEPROM WRITE command (Figure 7) using the same address as in Step 5. 8) 9) Wait 1ms. Return to Step 5 until all necessary bits have been set. 10) Read EEPROM to verify that the correct calibration/compensation coefficients have been stored. MAX1458 READ EEPROM Command The READ EEPROM command returns the bit stored at the memory location addressed by the lower seven bits of the data field (A0–A6). The higher bits of the data field (A7–A11) are ignored. Note that after a read command has been issued, the DIO lines become an output and the state of the addressed EEPROM location will be available on DIO 200µs (tREAD) after the falling edge of the 16th SCLK cycle (Figure 8). After issuing the READ EEPROM command, DIO returns to input mode on the falling edge of CS. Reading the entire EEPROM requires the READ EEPROM command be issued 128 times. SCLK DATA LSB DIO A0 LSB SCLK DATA LSB DIO A0 LSB A1 A2 A3 A4 A5 A6 0 0 0 0 MSB LSB 0 1 0 1 COMMAND MSB 0 MSB A1 A2 A3 A4 A5 A6 0 0 0 0 MSB LSB 0 0 1 0 COMMAND MSB 0 MSB 16-BIT COMMAND WORD – BEGIN EEPROM WRITE AT ADDRESS COMMAND 16-BIT COMMAND WORD – END EEPROM WRITE AT ADDRESS COMMAND Figure 7. Begin WRITE EEPROM and End WRITE EEPROM Timing Diagrams CS tMIN = 200µs 16 CLOCK CYCLES SCLK tREAD DIO X 1 0 1 0 U 0 A0 A1 A2 A3 A4 A5 A6 0 0 0 0 0 1 1 00 X EE DATA X INIT SEQUENCE READ EEPROM AT ADDRESS COMMAND DIO IS AN INPUT PIN DIO IS AN OUTPUT PIN Figure 8. READ EEPROM Timing Diagram ______________________________________________________________________________________ 11 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Writing to the Configuration and DAC Registers When writing to the configuration register or directly to the internal 12-bit DACs, the data field (D0–D11) contains the data to be written to the respective register. Note that all four DACs and the configuration register can be updated without toggling the CS line. Every register write command must be followed by a LOAD REGISTER command. Coefficient Initialization Select the resistor values and the PGA gain to prevent overload of the PGA and bridge current source. These values depend on sensor behavior and require some sensor characterization data, which may be available from the sensor manufacturer. If not, the data can be generated by performing a two-temperature, two-pressure sensor evaluation. The required sensor information is shown in Table 6 and can be used to obtain the values for the parameters listed in Table 7. __________Applications Information Power-Up At power up, the following occurs: 1) The DAC and configuration registers are reset to zero. 2) CS transitions from low to high after power-up (an internal pull-up resistor ensures that this happens if CS is left unconnected), and the EEPROM contents are read and processed. 3) The DAC and configuration registers are updated either once or approximately 400 times per second (as determined by the state of WE). 4) The MAX1458 begins accepting commands in a serial format on DIO immediately after receiving the INIT SEQUENCE. The MAX1458 is shipped with all memory locations in the internal EEPROM uninitialized. Therefore, the MAX1458 must be programmed for proper operation. Table 6. Sensor Information for Typical PRT PARAMETER Rb(T) TCR S(T) TCS O(T) OTC SENSOR DESCRIPTION Bridge Impedance Bridge Impedance Tempco Sensitivity Sensitivity Tempco Offset Offset Tempco Sensitivity Linearity Error as % FSO, BSLF (Best Straight-Line Fit) Minimum Input Pressure Maximum Input Pressure TYPICAL VALUES 5kΩ at +25°C 2600ppm/°C 1.5mV/V per PSI at +25°C -2100ppm/°C 12mV/V at +25°C -1000ppm/°C of FSO 0.1% FSO, BSLF 0 PSI 10 PSI S(p) PMIN PMAX Compensation Procedure The following compensation procedure was used to obtain the results shown in Figure 9 and Table 8. It assumes a pressure transducer with a +5V supply and an output voltage that is ratiometric to the supply voltage. The desired offset voltage (VOUT at PMIN) is 0.5V, and the desired FSO voltage (VOUT(PMAX) - VOUT(PMIN)) is 4V; thus the full-scale output voltage (VOUT at PMAX) will be 4.5V (refer to Figure 1). The procedure requires a minimum of two test pressures (e.g., zero and full scale) at two arbitrary test temperatures, T1 and T2. Ideally, T1 and T2 are the two points where we wish to perform best linear fit compensation. The following outlines a typical compensation procedure: 1) Perform Coefficient Initialization 2) 3) 4) 5) Perform FSO Calibration Perform FSOTC Compensation Perform Offset TC Compensation Perform Offset Calibration Selecting RISRC When using an external resistor, use the equation below to determine the value of RISRC, and place the resistor between ISRC and VSS. Since the 12-bit FSO DAC provides considerable dynamic range, the RISRC value need not be exact. Generally any resistor value within ±50% of the calculated value is acceptable. If both the internal resistors RISRC and RFTC are used, set the IRS bit at EEPROM address bit 7 high. Otherwise, set IRS low and connect external resistors as shown in Figure 10. RISRC ≈ 14 x Rb(T1) ≈ 14 x 5kΩ = 70kΩ where Rb(T) is the sensor input impedance at temperature T1 (+25°C in this example). 12 ______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner Table 7. Compensation Components and Values PARAMETER RISRC DESCRIPTION Internal (approximately 75kΩ) or usersupplied resistor that programs the nominal sensor excitation current. Internal (approximately 75kΩ) or usersupplied resistor that compensates FSO TC errors. Programmable-gain amplifier gain Input-referred offset correction DAC value Input-referred offset sign bit Internal resistor selection bit Offset correction DAC coefficient Offset sign bit Offset TC compensation DAC coefficient Offset TC sign bit FSO trim DAC coefficient FSO TC compensation DAC coefficient where S is the sensor sensitivity at T1, VBDRIVE is the sensor excitation voltage (initially 2.5V), and ∆ is the P maximum pressure differential. Then calculate the ideal gain using the following formula, and select the nearest gain setting from Table 2: APGA = OUTFSO SensorFSO 4V = = 106V/V 0.0375V MAX1458 RFTC APGA IRO IRO Sign IRS OFF COEF OFF Sign OFFTC COEF OFFTC Sign FSO COEF FSOTC COEF where OUTFSO is the desired calibrated transducer full-span output voltage, and SensorFSO is the sensor full-span output voltage at T1. In this example, a PGA value of 2 (gain of +95V/V) is the best selection. Determining Input-Referred OFFSET (IRO) The input-referred offset register is used to null any front-end sensor offset errors prior to amplification by the PGA. This reduces the possibility of saturating the PGA and maximizes the useful dynamic range of the PGA (particularly at the higher gain values.) First, calculate the ideal IRO correction voltage using the following formula, and select the nearest setting from Table 1: IROideal = - O( T1) x VBDRIVE ( T1) = - (0.012V/V) x 2.5V = - 30mV where IROideal is the exact voltage required to perfectly null the sensor, O(T1) is the sensor offset voltage in V/V at +25°C, and VBDRIVE(T1) is the nominal sensor excitation voltage at +25°C. In this example, 30mV must be subtracted from the amplifier front end to null the sensor perfectly. From Table 1, select an IRO value of 3 to set the IRO DAC to 27mV, which is nearest the ideal value. To subtract this value, set the IRO sign bit to 0. The residual output-referred offset error will be corrected later with the Offset DAC. Selecting RFTC When using an external resistor, use the equation below to determine the value for RFTC, and place the resistor between ISRC and FSOTC. Since the 12-bit FSOTC DAC provides considerable dynamic range, the RFTC value need not be exact. Generally, any resistor value within ±50% of the calculated value is acceptable. RFTC ≅ RISRC x 500ppm/ °C TCR - | TCS | 70kΩ x 500ppm/ °C ≅ = 70kΩ 2600ppm/ ° C - | -2100ppm/ ° C | [ ] This approximation works best for bulk, micromachined, silicon PRTs. Negative values for RFTC indicate unconventional sensor behavior that cannot be compensated by the MAX1458 without additional external circuitry. Selecting the PGA Gain Setting To select the PGA gain setting, first calculate SensorFSO, the sensor full-span output voltage at T1: SensorFSO = S x VBDRIVE x ∆ P = 1.5mV/V per PSI x 2.5V x 10 PSI = 0.0375V Determining OFFTC COEF Initial Value Generally, OFFTC COEF can initially be set to 0, since the offset TC error will be compensated in a later step. However, sensors with large offset TC errors may require an initial coarse offset TC adjustment to prevent the PGA from saturating during the compensation procedure as temperature is increased. An initial coarse offset TC adjustment is required for sensors with an offset TC greater than about 10% of the FSO. If an initial ______________________________________________________________________________________ 13 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 coarse offset TC adjustment is required, use the following equation: OFFTC COEF = ≅ = 4096 x ∆VOUT T Three-Step FSOTC Compensation Step 1 Use the following procedure to determine FSOTC COEF. Four variables, A–D, will be used. 1) Name the existing FSO DAC coefficient “A”. 2) Change FSOTC DAC to 3000. 3) Adjust FSO DAC until V BDRIVE ( T1) is equal to VBIDEAL(T1). 4) Name the existing FSO DAC coefficient “B”. 5) Readjust the offset voltage (by adjusting the Offset DAC), if required, to 0.5V. At this point, it is important that no other changes be made to the Offset or Offset TC DACs until the Offset TC Compensation step has been completed. Step 2 To complete linear FSOTC compensation, take data measurements at a second temperature, T2 (T2 > T1). Perform the following steps: 1) Measure the full-span output (measuredVFSO(T2). 2) Calculate VBIDEAL(T2) using the following equation: VBIDEAL ( T2) = VBDRIVE x ∆VBDRIVE T x 2.3 4096 x OTC x FSO x ∆T TCS x VBDRIVE x 2.3 x ∆T 4096 x -1000ppm/ °C x 4V ( ( () () ) ) -2100ppm/ °C x 2.5V x 2.3 = 1357 where OTC is the sensor offset TC error as a ppm/°C of OUTFSO (Table 6), ∆ is the operating temperature T range in °C, and OFFTC COEF is the numerical decimal value to be loaded into the DAC. For positive values, set the OFFTC sign bit high; for negative values, set the OFFTC sign bit low. If the absolute value of the OFFTC COEF is larger than 4096, the sensor has a very large offset TC error, which the MAX1458 is unable to completely correct. FSO Calibration Perform FSO calibration at room temperature with a fullscale sensor excitation. 1) Set FSOTC COEF to 1000. 2) At T1, adjust FSO DAC until VBDRIVE is about 2.5V. 3) Adjust Offset DAC (and OFFSET sign bit, if needed) until the T1 offset voltage is 0.5V (see O FFSET Calibration section). 4) Measure the full-span output (measuredVFSO). 5) Calculate the ideal bridge voltage, V BIDEAL (T1), using the following equation: VBIDEAL ( T1) = VBDRIVE x  desiredVFSO - measuredVFSO ( T1) 1 + measuredVFSO ( T1)    desiredVFSO - measuredVFSO ( T2)  1 +  measuredVFSO ( T2)   3) Set VBIDEAL(T2) by adjusting the FSO DAC. 4) Name the current FSO DAC coefficient “D”. 5) Change FSOTC DAC to 1000. 6) Adjust FSO DAC until V BDRIVE is equal to VBIDEAL(T2). 7) Name the FSO DAC coefficient “C”.     Step 3 Insert the data previously obtained from Steps 1 and 2 into the following equation to compute FSOTC COEF: FSOTC COEF = 1000 B - D + 3000 C - A + C- A Note: If VBIDEAL(T1) is outside the allowable bridge voltage swing of (VSS + 1.3V) to (VDD - 1.3V), readjust the PGA gain setting. If V BIDEAL (T1) is too low, decrease the PGA gain setting by one step and return to Step 2. If VBIDEAL(T1) is too high, increase the PGA gain setting by one step and return to Step 2. 6) Set VBIDEAL(T1) by adjusting the FSO DAC. 7) Readjust Offset DAC until the offset voltage is 0.5V (see OFFSET Calibration section). () (B - D) ( ( ) ) 1) Load this FSOTC COEF value into the FSOTC DAC. 2) Adjust the FSO DAC until VBDRIVE(T2) is equal to VBIDEAL(T2). This completes both FSO calibration and FSO TC compensation. 14 ______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner Offset TC Compensation The offset voltage at T1 was previously set to 0.5V; therefore, any variation from this voltage at T2 is an offset TC error. Perform the following steps: 1) Measure the offset voltage at T2. 2) Use the following equation to compute the correction required: NewOFFTC COEF = CurrentOFFTC COEF +  4096 VOFFSET ( T1) - VOFFSET ( T2)   2.3 VBDRIVE ( T1) - VBDRIVE ( T2)  4) If NewOFFTC COEF is negative, set the SOTC bit low; otherwise, set it high. Offset TC Compensation is now complete. MAX1458 OFFSET Calibration At this point the sensor should still be at temperature T2. The final offset adjustment can be made at T2 or T1 by adjusting the Offset DAC (and optionally the offset sign bit, SOFF) until the output (VOUT(PMIN)) reads 0.5V at zero input pressure. Use the following procedure: 1) Set Offset DAC to zero (Offset COEF = 0). 2) Measure the voltage at OUT. 3) If VOUT is greater than the desired offset voltage (0.5V in this example), set SOFF low; otherwise set it high. 4) Increase Offset COEF until VOUT equals the desired offset voltage. Offset calibration is now complete. Table 8 and Figure 9 compare an uncompensated input to a typical compensated transducer output. [ [ ] ]    Note: CurrentOFFTC COEF is the current value stored in the Offset TC DAC. If the Offset TC sign bit (SOTC) is low, this number is negative. 3) Load this value into the Offset TC DAC. Table 8. MAX1458 Calibration and Compensation Typical Uncompensated Input (Sensor) Offset ..........................................................................±80% FSO FSO ..................................................................................15mV/V Offset TC ......................................................................-17% FSO Offset TC Nonlinearity ..................................................0.7% FSO FSO TC.........................................................................-35% FSO FSO TC Nonlinearity.....................................................0.5% FSO Temperature Range...........................................-40°C to +125°C Typical Compensated Transducer Output VOUT ...................................................Ratiometric to VDD at 5.0V Offset at +25°C ......................................................0.500V ±5mV FSO at +25°C .........................................................4.000V ±5mV Offset Accuracy Over Temp. Range ..........±28mV (±0.7% FSO) FSO Accuracy Over Temp. Range .............±20mV (±0.5% FSO) UNCOMPENSATED SENSOR ERROR 30 0.8 0.6 20 ERROR (% SPAN) ERROR (% FSO) 0.4 COMPENSATION TRANSDUCER ERROR FSO 0.2 0 -0.2 -0.4 OFFSET 10 FSO OFFSET 0 -10 -0.6 -20 -50 0 50 TEMPERATURE (°C) 100 150 -0.8 -50 0 50 TEMPERATURE °(C) 100 150 Figure 9. Comparison of an Uncalibrated Sensor and a Temperature-Compensated Transducer ______________________________________________________________________________________ 15 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Ratiometric Output Configuration Ratiometric output configuration provides an output that is proportional to the power-supply voltage. When used with ratiometric A/D converters, this output provides digital pressure values independent of supply voltage. Most automotive and some industrial applications require ratiometric outputs. The MAX1458 provides a high-performance ratiometric output with a minimum number of external components (Figure 10). These external components include the following: • One power-supply bypass capacitor (C1) • Two optional resistors, one from FSOTC to ISRC, and another from ISRC to VSS, depending on the sensor type • One optional capacitor C2 from BDRIVE to VSS Test System Configuration The MAX1458 is designed to support an automated production pressure-temperature test system with integrated calibration and temperature compensation. Figure 11 shows the implementation concept for a lowcost test system capable of testing up to 12 transducer modules connected in parallel. Three-state outputs on the MAX1458 allow for parallel connection of transducers. The test system shown in Figure 11 includes a dedicated test bus consisting of five wires: • Two power-supply lines • One analog output voltage line from the transducers to a system digital voltmeter • Two serial-interface lines: DIO (input/output) and SCLK (clock) For simultaneous testing of more than 12 sensor modules, use buffers to prevent overloading the data bus. A digital multiplexer controls the chip-select signal for each transducer. +5V VDD BDRIVE C2 0.1µF OFFSET (IRODAC) C1 0.1µF LIMIT MAX1458 INP INM Σ PGA OUT SENSOR ISRC VDD RISRC RFTC RISRC VSS RFTC CONFIGURATION REGISTER 12-BIT D/A - OFFSET 12-BIT D/A - OFFSET TC 12-BIT D/A - FSO 12-BIT D/A - FSOTC A=1 FSOTC 128-BIT EEPROM CS WE SCLK DIO DIGITAL INTERFACE TEMP TEMP VSS Figure 10. Basic Ratiometric Output Configuration 16 ______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 DIGITAL MULTIPLEXER CS[1:N] CS1 CS2 CSN MODULE 1 CS BDRIVE INP INM MODULE 2 CS BDRIVE INP INM MODULE N CS BDRIVE INP INM MAX1458 MAX1458 SCLK DIO +5V VDD OUT VSS VDD SCLK DIO OUT VSS VDD SCLK DIO OUT VSS VOUT DVM SCLK DIO TEST OVEN Figure 11. Automated Test System Concept MAX1458 Evaluation _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ Development Kit __________________ To expedite the development of MAX1458 based transducers and test systems, Maxim has produced the MAX1458 evaluation kit (EV kit). First-time users of the MAX1458 are strongly encouraged to use this kit. The MAX1458 EV kit is designed to facilitate manual programming of the MAX1458 and includes the following: 1) Evaluation Board with a silicon pressure sensor. 2) Design/Applications Manual, which describes in detail the architecture and functionality of the MAX1458. This manual was developed for test engineers familiar with data acquisition of sensor data and provides sensor compensation algorithms and test procedures. 3) MAX1458 Communication Software, which enables programming of the MAX1458 from a computer (IBM compatible), one module at a time. 4) Interface Adapter and Cable, which allow the connection of the evaluation board to a PC parallel port. ______________________________________________________________________________________ MAX1458 17 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 Functional Diagram VDD BDRIVE OFFSET (IRODAC) Chip Information TRANSISTOR COUNT: 7772 SUBSTRATE CONNECTED TO VSS MAX1458 LIMIT PGA OUT INP INM Σ ISRC VDD RFTC RISRC VSS 128-BIT EEPROM CONFIGURATION REGISTER 12-BIT D/A - OFFSET 12-BIT D/A - OFFSET TC 12-BIT D/A - FSO 12-BIT D/A - FSOTC A=1 FSOTC CS WE SCLK DIO DIGITAL INTERFACE TEMP TEMP VSS 18 ______________________________________________________________________________________ 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner Package Information SSOP.EPS MAX1458 ______________________________________________________________________________________ 19 1%-Accurate, Digitally Trimmed Sensor Signal Conditioner MAX1458 NOTES 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. 20 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600 © 1998 Maxim Integrated Products Printed USA is a registered trademark of Maxim Integrated Products.