AD590 2-Wire, Current Output Temperature Transducer August 1997 Features Description • Linear Current Output . . . . . . . . . . . . . . . . . . . . 1µA/oK The AD590 is an integrated-circuit temperature transducer which produces an output current proportional to absolute temperature. The device acts as a high impedance constant current regulator, passing 1µA/oK for supply voltages between +4V and +30V. Laser trimming of the chip's thin film resistors is used to calibrate the device to 298.2µA output at 298.2oK (25oC). • Wide Temperature Range . . . . . . . . . . . -55oC to 150oC • Two-Terminal Device Voltage In/Current Out • Wide Power Supply Range . . . . . . . . . . . . . +4V to +30V • Sensor Isolation From Case • Low Cost Ordering Information NONPART LINEARITY TEMP. RANGE NUMBER (oC) (oC) PACKAGE PKG. NO. AD590IH ±3.0 -55 to 150 3 Ld Metal Can (TO-52) T3.A AD590JH ±1.5 -55 to 150 3 Ld Metal Can (TO-52) T3.A The AD590 should be used in any temperature-sensing application between -55oC to 150oC in which conventional electrical temperature sensors are currently employed. The inherent low cost of a monolithic integrated circuit combined with the elimination of support circuitry makes the AD590 an attractive alternative for many temperature measurement situations. Linearization circuitry, precision voltage amplifiers, resistance measuring circuitry and cold junction compensation are not needed in applying the AD590. In the simplest application, a resistor, a power source and any voltmeter can be used to measure temperature. In addition to temperature measurement, applications include temperature compensation or correction of discrete components, and biasing proportional to absolute temperature. The AD590 is particularly useful in remote sensing applications. The device is insensitive to voltage drops over long lines due to its high-impedance current output. Any well insulated twisted pair is sufficient for operation hundreds of feet from the receiving circuitry. The output characteristics also make the AD590 easy to multiplex: the current can be switched by a CMOS multiplexer or the supply voltage can be switched by a logic gate output. Pinout Functional Diagram + AD590 (METAL CAN) R1 260Ω + Q1 1 3 Q5 Q2 Q3 Q4 Q6 CASE Q7 - R2 1040Ω Q12 Q8 C1 26pF 2 CHIP SUBSTRATE R3 5kΩ R4 11kΩ Q10 Q9 8 1 R6 820Ω Q11 1 R5 146Ω - CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures. http://www.intersil.com or 407-727-9207 | Copyright © Intersil Corporation 1999 12-3 File Number 3171.1 AD590 Absolute Maximum Ratings TA = 25oC Thermal Information Supply Forward Voltage (V+ to V-) . . . . . . . . . . . . . . . . . . . . . . +44V Supply Reverse Voltage (V+ to V-) . . . . . . . . . . . . . . . . . . . . . . .-20V Breakdown Voltage (Case to V+ to V-) . . . . . . . . . . . . . . . . . . ±200V Rated Performance Temperature Range TO-52. . . . -55oC to 150oC Operating Conditions Thermal Resistance (Typical, Note 1) θJA (oC/W) θJC (oC/W) Metal Can Package . . . . . . . . . . . . . . . 200 120 Maximum Junction Temperature (Metal Can Package) . . . . . . . 175oC Maximum Storage Temperature Range . . . . . . . . . .-65oC to 150oC Maximum Lead Temperature (Soldering 10s) . . . . . . . . . . . . . 300oC Temperature Range . . . . . . . . . . . . . . . . . . . . . . . . . -55oC to 150oC CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied. NOTE: 1. θJA is measured with the component mounted on an evaluation PC board in free air. Electrical Specifications Typical Values at TA = 25οC, V+ = 5V, Unless Otherwise Specified PARAMETER TEST CONDITIONS AD590I AD590J UNITS 298.2 298.2 µA 1.0 1.0 µA/oK ±10.0 Max ±5.0 Max oC Without External Calibration Adjustment ±20.0 Max ±10.0 Max oC With External Calibration Adjustment ±5.8 Max ±3.0 Max oC Nominal Output Current at 2oC (298.2oK) Nominal Temperature Coefficient Calibration Error at 25oC Notes 1, 5 Absolute Error -55oC to 150oC, Note 7 Non-Linearity Note 6 ±3.0 Max ±1.5 Max oC Repeatability Notes 2, 6 ±0.1 Max ±0.1 Max oC Long Term Drift Notes 3, 6 ±0.1 Max ±0.1 Max oC/Month 40 40 pA/√Hz +4V < V+ < +5V 0.5 0.5 µA/V +5V < V+ < +15V 0.2 0.2 µA/V +15V < V+ < +30V 0.1 0.1 µA/V Case Isolation to Either Lead 1010 1010 Ω Effective Shunt Capacitance 100 100 pF Current Noise Power Supply Rejection Electrical Turn-On Time Note 1 20 20 µs Reverse Bias Leakage Current Note 4 10 10 pA +4 to +30 +4 to +30 V Power Supply Range NOTES: 2. Does not include self heating effects. 3. Maximum deviation between 25oC reading after temperature cycling between -55oC and 150oC. 4. Conditions constant +5V, constant 125oC. 5. Leakage current doubles every 10oC. 6. Mechanical strain on package may disturb calibration of device. 7. Guaranteed but not tested. 8. -55oC Guaranteed by testing at 25oC and 150oC. 12-4 AD590 Trimming Out Errors +10V The ideal graph of current versus temperature for the AD590 is a straight line, but as Figure 1 shows, the actual shape is slightly different. Since the sensor is limited to the range of -55oC to 150oC, it is possible to optimize the accuracy by trimming. Trimming also permits extracting maximum performance from the lower-cost sensors. The circuit of Figure 2 trims the slope of the AD590 output. The effect of this is shown in Figure 3. The circuit of Figure 4 trims both the slope and the offset. This is shown in Figure 5. The diagrams are exaggerated to show effects, but it should be clear that these trims can be used to minimize errors over the whole range, or over any selected part of the range. In fact, it is possible to adjust the I-grade device to give less than 0.1oC error over the range 0oC to 90oC and less than 0.05oC error from 25oC to 60oC. 35.7kΩ 97.6kΩ R2 5kΩ R1 2kΩ + VOUT = 100mV/ oC AD590 V- R1 = OFFSET R2 = SLOPE FIGURE 4. SLOPE AND OFFSET TRIMMING IDEAL I (µA) ACTUAL (GREATLY EXAGGERATED) FIGURE 5A. UNTRIMMED T (oK) FIGURE 1. TRIMMING OUT ERRORS +5V + + AD590 + R 100Ω 950Ω VOUT = 1mV/ oK R = SLOPE FIGURE 5B. TRIM ONE: OFFSET FIGURE 2. SLOPE TRIMMING IDEAL ACTUAL I (µA) TRIMMED FIGURE 5C. TRIM TWO: SLOPE T (oK) FIGURE 3. EFFECT OF SLOPE TRIM 12-5 AD590 Accuracy Maximum errors over limited temperature spans, with VS = +5V, are listed by device grade in the following tables. The tables reflect the worst-case linearities, which invariably occur at the extremities of the specified temperature range. The trimming conditions for the data in the tables are shown in Figure 2 and Figure 4. FIGURE 5D. TRIM THREE: OFFSET AGAIN FIGURE 5. EFFECT OF SLOPE AND OFFSET TRIMMING All errors listed in the tables are ±oC. For example, if ±1oC maximum error is required over the 25oC to 75oC range (i.e., lowest temperature of 25oC and span of 50oC), then the trimming of a J-grade device, using the single-trim circuit (Figure 2), will result in output having the required accuracy over the stated range. An I-grade device with two trims (Figure 4) will have less than ±0.2oC error. If the requirement is for less than ±1.4oC maximum error, from -25oC to 75oC (100oC span from -25oC), it can be satisfied by an I-grade device with two trims. I Grade Maximum Errors (oC) LOWEST TEMPERATURE IN SPAN (oC) NUMBER OF TRIMS TEMPERATURE SPAN (oC) -55 -25 0 25 50 75 100 125 None 10 8.4 9.2 10.0 10.8 11.6 12.4 13.2 14.4 None 25 10.0 10.4 11.0 11.8 12.0 13.8 15.0 16.0 None 50 13.0 13.0 12.8 13.8 14.6 16.4 18.0 - None 100 15.2 16.0 16.6 17.4 18.8 - - - None 150 18.4 19.0 19.2 - - - - - None 205 20.0 - - - - - - - One 10 0.6 0.4 0.4 0.4 0.4 0.4 0.4 0.6 One 25 1.8 1.2 1.0 1.0 1.0 1.2 1.6 1.8 One 50 3.8 3.0 2.0 2.0 2.0 3.0 3.8 - One 100 4.8 4.5 4.2 4.2 5.0 - - - One 150 5.5 4.8 5.5 - - - - - One 205 5.8 - - - - - - - Two 10 0.3 0.2 0.1 (Note 9) (Note 9) 0.1 0.2 0.3 Two 25 0.5 0.3 0.2 (Note 9) 0.1 0.2 0.3 0.5 Two 50 1.2 0.6 0.4 0.2 0.2 0.3 0.7 - Two 100 1.8 1.4 1.0 2.0 2.5 - - - Two 150 2.6 2.0 2.8 - - - - - Two 205 3.0 - - - - - - - NOTE: 9. Less than ±0.05oC. 12-6 AD590 J Grade Maximum Errors (oC) NUMBER OF TRIMS LOWEST TEMPERATURE IN SPAN (oC) TEMPERATURE SPAN (oC) -55 -25 0 25 50 75 100 125 None 10 4.2 4.6 5.0 5.4 5.8 6.2 6.6 7.2 None 25 5.0 5.2 5.5 5.9 6.0 6.9 7.5 8.0 None 50 6.5 6.5 6.4 6.9 7.3 8.2 9.0 - None 100 7.7 8.0 8.3 8.7 9.4 - - - None 150 9.2 9.5 9.6 - - - - - None 205 10.0 - - - - - - - One 10 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.3 One 25 0.9 0.6 0.5 0.5 0.5 0.6 0.8 0.9 One 50 1.9 1.5 1.0 1.0 1.0 1.5 1.9 - One 100 2.3 2.2 2.0 2.0 2.3 - - - One 150 2.5 2.4 2.5 - - - - - One 205 3.0 - - - - - - - Two 10 0.1 (Note 10) (Note 10) (Note 10) (Note 10) (Note 10) (Note 10) 0.1 Two 25 0.2 0.1 (Note 10) (Note 10) (Note 10) (Note 10) 0.1 0.2 Two 50 0.4 0.2 0.1 (Note 10) (Note 10) 0.1 0.2 (Note 10) Two 100 0.7 0.5 0.3 0.7 1.0 - - - Two 150 1.0 0.7 1.2 - - - - - Two 205 1.6 - - - - - - - NOTE: 10. Less than ±0.05oC. NOTES 1. Maximum errors over all ranges are guaranteed based on the known behavior characteristic of the AD590. 2. For one-trim accuracy specifications, the 205oC span is assumed to be trimmed at 25oC; for all other spans, it is assumed that the device is trimmed at the midpoint. 3. For the 205oC span, it is assumed that the two-trim temperatures are in the vicinity of 0oC and 140oC; for all other spans, the specified trims are at the endpoints. 4. In precision applications, the actual errors encountered are usually dependent upon sources of error which are often overlooked in error budgets. These typically include: a. Trim error in the calibration technique used b. Repeatability error c. Long term drift errors Trim Error is usually the largest error source. This error arises from such causes as poor thermal coupling between the device to be calibrated and the reference sensor; reference sensor errors; lack of adequate time for the device being calibrated to settle to the final temperature; radically different thermal resistances between the case and the surroundings (RθCA) when trimming and when applying the device. Repeatability Errors arise from a strain hysteresis of the package. The magnitude of this error is solely a function of the magnitude of the temperature span over which the device is used. For example, thermal shocks between 0oC and 100oC involve extremely low hysteresis and result in repeatability errors of less than ±0.05oC. When the thermalshock excursion is widened to -55oC to 150oC, the device will typIcally exhibit a repeatability error of ±0.05oC (±0.10 guaranteed maximum). Long Term Drift Errors are related to the average operating temperature and the magnitude of the thermal-shocks experienced by the device. Extended use of the AD590 at temperatures above 100oC typically results in long-term drift of ±0.03oC per month; the guaranteed maximum is ±0.10oC per month. Continuous operation at temperatures below 100oC induces no measurable drifts in the device. Besides the effects of operating temperature, the severity of thermal shocks incurred will also affect absolute stability. For thermal-shock excursions less than 100oC, the drift is difficult to measure (<0.03oC). However, for 200oC excursions, the device may drift by as much as ±0.10oC after twenty such shocks. If severe, quick shocks are necessary in the application of the device, realistic simulated life tests are recommended for a thorough evaluation of the error introduced by such shocks. 12-7 AD590 Typical Applications +15V +5V + + + (ADDITIONAL SENSORS) - + - - - AD590 VOUT (AVG) = (R) Σ i n VOUT = 1mV/ oK 333.3Ω 0.1% (FOR 3 SENSORS) 1kΩ FIGURE 8. AVERAGE TEMPERATURE SENSING SCHEME FIGURE 6A. The sum of the AD590 currents appears across R, which is chosen by the formula: R = 10kΩ --------------- , n OUTPUT CURRENT (µA) where n = the number of sensors. See Figure 8. HEATER ELEMENT 423 +15V 298.2 + RB AD590 LM311 3 - 218 R1 7 2 4 218oK (-55oC) 298.2oK (25oC) 423oK (150oC) R 0.1% C R2 ICL8069 1.23V R TEMPERATURE 1 VZERO 3 FIGURE 6B. FIGURE 6. SIMPLE CONNECTION. OUTPUT IS PROPORTIONAL TO ABSOLUTE TEMPERATURE FIGURE 9. SINGLE SETPOINT TEMPERATURE CONTROLLER The AD590 produces a temperature-dependent voltage across R (C is for filtering noise). Setting R2 produces a scale-zero voltage. For the celsius scale, make R = 1kΩ and VZERO = 0.273V. For Fahrenheit, R = 1.8kΩ and VZERO = 0.460V. See Figure 9. +15V + - 500µA + AD590 (AS MANY AS DESIRED) M - + + 4V < VBATT <30V + + - AD590 VOUT (MIN) - - 10kΩ 0.1% FIGURE 7. LOWEST TEMPERATURE SENSING SCHEME. AVAILABLE CURRENT IS THAT OF THE “COLDEST” SENSOR FIGURE 10. SIMPLEST THERMOMETER Meter displays current output directly in degrees Kelvin. using the AD590J, sensor output is within ±10 degrees over the entire range. See Figure 10. 12-8 AD590 The Kelvin scale version reads from 0 to 1999oK theoretically, and from 223oK to 473oK actually. The 2.26kΩ resistor brings the input within the ICL7106 VCM range: 2 general-purpose silicon diodes or an LED may be substituted. See Figure 12 and notes below. V+ R1 REF HI R2 REF LO R3 R ICL8069 1.235V ICL7106 R4 IN HI 12kΩ ZERO ADJ 5kΩ 1.000V R5 COMMON SCALE ADJ REF HI REF LO 5kΩ IN LO + V+ 7.5kΩ 15kΩ 402Ω AD590 ICL7106 26.1kΩ IN HI V- 1kΩ 0.1% FIGURE 11. BASIC DIGITAL THERMOMETER, CELSIUS AND FAHRENHEIT SCALES COMMON IN LO + R R1 R2 R3 R4 R5 oF 9.00 4.02 2.0 12.4 10.0 0 oC 5.00 4.02 2.0 5.11 5.0 11.8 AD590 V- FIGURE 13. BASIC DIGITAL THERMOMETER, KELVIN SCALE WITH ZERO ADJUST 5 ∑ R n = 28kΩ nominal n=1 ALL values are in kΩ. The ICL7106 has a VIN span of ±2.0V and a VCM range of (V+ -0.5V) to (V- +1V). R is scaled to bring each range within VCM while not exceeding VIN . VREF for both scales is 500mV maximum rending on the celsius range 199.9oC limited by the (short-term) maximum allowable sensor temperature. Maximum reading on the fahrenheit range is 199.9oF (93.3oC) limited by the number of display digits. See Figure 11 and notes below. This circuit allows “zero adjustment” as well as slope adjustment. the ICL8069 brings the input within the common-mode range, while the 5kΩ pots trim any offset at 218oK (-55oC), and set the scale factor. See Figure 13 and notes below. Notes for Figure 11, Figure 12 and Figure 13 Since all 3 scales have narrow VlN spans, some optimization of ICL7106 components can be made to lower noise and preserve CMR. The table below shows the suggested values. Similar scaling can be used with the ICL7126 and ICL7136. V+ 7.5kΩ 5kΩ 2.26kΩ 15kΩ SCALE ADJ REF HI REF LO SCALE VlN RANGE (V) RlNT (kΩ) CAZ (µF) K 0.223 to 0.473 220 0.47 C -0.25 to +1.0 220 0.1 F -0.29 to +0.996 220 0.1 ICL7106 IN HI For all: CREF = 0.1µF ClNT = 0.22µF COSC =100pF ROSC = 100kΩ COMMON 1.00kΩ IN LO + AD590 V- FIGURE 12. BASIC DIGITAL THERMOMETER, KELVIN SCALE 12-9 AD590 +15V 1kΩ ZERO SET + + NO. 1 AD590 44.2kΩ - - 10mV/ oC ICL7611 V+ + - 10kΩ 0.1% 118kΩ V- + 2.7315V - 10kΩ M +100µA - FIGURE 15. DIFFERENTIAL THERMOMETER FIGURE 14. CENTIGRADE THERMOMETER (0oC-100oC) The reference junction(s) should be in close thermal contact with the AD590 case. V+ must be at least 4V, while ICL8069 current should be set at 1mA - 2mA. Calibration does not require shorting or removal of the thermocouple: set R1 for V2 = 10.98mV. If very precise measurements are needed, adjust R2 to the exact Seebeck coefficient for the thermocouple used (measured or from table) note V1 , and set R1 to buck out this voltage (i.e., set V2 = V1). For other thermocouple types, adjust values to the appropriate Seebeck coefficient. See Figure 16. The ultra-low bias current of the ICL7611 allows the use of large value gain resistors, keeping meter current error under 1/ %, and therefore saving the expense of an extra meter 2 driving amplifier. See Figure 14. The 50kΩ pot trims offsets in the devices whether internal or external, so it can be used to set the size of the difference interval. this also makes it useful for liquid level detection (where there will be a measurable temperature difference). See Figure 15. V+ + 1µA/ oK - R2 40.2Ω VOUT = (T2 - T1) x (10mV/ oC) NO. 2 20kΩ FULL-SCALE ADJUST 100Ω 10kΩ - + (8V MIN) 115kΩ 10kΩ 741 5MΩ 50kΩ TC = 40µV/ oK SEEBECK COEFFICIENT = 40µV/ oK TYPE K + V1 = 10.98mV V+ + VOUT 1.235V V2 = 10.98mV R1 4521Ω 40.2Ω ICL8069 4.7µF FIGURE 16. COLD JUNCTION COMPENSATION FOR TYPE K THERMOCOUPLE 12-10 AD590 COLUMN SELECT ENABLE +15V +15V R (OPTIONAL) ROW SELECT 13 15 16 1 ENABLE 2 8 15 16 1 13 D GND V- 7 14 3 6 9 HI-0548 8-CHANNEL MUX 4 3 5 10 11 12 2 7 1 6 2 1 2 0 0 5 4 4 0 5 1 6 2 HI-0548 8-CHANNEL MUX 7 3 12 4 11 5 10 6 9 VGND 7 D 8 14 R (OPTIONAL) 10kΩ 0.1% AD590 (64) FIGURE 17. MULTIPLEXING SENSORS If shorted sensors are possible, a series resistor in series with the D line will limit the current (shown as R, above: only one is needed). A six-bit digital word will select one of 64 sensors. 12-11 VOUT 3 AD590 Die Characteristics DIE DIMENSIONS: PASSIVATION: 37 mils x 58 mils x 14 mils ±1 mil Type: PSG/Nitride PSG Thickness: 7kÅ ±1.4kÅ Nitride Thickness: 8kÅ ±1.2kÅ METALLIZATION: Type: Aluminum 100% Thickness: 15kÅ ±1kÅ Metallization Mask Layout AD590 All Intersil semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification. Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries for its use; nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see web site http://www.intersil.com 12-12