VRE3041 Low Cost Precision Reference THALER CORPORATION • 2015 N. FORBES BOULEVARD • TUCSON, AZ. 85745 • (520) 882-4000 FEATURES • 4.096 V Output ± 0.409 mV (.01%) PIN CONFIGURATION • Temperature Drift: 0.6 ppm/°C • Low Noise: 3µV p-p (0.1Hz-10Hz) N/C • Low Thermal Hysteresis: 1 ppm Typ. +VIN 1 2 • ±15mA Output Source and Sink Current N/C 3 • Excellent Line Regulation: 5 ppm/V Typ. GND 4 • Optional Noise Reduction and Voltage Trim VRE3041 TOP VIEW 8 NOISE REDUCTION 7 N/C 6 VOUT 5 TRIM FIGURE 1 • Industry Standard Pinout DESCRIPTION The VRE3041 is a low cost, high precision 4.096V reference that operates from +10V. The device features a buried zener for low noise and excellent long term stability. Packaged in an 8 pin DIP and SMT, the device is ideal for high resolution data conversion systems. The device provides ultrastable +4.096V output with ±0.4096 mV (.01%) initial accuracy and a temperature coefficient of 0.6 ppm/°C. This improvement in accuracy is made possible by a unique, patented multipoint laser compensation technique developed by Thaler Corporation. Significant improvements have been made in other performance parameters as well, including initial accuracy, warm-up drift, line regulation, and long-term stability, making the VRE3041 series the most accurate reference available. For enhanced performance, the VRE3041 has an external trim option for users who want less than 0.01% initial error. For ultra low noise applications, an external capacitor can be attached between the noise reduction pin and the ground pin. The VRE3041 is recommended for use as a reference for 14, 16, or 18 bit data converters which require an external precision reference. The device is also ideal for calibrating scale factor on high resolution data converters. The VRE3041 offers superior performance over monolithic references. SELECTION GUIDE Model VRE3041A VRE3041B VRE3041C VRE3041J VRE3041K VRE3041L Initial Error mV 0.410 0.614 0.820 0.410 0.614 0.820 Temp. Coeff. ppm/°C Temp. Range °C 0.6 1.0 2.0 0.6 1.0 2.0 0°C to +70°C 0°C to +70°C 0°C to +70°C -40°C to +85°C -40°C to +85°C -40°C to +85°C For package option add D for DIP or S for Surface Mount to end of model number. VRE3041DS REV. D JULY 2000 ABSOLUTE MAXIMUM RATINGS Power Supply ………………………-0.3V to +40V OUT, TRIM …………………………-0.3V to +12V NR……………………………………-0.3V to +6V Operating Temp. (A,B,C)……………0 °C to 70°C Operating Temp. (J,K,L)……………-40 °C to 85°C Out Short Circuit to GND Duration (VIN< 12V)…...Continuous Out Short Circuit to GND Duration (VIN< 40V)…….……5 sec Out Short Circuit to IN Duration (VIN< 12V)……...Continuous Continuous Power Dissipation (TA = +70°C)………...300mW Storage Temperature………………………..…-65°C to 150°C Lead Temperature (soldering,10 sec)…………………..250°C ELECTRICAL SPECIFICATIONS Vps =+10V, T = 25°C, Iout=0mA unless otherwise noted. PARAMETER Input Voltage Output Voltage (Note 1) SYMBOL CONDITIONS VIN VOUT MIN TYP 8 MAX UNITS 36 V VRE3041A/J 4.0956 4.096 4.0964 VRE3041B/K 4.0954 4.096 4.0966 VRE3041C/L 4.0952 4.096 4.0968 VRE3041A/J 0.3 0.6 VRE3041B/K 0.5 1.0 VRE3041C/K 1.0 2.0 V Output Voltage Temperature Coefficient (Note 2) TCVOUT Trim Adjustment Range ∆VOUT Figure 3 ±4 mV Turn-On Settling Time Ton To 0.01% of final value 2 µs 0.1Hz<f<10Hz 2.4 Output Noise Voltage en µVp-p 10Hz<f<1kHz 2 Note 4 1 ppm ∆VOUT/t 6 ppm/ 1khrs Supply Current IIN 3.5 Load Regulation ∆VOUT/ ∆IOUT Temperature Hysterisis Long Term Stability Line Regulation ∆VOUT/ ∆VIN Sourcing: 0mA ≤ IOUT ≤ 15mA Sinking: -15mA ≤ IOUT ≤0mA 4 4.0 8 12 8 12 8V ≤ VIN ≤ 10V 25 35 10V ≤ VIN ≤ 18V 5 10 ppm/°C µVRMS mA ppm/ mA ppm/V Notes: 1) The specified values are without external trim. 2) The temperature coefficient is determined by the box method. See discussion on temperature performance. 3) Line and load regulation are measured with pulses and do not include voltage changes due to temperature. 4) Hysterisis measured over the operating temperature range. VRE3041DS REV. D JULY 2000 TYPICAL PERFORMANCE CURVES VOUT vs. TEMPERATURE 1.00 0.75 0.75 0.75 0.50 0.50 0.50 0 -0.25 Lower Lower Limit Limit -0.25 Lower 0 20 30 40 50 60 -1.00 70 20 VOUT vs. TEMPERATURE 30 40 50 60 70 -1.00 1.0 1.0 1.0 Upper 0.5 Limit ∆Vout (mV) ∆Vout (mV) Limit 0 -0.5 Lower Limit -1.0 -1.5 -2.0 -50 -25 -1.5 -2.0 -50 -25 75 100 25 50 75 100 Temperature VRE3041K Temperature VRE3041J SUPPLY CURRENT VS. SUPPLY VOLTAGE Limit Lower Limit 0 25 50 75 100 (oC) Temperature VRE3041L QUIESCENT CURRENT VS. TEMP OUTPUT IMPEDIANCE VS. FREQUENCY Quiescent Current (mA) 5.0 4.0 3.0 0 5 10 15 20 25 30 35 40 Supply Voltage (V) Output Impediance ( Ω) 8.0 6.0 0 0 (oC) (oC) Upper 70 0 -1.5 -2.0 -50 -25 50 60 -0.5 -1.0 25 50 0.5 -1.0 0 40 VOUT vs. TEMPERATURE 1.5 Lower 30 VOUT vs. TEMPERATURE 1.5 -0.5 20 Temperature (oC) VRE3041C 1.5 0 0 Temperature (oC) VRE3041B 2.0 Limit Limit -0.50 2.0 Upper Lower 0 2.0 0.5 Limit -0.25 -0.75 0 Upper 0.25 Limit -0.75 Temperature (oC) VRE3041A ∆Vout (mV) Limit 0 -0.50 -0.75 -1.00 Upper 0.25 ∆Vout (mV) Upper Limit Limit Upper 0.25 -0.50 Supply Current (mA) VOUT vs. TEMPERATURE 1.00 ∆Vout (mV) ∆Vout (mV) VOUT vs. TEMPERATURE 1.00 6.0 4.0 2.0 0 -50 0 50 Temperature (oC) 100 Frequency (Hz) VRE3041DS REV. D JULY 2000 TYPICAL PERFORMANCE CURVES JUNCTION TEMP. RISE VS. OUTPUT CURRENT 30 20 c Vc = V 10 10 0 0 4 2 6 100 Ripple Rejection (dB) Junction Temperature Rise Above Ambient (oC) 40 8 RIPPLE REJECTION Vs. FREQUENCY(CNR=0µF) +10V A 0V 90 80 B 70 60 10 10 TURN-ON AND TURN-OFF TRANSIENT RESPONSE 100 1k 10k 1 µs/div Frequency (Hz) Output Current (mA) A: Vin, 10V/div B: Vout, 1V/div CHANGE IN OUTPUT VOLTAGE VS. OUTPUT CURRENT 100 400 60 40 100 1k Frequency (Hz) 10k 60 300 50 200 40 Vout (ppm) Vout (µV) 80 20 10 CHANGE IN OUTPUT VOLTAGE VS. INPUT VOLTAGE 100 0 -100 30 20 10 -200 0 -300 -10 -400 0 2 4 6 8 10 12 14 16 Iout(mA) -20 0 9 10 11 12 13 14 15 16 Vin(V) 0.1Hz to 10Hz Noise ∆Vout, 1µV/Div Output Noise Density (nV/√Hz) OUTPUT NOISE-VOLTAGE DENSITY vs. FREQUENCY 1 Sec/Div VRE3041DS REV. D JULY 2000 THEORY OF OPERATION BASIC CIRCUIT CONNECTION The following discussion refers to the schematic in figure 2 below. A FET current source is used to bias a 6.3V zener diode. The zener voltage is divided by the resistor network R1 and R2. This voltage is then applied to the noninverting input of the operational amplifier which amplifies the voltage to produce a 4.096V output. The gain is determined by the resistor networks R3 and R4: G=1 + R4/R3. The 6.3V zener diode is used because it is the most stable diode over time and temperature. Figure 3 shows the proper connection of the VRE3041 voltage reference with the optional trim resistor for initial error and optional capacitor for noise reduction. 8 + VIN 2 2 Optional Noise Reduction Capacitor + 6 8 CN 1µF + VOUT VRE3041 5 10kΩ 4 Optional Fine Trim Adjustment 6 - R1 Figure 3 External Connections R4 R2 5 R3 4 To achieve the specified performance, pay careful attention to the layout. A low resistance star configuration will reduce voltage errors, noise pickup, and noise coupled from the power supply. Commons should be connected to a single point to minimize interconnect resistances. Figure 2 Functional Block Diagram The current source provides a closely regulated zener current, which determines the slope of the references’ voltage vs. temperature function. By trimming the zener current a lower drift over temperature can be achieved. But since the voltage vs. temperature function is nonlinear this compensation technique is not well suited for wide temperature ranges. Thaler Corporation has developed a nonlinear compensation network of thermistors and resistors that is used in the VRE series voltage references. This proprietary network eliminates most of the nonlinearity in the voltage vs. temperature function. By adjusting the slope, Thaler Corporation produces a very stable voltage over wide temperature ranges. This network is less than 2% of the overall network resistance so it has a negligible effect on long term stability. Figure 3 shows the proper connection of the VRE3041 series voltage references with the optional trim resistor for initial error and the optional capacitor for noise reduction. PIN DESCRIPTION 1,3,7 N.C. Internally connected. Do not use 2 Vin Positive power supply input 4 GND Ground 5 TRIM External trim input. Leave open if not used. 6 OUT Voltage reference output 8 NR Noise Reduction VRE3041DS REV. D JULY 2000 TEMPERATURE PERFORMANCE THERMAL HYSTERISIS The VRE3041 is designed for applications where the initial error at room temperature and drift over temperature are important to the user. For many instrument manufacturers, a voltage reference with a temperature coefficient less than 1ppm/°C makes it possible to not have to perform a system temperature calibration, a slow and costly process. A change in output voltage as a result of a temperature change. When references experience a temperature change and return to the initial temperature, they do not always have the same initial voltage. Thermal hysterisis is difficult to correct and is a major error source in systems that experience temperature changes greater than 25°C. Reference vendors are starting to include this important specification in their datasheets. Of the three TC specification methods (slope, butterfly, and box), the box method is used commonly used. A box is formed by the min/max limits for the nominal output voltage over the operating temperature range. The equation follows: Vmax − Vmin • 10 6 T .C. = V • ( T − T ) max min nominal This method corresponds more accurately to the method of test and provides a closer estimate of actual error than the other methods. The box method guarantees limits for the temperature error but does not specify the exact shape and slope of the device under test. A designer who needs a 14-bit accurate data acquisition system over the industrial temperature range (-40°C to +85°C), will need a voltage reference with a temperature coefficient (TC) of 1.0ppm/°C if the reference is allowed to contribute an error equivalent to 1LSB. For 1/2LSB equivalent error from the reference you would need a voltage reference with a temperature coefficient of 0.5ppm/°C. Figure 4 shows the required reference TC vs. delta T change from 25°C for resolution ranging from 8 bits to 20 bits. 10000 1000 100 8 BIT ReferenceTC (ppm/°C) 10 10 BIT 12 BIT 1 14 BIT 16 BIT 0.1 18 BIT 0.01 20 BIT 1 10 100 Reference TC vs. ∆T change from 25°C for 1 LSB change VRE3041DS REV. D JULY 2000 MECHANICAL SPECIFICATIONS INCHES MILLIMETER MAX INCHES MILLIMETER DIM MIN MAX MIN DIM MIN MAX MIN A .110 .120 2.794 3.048 D1 .372 .380 9.45 MAX B .095 .105 2.413 2.667 E .425 .435 10.80 11.05 B1 .021 .027 0.533 0.686 E1 .397 .403 10.08 10.24 C .055 .065 1.397 1.651 E2 .264 .270 6.71 6.86 C1 .012 .020 0.305 0.508 P .085 .095 2.16 2.41 C2 .020 .040 0.508 1.016 S .045 .055 1.14 1.40 D .395 .405 10.03 10.29 9.65 D D1 E2 E1 E 1 A P C1 B S B1 C C2 VRE3041DS REV. D JULY 2000 MECHANICAL SPECIFICATIONS INCHES MILLIMETER MILLIMETER DIM MIN MAX MIN DIM MIN MAX MIN A .170 .180 4.318 4.572 E .425 .435 10.80 11.05 B .095 .105 2.413 2.667 E1 .397 .403 10.08 10.24 B1 .016 .020 0.406 0.508 E2 .264 .270 6.71 6.86 C .008 .011 0.203 0.279 G .290 .310 7.36 7.87 C1 .055 .065 1.397 1.651 L .175 .225 4.46 5.72 D .395 .405 10.03 10.29 P .085 .095 2.16 2.41 D1 .372 .380 S .045 .055 1.14 1.40 9.45 MAX INCHES 9.65 MAX D D1 E2 E1 E 1 P A C1 L C S B G B1 VRE3041DS REV. D JULY 2000