XTR106: 4-20mA Current Transmitter with Bridge Excitation And ...
XTR106: 4-20mA Current Transmitter with Bridge Excitation And ... XTR106: 4-20mA Current Transmitter with Bridge Excitation And ...
When using linearity correction, care should be taken toinsure that the sensor’s output common-mode voltage remainswithin the XTR106’s allowable input range of 1.1V to3.5V. Equation 6 in Figure 3 can be used to calculate theXTR106’s new excitation voltage. The common-mode voltageof the bridge output is simply half this value if nocommon-mode resistor is used (refer to the example inFigure 3). Exceeding the common-mode range may yieldunpredicatable results.For high precision applications (errors < 1%), a two-stepcalibration process can be employed. First, the nonlinearityof the sensor bridge is measured with the initial gain resistorand R LIN = 0 (R LIN pin connected directly to V REG ). Usingthe resulting sensor nonlinearity, B, values for R G and R LINare calculated using Equations 4 and 5 from Figure 3. Asecond calibration measurement is then taken to adjust R G toaccount for the offsets and mismatches in the linearization.UNDER-SCALE CURRENTThe total current being drawn from the V REF and V REGvoltage sources, as well as temperature, affect the XTR106’sunder-scale current value (see the Typical PerformanceCurve, “Under-Scale Current vs I REF + I REG ). This should beconsidered when choosing the bridge resistance and excitationvoltage, especially for transducers operating over awide temperature range (see the Typical Performance Curve,“Under-Scale Current vs Temperature”).LOW IMPEDANCE BRIDGESThe XTR106’s two available excitation voltages (2.5V and5V) allow the use of a wide variety of bridge values. Bridgeimpedances as low as 1kΩ can be used without any additionalcircuitry. Lower impedance bridges can be used withthe XTR106 by adding a series resistance to limit excitationcurrent to ≤ 2.5mA (Figure 5). Resistance should be added10BRIDGE TRANSDUCER TRANSFER FUNCTIONWITH PARABOLIC NONLINEARITY3NONLINEARITY vs STIMULUS9Bridge Output (mV)8765432100Positive NonlinearityB = +0.025B = –0.019Negative NonlinearityLinear Response0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1Nonlinearity (% of Full Scale)210–1–2–30Positive NonlinearityB = +0.025Negative NonlinearityB = –0.0190.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1Normalized StimulusNormalized StimulusFIGURE 4. Parabolic Nonlinearity.700µA at 5VV REF 5I TOTAL = 0.7mA + 1.6mA ≤ 2.5mAI REG ≈ 1.6mAV REF 2.5V REG3.4kΩ5V1/2OPA22771kΩ54V + IN14131R LIN1110V+1N4148350Ω10kΩR G3.4kΩ412Ω1/2OPA227710kΩR G125Ω32R GV – INXTR106B9E8Lin I OPolarity7I RET 1260.01µFI O = 4-20mABridge excitationvoltage = 0.245VApprox. x50amplifierShown connected to correct positivebridge nonlinearity. For negative bridgenonlinearity, see Figure 3b.FIGURE 5. 350Ω Bridge with x50 Preamplifier.12www.ti.comXTR106SBOS092A
to the upper and lower sides of the bridge to keep the bridgeoutput within the 1.1V to 3.5V common-mode input range.Bridge output is reduced so a preamplifier as shown may beneeded to reduce offset voltage and drift.OTHER SENSOR TYPESThe XTR106 can be used with a wide variety of inputs. Itshigh input impedance instrumentation amplifier is versatileand can be configured for differential input voltages frommillivolts to a maximum of 2.4V full scale. The linear rangeof the inputs is from 1.1V to 3.5V, referenced to the I RETterminal, pin 6. The linearization feature of the XTR106 canbe used with any sensor whose output is ratiometric with anexcitation voltage.ERROR ANALYSISTable I shows how to calculate the effect various errorsources have on circuit accuracy. A sample error calculationfor a typical bridge sensor measurement circuit is shown(5kΩ bridge, V REF = 5V, V FS = 50mV) is provided. Theresults reveal the XTR106’s excellent accuracy, in this case1.2% unadjusted. Adjusting gain and offset errors improvescircuit accuracy to 0.33%. Note that these are worst-caseerrors; guaranteed maximum values were used in the calculationsand all errors were assumed to be positive (additive).The XTR106 achieves performance which is difficult toobtain with discrete circuitry and requires less board space.SAMPLE ERROR CALCULATIONBridge Impedance (R B ) 5kΩ Full Scale Input (V FS ) 50mVAmbient Temperature Range (∆T A ) 20°C Excitation Voltage (V REF ) 5VSupply Voltage Change (∆V+) 5V Common-Mode Voltage Change (∆CM) 25mV (= V FS /2)ERRORSAMPLE(ppm of Full Scale)ERROR SOURCE ERROR EQUATION ERROR CALCULATION UNADJ ADJUSTINPUTInput Offset Voltage V OS /V FS • 10 6 200µV/50mV • 10 6 2000 0vs Common-Mode CMRR • ∆CM/V FS • 10 6 50µV/V • 0.025V/50mV • 10 6 25 25vs Power Supply (V OS vs V+) • (∆V+)/V FS • 10 6 3µV/V • 5V/50mV • 10 6 300 300Input Bias Current CMRR • I B • (R B /2)/ V FS • 10 6 50µV/V • 25nA • 2.5kΩ/50mV • 10 6 0.1 0Input Offset Current I OS • R B /V FS • 10 6 3nA • 5kΩ/50mV • 10 6 300 0Total Input Error 2625 325EXCITATIONVoltage Reference Accuracy V REF Accuracy (%)/100% • 10 6 0.25%/100% • 10 6 2500 0vs Supply (V REF vs V+) • (∆V+) • (V FS /V REF ) 20ppm/V • 5V (50mV/5V) 1 1Total Excitation Error 2501 1GAINSpan Span Error (%)/100% • 10 6 0.2%/100% • 10 6 2000 0Nonlinearity Nonlinearity (%)/100% • 10 6 0.01%/100% • 10 6 100 100Total Gain Error 2100 100OUTPUTZero Output | I ZERO – 4mA | /16000µA • 10 6 25µA/16000µA • 10 6 1563 0vs Supply (I ZERO vs V+) • (∆V+)/16000µA • 10 6 0.2µA/V • 5V/16000µA • 10 6 62.5 62.5Total Output Error 1626 63DRIFT (∆T A = 20°C)Input Offset Voltage Drift • ∆T A /(V FS ) • 10 6 1.5µV/°C • 20°C / (50mV) • 10 6 600 600Input Offset Current (typical) Drift • ∆T A • R B /(V FS ) • 10 6 5pA / °C • 20°C • 5kΩ/ (50mV) • 10 6 10 10Voltage Refrence Accuracy 35ppm/°C • 20°C 700 700Span 225ppm/°C • 20°C 500 500Zero Output Drift • ∆T A / 16000µA • 10 6 0.9µA/°C • 20°C / 16000µA • 10 6 1125 1125Total Drift Error 2936 2936NOISE (0.1Hz to 10Hz, typ)Input Offset Voltage V n (p-p)/ V FS • 10 6 0.6µV / 50mV • 10 6 12 12Zero Output I ZERO Noise / 16000µA • 10 6 0.035µA / 16000µA • 10 6 2.2 2.2Thermal R B Noise [√ 2 • √ (R B /2)/1kΩ • 4nV / √ Hz • √ 10Hz ] / V FS • 10 6 [√ 2 • √ 2.5kΩ /1kΩ • 4nV/ √ Hz • √ 10Hz ] / 50mV • 10 6 0.6 0.6Input Current Noise (i n • 40.8 • √2 • R B /2)/V FS • 10 6 (200fA/√Hz • 40.8 • √2 • 2.5kΩ)/50mV• 10 6 0.6 0.6Total Noise Error 15 15NOTE (1): All errors are min/max and referred to input, unless otherwise stated.TOTAL ERROR: 11803 33401.18% 0.33%TABLE I. Error Calculation.XTR106 13SBOS092Awww.ti.com
- Page 3 and 4: PIN CONFIGURATIONTop ViewV REGV - I
- Page 5 and 6: TYPICAL PERFORMANCE CURVESAt T A =
- Page 7 and 8: TYPICAL PERFORMANCE CURVES (CONT)At
- Page 9 and 10: APPLICATIONS INFORMATIONFigure 1 sh
- Page 11: Figure 4, but not peaking exactly a
- Page 15 and 16: V REF 50.01µFSee ISO124 data sheet
- Page 19: IMPORTANT NOTICETexas Instruments I
to the upper and lower sides of the bridge to keep the bridgeoutput <strong>with</strong>in the 1.1V to 3.5V common-mode input range.<strong>Bridge</strong> output is reduced so a preamplifier as shown may beneeded to reduce offset voltage and drift.OTHER SENSOR TYPESThe <strong>XTR106</strong> can be used <strong>with</strong> a wide variety of inputs. Itshigh input impedance instrumentation amplifier is versatileand can be configured for differential input voltages frommillivolts to a maximum of 2.4V full scale. The linear rangeof the inputs is from 1.1V to 3.5V, referenced to the I RETterminal, pin 6. The linearization feature of the <strong>XTR106</strong> canbe used <strong>with</strong> any sensor whose output is ratiometric <strong>with</strong> anexcitation voltage.ERROR ANALYSISTable I shows how to calculate the effect various errorsources have on circuit accuracy. A sample error calculationfor a typical bridge sensor measurement circuit is shown(5kΩ bridge, V REF = 5V, V FS = 50mV) is provided. Theresults reveal the <strong>XTR106</strong>’s excellent accuracy, in this case1.2% unadjusted. Adjusting gain and offset errors improvescircuit accuracy to 0.33%. Note that these are worst-caseerrors; guaranteed maximum values were used in the calculationsand all errors were assumed to be positive (additive).The <strong>XTR106</strong> achieves performance which is difficult toobtain <strong>with</strong> discrete circuitry and requires less board space.SAMPLE ERROR CALCULATION<strong>Bridge</strong> Impedance (R B ) 5kΩ Full Scale Input (V FS ) 50mVAmbient Temperature Range (∆T A ) 20°C <strong>Excitation</strong> Voltage (V REF ) 5VSupply Voltage Change (∆V+) 5V Common-Mode Voltage Change (∆CM) 25mV (= V FS /2)ERRORSAMPLE(ppm of Full Scale)ERROR SOURCE ERROR EQUATION ERROR CALCULATION UNADJ ADJUSTINPUTInput Offset Voltage V OS /V FS • 10 6 200µV/50mV • 10 6 2000 0vs Common-Mode CMRR • ∆CM/V FS • 10 6 50µV/V • 0.025V/50mV • 10 6 25 25vs Power Supply (V OS vs V+) • (∆V+)/V FS • 10 6 3µV/V • 5V/50mV • 10 6 300 300Input Bias <strong>Current</strong> CMRR • I B • (R B /2)/ V FS • 10 6 50µV/V • 25nA • 2.5kΩ/50mV • 10 6 0.1 0Input Offset <strong>Current</strong> I OS • R B /V FS • 10 6 3nA • 5kΩ/50mV • 10 6 300 0Total Input Error 2625 325EXCITATIONVoltage Reference Accuracy V REF Accuracy (%)/100% • 10 6 0.25%/100% • 10 6 2500 0vs Supply (V REF vs V+) • (∆V+) • (V FS /V REF ) 20ppm/V • 5V (50mV/5V) 1 1Total <strong>Excitation</strong> Error 2501 1GAINSpan Span Error (%)/100% • 10 6 0.2%/100% • 10 6 2000 0Nonlinearity Nonlinearity (%)/100% • 10 6 0.01%/100% • 10 6 100 100Total Gain Error 2100 100OUTPUTZero Output | I ZERO – 4mA | /16000µA • 10 6 25µA/16000µA • 10 6 1563 0vs Supply (I ZERO vs V+) • (∆V+)/16000µA • 10 6 0.2µA/V • 5V/16000µA • 10 6 62.5 62.5Total Output Error 1626 63DRIFT (∆T A = 20°C)Input Offset Voltage Drift • ∆T A /(V FS ) • 10 6 1.5µV/°C • 20°C / (50mV) • 10 6 600 600Input Offset <strong>Current</strong> (typical) Drift • ∆T A • R B /(V FS ) • 10 6 5pA / °C • 20°C • 5kΩ/ (50mV) • 10 6 10 10Voltage Refrence Accuracy 35ppm/°C • 20°C 700 700Span 225ppm/°C • 20°C 500 500Zero Output Drift • ∆T A / 16000µA • 10 6 0.9µA/°C • 20°C / 16000µA • 10 6 1125 1125Total Drift Error 2936 2936NOISE (0.1Hz to 10Hz, typ)Input Offset Voltage V n (p-p)/ V FS • 10 6 0.6µV / 50mV • 10 6 12 12Zero Output I ZERO Noise / 16000µA • 10 6 0.035µA / 16000µA • 10 6 2.2 2.2Thermal R B Noise [√ 2 • √ (R B /2)/1kΩ • 4nV / √ Hz • √ 10Hz ] / V FS • 10 6 [√ 2 • √ 2.5kΩ /1kΩ • 4nV/ √ Hz • √ 10Hz ] / 50mV • 10 6 0.6 0.6Input <strong>Current</strong> Noise (i n • 40.8 • √2 • R B /2)/V FS • 10 6 (200fA/√Hz • 40.8 • √2 • 2.5kΩ)/50mV• 10 6 0.6 0.6Total Noise Error 15 15NOTE (1): All errors are min/max and referred to input, unless otherwise stated.TOTAL ERROR: 11803 33401.18% 0.33%TABLE I. Error Calculation.<strong>XTR106</strong> 13SBOS092Awww.ti.com
Change language