AN-1271 APPLICATION NOTE One Technology Way • P.O. Box 9106 • Norwood, MA 02062-9106, U.S.A. • Tel: 781.329.4700 • Fax: 781.461.3113 • www.analog.com Optimizing the ADuCM350 for Impedance Conversion INTRODUCTION The ADuCM350 is an ultralow power integrated mixed-signal metering solution that includes a microcontroller subsystem for processing, control, and connectivity. The processor subsystem is based on a low power ARM® Cortex™-M3 processor, a collection of digital peripherals, embedded SRAM and flash memory, and an analog subsystem which provides clocking, reset, and power management capability. output voltage, the RTIA/CTIA combination, and the calibration resistor all need to be calculated. The maximum allowed current into the load dominates the calculation. This application note details how to set up the ADuCM350 to optimally measure the impedance of an RC sensor using a 2-wire measurement approach. To optimize the accuracy of the impedance measurement, the user must maximize the usage of the16-bit ADC range. To do this, the peak-to-peak excitation However, if there is a limitation on the load current, for example, to meet IEC 60601 standards in 2-wire, bioimpedance applications, then the user should calculate the maximum allowable current and use precautionary measures in the circuitry. If there is no limitation, then the user has the ability to maximize the amount of signal swing into the ADC from the transimpedance amplifier (TIA) to get the best SNR possible. Rev. A | Page 1 of 8 AN-1271 Application Note TABLE OF CONTENTS Introduction ...................................................................................... 1 Scenario 1: No Limitation on Load Current ..............................3 Revision History ............................................................................... 2 Scenario 2: Limited Load Current ..............................................4 Details................................................................................................. 3 Impedance Measurement Example in LabVIEW .....................6 Sensor Configuration ................................................................... 3 Impedance Measurement Example in Software Development Kit ...........................................................................8 Calculate the Minimum Ideal Impedance of the Sensor......... 3 RCAL Calculation ........................................................................ 3 REVISION HISTORY 3/14—Rev. 0 to Rev. A Changes to Figure 5 .......................................................................... 6 12/13—Revision 0: Initial Version Rev. A | Page 2 of 8 Application Note AN-1271 DETAILS To calculate the magnitude of YP SENSOR CONFIGURATION In the example described in this application note, the user wants to measure the impedance of an RC type sensor with the configuration shown in Figure 1 for a 1 kHz excitation signal. Magnitude = R 2 + I 2 |YP| = 0.00377 Invert this to get The sensor details are as follows: |ZP| = 265.16 Ω CP = 10 nF to 600 nF Calculate the phase of YP. RP = 10 kΩ Phase (rads) = atan (I/R) = 1.544 RS= 1 kΩ Phase (degrees) = Phase (rads) × 180 = −88.48° π Impedance of parallel RC = ZP = 265.16 ∠ −88.48 Now, add the RS series resistor. CP ZT = ZP + ZS RP ZP = 265.16 ∠ −88.48 = 7.03 –i265.06 ZS = 1000 ∠ 0 = 1000 + i0 RS 11993-001 Add the two complex numbers. ZT = 1007.03 – i265.06 = 1041 ∠ −14.75 Figure 1. Sensor RC Configuration CALCULATE THE MINIMUM IDEAL IMPEDANCE OF THE SENSOR The first step is to calculate the lowest unknown impedance of the sensor. This allows the user to calculate the highest current signal into the TIA. For the sensor in Figure 1, the impedance of the sensor is at its minimum when CP = 600 nF. To calculate the total impedance, ZT, the first step is to calculate the impedance of the CP capacitor. Z Cp 1 = −i265.26 = 2πfC P This is the lowest impedance realized by the part and it determines the maximum amount of current seen by the TIA. RCAL CALCULATION To calculate the RCAL value to calibrate the system, the lowest unknown impedance, Z, is used. If RCAL is equal to the magnitude of minimum impedance, the signal going into the DFT will be large. This improves repeatability and accuracy. Therefore, an RCAL of ~1041 Ω is used for this example. SCENARIO 1: NO LIMITATION ON LOAD CURRENT Where there is no limitation on current seen by load, the maximum signal swing is used to maximize the SNR of ADC results. where: f is an excitation frequency of 1 kHz. CP is 600 nF. • • Then, calculate the impedance of the parallel components RP || CP. • 1/ZP = 1/ZRp + 1/ZCp • or to simplify YP = YRp + YCp YRp = 1/10,000 YCp = 1/−i265.26 YP =1× 10-4 + i3.77 × 10−3 (C lags by 90°). Maximum voltage swing is 600 mV peak. Highest signal current into TIA = 600 mV peak/1041 Ω = 0.576 mA peak. Peak voltage at output of TIA (maximum allowed by the ADuCM350) = 750 mV peak. RTIA resistor to give peak 750 mV voltage for peak signal current: 750 mV/0.576 mA = 1.302 kΩ To improve the anti-aliasing performance and stability of the receive channel, an anti-aliasing capacitor is put in parallel with RTIA. The 3 dB point of 80 kHz is selected (this is the maximum bandwidth of the system). CTIA = Rev. A | Page 3 of 8 1 1 = = 1.5 nF 2πfRTIA 2 × π × 80 kHz × 1.302 kΩ AN-1271 Application Note AFE_DAC_OFFSET_UNITY SWIT CH MATRIX –900mV RCAL1 600mV p-p RCAL2 AFE8 AFE7 ×2 CLOSED LOOP GAIN = 2 D EXCITATION AMPLIFIER LOOP 0 DAC_ATTEN_EN = “0” DAC_ATTEN_EN DAC_ATTEN_EN –300mV VREFDAC 1.8V RCF 50kHz 1.0V DACP 12-BIT PGA DACN 0.6V DAC 600mV GAIN 1/0.025 REFNHIZ 0.6V RP AFE_DAC_GAIN DAC_CODE[11:0] DACCLK –3584 DECIMAL P 600mV N 600mV p-p –900mV –512 DECIMAL 3072 CODES p-p DECIMAL –300mV AFE5 CP 1 0.2V AFE6 1.7V AFE_DAC_OFFSET_ATTEN AFE4 500mV 1.2V p-p AFE2 1.85V DIFFERENTIAL 1.1V AFE1 2.25V p-p RS 1.5V p-p 0.9V 1.125V p-p 0V 0.35V GAIN = 1.5 4Ω T TIA_I RTIA 1.302kΩ 6Ω 0.9V TIA_O CTIA 1.5nF 4Ω VREF 1.8V ANTI-ALIAS FILTERING VBIAS = 1.1V ADC RANGE = 3.6V AFE3 ASSUMING VBIAS of 1.1V ADC 6Ω 1.1V 0.9V 1.125 V p-p ADC MUX 11993-002 ±576µA Figure 2. Signal Swings with No Limitation on Load Current SCENARIO 2: LIMITED LOAD CURRENT When there is a limitation on the load current, then a different approach is taken. In this example, the IEC 60601 bodily floating standard allows a maximum of 100 µA rms leakage. Thus, for this example, it is safe to assume that 50 µA rms/ 70.7 µA peak is the maximum current. From a single fault correction perspective, with regard to the bodily floating standard, the following is included on each leg: • • 1 µF dc CS blocking capacitor A series resistor representing some form of leads (RLEAD) Minimum impedance of the sensor remains at 1041 Ω. The series components now add to this minimum impedance seen by the ADuCM350 TIA. Calculating the impedance of extra circuitry in the network gives 200 Ω + 100 Ω +100 Ω + 1 µF + 1 µF Assume an excitation frequency of 1 kHz. Capacitors are in series, thus Connect an extra current limiting series resistance of 200 Ω to the drive leg, RLIMIT. CT = (C1 × C2)/(C1 + C2) where CT = 0.5 µF. ZC = 1/(2 πfCT) = 1/(2 × π ×1 kHz × 0.5 µF) = −i318.3 RLEAD 100Ω CS RLIMIT 200Ω RT = RLIMIT + RLEAD + RLEAD = 400 Total extra circuitry impedance = 400 –i318.3 1µF The minimum impedance seen by the TIA is the minimum sensor impedance plus the minimum extra circuitry impedance converter. RP RS RLEAD 100Ω CS ZT = (400 – i318.3) + (1007.03 – i265.06) = 1407 – i583.4 1µF = 1523 ∠ −22.5 11993-003 CP Figure 3. Sensor with Single Fault External Protection This is the minimum impedance seen by the ADuCM350. For safety reasons, reduce this by 20% to avoid unwanted overranging of ADC results. Rev. A | Page 4 of 8 Application Note AN-1271 Therefore, a minimum impedance of 1218.4 Ω is assumed. The Cortex-M3 flags any impedance measurement below value as an invalid result. Check the connections because the flag indicates that the ADC has overranged or encountered another error. With a bigger LSB size, there is less resolution of measurement, thus more quantization noise in creating the sinewave and measuring the response. Therefore, to allow a maximum of 70.7 µA peak with a minimum impedance of 1218.4 Ω, a sinewave amplitude is needed. Calculate the current seen by TIA. Continuing with this example, proceed using a 15 mV peak sinwave with attenuation enabled (DAC_ATTEN = 1). 15 mV peak/1218.4 Ω = 12.3 µA peak signal Then, calculate the RTIA and CTIA to optimize the ADC range where the RTIA resistor gives the peak 750 mV voltage signal current. 70.7 × 10−6 × 1218.4 Ω = 86 mV peak Note that the maximum allowed sinewave amplitude when the DAC attenuator is enabled, DAC_ATTEN = 1, is 15 mV peak. Because 86 mV peak exceeds this value, there are two options. The first option is to use 15 mV peak with a reduced signal-tonoise ratio. The second option is to disable the DAC attenuator and select 86.5 mV peak in nonattenuation mode. The downside of this is that the LSB size increases 40 times. 750 mV/12.3 µA = 60.98 kΩ To improve the anti-aliasing performance and improve the stability of the receive channel, an anti-aliasing capacitor is put in parallel to RTIA. The 3 dB point of 80 kHz is selected (the maximum bandwidth of the system). CTIA = 1/2πfRTIA Nonattenuator mode = 1/ (2π × 80 kHz × 60.98 kΩ) LSB size = 1.6 V/2 = 390 µV p-p = 195 µV peak 12 = 32.6 pF Attenuator mode LSB size = 1/40 (1.6 V/2 )= 9.76 µV p-p = 4.88 µV peak 12 AFE_DAC_OFFSET_UNITY SWIT CH MATRIX –607.5mV RCAL1 30mV p-p RCAL2 RLIMIT RLEAD 100Ω CS 200Ω 1µF –1.115V AFE8 AFE7 ×2 CLOSED LOOP GAIN = 2 D 0 DAC_ATTEN_EN = “1” DAC_ATTEN_EN DAC_ATTEN_EN –592.5mV VREFDAC 1.8V RCF 50kHz 1.0V DACP 12-BIT PGA DACN 0.6V DAC EXCITATION AMPLIFIER LOOP 600mV GAIN 1/0.025 REFNHIZ 0.6V RP AFE_DAC_GAIN DAC_CODE[11:0] DACCLK –3584 DECIMAL P 600mV N 600mV p-p –900mV –512 DECIMAL 3072 CODES p-p DECIMAL –300mV AFE5 –1.085V 30mV p-p 1 0.2V AFE6 CP AFE_DAC_OFFSET_ATTEN AFE4 AFE2 1.85V DIFFERENTIAL 1.1V AFE1 1.5V p-p 0.9V 1.125V p-p 0V 0.35V GAIN = 1.5 VBIAS = 1.1V 4Ω T TIA_I RTIA 61kΩ 6Ω 0.9V TIA_O CTIA 32.6F 4Ω ADC RANGE = 3.6V 1µF AFE3 VREF 1.8V ADC 6Ω ±12.3µA 1.1V ADC MUX Figure 4. Signal Swings When Limitation On Load Current Rev. A | Page 5 of 8 0.9V 1.125 V p-p 11993-004 RLEAD 100Ω CS 2.25V p-p RS ANTI-ALIAS FILTERING ASSUMING VBIAS of 1.1V AN-1271 Application Note IMPEDANCE MEASUREMENT EXAMPLE IN LABVIEW 11993-005 The ADuCM350 LabVIEW® GUI can measure impedance and rapidly prototype sensor measurement. Figure 5. LabVIEW Impedance Measurement AFE Control Rev. A | Page 6 of 8 AN-1271 11993-006 Application Note Figure 6. LabVIEW Impedance Measurement Rev. A | Page 7 of 8 AN-1271 Application Note IMPEDANCE MEASUREMENT EXAMPLE IN SOFTWARE DEVELOPMENT KIT The IAR embedded workbench based software development kit includes an ImpedanceMeasurement_2Wire example. This example verifies the performance of the impedance converter on the ADuCM350. The sequence for the example is measuring three unknown 2-wire impedances. • • • AFE3 to AFE4 AFE4 to AFE5 AFE5 to AFE6 The code has easy programmability for excitation frequency, excitation voltage, and RCAL value. It is possible to code other changes to the measurement into the measurement sequence. The readme section of this example project provides more details. 11993-005 Figure 7 shows an example measurement. Both real and imaginary components are calculated. Figure 7. Signal Swings When Limited On Load Current ©2013–2014 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. AN11933-0-3/14(A) Rev. A | Page 8 of 8