Freescale Semiconductor Application Note AN1668 Rev 2, 11/2006 Washing Appliance Sensor Selection by: Ador Reodique Sensor and Systems Applications Engineer INTRODUCTION North American washing machines currently in production use mechanical sensors for water level measurement function. These sensors are either purely mechanical pressure switch with discrete trip points or electromechanical pressure sensor with an on-board electronics for a frequency output. High efficiency machines require high performance sensors (accuracy, linearity, repeatability) even at lower pressure ranges. Benchmarks indicate that these performance goals is difficult to achieve using current mechanical pressure sensors. In Europe, where energy conservation is mandated, washing machine manufacturers have started to look at electronic solutions where accuracy, reliability, repeatability and additional functionality is to be implemented. North American and Asia Pacific manufacturers are also looking for better solutions. From surveys of customer requirements, a typical verticalaxis machine calls for a sensor with 600 mm H2O (24 “ H2O ~ 6 kPa) sensor with a 5% FS accuracy spec. Certain appliances call for a lower pressure range especially in Europe where horizontal axis machines are common. SENSOR SOLUTIONS For the typical 600 mm H2O, 5% FS spec, an off the shelf solution available today is the MPX10/MPX12, MXP2010 and the MPXV4006G sensor. The MPX10 (or the MPX12) is 10 kPa (40 “ H2O) full-scale pressure range device. It is uncompensated for temperature and untrimmed offset and full-scale span. This means that the end user must temperature compensate as well as calibrate the full-scale offset and span of the device. The output of the device must be amplified using a differential amplifier (see Figure 1) so it can be interfaced to an A/D and to obtain the desired range. Since the MPX10/MPX12 sensors must be calibrated, the implications of this device being used in high-volume production is expensive. Because the offset and full-scale output can vary from part to part, a two-point calibration is required as a minimum. A two point calibration is a time consuming procedure as well as possible modification to the production line to accommodate the calibration process. The circuitry must also accommodate for trimming, i.e., via © Freescale Semiconductor, Inc., 2006. All rights reserved. trimpots and/or EEPROM to store the calibration data. This adds extra cost to the system. The MPX2010 is a 10 kPa (40" H2O), temperature compensated, offset and full-scale output calibrated device. A differential amplifier like the one shown in Figure 1 should be used to amplify its output. Unlike the MPX10 or MPX12, this device does not need a two-point calibration but auto-zeroing can improve its performance. This procedure is easily implemented using the system MCU. The MPXV4006G is a fully integrated pressure sensor specifically designed for appliance water level sensing application. This device has an on board amplification, temperature compensation and trimmed span. An auto-zero procedure should be implemented with this device (refer to Application Note AN1636). Because expensive and time consuming calibration, temperature compensation and amplification is already implemented, this device is more suitable for high volume production. The MPXV4006G integrated sensor is guaranteed to be have an accuracy of ± 3% FS over its pressure and temperature range. For washing machine applications where low cost and high volume productions are involved, both the MPX2010 and MPXV4006G are recommended. Both solutions can be used in current vertical axis machines where the water level in the 600 mm H2O or 24 “ H2O range. In the following, a comparison is made between MPX2010 and MPXV4006G in terms of system and performance considerations to help the customer make a decision. EXPECTED ACCURACY OF THE MPX2010 SYSTEM SOLUTION The MPX2010 compensated sensor has an off the shelf overall RMS accuracy of ± 7.2% FS over 0 to 85°C temperature range. Auto-zeroing can improve the sensor accuracy to ± 4.42% FS. However, since this sensor does not have an integrated amplification, its amplifier section must be designed carefully in order to meet the target accuracy requirement. The MPX2010 compensated sensor has the following specifications shown on Table 1. Table 1. MPX2010 Specifications Min Characteristic Pressure Range Typ 0 Supply Voltage 10 Supply Current Max Unit 10 kPa 16 Vdc 6 Full Scale Span 24 Offset *1 25 mA 26 1 Sensitivity 25 Linearity 0.1 Temperature Hysterisis (*40 to 125°C) mV mV/kPa 1 *1 Pressure Hysterisis mV %VFSS %VFSS 0.5 %VFSS Temperature Effect on Span *1 1 %VFSS Temperature Effect Offset (0 to 85°C) *1 1 mV Input Impedance 1300 2550 W Output Impedance 1400 3000 W Response Time (10% to 90%) 1 ms Warm-Up 20 ms The sensor system errors is made up of the sensor errors, amplifier errors and A/D errors. In other words, εSystem = εSensor 2 + εAmplifier 2 + εADResolution 2 ( 1 ) Table 2 shows the MPX2010 with the errors converted to %VFSS. The expected maximum root mean squared error of the sensor is εSensor = 2 2 2 2 2 2 2 2 SpanCal + Lin + Phys + Thys + TCS + OffCal + Tco + OffStab = ± 7.19% FS. (2) With auto-zeroing, the offset calibration, temperature effect on offset and offset stability is reduced or eliminated, εSensor = SpanCal 2 + Lin 2 + Phys 2 + Thys 2 + TCS 2 (3) = +/- 4.42% FS. The sensor error is calculated using the full-scale pressure range of the device, 0 to 85°C temperature and 10 V excitation. In comparison with the MPXV4006G solution, the expected accuracy of the system (MPXV4006G + 8 bit A/D) with auto-zero is 3.1% FS. Table 2. MPX2010 Span, Offset and Calculated Maximum RMS Error* Span Errors (converted to %VFSS) Span Calibration Linearity Symbol Error Value SpanCal 4 Note Unit %VFSS Lin 1 %VFSS Pressure Hysterisis Phys 0.1 %VFSS Temperature Hysterisis Thys 0.5 %VFSS Tcs 1.5 %VFSS OffCal 4 %VFSS Tco 4 %VFSS OffStab 0.5 %VFSS Temperature Effect on Span Offset Errors (converted to %VFSS) Offset Calibration Temperature Effect on Offset Offset Stability Calculated Maximum RMS Errors RMS Error No Compensation* 7.19 %FS With auto-zero 4.42 %FS * This assumes the power supply is constant. AN1668 2 Sensors Freescale Semiconductor AMPLIFIER SELECTION AND AMPLIFIER INDUCED ERRORS instrumentation amplifier can be used. However, it requires additional components (3 op-amps and possibly a split power supply). An instrumentation topology shown in Figure 1 requires only a single supply and only 2 op-amps and 1% resistors. A differential amplifier is needed to convert the differential output of the MPX2010 sensor to a high level groundreferenced (single-ended). The classic three-op amp +VCC R2 R4 R+S1 R1 VREF 2 – 1 3 + R+S2 U1A R3 6 – 5 + 7 U1B VOUT_FS +VCC 3 4 S– Pressure Sensor X1 1 S+ 2 Figure 1. MPX2010 Amplifier Circuit The circuit uses a voltage divider R+S1 and R+S2 to provide the reference (level shift), U1A and U1B are noninverting amplifiers arranged in a differential configuration with gain resistors R1, R2, R3, and R4. Note that U1B is the main gain stage and it has the most gain. It is recommended to place a 0.015 µF capacitor in it's feedback loop (in parallel with R4) to reduce noise. The amplifier output can be characterized with the equation below: R4 Gain = -------- + 1 R3 R2 ⋅ R1 R2 ⋅ R4 Voffset = VREF ⎛ ---------------------⎞ – VSCM ⎛ ---------------------⎞ ⎝ R1 ⋅ R3⎠ ⎝ R1 ⋅ R3⎠ Vout = (S+ - S-) Gain + Voffset (4) –1 (5) (6) where (S+ - S-) = Sensor differential output + Sensor offset (7) Equation 4 is the differential gain of the amplifier and equation 5 is the resulting offset voltage of the amplifier. The above equations assume that the amplifier is close to ideal (high AOL, low input offset voltage and low input offset bias currents). Since an ideal op-amp is hard to come by, the customer should select an op-amp based on cost and performance. Below are some points to keep in mind when selecting an op-amp and designing the amplifier circuit. Note that the ratio R2*R4/R1*R3 controls the system offset as well as the common mode error of the amplifier. Mismatches in these resistors will result in an offset and common mode error which appear as offset. It is therefore recommended to use 1% metal film resistors to reduce these errors. Also, VREF source impedance should be minimized in comparison with R1 in order to reduce common mode error. Amplifier input offset and input bias currents can induce errors. For example, an input offset (Vio) of the amplifier can become significant when the closed-loop gain of the amplifier is increased. Furthermore, there is also a temperature coefficient of the input voltage offset which contribute an additional error across temperature. If the input bias current of the amplifier is not taken into account in the design, it can also become a source of error. A technique to reduce this error is to match the impedance the source impedance of what the opamp input pins sees. It is important to note that high performance op-amps are more expensive. An MC33272 op-amp has a low input offset and low input bias current which is suitable for the two-op amp amplifier design. We can see that there is a tradeoff between accuracy and cost when designing a solution with the MPX2010. When designing a system based on the MPX2010, it is important to take into account errors due to parametric variation of the sensor (i.e., offset calibration, span calibration, TcS, TcO), power supply and the inherent errors of the amplification circuit. The offset and span errors greatly determines the resolution of the system (which adds to the system error). Even though the system offset error can be AN1668 Sensors Freescale Semiconductor 3 nulled out by auto-zeroing, these errors must be accounted for when setting the system gain (refer to AN1556 for more details). This forces the total span of the system to be smaller, because we must reserve an extra headroom from the total span to account for amplifier and A/D variations (i.e., amp. sat. voltage, power supply variation, A/D quantization error, and gain errors). If these errors are not accounted for, it could, for example, result in non-linearity errors if the sensor span or offset error causes the amplified output of the sensor to reach the saturation voltage of the amplifier. As an example, a MPX2010 sensor system is designed which has a range of 600 mm H2O FS range with a ±5% FS RMS error. The system uses a +5.0 V ±5% linear regulated power supply, a MC33272 dual op-amp and a 1% resistors. Table 3 shows the resulting specification and component values for the system based on MPX2010 sensor. Table 3. MPX2010 Sensor System Values MPX2010 Sensor Design Parameter VCC Differential Gain Vout_FS VREF Description Value Units Reg Power Supply 5 V Gain 433 V/V Full Scale Span 3.02 V Offset Reference 0.66 V Ω Ω Ω Ω Ω Ω Parts List U1A,U1B MC33272 Op-amp R1 Gain Resistor 39.2K R2 Gain Resistor 90.9 R3 Gain Resistor 909 R4 Gain Resistor 392K R + S1 Level Shift Resistor 1K R + S2 Level Shift Resistor 150 X1 MPX2010 Table 4. Performance Comparison between MPX2010 and MPXV4006G Solution Error Contribution MPX2010 Solution Error (FS = 600 mm H2O) MPXV4006G Solution Error (FS = 612 mm H2O) ± % FS ± mm H2O ± % FS ± mm H2O Max Sensor Error 7.19 43 3.00 18 System Resolution (A/D + Amplification) 1.30 8 0.80 5 System Error (Sensor + A/D + Amplification) 7.3 44 3.10 19 System Error with Auto-Zero 4.6 28 t3 t19 Note that the error due to system resolution is higher for the MPX2010 solution (± 2 bit A/D accuracy). This is because the MPX2010 span is limited as discussed above. Also, this accuracy assumes that the amplifier does not induce significant errors. As noted MPXV4006G sensor has better overall accuracy. The system resolution is very good because of its large span (4.6 V versus 3.0 V typical). SUMMARY Several washing machine solutions were examined. The MPX10/12 solution can be expensive in terms of additional support circuitry and the added time and labor involved during the calibration procedure. The MPX2010 is good alternative for high volume manufacturing because is already calibrated. With this solution, however, the system amplifier design must be chosen and designed carefully in order to minimize the system error. This is a consideration when deciding to implement a high accuracy solution with the MPX2010 because the cost of the system will go up. The MPXV4006G solution is geared towards high volume manufacturing because trimming, compensation and amplification is already on board. Besides the system simplicity and using less component, the resolution and overall accuracy of this solution is better than the MPX2010 solution. In some cases, less components can actually improve the reliability and manufacturability the system. REFERENCES [1] Benchmark of Washing Machine Mechanical Sensor, Jack Rondoni, Freescale Semiconductor, Inc. Internal Document. [2] Mechanical Sensor Characterization, Ador Reodique, Freescale Internal Document. [3] AN1551 Low Pressure Sensing with the MPX2010 Pressure Sensor, Jeff Baum, Freescale Application Note. [4] AN1636 Implementing Auto-Zero for Integrated Pressure Sensors, Ador Reodique, Freescale Application Note. [5] AN1556 Designing Sensor Performance Specifications for MCU-based Systems, Eric Jacobsen and Jeff Baum, Freescale Application Note. AN1668 4 Sensors Freescale Semiconductor NOTES AN1668 Sensors Freescale Semiconductor 5 How to Reach Us: Home Page: www.freescale.com Web Support: http://www.freescale.com/support USA/Europe or Locations Not Listed: Freescale Semiconductor, Inc. Technical Information Center, EL516 2100 East Elliot Road Tempe, Arizona 85284 +1-800-521-6274 or +1-480-768-2130 www.freescale.com/support Europe, Middle East, and Africa: Freescale Halbleiter Deutschland GmbH Technical Information Center Schatzbogen 7 81829 Muenchen, Germany +44 1296 380 456 (English) +46 8 52200080 (English) +49 89 92103 559 (German) +33 1 69 35 48 48 (French) www.freescale.com/support Japan: Freescale Semiconductor Japan Ltd. Headquarters ARCO Tower 15F 1-8-1, Shimo-Meguro, Meguro-ku, Tokyo 153-0064 Japan 0120 191014 or +81 3 5437 9125 [email protected] Asia/Pacific: Freescale Semiconductor Hong Kong Ltd. Technical Information Center 2 Dai King Street Tai Po Industrial Estate Tai Po, N.T., Hong Kong +800 2666 8080 [email protected] For Literature Requests Only: Freescale Semiconductor Literature Distribution Center P.O. Box 5405 Denver, Colorado 80217 1-800-441-2447 or 303-675-2140 Fax: 303-675-2150 [email protected] AN1668 Rev. 2 11/2006 Information in this document is provided solely to enable system and software implementers to use Freescale Semiconductor products. There are no express or implied copyright licenses granted hereunder to design or fabricate any integrated circuits or integrated circuits based on the information in this document. Freescale Semiconductor reserves the right to make changes without further notice to any products herein. Freescale Semiconductor makes no warranty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does Freescale Semiconductor assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation consequential or incidental damages. “Typical” parameters that may be provided in Freescale Semiconductor data sheets and/or specifications can and do vary in different applications and actual performance may vary over time. All operating parameters, including “Typicals”, must be validated for each customer application by customer’s technical experts. Freescale Semiconductor does not convey any license under its patent rights nor the rights of others. Freescale Semiconductor products are not designed, intended, or authorized for use as components in systems intended for surgical implant into the body, or other applications intended to support or sustain life, or for any other application in which the failure of the Freescale Semiconductor product could create a situation where personal injury or death may occur. Should Buyer purchase or use Freescale Semiconductor products for any such unintended or unauthorized application, Buyer shall indemnify and hold Freescale Semiconductor and its officers, employees, subsidiaries, affiliates, and distributors harmless against all claims, costs, damages, and expenses, and reasonable attorney fees arising out of, directly or indirectly, any claim of personal injury or death associated with such unintended or unauthorized use, even if such claim alleges that Freescale Semiconductor was negligent regarding the design or manufacture of the part. Freescale™ and the Freescale logo are trademarks of Freescale Semiconductor, Inc. All other product or service names are the property of their respective owners. © Freescale Semiconductor, Inc. 2006. All rights reserved.