MCP3909 / dsPIC33FJ128GP206 3-Phase Energy Meter Reference Design (c) 2009 Microchip Technology Inc. DS51823A Note the following details of the code protection feature on Microchip devices: * Microchip products meet the specification contained in their particular Microchip Data Sheet. * Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the intended manner and under normal conditions. * There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchip's Data Sheets. Most likely, the person doing so is engaged in theft of intellectual property. * Microchip is willing to work with the customer who is concerned about the integrity of their code. * Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not mean that we are guaranteeing the product as "unbreakable." Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of our products. Attempts to break Microchip's code protection feature may be a violation of the Digital Millennium Copyright Act. If such acts allow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act. Information contained in this publication regarding device applications and the like is provided only for your convenience and may be superseded by updates. It is your responsibility to ensure that your application meets with your specifications. MICROCHIP MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND WHETHER EXPRESS OR IMPLIED, WRITTEN OR ORAL, STATUTORY OR OTHERWISE, RELATED TO THE INFORMATION, INCLUDING BUT NOT LIMITED TO ITS CONDITION, QUALITY, PERFORMANCE, MERCHANTABILITY OR FITNESS FOR PURPOSE. Microchip disclaims all liability arising from this information and its use. Use of Microchip devices in life support and/or safety applications is entirely at the buyer's risk, and the buyer agrees to defend, indemnify and hold harmless Microchip from any and all damages, claims, suits, or expenses resulting from such use. No licenses are conveyed, implicitly or otherwise, under any Microchip intellectual property rights. Trademarks The Microchip name and logo, the Microchip logo, dsPIC, KEELOQ, KEELOQ logo, MPLAB, PIC, PICmicro, PICSTART, rfPIC and UNI/O are registered trademarks of Microchip Technology Incorporated in the U.S.A. and other countries. FilterLab, Hampshire, HI-TECH C, Linear Active Thermistor, MXDEV, MXLAB, SEEVAL and The Embedded Control Solutions Company are registered trademarks of Microchip Technology Incorporated in the U.S.A. Analog-for-the-Digital Age, Application Maestro, CodeGuard, dsPICDEM, dsPICDEM.net, dsPICworks, dsSPEAK, ECAN, ECONOMONITOR, FanSense, HI-TIDE, In-Circuit Serial Programming, ICSP, Mindi, MiWi, MPASM, MPLAB Certified logo, MPLIB, MPLINK, mTouch, Octopus, Omniscient Code Generation, PICC, PICC-18, PICDEM, PICDEM.net, PICkit, PICtail, PIC32 logo, REAL ICE, rfLAB, Select Mode, Total Endurance, TSHARC, UniWinDriver, WiperLock and ZENA are trademarks of Microchip Technology Incorporated in the U.S.A. and other countries. SQTP is a service mark of Microchip Technology Incorporated in the U.S.A. All other trademarks mentioned herein are property of their respective companies. (c) 2009, Microchip Technology Incorporated, Printed in the U.S.A., All Rights Reserved. Printed on recycled paper. Microchip received ISO/TS-16949:2002 certification for its worldwide headquarters, design and wafer fabrication facilities in Chandler and Tempe, Arizona; Gresham, Oregon and design centers in California and India. The Company's quality system processes and procedures are for its PIC(R) MCUs and dsPIC(R) DSCs, KEELOQ(R) code hopping devices, Serial EEPROMs, microperipherals, nonvolatile memory and analog products. In addition, Microchip's quality system for the design and manufacture of development systems is ISO 9001:2000 certified. DS51823A-page 2 (c) 2009 Microchip Technology Inc. MCP3909 / dsPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Table of Contents Preface ........................................................................................................................... 7 Introduction............................................................................................................ 7 Document Layout .................................................................................................. 8 Conventions Used in this Guide ............................................................................ 9 Recommended Reading...................................................................................... 10 The Microchip Web Site ...................................................................................... 10 Customer Support ............................................................................................... 10 Document Revision History ................................................................................. 10 Chapter 1. Meter Overview 1.1 Introduction ................................................................................................... 11 1.2 Meter Design Parameters ............................................................................ 11 1.3 Power Calculations ....................................................................................... 12 1.4 Getting Started ............................................................................................. 13 Chapter 2. Hardware Description 2.1 Overview ...................................................................................................... 17 2.2 Analog Front End Circuitry ........................................................................... 18 2.3 Analog-To-Digital Conversion ...................................................................... 20 2.4 dspic33f Hardware Connection And Peripheral Usage ................................ 22 2.5 Power Supply ............................................................................................... 25 Chapter 3. Firmware 3.1 Overview ...................................................................................................... 27 3.2 Main Loop ..................................................................................................... 27 3.3 Calculation() - Calculating Electrical Parameters ......................................... 29 3.4 ADC Sampling Scheme For Calculations .................................................... 33 3.5 ReadING A/D Data Of The MCP3909 Device .............................................. 35 3.6 Communication Of UART Interface .............................................................. 37 3.7 Resource Configuration ................................................................................ 37 3.8 Description Of Project Files .......................................................................... 38 Chapter 4. Meter Calibration 4.1 Introduction ................................................................................................... 39 4.2 Current/voltage Calibration .......................................................................... 39 4.3 Apparent Power Calibration ......................................................................... 40 4.4 Phase Lag Calibration ................................................................................. 41 (c) 2009 Microchip Technology Inc. DS51723A-page 3 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design Chapter 5. PC Software 5.1 Overview And Installation ............................................................................. 43 5.2 Establish Communication ............................................................................. 44 5.3 Basic Parameters Output Screen ................................................................. 45 5.4 Phase A/B/C Harmonic Output Screen ........................................................ 45 5.5 Distortion Rate .............................................................................................. 46 5.6 Harmonic Power ........................................................................................... 46 5.7 Energy Accumulation ................................................................................... 47 5.8 Calibration Step 1 - Reset All Calibration ..................................................... 47 5.9 Linearity Calibration ...................................................................................... 48 5.10 Apparent Power Calibration ....................................................................... 49 5.11 Phase Lag Calibration ................................................................................ 50 Chapter 6. Meter Communications Protocol 6.1 Introduction ................................................................................................... 51 6.2 Test Connection Command .......................................................................... 52 6.3 Total Data Request ...................................................................................... 52 6.4 Status Register ............................................................................................. 54 6.5 Harmonic Content Command ....................................................................... 54 6.6 Total Harmonic Distortion (THD) Command ................................................ 55 6.7 Start Energy Measurement Command ......................................................... 56 6.8 Stop Energy Measurement Command ......................................................... 56 6.9 Harmonic Power Command ......................................................................... 57 6.10 Calibrate Meter Voltage/current Command ................................................ 58 6.11 Calibrate Phase Lag Command ................................................................. 59 6.12 Calibrate Apparent Power Command ......................................................... 59 6.13 Calibrate Energy Pulse Command ............................................................. 60 6.14 Reset All Meter Calibration Values Command ........................................... 60 6.15 Calibrate Meter Constant (Energy Pulse Output Constant) ....................... 61 Appendix A. Schematics and Layouts A.1 Introduction .................................................................................................. 63 A.2 Schematics And Pcb Layout ........................................................................ 63 Appendix B. Bill Of Materials (BOM) Appendix C. Power Calculation Theory C.1 Overview ...................................................................................................... 75 C.2 Synchronous Sampling And Quasi-synchronous Sampling ........................ 75 C.3 The Harmonic Analysis Algorithm Of Quasi-synchronous Sampling ........... 82 C.4 Measuring The Voltage/current Rms Value And Power Using Quasi-synchronous Sampling Algorithm ........................................................................ 84 C.5 Measuring Frequency .................................................................................. 87 C.6 Improving Measurement Precision Of Quasi-synchronous Sampling Algorithm ................................................................................................................. 89 C.7 Measuring Secondary Parameters .............................................................. 91 C.8 Apparent Power Of Each Phase And Total Apparent Power ....................... 91 C.9 Power Factor Of Each Phase And Total Power Factor ............................... 91 DS51723A-page 4 (c) 2009 Microchip Technology Inc. C.10 Active Energy And Reactive Energy .......................................................... 92 C.11 Positive/negative Active Energy, Positive/negative Reactive Energy And Four-quadrant Reactive Energy ............................................................. 92 C.12 Harmonic Components Of Current, Voltage And Total Harmonic Distortion ................................................................................................................. 94 C.13 Compensation For Ratio Error And Phase Lag ......................................... 95 C.14 Relationship Between Error And Current ................................................... 96 C.15 Ratio Error Compensation ......................................................................... 97 C.16 Phase Lag Compensation ......................................................................... 98 Appendix D. 50/60 Hz Meter Operation D.1 Firmware Versions ..................................................................................... 103 Worldwide Sales and Service .................................................................................. 104 (c) 2009 Microchip Technology Inc. DS51723A-page 5 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design NOTES: DS51723A-page 6 (c) 2009 Microchip Technology Inc. MCP3909 / dsPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Preface NOTICE TO CUSTOMERS All documentation becomes dated, and this manual is no exception. Microchip tools and documentation are constantly evolving to meet customer needs, so some actual dialogs and/or tool descriptions may differ from those in this document. Please refer to our web site (www.microchip.com) to obtain the latest documentation available. Documents are identified with a "DS" number. This number is located on the bottom of each page, in front of the page number. The numbering convention for the DS number is "DSXXXXXA", where "XXXXX" is the document number and "A" is the revision level of the document. For the most up-to-date information on development tools, see the MPLAB(R) IDE on-line help. Select the Help menu, and then Topics to open a list of available on-line help files. INTRODUCTION This chapter contains general information that will be useful to know before using the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design. Items discussed in this chapter include: * * * * * * Document Layout Conventions Used in this Guide Recommended Reading The Microchip Web Site Customer Support Document Revision History (c) 2009 Microchip Technology Inc. DS51823A-page 7 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design DOCUMENT LAYOUT This document describes how to use the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design as a development tool to emulate and debug firmware on a target board. The manual layout is as follows: This document describes how to use the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design as a development tool. The manual layout is as follows: * Chapter 1. "Meter Overview" - Summarizes the meter specifications and a quick getting started section * Chapter 2. "Hardware Description" - A detailed explanation of the different circuit blocks, their function, and implementation * Chapter 3. "Firmware" - All the calculations performed by the dsPIC33F are described here * Chapter 4. "Meter Calibration" - Explains how the meter is calibrated to accuracy * Chapter 5. "PC Software" - Includes screen shots of the viewer/calibration software included with the system * Chapter 6. "Meter Communications Protocol" - The UART commands used to communicate to the meter * Appendix A. "Schematics and Layouts" - Both PCB, SCH files are located here for the 2 board system * Appendix B. "Bill Of Materials (BOM)" - Part number and ordering information for all components of the energy meter * Appendix C. "Power Calculation Theory" - A detailed explanation of the theory behind the calculations described in Chapter 3. "Firmware" * Appendix D. "50/60 Hz Meter Operation" - Instructions on converting the meter for use in a 60 Hz line frequency environment DS51823A-page 8 (c) 2009 Microchip Technology Inc. Preface CONVENTIONS USED IN THIS GUIDE This manual uses the following documentation conventions: DOCUMENTATION CONVENTIONS Description Arial font: Italic characters Represents Examples Referenced books Emphasized text A window A dialog A menu selection A field name in a window or dialog A menu path MPLAB(R) IDE User's Guide ...is the only compiler... the Output window the Settings dialog select Enable Programmer "Save project before build" A dialog button A tab A number in verilog format, where N is the total number of digits, R is the radix and n is a digit. A key on the keyboard Click OK Click the Power tab 4`b0010, 2`hF1 Italic Courier New Sample source code Filenames File paths Keywords Command-line options Bit values Constants A variable argument Square brackets [ ] Optional arguments Curly brackets and pipe character: { | } Ellipses... Choice of mutually exclusive arguments; an OR selection Replaces repeated text #define START autoexec.bat c:\mcc18\h _asm, _endasm, static -Opa+, -Opa0, 1 0xFF, `A' file.o, where file can be any valid filename mcc18 [options] file [options] errorlevel {0|1} Initial caps Quotes Underlined, italic text with right angle bracket Bold characters N`Rnnnn Text in angle brackets < > Courier New font: Plain Courier New Represents code supplied by user (c) 2009 Microchip Technology Inc. File>Save Press <Enter>, <F1> var_name [, var_name...] void main (void) { ... } DS51823A-page 9 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design RECOMMENDED READING This user's guide describes how to use MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design. Other useful documents are listed below. The following Microchip documents are available and recommended as supplemental reference resources. MCP3909 Data Sheet, "Energy Metering IC with SPI Interface and Active Power Pulse Output" (DS22025) This data sheet provides detailed information regarding the MCP3909 device. AN994 Application Note "IEC61036 Meter Design using the MCP3905/6 Energy Metering Devices" (DS00994) This application note documents the design decisions associated with using the MCP390X devices for energy meter design and IEC compliance. THE MICROCHIP WEB SITE Microchip provides online support via our web site at www.microchip.com. This web site is used as a means to make files and information easily available to customers. Accessible by using your favorite Internet browser, the web site contains the following information: * Product Support - Data sheets and errata, application notes and sample programs, design resources, user's guides and hardware support documents, latest software releases and archived software * General Technical Support - Frequently Asked Questions (FAQs), technical support requests, online discussion groups, Microchip consultant program member listing * Business of Microchip - Product selector and ordering guides, latest Microchip press releases, listing of seminars and events, listings of Microchip sales offices, distributors and factory representatives CUSTOMER SUPPORT Users of Microchip products can receive assistance through several channels: * * * * Distributor or Representative Local Sales Office Field Application Engineer (FAE) Technical Support Customers should contact their distributor, representative or field application engineer (FAE) for support. Local sales offices are also available to help customers. A listing of sales offices and locations is included in the back of this document. Technical support is available through the web site at: http://support.microchip.com DOCUMENT REVISION HISTORY Revision A (November 2009) * Initial Release of this Document. DS51823A-page 10 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 1. Meter Overview 1.1 INTRODUCTION The MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design is a fully functional energy meter with many advanced features such as harmonic analysis, per phase distortion information, voltage sag detection, four quadrant energy measurement, and active and reactive power calculation. It uses Microchip's powerful 16-bit dsPIC33F Microcontroller Unit (MCU). The MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design is unique in the fact that all calculations take advantage of the dsPIC33F DSP engine, and all output quantities are calculated in the frequency domain through the use of direct fourier transforms (DFT). This approach yields a large volume of outputs for a variety of meter designs, from simple active power only energy meters, to advanced energy meters requiring harmonic analysis. Another significant advantage of this design, is that the dsPIC firmware implements a quasi-synchronous sampling algorithm, eliminating the need for external zero-crossing detection and PLL (Phase Locked Loop) circuit for the synchronization of ADC samples to line frequency. The line frequency is measured in software and corrected for measurement errors caused by frequency fluctuations in the power grid. This additional processing on the dsPIC reduces the overall meter cost by eliminating the requirement for a PLL circuit. 1.2 METER DESIGN PARAMETERS * * * * * * * * * * * * * * * * Accuracy Class: 0.2S Rated Current Ib: 3 X 5(20)A 3-phase 4-wire System Line Frequency Range: 47-53 Hz or 57-63 Hz (firmware option, see Appendix D. "50/60 Hz Meter Operation") ADC Sampling Rate: 12.8 ksps to 3.2 ksps Voltage Input: - 3 x 220/380V - 3 x 57.7/100V (3-Phase, 4-Wire) Starting Current: 0.001 IB Active Power Measurement Range: 0-13200W, Precision Class: 0.2. Reactive Power Measurement Range: 0-13200VAR, Precision Class: 0.2. Power Factor (PF) Precision Class: 0.2. Frequency Measurement: Precision Class: 0.2, Max. Error 0.1 Hz Harmonic Component Measurement of Voltage Input: 2ND-31ST Harmonic Harmonic Component Measurement of Current Input: 2ND-31ST Harmonic Creeping: Anti-creeping Design (<0.0008 IB) Two Pulse Outputs: Total Phase Active Power, Total Phase Reactive Power Pulse Constant: 3200 Imp/kWh (c) 2009 Microchip Technology Inc. DS51723A-page 11 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 1.3 POWER CALCULATIONS A summary of all the calculations performed by this energy meter are summarized below. Chapter 3. "Firmware" provides an explanation on the firmware implementation, Appendix C. "Power Calculation Theory" is included to show the theory behind this firmware. * * * * * * * * * * * * * * * * * * * * * * DS51723A-page 12 Power Grid Frequency RMS Voltage Of Each Phase RMS Current Of Each Phase RMS Neutral Current Active Power Of Each Phase Reactive Power Of Each Phase Apparent Power Of Each Phase Power Factor Of Each Phase Fundamental Active Power Of Each Phase Fundamental Reactive Power Of Each Phase Harmonic Active Power Of Each Phase Harmonic Reactive Power Of Each Phase Total Active Power: - The Algebraic Sum Of Active Power Of Three Phases Total Reactive Power: - The Algebraic Sum Of Reactive Power Of Three Phases Total Apparent Power: - The Algebraic Sum Of Apparent Power Of Three Phases Total Power Factor Phase Missing / Line voltage sag detection and alarm Total Active Energy: - The Algebraic Sum Of Positive/negative Active Energy Positive/negative Active Energy Positive/negative Reactive Energy Four-quadrant Reactive Energy Voltage/current Harmonic Content Of Each Phase (c) 2009 Microchip Technology Inc. Meter Overview . Current Transformer UART Interface ICD Interface dsPIC33 FIGURE 1-1: 1.4 MCP3909 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design. GETTING STARTED To describe how to use the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design, the following example is given using a 4-wire, 3-phase, 220VAC line voltage and connection. The rated current of the energy meter is 5(20)A. The energy meters are not shipped fully calibrated, and a full calibration should be performed to show the true meter accuracy. See Chapter 4. "Meter Calibration" for more information. All connections described in this section are dependent on the choice of current sensing element and a secondary external transformer may be required in higher current meter designs. (c) 2009 Microchip Technology Inc. DS51723A-page 13 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design Step 1: Connect the meter to 220V line and load The diagram below shows where the voltage and current connections should be made. It is not required to connect all 3 phases for the meter to be operational. 220VAC should be placed between either VA, VB, VC and NIN, NOUT. The AC load for a given phase should then be connected to the IIN and IOUT of a given phase. IAIN VA IAOUT FIGURE 1-2: IBIN VB IBOUT ICIN VC ICOUT Meter Case Bottom. Step 2: Turn On Line/Load Power to the Meter Turn on the power to the energy meter. D1 should be lit showing the meter has power. At this point, if a load is connected and the meter is measuring power, the power LED, D1, should be blinking. DS51723A-page 14 (c) 2009 Microchip Technology Inc. Meter Overview Step 3: Connect the RS-232 Cable 1. Connect the RS-232 cable from the energy meter to a Personal Computer (PC), using either COM1, COM2, or COM6. Step 4: Run the PC Calibration Software After installing and running the PC energy meter software on a PC running a WindowsTM Operating System, and selecting the proper comm port for RS-232 communication, the following screen should show real-time meter results. The following chapters include more detail on the firmware, calculation, and PC software. . FIGURE 1-3: "PM_Viewer" or Power Meter Viewer PC Software. (c) 2009 Microchip Technology Inc. DS51723A-page 15 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design NOTES: DS51723A-page 16 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 2. Hardware Description 2.1 OVERVIEW Figure 2-1 is the basic hardware block diagram of the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design. The hardware includes the dsPIC33F and ICD 2 interface, analog signal conditioning for 3-phase voltage/current inputs and current using the MCP3909 Energy Meter IC, neutral current measurement using external op-amp on-board dsPIC33F ADC, UART interface to PC, ICD2 interface for MCU programming, and power supply circuits. Note there are two PCBs comprising this energy meter, the power supply PCB, and the MCU/AFE PCB. Refer to Appendix A. "Schematics and Layouts" for more information. Three-phase voltage and current signals are connected to the meter through transformers, and connected to the MCP3909 A/D converter through a simple signal conditioning circuit. The MCP3909 device samples the signal and performs the analog-to-digital conversion (ADC). The MCP3909 device sends the digital conversion results to the dsPIC device via the SPI interface. The MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design uses a quasi-synchronous sampling algorithm via the dsPIC33F, therefore eliminating the need for voltage zero-crossing detect and clock generating circuit, which are otherwise needed in a synchronous sampling algorithm. The clock of the MCP3909 device is an active external 3.2768 MHz crystal. . Phase A CT Phase C CT FIGURE 2-1: (c) 2009 Microchip Technology Inc. MCP3909 Phase C PT 3.2768 MHz ADC UART I/O dsPIC33FJ64GP206 SPI SPI +5V Power Supply RS232 ICD2 Interface Phase B CT CLK Gain Control MCP3909 Phase B PT MCP3909 Phase A PT Op Amp UART Interface Neutral line CT +3.3V Power Supply Hardware Block Diagram. DS51723A-page 17 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 2.2 ANALOG FRONT END CIRCUITRY For safety, current transformers are used between the voltage and current input signals and the measurement module to isolate it from the 3-phase power supply. The transformer used for the voltage path is a 1:1 transformer, SPT204B from Beijing Singure Measurement & Control Technology Co., with non-linearity less than 0.1%, and rated input/output current of 2 mA/2 mA. The transformer used on the current side is a SPT254FK, from Yhehua Shanghai, with rated input /output current of 20A/2 mA, non-linearity less than 0.1%, and linear range of 0-20 A. See Appendix B. "Bill Of Materials (BOM)" for more information on the input circuitry. Using phase A as an example for the voltage signal path, a 150 k (R1) resistor is used before the CT to transform the signal to an appropriate current. After the CT, burden resistors R125 and R126 are needed to transform the current signal to a differential voltage signal for the MCP3909 device to sample. Signals are coupled into the MCP3909 device's signal input port via R110 and R111. C111 and C112 are used to filter high-frequency signals. Using phase A as an example for the current signal path, transformation of the current signal is similar to that of the voltage signal. The burden resistors R125/R126 and R116/R117 are chosen to be 20 for the current channel and 100 for the voltage channel after considering the following 3 factors: * The MCP3909 device's differential voltage input range: 1V for voltage channel and 0.705V for current channel * Maximum current/voltage for the meter: rated current of 5A, maximum current of 20A, and maximum voltage of 300V) * Transformer ratio for the current and voltage transformer. A non-isolated voltage input circuit is included. In practice, a voltage divider network of resistors is often used for sampling AC voltage input. This measuring method is therefore included in the hardware design. In Figure 2-2, voltage divider resistors R3, R4, and sampling resistor R1 construct a network for sampling AC voltage. CTA-1 1.0 k SCT220B CH0+ R125 CURRENT R126 R110 20 20 1 nF C111 1.0 k CTA-2 CH0T103 150 k PA R111 1.0 k SPT204B R1 CH1+ R116 VOLTAGE 1 nF C112 R117 R108 100 100 1 nF C109 1.0 k N CH1+ T1 R109 Jumper 499 k 499 k R4 R3 1 nF C110 MCP3909 PA VOLTAGE (Non-Isolated Option) R1 1 k C3 33 nF N FIGURE 2-2: DS51723A-page 18 Input Signal Conditioning Circuit (Phase A). (c) 2009 Microchip Technology Inc. Hardware Description 2.2.1 Burden Resistor Temperature Coefficient The high precision class 0.2S requirement for the energy meter makes it crucial to select proper burden resistors for the output of the current transformers. Metal film resistors with low inherent noise and temperature coefficient are ideal. Given that the secondary current of the CT is I, then the input voltage of the MCP3909 device is U = IR, where R is the resistance of R125 and R126 (Using Phase A as an example). If the temperature varies by T, and the temperature coefficient of sampling resistor R is ppm/C, then the output voltage is: EQUATION 2-1: U' = I ( R + T x x R ) The voltage variation is: EQUATION 2-2: U = U' - U = I x T x x R = U x T x This relationship shows that the output voltage variation caused by temperature variation is in proportion to the temperature coefficient of the burden resistor. In addition, a smaller temperature coefficient benefits meter start stabilization after startup. It takes a longer time for resistors with larger temperature coefficient to stabilize. Therefore, accurate measurements would require a longer wait after power-up. This affects the efficiency, or speed of meter calibration. (c) 2009 Microchip Technology Inc. DS51723A-page 19 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 2.3 ANALOG-TO-DIGITAL CONVERSION This meter design uses Microchip's MCP3909 energy meter ICs. Small-signal current inputs can be amplified by the programmable gain amplifier inside the MCP3909 device. The programmable range for the MCP3909's PGA is 16:1 V/V. The MCP3909 PGA gain can be configured by G0 and G1 (Pin 15 and 16 of the dsPIC33F). The MCP3909 device's clock is provided by a 3.3768 MHz active crystal (for 50 Hz system, see Appendix D. "50/60 Hz Meter Operation" for 60 Hz line frequency information). The MCP3909 device's output data rate is 12.8 ksps. The clock lines and MCLR lines of all 3 MCP3909 devices are connected together, which ensures that the 3 phases are strictly synchronous. 3.3768 MHz (50 Hz Version) X100 MCLK input MCP3909 SDO SDO MCP3909 SDO MCP3909 To dsPIC33F IRQ IC1 (Input Capture) DR Pulse fSAMPLE1 = 12.8 ksps (Active Power) fSAMPLE2 = 6.4 ksps (Reactive Power, RMS Current / Voltage) fSAMPLE3 = 3.2 ksps (Harmonic Analysis, Distortion) tLINE_CYC Phase A,B,C I & V Data SDO DR 16 bits x 6 ADCs DR tSAMPLE FIGURE 2-3: DS51723A-page 20 Clock Generation, Sampling Times and Calculation Frequencies. (c) 2009 Microchip Technology Inc. Hardware Description 2.3.1 Samples And Processing Input capture IC1 on the dsPIC33F is used to detect if A/D conversion is complete. However, not all MCP3909 device samples are stored in the MCU, depending on the parameter being calculated. The ADC conversion rate of the MCP3909 device is determined by the frequency of the master clock (3.378 MHz for the case of a 50 Hz line), and the output data rate is MCLK/256 or 12.8 ksps. After each conversion is complete, a Data Ready signal is generated by the SDO of the MCP3909 device. The signal is fed into IC1, allowing the Interrupt Service Routine (ISR) of IC1 to read the data. When the MCP3909 device outputs data, it first sends an ADC result of the voltage channel, then an ADC result of the current channel, with MSB first. As noted, not all MCP3909 device samples are used for calculating all the parameters. In practice, 6.4 ksps sampling rate is required, which means only 1 output data is used for every 2 data sampled. For 50 Hz input signal, 6.4 ksps sampling rate will take 128 samples for each cycle. For example, the active power metering is computed based on this condition. But for other parameters for which precision is not critical, such as reactive energy, voltage, current and frequency, the sampling rate may be reduced to save data storage space and processing time. In this design, the 3.2 ksps sampling rate is used, which means only 1 result is stored for every 4 ADC conversions. After each conversion, a positive pulse with the width of 4 clock cycles is output by the SDO pin of the MCP3909 device. IC1 is used to detect the falling edge of the pulse and generate an interrupt for every 2 falling edges, i.e., 1 data is read for every 2 conversions, thus realizing 6.4 ksps sampling rate. (c) 2009 Microchip Technology Inc. DS51723A-page 21 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 2.4 DSPIC33F HARDWARE CONNECTION AND PERIPHERAL USAGE Table 2-1 is the pin allocation for the dsPIC33FJ64GP206 MCU. TABLE 2-1: dsPIC Pin FUNCTIONAL ALLOCATION FOR dsPIC33F PINS Pin Function Name Corresponding Name in Schematic Diagram Functional Description 62, 1 RG14, RG15 G0A, G1A MCP3909's gain control for Phase A 2, 3 RC1, RC2 G0B, G1B MCP3909's gain control for Phase B 13, 12 RB3, RB4 G0C, G1C MCP3909's gain control for Phase C 14 RB2 CSA MCP3909's chip-select signal corresponding to Phase A 15 RB1 CSB MCP3909's chip-select signal corresponding to Phase B 16 RB0 CSC MCP3909's chip-select signal corresponding to Phase C 4, 5, 6 SCK, SDI, SDO SPI I/F Interface signal of SPI2. SPI interface operates under master mode, used for the MCP3909 device communications 42 INT1 SDO SDO line of SPI interface, for detecting MCP3909's A/D conversion complete flag 49-51 RD1-RD3 PULSE1, PULSE2, PULSE3 PULSE1 is total phase active power pulse output, PULSE2 is total phase reactive power pulse output. PULSE3's function is to be determined. 61 RG0 AD_MCLR Master clear signal of 3 MCP3909 devices (tied together) 53 RC13, RC14 LED1, LED2 LED drive pins, can be used as energy pulse output indicator, for meter calibration. Its function is similar to those of PULSE1 and PULSE2 36, 37 SDA1 / SCL1 SDA/SCL I2CTM interface, used to read/write EEPROM externally 33, 34 U1TX / U1RX RF3/RF2 UART interface 33, 34, SDO1/SDI1/SCK1 35 RF3/RF2/RF6 SPI1 interface, can be designed by customers, used for communication with host MCU to obtain measurement results and calibrate a meter. Its function is the same as UART interface, but have a faster communication rate and higher efficiency. SPI operates in slave mode. If UART interface is used to communicate with host MCU, then this interface cannot be used. 27 AN12 Current_N Detect neutral current 28 AN13 Ref_V Detect boost voltage of neutral current 17, 18 ICSPCLK, ICSPDAT ICSP I/F Online debugging/programming interface 7 MCLR Master clear input MCLR DS51723A-page 22 (c) 2009 Microchip Technology Inc. Hardware Description 2.4.1 UART and SPI1 Interface The UART and SPI1 interfaces are multiplexed. Through the UART or SPI1 interface, the host MCU can communicate with the metering front-end to perform calibration or obtain metering results. The SPI interface may also be used if high-speed data transfer is desired. In this case, the SPI interface of the dsPIC device works in the slave mode. The UART and SPI1 share a common pin, so only one of the two interfaces can be used at a time. Since the reference design uses a PC to simulate the host MCU, the UART interface is chosen as the communication interface. SPI and RS232 interfaces are not isolated from the PC. A general-purpose transceiver device, MAX232, is used for the UART interface. 2.4.2 Energy Pulse Output Interface Three sets of outputs for energy measurement pulses are available in this design, corresponding to the I/O pins of RD1-RD3. Two of them, output total active energy and total reactive energy, respectively, and the other is not yet specified. Outputs are isolated by a photo-electronic coupling device, U3. The photo-electronic coupler is active when corresponding I/O pin is high. In addition, the design also provides two sets of LED outputs for energy meter calibration. The output pins for these LEDs are RC13 and RC14. The LED is on when the output is low. Figure 2-4 is the circuit of energy pulse output interface. 1 k RD3 R301 Total Active 1 k RD2 R301 Total Reactive 1 k RD1 R301 (not used) 3.3V 470 R314 D303 Total Active RC13 3.3V 470 R310 Total Reactive D302 RC14 FIGURE 2-4: (c) 2009 Microchip Technology Inc. Energy Output Pulse Configuration. DS51723A-page 23 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 2.4.3 Neutral Current Detection Detection of neutral wire tampering is performed by the on-chip A/D of the dsPIC device. The purpose of the detection is to prevent electricity theft, balance 3-phase signals and detect electricity leakage. Since the precision is not critical, a dsPIC on-chip A/D is sufficient. Figure 2-5 shows the circuit for neutral current detection. The neutral input uses R128 for sampling. To bias the AC signals to the A/D measurement range, a 3.3V power supply is divided by R130 and R129 and connected to the emitter-follower of the MCP6002 device to output a boost voltage REV_V of 1.65V. The biased voltage is then connected to the CT in series. The CT's sampling voltage is connected to Op-amp B of the MCP6002 device via R127 as emitter-follower output, generating the sampling voltage, Current_N, for the current. Both Current_N and 1.65V VREF signals are sampled and measured by the dsPIC on-chip A/D. . Neutral Wire Input T100 Secondary 3.3V R130 4.7 k U101_A + 470 4.7 k R128 R127 U101_B + Current_N R129 4.7 k 1.65V VREF FIGURE 2-5: DS51723A-page 24 Circuit of Neutral Line Detection. (c) 2009 Microchip Technology Inc. Hardware Description 2.5 POWER SUPPLY The power supply used in this design provides 3.3V and 5V. Since the energy meter for a 3-phase 4-wire system is required to operate properly when any one phase is active, a switching power supply module is used for convenience. T4 is the switching power supply module. Prior to the input of this module, additional protection circuitry is included with the meter design. Figure 2-6 shows the input to the switching power supply module and the additional filtering and protection circuitry. In Figure 2-6, R5 is the integrated ferrite bead, C1, C4, C6, RV1, RV2 and RV3 are CBB capacitors and varistors. They are used to improve anti-surge performance of the system. . 1 R5 JP4 1 JP5 1 1 2 3 4 A B C N A B C N 8 7 6 5 PA PB PC RV1 JP6 C1 0.1U COILS 1 RV2 C4 0.1u RV3 C6 0.1u N N 12V T4 N PC PB PA FIGURE 2-6: 1 2 3 4 N Vo2 C G2 B Vo1 A G1 8 7 6 + C2 100uf JP6 1 2 3 Header 3 5 Switching Power Supply Module (T4) and Additional Input Protection Circuitry. (c) 2009 Microchip Technology Inc. DS51723A-page 25 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design An LDO is connected in series at the output of the switching power supply to obtain a more stable power supply. Figure 2-7 shows this circuit. Microchip's MCP1701 device and MCP1700 device, low drop-out high efficiency LDOs, are selected for use. D301 is the power LED for the meter and is active when the meter is connected to the proper line input voltage. J1 MCP1701(5V-SOT89) L302 2 1 2 5V_IN INDUCTOR L301 INDUCTOR R316 470 C322 CAP D301 LED 3 DS51723A-page 26 Vout C336 0.1uF + C338 100uf U306 MCP1700(3.3V-SOT23) 2 Vin Vout Gnd 1 C323 CAP FIGURE 2-7: Vin VSS 1 + C337 100uf 5V 3 3.3V C324 CAP + C339 47uF 5V and 3.3V LDO Modules. (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 3. Firmware 3.1 OVERVIEW This section discusses the dsPIC firmware structure, peripheral resources, important program flows, and explanations of project files included in the firmware zip files included with the system. See Appendix D. "50/60 Hz Meter Operation" for converting to 60 Hz code. * * * * * 3.2 Calculate all electrical parameters in frequency domain MCP3909 device communication Detect voltage/current phase order, and determine missing phases Generation imp/kWh power pulse UART communication MAIN LOOP The main loop of the entire dsPIC33F program is shown in Figure 3-1. Main Program Initialize on-chip peripherals and variables and MCP3909 device Process UART comm. commands Calculation() Sampled 3 cycles? Yes Compute elec. parameters Yes Compute neutral line current No Neutral data acquisition complete? No Clear WDT FIGURE 3-1: (c) 2009 Microchip Technology Inc. Main Loop Chart. DS51723A-page 27 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design After system power-up, the MCU enters the initialization process, which includes proper configuration of the I/O ports and on-chip peripherals (such as timer, UART, SPI and IC). At the same time, the system control parameters can be loaded from the external EEPROM and variables are all initialized. Since most tasks of this system are accomplished through interrupts, only three tasks are carried out in the system main loop, which are interpreting/processing UART communication protocol, calculating parameters, and detecting neutral current. The UART communication is performed in function UART_process(), the executing frequency of which depends on the polling frequency of the upper computer. Parameters are calculated by the function Calculation(), which is executed once every 3 cycles of the power grid. Neutral current is detected by function ComputeNeutralCurrent(), and computing is performed once every 16 cycles of the power grid. DS51723A-page 28 (c) 2009 Microchip Technology Inc. Firmware 3.3 CALCULATION() - CALCULATING ELECTRICAL PARAMETERS All power parameters are calculated with the function Calculation(), which is executed once every 3 cycles of the power grid. As shown in Figure 3-2, all calculations are performed post DFT (direct fourier transform), in the frequency domain. * * * * * * * * * * * RMS Voltage/current Of Each phase Phase Angle Measuring Line Frequency Active, Reactive, Apparent Power Of Each Phase Positive and Negative Active Power Positive/negative Reactive Power 4-quadrant Reactive Energy Total Active Power, Total Reactive Power, and Total Apparent Power Total Power Factor Voltage and Current Distortion Of Each Phase Voltage and Current Harmonic Contents Of Each Phase Note: Algorithms for all calculations are shown in Appendix C. "Power Calculation Theory". Calculate function Select sync. window function and sine/cosine table phase sequence 1~3 loop Voltage signal and sync. window process DTF transformation Calculate voltage RMS value Calculate current RMS value Calculate current harmonic Active/reactive power calculation and compensation phase sequence 1~3 end Calculate combined power Combined energy accumulation Calculate voltage harmonic Current signal and sync. window process DTF transformation Calculate frequency, determine phase sequence Update loop array pointer and data length End of function calculation FIGURE 3-2: (c) 2009 Microchip Technology Inc. Calculation() Flow Chart. DS51723A-page 29 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design Each block can be categorized into one of three types of calculations: 1. Calculation for an individual phase 2. Calculation for all 3 phases 3. Calculation of power accumulation (Energy) 3.3.1 Process Quasi-Synchronization Window The first blocks of the calculation flow are to determine how many samples to use for the quasi-synchronization sampling algorithm. See Appendix C. "Power Calculation Theory" for more information on this approach. The firmware selects the proper quasi-synchronization window function (array of data) and corresponding sine/cosine table according to the number of sampling points in present current cycle. The number of sampling points is obtained from the last calculation of frequency function. As the line frequency fluctuates in slow motion and typically varies by a small amount over three (3) line cycles, the period of line frequency measurements of the previous cycle can be used to determine data length of sampling. The quasi-synchronization window function is an array established in advance, and its length is the same as that of the sampling data, which is obtained by the weight coefficient multiplied by 32768. In this design, the method of quadrature by complexification echelon is used, corresponding weight coefficient is calculated with three (3) iterations. Three iterations implies that the length of the input original data is equal to the number of sampling points in three cycles. The number of sampling points in each cycle is usually different from the input signal cycle, but it will be close to an integral multiple of the input signal cycle. For example, at a sampling rate of 3.2 ksps, 50 Hz input signal corresponds to 64 sampling points for each cycle, and 50.1 Hz input signal would also be close to 64 sampling points for each cycle. Again, 51 Hz input signal would be close to 63 sampling points for each cycle. Therefore, at different input frequencies, the corresponding numbers of sampling points in each cycle are different. Consequently, the corresponding quasi-sync window function and sine/cosine table need to be established according to different numbers of sampling points. The sine/cosine table is established by evenly dividing a cycle into a number of segments equal to the number of sampling points, calculating corresponding sine/cosine values and then multiplying the values by 32768. The purpose of the multiplication is to change the original operation of floating point numbers into that of fixed-point numbers. Adjustments will be performed in the final stage of calculation. The processing of the original signal being processed by the quasi-synchronization window function is actually a process of array multiplication, i.e. the original input signal is multiplied with a corresponding array of window functions. It's accomplished by the function qusi_syn_wnd(), which is written in assembly to take full advantages of DSP features. 3.3.2 DFT Transformation A Direct Fourier Transform (DFT) is performed on the collected sets of data. Processing of the original signal by quasi-synchronization window can effectively reduce spectrum leakage caused by non-entire-cycle sampling during DFT transformation. The data length is not a power of 2, therefore, the FFT algorithm cannot be used in DFT transformation. DFT transform is accomplished by function DFT(), which is written in assembly to take full advantages of the DSP feature of accumulated multiplication. Since an FFT algorithm cannot be used, and it takes longer to perform a DFT calculation, this is the most time-consuming process in the entire system. DS51723A-page 30 (c) 2009 Microchip Technology Inc. Firmware 3.3.3 Calculating RMS Voltage/Current After the data set of either a voltage or current signal (of each phase), has been DFT transformed, the voltage or current magnitudes of the different harmonics can then be calculated. The total effective voltage or current (RMS) can be obtained by further calculaton, by simply combining the results of the individual harmonics (including the fundamental or the 1ST harmonic). Computing the magnitude of the voltage or current is accomplished by function ComputeMagnitude(). The result, called amplitude, is a long integer, and is the squared magnitude of voltage or current. To speed up the computation, fixed-point operation is used. The ComputeMagnitude() routine is written in assembly language. After the magnitude is computed, there is an adjustment process which is based on a floating-point operation. The limited number of computations will not affect the operation speed, and will instead greatly improve precision. Parameter ratio1 in the firmware is a coefficient related to the number of sampling points (see Equation 2-2). Division is accomplished by a simple shift operation in firmware. If the sampling cycle is not a power of 2, it cannot be accomplished by shifting. However, division by shifting can be accomplished by multiplying a compensating coefficient Coeff.data.linear.V_channel[]. This is the calibration coefficient for ratio errors. Since the current signal has a wide dynamic range, for small signals, the ADC output data range is small and is limited by DSP's bit resolution (16-bit MCU). If division by shift is used in the same way that is used for large signals in computation, precision may be affected. Therefore, for computing magnitudes of small signals, ComputeSmallMagnitude() function is used instead. This function is similar to ComputeMagnitude(), the only difference being that the shift length is shortened in division operation, and will be compensated during data adjustment. The computation precision will not be affected as the data adjustment process uses float-point operation. 3.3.4 Calculating Harmonics Computing harmonics is accomplished in assembly language, by the function ComputeHarmonic(). The computation is based on Equation C-62, in Appendix C. "Power Calculation Theory". The result is the ratio of the magnitude of K-th harmonic to fundamental magnitude, and is given in a percentage. Note: 3.3.5 Since the output of ComputeHarmonic() is the squared harmonic magnitude, extraction of square root is needed in computation. The calculated harmonic content is stored as a fixed-point number, and the actual value stored is the harmonic content multiplied by 10. Calculating Power Computing power is accomplished in assembly language by the function ComputePower() based on Equation C-39 and Equation C-40 in Appendix C. "Power Calculation Theory". After the ComputePower() function is complete, an adjusting process for computed power is required. First, the computed result is adjusted according to the gain of current amplifier. Then the calibrating coefficient ratio2 is determined according to present number of sampling points. Additional compensation to the power calculation is required, for phase compensation. This compensation is based on the present load current. The difference between signal frequency and the central frequency is also taken into consideration. Consequently computed power is compensated. (c) 2009 Microchip Technology Inc. DS51723A-page 31 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 3.3.6 Active Energy Accumulation Energy accumulation is done by calculating the total energy, which is the algebraic sum of energy of each phase. Active energy is obtained by accumulating the multiplication of voltage and current of each sample, which ensures the high accuracy of measurement. 3.3.7 Reactive Energy Accumulation The required measurement accuracy of reactive energy low, so in this design, it is obtained by accumulating the product of the present measured reactive power and the time interval between two measurements. 3.3.8 Output Pulse Generation Refer to Section 2.4.2 "Energy Pulse Output Interface" for pulse output. To ensure the uniformity of output pulses, the calculation is divided in the measurement cycle into in a number of equal sections, and accumulate them. For simplification and lowering computation complexity, a counter is used to substitute the process of accumulation. The counter is only enabled when accumulated energy approaches to the threshold of the pulse output. 3.3.9 Line Frequency Calculation Frequency calculation is based on Equation C-52 and Equation C-53, in Appendix C. "Power Calculation Theory". The dsPIC33F collects 3-line cycles worth of data. The first two cycles of data of all sampled data is analyzed, and then the frequency of two successive cycles is used. The data of two successive cycles are transformed via DFT for the fundamental, which is accomplished by assemble function DFT_Fundamental(). This is followed by the computation of the initial phase angle of the first two line cycles. Then the phase lag and frequency offset of the two line cycles of signal can be calculated. When measuring frequency, only the first two cycles of data are used. It must be assumed the input frequency is 50 Hz and the chosen appropriate sine/cosine table to carry out DFT transform for fundamentals of the 1st and 2nd cycles of data. See Appendix D. "50/60 Hz Meter Operation" for 60 Hz firmware. Frequency offset is calculated by determining the initial phase angle for each line cycle. The greater the frequency offset, the greater the measurement error. Since one of the 3 phases may be missing, if the voltage magnitude for phase A is less than the threshold, it is necessary to switch to phase B. Consequently, if sufficient voltage magnitude of phase B is not detected, it is necessary to switch to phase C. The basic algorithm for measuring line frequency is based on the method described in Appendix A, Section C.4 "Measuring The Voltage/current Rms Value And Power Using Quasi-synchronous Sampling Algorithm". Frequency will be measured once for every 3 times the data is sampled. DS51723A-page 32 (c) 2009 Microchip Technology Inc. Firmware 3.4 ADC SAMPLING SCHEME FOR CALCULATIONS The ADC conversion rate of the MCP3909 device is determined by the frequency of master clock, MCLK, and the rate will be MCLK/256. After each conversion is complete, a DataReady signal (4-CLK length) is generated by the SDO of the MCP3909 device. The signal is fed into IC1 (Input Capture 1 on the dsPIC33F), allowing the Interrupt Service Routine (ISR) of IC1 to invoke data-read function of the MCP3909 device. When the MCP3909 device outputs data, it first sends the ADC result of the voltage channel, then that of the current channel, with the MSB first. The frequency of the master clock, MCLK, of the MCP3909 device is 3.2768 MHz, and ADC outputs @12.8 ksps. In practice 6.4 ksps sampling rate is used in the program, which means only 1 output data is used for every 2 data sampled. For a 50 Hz input signal, a 6.4 ksps sampling rate will take 128 samples for each cycle. The active power calculation is computed based on this condition. The other parameters for which precision is not critical, such as reactive energy, voltage, current and frequency, the sampling rate may be reduced to save data storage space and processing time. In this design, the 3.2 ksps sampling rate is used, which means only 1 result is stored for every 4 ADC conversions. In the program, sampling and calculation are carried out concurrently, and data is stored in the cyclic array in the dsPIC33F RAM. A calculation may be performed after either 1 cycle, 2 cycles or 3 cycles of data are sampled, which can be configured in the program. The user should note that frequent calculations will increase the measurement precision at the price of system overhead and response speed, therefore making proper tradeoffs based on practical requirement. In this design, 3 cycles of signals are sampled before an AC electrical parameter calculation is performed. Refer to Figure 3-3. Sampling Cycle n Sampling Cycle n+1 Sampling Cycle n+2 Idle Idle Idle FIGURE 3-3: 3.4.1 Sampling Cycle n+3 Calculate n,n+1,n+2 Sampling Cycle n+4 Idle AC Signal Sampling alnd Computing. Processing IC1 Interrupt Input capture IC1 is used to detect if the A/D conversion is complete. After each conversion, a positive pulse the width of 4 clock cycles is outputted by the SDO pin of the MCP3909 device. IC1 is used to detect the falling edge of the pulse and generate an interrupt for every 2 falling edges, i.e., 1 data is read for every 2 conversions, thus realizing 6.4 ksps sampling rate. In addition to reading the data of the MCP3909 device, the IC1 interrupt service routine (ISR) also controls the energy pulse output generation. Energy pulse processing consists of active/reactive energy pulse processing. For the pulses to be outputted more uniformly, the clock resolution used to generate the pulses must be as high as possible. The interval of the IC1 interrupt is 156.25 s, therefore, the resolution generated by the pulse can be up to 156.25 s. (c) 2009 Microchip Technology Inc. DS51723A-page 33 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design The energy pulse processing program only begins when the level is close to outputting pulse level. To simplify the process and shorten the ISR execution time, a counter is used in place of the energy accumulation function for each pulse and to determine if a pulse will be outputted. When the count is greater than the threshold of pulse output, an energy pulse will be outputted, and the appropriate amount of energy will be subtracted from the energy accumulating register. Output toggling will then be processed. Once the width of the output pulse exceeds 80 ms, the pulse output will be turned off. The program flow chart is shown in Figure 3-4. ICI Interrupt Service routine Call MCP3909 data read program Processing falling edge of active energy? Yes Update pulse width counter, if pulse width > 80 ms, toggle pulse output level. End pulse output process. No Processing rising edge of active energy? Yes Update pulse width counter, if counter > flip threshold, output pulse and update energy accumulation register. No Processing falling edge of reactive energy? Yes Update pulse width counter, if pulse width > 80 ms, toggle pulse output level. End pulse output process. No Processing rising edge of active energy? Yes Update pulse width counter, if counter > flip threshold, output pulse and update energy accumulation register. No Return FIGURE 3-4: DS51723A-page 34 IC1 Interrupt Service Routine. (c) 2009 Microchip Technology Inc. Firmware 3.5 READING A/D DATA OF THE MCP3909 DEVICE All three MCP3909 devices use the same clock source and reset signal, so all 6 A/D channels of the 3 MCP3909 devices are synchronous. Only a single Data Ready (SDO) signal of any of the MCP3909 device is required to read A/D data of the 3 phases in turn. This module is invoked by IC1 interrupt triggered by the "data ready" signal on the SDO of the MCP3909 device. IC1 is set to generate an interrupt for every two falling edges. Therefore, only one of the two sampling data of the MCP3909 device is read.The flow of reading the MCP3909 device's data is as follows: * Retrieve all values of 3-phases, both current channel and voltage channel data. Bits 0-15 of each phase data are voltage channel data, bits 16-31 are current channel data * Accumulate the active power of each phase. On every other interrupt, the current and voltage values are stored into RAM in the cyclic sampling array * Update the pointer of sampling array and length of sampling data. If the length of sampling data is 3-line cycles long, set the sampling complete flag, and then the calculation function Calculate() will be called by the main flow to start computing all corresponding parameters. Read MCP3909 data Select phase A of The MCP3909 device, clear SPI flag Read phase A data, No accumulate active energy of phase A and save data to array Even count data read? Yes Read phase A data and accumulate active energy of phase A Read phase B data and accumulate active energy of phase B Read phase C data and accumulate active energy of phase C Read phase A data, accumulate active energy of phase A and save data to array Read phase A data, accumulate active energy of phase A and save data to array Update array pointer, sample pass count flag and data length End of sampling of this cycle? y No Set data sampling complete flag End FIGURE 3-5: (c) 2009 Microchip Technology Inc. Flow Chart of Read A/D Data. DS51723A-page 35 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 3.5.1 Initialize and Configure MCP3909 Operation Mode The task of this module is to enable the MCP3909 device to enter the "Channel Output" mode. This design uses the "Channel Output" mode of the MCP3909 device. In this mode, the current and voltage data channels measured by the ADC is sent through the MCP3909 device's SPI port. To enable the MCP3909 device to enter the Channel Output mode, certain instructions must be sent to the device via the SPI interface within a specified time (32CLK) after resetting the MCP3909 device. * Enable all MCP3909 devices: enable ADCS1, ADCS2 and ADCS3, and configure MCU's SPI to 8-bit mode * Reset the MCP3909 device through the RESET pin. The pin must be pulled low for no less than 1 clock cycle of the MCP3909 device * After the RESET pin is pulled high, wait for 4 clock cycles for the MCP3909 device pin functions to reset * Send Instruction 0x94 to the MCP3909 device through the SPI Interface * Configure the SPI interface to 16-bit mode and strobe the MCP3909 device for Phase A Initialize MCP3909 Strobe all MCP3909 devices Reset the MCP3909 devices Wait for 4 CLK cycles Send instruction 0xac Set SPI to 16-bit mode Strobe phase A of the MCP3909 device End FIGURE 3-6: DS51723A-page 36 Initializing the MCP3909 Device Flow Chart. (c) 2009 Microchip Technology Inc. Firmware 3.6 COMMUNICATION OF UART INTERFACE The UART interface is used to communicate with the upper computer (MCU or PC). Via the UART interface, the upper computer reads the measured parameters of the power grid, and may also send system parameters and calibration parameters to the target board as well. The communication interface is a bidirectional interface based on UART, using master/slave half-duplex mode. The baud rate is 19,200 bps, with 1 start bit, 8 data bits and 1 stop bit. Communication is done by frames with non-fixed-length frame structure, definition of which is shown in Table 3-1. The Communication protocol is specified in a master-slave structure. The system in this design is the slave, and the upper computer is the master. The master sends commands to the slave, and slave responds to the master. Each command is defined in Chapter 6. "Meter Communications Protocol". TABLE 3-1: FRAME STRUCTURE OF COMMUNICATION PROTOCOL Sync Field Command Type Data Length Data Field Checkout Byte End Byte 1 byte 1 byte N bytes 1 byte 1 byte 2 bytes 3.7 RESOURCE CONFIGURATION Details of the MCU resources used in this design and their configurations are listed in Table 3-2. TABLE 3-2: CONFIGURATION OF MCU RESOURCES Resource Name Interrupt Priority System Clock Timer Interrupt Functional Description Fcy = 29.4912M, provided by an external 7.3728 Mz timer through an internal PLL frequency doubler. Timer2 1 System clock, used for timing. Its cycle is 10 ms. The interrupt flag may be set in the IRS. Used to extend the indication of timer. Also used to deal with UART reception overtime. Timer3 none Used to detect ADC's sampling synchronization of neutral current. After the frequency of the power grid is measured, the period of TMR3 is adjusted accordingly. 16 points are sampled by ADC for each cycle of power grid. TMR2 1 ditto IC1 5 Driven by a 3.2768 MHz clock. The MCP3909 device can generate 12.8 ksps of data output. Sampling input capture. An interrupt is generated for every two MCP3909 device samplings. 6.4 ksps sampling rate is realized. In fact, active power is cumulated at 6.4 ksps sampling rate (128 sampling points each cycle at 50 Hz), but other parameters are cumulated at 3.2 ksps sampling rate UART RX 2 Receive data of UART communication UART TX 2 Transmit data of UART communication ADC 2 Detect current of neutral line SPI2 none Used in communicating with the MCP390X device - set the MCP390X device's modes and read A/D results SPI1 none Unused, but the interface is reserved and may be used to communicate with upper computer in substitution of the UART interface (c) 2009 Microchip Technology Inc. DS51723A-page 37 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 3.8 DESCRIPTION OF PROJECT FILES TABLE 3-3: FILE DESCRIPTION File Name Description main.h main.c Main program global.h global.c Mainly define important system macros, key data structures, and declare global variables. MCP390x.h Declare macros, constants, local global variables, some of the global variables and functions used in the MCP390X device. MCP390x.c Functions involved with the MCP390X device, including set SPI, initialize the MCP390X device and read data. calcu.h Declare macros, constants, local global variables, some of the global variables and functions used in calcu.c. calcu.c Main module to calculate parameters, including calculate frequencies, current/voltage RMS, power, power factors and energy, and analyze harmonics. uart_comm.c Declare macros, constants, local global variables, some of the global variables and functions used in uart.c. uart.c Receive, transmit, process protocol and so forth for UART communication. Calibrate.c Program for Ratio error calibration, power calibration and phase lag calibration, it stores and initializes calibration data. Calibrate.h Declare constants, local variables and global variables used in calibration. Adc.c Adc.h On-chip ADC operation, detecting the current of neutral wire. I2Csubs.h I2Csubs.c Control EEPROM of off-chip I2C interface. interrupt.h Declare macros, constants, local global variables, some of the global variables and functions used in interrupt.c. interrupt.c Set interrupts and ISRs. Asmcode.c Some assemble functions used in calculation. DS51723A-page 38 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 4. Meter Calibration 4.1 INTRODUCTION Meter calibration consists of using standard electrical power equipment that supplies the power to the meter and calculates the error and correction factor at each calibration point. This equipment must be accurate in order to calibrate the energy meter. The supplied PC software is then used to send calibration commands and correction factors down to the dsPIC33F, completing meter calibration. Why Is Calibration Necessary? An energy meter usually consists of errors due to transformers, VREF tolerance, ADC gain errors, and other passive component errors. Energy meters are factory calibrated before shipping to eliminate the impact from such elements and reduce the error. The non-linearity and inconsistency of signals in the path of sampling circuit and A/D conversion circuit cannot be ignored in high-accuracy measurement. The impact needs to be corrected to improve measurement accuracy. The calibration described in this chapter are calibrated with the help of the PC software PM_Viewer, described in detail in the next chapter. To summarize the process, the measurement error is fed into the software, and the data is then sent to the meter to via the UART. The details of this procedure are detailed in the next chapter, "PC Software". 4.2 CURRENT/VOLTAGE CALIBRATION Current and voltage calibration is a ratio error calibration from the upper computer by sending commands and data for correction to the MCU. The dsPIC33F will call a firmware module after receiving the command from the host PC. The flow is as follows: 1. Determine the phase to be calibrated and the magnitude of current and voltage being applied to the meter, and read measurement (RMS) values of that channel. 2. Calculate the calibration coefficient of the ratio error by the ratio of standard value received to the measured value. 3. Multiply the original coefficient by calibration coefficient and obtain the calibration coefficient after correction. 4. Store the final calibration coefficient after correction into EEPROM. Note: Voltage and current calibration is a two step process using 100% and 10% IB. Since the dynamic range of the voltage channel is usually very small, single-point calibration is enough to meet the accuracy requirements for full range. However, the dynamic range of the current channel is larger, and the transformer has different ratio errors at different current loads. The MCP3909 device's current channel, CH0, contains a PGA with gain options of 1, 2, 8, 16. For high-accuracy energy meters, current ratio error needs to be segmented and calibrated for different current loads. The ratio error calibration of current channel uses a two-point calibration method. One point is calibrated when the load is at the rated current (IB) and the PGA gain is 1. The second point is calibrated under small-signal input condition (0.1 IB) and the PGA gain is 16. (c) 2009 Microchip Technology Inc. DS51723A-page 39 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 4.2.1 Current/Voltage Calibration Process The process of calibration is as follows: 1. Supply the meter with balanced load, PF = 1.0, VCAL = 220V, IB = 5A. 2. Load the dsPIC33F with the proper correction factors. Automatically done using the PC software. See Chapter 5. "PC Software". Repeat these following steps for the second point at 10% IB: 1. Set the three-phase balance input conditions PF = 1.0, VCAL = 220V, ICAL = 0.1 IB or 500 mA. 2. Load the dsPIC33F with the proper correction factors. Automatically done using the PC software. See Chapter 5. "PC Software". If accuracy is not critical, the single-point calibration method can be used. The number of calibration points can be defined in the header file of the program. 4.3 APPARENT POWER CALIBRATION Apparent Power calibration function is implemented by the upper computer by sending the commands. Before the power calibration process can be entered, power calibration mode command needs to be sent first. The error data of the calibration workbench and channel information to be calibrated are sent to the metering front-end. When the front-end receives the command, it calls this module. The flow is as follows: 1. Determine the phase to be calibrated according to the parameters received. 2. Calculate new power calibration coefficient according to the error value received and the measured value, together with the original power calibration coefficient. 3. Store the coefficient after correction into the EEPROM. 4.3.1 Apparent Power Calibration Process The process of calibration is as follows: 1. Set the input condition as: Phase A PF = 1.0, VCAL = 220V, input current is the current when region N is being calibrated, the voltage and current inputs of phase B and C are zero. 2. Choose the energy pulse output to be the apparent power output mode Refer to Chapter 6. "Meter Communications Protocol". At this time, the energy pulse is the accumulated multiplication of power and time. 3. Load the dsPIC33F with the proper correction factors. This is automatically done using PC software. See Chapter 5. "PC Software". 4. Repeat steps 1 - 3 for phase lag calibrations for all current regions of phase A. Note: At this time, the phase lag has not been calibrated, so when the input PF = 1.0, the measured value of the reactive power isn't equal to zero. 5. Repeat the above steps for Phases B and C. DS51723A-page 40 (c) 2009 Microchip Technology Inc. Meter Calibration 4.4 PHASE LAG CALIBRATION The phase lag calibration function is implemented by the upper computer by sending the proper commands via the UART. When calibrating phase lag, error from the calibration equipment and channel information to be adjusted are sent to the dsPIC33F energy meter. When the front-end receives the command, it calls this module. The flow is as follows: 1. Determine the phase to be calibrated according to parameters received. 2. Calculate new phase lag calibration coef. according to the error value received and the measured value. 3. Store the coefficient after correction into the EEPROM. This meter design supports single, two, and five point calibration for phase lag error correction. The purpose of phase lag calibration is to eliminate the impact of phase lag introduced by the current transformer (CT), and voltage transformer (PT) over the power measurement range. The voltage transformer usually has a constant load, thereby introducing a phase lag that varies insignificantly. The dynamic range of current is larger, and under different current loads, phase lags caused by CT vary greatly. In order to meet the requirements of measurement accuracy in the entire range, it is usually necessary to segment the phase lag and calibrate. In this design, current is partitioned into 5 regions. TABLE 4-1: CURRENT REGIONS FOR PHASE CALIBRATION Region Current Range 1 0 - 0.075 IB 2 0.075 IB - 0.2 IB 3 0.2 IB - 0.75 IB 4 0.75 IB -1.5 IB 5 1.5 IB - 4.0 IB The partition limit for the current region can be modified in the header file of the program. If accuracy is not critical, single-point calibration and two-point calibration can be used to improve the efficiency of meter calibration. Single, Two, or Five Point Calibration Single-, two- or five-point calibration method can be configured by modifying the header file. When using the single-point calibration, the phase lag compensation values of all regions are the same; When using two-point calibration, the compensation values of region 1 and 2 (0-0.075 IB, 0.075 IB - 0.2 IB) are the same, and the phase lag compensation values for region 3, 4 and 5 (0.2 IB - 0.75 IB, 0.75 IB - 1.5 IB, 1.5 IB 4.0 IB) are the same. (c) 2009 Microchip Technology Inc. DS51723A-page 41 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 4.4.1 Phase Lag Calibration Process The process of phase calibration is as follow: 1. Setup input condition: Phase A, voltage input 220V, current input is the current for region 1, voltage and current inputs of phase B and phase C are zero. 2. Load the dsPIC33F with the proper correction factors. This is automatically done using the PC software. See Chapter 5. "PC Software". 3. Repeat steps 1 and 2 for phase lag calibrations for all current regions of phase A. Note: If the power metering error still can not meet the requirement, the meter can be calibrated a few more times. When doing so, simply input a new error value into the front-end of the meter. 4. Repeat for Phases B & C. DS51723A-page 42 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 5. PC Software 5.1 OVERVIEW AND INSTALLATION The PC software "PM_Viewer" or "Power Meter Viewer" has two main functions: view the calculated parameters and calibrate the meter. The PC software has seven output display screens, or "work modes", selected from the toolbar pull-down menu. * * * * * * * Basic Parameters Phase A Harmonic Phase B Harmonic Phase C Harmonic Distortion Rate Harmonic Power Energy Accumulation In addition, the PC software has four calibration screens, selected from the toolbar pull-down menu. * * * * Reset All Calibration Linearity Calibration Apparent Power Calibration Phase Lag Calibration 5.1.1 System Required * HDD space > 25 MB * Microsoft Windows OS98 or later * Hardware COM interface 5.1.2 1. 2. 3. 4. Installation Unzip PM_Viewer setup.zip. Double click on setup.exe. Finish the installation according the prompt. To PM_Viewer.exe - Start -> Program -> Energy Meter ->PM_Viewer.exe. (c) 2009 Microchip Technology Inc. DS51723A-page 43 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 5.2 ESTABLISH COMMUNICATION 1. Open PM_Viewer. 2. Click on Comm Port selection, and select the com port (com1, com2, or com3) that you will use on the menu, noting the baud rate is 19200 bps in 1-8-1 format, and can not be changed. 3. Click on the Link to establish communication with the demo. "Communication OK!" will be displayed on the bottom, if communication is established. FIGURE 5-1: DS51723A-page 44 Establising Communications. (c) 2009 Microchip Technology Inc. PC Software 5.3 BASIC PARAMETERS OUTPUT SCREEN FIGURE 5-2: 5.4 Basic Parameters Work Mode Screen. PHASE A/B/C HARMONIC OUTPUT SCREEN FIGURE 5-3: (c) 2009 Microchip Technology Inc. Phase N Harmonic Work Mode Screen. DS51723A-page 45 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 5.5 DISTORTION RATE FIGURE 5-4: 5.6 Distortion Mode Screen. HARMONIC POWER FIGURE 5-5: DS51723A-page 46 Harmonic Power Work Mode Screen. (c) 2009 Microchip Technology Inc. PC Software 5.7 ENERGY ACCUMULATION FIGURE 5-6: 5.8 Energy Accumulation Work Mode Screen. CALIBRATION STEP 1 - RESET ALL CALIBRATION 1. Select Reset All Calibration from the toolbar menu. 2. Meter Calibration Values are Reset. FIGURE 5-7: (c) 2009 Microchip Technology Inc. Reset All Calibration Command. DS51723A-page 47 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 5.9 LINEARITY CALIBRATION 1. 2. 3. 4. 5. 6. DS51723A-page 48 Select Channel, either Voltage or Current. Select Phase A, B or C. Select Region, either 100% or 10%. Using a standard meter, supply the input conditions given here. Enter the error recorded from the standard meter here. Click the Set button. (c) 2009 Microchip Technology Inc. PC Software 5.10 APPARENT POWER CALIBRATION 1. 2. 3. 4. 5. 6. 7. 8. Select Phase A, B or C. Select Region n. Using a standard meter, supply the input conditions given here. Click the Set Apparent button. Enter the error recorded from the standard meter here. Click the Set button. Repeat steps 2-5 for the different regions. Repeat for other 2 phases. (c) 2009 Microchip Technology Inc. DS51723A-page 49 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 5.11 PHASE LAG CALIBRATION 1. 2. 3. 4. 5. 6. 7. DS51723A-page 50 Select Phase A, B or C. Select Region n. Using a standard meter, supply the input conditions given here. Enter the error recorded from the standard meter here. Click the Set button. Repeat steps 2-5 for the different regions. Repeat for other 2 phases. (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Chapter 6. Meter Communications Protocol 6.1 INTRODUCTION The UART interface is used to communicate with the upper computer (MCU or PC). Via the UART interface, the upper computer reads the measured parameters of the power grid, and may also send system parameters and calibration parameters to target board as well. The communication interface is a bidirectional interface based on UART, using master/slave half-duplex mode. The baud rate is 19,200 bps, with 1 start bit, 8 data bits and 1 stop bit. Communication is done by frames with non-fixed-length frame structure, definition of which is shonw in Table 6-1. Communication protocol is specified in master-slave structure. The system in this design is the slave, and the upper computer is the master. The master sends commands to the slave, and the slave responds to the master. UART communication uses half-duplex mode. The data format is 8-1-1 and the rate is 19,200 bps. The PC is the host computer, and the target board is the slave. There are 14 command strings that the meter uses. These command strings are defined in Table 6-1 TABLE 6-1: COMMAND STRINGS Command Description Test Connection Command 0x41 Total Data Request 0x42 Harmonic Content, Phase A 0x43 Harmonic Content, Phase B 0x44 Harmonic Content, Phase C 0x45 Total Harmonic Distortion 0x46 Energy 0x47 Stop Energy Measurement and Clear Energy Values 0x48 Harmonic Power 0x49 Write Calibration Values to Meter 0x62 Write Phase Lag Calibration Values to Meter 0x63 Write Power Calibration Values to Meter 0x64 Write Energy Pulse Configuration - Active/Apparent 0x65 Reset All Calibration Values 0x66 Write Energy Pulse Constant 0x67 (c) 2009 Microchip Technology Inc. DS51723A-page 51 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design Variant-length frame structure is used and data communiate in bytes. The command protocol structure is defined as following. TABLE 6-2: TYPICAL PROTOTOL START Command Data Length Data Field Check Sum STOP 2 Bytes 1 Bytes 1 Byte N Bytes 1 Byte 1 Byte * The START word has 2 bytes, which are 0x00, 0xFF ( PC to target board) or 0xFF, 0x00 (target board to PC) * Command word is 1 byte which indicates the type of the command * Data length word is 1 byte that indicates the length of data field * The data field word has multiple byte(s) that varies with command types * Checksum word is a single byte, whose content equals to the XOR value of all bytes sent before it * Stop word is 1 byte with the content of 0xE0 6.2 TEST CONNECTION COMMAND This command is sent from the PC to the meter to setup and test the connection. TABLE 6-3: START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x41 0x00 0x00 XX 0xE0 TABLE 6-4: 6.3 PC TO METER (7 BYTES) METER RESPONSE (8 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x42 0x02 0xA5, 0X5A XX 0xE0 TOTAL DATA REQUEST This is the main command retrieves all the calcuated data from the dsPIC33F. This command gathers data from all 3 phases including total energy, power, and power factor data. TABLE 6-5: START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x42 0x00 0x00 XX 0xE0 TABLE 6-6: DS51723A-page 52 PC TO METER (7 BYTES) METER RESPONSE (104 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x43 0x62 98 bytes XX 0xE0 (c) 2009 Microchip Technology Inc. Meter Communications Protocol TABLE 6-7: METER RESPONSE, TOTAL REQUEST FRAME, DETAILED DESCRIPTION Data Field Byte Name Value 1,2 Status See Definition Below 3-6 Frequency Float, 4 Bytes Total 7-10 Phase A Voltage Float, 4 Bytes Total Phase B Voltage Float, 4 Bytes Total 94-98 (c) 2009 Microchip Technology Inc. Phase C Voltage Float, 4 Bytes Total Phase A Current Float, 4 Bytes Total Phase B Current Float, 4 Bytes Total Phase C Current Float, 4 Bytes Total Neutral Current Float, 4 Bytes Total Active Power, Phase A Float, 4 Bytes Total Reactive Power, Phase A Float, 4 Bytes Total Apparent Power, Phase A Float, 4 Bytes Total Power Factor, Phase A Float, 4 Bytes Total Active Power, Phase B Float, 4 Bytes Total Reactive Power, Phase B Float, 4 Bytes Total Apparent Power, Phase B Float, 4 Bytes Total Power Factor, Phase B Float, 4 Bytes Total Active Power, Phase C Float, 4 Bytes Total Reactive Power, Phase C Float, 4 Bytes Total Apparent Power, Phase C Float, 4 Bytes Total Power Factor, Phase C Float, 4 Bytes Total Total Active Power Float, 4 Bytes Total Total Reactive Power Float, 4 Bytes Total Total Apparent Power Float, 4 Bytes Total Total Power Factor Float, 4 Bytes Total DS51723A-page 53 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 6.4 STATUS REGISTER This register contains the gain register. REGISTER 6-1: STATUS REGISTER R-0 R-0 R-0 R-0 R-0 R-0 R-0 R-0 CPO VPO PHC_S1 PHC_S0 PHB_S1 PHB_S0 PHA_S1 PHA_S0 bit 7 bit 0 Legend: R = Readable bit W = Writable bit U = Unimplemented bit, read as `0' -n = Value at POR `1' = Bit is set `0' = Bit is cleared bit 7 CPO: Current Phase Order 1 = Problem Detected 0 = Normal bit 6 VPO: Voltage Phase Order 1 = Problem Detected 0 = Normal bit 5:4 PHC_S: Phase C Status 11 = High Votage 10 = No Input 01 = Low Voltage 00 = Normal bit 3:2 PHB_S: Phase C Status 11 = High Votage 10 = No Input 01 = Low Voltage 00 = Normal bit 1:0 PHA_S: Phase C Status 11 = High Votage 10 = No Input 01 = Low Voltage 00 = Normal 6.5 x = Bit is unknown HARMONIC CONTENT COMMAND TABLE 6-8: START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x43 0x00 0x00 XX 0xE0 TABLE 6-9: DS51723A-page 54 PC TO METER (7 BYTES) METER RESPONSE (134 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x44 0x80 128 bytes XX 0xE0 (c) 2009 Microchip Technology Inc. Meter Communications Protocol TABLE 6-10: HARMONIC ANALYSIS DETAILED DESCRIPTION Data Field Byte 6.6 Description Value 1,2 Fundamential or 1st Harmonic, Voltage Content Unsigned Int, 2 bytes 3,4 2nd Harmonic, Voltage Content Unsigned Int, 2 bytes 5-56 3-31st Harmonic, Voltage Content Unsigned Int, 2 bytes 57,58 Total Voltage Harmonic Content (not including Fundamental) Unsigned Int, 2 bytes, / 1000 * (100%) 59,60 Fundamential or 1st Harmonic, Current Content Unsigned Int, 2 bytes 61,62 2nd Harmonic, Current Content Unsigned Int, 2 bytes 63-126 3-31st Harmonic, Current Content Unsigned Int, 2 bytes 127,128 Total Current Harmonic Content (not including Fundamental) Unsigned Int, 2 bytes, / 1000 * (100%) TOTAL HARMONIC DISTORTION (THD) COMMAND TABLE 6-11: PC TO METER (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x46 0x00 0x00 XX 0xE0 TABLE 6-12: METER RESPONSE (29 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x47 0x18 24 bytes XX 0xE0 TABLE 6-13: Data Field Byte TOTAL HARMONIC DISTORTION DESCRIPTION Description Value 1-4 Total Harmonic Distortion of Phase A Voltage Float, 4 Bytes Total 5-8 Total Harmonic Distortion of Phase B Voltage Float, 4 Bytes Total 9-12 Total Harmonic Distortion of Phase C Voltage Float, 4 Bytes Total 13-16 Total Harmonic Distortion of Phase A Current Float, 4 Bytes Total 17-20 Total Harmonic Distortion of Phase B Current Float, 4 Bytes Total 21-24 Total Harmonic Distortion of Phase C Current Float, 4 Bytes Total (c) 2009 Microchip Technology Inc. DS51723A-page 55 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 6.7 START ENERGY MEASUREMENT COMMAND TABLE 6-14: PC TO METER (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x47 0x00 0x00 XX 0xE0 TABLE 6-15: METER RESPONSE (42 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x48 0x24 36 bytes XX 0xE0 TABLE 6-16: TOTAL HARMONIC DISTORTION DESCRIPTION Data Field Byte 6.8 Description Value 1-4 First quadrant reactive energy Float, 4 Bytes Total 5-8 Second quadrant reactive energy Float, 4 Bytes Total 9-12 Third quadrant reactive energy Float, 4 Bytes Total 13-16 Fouth quadrant reactive energy Float, 4 Bytes Total 17-20 Forward Reactive energy Float, 4 Bytes Total 21-24 Reverse Reactive Energy Float, 4 Bytes Total 25-28 Forward Active Energy Float, 4 Bytes Total 29-32 Reverse Active Energy Float, 4 Bytes Total 33-36 Reverse Active Energy Float, 4 Bytes Total STOP ENERGY MEASUREMENT COMMAND TABLE 6-17: START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x48 0x00 0x00 XX 0xE0 TABLE 6-18: DS51723A-page 56 PC TO METER (7 BYTES) METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x49 0x00 0 bytes XX 0xE0 (c) 2009 Microchip Technology Inc. Meter Communications Protocol 6.9 HARMONIC POWER COMMAND TABLE 6-19: PC TO METER (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x49 0x00 0x00 XX 0xE0 TABLE 6-20: METER RESPONSE (54 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x4A 0x30 48 bytes XX 0xE0 TABLE 6-21: Data Field Byte 1-4 HARMONIC POWER MEASUREMENTS Description Fundamental Active Power Of Phase A Value Float, 4 Bytes Total 5-8 Fundamental Reactive Power Of Phase A Float, 4 Bytes Total 9-12 Fundamental Active Power Of Phase B Float, 4 Bytes Total 13-16 Fundamental Reactive Power Of Phase B Float, 4 Bytes Total 17-20 Fundamental Active Power Of Phase C Float, 4 Bytes Total 21-24 Fundamental Reactive Power Of Phase C Float, 4 Bytes Total 25-28 Harmonic Active Power Of Phase A Float, 4 Bytes Total 29-32 Harmonic Reactive Power Of Phase A Float, 4 Bytes Total 33-36 Harmonic Active Power Of Phase B Float, 4 Bytes Total 37-40 Harmonic Reactive Power Of Phase B Float, 4 Bytes Total 41-44 Harmonic Active Power Of Phase C Float, 4 Bytes Total 45-48 Harmonic Reactive Power Of Phase C Float, 4 Bytes Total (c) 2009 Microchip Technology Inc. DS51723A-page 57 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 6.10 CALIBRATE METER VOLTAGE/CURRENT COMMAND TABLE 6-22: PC TO METER (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x63 7 0x07 XX 0xE0 TABLE 6-23: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x64 0 0x00 XX 0xE0 TABLE 6-24: CALIBRATION OF GAIN AND OFFSET VALUES Data Field Byte DS51723A-page 58 Description Value 1 Phase Select 0x01 = Phase A 0x02 = Phase B 0x03 = Phase C 2 Range Select 0x01 = 10% 0x02 = 100% 3 Channel Select 0x00 = Current 0x01 = Voltage 4-7 Correction Factor (Error Being Calibrated Out) Float, 4 Bytes Total (c) 2009 Microchip Technology Inc. Meter Communications Protocol 6.11 CALIBRATE PHASE LAG COMMAND TABLE 6-25: PC TO METER (12 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x63 6 0x06 XX 0xE0 TABLE 6-26: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x64 0 0x00 XX 0xE0 TABLE 6-27: CALIBRATION OF PHASE LAG Data Field Byte 6.12 Description Value 1 Phase Select 0x01 = Phase A 0x02 = Phase B 0x03 = Phase C 2 Range Select 1-7 3-6 Correction Factor (Error Being Calibrated Out) Float, 4 Bytes Total CALIBRATE APPARENT POWER COMMAND TABLE 6-28: PC TO METER (12 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x64 6 0x06 XX 0xE0 TABLE 6-29: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x65 0 0x00 XX 0xE0 TABLE 6-30: CALIBRATION OF POWER Data Field Byte Description Value 1 Phase Select 0x01 = Phase A 0x02 = Phase B 0x03 = Phase C 2 Current Range Select 1-7 3-6 Correction Factor (Error Being Calibrated Out) Float, 4 Bytes Total (c) 2009 Microchip Technology Inc. DS51723A-page 59 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design 6.13 CALIBRATE ENERGY PULSE COMMAND TABLE 6-31: PC TO METER (12 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x65 2 0x02 XX 0xE0 TABLE 6-32: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x66 0 0x00 XX 0xE0 TABLE 6-33: CALIBRATION OF POWER Data Field Byte 6.14 Description Value 1 Phase Select 0x01 = Phase A 0x02 = Phase B 0x03 = Phase C 2 Energy Output Mode 0x00 = Active Power 0x01 = Apparent Power RESET ALL METER CALIBRATION VALUES COMMAND TABLE 6-34: START Command Data Length Data Field Check Sum STOP 0x00, 0xFF 0x66 4 0x04 XX 0xE0 TABLE 6-35: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x67 0 0x00 XX 0xE0 TABLE 6-36: Data Field Byte DS51723A-page 60 PC TO METER (12 BYTES) RESET METER OPTIONS Description Value 1 Command Type 0x55 - Reset All 0xA1 = Reset 0xA2 - Reset Power Calibration Only 0xA3 - Reset Phase Calibration 2 Phase Select 0x01 = Phase A 0x02 = Phase B 0x03 = Phase C 3 Current Range Select 4 Reserved (c) 2009 Microchip Technology Inc. Meter Communications Protocol 6.15 CALIBRATE METER CONSTANT (ENERGY PULSE OUTPUT CONSTANT) TABLE 6-37: PC TO METER (9 BYTES) START Command Data Length Data Field Check Sum STOP 0x00, 0xFF 0x67 2 0x02 XX 0xE0 TABLE 6-38: METER RESPONSE (7 BYTES) START Command Data Length Data Field Check Sum STOP 0xFF, 0x00 0x68 0 0x00 XX 0xE0 TABLE 6-39: Data Field Byte 1-2 (c) 2009 Microchip Technology Inc. ENERGY CONSTANT OPTIONS OPTIONS Description Energy Constant Value Range of 10064000 (Decimal) DS51723A-page 61 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design NOTES: DS51723A-page 62 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Appendix A. Schematics and Layouts A.1 INTRODUCTION This appendix contains the following schematics and layouts for the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design. * * * * * * * A.2 Power Supply Board Schematic Main Board Schematic - Page 1 Main Board Schematic - Page 2 Power Supply Board - Assembly Drawing Power Supply Board - Composite Drawing Main Board - Assembly Drawing Main Board - Composite Drawing SCHEMATICS AND PCB LAYOUT FIGURE A-1: LAYER ORDER Top Layer Bottom Layer (c) 2009 Microchip Technology Inc. DS51723A-page 63 DS51723A-page 64 A B N PC N PC N PB N PB N PA N 1 150k T3 499K 1% 499K 1% R11 R13 R14 150k T2 499K 1% 499K 1% R7 R8 R9 150k T1 499K 1% 499K 1% R2 R3 R4 SPT204B R10 1K 1% SPT204B R6 1K 1% SPT204B R1 1K 1% C7 33N Jumper W3 C5 33N Jumper W2 C3 33N Jumper W1 PC- PC+ PC- PB- PB+ PB- PA- PA+ PA- 2 2 PA PB PC N 1 JP6 1 JP5 1 JP4 1 JP1 4 3 2 1 1 2 3 4 PC+ PC- PB+ PB- PA+ PA- N 0 R12 DB-12CY A B C N T4 COILS A B C N R5 5 6 7 8 8 7 6 5 Header 10 1 2 3 4 5 6 7 8 9 10 JP3 G1 Vo1 G2 Vo2 A B C N 3 3 Date: File: A4 Size RV2 Header 3 1 2 3 JP6 C4 0.1u C6 0.1u N RV3 N 1-1 2.0 Revision A B C D 4 30-Nov-2007 Sheet of X:\AIPD Eval Boards\102-00011 thru 00020\102-00018\PCB Drawn By: Files (Protel )\PCB-R5\PM33_ Number Power supply for 3phs power meter RV1 Title + C2 100uf 12V C1 0.1U PA PB PC three phase voltage input 2R = Vour*Rin/(Kpt*Vin) 4 FIGURE A-2: C D PA 1 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design POWER SUPPLY BOARD SCHEMATIC (c) 2009 Microchip Technology Inc. (c) 2009 Microchip Technology Inc. A C104 1 R129 4.7K 1% R130 4.7K 1% 3.3V REF_V SCT220B REV_V 4 8 6 2 1 2 3 4 1K R111 R127 4.7K 2 MCP6002 VDD OutB InBInB+ U101 R128 470 OutA InAInA+ Vss 20 R126 1.65V Neutral line current detection T100 R110 R107 R106 R103 20 2R = Vout/(Kct*Iin) R125 20 R124 20 R123 20 R102 R109 1nf 8 7 6 5 1nf 1K 1K C100 1nf 3.3V C112 1nf C111 C108 1nf C107 1nf C103 1K 1nf 1K 1K 1nf Current_N C128 0.1uF PAC- PAC+ PBC- PBC+ PCC- PCC+ PAV+ PAV- Current_N 3 5V 5V 5V AD_MCLR R118 C127 0.1 10 C125 10 C116 10 C120 R119 R120 C114 C118 + 4 8 6 2 4 8 6 2 20 R122 R121 100 R117 1nf C110 C109 1nf PBV+ PBV- PCV+ PCV- 5V 5V 5V C117 C115 C119 C124 C123 1u PAC+ PACPAVPAV+ AD MCLR 0.1 C126 500 FB1 C122 PBC+ PBCPBVPBV+ AD_MCLR 500 FB2 PCC+ PCCPCVPCV+ AD MCLR C121 1uf C113 500 FB3 + SCT220B T103 SCT220B T102 4 8 6 2 1K R108 1K C106 1nf C105 1K 1nf C102 C101 1nf + SCT220B T101 1K R105 100 100 R104 R101 1K 100 2R = Vour*Rin/(Kpt*Vin) R116 R115 R114 100 3 + B C 3 2 1 3 2 1 100 R113 R112 R100 + J8 J7 3 2 1 2 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 U104 4 DVDD HPF AVDD NC CH0+ CH0CH1_ CH1+ MCLR REFi/o AGND F2 U102 DVDD HPF AVDD NC CH0+ CH0CH1_ CH1+ MCLR REFi/o AGND F2 U103 3.2768M VCC CLKOUT GND X100 Fo0 Fo1 HFo DGND NEG NC OSC2 OSC1 G0 G1 F0 F1 MCP3909 Fo0 Fo1 HFo DGND NEG NC OSC2 OSC1 G0 G1 F0 F1 MCP3909 Fo0 Fo1 HFo DGND NEG NC OSC2 OSC1 G0 G1 F0 F1 MCP3909 7 5V_IN 14 DVDD HPF AVDD NC CH0+ CH0CH1_ CH1+ MCLR REFi/o AGND F2 4 24 23 22 21 20 19 18 17 16 G0A 15 G1A 14 13 24 23 22 21 20 19 18 17 16 15 14 13 24 23 22 21 20 19 18 17 16 15 14 13 8 G0B G1B SCK G0C G1C CSC SDI SDO AD_CLK 5 G0A 5 CSC CSB CSA SDO SDI SCK R132 4.7K GAIN 3905 CS SPI I/F R134 4.7K R135 4.7K G1C R136 4.7K GAIN 6 A B C D Date: File: B Size 2-1 Revision 1.0 6 30-Nov-2007 Sheet of Jemmey Huang X:\AIPD Eval Boards\102-00011 thru 00020\102-00018\PCB Drawn By: Files (Protel )\PCB-R5\PM33_ Number Front-end of three Phase Power Meter Design Title R131 4.7K R133 4.7K Gain selection G1 G0 Gain 0 0 1 0 1 2 1 0 8 1 1 16 G1A J6 G0B + G1B FIGURE A-3: G0C D 1 Schematics and Layouts MAIN BOARD SCHEMATIC - PAGE 1 DS51723A-page 65 AD_CLK A B C D J1 2 1 6 5 4 3 2 1 CON6 J2 MCLR 1 L301 INDUCTOR INDUCTOR 5V_IN D301 LED R316 470 510 R313 3.3V S301 SW-PB ICSPCLK ICSPDAT L302 C340 0.1uf R321 10k 3.3V F = 50Hz * 64 * 256 = 819.2 KHz F = 50Hz * 128 * 256 = 1638.4 KHz * 3905 clk range 1M to 4M AD CLK = Fac * SampleRate * Kad 1 2 1 2 + C337 100uf GAIN 3909 CS SPI I/F MCLR C328 3 2 C323 CAP C322 CAP G0A G1A G0B G1B G0C G1C CSA CSB CSC SDI SDO SCK 3.3V C330 100 100 100 3.3V C331 Vout U306 MCP1700(3.3V-SOT23) 2 Vin Vout Vin 3 3.3V U305 MCP1701(5V-SOT89) R319 R318 R317 3.3V 2 C332 C324 CAP C336 0.1uF G1C G0C CSA CSB CSC GND 3.3V + C339 47uF RG15 AN16/RC1 AN17/RC2 RG6 RG7 RG8 MCLR SS2/RG9 Vss Vdd AN15/RB5 AN4/RB4 AN3/RB3 AN2/RB2 PGC3/RB1 PGD3/RB0 3.3V + C338 100uf 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 C335 5V 3.3V AD_MCLR C333 G1A G0B G1B SCK2 SDO2 SDI2 MCLR 3.3V 3 C329 3 1 2 3 4 24LC04B A0 VCC A1 WP A2 SCL Vss SDA U309 Current_N REF_V 2 VSS 1 Gnd 1 3.3V G0A 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 8 7 6 5 3.3V 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 SDA R311 1k 4 D302 D303 SCL R312 1k 4 C334 R309 100 SDO GND OSC2 OSC1 3.3V SCL SDA SCK1 U1RX(SDI1) U1TX(SDO1) DSPIC33FJ128GP706 U307 RC14 RC13 RD0 IC4/RD11 RD10 RD9 RD8 Vss OSC2/RC15 OSC1/RC12 Vdd SCL1/RG2 SDA1/RG3 INT0/RF6 RF2 RF3 RG13 RG12 RG14 RG0 RG1 RF1 RF0 Vdd Vddcore RD7 RD6 RD5 RD4 RD3 RD2 RD1 DS51723A-page 66 PGC1/RB6 PGD1/RB7 AVdd AVss AN8/RB8 AN9/RB9 AN10/RB10 AN11/RB11 Vss Vdd AN12/RB12 AN13/RB13 AN14/RB14 AN15/RB15 RF4 RF5 R308 100 LED LED R307 100 X303 470 R310 470 R314 C346 C345 0.1uf 3.3V C303 C302 0.1uf C301 0.1uf 5 3 4 1 U1RX/SDI1 12 9 11 U1TX/SDO1 10 0.1 C342 0.1 C341 100 R306 R303 1K R302 1K R301 1K 5 5 R1IN R2IN T1OUT T2OUT VCC GND V- V+ 13 8 14 7 16 15 6 2 AD_CLK 0.1 0.1 C344 3.3V C343 U1TX/SDO1 U1RX/SDI1 SPI1_SCK 1 6 2 7 3 8 4 9 5 DB9 J5 1 2 3 4 J4 6 5 4 3 2 1 J3 Date: File: B Size 2-2 Revision 1.0 A B C D 6 30-Nov-2007 Sheet of Jemmey Huang X:\AIPD Eval Boards\102-00011 thru 00020\102-00018\PCB Drawn By: Files (Protel )\PCB-R5\PM33 Number Front-end of three Phase Power Meter Design Title MAX232 R1OUT R2OUT T1IN T2IN C2- C1C2+ C1+ U308 AD_CLK U303 TLP521-1 U302 TLP521-1 U301 TLP521-1 Output Pulse 1: active power 2: active power + 3: Reactive power 4: Reactive power + 5: TBD 6: TBD + 6 FIGURE A-4: ICSPCLK 17 ICSPDAT 18 3.3V 19 GND 20 21 22 23 24 GND 25 3.3V 26 Current_N 27 28 29 30 31 32 1 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design MAIN BOARD SCHEMATIC - PAGE 2 (c) 2009 Microchip Technology Inc. Schematics and Layouts FIGURE A-5: POWER SUPPLY BOARD LAYOUT - ASSEMBLY DRAWING (c) 2009 Microchip Technology Inc. DS51723A-page 67 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design FIGURE A-6: DS51723A-page 68 POWER SUPPLY BOARD LAYOUT - COMPOSITE DRAWING (c) 2009 Microchip Technology Inc. Schematics and Layouts FIGURE A-7: MAIN BOARD LAYOUT - ASSEMBLY DRAWING (c) 2009 Microchip Technology Inc. DS51723A-page 69 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design FIGURE A-8: DS51723A-page 70 MAIN BOARD LAYOUT - COMPOSITE DRAWING (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Appendix B. Bill Of Materials (BOM) TABLE B-1: Qty BILL OF MATERIALS - POWER SUPPLY (BOTTOM BOARD) Reference Description 3 1 3 3 C1, C2, C3 C4 C5, C6, C7 J4, J5, J6 CAP 0.1UF 305VAC EMI SUPPRESS CAP 470UF 25V ALUM LYTIC RADIAL DO NOT POPULATE CONN HEADER 3POS .100 VERT TIN 4 HOOK-UP WIRE 18AWG STRAND RED 1 JP1, JP2, JP3, JP4 JP1, JP2, JP3, JP4 JP6 CONN HEADER 3POS .156 VERT TIN 1 JP6 CONN HOUSING 3POS .156 W/O RAMP 3 JP6 CONN TERM FEMALE 18-24AWG TIN 1 PCB 3 P4, P5, P6 9 P4, P5, P6 RoHS Compliant Bare PCB, dsPIC33F and MCP3909 3-Phase Energy Meter (Power) Bottom Bd. CONN HOUS 3POS .100 W/RAMP/RIB (Connects to Above) CRIMP TERM FEMALE 22-30AWG TIN 1 1 6 R1 R2 R3, R4, R5, R6, R7, R8 R9, R10, R11 4 CONN RING TERM #6 18-22AWG Part of T1 DO NOT POPULATE DO NOT POPULATE Manufacturer Part Number EPCOS Inc Panasonic(R) - ECG -- Molex(R)/Waldom Electronics Corp Alpha Wire Company Molex/Waldom Electronics Corp Molex/Waldom Electronics Corp Molex/Waldom Electronics Corp Molex/Waldom Electronics Corp Microchip Technology Inc. B32922A2104M ECA-1EM471 -- 22-23-2031 Molex/Waldom Electronics Corp Molex/Waldom Electronics Corp Sanki -- -- 22-01-3037 3055 RD005 19070-0040 26-48-1035 09-50-7031 08-50-0105 104-00158 08-65-0805 -- -- -- 5033ED150K0F12AF5 Vishay(R)/Phoenix Passive Components 3 R12, R13, R14 DO NOT POPULATE -- -- 3 RV1, RV2, RV3 VARISTOR 300V RMS 14MM RADIAL EPCOS Inc S14K300E2 1 T1 3P AC-DC converter, 220V - 12V/5V Sanki DB-12CY220 3 T2, T3, T4 2mA/2mA Current transformer Xinge SPT204 3 W1, W2, W3 DO NOT POPULATE -- -- Note 1: The components listed in this Bill of Materials are representative of the PCB assembly. The released BOM used in manufacturing uses all RoHS-compliant components. 3 RES 150K OHM METAL FILM .50W 1% (c) 2009 Microchip Technology Inc. DS51723A-page 71 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design TABLE B-2: Qty 13 6 26 4 BILL OF MATERIALS - TOP BOARD Reference C100, C101, C102, C103, C104, C105, C106, C107, C108, C109, C110, C111, C112 C113, C114, C115, C116, C124, C125 C117, C118, C119, C120, C126, C127, C128, C301, C302, C303, C322, C323, C324, C328, C330, C331, C332, C333, C334, C335, C336, C340, C341, C342, C343, C344 C121, C122, C123, C329 C337, C338, C339 Description Manufacturer Part Number (R) CAP 1000PF 50V CERAMIC X7R 0805 Kemet Electronics C0805C102K5RACTU Corp CAP CER 10UF 16V X7R 1206 Murata Electronics(R) GRM31CR71C106KAC7L North America Panasonic(R) - ECG ECJ-2VB1E104K CAP .1UF 25V CERAMIC X7R 0805 CAP 1.0UF 10V CERAMIC X7R 0805 Kemet Electronics C0805C105K8RACTU Corp 3 CAP 470UF 25V ALUM LYTIC Panasonic - ECG ECA-1EM471 RADIAL 2 C345, C346 CAP 22PF 50V CERM CHIP 0805 Panasonic - ECG ECJ-2VC1H220J SMD 3 6" Wire Thru CT HOOK-UP WIRE 16AWG STRAND Alpha Wire 3057 RD005 RED Company 6 CT Wire terminals CONN RING TERM #6 18-22AWG Molex/Waldom 19070-0040 Electronics Corp 1 D301 LED 3MM ALGAAS RED CLEAR LITE-ON INC LTL-4266N 1 D302 LED 3MM GREEN CLEAR LITE-ON INC LTL-4236N 1 D303 LED 3MM YELLOW CLEAR LITE-ON INC LTL-4256N 3 FB1, FB2, FB3 FERRITE SMT luying STBL2012-121 1 J1 CONN HOUSING 3POS .156 W/O Molex/Waldom 09-50-7031 RAMP Electronics Corp 1 J1 CONN TERM FEMALE 18-24AWG Molex/Waldom 08-50-0105 TIN Electronics Corp 1 J1 HOOK-UP WIRE 22AWG STRAND Alpha Wire 3051 RD005 RED Company 1 J1 HOOK-UP WIRE 22AWG STRAND Alpha Wire 3051 BK005 BLACK Company 1 J2 CONN MOD JACK 6-6 VERT PCB Tyco 5520258-3 50AU Electronics/Amp 1 J3 CONN HEADER 6POS .100 VERT Molex/Waldom 22-27-2061 TIN Electronics Corp 1 J4 CONN HEADER 4POS .100 VERT Molex/Waldom 22-27-2041 TIN Electronics Corp 1 J5 DB9 Female vertical Tyco 747091-2 Electronics/Amp 3 J6,J7 & J8 CONN HOUS 3POS .100 -- -- W/RAMP/RIB (Connects to J4,J5 & J6 of Lower Board) 9 J6,J7 & J8 CRIMP TERM FEMALE 22-30AWG Molex/Waldom 08-65-0805 TIN Electronics Corp Note 1: The components listed in this Bill of Materials are representative of the PCB assembly. The released BOM used in manufacturing uses all RoHS-compliant components. DS51723A-page 72 (c) 2009 Microchip Technology Inc. Bill Of Materials (BOM) TABLE B-2: Qty 3 BILL OF MATERIALS - TOP BOARD Reference Description 1 HOOK-UP WIRE 22AWG STRAND GREEN J6,J7 & J8 HOOK-UP WIRE 22AWG STRAND RED J6,J7 & J8 HOOK-UP WIRE 22AWG STRAND BLACK L301, L302 60ohm bead PCB RoHS Compliant Bare PCB, dsPIC33F and MCP3909 3-Phase Energy Meter (Power) Bottom Bd. R100, R101, R102, RES 1.00K OHM 1/8W 1% 0805 SMD R103, R104, R105, R106, R107, R108, R109, R110, R111, R301, R302, R303, R311, R312 R112, R113, R114, RES 100 OHM 1/8W 1% 0805 SMD R115, R116, R117, R307, R308, R309, R317, R318, R319 R118, R119, R120 RES 10.0 OHM 1/8W 1% 0805 SMD R121, R122, R123, RES 20.0 OHM 1/8W 1% 0805 SMD R124, R125, R126 R127, R131, R132, RES 4.7K OHM 1/8W 5% 0805 SMD R133, R134, R135, R136 R128, R310, R314, RES 470 OHM 1/8W 5% 0805 SMD R316 R129, R130 RES 4.70K OHM 1/8W 1% 0805 SMD R313 RES 510 OHM 1/8W 5% 0805 SMD R321 RES 10.0K OHM 1/8W 1% 0805 SMD S301 SWITCH TACT 6MM MOMENTARY 250GF X100 3.2768Mhz Crystal X303 7.3728Mhz Crystal U305 2uA Low Dropout Positive Voltage Regulator U306 Low Quiescent Current LDO 1 U307 1 U308 1 U309 3 3 2 1 17 12 3 6 7 4 2 1 1 1 1 1 1 J6,J7 & J8 High-Performance, 16-bit Digital Signal Controllers IC DRVR/RCVR MULTCH RS232 16SOIC 4K I2CTM Serial EEPROM Manufacturer Part Number Alpha Wire Company Alpha Wire Company Alpha Wire Company Jones -- 3051 GR005 B60 104-00159 Panasonic - ECG ERJ-6ENF1001V Panasonic - ECG ERJ-6ENF1000V Panasonic - ECG Panasonic - ECG ERJ-6ENF10R0V ERJ-6ENF20R0V Panasonic - ECG ERJ-6GEYJ472V Panasonic - ECG ERJ-6GEYJ471V Yageo Corporation Panasonic - ECG Panasonic - ECG E-Switch 9C08052A4701FKHFT ERJ-6GEYJ511V ERJ-6ENF1002V TL1105BF250Q Koan Koan Microchip Technology Inc Microchip Technology Inc Microchip Technology Inc Texas Instruments DIP-8-3.2768M HC-49S-SMD-7.3728M MCP1701T-5002I/MB 3051 RD005 3051 BK005 MCP1700T-3302E/TT dsPIC33FJ128GP206 MAX3232CDR Microchip 24LC04B-E/SN Technology Inc 3 U102, U103 U104 Energy Metering IC with SPI Interface Microchip MCP3909-I/SS and Active Power Pulse Output Technology Inc 1 U101 1MHz, Low Power Op-Amp Microchip MCP6002-I/SN Technology Inc 3 T101 T102, T103 5A/5mA Current transformer Xinge SCT954F 1 T100 " DO NOT POPULATE -- -- Note 1: The components listed in this Bill of Materials are representative of the PCB assembly. The released BOM used in manufacturing uses all RoHS-compliant components. (c) 2009 Microchip Technology Inc. DS51723A-page 73 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design NOTES: DS51723A-page 74 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Appendix C. Power Calculation Theory C.1 OVERVIEW This MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design is unique in that all calculations are done in the frequency domain. This is easily realized using the DSP engine core of the advanced 16-bit MCU, the dsPIC33F. In addition to performing direct fourier transforms (DFTs) on all the input channel measurements, an additional firmware function is included, quasi-synchronous sampling. C.2 SYNCHRONOUS SAMPLING AND QUASI-SYNCHRONOUS SAMPLING The fundamentals of quasi-synchronous sampling and corresponding methods for measuring AC electrical parameters are discussed in this section. Typically, a synchronous sampling method is used for measuring electrical parameters. The method requires synchronization between sampling intervals and power grid frequency. For these types of meter designs, an external hardware PLL circuit is used to track power grid frequency, and the clock of the MCP3909 device is automatically updated to change the sampling frequency. Since the PLL output frequency drops behind the power grid frequency, a synchronous error exists in the system, and fully synchronous sampling is hard to achieve. In addition, as non-sine waves exist in the power grid, which may affect zero-crossing detection, when such conditions worsen, it may cause PLL failure, preventing the system from working normally. For the MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design presented here, the dsPIC33F performs additionall calculations that eliminates the need for a costly external PLL circuit. This quasi-synchronous sampling method has an advantage in the engineering practice, which actually is periodic sampling, without synchronizing with the power grid frequency. Therefore, the zero-crossing detection and PLL circuit can be reduced to lower the hardware complexity. The tradeoff is the increased software requirements of the system, easily realized using the powerful dsPIC33F. For DFT or FFT harmonic analysis of periodic signals, a Fourier transform may only bring accurate spectrum analysis when the sampling points satisfy N > 2M for each line cycle and strict synchronous sampling is realized. (Where M is the maximal harmonic order of periodic signals, and N is the number of samples per line cycle). Otherwise, if N 2M, it will cause spectrum aliasing. In addition, if strict synchronous sampling cannot be realized, spectral-leakage will occur (the Hurdle Effect). However, in the quasi-synchronous sampling method emplyed here, strict synchronization between sampling intervals and the period of sampled signal is not guaranteed but is overcome through post-processing and iteration of the collected data. To reduce the error caused by this problem and to obtain better accuracy when measuring the fundamental and harmonics of higher orders, an increase in the number of iterations when processing data to improve accuracy is performed. The iterations can effectively reduce the impact of synchronization error over the measurement accuracy, and is one of the methods to realize accurate measurement of the frequency and harmonics under steady-state conditions. (c) 2009 Microchip Technology Inc. DS51723A-page 75 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design TN*Ts V t Ts FIGURE C-1: C.2.1 Quasi Synchronous Sampline. Basic Idea of Quasi-Synchronous Sampling Assuming that the average of a periodic signal in one cycle is g(t), EQUATION C-1: ( TO + T ) T 1 1 g ( t ) = --- g ( t )dt = --- T T 0 g ( t )dt TO Make t = x/, then, EQUATION C-2: 2 1 g ( t ) = ------ 2 f ( x ) dx = f(x) 0 where f(x) = g(x/), and the period is 2. If the entire period sampling cannot be realized while a sampling frequency deviation exists, then: EQUATION C-3: (2 + ) 1 f ( x ) ----------------- 2 + ( + 2 + ) 1 f ( x ) dx ----------------- 2 + 0 f ( x ) dx We have: EQUATION C-4: ( + 2 + ) 1 1 F ( ) = ----------------- 2 + f ( x ) dx The value of F1() is a function of and also a function with 2 as its perod. The non--synchronous sampling error E = f(x)- F1(). As F1() is function with 2 as its period, its value may be averaged through integration within the range of 0-2, and it can be deduced that f(x) = F1(). DS51723A-page 76 (c) 2009 Microchip Technology Inc. Power Calculation Theory Assuming that the integral starts at , then: EQUATION C-5: ( + 2) 1 f ( x ) = F ( ) = ------ 2 1 1 F ( ) d Likewise, as strict integration cannot be realized in the entire cycle, so: EQUATION C-6: ( + 2 + ) 1 f ( x ) = F ( ) ----------------- 2 + 1 1 F ( ) d Similarly, the integral value of above equation is related to with 2 as its period, let's denote it as F2(). If it won't confuse people, we'll write F1() and F2() as F1(x) and F2(x), and a recurrence formula can be obtained as the following: EQUATION C-7: (x + 2 + ) n 1 F ( ) = ----------------- 2 + F n-1 ( x ) dx x It can be proved that, EQUATION C-8: n lim F ( ) = f ( x ) n- In practical applications, it is necessary to sample the continuous analog signals and process the data obtained with discrete algorithms. The quasi-synchronous recursive process mentioned above can be expressed as follows: For Equation C-4, the integral interval [x0, x0 + n x (2 + )] whose width is n x (2 + ) can be divided equally into n x N sections, which results in n x N + 1 sampled data, f(xi), (i=0,1,...,nxN), and we can iterate as follows: (c) 2009 Microchip Technology Inc. DS51723A-page 77 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design First iteration: N 1 1 F 0 = ------------- N i i f ( xi ) i=0 i=0 N+1 1 1 F 1 = --------------N+1 i i f ( xi ) i=1 i=1 ... Nxn F 1 (n - 1) x N 1 - = --------------------------------Nxn i = (n - 1) x N i i f ( xi ) i = (n - 1 ) x N Second iteration: N 2 1 F 0 = ------------- N i i Fi 1 i=0 i=0 N+1 F1 2 1 = -------------- N+1 i i Fi 1 i=1 i=1 ... N x (n - 1) F 2 (n - 2) x N 1 - = --------------------------------N x (n - 1) i = (n - 2) x N i i Fi 1 i = (n - 2) x N DS51723A-page 78 (c) 2009 Microchip Technology Inc. Power Calculation Theory Third iteration: N F0 3 1 = ------------- N i i Fi 2 i=0 i=0 N+1 F1 3 1 = -------------- N+1 i i Fi 2 i=1 i=1 ... N x (n - 2) F 3 (n - 3) x N 1 - = --------------------------------N x (n - 2) i = (n - 3) x N i i Fi 2 i = (n - 3) x N ... (c) 2009 Microchip Technology Inc. DS51723A-page 79 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design N-th iteration: N F0 n 1 = ------------- N i i Fi n-1 i=0 i=0 Where i is the weight coefficient which is decided by the digital quadrature formula. Complex rectangular quadrature algorithm or complex trapezoidal quadrature algorithm is usually used in quasi-synchronous sampling. Figure C-2 shows a 3-cycle interative process. Original data f0 f1 f2 ...... First Iteration F01 F11 F21 Second Iteration F02 F12 Third Iteration F03 FIGURE C-2: fN fN+1 fN+2 ...... FN1 F1N+1 F1N+2 ...... f2N f2N+1 f2N+2 ...... f3N ...... F12N F22 ...... FN2 3-Cycle Iterative Process. In practical applicatons, a frequency offset is usually small, and good results may usually be obtained through 3-5 iterations. As mentioned above, the iterative process will result in a group of weight coefficients i, called weight coefficients of quasi-synchronous algorithm, they may be deduced from the numeric quadrature formula. The relationship between the iterative result and original data is shown in Equation C-9. EQUATION C-9: nxN F0 n 1 = -------------- nxN i i f ( xi ) i=0 i=0 nxN 1 = -----n- N i f ( xi ) i=0 nxN = Ri f ( xi ) i=0 DS51723A-page 80 (c) 2009 Microchip Technology Inc. Power Calculation Theory Where: EQUATION C-10: 1 R i = -----n- i N i = 0 ~ n x N Equation C-10 is called the quasi-synchronous window function. After determining the sampling points, the number of iterations and numerical quadrature method, the coefficients of quasi-synchronous window function will become definite, and a quasi-synchronous window function arrays may be established in advance. Using the quasi-synchronous window function to carry out the weighted process of the original data is equivalent to carrying out data synchronization one time, and the algorithm realization is also very simple that only a multiplication of the original data and quasi-synchronous window function arrays is required. After processing, the new periodic signal will have the same period and frequency components as the original signal, and the synchronization error of the new signal becomes smaller. Figure C-3 is a schematic of the quasi-synchronous window function characteristics in the time domain and data processing. In Figure C-3, the red curve is the characteristic of the window function, and the blue curve is input signal, and the green curve is the output signal. Window function Amplitude Raw data Data processed Sampling point (1-193) Time 3.2 ksps FIGURE C-3: Quasi-Synchronous Window Function Characteristics Curve and Data Processing. (c) 2009 Microchip Technology Inc. DS51723A-page 81 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.3 THE HARMONIC ANALYSIS ALGORITHM OF QUASI-SYNCHRONOUS SAMPLING Periodic signal can be expressed as trigonometric Fourier series or exponential Fourier series. A periodic signal with a period of T can be expressed as : EQUATION C-11: 0 f ( t ) = ------- + 2 ( k cos ( k t ) + cos ( k t ) ) k=1 where: EQUATION C-12: T 2 a k = --- f ( t ) cos ( k t )dt T 0 EQUATION C-13: T 2 b k = --- f ( t ) sin ( k t ) dt T 0 or as: EQUATION C-14: f ( t ) = a0 + ck sin ( k t + k ) k=1 where the relationship between ak, bk ck and k is: EQUATION C-15: ck = 2 2 ak + bk EQUATION C-16: a bk k = atan ----k- EQUATION C-17: a k = c k sin k DS51723A-page 82 (c) 2009 Microchip Technology Inc. Power Calculation Theory EQUATION C-18: b k = c k cos k Make g(t) = f(t) * cos(K**t), it can be proved that g(t) is also a periondic function with T as its period. Averaging g(t) in the range of [0 ~ T] results in: EQUATION C-19: T 1 g ( t ) = --- f ( t ) cos ( k t )dt T 0 So ak = 2 x g(t). Therefore, ak can be obtained if only g(t) can be calculated. EQUATION C-20: nxN 2 a k = 2g ( t ) = -----n- N i gi i=0 nxN 2 = ------n N - i i fi cos k ----N 2 i=0 EQUATION C-21: nxN 2 b k = -----n- N 2 - i i fi sin k ----N i=0 Where N, n and i are constants. The equation may therefore be written as: EQUATION C-22: nxN ak = 2 - i Ii fi cos k ----N i=0 EQUATION C-23: nxN bk = 2 Ii fi sin k -----N- i i=0 Where: EQUATION C-24: 2 I i = -----n- 1 = R 1 x 2 N (c) 2009 Microchip Technology Inc. (i = 0 n x N) DS51723A-page 83 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.4 MEASURING THE VOLTAGE/CURRENT RMS VALUE AND POWER USING QUASI-SYNCHRONOUS SAMPLING ALGORITHM From Equation C-11, a periodic voltage can be expressed as: EQUATION C-25: U a0 U ( t ) = ----------+ 2 ( uak cos ( k t ) + ubk sin ( k t ) ) k=1 So the voltage fundamental and the voltage of each harmonic can be expressed as: EQUATION C-26: U k ( t ) = u ak cos ( k t ) + u k sin ( k t ) From Equation C-25, the fundamental voltage and voltage of each harmonic can also be expressed as: EQUATION C-27: U k ( t ) = u ck sin ( k t + uk ) Where: EQUATION C-28: u ck = 2 u ak + u bk 2 EQUATION C-29: u u bk ak uk = atan ------- The voltage RMS value of each harmonic, then can be expressed as shown in Equation C-31 with its initial phase angle shown in Equation C-29. EQUATION C-30: 1 U k = ------- u ck = 2 DS51723A-page 84 2 2 u ak + u bk ---------------------------2 (c) 2009 Microchip Technology Inc. Power Calculation Theory The relationship between uak, ubk and Uk can be expressed as: EQUATION C-31: u ak = 2 U k sin uk u bk = 2 U k cos uk EQUATION C-32: Total effective voltage can be expressed as: EQUATION C-33: U total = Uk 2 k=0 Similarly, the effective values and initial phase angles of fundamental current and current of each other harmonic can be expressed as: EQUATION C-34: 2 2 i ak + i bk ------------------------2 1 I k = ------- i ck = 2 EQUATION C-35: i i bk ak ik = atan ----- The relationship between iak, ibk and Ik can be expressed as: EQUATION C-36: i ak = 2 I k sin ik i bk = 2 I k cos ik EQUATION C-37: Total current RMS can be expressed as: EQUATION C-38: I total = Ik 2 k=0 (c) 2009 Microchip Technology Inc. DS51723A-page 85 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design According to the definition of power measurement, the active power and reactive power of the fundamental and each other harmonic can be expressed as: EQUATION C-39: P k = U k I k cos ( uk - ik ) 1 = --- U k I k ( sin uk sin ik + cos uk cos ik ) 2 EQUATION C-40: Q k = U k I k sin ( uk - ik ) 1 = --- U k I k ( sin uk cos ik - cos uk sin ik ) 2 Substituting Equation C-31, C-32, C-33 and C-37 into Equation C-39 and C-40, the power of each harmonic component can be obtained with the following: EQUATION C-41: 1 P k = --- ( u ak i ak + u bk i ik ) 2 EQUATION C-42: 1 Q k = --- ( u ak i ak - u bk i ik ) 2 Total active power and reactive power can be expressed as: EQUATION C-43: P total = Pk k=0 EQUATION C-44: Q total = Qk k=0 DS51723A-page 86 (c) 2009 Microchip Technology Inc. Power Calculation Theory C.5 MEASURING FREQUENCY There are many ways to measure frequency, with the most common being counting the signal cycle. In this method, a counter increments each time a zero-crossing is detected. Based on the counts, the width of a cycle can be measured. If the zero-crossing is accurate and the counter precision is high enough, cycle counting can be a simple and practical method. But if the input signal has large harmonic components, causing distortion around zero-crossing, then this approach may produce large errors. Another method is to analyze and process the sampled data and calculate the frequencies. Analysis may be carried out in time domain, such as digital differential ND and interpolation method; or may be carried out in frequency domain after DFT transformation, such as gravity center method, spectrum zoom method and phase difference method, among which the phase difference method is the most common one. It is not sensitive to signal distortion around zero-crossing points. The basic idea of the phase difference method is: if the rough range of to-be-measured signal frequency is known, then we may assume a frequency that is close to the actual frequency and then acquire an array of samples based on the assumed frequency. In the sampled data, the phase of the 1st cycle and the subsequent N-th cycle are meaured and their difference may be calculated. Then the phase difference may be used to calculate the difference between the actual freqency and the assumed frequence, thus figuring out the actual frequency. If the frequency f0 to be measured is known to be a definite value f, i.e., f0 = f + f, f << f, then from Equation C-27, the fundamental signal can be expressed as: EQUATION C-45: U 1 ( t ) = u c1 sin ( t + u1 ) = u c1 sin ( 2 f 0 t + u1 ) If: EQUATION C-46: T = 1--f EQUATION C-47: T 2 u a = --- U 1 ( t ) cos ( t ) dt T 0 T 2 = --- u c1 sin ( 2 f 0 t + u1 ) cos ( 2 ft ) dt T 0 EQUATION C-48: T 2 u b = --- U 1 ( t ) sin ( t ) dt T 0 T 2 = --- u c1 sin ( 2 f 0 t + u1 ) sin ( 2 ft ) dt T 0 (c) 2009 Microchip Technology Inc. DS51723A-page 87 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design We get: EQUATION C-49: f ( f + f ) sin ( u1 ) sin --------- f 2u c1 u a = ----------- ------------------------------------------------------------------------ ( 2f + f ) f T EQUATION C-50: f f cos ( u1 ) sin --------- 2u c1 f u b = ----------- -------------------------------------------------------T ( 2f + f ) f As f << f, from Equation C-49 and C-50, we have: EQUATION C-51: u a sin ( u1 ) ----- ---------------------u b cos ( u1 ) Therefore, EQUATION C-52: , (ub = 0, u a > 0) 2 3 , (ub = 0, u a < 0) 2 -1 u a u1 tg ( u ), (ub > 0, u a > 0) b tg -1 ( u a ) + , (ub > 0, u a < 0) ub -1 u a tg ( ) + 2 , (ub < 0, u a > 0) ub Assuming that the signal's initial phase angle measured in the 1st cycle is 1 and in the N-th is N, then the difference between actual frequency and the assumed frequency is: EQUATION C-53: ( N - 1 ) f , ( N - 1 < ) 2 N ( - + 2 ) f 1 f N , ( N - 1 < - ) 2 N ( N - 1 - 2 ) f , ( - > ) N 1 2 N DS51723A-page 88 (c) 2009 Microchip Technology Inc. Power Calculation Theory C.6 IMPROVING MEASUREMENT PRECISION OF QUASI-SYNCHRONOUS SAMPLING ALGORITHM When using the quasi-synchronous sampling method for harmonic analysis and calculation of power as well as voltage and current, strict restrictions apply for the algorithm and compensation, (i.e., the frequency offset must not exceed 1% of the central frequency). Precision of the result increases as the frequency offset gets less. Measurement accuracy is not guaranteed, if this condition can not be met. Figure C-4 shows the quasi-synchronization algorithm using 3 iterations with input signal ranging from 47.5 Hz to 52.5 Hz. The algorithm is for calculating the active power, the reactive power and the relative error of current and voltage. Figure C-4 shows that the algorithm works well when the frequency falls in the range of 47.5 Hz to 52.5 Hz. As the frequency deviates from the range, the error increases significantly. Therefore, the algorithm needs to be improved to fit into more applications with a more relaxed restriction. FIGURE C-4: Quasi-sync Algorithm Error Analysis of 3 Iterations. The quasi-sync sampling algorithm has relative high accuracy in frequency measurement and the error can be less than 0.005 Hz. If the frequency range to be measured can be segmented to make the frequency input closest to the multiple of cycle point, and processed using appropriate quasi-sync window function and sine/cosine tale, then the algorithm can be used for a much wider range of frequency . Figure C-5 is the error analysis of the improved 3-iteration quasi-synchronous algorithm at 3.2 ksps. It shows that the relative error for each result can be well controlled when the frequency of the input signal falls in the range of 47.5 Hz to 52.5 Hz. Figure C-5 clearly shows that the relative errors of the current, voltage, active power and reactive power in the entire frequency range are less than 0.08%. Also, when the input frequency is around the multiple of cycle frequencies (52.459 Hz, 51.613 Hz, 50.794 Hz, 50.0 Hz, 49.231 Hz, 48.485 Hz and 47.761 Hz), the calculateion error is (c) 2009 Microchip Technology Inc. DS51723A-page 89 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design minimal (<0.01%). When the input frequency deviates from the multiple of cycle frequencies, the calculation error increases rapidly. As the calculation error is related to the frequency offset to the multiple of freqency point, the calculation error caused by frequency offset can be corrected. Figure C-6 is the error analysis after frequency offset correction using simple parabolic interpolation. Calcultion errors for all parameters are shown to be less than 0.015% after the correction. DS51723A-page 90 FIGURE C-5: Iterations. Error Analysis Of Improved Quasi-sync Calculation Using 3 FIGURE C-6: Compensation. Calculation Error Analysis After The Frequency Offset (c) 2009 Microchip Technology Inc. Power Calculation Theory C.7 MEASURING SECONDARY PARAMETERS The methods of measuring parameters such as RMS values of voltage and current, active power, reactive power and frequency have been discussed in previous sections. These are primary parameters that need to be calculated from the original data. There are some other parameters called secondary parameter, such as power factor of each phase, total reactive power, total active power, total power factor, harmonic components and cumulative energy. They are obtained indirectly from primary parameters. The measurement of secondary parameters is discussed in this section. C.7.0.1 TOTAL ACTIVE POWER AND TOTAL REACTIVE POWER For 3-phase 4-wire systems, 3-phase total active power and recative power are the sum of power of each phase, respectively, which can be expressed as: EQUATION C-54: P = PA + P B + PC EQUATION C-55: Q = QA + QB + QC C.8 APPARENT POWER OF EACH PHASE AND TOTAL APPARENT POWER Apparent power is defined as: EQUATION C-56: S = C.9 2 Q +p 2 POWER FACTOR OF EACH PHASE AND TOTAL POWER FACTOR Power factor is defined as the ratio of active power to apparent power. The definition can be represented as shown in Equation C-57: EQUATION C-57: P PF = ----------------------2 2 P +Q (c) 2009 Microchip Technology Inc. DS51723A-page 91 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.10 Active Energy AND REACTIVE ENERGY Active energy is defined as the integral of active power over time, which is: EQUATION C-58: N T W = 0 P ( t ) dt u(k ) i( k) t = k=0 In this design, active energy is obtained from multiplying the voltage by the current sampled each time. The phase angle difference is compensated after each power measurement is completed. For reactive power, the cumulative reactive energy over a time period can be calculated by measuring the average power and calculating the time interval between 2 measurements. EQUATION C-59: T Vr = 0 Q ( t )dt C.11 POSITIVE/NEGATIVE ACTIVE ENERGY, POSITIVE/NEGATIVE REACTIVE ENERGY AND FOUR-QUADRANT REACTIVE ENERGY In the measurement plane, the horizontal axis denotes voltage vector U (fixed on the horizontal axis). The instantaneous current vector is used to represent the power transfer, and has a phase angle against vector U. is positive in counter-clockwise direction. Power exchange can be defined in 4 scenarios: * Quadrant I (P>0, Q>0): active energy and reactive energy are sent out at the same time; * Quadrant II (P<0, Q>0): active energy is sent in while reactive energy is sent out; * Quadrant III (P<0, Q<0): active energy is sent in while reactive energy is absorbed; * Quadrant IV (P>0, Q<0): active energy is sent out while the reactive energy is absorbed. 1. Positive active energy and negative active energy: accumulated active energy can be defined as positive and negative depending on the direction of active current. When the direction of active current is positive (from power grid to loads), active energy is positive (where active power P>0, corresponding to quadrants I and IV, which means that loads are drawing energy from grid). When current moves from loads to power grid, it is defined as negative active energy (where active power P < 0, corresponding to quadrants II and III, which means energy is provided to grid). Usually only positive active energy is taken into account in active energy, but in practice negative active energy may be taken into account as well, if necessary. 2. Positive reactive energy and negative reactive energy: If reactive power Q > 0 (corresponding to quadrants I and II), it means power grid is providing reactive energy to loads, so the energy is defined as positive reactive energy. When reactive power Q < 0 (corresponding to quadrants III and IV), it means that loads are providing reactive energy to power grid, so the energy is defined as negative reactive energy. DS51723A-page 92 (c) 2009 Microchip Technology Inc. Power Calculation Theory 3. Four-quadrant reactive energy: metering reactive energy in positive/negative reactive energy cannot truly reflect the status of reactive energy, whereas 4-quandrant reactive energy measuring gives a true picture of energy exchange. Reactive energy in four quadrants represents four different reactive energy (see Figure C-7). And the reactive energy is accumulated depending on which quadrant it is located. Reactive in (+R) I (RL) II (RC) Current vector Active in (+A) Active out (-A) Voltage vector III (-RL) IV (-RC) Reactive out (-R) Where: AR--R --R -- L = is active energy, R = is reactive energy RL = is inductive reactive energy RC = s capacitive reactive energy 2.4 FIGURE C-7: (c) 2009 Microchip Technology Inc. C A Definition Of 4 Quadrants To Measure Electrical Energy. DS51723A-page 93 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.12 HARMONIC COMPONENTS OF CURRENT, VOLTAGE AND TOTAL HARMONIC DISTORTION In Section C.3 "The Harmonic Analysis Algorithm Of Quasi-synchronous Sampling", we discussed the measuring of current and voltage signals for each order of harmonics. 3 parameters are used to show to what extent a distorted wave deviates from a sine wave. They are: harmonic content, total distortion and harmonic ratio of the k-th harmonic. The term harmonic content means the root of square of effective values for all harmonics, which is defined as: EQUATION C-60: N UH = Uk 2 k=2 The total voltage distortion ratio of harmonics is the ratio of harmonic content to the fundamental in percentage, which is defined as: EQUATION C-61: THD U UH = ------- x 100% UI N ( THDUk ) 2 k=2 The k-th harmonic ratio for voltage is the ratio of the k-th harmonic to the fundamental in percentage, defined as: EQUATION C-62: Uk THD Uk = ------ x 100% U1 Similarly, harmonics of each order for the current and total distortion ratio can be sorted out. DS51723A-page 94 (c) 2009 Microchip Technology Inc. Power Calculation Theory C.13 COMPENSATION FOR RATIO ERROR AND PHASE LAG The error of the current transformer (CT) is a complex, which can usually be expressed by two orthogonal parts, current error (f) and phase lag (). EQUATION C-63: = f+j The current error, also known as ratio error, can be written in percentage as: EQUATION C-64: ( kn I2 - I1 ) f = 100 ------------------------I1 Where: Kn = I1 = I2 = the rated current ratio the primary current the secondary current that passes I1 under the test condition Phase lag, also known as angle error, is the phase difference between primary and secondary current vector, in 'minute'. Different current transformers have different errors. Current transformers are classified into different accuracy classes depending on their error magnitudes. The accuracy class of a transformer is nominated by the percentage of the maximal ratio error allowed for a given rated current. Accuracy classes and corresponding error limits for a current transformer are listed in Table C-1. Accurate Class TABLE C-1: ACCURACY CLASS AND ERROR LIMIT FOR THE CURRENT TRANSFORMER Ratio Error (%) Phase lag Rated Current (%) 1 (Grad) Rated Current (%) 5 20 100 120 0.1 0.4 0.2 0.1 0.1 0.2 0.75 0.35 0.2 0.2 30 15 10 10 0.9 0.45 0.3 0.3 0.5 1.5 0.75 0.5 0.5 90 45 30 30 2.7 1.35 0.9 0.9 3 1.5 1.0 1.0 180 90 60 60 5.4 2.7 1.8 1.8 1 Rated Current (%) 50 120 3 3 3 5 5 5 (c) 2009 Microchip Technology Inc. 1 (%) Rated Current (%) 5 20 100 120 15 8 5 5 1 5 20 100 120 0.45 0.24 0.15 0.15 Phase difference Phase difference of class 3 and class 5 are not specified. DS51723A-page 95 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.14 RELATIONSHIP BETWEEN ERROR AND CURRENT For a given load and frequency, the absolute ratio error and angle error increase when the primary current decreases from the rated value for un-compensated current transformer. The reason is that with the decrease of the secondary current, the magnetic permeability of the iron core varies non-linearly, resulting in less reduction in the field ampere turns. Figure C-8 is a typical curve for load current and the phase lag of a CT. Generally, phase lag of the output signal is great when the load current is small, and also increases at a fast rate. Phase Lag Current Load FIGURE C-8: DS51723A-page 96 Current Load Versus CT Phase Lag. (c) 2009 Microchip Technology Inc. Power Calculation Theory C.15 RATIO ERROR COMPENSATION As non-linearity and inconsistency exist in the sampling circuit (including CT and back-end shunt resistors) and the ADC circuit, it is necessary to compensate for ratio error to the system. Figure C-9 is the transfer link of the voltage and current channels. The compensation for ratio error is quite simple. It compares the measured value against the actual input value under certain input conditions and obtains a correction coefficient. EQUATION C-65: Calibration Value Correction coefficient = Coefficient before correction x ------------------------------------------Measured Value Voltage secondary Sampling resistor Input Voltage R0 R0 Voltage primary Sampling resistor CT 1:1 ADC R0 R2 Input Current CT 2000:1 ADC R2 Current secondary Sampling resistor FIGURE C-9: Error Caused By Sampling Circuit And ADC. Since current has a large dynamic range, for a meter which requires high accuracy (0.2s and 0.5s), the multi-point calibration method is needed to meet input requirement for full range. The MCP3909 device's current channel includes an adjustable gain amplifier. The ratio error must be recalibrated for different amplification, but only needs to be calibrated once under the same amplification conditions. (c) 2009 Microchip Technology Inc. DS51723A-page 97 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design C.16 PHASE LAG COMPENSATION Phase lag of a CT has no effect on the metering of RMS current/voltage and apparent power, but will affect the metering of power, since the phase lag will change the phase relationship between the input current and the voltage. This will result in a deviation of the calculated active power from the calculated reactive power. Figure C-10 shows how a transformer's phase lag affects the measured results under both inductive and capacitive loads. Let's assume that the output of CT has no phase lag from the input voltage, while the CT has a phase lag from the input signal. With inductive loads, the phase angle increases between the volatge and the current because of the phase lag induced by the CT, resulting in a decrease of the measured active power and an increase of the reactive power. While with capacitive loads, the phase angle between the voltage and the current decreases because of the phase lag induced by the CT, resulting in a decrease of the measured reactive power and an increase of active power. There are many methods for phase lag compensation. In this design the result correction method is used. It compensates with a coefficient after the active power and reactive power are figured out, which has a small amount of calculation. Inductive Load Inductive Load CT Output current Input voltage Input current CT Output current Input voltage Input current Capacitive Load Capacitive Load FIGURE C-10: Measurement Change Caused By Transformer Phase Lag. Assuming that the phase lag of CT is i, of PT is u, after PT and CT, the variation of phase lag between current and voltage is: = u - i. S Q' Q f F P' P FIGURE C-11: DS51723A-page 98 Principle Of Phase Lag Correction. (c) 2009 Microchip Technology Inc. Power Calculation Theory Given that the phase lag between the input voltage and current is , after PT and CT, the actual measured active power is P', reactive power is Q'. RMS current is I, RMS voltage is V, input apparent power is S, actual input active power is P and reactive power is Q, then Figure C-11 can be drawn based on the principles of power triangle. EQUATION C-66: P' = V I cos ( + ) = P cos - Q sin EQUATION C-67: Q' = V I sin ( + ) = Q cos - P sin From the above 2 equations, we have: EQUATION C-68: P = k 1 P' + k 2 Q' EQUATION C-69: Q = k 1 Q' - k 2 P' Where: EQUATION C-70: k 1 = cos EQUATION C-71: k 1 = sin By setting up certain input conditions, can be measured, and then k1 and k2 can be calculated. In this design, we use an input of 0.5L for calibration. With this condition, can be calculated using the difference between the meaured and the actual input value of active power. For accurate calculations, the user may also use the difference between the measured cumulative energy and the actual cumulative energy to calculate the difference. EQUATION C-72: P' = a cos ----------- - --- = a cos ( 0.5 ( 1 + err ) ) - --- 2P 3 3 Where err is the error rate of the energy measurement, which results from calculating the error between the actual energy measured by a standard meter and the energy measured by the dsPIC devices. The error can be obtained from the output of the meter calibration workbench. EQUATION C-73: - P- = ------P- x 100 err = P' ------------P P (c) 2009 Microchip Technology Inc. DS51723A-page 99 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design Since the phase lag of a CT's output signal is related to the magnitude of current, different correction coefficients, K, can be set according to different RMS current values. In this design, 5 calibration points can be set. If it does not require high-precision, fewer points can be set to simplify calibration. If one-time calibration cannot meet the precision requirement, more calibrations can be done. The new angle error may still be calculated using Equation C-72. The new correction coefficient is: EQUATION C-74: k' 1 = cos ( 1 + 2 ) = k 1 cos 2 - k 2 sin 2 EQUATION C-75: k' 2 = sin ( 1 + 2 ) = k 2 cos 2 - k 1 sin 2 C.16.0.1 PHASE LAG COMPENSATION WHEN FREQUENCY VARIES For the same current intensity, the signal delay caused by the CT is the same. When the frequency of the input signal varies, the phase lag will be different. Normally, calibration is done at 50 Hz. When the frequency varies, if the same phase lag compensation coefficient for 50 Hz is still used, it will cause an error in the power measurement. In most cases, the frequency varies in a small range (test specification requires 2.5%), so it has little effect on the measurement accuracy. For meters with an accuracy of 0.5s or above, this measurement error can be ignored. But for 0.2s meters, the error cannot be ignored and the frequency variation needs to be corrected during calculation. The phase lag compensation coefficient k1 and k2 are corrected during calculation. Assuming that the freqnency is 50 Hz, the signal delay caused by CT is t, then after correction, the compensation coefficient k1 and k2 will be: EQUATION C-76: k 1 = cos = cos ( t 50 2 ) EQUATION C-77: k 2 = sin = sin ( t 50 2 ) When frequency varies, assuming that the frequency offset is f, i.e. the input signal frequency is 50 + f, then the compensation coefficient will be: EQUATION C-78: k' 1 = cos = cos ( t ( 50 + f ) 2 ) EQUATION C-79: k' 2 = sin = sin ( t ( 50 + f ) 2 ) DS51723A-page 100 (c) 2009 Microchip Technology Inc. Power Calculation Theory To avoid complexity in calculation and to maximize the correction precision, the following equations may be used to approximate K1 and K2 when the phase angle is small. EQUATION C-80: f 2 k' 1 = cos cos - ------ sin 50 f = k 1 - ------ k 2 k 2 50 EQUATION C-81: f k' 2 = sin sin - ------ sin cos 50 f = k 2 - ------ k 1 k 2 50 (c) 2009 Microchip Technology Inc. DS51723A-page 101 MCP3909 / dsPIC33F 3-Phase Energy Meter Reference Design NOTES: DS51723A-page 102 (c) 2009 Microchip Technology Inc. MCP3909 / DSPIC33F 3-PHASE ENERGY METER REFERENCE DESIGN Appendix D. 50/60 Hz Meter Operation D.1 FIRMWARE VERSIONS There are two versions of firmware for the meter due to the quasi-synchronous sampling scheme employed by the dsPIC33F firmware. The design covers the frequency of rated frequency 2.5 Hz. To change the rated frequency, just change the definition in the beginning of cacul.h - #define STD_FREQ 50.0. This is provided for download from Microchip's website, file names and checksums below. At the same time you need to change the crystal to provide clock for metering IC. TABLE D-1: FIRMWARE FILES Line Frequency Firmware Name Hex File Checksum YXX VALUE 50 Hz 60 Hz PM_1_50.ZIP PM_1_60.ZIP TBD TBD 3.857 MHz TBD (c) 2009 Microchip Technology Inc. DS51723A-page 103 WORLDWIDE SALES AND SERVICE AMERICAS ASIA/PACIFIC ASIA/PACIFIC EUROPE Corporate Office 2355 West Chandler Blvd. Chandler, AZ 85224-6199 Tel: 480-792-7200 Fax: 480-792-7277 Technical Support: http://support.microchip.com Web Address: www.microchip.com Asia Pacific Office Suites 3707-14, 37th Floor Tower 6, The Gateway Harbour City, Kowloon Hong Kong Tel: 852-2401-1200 Fax: 852-2401-3431 India - Bangalore Tel: 91-80-3090-4444 Fax: 91-80-3090-4080 India - New Delhi Tel: 91-11-4160-8631 Fax: 91-11-4160-8632 Austria - Wels Tel: 43-7242-2244-39 Fax: 43-7242-2244-393 Denmark - Copenhagen Tel: 45-4450-2828 Fax: 45-4485-2829 India - Pune Tel: 91-20-2566-1512 Fax: 91-20-2566-1513 France - Paris Tel: 33-1-69-53-63-20 Fax: 33-1-69-30-90-79 Japan - Yokohama Tel: 81-45-471- 6166 Fax: 81-45-471-6122 Germany - Munich Tel: 49-89-627-144-0 Fax: 49-89-627-144-44 Atlanta Duluth, GA Tel: 678-957-9614 Fax: 678-957-1455 Boston Westborough, MA Tel: 774-760-0087 Fax: 774-760-0088 Chicago Itasca, IL Tel: 630-285-0071 Fax: 630-285-0075 Cleveland Independence, OH Tel: 216-447-0464 Fax: 216-447-0643 Dallas Addison, TX Tel: 972-818-7423 Fax: 972-818-2924 Detroit Farmington Hills, MI Tel: 248-538-2250 Fax: 248-538-2260 Kokomo Kokomo, IN Tel: 765-864-8360 Fax: 765-864-8387 Los Angeles Mission Viejo, CA Tel: 949-462-9523 Fax: 949-462-9608 Santa Clara Santa Clara, CA Tel: 408-961-6444 Fax: 408-961-6445 Toronto Mississauga, Ontario, Canada Tel: 905-673-0699 Fax: 905-673-6509 Australia - Sydney Tel: 61-2-9868-6733 Fax: 61-2-9868-6755 China - Beijing Tel: 86-10-8528-2100 Fax: 86-10-8528-2104 China - Chengdu Tel: 86-28-8665-5511 Fax: 86-28-8665-7889 Korea - Daegu Tel: 82-53-744-4301 Fax: 82-53-744-4302 China - Hong Kong SAR Tel: 852-2401-1200 Fax: 852-2401-3431 Korea - Seoul Tel: 82-2-554-7200 Fax: 82-2-558-5932 or 82-2-558-5934 China - Nanjing Tel: 86-25-8473-2460 Fax: 86-25-8473-2470 Malaysia - Kuala Lumpur Tel: 60-3-6201-9857 Fax: 60-3-6201-9859 China - Qingdao Tel: 86-532-8502-7355 Fax: 86-532-8502-7205 Malaysia - Penang Tel: 60-4-227-8870 Fax: 60-4-227-4068 China - Shanghai Tel: 86-21-5407-5533 Fax: 86-21-5407-5066 Philippines - Manila Tel: 63-2-634-9065 Fax: 63-2-634-9069 China - Shenyang Tel: 86-24-2334-2829 Fax: 86-24-2334-2393 Singapore Tel: 65-6334-8870 Fax: 65-6334-8850 China - Shenzhen Tel: 86-755-8203-2660 Fax: 86-755-8203-1760 Taiwan - Hsin Chu Tel: 886-3-6578-300 Fax: 886-3-6578-370 China - Wuhan Tel: 86-27-5980-5300 Fax: 86-27-5980-5118 Taiwan - Kaohsiung Tel: 886-7-536-4818 Fax: 886-7-536-4803 China - Xiamen Tel: 86-592-2388138 Fax: 86-592-2388130 Taiwan - Taipei Tel: 886-2-2500-6610 Fax: 886-2-2508-0102 China - Xian Tel: 86-29-8833-7252 Fax: 86-29-8833-7256 Thailand - Bangkok Tel: 66-2-694-1351 Fax: 66-2-694-1350 Italy - Milan Tel: 39-0331-742611 Fax: 39-0331-466781 Netherlands - Drunen Tel: 31-416-690399 Fax: 31-416-690340 Spain - Madrid Tel: 34-91-708-08-90 Fax: 34-91-708-08-91 UK - Wokingham Tel: 44-118-921-5869 Fax: 44-118-921-5820 China - Zhuhai Tel: 86-756-3210040 Fax: 86-756-3210049 03/26/09 DS51723A-page 104 (c) 2009 Microchip Technology Inc.