With the rapid development of the electronics industry, various instruments and meters are widely used in all aspects of industrial control and social life, among which power instruments are particularly prominent.
Reliability requirements for power meters are a key aspect of smart meter technology standards. The standard mandates an average lifespan of at least 10 years for power meters, making reliability design crucial during the design and development process. The probability of a device fulfilling its intended function under specified conditions and within a specified time is called the Mean Time Between Failures (MTBF), also known as the Mean Time Between Failures. MTBF is a common indicator of reliability. The reliability design of power meters aims to improve the product's MTBF and ensure its normal operation.
1 Hardware Reliability Design
1.1 Power supply anti-interference design
According to engineering statistics, 70% of interference in power instrumentation systems enters the system through power supply coupling. Therefore, improving power supply quality is crucial for the reliable operation of the entire system. Since the system's power supply is generally converted from mains power, the anti-interference design of the power supply section mainly focuses on filtering at the power input port and suppressing transient interference. Figure 1 shows a typical power supply anti-interference design, where RV1 is a thermistor, VZ1 is a varistor, and LA1 is a common-mode choke. This circuit can effectively suppress surges and burst pulse interference.
Figure 1 Power supply anti-interference design circuit
In addition, modular power supply is another principle in power supply design. The advantage of this design is that it can effectively prevent high-voltage equipment from interfering with other modules in the system, thereby improving the reliability of the entire system.
1.2 Grounding Design
The design of the grounding system directly affects the overall product's anti-interference capability. A good design can block interference from the external environment and effectively suppress internal coupling noise. The following two considerations can improve system reliability:
(1) Digital Ground and Analog Ground. Because digital signals have steep edges, the ground current in digital circuits exhibits pulse-like changes. Therefore, in power instrumentation systems, analog ground and digital ground should be designed separately, connected only at one point, and the analog and digital circuits on the circuit board should be connected to their respective "grounds". This effectively prevents the pulse signal of the digital circuit's ground current from coupling into the analog circuit through the common ground impedance, thus avoiding transient interference. This interference is more pronounced when high-frequency, large signals are present in the system.
(2) Single-point and multi-point grounding. In low-frequency systems, grounding generally adopts a combination of parallel single-point grounding and series single-point grounding to improve system performance. Parallel single-point grounding refers to the ground wires of multiple modules being connected to one point, with the location of each module related to its own current and resistance. The advantage of this grounding method is that there is no coupling interference from the common ground resistance, but the disadvantage is that too many ground wires are used. Series single-point grounding refers to multiple modules using the same section of ground wire. Because the equivalent resistance of the current on the ground wire will generate a voltage drop, the potential of the connection point between the module and the ground wire to the ground is different. Changes in the current of all modules will affect the potential at the connection point, causing changes in the circuit output, and ultimately leading to coupling interference from the common ground resistance. This method has the advantages of simple wiring. Multi-point grounding is often used in high-frequency systems. The principle is that the ground wires of each module are connected to the ground busbar as close as possible. The advantages are short ground wires, low impedance, and no interference noise caused by the common ground impedance.
1.3 Isolation Design
One of the main purposes of isolation design is to isolate noise sources from sensitive circuits. The characteristic of isolation design is that power instruments maintain signal communication with the operating environment without electrical interaction. The main methods for achieving isolation design include transformer isolation, opto-isolation, relay isolation, isolation amplifiers, and wiring isolation.
(1) Transformer isolation. Pulse transformers have the characteristics of few turns, small winding distributed capacitance (only a few picofarads), and primary and secondary windings wound on both sides of the magnetic core respectively. They can be used as isolation devices for pulse signals to achieve isolation of digital signals.
(2) Opto-isolation. Adding an optocoupler can suppress spikes and various noise interferences. Opto-isolation eliminates electrical interaction between the host computer system and the power meter's communication port, improving the system's anti-interference performance. Optocouplers can isolate digital signals but are not suitable for analog signals. Common methods for isolating analog signals include: ① converting to an opto-isolation circuit (this circuit is complex); ② differential amplifiers (the isolated voltage is relatively low); ③ isolation amplifiers (although they have good performance, they are expensive).
(3) Relay isolation. Since there is no electrical connection between the coil and the contacts of a relay, the coil can be used to receive signals and then transmit signals through its contacts. This can effectively solve the problem of contact between high-voltage and low-voltage signals and complete interference isolation.
(4) Wiring isolation. Isolation is achieved through the layout of the circuit board, mainly the isolation between high-voltage and low-voltage circuits.
1.4 Printed Circuit Board Anti-interference Design
Printed circuit boards (PCBs) are the carriers of circuit components, providing electrical connections between them. The quality of PCB design directly affects the system's anti-interference capability. The following principles are generally followed when designing PCBs:
(1) When wiring the crystal oscillator, try to get it as close as possible to the pins of the central processing unit. Ground and fix its casing. Finally, use a ground wire to isolate the clock area. This method can avoid many difficult problems.
(2) Under the condition of meeting the system performance requirements, the central processing unit should use a low-frequency crystal oscillator as much as possible, and the digital circuit should be as low-speed as possible.
(3) Unused input and output port resources of the central processing unit should not be left unused. They should be connected to the system power supply or grounded. The same applies to other chips.
(4) The connection between high-frequency components should be shortened as much as possible, and components with input and output functions should be kept as far apart as possible. Components that are easily interfered with should not be placed too close.
(5) Current loops should not appear in low-frequency and weak signal circuits. If it is impossible to avoid them, the loop should be made as small as possible to reduce induced noise.
(6) 90° bends should be avoided when wiring the system to prevent high-frequency noise emission;
(7) The input and output lines in the system should not be parallel as much as possible, and a ground wire should be added between the two wires. This can effectively prevent feedback coupling.
2 Software Reliability Design
2.1 Digital Filter Design
Currently, various metering chips are widely used in power meters. The central processing unit (CPU) communicates with these chips via serial peripheral interfaces or universal asynchronous transceivers (UASTs) to obtain power system operating parameters. If the bus is interfered with during communication, or if the metering chip is in an abnormal state, the CPU will receive erroneous data. Therefore, incorporating filtering into the software program is crucial. For ordinary power parameters, the mean method can be used. When calculating the effective value, five to six data points are collected, the maximum and minimum values are removed, and then the average is calculated. For energy data, the dynamic range of energy per unit time can be estimated based on the meter's rated operating environment. If abnormal energy data occurs, the software can discard the data. Other methods include the median method, the arithmetic mean method, and the first-order low-pass filter method. Practice has shown that the use of software filtering can maximize the reliability of each parameter reading.
2.2 Data Redundancy Design
To improve system reliability, multiple backups can be designed for system settings and calibration parameters. If one set of data becomes corrupted, another backup set can be activated. To ensure data security and increase the probability of data surviving erroneous operations, several sets of data should be stored separately.
2.3 Redundancy Design for Data Validation and Operation
When the CPU writes setting parameters or calibration parameters to storage, it may be interfered with, resulting in erroneous data being written. However, the CPU cannot determine the correctness of the written data in this situation. To ensure correct data writing, the software program performs a checksum calculation on the data to be written and writes the checksum to storage as well. After each write operation, a read operation is performed, and the read data is checked against the written checksum. If the two sets of data do not match, the write operation is repeated until the data is correctly written. If the set number of rewrites is exceeded, a write error is displayed.
2.4 Software Trap Design
Software traps are a form of instruction redundancy used to catch programs that "run away." When interfered with by noise signals, the system program may deviate from its normal operating trajectory. To stabilize the "runaway" program, designers incorporate traps into the program. A software trap uses a boot instruction to forcibly redirect the captured program to a specific address and handle errors in the disordered program. For programs that have become disordered due to interference, multi-byte instructions are the most dangerous because erroneous pointers can "run away" between multiple byte instructions, thus executing instructions at a deeper, unknown depth. Compared to multi-byte instructions, single-byte instructions can straighten out disordered pointers, allowing them to run in the correct order, effectively suppressing the disordered phenomenon. Based on the above principles, a software trap can be formed into a program. Typically, to improve the capture rate of "runaway" programs, two no-operation instructions can be added before the boot instruction, specifically in the following form:
--NOP-- --NOP-- JUMP ERROR
In a program, a jump error redirects the program to an error handling routine. Using software traps in four locations—large unused areas of read-only memory, unused interrupt vector regions, breaks in the program structure, and the beginning and end of tables—is most effective.
2.5 Software Watchdog Design
A watchdog timer uses a combination of hardware and software to prevent programs from entering an infinite loop. The hardware foundation of the watchdog timer is an independently running counter with a timer period of T. The CPU's reset pin is connected to the counter's timer output pin, and the CPU controls the counter's reset. During normal system operation, the watchdog timer is reset within time intervals less than T, preventing overflow. However, when the system malfunctions or operates abnormally, the CPU's timing logic is disrupted, preventing the counter from being reset within period T, ultimately leading to overflow. In this case, the watchdog timer generates a reset signal, which is sent to the CPU to reset it. This design allows the system to overcome temporary interference, enhancing system reliability.
3 Other precautions
3.1 Selection and Control of Components
Components are the basic units that make up power meters. The reliability level of power meters depends primarily on the reliability level of their components. Major components include: central processing unit (CPU), metering chip, digital display, LCD screen, electrolytic capacitors, varistors, current transformers, voltage transformers, crystal oscillators, surface mount capacitors, surface mount resistors, optocouplers, and batteries. Acrel has strict requirements for component selection: selected components must have reliability indicators; component procurement requires specific model specifications, suppliers, and procurement channels; and purchased components must undergo incoming inspection, warehousing inspection, and pre-use inspection.
3.2 Margin Design
The margin design ensures that the operating stress of the components is appropriately lower than the rated value of the components, thereby reducing the basic failure rate and improving the reliability of the components.
3.3 Redundancy Design
Redundant design uses one or more identical units connected in parallel, so that if one unit fails, the others can still ensure the system continues to function normally. High-reliability connections should be selected for main signal lines and cables. Redundancy techniques can be used for switches, connectors, etc., if necessary, such as parallel connection or utilizing all redundant contacts.
3.4 Reliable production process
Reliable manufacturing processes hinge on electrostatic discharge (ESD) protection and moisture control. Introducing an ESD protection system into the production environment effectively prevents damage to metal-oxide-semiconductor (MOSFET) field-effect transistors (FETs) and integrated circuit chips. High humidity allows water molecules to penetrate materials, creating leakage paths between conductors and reducing insulation resistance and voltage withstand capability. Conversely, excessive dryness can cause materials to become brittle and generate static electricity. Acrel has implemented a series of control measures in its production environment, such as dehumidification and air conditioning systems, to maintain a stable environment and minimize the impact of adverse external factors. Furthermore, some product circuit boards undergo a pre-painting process before leaving the factory to further prevent moisture absorption and improve the reliability of power instruments.
3.5 High-temperature aging treatment
Potential problems and performance defects in the welding and assembly of components in power meters can be revealed in advance through high-temperature aging. After treatment, the products undergo normal electrical parameter testing to screen out and remove components with variable values or that have failed, eliminating potential problems before the products are sold, thus ensuring that the products leaving the factory can withstand the test of time.
3.6 Maintenance
To facilitate future maintenance, the design and layout of components should take into account future maintenance work. Software design should fully consider potential system errors and list all possible errors so that when an error occurs, designers and maintenance personnel can identify and resolve the problem, thereby improving the reliability of system operation.
4 Reliability Testing Methods
4.1 Hardware Reliability Testing
Based on the anti-interference design of the system hardware, the main hardware test contents include: electrostatic discharge immunity, radio frequency electromagnetic field immunity, fast transient burst immunity, conducted disturbance immunity induced by radio frequency field, surge immunity, attenuated oscillation wave immunity, and radio interference suppression.
In addition, Acrel has introduced high-acceleration life testing and high-acceleration stress screening equipment, which use stresses higher than those present in the environment to accelerate the discovery of problems and improve the design and manufacturing process of power instruments.
4.2 Software Reliability Testing
Black-box testing techniques are generally used for embedded software reliability testing. The general process for software reliability testing is as follows: define reliability objectives, develop a test plan, perform development operations, enter the test preparation phase, execute reliability tests, analyze and evaluate the data, and generate a reliability test report based on the test data.
5. Conclusion
This article, drawing on Acrel's years of experience in manufacturing power meters, illustrates the importance of reliability in power meter design. It analyzes the types of interference sources and their causes during meter use, focuses on the hardware and software solutions adopted to address interference, and finally introduces a series of verification experiments required after the product design is completed to prove the feasibility of the meter design and ensure the high reliability of the product.
References
[1] Meng Deyi. Research on the reliability and anti-interference design of instruments and meters [J]. Science & Technology Communication, 2014, (6): 156, 169.
[2] Cai Fanghui. Reliability design of smart meters [C]. The 24th China Electrical Instruments and Meters Industry Development Forum and Exhibition, 2012: 52-54.
[3] Hou Yongchun. Research on anti-interference technology of intelligent instruments [J]. Electronic Testing, 2014, (6): 72-44.
[4] Qi Xiaoyun. A brief discussion on several technical safeguards for improving the reliability of smart meters [J]. Electronic Technology and Software Engineering. 2013, (23): 153-154.
[5] Qi Jifei, Zheng Liya. Design method to improve the reliability of smart power metering [J]. Ordnance Automation. 2008, 27(4):80-82.
[6] Li Chuanwei. Reliability analysis and anti-interference design of instruments and meters [J]. China Educational Technology Equipment. 2005, (10): 34-38.
[7] Sun Huaiyi. Reliability design and evaluation technology of intelligent instruments [J]. Industrial Control Computer. 2001, (9): 34-38.
[8] Li Lei, Xue Jianbo. Software anti-interference measures for intelligent instruments and meters [J]. Journal of Xinxiang Teachers College, 2005, 18(1):60-61.
[9] Li Yan, Sun Xiaohua, Wang Fushun. Research and design of hardware anti-interference for intelligent instruments and meters [J]. Science, Education and Culture, 2009, (22):276-277.
[10] Q/TDEI 04-2007, PZ series AC programmable digital display smart meter [S].
[11] GB/T 28171-2011, Reliability Testing Methods for Embedded Software [S].
[12] Ren Zhicheng, Zhou Zhong. Principles and Application Guide of Digital Instruments for Power Measurement [M]. Beijing: China Electric Power Press. 2007. 4
