1. Circuit design and EMC component selection
At the beginning of a new design and development project, the correct selection of active and passive components and perfect circuit design technology will help obtain EMC certification at the lowest cost and reduce the additional cost, volume and weight of the product due to shielding and filtering. These technologies can also improve the integrity of digital signals and the signal-to-noise ratio of analog signals, and can reduce the reuse of hardware and software at least once, which will also help new products meet their functional technical requirements and be put on the market as soon as possible. These EMC technologies should be regarded as part of the company's competitive advantage and help the company obtain the greatest commercial benefits.
1.1 Digital devices and EMC circuit design
1.1.1 Device Selection
Most digital IC manufacturers can produce at least a series of devices with low radiation, and can also produce several I/O chips with ESD resistance. Some manufacturers supply VLSI with good EMC performance (some EMC microprocessors have 40dB lower radiation than ordinary products); most digital circuits use square wave signals for synchronization, which will produce high-order harmonic components, as shown in Figure 1. The higher the clock rate, the steeper the edge, and the higher the frequency and harmonic emission capabilities. Therefore, under the premise of meeting the product technical indicators, try to choose a low-speed clock. Never use AC when HC can be used, and don't use HC if CMOS4000 can work. Choose integrated circuits with high integration and EMC characteristics, such as:
* The power and ground pins are close
* Multiple power and ground pins
* Small output voltage fluctuation
* Controllable switching rate
* I/O circuit matching with transmission line
* Differential signal transmission
* Low ground reflection
* Immunity to ESD and other interference phenomena
* Small input capacitance
* Output stage drive capability does not exceed the requirements of the actual application
* Low power transient current (sometimes called penetration current)
The maximum and minimum values of these parameters should be specified by their manufacturers. Devices with the same model and indicators produced by different manufacturers may have significantly different EMC characteristics, which is very important for ensuring that the products produced successively have stable electromagnetic compatibility compliance.
Manufacturers of high-tech integrated circuits can provide detailed EMC design instructions, such as Intel's Pentium MMX chip. Designers should understand these and strictly follow the requirements. Detailed EMC design suggestions show that the manufacturer is concerned about the real needs of users, which must be considered when selecting devices. In the early design stage, if the EMC characteristics of the IC are not clear, various EMC tests can be performed on a simple functional circuit (at least the clock circuit must work), and the operation should be completed as much as possible under high-speed data transmission. Emission tests can be easily performed on a standard test bench by connecting the near-field magnetic field probe to a spectrum analyzer (or broadband oscilloscope). Some devices are obviously much less noisy than others. The same probe can be used for immunity testing and connected to the output of the signal generator (continuous RF or transient). However, if the probe is specially matched with the instrument (not just a simple short-circuit ring or wire), first check whether its power handling capacity meets the requirements. During testing, the near-field probe needs to be close to the device or PCB board. In order to locate the "key detection point" and maximize the probe direction, the entire area should be scanned horizontally and vertically (so that the probes are perpendicular to each other in all directions), and then the scan should be concentrated on the area with the strongest signal.
Figure 1 Spectrum of an ideal 60MHz square wave with 1ns rise/fall time
1.1.2 IC socket should not be used
IC sockets are very bad for EMC. It is recommended to solder surface-mount chips directly on PCB. IC chips with shorter leads and smaller size are better. ICs with BGA and similar chip packages are currently the best choice. The emission and sensitive characteristics of programmable read-only memory (PROM) installed on the socket (even worse, the socket itself has a battery) often make an originally good design worse. Therefore, surface-mount programmable memory directly soldered to the circuit board should be used.
Motherboards with ZIF sockets and spring-mounted heat sinks on the processor (to facilitate upgrading) require additional filtering and shielding, but even so, it is beneficial to choose a surface-mount ZIF socket with the shortest internal leads.
1.1.3 Circuit Technology
* Use level detection (not edge detection) for inputs and buttons
* Use a digital signal with the slowest possible leading edge rate (without exceeding the distortion limit)
* On PCB prototypes, allows for control of signal edge speed or bandwidth (e.g. using soft ferrite beads or series resistors on the driver side)
* Reduce the load capacitance so that the open collector driver close to the output is easy to pull up, and the resistance value is as large as possible
* The processor heat sink is isolated from the chip by thermally conductive material, and RF grounding is performed at multiple points around the processor.
* High quality RF bypassing (decoupling) of the power supplies is important at each supply pin.
* High-quality power monitoring circuits must be immune to power interruptions, sags, surges, and transient disturbances
* Requires a high quality watchdog
* Never use programmable devices in watchdog or power monitoring circuits
* Power monitoring circuits and watchdogs also require appropriate circuit and software techniques so that they can adapt to most unexpected situations, depending on the critical state of the product
* When the rise/fall time of the logic signal edge is shorter than the time it takes for the signal to transmit a round trip in the PCB trace, transmission line technology should be used:
a. Experience: The time it takes for a signal to travel back and forth in each millimeter of track length is equal to 36 picoseconds.
b. To obtain the best EMC characteristics, use transmission line technology for much shorter traces than the experience in a.
Some digital ICs generate high-level radiation, and their small metal boxes are often soldered to the PCB ground line to achieve shielding effect. Shielding on PCB is low cost, but it is not applicable to devices that require heat dissipation and good ventilation.
The clock circuit is usually the main source of emission, and its PCB track is the most critical point. The layout of the components should be done well to make the clock routing as short as possible, while ensuring that the clock line is on one side of the PCB but not through the vias. When a clock must go through a long path to reach many loads, a clock buffer can be installed next to the load, so that the current in the long track (wire) is much smaller. Here, relative distortion is not important. The clock edge in the long track should be as smooth as possible, and even a sine wave can be used, and then shaped by the clock buffer next to the load.
1.1.4 Spread Spectrum Clock
So-called "spread spectrum clocking" is a new technology that can reduce radiated measurements, but it does not actually reduce the instantaneous emission power, so some fast-response devices may still experience the same interference. This technology modulates the clock frequency by 1% to 2%, spreading out the harmonic components so that the peaks in CISPR16 or FCC emission tests are lower. The measured emission reduction depends on the bandwidth and the integration time constant of the test receiver, so it is a bit speculative, but the technology has been accepted by the FCC and is widely used in the United States and Europe. The modulation must be kept within the audio range so that the clock signal is not distorted. Figure 2 shows an example of the improvement in clock harmonic emissions. Spread spectrum clocking cannot be used in communication networks that require strict timing, such as Ethernet, fiber, FDD, ATM, SONET, and ADSL.
Most of the problems with emissions from digital circuits are due to synchronous clock signals. Asynchronous logic (such as the AMULET microprocessor, being developed at UMIST by Professor Steve Furbe's group) will greatly reduce emissions while also achieving a true spread spectrum effect, rather than just focusing on the clock harmonics to produce emissions.
Figure 2: Radiation reduction due to clock spread spectrum
1.2 Analog Devices and Circuit Design
1.2.1 Selecting Analog Devices
Selecting analog devices from an EMC perspective is not as straightforward as selecting digital devices. Although we also hope that emissions, conversion rates, voltage fluctuations, and output drive capabilities are as small as possible, for most active analog devices, immunity is a very important factor, so it is quite difficult to determine clear EMC ordering characteristics.
The same model and specification of operational amplifiers from different manufacturers can have significantly different EMC performance, so it is very important to ensure the consistency of performance parameters of subsequent products. Manufacturers of sensitive analog components provide EMC or signal-to-noise processing tips or PCB layout on circuit design, which shows that they care about the needs of users, which helps users weigh the pros and cons when purchasing.
1.2.2 Preventing demodulation problems
Most of the noise immunity problems of analog devices are caused by RF demodulation. Each pin of the op amp is very sensitive to RF interference, regardless of the feedback line used (see Figure 3). All semiconductors have a demodulation effect on RF, but the problem is more serious in analog circuits. Even low-speed op amps can demodulate signals at mobile phone frequencies and above. Figure 4 shows the test results of actual products.
To prevent demodulation, analog circuits must remain linear and stable when in an interference environment, especially the feedback loop, which must be linear and stable over a wide bandwidth. This often requires buffering the capacitive load and using a small series resistor (about 500) and an integral feedback capacitor of about 5PF in series.
When performing stability and linearity tests, inject a small but extremely steep rising edge (<1ns) square wave signal at the input (it can also be fed to the output and power supply through a capacitor). The fundamental frequency of the square wave must be within the expected frequency band of the circuit. The circuit output should be checked for impact and ringing using a 100MHz (at least) oscilloscope and probe. The same applies to audio or instrument circuits. For higher-speed analog circuits, select an oscilloscope with a wider frequency band and pay attention to the technique of using the probe.
Overshoot exceeding 50% of the signal height indicates that the circuit is unstable and should be effectively attenuated. Any long-term ringing (more than two cycles) or sudden oscillation of the signal indicates poor stability.
The above tests should be performed without filters at both the input and output ends. You can also use frequency sweep instead of square wave, and spectrum analyzer instead of oscilloscope (it is easier to see the resonant frequency).
Figure 3 Demodulation occurs in any semiconductor device, and all leads are sensitive
Figure 4: The operational amplifier can effectively demodulate the RF signal.
1.2.3 Other analog circuit technologies
After obtaining a stable and linear circuit, all its connections may need filtering. The digital circuit part of the same product will always induce noise to the internal connection, and the external connection will be subjected to the interference of the external electromagnetic environment. Filters will be introduced later.
Never try to use active circuits to filter and suppress RF bandwidth to meet EMC requirements, only use passive filters (preferably RC type). In op amp circuits, integral feedback is only effective in the frequency range where its open loop gain is much greater than the closed loop gain, but at higher frequencies, it cannot control the frequency response.
Avoid circuits with high input and output impedances. Comparators must have hysteresis characteristics (positive feedback) to prevent output malfunctions due to noise and interference, and to prevent oscillations near the switching point. Do not use comparators with output transitions that are much faster than necessary, and keep dv/dt low.
For high-frequency analog signals (such as RF signals), transmission line technology is necessary, depending on its length and the highest frequency of communication. Even for low-frequency signals, if transmission line technology is used for internal connections, its immunity will be improved.
The circuits in some analog integrated circuits are extremely sensitive to high field strengths. In this case, they can be shielded with a small metal shell (if heat dissipation allows) and the shielding box can be soldered to the PCB ground plane.
Similar to digital circuits, analog devices also require high-quality RF bypassing (decoupling) for the power supply, but low-frequency power supply bypassing is also required because the power supply noise rejection rate (PSRR) of analog devices is very weak for frequencies above 1kHz. RC or LC filtering is necessary for each analog power supply pin of each op amp, comparator or data converter. The corner frequency and transition band slope of these power supply filters should compensate for the corner frequency and slope of the device PSRR to obtain the desired PSRR within the frequency band of concern.
RF design is rarely covered in general EMC design guidelines. This is because RF designers are generally familiar with most continuous EMC phenomena. However, it should be noted that local oscillator and IF frequencies generally have large leakage, so shielding and filtering need to be emphasized.
Previous article:Implementation of a compact power manager with USB OTG and overvoltage protection
Next article:Power Manager Simplifies Li-Ion Battery Charging
Recommended ReadingLatest update time:2024-11-17 01:57
- Popular Resources
- Popular amplifiers
- MathWorks and NXP Collaborate to Launch Model-Based Design Toolbox for Battery Management Systems
- STMicroelectronics' advanced galvanically isolated gate driver STGAP3S provides flexible protection for IGBTs and SiC MOSFETs
- New diaphragm-free solid-state lithium battery technology is launched: the distance between the positive and negative electrodes is less than 0.000001 meters
- [“Source” Observe the Autumn Series] Application and testing of the next generation of semiconductor gallium oxide device photodetectors
- 采用自主设计封装,绝缘电阻显著提高!ROHM开发出更高电压xEV系统的SiC肖特基势垒二极管
- Will GaN replace SiC? PI's disruptive 1700V InnoMux2 is here to demonstrate
- From Isolation to the Third and a Half Generation: Understanding Naxinwei's Gate Driver IC in One Article
- The appeal of 48 V technology: importance, benefits and key factors in system-level applications
- Important breakthrough in recycling of used lithium-ion batteries
- Innolux's intelligent steer-by-wire solution makes cars smarter and safer
- 8051 MCU - Parity Check
- How to efficiently balance the sensitivity of tactile sensing interfaces
- What should I do if the servo motor shakes? What causes the servo motor to shake quickly?
- 【Brushless Motor】Analysis of three-phase BLDC motor and sharing of two popular development boards
- Midea Industrial Technology's subsidiaries Clou Electronics and Hekang New Energy jointly appeared at the Munich Battery Energy Storage Exhibition and Solar Energy Exhibition
- Guoxin Sichen | Application of ferroelectric memory PB85RS2MC in power battery management, with a capacity of 2M
- Analysis of common faults of frequency converter
- In a head-on competition with Qualcomm, what kind of cockpit products has Intel come up with?
- Dalian Rongke's all-vanadium liquid flow battery energy storage equipment industrialization project has entered the sprint stage before production
- Allegro MicroSystems Introduces Advanced Magnetic and Inductive Position Sensing Solutions at Electronica 2024
- Car key in the left hand, liveness detection radar in the right hand, UWB is imperative for cars!
- After a decade of rapid development, domestic CIS has entered the market
- Aegis Dagger Battery + Thor EM-i Super Hybrid, Geely New Energy has thrown out two "king bombs"
- A brief discussion on functional safety - fault, error, and failure
- In the smart car 2.0 cycle, these core industry chains are facing major opportunities!
- Rambus Launches Industry's First HBM 4 Controller IP: What Are the Technical Details Behind It?
- The United States and Japan are developing new batteries. CATL faces challenges? How should China's new energy battery industry respond?
- Murata launches high-precision 6-axis inertial sensor for automobiles
- Ford patents pre-charge alarm to help save costs and respond to emergencies
- [TI mmWave Radar Review] + Belated AWR1443 EVM Board Unboxing
- Watch Shuige's video for a reward | MIPI D-PHY transmitter physical layer consistency test
- Summary of Problems and Solutions in CCS6 Compilation
- LabView Notes 2: Two methods to implement traffic light experiments based on LabView
- [STM32WB55 review] +thread trial 1
- Guess the question about the list of materials for the undergraduate group of the electronic competition: Sound source localization system
- 【ST NUCLEO-G071RB Review】_01_First impression
- Why use discrete components for preamplification?
- Question: How can a single-chip microcomputer detect when a telephone is hung up?
- MSP430IO Driver