ALLPCB
1. Why is EMC design necessary for products?
To ensure product functionality, reduce debugging time, meet EMC standards, and prevent electromagnetic interference with other devices in the system.
2. What are the key aspects of EMC design for a product?
EMC design involves several areas: circuit design (including component selection), software design, printed circuit board (PCB) layout, shielding structures, signal line/power line filtering, and circuit grounding schemes.
3. Why are decibels (dB) commonly used in EMC? What is 10mV in dBmV?
Because the amplitude and frequency ranges involved are very wide, using a logarithmic scale on graphs is more practical. The dB is the unit for this logarithmic representation. 10mV is equal to 20dBmV.
4. Why can't a spectrum analyzer observe transient disturbances like electrostatic discharge (ESD)?
A spectrum analyzer is a narrowband swept-frequency receiver, meaning it only receives energy within a specific frequency range at any given moment. Transient disturbances like ESD are pulse-type interferences with a very broad spectrum but extremely short duration. Therefore, a spectrum analyzer can only capture a small fraction of the total energy, failing to reflect the actual interference profile.
5. When diagnosing EMI issues on-site, near-field probes and a spectrum analyzer are often needed. How can a simple near-field probe be made from a coaxial cable?
Strip back the outer layer (shield) of the coaxial cable to expose the center conductor. Form the exposed center conductor into a small loop with a diameter of 1-2 cm (1-3 turns) and solder the end of the loop back onto the cable's shield.
6. A device originally had an electromagnetic radiation emission of 300 mV/m. After adding a shielded enclosure, the emission dropped to 3 mV/m. What is the shielding effectiveness of this enclosure in dB?
The shielding effectiveness of this enclosure is 40 dB.
7. When designing a shielded enclosure, what factors determine the choice of shielding material?
From an electromagnetic shielding perspective, the main consideration is the type of wave being shielded. For electric field waves, plane waves, or high-frequency magnetic field waves, most common metals are sufficient. For low-frequency magnetic field waves, materials with high magnetic permeability are required.
8. Besides the shielding material, what other factors affect an enclosure's shielding effectiveness?
Two main factors: discontinuities in the enclosure's conductivity, such as apertures and seams; and conductors that penetrate the enclosure, such as signal cables and power lines.
9. What are the key considerations when shielding a magnetic field radiation source?
Since magnetic field waves have very low wave impedance, reflection loss is minimal, and shielding relies primarily on absorption loss. Therefore, a material with high magnetic permeability should be chosen. Additionally, during structural design, the shield should be placed as far as possible from the radiation source (to increase reflection loss), and apertures or seams near the source should be avoided.
10. A principle in shielding design is to keep internal cables away from seams and apertures. Why?
A magnetic field always exists around a cable, and this field can easily leak through apertures, regardless of frequency. When a cable is close to a seam or aperture, this magnetic field leakage occurs, reducing the overall shielding effectiveness.
11. Measuring the human body's biomagnetic information is a new medical diagnostic method that requires a magnetically shielded room. This room must shield fields from DC (static) up to 1 GHz. Propose a design for such a room.
First, consider the shielding material. To shield very low-frequency magnetic fields, a high-permeability material like Permalloy is necessary. Since the permeability of Permalloy decreases after processing, it must be heat-treated. Therefore, the shielded room should be a modular, panel-based assembly. Panels are fabricated to design specifications, then heat-treated, transported to the site, and carefully assembled. The joints between panels must overlap to form a continuous magnetic path. This construction provides good shielding for low-frequency magnetic fields but is susceptible to high-frequency leakage through the seams. To compensate, a second shield made of welded aluminum plates is built around the Permalloy structure to shield high-frequency electromagnetic fields.
12. What is a waveguide-below-cutoff panel (honeycomb vent), where is it used, and what are the precautions for its use?
It is a panel composed of an array of many waveguide-below-cutoff tubes. It is used in applications requiring high shielding effectiveness and significant ventilation. When installing, it is crucial to use an EMI gasket between the honeycomb panel and the chassis or to weld it in place to ensure a proper conductive seal.
13. A device with a plastic enclosure fails EMC testing due to excessive radiation. The developers sprayed the inside of the enclosure with conductive paint, but there was no significant improvement. What are the possible reasons?
The original plastic enclosure may have too many or too large apertures, causing significant leakage. The seams might not be properly sealed (either due to poor contact or because conductive paint was not applied at the joints). Additionally, the cables (signal and power) connected to the device likely lack proper filtering, allowing them to act as antennas and radiate noise.
14. What types of transparent shielding windows are available, and what should be considered when using them?
There are two main types: windows made of glass with an embedded metal mesh, and windows with a thin metallic film coated onto the glass. When using either type, it is essential that the metal mesh or conductive coating makes good electrical contact with the main body of the shielded enclosure.
15. When using a metal mesh shielding glass on a CRT display, annoying moiré patterns can appear. How can this be reduced?
Rotate the orientation of the wire mesh so that its weft lines form a 15-20 degree angle with the CRT's scan lines.
16. What are the two critical properties of an EMI gasket? List several types and their typical applications.
The two essential properties of an EMI gasket are elasticity and conductivity. Common types include: fingerstock, wire mesh gaskets, conductive elastomers, fabric-over-foam gaskets, and spiral gaskets. Avoid fingerstock in applications with tangential sliding contact. Use conductive elastomers when an environmental seal is also required. Fabric-over-foam gaskets are suitable for many other applications. Wire mesh gaskets can be used for lower-frequency shielding. Spiral gaskets can be used when over-compression can be reliably prevented.
17. What precautions should be taken when using EMI gaskets?
The panel should be thick enough to resist bending under the gasket's spring force, which could create larger gaps. For thinner panels, use a closer spacing for fastening screws. Implement compression stops to prevent over-compression. Select compatible metals to minimize galvanic corrosion.
18. A metal rod must pass through a shielded enclosure. How can this be done without compromising the shielding effectiveness?
Ensure the metal rod is reliably bonded to the enclosure's chassis around its entire circumference using a material like beryllium copper fingerstock.
19. What is the primary function of a power line filter? What are the key parameters to consider during selection, and what are the main precautions for installation?
A power line filter's function is to suppress conducted emission currents from traveling along the power line. Key selection parameters include insertion loss (common-mode and differential-mode), rated current, rated voltage, and effective frequency range. During installation, it is crucial to ensure a good RF ground, isolate the input and output wiring, and prevent the filtered output wiring from being re-contaminated by noise.
20. Why is the high-frequency performance of a power line filter so important?
Poor high-frequency performance can lead to excessive radiated emissions or increased susceptibility to pulse-type disturbances.
21. A principle of EMC structural design is to keep filtered power lines away from signal cables. Why?
If power lines are close to signal cables, high-frequency signals on the signal cables can couple onto the power lines (especially the filtered section), causing conducted emissions to exceed limits.
22. When selecting a power line filter, why is it not always best to choose the smallest one?
A filter's size is largely determined by its internal inductors. A smaller filter must use smaller inductors, which may have lower inductance and thus poorer low-frequency filtering performance. Additionally, the close proximity of components inside a small filter can degrade its high-frequency performance due to parasitic coupling.
23. What is filter insertion loss, and what measurement method provides the most conservative (worst-case) result?
Insertion loss is the loss of current or voltage resulting from inserting a filter into a circuit. An EMI filter should have the highest possible insertion loss at the interference frequencies. To get the most conservative result, measure insertion loss using a source and load impedance ratio of 0.1:100 (or 100:0.1). This represents the worst-case conditions and thus provides the most reliable data.
24. Generally, AC line filters can be used in DC applications, but DC line filters must not be used in AC applications. Why?
DC filters use DC-rated bypass capacitors. If used on an AC line, they can overheat and fail. If the DC capacitor's voltage rating is too low, it can also be damaged by breakdown. Even if these failures don't occur, the common-mode bypass capacitors in DC filters often have large capacitance values, which would cause excessive leakage current on an AC line, violating safety standards.
25. What is the role of a signal line filter? What are the different mounting types, and how do you choose?
A signal line filter reduces unwanted high-frequency components (mainly common-mode) on a signal line, thereby reducing cable radiation or preventing the cable from acting as an antenna to receive and conduct external EMI into the enclosure. Mounting types include PCB-mount and panel-mount. PCB-mount filters are suitable for lower-frequency filtering, while panel-mount filters are necessary for higher-frequency applications.
26. A signal line carries a signal with a maximum frequency of 30 MHz. Measurements show a 120 MHz common-mode interference current on this line. Predictions indicate that suppressing this current by 30 dB will meet EMC standards. How many orders (poles) are needed for the low-pass filter?
The low-pass filter's cutoff frequency should be 30 MHz, and its insertion loss at 120 MHz must be greater than 30 dB. The insertion loss of an N-order filter increases at a rate of 6N dB per octave. From 30 MHz to 120 MHz is two octaves (30 -> 60 -> 120). Therefore, an N-order filter with a 30 MHz cutoff will have an insertion loss of 12N dB at 120 MHz. To ensure 12N > 30, N must be at least 3. A third-order low-pass filter is required.
27. Why are three-terminal capacitors better for EMI filtering?
EMI often occurs at high frequencies, where filter performance is critical. A three-terminal capacitor cleverly uses the lead inductance on one of its electrodes to form a T-type low-pass filter. This cancels the negative effects of lead inductance found in conventional two-terminal capacitors, improving high-frequency filtering performance. This makes them more suitable for interference suppression.
28. Why are feedthrough capacitors considered ideal devices for interference filtering?
A feedthrough capacitor is a type of three-terminal capacitor. However, because it is mounted directly on a metal panel, its ground inductance is much lower, virtually eliminating the effect of lead inductance. Furthermore, its input and output terminals are isolated by the metal panel, which prevents high-frequency coupling. These two features give the feedthrough capacitor a filtering performance that is close to an ideal capacitor.
29. How do ferrite cores for EMI suppression differ from traditional inductor cores? What happens if they are used interchangeably?
Cores for traditional inductors are made from materials with very low loss to create high-Q inductors. In contrast, cores for EMI suppression are made from high-loss materials. An "inductor" made with such a core behaves more like a resistor at high frequencies. Using them interchangeably leads to poor results. If a suppression core is used for a standard inductor, the resulting Q-factor will be very low, making resonant circuits ineffective or causing excessive signal loss. If a low-loss inductor core is used for EMI suppression, it can resonate with parasitic capacitance in the circuit, potentially amplifying interference at a specific frequency.
30. A 470 pF bypass capacitor has two leads, each 2 mm long. At what frequency will this capacitor provide the best filtering effect? (Estimate lead inductance at 1 nH/mm).
A capacitor provides the best filtering at its series resonant frequency, where its impedance is lowest. The resonant frequency is calculated as: f = 1 / [ 2π * sqrt(LC) ] = 1 / [ 2 * 3.14 * sqrt( (4 * 10^-9 H) * (470 * 10^-12 F) ) ] ≈ 116 MHz. Therefore, this capacitor is most effective at 116 MHz.
31. Why should the rated voltage of a signal line filter on an external cable be greater than 200V, even though the signal voltage is typically only a few volts?
External cables are exposed to high-amplitude transient disturbances like surges and ESD. The filter's capacitors must have a voltage rating high enough to withstand these high-voltage impacts without being damaged.
32. What is a common-mode choke, and how is it wound?
A common-mode choke is an inductor that presents a high impedance only to common-mode currents. It is wound such that the magnetic flux generated by the differential-mode currents in the two conductors cancels out. For high-voltage applications, the forward and return conductors should be wound separately to ensure adequate insulation. For low-voltage applications, they can be wound together as a bifilar pair.
33. When a device fails radiated emissions testing, a common fix is to place a ferrite core on the cable. If this is done and there is no improvement, what does it suggest, and what should be done next?
There are two possibilities. First, the original common-mode circuit impedance might already be high, so the additional impedance from the ferrite core is negligible in comparison. Second, there may be other significant radiation sources in the system, so reducing radiation from this one cable has a minimal effect on the total (when measured in dB). If the first case is true, adding bypass capacitors at the cable port can help lower the common-mode circuit impedance. If the second case is the cause, other radiation sources must be identified and addressed.
34. When many conductors pass through a panel, filtered connectors or filter array plates are often used. What precautions are needed during their installation?
An EMI gasket must be installed between the filtered connector or filter array plate and the chassis panel, or the seam must be sealed with conductive tape. This prevents electromagnetic leakage through the gap.
35. In EMI analysis, how is "ground" typically defined?
Ground is defined as the signal return path.
36. What is the fundamental cause of ground interference problems?
The fundamental cause is ground impedance. Because ground conductors have impedance, current flowing through them creates a voltage drop, resulting in different ground potentials at different points. Circuit design assumes that ground is an equipotential reference plane. The discrepancy between this assumption and reality leads to various ground-related problems.
37. Why do some imported devices have ground connections made through capacitors or inductors?
This is done to create different grounding structures for signals of different frequencies. For example, a capacitor can provide a low-impedance ground path for high-frequency signals while isolating DC or low-frequency grounds.
38. List several methods for reducing the RF impedance of a ground connection.
Use conductors with a large surface area to reduce high-frequency resistance (skin effect); keep conductors as short as possible to minimize both resistance and inductance; plate the conductor surface with silver to reduce surface resistance; use multiple parallel conductors to reduce total inductance.
39. What is "bonding," and what are some common methods?
Bonding is the creation of a low-impedance (at RF) connection between metal components. Common methods include welding, riveting, screw connections, and using EMI gaskets.
40. How can galvanic corrosion at a bonding point be prevented?
Select metals with similar electrochemical potentials, or apply an environmental seal to the contact area to keep out electrolytes (like moisture).
41. When designing a circuit or PCB for EMC, which signals are considered "critical" and require special attention?
From an emissions perspective, critical signals are periodic signals like local oscillators, clocks, and low-order address lines. From a susceptibility perspective, critical signals are those highly sensitive to external EMI, such as low-level analog signals.
42. Why do digital circuits often have significant noise voltage on their ground and power lines? How can this be reduced?
Digital circuits draw large, instantaneous currents when switching. As this transient current flows through the inductance of the power and ground traces, it generates a significant voltage spike (ground bounce or power rail droop). To reduce this noise, you can either decrease the inductance of the power and ground paths (by using a ground grid, ground plane, or power plane) or use appropriate decoupling capacitors to supply this transient current locally.
43. In practice, multi-stranded wires are sometimes twisted together to serve as a high-frequency conductor. Why is this said to reduce the RF impedance of the wire?
This technique increases the total surface area of the conductor, which reduces its resistance at high frequencies due to the skin effect.
44. Why do PCBs completed with auto-routing software often have higher radiated emissions?
Auto-routing software generally cannot guarantee that periodic signals (like clocks) will have the smallest possible loop area. Large current loops act as efficient antennas, leading to stronger radiation.
45. What are the primary measures for reducing radiated emissions from a PCB?
Minimize the loop area of signals that are prone to radiation (i.e., periodic signals). If the PCB has external cables, keep high-radiation circuits away from the I/O area. Establish a "clean" ground area at the I/O section to minimize common-mode voltage on the cables.
46. How can component selection help reduce electromagnetic radiation?
Choose integrated circuits with lower power consumption, slower signal rise/fall times (when possible), and a higher level of integration (to keep high-speed signals within the chip).
47. When using a multilayer PCB, can you use two separate ground planes—one for digital ground and one for analog ground—to prevent interference? Why or why not?
No, this is not a good practice. The large parasitic capacitance between two parallel ground planes will cause significant crosstalk, coupling noise between the digital and analog sections.
48. In PCB layout, why should high-frequency circuits be placed far away from I/O cable interfaces?
To prevent high-frequency signals from coupling onto the cables, which would create common-mode voltage (and current) and lead to strong common-mode radiation from the cables.
49. What is the key consideration when placing power supply decoupling capacitors on a digital circuit board?
The loop area formed by the decoupling capacitor, the IC's power pin, and the IC's ground pin must be as small as possible.
50. Cables connecting two shielded enclosures are a major source of radiation. To reduce this, shielded cables are used. What condition must be met for a shielded cable to be effective?
The cable's shield must be terminated with a 360-degree bond to the chassis of both shielded enclosures. This follows the principles of the "dumbbell model" for shielded interconnects.
51. A ferrite core is an effective device for suppressing common-mode radiation from cables. What are the key considerations when using one?
First, select a ferrite material specifically designed for EMI suppression. Second, the inner diameter of the core should be as small as possible to fit tightly around the cable. The outer diameter and length should be as large as space allows. Winding the cable through the core multiple times improves low-frequency effectiveness but can degrade high-frequency performance. The ferrite core should be placed as close as possible to the ends of the cable.
52. What should be considered when using twisted-pair wiring to improve immunity to magnetic fields?
The circuits connected at both ends of the twisted pair should not be grounded at both ends, as this would create a ground loop and a second return path for the signal. Ideally, they should be used with a balanced circuit.
53. If an inductive load is controlled by a mechanical switch, arcing and EMI will occur when the switch opens or closes. Is the interference more severe when the switch closes or when it opens?
The interference is more severe when the switch opens.
54. Why can't a transient voltage suppressor (TVS) replace a filter to prevent circuit malfunctions?
A TVS device only clamps the peak voltage of a high-amplitude pulse, leaving a residual pulse that, while lower in amplitude, still contains a wide range of high-frequency components that can disrupt circuit operation. Therefore, it cannot replace a filter for preventing functional issues.
55. What is the main consideration when installing a transient suppression device?
Ensure the current path through the suppressor has the lowest possible impedance. This means the connecting leads must be kept as short as possible. The same placement principles that apply to bypass capacitors also apply to transient suppressors.
56. Describe the mechanisms by which electrostatic discharge (ESD) affects a circuit.
There are two main mechanisms. First, the ESD current can flow directly into the circuit, disrupting its operation or even causing physical damage. Second, the rapid discharge creates a strong, transient electromagnetic field in the vicinity of the discharge path, which can inductively or capacitively couple into the circuit and cause interference.
57. Why do ESD tests often cause problems when an enclosure is not continuously conductive?
Discontinuities in an enclosure's conductive path force the discharge current to find an alternate route. This alternate path may be the circuit itself, allowing ESD current to flow into sensitive components. Additionally, as the current flows across a discontinuity (like a slot or gap), it can generate a strong localized electromagnetic field, disrupting nearby circuit operation.
58. Why does a device with good immunity to RF interference generally also have good immunity to ESD?
Because an ESD event also generates a high-frequency electromagnetic field, similar in nature to an RF interference field.
