Common Mode Chokes for EMI Suppression

Introduction to Common Mode Noise

Common mode noise refers to the interference that appears in phase and with equal amplitude on both lines of a differential signal or a power line with respect to a common reference (typically ground). It’s a major source of EMI because it can easily radiate from cables or PCB traces, especially when long conductors act as antennas.

Common Mode Chokes

A common mode choke (CMC) is a type of passive filter made by winding two or more conductors (typically for a differential signal) on a magnetic core in such a way that:

Differential signals pass unaffected.

Common mode signals (noise present equally on both lines) induce magnetic flux in the same direction, which creates impedance and attenuates the noise.

Structure and Working Principle

A common mode choke is essentially a toroidal or cylindrical magnetic core (often ferrite or powdered iron) with two windings of wire wound around it—usually side by side or bifilar wound.

Key components:

Magnetic Core:
- Typically ferrite material with high permeability.
- Acts as the medium for magnetic flux.
Windings (Coils):
- Two insulated copper wires are wound around the core.
- The wires carry current for the differential signal (i.e., one wire for signal, one for return).
Pins / Terminals:
- The choke is mounted on a PCB, with four pins—two for each coil end.

Working Principle of a Common Mode Choke

The common mode choke blocks common mode currents (noise or interference that flows in the same direction on both lines) but allows differential mode currents (normal signal current that flows in opposite directions) to pass freely.

Here’s how it works:

For differential mode currents:
- Current flows in opposite directions in the two windings.
- Their magnetic fields cancel each other out in the core.
- Result: Minimal inductance → signal passes through with little impedance.
For common mode currents (e.g., EMI or RF noise):
- Current flows in the same direction through both windings.
- Their magnetic fields add up, reinforcing the magnetic flux in the core.
- Result: High inductance → blocks or attenuates noise effectively.

Think of a CMC as a filter that says:

"If the currents are opposite (real signal), I’ll let them through with no problem."
"If the currents are the same (noise), I’ll choke them with high impedance."

Applications in PCB and System Design

Power lines: To suppress switching noise from DC-DC converters.

USB, HDMI, Ethernet: To block radiated and conducted emissions from high-speed I/O.

CAN/LIN buses: To reduce susceptibility and emissions in automotive networks.

Audio circuits: To remove hum and external noise from cables.

Design Considerations for Common Mode Chokes in EMI Suppression

Effective use of common mode chokes (CMCs) requires thoughtful integration into PCB and system-level designs. The following key design considerations ensure optimal EMI mitigation without compromising signal integrity or power delivery.

Common mode chokes (CMCs) are vital components in suppressing electromagnetic interference (EMI), particularly common mode noise that tends to couple onto cables and radiate from electronic systems. To achieve optimal performance, careful attention must be paid to several design factors that influence both EMI suppression and signal integrity.

1 Frequency Characteristics and Impedance Profile

The effectiveness of a CMC is highly dependent on its impedance profile across frequency. An ideal CMC offers high impedance to common mode noise while remaining transparent to differential mode signals. Design selection should be guided by the spectral content of the unwanted emissions. In most digital systems, common mode noise appears predominantly in the 10 MHz to 1 GHz range. Therefore, chokes must be selected based on insertion loss data provided by the manufacturer, ensuring that they provide sufficient attenuation in this band.

Example: In USB 2.0 applications, significant emissions occur between 30 MHz and 300 MHz. A CMC designed to provide high impedance within this band will help in meeting EMI compliance requirements such as CISPR 32 or FCC Part 15.

2 Current and Voltage Handling Capability

Another critical consideration is the current-carrying capability of the CMC. The selected choke must support the continuous operating current of the line without causing magnetic saturation of the core. Saturation significantly reduces the common mode impedance and thereby degrades EMI suppression performance. Additionally, the choke’s voltage rating must exceed the peak differential voltage on the line to prevent insulation failure or dielectric breakdown between windings.

3 Differential Mode Transparency

While the primary purpose of a CMC is to suppress common mode noise, it is equally important to ensure that differential mode signals are not attenuated or distorted. This is particularly relevant for high-speed data lines such as USB 3.0, HDMI, or Ethernet, where signal integrity is paramount. In such applications, low differential impedance and symmetrical winding structures are essential to minimize skew and avoid signal degradation.

4 Optimal Placement Strategy

The physical placement of the CMC on the PCB significantly impacts its effectiveness. It should be placed as close as possible to either the source of EMI (e.g., switching regulators) or the I/O connectors (e.g., USB or RJ45 jacks). This minimizes the path length through which common mode currents can radiate. In systems with well-defined subsystem boundaries, placing the choke at the interface helps isolate noise between domains.

5 PCB Layout Guidelines

Proper PCB layout is essential to maintain the effectiveness of the CMC. The traces connected to the choke must be routed symmetrically to preserve differential balance. Any mismatch can lead to mode conversion, wherein differential signals generate additional common mode noise. The loop area between the signal paths and their return currents should be minimized to reduce parasitic inductance. Moreover, ground stitching vias should be included near the choke to provide a low-impedance return path for high-frequency currents.

6 Core Material and Winding Structure

The core material plays a crucial role in determining the frequency response of the choke. Ferrite materials are widely used due to their high permeability and frequency-selective damping properties. The winding structure, especially bifilar winding, enhances common mode suppression by ensuring tight magnetic coupling while allowing differential signals to pass with minimal interference.

7 Thermal and Mechanical Considerations

As passive components, CMCs also dissipate heat due to core losses and copper resistance. Therefore, they should be selected with a current derating margin to account for ambient temperature, self-heating, and aging effects. The choice between surface-mount and through-hole packages depends on current requirements, thermal dissipation, and mechanical robustness. Surface-mount components are preferred for compact designs and automated assembly processes.

8 Multi-Channel and System-Level Integration

In systems with multiple parallel data lines (e.g., differential pairs in HDMI or Ethernet), multi-channel CMCs can reduce board space and ensure uniform filtering across all lines. However, inter-channel crosstalk and imbalance must be considered during layout and simulation. Additionally, CMCs should be included as part of a comprehensive EMI strategy, integrated with other filtering elements such as ferrite beads and LC filters.

9 EMC Compliance and Validation

Ultimately, the choice and integration of CMCs must be validated through EMI pre-compliance and compliance testing. Techniques such as Line Impedance Stabilization Network (LISN) measurements, spectrum analysis, and near-field probing help assess the effectiveness of common mode suppression. Iterative tuning of choke parameters and placement is often required to meet stringent regulatory limits.

Comparison with Other Filters

Filter Type	Best For	Blocks	Notes
Common Moe choke	Common Mode EMI	Common mode Noise	Transparent to differential signals
Ferrite Bead	Broadband Suppression	High Frequency Noise	Acts as lossy inductors
LC Filter	Specific frequency range	Both Common and differential Noise	Requires careful tuning

Example

Consider a USB 3.0 line:

Without a CMC: Common mode currents induced by fast switching can radiate via the cable, failing EMI tests.
With a CMC: Noise is attenuated before reaching the cable, ensuring compliance with standards like CISPR 32.

Automotive Pulses

Automotive pulses refer to specific transient voltage waveforms that occur in a vehicle's electrical system. These pulses are defined by standards (like ISO 7637, SAE J1113, or LV 124) and used primarily for EMC (Electromagnetic Compatibility) testing of automotive electronic components and systems.

These tests simulate real-world disturbances (like load dumps, switching transients, inductive kicks) to ensure components can survive and operate correctly in harsh automotive environments.

Modern vehicles contain many ECUs, sensors, and actuators. They're exposed to:

Alternator spikes
Battery disconnection
Relay switching
Ignition transients
Electrostatic discharge (ESD)

To ensure reliable operation, these pulses are simulated and tested in labs.

Pulses as per ISO 7637-2

ISO 7637-2 (for 12V and 24V systems)

The most widely used for electrical transients on supply lines.

Pulse	Description	Typical Cause
Pulse 1	Negative spike	Battery disconnect while inductive loads are ON
Pulse 2a/2b	Positive/negative spikes	Switching of inductive loads
Pulse 3a/3b	Fast transients	Arising from relay contact bounce
Pulse 4	Voltage drop	Engine cranking
Pulse 5	Load dump	Battery disconnect while alternator is charging

Pulse Characteristics (ISO 7637-2)

Pulse 1

Negative Spike
Cause: Battery disconnection from Inductive Loads (motor, Solenoid)
Affect: Power supply circuit, Microcontroller

Parameters	Nominal 12 V system	Nominal 24 V system
Us	−75 V to −150 V	−300 V to −600 V
Ri	10 Ω	50 Ω
td	2 ms	1 ms
tr	1 μs	3 μs
t1	≥0.5 s
t2	200 ms
t3	<100 μs

Pulse 2a

Positive Spike
Cause: Sudden interruption of current due to sudden disconnection of large loads
Affect: Voltage regulators and semiconductors devices

Parameters	Nominal 12V & 24V system
Us	+37 V to +112 V
Ri	2 Ω
td	0.05 ms
tr	1 μs
t1	0.2 s to 5 s

Pulse 2b

Negative Spike
Cause: Alternator field decay, Alternator’s field winding de-energises after engine shutdown
Affect: ECU Reset

Parameters	Nominal 12 V system	Nominal 24 V system
Us	10 V	20 V
Ri	0 Ω to 0.05 Ω
td	0.2 s to 2 s
tr	1 ms
t12	1 ms
t6	1 ms

Pulse 3a

Fast Transients – Negative
Cause: Relay chatter (very fast edge), switching processes. Characteristics of these transients are influenced by distributed capacitance and inductance of the wiring harness.
Affect: Signal Interference and component malfunction

Parameters	Nominal 12 V system	Nominal 24 V system
Us	−112 V to −220 V	−150 V to −300 V
Ri	50 Ω
td	150 ns ± 45 ns
tr	5 ns ± 1.5 ns
t1	100 μs
t4	10 ms
t5	90 ms

Pulse 3b

Fast Transients – Positive
Cause: Relay chatter (very fast edge), switching processes. Characteristics of these transients are influenced by distributed capacitance and inductance of the wiring harness.
Affect: Sensors and controllers

Parameters	Nominal 12 V system	Nominal 24 V system
Us	+75 V to +150 V	+150 V to +300 V
Ri	50 Ω
td	150 ns ± 45 ns
tr	5 ns ± 1.5 ns
t1	100 μs
t4	10 ms
t5	90 ms

Pulse 4

Cranking Pulse - Slow Negative
Cause: Supply voltage reduction caused by energising starter motor of internal combustion engine.
Affect: Sensors and controllers

Parameters	Nominal 12 V system	Nominal 24 V system
Us	− 6V to − 7V	− 12V to − 16V
Ua	− 2.5V to − 6V with \|Ua\| ≤ \|Us\|	− 5V to − 12V with \|Ua\| ≤ \|Us\|
Ri	0 Ω to 0.02 Ω
t7	15 ms to 40 ms	50 ms to 100 ms
t8	≤ 50ms
t9	0.5 s to 20 s
t10	5ms	10ms
t11	5ms to 100ms	10ms to 100ms

Pulse 5a – Load Dump

Unsuppressed Alternator Surge- Without Protection
In the event of a discharged battery being disconnected while the alternator is generating charging current and with other loads remaining on the alternator circuit at this moment.
Affect: ECU, Sensor, Power circuit

Parameters	Nominal 12 V system	Nominal 24 V system
Us	65V to 87V	123V to 174V
Ri	0.5 Ω to 4 Ω	1 Ω to 8 Ω
td	40ms to 400ms	100ms to 350ms
tr	10ms

Pulse 5b – Load Dump

Suppressed Alternator Surge- with protection (like Zener diode, TVS diode)
In the event of a discharged battery being disconnected while the alternator is generating charging current and with other loads remaining on the alternator circuit at this moment.
Affect: ECU, Sensor, Power circuit

Parameters	Nominal 12 V system	Nominal 24 V system
Us	65V to 87V	123V to 174V
Us*	As specified by manufacturer
td	40ms to 400ms	100ms to 350ms

Testing Setup

Usually tested in the lab using an EMC test bench with:

Pulse generators
Coupling/decoupling networks
Oscilloscopes
Electronic loads or real DUTs

The DUT (Device Under Test) must withstand or operate normally depending on the pulse.

Other Standards

SAE J1113 – Similar to ISO 7637, North American usage
LV 124 / LV 148 – German standards for 12V/48V systems (used by BMW, VW, Daimler, etc.)
ISO 16750-2 – Broader set including electrical load, jump start, reverse polarity, etc.

Conducted Emission

Conducted emissions refer to electromagnetic disturbances that are transmitted through conducting wires or cables connected to an electrical or electronic device. These emissions can propagate along power lines, signal lines (such as data cables), and any other conductive paths that are connected to the device.

Here's a detailed breakdown of conducted emissions:

Sources:

Conducted emissions originate from various sources within electronic equipment:

Power Supplies: Switching power supplies and transformers can generate conducted emissions due to switching transients and harmonics.
Electronic Circuits: Digital circuits, especially those with fast switching speeds, can produce conducted emissions through power and signal lines.
Motors and Drives: Electric motors and motor drives can introduce conducted emissions into power lines due to switching frequencies and power modulation.
Cables and Connectors: Poorly shielded or improperly grounded cables can act as antennas, radiating conducted emissions.

Frequency Range:

Conducted emissions typically cover a wide frequency range, often from a few kilohertz (kHz) up to several hundred megahertz (MHz), depending on the nature of the emitting device and the frequency of the signals or power being processed.

Low Frequency Range (LF): Typically below 150 kHz, includes power line harmonics and switching noise from power supplies.
Radio Frequency Range (RF): Spans from 150 kHz to several hundred MHz, encompassing emissions from digital circuits, clock signals, and other high-frequency components.

Measurement and Testing:

Conducted emissions are typically measured using specialized equipment such as spectrum analyzers and conducted emission measurement receivers. Testing is conducted with the device under test (DUT) connected to a standardized test setup that simulates real-world operating conditions. Measurements are taken across specified frequency ranges to ensure compliance with applicable EMC standards.

Test Setup

Test Instrument

Below are the some Major Test equipment, required for Conducted Emission Testing

EMI Receiver
Line Impedance Stabilization Network (LISN)
Pulse Limiter
RF Cables

Test Parameters

Frequency Range: 9kHz to 30MHz (depends on the Product standard)
Detector: Quasi-Peak, Average

Test Procedure

Setup EUT (Equipment Under Test) at the designated place, 0.8m high on the insulated table.
Table must be kept on the Ground Reference plane and at a distance of 0.4m from the Vertical coupling plane.
EUT should be energised through LISN with the rated Voltage and current and to be set in the highest configuration to achieve max level of emission from it.
RF cable should be connected between LISN and EMI Receiver.
Set the desired frequency range, limit lines and other parameters in EMI Receiver to receive the EMI signal from the EUT.
Select the peaks closure to limit line and measure the Q-Peak and Average values for final measurement.
Repeat step 5 & 6 for every individual line and neutral.
Capture the data like emission level in dbµV, Limit at that particular frequency, calculate the margin to mention in the final report.
There are different limit lines as per different product standards.

Most of the standards refers to the below limits lines based on the end use of the product.

Conducted Emission limits_Class A

Frequency Range (MHz)	Quasi Peak (dbμV)	Average (dbμV)
0.15 - 0.5	79	66
0.5 - 30	73	60

Conducted Emission limits_Class B

Frequency Range (MHz)	Quasi Peak (dbμV)	Average (dbμV)
0.15 - 0.5	66-56	56-46
0.5 - 5.0	56	46
5.0 - 30	60	50

Electromagnetic Emissions from the DUT should be less than the limits mentioned above.

Conducted Emission Spectrum with both the Class A and Class B limits.

In summary, conducted emission testing ensures that devices meet regulatory EMC limits, contributing to reliable and interference-free operation across electrical and electronic systems.

Electrical Noise

Electrical noise is any unwanted signal superimposed on a desired electrical signal that can distort, interfere, or reduce signal fidelity. In high-speed circuits and EMC analysis, this noise is typically classified as:

Differential-Mode (DM) Noise
Common-Mode (CM) Noise

These two have different origins, transmission paths, and countermeasures.

1. Differential-Mode Noise – “Normal-mode noise”

Differential-mode noise is the voltage difference between two conductors of a signal or power line. It represents the intentional signal path where the noise rides in opposition across the two conductors.

Current Flow:

The noise flows in opposite directions in a loop between the two wires. If one conductor has a +5V spike and the other has -5V, the differential noise is 10V.

Example Use Cases:

Power rails: +V and GND in DC systems
Communication: USB, HDMI, Ethernet pairs
Analog signals: Sensor lines in instrumentation

Typical Sources:

Switch-mode power supplies (SMPS) – due to rapid switching
High-speed data lines – signal reflections, ringing
Magnetic coupling – between traces or cables
Load changes – from motors, solenoids, or relays

Mitigation Strategies:

Differential-mode filters: Series inductors with capacitors across lines
Matched impedance design: Prevents signal reflection and ringing
Twisted pairs: Cables twisted to cancel out opposing fields
Short trace lengths: Reduces antenna effect

2. Common-Mode Noise – “Ground-referenced noise”

Common-mode noise appears equally and in phase on both conductors relative to a common reference point, usually system ground or chassis.

Current Flow:

Both wires carry the noise in the same direction, and the return path flows through the ground plane, earth, or shielding.

Real-World Example:

Long USB cables pick up RF noise equally on both data lines.
AC power lines exposed to EMI from nearby radio towers or industrial machines.

Typical Sources:

Electrostatic coupling: From nearby high-voltage lines
Radiated emissions: Antenna-like behavior of cables or traces
Ground potential differences: Between system components (e.g., USB ground loop)
Parasitic capacitance: Between PCB traces and the chassis

Mitigation Strategies:

Common-mode chokes: High impedance to CM signals, low to DM signals
Shielded cables and connectors: With 360° termination to chassis
Isolated grounds: Breaks in ground loops
Ferrite beads/clamps: On external cables

3. Visualizing the Current Paths

Noise Type: Differential Mode

Current Direction: Opposite directions on each line, Reference Point: Across the pair, Return Path: One conductor to another

Noise Type: Common Mode

Current Direction: Same direction on both lines, Reference Point: Ground or chassis, Return Path: Through system or earth ground

4. Importance in EMC Testing

Emissions:

Differential-mode emissions are mostly conducted (via power/signal lines).
Common-mode emissions often become radiated, as CM currents form large loop areas (acting like antennas).

Compliance Implications:

Regulatory bodies (e.g., FCC, CISPR, IEC) test both emission types.
CISPR 22/32 and FCC Part 15 focus heavily on common-mode conducted emissions in lower frequency bands (150 kHz–30 MHz).

5. Filtering Techniques Comparison

Filter Type: CM Filter

Effective Against: Common-mode noise, Components Used: Common-mode choke, Y-capacitors

Filter Type: DM Filter

Effective Against: Differential-mode noise, Components Used: Series inductors, X-capacitors

Notes:

Y-capacitors connect from line to ground (handle CM).
X-capacitors go across the line pair (handle DM).

6. Example: SMPS Input Filtering

In a switch-mode power supply, both noise types are generated:

Differential noise arises from the switching node oscillations.
Common-mode noise results from parasitic capacitance between high-frequency switching nodes and the chassis.

The input filter typically includes:

X-capacitor across line and neutral (DM)
Y-capacitors from line/neutral to ground (CM)
Common-mode choke for both conductors

7. Application-Specific Impacts

High-Speed Digital Systems:

Differential signaling (LVDS, HDMI, USB) relies on clean DM paths. Noise degrades data integrity (eye diagrams, jitter).
CM noise can cause cross-talk between pairs or fail EMC radiated tests.

Automotive:

CM noise is a major concern due to long wire harnesses acting as antennas.
Standards like ISO 11452 and CISPR 25 require thorough CM filtering.

Medical Devices:

Safety and immunity to external EMI are critical—isolation transformers, CM chokes, and filtering are used to ensure patient safety and compliance (e.g., IEC 60601-1-2).

Conclusion

Aspect	Differential-Mode Noise	Common-Mode Noise
Flow Direction	Opposite on conductors	Same on both conductors
Reference	Between lines	Against ground or chassis
Typical Source	Internal circuit switching	External EMI, parasitic coupling
Testing Concern	Conducted emissions	Radiated + conducted emissions
Mitigation	X-caps, twisted pairs, impedance match	CM chokes, Y-caps, shielding

Both types of noise must be addressed for:

Regulatory compliance
Signal integrity
Functional reliability

Designers should always measure both types during EMC testing and implement layered filtering and shielding strategies.