Electronics Design AUWhat Is the Difference Between Conducted and Radiated Emissions?
Last updated 26 June 2026 · 8 min read
Direct Answer
Conducted emissions are high-frequency noise that a product injects back into its supply wiring or signal cables — measured at 150 kHz to 30 MHz using a Line Impedance Stabilisation Network (LISN). Radiated emissions are electromagnetic fields that propagate through the air from the product and its cables — measured from 30 MHz upward using a calibrated antenna in a controlled test environment. The two are related: conducted noise on cables often becomes radiated noise when those cables act as antennas at higher frequencies.
Detailed Explanation
When a product generates electromagnetic interference, that interference travels by one of two paths: along wires (conducted) or through the air (radiated). Regulatory standards separate these into two distinct test categories because they have different mechanisms, different frequency ranges, different measurement methods, and partially different mitigation strategies.
Understanding the distinction is essential for reading EMC test reports, interpreting pre-compliance scan results, and diagnosing which part of your design is responsible for a failing measurement. For an introduction to the EMC regulatory framework and why it applies to products sold in Australia, see the EMC topic hub. For the end-to-end process of getting RCM marking — from standards selection and NATA-accredited lab testing through to Declaration of Conformity — see RCM certification in Australia. For guidance on which specific Australian EMC standard applies to your product category — CISPR 32 for most commercial electronics, IEC 61000-6-4 for industrial equipment — see which Australian EMC standard applies to your product.
Conducted Emissions
Definition: Conducted emissions are high-frequency noise currents or voltages that a product injects back into its supply wiring (mains, USB VBUS, or DC supply leads) or signal cables. They propagate as electrical signals along the conductor.
Frequency range: 150 kHz to 30 MHz for mains-connected and DC-powered equipment in CISPR 32 and related standards. Some standards extend the lower limit to 9 kHz for power line communications environments.
How they're measured: A LISN (Line Impedance Stabilisation Network) is connected between the supply and the equipment under test. The LISN provides a standard 50Ω measurement impedance at the RF test port and isolates the test from the mains supply impedance, making measurements reproducible. A calibrated spectrum analyser measures the voltage at the LISN's RF port — this is the conducted emissions noise spectrum.
What causes them: switching power supplies are the dominant source of conducted emissions in most electronic products. The switching current drawn from the supply during each MOSFET turn-on event has harmonic content extending from the switching frequency (typically 100 kHz to 5 MHz) through many harmonics. Without filtering, this harmonic-rich current flows back into the supply cables.
The main paths for conducted noise:
- Differential-mode (DM) noise: current that flows in one supply line and returns via the other — suppressed by X-rated capacitors (line-to-line) and differential-mode inductors.
- Common-mode (CM) noise: current that flows the same direction in both supply lines and returns via the safety earth or chassis — suppressed by Y-rated capacitors (line-to-earth) and common-mode chokes.
For a product powered by a switching regulator, conducted emissions are reduced by:
- Using a correctly designed EMC input filter at the power supply input.
- Keeping the switching loop compact on the PCB — see how should you lay out a buck converter PCB? for the specific layout rules.
- Providing good decoupling at the IC's supply pins to prevent noise from propagating back to the input.
Radiated Emissions
Definition: Radiated emissions are electromagnetic fields that propagate through the air from the product, its PCB, and its attached cables. They do not require physical contact with other equipment to cause interference — they propagate outward and can affect nearby receivers.
Frequency range: 30 MHz and above (typically tested to 1 GHz for most product categories; up to 6 GHz or 40 GHz for products with RF components or faster digital logic).
How they're measured: in a semi-anechoic chamber (a shielded room with absorbing material on walls and ceiling, reflective floor) or an Open Area Test Site (OATS). A calibrated broadband antenna at a defined distance (3 m or 10 m depending on the standard) measures the electric field strength at each frequency. The product is rotated and the antenna is scanned at multiple heights to find the worst-case orientation.
What causes them: any conductor carrying a high-frequency current can radiate. The primary sources are:
- PCB traces and switching loops: a small loop antenna is formed by a switching current path and its return path. The radiation is proportional to loop area × current × frequency².
- Cables and interconnects: cables longer than approximately λ/20 at the noise frequency become efficient antennas. A 1 m USB cable at 150 MHz has λ = 2 m, so the cable is λ/2 — a half-wave antenna. Common-mode currents on cable shields or signal cables are the dominant radiated emissions source in many designs.
- PCB oscillator and clock harmonics: the crystal oscillator and MCU clock harmonics appear as narrowband peaks in the radiated spectrum. Each harmonic is a potential for a narrowband emissions failure if the layout couples the clock signal to cable antennas.
- Antenna intentional radiators (for products with Wi-Fi, BLE, LoRa): the intentional radio emissions are tested separately under the radio certification framework; what matters here is that the PCB layout of the host board doesn't allow the radio module to radiate more than its certified limits.
The Relationship Between Conducted and Radiated Emissions
Conducted and radiated emissions are not independent. Noise currents that exceed the conducted emissions limits at the lower end of the frequency range — typically the switching frequency and its immediate harmonics — often appear at higher harmonics as radiated emissions above 30 MHz. The mechanism is:
- Switching noise is generated at the power supply.
- Some noise reaches the supply cables (conducted path).
- Above 30 MHz, the cable (now comparable to an efficient antenna length) radiates that noise into the air (radiated path).
This coupling means that improving the conducted emissions filter at the power supply input often improves radiated emissions as well, particularly in the 30–100 MHz range where this transfer is most direct.
Conversely, radiated emissions above 100 MHz often originate from the PCB itself (fast clock edges, GPIO signals) rather than from the supply cables. These require layout-based solutions — edge rate control, controlled impedance traces, ground plane integrity — rather than input filtering.
For a complete treatment of how to address both categories through PCB layout and filtering, see how do you reduce EMI in PCB design?.
For products being developed through to RCM compliance with ACMA in Australia, Zeus Design's engineering team covers design-for-EMC, pre-compliance scanning, and certification strategy — contact Zeus Design to discuss your EMC compliance requirements.
Design Considerations
- Filtering the power supply input covers both emission types: a well-designed EMC input filter at the mains or DC supply port reduces conducted emissions directly and also reduces the common-mode current on supply cables, which in turn reduces radiated emissions from those cables. Investing in the input filter early pays dividends in both test categories.
- Cable routing affects radiated emissions outside the designer's control: in the test lab, cables are arranged in a standardised configuration. In the field, users route cables differently — sometimes dramatically increasing cable radiation. Filtering at every cable interface (ferrite beads, common-mode chokes at connector entry points) provides immunity to cable routing variation.
- Common-mode vs differential-mode noise has different filter solutions: differential-mode noise (between supply lines) is reduced by X-capacitors and DM inductors; common-mode noise (both lines to earth) is reduced by Y-capacitors and CM chokes. Misdiagnosing which mode dominates leads to ineffective filtering. A LISN measurement can distinguish the modes by comparing measurements made between L and N versus between each line and earth.
- Pre-compliance scanning identifies the dominant emissions mechanism before formal testing: a near-field scanning probe (probe held close to the PCB surface) can identify which area of the board is the dominant radiator. This directly guides layout changes or shielding decisions without the ambiguity of a full radiated emissions measurement on a commercial test range. For a step-by-step procedure covering equipment selection, scan setup, and how to correlate frequency peaks back to specific circuit blocks, see how to conduct EMC pre-compliance testing.
Common Mistakes
- Adding only differential-mode filtering to a product with dominant common-mode noise: if the conducted emissions failure is driven by common-mode currents (equal magnitude, same direction on both supply lines), adding X-capacitors (differential-mode) and DM inductors has minimal effect. Diagnosing the mode before designing the filter avoids this waste.
- Assuming cable emissions will be low because the PCB emissions are low: a PCB that passes radiated emissions in a test fixture without cables can fail dramatically when cables are connected, if the common-mode current on those cables is high. Always test with a representative cable configuration.
- Neglecting shielded enclosure ground bonding: a metal enclosure provides shielding only if it is bonded to the PCB ground at multiple points with low-impedance connections. A metal enclosure with a single bolt-down PCB standoff forms a resonant structure rather than a uniform shield — at frequencies where the enclosure dimension approaches λ/2, it actually amplifies emissions rather than reducing them.
- Using conducted emissions limits as the only guide during pre-compliance: products that pass the conducted emissions limits can still fail radiated limits due to PCB-originated radiation above 30 MHz that has no conducted path. Both categories require independent assessment during pre-compliance scanning.
Frequently Asked Questions
- Why does 30 MHz mark the boundary between conducted and radiated emissions?
- 30 MHz is approximately the frequency at which a standard mains cable (1–2 m long) becomes an efficient antenna — a half-wavelength at 30 MHz is 5 m, so a 1 m cable segment approaches λ/10. Below 30 MHz, cable radiation is inefficient and the dominant path for noise is conduction along the cable. Above 30 MHz, cables radiate efficiently and radiated field measurement becomes the more sensitive indicator. The LISN also becomes less accurate above 30 MHz, making antenna measurement the practical choice at higher frequencies.
- What is a LISN and why is it used for conducted emissions testing?
- A LISN (Line Impedance Stabilisation Network) is a network inserted between the mains supply and the equipment under test. It serves two purposes: it presents a defined, standardised impedance (typically 50Ω in parallel with 50 µH + 5Ω for CISPR measurements) to the equipment at the test frequencies, so measurements are reproducible regardless of the mains supply impedance; and it provides an RF-isolated tap point where a spectrum analyser can measure the noise voltage injected by the equipment without picking up mains-borne interference from the building supply.
- Can a product pass conducted emissions but fail radiated emissions, or vice versa?
- Yes — both are possible. A product with a switching power supply that has effective conducted filtering at the mains port can still fail radiated emissions if the switching loop on the PCB is large and radiates directly at frequencies above 30 MHz. Conversely, a product with poor conducted noise suppression can pass radiated testing if the cables are short and not efficient antennas at the test frequencies. This is why EMC standards require both tests — they are complementary, not redundant.
References
Related Questions
How Do You Reduce EMI in PCB Design?
PCB EMI starts with the switching loop and ground plane. This guide covers layout techniques and filtering approaches that make the biggest difference.
How Do You Conduct EMC Pre-Compliance Testing?
Pre-compliance EMC testing finds emission problems before the formal lab test. Covers equipment, test setup, near-field probing, and how to interpret results.
How Should You Lay Out a Buck Converter PCB?
PCB layout is as critical as component selection for buck converters. Learn the switching loop, SW node size, input cap placement, and thermal rules.
How Do You Design PCB Power and Ground Plane Layouts?
PCB power and ground planes distribute power and provide a low-impedance return path for every referenced signal. Here's how to design them well.
Linear vs Switching Regulator: Which Should You Use?
Choose a linear (LDO) regulator for low-noise designs with a small voltage drop; use a switching regulator when efficiency matters more than simplicity.
How Do You Get RCM Certification for a New Electronic Product in Australia?
How to get RCM marking in Australia: standards selection, NATA lab testing, Declaration of Conformity, and ACMA registration — explained step by step.