The magnetic field and electric field relation dictates that a changing electric field (driven by voltage) generates a magnetic field, and a changing magnetic field (driven by current) generates an electric field, binding them together as electromagnetic interference (EMI) in practical circuits. Understanding this coupling changes how you route cables, select shielding, and design filters, because ignoring it turns your wiring into unintended antennas. Most hobbyists and technicians commonly confuse electric field shielding—which only requires a simple conductive foil barrier tied to ground—with magnetic field shielding, which demands geometric cancellation or high-permeability metals.
The Core Relationship: Voltage Pushes, Current Circulates
To make decisions on the bench, strip away the abstract physics and look at the circuit variables. An electric field (E-field) is created by a difference in potential (voltage) between two points. It pushes outward from the conductor and terminates on another conductor or ground. A magnetic field (B-field) is created by the flow of charge (current). It circulates in concentric loops around the conductor.
According to Maxwell’s equations, summarized well in this Electronics Tutorials guide on Electromagnetism, these fields are not independent. When voltage changes rapidly ($dv/dt$), the expanding and collapsing E-field induces a current in nearby wires (capacitive coupling). When current changes rapidly ($di/dt$), the expanding and collapsing B-field induces a voltage in nearby loops (inductive coupling). In high-speed digital logic or switching power supplies, both are changing simultaneously, creating composite electromagnetic waves.
If you wrap a cable in aluminum foil, you have built a Faraday cage that blocks E-fields. However, low-frequency B-fields will pass right through the aluminum. To block magnetic fields, you must either use a thick, high-permeability material (like mu-metal) or cancel the field geometrically by twisting the wire pairs so the induced voltages oppose each other.
The Math on the Bench: A Worked Crosstalk Example
Let’s quantify exactly what this relation does to a real installation. Suppose you are routing a 5V analog sensor cable parallel to a 120V AC mains cable powering a 15A space heater. Both cables run together for 2 meters, spaced 10mm apart.
- Mutual Inductance ($M$): For unshielded parallel wires at this spacing, $M$ is roughly $1 \mu H$ per meter. For 2 meters, $M = 2 \mu H$.
- Mains Current: $I(t) = 15\sqrt{2} \sin(2\pi \cdot 60t)$. The peak current is 21.2A.
- Rate of Change ($di/dt$): The maximum rate of change occurs at the zero-crossing. $di/dt_{max} = I_{peak} \cdot \omega = 21.2 \cdot (2\pi \cdot 60) \approx 7992 \text{ A/s}$.
Using Faraday’s law of induction, the induced noise voltage on your sensor cable is:
V_noise = M × (di/dt_max)
V_noise = 2 × 10^-6 H × 7992 A/s ≈ 16 mV
What this means for your circuit: A standard 10-bit ADC on a 5V Arduino or ESP32 has a resolution of about 4.88 mV per step. That 16 mV of 60Hz magnetic crosstalk will cause your sensor readings to jitter by roughly 3 to 4 steps, completely ruining precision measurements. This is why the Analog Devices MT-095 Tutorial on Grounding and Shielding heavily emphasizes physical separation and loop-area reduction over simply adding software filters.
Where You Meet This in Practice
You will encounter the E-field and B-field relation in three primary areas of electrical and electronics work:
- PCB Layout and Ground Planes: High-speed digital traces (like ESP32 SPI lines) generate intense E-fields. A solid ground plane directly beneath the trace provides a low-impedance return path, confining the E-field to the microscopic dielectric gap and preventing it from radiating. Simultaneously, keeping the outbound and return traces close together minimizes the loop area ($A = length \times width$), which drastically reduces the B-field radiation.
- Switching Power Supply Snubbers: A MOSFET switching 300V DC at 100kHz creates a massive $dv/dt$. This E-field couples through parasitic capacitance to the chassis, causing common-mode EMI. An RC snubber slows the $dv/dt$, trading a tiny bit of efficiency for a massive reduction in electric field radiation.
- Motor Drive Cabling: VFDs (Variable Frequency Drives) and stepper motor choppers push high currents with fast rise times ($di/dt$). The resulting B-fields will induce voltage in any nearby unshielded control wiring. This is why industrial standards mandate physical separation or steel conduit (which provides high magnetic permeability) for motor leads.
Decision Tree: Choosing the Right Shielding and Routing
Use this decision matrix to select the correct cable and shielding strategy based on the dominant field type in your environment.
| Noise Source Profile | Coupling Type | Symptom on Scope | Concrete Solution & Part Pick |
|---|---|---|---|
| High Voltage, Low Current (e.g., 120V AC mains, switching nodes, fluorescent ballasts) |
Electric Field (Capacitive) |
High-frequency spikes, ringing, amplitude scales with $dv/dt$. | Foil shield grounded at ONE end. Pick: Belden 8760 (Foil shielded pair). Ground the drain wire at the panel, leave it floating at the sensor. |
| Low Voltage, High Current (e.g., Motor drives, solenoids, heater contactors) |
Magnetic Field (Inductive) |
Low-frequency hum (50/60Hz or PWM fundamental), scales with $di/dt$. | Twisted pair + high-permeability ferrite. Pick: Belden 9841 (STP) + Fair-Rite 2643625002 ferrite bead clamped over the cable. |
| Mixed / Unknown Environment (e.g., Factory floors, dense control panels) |
Electromagnetic (Both) |
Broadband hash, unpredictable spikes, communication dropouts. | Shielded Twisted Pair (STP) + Common Mode Choke. Default Pick: Belden 9841 routed in steel conduit, terminated through a Würth 744222 common-mode choke on the PCB. |
If you ground a foil shield at both ends in a high-magnetic-field environment, the shield itself becomes a conductor. The changing B-field will induce a current to flow through your shield (a ground loop), turning your shield into a noise injector. Always default to single-point grounding for foil shields unless you are using a dedicated high-frequency coaxial design.
Frequently Asked Questions
Why does my aluminum foil shield make magnetic noise worse?
When a changing magnetic field hits a conductive but non-magnetic material like aluminum, it induces swirling "eddy currents" inside the foil. These eddy currents generate their own secondary magnetic fields. While this provides some high-frequency attenuation (due to the skin effect), at lower frequencies (like 60Hz mains hum), the aluminum is essentially transparent to the B-field, and the shield just adds parasitic capacitance to the circuit, potentially worsening high-frequency E-field coupling.
How far apart should I route low-voltage sensor wires from AC mains?
The NEC and general industrial best practices (as detailed in resources like All About Circuits) recommend a minimum of 12 inches (300mm) of separation between unshielded low-voltage control wiring and 120V/240V AC power wiring. If you must cross them, cross at a strict 90-degree angle to minimize the parallel run length and reduce mutual inductance to near zero.
Does a ferrite bead block electric or magnetic fields?
A standard snap-on ferrite bead primarily combats magnetic fields by increasing the inductance of the cable, which chokes high-frequency common-mode currents ($di/dt$). It does not act as a Faraday cage and will not block an external electric field from capacitively coupling onto the wires. For E-field protection, you must rely on physical spacing or a grounded conductive shield.
