Ferrite Bead Selection for EMI Suppression on PCBs
A ferrite bead looks like an inductor and is often drawn like one on a schematic, but it behaves differently. It is specified in ohms rather than henries because the important figure is the impedance it presents at a given frequency, not the inductance it stores. Used correctly, a bead turns high frequency noise into a small amount of heat and stops it from travelling along a cable. Used carelessly, it creates a resonance that makes emissions worse.
Impedance, Not Inductance
Manufacturers rate a bead by its impedance at 100 MHz, written as something like 600 Ω at 100 MHz. At low frequency the part looks inductive and passes the signal; as frequency rises, losses in the ferrite material dominate and the impedance becomes largely resistive. That resistive behaviour is what absorbs noise instead of reflecting it back into the circuit.
The distinction matters because a bead that is chosen only by its inductance value will have unpredictable loss behaviour. Always work from the impedance curve, which the supplier publishes as impedance against frequency with the DC current as a parameter.
Choosing the Impedance Value
Match the bead to the noise band you are trying to suppress. A part with a peak impedance near 100 MHz is suitable for switching noise from a converter running in the hundreds of kilohertz to a few megahertz, while higher frequency interference needs a bead whose curve peaks higher. The self resonant frequency of the bead sits above the useful range, and beyond it the part becomes capacitive and stops suppressing anything.
Where a broad band of noise has to be attenuated, two beads with different characteristics in series often work better than one very large part. The total series resistance added to the circuit also matters, because a high impedance bead in a supply line will drop voltage under load, and that drop has to be accounted for in the rail budget.
DC Bias and Saturation
A ferrite bead carrying substantial direct current loses much of its impedance. The magnetic material saturates, and a part specified at 600 Ω may fall to 100 Ω or less at its rated current. The curve provided by the manufacturer shows this derating, and it should be consulted before the bead is placed on a power rail.
Larger beads, or beads with a bigger cross section, saturate less for the same current. Physical size helps in another way too: a long, thin bead shape generally gives more loss than a short, thick one of the same volume, and a smaller inner diameter concentrates the field more effectively.
Placement on a Board
Place the bead as close as possible to the source of the noise, not at the connector. A bead at the source keeps the noise current confined to a small loop and stops it from coupling into other parts of the board. When the goal is to protect a cable from carrying noise out of the enclosure, beads placed at the cable entry also help, and the two positions serve different purposes.
The return path is as important as the bead itself. A bead in the supply line without a corresponding low impedance return creates a loop antenna, so the ground plane under the bead and the decoupling capacitor next to it have to be planned together. The general principles of keeping loops small are covered in the notes on EMI suppression design principles.
Beads Compared With Inductors
An inductor stores energy and is used in filters and converters where a defined inductance is required, such as the output stage of a switching regulator or a resonant circuit. A bead dissipates energy and is used to attenuate unwanted frequencies. Choosing an inductor for EMI suppression often produces a resonant peak with the surrounding capacitors, which can amplify the very noise it was meant to remove.
A bead has almost no DC resistance compared with a wound inductor of similar impedance, and it is far smaller, which is why it dominates on dense boards. Where a filter must have a defined roll-off, the inductor is still the correct part, and the difference is worth stating explicitly in the schematic so that the two are not substituted during purchasing. Known interactions between traces and radiation are described in the discussion of radiated EMI in switching regulator layout.
Where Beads Are Used
Typical applications include DC to DC converter inputs and outputs, USB and other data lines, audio rails, sensor supplies and the interface between a board and a cable. In each case the bead is part of a filter network with capacitors on either side, and the impedance of the bead at the noise frequency should be much larger than the impedance of the surrounding circuit for the filter to work.
Beads are also used in differential pairs, but there they must be placed carefully. A bead in one leg of a pair and not the other converts differential noise into common mode noise, which is usually worse for emissions.
Practical Selection Checklist
Start with the frequency band of the noise, then choose the impedance curve that peaks there. Check the DC bias derating curve at the actual current, confirm the DC resistance is acceptable for the rail, and verify the physical size against the placement area. Finally, confirm that the part has a defined loss characteristic rather than only an inductance value.
After assembly, measure the emissions with the bead fitted and removed. A part that improves one band while making another worse is a sign that a resonance has been created, and the fix is usually a smaller bead or a different capacitor value rather than a larger component.
Filter Integration and Layout
A bead does not work alone. It works with the capacitors on either side of it, and the choice of those capacitors determines the frequency range over which the filter is effective. A bead with a 100 nF capacitor forms a low pass network whose corner frequency can be computed directly, and adding a second capacitor of a different value extends the attenuation across a wider band.
The impedance of the source and the load also matters. A bead works well in a low impedance circuit, such as a power rail, and poorly in a high impedance signal path where the surrounding resistance already limits the current. That is why the same component can be very effective on a converter input and almost useless on a sensitive analogue input.
Layout decides whether the filter behaves as designed. The capacitor ground must return to the same reference as the source, with a short path and no shared impedance with the noisy return. Keeping the bead and its capacitors in a compact group, close to the connector or the device they protect, follows the same reasoning used in DC-DC converter layout and routing.
FAQ
Why is a ferrite bead rated in ohms? Because it is used as an impedance, not as an energy storage element. The ohmic value states the impedance at a reference frequency such as 100 MHz, which is more useful for filter design than an inductance figure.
Can a bead be used in a high current rail? Yes, if the derating curve is respected. The impedance falls as current rises, so a part chosen for its small signal performance may be almost useless at the working current.
What happens if a bead is placed on only one line of a differential pair? The pair becomes unbalanced. Common mode noise increases and emissions usually get worse, so beads on differential lines should be applied symmetrically or not at all.
