Capacitors

Decoupling and Bypass Capacitors: Purpose and Typical Values

What bypass and decoupling capacitors do, how the two terms differ, and typical values such as 100 nF ceramics beside IC supply pins with bulk 10 to 100 µF.

On this page
  1. What bypass and decoupling capacitors do
  2. Bypass vs decoupling: what the terms mean
  3. Typical values
  4. Placement matters
  5. Bypass capacitors in amplifiers
  6. Common mistakes
  7. Key takeaways

What bypass and decoupling capacitors do

Digital chips and amplifiers draw current in short bursts. Every time a logic gate switches, or an output stage swings, it pulls a sudden pulse of current from the power supply. The wires and copper traces between the supply and the chip have some inductance and resistance, so a sudden current pulse causes a small dip or spike in the supply voltage at the chip.

A capacitor placed close to the chip’s power pins supplies that pulse locally. It also shunts high-frequency noise on the supply to ground. This keeps the supply voltage steady at the chip and stops noise from spreading into other parts of the circuit.

Bypass vs decoupling: what the terms mean

The two words are often used for the same part, and that causes confusion.

  • Bypass describes what the capacitor does to a particular component: it gives AC signals or noise a low-impedance path around that component to ground. A capacitor across an amplifier’s emitter resistor is an example.
  • Decoupling describes the job of isolating one part of a circuit from the noise or current demands of another part, usually through a shared supply. A capacitor on a chip’s supply pin is a decoupling capacitor, because it stops that chip’s switching from disturbing the rest of the board.

In practice the same 100 nF capacitor on a chip’s supply pin is called either one. Datasheets often say “bypass (decoupling) capacitor” to cover both.

Typical values

Small ceramic values (100 nF)

A 100 nF (0.1 µF) ceramic capacitor placed on each power pin of a chip is the most common choice for digital logic and microcontrollers. It handles high-frequency switching noise, which is the main problem near the chip. Ceramic capacitors have low inductance, which helps at these frequencies. Read the value with the code guide in capacitor codes and markings, where 104 means 100 nF.

Bulk values (10 to 100 µF)

Larger capacitors are used to hold the supply up during bigger, slower current demands, and to smooth ripple from a regulator or rectifier. Common bulk values are about 10 µF to 100 µF, often placed at the regulator output or once per board section. Ceramic, tantalum or electrolytic types can be used. Electrolytics are polarized, so check the orientation first. See capacitor polarity.

Very small values for high frequencies

Some high-speed or radio-frequency circuits use smaller values, such as 10 nF or 1 nF, on specific pins. The right value depends on the frequency. Many designs use a mix: a small ceramic for high frequencies and a larger one for bulk, placed in parallel. Parallel values add their capacitance, as explained in capacitors and inductors in series and parallel.

Treat these numbers as typical starting points. Always follow the datasheet or reference design for the chip you are using.

Placement matters

For a bypass or decoupling capacitor, placement often matters more than the exact value. The capacitor only helps if it can deliver current faster than the supply can. That depends on the trace and via inductance between the capacitor and the chip.

  • Place the capacitor as close to the power pin as you can.
  • Keep the traces short and wide.
  • Connect the capacitor’s ground end to the ground plane with a short via.
  • Use the smallest package that gives the value you need at high frequency, where the board allows it.

A capacitor 2 cm away from a pin can behave like a much smaller one, because the trace inductance dominates at high frequencies.

Bypass capacitors in amplifiers

In a common-emitter transistor amplifier, the emitter resistor provides stability and sets the gain, but it also reduces the gain for AC signals. A bypass capacitor across the emitter resistor shunts the AC signal around the resistor, so the gain goes up for AC while the DC bias stays the same.

Without that capacitor, the amplifier still works. It simply has lower gain, because the emitter resistor stays in the AC path. This is a useful trade-off in some designs, which give stability and linearity at the cost of gain.

Common mistakes

  • Putting the capacitor far from the pin. The trace inductance then blocks the high-frequency current it is meant to supply.
  • Using one large electrolytic and nothing else. A large electrolytic is poor at high frequencies, so add a small ceramic.
  • Ignoring the voltage rating. Choose a rating with margin for the supply voltage.
  • Assuming a bypass capacitor improves sound quality. A bypass capacitor’s role is electrical. Claims about sound quality are subjective, and the effect depends on the circuit.

Key takeaways

  • Bypass and decoupling capacitors supply local current bursts and shunt noise to ground.
  • The two terms overlap. Decoupling emphasizes isolating a circuit section, and bypass emphasizes a path around a component.
  • A 100 nF ceramic on each chip’s power pin is the usual starting point. Add a 10 to 100 µF bulk capacitor per regulator or board section.
  • Placement and short connections matter as much as the value.
  • A missing emitter bypass capacitor lowers AC gain but does not stop the amplifier from working.

The complete guide to capacitors

Capacitors Explained: How They Behave in DC and AC Circuits
Read the guide

Keep exploring

Guides from other topics

All topics →