A capacitor is a passive two-terminal electronic component that stores electrical energy in an electric field. Physically, it consists of two conductive plates separated by an insulating material called a dielectric. But what does a capacitor do in a practical circuit? It resists sudden changes in voltage, blocks direct current (DC) while allowing alternating current (AC) to pass, and filters out high-frequency noise. Think of it as a flexible rubber membrane stretched across a water pipe: it stretches to absorb sudden pressure spikes and pushes back when the pressure drops, smoothing out the flow.
The Core Physics: What a Capacitor Actually Does in a Circuit
The fundamental governing equation of a capacitor is I = C × (dV/dt). This means the current flowing through a capacitor is proportional to the rate of change of the voltage across it. If the voltage is constant (pure DC), the current is zero. If the voltage changes rapidly, the capacitor sources or sinks a large current to oppose that change.
Let us look at a concrete bench example. Imagine a 5V microcontroller rail powering an ESP32. When the WiFi radio transmits, it pulls a sudden 50mA current spike lasting 10µs. If we rely solely on a 100nF (0.1µF) ceramic decoupling capacitor, the voltage dip is calculated as:
ΔV = (I × Δt) / C = (0.05A × 0.00001s) / 0.0000001F = 5V
The rail would momentarily dip to 0V, causing a brownout reset. This is why we place a bulk 10µF capacitor in parallel. With 10µF, the dip becomes (0.05 × 0.00001) / 0.00001 = 0.05V (50mV), which the microcontroller easily rides through. This dual-capacitance approach is why you see both large and small capacitors clustered around IC power pins.
Capacitor Types: Which Dielectric for Which Job?
Selecting the right capacitor is not just about matching the microfarad value; the dielectric material dictates the component's behavior under temperature, voltage, and frequency stress. According to KEMET's ceramic capacitor guide, the dielectric class fundamentally alters capacitance under applied DC bias.
| Type | Dielectric | Tolerance | Tempco (Temp Stability) | Typical Use Case |
|---|---|---|---|---|
| MLCC (Class I) | C0G / NP0 | ±1% to ±5% | ±30 ppm/°C (Excellent) | RF filters, precision timing, resonant tanks |
| MLCC (Class II) | X7R / X5R | ±10% to ±20% | ±15% over -55°C to +125°C | General decoupling, bulk bypass, non-critical coupling |
| Aluminum Electrolytic | Aluminum Oxide | ±20% typically | Poor (High drift, limited life) | Power supply bulk filtering, audio coupling |
| Tantalum | Tantalum Pentoxide | ±10% to ±20% | Moderate (Stable but temp sensitive) | Space-constrained bulk filtering, medical/military |
| Film (Polyester/PP) | Mylar / Polypropylene | ±5% to ±10% | Good (Self-healing properties) | High voltage snubbers, AC line filtering, motor run |
Selection Rule of Thumb: Use C0G/NP0 for anything involving signal timing or RF. Use X7R/X5R for power rail decoupling. Use Aluminum Electrolytic for high-capacitance bulk energy storage at low frequencies. Never use X7R in precision analog signal paths, as its capacitance drops significantly under DC bias voltage and it exhibits piezoelectric microphonic noise.
Decoding the Markings: How to Read Capacitor Codes
Through-hole and large SMD capacitors usually have their values printed plainly (e.g., "470µF 25V"). However, small ceramic capacitors use the EIA 3-digit code system, which trips up many beginners.
The first two digits are the significant figures, and the third digit is the multiplier (number of zeros), expressed in picofarads (pF).
- 104: 10 × 10^4 pF = 100,000 pF = 100nF = 0.1µF
- 103: 10 × 10^3 pF = 10,000 pF = 10nF
- 471: 47 × 10^1 pF = 470 pF
- 225: 22 × 10^5 pF = 2,200,000 pF = 2.2µF
Tolerance and Voltage Codes:
Following the numeric code, you may see a letter indicating tolerance: J (±5%), K (±10%), or M (±20%). Voltage ratings are often omitted on tiny ceramics. If a datasheet specifies a 50V rating for a specific 0805 X7R part, you cannot assume a random 104 capacitor from a bin is rated for 50V; many generic 104 caps are only rated for 16V or 25V.
Failure Modes: Visual Symptoms and Bench Testing
Capacitors do not last forever. Electrolytic capacitors contain liquid electrolyte that slowly evaporates, while ceramics suffer from mechanical stress. Here is what to look for when troubleshooting a dead board.
- Aluminum Electrolytic (Bulging/Venting): The top cross scores open to vent gas. You will see a domed top or a brown, crusty electrolyte leak at the base. This indicates the internal liquid has boiled due to excessive ripple current or reverse bias.
- MLCC Ceramic (Flex Cracking): Common in large SMD packages (1206 and above). If the PCB flexes during assembly or operation, the rigid ceramic cracks internally. Visually, the solder joint might look fine, but the part acts as a dead short. This is why board designers add PCB slots or use smaller package sizes in high-stress zones.
- Tantalum (Thermal Runaway): Tantalum capacitors fail catastrophically. If subjected to voltage spikes exceeding their rating or reverse polarity, they ignite. Visually, you will find a charred, melted black lump on the board. According to Cornell Dubilier's technical guides, derating tantalum voltage by 50% is standard practice to prevent these fires.
- ESR Drift (The Silent Killer): As electrolytics age, their ESR rises. A 1000µF cap might still read 950µF on a standard multimeter, but its ESR might have climbed from 0.05Ω to 5Ω. In a 100kHz switch-mode power supply, that 5Ω ESR creates massive heat and voltage ripple, causing the supply to shut down.
Safe Substitution: When You Don't Have the Exact Part
When your parts bin lacks the exact BOM component, you can substitute safely if you follow the hierarchy of capacitor parameters.
- Voltage Rating: You can always substitute a higher voltage rating for a lower one (e.g., using a 50V cap instead of a 25V cap). Never substitute a lower voltage rating. Note that higher voltage caps often have higher ESL/ESR and larger physical footprints, which may affect high-frequency performance.
- Capacitance Value: For bulk filtering and decoupling, substituting a higher value (e.g., 10µF instead of 4.7µF) is generally safe and often beneficial. For timing circuits, oscillators, and active filters, the exact value is critical; a 10% deviation will shift your frequency response.
- Dielectric Class: Never substitute a Class II (X7R) for a Class I (C0G) in a timing or filter circuit. The X7R will introduce distortion and temperature drift. You can, however, swap an X5R for an X7R in general power bypass applications, provided the operating temperature stays within the X5R's -55°C to +85°C range.
- Polarity: You can replace a polarized electrolytic with a non-polarized film or ceramic capacitor of the same value (if you can find one that fits), but never replace a non-polarized capacitor with a polarized one in an AC or bipolar signal path.
Frequently Asked Questions
What does a capacitor do in a DC circuit versus an AC circuit?
In a pure DC circuit, a capacitor acts as a temporary energy reservoir. It draws current only while charging or discharging to match the source voltage, effectively blocking steady-state DC current once fully charged. In an AC circuit, the continuously reversing voltage causes the capacitor to constantly charge and discharge. This allows alternating current to 'flow' through the circuit, with the capacitor presenting a frequency-dependent resistance called capacitive reactance (Xc = 1 / (2πfC)). Higher frequencies pass more easily than lower frequencies.
Can I replace a polarized capacitor with a non-polarized one?
Yes, with caveats. A non-polarized capacitor (like a film or ceramic type) can safely replace a polarized electrolytic capacitor in a DC circuit, provided you can match the capacitance and voltage ratings. Non-polarized parts are inherently safer regarding reverse-voltage installation errors. However, non-polarized capacitors with high microfarad values (e.g., 1000µF) are physically massive and expensive compared to electrolytics, making the swap impractical for bulk power supply filtering.
Why do we put a 100nF capacitor next to every IC VCC pin?
This practice, detailed extensively in the Texas Instruments application note on decoupling, addresses high-frequency transient noise. When digital logic gates inside an IC switch simultaneously, they draw sharp, nanosecond-scale current spikes from the power rail. The physical trace inductance of the PCB prevents the main power supply from responding fast enough to these spikes. A 100nF (0.1µF) ceramic capacitor placed as close as physically possible to the VCC and GND pins acts as a localized, high-speed energy reservoir, supplying the instantaneous current and preventing the logic voltage from dipping below the chip's threshold.
