The current through a capacitor is not a flow of electrons crossing the dielectric; it is a displacement current driven by the rate of voltage change over time, defined by the equation i = C(dv/dt). In DC steady-state, this current is effectively zero (save for microamp leakage). But in AC circuits or switching DC power supplies, this displacement current manifests as RMS ripple current. If you ignore the RMS current rating and treat a capacitor merely as a 'voltage bucket,' the internal equivalent series resistance (ESR) will dissipate that current as heat, boiling the electrolyte and destroying the component.
The Bench Failure: When Ripple Current Melts an Input Filter
To understand why calculating the current through a capacitor matters, let us look at a real-world bench failure involving a 12V to 3.3V buck converter designed to drive a 5A microcontroller load.
The Setup: The designer needed an input filter capacitor for the buck converter. They selected a standard, off-the-shelf 470µF 25V aluminum electrolytic capacitor. The switching frequency was 500 kHz.
The Numbers: The RMS input ripple current for a buck converter is calculated as: I_in(rms) = I_out × √[ (V_out/V_in) × (1 - V_out/V_in) ]. Plugging in our numbers (5A out, 3.3V out, 12V in), the RMS current through the input capacitor is 2.23A. The generic 470µF capacitor chosen had a maximum ripple current rating of just 0.85A at 105°C and an ESR of 0.12Ω.
The Outcome: Power dissipation inside the capacitor is P = I² × ESR. That yields (2.23)² × 0.12 = 0.59W of pure heat trapped inside a small radial can. Within four hours of burn-in testing, the capacitor's internal core temperature exceeded 130°C. The electrolyte vaporized, the pressure vent popped with a loud hiss, and the capacitor failed open, taking the switching IC with it due to input voltage ringing.
What Went Wrong: The designer looked only at capacitance and voltage rating, entirely ignoring the RMS current through the capacitor. The fix was swapping the generic part for a low-ESR polymer or high-ripple electrolytic (like the Panasonic FR series), which handles 3.5A of ripple and features an ESR under 0.02Ω.
The Physics: Reactance and Displacement Current
When an alternating voltage is applied, the capacitor continuously charges and discharges. The opposition to this AC current is called capacitive reactance (Xc), measured in ohms. According to All About Circuits, reactance is inversely proportional to both frequency and capacitance:
Xc = 1 / (2πfC)
At 60Hz mains frequency, a 1µF capacitor has a reactance of roughly 2,652Ω, passing only a few milliamps. But at a 500kHz switching frequency, that same 1µF capacitor's reactance drops to 0.31Ω, allowing massive high-frequency ripple currents to flow. This is why high-frequency switching supplies require capacitors specifically engineered to handle high RMS currents without thermal runaway.
Capacitor Types: Which Dielectric for Which Job?
Selecting the right component requires matching the dielectric material to the circuit's current, frequency, and temperature demands. Here is the selection matrix for the most common bench components.
| Type | Construction / Dielectric | Tolerance | Tempco (Temp Stability) | Ripple Current Handling | Typical Use Case |
|---|---|---|---|---|---|
| Aluminum Electrolytic | Etched aluminum foil, liquid/solid electrolyte | ±20% | Poor (dries out over time) | Moderate (High in 'Low-ESR' series) | Bulk filtering, low-frequency smoothing |
| MLCC (Ceramic X7R) | Multi-layer ceramic, nickel electrodes | ±10% to ±20% | Moderate (capacitance drops with DC bias) | Excellent (Virtually zero ESR) | High-frequency decoupling, switching supply bypass |
| MLCC (Ceramic C0G/NP0) | Class 1 ceramic dielectric | ±1% to ±5% | Excellent (0 ±30ppm/°C) | Excellent | RF tuning, precision timing, audio signal paths |
| Tantalum | Tantalum metal anode, manganese dioxide cathode | ±10% to ±20% | Good | Poor (Prone to thermal runaway) | Space-constrained bulk filtering (low ripple only) |
| Film (Polypropylene) | Metallized plastic film, wound or stacked | ±1% to ±5% | Excellent | Very High (Self-healing properties) | AC line filtering, motor run, high-power audio crossovers |
Decoding the Silk Screen: Reading Capacitor Markings
Misreading a capacitor code is a fast track to a dead board. Here is how to decode the physical markings on the three most common packages you will encounter at the bench.
Ceramic MLCC (The 3-Digit Code)
Through-hole and small SMD ceramics use a 3-digit picofarad code. The first two digits are the significant figures, and the third is the multiplier (number of zeros).
- 104: 10 followed by four zeros = 100,000 pF = 100 nF = 0.1 µF.
- 473: 47 followed by three zeros = 47,000 pF = 47 nF.
- Letter Suffix: A trailing letter indicates tolerance. 'J' = ±5%, 'K' = ±10%, 'M' = ±20%.
Aluminum Electrolytic (The Polarity Stripe)
Electrolytics are polarized. The cylindrical can will have a prominent stripe running down the side. The stripe always denotes the NEGATIVE lead. The capacitance (e.g., 470µF) and maximum DC voltage (e.g., 25V) are printed in plain text. Never apply reverse voltage; it will cause a rapid exothermic venting event.
Tantalum (The Dangerous Exception)
Tantalum capacitors (often yellow or orange SMD beads) are also polarized, but their marking convention is the exact opposite of aluminum electrolytics. The stripe on a tantalum capacitor denotes the POSITIVE anode. Reversing a tantalum does not just pop a vent; it typically results in a violent, self-sustaining chemical fire.
Failure Modes and Visual Autopsies
When troubleshooting a dead board, capacitors often leave physical evidence of their demise. According to Cornell Dubilier's application guides, understanding these visual symptoms tells you whether the failure was caused by overvoltage, overcurrent, or age.
- Electrolytic Bulging/Venting: The top of the can domes upward, or the cross-scored vent on top splits open. You may see crusty brown or black electrolyte leaked onto the PCB. Cause: Excessive RMS ripple current generating internal heat, or reverse polarity.
- Electrolytic 'Sucking In': The bottom rubber bung is pulled inward, and the ESR measures incredibly high on an LCR meter, though it looks fine visually. Cause: Electrolyte evaporation due to old age and proximity to hot components (like heatsinks).
- Ceramic Cracking: MLCCs show microscopic fractures near the PCB solder pads, often leading to a dead short. Cause: Mechanical board flexure during assembly or piezoelectric stress from high AC ripple voltages.
- Tantalum Charring: The SMD bead is cracked open, discolored black, and smells like burnt plastic. Cause: Exceeding the surge current rating or a momentary reverse voltage spike.
The Substitution Matrix: Swapping Parts Safely
You are at the bench, the exact BOM part is out of stock, and you need to keep the prototype running. Follow these numbered rules to substitute safely without compromising the current handling or circuit stability.
- Voltage Rating Rule: You can always substitute a capacitor with a higher DC voltage rating. A 50V cap can safely replace a 25V cap. Never substitute a lower voltage rating, even if the nominal rail voltage seems safe; transient spikes will breach the dielectric.
- Capacitance Variance: For bulk power supply filtering, a ±20% variance in capacitance is acceptable. For timing circuits (like a 555 timer oscillator) or resonant filters, you must match the exact value and use a tight tolerance (C0G or Film) part.
- The ESR Trap: If you are replacing an input/output capacitor on a switching regulator, you must replace it with a Low-ESR equivalent. Swapping a Low-ESR Panasonic FM series with a standard generic electrolytic will cause the regulator's control loop to go unstable and the capacitor to overheat from ripple current.
- Dielectric Swaps: Never swap an X7R ceramic for a C0G/NP0 in a high-voltage DC bias application. X7R loses up to 50% of its rated capacitance when DC voltage is applied; C0G does not. If you swap them in a precision analog filter, your cutoff frequency will shift drastically.
By treating the current through a capacitor as a primary design parameter—rather than an afterthought—you eliminate the most common source of mysterious thermal failures on the workbench. Always calculate the RMS ripple, verify the ESR, and read the polarity stripe twice before applying power.
