The short answer is no, DC current cannot flow through a capacitor's dielectric, while AC current appears to flow through it via displacement current. However, on a workbench, the real question isn't just about textbook physics—it's about how much current a specific capacitor chemistry can handle before it overheats, leaks, or explodes. When you apply voltage to a capacitor, you are dealing with three distinct current phenomena: charging/discharging current, AC ripple current, and DC leakage current.
Understanding these current paths is the difference between a power supply that runs for ten years and one that vents electrolyte across your PCB in a week. Below, we break down the physics, map the real-world current limits of common dielectrics, and provide the exact selection and substitution rules you need for your next build.
The Physics: Displacement Current vs. True Conduction
A capacitor consists of two conductive plates separated by an insulating dielectric. Because the dielectric is an insulator, true electron flow (DC conduction) is blocked. When you connect a DC source, electrons pile up on one plate and repel electrons from the other, creating a brief transient charging current. Once the capacitor reaches the source voltage, current drops to zero, save for a tiny DC leakage current that sneaks through the imperfect dielectric.
With AC voltage, the polarity constantly reverses. The capacitor charges and discharges continuously. No electrons cross the dielectric gap, but the continuous movement of electrons in the external circuit creates what James Clerk Maxwell termed displacement current. To the rest of your circuit, it behaves exactly as if AC current is flowing straight through the component. As detailed in the All About Circuits AC reactance guide, this opposition to AC flow is called capacitive reactance ($X_c = \frac{1}{2\pi fC}$), which drops as frequency or capacitance increases.
Real-World Current Handling & Leakage Specs (10µF Equivalents)
Theory assumes ideal components; reality introduces Equivalent Series Resistance (ESR) and dielectric absorption. Here is how a standard 10µF capacitor behaves across different chemistries when subjected to real-world currents.
| Capacitor Type | Max Ripple Current (100kHz) | Typical DC Leakage | Typical ESR | Dielectric Breakdown Risk |
|---|---|---|---|---|
| Aluminum Electrolytic | ~150 mA | 3.0 µA to 10 µA | 500 mΩ - 2 Ω | Moderate (Venting) |
| Solid Tantalum | ~80 mA | 0.5 µA to 2 µA | 100 mΩ - 500 mΩ | High (Thermal Runaway/Fire) |
| MLCC (X7R Ceramic) | > 2.0 A | < 10 nA | 2 mΩ - 10 mΩ | Low (Flex Cracking) |
| Metallized Film (PP) | ~500 mA | < 1 nA | 10 mΩ - 50 mΩ | Very Low (Self-healing) |
Capacitor Type Selection: Which Dielectric for Which Job?
Choosing the right capacitor isn't just about hitting the target microfarad value; it's about matching the dielectric's physical construction to the current and thermal environment of your circuit. Referencing SparkFun's component guidelines, here is the selection matrix for the four most common through-hole and SMD families.
| Type | Construction / Dielectric | Typical Tolerance | Temperature Coefficient | Best Typical Use |
|---|---|---|---|---|
| Aluminum Electrolytic | Etched aluminum foil with liquid/polymer electrolyte | ±20% | High capacitance drop at -40°C | Bulk power supply filtering, low-frequency audio coupling |
| Solid Tantalum | Sintered tantalum powder with manganese dioxide/polymer cathode | ±10% to ±20% | Stable across -55°C to +125°C | Space-constrained DC decoupling, medical/aerospace hold-up |
| MLCC (Ceramic) | Alternating layers of metal electrodes and ceramic dielectric | ±10% (X7R) to ±22% (Y5V) | X7R: ±15% over temp; C0G: ±30ppm/°C | High-frequency bypass, SMPS output filtering, RF tuning |
| Film (Polypropylene) | Metallized plastic film wound into a cylinder | ±1% to ±5% | Extremely stable (-200 ppm/°C) | EMI suppression (X/Y class), motor run, precision audio crossovers |
Decoding Physical Markings and Capacitance Codes
Through-hole electrolytics and large film caps usually have their values printed plainly (e.g., "100µF 50V"). But when you are scavenging parts or reading SMD MLCCs, you will encounter the 3-digit EIA (Electronic Industries Alliance) code. Here is how to read them without guessing.
The 3-Digit EIA System
The first two digits are the significant figures. The third digit is the multiplier (number of zeros to add), and the base unit is always picofarads (pF).
- 104: 10 followed by four zeros = 100,000 pF = 100 nF = 0.1 µF. (The most common bypass cap in existence).
- 223: 22 followed by three zeros = 22,000 pF = 22 nF.
- 471: 47 followed by one zero = 470 pF.
- 100: 10 followed by zero zeros = 10 pF. (Note: this is different from 100µF printed on an electrolytic).
Tolerance and Voltage Letters
You will often see a letter trailing the numeric code. This denotes tolerance or voltage rating depending on the manufacturer standard:
- J: ±5% tolerance
- K: ±10% tolerance
- M: ±20% tolerance
- 1H / 1V / 1E: EIA voltage codes where 1H = 50V, 1V = 35V, 1E = 25V, 1C = 16V, 1A = 10V.
Failure Modes and Visual Symptoms
When you force too much current through a capacitor—either by exceeding its ripple current rating or by subjecting a polarized part to reverse DC current—the internal dielectric breaks down or the electrolyte boils. Here is what to look for when debugging a dead board.
Solid tantalum capacitors are polarized and highly sensitive to reverse voltage and current spikes. When they fail, the manganese dioxide dielectric can ignite, resulting in a violent, self-sustaining flame that will scorch your PCB. Never substitute a tantalum for an electrolytic in high-inrush circuits without adding series resistance or using a polymer-tantalum variant.
Visual Symptom Checklist
- Aluminum Electrolytic (Bulging/Venting): The top scoring (the Krupton vent) splits open, or the rubber bung at the bottom pushes out. You will see a crusty brown or black residue (dried electrolyte) on the PCB. Cause: Excessive ripple current generating internal heat, or reverse polarity.
- MLCC (Flex Cracking): There is rarely a visual symptom to the naked eye. Under a microscope, you will see a hairline fracture near the solder termination. Cause: Mechanical board flexing during depaneling or connector insertion, causing the piezoelectric ceramic to snap. The resulting short circuit allows massive DC current to flow, eventually burning a hole through the PCB.
- Tantalum (Charring/Explosion): The epoxy casing is cracked, blackened, or completely missing, leaving a charred crater. Cause: Forward current spikes exceeding the part's surge rating, or reverse DC voltage.
- Film Capacitor (Swelling/Discoloration): The plastic dip or epoxy coating is yellowed or bulging. Cause: Sustained overvoltage or high-frequency AC current exceeding the $dV/dt$ rating, causing internal corona discharge.
Safe Substitution: What to Do When the Exact Part is Missing
You are repairing a board or building a prototype at 2 AM, and you don't have the exact BOM part. Substituting capacitors is generally safe if you follow the hierarchy of dielectric stress, but there are critical edge cases where a "better" spec will break your circuit.
The Golden Rules of Substitution
- Voltage Rating: Always go equal or higher. Substituting a 50V cap for a 16V cap is perfectly safe (and often improves reliability). Never go lower.
- Capacitance Value: For power supply bypassing and bulk filtering, going higher (e.g., using 22µF instead of 10µF) is usually fine. For timing circuits (like a 555 timer oscillator) or active filters, you must match the exact value or the frequency response will shift.
- Temperature Rating: Swapping an 85°C electrolytic for a 105°C variant increases lifespan exponentially. Never put an 85°C part in a 105°C rated design.
The ESR Trap in LDO Regulators
The most common substitution mistake involves Low Dropout (LDO) linear regulators. Older LDOs (like the classic LM1117) require a specific amount of Equivalent Series Resistance (ESR) on the output capacitor to maintain control loop stability. The datasheet usually specifies an ESR window (e.g., 0.1Ω to 1.0Ω).
If your BOM calls for a 10µF Aluminum Electrolytic (which naturally has high ESR) and you substitute it with a 10µF MLCC Ceramic (which has near-zero ESR), the LDO's feedback loop will lose its phase margin. The regulator will break into high-frequency oscillation, and your "clean" DC output will become a noisy AC mess. If you must use a low-ESR ceramic in place of an electrolytic for an older LDO, add a small 0.5Ω series resistor to artificially recreate the required ESR.
