To check a capacitor with a multimeter, you must first safely discharge it, isolate it from the circuit, set your meter to the capacitance (F) or resistance (Ω) mode, and measure across the terminals. A good reading falls within the manufacturer's specified tolerance—typically ±20% for aluminum electrolytics and ±10% for film capacitors. If the reading is significantly low, reads as an open circuit (OL), or shows a dead short (near 0Ω), the component has failed.
Meter Setup and Safety Category (CAT) Requirements
Before you touch a single probe to a component, you need to configure your digital multimeter (DMM) correctly and verify your safety category. Testing capacitors in power supplies, HVAC units, or appliance control boards exposes you to stored lethal energy and mains-adjacent voltages.
For any measurement involving mains-powered equipment, your meter must carry the appropriate safety rating. A CAT II meter is required for testing appliance boards and power tools, while a CAT III meter is mandatory for HVAC run capacitors and hardwired distribution panels. Never use a CAT I hobbyist meter on mains-adjacent circuits.
Meter Setup Block
- Dial Position: Set to Capacitance (marked as F, mF, µF, or nF). If your meter lacks a capacitance setting, use the Resistance (Ω) mode as a secondary diagnostic.
- Lead Jacks: Black lead in COM, Red lead in the V/Ω/F (or dedicated capacitance) jack. Note: Some older meters require moving the red lead to a specific 'mA/µF' jack.
- Range Selection: Use Auto-ranging if available. For manual ranging, select the next highest decade above the capacitor's rated value (e.g., select the 2mF range for a 470µF capacitor).
- Zeroing: Touch the probes together and press the 'REL' (Relative) or 'ZERO' button to null out the inherent capacitance of your test leads (usually 0.1nF to 0.5nF).
Expected Readings: Good vs. Bad Capacitors
Knowing what a good reading looks like numerically is the difference between a successful repair and chasing ghosts. The table below provides real-world benchmarks for the most common capacitor chemistries you will encounter on the bench. According to Fluke's testing guidelines, a capacitor measuring outside its rated tolerance band should be replaced, especially in switching power supplies where ripple current tolerance is critical.
| Component Type & Rating | Good Reading (In Tolerance) | Degraded (Replace Soon) | Failed (Short / Open) |
|---|---|---|---|
| Aluminum Electrolytic (470µF, 25V, ±20%) |
376µF – 564µF | 300µF – 375µF (High ESR likely) | < 10Ω (Short) or OL (Open) |
| Ceramic Disc (104 / 100nF, ±10%) |
90nF – 110nF | 70nF – 89nF (Dielectric cracking) | OL (Open) or < 1kΩ (Leaky) |
| Metallized Film (0.1µF, 400V, ±5%) |
0.095µF – 0.105µF | 0.080µF – 0.094µF | OL (Internal fuse blown) |
| Tantalum Polarized (10µF, 16V, ±10%) |
9.0µF – 11.0µF | N/A (Tantalums fail catastrophically) | < 5Ω (Dead Short / Burnt) |
Step-by-Step: Testing Capacitance and Resistance
Follow this procedure to accurately check a capacitor with a multimeter. This assumes you are testing an out-of-circuit or isolated component.
- Discharge the Capacitor: Connect a bleeder resistor across the terminals for 5 to 10 seconds. Verify it is at 0V DC using your multimeter's voltage function before switching to capacitance mode.
- Isolate the Component: If the capacitor is soldered to a PCB, desolder and lift at least one leg. Testing in-circuit will result in parallel impedance paths, giving you wildly inaccurate readings.
- Probe Placement:
- Polarized (Electrolytic/Tantalum): Place the red probe on the anode (positive/long leg) and the black probe on the cathode (negative/stripe side).
- Non-Polarized (Ceramic/Film): Probe placement does not matter; place one probe on each lead.
- Read and Stabilize: Hold the probes firmly. Small ceramic caps will read instantly. Large electrolytics (1000µF+) require the meter to output a charging current; wait 3 to 8 seconds for the display to stabilize on the final value.
- Secondary Resistance Check (Optional): Switch the dial to Ohms (Ω). Place probes across the leads. A healthy, uncharged capacitor will start at a low resistance and steadily climb until it reads 'OL' (Open Loop) as it charges from the meter's internal battery. If it stays at 0Ω, it is shorted. If it immediately reads OL without climbing, it is internally open.
Common Mistakes That Give Misleading Readings
Even with a high-end bench meter like a Keysight U1252B or a Fluke 87V, user error can mask a failed component. Watch out for these four bench pitfalls.
1. Testing In-Circuit (The Parallel Path Error)
Capacitors are almost always placed in parallel with resistors, ICs, or inductors for filtering and decoupling. If you measure a 100nF ceramic capacitor while it is still soldered to the board, the multimeter will measure the combined parallel impedance of the entire local circuit. The reading will likely show as a dead short or a massively inflated capacitance value. Always lift one leg.
2. Finger Capacitance on Small Values
The human body acts as a dielectric. If you hold the metal tips of the probes with your bare fingers while testing a small ceramic or mica capacitor (e.g., 22pF or 100pF), your body will introduce 50pF to 100pF of stray capacitance into the measurement. For components under 1nF, use alligator clip leads or a dedicated component test socket to keep your hands out of the circuit.
3. Ignoring Equivalent Series Resistance (ESR)
This is the most dangerous trap in power supply repair. A standard multimeter measures capacitance by timing a charge/discharge cycle. An old, dried-out 470µF electrolytic capacitor might still read 465µF on your DMM, tricking you into thinking it is healthy. However, its internal electrolyte has evaporated, pushing its ESR from a normal 0.1Ω up to 15Ω. In a high-frequency switching supply, that 15Ω ESR will cause massive ripple voltage and overheating. A standard DMM cannot measure ESR; you need a dedicated ESR meter or an LCR meter (like the DER EE DE-5000L) to catch this failure mode.
4. Forgetting to Zero the Test Leads
Standard silicone test leads possess their own parasitic capacitance, typically between 0.2nF and 0.8nF depending on length and shielding. If you are testing high-frequency bypass capacitors (e.g., 1nF or 2.2nF), failing to short the probes and press the 'REL' button will result in a reading that is 20% to 40% higher than the actual component value, leading you to falsely reject good parts.
For deeper theory on how dielectric absorption and leakage current affect these measurements, consult the component testing primers at All About Circuits. Remember that while a multimeter is an excellent first-line diagnostic tool for shorts, opens, and gross capacitance loss, it is not a substitute for an LCR meter when diagnosing subtle dielectric degradation in precision analog or RF circuits.
