To check a capacitor with a multimeter, set your dial to the capacitance mode (⊣⊢), ensure the component is fully discharged, and connect the probes directly to the leads. A good reading falls within ±20% of the printed microfarad (µF) value. If the reading is significantly lower, the electrolyte has dried out; if it reads 'OL' (overload) or zero ohms, the capacitor is open or shorted and must be replaced.
While a basic pass/fail check is straightforward, diagnosing subtle failures like high Equivalent Series Resistance (ESR) or dielectric leakage requires a more rigorous approach. This guide provides the exact meter setups, numerical baselines, and decision frameworks used by bench technicians to confidently evaluate both PCB-mounted electrolytics and heavy-duty HVAC run capacitors.
Safety First: Discharging and CAT Ratings
Before your probes touch the terminals, you must safely bleed off stored energy. Do not short the terminals with a screwdriver; the massive current spike can vaporize the screwdriver tip, damage the capacitor's internal foil, and spray shrapnel.
- The Proper Discharge Tool: Use a 20kΩ, 5-watt wirewound resistor mounted on an insulated fiberglass stick. Clamp the resistor leads across the capacitor terminals for 10 to 15 seconds.
- Verify Dead: Switch your multimeter to DC Voltage. Place the probes across the terminals. The reading must be below 0.1V before proceeding to capacitance or resistance tests.
- Safety Category (CAT): For general DC electronics and PCB repair, a CAT II meter is sufficient. For HVAC compressors, motor run/start capacitors, and appliance mains boards, you must use a CAT III 600V meter (such as the Fluke 117 or Brymen BM235) to protect against transient voltage spikes.
Meter Setup and Probe Placement
Accurate capacitance measurements require eliminating parasitic capacitance from your test leads and ensuring proper polarity for electrolytic components.
Meter Setup Block
- Dial Position: Capacitance (⊣⊢). If your meter lacks this, use the highest Ohms (Ω) range for a basic charge/short test.
- Lead Jacks: Black lead to COM. Red lead to the V/Ω/Hz/Cap jack (check your specific meter's bezel diagram, as some dedicated capacitance jacks exist on older models).
- Range: Use Auto-range for unknown values. For known large capacitors (e.g., 1000µF), manually set the range to the 2000µF or 20mF setting to prevent the meter from 'hunting' through lower ranges, which can take up to 30 seconds to stabilize.
- Zeroing: Short the probe tips together and press the REL (Relative) or NULL button. This subtracts the ~0.05nF to 0.1nF parasitic capacitance of the leads, which is critical when measuring sub-1µF ceramic or film capacitors.
Probe Placement
- Identify Polarity: For aluminum electrolytic capacitors, the negative (cathode) lead is marked by a contrasting stripe with minus signs on the sleeve. The positive (anode) lead is usually longer on new, unclipped components.
- Connect Probes: Place the Red probe on the Anode (+) and the Black probe on the Cathode (-). Reversing polarity on some meters will yield an 'OL' error or a negative value, even if the capacitor is healthy.
- Non-Polarized Caps: For ceramic, mica, and film capacitors, probe placement does not matter.
- Hold Steady: Maintain firm contact. Large capacitors take 5 to 15 seconds to charge from the meter's internal test voltage and display a stable final number.
Testing Capacitance: The Numerical Baseline
Capacitors degrade over time, primarily due to electrolyte evaporation. The industry-standard tolerance for a 'good' capacitor is ±20% of its labeled value. Below is the spec-sheet baseline for common component sizes.
| Labeled Value | Good Reading (±20%) | Bad Reading (Fail State) | Typical Failure Mode |
|---|---|---|---|
| 10 µF (Electrolytic) | 8.0 µF to 12.0 µF | < 8.0 µF or 'OL' | Electrolyte dry-out, internal open |
| 470 µF (Electrolytic) | 376 µF to 564 µF | < 376 µF | Dry-out, excessive ripple current damage |
| 35 µF (HVAC Run Cap) | 31.5 µF to 38.5 µF | < 31.5 µF or Short | Dielectric breakdown, oil leakage |
| 100 nF / 0.1 µF (Ceramic) | 0.08 µF to 0.12 µF | 0.00 µF or Short | Micro-cracking, solder bridge short |
Source reference: Tolerance bands and testing baselines align with Fluke's official electrical testing guidelines.
Testing Resistance and Leakage (When Capacitance Lies)
A standard multimeter capacitance test applies a low-frequency, low-voltage DC charge. This means a capacitor can read a perfect 470µF on your DMM but still fail catastrophically in a 100kHz switching power supply due to high Equivalent Series Resistance (ESR) or internal leakage. While true ESR requires a dedicated 100kHz ESR meter (like the Signstek MESR-100), your multimeter can catch massive leakage and dead shorts.
The Ohms Leakage Test
- Set your multimeter to the highest Ohms range (e.g., 20MΩ or 200MΩ).
- Connect the probes (Red to +, Black to -).
- Watch the display: A healthy capacitor will initially show a low resistance as it draws current to charge, then the resistance value will steadily climb until it reaches 'OL' (infinity).
- The Verdict: If the resistance climbs to OL, the dielectric is intact. If it stops climbing and settles at a fixed value (e.g., 150kΩ), the capacitor is leaky and must be replaced. If it stays at 0.0Ω, it is dead-shorted.
Decision Tree: Pass, Fail, or Replace?
Use this decision matrix to determine your next step. Never reinstall a component that fails any single parameter in this tree.
| Symptom / Reading | Verdict | Action & Concrete Replacement Pick |
|---|---|---|
| Reads ±20% of label; Ohms test climbs to OL. | PASS | Keep in circuit. No action required. |
| Reads >20% below labeled µF value. | FAIL (Dry) | Replace with same µF, equal/higher voltage, 105°C rated low-ESR series (e.g., Panasonic FR or Nichicon PW). |
| Reads 'OL' immediately on Capacitance mode. | FAIL (Open) | Replace. Check for physical lead separation at the rubber bung. |
| Reads 0.0Ω or near-zero on Ohms mode. | FAIL (Short) | Replace and test downstream diodes/MOSFETs; a shorted cap often takes out the rectifier. |
| Ohms test stops climbing and settles < 1MΩ. | FAIL (Leaky) | Replace. Clean the surrounding PCB area with 99% isopropyl alcohol to remove leaked electrolyte. |
| Visual: Domed top, split vent, or brown crust. | FAIL (Physical) | Replace immediately. Do not bother testing; the internal pressure relief has already triggered. |
Common Mistakes That Give Misleading Readings
Even with a high-end bench meter, operator error can make a bad capacitor look good, or a good capacitor look bad. Avoid these three bench pitfalls:
1. Testing In-Circuit (The Parallel Trap)
Never trust a capacitance reading taken while the capacitor is still soldered to the board. Other components in parallel (especially inductors, transformers, and other capacitors) will skew the meter's test current. A 10µF cap might read 450µF because the meter is actually measuring the parallel impedance of the entire power rail. Rule: Always desolder at least one leg of the capacitor, lifting it completely clear of the PCB pad before testing.
2. Touching the Metal Probe Tips
When measuring small values (under 1µF), the human body acts as an antenna and a capacitor. If your fingers touch the metal probe tips or the bare component leads, your body's parasitic capacitance (often 50pF to 200pF) will add to the reading. A 100pF ceramic capacitor might falsely read 250pF. Hold only the insulated plastic probe barrels.
3. Ignoring the Voltage and Temperature Ratings
When a capacitor fails and you go to your parts bin, matching the µF value is only half the job. If you replace a 25V, 85°C capacitor with a 16V, 85°C equivalent because 'the 5V rail only pulls 5 volts,' you are setting up a repeat failure. Voltage spikes and ripple currents will exceed the 16V dielectric limit. Always match or exceed the original voltage rating, and default to 105°C low-ESR replacements for power supplies.
