Capacitor Calculator
Decode any ceramic capacitor marking, convert values in every unit from pF to farads, and compute the charge and energy a capacitor stores — with tolerance ranges, worked formulas, a conversion chart, the full EIA code reference, and a complete toolset for series/parallel banks, plate geometry and HVAC run-capacitor testing. Below the calculator: capacitor families compared, reactance and RC timing explained, energy sizing worked out, dielectric constants tabled, and a field guide to testing and selection.
Code to capacity converter
Read the 3-digit code printed on the ceramic capacitor.
10 × 10⁴ pF = 100 000 pF
Capacity to code converter
Find the marking for a known capacitance value.
Capacitor parameters — charge & energy
Q = C × V and E = ½ × C × V², computed live.
Capacitor conversion chart
The fifteen values that cover most of the capacitors you will ever hold, in every unit and with the marking you will find on the part. 1 000 pF = 1 nF and 1 000 nF = 1 µF — the chart applies that division at every step.
| Picofarad (pF) | Nanofarad (nF) | Microfarad (µF) | Capacitor code |
|---|---|---|---|
| 10 pF | 0.01 nF | 0.00001 µF | 100 |
| 22 pF | 0.022 nF | 0.000022 µF | 220 |
| 100 pF | 0.1 nF | 0.0001 µF | 101 |
| 220 pF | 0.22 nF | 0.00022 µF | 221 |
| 470 pF | 0.47 nF | 0.00047 µF | 471 |
| 1 000 pF | 1 nF | 0.001 µF | 102 |
| 2 200 pF | 2.2 nF | 0.0022 µF | 222 |
| 10 000 pF | 10 nF | 0.01 µF | 103 |
| 22 000 pF | 22 nF | 0.022 µF | 223 |
| 47 000 pF | 47 nF | 0.047 µF | 473 |
| 100 000 pF | 100 nF | 0.1 µF | 104 |
| 220 000 pF | 220 nF | 0.22 µF | 224 |
| 1 000 000 pF | 1 000 nF | 1 µF | 105 |
Need a value that is not in the chart? The capacitance converter converts any number across all five units live — and shows the closest EIA code for it.
Capacitor formula
A capacitor stores energy in an electric field between two conductive plates. Its defining relationship links the charge it holds to the voltage across it:
Where C is capacitance in farads, Q is charge in coulombs and V is voltage in volts. One farad stores one coulomb at one volt — a huge unit, which is why practical capacitors are rated in microfarads (µF, millionths), nanofarads (nF) or picofarads (pF, trillionths). The charge section of the calculator above runs this formula in both directions: enter C and V, read the stored charge and the energy E = ½CV².
Worked example: a 100 nF (0.1 µF) capacitor charged to 12 V holds Q = 0.1×10⁻⁶ × 12 = 1.2 µC of charge and E = ½ × 0.1×10⁻⁶ × 144 = 7.2 µJ of energy. Small numbers, real physics — and the same math scales up to the multi-farad supercapacitors used for backup power.
Capacitor code system
Ceramic and film capacitors are too small to print “0.0000001 F” on, so they use a 3-digit EIA code. The first two digits are the significant figures of the value in picofarads; the third digit is the power of ten to multiply by — in practice, the number of zeros to append:
Codes with only one or two digits are read directly in picofarads — a part marked 22 is 22 pF. Values below 10 pF use decimal notation with an R: 4R7 means 4.7 pF. The multiplier digit rarely goes beyond 6 (10 µF), because larger values move to electrolytic parts printed with their value in µF.
From capacitance back to code
The reverse direction matters just as often: you have a value from a BOM or a calculation and need the code the factory would print. Round the value to two significant figures, convert to picofarads, and strip the power of ten:
Note the last row: 1.24 µF has three significant figures, but the code system only carries two, so it rounds to 1.2 µF → 125. That rounding is exactly what the converter above does when you switch it to capacity-to-code mode.
Voltage is the second number on the part: a capacitor marked “220 µF 25 V” stores 220 µF and works safely up to 25 V — exceeding the voltage rating destroys the dielectric even when the capacitance is fine. Electrolytics and film parts print both values openly; ceramics squeeze only the code plus a voltage band onto the body. Tolerance letters such as K or M ride along with the code.
Reading the capacitor color code on banded parts
Before printing became cheap enough for parts this small, ceramic disc and mica capacitors carried their value as painted color bands instead of digits. The digit colors follow the resistor sequence — black 0, brown 1, red 2, orange 3, yellow 4, green 5, blue 6, violet 7, grey 8, white 9 — but the multiplier band counts in picofarads, so brown-black-orange paints 10 × 10³ pF: 10 nF, the same value as the printed code 103.
An extra band after the multiplier marks the tolerance class, and some mica parts add a sixth band for the temperature-coefficient letter. Banded parts you meet today are usually vintage or military stock — translate the bands into the equivalent 3-digit code and enter it in the converter above to confirm the value in pF, nF and µF.
What is the capacitor tolerance code?
The letter that follows the numeric code is the capacitance tolerance — the guaranteed window the real value falls inside. Tolerances matter: a timing circuit needs a ±5% part, while a decoupling capacitor tolerates ±20% happily. The full EIA letter set:
| Letter | Tolerance | Typical use |
|---|---|---|
| B | ±0.1 pF | Precision / RF |
| C | ±0.25 pF | Precision / RF |
| D | ±0.5 pF | Precision / RF |
| F | ±1% | Timing, filters |
| G | ±2% | Timing, filters |
| J | ±5% | General purpose |
| K | ±10% | General purpose |
| M | ±20% | General purpose |
| Z | +80% / −20% | Large electrolytics |
Reading 104K: 100 nF with a ±10% tolerance, so the real capacitance is guaranteed between 90 nF and 110 nF. The converter above shows this range for any code and tolerance letter you enter.
Every capacitor family at a glance
“Capacitor” names a whole zoo of parts. The formula never changes, but the materials do — and the material decides the capacitance range you can buy, the voltage the part survives, how stable the value stays, and how long it lives. Ceramic C0G and electrolytic aluminum are barely the same species: one holds picofarads rock-steady for RF resonators, the other holds millifarads for power supplies. Before you calculate, make sure you are comparing parts from the right row:
| Family | Capacitance | Voltage | Tolerance |
|---|---|---|---|
| Ceramic C0G / NP0 Timing, resonant circuits, RF, precision filters | 1 pF – 1 µF | 16 V – 3 kV | ±5% (or better) |
| Ceramic X7R / X5R (Class II) Decoupling, bypass, coupling — never precision timing | 1 nF – 100 µF | 6.3 – 100 V | ±10–20% |
| Film (polyester / polypropylene) Audio crossovers, snubbers, motor run, sample-and-hold | 1 nF – 100 µF | 50 – 2 000 V | ±1–10% |
| Aluminum electrolytic Bulk smoothing in power supplies, energy buffering (polarized) | 0.1 – 100 000 µF | 6.3 – 630 V | ±20% (Z) |
| Solid polymer Motherboard VRM output, low-ESR replacement for electrolytics | 4.7 – 1 000 µF | 2 – 35 V | ±20% |
| Tantalum Regulator output, compact portable devices | 0.1 – 1 000 µF | 2.5 – 50 V | ±10–20% |
| Supercapacitor (EDLC) Memory backup, energy harvesting, ride-through power | 0.1 – 3 000 F | 2.3 – 2.7 V per cell | ±20% |
Stability is the hidden column: C0G ceramics and film hold their value across voltage and temperature, X7R ceramics silently shed 20–80% of their capacitance under DC bias, and electrolytics drift as their electrolyte dries. Match the family to the job — precision timing gets a stable dielectric, bulk smoothing gets cheap farads.
Capacitive reactance: why capacitors block DC and pass AC
Fill and drain a capacitor repeatedly and it never gets in the way much — change the voltage slowly and it behaves like an open circuit. That frequency-dependent opposition is called capacitive reactance, and it is the single most useful number for coupling, filtering and decoupling design:
Where f is frequency in hertz and C capacitance in farads; the result is in ohms. A 100 nF capacitor — the value sitting next to almost every IC — reads across the whole spectrum like this:
| Frequency | Where you meet it | X꜀ of 100 nF |
|---|---|---|
| 10 Hz | Audio sub-bass | 159 kΩ |
| 50 Hz | Mains (EU/CN) | 31.8 kΩ |
| 60 Hz | Mains (US) | 26.5 kΩ |
| 1 kHz | Audio mid-band | 1.59 kΩ |
| 100 kHz | Switching converters | 15.9 Ω |
| 1 MHz | RF / fast decoupling | 1.59 Ω |
Two practical consequences fall straight out of that table. First, a coupling capacitor is sized so its reactance is small compared with the load at the lowest frequency you care about — 100 nF into a 10 kΩ audio input passes bass almost untouched. Second, X-rated mains capacitors exploit the same math deliberately: at 50 Hz a capacitor draws a small, safe, phase-shifted current, which is how capacitor-powered LED bulbs and indicator lamps drop the mains without a transformer.
The fine print: real impedance is |Z| = √(ESR² + X꜀²), and every real part also carries a little parasitic inductance (ESL). Below the self-resonant frequency the capacitor behaves like a capacitor; above it, the inductance wins and impedance climbs again. That is why wideband decoupling stacks several values in parallel instead of betting on one big capacitor.
RC charging and the time constant τ = RC
Connect a capacitor to a supply through a resistor and the voltage across it climbs along an exponential curve — fast at first, then slower, never quite reaching the rail. One number describes the whole curve: the time constant τ (tau), which is just resistance times capacitance:
| Elapsed | Charged | Rule of thumb |
|---|---|---|
| 1 τ | 63.2% | One time constant — the defining landmark |
| 2 τ | 86.5% | Most meter tests land here |
| 3 τ | 95.0% | Fast approximations usually stop here |
| 4 τ | 98.2% | Functionally full for slow circuits |
| 5 τ | 99.3% | The engineering definition of 'charged' |
Worked example: 100 µF charging through 1 kΩ from a 12 V supply gives τ = 1 000 × 0.0001 = 0.1 s. After one time constant (100 ms) it sits at 12 × 0.632 = 7.58 V; after half a second (5τ) it is within 0.7% of full. The discharge curve is the mirror image — 36.8% remaining after 1τ, effectively empty after 5τ — which is why the same R × C product sets the blink rate of LEDs, the delay on reset circuits, and the soft-start ramp on motor drives.
Sizing a delay: work backwards from the wait you want. For a 2-second delay to the 95% point (3τ), pick C = 100 µF and R = 2 ÷ (3 × 0.0001) ≈ 6.8 kΩ. One multiplication, one division — no simulation required.
Sizing capacitors for energy and hold-up
The charge section of the calculator above computes stored energy directly, but design problems usually run the other way: you know how much energy the load needs and must find the capacitor. The usable energy between a charged voltage V₁ and the lowest voltage your load tolerates V₂ is:
Worked example — ride-through sizing: a 48 V industrial controller drawing 100 W must survive a 16 ms mains dropout. Energy needed: E = 100 × 0.016 = 1.6 J. If the bulk capacitor charges to 54 V and the converter stops working at 42 V, then C = 2 × 1.6 / (54² − 42²) = 3.2 / 1 152 ≈ 2 800 µF. Real designs install 4 000 µF or more — tolerances, aging and margin all eat into the number.
The same equation in reverse explains supercapacitor sizing. A 10 F supercapacitor charged to 2.7 V holds E = ½ × 10 × 2.7² ≈ 36 J — enough to run a small radio module for minutes, which is why EDLCs own the memory-backup and energy-harvesting niches. Their low 2.7 V cell rating means series strings need balancing resistors, the same rule high-voltage electrolytic banks follow.
Constant-current shortcut: when the load draws a steady current instead of constant power, the squared voltages collapse to t = C (V₁ − V₂) / I. A 1 F capacitor feeding 50 mA from 5 V down to 3 V lasts t = 1 × 2 / 0.05 = 40 s. For constant-power loads (most DC-DC converters), stick with the energy equation — current rises as voltage falls.
Series and parallel banks: the rules behind the calculator
Parallel capacitors share the same voltage and add their capacitance — more stored energy, lower combined ESR, more ripple current shared around. Series capacitors carry the same current and divide the voltage, combining by reciprocals — a bank that always measures less than its smallest member:
Series strings divide voltage — unevenly
Charge is shared, so voltage lands inversely on capacitance: V₁ = V_total × C₂ / (C₁ + C₂). Put a 10 µF and a 100 µF capacitor in series across 900 V and the little one takes 900 × 100 / 110 ≈ 818 V while the big one idles at 82 V — instantly over a 450 V rating. That asymmetry is why high-voltage banks string matched capacitors and hang an equalizing resistor (typically 100 kΩ–1 MΩ) across each part to swamp leakage differences.
The classic trick: two 100 µF / 450 V electrolytics in series yield 50 µF at 900 V — the standard way to build bulk capacitance on high-voltage buses. In the other direction, paralleling four 47 µF parts gives 188 µF and quarters the ESR, which is exactly how low-impedance DC-link banks are assembled.
The series & parallel calculator runs both rules for up to 12 capacitors with mixed units and shows the voltage-sharing note alongside the total.
Series, parallel, or series-parallel: picking the arrangement
Pure parallel banks buy capacitance; pure series strings buy voltage. When a design needs both at once — say a DC link that wants hundreds of µF across an 800 V bus — the two rules stack into a series-parallel matrix: identical capacitors wired into series strings, with the strings then paralleled. Four 100 µF / 450 V parts in a 2×2 grid give 100 µF at 900 V, and doubling the grid doubles the capacitance again. Keep every position on the same part number, fit balancing resistors across each series string, and sanity-check the total in the series-parallel capacitor calculator before you commit the layout.
Testing a run capacitor under load: the 2652 formula
Bench meters test a naked capacitor at low voltage, but HVAC capacitors fail in ways that only show up at real voltage, real temperature and real current. Technicians therefore measure the part while the system runs — clamp the ammeter on the start-winding lead between capacitor and motor, measure the voltage across the capacitor terminals, and let the motor itself be the test instrument:
50 Hz systems: µF = (3183 × Amps) ÷ Volts
Worked example: the clamp reads 1.5 A, the meter across the capacitor reads 370 V. µF = 2652 × 1.5 ÷ 370 ≈ 10.7 µF. Against a rated 10 µF ±6% part that is +7.5% — inside the “marginal” band. Grading follows the industry rule: within ±5% of rated is good, ±5–10% is marginal (replace before summer peak), beyond ±10% is replace now.
HVAC run capacitors are typically rated 370–440 V (start capacitors up to 600 V), and the motor's start winding depends on the µF value: replace with the same rating, never a random larger value. The under-load calculator automates the formula, switches between 60/50 Hz and grades your reading against the rated value instantly.
Live-voltage warning: terminals sit above 300 V while the system runs, and stored charge persists after power-off. Discharge through a 20 kΩ 5 W resistor before touching, never short the terminals, and keep one hand in your pocket. This test is for trained HVAC professionals.
Dielectric constants: the material multiplier
Three things set a capacitor's value from pure geometry: bigger plate area raises capacitance, a thinner gap raises it further, and the insulating material between the plates multiplies the whole thing by its dielectric constant K. That multiplier spans five orders of magnitude across real materials — which is why a barium-titanate MLCC fits 10 µF into a grain of rice while an air-gap tuning capacitor strains to reach 500 pF:
Working in imperial units? The same law becomes C (pF) = K × 0.2248 × A(in²) / d(in) — handy for PCB etched RF capacitors, where a 1 cm² plate pair 1 mm apart in air is only about 0.088 pF. High-K ceramics are wonderful for density but carry the Class II trade-offs above: drift, aging and DC-bias droop. Water's high K is a textbook trap — it is polar, leaky and useless as a dielectric.
The parallel plate calculator runs the formula with 25 material presets, accepts mm²/µm or sq in/mil inputs and warns when fringing fields start to skew the result.
How to choose the right capacitor
Calculated value in hand, the datasheet still asks six questions before the part is real. Answer them in order and the selection writes itself:
Capacitance & tolerance
Derive the value from the formula first, then pick the closest EIA series part. Timing and filter circuits need ±5% (J) or better; bulk smoothing and decoupling happily accept ±20% (M).
Voltage rating — derate it
Run electrolytics at 50–70% of their rating and tantalums below 50%: a 16 V rail wants a 25 V electrolytic or 35 V tantalum. Higher voltage ratings are always safe, just bigger and pricier.
Temperature coefficient
C0G/NP0 barely moves — use it for timing and resonant parts. X7R handles −55 to +125 °C but drifts; Y5V swings −80% and should be avoided where the value matters.
ESR & ripple current
Heat inside the part is P = I²RMS × ESR. Switching supplies need low-ESR polymers or ceramics; an old electrolytic whose ESR has doubled is the classic cause of bulging and boot glitches.
Frequency behaviour
Every capacitor goes self-resonant where its parasitic inductance cancels X꜀. Wideband decoupling mixes values — 100 µF bulk + 10 µF ceramic + 100 nF at the IC pin — each covering its own band.
Lifetime & environment
Electrolytic life roughly doubles per 10 °C cooler: a 105 °C-rated cap run at 65 °C lasts about 8× longer. Cold raises ESR 5–15×, which is why outdoor electronics fail on winter starts.
The order matters: voltage derating and dielectric stability eliminate far more candidates than the capacitance value itself. A 10 µF timing capacitor exists in exactly one family (C0G is too small, so film it is), while a 10 µF decoupling capacitor is any cheap X7R on the shelf.
Making capacitors last — and staying safe around them
Most capacitor failures are slow recipes rather than accidents: heat, voltage stress and ripple current add up over years. The countermeasures are just as dull and just as effective:
Discharge before touching. Bleed stored charge through a 20 kΩ 5 W resistor — never a screwdriver. Switch-mode power supply bulk caps hold lethal voltage for minutes after the plug is pulled.
Keep them cool. Every 10 °C rise halves an electrolytic's life. Distance from heat sources, airflow, and generous voltage derating all buy years; a 105 °C-rated part run at 65 °C lasts roughly 8× longer than at 105 °C.
Respect polarity. Reverse-biased electrolytics and tantalums vent, catch fire or fail as a short. Tantalums additionally want their working voltage halved — they forgive almost nothing.
Watch the DC bias on Class II ceramics. X7R/X5R parts can shed 60–80% of their capacitance at rated voltage. Check the manufacturer's bias curves or oversize the nominal value before relying on the nameplate figure.
Balance series strings. Leakage varies part to part, so a series stack drifts voltage onto its weakest member. Equalizing resistors across each capacitor keep the division honest.
Re-form long-stored electrolytics. Electrolytics that sat unpowered for years draw excess leakage current when abruptly re-energized. Ramp the voltage gently or reform them at rated voltage through a current-limited supply.
Bulged, leaking or domed tops? That is an electrolytic dying of heat and age — ESR has climbed, ripple heating accelerates, and failure follows. Replace with the same capacitance, an equal-or-higher voltage rating, and a temperature rating of 105 °C rather than 85 °C where space allows.
Sixteen dedicated capacitor calculators — one for every question
Every capacitor question has one calculator that owns it. Decode the marking, convert units, combine banks of parts, compute capacitance from geometry, test an HVAC run capacitor live, calculate energy in joules, RC time constant, charge in coulombs, reactance and impedance, color code bands, or size motor start and run capacitors. Every tool runs the physics directly — no hidden constants, no black boxes.
Cap Voltage Divider
Split voltage across 2–4 series capacitors instantly: Vn = Q/Cn with equal charge Q = Ceq·Vtotal, plus per-cap bars, charge and rating warnings.
OpenPower Supply Cap
Size rectifier bulk, buck-converter output and hold-up capacitors: C = I/(f·ΔV) and C = 2Pt/(V₁²−V₂²), with a ripple and voltage-rating table.
OpenLC Resonance
Get f₀ = 1/(2π√LC) for parallel or series LC tanks, plus Q factor, bandwidth and a capacitor self-resonance (SRF) warning, from any L and C.
OpenRC Filter Cap
Solve C = 1/(2πfR) for low-pass, high-pass, band-pass and notch RC filters, with a live Bode plot, gain readout and an R×C pairs table.
OpenCoupling & Bypass
Size AC coupling, input/output, emitter and cathode bypass capacitors from impedance and cutoff: C = 1/(2πfR), with an audio stage value table.
OpenCrystal Load Cap
Calculate C1 and C2 for a crystal oscillator: C = 2CL − Cstray, with PCB stray cap, transistor input cap and an AT-cut package comparison table.
OpenMotor Capacitor Sizing
Size start and run capacitors for split-phase, cap-start, cap-run and dual-value single-phase motors with HP/voltage inputs and standard ranges.
OpenCapacitor Color Code
Decode 3-band, 4-band and 5-band capacitor color codes to pF/nF/µF with tolerance and voltage, plus a complete reference table.
OpenCapacitive Reactance
Compute Xc in Ω/kΩ/MΩ from frequency and C, with parallel-R || C and series-R + C impedance, phase angle and admittance.
OpenCapacitor Charge
Calculate stored charge Q = CV in C, mC, µC, nC, pC from capacitance and voltage, plus peak displacement current over any R.
OpenRC Time Constant
Get τ in seconds, ms, µs, ns from any R and C pair, plus the 5τ fully-settled time, charge/discharge curve values and cutoff frequency.
OpenCapacitor Energy
Compute energy stored in any capacitor from E = ½CV² with live J/mJ/µJ/nJ/Wh outputs, plus the Q²/(2C) and ½QV forms.
OpenHVAC Capacitor Under Load
Test an HVAC run capacitor under load: enter amps, volts and 60/50 Hz to get µF = 2652 × A ÷ V, graded good/marginal/replace against the rated value.
OpenParallel Plate Capacitor
Compute capacitance from plate area, separation and dielectric: C = K·ε₀·A/d with 25 real materials, mm²/µm to sq in/mil inputs and fringing warnings.
OpenSeries & Parallel Capacitors
Combine up to 12 capacitors in series or parallel with per-cap unit selectors. Live total capacitance, the rule behind it and voltage-sharing notes.
OpenCapacitance Converter
Convert capacitance between pF, nF, µF, mF and F live, and translate any value to its closest EIA 3-digit code with the tolerance range shown.
OpenThe full toolset with direction cards, FAQ and related guides lives on the dedicated calculators hub. Prefer step-by-step reading? The guide library covers every topic above in depth.
Capacitor FAQ
What does the capacitor code 104 mean?▼
The first two digits are the value in picofarads and the third digit is the number of zeros to append. Code 104 means 10 followed by four zeros: 100 000 pF, which equals 100 nF or 0.1 µF — the most common decoupling capacitor value in electronics. Enter 104 in the converter above to see the tolerance range too.
How do I convert pF to nF and µF?▼
Divide by 1 000 at each step: 1 000 pF = 1 nF, and 1 000 nF = 1 µF. So a 22 000 pF capacitor is 22 nF or 0.022 µF. The conversion chart above lists the fifteen most common values, and the dedicated capacitance converter tool accepts any value in any unit and displays it in all five at once.
What is the formula for a capacitor?▼
The defining relationship is Q = C × V: the stored charge (Q, in coulombs) equals the capacitance (C, in farads) times the voltage (V, in volts). Rearranged, C = Q / V — one farad is the capacitance that stores one coulomb at one volt. The energy stored is E = ½ × C × V², which the parameters section above computes alongside the charge.
What do the tolerance letters J, K and M mean on a capacitor?▼
The letter after the numeric code is the EIA tolerance class: J = ±5%, K = ±10%, M = ±20%. Precision codes exist too — B = ±0.1 pF, C = ±0.25 pF, D = ±0.5 pF, F = ±1%, G = ±2% — and Z marks an asymmetric +80%/−20% part typical of large electrolytics. A 104K capacitor is therefore 100 nF with a guaranteed range of 90–110 nF.
Why do capacitors use codes instead of printing the value?▼
Ceramic capacitors are physically tiny, and '0.0000001 F' does not fit on a 2 mm package. The 3-digit code compresses any standard value into three characters: two significant digits plus a power-of-ten multiplier. Values below 10 pF use a decimal R notation instead (4R7 = 4.7 pF), and 1–2 digit codes are read directly in pF.
How is capacitance measured in a circuit?▼
Out of circuit, a multimeter with a capacitance mode charges the capacitor through a known current and times the voltage ramp. In circuit — for HVAC run capacitors — technicians measure the current through the start winding and the voltage across the capacitor while the motor runs, then compute µF = 2652 × amps ÷ volts at 60 Hz. The capacitor under load calculator on this site runs that formula for you and grades the reading against the rated value, and the guides walk through the method step by step.
How do I calculate capacitance from physical dimensions?▼
For a parallel-plate capacitor, C = K × ε₀ × A / d, where K is the dielectric constant of the material between the plates, ε₀ = 8.854 × 10⁻¹² F/m, A is the overlapping plate area and d the separation. A 1 cm² plate pair 1 mm apart in air is only about 0.088 pF — which is why real capacitors use huge surface areas rolled into tiny cans. The parallel plate calculator on this site runs the formula with 25 dielectric presets, from air and FR-4 to barium titanate.
What happens when I connect capacitors in series or parallel?▼
Capacitors in parallel add directly: C_total = C₁ + C₂ + C₃… — the same rule as resistors in series. Capacitors in series combine by reciprocals: 1/C_total = 1/C₁ + 1/C₂ + 1/C₃…, the mirror of the resistor rule, and the total is always smaller than the smallest capacitor. Series strings also divide voltage across the capacitors, which is how high-voltage banks are built. The series & parallel calculator handles both rules for up to 12 capacitors with mixed units, and the guide library shows the worked examples you can check by hand.
Can I replace a capacitor with a higher µF value?▼
For HVAC run capacitors, stay within about ±5–10% of the rated µF — the motor's start winding current depends on it, and an oversized capacitor overheats the winding while an undersized one loses torque. The voltage rating, however, should be equal or higher than the original; a higher voltage rating is always safe. For decoupling and timing circuits, follow the datasheet — timing values shift proportionally with capacitance.
Why does my multimeter read a different value than the label?▼
Several honest reasons: the meter's tolerance is often ±2–5% itself, many capacitors are only guaranteed to ±10% or ±20%, and capacitance drifts with age and temperature. Electrolytics typically lose 10–20% of capacitance over their life. That is exactly why the under-load test method exists — it measures the capacitor while it does real work, at real temperature and real voltage.
Is it safe to test a capacitor with a multimeter?▼
In-circuit HVAC testing is live-voltage work: terminals can sit above 300 V even when the unit runs on 240 V mains, and stored charge persists after power-off. Discharge the capacitor through a 20 kΩ 5 W resistor before touching it, never short the terminals directly, and use one hand when probing. If you are not trained for live electrical work, the safer route is a bench test with the capacitor removed.
What is the difference between µF and uF and mF?▼
They are all the same unit spelled differently: µF (microfarad) is often typed uF because µ is hard to produce on a keyboard. Watch out for mF, which formally means millifarad (1 000 µF) but is occasionally misused for microfarad on old or cheap markings — when the context is motor or HVAC capacitors, an 'mF' label almost always means µF. Our converter treats every unit formally: pF, nF, µF, mF and F.
What is capacitive reactance and how do I calculate it?▼
Capacitive reactance is the opposition a capacitor shows to AC, and it falls as frequency rises: X꜀ = 1 / (2πfC). A 100 nF capacitor looks like 31.8 kΩ on a 50 Hz mains line but only 1.59 Ω at 1 MHz. That falling impedance is why capacitors block DC while passing AC signals, and why they make effective noise shunts. Real parts also carry a little series resistance (ESR) and inductance (ESL), so total impedance is |Z| = √(ESR² + X꜀²) until self-resonance, above which the inductance takes over.
How long does a capacitor take to charge?▼
Charging is exponential, not linear: V(t) = V₀ × (1 − e^(−t/τ)), where τ = R × C is the time constant in seconds. The capacitor reaches 63.2% of the supply after 1τ, 86.5% after 2τ, and 99.3% after 5τ — engineers call 5τ 'fully charged'. A 100 µF capacitor fed through a 1 kΩ resistor has τ = 0.1 s, so it is effectively full in half a second. Multiply R by C once and you have the whole curve.
Why does my X7R capacitor measure far below its rated value?▼
Class II ceramics (X7R, X5R, Y5V) lose capacitance under DC bias — the ferroelectric domains in barium-titanate-based dielectric stop polarizing as the electric field strengthens. A 100 µF X7R rated at 50 V can deliver only 30–50 µF near rated voltage. Capacitance meters test at nearly 0 V, so they read the nameplate value. Check the manufacturer's capacitance-vs-voltage curve, use C0G/NP0 for voltage-stable timing, or simply size up the nominal value to keep the effective capacitance you need.
How do I size a capacitor for energy storage or hold-up?▼
Use E = ½C(V₁² − V₂²), where V₁ is the charged voltage and V₂ the lowest voltage your load tolerates. Example: a 48 V, 100 W controller that must ride through a 16 ms dropout needs E = 1.6 J; from a 54 V charge down to 42 V, C = 2 × 1.6 / (54² − 42²) ≈ 2 800 µF — install 4 000 µF or more for margin. For constant-current loads the simpler t = C(V₁ − V₂)/I applies, and supercapacitors take over where electrolytics run out of farads.
How does temperature affect capacitors?▼
Three ways. Heat shortens life — every 10 °C rise roughly halves an electrolytic's lifetime, which is why a 105 °C-rated part run cool lasts about eight times longer. Cold raises ESR dramatically (5–15× at −40 °C for electrolytics), which can trip a converter on ripple. And Class II ceramics drift strongly across their range while losing extra capacitance under DC bias. Precision timing across temperatures is what C0G/NP0 ceramic and film parts are for.
What is ESR and why does it matter?▼
Equivalent series resistance is the tiny resistance inside the capacitor — leads, plates and dielectric losses lumped into one number. Every amp of ripple current heats the part by P = I²RMS × ESR, so a cap handling 10 A RMS across 50 mΩ dissipates 5 W of heat. High ESR causes voltage ripple, self-heating, and a death spiral in electrolytics where heat evaporates electrolyte, raising ESR further. Power and switching applications always specify low-ESR parts — polymer, ceramic or film.
How do I read a capacitor color code?▼
Read the bands left to right as if on a resistor, but treat the multiplier as picofarads: the first two bands are significant digits, the third multiplies by ten to that power, and an optional fourth band gives tolerance. Brown-red-orange, for example, decodes to 12 × 10³ pF = 12 nF — the same value a printed 123 code would mark. Some vintage mica parts use color dots instead of bands with identical meanings. Decode the bands into the equivalent 3-digit code and enter it in the converter above to see the value in pF, nF and µF with its tolerance range.
Is this capacitor calculator app free to use?▼
Yes — every tool on the site is a free web app that runs entirely in your browser: nothing to install, no account, no paywall. Open it on a phone, tablet or desktop, bookmark it or add it to your home screen, and it behaves like a native capacitor calculator app. The code converter, series and parallel bank tool, parallel-plate calculator and HVAC under-load tester all compute locally on your device, so results update instantly as you type.
What is a series-parallel capacitor bank and when do I need one?▼
It stacks both connection rules: capacitors are wired in series to raise the voltage rating, and identical series strings are then wired in parallel to build the capacitance back up. Example: six 22 µF / 250 V film capacitors arranged as two strings of three — each string measures about 7.3 µF and stands off 750 V, and paralleling the two strings gives roughly 14.7 µF at 750 V. Use matched parts from the same batch, fit a balancing resistor across every string, and model the bank in the series & parallel calculator before building it.
Decode your capacitor now
Read the three digits off the part, type them into the converter, and get the value in every unit with the tolerance range — free, instant, and entirely in your browser.
Open the Calculator