How do you calculate capacitors in parallel?

Add them: Ceq = C1 + C2 + … + Cn. Convert everything to the same unit first. There are no reciprocals — parallel capacitance is a plain sum, the opposite of the series case.

Does capacitance increase in parallel?

Yes. The total is the sum, so it is always larger than the largest capacitor in the group. Physically it is equivalent to increasing the plate area, which stores more charge at the same voltage.

What is the voltage rating of capacitors in parallel?

The lowest rating in the group. Ratings do not add in parallel because every capacitor sees the full applied voltage. A 16 V part in parallel with a 450 V part gives a 16 V bank.

Why use several small capacitors instead of one large one?

For lower ESR and better high-frequency behaviour. Parallel loss resistances combine like parallel resistors, so the bank responds faster than one large part, and a mix of sizes covers a wider frequency range. Splitting ripple current between parts also reduces the heating in each one.

Can you mix capacitor types and values in parallel?

Yes, and it is normal practice — a bulk electrolytic with ceramics beside it is the standard decoupling arrangement. Keep the polarity right on the electrolytics and remember that the lowest voltage rating governs the whole bank.

Parallel Capacitance Calculator

Capacitors in parallel add. The equivalent parallel capacitance is the plain sum of every capacitor across the same two nodes, they all see the same voltage, and the total charge stored is the sum of what each one holds.

Parallel Capacitance Formula

Capacitors in parallel are all connected across the same two nodes, so they all have the same voltage V across them. Each one then stores the charge its own capacitance implies at that voltage.

The total charge stored by the group is the sum of the individual charges, because the capacitors are independent stores fed from the same pair of nodes:

Q = Q_{1} + Q_{2} + Q_{3}

Substituting Q = C × V for each term, with the same V throughout:

C_{e q} \times V = C_{1} \times V + C_{2} \times V + C_{3} \times V

The voltage divides out, leaving the simplest formula on the site - parallel capacitance is a plain sum:

C_{e q} = C_{1} + C_{2} + \dots + C_{n}

Capacitors in parallel add, like resistors in series. No reciprocals anywhere.

Physically, putting capacitors side by side is like widening the plates: more area at the same separation stores more charge per volt. That is why the values simply add, and why the total is always larger than the largest capacitor in the group.

The voltage rating does not add

This is the trap. Capacitance adds in parallel, but the working voltage of the bank is the LOWEST rating in it, not the sum and not the average. Put a 16 V capacitor in parallel with a 450 V one and the bank is a 16 V bank — the 16 V part will fail first and, if it is an electrolytic, loudly.

Worked example

Given

C1 = 10 µF bulk
C2 = 100 nF decoupling
C3 = 100 nF decoupling

Working

Ceq = C1 + C2 + C3
Ceq = 10 µF + 0.1 µF + 0.1 µF

Answer10.2 µF

The two 100 nF parts barely change the total — and that is not why they are there. See the note on ESR below.

Why Parallel Capacitors Are Not About Capacitance

The worked example above is the standard decoupling arrangement, and it shows why the total capacitance is often the least interesting number. Adding 100 nF to 10 µF changes the total by one percent. Nobody fits those parts for the one percent.

Lowering ESR and impedance

Every real capacitor has series resistance and series inductance. In parallel those combine like resistors in parallel, so the bank has lower ESR than any single member. That is what lets a bank supply a fast current step that one large capacitor could not.

\frac{1}{E S R_{e q}} = \frac{1}{E S R_{1}} + \frac{1}{E S R_{2}} + \dots

Loss resistance falls in parallel, which is the real reason to use several parts.

Covering different frequencies

A large electrolytic has useful impedance at hundreds of hertz and is inductive by a megahertz. A small ceramic is still capacitive at tens of megahertz but holds little charge. In parallel each covers the band the other cannot, which is why decoupling is specified as a mix of values rather than one big part.

Sharing ripple current and heat

Ripple current heats a capacitor through its ESR, and heat is what kills electrolytics. Splitting the same ripple across several parallel parts divides the current, and because dissipation goes as the square of current, it divides the heating in each part by much more.

Sizing a Parallel Bank

Where the capacitance does matter, the usual question is how much is needed to hold a rail up during a load step. For a current I drawn for a time t with an allowable droop ΔV:

C \geq \frac{I \times t}{Δ V}

That gives the bulk figure the bank has to add up to. Enter the parts you have in the calculator above to check the total, or use the find-missing-value mode to work out what still has to be fitted to reach it.

For a mains-frequency reservoir after a rectifier the same idea applies with t set by the ripple period, which is 10 ms for a full-wave rectifier on 50 Hz and 8.3 ms on 60 Hz.

Practical Cautions With Parallel Capacitors

The bank voltage rating is the lowest rating present, never the sum. One under-rated part defines the whole bank.
Polarity has to match on electrolytics. One fitted backwards in a parallel bank will conduct, heat and vent.
A large parallel bank draws a large inrush current at switch-on, which can weld relay contacts and trip supplies. That is what soft-start circuits exist for.
A charged bank stays dangerous after power is removed. Capacitance in parallel means more stored energy at the same voltage.
Class 2 ceramics lose capacitance under DC bias, so a bank of X7R parts may measure well below its marked total in circuit.

Common mistakes

Using the reciprocal formula. Parallel capacitors add; it is series that uses reciprocals.
Assuming the voltage rating adds. The bank is limited by its lowest-rated member.
Mixing prefixes before adding. 10 µF + 100 nF is 10.1 µF, not 110 of anything.
Expecting the total capacitance to explain why a bank of mixed values is used. The reason is usually ESR and frequency coverage, not the capacitance sum.
Fitting an electrolytic backwards in a parallel bank, where it will conduct rather than store.
Ignoring inrush current when a large bank is switched on.

Frequently asked questions

How do you calculate capacitors in parallel?: Add them: Ceq = C1 + C2 + … + Cn. Convert everything to the same unit first. There are no reciprocals — parallel capacitance is a plain sum, the opposite of the series case.
Does capacitance increase in parallel?: Yes. The total is the sum, so it is always larger than the largest capacitor in the group. Physically it is equivalent to increasing the plate area, which stores more charge at the same voltage.
What is the voltage rating of capacitors in parallel?: The lowest rating in the group. Ratings do not add in parallel because every capacitor sees the full applied voltage. A 16 V part in parallel with a 450 V part gives a 16 V bank.
Why use several small capacitors instead of one large one?: For lower ESR and better high-frequency behaviour. Parallel loss resistances combine like parallel resistors, so the bank responds faster than one large part, and a mix of sizes covers a wider frequency range. Splitting ripple current between parts also reduces the heating in each one.
Can you mix capacitor types and values in parallel?: Yes, and it is normal practice — a bulk electrolytic with ceramics beside it is the standard decoupling arrangement. Keep the polarity right on the electrolytics and remember that the lowest voltage rating governs the whole bank.

Assumptions and limitations for Parallel Capacitance Calculator are listed on the About page. Every worked example on this site is checked against the same solver the calculator uses.