Looking at the typical application example on this datasheet you can see that for COUT they specify 2 x 10 μF capacitors to be used although the schematic only shows the symbol for a single capacitor on COUT.

MCP16301/H Typical Application Example

Why doesn't the schematic show 2x capacitors, and should they be wired in series or parallel?


4 Answers 4


In table 5.2 (page 18), they say that Cout should be 20µF.

So it is definitively 2*10µF in parallel.

Why putting 2 identical capacitors in parallel?

  1. to increase the ripple current capability (ie be able to provide more current).

  2. to reduce the ESR (internal series resistor) of the "global capacitor": this means that there will be less voltage difference between when the capacitor is charging and when it is discharging, reducing the voltage ripple.

You could of course use a single 20µF capacitor, but then you have to make sure that it has ripple current at least twice the one of a "normal" capacitor, and ESR not more than half of the one from a "normal" capacitor.

PS: putting capacitors (same value or not) in parallel is very frequent.

Putting them in series is very rare (the only thing you gain is doubling the voltage rating (provided you can somehow ensure that the voltage is equally split, you might need to add a voltage divider with the midle point connected to the middle of the capacitors), but the capacity is divided by 2.

  • \$\begingroup\$ Thank you for explaining the reasons as to why they would use them in parallel in a clear and easy to understand way. \$\endgroup\$
    – Jacob Wood
    Commented Jan 29, 2022 at 10:05
  • 3
    \$\begingroup\$ you may also find that 20uF (or 22uF) is more expensive than two 10uF parts. \$\endgroup\$ Commented Jan 29, 2022 at 11:42
  • 1
    \$\begingroup\$ Putting capacitors in series is common in certain applications. A wide-bandwidth attenuator would use a capacitive divider, for instance (as in the inputs to any oscilloscope). And of course the purpose you noted of increasing voltage rating; I somewhat frequently use two 550 volt capacitors in series (along with a resistor string voltage divider to keep them balanced) to build test fixtures that can operate at 1000 volts. And there are NPC converters, which keep two series capacitors balanced through active control to create a neutral point. \$\endgroup\$
    – Hearth
    Commented Jan 29, 2022 at 16:42
  • 9
    \$\begingroup\$ The other big reason to double them is to simplify the BOM. Using two 10uF caps means not having to add a 20uF cap reel to the pick-and-place machine if you're already using 10uF caps elsewhere on the board. \$\endgroup\$
    – J...
    Commented Jan 29, 2022 at 17:58
  • 6
    \$\begingroup\$ Sometimes you use capacitors in series in high reliability systems to guard against short-circuit failure. \$\endgroup\$
    – John Doty
    Commented Jan 29, 2022 at 22:04

It means to parallel them to achieve the required amount of capacitance.

In switching mode regulators, it is better to have multiple smaller capacitors in parallel than to have a single large capacitor.

It's not drawn in the diagrams as they simply show components and their values like a block diagram, it does not show how you must implement it, sometimes you can just use one capacitor, sometimes you need multiple, depending on your specific requirements for size, cost, BOM optimization, PCB area, etc.

  • \$\begingroup\$ Great to know why only one capacitor was shown. Thank you. \$\endgroup\$
    – Jacob Wood
    Commented Jan 29, 2022 at 10:06

One technique I have seen in the past, (though this would be a lazy way to draw it), is to have two capacitors in parallel, spaced some distance apart.

For example, assume a power conversion circuit on a PCB where the trace between the edge of the PCB and the components, (like an IC) that does the actual work is not trivially short. Though it might be a heavy trace, there might still be measurable ripple across it. Therefore a capacitor is placed at each end creating what could be called a pi-filter.

However, in this case it is likely that the underlying reason is that they already have reels of the smaller value, or the circuit is picky enough that reducing the series resistance is desirable.


Another small advantage is that twin caps in parallel will still provide some level of functionality if one fails. This might also be done with diodes.

Whether that lesser capacitance is sufficient to function as required, depends on the circuit overall.

Adding redundancy will increase the BOM and add complexity to the board in additional tracks and pads, but does offer additional protection.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.