Sizing capacitors for smoothing signals

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 3K views
sherrellbc
Messages
83
Reaction score
0
In general, by what considerations guide the decision to size a capactior to smooth voltage signals?

For example, if I were to use a 3.3V regulator in a circuit design, what size cap should be set in place to ensure a smooth regulated voltage? My attempt at answering my own question would bet to get an appropriate voltage rating (~10V, for this case maybe?) and, presummably, larger capacitance is better? I know that smaller capacitances will discharge very quickly if the need ever arises (i.e. fluctuations in the output voltage require cap discharge).
-What keeps a capacitor from discharging back back into the regulator?

However, at the same time, I know that discharge rates of capacitors depend largely on the load across its terminals. That being said, is a direct analysis per design in order for capacitance sizing?

What about the use of dielectric capacitors versus ceramic (non-polar) capacitors?
 
Engineering news on Phys.org
What keeps a capacitor from discharging back back into the regulator?
The regulator (or the connection between regulator and capacitor), hopefully.

and, presummably, larger capacitance is better?
Usually, yes.

what size cap should be set in place to ensure a smooth regulated voltage?
That depends on the fluctuations you expect.
Maximal current for the capacitors is also a design choice, and depends on the load.
 
mfb said:
The regulator (or the connection between regulator and capacitor), That depends on the fluctuations you expect.
Maximal current for the capacitors is also a design choice, and depends on the load.

Where would such fluctuations originate from? I was considering placing smoothing capacitors across the terminals of the a voltage regulator to ensure output is kept very clean and any high frequency fluctuations (inherent circuit noise?) is kept filtered out.

For example, say mains voltage is stepped down to around 12V, which is then in turn clamped to 5V and again to 3.3V for various applications. What order of fluctuations are to be expected in a configuration like this? If each clamping stage is filtered by such a capacitor that is the purpose of this thread.

What about if the initial supply were to come from a battery. Are capactiors required in this case, or is the voltage supplied by batteries very clean to begin with?
 
sherrellbc said:
Where would such fluctuations originate from?
That depends on your setup.
A variable load, a variable input voltage for the regulator (no regulator is perfect...), induced electric fields in the circuit, noise in the circuit components, ...

For example, say mains voltage is stepped down to around 12V, which is then in turn clamped to 5V and again to 3.3V for various applications. What order of fluctuations are to be expected in a configuration like this? If each clamping stage is filtered by such a capacitor that is the purpose of this thread.
It is impossible to answer it without analyzing the whole circuit.

What about if the initial supply were to come from a battery. Are capactiors required in this case, or is the voltage supplied by batteries very clean to begin with?
I would not expect significant variations from the source, but what about your load?