In some circuits often can be seen in the side of the large capacitor in parallel with a small capacitance, why?
Because a large capacitor is generally relatively large size, and usually use a multi-layer winding way to create (such as electrolytic capacitors), which led to large distributed inductance capacitance is relatively large (also called equivalent series inductance, the English referred ESL). Inductive impedance of the high frequency signal is large, so large capacitance high-frequency performance is not good.
Some small-capacity capacitor is just the opposite, due to the small capacity, so the volume can be made very small (to shorten the lead is reduced ESL, because the length of wire can also be seen as an inductor), and often use plate capacitor structure , so that there is little small capacitor in the ESL, so that it has a very good high frequency performance, but because of the small capacity of the reason, the impedance of the low frequency signal is large. So, if we have to make the low-frequency, high-frequency signal can be a good pass, and then on the use of a large capacitor capacitance and a small way.
In the power supply filter circuit, generally use large-capacity electrolytic capacitor with small capacitance ceramic capacitors in parallel, in order to achieve low and high frequencies have better filtering effect. In a digital circuit, usually given in parallel with a 0.1uF on each chip supply pin capacitance to ground (this is called the decoupling capacitor capacitance, of course, can also be understood as a power supply filter capacitor. It is closer to the position of the chip as possible ), because the signal in these areas are mainly high-frequency signals, the use of smaller capacitor filter on it.