Capacitors
A capacitor is analogous to a spring. It is capable of accumulating charge,
and thus of storing energy as well. In fact, capacitors used to be called
condensers or accumulators.
Earlier we described a hydraulic spring:
In this hydraulic spring, the idea of fluid volume "accumulating" can
be seen in the figure. As the membrane distorts, and develops a pressure
across it, there is an internally a "displaced" volume of fluid (shown
dotted) which is the total accumulated flow that has gone down the pipe.The
volume of fluid being stored in the spring is equal to the volume of fluid
which has been displaced due to the stretch of the spring.
An electrical capacitor is of this kind. It consists of two metal plates
separated by a thin insulator, just as is suggested by its schematic symbol:
Charge cannot cross the insulator, just as hydraulic fluid cannot cross
the membrane. But positive charge can accumulate on one plate, exactly
balanced by negative charge on the other. This physical separation of a
nontrivial amount of charge is tolerable because the positive and negative
charges remain in close proximity to each other, and they attract so they
are reasonably happy about that. Still, as more charge gets segregated
in this way, as current flow in from the left and out on the right, an
increasing voltage develops across the capacitor. This is analogous to
a pressure developing across the distended membrane in the hydraulic spring
as fluid flow enters the left and exits the right, accumulating in the
dotted "pouch".
The constitutive law of real capacitors is not typically as linear as
that of resistors, but to the extent that it is approximately linear we
use this constitutive law:
C is the capacitance, measured in farads. A 1 farad
capacitor is a whopping big one. More typical are microfarads, 10^{6}
farad, abbreviated mF, and picofarad, 10^{12}
farad, abbreviated pF. You might expect mF to mean millifarad but it usually
doesn't. It usually means microfarad as well, written by people who don't
have Greek letters available. mmF means mF too.
nF, or nanofarad, 10^{9} farad, is somewhat unusual.
Really, it's hard to read capacitor values if you don't already have
a rough idea of the answer based on the physical size and type. "222" on
a tiny disk capacitor, for instance, is a cryptic 22 * 10^{2} pF,
encoded much the way the three color bands on a resistor are.

Capacitance is defined differently from spring constants (F = kx) , unfortunately.
Higher spring constants meant stiffer springs: more force for the same
displacement. Higher capacitances are lessstiff capacitors: less voltage
for the same charge. The word capacitance seems to suggest capaciousness:
higher values indicate more capacity for accumulating charge, without a
lot of voltage building up.
Recall, Charge is related to current just as volume is related to
hydraulic flow:
q = i
dt
i = q'

(2)

So, we can also write (1) as

or 

(3) 
Parallel and Series
Similar to what we did for resistors, we can examine the effective capacitance
of simple circuits of capacitors in series and an parallel. The same rules
outlined before for elements in series and parallel hold. Here
is an example of capacitors in parallel.
Energy
Suppose a capacitor of C farads presently contains a charge of q
coulombs. Therefore the voltage across it is V=q/C. If we charge
it some more, by injecting a current i, the power flowing into the
capacitor is P=iV, just as it would be for any component (not just
a capacitor). If we do this for a short interval of time dt (short
enough that the voltage doesn't change significantly), the energy we have
put into the capacitor is DE = P dt = V i
dt.
In the same infinitesimal moment we have increased the charge of the
capacitor from q to q+ i dt. The amount of charge we have
added is dq = i dt. Therefore the amount of energy we have added
is DE = V dq.
On a graph of the constitutive law of the capacitor, V dq has
a geometric interpretation: it is the area under the curve, from q
to q + dq. (This is so even if the constitutive law is not linear.)
In charging the capacitor from q_{0} to q, the
total energy added to the capacitor is
In the case of a linear capacitor (V = q/C), changing up all the way from
from q_{0}=0, we can perform the integral and find
E = q^{2} / 2C, or
E = CV^{2}/2

(5)

This is the energy stored in the capacitor. It can be recovered if the
capacitor is discharged into a resistor, for instance. Capacitors have
a maximum voltage rating, much like resistors have a maximum power
rating. A capacitor can be charged up to only a certain voltage before
the insulating material between the two plates begins to break down  the
maximum voltage rating of the capacitor is usually noted on the capacitor
itself.
Recognizing a capacitor
Below are pictures of some capacitors. Capacitors are typically labelled
with text which indicates their capacitance, and a voltage rating. However,
the text is often not clearly marked and can be quite cryptic, as in the
examples below. More pictures of capacitors here.
To see inside a capacitor to see what it is made of, click here.