The bipolar transistor is a three terminal component made from three layers of alternating semiconductor material. The layers form two PN junctions and in some ways work like two diodes connected in series with pointing away from the point where they connect, this point being the third terminal of the transistor.
The terminals in the transistor are called Collector, Base and Emitter.
Although similar in construction, the particular way in which the layers in a transistor are arranged give it some interesting properties.
The bipolar transistor functions as what is called a current controlled current regulator. When a small current flows through the forward biased base-emitter junction, a large current is also allowed to flow from collector to emitter.
This seems counter-intuitive with the way you learned about diodes; a reverse biased diode should not allow current through it. This emergent property of the transistor is what gives it most of its uses, since a little input current at the base generates a large output current through the transistor, in essence it amplifies the current using an external power source.
There are two types of bipolar transistors, PNP and NPN, named after the combination of material types that make them up, they differ in polarity of voltages applied. The explanations given are for NPN transistors, use the reverse polarities for PNP transistors.
Transistor Operation
With collector connected to a more positive voltage than emitter and no current flowing into the base of the transistor, no current flows from collector to emitter, and the transistor is said to be in cutoff.
When the voltage applied to the base is slightly higher than the junction voltage of the base emitter junction, some current starts to flow from base to emitter, as well as from collector to emitter. The current that flows through the collector is roughly the same as the current going into the base times the current gain of the transistor (Typically written as hfe or B [beta]).
Consider a transistor's collector connected directly to the voltage source and the emitter connected to ground, by kirchoff's second law you can see that the voltage the transistor gets will always be equal to the supplied voltage.
In cutoff, no current flows through the transistor, so the voltage source "sees" an infinite resistance, that is equivalent to an open switch.
With only a voltage of a little over the junction voltage of base emitter, let's say enough for 1mA to flow and a beta of 100, we get a collector current of roughly 100mA. So in theory, if we supply 100mA we should get a collector current of roughly 10A right?
In theory, yes, that should be possible. In practice however, current flowing through any conductor generates heat, and with small transistors even a current of less than 500mA could be enough to create enough heat to burn and destroy the transistor. There's also the fact that any voltage source has a limit on the amount of current it can supply.
Let's now consider another similar circuit, now instead of being connected directly to voltage ground, we use a resistor of 90 ohms as a load the collector. Let's also use the base current again from the previous example, 1mA and a voltage source of 10v.
The theoretical collector current should be Ic = B Ib = 100 1mA = 100mA.
With 100mA flowing through it, the resistor gets an induced voltage of 9v, close to our voltage supply, with 1v across the transistor, we account for all 10v of supply. But what happens if we increase the base voltage to 2mA?
In theory, collector current should be Ic = 100 2mA = 200mA.
With 200mA flowing through it, the resistor should get an induced voltage of 18v, which is clearly higher than our supply voltage. To compensate, the transistor should have to be 8v lower than ground potential, which it simply cannot do.
What happens in this situation is that the transistor will try to keep the voltage across it as close to ground as it can to accommodate the current that should be flowing through its collector. The base current at which the transistor cannot lower the voltage across it , in other words the transistor is fully on, is called the saturation current, and the state itself called saturation.
