1. A tiny amount of current flowing into "B" allows a large amount to flow from "C" to "E" so we have an "amplification effect". We can control a BIG flow of current with a SMALL flow of current. If we continually change the small amount of water flowing into "B" then we cause corresponding changes in the LARGE amount of water flowing from "C" to "E".
For example, if we measure the current flow in gallons/minute: Suppose 1 gallon/minute flowing into "B" allows 100 gallons/minute to flow from "C" to "E" then we can say that the transistor has a "gain" or "amplification" factor of 100 times. In a real transistor we measure current in thousandths of an Ampere or "milliamps". So 1mA flowing into "B" would allow 100mA to flow from "C" to "E".
2. The amount of current that can flow from "C" to "E" is limited by the "pipe diameter". So, no matter how much current we push into "B", there will be a point beyond which we can't get any more current flow from "C" to "E". The only way to solve this problem is to use a larger transistor. A "power transistor".
3. The transistor can be used to switch the current flow on and off. If we put sufficient current into "B" the transistor will allow the maximum amount of current to flow from "C" to "E". The transistor is switched fully "on".
If the current into "B" is reduced to the point where it can no longer lift the black plunger thing, the transistor will be "off". Only the small "leakage" current from "B" will be flowing. To turn it fully off, we must stop all current flowing into "B".
In a real transistor, any restriction to the current flow causes heat to be produced. This happens with air or water in other things: for example, your bicycle pump becomes hot near the valve when you pump air through it. A transistor must be kept cool or it will melt. It runs coolest when it is fully OFF and fully ON. When it is fully ON there is very little restriction so, even though a lot of current is flowing, only a small amount of heat is produced. When it is fully OFF, provided we can stop the base leakage, then NO heat is produced. If a transistor is half on then quite a lot of current is flowing through a restricted gap and heat is produced. To help get rid of this heat, the transistor might be clamped to a metal plate which draws the heat away and radiates it to the air. Such a plate is called a "heat sink". It often has fins to increase its surface area and, thereby, improve its efficiency.
This is the symbol used to represent an "NPN" transistor.
You can distinguish this from a "PNP" transistor (right) by the arrow which indicates current flow direction.
Electronic Repair Information: How You Can Test Electronic Components Like A Professional! Click Here!
The difference between PNP and NPN transistors is that NPN use electrons as carriers of current and PNP use a lack of electrons (known as "holes"). Basically, nothing moves very far at a time. One atom simply robs an electron from an adjacent atom so you get the impression of "flow". It's a bit like "light pipes". In the case of "N" material, there are lots of spare electrons. In the case of "P" there aren't. In fact "P" is gasping for electrons. Clear as mud isn't it?
OK, bear in mind that the Base is only a few atoms in thickness - almost a membrane - so any electrons allowed into the base "membrane" act as a catalyst to allow other electrons to break through from emitter to collector.
Imagine a pool of water near the edge of a table. It rests there with surface tension holding it in place. Now put one tiny drop of water on the table edge and let it touch the pool of water. Suddenly, the pool drains onto the floor as gravity takes over! Your tiny drop provided the catalyst to get it moving. So the base electrons do a similar job for the "pool" of electrons in the emitter - helped by the "gravity suction" of the power supply voltage on the collector.
A transistor doesn't "increase" current. It simply allows power supply current to pass from collector to emitter* - the actual amount depends on the (small) current allowed to flow into its base. The more electrons you allow into the base, the more (x 100) that flow from collector to emitter . I put "x 100" because that is the typical gain (amplification factor) of a transistor. For example, one electron put into the base could allow 100 to escape from collector to emitter.
The best way to understand this is to get your soldering iron and start building!
Introduction To Basic Electronics Hands-On Mini Course Click Here!
* The purist might argue that current flows from emitter to collector - dependent on whether we are discussing electron flow or "hole" flow. I don't want to get involved in the physics of current flow. You don't need to know this to design a circuit.
* This discussion relates to Bipolar transistors. Other types of transistor such as "FETs" (Field Effect Transistors) are in common use and work in a slightly different way in that the voltage applied to the "gate" terminal controls the current flowing from "cathode" terminal to "anode" terminal. In effect, a FET is simply a semiconducting (one-way) resistor whose value is controlled by the voltage applied to its "gate".
* OK, having told you all that, I now have to point out that the above description is basically WRONG! What I've described is based on what is called "the beta model" of a transistor. A transistor actually relies on base voltage input - the current input is incidental. If you are at college, your teacher will explain this fully with lots of mathematical equations that will let you design anything at all. However, my description will let you design simple circuits with a minimum of effort and they will almost certainly work. Unless you are going to take up design as a profession, this is all you need to know.