Tim Edwards continues this series for beginners in electronics with some practical applications using semiconductors

Last month we concentrated on soldering techniques and identification of resistors and capacitors. We now move on to discuss semi-conductor devices.

Semi-conductors are so called because in their pure state the semi-conductor elements, such as germanium and silicon, are mid-way in their characteristics between the conductors (eg, aluminium and copper), and insulators. Semi-conductor devices are all based on the PN junction. The P and N indicate the way in which the pure semi-conductor has been altered to give the desired effect. Putting P type and N type silicon together gives the PN junction. The basic characteristic resulting from this is that current will flow in one direction and not the other. This gives the simplest semi-conductor devices, the diodes.

Figure 1 shows the symbol for a diode. The direction of the arrowhead part of the symbol indicates the direction of conventional current flow — from positive to negative. For current flow, the anode of the device (the P side of the PN junction), is made more positive than the cathode (the N side of the junction). Polarity identification is usually indicated by a painted band at one end of the package (Fig. 1(b)). This is the cathode. Care must be taken not to be confused by some manufacturers who place a small + next to the marking band. This indicates the direction in which current flows and does not denote the anode of the device.

When using diodes, as with any other device, pay close attention to the maximum ratings given by the manufacturers. Many of the component suppliers have catalogues which give all the relevant details. For diodes, the main ratings are for PIV, Peak Inverse Voltage and current. The PIV rating refers to the reverse voltage applied. If this is exceeded the device will be destroyed. Similarly, if too much current is passed through the diode the heat generated will destroy it. The best way to avoid these problems is to 'over-design' your circuit and choose a device with ratings that give a generous margin of safety.

The most common diodes nowadays are made from silicon and only in certain circumstances are germanium diodes still used. Generally, the silicon device provides better characteristics — it turns on faster in the forward conducting mode and will have a smaller leakage current under reverse biased conditions. However they do have a greater forward voltage drop across them — the forward voltage that needs to be applied to the device before it will conduct significant current. In the silicon diode this voltage is between 0.6V to 0.7V, while the germanium diode requires a forward bias of about 0.4V. This means that germanium diodes may be more suitable in low voltage circuits.

Crossing Bridges

You will soon use the rectification properties of diodes if you decide to power your circuits from the mains. A diode circuit of one form or another will be required to convert the alternating voltage into a direct voltage.

Figure 2(a) shows the configuration of diodes to form a bridge circuit. This is the best rectification circuit to use to cope with all eventualities. With the advent of cheap, ready-made bridges, it is often simpler to buy a package than make a circuit from diodes. Fig. 2(b) shows the pin configuration of a common low voltage bridge.

Once you have a direct voltage, this must be smoothed and regulated. It is not intended to go into details of this here, but one way of regulating the output of the bridge to give a constant voltage is to use a Zener diode circuit. Fig. 3(a) shows the symbol for the Zener diode and (b) the Zener diode in use. Note in this case that the diode is used under reverse biased conditions.

The Zener diode is manufactured such that the PN junction will give a nondestructive breakdown when the reverse voltage exceeds the chosen rating. If, for example the Zener is rated at 5V6 then the reverse voltage cannot exceed 5.6V. So we have a simple voltage regulator. The series resistor is necessary to limit the maximum current that will flow through the device. Again for identification purposes the manufacturers paint a band at the cathode end of the device.

Light Matters

The final type of diode we shall look at here is the Light Emitting Diode (LED). These diodes use special semiconductors modified to give the various colours (eg gallium phosphide — GaP; gallium arsenide phosphide — GaAsP). A typical circuit used to drive a LED is shown in (c). Note that the diode is forward biased — the anode more positive than the cathode. Typical current consumption to give a reasonable light output is 10 to 40mA. The diagram also shows an optional reverse diode placed across the LED. This is only needed if there is any chance of a significant reverse bias occurring across the LED. If a reverse voltage of more than about 3V is applied to a LED, destructive breakdown occurs. With the diode placed in parallel with the LED, the reverse voltage across it cannot exceed about 0.6V since this is the forward voltage drop across a conducting silicon diode. Try it for yourself (A LED's cathode is on the flat-edged side).

Up The Junctions

A mention was made earlier of the PN junction in diodes. If another PN junction is built into the device, the result is a bipolar transistor. There are two types, PNP and NPN referring to the way round in which the junctions are formed. It is important to know which type you are using in order that the correct polarity voltage is applied to them. Fig. 4 shows the symbols for bipolar transistors. It is important to remember that this type of transistor is a current controlled device. Current flow between the emitter and base causes a very much larger current to flow between collector and emitters. Hence the reason that a transistor can be used as an amplifying device. For a NPN transistor to function, the base must be more positive than the emitter. If not, the device shuts off and exhibits a very high resistance between collector and emitter. The collector should remain more positive than the emitter for a NPN transistor. Using a PNP device is the same principle but the polarities are reversed. The collector is negative with respect to the emitter and the base is also negative to the emitter. All you need to do now is to identify the connections to the device you have chosen to use. Fig. 5 shows a few transistor package outlines, but there are many more. As mentioned previously many commercial catalogues will contain information on the devices supplied. Alternatively, a transistor selector handbook can be invaluable, particularly if you come across the more unusual types. The transistor selector will give all the parameters applying to that device (such as gain, maximum frequency of operation, maximum voltages that can be applied to the device, power dissipation capability and maximum current). You really need to be able to understand some of the terminology and do some simple calculations to avoid disappointments through destroyed components.

Table 1 shows some of the more common parameters used to describe a transistor. You may wish to let someone else do the design work and simply follow their choice of device. However it can often be very useful to know what requirements are needed in a particular situation. Then a different device can be substituted, perhaps avoiding the delay while waiting for a component supplier to send that particular device. Remember, that just because the circuit diagram states say 'BC 108' for the transistor, this does not mean that this is the only transistor that will work. In fact there are quite likely to be tens or even hundreds of different devices that could be substituted perfectly adequately.

So, what do you need to look for? Well, first look at the voltage requirements of the circuit and check this against the transistor ratings, particularly the collector-emitter rating _Vceo. Usually, if the _Vceo rating is greater than the power supply voltage this will be sufficient. Now see if the current rating of the transistor is sufficient and that the current gain is adequate, although this is not normally critical as the gain of the circuit is often set by other components. Don't forget to make a quick check on the power dissipation required. The transistor may have an ample current rating and still fall short of the power rating. Remember, that DC power is current times voltage. For example, say a transistor is to be biased such that the collector-emitter voltage is 15V and the current through it is 20mA. The power dissipated by the device is then 15Vx20mA = 300mW. So if you use a 200mW transistor then don't expect it to last more than a few seconds. Lastly, don't forget the obvious. A PNP transistor cannot be substituted for a NPN. The simple common emitter amplifier of Fig. 6 shows how the transistor is used in a circuit.