IAN LANG ELECTRONICS
A Darlington Pair (aka Darlington Transistor) is a structure in which two n-p-n bipolar junction transistors are compounded in such a way that the second amplifies a signal that has already been amplified by the first. This gives a very high current gain and is particularly useful where very small, very weak signals are to be passed from one part of a device and must be clearly received by the next. The Darlington Pair can be made from discrete components but is now more usually found in a T092 packaging, or it can be put onto an integrated circuit. The symbols are shown below, on the left the integrated symbol and on the right the schematic for a pair built from discrete components.
Darlington was an American engineer who worked for Bell Laboratories from 1929 until 1971. His contributions included various devices for improving the accuracy of weaponry that was used in World War II and beyond and advancements in the field of network theory. His particular field was in radar, but he is best remembered for his invention of a linked system of transistors which was patented in 1953, as transistors at the time had a poor gain. This was the Darlington Pair, although experiments at the time proved at least three could be linked this way.
In both cases you will note the shared collector. This is a great aid to designers when making integrated circuits as it saves a good deal of space.
Feeding Q2 from the emitter of Q1 brings its own problems in that there are now two junctions to consider and the breakover voltage for an on bias must be twice that of a single transistor, so 2 x 0.7V or 1.4V. Only a tiny current is needed to switch on. The output can be taken from the junction of the two collectors and is hFE 1 x hFE 2 (hFE being the gain of the transistors).
This of course gives the device a very high current gain, it could be as much as 10,000. This means that Q2 has to be able to handle more current than Q1 in normal operations.
A greater problem when used as a switch is that of switching speed. Since the base current of Q2 cannot actively be inhibited by the emitter of Q1 the device can be slow to switch off. A method of alleviating this is to provide a low-impedance path to ground to discharge the base emitter junction of Q2, and a common method is to place a resistor of a value of a few hundred ohms between the base and emitter terminals of the second transistor.
Below is a typical circuit featuring a Darlington driver for passing power from a small input to a larger load requiring more:
Like all transistors the base junction can be easily destroyed by overcurrent and so a resistor is placed there to limit the incoming current. It need not be too big, perhaps 1.2k. As you can see, the load is taken at the shared collector and as such it should be noted that the output will be the inverse of the input.
Below is a clever way of using a Darlington with an emitter follower designed by Meccano:
You will notice that a diode has been placed over the terminals of the motor. The motor is an inductive device and as such produces a back emf. This can destroy the second transistor. The diode serves to short out the back emf and prevent destruction from happening.
In summary then the Darlington Pair provides a solution to the requirements of boosting current or accepting a smaller current at the input that a single transistor alone cannot achieve. This comes at the expense of the need for a greater breakover voltage and a slower switching time.
It is as can be seen a motor speed control used in some of their models. It illustrates some of the important points we have made; the TIP31is a high power transistor, and both share a collector. The reason for doing this is it gives a greater span than attaching the potentiometer directly to the motor- a small change in the input gives a large change in the output. This makes the motor react very sensitively. However it should be noted that since the transistors can never be absolutely linear, the middle range will be more subtle than the ends.