The transistor as current source
In part 3 of this tutorial we looked briefly at several current sources, which contain integrated circuits. In this part, we will look in more detail at a simple current source built from a transistor, a zener diode and two resistors.
The basic principle for this circuit is to start with a fixed reference voltage. This voltage is then translated to a fixed current.
Simple Voltage Stabilisation
Let's start with that fixed voltage. We don't need a high precision stabilisation for an LED. So we can use a zener diode for this. Zener diodes are normal diodes, really. They behave like any normal silicon diode. And just like any standard diode, when a voltage is applied in reverse direction, no current flows.
But like any other silicon diode, there comes a point when the reverse voltage gets to high. The diode then "breaks down". It starts to conduct. What is different with zener diode is that that the break down occurs at a defined, relatively low voltage.
Here is a circuit, which shows the behaviour of a zener diode with a break down voltage of 3.3 volts.
There is a resistor R1 to limit the current through the zener diode. As explained, the diode is connected "in reverse". If we apply voltages between 2 and 15 volts, the voltage over the zener diode remains more or less the same - about 3.2 volts.
Adding a Transistor
Now for the interesting bit: the constant voltage is converted to a constant current. We add a transistor.
The transistor has an interesting property: Over its base emitter-diode (indicated by the little arrow in its symbol), the voltage drops by about 0.6 volts - just like with every normal silicon diode. But in order to get that voltage drop, we need some current to flow. So we add a resistor.
Now what is the voltage over the new resistor R2? Let's start again with the zener diode. The voltage there was 3.2 volts (at the label "Uref"). Over the base-emitter diode of the transistor we have another 0.6 volts, leaving us with 2.4 volts.
Now we have a resistor with a fixed voltage over it and that means we know the current through it: I = U / R. In our case it is 2.4 volts / 120 Ohms = 20 mA.
The Full Monty
What if there was additional current coming through the collector-emitter path of the transistor? Well, R2 would still follow Ohm's law and the sum of currents would remain constant. Where the current rising, the voltage over the base-emitter diode would decrease. That would reduce the current into the base of the transistor. Since the current through the collector-emitter path of the transistor is controlled by the base current, that current will also be decreased.
So we now have a current source that can drive a LED and keeps the current constant over a wide range of input voltages. See the following diagram:
This circuit needs only four easily available components. The current can be controlled by selecting R2. Take care not to operate that resistor beyond its rated power (P = U x I).
Select R2 with the following formula: R2 = (Uzener - 0,6 V) / I, where Uzener is the break down voltage of the zener diode and I is the desired current.
Now all that is left is a look at how microcontrollers can control LEDs and why this can be extremely simple. Read all about it in part 5.