Current Sources
There are different methods of simulating a dc current source for integrated circuit amplifier biasing. One type of current source used to provide a fixed current is the fixed bias transistor circuit. The problem with this type of current source is that it requires too many resistors to be practically implemented on IC. The resistors in the following circuits are small and easy to fabricate on IC chips. When the current source is used to replace a large resistor the Thevenin resistance of the current source is the equivalent resistance value.
A simple current source
The simple two transistor current source shown in fig. 1 is commonly used in ICs.
Fig. 1
A reference current is the input to a transistor connected as a diode. The voltage across this transistor drives the second transistor, where R_{E} = 0. Since the circuit has only one resistor, it can be easily fabricated on an IC chip.
The disadvantage of this circuit is that the reference current is approximately equal to the current source. In this circuit, Q_{2} is in linear mode, since the collector voltage (output) is higher than the base voltage. The transistor Q_{1} and Q_{2} are identical devices fabricated on the same IC chip. The emitter currents are equal since the transistors are matched and emitters and bases are in parallel. If we sum the currents of Q_{2}, we obtain.
I_{B} + I_{C} =I_{E}
So
Summing currents at the collector of Q_{1} we obtain
If β is large, the current gain is approximately unity and the current mirror has reproduced the input current. One disadvantage of this current source is that its Thevenin resistance (R_{TH}) is limited by the r o (1 / h_{oe}) of the transistor. That is
Widlar Current Source
Large resistors are often required to maintain small currents of the order of few ľA and these large resistors occupy correspondingly large areas on the IC chip. It is therefore, desirable to replace these large resistors with current sources. One such device is the Widlar current source as shown in fig. 2.
The two transistors are assumed perfectly matched. For the base circuit,
(E4)
For a forward biased baseemitter junction diode, the emitter current is given by
Since i_{E} ≈ i_{C} = I_{C} and n = 1
and (E5) 
Fig. 2 
Substituting V_{BE1} and V_{BE2} from (E5) to (E4), we get
(E6)
We have assumed that both the transistors are matched so that I_{CO}, β and V_{T} are the same for both the transistors. Thus
Hence, (E7)
where, (E8)
For design purposes, I_{C1} is usually known since it is used as the reference for all current sources on the entire chip and I_{C2} is t he desired output current. The Widlar circuit can also be used to simulate a high resistance.
Example1
Design a Widlar current source to provide a constant current source of 3 ľA with V_{CC} = 12V, R_{1} = 50 kO, β =100 and V_{BE} = 0.7V
Solution:
The circuit is given in fig.2 . Applying KVL to the Q_{1} transistor we get,
Using the equation (E7) we can calculate R_{2}
or R_{2} = 36 kΩ
GOTO >> 1  2  3  Home
