What is the need for adding a current source (\$I_\text{E}\$) at the output of an emitter follower topology at the emitter leg of the transistor?

enter image description here

  • 1
    \$\begingroup\$ Where did the picture come from? \$\endgroup\$
    – Andy aka
    Jan 20, 2017 at 18:11
  • \$\begingroup\$ Wikipedia, en.wikipedia.org/wiki/Common_collector#/media/… \$\endgroup\$
    – G.Hajj
    Jan 20, 2017 at 18:23
  • 1
    \$\begingroup\$ The current source reduces 2HD caused by the transistor being better able to source current via the collector than sink it via RE. \$\endgroup\$
    – user207421
    Jan 20, 2017 at 21:19

3 Answers 3


When biasing a transistor, you want to Q point to be stable. Since \$\beta\$ varies a lot, you want to design in such a way that changes in \$\beta\$ do not disturb the bias point.

By placing a current source there, \$I_E\$ will not depend on \$\beta\$ (and \$I_C\approx I_E\$) and that way you can keep the transistor in the active region regardless of changes in \$\beta\$.


what is the need for adding a current source (IE) at the output of an emitter follower topology at the emitter leg of the transistor?

  1. increasing the gain closer to 1x;
  2. setting the bias point for the BJT;
  3. maintaining better linearity towards the ground;
  4. lower production cost in a chip - a CCS is much easier to make than a resistor on a chip.
  5. ...

Basically, very limited benefits, so they are not as widely used.


The wiki page (on common collector topology) that the picture in the question comes from says this: -

Figure 4: NPN voltage follower with current source biasing suitable for integrated circuits.

In other words, inside a linear integrated circuit, current source biasing is usually preferred over the use of a resistor. In some instances this is beneficial for maintaining stability against temperature changes however, in this instance, the use of a bias current at the emitter is less advantageous. The wiki article also says this: -

sometimes an active current source is used instead of RE (Fig. 4) to improve linearity and/or efficiency1.

At the end of the quote it links to the design of a 20 watt class A amplifier but it's all rather tenuous given the type of output stage it has and the amount of negative feedback it uses: -

enter image description here

So, my personal opinion is that the use of a current source in the emitter is justified inside a chip (because they are easier to fabricate) but on an amplifier like this it might be pushing the boat out a little too far in favour of the audiofools I mean philes.

  • \$\begingroup\$ The current source makes the class A Amplifier more efficient than it would be with a resistor .The circuit as shown would be less than 25% efficient .If the current source was somehow modulated with the Audio signal then your theory max efficiency would be 50% . \$\endgroup\$
    – Autistic
    Jan 20, 2017 at 20:28
  • \$\begingroup\$ I agree but the wiki article is tenuous in its justifications. \$\endgroup\$
    – Andy aka
    Jan 20, 2017 at 21:08
  • \$\begingroup\$ that circuit has three "ccs": 1) the lower leg of the output stage; 2) the bootstrap circuit; 3) that 6.8k resistor. \$\endgroup\$
    – dannyf
    Jan 20, 2017 at 22:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.