0
\$\begingroup\$

From this online lecture (page 190-20), I see that the negative feedback (Rs in common source degeneration) helps increases linearity range.

I am wondering if anybody know an explanation for this intuitively or mathematically.

Another question is what is the difference between the left and right hand pictures in the image below? For small signal, they are exactly the same.

Thank you. enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ The feedback provides a correction signal that you use to compare the actual output to the desired output. Then that error signal is used to improve the output in regard to how it tracks the input. \$\endgroup\$
    – Daniel
    Oct 16, 2016 at 5:54

1 Answer 1

5
\$\begingroup\$

Suppose you have a nonlinear transfer function \$\frac{v_O}{v_{IN}} = a(v_{IN})\$

If you add negative feedback 1/m such that:

\$ \frac{v_O}{v_{IN}} = a(v_{IN})(v_{IN} - \frac{ v_O}{m})\$

so

\$ \frac{v_O}{v_{IN}} = \frac {a(v_{IN})}{1+\frac{a(v_{IN})}{m}}\$

For \$\frac{a(v_{IN})}{m} \gg 1, \frac{v_O}{v_{IN}} \approx m \$ (linear gain of m)

For \$\frac{a(v_{IN})}{m} \ll 1, \frac{v_O}{v_{IN}} \approx a(v_{IN}) \$ (nonlinear with negligible feedback)

As you can see, the closed loop gain must be reduced significantly from the open-loop gain: \$m \ll a(v_{IN}) \$ for there to be a large reduction in nonlinearity.

\$\endgroup\$
3
  • \$\begingroup\$ Thanks. I understand how negative feedback increases linearity but I would like to know why it increases "linearity range" not just linearity. \$\endgroup\$
    – emnha
    Oct 16, 2016 at 6:36
  • \$\begingroup\$ So the linearity range is the range over which the transfer function is linear to a certain degree. Increase the linearity and the range increases, at least until you run out of gain as I indicated above. \$\endgroup\$ Oct 16, 2016 at 6:43
  • \$\begingroup\$ After you apply a local feedback the stage gain drops and this is why the "linearity range" increases. We know need a more voltage at the input to "move" out of his "linear region" because of a "less" gain. \$\endgroup\$
    – G36
    Oct 16, 2016 at 11:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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