0
\$\begingroup\$

I am trying to figure out how to calculate the sensitivity of a circuit, especially the sensitivity of a damping ratio (how much the damping ratio changes as each component value varies).

enter image description here

Expression for zeta is quite long and complicated, so I prefer not to take partial derivative. I would like to do something like:

calculate the value of zeta when R = 100 ohms.

calculate the value of zeta when R = 101 ohms.

then divide the change in zeta by the change in R.

In this case, how would I normalize this sensitivity value (x/y part)?

\$\endgroup\$
1
  • \$\begingroup\$ In this case your "y" is zeta and your "x" is R. So just subsitute the values in. \$\endgroup\$
    – The Photon
    Commented Oct 2, 2019 at 15:56

1 Answer 1

1
\$\begingroup\$

In this case your "y" is zeta and your "x" is R. So just subsitute the values in.

$$S_R^\zeta \approx \frac{R}{\zeta}\frac{\Delta \zeta}{\Delta R}$$

\$\endgroup\$
4
  • \$\begingroup\$ Would I have to substitute original values or R and zeta for normalizing? \$\endgroup\$
    – user207787
    Commented Oct 2, 2019 at 15:58
  • \$\begingroup\$ @user207787, what do you mean by "original" values? You only have two "experimental" values: \$\zeta(R=100)\$ and \$\zeta(R=101)\$. Use those to calculate \$\Delta \zeta\$. \$\endgroup\$
    – The Photon
    Commented Oct 2, 2019 at 16:00
  • \$\begingroup\$ Sorry, I meant "Would I have to substitute original values *for R and zeta for normalizing?". For R/zeta, do I have to plug in 100 or 101? \$\endgroup\$
    – user207787
    Commented Oct 2, 2019 at 16:11
  • \$\begingroup\$ @user207787, if it makes a difference, you should use a smaller difference between your two R values, like 99.9 and 100.1 or something. \$\endgroup\$
    – The Photon
    Commented Oct 2, 2019 at 16:12

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.