0
\$\begingroup\$

Question is about the gain peak (or maybe more suitable would be growth) ocurring in higher frequencies (>20kHz) in Chebyshev filter 4-th order in simulation (simulated using Bode Analyzer in NI Multisim) - what's the reason for this? Theory shows that gain is going still down even up to -150dB, however simulation shows minimum at about -120dB. Is that because of some capacitance effects ocurring in op-amps or something else? I found similar problem to mine, but the grounding there and op-amp was quite different (Unexpected behavior of the Sallen-Key second order filter). Below theoretical gain based on formula: $$ A(f)_{dB}=20\log{\left\lvert\frac{k_0}{(1 + a_1(\frac{j(2\pi f)}{\omega_g}) + b_1(\frac{j(2\pi f)}{\omega_g})^2)(1 + a_2(\frac{j(2\pi f)}{\omega_g}) + b_2(\frac{j(2\pi f)}{\omega_g})^2)}\right\rvert} $$ and chart for this: enter image description here The result of the simulation is shown below. enter image description here In higher frequencies like 1MHz gain continues to go up. The circuit was constructed based on the Sallen-Key scheme - schematics of the cicuit below. enter image description here

\$\endgroup\$
  • \$\begingroup\$ Welcome to the real world, where opamps have finite I/O impedances, limited bandwidths, even RLC elements behave strangely with frequency. Not to mention the methods of measuring, or the questionable use of that particular opamp. \$\endgroup\$ – a concerned citizen Jun 2 at 11:03
3
\$\begingroup\$

Is that because of some capacitance effects ocurring in op-amps or something else?

Your sim uses the cranky old 741 op-amp and that device fails to become effective even as a unity gain amplifier around 1 MHz. You have about 20 dB pass band gain and so you are asking each 741 to roughly provide a gain of 3 and now the effective point where the 741 gives up the ghost is going to be around 300 kHz.

But, because your circuit is trying to produce a fairly high Q using the feedback capacitors C2 and C4, the new point in the spectrum where the 741 falls over is going to be about ten times lower in frequency i.e. circa 30 kHz.

Just ballpark numbers but you can prove this to yourself by using an op-amp with a gain-bandwidth product of 10 MHz or above.

| improve this answer | |
\$\endgroup\$

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.