Question about the output waveform of low-pass filter(op-amp)

Thread Starter

William.A.

Joined Dec 12, 2020
8
Hi guys I found a interesting phenomenon in the output waveform of a low-pass filter.

Why the first peak is always higher than the others when I set the input signal frequency as 6000Hz or more larger? The larger frequency then the difference between the first peak and others become more obvious.

The second question is why the peak to peak value keep decreasing as the frequency is increasing?

**The cut-off frequency of this circuit is around 15915Hz by my calculation.

Could anyone explain this phenomenon to me?
Screenshot 2021-10-22 161209.png
Screenshot 2021-10-22 161228.png
 

Attachments

LvW

Joined Jun 13, 2013
1,752
Obviously, the frequency is too large. The limited slew rate is responsible for the waveform.
Reduce the input amplitude to some mV only - and you will have a sinusoidal waveform at the output.
Regarding start of the TRAN simulation (time domain):
For any simulation in the time domain (and for all real-time measurements as well) , the first amplitude peak is never according to the expected value because the circuit has not reached its steady-state condition.
 

Papabravo

Joined Feb 24, 2006
21,159
It has nothing to do with noise. It is a linear AC analysis about a fixed DC operating point. It plots the magnitude, in dB, of of the response and the phase in degrees.
Notice the AC 1 next to the Voltage source V1. That sets the small signal AC input to 1 so that when you compute the magnitude of the output in dB with respect to 1 the computation is simplified.
 

LvW

Joined Jun 13, 2013
1,752
It has nothing to do with noise. It is a linear AC analysis about a fixed DC operating point. It plots the magnitude, in dB, of of the response and the phase in degrees.
Just for clarification (and in order to avoid misunderstandings on the questioners side):
Such a linear AC analysis is a small-signal analysis in the frequency domain. This analysis cannot show any large-signal induced distortions or slewing effects. Therefore, no results about output waveforms can be derived from such simulations.
 

Papabravo

Joined Feb 24, 2006
21,159
Just for clarification (and in order to avoid misunderstandings on the questioners side):
Such a linear AC analysis is a small-signal analysis in the frequency domain. This analysis cannot show any large-signal induced distortions or slewing effects. Therefore, no results about output waveforms can be derived from such simulations.
That is correct. I should also point out, for the TS's benefit, that a transient analysis shows transient behavior. Transient behavior is different than steady state behavior because even though you have the same governing differential equation, the initial conditions are different. You may also notice that a transient analysis has an option to ignore data at the beginning of a simulation for some period of time for the reason that the initial transient behavior may be of no particular interest.
 
Top