Hi there,
currently I am simulating soft clipper circuits for audio amplifier in SPice (Microcap). The characteristic curve of a circuit gives a detailled information of the nonlinear behaviour.
When using the DC-Analysis (DC-Sweep) all capacitors are replaced by an open. The characteristic curve can just be estimated when the nonlinear transfer element is DC-coupled, which means without any coupling or feedback capacitors in the signal path.
The clipper-circuits, that I investigate contain capacitors.
Does anyone has an idea to avoid this restrictions concerning the DC-Analysis?
I guess the use of a transient analysis could be a possibilty, but how?
It would be great to do the processing and analysis only in Microcap.
I thank you in advance!
Johannes
currently I am simulating soft clipper circuits for audio amplifier in SPice (Microcap). The characteristic curve of a circuit gives a detailled information of the nonlinear behaviour.
When using the DC-Analysis (DC-Sweep) all capacitors are replaced by an open. The characteristic curve can just be estimated when the nonlinear transfer element is DC-coupled, which means without any coupling or feedback capacitors in the signal path.
The clipper-circuits, that I investigate contain capacitors.
Does anyone has an idea to avoid this restrictions concerning the DC-Analysis?
I guess the use of a transient analysis could be a possibilty, but how?
It would be great to do the processing and analysis only in Microcap.
I thank you in advance!
Johannes
Johannes,
Not knowing the details of your circuit, I would say you need to use .tran analysis. To determine the frequency response (Bode plot) you need to run the analysis for each frequency of interest and, in effect, build the plot frequency by frequency. Yes, it's a pain, but it will work. .AC and .DC will not give useful results.
Rick
Not knowing the details of your circuit, I would say you need to use .tran analysis. To determine the frequency response (Bode plot) you need to run the analysis for each frequency of interest and, in effect, build the plot frequency by frequency. Yes, it's a pain, but it will work. .AC and .DC will not give useful results.
Rick
I like the idea of using transient analysis for that. Try a few different amplitudes of sine wave input so you can see what the actual output waveform looks like as you approach, and go beyond, the clipping level.
At the same time you can do fourier analysis to see how much distortion there is and what the spectrum looks like.
It's probably a good idea to check the clipped waveform at very low and high frequencies as well, just to make sure there are no funnies.
At the same time you can do fourier analysis to see how much distortion there is and what the spectrum looks like.
It's probably a good idea to check the clipped waveform at very low and high frequencies as well, just to make sure there are no funnies.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.