Dear all,


I am working on the design of an active band pass filter with a high quality factor, and I have encountered a problem with my circuit in which I solicit your help.

In the literature, it is said that when the quality factor of a filter is very large there is a possibility that the circuit oscillates, ie, the structure does not function as a filter but rather as an oscillator. Therefore, to check this, we must attack the structure by a square signal and see the nature of the output signal: is there always a square signal or we have oscillations.

I tested this principle on my circuit but I do not find the good outcome. Therefore, I tried to check it on a simple RC circuit that works like a bandpass filter, but also I cannot find a square signal in the output, I have rather a distorted signal (attached figure).

Therefore, I would like to know if this method is correct. If so, why the RC bandpass filter not follow this principle? Moreover, is there another method to check the operation of a bandpass filter with a high quality factor?

Thanks for your help in advance.

 

hi jn2,
I assume you are familiar with Fourier analysis of a square signal regarding its sinusoidal structure.?

Your pass band filter is modifying some of the sine wave components, so the wave will no longer be 'square'.

E

 

So what is the solution from your experience???

 

hi,
Please post your pass band filter design.
E

 

I tested this principle on my circuit but I do not find the good outcome. Therefore, I tried to check it on a simple RC circuit that works like a bandpass filter, but also I cannot find a square signal in the output, I have rather a distorted signal (attached figure).

The output of your RC circuit looks perfectly normal to me, and is exactly what I'd expect to see; given that it's a bandpass filter, I would NOT expect to see a square wave on the output. The highpass portion of the bandpass characteristic is evident from the fact that the tops and bottoms of the waveform are sloping, rather than level; and the lowpass portion is evident from the low-to-high and high-to-low transitions, which are smoothed and rounded rather than abrupt.

In the literature, it is said that when the quality factor of a filter is very large there is a possibility that the circuit oscillates, ie, the structure does not function as a filter but rather as an oscillator. Therefore, to check this, we must attack the structure by a square signal and see the nature of the output signal: is there always a square signal or we have oscillations.

That is incorrect; what you'd expect to see is a filtered version of the square wave, or oscillations. You will not see a perfect square wave coming out of a bandpass filter.

 

Hello,

Thank you for your help and your answers.

So, you say that it is impossible to see a square signal at the output of a bandpass filter, and that the signal I found is correct.

 

Hello,
So, you say that it is impossible to see a square signal at the output of a bandpass filter, and that the signal I found is correct.

hi,
The output will appear as shown in your image.
If you post your full filter circuit I will run a LTSpice simulation.
E

 

Here is the diagram of the filter RC.

 

So, you say that it is impossible to see a square signal at the output of a bandpass filter, and that the signal I found is correct.

Yes. Your circuit is operating normally, and its output is exactly what you should expect to see.

 

Hello,

You made a filter like in figure 5 of the attached PDF.
(only high and low pass are exchanged).

Bertus

 

hi jn2,
This is from LTSpice.
I have superimposed your image, its a close match.
E

 

We still do not know anything about your bandpass circuit requirements or design.
Filter order
Gain
corner frequencies
Topology (filter type)
Q
The first thing to do when a circuit does not work as expected is to LOOK AT THE SCHEMATIC, but we cant.. Do you have one?

ak