I made this circuit yesterday:

And I used pots for R1 and R2. My question is, when I turn the pots, what am I affecting? Cutoff frequency, resonance? And what would I be affecting if I altered the value of the third resistor at the emitter of the transistor? Finally, where should I take the output from? Yesterday when I was experimenting with this circuit I took the output from the emitter, and got some interesting results, but I don't quite know what I'm doing here lol. How does this circuit work?

 

Changing R1 and R2 change the cutoff frequency *and* the shape or slope of the filter rolloff. Changing the emitter resistor does not affect anything unless it is too large (this affects the source impedance for the feeback through C1) or too small (this affects the current through the transistor and can cause it to overheat.

There are many web sites that explain active filters, but most of them assume a basic knowledge of AC circuits. Basically, this is a two pole filter, R1-C1 and R2-C2. Because the two filter poles are in series, the second one acts as a load on the first one. This interaction makes it difficult to get the best possible filter performance. The feedback from the emitter is called bootstapping. This technique raises the apparent input impedance at R2 so there is less interaction between the two sections. Now the component values in the two sections can be closer to their theoretically perfect values for better performance and much easier calculations.

ak

 

Changing R1 and R2 change the cutoff frequency *and* the shape or slope of the filter rolloff. Changing the emitter resistor does not affect anything unless it is too large (this affects the source impedance for the feeback through C1) or too small (this affects the current through the transistor and can cause it to overheat.

There are many web sites that explain active filters, but most of them assume a basic knowledge of AC circuits. Basically, this is a two pole filter, R1-C1 and R2-C2. Because the two filter poles are in series, the second one acts as a load on the first one. This interaction makes it difficult to get the best possible filter performance. The feedback from the emitter is called bootstapping. This technique raises the apparent input impedance at R2 so there is less interaction between the two sections. Now the component values in the two sections can be closer to their theoretically perfect values for better performance and much easier calculations.

ak

Ok. But because there is feedback, doesn't that create resonance? If so, is there any way to alter the resonance, say maybe put a pot between the emitter and c1?

 

Ok. But because there is feedback, doesn't that create resonance?

No. Some feedback configurations with only resistors and capacitors can form a resonant circuit, but not this one.

There is nothing automatic about resonance. It can be a property of a circuit that does not have any feedback. Also, the vast majority of circuits that do have feedback, but only negative feedback, do not exhibit resonance. Note that I'm speaking of circuits with only resistors and capacitors. When inductors and capacitors are involved, resonance is more likely to be present, but still not always.

ak

 

It should also be noted that the feedback through C1 affects the filter damping and rolloff at the corner frequency, which can give a sharper rolloff, than two passive RC filters in series can achieve.

 

It should also be noted that the feedback through C1 affects the filter damping and rolloff at the corner frequency, which can give a sharper rolloff, than two passive RC filters in series can achieve.

Yeah, based on the level of the question I decided not to get into all of that. The peaking can look like some kind of resonance effect, but in the context of this thread I thought it might be confusing. Now I'm going to have to go back to the books for some s-plane refresh.

ak