You cannot get it more simple than this, a 1 transistor CW filter. This filter is build according to the (surprise !) Multiple Feedback BandPass Filter (MFBPF) topology. Below, you see the circuit, originally designed by KE3IJ (alas now SK).
Transistor R1, R2 and R3 together with C1 and C2 form the standard MFBPF topology. R3 is part of the feedback network and doubles as the resistor to set the bias current. The open-loop gain of a single transistor is much smaller than of an opamp, of course. So the measured values for center frequency fc, Q and especially gain might differ somewhat.
It was build dead-bug style on a piece of copper clad board. Turns out there was an error in the circuit and fc was set around 485Hz (not 700Hz as advertised). Also I wanted to have a gain of ≈ 1x. So I changed the values from R1 and R2 to 82k and 2k2. The simulated gain is now -2 dB (≈ 0.79x).
Below the measured actual response. This is from two stages, so that I could compare it to the previous designs. Gain is circa -5 dB and center frequency fc is circa 830 Hz.
All in all a pretty nice filter that can quickly be build. it is nice variation on the MFBPF topology theme and might come quit useful. Even though the overall gain is little lower then aimed for!
I like the simplicity! Can the frequancy be adjustable using a potmeter?
Hi Jan,
just with a Multiple Feedback Bandpass Filter build with operational amplifiers, one can change the center frequency by varying the value of resistor R2. The passband bandwidth will stay the same, so the Q will change accordingly.
Kind regards,
cor / pa3cor
thx, I will breadboard it!