Active bandpass filter - unwanted oscillations ( noise)

[electronx]

I designed a bandpass filter with a cut-off frequency of 0.1hz and 110khz when I applied a 1khz -20khz sine wave to the input of my band pass filter, I saw oscillations (ringing) at the zero cross points.

(It is obvious that there is a higher frequency harmonic here)

When I do an fft analysis, I see oscillations between 4-5 MHz. Around -40dbV, at a level of approximately 8-10mV.I think one of the filters is producing an oscillation. (I tried with different power supplies. The noise is not coming from the power supply.)What is the solution to this or what is the point I missed? capacitance value selection?
When I did AC sweep in Spice simulations and observe in frequency domain  there was no problem, everything was perfect.
maybe capacitance value selection, resistance value selection, pcb design, factors that I cannot foresee or other factors that I do not know theoretically







first High pass filter (opa192)
order Active High Pass Filter Topology: Sallen-Key Stage Gain:1 Cutoff Frequency:0.1Hz

-----------------------------------------------------------------------------------------------------------------
lowpass pass filter THP210
Multiple-Feedback Topology 2. order Active Low Pass Filter (Gain:1) fully differential Output Cutoff Frequency:110KHz

[tggzzz]

It is good to see someone exploring, questioning, thinking and asking questions.

Show a photo of your implementation's construction and your measurement technique. Implementations sometimes diverge significantly from the textbook behaviour and from simulated behaviour.

For a simple circuit illustrating one gross and one subtle divergence, see https://entertaininghacks.wordpress.com/2024/03/16/practical-traps-with-a-one-transistor-audio-amplifier-solderless-breadboards-and-oscilloscopes/ That example also uses an Analog Discovery.

[moffy]

First suspect is R11 introducing some unwanted delay, try reducing it or eliminating it. Second output stage oscillation through the transition region, try a resistive load and or a Sobel network, basically an RC snubber to dampen the output oscillation.

[electronx]

The r11 resistor is used to limit the current entering the power rails with schottky clamp (input protection) diodes. (yes, there is no clamp diode here, but it is not placed on the pcb). I will check the factors that cause delay in the filter response.

[Wolfram]

Ferrite beads and ceramic caps on the supply rails can form an underdamped resonator. It could be worth to rule this out as a cause by measuring the supply rails together with the output signal, and ideally add some damping or remove the ferrite beads in any case.

[Sensorcat]

In general, it is useful to look at signals between the blocks whenever you want to debug something with functional blocks connected. Since your filter has two chained blocks, you can look at Vmid. Is the oscillation present there? Yes, problem with 1st block, no, 2nd block. If you're unlucky, the problem is somewhat distributed between the blocks, which means you have to ask which common cause can interfere with both blocks.

[Benta]

I can't read the schemayic text (too small), but are those amps unity-gain stable?

[electronx]

I can't read the schemayic text (too small), but are those amps unity-gain stable?

Hi Benta, thanks for your return, I was waiting for your comment (I saw that you spent a lot of time on active filters)

If you right click on the photo and say open in new tab and left click on the image, I think you can see the photo in a big size.

If the unity gain stable ratio is from the existing graphs (unity gain ratios are well above the frequencies I work with.)

 opa 192 unity gain stability (closed loop)



thp210  unity gain stability (closed loop)