C61 is also part of the second-order low-pass filter and it probably provides the charge when the ADC connects its input sampling capacitor to its input to take a sample.
In fact when you analyse the output impedance of the circuitry to the left of C61, you will find it is equivalent to an inductance of 2 R19 R13 C5 in parallel with a resistance 2 R19 R13/(R19 + R13). Together with C61, it behaves similar to a second-order LRC low-pass.
In fact when you analyse the output impedance of the circuitry to the left of C61, you will find it is equivalent to an inductance of 2 R19 R13 C5 in parallel with a resistance 2 R19 R13/(R19 + R13). Together with C61, it behaves similar to a second-order LRC low-pass.
Last edited:
If I did the math correctly, C5 = C6 = 18 nF and C61 = 39 nF (both NP0/C0G or some other good quality cap, definitely no X7R, X5R or other ceramic class 2 stuff) should result in about your desired roll-off, see the attachment. The normal cut-off frequency is around 657 kHz and the -0.3 dB point well over 300 kHz. I don't know whether the increased inductance will have any side effects on the performance of the PCM4222, you will just have to try and see.