Hello. I'm using an ADA4940 amplifier in a MFB filter, as an antialiasing filter for ADC AD1978. Here are the schematics :
As we are concerned about noise, this circuit allows a fc frequency at 6.5kHz and a -15dB attenuation at 16kHz (half of the sampling frequency for the ADC).
This circuit works well as we have 20-25µVrms of noise in our whole audio path, as expected. But on some lanes we have a disparity issue, and the noise can grow up to 60-70µVrms (x3).
On these lanes we identified that the noise is sensitive to the temperature of the ADA4940. The feedback capacitors seem also have an impact on the phenomenon : changing the capacitors induces the noise to come back to its 'normal' value.
Here are my questions :
1/ Do you think the circuit is well used, in particularly concerning the feedback path ?
2/ It seems that ADA4940 is a current-feedback mode amplifier, is there some precautions to implement capacitor feedbacks (range of values, ferrites, etc) ?
3/ We also noticed a ~200MHz oscillation on the ADA4940 output, which confirm us that the amplifier has instability. What could be the cause ?
Thank you for your feedbacks,