LM4670: Similar issue to "LM4670: Output noise without input"

As a somewhat-related aside, what do Cin and Rin represent in equations 4 and 5 in the LM4670 datasheet?  I'm checking through this design that someone else did and trying to understand their choices while I debug this problem.

Hi, Steve,

I couldn't find any information about the case that you referred. However, these are the common noise issues that you may verify in your design:

- It is important to verify that the power supplies are correctly regulated and there are no ripple or noise in the supply. Any noise activity will affect the output behavior. Please use capacitors around 10uF and complement it with capacitors of 0.1uF. This capacitors array should be enough to mitigate the power supply noise.

- If the problem persists, it is a good practice to place all the inputs (in this case, your single ended inputs) to GND. If the output still has a noise issue, you will need to focus on the output lines or even the ground plane, which could be also affected by an external component in the board.

- If the noise is present after the practice mentioned above, there could be an impedance difference at both inputs. It is important to have both inputs with the same passive components. Any difference may cause pop issues or even noise events. Small capacitors of the pF order may be added to reduce parasitic signals. Please route them from the analog inputs to ground.

Regarding your question about the equations 4 and 5, they represent a resistors and capacitors array. When many single-ended inputs are used, like in figure 30, the equivalent input capacitor and resistor of the non used input will be determined by the equations 4 and 5.

Please let me know if you have additional questions or comments.

Best regards,
Luis Fernando Rodríguez S.

Sorry, I think I pasted the wrong link; this is the issue in question: https://e2e.ti.com/support/audio/f/6/t/618222

Hi, Steve,

I have no details about the resolution of the thread you mentioned. It seems that it was not concluded or the conversation continued via email or phone.

Could you verify if your design respects all the recommendations/best practices that I mentioned in my previous post, please?

Best regards,
Luis Fernando Rodríguez S.

Hi Luis,

Thanks for checking on this for me.  It appears that we had too much capacitance at the output for the amplifier to tolerate.  Removing that capacitance and using ferrites for EMC seems to help quite a bit.  Do you have any recommendations for a maximum capacitance that can appear at the outputs of this amplifier?  (Did we just miss it in the datasheet somewhere?)

Thank you,

Steve

Hi Luis,

Thank you for your response!  We were able to pass radiated EMC testing with the caps removed and series ferrites in place.  I was concerned that we'd have to add capacitors to better filter the output, but that wasn't necessary.  Still, it's good to have a rule of thumb in place in case it's needed for something else.

Regards,

Steve