Does the Input and output filter required for the codec

Hello Avnendra,

I will start with the ADC side.

To obtain the best possible performance you have to remove any out of band energy (frequencies) from getting into the ADC. The better the filter the better the performance. These converters are oversampling so that does remove some of the strict requirements of the filter but you still need to filter to get the best specs. Then you are correct in that the other purpose is to make the single ended signal differential to drive the ADC. You certainly can simply cap couple one of the inputs to ground and only drive the other ADC input. You will not be able to drive the ADC to full scale 0dBFS. The best you will get is just below -6dBFS. 

Then you can use a simple RC filter for the input and that will get rid of any RF that might cause problems. This will work and you will get decent signal into the ADC but do not expect datasheet specs. It does not seem like you are expecting perfect performance. 

For the DAC side it is similar. On the DAC side the problem is that the DAC output is oversampled so there is a lot of out-of-band energy and that is the purpose of the filter. On the ADAU1962A datasheet I showed a passive filter on the evaluation board. Here is the schematic:

Again, it is not as good as an active filter but does a decent job. This schematic is setup single ended so you will have to stuff R40, R41 and C30 and remove R119 and you will have differential outputs. If you need single ended then use it like it is. Keep in mind the max output will be 6dB less. 

Dave T

Hello Avnendra,

I will start with the ADC side.

To obtain the best possible performance you have to remove any out of band energy (frequencies) from getting into the ADC. The better the filter the better the performance. These converters are oversampling so that does remove some of the strict requirements of the filter but you still need to filter to get the best specs. Then you are correct in that the other purpose is to make the single ended signal differential to drive the ADC. You certainly can simply cap couple one of the inputs to ground and only drive the other ADC input. You will not be able to drive the ADC to full scale 0dBFS. The best you will get is just below -6dBFS. 

Then you can use a simple RC filter for the input and that will get rid of any RF that might cause problems. This will work and you will get decent signal into the ADC but do not expect datasheet specs. It does not seem like you are expecting perfect performance. 

For the DAC side it is similar. On the DAC side the problem is that the DAC output is oversampled so there is a lot of out-of-band energy and that is the purpose of the filter. On the ADAU1962A datasheet I showed a passive filter on the evaluation board. Here is the schematic:

Again, it is not as good as an active filter but does a decent job. This schematic is setup single ended so you will have to stuff R40, R41 and C30 and remove R119 and you will have differential outputs. If you need single ended then use it like it is. Keep in mind the max output will be 6dB less. 

Dave T

Hi Dave,

Thank you for your prompt response. I have attached a schematic diagram based on your suggestion for your review. Could you kindly examine the design and provide your feedback on its feasibility? Please note that this is still a work-in-progress, and we intend to add the digital pins in the next stage. Apart from that, we believe the rest of the design is complete.

I look forward to your input and appreciate your time in reviewing the schematic.

Best regards,
Avnendra

Hello Avnendra,

I will mention a few things...

Power supplies, You do not show where these are coming from and how the AVDD and DVDD supplies are isolated from each other. 

You show the 100nf bypass caps but you do not show any bulk capacitors like a 22uf. You should have one for the AVDD and DVDD to help filter out any noise on the power. Only one is needed for each supply and it does not have to be super close to the part. Just on the board somewhere. 

The circuit you show for the ADC input is not cap coupled. Since you do not show the Opamp input circuit I am not sure if the input is cap coupled or referenced to the CM. You need to do either. 

You need to install the jumpers for the unused ADC inputs. You have a cap to ground which is good. 

For the unused DAC outputs. You certainly can ground them through the 10K resistor like you show. It does not HAVE to be done but it is good to keep RF out of the part and from radiating from the DAC.

I have an issue with the passive DAC output filter. You show a 4.9K ohm resistor to ground on each output pin. This is a rather heavy load for no reason. I would use something like a 47K ohm. The purpose of that resistor is to prevent the leakage of the coupling cap to allow the voltage to creep up on the pin when nothing is connected. Then when you plug something in you can get a POP! So this resistor keeps the voltage from creeping up due to leakage from the capacitor. You do not want to load down the output of the DAC with a 4.9K when a 47K would do the job. 

Depending on the leakage of the capacitor a 100K might do the job! 

Otherwise, it looks good. 

I suggest you look at the PCB layout guidelines document that I have up on the home page of the Audio forum space. 

It is important to follow the recommendations. 

Thanks,

Dave T

Hi Dave,

Thank you for your detailed feedback on the previous schematic. I have addressed the concerns you raised and would appreciate your review of the revised design.

1. I have updated the schematic to include the power supply connections. Please review the new schematic and let me know if it meets the requirements for the codec.

2. Referring to your initial comment, I understood that we might be able to eliminate the input op-amp. If we were to connect the input directly from the headphone jack, I would like to clarify whether the attached schematic would still be valid and feasible. Alternatively, could you please advise on a method to convert the single-ended input to differential-ended without using an op-amp for this codec?

3. I apologize for the mistake in the output DAC filter circuit. I have corrected the schematic according to your feedback. Please confirm that the revised design is accurate.

Thank you for your time and expertise in reviewing the revised schematic.

Best regards,
Avnendra

Hello Avnendra,

Looks good!

Thanks for letting us review it.

Dave T