• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 207 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PCM5252: High noise floor when output active

Simone Matthys
Prodigy 10 points
Part Number: PCM5252
Other Parts Discussed in Thread: PCM5242, TAD5142, TAD5112

Hello

I'm currently developing a new DAC circuit for our audio amplifiers as the old DAC is obsolete.

We've been trying out different ICs to handle the conversion, and the PCM5252 seems like the perfect candidate.

I've designed a prototype board and used the LT3042 LDO @ 3.3V as a power supply. I've followed some example circuits and got the DAC working in Hardware Mode.

The issue I currently have is, when the I2S data is enabled but the generated sine wave is at -150dBV or lower, I have a noise floor of -99dBV, which is about 12uVrms.

When the I2S is disabled or the Soft mute function is enabled, the output noise floor drops to about -115dB, which is the lowest my audio analyser (Quantasylum QA403) can get. So this means the muting function is working as intended.

However, when playing a 1kHz Sine @ 2Vrms, the measured SNR of 99dBv is much lower than the 114dB described in the data sheet.

Does someone have any idea what the origin/source of the noise could be? I've tried using both LT3042 LDO en MIC5202-3.3 but this didn't change anything noise wise.

Any help is much appreciated!

Kind regards

Simone Matthys

  • 0
    TI__Mastermind 25540 points

    Hi Simone,

    The number you are reporting is very different form datasheet. Since you are not using the official  EVM and is a prototype board, most probably you have an issue with your board ( layout, grounding issues, noise coupling. I  have seen  bad capacitor  caused increased noise floor too.)

    You can use the EVM schematic for PCM5242 as a guidance (SLAU592A). These 2 parts  are almost identical devices, so you can use it as a guide. The layout layers shown in the document as well . Another option   is to get 5242  EVM and use it for PCM5252 since the pinout is identical.

    You may be  interested in our new line of DACs that are very advanced and high performance:  The TAD5xxx family such as TAD5142 or TAD5112 to name a couple.

    Let me know if you have any questions.

    Regards,

    Arash

  • 0 in reply to Arash Loloee
    Prodigy 10 points

    Hi Arash

    Using almost exactly the same board layout we tested another DAC, and this IC had a noise floor of -115 dB with the same configuration (I2S active but signal level as low as possible). I've also looked at the EVM schematic and layout and concluded that these are almost identical to our prototype. The only thing that caught my attention is that the EVM uses a DVDD of 1.8V instead of 3V3. The1.8V is generated by an LDO which is powered by 3V3 coming from another LDO. 

    Our DVDD is connected to the same 3V3 rail that is connected to AVDD and CPVDD. Could this be the cause of the noise we're seeing?
    I saw another post on the forum that also has noise issues with the PCM5242 on their custom pcb, and they were also using just a 3V3 supply rail. Perhaps it's the same issue?

    Do you suggest using a 1V8 LDO for the DVDD to minimize noise or shouldn't this be causing these issues?

    Thank you for you response, inspecting the EVM schematics was very interesting!

    Kind regards

    Simone

  • 0 in reply to Simone Matthys
    TI__Mastermind 25540 points

    Hi Simone, 

    The internal digital core of the PCM5252 devices needs 1.8-V supply. However all plots in the datasheet are with AVDD=DVDD=CPVDD=3.3V. So, the part should  work with DVDD of  3.3V as well as 1.8V , however,  if LDOO is the Output from the on chip LDO , it needs  a 0.1-µF decoupling cap and i assume  you verified there is a  cap and it  is okay.

    I would suggest to try with 1.8V for DVDD (with proper decoupling the LDOO pin) or try to power LDOO pin  with an external 1.8V supply and keep your DVDD at 3.3V. These 2 tests should give you a better indication of the source of the issue. Most probably the issue is either the actual implementation of the layout of the board or could be a  noisy supply.

    Regards,

    Arash

  • 0 in reply to Arash Loloee
    TI__Mastermind 25540 points

    Hi Simone, I am closing this issue but feel free to open it if you have any questions.

    Regards,

    Arash