I was using PCM2912 for my application but it was stated in TI errata that it has packet size problem so i switched to PCM2912A.
But in that also i am getting the same noise problem. So can anyone please suggest me what could be the possible reasons?
I have saperated both analog and digital ground used all TI suggested component values.
Can any TI employee Plz look in to this matter?
This is very serious matter.
Or else can any body give me PCM2912A EVM PCM designs?
Thanking in advance.
Or else can any body give me PCM2912A EVM PCB designs?
Here is the PCM2912A EVM's user guide:
It has schematics and PCB layouts for proper reference. Please let me know if you have any specific questions after referencing our design.
One area to note is that the capacitance on the Vcc pins should followed according to the EVM design. Any variation may cause the power supply rails rails (which are generated via internal voltage regulators) to be noisy. This can degrade overall performance.
thanx nick for the reply,
I have gone through the guide and rechecked my designs . It matches with what ti has suggested in the guide.let me take time to describe the exact problem.
I have given a sine wave signal to the microphone input and recorded it using PCM2912.The recorded waveform shows noise at regular intervals in the waveform.The sound card of the test PC has been disabled.The same card was tested on other machines and gave similar performance.
While listening to audio using the card(PCM2912) the same noise could be heard at regular intervals.So its for sure the problem is not with the recording but with the design of the card. Can you through some light in what might be causing this noise.
Running an FFT on this recorded sine wave will shed a lot of light on what is causing this issue. Common noise frequencies can include:
- 60Hz (or applicable AC power frequency)
- 1kHz - the USB data transfer rate used in recording
However, based on your description of interruptions at "regular intervals," it may be possible that the clock frequency for the chip is slightly off, thus causing glitches and artifacts in your recorded signal. Please verify that the clock frequency is within the recommended bounds outlined by the datasheet. If a crystal is used, please verify that the load capacitors are within the bounds outlined by the crystal datasheet. Variations in load capacitors will "pull" the frequency of the crystal-oscillator circuit and can result in inaccuracies in system timing.
Thanks for your reply. It does work.
Actually the problem was with accuracy of crystal. As i change it to accurate value noise was gone.
Thank you for supporting me in this problem.[:)]
That's great news! Glad to help.
I have been having the same problem, and it is because I used a ceramic resonator. I haven't replaced it yet but I will in the next couple days.
I deliberately made my board with poor grounding so I could see how well the devices worked on a board that had a poor layout. (2 sided board routed ground and power)..
It worked fine except for the regular low frequency noise. The frequency of the noise changed when I probed the resonator, so I'm sure this is the same problem. I have a single channel Class D amp following the PCM2912 to give 1W out.
I used the TPA2006. this works fine with a medium sized speaker but does not do well with the small ones. for them it needs an output filter.
I have an output filter but am curious what would be recomended when going form the PCM2912 to a class D amp. I think there needs to be a filter on the output of the 2912 before feeding theat signal to another AMP. Is an inductor helpful? if any can comment they would be appreciated.
Typically an RC filter is used before the class D amp to prevent any out of band noise from folding back into the audible spectrum. Also note the filter pins can be used to implement an LPF per page 22 of the d/s.
thank you Patrick,
I currently have the filter pins not populated but will put the caps back on them. The crystal did fix my problem as with the earlier post. I confess if I had read the data sheet all the way through I would have used a crystal. The sound is good.
change of plan....now I need to SPI instead of USB.. I'll have to rig up a SPI codec or SPI DAC to replace PCM 2912. any suggestions for what TI part to use? small is important, not sound quality.
Does anyone know if I can drive single 8 ohm speaker instead of the headphone output ?
using the deferential output and dropping the HGND ?
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.