TAS5751M: PWM output not changing between AD and BD modulation - no audio output

Hi Mark,

So the PWM output is switching on your board, which means the device is intialized successfully and i2c communication is working. Could you please try to enlarge the volume and the input signal amplitude? Make sure EQ and AGL are disabed. Please check whether there is any Error in the register 0x01. Are you using 4-wire I2S input or 3-wire I2S input mode for TAS5751M? For the input clock requirement, please find this wiki:

About the AD/BD mode configuration, I think you could try to dump the register configuration on EVM board by PPC3 GUI and use it for your board. I believe there shouldn't be any difference between EVM board and your board. To dump the register config, please find the procedure in section 3.8 in the EVM user's guide: http://www.ti.com/lit/ug/slou440/slou440.pdf

Best regards,

Shawn Zheng 

Hello Shawn-

I used 3 wire (SCLK, LRCLK, SDIN) and 4 wire (MCLK, SCLK, LRCLK, SDIN) I2S signal. I purposely put 100 ohm resistors between ADC and the amp to probe the signals. So it was very easy to remove MCLK signal - the results are the same.

Register 0x01is the device ID, I am not sure how to use this for diagnostics. I check register 0x00 (returns 0x6C) to confirm clock signals are ok and register 0x02 (returns 0x00) to check for any errors.

I can take the same I2S signal and pass it to another amplifier with no problem. With the same/similar volume settings on the EVM I can observe PWM signal and hear audio from a speaker. So I am not sure the input levels could be a problem. I included the I2C initialization from the I2C bus to show exactly what is being initialized. Just using the default values and the values I included should have the proper plumbing for the audio chain, AGL and mixers etc. - The EQ and limiters should be "off". I have tried two different PCBA of this design but cannot get audio out on either. For some reason, the switching remains constant at 50% duty cycle and absolutely no sound from the speaker, not even a *hiss*

Below is the I2C traffic. I copied the mux/mixer settings from EVM and included it just in case. I also poll the 0x00 and 002 registers to check for errors (see below)

Can there be a hardware error? For example some external PLL R's and C's that could lock up the PWM to 50% duty cycle? The majority of the parts on the PCBA are 0402 so very difficult to read the values and confirm the assembly. 

Also, when using the PPC3 tool, since my board only has stereo (no mono amp), the GUI tool reports error when loading the audio processing tab. When I look into the I2C traffic, the error occurs when the PPC3 tool tries to initialize the sub amplifier (not exist in my design). In the PPC3 tool I adjusted the volume to +24dB and still no audio signal from the PWM

This is the error I received when entering the process flow although the I2C communication still works when I adjust any single parameter.

Shawn-

The startup sequence specified in section 8.2.1.2.5.1 has some specifics I cannot follow.

The PVDD, AVDD, DVDD all come up roughly the same time - within 0.5msec. PVDD is regulated down to AVDD/DVDD. The only control I have is the RST, PDN and I2C communication timing. From quick measurement, DVDD/AVDD lags PVDD by about 250~500uSec (depending on which threshold we consider). Could the TAS5751M be behaving the way it is because of PVDD/DVDD timing?