TAS5782MEVM: Idle Mode Current Consumption

John Fedak

Part Number: TAS5782MEVM
Other Parts Discussed in Thread: TAS5782M,

Hi Audio Team,

My customer is using the TAS5782EVM and is looking at reproducing the idle power consumption information in the datasheet on page 20. In particular the 37mA on the PVDD rail in the 24V PVDD, 4 Ohm load, 768kHz, Idle configuration.

In 2 different configurations, this was measured on the EVM with one (of the two) device enabled (and unmuted) and in both cases the EVM is showing around 120mA PVDD (24V) current. This is a little more than 3x higher than stated.

In the first configuration, the EVM was connected to the PPCMB (Rev F, +4V powered via an external supply) and using the EVM GUI to control the status of the device.

In the second configuration, the EVM was separated from the PPCMB and the I2C/I2S/3.3V were provided directly from an alternate source.\

In both cases a resistive 4Ohm load was used.

We are very interested for your thoughts on why the PVDD current draw would be so much higher than expected.

Please let me know if you have any questions, thank you!

Regards,

~John

over 7 years ago

0 Jeff Kohnle over 7 years ago

TI__Expert 6985 points

Hello John! Andy will follow up on your question, but can you also verify what gain setting is in place for your configuration? Thanks, Jeff

0 John Fedak over 7 years ago in reply to Jeff Kohnle

TI__Expert 6715 points

Hi Jeff,

Looking at the EVM schematic, I see that it is populated with the 20dB gain setting (and the datasheet states 26dB for the spec). I will check with the customer to see if this has been modified.

Interested to understand if the lower gain setting would cause a (significant) increase in the idle current.

Regards,

~John

0 John Fedak over 7 years ago in reply to John Fedak

TI__Expert 6715 points

Hi Jeff, Andy,

The customer has pointed out to me that the EVM schematic and the datasheet don't match on the gain calculations. Based on the datasheet table 14, the populated (750k and 150k) resistors are set for 26 dB of gain - matching the idle current spec that is looking to be replicated.

Regards,
~John

0 Andy Liu (BMS) over 7 years ago in reply to John Fedak

TI__Guru 52115 points

Hi John,

The 14dB and 20dB gain settings in the EVM schematic are equivalent to the 20dB and 26dB gain setting in the datasheet. The reason for the difference is that they have different definitions. We can update the EVM schematic to make them consistent.

Andy

0 Andy Liu (BMS) over 7 years ago

TI__Guru 52115 points

Hi John,

If I remember correctly, the idle power consumption data on Page 20 were not taken using a TAS582MEVM since that board needs quite a few hardware changes if used for idle PVDD current measurement. Let me check with my colleague to find out which board was used and how we got these data.

Andy

0 Jeff Kohnle over 7 years ago in reply to Andy Liu (BMS)

TI__Expert 6985 points

Hello John! As discussed Andy will close back with you on the feedback from the design team. Thanks, Jeff

0 Andy Liu (BMS) over 7 years ago in reply to Andy Liu (BMS)

TI__Guru 52115 points

Hi John,

I have confirmed with my colleague that he used another board which we call MTS board to take the measurement data on Page 20.

In the e2e post, you mentioned the customer tried two different configurations. Did they remove one of the two TAS5782M devices on the EVM in the second configuration? What’s the I2S input to TAS5782M? Provided by AP? What are the commends sent to TAS5782M via i2c?

Andy

0 Andy Liu (BMS) over 7 years ago in reply to Andy Liu (BMS)

TI__Guru 52115 points

Hi John,

I am checking with my colleague again which LC filter we used when we took data. It's probably different from the LC filter on the TAS5782MEVM. Different LC filters may lead to different idle current based on our experience.

Andy

0 John Fedak over 7 years ago in reply to Andy Liu (BMS)

TI__Expert 6715 points

Hi Andy,

The second device was not removed, however the current was 3x higher. Yes, an AP provided the I2S. Are there any I2C settings to pay close attention to?

Please let us know the feedback from your colleague on the best way to achieve this idle spec on the EVM.

Thank you!

Regards,

~John

0 Andy Liu (BMS) over 7 years ago in reply to John Fedak

TI__Guru 52115 points

Hi John

Unfortunately I have not yet got feedback from my colleague who took the data yet since he is on vacation this week. For now I would suggest your customer replace the 4.7uF inductors with 10uF inductors. They should be able to see much better results.

Andy

0 Jeff Kohnle over 7 years ago in reply to Andy Liu (BMS)

TI__Expert 6985 points

Hello John! Did this recommendation help your situation? Thanks, Jeff

0 John Fedak over 7 years ago in reply to Jeff Kohnle

TI__Expert 6715 points

Hi Jeff,

Thank you for following up. I believe the inductor change will help with idle current, however we still need to understand the exact settings for the idle current measurements in the datasheet to re-create in this system.

Regards,

~John

0 Jeff Kohnle over 7 years ago in reply to John Fedak

TI__Expert 6985 points

Hello John! I believe the difference is that some DS data is pulled from a test board (MTS), which would not necessarily be exactly configured like the actual EVM. Idle currents are tough to measure on an EVM, and so in this case the MTS board was used and was configured as:
Frequency LC Filter
384k 10uH/0.68uF
768k 4.7uH/0.68uF

Audio input -999dBFS, 1kHz

Registers at default, but 0x02=0x00 to put device in play mode.

Andy, Do you have any other points that would help John here.

Thanks, Jeff

Audio

Audio forum

TAS5782MEVM: Idle Mode Current Consumption