AM623: In boot the GPIO output issue

Hello Neil-chen

Please confirm the VDDSHV_SDIO connection - where is the signal connected?

There is an ferrite for the LDO at the output.

Please disconnect and check.

Regards,

Sreenivasa

Hi Kallikuppa

I don't quite understand what you mean. Do you mean where the VDDSHV_SDIO voltage is connected? If so, we have it connected to the MMC1_DAT0~3 pull-up and the AM623 VDDSHV5.

If not, please give me some guidance on what to provide.

Thanks.

Hello Neil-chen

Thank you.

How about the SDCD and SDWP?

Can you share the schematics for the VDDSHV_SDO connection.

Can you isolate the LDO output and check if the voltage can change  with the EN input controlled.

Regards,

Sreenivasa

Hi Kallikuppa

SDCD it's normal, we do not connect to MMC1_SDWP. And Nothing changes after isolate LDO .

And the VSEL_SD & VDDSHV_SIO state is as follows (In power on)

Hello Neil-chen

Thank you.

Can you please share the schematics pages in searchable PDF.
I cant see the MMC1_ SOC interface section.

I see the MMC1_CLK pulled up. Can you confirm you do not have a pulldown near to the SOC MMC1_CLK output pin. A pulldown is recommended.

SDCD is connected to VDDSHV_SDIO

Please remove the connection and connect the SDCD pullup to 3.3V SYS voltage .

The SOC supply picture cannot be viewed.

I am attaching the SK schematics.

Regards,

Sreenivasa

Hi Kallikuppa

The MMC_CLK pull-up resistor is not installed. I also modified it according to your suggestion, but the output signal result is the same.

Currently, there is no way to change VDDSHV_SDIO from 3.3V to 1.8V like the EVM. Also, GPIO0_31 cannot be changed from high to always low like the EVB.

The SOC_MMC show as:

Power Bank:

Thanks.

Hello Neil-chen

Thank you.

neil chen said:

Currently, there is no way to change VDDSHV_SDIO from 3.3V to 1.8V like the EVM. Also, GPIO0_31 cannot be changed from high to always low like the EVB.

In case you expect this to be a software issue, i can reassign to the software expert.

If you can share the PDF schematics, i can do a quick review before assigning.

Can you please remove the ferrite at the LDO putput and measure,

Regards,

Sreenivasa

Hi Kallikuppa

So you can't view the diagram above? Or do you need the circuit diagrams?
I've removed the beads, but the situation is the same.

Thanks

Hello Neil-chen

Thank you.

With partial diagram provided i can only provide some guidelines.

If you can share searchable PDF of the board schematics, i can review and add my quick inputs.

Regards,

Sreenivasa

Hi Kallikuppa

Sorry for the late reply, I sent you the PDF file via message.

Thanks.

Hello Neil-chen

Thank you.

Let me review and update.

Regards,

Sreenivasa

Hi Kallikuppa

How else can I provide you with a PDF file?

Thanks.

Hello Neil-chen

You could attach the schematics over the private chat.

Regards,

Sreenivasa

Hi Kallikuppa

I sent you a message again, please check again

Thanks.

Hello Neil-chen

Thank you for the schematics.

Let me review and update.

Regards,'

Sreenivasa

Hello Neil-chen

Thank you for the schematics.

SoC_VDDSHV5_SDIO

Not sure why this signal is connected to VDDSHV6

You need to connect VDDSHV6 to a fixed 3.3V 

The FPGA supply seems to be available before the SOC supply, and this is violation of fail-safe operation

Quick check

DNI the below pullups

disable McASP interface 

Check the SD card IO level and the voltage change functionality

Note that there are few additional errors on the board.

Regards,

Sreenivasa

Hi Kallikuppa

However, we have modified FPGA_3V3 to use 3.3V_SYS instead of 3.3V_source. We have confirmed that FPGA_3V3 will be powered later than 3.3V_source, which means it will be powered later than the CPU. Also, will using the McASP interface on MMC2 affect performance? Because the SD card we use uses MMC1.

Thanks.

Hello Neil-chen

Thank you.

I am not sure i understand your inputs but the current configuration is not a valid configuration.

The pullups connected to MMC2 and the attached device IO interfaced to MMC2 IOs needs to be powered from the same supply source connected to VDDSHV6 which is expected to be a fixed 3.3V and not a switched 1.8V/3.3V.

Regards,

 Sreenivasa

Hello Neil-chen

Additional inputs:

You may have to configure the MMC1 SD card interface to operate in legacy mode (fixed 3.3V) for the current board version if that is an option for you.

Please follow the below thread

(+) OCTVO-3P-OSD62X: Issue with VDD_SD dropping to 1.8V when SD card inserted (OSD62x-PM with AM625) - Processors forum - Processors - TI E2E support forums

Regards,

Sreenivasa

Hello Neil-chen

Checking if you made some progress.

Regards,

Sreenivasa