AM62A7: Custom board unable to connect through USB or UART

Here's the screenshot of sequencing.  Ch1 is SoC_DVDD1V8, Ch2 is VDD_LPDDR4, Ch3 is VDD_CORE.

Here's the screenshot of the additional channels of sequencing.  Ch1 is VDD_CORE, Ch2 is MCU_OSC, Ch3 is MCU_PORz.

MCU_OSC early rise looks to correlate with SoC_DVDD1V8 booting up shown below. Ch1 is SoC_DVDD1V8 and Ch2 is MCU_OSC.

Hello Josh Fox,

Thank you for the additional information. I will need 1-2 more days to familiarize myself with your issue.

Thanks,

Stan

Thank you Stanislav Stilyanov , please let me know what information I can provide to help troubleshoot.

Hi Josh,

Just wanted to check you have followed power sequencing diagram in section 6.12.2.2 Power Supply Sequencing of the datasheet?

It's a bit hard for me to understand how your supplies 1-5 match with  SoC inputs.

Thanks,

Stan

Hi Stanislav Stilyanov , yes I did follow the sequencing diagram in section 6.12.2.2.  Basically everything is powered by 1.8V aside from VDD_CORE and LPDDR.  One thing that I see is incorrect in my design is that VDD_CANUART is connected to 1.8V with VDDSH_CANUART conflicting with note 10 which states: "VDD_CANUART shall be connected to the same power source as VDD_CORE, VDDA_CORE_CSIRX0, VDDA_CORE_USB, and VDDA_DDR_PLL0 when not using Partial IO low power mode." 

Unfortunately, I do not have the soft reset pins available on this board which may help ruling out the power supply ramp up per [FAQ] Board bring up tips for Sitara devices (AM64x, AM243x, AM62x, AM62L, AM62Ax, AM62D-Q1, AM62Px).

I took a look at MMC0_CLK which is the clock for the Flash memory which is the default bootmode.  It appears to be starting, though at the oscillator frequency and quite noisy so I wonder if it is just noise coupling.  

In the scope plot below Ch 1 is MCU_PORz, Ch2 is RESETSTATz (which is used to enable the flash out of reset), Ch3 is MMC0_CLK

Hi Josh,

Can you please try UART as primary boot peripheral and see if you get something in a serial terminal @ 115200 kbaud, 8-n-1 ?

UART primary boot selection:

Thanks,

Stan

Hi Stanislav Stilyanov , 

Unfortunately not, this is a custom board with 0201 resistors setting the boot mode and I worry I would cause irreparable damage.  Is there anything else that I can try?

Thanks,

Josh

Josh,

Can you please post the part of the schematic related to USB?

Thanks

Stanislav Stilyanov , of course.

Here is the AM62A7 connection to a Hirose connector.  The choke shown ended up being populated during manufacturing of the board so both 0 ohms and choke are on the board.

Below is the Hirose to USB-C connection which is its own flexboard.

I sincerely appreciate your help debugging this problem.

Josh,

Can I take a look at the VBUS resistor divider to USB0_VBUS?

Also, you never mentioned a 3.3-V power supply in the sequence?

Do you have proper 3.3 volts connected to VDDA_3P3_USB

Stanislav Stilyanov here is the resistor divider for USB0_VBUS.  The product is generally battery powered, though all bench testing during bring up has been through USB 5V, so I can see this as a problem during normal use but that's a problem for another day.

VDDA_3P3_USB is connected to a LDO 1.8V since the device will be used with Li-Ion battery 4.2-3.2V.  I will continue to look for the reference that mentioned this connection; I'm having difficulty finding it right now.

Thanks Stanislav Stilyanov for your guidance.  I'll make the recommended changes and order new samples.  

Stanislav Stilyanov , thanks for the clarification.  

Regarding making UART primary boot through reworking the board, do you expect that UART will operate despite the VDDA_3P3_USB error?

Thank you.

Hi Josh,

Yes, I would try switching to UART, even though pull resistors are miniature, to see if UART boot happens. USB PHY power supply should not affect LVCMOS pins.

Regards,

stan