cygpio.exe script is not working with new batch of CY7C65215-32LTX

[EfMi_2218266]

Hello,

We are using the cygpio.exe script with the CY7C65215-32LTX on our boards.

Recently we have a new batch of boards and this command is failed to configure the gpio (see attached picture ''Cygpio_failed.PNG'), we compare on the same PC two boards one with old batch of devices (command succeeded) and one board with new batch (command failed).

We also tried manually using the ''USB serial configuration utility'' to configure the same gpio on both batch the gpio is working as expected.

We compare the version of both devices and its identical (see picture ''Cypress_device_configuration.png'').

The only difference we can see is the marking on the devices:

Old batch marking:
CY7C652
1532LTXI
2025 PHI
622477 C
103

New batch marking:
CY7C652
1532LTXI
2025 PHI
622477 C
311

Can you please advice what the last line at the marking refereeing to?
Can you please advice what can be the root cause for this behavior of the new batch of devices? do you have a solution?

Thank you

Best regards

Efi

[MallikaK_22]

Hi Efi,

Would it be possible to do a swap test, i.e. place the silicon from the working board on the non-working board and vice-versa?

Does the failure follow the silicon?

[EfMi_2218266]

Hi Mallika,

It can be done, but since the PCBs are identical and the difference between them is the assembly batch I think the issue will follow the silicon.

This issue is very strange: while we sending the command manually (using the "utility") the gpio is changed but when using cygpio.exe it doesn't, Do you have another idea why we see different results using different methods?
Can you advice what is the last number at the marking refer to?

Thank you

Regards

Efi

[EfMi_2218266 · Accepted solution]

Hi

After some debug we found out that the default value of the gpio should be as output and not tristate.

After we configure the gpios as output (0/1) the script is working properly.

Regards

Efi