Issue while reading SO/QS registers causes outputs to flicker on MC17XS6500EEK (MPC5775B SPI)

[bharathn3402]

I am controlling an MC17XS6500EEK HSD from an MPC5775B MCU via SPI.
Basic output control works fine, but I have a problem when I try to periodically read status registers.

What I am currently doing

1. On startup I initialize the HSD:

   • Configure outputs

   • Set SOA mode during initialization

   • Enable watchdog toggle periodically

2. Outputs operate normally based on my application logic

3. Every 1 second I read status:

   • Write the SO/QS register address into the INI register

   • Read back the SO/QS response

   • If a fault is detected, I write the specific channel address to the INI register and read that response

   • When done, I restore the original INI register value from step 1

   • Watchdog is toggled every time

Problem

When I do this, the outputs flicker and it looks like the device re-initializes.
It seems modifying the INI register at runtime causes side effects on the configured outputs.

My question

Is it allowed/safe to change the INI register after initialization just to read SO/QS registers?
If not, what is the recommended SPI sequence to read status (SO/QS and channel faults) without disturbing the configured outputs or causing flicker/reset?

Any guidance or reference to an application note would help.

[davidtosenovjan]

The flickering outputs are likely caused by unintended reinitialization due to how the SO/QS registers are being accessed.
To avoid this, please ensure each SPI transaction includes a properly formatted 16-bit input word with:

- Correct watchdog bit (D15),
- Valid parity (D14),
- Proper register address and bank selection.

Avoid modifying the INI register outside of the SPI frame. Instead, use the standard SPI protocol as described in NXP's AN5106 application note. This ensures stable operation without triggering output resets.

https://www.nxp.com/docs/en/application-note/AN5106.pdf

[bharathn3402]

Hi @davidtosenovjan ,

In the MC17XS6500EEK HSD, how can we poll fault information? Is there any sample code available for this?

Currently, I’m initializing the HSD with the SOA mode bits configured for I/O status, and using the I/O status SO response address as the SO register address. My approach is to periodically poll the Quick Status Register (QSR), and if any faults are detected, I then poll the individual channel to get detailed fault information and handle it accordingly.

However, when I poll for channel-specific faults, the HSD seems to reset the faults automatically. Could you please share any sample code or guidance on how to properly retrieve fault information without resetting the fault status?