IWR6843: I2C driver fail after Warm reset

Swapnil

Part Number: IWR6843

Hello Community,

I am using I2C for reading and writing into EEPROM. Before Watchdog reset I don't get any problem with read or write, but after Watchdog reset when I try to write or read into EEPROM, sometimes the I2C transfer is not successful. But it's not always the case.

After the power-on reset, this problem never occurred. I2C communication was always functional after the power-on reset.

Thanks and Regards,

Neil

over 2 years ago

0 Angie Mitchell over 2 years ago

TI__Genius 10730 points

Hi Neil,

Thank you for reaching out! Which version of our SDK and which of our devices are you currently using?

Do you see this problem when using the NRST (soft) reset?

Thank you,

Angie

0 Swapnil over 2 years ago in reply to Angie Mitchell

Intellectual 550 points

Hello Angie,

Thank you for the response.

I am using SDK version 3.5.0.4. with IWR6843 on a custom PCB. I see this problem after using soft reset (Watchdog timer reset). My guess would be before Watchdog reset I2C driver does not close since the reset comes during operation, hence after reset when I am trying to use I2C it fails sometimes, but I don't know if this is the correct assumption.

When I close I2C before reset and use it after reset to read or write EEPROM it functions without fail.

Is there any possibility to correctly reinitializing the I2C driver after the watchdog reset?

Thanks and Regards,

Neil

0 Angie Mitchell over 2 years ago in reply to Swapnil

TI__Genius 10730 points

Hi Neil,

My apologies in my delay in a response here, I am checking with our team internally to learn more about this behavior. Please give me 2 more days to look into this and feel free to ping this thread if you do not hear back from me by then. I agree with you that the soft / warm reset must not be closing the i2c driver properly but I want to see if someone internally knows a work around to fix this behavior.

Thanks,

Angie

0 Swapnil over 2 years ago in reply to Angie Mitchell

Intellectual 550 points

Hello Angie,

No problem and thank you for looking into this.

Regards,

Neil

0 Angie Mitchell over 2 years ago in reply to Swapnil

TI__Genius 10730 points

Hi Neil,

I am so sorry for the delay and our team is still looking internally to confirm this behavior.

Angie

0 Swapnil over 2 years ago in reply to Angie Mitchell

Intellectual 550 points

Hello Angie,

Is there any feedback from your team about I2C behavior after the warm reset?

Regards,

Neil

0 Angie Mitchell over 2 years ago in reply to Swapnil

TI__Genius 10730 points

Hi Neil,

Sorry for the delay here. Our team tested this behavior without the SDK and did not see this issue. Which version of the SDK are you currently using?

Thanks,

Angie

0 Swapnil over 2 years ago in reply to Angie Mitchell

Intellectual 550 points

Hello Angie,

Thank you for testing this behavior. I am using SDK version 03.05.00.04.

The I2C diver fails sometimes after a warm reset from the watchdog, but it's not the case all the time. So rigorous testing on I2C driver is required after warm reset to confirm the behavior.

Thanks and regards,

Neil

+1 Angie Mitchell over 2 years ago in reply to Swapnil

TI__Genius 10730 points

Hi Neil,

Our team confirms that it seems that the slave does not detect the stop condition after the warm reset. As a workaround, we suggest creating the stop condition before starting any communication on the I2C. This way, slave is in init state before any transaction starts after warm reset.

Thank you,

Angie

0 Swapnil over 2 years ago in reply to Angie Mitchell

Intellectual 550 points

Thank you Angie. I will use this workaround.

Sensors

Sensors forum

IWR6843: I2C driver fail after Warm reset