Hi TI experts,
I have an application that I2C speed 1MHz with a slave device.
The normal I2C read waveform with another host as below:
You can see a little stretch after 8th SCK for waiting slave ACK.
And LP-MSPM0G3507 I2C read waveform as below:
It don't have stretch and I get NACK interrupt.
I have enabled controller clock stretching and I2C configuration as below:
Am I missing other settings?
BTW, It works when I slow down to 400KHz. (the waveform no stretch)
Thanks.
BR,
Nian
Hi,
I have tested the stretch function.
With the stretch enable the SCL frequency is down to around 695kHz. And without it, the SCL frequency is 1MHz. So the stretch function is working.
The long wait time on the ACK status is not related to the stretch function.
Best regards,
Cash Hao
Hi,
Oh sorry, I am misunderstanding the stretch thing.
These are what I get from more test. When I disable the stretch function on the Target side. There is no delay after the Read commend. And if I enable the stretch function on the Target side. There will be a delay after the Read commend. And I also changes the MCLK for the Target device, trying to act as a slow I2C node. And the stretch is longer according to the test result.
Best regards,
Cash Hao
Hi Cash,
Thanks for your more test.
From your results, the stretch function of MSPM0G is good.
The stretching of your test occurs after slave ACK. (before read data)
However, the stretching of my device occurs before slave ACK. (between SCK 8th and 9th)
Could the MSPM0G still handle it?
BR,
Nian
**Attention** This is a public forum
