- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Original question:
Hello,
I was able to communicate with the SN65DSI85 EVM using I2C address 0x2D and now I am trying to communicate using the I2C from a microcontroller. I get an io bus address failure when I use 0x2D as the I2C address. Can you confirm?The datasheet (sec 7.5.1 Local I2C interface overview) mentions that the Addr = 1, the address cycle is 0x5A (write) and 0x5B (read) but I dont set these address with Aardvark and I am able to read correctly from the EVM. Can you confirm if I'm missing something?
Thanks
Ranga
Hi I.K.,
I checked the waveform with the micro-controller and it does shift the address to the left but I still get IO address failure, I am trying if there is any other timing violation that might be causing this issue? I have included a scope plot with the NAK.
Hello I.K.,
I capture the plot with Aardvark and I find that the data with the micro-controller is about 1/4 cycle delayed compared to the plot captured with Aardvark. Can you confirm if the data from my micro controller is violating any timing spec requirement?
Thanks
Ranga
Hi Ranga,
From the first provided waveform it looks like there actually may be a rise/fall time violation. The maximum is 300ns. You can reference the fast-mode timing parameters in the table below:
Regards,
I.K.
Hello I.K.,
I measured the rise/fall time again @ 400kHz and it does not seem to violate the timing spec, I'm still debugging the issue to find what might be causeing the "IO address failure".
