TMS320F28377D: tms320f28377d

Hi Sharon,

You may want to reference the C2000WARE I2C EEPROM example 'i2c_ex4_eeprom_polling' or 'i2c_ex6_eeprom_interrupt' in the directory location below. It's a newer example and used the driverLib functions instead of bit-field (which can be more user friendly).

C:\ti\c2000\C2000Ware_4_00_00_00\driverlib\f2837xd\examples\cpu1\i2c

In your waveform I only see the write and command portion. For reading, I would expect a repeated start condition with slave address + read bit followed by data being read from the EEPROM. I strongly recommend trying out the other examples and SW framework.

Best,

Kevin

Hi

Thankyou for the reply.

I have checked with all those examples. 

Even when i call the functions of write and read the waveform looks as below

I think that my issue is, after write, stop signal is generated in software but not reflecting in the waveform.

Hi Sharon,

From the 24LC256 datasheet below which operation are you trying to execute exactly? From your shared waveform it looks like a 'CURRENT ADDRESS READ'. If so, are you expecting to read back 0xFF?

https://ww1.microchip.com/downloads/aemDocuments/documents/MPD/ProductDocuments/DataSheets/24AA256-24LC256-24FC256-256-Kbit-I2C-Serial-EEPROM-20001203X.pdf

In your waveform the 2nd byte has a NACK at the end (F2837x device reading byte from 24LC256) which may be generated instead of a STOP condition if only reading the single byte.

Sharon Mathew said:

I think that my issue is, after write, stop signal is generated in software but not reflecting in the waveform.

What's shown in the latest waveform is a read operation. Do you mean in the original waveform?

Best,

Kevin

Hi Kevin

Am trying for a random read from 24LC256. That means initially i will call write function to write the slave addresses, page addresses and data.

Then i call read fuction which initially contains a write for slave addresses and page addresses. Followed by a read to get data. But whats happening is first write is done and second write is skipped and followed by a read with NAK.

Is there anything to do for repeated write? Am using RM = 0.

Hi Sharon,

I think what's happening is the 24LC256 is not ready to be read from yet. It may take a few read attempts before the 24LC256 responds with an ACK and the F2837x can read the data from it.

Have you tried the 'i2c_ex4_eeprom_polling' example in C2000WARE. without making any changes yet? This example should work with no changes and handles the read commands properly. See below from example:

    //Wait for EEPROM write cycle time
    //This delay is not mandatory. User can run their application code instead.
    //It is however important to wait for EEPROM write cycle time before you initiate
    //another read / write transaction
    DEVICE_DELAY_US(EEPROM.WriteCycleTime_in_us);

Best,

Kevin