K60 I2C-EEPROM write and read

[mehdikarimibiuk]

I have MK60DN512VLL10 that has its I2C1 connected to EEPROM (ST Microelectronics - datasheet pdf) M24256-BWDW6

I removed the pull up on WC signal above.

following the above write sequence in my application I have:

Then based on the read sequence:

My read looks like:

But this returns me 0xff (printf result is 0xff 0x0 0x0 0x0 0x0), that means either I did not write or I cannot read?

As you can see from the above code snippets, my I2C1 is interrupt based:

Can someone tell me what is wrong? And why I cannot write/read EEPROM?

Thanks

Mehdi

[trytohelp]

Hi Medhi,

The problem has been handle through Service Request process.

The SR is 1-1342499221

Have a great day,
Pascal

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[mehdikarimibiuk]

Just to add that as I stated in my last post above, I could write and read but I cannot write and read in sequence. Is there such a rule for I2C EEPROMs to write and read separately?

[trytohelp]

Hi Mehdi,

I received feedback for Proceesor Exper team which could explain the EEPROM behavior.

See below:

It is a feature of the EEPROM device.

Write cycle takes up to 5ms.

It is not responsibility of I2C bus protocol in general.

Have a great day,
Pascal

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[trytohelp]

Hi Mehdi,

Thanks for the feedback and survey.

I shared this information with Processor Expert team and development.

Have a great day,
Pascal

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[mehdikarimibiuk]

Hi @Jorge_Gonzalez

Can you please give me an example based on the EEPROM read/write format above.

based on Figure8 and Figure9 that I showed above, would the following work:

uint8 buffer[4];

buffer[0] = 0x57 //address of the slave

buffer[1] = 0x00 // first byte address

buffer[2] = 0x00 //second byte address

buffer[3] = 0x55 // one byte data

I2C1_SelectSlaveDevice(I2C1ptr, LDD_I2C_ADDRTYPE_7BITS, passin[0]);

Sent = 0;

I2C1_MasterSendBlock(I2C1ptr, passin , sizeof(passin), LDD_I2C_SEND_STOP); //LDD_I2C_SEND_STOP LDD_I2C_NO_SEND_STOP

while (Sent == 0){}

I tried this and it did not work. Do you have any suggestions?

[Jorge_Gonzalez]

Hi Medhi:

Each call to MasterSendBlock() generates a new start signal together with the address, as you see in your scope.

To avoid this try placing together all your data. If you look at Figure 8 above, you may use a unique buffer with [Byte addr + Byte addr + Data in] in the same buffer.

As reference here you have some code:

uint8  pass_buff[3];

uint8  passin;

pass_buff[0] = 0x00;

pass_buff[1] = 0x00;

pass_buff[2] = 0x55;

passin = 0x57;

I2C_SelectSlaveDevice(I2C1ptr, LDD_I2C_ADDRTYPE_7BITS, passin);

Sent = 0;

I2C1_MasterSendBlock(I2C1ptr, &pass_buff , sizeof(pass_buff), LDD_I2C_SEND_STOP);

while (Sent == 0){}

Regards!

Jorge Gonzalez

[Jorge_Gonzalez]

The whole transmission frame is complete according to your scope. Then we need to track down the problem with the debugger.

You can start by setting a breakpoint inside of the OnMasterBlockSent(), just to confirm that it never reaches that line.

Next step is to analyze the code in the interrupt routine PE_ISR(I2C1_Interrupt). I know it is a lot of code, but the problem resides in there.

[mehdikarimibiuk]

n/a

[Jorge_Gonzalez]

I think that should be working.

So, in your last code there are 3 while loops, if I understood correctly you can already passed the first one and now you are stuck in the second while loop?

I do not know what is the cause, but again it would be helpful to check it with the scope.

[mehdikarimibiuk]

Hi Jorge,

So I checked it with scope. Here is what's happening, I cannot do write and read in sequence but I can do them separately. I don't know why? Do you have any idea why?

I wrote to address 0x0000 first time. I wrote the value 0x55. So my 0x55 stored at address zero of my eeprom. I removed write and do the read of that byte. When I do that I can read 0x55 back, which means it wrote to that address and stores it there coreclty.

But do you have any idea why I cannot do this in sequence and it get stuck in the while loop for read?

Here is my latest code:

[Jorge_Gonzalez]

Your code seems OK to me.

Check the bus with your scope to verify the activity.