I2C Trouble

[TucsonDon]

I am trying to setup a rtc using I2C but am having issues in getting it to work. I used the code that was generated by MPLab 4.01 code config. Using on an EasyPICv7 board with a PIC18F46K22 and mikro click rtc.

If I run the code with the debugger I can see on the scope (picture) that it starts but when I pause the program after no further activity the pc is at line 663 of the i2c1.c file (while(i2c1_state != S_MASTER_IDLE); )


I don’t think that the master is getting an acknowledgment but unsure as to why.
I have tried to provide all the info but if I missed something forgive me.

 

[Pommie]

When ever I use I2C in the debug environment it hangs due to it being interrupted mid transmission. So, in my init routine I check for the data line being held low and send clock pulses until it is released.
Here's my init code,

void I2cInit(){
    I2cDataTris=1;              //Make data input
    I2cClockTris=1;             //and clock
    while(!I2cData){            //if something holding line low
        I2cClock=0;             //then pulse clock line
        I2cClockTris=0;
        NOP();
        NOP();
        I2cClockTris=1;
    }
    RC0PPS=0x18;        //RC0=SCK/SCL
    RC1PPS=0x19;        //RC1=SDO/SDA
    SSP1CLKPPS=0x10;    //SSP1CLK = RC0
    SSP1DATPPS=0x11;    //SSP1DAT = RC1
    SSPADD=80;      //8000/100;
    SSPSTAT=8;      //I2C Master clock = Fosc/(4*SSPADD)
    SSPCON1=0x28;
    SSP1IF=0;
    BCL1IF=0;
    SSP1CON1bits.SSPEN=1;
}

Mike.

 

[Ian Rogers]

In my experience, Interrupts on I2C and SPI are not really necessary.. Especially at high speeds..

I read / write the clock very quickly and it doesn't affect program flow much at all... Clocks only need to be read once / twice a second as they only store seconds...