CONTROLSUITE: I2C routine

William, 

Which device are you trying this on? 

Instead of the lines starting from  "I2caRegs. I2CMDR. all = 0x2C20;" please try using the folllowing lines of code

//
// send Start condition
//
I2caRegs.I2CMDR.bit.STT = 0x1;

//
// Read the received data into RX buffer
//

Uint16 count =0;
while(count < msg->NumOfBytes)
{
    if(I2caRegs.I2CSTR.bit.RRDY ==0x1)
   {
     *(msg->MsgBuffer + count) = I2caRegs.I2CDRR.all;
      count++;
   }
}

//
// Send STOP condition
//
I2caRegs.I2CMDR.bit.STP = 0x1;
while(I2caRegs.I2CMDR.bit.STP != 0x0);

Best Regards

Siddharth

Dear Siddharth

Customers use the way you advice and he has  two questions that need help.

1、 If i2caregs is not set during initialization I2COAR = 0x0050; Unable to receive data. 

2、 After reading a data from DRR (it is correct), RRDY is always 0 and cannot continue to read data. 

thank you

Best Regards,
William

William,

I2COAR is used to specify own address. Which slave address is it trying to communicate with? If it does not find any such address on the bus, then it won't receive the data. 

RRDY should be set each time it receives a new data. Are you not observing this?

Best Regards

Siddharth

Dear Siddharth

Thanks for your reply , our customers has still query below when he understand you reply before.

As stated in the reply, if there is no error in sending the requested EEPROM information, the EEPROM should send the requested data bytes, the DRR register should be read a byte of data, and the EEPROM sends another byte of data until the requested CNT bytes are sent, and the internal CNT is reduced to 0. But when I first received a byte of data, the RRD was 1, and then it was always 0, indicating that the DRR received only one byte. I don't know why? On the other hand, I don't know why I have to set oar to receive data? Only one EEPROM is connected to my I2C bus. I don't think the reliability of I2C using query method is good. How about the reliability of analog GPIO? Can it be used in FOC project? Because I didn't make a mistake in this way.

Thanks again.

Best Regards,
William

Hi Williams, 

Request the customer to look at the I2C Status register I2CSTR register and see if anything is flagged.  I2COAR is used to set own address. Whenever an address is sent on the I2C bus , the own address is used to determine whether it is intended for the particular slave.  

The RRDY bit should be set each time it receives a new data. What is the configuration in the I2CMDR register?

Best Regards

Siddharth

Hello Siddharth

I have a new problem to be solved urgently! Tms320f28069f chip. The routine adopts motorware FOC hvkit lab10e. I hope to enable tz1-tz3 interrupt in epwm control. I modified Hal Void Hal in C_ Setupfaults (hal_handle), and the interrupt vector and interrupt function have been set. The code is as follows:

The routine has called Hal in main()_ setupFaults(). I found a strange phenomenon: when running the program, immediately enter the Tz interrupt program, after examination, found in main () middle note.
HAL_ Hal is called when setparams()_ When setparams(), PWM - > tzfrc | = PWM_ TZFRC_ OST_ BITS; Make tzflg = 0x4 and generate a software mandatory event. I don't understand what this means. What should I do to enable tz1-tz3 normally? Wait for your reply!

Continue to do the epwm TZ interrupt experiment. After entering the interrupt, clear the interrupt, exit the interrupt, enter the interrupt again, and keep reciprocating. I read the state of gpio12-14 in GPIO mode to determine that it is a high-level state, and check it with the instrument. Why can't I clear the interrupt and generate repeated entry interrupt? Here is my interrupt function code:

interrupt void igbtTz1ISR(void)
{
Temp_count ++;
// HAL_tzClearFlag(halHandle, PWM_Number_1, PWM_TripZoneFlag_Global);
// HAL_tzClearFlag(halHandle, PWM_Number_1, PWM_TripZoneFlag_CBC);
HAL_tzClearFlag(halHandle, PWM_Number_1, PWM_TripZoneFlag_OST);
HAL_tzClearFlag(halHandle, PWM_Number_1, PWM_TripZoneFlag_Global);
HAL_pieAckInt(halHandle, PIE_GroupNumber_2);

return;
}

Thank you.

Best Regards,
William

William, 

Will forward your query to PWM expert.

Best Regards

Siddharth

Dear TI 

Thank you very much for your great support and also thanks to TI Technology for the reply. But this still doesn't solve my problem. Calling HAL_tzClearFlag() in my interrupt function also calls PWM_clearTripZone(); modify this function according to the scheme proposed by TI technology, the code is as follows:
void PWM_clearTripZone(PWM_Handle pwmHandle, const PWM_TripZoneFlag_e tripZoneFlag)
{
PWM_Obj *pwm = (PWM_Obj *)pwmHandle; ENABLE_PROTECTED_REGISTER_WRITE_MODE;
// set the bits
// pwm->TZCLR |= tripZoneFlag;
pwm->TZCLR = (PWM_TZCLR_OST_BITS | PWM_TZCLR_CBC_BITS | PWM_TZCLR_INT_BITS); DISABLE_PROTECTED_REGISTER_WRITE_MODE; return;
} // end of PWM_clearTripZone() function
After testing, the Tz interrupt is still being entered frequently.
Register state during test:
1 Use breakpoints to check that TZFLG is 0x5 when entering an interrupt, and 0x0 when exiting an interrupt
TZSEL=0x0100
TZCTL = 0x000A
TZEINT=0x4
ETSEL=0x0900
ETPS=0x0101
ETFLG=0x4

It means that the TZ configuration is correct and the data feedback is correct, but a new interrupt is generated immediately after exiting the interrupt, TZFLG= 0x5, indicating that an interrupt event has occurred. TZFLG=0x4 of pwm2/pwm3 (because TZ2 and TZ3 are set during initialization), but no interrupt is generated, indicating that pwm2/pwm3 also generated OST events. CMPCTL is always changing. In addition, an experiment was done to verify the possible problems with the TZ1 pin, set TZ1 (GPIO12) to XINT1, it can normally generate interrupts, and will not enter interrupts frequently, therefore, the pin problem can be ruled out. TZ frequently goes into the possibility of interruptions.
In view of the above situation, it means that the event is really triggered. In addition to the TZ1/2/3 pin triggering of the TZ event, there is also a DC (digital comparator) and Force that can be triggered. I don't know if the DC can be turned off, it will generate a normal pwm signal. Affect?

Best Regards,
William