Simple custom I2C interface controller from MSP430 Master to a digital camera

Leo is right! Just had a look in the user's guide - even the chapter of the USI module that can handle 16 bit transfers says that USI16B has to be cleared for I²C which means 8 bits only. And this makes sense, of course, because you have to handle the (N)ACK from the other side.

Dennis

Dan,

I²C is I²C - I don't think you have to modify anything to get it working. Just configure the USCI of the MSP430F5438A for I²C operation.

Samples for doing this with your MSP430F5438A can be found here:

• msp430x54xA_uscib0_i2c_04.c           USCI_B0 I2C Master RX single bytes from MSP430 Master
• msp430x54xA_uscib0_i2c_05.c           USCI_B0 I2C Slave TX single bytes to MSP430 Slave
• msp430x54xA_uscib0_i2c_06.c           USCI_B0 I2C Master TX single bytes to MSP430 Slave
• msp430x54xA_uscib0_i2c_07.c           USCI_B0 I2C Slave RX single bytes from MSP430 Master
• msp430x54xA_uscib0_i2c_08.c           USCI_B0 I2C Master TX multiple bytes to MSP430 Slave
• msp430x54xA_uscib0_i2c_09.c           USCI_B0 I2C Slave RX multiple bytes from MSP430 Master
• msp430x54xA_uscib0_i2c_10.c           USCI_B0 I2C Master RX multiple bytes from MSP430 Slave
• msp430x54xA_uscib0_i2c_11.c           USCI_B0 I2C Slave TX multiple bytes to MSP430 Master
• msp430x54xA_uscib0_i2c_12.c           USCI B0 I2C Master TX/RX multiple bytes to/from MSP430 Slave Repeated Start
• msp430x54xA_uscib0_i2c_13.c           USCI B0 I2C Slave RX/TX multiple bytes from/to MSP430 Master Repeated Start

You can send as many bytes as you want, but the I²C interface has a (N)ACK between every byte.

Dennis

There is always an address, even with only one slave connected. This slave address is given by the slave itself. It will respond only when beeing addressed. Where are your examples from? Those sub-routines maybe come from Energia? Not sure, have never used it. I prefer writing everything myself. And those "constants" you are mentioning are definitions/macros. They are easier to read expressions for registers, bits in registers or actions on registers.

Just an example:

#define LED_ON (P1OUT |= 0x01)

void main( void )
{
  ...
  LED_ON;
  ...
}

Same for individual bits or bit fields in a register - what is easier to read?

UCA0CTL1 |= UCSWRST;

or

UCA0CTL1 |= 0x01;

Both is the same, but the first one lets you know that it is the software reset bit - the 0x01 only tells you that it is bit 0, but who knows all the bits in a register?

The reason for the possibility of using UCSWRST is a definition in the header file of the processor you will include in your project which simply says:

#define UCSWRST 0x01

In your case these expressions will define registers and bits of them.

Dennis