ADS8688 flow chart

Hi,

I see the state diagram just showing how to switch between modes. I would like see how to make the ADS8688 works from the start. After reading the datasheet, I have the code sequence flow as follow:

First, 

1. Sending reset 8500h to command register

Second, initialize the ADS8688. write to program register with commands

1. The auto scan seq is left default to scan through all channels .
2. Channel power down is left default to enable all ADC channels.
3. Feature select is left default.
4. Range select for all ADC channels = 0 to 0.25V x Vref

• Writing to Channel 0. Data  = 0505h. Adding 1 for write to the 8th bit, the data send to SDI is 0B05h

.

.

• Writing to Channel 7. Data  = 0C05h. Adding 1 for write to the 8th bit, the data send to SDI is 1905h

Third, sending command to command register.

1.  Following figure 82, page 41 datasheet, sending A000h to the command register for AUTO_RST mode.
2. sending a dummy bytes (FFFFh) to the SDI in order to receive data from the SDO.
3. Sending the NO_OP (0000h) to the command register.
4. Sending a dummy bytes (FFFFh) to the SDI in order to receive ADC conversion data from channel 0.

Repeat 4 and 5 until channel 7.

I use 8 bit 8051 MCU to interface with the ADS8688. following is some of the code sequences

// CS = chip select bit

// SPI0DAT = SPI buffer

// SPIF = SPI interupt flag

// NOP = no operation 

CS = 0;                              //Enable the slave select

 //Sending command to program register to config the channel 0 range = 0.25V x Vref

SPI0DAT = 0x0B;           // First byte

while(SPIF == 0);           //Wait for the transmission complete

SPIF = 0;                         //Reset the flag

SPI0DAT = 0x05;           // Second byte

while(SPIF == 0);           //Wait for the transmission complete

SPIF = 0;                         //Reset the flag

// Sending a dummy byte to complete 24 clocks
SPI0DAT = 0xFF;         
while (SPIF==0);          //Wait for the transmission complete
SPIF = 0;                        //Reset the flag

CS = 1;                           //Deselect CS

NOP  = 0;                       // Adding few ns delay before activate the slave select

NOP  = 0;

CS = 0;                           // Select CS

// Repeat previous code (different data according to table 15 datasheet) to complete range setting for all channels

CS = 1;                           // Deselect CS

// sending command to command register to select AUTO_RST (according to state diagram)

CS = 0;                           // select CS

SPI0DAT = 0x85;
while (SPIF==0);
SPIF = 0;
SPI0DAT = 0x00;
while (SPIF==0);
SPIF = 0;

// Sending dummy bytes to complete the 32nd clocks cycle
SPI0DAT = 0xFF;
while (SPIF==0);
SPIF = 0;
SPI0DAT = 0xFF;
while (SPIF==0);
SPIF = 0;

CS = 1;                           // Deselect CS

NOP  = 0;                       // Adding few ns delay before activate the slave select

NOP  = 0;

// sending command to command register with NO_OP to continue sequence control

CS = 0;                           // select CS

SPI0DAT = 0x00; 
while (SPIF==0); 
SPIF = 0; 
SPI0DAT = 0x00; 
while (SPIF==0); 
SPIF = 0;

// Sending dummy bytes to complete the 32nd clocks cycle with interrupt enable to read data at the SDO line

EIE1 |= 1;                              // Enable SPI interrupt
SPI0DAT = 0xFF; 
while (SPIF==0); 
SPIF = 0; 
SPI0DAT = 0xFF; 
while (SPIF==0); 
SPIF = 0;

CS = 1;                           // Deselect CS

//repeat with NO_OP until all the channel ADC count are acquired.

  The data for ADC channels receives  is not right and not consistence. Please advise if you have any recommendation. This is the first time I deal with this ADS8688.

Thanks.

Hi Shridhar,

I did try the flow you recommended. The received data on the SDO line does not relevant with the voltage input. I have almost the same data every time I try to read the ADC register but the 16 bit count is way up somewhere, not even near the calculated value. For example, there is 1V at the ADC input, The converted data I receive is 22785 counts instead of 6400 counts. When I increase the input voltage to 2V, the count is drop instead of increase.

I think the problem I have is similar to another thread "ADS8688 Problem". I monitoring that thread but have not seen any solution yet.

Also, there is another thread I create "ADS8688 SPI Clock" showing another problem I have when trying to read the content of a register. 

The MCU always read the last bit on the SDO line with the value of 0. e.g. register 01h if default is equal to FFh. when I try to read it, the data I receive always = FEh (11111110)

I finally got it something positive from this ADS8688. However, the received data is shifted left by 1.

For example, I set the range = 0 to 5.12V. I adjust the ADC input  =  2.084V. The 16 bit count supposed to be equal to 2.084 / 78.125uV = 26675 counts. However, the data I receive is 53350 count.

Also, I notice that, the receive count = 65534 @ half of full range (2.56V). If I increase the input voltage pass 2.56V, the ADC count start over from 0 and reach 65534 when input voltage reach full range.

Is there a faster way to get help from TI ? 

I think the SPI logic of the ADS8688 has a bug because it samples the SDI line AND clocks data on the SDO line on the same falling edge! Which is not correct SPI behavior.  Correct behavior for CPOL = 0 CPHA=1 according to wikipedia:

en.wikipedia.org/.../Serial_Peripheral_Interface_Bus

At CPOL=0 the base value of the clock is zero

• For CPHA=0, data are captured on the clock's rising edge (low→high transition) and data is output on a falling edge (high→low clock transition).
• For CPHA=1, data are captured on the clock's falling edge and data is output on a rising edge.

At CPOL=1 the base value of the clock is one (inversion of CPOL=0)

• For CPHA=0, data are captured on clock's falling edge and data is output on a rising edge.
• For CPHA=1, data are captured on clock's rising edge and data is output on a falling edge.