ADS124S08EVM: Interfacing problem with ADS124S08 EVM board to arduino UNO

Hi Shushmita,

Make sure that there is a ground connection between the EVM and the UNO. Also verify that the RESET pin is truly high, and make sure that MOSI is connected to J3 pin labled SDI and MISO is connected to pin labeled SDO.

I would suggest that you verify that the communication is working as expected by reading back the registers after they are written and monitoring the communication with a logic analyzer of oscilloscope. You may also want to toggle CS between commands just to make sure that the device is not stuck in a weird state. For example, you are not giving any delay following the setting of the RESET pin high. You must wait a minimum of 4096 tCLK periods (1ms minimum) before the first edge of SCLK after the RESET pin goes high (see section 7.6 on page 14 of the ADS124S08 datasheet under RESET PIN). If communication is taking place prior to the completion of the delay time there could be corrupted communication.

I'm not quite sure I follow your 'loop' code. You have a for loop that never has an ending point, and the only time you issue the START command is when i=0. So if communication is corrupted, the device may never properly decode the START command. In the collection portion of the loop, you issue the RDATA command then wait 1000us. There is no reason to wait after the RDATA command as 4 more conversions will complete before you read the data.

Best regards,
Bob B

Hi Sushmita,

See my comments below.  It appears that you have the RDATA and START commands swapped.  You also need to make sure that you wait for each conversion to be competed before reading the results.

Best regards,

Bob B

Sushmita Chaudhary said:

Hi Bob,

Thank you for your explanation. Actually I have changed my code also,as you said about my  unending for loop that is my mistake,that also I changed. And also I have given the delay after RDATA command. But when I have uploaded the code on arduino, I observed on serial monitor that only 3 digit data is coming and our ADC is 24 bit so it should  range from 0 to 2^23 but data is coming between 0 to 255. My code is as follows:

#include <SPI.h>

#include <TimerOne.h>

#define START 6

#define RESET 7

const int CS = 4;

int stat;

int stat1;

int stat2;

SPISettings settings1(4000000, MSBFIRST, SPI_MODE1);

void setup() {

SPI.begin();

pinMode(CS,OUTPUT);

pinMode(RESET, OUTPUT);

pinMode(START,OUTPUT);

digitalWrite(RESET , HIGH);

delay(1000); // [Bob] This delay must be long enough for the device to start prior to beginning any communication if delay refers to 1ms per count, then the delay will be sufficient.

Serial.begin(9600);

SPI.beginTransaction(settings1);

digitalWrite(CS, LOW);

digitalWrite(START,LOW);

}

void loop() {

for(int i=0;i<=5;i++)

{

if(i==0){

SPI.transfer(0x42); // WREG register adress

SPI.transfer(0x02); // WREG byte number -1

SPI.transfer(0x24); // WREG write AIN2-AIN4

SPI.transfer(0x07); 128 gain

SPI.transfer(0x1E); // DATA RATE 4000 SPS

SPI.transfer(0x12); //RDATA command [Bob] this command is meaningless for two reasons.  One is no conversion has yet taken place and the second reason is the command was not completed.  RDATA is a command to be followed by 3-bytes of data read from the device.  What should have been sent here should have been the START command (0x08) to start a conversion.  The command here should be:

SPI.transfer(0x08); // start converting command

 

delayMicroseconds(1000);

}

else

{

SPI.transfer(0x08); // start converting command [Bob] Here you start the conversion but give no time for the conversion to complete, so at this point any data read from the device will be meaningless.  Here you should have issued the RDATA command to read the data, followed by 3 bytes of NOPs to retrieve the data.  So you have the START and RDATA commands swapped in the code.  Also, as you are not using CS the previous command was not completed so this is actually the first retrieval byte of the RDATA command.  When holding CS low you should use the SPI timeout feature which is enabled by setting the TIMEOUT bit to 1 in the system control register (0x09).  This will allow an incomplete command to be canceled resetting the SPI interface after the timeout period (section 9.5.1.6 on page 63 of the ADS124S08 datasheet.)

The command should be:

SPI.transfer(0x12); //RDATA command 

stat = SPI.transfer(0) ; // get data from miso [Bob] The original RDATA command was still incomplete, so this would actually be the middle byte of 3 from the ADS124S08.  The remaining portions of the command should be:

stat = (stat<<8) | SPI.transfer(0); // Using the RDATA command the first save to stat is the MSB, and this is the mid-byte

stat = (stat<<8) | SPI.transfer(0); // Using the RDATA command the save to stat includes the LSB, which is now 24 bits of 32...now you need sign-extension for negative values.

// Add sign-extension code here to achieve negative values

Serial.println(stat); // [Bob] As you are only capturing 1 byte in your original code you are only transferring one byte.  As 'stat' is declared as an integer, you must also sign-extend the 24 bit binary 2's complement result to 32 bits.

delayMicroseconds(250); // [Bob] You will also need an additional delay to allow the next conversion period to end.

}

}

SPI.endTransaction();

}

 

 

Hi Sushmita,

Your code appears to be ok, however in some way you need to verify your communication and that the device is operating.  Using an oscilloscope or logic analyzer will help you to determine if you are communicating correctly.  You cannot short-cut this step.  Please send me shots of your communication.  Check DRDY with an scope or analyzer to make sure that it is pulsing at the data rate you have programmed to the device.  If it is not, then you are either missing a supply, a required pin setting is not as it should be (possible wiring issue), or you are not communicating to the device properly.

Best regards,

Bob B