ADS1148 Thermistor

Oscar,

You could connect the thermistors to the device like this:

Shown in this diagram is the measurement of four RTDs (thermistors would be measured in the same way). The measurements are ratiometric, so that the ADC value becomes a ratio between the RTD value and the reference resistor RREF.

To make the measurement, let's use RTD1 as an example. First, set the IDAC to AIN0. This will drive both the RTD1 and RREF. As long as there is no leakage current, then the current through each will be the same. The ADC result will be:

ADC result = (RTD1/RREF)*2^15

Again, this measurement is ratiometric. The result is a ratio between the resistors, and you do not need to calculate the voltages to get a result. There are two important things to note. First, the accuracy of the measurement is directly related to the accuracy of RREF. To make this measurement you would want to use a precision resistor for RREF. Second, won't be able to measure a voltage larger than the voltage on REFP0/REFN0. Therefore, RREF needs to be larger than the largest expected value on the RTD. If you are measuring a 10k RTD (or thermistor) that has a max value of 15k, then RREF needs to be 15k or larger. You would probably need to use the 100uA setting for the IDAC to keep the input within the range of operation.

Since you are using a two-wire element, the resistance in the lead wire is always an error term. The way that it is drawn, you won't be able to use any input filtering. Because input filtering usually requires series resistance, the resistance reacting with the excitation current would give an additional error. There are alternate connections that would allow for filtering, but you would only be able to make three measurements and need to use the IEXC1 and IEXC2 pins, and alternate input to drive the RTD. This would bypass the series resistance so that it doesn't contribute to the error.

Going back to the original diagram, once you make the first measurement, you move to the next RTD. The IDAC is connected to AIN2 and you make a similar measurement.

Joseph Wu

Thanks for your answer Joseph, helped me alot.

I Have a question about the analog power supply, I understand that the power supply would have to be unipolar in this case, is this correct? also I would know how is the connection that you mention for filtering three sensors using IEXC1 and IEXC2 pins?

Best regards

Oscar

Hi Joseph Wu,

I am testing SPI communication in ADS1148evm and implementing the diagram showed above with a microcontroller(Renesas R7F124). I am using the pins: CS, SCLK, DIN, DRDY, DOUT, START and RESET.

I want get the ADC result = (RTD1/RREF)*2^15
But always I am getting value = 0xFFFF in my RxSPI variable.

Could you help me?

This is my code:

    tx_thread_sleep (1);    //Delay 10ms
    g_ioport.p_api->pinWrite(IOPORT_PORT_01_PIN_04,HIGH);           //START pin high;

    
    g_spi0.p_api->open(g_spi0.p_ctrl, g_spi0.p_cfg);  //Opening and configuring SPI
    tx_thread_sleep (1);

    g_ioport.p_api->pinWrite(IOPORT_PORT_01_PIN_07,LOW);           // CS LOW
    tx_thread_sleep (1);

   
    g_ioport.p_api->pinWrite(IOPORT_PORT_01_PIN_06,LOW);           //Reset the ADS1148;
    tx_thread_sleep (1);
    g_ioport.p_api->pinWrite(IOPORT_PORT_01_PIN_06,HIGH);           // RESET pin goes high
    tx_thread_sleep (1);

    
    spi_Buffer_Write[0]=0x17;                   
    g_spi0.p_api->write(g_spi0.p_ctrl,&spi_Buffer_Write[0],1,SPI_BIT_WIDTH_8_BITS);      //Send SDATAC
    tx_thread_sleep (1);

//WREG command;

spi_Buffer_Write[0]=0x40;             
    spi_Buffer_Write[1]=0x03;               //Second Command Byte (Number of bytes to be written – 1.)

    spi_Buffer_Write[2]=0x01;               //MUX0   Burn-out current source off,  AIN0 +,  AIN1 -.
    spi_Buffer_Write[3]=0x00;               //VBIAS  00000000: Bias voltage is not enabled (default)
    spi_Buffer_Write[4]=0x20;               //MUX1   Internal oscillator,   Internal reference on, REFP0 and REFN0, Normal Op
    spi_Buffer_Write[5]=0x05;               //SYS0   PGA = 1, DR = 20 SPS

    g_spi0.p_api->write(g_spi0.p_ctrl,&spi_Buffer_Write,3,SPI_BIT_WIDTH_8_BITS);
    tx_thread_sleep (1);//10 ms

    //IDAC0—IDAC Control Register 0 (offset = 0Ah)
    spi_Buffer_Write[0]=0x4A;               //0Ah IDAC0   //First Command Byte: 0100 rrrr, where rrrr is the address of the first register to be written.
    spi_Buffer_Write[1]=0x04;               //Second Command Byte (Number of bytes to be written – 1.)

    spi_Buffer_Write[2]=0x02;               //0000, 0: DOUT/DRDY only as Data Out, 010: 100 μA
    spi_Buffer_Write[3]=0x2F;               //IDAC1   10x0: IEXC1 (ADS1148 only), 10x1: IEXC2 (ADS1148 only)
    spi_Buffer_Write[4]=0x00;               // GPIO Configuration Register Field Descriptions (ADS1148)
    spi_Buffer_Write[4]=0x00;               // GPIO Configuration Register Field Descriptions (ADS1148)
    spi_Buffer_Write[4]=0x00;               // GPIO Configuration Register Field Descriptions (ADS1148)

    g_spi0.p_api->write(g_spi0.p_ctrl,&spi_Buffer_Write,3,SPI_BIT_WIDTH_8_BITS);
    tx_thread_sleep (1);

    g_ioport.p_api->pinWrite(IOPORT_PORT_01_PIN_07,HIGH);           // CS high.
    tx_thread_sleep (1);

while (1)
    {

if(DRDY==0)

            spi_Buffer_Write[0]=0xFF;               //Nop
            spi_Buffer_Write[1]=0xFF;               //Nop

            g_spi0.p_api->writeRead(g_spi0.p_ctrl,&spi_Buffer_Write,RxSPI,2,SPI_BIT_WIDTH_8_BITS);
            tx_thread_sleep (1);

DRDY=1;

}

Hello Joseph

I have a question, why the ADC resolution in your equation (ADC result = (RTD1/RREF)*2^15) is 2^15 if the ADS1148 has a 16 bits ADC?

Best regards

Oscar