FXLN8371Q - Bare metal example project

Document created by Tomas Vaverka Employee on Feb 3, 2015
Version 1Show Document
  • View in full screen mode

Hi Everyone,

 

In this document I would like to present a simple bare-metal example code/demo for the FXLN8371Q Xtrinsic three axis, low-power, low-g, analog output accelerometer. I have created it while working with the Freescale FRDM-KL25Z development platform and FXLN8371Q breakout board. The FreeMASTER tool is used to visualize the acceleration data that are read from the FXLN8371Q through ADC.

 

This example illustrates:

 

1. Initialization of the MKL25Z128 MCU (mainly ADC, PORT and PIT modules).

2. Simple offset calibration.

3. Accelerometer outputs reading using ADC and conversion of the 10-bit ADC values to real acceleration values in g’s.

4. Visualization of the output values in the FreeMASTER tool.

 

1. The FXLN8371Q breakout board (schematic is attached below) needs to have the following pins connected to the FRDM-KL25Z board:

 

J4-1 (VDD) => J9-4 (P3V3)

J4-2 (XOUT) => J10-2 (PTB0/ADC0_SE8)

J4-3 (YOUT) => J10-4 (PTB1/ADC0_SE9)

J4-4 (ZOUT) => J10-6 (PTB2/ ADC0_SE12)

J4-6 (GND) => J9-14 (GND)

J3-3 (ST) => J9-14 (GND)

J3-4 (EN) => J9-4 (P3V3)

 

The PIT (Periodic Interrupt Timer) is used to read the output data periodically at a fixed rate of ~200Hz. The MCU is, therefore, configured as follows.

 

void MCU_Init(void)

{

     //ADC0 module initialization

     SIM_SCGC6 |= SIM_SCGC6_ADC0_MASK;        // Turn on clock to ADC0 module

     SIM_SCGC5 |= SIM_SCGC5_PORTB_MASK;       // Turn on clock to Port B module

     PORTB_PCR0 |= PORT_PCR_MUX(0x00);        // PTB0 pin is ADC0 SE8 input

     PORTB_PCR1 |= PORT_PCR_MUX(0x00);        // PTB1 pin is ADC0 SE9 input

     PORTB_PCR2 |= PORT_PCR_MUX(0x00);        // PTB2 pin is ADC0 SE12 input

     ADC0_CFG1 |= ADC_CFG1_ADLSMP_MASK | ADC_CFG1_MODE(0x02);             // Long sample time, single-ended 10-bit conversion   

             

         //PIT module initialization

     SIM_SCGC6 |= SIM_SCGC6_PIT_MASK;         // Turn on clock to PIT module

     PIT_LDVAL0 = 52400;                      // Timeout period = ~5ms (200Hz)

     PIT_MCR = PIT_MCR_FRZ_MASK;              // Enable clock for PIT, freeze PIT in debug mode

                    

         //Enable PIT interrupt on NVIC    

     NVIC_ICPR |= 1 << ((INT_PIT - 16) % 32);

     NVIC_ISER |= 1 << ((INT_PIT - 16) % 32);

}

 

2. The simplest offset calibration method consists of placing the board on a flat surface so that X is at 0g, Y is at 0g and Z is at +1g. 16 samples are recorded and then averaged. The known sensitivity needs to be subtracted from the +1g reading to calculate an assumed 0g offset value for Z.

 

void Calibrate(void)

{

     unsigned int Count = 0;

             

     do                   // Accumulate 16 samples for X, Y, Z

     {

        ADC0_SC1A = ADC_SC1_ADCH(0x08);               // Process ADC measurements on ADC0_SE8/XOUT                                                

               while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};

        Xout_10_bit += ADC0_RA;                        

                                                                    

        ADC0_SC1A = ADC_SC1_ADCH(0x09);               // Process ADC measurements on ADC0_SE9/YOUT

               while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};

        Yout_10_bit += ADC0_RA;                        

                                                                          

        ADC0_SC1A = ADC_SC1_ADCH(0x0C);

               while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};    // Process ADC measurements on ADC0_SE12/ZOUT      

        Zout_10_bit += ADC0_RA;

                                               

        Count++;                  

     } while (Count < 16);

             

     X_offset = (float) Xout_10_bit / 16;                             // Compute X-axis offset by averaging all 16 X-axis samples

     Y_offset = (float) Yout_10_bit / 16;                             // Compute Y-axis offset by averaging all 16 Y-axis samples

     Z_offset = ((float) Zout_10_bit / 16) - SENSITIVITY_2G;          // Compute Z-axis offset by averaging all 16 Z-axis samples and subtracting the known sensitivity  

             

     PIT_TCTRL0 = PIT_TCTRL_TIE_MASK | PIT_TCTRL_TEN_MASK;          // Enable PIT interrupt and PIT                                     

}


3. Reading the FXLN8371Q datasheet, the output voltage when there is no acceleration is typically 0.75V and it should typically change by 229mV per 1g of acceleration in ±2g mode. The signal from a 10-bit ADC gives me a number from 0 to 1023. I will call these “ADC units”.  0V maps to 0 ADC units, VDDA (2.95V on the FRDM-KL25Z board) maps to 1023 ADC units and let’s assume it is linear in between. This means that zero acceleration on an axis should give me a reading of 260 ADC units (0.75V / 2.95V x 1023 ADC units) on the pin for that axis. Also, a change of 1 ADC unit corresponds to a voltage difference of 2.95V / 1023 ADC units = 2.884mV/ADC unit. Since the datasheet says a 1g acceleration typically corresponds to 229mV voltage difference, I can easily convert it to ADC units/g:

 

229 mV/g = 229 mV/g x (1023 ADC units) / 2.95V = 79.4 ADC units/g = SENSITIVITY_2g

 

Using this calculated sensitivity and measured offset, I convert the 10-bit ADC values to real acceleration values in g’s in the PIT ISR as follows.

 

void PIT_IRQHandler()

{

     ADC0_SC1A = ADC_SC1_ADCH(0x08);                    // Process ADC measurements on ADC0_SE8/XOUT                                        

         while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};

     Xout_10_bit = ADC0_RA;

                                  

     ADC0_SC1A = ADC_SC1_ADCH(0x09);                    // Process ADC measurements on ADC0_SE9/YOUT      

         while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};

     Yout_10_bit = ADC0_RA;

                                        

     ADC0_SC1A = ADC_SC1_ADCH(0x0C);                    // Process ADC measurements on ADC0_SE12/ZOUT      

         while(!(ADC0_SC1A & ADC_SC1_COCO_MASK)){};

     Zout_10_bit = ADC0_RA;

             

     Xout_g = ((float) Xout_10_bit - X_offset) / SENSITIVITY_2G;         // Compute X-axis output value in g's

     Yout_g = ((float) Yout_10_bit - Y_offset) / SENSITIVITY_2G;         // Compute Y-axis output value in g's

     Zout_g = ((float) Zout_10_bit - Z_offset) / SENSITIVITY_2G;         // Compute Z-axis output value in g's        

      

     PIT_TFLG0 |= PIT_TFLG_TIF_MASK;          // Clear PIT interrupt flag

}


4. The calculated values can be watched in the "Variables" window on the top right of the Debug perspective or in the FreeMASTER application. To open and run the FreeMASTER project, install the FreeMASTER application and FreeMASTER Communication Driver.


CW Variables.JPG.jpg

 

FreeMASTER Variables.JPG.jpg

 

I guess this is enough to let you start experimenting with the FXLN83xxQ family of analog accelerometers. Attached you can find the complete source code written in the CW for MCU's v10.6 including the FreeMASTER project.

 

If there are any questions regarding this simple application, do not hesitate to ask below. Your feedback or suggestions are also welcome.

 

Regards,

Tomas

Outcomes