16 bit value in 2's compliment, confusing

Question

I am working on reading data from a gyroscope and I am getting the data back just fine but the values are not as expected. I am thinking this is due to an error in my coding after searching around for a solution I came up with a post HERE that states:

Make sure you are reading the output registers correctly, the data is a 16-bit value in 2's complement (i.e the MSB is the sign bit, then 15 bits for the value)

This is very confusing to me and I am unsure if my coding is reading the value as its supposed to. I am using the wiringPi I2C library to convert existing code for the Arduino to run on the Raspberry Pi. I have my code below I am hoping someone can tell me if they see the proper attempt to read the 16 bit value in 2's compliment. Inside the getGyroValues function is the only place I could see this happening. Is my code reading the values correctly?

#include <stdio.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
#include <stdlib.h>
#include <stdint.h>
#include <time.h>

#include <wiringPi.h>
#include <wiringPiI2C.h>

#define CTRL_REG1 0x20
#define CTRL_REG2 0x21
#define CTRL_REG3 0x22
#define CTRL_REG4 0x23


int fd;
int x = 0;
int y = 0;
int z = 0;
int main (){



    fd = wiringPiI2CSetup(0x69); // I2C address of gyro
    wiringPiI2CWriteReg8(fd, CTRL_REG1, 0x1F); //Turn on all axes, disable power down
    wiringPiI2CWriteReg8(fd, CTRL_REG3, 0x08); //Enable control ready signal
    wiringPiI2CWriteReg8(fd, CTRL_REG4, 0x80); // Set scale (500 deg/sec)
    delay(100);                    // Wait to synchronize

void getGyroValues (){
    int MSB, LSB;

    LSB = wiringPiI2CReadReg16(fd, 0x28);
    MSB = wiringPiI2CReadReg16(fd, 0x29);
    x = ((MSB << 8) | LSB);

    MSB = wiringPiI2CReadReg16(fd, 0x2B);
    LSB = wiringPiI2CReadReg16(fd, 0x2A);
    y = ((MSB << 8) | LSB);

    MSB = wiringPiI2CReadReg16(fd, 0x2D);
    LSB = wiringPiI2CReadReg16(fd, 0x2C);
    z = ((MSB << 8) | LSB);
}

    for (int i=0;i<10;i++){
    getGyroValues();
    // In following Divinding by 114 reduces noise
    printf("Value of X is: %d\n", x / 114);
    printf("Value of Y is: %d\n", y / 114);
    printf("Value of Z is: %d\n", z / 114);
    int t = wiringPiI2CReadReg8(fd, 0x26);
    t = (t*1.8)+32;//convert Celcius to Fareinheit
    int a = wiringPiI2CReadReg16(fd,0x2B);
    int b = wiringPiI2CReadReg16(fd,0x2A);
    printf("Y_L equals: %d\n", a);
    printf("Y_H equals: %d\n", b);
    int c = wiringPiI2CReadReg16(fd,0x28);
    int d = wiringPiI2CReadReg16(fd,0x29);
    printf("X_L equals: %d\n", c);
    printf("X_H equals: %d\n", d);
    int e = wiringPiI2CReadReg16(fd,0x2C);
    int f = wiringPiI2CReadReg16(fd,0x2D);
    printf("Z_L equals: %d\n", e);
    printf("Z_H equals: %d\n", f); 

    printf("The temperature is: %d\n\n\n", t); 
    delay(500);
}
};

I am wondering if my code is reading the values correctly or not? — Yamaha32088, Mar 03 '13 at 22:41
In `getGyroValues`, you call `wiringPiI2CReadReg16` and you're shifting the MSB by only eight bits. I have no idea if your code is reading the values correctly, but this looks wrong. — Brian Cain, Mar 03 '13 at 22:48

score 1 · Accepted Answer · answered Mar 03 '13 at 22:44

Here's how you could easily combine the most and the least significant bytes, representing the halves of a 2's complement 16-bit integer into an int:

int Bytes2Short(unsigned char msb, unsigned char lsb)
{
  long t = msb * 0x100L + lsb;
  if (t >= 32768)
    t -= 65536;
  return (int)t;
}

score 1 · Answer 2 · answered Mar 04 '13 at 02:44

Use Alexey Frunze's method for clarity, but here's a version with no tests (which may compile to faster inline code in some situations). This pattern works for other sizes of unsigned bit patterns as well—for instance, to handle 12-bit integers, use 0x0800, for 17-bit integers use 0x10000, and so on.

I bundled it with a little driver main to show the outputs.

#include <stdio.h>

int comb(unsigned char msb, unsigned char lsb) {
        long t = ((long)msb << 8) | lsb;
        if (t >= 32768)
                t -= 65536;
        return t;
}

int comb2(unsigned char msb, unsigned char lsb) {
        unsigned long t = ((unsigned long)msb << 8) | lsb;
        t ^= 0x8000UL;
        return (int)((long)t - (long)0x8000);
}

int main(void) {
        printf("%6s %6s\n", "comb", "comb2");
        printf("%6d %6d\n", comb(0, 0), comb2(0, 0));
        printf("%6d %6d\n", comb(127, 255), comb2(127, 255));
        printf("%6d %6d\n", comb(128, 0), comb2(128, 0));
        printf("%6d %6d\n", comb(255, 255), comb2(255, 255));
        return 0;
}

16 bit value in 2's compliment, confusing

2 Answers2