BNO055 calibration. Am I missing any configuration register?

I am trying to get some reading from the IMU BNO055, the implementation from ADA Fruit below.

https://cdn-learn.adafruit.com/downloads/pdf/adafruit-bno055-absolute-orientation-sensor.pdf

It seems like I am not able to pass the calibration step.

The BNO055 manual can be found below: https://cdn-shop.adafruit.com/datasheets/BST_BNO055_DS000_12.pdf

I am not using libraries but communicating using I2C directly and reading and writing from the device registers.

The configuration I am doing is as follows:

// The opening of the device and so on, works, so the line below is not a problem.
    if (ioctl(data_exchange->file_imu_sensor, I2C_SLAVE, data_exchange->imu_addr) < 0) 
     {            
     printf("Impossible to communicate with the IMU in the i2c. Make sure that the address you have is correct!\n");                 
     exit(1);
     }
    // Here is where I start with the configuration. First I reset the device.
    //SYS_TRIGGER Register. ADD 3F
    //CLK_SEL (0 internal oscillator) REST_INT(reset interrumptions)RST_SYS(1 to reset) x x x x Self_TEST(1 to make a self test.)
    // 00xxxx0    
    buf[0] = 0x3F;    
    buf[1] = 0x20; //00100000    
    write_buf(data_exchange,buf,3);    
    printf("IMU WAS COMMANDED TO RESET");    
    sleep(1); //It needs some time to reset. With one sec should be sufficient.

    // Here I start with the real configuration.
    // G range 2G xxxxxx00
    // BW 7.81Hz  xxx000xx
    // Op mod normal 000xxxxx
    //So I will write a 0 to the register ACC_condfi    
    buf[0] = 0x0D;    
    buf[1] = 0x00;    
    write_buf(data_exchange,buf,3);

    // Now unit selection.UNIT_SEL Page 30.
    //Accelaration m/s2 xxxxxxx0b
    // Magnetic field in micro Teslas (always)
    // Angular rate defrees ps xxxxxx0xb
    // Euler angles Degress> xxxxxxxb
    // Quaternion in Quaernion units always.
    //Temperatyre deg xxx0xxxxb
    // NPI The data output format can be selected by writing to the UNIT_SEL register, this allows user to switch between the orientation definition described by Windows and Android operating systems
    // Windows> 0xxxxxxxb
    // Android> 1xxxxxxxb. Page 30.
    // Bits 5 and 6 are reserved. So it does not matter. So we write a 0.    
    buf[0] = 0x03;    
    buf[1] = 0x00;    
    write_buf(data_exchange,buf,3);

    //We now need to set the operation mode:
    //Register OPR_MODE Page 21.
    //Fusion mode NDOF. xxxx1100b    
    buf[0] = 0x1C;    
    buf[1] = 0x0C; //00001100    
    write_buf(data_exchange,buf,3);

    // GYR_Config_0    
    buf[0] = 0x0B;    
    buf[1] = 0x00; //00001100    
    write_buf(data_exchange,buf,3);


    // PWR_MODE. Normal mode is xxxxxx00b    
    buf[0] = 0x3E;    
    buf[1] = 0x00; //00000000    
    write_buf(data_exchange,buf,3);

Now, I am supposed to wait for the internal calibration of the chip to be performed. But it seems like it does not finish doing that.

I am checking it by means of checking the register 0x35, which should be >0 if the calibration was finished.

 char buf[10];
 int status=0;         
 read_buf(dataset_pointer,0x35,3,buf);         
 status=buf[0];
 if(status>0){
         //Here is where I perform all the reading and so on.
    }

I also followed this code:

https://github.com/fgallardo/BNO055/blob/master/BNO055%2BBMP280_Basic_AHRS_t3.ino

Which also writes a 1 to the PAGEID register (0x07) and then a 0 after writing to the configuration registers.

Am I missing something in the configuration? I checked on the Internet and I found some suggestions (as per the IMU manual, too) about making an 8 symbol in the air with the device, which I did, but the calibration is not finished anyway.

I don't think it is a power issue (I found some references about that in the Bosh forums) because I connected VIN to 5.0V, 3VO to 3.3V and GND to GND.

Any comment on this would be of great help. Am I configuring the device wrong? Am I missing any register?

Thanks!