• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 80 subscribers
  • Views 578 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ez430 chronos acceleration sensor, axis positions

iggarpe
Prodigy 90 points

Hi,

I just received my ez430 chronos, and while reviewing the documentation I realized I just can't find information on how the acceleration sensor axis are positioned. The default firmware shows this behaviour (could please someone confirm this is the expected behaviour?):

1- With the watch screen in horizontal position looking upwards, the X and Y readings are approximately zero.

2- If you tilt the watch sideways the Y reading varies, going up to 0.85 and then back to zero when the screen is again horizontal but looking downwards.

3- If you tilt the watch backwards (i.e. rotate screen towards you) the X reading varies, going up to 0.88 and then back to zero when the screen is again horizontal but looking downwards.

(so far this makes sense to me)

4- In the position described in (1) the Z reading is approx. 0.40, if you start to turn the watch towards you the reading goes down to zero, which is reached when the screen is facing you (about 45º turn from the initial position). If you continue turning the watch until the screen is facing downwards, the reading is 1.40.

Any help understanding why the Z reading behaves so differently from the other axis will be very welcome.

Thanks.

  • 0
    Guru 227245 points

    iggarpe said:
    Any help understanding why the Z reading behaves so differently from the other axis will be very welcome.

    I guess there is some offset on the Z axis meant to calibrate for gravity. If you just subtract 0.4 from the reading, the results will cling to the other directions.

  • 0 in reply to Jens-Michael Gross
    Prodigy 90 points

    That doesn't actually make much sense. In the static rotation readings the only force acting is gravity... and gravity is actually what you are measuring on each axis subsensor, multiplied by the cosine of the angle between that subaxis and the horizont.

  • 0 in reply to iggarpe
    Guru 227245 points

    Right. So initially, the z axis should measure 1G (whatever numerical reading this results in) while the others are showng 0. If you flip the clock, teh Z axis should read -1G.
    X and Y should read 0 and +-1G when being turned by 90 or 270°. You didn't post the 270° value, so I don't know about range and sign and symmetry (I don't know the sensor).

    The difference between Z axis 0° and 180° seems to indicate an offset on the Z axis output signal.

    Anyway, when using the sensor for acceleration only (but NOT including rotation), it might make some sense to calibrate the default 1G so it will show relative acceleration values for z direction too.

    Extracting orientation and accelleration/rotation speeds from plain absolute sensor reading when allowing movements in all 6 orientations (12 directions) is a major mathematical task. Not impossible, of course, but rather complex. So in most applications, hotizontal orientation will be assumed (holding the clock horizontally while looking at it)

    There is, however, the possibility that the Z axis sensor has been damaged by dropping the clock flat on the floor. This kind of shock might introduce and offset or nonlinearity in one axis while the others are unharmed. It of course depends on the type of the sensor mechanism.

  • 0
    Prodigy 40 points

    If there's still any interest in this subject, please take a look at:

    http://e2e.ti.com/support/microcontrollers/msp43016-bit_ultra-low_power_mcus/f/166/p/139739/503414.aspx

     

**Attention** This is a public forum