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.

TMAG5170-Q1: Angle detection accuracy

Part Number: TMAG5170-Q1
Other Parts Discussed in Thread: TMAG5173-Q1, TMAG6180-Q1

Hello TI  expert,

I saw TI also have Hall sensor  with automotive grade , does Ti has Hall sensor angle detection less than ±0.5° in full temperature range?

Basic requirement:

my product need flip left and right ± 25° from Z axis, the angle detection accuracy must be less than ±0.5° in full

 temperature range and life time.

As I saw the TMAG5170-Q1 can't meet my requirements, so Can you recommend a suitable angle hall sensor to me ?

Thanks!

  • Thanks for reaching out on E2E.

    A number of factors will influence the overall angle accuracy including the mechanical configuration.  To achieve the best results, it is typically recommended to place a diametric magnet on the axis of rotation, and to similarly place the sensor adjacent on the axis of rotation.  The result will be a rotating magnetic field with constant magnitude.  This is the best case to achieve what you are looking for.  

    This position is also the most forgiving for mechanical tolerances.  The ultimate goal when calculating the angle is to achieve sine and cosine outputs with equal amplitude, and perfectly 90 degrees separated in phase.  If there is amplitude mismatch you may see some non-linearity in your angle calculation.

    The integrated CORDIC in TMAG5170-Q1 has a resolution of 0.25 degrees, and in TMAG5173-Q1 the resolution is 0.0625 degrees.  However, in both cases, the angle calculation can be done in the MCU rather efficiently.  Example code for this process is available in TMAG5170-CODE-EXAMPLES.

    Another option to consider if analog outputs are acceptable would be TMAG6180-Q1, which is an AMR based angle sensor for on-axis rotation, which is more accurate, but more limited in placement and requires you to use an ADC to sample the outputs. 

    Thanks,

    Scott