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.

DRV5055: DRV5055 - array

Part Number: DRV5055
Other Parts Discussed in Thread: DRV5056, , DRV5057

Hi All,

I'm evaluating the use of AMR sensors to sense the absolute linear position of an element. The stroke is about 40cm, not enough to be covered by a single IC, therefore I would like to get hints about the usage of multiple ICs in array configuration.

Is there any app note or tutorial where I can get some hints about a such configuration (sensors pitch vs position accuracy, ...)?  

Thanks

regards

  • Hello Pasquale,

    Thank you for using the TI forum. We do not have any documentation on this currently, however, I can still give you a couple ideas.

    In general, when doing linear position sensing via an array of sensors, each sensor will give the same voltage output at 2 different physical locations. The sensors to the left and right of that sensor are then used to determine which of those two locations you are at. A lookup table can then convert this into a distance for the system.  Here is how that works in two common configurations:

    1) Using unidirectional linear Hall sensors (DRV5056):

    For this configuration, the magnet and sensors are setup like this:

    Which gives an output similar to this for 3 sensors as the magnet moves across:

    In this setup, if you are using the middle sensor/curve for measurement, then to determine which side of the peak you are on, you see which of the two sensors to the sides has a higher voltage.  

    As for spacing, you can be flexible with how far to space the sensors, but you need to have them close enough that when you are near the peak output of one sensor, you can still tell which of the two side sensors has a higher voltage with reliable accuracy.

    More sensors can be added as necessary, keeping equal spacing between all sensors. 

    2) Using bidirectional linear Hall sensors (DRV5055, DRV5057)

    For this configuration, the magnet and sensors are setup like this:

    Which gives an output similar to this for 3 sensors as the magnet moves across:

    In this setup, if you are using the middle sensor/curve for measurement, then to determine which side of each peak/trough you are on, you see if the neighboring sensor is above or below the mid-point (for this configuration, for a peak you look at the curve to the right, and fora trough you look at the curve to the left). 

    Spacing for this setup is determined by the length/thickness of the magnet because you need the the neighboring sensors mid-points to be at the peak/trough of the sensor being used for positional measurement. What this means for spacing is that when the sensor being used for positional measurement is directly underneath the center of the magnet, then the sensors to each side will be underneath the outer edge of the magnet (exact positioning under the edge can change a little depending on magnet shape and sensing distance).  

    To expand this method beyond 3 sensors there are two options:

    • Similar to the method above, you can continue to add more sensors at equal spacing as needed. 
    • Instead of adding more sensors at equal spacing, you can add small groups of 3 sensors spaced according to the desired resolution (the 3 individual sensors in each group will be spaced according to the magnet thickness as described above, and the spacing between each group is can be chosen based on desired resolution).  In this setup, you will only ever use the middle sensor from each group to determine location.

    I hope this helps get you started.