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.

LDC0851: Box orientation detection- 0 or 180deg

Part Number: LDC0851

Hi,

One of my customers has designed a Pizza box sized switch. This is a standalone box and does not plug into any backplane. The power to this is through AC power chord. They need to detect the orientation of this box. It essentially has only 2 orientations- 0 or 180 when flipped.
They need to sense this orientation in order to correct the front display according to the orientation.

Can we use the LDC/FDC here in order to achieve this? Or any sensing solution form TI?

  • Hello Mahesh,

    To detect 0 or 180 degrees, you could simply attach a metal surface to half of the box and place the SENSE and REF coils on either side. I've sketched up a quick drawing below.

    Regards,

    Luke LaPointe

  • Hi Luke,

    Thanks for the input. However, when you attach a metal coil, you are introducing a metal moving part which is susceptible to damage. On top of that, the rotation of the metal piece is not reliable right? What if it gets stuck at a constant position in spite of rotation?
    Can we suggest something that is devoid of these moving parts?
  • Hello Mahesh,
    The solution I proposed had the metal target separate from the LDC coils for a contactless solution. Note the LDC devices detect the movement of conductive objects. If the conductive object is not moving then there will be no detection. Now if your box is able to rotate 180 degrees then I was suggesting that the metal be attached to the box and the LDC coils would be stationary. If this is not possible or not describing the situation properly, then I would ask you to provide more details about the system. Photograph, block diagram, descriptions, etc. all would be helpful.
    Note, for more information about the LDC coil design (typically an air core inductor drawn as spiral traces on a PCB), see the following application note: www.ti.com/.../snoa930
    Thanks,
    Luke
  • Hi Luke,

    The customer's/my concern is only about the moving parts and their reliability. If we can sense the orientation without having to make something actually move, it is more robust and error free. Here is the typical picture of the device. The display block is upright now and if the box is  flipped(180 deg rotation), the Display will also flip 180.

    Let me know what else we have.

  • Hello Mahesh,
    The LDC0851 does not need something to move to make a determination of the orientation. If using the scheme I mentioned above, then when the box is stationary in one orientation, the LDC0851 will output a 0. If the box is installed 180 degrees incorrectly, the output will be a 1. The key is that the two orientations should have a difference in metal to sensor coverage. For the 0 output state, the LSENSE coil should be covered by metal with the LREF coil uncovered. Similarly, for a 1 output state, the LREF coil should be covered by metal and the LSENSE coil should be uncovered. It is your choice on metal/sensor placement, but if you follow this rule above, you can distinguish between the two orientations without requiring any movement.
    Regards,
    Luke LaPointe
  • Luke,

    The customer went ahead with a Tilt switch based approach to detect the orientation.

  • Hello Mahesh,

    Thank you for the update. Please also note that the same approach (in terms of alignment) could be for the LDC0851, which is similar to an open/close application. Please take a look at the following 2 blogs for more info: 

    Regards,

    Luke