• State Resolved
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 71 views
  • Users 0 members are here
Support feedback
Options
Options
Related

Original question:

LDC1612: Product selection and sensing range

LDC1612: Coil diameter selection

runjie zuo
Prodigy 10 points
Part Number: LDC1612

Tool/software:

Ti Team

Recently, our customer has shown interest in eddy current sensors. Their requirement is a sensing range of 2.8mm to 6.0mm. The target approaches the coil axially, with the target material being manganese-chromium alloy like 16MnCr5. On our end, we will use an MCU to parse the I2C data from an LDC1612 and output a PWM signal based on this data—90% duty cycle at 2.8mm and 10% at 6mm. To achieve higher resolution and accuracy for the sensor, could you please suggest a reasonably appropriate coil diameter for this range? Thank you.

  • +1
    TI__Mastermind 43440 points

    Hello Runjie,

    It looks like 16MnCr5 has a relative conductivity of 7.3 to 8.3%.  This is on the lower end, but comparable with steel, which ranges from 3-15%.  This will work, but will have a reduced shift in inductance compared with higher conductivity metal such as copper or aluminum.

    Generally, the range will be about half the diameter of the coil, so likely about 12mm diameter coil will be okay.  For highest sensitivity, it is recommended to keep the distance to 20% of the coil diameter.  This would indicate a coil diameter of 30mm, but it is also important to also have a similar or larger sized metal target.  I recommend the two app notes below.

    Coil Design: https://www.ti.com/lit/an/snoa930c/snoa930c.pdf

    Target Design: https://www.ti.com/lit/an/snoa957b/snoa957b.pdf       

  • +1 in reply to Eddie LaCost
    Prodigy 10 points

    Thanks Eddie

    The coil diameter will be designed at 17mm due to space constraints.

    Additionally, we would like to omit the external crystal oscillator to reduce costs. However, I have some doubts regarding the internal clock of the LDC1612. The Clock Temperature Coefficient mean is quite low, but the range of fINTCLK is 35MHz to 55MHz, with a typical value of 43.4MHz. Is this due to chip manufacturing variations, meaning that the internal clock frequency of most chips is around 43.4MHz, while a small portion of chips have internal clock frequencies closer to 35MHz or 55MHz?

  • +1 in reply to runjie zuo
    TI__Mastermind 43440 points

    See the thread below which addresses this.  Your understanding is correct though.  Typical value is 43.4MHz, but there will be a portion of devices that can be higher or lower.  

     LDC 1614 - Clock Frequency Variation