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.

FDC2214: Selection of Capacitor and Inductor in LC Tank

Part Number: FDC2214
Other Parts Discussed in Thread: LDC1614

Dear TI experts,

My application needs to drive two channels of the FDC2214 at different frequencies, for example, one 3MHz and the other 300kHz.

If I want the signal-to-noise ratio of the two channels to be as close as possible, besides fSENSOR = 1/2π√(LC), are there any restrictions on the values of L and C??

Also, do I need to consider Rs, Rp and Q in the FDC calculation tool?

Although the inductance in FDC is fixed, will the requirement of FDC for Rp be similar to that in LDC1614 datasheet?

Best regards,

Will

  • Hello Will, 

    Depending on what values you have for your LC tanks, your capacitive sensor may not make the same amount of dynamic change in the data due to the large frequency difference between them. You will need to consider how much of an impact your sensor has on the capacitance for the LC tank and the amount of change in data you get for each set of LC tank values. There aren't really restrictions on these values as long as you have a resulting LC tank in the correct frequency range. The Rp/Rs impacts the loss in the oscillation and how much current is needed for the FDC2214 to keep a stable oscillation on the sensor. It is best to keep the Q factor high to reduce the impact of noise but the FDC can work with a sensor Q factor in the range of 5 to 100. 

    The Rp isn't usually a concern for FDC because you do not have the planar spiral coils like you do with the LDC1614. 

    Best Regards, 

    Justin Beigel 

  • Hi Justin,

    In my experiments I used 18uH and 33pF to drive 3MHz and 680uH and 33pF to drive 300kHz.

    The sensor itself also has about 120pF of capacitance. I'm not sure if there is any potential problem with this configuration, but the measurement works fine.

    Regarding the range of Q, can it be higher than 100?

    Best regards,

    Will

  • Hello Will, 

    A Q factor higher than 100 can still work with the device but a very high Q for the LC tank can lead to the device struggling to lock onto the resonant frequency. 

    Best Regards, 

    Justin Beigel

  • Hi Justin,

    So when the Q exceeds 100, does the higher Q limit the sensitivity?

    For Rs in the FDC calculation tool, can its value be the DC resistance of the inductor? Or it require a test at a specific frequency with an impedance analyzer?

    In addition, since FDC needs to use realtively large L and C to drive at low frequencies, the Rp obtained by the FDC calculation tool is usually greater than 100kohm.

    This is outside the 1kohm to 100kohm range described in the LDC1614 datasheet, will this cause any problems?

    Best regards,

    Will

  • Hello Will, 

    No, The sensitivity does not degrade unless the device has a hard time determining the resonant frequency. If your sensor has a stable oscillation, then it is not a concern. 

    When you say FDC calculation tool, are you referring to the FDC211x/FDC221x Current Consumption Estimator (Rev. A)

    Having an Rp above 100kOhm means that the LC tank requires very little current to maintain a stable oscillation. The concern here would be setting the IDRIVE register so that the oscillation peak voltage is still within the range of 1.2V to 1.8V. As long as that voltage is still achieved, then there is no concern. 

    Best Regards, 

    Justin Beigel

  • Hi Justin,

    Yes, the FDC calculation tool refers to FDC211x/FDC221x Current Consumption Estimator (Rev. A). 

    Sorry, I also have a question about shield configuration in chapter 10.1.2 of the datasheet.

    For an actively driven shield, is the method described in the text for a single-ended sensor configuration? Because the buffered signal of INxA should not be able to protect INxB due to the 180° phase difference.

    For passive shielding, if I use a differential configuration to connect two parallel coplanar electrodes to INxA and INxB respectively and introduce a ground plane at the bottom of them. Assuming the sensor electrodes and ground are 1mm apart. When a sensor is used to detect an object above it, is the electric field between INxA and INxB used for sensing, or the electric field between INxA/B and the ground plane?

    Best regards,

    Will