• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 203 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

FDC2212: Sensing of small (<100pF) rapidly changing capacitances

Eric DeYoung
Prodigy 20 points
Part Number: FDC2212
Other Parts Discussed in Thread: FDC1004,

Tool/software:

I'm currently using the FDC1004 capacitive sensor in a design, and I'm running into limitations due to the maximum sample rate of 400Hz.  I'm using a custom rigid-flex sensor with active shields,  very similar to the Out of Phase Liquid Level sensor example with some minor dimensional tweaks (https://www.ti.com/lit/an/snoa925/snoa925.pdf?ts=1747250607381)  The data I'm getting is great, but the sample rate is a bit too slow to accomplish the customers goals.  I've been looking at the FDC2212 and I find it interesting because of the higher sample rate, but it doesn't have some of the integrated features - specifically the active shield.  Is it possible to use the device in a similar fashion to the FDC1004 OoP level sensor using external op amps to derive the shields?  Is a significantly higher sample rate possible while still maintaining sub pF resolution in the ~0-100pF range?  The sensed material is non-conductive.

Thanks!

  • 0
    TI__Mastermind 43730 points

    Hi Eric,

    Yes, you should be able to create an active shield driver based on external op amps to use with the FDC2212.  We do not have a reference design for this, but there are some details in the thread below that could be useful.

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/872784/fdc2214-remote-sensor-penetration-sensitivity-and-shield 

  • 0 in reply to Eddie LaCost
    Prodigy 20 points

    Eddie,

    That link has been very helpful.  I expanded the spreadsheet to do some calculations for some of the other registers and had a couple more questions:

    1.  There's an example for calculating the SETTLECOUNT on page 43 of the datasheet.  Vpk is 1.68, FREFx is 40MHz, C is 53pF, and IDRIVE is 15.  Using units of Hz, Farads, and the raw IDRIVE register value (15) the decimal place seems to be off.  Is one of the units (perhaps the IDRIVE) supposed to be different?

    2.  In order to use the out of phase method as I am with the FDC1004, does the following sensor layout make sense?  2 Channels would be used, with the active shields being driven by a op-amp buffered version of the complementary input?

    Thank You!

  • 0 in reply to Eric DeYoung
    TI__Mastermind 43730 points

    Eric,

    I am looking into this and will provide feedback shortly.

  • 0 in reply to Eddie LaCost
    TI__Mastermind 40295 points

    Eric,

    It looks like the value for IDRIVE should be in amps according to the legend below equation (4).
    Table 41 of the data sheet shows the corresponding value for the raw value of IDRIVE = 15 is 0.146mA.

    Using this value in equation (4) along with the other given values:
    CHx_SETTLECOUNT = 1.68 × 40e6 × 53e-12 × π2/(32 × 1.46e-4) = 7.52 → 8 (round up to the next largest integer).

    Your sensor stack-up looks okay, as long as it is for a fully differential system.
    If I understand your proposal, the system should look like Figure 56 in the data sheet, except LEVEL and REFERENCE should have two separate sensors each, with one tied to INxA and the other ties to INxB.

    I'm not sure if we can say it is OOP, but it should work. 
    Be sure to pick driver amps with rail-to-rail outputs that can dive at the frequencies of interest, and that can dive capacitive loads.

    Regards,
    John