• State TI Thinks Resolved
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 132 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

FDC2214: Decreasing Capacitance Value

specia special
Prodigy 20 points
Part Number: FDC2214
Other Parts Discussed in Thread: FDC1004

Tool/software:

Hello ,

While using this sensor, I found an issue about decreasing cap value while detecting as shown below

Any thoughts for this issue? Is this abnormal behavior of the sensor? Or I did incorrect setup? 

Why did the cap value decrease instead of increase?

Thanks!

Specia

  • 0
    Prodigy 20 points

    I'm curious about this, since sensed another liquid increase the capacitance value, but for some different liquids it gave decreasing value.

    Is it possible to compare the decreasing value with result of LCR meter measurements? But FDC2214 measures using GND reference, so how could I replicate the measurement using LCR meter? any ideas?

  • 0 in reply to specia special
    TI__Mastermind 35585 points

    Specia,

    Its hard to say what might be going on. 
    Could you say more about your setup?

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 20 points

    Hi John,

    This is my setup.

    For liquids with very low conductivity, the cap value is increased.
    However for several liquids with high conductivity, the cap value is decreased (drop capacitance reading value).

  • 0 in reply to specia special
    TI__Mastermind 35585 points

    Specia,

    The decreased capacitance with a conductive liquid makes sense.

    The basic capacitance equation expresses stored charge in terms of the capacitance and voltage: Q = C·V

    If the liquid is non-conductive, charge will be stored on the electrode and the capacitance will appear to be fixed.

    If the liquid is conductive, the charge will leave the electrode and to flow thru the liquid, thereby making the capacitance appear to be less.
    A rough analogy is the current flow thru the conductive liquid is like having a finite parallel resistor across the otherwise ideal capacitor's electrodes. 

    We assume nominal device operation will be with a non-conductive liquid.

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 20 points

    Hi John, what do you mean with 'nominal device operation'?

    I guessed the equivalent circuitry of my setup is something like that. Is that correct? If that's incorrect could you please provide the proper circuit modeling from your side?

    Actually I have two different electrodes. Meanwhile I use electrode A, for liquids with low conductivity and high conductivity, it provides an increased cap value.

    However, if I use electrode B, for liquids with low conductivity the cap value increases, while for liquids with high conductivity the cap value decreases.

    I have checked the electrodes using an LCR meter, and it turns out that electrode A has a resistance around 16Kohm, electrode B has a resistance around 42Kohm.

    What do you think if my electrode contributes high resistance, thereby resisting the current drive of the FDC2214? Any thoughts about this case? Does the decreased cap value make sense for the electrode B case and liquids with high conductivity? Could you please give explanation about this case?

    Or any suggestions? 

    All the best,
    Specia

  • 0 in reply to specia special
    TI__Mastermind 35585 points

    Specia,

    Nominal device operation means room temperature with a non-conducting liquid, like distilled water.

    For the FDC to work properly, the sensor waveforms and frequencies need to be within ranges given in the data sheet.

    The data sheet gives the following limits on the sensor amplitude and frequency:

    If the resistance of the electrodes and/or the conductivity of the liquids don't support these limits, then the FDC will not give accurate results.

    For example, if the liquid is too conductive, the sensor waveform amplitude may become less than the V_SENSORMIN spec (1.2V), and the device will output less-than-accurate results.

    You can monitor the sensor signal/exciter using a high impedance oscope probe. The best way we have found is to place a leaded 1k resistor between the probe tip and the test point. You should see half-sinusoids on the sensor pins. The half-sinusoids should have frequencies and amplitudes that conform to the above limits. 

    Just so you know, we recommend using the FDC1004 instead of the FDC2x1y family of devices.
    Users have struggled with FDC2x1y devices due to EMI.
    The FDC1004 supports active shields which help with EMI management.

    There is a comprehensive list of FDC1004 app notes and other collateral on the E2E FDC1004 Frequently Asked Questions page.
    Several of the app notes are devoted to using the FDC1004 for liquid level detection.
    There's also an FDC1004 overview on the FAQ page that gives high-level info, as well as a detailed look at key specs and features, including active shields.
    Please let me know your feedback on this page.

    regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 20 points

    Hi John,

    I realized that this issue was caused by the electrode. The electrode has resistance too high, more than 42k ohm. I think it makes sense that the resistance of the electrode will resist the current drive. That's why while I tried to set the current drive more even max 1.5mA, the cap value increases.

    Best,
    Specia

  • 0 in reply to specia special
    TI__Mastermind 35585 points

    Specia,

    Thanks for the update.
    Were you able to look at the sensor waveform with a high impedance probe and an oscilloscope?

    Regards,
    John