Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 751 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

FDC1004: Sensor design, optimizing sensitivity within range

Love Kull
Prodigy 20 points
Part Number: FDC1004

Hi,

Working on a design for a OoP sensor on a 4-layer PCB like in this thread: https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1240016/fdc1004-sensor-design-on-4-layer-pcb-submerged-in-liquid?tisearch=e2e-sitesearch&keymatch=FDC1004#

The sensor is about 500 mm long.

Using a Cu-foil tape mockup on a polystyren core (images below) I get quite high capacitance in air (about 5pF).

If submerged in water the FDC1004 overflows at about 50 mm depth. If putting a thin PTFE-tube around the sensor I get about 9pF at max level, I guess it is ok with the resolution the FDC1004 has but is there some way to optimize the sensitivity when designing capacitive sensors? I have read the guidelines but would like to know more, a simulation software would be nice Slight smile

I would like to use PTFE heat shrink tubing around the final sensor (4-layer Tg170 PCB) if the PCB can handle the temperature for shrinking the tube Thinking, but then there will be no air gap between sensor and the liquid (like in the image above) and I guess the sensitivity will be to high. I could go with narrower electrodes and greater spacing between them but I guess it would reduce the sensitivity as well. It would be nice to have a simulation software to experiment with before I send the design to production.

Finally a question about the OoP design,

Looking at the OoP reference design, the shields are a little bit bigger than the electrodes:

And the CHLD2 is used both as shield and sensor electrode, isn't the larger shield interfering with the sensor (same potential as the sensor electrode)? Should/can I design the sensor with the same width for the sensor electrodes and the shields?

Best regards, L Kull

  • 0
    TI__Mastermind 35495 points

    L,

    Sorry, but we do not have any simulation software to offer that this time.
    I will pass your suggestion on to management.

    For the shrink tubing, please pay attention to the relative permittivity of the material. 
    We don't have any detailed guidance to offer, but if εr is too high, it could contribute to saturated readings.

    The SHLDs are drawn larger than the sensors because they are acting as shields and need to overlap the sensors to be most effective. 
    They may be okay if they are the same size as the sensors, but this could depend on the application and the environment. 
    The shields should not interfere with the sensors because they are driven with the sensor's waveform. As a result, the SHLDs should not load the sensor and not affect it's  capacitance. 

    In the OOP technique the SHLD does not act as a sensor. It serves as a signal source that is equal-and-opposite to the sensor's signal, but does not directly contribute to the sensor reading. Ideally, this should force the liquid being measured to a point of zero potential. If the sensor signal is V(sensor) and the SHLD signal is equal-and-opposite, it can be represented by -V(sensor). The potential of the liquid is then the sum (common-mode) of the two signals, which is zero (e.g. ground). If a large grounded object gets close to the tank, it will not affect the liquid's potential since it is already ground.

    Compare this to the conventional technique, where a single sensor has an active waveform that is referenced to ground thru the liquid. The potential in the liquid would be non-zero. The disadvantage  is if a large grounded (like a hand) gets close to the liquid, it could change the liquid's potential and skew the capacitance readings. 

    One work-around is if the tank can be made of a grounded conductor, it can then act as a shield.
    But this in turn demands that the sensors be placed directly in the liquid to be measured, which has its own requirements (see Liquid Level Sensing with the Immersive Straw Approach).

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

    Thank you John,

    I now understand the OoP technique better.

    To clarify, the sensor is supposed to be submerged into the liquid.

    The PTFE tube I use on my mockup to lower the sensibility by creating an air gap between sensor and liquid (see image) is kind of thin (0.3 mm) and I don't see that it is affecting the sensor reading that much when I put it on. But if I would shrink it and the air gap would disappear then the sensitivity would be too high and the chip would saturate.

    Do you have any ideas how to lower the sensitivity not using an air gap?

    Could I design the sensor in some other way to focus the detection (dual CHLD2 on both sides of the CH1)? Would that work and would that lower the sensitivity? Or would it mess with the OoP theory if there is greater area with CHLD2 than CH1 therefor the sum would not be zero?

    You write that "In the OOP technique the SHLD does not act as a sensor", does that mean that the sensors sensitive area/volume is focused around the CH1 plane and not symmetrically between CH1 and CHLD2?

    Using OoP, does distance between the CH1 and CHLD2 electrodes in the same planes have any effect on the "dry" capacitance and/or sensitivity? With the conventional design I found that the distance affected the "dry" capacitance and sensitivity a lot.

    Using OoP is it only the width of the CH1 foil that affects the sensitivity or does the width of the adjacent CHLD2 also affect it? Do they have to be equal width to have zero sum?

    To maximize the sensitivity of the RE and RL sensors on the back of the level sensor, could I use wider foils for them (not wider than the shields though), as long as the shields behind them has equal width as the level sensor shield widths?

    Aiming for a sensor that has low "dry capacitance" and just the perfect sensitivity to maximize the resolution Slight smile I know it is probably a utopia but would like to try Stuck out tongue closed eyes

    Best regards, L

  • 0 in reply to Love Kull
    TI__Mastermind 35495 points

    L,

    I will look into your latest questions and provide an update by COB on Tuesday.

    Regards,
    John

  • 0 in reply to Love Kull
    TI__Mastermind 35495 points

    L,

    One way to lower the sensitivity is to reduce the area of the sensor. for example, some customers have used wire as a long-ish, thin wire as a sensor.

    The SHLDx pins are purely outputs, not inputs, so they do not gather data. That is what I meant when I said SHLD does not act as sensor. You can think of the CH1 and SHLD2 balancing each other since their waveforms are equal-and-opposite.

    I think it is reasonable to say the distance between CH1 and SHLD2 will impact the measured capacitance, and therefor the sensitivity. 
    Plate capacitors are dependent on the "plate" geometries and separation,

    The goal should be to make the CH1 and SHLD2 foils the same shape, dimensions and relative orientation.

    Sorry, but I don't follow what is said in "To maximize the sensitivity of the...."

    Regards,
    John