• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 816 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.

FDC2112: or FDC1004 for liquid level sensing application

Bill Allai
Prodigy 40 points
Part Number: FDC2112
Other Parts Discussed in Thread: FDC1004,

We are looking for guidance on a liquid level sensing application. Our liquid of interest is a refrigerant with a dielectric constant of 3.0. Our initial plan was to attach copper plates within the liquid and wire out to an external sensor board. The depth of the metal tank is around 3" and we expect the capacitance to change from 25pf to 250pf, not accounting for any cable parasitics. We are looking for +/- 5pf accuracy.

It isn't clear to us if the FDC1004 or FDC2112 is the right path or if neither is appropriate. We would love some guidance on this and any thoughts on detecting level change in a refrigerant such as ours, with a relatively low dielectric constant.

Thank You!

  • 0
    TI__Mastermind 40295 points

    Bill,

    Is the refrigerant conductive?

    John

  • 0 in reply to John Miller - Sensing
    Prodigy 40 points

    Hi John, it is not conductive, I should have mentioned that previously.

  • 0 in reply to Bill Allai
    TI__Mastermind 40295 points

    Bill,

    We don't have much experience with non-water-based liquids, but if the refrigerant has the right properties, it should work fine.

    We advise using the FDC1004 over the FDC2xxx series of parts because customers have struggled getting to production with FDC2xx devices, mostly due to EMI. THE FDC1004 has built-in active shield drivers which help with EMI management, and FDC2xx devices do not.

    All that being said, the 25pF to 250pF measurement range is beyond the FDC1004's range with a single sensor (0pF - 115pF), so to use that device you may need to have multiple sensors measuring the liquid height. For example, one sensor could measure the bottom third of the liquid, another could measure the middle third, and a top sensor could measure the top third.

    And after trashing the FDC2xxx devices: if you can shield your sensors (e.g. with a grounded metal tank) and all of your cables and still get the desired measurement range and accuracy, you may have some luck with those  devices.
    The potential advantage may be a single sensor, rather than multiple sensors.
    Robust, grounded shielding will be a key factor.

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 40 points

    Thanks John! Sorry if this is a stupid follow-up question. Let's say we use multiple FDC1004(s). I assume each of them can measure the same two wires (capacitance). Meaning, it would be OK if the first sensor technically is measuring a capacitance of say 250pF? I assume it will just saturate and generate a max output but won't have any parameter violations, is that true? I wasn't sure if there would be a datasheet/part issue if we attempt to measure > 115 pF per sensor. Hope that made sense, thanks again!

  • 0 in reply to Bill Allai
    TI__Mastermind 40295 points

    Bill,

    The FDC1004 actively drives the sensor, so sharing a sensor between different FDCs probably would not work.
    A single FDC1004 sequentially reads thru its own channels, so different channels on an FDC1004 might be able to share the same connection, but I have never tried that and am not sure it would work. One alternative to consider might be to "stack" your sensors where each sensor is connected to a different FDC1004 input and independently measures a predefined percentage of the tank fluid level.

    So the sensor near the bottom of the tank would measure the fluid from tank/empty to x% full at which point that sensor reading would be maxed out.
    Once the level is past x% full, the fluid level would engage the second sensor, and as the fluid level rises, that sensor's readings would change, and would  then max out. 
    The rising fluid would then engage the third sensor and so on.

    If many small sensors are okay from an implementation perspective, each sensor might be able to measure ΔC = 30pF, with CAPDAC = 15pF.
    If a few larger sensors are preferred, then ΔC > 30pF, and CAPDAC would need to vary as the fluid level rises, up to a max of 100pF.

    I hope this helps. 

    Regards,

    John