• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 2160 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.

FDC2214 how to improve the result

na na78
Guru 16770 points
Other Parts Discussed in Thread: FDC2214, FDC2214EVM

Hi

We have problem about measurement using FDC2214.

Application : Drone

We are about to use FDC2214 for liquid level detection(the remaining amount of fuel).


Following figure shows the actual situation. The target liquid is gasoline and the electrode is copper foil (level and reference).

First of all, we tried desk test, and we got good result (stable result).
In desk test, there were no frames (only tank).

But test with actual situation, we could not get stable result.
We are guessing environmental noise, influene of frames, and so on.

The following is our schematic of FDC2214.

What do you think what should we care about to improve measurement result. Do you have any idea?

Bestregards

  • 0
    TI__Expert 8050 points

    Hello,

    How different are your measurement results under the two test conditions? What are the dimensions of the tank and sensors? Is the drone operational?

    Regards,

    Yibo

  • 0 in reply to Yibo Yu
    Guru 16770 points

    Hi Yibo

    Thank you for your reply.

    The dementions of the tank is H:60mm x W:70mm x D:120mm. The drone is not operational.

    We tried to test with FDC2214EVM and we found a probable cause.

    EVM#1

    Raw Code

    Water Level (cm)

    Level Sensor (CH2)

    Ref(CH3)

    Diff (|Level - Ref|)

    Empty

    32122032

    32575827

    453795

    1 (Ref Level)

    32019650

    32492287

    472637

    2

    31944384

    32476240

    531856

    3

    31870669

    32469905

    599236

    4

    31790109

    32467279

    677170

    5

    31703187

    32464925

    761738

    EVM#2

    Raw Code

    Water Level (cm)

    Level Sensor (CH2)

    Ref(CH3)

    Diff (|Level - Ref|)

    Empty

    32248536

    32097220

    151316

    1 (Ref Level)

    32154771

    32020064

    134707

    2

    32063412

    32000532

    62880

    3

    31988816

    31995396

    6580

    4

    31914748

    31993604

    78856

    5

    31826922

    31992062

    165140

    As shown in upper tables, there is variability in each board. During measurement, the test environment was fixed (tank, cable, sensor and so on). Simply, we just changed EVM and tested.

    Why does differences occure between EVM?

    In this case, the level sensor (CH2) indicates larger raw data than Ref (CH3) for EVM#1, however, the Ref(CH3) indicates larger raw data than level sensor (CH2) at first, but going to reverse gradually for EVM#2.

    We want to have the measurement result indicates same tendency, not depend on the EVM.

    Do you have any idea?

    BestRegards

  • 0 in reply to na na78
    Guru 16770 points

    Hi

    We found two types of solution for Level detection using FDC2214.

    One is single-ended measurement shown in the datasheet, which is using three electrodes (Reference environmental, Reference liquid and Level sensor).

    Another is differential sensor configuration that is shown in SNOA935A application report. (Our trial is also using differential sensor configuration).

    Which way is better for fuel gauge application?

    Thank you for your help!

    BestRegards

  • 0 in reply to na na78
    TI__Expert 8050 points
    Hello,

    From your schematic, you are using the differential approach, which is the right way to go. The figure in the datasheet is a simplified drawing.

    I've examined your data — the raw codes look reasonable to me. You should look at the delta values of the level and reference sensors, not the difference of the absolute raw codes. For example, if we look at the level sensor only, the delta from empty to full for EVM1 is 32122032-31703187=-418845, for EVM2 is 32248536-31826922=-421614. The difference between them is less than 1%. Absolute codes can differ due to variations in the placement of the board and setup, etc., but the changes in capacitance due to liquid should be repeatable.

    Best regards,
    Yibo
  • 0 in reply to Yibo Yu
    Guru 16770 points
    Hi Yibo

    Thank you for your reply.

    I have one more question. Can you please answer?

    What is the advantage and disadvantage for single-ended measurement and differential measurement?

    Why is the differential way recommended?

    Bestregards
  • 0 in reply to na na78
    Guru 16770 points

    Hi Yibo

    We did measurement again having the delta values of the level and reference sensors as you mentioned.

    The following data that we got is for water.

    #1 Test1 Capacitance(pF) 　
    Water Level (cm) Level Sensor (CH2) Clev-Clev(0) Ref(CH3) Crl-Cre Level
    Empty 61.701 0 60.017 0 　
    1 (Ref Level) 62.114 0.413 60.345 0.328 1.259146341
    2 62.435 0.734 60.41 0.393 1.867684478
    3 62.782 1.081 60.44 0.423 2.555555556
    4 63.079 1.378 60.451 0.434 3.175115207
    5 63.417 1.716 60.461 0.444 3.864864865

    #2 Test1 Capacitance(pF) 　
    Water Level (cm) Level Sensor (CH2) Clev-Clev(0) Ref(CH3) Crl-Cre Level
    Empty 62.535 　 60.851 　 　
    1 (Ref Level) 62.911 0.376 61.162 0.311 1.209003215
    2 63.23 0.695 61.239 0.388 1.791237113
    3 63.574 1.039 61.269 0.418 2.485645933
    4 63.894 1.359 61.272 0.421 3.228028504
    5 64.24 1.705 61.284 0.433 3.937644342

    #1, #2 means the different EVM and we had test twice by each.  According to the result, we can consider they have tendency for the measurement of water level sensing.

    While, we got the following data for gasoline.

    #1 Test1 Capacitance(pF)
    Water Level (cm) Level Sensor (CH2) Clev-Clev(0) Ref(CH3) Crl-Cre Level #1 Test1 (Up)
    Empty 62.026 　 60.287 　 　
    1 (Ref Level) 62.047 0.021 60.288 0.001 21
    2 62.075 0.049 60.279 -0.008 6.125
    3 62.114 0.088 60.272 -0.015 5.866666667
    4 62.166 0.14 60.287 0 　
    5 62.216 0.19 60.275 -0.012 15.83333333

    #2 Test1  Capacitance(pF)
    Water Level (cm) Level Sensor (CH2) Clev-Clev(0) Ref(CH3) Crl-Cre Level #2 Test1 (Up)
    Empty 62.883 　 61.132 　 　
    1 (Ref Level) 62.931 0.048 61.183 0.051 0.941176471
    2 62.983 0.1 61.208 0.076 1.315789474
    3 63.03 0.147 61.214 0.082 1.792682927
    4 63.068 0.185 61.211 0.079 2.341772152
    5 63.115 0.232 61.201 0.069 3.362318841

    For gasoline, we got different result by each board and test.

    So we couldn't see tendency that is acceptable for level detecting.

    Do you have any ideas for improvement.

    Is it difficult to detect gasoline?

    BestRegards