• State TI Thinks Resolved
  • Locked Locked
  • Replies 16 replies
  • Answers 3 answers
  • Subscribers 31 subscribers
  • Views 1675 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.

FDC1004: Saturation and problems reading from sensor

Mohammed Hammouda
Prodigy 10 points
Part Number: FDC1004


Hello, 

Thank you for setting up this forum. It is very helpful and the contributors are patient and very very helpful.

We want to measure the volume of water in a small plastic (PLA 3D printed for now) container (less than 50 ml or 1.7 oz ). We could not access the same Ti sensor we sent the files in https://www.ti.com/tool/TIDA-00317 to a PCB manufacturer which gave us, what seemed to us, a high quotation for it so we decided to make our own using copper tape and acid. However, based on our container ours is smaller than the TIDA 00317. Ours is about 10 cm in length (including the tail part) x 4 cm in width. Please check the photos below. Hence, due to imperfections, mending work needed to be done, hence the soldering and such.  However, to the best of our knowledge we made sure that connections are correct and no discontinuities in the copper or soldering exist. 

Below you can see the interface on the GUI. We immediately get a saturated channel. Even when we put water in the container we get the following.

Finally, because our sensor dimensions are different we used the CAPDAC feature as follows. I got the idea of using it from a recent post I saw here: https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1078277/fdc1004qevm-with-tida-00317?tisearch=e2e-sitesearch&keymatch=FDC1004#

We get an increase when filling the liquid but a very very small increase as shown.

Note: we did a free trials before, however using a breadboard, and not soldering, we some how got less of the saturation problem. Still we did not get a reading.

Moreover, can you please inform me what things should I be looking for when decreasing the size of the sensor. Also, the TIDA 00317 with the TIDU736A document show 4 layers and an ENIG requirement,  I assume this is why the quotation we received from the PCB manufacturer was high. As a trial can we use a normal FR4 rigid sensor or a flexible one with 2 layers only and normal plating instead of ENIG?

Finally, In the US at least, what would be the expected price for this sensor? We want to compare it to the price we got if possible.

Any feedback is very much appreciated. 

Thank you very much for your time and assistance in advance.

  • 0
    TI__Mastermind 40295 points

    Mohammed,

    We have some app notes that might help with using and troubleshooting liquid-level monitoring systems using the FDC1004.
    You can see a complete list at the FDC1004 Capacitive Sensing FAQ page.

    The FDC1004 can be used with a variety of sensor configurations. Some users have implemented FR4-based sensors, and others have used combinations of copper tape layered with insulators. Pricing of the sensors will vary from vendor-to-vendor. We don't have pricing guidelines to offer since those are determined by the vendors themselves.

    What is the liquid you are trying to work with? Too much conductivity - caused by dissolved salts or other minerals or contaminants - can cause errors.

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply Mr.Miller.

    I took a look at the FAQ you sent me. However, I am still unable to pinpoint our errors. 

    Thank you for the information regarding the configuration and pricing.

    For now I am working with normal tap water. Does using saline provide unreadable results or can it be compensated by offsets or so?

    Finally, can you please provide comments on our sensor and results?

    Thank you.

    Mohammed

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed,

    Tap water should be okay.
    What capacitance readings do you get with no water in the tank, and with the tank full?
    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply Mr.Miller,

    My first screenshot of the GUI data is immediately after I connect and while the tank is empty. 

    In my second one with the graph, I start adding water at about 225 up to about 260, the small ramp you see there. Then once I stop adding water, and before emptying, I get the drop you see at about 260 and it continues at that level.

    In order to remove saturation I tried the CAPDAC method shown in the third screenshot of the GUI. And the corresponding, second graph with MEAS1 only, shows when I start adding water at about 2320 and then I stop at about 2365. The moment I stop it drops steeply, not to the start point but to a slightly higher point. 

    Looking forward to your reply.

    Thank you again.

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed, 

    Just to confirm, in your screenshots of the plots (#2 & #4 above), it looks like the capacitance increases while you are filling the tank, but drops back down to the same capacitance (as with no water) when you stop adding water to the tank? 

    As an experiment, can you try a capacitance reading of dry and wet sensors when they are out of the tank?

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    TI__Mastermind 40295 points

    Mohammed,

    One more thing: would you please measure he resistance between the sensor connections (out of the tank) with them dry and wet?

    John

  • 0 in reply to John Miller - Sensing
    TI__Mastermind 40295 points

    Mohammed,

    Please check your CIN3 connection on the PCB. Its hard to tell from the image, but it looks like it is not connected to anything.
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply. 

    Apologies if I was not clear. These sensors are out the container, we attached them to the outer wall surface of a 2 mm thick container made from PLA.

    Yes as soon as we stop adding water it drops to a level very very much slightly higher than the starting empty capacitance.

    Do you mean we should just deattach them from the container and measure while dry and then sprinkle them with water and measure that?

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Mr. Miller,

    CIN3 is connected from the other side of the FDC1004EVM. So it is connected. 

    Just to clarify, our setup is the same setup as the one provided in the TIDU736A document. The only difference is that our sensor is smaller than the measurnments provided in the TIDA00317 document.

    Thank you again.

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed,

    The puzzling part of this is that it looks okay while the tank is filling, and that it reverts back to what looks like an empty tank when the filling stops, as shown in the third and fourth screen shots.

    Can you monitor the sensor waveforms with a high impedance probe and an oscilloscope while you are performing your tests?

    And are the copper sensors in direct contact with the water? In the images, it looks like they have a plastic covering, but it is hard to tell for sure.

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply again.

    As you could see in the images, the MEAS2 (orange) is saturated, is that normal? 

    Also what should I be looking for on the oscilloscope.

    Yes we have a plastic cover on the sensor. Hence, yes the copper is not in direct contact with the water. There is the container  wall and a thin plastic cover over the coppers. Possibly just duct tape. 

    The sensor is not at all in contact with the water. Is is attached on the outside wall of the container.

    Moreover,  while it is filling it is only increasing by about 1 pf. Is that because the sensor is small in dimensions?

    Thank you.

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed,

    The data sheet describes the typical behavior of sensor waveform. It should be a 25kHz, 2.4Vpp signal with a 1.2V DC offset.
    If it doesn't look like that - or close to that - then that may help us identify the root cause of the problem.

    We don't have the analysis tools to help with the design of the capacitive sensors, so it is hard to say if the smaller sensors are the root cause of what you are seeing.

    I'm not sure if you saw it on the FAQ page, but there is an app note about calibrating capacitive sensors for liquid-level sensing.
    You can see it here.

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply again, 

    Using an oscilloscope between the GND and the pins we get the following:

    GND-SHLD1 

    25kHz, 2.4Vpp, 1.2V DC offset and here is the screenshot from the oscilloscope:

    GND-CIN1

    25kHz, 520mV Vpp, 40mV DC offset

    GND-CIN2

     25kHz, 580mV Vpp, 0 DC offset

    GND-CIN3

    25kHz, 520mV Vpp, 0 DC offset

    GND-SHLD2

    25kHz, 1.2V Vpp, 360mV DC offset

     

    Looking forward for your feedback.

    Thank you,

    Mohammed

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed,

    I measured the signals on C1 and SHLD1 on one of our EVMs for comparison. 
    It looks like your signals - and DC offsets - are a little low compared to our EVM. 
    My guess at this point is that your sensor is loading the pins in some strange way and this is impacting the device's function.

    Pin freq Vpp VDC
    C1 25kHz 1.5 0.56
    SHLD1 25kHz 2.5 1.26

    C1                                                                                                   SHLD1

          

    Regards,
    John

  • 0 in reply to John Miller - Sensing
    Prodigy 10 points

    Thank you for your reply. 

    What would be the reason for that. Is there any chance that our hand-made sensor and its rough work could be th reason. I mean do you think that getting a pcb professionally made would solve the issue? 

    Final question, may that explain the behaviour why after we stop adding water we get a sudden drop in signal?

    Thanks again.

    Mohammed

  • 0 in reply to Mohammed Hammouda
    TI__Mastermind 40295 points

    Mohammed,

    It may very well be the sensor, but I can't say for sure. 
    Let me suggest one more experiment. Try filling the tank to different levels, such as low, medium, and completely full.

    As you reach each of the three points, stop adding water and see if the signal remains constant, or starts to drop. 

    If you can fill to the low level, and then stop for a little while, and the signal doesn't decrease, then the sensor is okay there.
    If you then fill to the medium or high level, stop for a bit, and the signal level decreases, then I'd say there is a leak in the sensor that allows the metal parts to contact the water.

    Some of our app notes available on-line might help with the sensor design & build.
    If you haven't looked at them already, you can find them at the FDC1004 Capacitive Sensing FAQ page.