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.

FDC1004EVM: Issues with independent readings of capacitive sensors

Part Number: FDC1004EVM
Other Parts Discussed in Thread: FDC1004

Hi Team,

I am trying to simultaneously readout from 3 capacitive pressure sensors using the FDC1004EVM board. I have each sensor connected to one of the Cin pins and GND. On the GUI, I set CHA to Cinx (Cin1 for example), and have tried to set CHB to either disabled, CAPDAC, or to Cinx. For all of those configurations, I have the same issue. It seems that the other channels influence each other. For example, when I plug in a sensor to Cin1, then plug in another sensor to Cin4, Cin1’s capacitance changes immediately. The same thing happens when they are already plugged in and I simply enable the channel. 

My first question is, what should the settings be for CHA and CHB for each measurement channel in the GUI? Secondly, is the shield the only chance I have at resolving the channel dependency issue?
Thanks in advance!


Kind Regards,

Jejomar
  • Jejomar,

    I am sorry you are experiencing problems with the FDC1004.

    Could you give a little more information on the end application you are supporting with the sensors and the device?
    - Is this a liquid-level sensing application or something else?

    What are the sensor dimensions and relative spacings w/respect to one another?
    - A snapshot would be fine as long as a scale reference is included.

    What are the sensors' nominal capacitances?

    Is the active shield a workable option for your application/configuration?

    Regards,
    John

  • Hi John,

    Many thanks for looking into this. I am using the sensors as pressor sensors embedded along the inside of a tube wall. It serves a purpose in measuring air flow rates passing through the tube.

    The sensors are 1 cm x 1 cm and are about 6.5 cm apart from each other. 

    The nominal capacitances have changed since I added shielded twisted cables. Prior to the new cables, the nominal capacitances were around 10 pF. With the shielded twisted cables, they are much higher. I usually have to set the offset capacitance to around -13 pF for each sensor and that will yield a 10 pF reading. This would imply a nominal capacitance of 23 pF. 

    I am not very familiar with shielding if I’m being honest but I do have the shield wrapped around the twisted pair cables, so I wasn’t sure if I should connect that to the shield pins or not. 

    Lastly, I am having another problem with one of my sensors. 2 of them show good readings (changes in decimal places as samples are collected). The third sensor does not. The capacitance will drop from 15.999999 to 12.xxxx after I apply the offset capacitance, but after that, it never changes value. I thought it was the sensor, so I made a new one but I still got the same result. I then tried to switch out the not working sensors and put in one of the other 2 that are working, and it works fine. This tells me that it’s not the channel. The sensors are exactly the same, so I am not sure why this freezing would occur. Do you happen to know a cause for this?


    Kind Regards,

    Jejomar

  • Hello Jejomar, 

    The shielding is useful to reduce the impact of noise or external impacts on the sensors. This E2E post talks about the benefits of using active shielding. Using a shielded twisted cable, it is beneficial to connect to the shield pins and you can use this to include a shield around your sensor as well. The shielding can also reduce the impact the sensors have on each other in some cases. For the input channels in your measurements, have you tried taking a differential measurement where CH1 is set as CHA and CH2 (or CH4) is CHB? 

    In regards to the stuck capacitance, when you switched out the sensors, did you plug the originally frozen sensor into one of the other channels to ensure that it was not working there as well? 

    Thank you, 

    Justin Beigel

  • Hello Justin,

    I have tried to take a differential measurement and it works fine, but that limits me to only 2 channels on the board. I need to use 3 readouts, as I have 3 sensors.

    For the stuck capacitance, I did plug the original frozen sensor into all of the other channels and it does the same thing. It’s so strange. It seems like all things are pointing to the sensors, but like I said I make them all as similar as possible and I even made another one to rule out whether or not that sensor was just a dud. However, the same thing happened with the new one. I’m not sure what could be causing that.


    Kind Regards,

    Jejomar

  • Jejomar,

    Sorry you are still experiencing problems.

    If possible, would you please measure the DC voltage at the sensors and at the FDC1004 pins when they are connected?
    Also, if it is possible, please measure the resistance of the cables and sensors?

    Regards,
    John

  • Hello John,

    Glad to inform you that the issue has been resolved. When we connected the shielding, we no longer had crosstalk issues between the channels.

    Many thanks to you and Justin's help!



    Kind Regards,

    Jejomar

  • You are most welcome Jejomar. 
    I am glad we could help.
    Please let us know if you have any more questions or concerns.
    John