• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 3601 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 differential measurement result

Yingying Sun
Intellectual 1155 points
Other Parts Discussed in Thread: FDC1004, TMP112, TIDA-00220

Hi,

I measured the FDC1004 EVM and configureas below:

 meas1 as single-ended, CHA as ch 1, CHB disabled; result shows capactance1 =0.4618pf ; 

meas4 as single-ended, CHA as ch4, CHB disabled; result shows capactance 4=0.3730pf ;

 meas2 as differential mode,CHA as ch1, CHB as ch4; result shows capactance =-1.2199pf; but capactance1-capactance4=0.4618pf - 0.3730pf≠-1.2199pf, why?

configuration and register is shown below:

 Best Regards

Janet Sun

  • 0
    TI__Expert 7400 points

    Hi Janet:

    The configuration settings that you posted looks fine. Are you using the EVM with the sensor portion broken off? I just tested it without it and it seems to be fine with and without external sensors connected to it. The measurement is within 10-20fF. If you are using it WITH the sensor portion that came with the board then I see similar issues. I believe it may have to do with how the sensor portion was designed.

    Let me look more into this and I'll get back to you within the next few days.

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    TI__Intellectual 1155 points

    Hi, David

    Thank you for your quick response. The problem I posted is tested with our sensor portion of our EVM.  if I broken the sensor portion,the differential measurement works fine as you said.  I am looking forward to get your progress with the study about the sensor portion and how it affect with the diffeential measurement. Thanks very much.

     

    Best Regards

    Janet Sun

  • 0 in reply to Yingying Sun
    TI__Expert 7400 points

    Hi Janet:

    Sorry for the delay. The sensor portion of the EVM is the cause of the differential measurement inconsistency. The reason why is the way the single ended and differential measurements relate to the configurations of the shields. On the EVM, CIN1 is paired with SHLD1 and CIN4 is paired with SHLD2. In single-ended measurements, SHLD1 and SHLD2 are internally shorted (described in the datasheet) thus acting as a single shield. In differential mode, they are driven with a shift in phase, thus means that the sensor is in 2 different conditions.

    So on the EVM, you cannot correlate the differential measurement with the difference between 2 single ended measurements due to the topology of the shields.

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    TI__Intellectual 1155 points

    Hi, David

    Thank you for your reply. Now, I have understood the reason of the difference. One more thing, my customer found  that when he tested the EVM with sensor portion with temperature range of 25~75C,  the differential measurment result also drift with temperature. If in the single mode, both channels drift almost the same. Is it reasonable? I can't explain why the differential mode drift with temperature.

     

    Thanks

    Janet Sun

  • 0 in reply to Yingying Sun
    TI__Expert 7400 points

    Hi Janet:

    I believe it is the same reason, the shield configurations. You should not see too much of a drift with temperature when in differential mode if you use non-shielded electrodes. In single mode, the drift should be comparable.

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    TI__Intellectual 1155 points

    Hi,David

    What if the customer want to measure the differential capacitance and add the shield at the same time? Do we have a way to realize that by design? Customer  don't want to do the minus in the MCU(measure two channels in single mode and minus in the MCU to prevent temperature drift) which will cost too much power and memory. And the shild is also important in preventing disturbance from outside.

  • 0 in reply to Yingying Sun
    TI__Expert 7400 points

    OK so if you are still using the EVM sensor portion to look at drift, you will still see it in differential mode. The reason is because of the size of the shields are different, which also affects capacitances. If you make the sensor/shield pairs symmetrical, then the temperature drift should be the same and cancel out if using differential mode, but you would still see the problem as stated initially about singled mode measurements versus differential measurements.

    I would suggest you to break that portion off and just use copper tape to do quick prototypes and evaluations based on your application.

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    TI__Intellectual 1155 points

    Hi, David

    The customer does need the shield and it is not easy to get them symmetrical enough. Do we have some way to solve the problem if they want to do the differential measurment?

    Besides, I connected the shield pins together in our EVM together to measure in differntial mode, the value is not equal to CHI1-CH4 and it also drift with temperature. It doesn't make sense to me. Can you explanin ? Since I have connected them together , it will not matter whether short or not internally in single and differential mode. I am really confused. Thank you very much.

    Best Regards

    Janet Sun

     

  • 0 in reply to Yingying Sun
    TI__Expert 7400 points

    Yes there are some ways in terms of algorithms and external circuitry. This has been tested out with different geometries with the proximity detection reference design. It uses the TMP112 to adjust the gains of the FDC1004 channels to make it more linear (since the geometries are different).

    http://www.ti.com/tool/TIDA-00220?keyMatch=FDC1004&tisearch=tidesigns

    With the EVM, you CANNOT connect the shield pins together to measure differential mode. In differential mode, both channel measurements are obtained at the same time so the SHLD pins cannot be shorted. SHLD1 and SHLD2 are out of phase of each other and are paired with CINx and CINy (please see the datasheet for more about this). If you connect them together, it will not read the capacitance correct because the SHLDs paired with the channels are not the same waveforms....

    We can discuss more about this offline if it is still not clear because it has to do with the internals of the part and how it is processed

    Thanks

    -David Wang

    Capacitive Sensing Applications

  • 0 in reply to David Wang11
    TI__Intellectual 1155 points

    David

    Thank you very much and I will discuss with you offline after I get more information from the customer. Thanks again.

     

    Janet Sun

  • 0 in reply to Yingying Sun
    TI__Expert 7400 points

    The reason for not being able to take differential measurements with temperature drift on the EVM sensor portion is because there is no ground electrode or plane. The field lines will terminate at the nearest ground potential thus causing variations in capacitance measurements since capacitance is dictated by distance of the CIN electrode and ground and change in dielectric from the change in temperature.

    To implement a reference sensor with different geometries, you cannot simply take a differential measurement. The reference measurements need to be scaled in relations to the primary electrode to be able to do a differential measurement.