5 Replies Latest reply on Dec 20, 2019 1:21 AM by BragadeeshV_41

    Shield is taking signal

    LaKa_4300241

      Hi

       

      I am using a sensor PCB with two loops, outer loop is GND and inner loop is sensing loop. The shield is hatched at the back of PCB so that it should not sense.

       

      I am using kit CY8CKIT-145-40XX

       

      pin 4.0 is connected to shield in PCB

      pin 1.4 is connected to sensor in PCB

      pin GND is connected to GND in PCB.

       

      I want the PCB that shield should not take signal. I am sure that wires are not touching PCB.

      Pictures are attached in the mail for better understanding. The picture that i have uploaded have only one loop on PCB but i am using PCB with two loops.

        • 1. Re: Shield is taking signal
          BragadeeshV_41

          Hi LaKa_4300241,

           

          Are you observing increase in raw counts when you touch the shield from the bottom?

           

          Can you please use Tuner GUI to send us the raw counts/ baseline graph showing the following?

           

          1. When there is no hand near the proximity.

          2. When there is hand in the proximity region and then remove your hand again

          3. Touching the shield

          4. Touching the shield from button and place your hand in the proximity region?

           

          Regards,

          Bragadeesh

          • 2. Re: Shield is taking signal
            LaKa_4300241

            Hi Bragadeesh,

             

            I have tried to take the pics and attached file to the following mail,

             

            1. When there is no hand near the proximity.

            2. When there is hand in the proximity region and then remove your hand again

            3. Touching the shield

            4. Touching the shield from button and place your hand in the proximity region?

            • 3. Re: Shield is taking signal
              BragadeeshV_41

              Hi LaKa_4300241,

               

              Apologies for the delayed response.

              Please probe the shield and proximity sensor pin share us the oscilloscope waveform. Ensure that the shield is charging and discharging completely. If it is not, reduce the sense clock frequency until the sensor charges and discharges completely. You can also use the Fs <= 1/10RC rule to set the max sense clock frequency that will charge and discharge the sensors and shield.

              R = 1.06 ohms (500 ohms internal resistance and 560 ohms external series resistance) . C is Cp of the sensor

              Also ensure that the settling time for the shield waveform to reach 97 % of Vref is within 90 % of the ON time of the sensor.

               

              Regards,

              Bragadeesh

              • 4. Re: Shield is taking signal
                LaKa_4300241

                Hi Bragadeesh,

                 

                When looking into your answer I can't quite follow your suggestion in regard to formula Fs <= 1/10RC?

                What is the reason for using 10 in the denominator?

                The formula for frequency is, f = 1 / (2*pi*R*C), is your formula corresponding to it or has it some other origin?

                 

                In attached files you can find an output voltage graph collected with oscilloscope from the CY8CKIT-145-40XX kit.

                In the graphs the red line is shield and blue is signal.

                In the graph the shield is driven with 3.0 V. Which is greater than in data sheet stated Vref = 2.743 V.

                Data collected for 4.7V <= VDDA for a Fourth-generation CapSense devices, CSD sensing method, stated in data sheet, Document Number: 002-22299 Rev.**

                What is the reason for this higher voltage?

                Can the parameter setup effect shield voltage, giving a higher voltage?

                Do you have any suggestions to overcome the issue?

                 

                Another thing that was observed is that the oscilloscope reading for the signal (blue) have a decreasing voltage throughout the measurement.

                Is this as it should be or is something strange with this?

                 

                Used parameter setup:

                 

                I hope that this can give you information to understand why there is an issue with the measurements when an object is influencing the measurement when it is situated behind the shield.

                 

                 

                Any suggestions ?

                • 5. Re: Shield is taking signal
                  BragadeeshV_41

                  Hi LaKa_4300241,

                   

                  The time taken by a capacitor to completely charge is 5 RC, where RC is the time constant.

                  For your reference: https://www.electronics-tutorials.ws/rc/rc_1.html

                   

                  Now taking into account both charging and discharging cycle, we arrive at 2 * 5RC = 10 RC. We need to choose Fsw such that it is able to charge and discharge the capacitor completely.

                   

                  It looks like you have used a pin that is connected to some other component in the kit. Please refer to the kit schematic and make sure that your shield pin is not connected to any other component (pullups for example).

                   

                  Regards,

                  Bragadeesh