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.

MSP430FR2672: Shielding for CapTIvate

Part Number: MSP430FR2672
Other Parts Discussed in Thread: FDC1004, OPA354

Hello team,

FDC1004 requires active shielding in order to avoid EMI noise.

Why MSP430 CapTIvate don't require shields? Or do they?

Best Regards,

Kei Kuwahara

  • Hi Kei,

    For both FDC1004 and MSP430 Captivate MCUs, actively driven shields will be helpful, but may not be needed in every case.  Active shields are used rarely in Captivate HMI applications, buth they do help in reducing the large parasitic capacitances that can impact the sensitivity of the electrode.  An example is a proximity sensor that needs to detect an object 10's of cm away.  Without the active shield, nearby parasitics would reduce the sensitivity below the resolution of the Captivate peripheral.  Also, directional sensitivity. Say you want the sensor to be sensitive in only one direction and not the other. If you imagine a sensor layout on top copper PCB layer, it will be sensitive to a finger or hand both from the top and bottom side of the PCB.  Adding an active shield on the backside of the PCB completely blocks any sensitivity from the backside, making the sensor only detect changes in capacitance on the top side.  In most cases using Captivate though, these parasitic capacitances can be mitigated with proper PCB design/layout and and active shield is not required.

    Now, if an active shield is required, Captivate does not implement an active shield directly.  You need to use an external, high BW OPA (OPA354) to buffer the electrode signal and drive the shield at same voltage and phase as the signal.  The OPA adds cost and you can only shield one sensor for each OPA.

    Another thing to consider is signal to noise ratios and resolution differences between Captivate (~13-bit) and FDC1004 (24-bit).  What I mean is if trying to measure very, very small changes in capacitance (10's of femto-Farads) then the FDC1004 is the way to go due to its very high resolution.  But those very small signals could be the same order, or even smaller, than the surrounding noise, hence the SNR (signal to noise ratio) is small (bad).  So shielding is pretty much required to keep the noise out of the signal.

    But, if you are measuring a change of several pico-farads then the signals are quite large compared to the noise, hence the SNR is large (good), and therefore you don't need 24-bits of resolution and the MSP430 Captivate could do the job.

    What is the application?

  • Hello Dennis-san,

    Thank you for your detailed answer.

    The end equipment is eye inspection machine (Medical). Proximity sensor is considered for stopping the camera movement at 10mm away from the patient's face.

    Best Regards,

    Kei Kuwahara

  • Hi Kei,

    My apologies for responding late.  For this type of application, I would suggest the customer use one of the FDC products.

**Attention** This is a public forum