• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 79 subscribers
  • Views 545 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.

MSP430FR2522: Touch senstivity of 3 element slider

GerryP
Prodigy 160 points
Part Number: MSP430FR2522
Other Parts Discussed in Thread: EVM430-CAPMINI, MSP430FR2512

Hi,

I am evaluating the MSP430FR2522 for use with an application that has two 3-element capacitive touch sliders. Due to space restrictions the sliders are very small, an overall area of around 8.5mm x 21mm for the 3 elements. The effective area of each element is around 38mm2. For testing a 3mm thickness acrylic sheet is used to be representative of the overlay in the eventual product design. I believe we have about 9 to 11pF parasitic capacitance and 0.2pF touch capacitance change with our touch slider PCB.

The application was previously using the Cypress CY8CM-BR3106S but for various reasons we want to migrate to the MSP430FR2522. The Cypress part was giving a sensible touch sensitivity with gain in hand but the MSP430FR2522 seems to be struggling to achieve enough sensivity, even with what I believe to be the parameters set for maximum sensitivity using the CapTIvate Design Centre. We are actually using an EVM430-CAPMINI eval board with the eval board button tracks cut and our slider PCB neatly wired to it instead. The MSP430GR2512 has been refitted with an MSP430FR2522 and re-programmed accordingly.

The Conversion Gain is set to 100 but the Conversion Count can't be set to more than 1800 otherwise the CapTIvate Design Centre reports an error that the Proximity or Touch threshold is too low. Does anyone know how the sensitivity can be increased without running into this problem?

Has anyone any experience of achieving adequate performance with the MSP430FR2512 or 2522 with the above mention touch/parasitic capacitance, or experience of comparing the Cypress solution with the MSP430?

Best regards,

Gerry

  • 0
    TI__Intellectual 2585 points

    Hi Gerry, 

    Thanks for reaching out and using Captivate. 

    For the sesitivity, MSP430FR25xx can reach 10fF detection. So I think there should be no difference as Cypress solution. 

    From your description, the connection between 2512 and pad may cause some noise issue and maybe this would be a reason that you cannot get the result from CDC. And I also doubted the CDC design. Have you changed the sw design by CDC when you connect your board? 

    - I want to check your CDC design and data log to confirm your design is correct or not. Could you send me the CDC file and data log?

    B.R

    Hawken

  • 0 in reply to Hawken Li
    Prodigy 160 points

    Hi Hawken,

    Thanks for your reply. After reconfiguring the project in CDC for my 3 element slider, I auto-generated the code, re-compiled it with IAR and downloaded it to the 2512 on the EVM430-CAPMINI using an MSP-FETU430IF debug interface. 

    I have attached the CDC project file (the .ser) and the data log files captured as I was sliding my finger over the slider a few times.

    I note that the bar chart reflects a touch reading on the elements as I slide my finger across but the slider position reading doesn't always register a sensible value, presumably because the touch readings to rather low or noisy.

    Let me know if I've attached the files correctly or if you need anything else. Thanks for your help.

    Best regards.

    Gerry

    MSP430FR2512 Slider.zip

  • 0 in reply to GerryP
    Prodigy 160 points

    Hi Hawken,

    After some further experimentation with the EVM430-CAPMINI and our slider PCB, I've realised that the even though a high Conversion Count setting generates the warning window about proximity and touch thresholds being below the recommended minimum, it doesn't stop the programming of the high conversion count. I somehow thought this was an error and prevented the use of higher Conversion Counts. I was setting the conversion count to 1800 in order to keep the thresholds within the minimum but I can set the Conversion count to 3000 and accept that the proximity and touch thresholds are 0.3% and 0.51% respectively, and the slider position seems to work reasonably. There's a bit more noise but I assume I can apply some Count Filter Beta to reduce the noise at the expense of slider response time?

    Best regards,

    Gerry

  • 0 in reply to GerryP
    TI__Intellectual 2585 points

    Gerry, 

    I think you can try 3000 as a temporary solution. But I think it would be not necessary when you make a real PCB. From the raw counter from your data, the pads' sensitivity is not equal. It will cause the final result is not so linear. 

    I would suggest you to make a PCB to reduce the connection noise. 

**Attention** This is a public forum