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.

CAPTIVATE-FR2633: Detection of ferrite magnet using capactive sensing

Part Number: CAPTIVATE-FR2633
Other Parts Discussed in Thread: MSP430FR2676

I am newbie to electronics and this is my first post. Please forgive me if I have posted it in incorrect forum  :).

I have made a scaled version of 64 channel PCB provided in development platform (https://www.ti.com/lit/ug/tiduaz5/tiduaz5.pdf).

I have explained everything in this video : drive.google.com/.../view

Queries:

-I want to detect ferrite magnet instead of human touch can please help me on how do I configure it?
-The software kinds of hangs as soon as if the surface if touch, Can I know the reason?

I have one 64 channel array with IC from renesas (www.renesas.com/.../r01ds0273ej0300-rx130.pdf) which is used in one such 64 array channel product. It works only with iron core touch sensing and is not affected by any other element, I want to achieve the same using TI module. Can you let me know if this is possible?

  • Dhiraj,

    thanks for contacting the e2e forum. You are in the right forum.

    Interesting application, as I don't have an quick answer, I will check and see what can we do.

    What kind of ferrite to you use?

    Thanks for the video, it helped a lot to understand the implementation.

    Regarding the issue where the system stops when directly touch that's correct.. Please consider having an thin overlay (plastic, plexi) on top.

    Please stick that on the PCB (e.g. double sided tape or epoxy) to avoid any movement

    The mutual design you are using is very sensitive to these kind of movements

    I will check in detail and come back to you

    Regards

    Kostas

  • Hello Kostas,

    Thanks for the reply.

    What kind of ferrite to you use?

    - I am using heteroferrite magnet (also called Permanent ferrite), 

    Dimension is 1.2mm thickness *14mm diameter.

    Please stick that on the PCB (e.g. double sided tape or epoxy) to avoid any movement

    - Tried this it works fine when I put some layer.

  • Hi Dhiraj,

    thanks for the info.

    I am not an expert in material, I have some ferrites around at least to do some test, but not the one you are going to use.

    The issue in general is, that the object to be detected (finger or any material) needs to have a GND reference to generate the delta in C.

    In this case if the material is floating without any coupling to GND it will not generate sufficient delta C to be recognized as a touch (very much dependent on the material composition and geometry).

    I have done some tests using a standard ferrite magnet, (the ones used for pin walls), using the small TI 64 button board. The settings I used: Conversion Count: 500 and Conversion Gain: 100. I can see count change around 20-30 counts, while my finger generates a count of around 400.

    Now it depends on the application. As you have a much larger board with larger electrodes, you need to check the behavior on this board.

    To you see a possibility to modify the mechanics in a way to have the ferrite somehow coupling to the GND?

    I will try to do some additional tests, but it's not easy without GND reference.

    One suggestion would be to use the new generation of the CapTIvate devices e.g. MSP430FR2676.  This family (GEN2) has a higher noise immunity which would allow you to work with lower touch and proximity detection thresholds.

    Regards

    Kostas

  • Hello Kostas,

    Thanks again for reply.

    In my case there is an action figure plastic model which has ferrite magnet in the base and the user have no contact with ferrite magnet directly.

    I have one product which is using Renesas IC (https://www.renesas.com/in/en/doc/products/mpumcu/rx100/r01ds0273ej0300-rx130.pdf), I checked the connection and each trace thoroughly. They are connected to capacitive measurement pins of this IC. Also It uses 8*8 mutual capacitance matrix. I am very sure about these two points.

    Their action figure have one iron disc at bottom and the figure is made of plastic, so there is no direct ground contact, but I am really confused how is this working. Even if I completely put my finger on the surface, there is no detection of human touch, but when I keep iron disc or aluminium foil it detects very accurately. Can this be because of top graphic PET layer? or Is it because when person lift the action figure there is indirect GND connection via human finger touch to action?

    I have done some tests using a standard ferrite magnet, (the ones used for pin walls), using the small TI 64 button board. The settings I used: Conversion Count: 500 and Conversion Gain: 100. I can see count change around 20-30 counts, while my finger generates a count of around 400.

    - Let me try same setting and update you with video.

    To you see a possibility to modify the mechanics in a way to have the ferrite somehow coupling to the GND?

    - Can you provide some suggestion with respect to application I mentioned above? All I can think is having top layer of action figure such that it can create contact wth electrode. But actually what I require is 

    how Renesas IC working. (no detection of human touch).

  • Hi Dhiraj,

    thanks for the info.

    Basically we do not do any competition analysis in this forum, therefore I can not look in the referenced document.

    If you have any drawings or pictures of the described setup above

    (Their action figure have one iron disc at bottom and the figure is made of plastic, so there is no direct ground contact, but I am really confused how is this working. Even if I completely put my finger on the surface, there is no detection of human touch, but when I keep iron disc or aluminium foil it detects very accurately. Can this be because of top graphic PET layer? or Is it because when person lift the action figure there is indirect GND connection via human finger touch to action?)

    I would be interested to see for a better understand of the solution.

    You say that when you put the finger on it is nor detected, but when you put the iron or aluminum? Do you hold the iron with your fingers during the detection What about the "graphic PET layer"? is it a special  overlay or just plastic? How thick is that?

    Sorry for the additional questions, but I feel I do not fully understand the setup an functionality.

    Regards

    Kostas