0
\$\begingroup\$

I have a board comprised of electromagnets array, to drive certain objects having magnets attached to them on its surface. Using a microcontroller I plan to move those objects by switching the polarities of electromagnets to drive the objects on a certain path.

Although the board will be fully automated, there will be no human intervention. In case there is any intervention, I want to make it so that the board detects the external changes and notifies (it can be any minor indicator).

The reason for needing the positions of the objects with their respective IDs is to make the board reset to its original position. Which isn't possible without knowing what unsettled position it is in right now.

Below is a snap of how I envisioned it to look like:

enter image description here

I have researched a bit. This is a similar use-case in DIY projects like Automated Chessboards, where individuals have used magnetic sensors (Reed switch). Image taken from Automated Chessboard by Greg06(https://www.instructables.com/Automated-Chessboard/):

enter image description here

The problem in my case is, that this only provides the positions of the pieces, not its ID.

In a spree of my research, I found, that RFID can be used to detect the IDs if RFID tags are placed in front of the Antena.

From there I thought of using RDM6300/RDM630 as it's inexpensive to test around with.

Obviously, I can't place hundreds of these RFID modules, it will be way out of my budget. But then I thought of these techniques like Charlieplexing or Multiplexing.

So my question is:

Is there a way to extend/connect multiple Antennas out of any RFID module using Charlieplexing or multiplexing? (How feasible it is.) I can place these antennas over each electromagnet by reducing the size of those antennas (A coil of 5mm diameter). Then sample them one by one to detect pieces, the Antenna on which the tag is detected is labeled as a positiosnId which could be coupled with the detected ID from the tag.

Below is the image, of how it would look like if I place the Antenna coil over each electromagnet:

enter image description here: enter image description here

Note: Please ignore the polarity of the magnet, it will be switched of in the Scanning pass.

I am just not sure, what will be the effects of adding components like resistors and Diodes to achieve Charliplexing could have on the detection capabilities of the module. Because those components will be a part of the Antenna to activate it. If charlieplexing can be used, I can work with just a single RDM6300.

I don't care if it reduces the effective range of the antenna, as I need the detection slot to be 25 cubic millimeters ( that is a little volume above a single Electromagnet.)

One essential thing: I cannot spend more than 40-50 dollars on the detection component of the project, it's not a hard limit, but I am looking to work that will be less expensive.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Be aware that RFID modules tags and readers don't quite like nearby metal, it can adversely affect performance, so this may be an issue to watch out for. \$\endgroup\$
    – jcaron
    Commented Dec 24, 2023 at 13:45

1 Answer 1

0
\$\begingroup\$

No matter the approach, you’d have to prototype it.

Since RFID interrogation signal is all AC, there should be no problem using diodes as AC switches. DC bias would control them. The row and column lines need AC isolation from the DC drive sources - say using suitable chokes.

An alternative approach is to float the bias voltage and bias switches using a small isolated DC-DC module. That way there is no significant AC path between diode bias circuit and the ground of the RFID module. So capacitive coupling is all that’s needed to make DC disappear from the POV of the RFID module.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.