I have an 8 x 8 grid and would like to detect where on the grid I place a metal object. I'm currently working with the TDA0161 metal detection IC, and so far have the following setup with detection range of 5mm:

enter image description here

I don't fancy using 64 of these ICs and am experimenting with multiplexing the coils.

I'm thinking of using 8 ICs each with 8 coils, and then using an analog switch to switch between the coils checking each position.

So far I have tried the to use the the CD4051BC, with the common I/O connected to one end of the coil in above picture, and each 8 to the other end and then to each I/O of the CD4051BC as you can see in the schematics below:

enter image description here

The above schematics however doesn't work.

When I use the first schematics with a single coil, it works perfect, but when using the analog switch nothing. When I measure the inductance across the single coil I get about 300 uH but when I measure the coil after it has passed through the analog switch, I get some random high number over 10 H.

The datasheets for the two ICs:

TDA0161: http://www.gremlyn.plus.com/ahme/CD00000119.pdf

CD4051BC: http://pdf.datasheetcatalog.com/datasheet/fairchild/CD4052BC.pdf

  • \$\begingroup\$ Maybe there's a better way of proximity sensing. Capacitance? ? Magnetic Hall sensor? Chess board?? \$\endgroup\$ – Sunnyskyguy EE75 Apr 14 at 19:57
  • \$\begingroup\$ I have been touching on several ways of proximity sensing. In the start I used a magnet, instead of a piece of metal, an I tried hall sensors and reed contacts to locate them, but for other reasons with my project I had to ditch the magnets, and use metal instead. I'm kind of locked on the proximity sensing choice by now, so I'm hoping I can get this to work. \$\endgroup\$ – Benja0906 Apr 14 at 20:07
  • \$\begingroup\$ When you have a spiral pair and couple it with a magnet bridging the coil anywhere, the inductance changes or the capacitance changes as long as gap to target is less but a reed relay is too insensitive. compared to a Hall sensor, so define your MUST HAVEs and nice to have 1st and material limitations. \$\endgroup\$ – Sunnyskyguy EE75 Apr 14 at 20:21
  • \$\begingroup\$ Maybe RFID technology with unique RFID on each piece \$\endgroup\$ – Sunnyskyguy EE75 Apr 14 at 20:36
  • \$\begingroup\$ Then I would still have to switch between different coils right? The problem I'm having is not so much how to detect the object, but rather switching between different coils, to make my circuit more effective, because after all I could just use 64 TDA0161 IC's, if I'm not mistaking. \$\endgroup\$ – Benja0906 Apr 14 at 20:46

Use 8 narrow coils, in the X direction. Use 8 narrow coils in the Y direction. And measure the Q of each of the 16 coils. Thus you have 16 oscillators, and the two coils that are de-Q'd and will NOT oscillate are indicating the X-Y location.

  • \$\begingroup\$ Sure it could work, but unfortunately not in my case. I have a very restricted physical space, and there is simply not space enough for the coils to travel above each other as they would with this method. Thanks for the suggestion however. \$\endgroup\$ – Benja0906 Apr 14 at 23:00

This table from the datasheet ought to explain the problem:

enter image description here

The analog switch has (best case) several hundred ohms of resistance.

The TDA0161 uses an external inductor and capacitor in an oscillator.

Your analog switch is inserting hundreds of ohms of resistance into a resonant circuit.

I expect that pretty well kills any oscillation dead in its tracks.

  • \$\begingroup\$ I did see that, but wasn't 100% sure that was the problem, thank you for clarifying that. Do you have any suggestion on how I should approch swithcing between different coils? \$\endgroup\$ – Benja0906 Apr 14 at 20:43
  • \$\begingroup\$ ActualIy it's a parallel resonance circuit so series R is not the limiting factor but the voltage may exceed the supply rail which is not good for the analog IC ( if sharing Vdd.) Haven't you measured anything yet@Benja0906 ? \$\endgroup\$ – Sunnyskyguy EE75 Apr 14 at 20:57
  • \$\begingroup\$ It doesn't seem to exceed the supply rail. When no metal is present it lays steady at around 2V, and when metal comes in contact it jumps and stays at around 3V. The supply voltage is 5V. \$\endgroup\$ – Benja0906 Apr 14 at 23:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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