0
\$\begingroup\$

There is a project at

http://www.instructables.com/id/DIY-3D-Controller/

to make a DIY 3D controller with capacitors to infer the 3D position of the object being tracked (hand) without using any wiring to hand. My question is it possible to build a 3D sensor that uses changes in magnetic files and inductors (instead of the capacitors) to also build a hands free 3D tracker?-probably the hand would need a metal object possibly a finger ring to create a change in the three or more magnetic fields. So, if possible, whats the circuit for the 3D sensor using changes in magnetic fields instead? Also is there a simple circuit to do it like known capacitor circuit mentioned above? As a guess it could probably be done with two magnetic fields if it works something like 2d vision tracking where two 2d pictures (in suitable planes and orientations) can be used to determine the 3d position of the object.

\$\endgroup\$
1
  • \$\begingroup\$ Please read electronics.stackexchange.com/help/how-to-ask . Questions should be specific and not open ended, so a specific answer can be voted on. If you want to ask opinion or broad questions, you may want to find another forum. \$\endgroup\$
    – Voltage Spike ♦
    Commented Aug 15, 2016 at 19:56

1 Answer 1

Reset to default
0
\$\begingroup\$

You could use three copper coils that are physically separated and have the user wear an iron or steel ring as you suggested.

As the ring moves nearer each coil the inductance will increase very slightly. And as it moves away it would decrease.

You would need to characterize how much the inductance actually varies with position. Unless the distance between the coil was small (say less than a few inches) my guess it that the inductance woudld vary by less than 1% of its initial value. Which means that your detector circuit would need to be very sensitive.

To measure the inductance change you could run a square wave or sinewave through the inductor and a series resistor of known value. Use a peak detector circuit to measure the peak inductor voltage. The measured voltage should be approximately propotional to the inductor value when considering such small changes.

By recording the voltage measurements for various positions you could build a rough lookup table of value vs position (say 64 entries). From the rough table you can determine position for any voltage readings by interpolating the table entries.

The capacitive sensor you linked is probably a lot easier since the human body makes a pretty good capacitor and therefore you can get a pretty good reading of the proximity to a plate.

This inductance to digital converter chip may also be useful. LDC1612 http://www.ti.com/lit/ds/symlink/ldc1612-q1.pdf

\$\endgroup\$
8
  • \$\begingroup\$ I found your reply interesting and informative. I would vote it up but Im a new user and I dont have that privilege. So basically it could also be made with a few components: resistors and inductors and an arduino-and the arduino could generate the square wave pulse too. \$\endgroup\$
    – qwerty10
    Commented Aug 15, 2016 at 20:03
  • \$\begingroup\$ "And You would need to characterize how much the inductance actually varies with position." So you mean a formula exists that links the distance and inductance. But what is it? \$\endgroup\$
    – qwerty10
    Commented Aug 15, 2016 at 20:04
  • \$\begingroup\$ @qwerty10: You can accept the answer instead but usual advice is to wait a day or two to encourage others to contribute answers. You'll find that even with a more standard application you'll get several points of view and learn a few more things. Don't forget to come back and accept the best. \$\endgroup\$
    – Transistor
    Commented Aug 15, 2016 at 20:05
  • \$\begingroup\$ Maybe to simplify things a plastic pen with a metal ball bearing on the end can be used instead of a metal finger ring. \$\endgroup\$
    – qwerty10
    Commented Aug 15, 2016 at 20:40
  • \$\begingroup\$ @qwerty10 characterization would involve building it and then making measurements. The measurements could be plotted on a graph and then a best fit line could be made. The equation for that line would be your formula. There would be one formula for each axis (and I suspect they would all be identical if the device was built symetrically). \$\endgroup\$
    – user4574
    Commented Aug 16, 2016 at 17:27

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