Author Topic: Capacitive sensor guard traces  (Read 409 times)

0 Members and 1 Guest are viewing this topic.

Offline carnaval

  • Newbie
  • Posts: 2
  • Country: 00
Capacitive sensor guard traces
« on: October 01, 2019, 08:33:16 pm »
Hey folks

I'm trying to design a capacitive displacement sensor for fun and I have a couple questions related to guard electrodes.
The setup is quite standard, there are two PCBs that are moving wrt each other. The top board has all the electronics and is measuring the capacitive coupling between two copper fills (see schematic).

Now, for this to work I want the capacitance formed by the two electrodes to be proportional (as much as possible) to their area of overlap. My understanding is that it's a good idea to add guard traces to avoid boundary effects and get some noise shielding.

For the top board I can directly connect the guards to a low impedance reference potential (ground), however I'm not sure what I should be doing for the bottom board.
Since it's moving around, it has no direct connection to the main circuit (the signal is sent back through another pair of electrodes with a constant overlap that are not on the schematic).

So I see only a couple options :

- Don't put guards around the bottom electrode, but it seems to me that I would get boundary effects (for example on the right in the drawing)

- Have a "ground electrode" on both boards facing each other and connect the bottom guards to that, however the coupling between them would only be a couple pF at most and it would require reducing the area of everything else. That would make the whole thing more sensitive to noise.

- Something I'm overlooking ?

I'm also wondering about having a "ground" plane on the bottom side of the bottom board. Sounds good in principle for shielding but wouldn't it make thing worse if its only connection to the circuit ground was through a ~5pF capacitance ?

I have to admit that I'm a bit in over my head with all this. I've been trying to get a better feel for it using field solvers but it's either quite a hassle or expensive ;-)

Offline carnaval

  • Newbie
  • Posts: 2
  • Country: 00
Re: Capacitive sensor guard traces
« Reply #1 on: October 05, 2019, 04:07:51 pm »
For example in this picture (from some paper) I don't understand to what potential you would connect the guard electrodes on the rotor.

It seems to me that if you leave them floating they would just act as parasitic capacitance and spread the signal around.

Offline glentek

  • Contributor
  • Posts: 16

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo