# Capacitive sensor circuit

I'm designing a capacitive proximity sensor for detection of human presence. My requirement is that I should be able to detect presence by monitoring change in capacitance at the highest possible distance for my plate size and configuration (I'm looking at a target range of 100-120 mm). I have a DC source of 24V. Additionally, I would also like to calculate approximate distance of the target from the sensor once the target is within range. Most importantly, the detection distance should be consistent and should have low susceptibility to noise. Looking at different approaches for measuring capacitance, I've found the following to be most commonly mentioned:

• RC decay circuit
• Oscillator circuit using a 555 timer
• Op-amp integrator circuit

Which one of these best suits my application? Or is there any other approach that I've missed? What are the advantages and disadvantages of the above methods?

• It seems to me that you want to detect the presence of something or someone and, possibly measure the distance of that object yet you are fixated on using capacitive sensing. Why? Sep 10, 2013 at 7:19
• Because the surface from which I wish to measure proximity is large, changes orientation and is approachable from multiple angles. A toroid shaped object can serve as an approximate likeness to the geometry I'm talking about. In case of such a geometry, I'm choosing capacitive sensing because I can do with a single sensor instead of multiple IR or ultrasonic sensors at various angles and also, I need to be able to sense proximity at less than 40-50 mm as well. Sep 10, 2013 at 7:31
• A big object at some distance compared to a smaller object nearby could give exactly the same change in capacitance - how would you differentiate those two scenarios to determine object distance? I have an idea how to do this but I'm not sure you have considered this so far. Sep 10, 2013 at 8:12
• @Andy aka: Yes, I've considered this. I did say in my previous comment that I can do with one capacitive sensor. But when you consider the problem that you just mentioned, it makes sense to break the one large transducer plate into smaller ones. A big object will then couple with more transducer plates while a smaller object will couple only with one or two. So, having say, four transducer plates instead of one can solve this. Sep 10, 2013 at 8:20
• Or you use a high enough frequency where dielectric losses can be measured along with delta C. Bigger dielectric losses possibly equate to bigger mass Sep 10, 2013 at 9:35