Response delay in Capacitive Humidity Sensors

[apj_anshul]

Hello,

A project I am working on currently requires me to test how capacitors in our circuit react to change in Humidity. As expected, the experimental results suggest that capacitance goes up with increasing humidity, as the electric permitivitty of air surrounding capacitor increases.

However, I also see a curious thing - there is a delay of a few minutes before capacitor responds to change in Humidity. I am conducting these experiments in a closed humidity chamber. I can tell the delay comes from the capacitor because even when chamber's relative humidity display stabilizes to a constant RH value, the capacitor still takes some sizable amount of time before it settles down. This delay is also observed in commercial humidity sensors that use capacitors.

One would expect the capacitance to change immediately, as permittivity of air changed as soon as humidity changed. What causes this delay then?


P.S. Btw, my capacitors are simple two plate capacitors, with no dielectric between them except air. So, any change in capacitance can be safely assumed to come only from changing properties of air.

 

[truesearch]

What are you actually measuring that enables you to say that the capacitance is changing?

 

[apj_anshul]

I am using AD7150 board to measure capacitance (analog[dot]com/static/imported-files/data_sheets/AD7150.pdf). As far as I know, the board applies a excitation voltage to the capacitor and reads the charge to determine capacitance. I then scan the hex values returned by the board and convert them into capacitance. The manufacturer rates the board's response time at 10ms.

 

[256bits]

I would suspect that the 10ms response time is that related to a fast change in capacitance such as when a finger comes close to disrupt the capacitance field of the sensor and not to slow changes in environment brought about by temperature or humidity.

Perhaps the board up is incorrectly to detect these slow changes or perhaps slow changes are impossible to detect with this board through its own environmental adaptive internal circuitry and algorithms, which is needed so that the envionment does not falsely trigger the sensor and board in a real life application such as a touch screen or keyboard.

You could try another arrangemnet of capacitor such as a flat spiral or similar so that the humidity acts on all of the capacitor and does not have to work its way into the space between two flat plates ( which should be pF but you do not say specs of your capacitor ) as one way to test if that affects your readings.

Just shooting somewhat blindly here to give you some ideas to think about if yo have not already.

 

[marcusl]

What is the separation d, and ratio of plate area A to d? If d is small, and A/d very large, then it can take considerable time for humid air to diffuse into the gap and equilibrate.