Hello, I'm trying to make a capacitance to voltage converter. I applied an AC source to a variable capacitor to generate an rms voltage proportional to the capacitance value. I then put that signal into a differential amplifier along with a rms voltage dependent on a reference capacitor. The differential amplifier tells me the voltage difference between the two signals. However, when I vary the capacitance slightly it takes about 5 seconds for the final rms output voltage to adjust. How can I minimize this delay?

I attached a picture of my circuit.

Thank you,
CD

 

You won't get many bees with .doc vinegar. Use .png or such for honey.

Here's the OP's diagram

 

One problem you have is that the "source" impedance from the capacitors in your bridge is about 221K Ohms. You want low source impedance going into the differential amplifier.

Also, you want a REAL bridge. Drive the bridge from a single AC source, or you will get potentially big errors.

 

The path to ground for both caps is very high impedance. You might add a bit more resistive load to speed up the approach to equilibrium.

 

I added some resistive load to speed up the approach to equilibrium.
I also switched my circuit around to use the same AC source. Why do I want a low source impedance going into the differential amplifier and how can I achieve that?

 

It is always best to give us the big picture.
What are you trying to measure?
If you wish to measure capacitance there are easier and better ways to do it.

 

I'm trying to convert a capacitance to a voltage value. How would you recommend doing it?

 

I don't know what you are doing with the amplified signal, but it looks like you are shooting for a gain of 100. The op amp gain equations assume you have very low impedance going into the 10K resistors. If that is not the case [as here], you have to include the added source impedance into the gain equation.

You may want to buffer the two bridge signals with another op amp or perhaps an emitter follower on each leg would be sufficient.

How accurate do you need to be with the unknown capacitor? There are other ways to do this that might perform better. It's good to look at multiple options. There's more than one way to skin a cat.

 

My capacitance values range from 10pF to 851pF. I'd like to get the best accuracy possible.

 

You still have not answered the question.
Why are you converting capacitance to voltage?
Are you trying to measure capacitance?

 

I need my values to be read by a daq device that reads voltage.

 

Ok. Then what I would do is measure the capacitance with an MCU and output an analog voltage via a DAC. Seems like a waste of a step.

Another solution is to use a F/V converter.

 

You can get a more accurate measurement of capacitance using the MCU (As MrChips said) measuring the period of a capacitor oscillator, here is an example of a high resolution version as the micro adds up thousands of periods to derive a cap value;

http://www.romanblack.com/onesec/CapMeter.htm