# Twin T oscillator + sensor

#### shreyas_bhat

Joined Jul 26, 2004
47
Hi Folks
am trying to build a sensor system. It consists of capacitive sensors (20 nF) which change capacitance when they sense a particular substance.
To measure this change in capacitance I am using them as part of a Twin T oscillator network, so that when the capacitance changes, the oscillator frequency also changes. I now need to build a circuit to measure the frequency change.

I am not able to attach an image of the system that I am building. However, I'll try to explain. Could I use the following set up to detect the frequency change.

Twin T oscillator ---------->Phase detector------>Low pass filter----> Voltage out
I I
I I
Freq. out <------------------VCO-----------------------------

Could this be achieved straight using a PLL IC.

Please send me all the inputs. I would really appreciate if you could send me a block diagram of the circuit that I would need. In case you are unable to send it through this, you could email it to me at bhatshreyas@gmail.com

Thanks
Shreyas

#### hgmjr

Joined Jan 28, 2005
9,027
Can you reveal the total range of capacitance change between the minimum and the maximum value? For example, the capacitor changes from a minimum of 18nF to a maximum of 22nF.

The magnitude of the capacitance change will dictate the most appropriate circuit to use.

#### shreyas_bhat

Joined Jul 26, 2004
47
The range would be from 19nF to 21.5nF. Please note that even a slight variation of capacitance in this range (say 0.02 nF - 0.05 nF) would have to be picked up as a change in frequency.

Thanks

Originally posted by hgmjr@Mar 23 2005, 08:41 PM
Can you reveal the total range of capacitance change between the minimum and the maximum value? For example, the capacitor changes from a minimum of 18nF to a maximum of 22nF.

The magnitude of the capacitance change will dictate the most appropriate circuit to use.
[post=6337]Quoted post[/post]​

#### hgmjr

Joined Jan 28, 2005
9,027
Are there any special constraints to the design. For example, are you looking for something portable or will you be using this measurement instrument in the lab where you have access to AC power?

Do you have any special display requirements?

How often do you need the measurement to update? Once per second, 10 times per second, once a minute?

How far away from this measuring instrument will the container of "particular substance " be located?

From your response above, you are looking at detecting changes of less than 1% in the capacitive transducer.

#### p8ntman442

Joined Mar 30, 2005
1
I too am interested in a way of detecting fractions of nano farad capacitance change. Mine is for a nano gas detecting sensor project. Where the change in gas between two plates creates a change in capacitance. the only thing is the change is in the pico farad range. My thoughts are to add several sensors in series therefore the total capacitance change is larger and easier to detect.

#### shreyas_bhat

Joined Jul 26, 2004
47
Yes, I shall be making measurements close to an AC course. No special display requirements.
I'm planning to observe it on an impedance analyser.
Measurement has to be made 10 times / sec.
The substance flows between the plates of the capacitor. However the circuitry will be isolated from the substance.

Originally posted by hgmjr@Mar 24 2005, 07:26 PM
Are there any special constraints to the design. For example, are you looking for something portable or will you be using this measurement instrument in the lab where you have access to AC power?

Do you have any special display requirements?

How often do you need the measurement to update? Once per second, 10 times per second, once a minute?

How far away from this measuring instrument will the container of "particular substance " be located?

From your response above, you are looking at detecting changes of less than 1% in the capacitive transducer.
[post=6367]Quoted post[/post]​

#### hgmjr

Joined Jan 28, 2005
9,027
Originally posted by shreyas_bhat@Apr 3 2005, 08:33 PM
Yes, I shall be making measurements close to an AC course. No special display requirements.
I'm planning to observe it on an impedance analyser.
Measurement has to be made 10 times / sec.
The substance flows between the plates of the capacitor. However the circuitry will be isolated from the substance.
[post=6691]Quoted post[/post]​
As an alternate to the Twin-T oscillator technique, I believe there may be some significant merit in exploiting the relationship dv/dt = I/C.

Using a constant current source applied to the capacitive transducer, a variation in the capacitance would result in a change in the slope of the voltage ramp. You would need a bit of additional circuitry of course to produce a reference ramp that could then be used to nullify the component of dv/dt associated with the fixed portion of the capacitance. More circuitry would be needed to turn the difference in ramp rates between the reference ramp and the transducer generated ramp into a final voltage that would represent the magnitude of the quantity you are trying to measure. I don't think this additional circuitry would be that exotic.

I believe this technique is patented.

I can elaborate if you are interested.

#### shreyas_bhat

Joined Jul 26, 2004
47
Hi there
i apologise for getting back on this so late. But, if you can elaborate the technique u mentioned, that would be of immense help.

Thanks for all ur help so far

Shreyas

Originally posted by hgmjr@Apr 6 2005, 06:59 AM
As an alternate to the Twin-T oscillator technique, I believe there may be some significant merit in exploiting the relationship dv/dt = I/C.

Using a constant current source applied to the capacitive transducer, a variation in the capacitance would result in a change in the slope of the voltage ramp. You would need a bit of additional circuitry of course to produce a reference ramp that could then be used to nullify the component of dv/dt associated with the fixed portion of the capacitance. More circuitry would be needed to turn the difference in ramp rates between the reference ramp and the transducer generated ramp into a final voltage that would represent the magnitude of the quantity you are trying to measure. I don't think this additional circuitry would be that exotic.

I believe this technique is patented.

I can elaborate if you are interested.
[post=6757]Quoted post[/post]​