Charge filter for piezoelectric sensor

[spadecrank]

charge amplifier project

Hi all,

I am trying to design a three pole low pass RLC Butterworth filter for a piezoelectric transducer, the piezoelectric transducer's output is charge instead of voltage. The piezoelectric transducer can be modeled as a voltage source in series with a capacitor. When I put a typical three pole low pass RLC Butterworth filter topology together with this model, it would not give me expected low pass result. Anybody has idea about it ? Thank you

 

[TekUT]

Could you post the circuit schematic that are you simulating?

Regards
Powermos

 

[spadecrank]

I simulated a 7 pole low pass butterworth, its similar to 3 or 5 pole topology[/url]

 

[kender]

Would you better off if you convert the charge to voltage using the charge amplifier topology and then filter the voltage?

 

[spadecrank]

The filter is supposed to bridge the sensor and the charge amplifier, which is the requirement of my project. Thanks for the advice anyway.

 

[kender]

The filter is supposed to bridge the sensor and the charge amplifier, which is the requirement of my project.

Why such requirement? How (word-for-word) is it formulated?

Thanks for the advice anyway.

You're welcome.

 

[FvM]

There is a basic fault in problem formulation to my opinion. If you describe the sensor as a charge generator, then the equivalent circuit is a current source with a parallel capacitor. In this case, you can make the sensor capacitance part of a passive low pass filter in a shunt element first topology.

When you assume a voltage source with a series capacitance and conncet a passive filter with a real impedance, then you create an additional high pass and won't never achieve the intended characteristics.

 

[spadecrank]

Thanks FvM, your suggestion makes a lot of sense. But how am I going to get a current output from the filter since the next stage of the filter is a charge amplifier ?

 

[FvM]

The charge amplifier input is low impedance, so the filter output can be connected through it's nominal load impedance (usually a resistor connected to ground) to the charge amplifier. Connecting a load is necessary to achieve the nominal frequency characteristic.

 

[spadecrank]

The charge amplifier input is low impedance, so the filter output can be connected through it's nominal load impedance (usually a resistor connected to ground) to the charge amplifier. Connecting a load is necessary to achieve the nominal frequency characteristic.

Is the input of a charge amplifier supposed to be high impedance ?
please check the online reference http://people.na.infn.it/~barbarin/Biblioteca%20rivelatori%20ed%20elettronica%20(per%20studenti)/rivelatori/hamamatsu/Characteristics_and_use_of_Charge_Amplifiers.pdf

Now my confusion is that the filter accepts charge signal and output a voltage signal, so it can not be directly connected to a charge amplifer which accepts charge signals?

 

[LvW]

Is the input of a charge amplifier supposed to be high impedance ?

I suppose, your question arises because the wording of the referenced article is somewhat misleading.
The amplifier itself (without capacitice feedback) is an high impedance device;
The amplifier with feedback (and because of high open loop gain) has a low impedance input at the inverting input.