Hi All,

I plan to build a charge and voltage amplifier for piezoelectric sensors (patches). The maximum frequency of the output signal would be 500kHz, with the output amplitudes ranging from a few millivolts to 0.3 V. I have chosen OP37EPZ as its slew rate is 17microV/s and GBW product is 63 MHz. Though the theoretical calculations show that this would do the job, I would like to hear your suggestions.

I saw another option (LT1221CN8), with a better slew rate and GBW product, but then its input bias current and offset voltage are bad. Not sure how this would affect my output.

Please give your thoughts about this. OR, suggest an opamp more suited for my purpose.


Circuits:
https://www.allaboutcircuits.com/te...charge-amplifiers-for-piezoelectric-sensor-s/

OpAmp:
https://www.digikey.com/en/products/detail/analog-devices-inc/OP37EPZ/671220
https://www.digikey.com/en/products/detail/analog-devices-inc/LT1028CN8-PBF/889830

Best,
Shiva

 

I think that they are both poor choices.
First, the OP37 is NRND (Not recommended for new designs).
Both devices have large input bias current (100nA) - where is that going to come from in a charge amplifier?
The current noise is poor (1pA/√Hz) Multiply that figure by the input resistance .
The minimum power supply is large (±4.5V)

You have chosen two devices that are designed to produce very low noise with low impedance sources, and you have one of the highest impedance sources available.
You need JFET amplifiers with current noise figures in the low fA/√Hz region, and almost zero bias current.
Do you really expect signals of 500kHz from a mechanical transducer?

 

What is the value of the piezo signal (charge generated)?

 

Hi Ian,

Thanks for your response. We currently use the disc to generate and measure 125kHz, 250kHz and 400kHz tone burst signals. We would be using it for 500KHz.

 

What is the value of the piezo signal (charge generated)?

Not sure about it. We have some amplifiers from Olympus which we used to measure the output voltage.

 

Hi Ian,

Thanks for your response. We currently use the disc to generate and measure 125kHz, 250kHz and 400kHz tone burst signals. We would be using it for 500KHz.

Not quite what I thought you meant by a sensor. I was thinking of something that produces a voltage when it is moved.
Is it part of a transmitter - receiver system?

 

Not quite what I thought you meant by a sensor. I was thinking of something that produces a voltage when it is moved.
Is it part of a transmitter - receiver system?

What you thought is correct. I meant to say that we have used these sensors to detect structural vibrations at 125 kHz and more.

 

As noted, you should use a high input-impedance, low-noise op amp.
Take a look at the TL074xH series, for example.
You should likely configure it in a non-inverting gain configuration.

 

I think that they are both poor choices.
First, the OP37 is NRND (Not recommended for new designs).
Both devices have large input bias current (100nA) - where is that going to come from in a charge amplifier?
The current noise is poor (1pA/√Hz) Multiply that figure by the input resistance .
The minimum power supply is large (±4.5V)

You have chosen two devices that are designed to produce very low noise with low impedance sources, and you have one of the highest impedance sources available.
You need JFET amplifiers with current noise figures in the low fA/√Hz region, and almost zero bias current.
Do you really expect signals of 500kHz from a mechanical transducer?

Hi Ian,

I found OPA656/OPA657. These are discussed as ideal choices for ultrasonic sensors in a thesis. I think this fits your recommendation as well. What do you think?

 

Hi Ian,

I found OPA656/OPA657. These are discussed as ideal choices for ultrasonic sensors in a thesis. I think this fits your recommendation as well. What do you think?

Seems like a good choice. Pricey, though! I agree with @crutschow : Good old TL071 types would probably do the job.

 

Seems like a good choice. Pricey, though! I agree with @crutschow : Good old TL071 types would probably do the job.

I'll check them out as well.

I have a query: all these high freq opamps have a relatively high input offset voltage (compared to the general-purpose ones I listed earlier). Will this affect the output of the circuit (Charge and voltage amplifier) in any way?

 

As noted, you should use a high input-impedance, low-noise op amp.
Take a look at the TL074xH series, for example.
You should likely configure it in a non-inverting gain configuration.

Thanks! Looks like these should do.

 

I'll check them out as well.

I have a query: all these high freq opamps have a relatively high input offset voltage (compared to the general-purpose ones I listed earlier). Will this affect the output of the circuit (Charge and voltage amplifier) in any way?

You young folks have never had it so good. A few decades ago, that was a good offset voltage!
It shouldn't be a problem as there is no DC gain which would turn it into a bigger problem.

 

You young folks have never had it so good. A few decades ago, that was a good offset voltage!
It shouldn't be a problem as there is no DC gain which would turn it into a bigger problem.

Thanks A Lot for this clarification and for your suggestions.
This has greatly helped me. Have a great day!!!

 

As noted, you should use a high input-impedance, low-noise op amp.
Take a look at the TL074xH series, for example.
You should likely configure it in a non-inverting gain configuration.

Why the non-inverting configuration over inverting?

 

I don't want to speak for anybody else, but I would imagine the input impedance.