Maker Pro
Maker Pro

Amplify signal from piezoeletric

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
Hello!

I am currently constructing a cough-sensor in which I used a ceramic piezoeletric sensor with a crystar part of about 1.3cm of diameter (obtained from a beepin-card:
). I now want to be able to use it to arduino.
But I have 2 problems:
1) The signal has too much low voltage. I have tried to amplify it with an OP07 and with a charge amplifier (as shown below) but I have failed!
The signal has an amplitude of around 200 mV, with a frequency ranging from 0.6Hz to 4 Hz.
What kind of amplifier (it has do be breadboard compatible!) can I use and what kind of montage?
2) How do I offset the signal? I have also tried a simple resistor summation circuit but it failed also!
upload_2017-6-28_22-23-51.png
Thanks in advance!
 
Last edited:

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The piezo produces a voltage output with almost no current. Your opamp is inverting with an extremely low input resistance so its input is a current, not a voltage.
If you make the opamp non-inverting with a high input resistance then the piezo will work fine until it is dropped then its very high output voltage spike will destroy the input of the opamp.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I think the charge amplifier should work, I used one on a scanning electron microscope 40 years ago to detect radiation.

What are the values of R and Cf?
What input specification does the amp have? A fet input amp would be preferred.
What is the capacitance of the detector.
I would try Cf of 100pF or less, this sets the gain.
I would try R of 10MΩ.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
The piezo produces a voltage output with almost no current. Your opamp is inverting with an extremely low input resistance so its input is a current, not a voltage.
If you make the opamp non-inverting with a high input resistance then the piezo will work fine until it is dropped then its very high output voltage spike will destroy the input of the opamp.
You mean inserting a high valued resistor between the non-inverting input and the piezo?
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
You mean inserting a high valued resistor between the non-inverting input and the piezo?
A resistor in series with the non-inverting input if a non-inverting opamp amplifier is useless. The non-inverting input must be biased at half the supply voltage with a high value resistor. If the supply is plus and minus then the bias resistor connects to 0V.
 

Attachments

  • opamp's gain.PNG
    opamp's gain.PNG
    7.6 KB · Views: 127

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
A resistor in series with the non-inverting input if a non-inverting opamp amplifier is useless. The non-inverting input must be biased at half the supply voltage with a high value resistor. If the supply is plus and minus then the bias resistor connects to 0V.
1) What about the other capacitors? what value should I give them?
2) Also, the capacitor C2 is connected to ground?
3) R6 and C2 are the ones defining the cut-off frequency right?
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The input capacitor is not needed if the input is from a piezo because a piezo is already a capacitor.
The C2 capacitor from the negative feedback divider to ground blocks DC but passes AC so that the opamp does not amplify its DC input offset voltage. Yes, R6 and C2 are a highpass filter producing the cutoff frequency. The series output capacitor also produces a highpass filter with its load resistance.

Another opamp must be used for the lowpass filter you want.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
The input capacitor is not needed if the input is from a piezo because a piezo is already a capacitor.
The C2 capacitor from the negative feedback divider to ground blocks DC but passes AC so that the opamp does not amplify its DC input offset voltage. Yes, R6 and C2 are a highpass filter producing the cutoff frequency. The series output capacitor also produces a highpass filter with its load resistance.

Another opamp must be used for the lowpass filter you want.

Thank you very much for the explanation! And I'm sorry for my complete unknown!

What is the load resistor u talk that produces the highpass filter?
What whay do you mean by "Another opamp must be used for the lowpass filter you want."? I can't use OP07? If so, what amplifier do you recommend me to? (that I can use in breadboard!)
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The load resistor is whatever this preamp opamp drives. Maybe its load is the high value bias resistor of a second opamp that is your lowpass filter.

A non-inverting opamp makes a poor lowpass filter when its negative feedback resistor has a capacitor in parallel because it can only reduce its high frequency gain to 1 like its DC gain. An inverting opamp that has its negative feedback resistor with a capacitor in parallel can reduce high frequencies very well to almost zero.

A solderless breadboard has many intermittent contacts. Its many rows of contacts and many wires all over the place have capacitance between them which can cause many circuits to oscillate instead of amplify and since they act as antennas they pickup all kinds of interference. Solder the parts on a compact pcb or stripboard with its strips cut to length instead.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
The load resistor is whatever this preamp opamp drives. Maybe its load is the high value bias resistor of a second opamp that is your lowpass filter.

A non-inverting opamp makes a poor lowpass filter when its negative feedback resistor has a capacitor in parallel because it can only reduce its high frequency gain to 1 like its DC gain. An inverting opamp that has its negative feedback resistor with a capacitor in parallel can reduce high frequencies very well to almost zero.

A solderless breadboard has many intermittent contacts. Its many rows of contacts and many wires all over the place have capacitance between them which can cause many circuits to oscillate instead of amplify and since they act as antennas they pickup all kinds of interference. Solder the parts on a compact pcb or stripboard with its strips cut to length instead.

Got it!
But since I only want to amplify my signal, I don't have to bother about the filters!
I am using solderless breadboard only for prototyping reasons, the next step is soldering (which I have learned yet!).
I will then try to use this circuit with an OP07! Thank you very much for your help! Hope it works :)
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
You said you are making a cough sensor with only low frequencies no higher than 4Hz. But a cough contains many higher frequencies.
Your circuit shows a lowpass filter of R with Cf in parallel creating its cutoff frequency. The circuit I showed with a non-inverting opamp does not and cannot have a good lowpass filter, a second opamp must be used to cut high frequencies.

An OP07 opamp is old with a poor high frequency response. I do not know if it is stable enough to be used on a solderless breadboard since the high stray capacitance might cause oscillation.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
Well, the signal is caused by movement of the thorax, I don't see it getting more than 15Hz (less than 66 ms) tho! Thanks for the warning :)

What opamp do you recommend me? I don't know many..

About the filtering, it can be done with the usual montage with ampops? or do I need a special opamp like in the amplification stage?
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
I did not see the video when I first replied. I thought the piezo was used as a microphone but the video shows it used to show breathing pressure on it. Maybe you are using it to show the pressures caused by coughing? The video shows ordinary wires as a cable which will pickup mains hum and other interference, it should be a shielded audio cable.

Your opamp is not powered. Since its (+) input is biased at 0V then the power supply to the opamp should have a positive voltage, 0V and a negative voltage. If your power supply only has a positive voltage and 0V then the opamp must have its (+) input biased with a voltage divider to feed half the supply voltage to the bias resistor.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
If your power supply only has a positive voltage and 0V then the opamp must have its (+) input biased with a voltage divider to feed half the supply voltage to the bias resistor.
Can you please sketch me this? and what resistors should I use?
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
Here is a simple non-inverting opamp with a single positive power supply:
 

Attachments

  • opamp.png
    opamp.png
    9.4 KB · Views: 148

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
I have never applied a voltage to the input of an opamp that is higher than its supply voltage so I do not know if the input is destroyed if you limit the over-voltage with a Schottky diode to about 300mV too high.The datasheets of most opamps do not say to use Schottky diodes, the datasheets say do not exceed the supply voltage.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
I am a bit unsure of what value to choose the the series output capacitor.
For the high-pass filter I chose C = 2.4uF and R = 100k in order to make a cut-off frequency of 0.66 Hz.
But I don't know what value to choose for the other capacitor...
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
If your feedback to ground resistor is 100k then the Rf must be an extremely high value to have any gain.
You still did not say what resistance this circuit is feeding so the value of the output capacitor cannot be calculated.
 

Diogo

Jun 28, 2017
29
Joined
Jun 28, 2017
Messages
29
Yeah, I know, but I don't want a gain much more that 5 :)
I have no idea what you mean by "resistance this circuit is feeding" sorry. I intend to connect it to an arduino!
 
Last edited:
Top