motion/vibration sensor with piezo

[Rickharris]

The problem I think we have is that the piezo is a very simple device - It can produce sound from an applied electrical signal and can produce a spike of electricity from a sharp shock - Gentle movement doesn't effect them

- There are many products that are packaged in a similiar round package - a gyro for model helicopters for example.

[boriz]

This circuit should get you moving in the right direction:


A piezo is very much like a capacitor. In the above circuit, R1 should be a preset POT included on the circuit board and should be adjusted to the point where the transistor just starts to conduct (the collector voltage just starts to drop from the +V rail voltage). A fixed resistor would be better, but that would require a particular transistor. Using the POT, you can adjust it for any transistor you have handy. C1 is a negative feedback capacitor, there to ‘roll off’ the frequency response, so that the sensor is less sensitive to higher frequencies. You don’t want it to respond to sounds do you? The value of C1 shown here is just a quick guess. Try different values until your sensor responds to vibration but not to sound.

The piezo disk will need a little mod. Glue one edge to your project (or window frame or dog), fix a bolt to the other edge like this:


The weight on the head of the bolt will cause a twisting or bending force depending on the direction of movement, but it will respond to movements in any direction.

Make sure that:
A) The edge you glue to the project is opposite from the bolt. In the above diagram, that would be near the point where the leads meet the piezo disk.
B) The rest of the disk (and the bolt) are not touching anything and are free to move/flex.

[alexthefox]

Originally Posted by boriz

This circuit should get you moving in the right direction:


A piezo is very much like a capacitor. In the above circuit, R1 should be a preset POT included on the circuit board and should be adjusted to the point where the transistor just starts to conduct (the collector voltage just starts to drop from the +V rail voltage). A fixed resistor would be better, but that would require a particular transistor. Using the POT, you can adjust it for any transistor you have handy. C1 is a negative feedback capacitor, there to ‘roll off’ the frequency response, so that the sensor is less sensitive to higher frequencies. You don’t want it to respond to sounds do you? The value of C1 shown here is just a quick guess. Try different values until your sensor responds to vibration but not to sound.

The piezo disk will need a little mod. Glue one edge to your project (or window frame or dog), fix a bolt to the other edge like this:


The weight on the head of the bolt will cause a twisting or bending force depending on the direction of movement, but it will respond to movements in any direction.

Make sure that:
A) The edge you glue to the project is opposite from the bolt. In the above diagram, that would be near the point where the leads meet the piezo disk.
B) The rest of the disk (and the bolt) are not touching anything and are free to move/flex.

thank you. i will try to do what you show.

[alexthefox]

i find this in google : http://images.google.it/imgres?imgur...%3Dit%26sa%3DN

[boriz]

oops!

Assuming your +v is 5v, R1 should be 100k NOT 10k. Sorry.

[alexthefox]

Originally Posted by boriz

oops!

Assuming your +v is 5v, R1 should be 100k NOT 10k. Sorry.

it not work. i cant ee the different.

[boriz]

Ok. Try this:


I just tried it and it works for me. And it seems to need no roll-off. Prolly coz direct physical distortion of the transducer creates much more output than any sound could.

If you can’t get enough output from this for your project, use another transistor stage or an op-amp.

I’m not going to design this project for you. I’m just trying to help you get started.

[Dippy]

I did a very similar circuit to Boriz's years ago, except I used an N chan MOSFET (ZVN something) and biased the gate with an LED+1M res. (Not all LEDs are sutiable, experimatation needed). The LED+1M povided a 'floppy' bias voltage so that the feeble piezo output comforatbaly took it over the gate threshold.
It worked well, and was very sensitive from 0oC to 50oC.
I may have included a zener I can't remember.

I used a different set up with the disk piezo as I needed a vertically biased vibration sensor to use on wooded steps. Overall it was pretty good and even a gentle step-up onto step triggered it, whereas sideways taps on edge of step (unless vigorous) were usually ignored.

PS: Boriz, do you say "prolly" when speaking to people? Do they roll their eyes when you say it? Are you hoping it will be put in the Dictionary?

[boriz]

No. But it's easier to type. Does it bother you?

[alexthefox]

Originally Posted by boriz

Ok. Try this:


I just tried it and it works for me. And it seems to need no roll-off. Prolly coz direct physical distortion of the transducer creates much more output than any sound could.

If you can’t get enough output from this for your project, use another transistor stage or an op-amp.

I’m not going to design this project for you. I’m just trying to help you get started.

and i will to thank you for your help. i try to test it and write here the result asap.