Science Buddies: "Ask an Expert"

Hello,

I am beginning to work on a school science project involving the use of (prewired) piezoelectric sensors (most likely 35 - 41 mm wide and .35 mm thick "piezoelectric elements," as they are called). I am planning to find out how the type of mechanical stress applied to these sensors affects output voltage (among these I am most likely to choose compression, bending, torsion, and shear). Despite having done a decent amount of research concerning the piezoelectric effect, I have run into some problems/specifics that are a bit over my field of knowledge. These problems include the following:

- Does the physical size (dimensions, as in an arbitrary object like a rock) of an applied mechanical force influence output voltage for a piezoelectric sensor? ex. would a "large" rock weighing 10 grams produce a higher voltage from the piezo sensor than a "small" rock weighing 10 grams? And if so or if not, why?

- What is an efficient, inexpensive way to simulate these varying types of mechanical stress on individual piezoelectric sensors while at the same time maintaining a constant level (magnitude) of force? ex. a constant 50 grams for each type of stress (I plan to have a different experimental group for each type).

- What is a practical, safe range of force (in metric units) to apply to a piezoelectric sensor that is around 35 - 41 mm wide and .35 mm thick (taking into consideration output voltage and flexibility)?

I would also greatly appreciate any recommendations or tips that could help guide me toward the right path in my experiment, whether it be a note on something that I am missing or could improve on, additional facts/information I could use, or a helpful source for relevant information on the topics I am working with.

Thanks!

bvionis wrote:Hello,

I am beginning to work on a school science project involving the use of (prewired) piezoelectric sensors (most likely 35 - 41 mm wide and .35 mm thick "piezoelectric elements," as they are called). I am planning to find out how the type of mechanical stress applied to these sensors affects output voltage (among these I am most likely to choose compression, bending, torsion, and shear). Despite having done a decent amount of research concerning the piezoelectric effect, I have run into some problems/specifics that are a bit over my field of knowledge. These problems include the following:

- Does the physical size (dimensions, as in an arbitrary object like a rock) of an applied mechanical force influence output voltage for a piezoelectric sensor? ex. would a "large" rock weighing 10 grams produce a higher voltage from the piezo sensor than a "small" rock weighing 10 grams? And if so or if not, why?

- What is an efficient, inexpensive way to simulate these varying types of mechanical stress on individual piezoelectric sensors while at the same time maintaining a constant level (magnitude) of force? ex. a constant 50 grams for each type of stress (I plan to have a different experimental group for each type).

- What is a practical, safe range of force (in metric units) to apply to a piezoelectric sensor that is around 35 - 41 mm wide and .35 mm thick (taking into consideration output voltage and flexibility)?

I would also greatly appreciate any recommendations or tips that could help guide me toward the right path in my experiment, whether it be a note on something that I am missing or could improve on, additional facts/information I could use, or a helpful source for relevant information on the topics I am working with.

Thanks!

When piezoelectric material is placed under mechanical stress, a shifting of the positive and negative charge centers in the material takes place, which then results in an external electrical field. When reversed, an outer electrical field either stretches or compresses the piezoelectric material.

The principle of operation of a piezoelectric sensor is that a physical dimension, transformed into a force, acts on two opposing faces of the sensing element. The detection of pressure variations in the form of sound is the most common sensor application, which is seen in piezoelectric microphones and piezoelectric pickups for electrically amplified guitars. Piezoelectric sensors in particular are used with high frequency sound in ultrasonic transducers for medical imaging and industrial nondestructive testing.

I appreciate the help, but I will be honest: this does not help me. I am not necessarily in need of information concerning the piezoelectric effect and the theory itself.

It is the specific questions posted that I need an answer to (or any comments, advice, etc. on my actual science project).

Again, I appreciate your taking the time to respond to my post with the intention of helping me, but it is the above-stated material that I am looking for.

P.S. - I don't recommend directly quoting from Wikipedia without at least saying that you did!

Hi bvionis,
This sounds like an interesting experiment.

In regards to your questions, the behavior of the piezoelectric sensor will largely be a function of its design and packaging of the pre-wired sensor. (Do you have a specific manufacturer part number that you are planning to work with?)

Here's an example of a typical pre-wired sensor that you can measure the analog voltage output.
https://www.mouser.com/datasheet/2/418/ ... 129857.pdf

As you can see, a sensor in the range you described (20mm x 40mm) can handle 0 to 100PSI. It really depends on the specific model that you plan to use.

In terms of simulating the varying type of mechanical stress, I think we are probably looking at load cells / strain gauge / torque meter. Does your school have a automotive lab / shop? You might be able to get access to these equipment there.

Good Luck with your experiment and please post again if there's anything else we can help answer.
Willey

Hi bvionis,

In addition to the other replies here, I wanted to mention that it's important that you understand the difference between force and stress. In the metric system force is measured in Newtons [N] and stress is measured in the same units as pressure - force per unit area, or newtons per square meter [N/m^2]. One newton per meter squared is equal to one Pascal [Pa].

This is very important because you can apply the same force to an object in different ways or locations to result in very different stresses, and stress is what determines whether or not a material will break. In the case of piezoelectric material, it's the stress (and resulting deformation of the material), not the force, that will generate the voltage.

To take a simple example, imagine standing flat on your feet vs standing on your toes. Your weight (the force) is the same in both situations. However, the surface area of your tiptoes is much smaller than your feet, so the pressure (stress) is much higher.

Another slightly more complicated example - imagine clamping one end of a ruler to a desk, leaving the other end sticking out, then pushing down on the ruler with your finger. You can apply the same force anywhere along the ruler with your finger, but the internal stresses in the ruler (and the amount it bends) will be higher if you push it farther out (towards the free end).

In your case, you need to be careful because applying 50 grams of force in the different configurations you mentioned (compression, bending, torsion, and shear) will NOT necessarily result in the same stress in the material, so you can't really compare them directly. A specific type of piezoelectric sensor may also only be designed to measure certain types of loads (e.g. bending or compression but not torsion or shear).

That's a VERY quick overview of what is really a college or even graduate level engineering topic. In my case - I learned about force and stress etc. in my junior year of a mechanical engineering degree; and didn't learn about piezoelectric materials until graduate school. So don't hesitate to write back if you have more questions.

Thanks,

Ben

PS - if you're curious about a cool application for piezoelectric materials - I did my PhD on the "Robobee" project - the wings are flapped by tiny piezoelectric beams (so like a sensor working in reverse - instead of applying a force and generating a voltage, you apply a voltage and it generates a force): https://www.google.com/search?q=robobee ... 80&bih=947

Thank you Willey and Ben. You two helped me immensely.

I am trying to figure out what the metric unit for stress is, considering the fact that for this project I am required to have my units metric (not SI). I have attempted to research for an answer but cannot seem to find a definite unit.

Also, would something as simple as a set of small weights work to simulate a stress (producing compressive deformation/strain) on one of the above-mentioned kind of piezoelectric sensors/elements? (related to my second experiment.)

Thank you!

P.S. - I also looked at the Robobee project you worked on @bfinio (I have actually already heard about it via tech news sites, etc.). Saying that it is admirable work would be an understatement!

Hi bvionis - two answers:

-The metric and SI units for stress are the same - the pascal [Pa]. One pascal is equal to one newton per square meter [1 Pa = 1 N/m^2].
- The answer to your second question is "it depends on the sensor." You would need to look at the data sheet and information for the type of sensor you plan to buy to figure out how you can attach a load to it. Unfortunately I don't have experience working with commercial piezoelectric sensors directly (for the Robobee project we laser cut our own piezoelectic pieces from larger sections of the material). For example, I'm not sure if there are sensors that essentially work like a bathroom scale where you can just place weights on top of them (it sounds like that's what you'd like to do). If you have a link to the specific type of sensor you plan to buy then LeungWilley or I could probably take a look.

Here is the link for the kind of sensors I am likely to purchase (either the 35 mm ones or 41 mm ones posted in this link):

https://www.amazon.com/15-Pieces-Elemen ... 52&sr=8-25

I have ordered a sample of these and have found that a simple compressive stress applied (something like my finger) can produce an output voltage (I just connected the prewired leads to a multimeter); however, I have not yet purchased any weights or anything similar to actually provide myself with a consistent (compressive) force. I am wondering if anyone knows any affordable, practical way to attain something that will provide me with the option of having varying stress-inducing weights, etc. that increment in their values (like a pack of weights that would conveniently serve this purpose. This part I am not too worried about, as I can continue to look for some, or do as discussed below).

I do not believe that any small weights that I could use would be commercially measured by the magnitude of the stress they cause, but I was wondering if just simply calculating the area of a weight and finding the stress from there could give me an accurate measurement of the stress that weight will exert. (I am not currently talking about the above-stated problem; I have decided that, considering the time and resources available to me, it will not be practical to attempt to simulate differing stress types. I am just talking in general for another differing problem.) I am planning for the incrementing stress values to correspond to different experimental groups (for accuracy purposes).

Hi bvionis,

For many of our projects where people need to put weights on something, we suggest just using coins. They're readily available and actually pretty cheap for the amount of weight they provide. For example, this paper bridge project: https://www.sciencebuddies.org/teacher- ... -bridges-2

Otherwise, I remember using these types of weights in science class, but they are probably much more expensive for an equivalent amount of weight: https://www.amazon.com/Ajax-Scientific- ... B00EPQMEWC

However, you need to be careful about how you calculate stress, and whether you use the area of the weight or the area of the piezoelectric disk. For example, say you stack a pile of pennies directly on top of the white part of the piezoelectric sensor. A penny's diameter is smaller than the piezoelectric disk. If you wanted to calculate the contact stress between the surface of the penny and the surface of the disk, you would divide the weight of the penny by the area of the penny. However, assuming the top part of the piezoelectric disk is rigid enough, to calculate the internal stress you would use the area of the white part of the sensor (which from the pictures I'm assuming is the "top" part). In this case it's the internal stress in the piezoelectric material that generates the voltage, so that's the value what you want.

Put another way - imagine that you are standing on top of something fairly rigid, like a large piece of wood, that is resting on the ground. To calculate the contact stress between your feet and the wood, you would divide your weight by the surface area of your feet. However, to calculate the contact stress between the wood and the ground, you would divide your weight (plus the weight of the wood, if it's heavy enough that you can't neglect it) by the area of the wood. In this case you want to do the latter.

Hope that helps,

Ben