Would a piezo disk from a Motorola tweeter make a good strain guage?

[tauro0221]

Hi,
I just forgot the link where you can buy it.
here it is the link :https://www.sparkfun.com/tutorials/389

 

[kelticwizard]

I am completely blown away by the enormous number of responses and the information they contain, both about the project and helping with the foot as well.

Getting the piezo from a greeting card is a great idea.

The digital multimeter I have has a "data hold" which holds the info in case the leads get disconnected. Very nice idea but I don't think this a peak indicator-especially since some digital multimeters also have a "max hold" which does appear to be a peak indicator.

There is an analog meter in a store nearby, I'll check it out. Visually following a quick voltage spike is easier in an analog than a digital multimeter.

Just as a quick test, I plan to tape the piezo to my heel and then take a few steps barefoot. If I get a reading, I'll put on a pair of sneakers and see if the reading shows an appreciable drop. If it does, I think I'll be able to do an approximate reading of several brands. If not, then it's time to consider the many alternatives I have been presented with here.

 

[mickeymoose]

Why not use a strain gauge with an amplifier and an analog display: http://www.vishaypg.com/docs/11224/125un.pdf available from Digikey. E

 

[kelticwizard]

Well, I got the piezo out of a Radio Shack piezo transducer. I soldered it to a length of old speaker wire, and soldered the other speaker wire end to a 1/8 Walkman plug. I put the plug in the mic input of my netbook, which I can carry to the store.

I took the now connected piezo element and duct taped it to the bottom of my heel. It stuck to my sock admirably.

I decided to experiment at home before I went to the store to buy sneakers. I thought first I would test stepping in my stockinged foot on the carpet, and then in my sneaker on the carpet, to see how the readings differ.

I tried downloading Oscillloscope 2.51, a freeware app, to see the waveform. Definitely something was happening, but as I walked on the carpet with my shoe on and off, nothing was happening on the screen which made any sense. Indeed, the greatest reading was when I lifted my foot off the ground.

Better results came when I downloaded Peak Level Meter. Audio Level Meter

I gives an adjustable peak level and is great for freeware. The thing peaks about where you would expect it as you take a step, but it looks like I get similar readings from when I walk barefoot on the carpet and when I have the sneaker on.

I suspect that the piezo element might not do the trick in this application, and I need a new type of element. Let me check the many excellent recommendations given in this thread.

 

[JMFahey]

Well, I got the piezo out of a Radio Shack piezo transducer. I soldered it to a length of old speaker wire, and soldered the other speaker wire end to a 1/8 Walkman plug. I put the plug in the mic input of my netbook, which I can carry to the store.

I took the now connected piezo element and duct taped it to the bottom of my heel.

You need some kind of shielded wire ***and*** a Buffer.
I commented on my experiments using a 10Meg input one, a 1Meg one would do, sort of, but your mic input is around 600 ohms, what were you expecting?😕

It stuck to my sock admirably.

Easy to solve.
Treat your socks to some:

Ok, ok, don't me, I couldn't resist 😛
Back to DIY stuff:

I decided to experiment at home before I went to the store to buy sneakers. I thought first I would test stepping in my stockinged foot on the carpet, and then in my sneaker on the carpet, to see how the readings differ.

I tried downloading Oscillloscope 2.51, a freeware app, to see the waveform. Definitely something was happening, but as I walked on the carpet with my shoe on and off, nothing was happening on the screen which made any sense.

Scopes are great, but offer a "too raw" kind of data.
Often screen images don't seem to make sense.
"Everything is there", but often not in an easy to interpret or analyze way.

Indeed, the greatest reading was when I lifted my foot off the ground.

Not surprising; the small transducer capacitance and the low input impedance make an integrator, you are only seeing the highest frequency peaks, while the "pulse width" of a step, what you are trying to measure, is , say, in the order of a second.
To get an acceptable waveform you should be able to load your transducer so you have reasonably flat response down to 1 Hz or thereabouts.

Better results came when I downloaded Peak Level Meter. Audio Level Meter

I gives an adjustable peak level and is great for freeware. The thing peaks about where you would expect it as you take a step, but it looks like I get similar readings from when I walk barefoot on the carpet and when I have the sneaker on.

Well, peak meters will measure peak applied pressure (although the experiment is flawed because you are actually measuring the differential of the applied pressure.
What you actually want to see is the envelope of the pressure waveform between your foot and the shoe, so you can see (literally) differences in softness or damping among different sole materials or construction.
Your experiment is valid, but you must optimize it a bit further.

I suspect that the piezo element might not do the trick in this application, and I need a new type of element. Let me check the many excellent recommendations given in this thread.

The piezo is fine, in fact I suspect it's the best element for this type of application, unless you consider reading a strain gauge in real time and plot the strain vs. time curve.

Rather than switch transducers , you need to redesign both the "pickup head interface" and the way you will display that data.

Of course, it will take some time and effort, don't know if it's justified in this case, although in an University setting it would make for an interesting experiment, even a "last project thesis" to get a Degree or something.
FWIW, it's a perfect example of Biometrics 🙂

 

[jneutron]

Why not take apart a pair of kids shoes, the ones with the LED's?

Use the guts.

jn