Piezoelectrical Project

[slakkie]

I am working on a force plate to see if my students are improving on their martial arts classes.

The force plate is a piece of thick plastic pith lots of padding. On the piece of plastic, we placed piezo electric sensors which give out voltage when the student hits the force plate.

The problem is, when I connect this to a computer data aquisition tool, the voltage just shows once and then disappears.

Do I need an osciloscope for this or is there any electronic kit that can help me like a logger tool that will log every few milliseconds?

 

[slakkie]

Oh I forgot, I am using a Velleman data usb aquisition card to read the sensor data.

 

[duffy]

The piezo sensors just give a spike when you hit it, then it goes away. An oscilloscope would be very helpful, but if your data aq. card can read the highest point on the spike, that's all you want anyway.

 

[slakkie]

Hi,

But it does not even read the spike or the highest point. Do you know perhaps of a company who sells a decent aquisition tool! I currently have Velleman k8055.

Thanks!!!

 

[duffy]

That one should work. How many piezos do you have and how are they connected to the Velleman's 2 analog inputs?

 

[stevez]

The duration of the peak value of the spike may be too short for the DAC to see. What you might do is add some means to get the peak to remain long enough for the DAC to see. I've seen op amp peak detectors where a capacitor is charged in proportion to the input voltage and it stays charged until leakage (or a deliberate reset) discharges the capacitor.

I do wonder if the peak is what you really want. I know little about martial arts, boxing etc. Is the peak velocity of a fist more significant than a fist with the inertia of the arm, shoulder/other body mass. Not sure which is what you want but those things may not be well indicated by just the peak.

 

[slakkie]

Thanks. Yeah regarding Martial arts, I would like to have an oscilloscope type effect using a PC oscilloscope aquisition tool but it is way too expensive. There is however a K112 Pocket sampler that I might be able to get to work it reads every few milliseconds. It might be a little easier to get the highest reading and then an average reading which will be much more usefull. I'm just trying to port the C++ and assembler code to windows xp delphi.

Thanks for your help!

 

[slakkie]

That one should work. How many piezos do you have and how are they connected to the Velleman's 2 analog inputs?

I am going to have a piezo connected to positive on my k8055 and negative to ground of analog input. I think the digital inputs are just counters of pulses or similar.

**broken link removed**

They say this thing can give lots of voltage especially if it is on plexiglass or some material that can bend alot. So I might fry my k8055. I might change potentiometer and get some resistors or something similar for one of our stronger students.

 

[slakkie]

That one should work. How many piezos do you have and how are they connected to the Velleman's 2 analog inputs?

I am just going to use 1 piezo sensor for now and later a few others connected in parallel mode.

 

[duffy]

The duration of the peak value of the spike may be too short for the DAC to see. What you might do is add some means to get the peak to remain long enough for the DAC to see. I've seen op amp peak detectors where a capacitor is charged in proportion to the input voltage and it stays charged until leakage (or a deliberate reset) discharges the capacitor.

I do wonder if the peak is what you really want. I know little about martial arts, boxing etc. Is the peak velocity of a fist more significant than a fist with the inertia of the arm, shoulder/other body mass. Not sure which is what you want but those things may not be well indicated by just the peak.

That last part is why I'm thinking a peak detector wouldn't be quite right. In addition to the peak, you would want to see the total area under the curve.

I actually have one of those red MSI sensors. I just hooked it up and smacked it around a little. Barely got more than a volt, and that's with no padding. "Lots of voltage" may mean something very different than you think - that's a HUGE voltage for a sensor, but not so much for damaging things.

The duration of the pulse was in the millisecond range, and it's AC.

If you have one side hooked to the positive supply and the other to analog ground, you're doing it wrong. Connect one side to analog ground and the other to the analog input pin. No positive supply at all. See how that reads.

What is the fastest samples/second on your Velleman? It ought to read in the khz range.

I think the peak would correspond to impact speed, and the area under the curve is going to give an indication of the total energy delivered by the blow.

Calibrate it by swinging a known weight on a rope from a known height, maybe.

 

[slakkie]

Yup. Good idea.

But these people at imagesco told me I can get alot of voltage by putting the sensor on a flexible object that can bend. The more the sensor bends, the better. Maybe they just want my money!

So what I can do is now hook up 3-5 sensors to the analog input and then check out the readings.

Or maybe I should try look for another supplier.

You know what.... If you go import these force plates from overseas in fong kong, they go for $29 and if I buy in bulk there go for $15 each. HOW DO THEY DO IT!!! LOL. But it's more fun doing the project yourself! Thanks guys

 

[slakkie]

That last part is why I'm thinking a peak detector wouldn't be quite right. In addition to the peak, you would want to see the total area under the curve.

I actually have one of those red MSI sensors. I just hooked it up and smacked it around a little. Barely got more than a volt, and that's with no padding. "Lots of voltage" may mean something very different than you think - that's a HUGE voltage for a sensor, but not so much for damaging things.

The duration of the pulse was in the millisecond range, and it's AC.

If you have one side hooked to the positive supply and the other to analog ground, you're doing it wrong. Connect one side to analog ground and the other to the analog input pin. No positive supply at all. See how that reads.

What is the fastest samples/second on your Velleman? It ought to read in the khz range.

I think the peak would correspond to impact speed, and the area under the curve is going to give an indication of the total energy delivered by the blow.

Calibrate it by swinging a known weight on a rope from a known height, maybe.

I don't know where to check for samples/second but it's probably very slow! That is what I am worried about lol!

1 KHz is equivalent to how many per second? Remember this Windows XP and not DOS!

 

[duffy]

You need 2k samples/sec to read a 1khz signal by Nyquist. There should be a setting in the software for it somewhere, and the speed is not going to be governed by the operating system, it will be controlled by the PIC chip, and that's...

...awww, WTF? "20ms conversion time" - ? Are they KIDDING ME? That PIC chip can do conversions a hundred times faster!!!
%&$#@!!

Sorry bud. Looks like someone screwed the pooch on the interface. The chip is willing, but the software is weak.

Ok, so plan B - send the output of the piezo through a pair of diodes to a cap. I found this schematic online, shows the circuit for a charge pump, which is what will work here -

**broken link removed**

Replace C1 with your piezo. Connect the part that says "hand-held probe" to analog ground. Use those diodes shown, or 1N914 diodes, or any general purpose small signal diode. Make C2 bigger, say .22µf. Connect the Velleman in place of the meter, (+) is the A1 analog input, (-) goes to to analog ground. Your input impedance on the Velleman is 100k to 200k, that will stretch this pulse out to the range that you can capture it.

A side benefit is that the impact waveform will be all positive now, no AC to interpret. You can adjust the amplitude with ATT1 on the Velleman.

 

[slakkie]

Thanks for all yer help!

 

[duffy]

Sure. You get it going?