Transistor Switch behaviour

[prprog]

I built the attach circuit. It work as expected, if I press the Piezo the LED is set on for a short moment. It does respond on how hard I press the Piezo. Which changes do I have to make to the circuit to make the LED ON as long as I press/touch the Piezo sensor?

Thanks,
PRPROG

 

[MikeMl]

A piezo generates a voltage while it is being displaced (it is a rate-of-change dependent device). If it produced a voltage continuously while displaced, you could use it to make a perpetual motion machine.

 

[prprog]

A piezo generates a voltage while it is being displaced (it is a rate-of-change dependent device). If it produced a voltage continuously while displaced, you could use it to make a perpetual motion machine.

Thanks MikeMI. Can we consider pressure a rate-of-change? So as long as the piezo is being press does it generate a voltage? (it might generate different voltage but can we buffer..or charge it to capacitor and used it as a way to activate the transistor?

Thanks,
PRPROG

 

[MikeMl]

Thanks MikeMI. Can we consider pressure a rate-of-change?

No, only if the pressure is pulsing. If the rate of displacement of the piezo is zero, the output is zero.

So as long as the piezo is being press does it generate a voltage? (it might generate different voltage but can we buffer..or charge it to capacitor and used it as a way to activate the transistor?

No, read my previous post, again.

 

[crutschow]

Depending upon how long you want it to operate, you may be able to connect it to a MOSFET input which has a very high input impedance. You would still need a high value resistor (say 10MΩ) from the MOSFET gate to ground for a DC path. If you connect the piezo directly to a digital DC multimeter, that will tell you how much voltage you get and how long it takes to decay when you press the button (standard digital multimeters typically have a 10MΩ input).

 

[dr pepper]

If you were to use a d type flip flop such as a 4013 you can feedthe flip flops clock with your circuit and then you'd have a press on then press off circuit which is close to your needs.

If you want press and hold for on it can be done but you'd need some more complex electronics, an ac signal source which can be applied to the piezo and then some way of measuring the o/p, pressure on the piezo is going to change its resonance and affect the o/p, however this would entail an amount of 'messing'.

 

[ronv]

I have used a resistive sensor like this:

https://www.sparkfun.com/products/9375

 

[prprog]

I have used a resistive sensor like this:

https://www.sparkfun.com/products/9375

Thanks. Great suggestion. Will considered this as a an alternate way to get a continöus preasure.
I am also experimenting with a different transistor type; a FET transistor . Will report shortly the findings.

Thanks,
PRPROG

 

[alec_t]

Instead of using a piezo you could just rely on finger conductivity to make a touch-sensitive switch.

 

[prprog]

Another option - FET

I built this FET switch based on sometime I read about FET behavior as switch. Well...It did switch the LED on and it stay on as long as I press the Piezo, but it is unreliable. It stay a few seconds after I release the pressure on the piezo. It also turn on the LED at any time...without even pressing the piezo. How can I retain the "stay on LED as I press the Piezo" behavior but make it reliable.

Thanks,
PRPROG

 

[()blivion]

FYI, the MPF102 is a JFET, not a MOSFET. Slightly different parts, but for this, I would assume it would be critical.

Incidentally, *IF* the Pizeo element works the way I think it does, then it puts out a voltage of one polarity when pressed down, then when released outputs the same voltage, but of opposite polarity. With this in mind, something like a peak detector could do what you want maybe.

 

[alec_t]

How can I retain the "stay on LED as I press the Piezo" behavior but make it reliable.

Follow DrPepper's suggestions in post #6.

 

[dr pepper]

Thanks mr t

 

[prprog]

Follow DrPepper's suggestions in post #6.

OK. The behavior I need is the one described on the second paragraph:
" If you want press and hold for on it can be done but you'd need some more complex electronics, an ac signal source which can be applied to the piezo and then some way of measuring the o/p, pressure on the piezo is going to change its resonance and affect the o/p, however this would entail an amount of 'messing'. "

Honestly I have no idea where to start . Can you kindly provide a suitable circuit I can start exploring this option?
Thanks,
Victor

 

[dr pepper]

No but I can describe what I meant.

depends on the size of the piezo but they usually resonate at 2kc, you could measure it scientifically but I'd make up a 555 variable frequency astable oscillatpr circuit on breadboard and try it, the sound will change/get louder on resonance.
If you apply the o/p of the 555 to the piezo through say a 10k resistor, then from the piezo rectify the ac using a 1n4148 diode and a 100nf cap, you can then apply this rectified dc voltage to a lm393 comparator.
Touching the piezo will shift the resonant freq away from the 555's o/p and the voltage will drop triggering the o/p of the comparator.
One drawback that could be major is that the piezo will make a beeping sound all the time.

This might sound tricky but its fairly trivial, maybe one the guys will run it through spice for you, that is if they have the piezo definition file.

Theres probably an even simpler way of doing it too.

 

[alec_t]

Depending on the piezo you have, if it produces an output as ()blivion mentioned in post #11, and if the output is >20V or so then the following circuit SHOULD work in theory (I haven't tried it in practice). To make the circuit more sensitive the 10k resistors can be increased in value to, say, 100k :-

U1a is an inverter. U1b and U1c are cross-coupled to form a latch. A positive-going pulse from the piezo sets the latch one way and a negative-going pulse sets it the other way.

 

[ronv]

Do piezos work that way. I always thought they just put out a ringing waveform, both + and - .

 

[alec_t]

Seems they do work that way. I've just tried a piezo attached to a DMM set on the 200V range. I got ~30V decaying to 0V in ~1 sec with a steady press, then ~-30V decaying to 0V in ~1 sec on release. The decay times and actual voltages are unreliable because of the response time of the DMM, but the waveform is generally as ()blivion mentioned. When I have the time I'll dig out a scope and verify it, also check for ringing.

 

[Roff]

Here is a good explanation of how piezo sensors work.

 

[MikeMl]

Here is a good explanation of how piezo sensors work.

I guess I used words that were too big back in post #2 and #4...