audiopiezo-buzzerpiezoelectric-effectspeakers
Can piezo-buzzer be interfaced with MCU same as normal 5V continuous buzzer?
I am talking about the flat thinnest buzzer like this.
We need to feed PWM to this to create different tones right? If yes, do we need to provide PWM at audio frequency or any other resonant frequency?
I can't find any proper datasheet for such parts, so finding exact parameters are difficult.
Is there anything I can do in my circuit in order to make the horn louder other than using a higher voltage which is not an option?
No, there isn't. With a 24 volt peak-to-peak square wave into the voice coil, that's the same as feeding 12 volts DC into a 6 ohm resistor, which means that - under those conditions - the power your voice coil is dissipating is 24 watts.
Also, there's a problem with the data sheet, since that 1 ampere current limit translates to 12 watts, not their specified 20 watts max.
Will going to a 4 ohm speaker be louder than an 8 ohm speaker in this case?
Maybe, maybe not. The speaker's impedance doesn't determine how loud it is, what matters, basically, is the geometry of the horn, how far the diaphragm can flex, and how much current you can put into the voice coil before it melts.
If you want to use the audio jack output of a phone, you can't use frequencies above the hearing limit. Because the phones all contain a DAC followed by a low-pass filter that attenuates everything above 20kHz or so. So the phone output will never be able to produce ultrasounds.
As a consequence, the simplest is to use old-fashioned DTMF, as you suggested yourself. There are a bunch of chips able to do that, for example the Microsemi MT8870. The figure 9 in the datasheet shows a typical configuration that should fit your needs, and you'll see there are very few external components. And this way, the software arduino part does not need to do the actual DTMF decoding, and you get 16 possible output states (if it's not enough, you could use sequences of multiple tones).
By the way, nice idea, overall. Very simple and effective.
Best Answer
No, what you show needs to be driven differently than a "buzzer". It's not a buzzer, just a piezo transducer. These things don't oscillate on their own. You have to drive it with the signal you want to emit, not just apply power as with a buzzer.
To know how to drive this buzzer, read it's datasheet. The supplier you bought these from should be able to supply a datasheet. If not I wouldn't trust them.
If you can't get hold of a datasheet for these parts, look around on web sites of reputable suppliers for something similar. At least that will give you a rough idea.
Piezo elements look mostly capacitive to the driving circuit, but can also exhibit kickback like inductors and can generate high voltages when subjected to mechanical shocks.
For starters, connect one of these between a microcontroller PWM output pin and ground. Add reverse Schottky diodes to power and ground to protect the microcontroller output. Drive it with 1 kHz square wave and see what you get. Look at the voltage and see how clean it still is after being loaded by the piezo. If it's a mess, then that's a good clue that the micro output isn't strong enough to drive the piezo well at that voltage.
You can try putting a double emitter follower (NPN and PNP) on the output of a opamp to drive the piezo from a higher voltage and with lower impedance. What you show can probably take a few 10s of volts without damage, but of course without a datasheet we don't really know what the limits are.