• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Driving a Piezo Tweeter ?

Status
Not open for further replies.

ozarkshermit

New Member
Howdy:

I am attempting to build an ultrasonic "Dog Chaser" using a classic push-pull driver (circuit attached), and a cheap Piezo Tweeter. I use a PIC to output varying frequencies of 15 - 17 - 21 - 23 - 25 - 27 - 30 and 33 KHz, in two and 1/2 second bursts, then repeating. The output from the driver is definitely functioning, changing frequencies as it should. (I can see the proper frequency output on a Scope) The thing is - I can hear something at all of the different frequencies, which seems to tell me that the tweeter is not oscillating at the driven frequency.. with these old ears I would doubt that I would hear even 15KHz.

The speaker is connected as shown, using 9 volts (not 12 as shown) across the driver. The PIC outputs (which are out of phase to each other) are connected to the 1K resistors feeding the transistor bases. The PIC shares a common ground with the driver.

The Tweeter has the specs shown below:

*Similar to KSN1016 *Power handling: 50 watts RMS/75 watts max *Frequency response: 3,500-27,000 Hz *SPL: 94 dB *Dimensions: A: 5-3/4", B: 2-3/4", C: 2".

I realize the top frequency response shows 27 Khz, but the lower ones should be OK.

Should I be applying a higher voltage to the tweeter so as to drive it harder? Or is there something simply stupid with my approach?

Thanks

Ken
 

Attachments

Hero999

Banned
That will work.

You don't need any base resistors though as they're emitter followers.

Obviously you'll need more buffering and level shifting to drive this from a microcontroller.
 

Sceadwian

Banned
ozark you're probably just hearing harmonics. I can almost guarantee you're going to need more than 9 volts to get full power out of that piezo though, check the voltage rating on the data sheet, or find out what it's impedence at resonance is. Even tiny low power piezo's can handle / and or need 20+ volts to produce full volume. Try to keep the bursts short too, 50 watts at 20+khz can damage hearing, even if you can't hear it.
 

Sceadwian

Banned
hero 9 volts is going to give you 50 watts out of a piezo tweeter??
 

audioguru

Well-Known Member
Most Helpful Member
The piezo tweeter makes a horrible high frequency speaker. Its frequency response is full of peaks (resonances) and notches and it has such high distortion that the harmonics beat and create lower frequencies (intermodulation distortion).

A 50W into 8 ohms amplifier has an RMS voltage of 20V. The transistor circuit will provide it when its supply is about 11V and the signal is a square-wave.
The piezo tweeter has an average output of 94dB at 1W/1m so at 8m to 10m distance it will not be loud.

Dogs will ignore the sounds.
 

ozarkshermit

New Member
Thanks for the info:

Actually, I will be using more than 9 volts to drive the tweeter. For now I just bread-boarded the circuit using some wimpy TO-92 type transistors, and they can't handle what I would eventually use.

I have been thinking about using a LM1875 or the like for an amp, with perhaps a series R-L input to the tweeter. I am totally at a loss to determine the input voltage to drive the LM1875 however, and since I would be driving it from a PIC (5 volts input to the amp), I think that might be too high.

Any suggestions?

I can cope with the digital stuff, but am lost when it comes to the linear part.

Ken
 

ozarkshermit

New Member
Thanks Nigel

That's what I was thinking , perhaps two resistors for a voltage divider, and maybe add a pot to "fine adjust" the input level.

Ken
 

audioguru

Well-Known Member
Most Helpful Member
The piezo tweeter uses hardly any power because its current is very low. Therefore an amplifier is not needed, the four little power transistors will be fine.

The dogs will not care. As soon as they hear the sound and nothing bad happens then the next time they will ignore it.
 

HarveyH42

Banned
Everybody enjoys the building, it's a fun learning experience. But Audioguru is right, the dog isn't going to respond in a predictable manner, unless trained to do so. Wouldn't depend on something like this, with a dog that is known to attack, most likely you'll only provoke it.

I've been having some fun with my Lab puppy past few months. When I get bubble wrap or air filled bags as packaging, I pop them. He knows that good things also come in plastic wrapping. He runs from the pops and bangs, but returns because he hears me with the plastic. He doesn't run if he smells food, even if I pop some bubble wrap.

Most dogs raised around people, will respect you, unless you come off as weak and afraid, or trained to attack. Some dogs are messed up in the head (just like some people), from abuse, confined in a small filth space (cage/short chain). When they break free, look out, because they are insane. Even owner will have no control.
 

Sceadwian

Banned
Deer, bugs, rats, dogs. A post like this seems to crop up every few weeks.
 

Sceadwian

Banned
A buzzer isn't particularly effective at warding off animals =) Who's dog is it? Report the owner.
 

ozarkshermit

New Member
Thanks

Well, I guess I should just listen to the personal experiences of all the posters here who have attempted this, and found out that dogs are not bothered by ultrasonic frequencies after all.

Ken
 

Sceadwian

Banned
Oh it bothers them, but if it doesn't cause them pain or they don't have a negative reason to avoid that area they'll ignore it after a while. Might work for a single dog on a lawn like that he'll go next door, but I'm more inclined to do what some other posters have which is use a motion sensor to turn on a sprinkler head. I don't want to put you off your project ozark, but there are better ways.
 
Last edited:

unclejed613

Well-Known Member
Most Helpful Member
i've built these devices before. it will definitely give a human a headache. the piezo "super-tweeters" marketed by radio shack back in the 80's were rated at 20Vrms max. piezo elements do have a nasty habit of intermodding against their own resonances, so i f you feed in 20khz, and one of the device's resonant peaks is at 15khz, you get a 5khz intermod product out. it gets worse as you drive it harder, because there's enough vibration in the crystal to set up resonant vibrations even though you are driving it at another frequency. you can actually see the voltages from these intermod products at the terminals. you can further identify them by changing the back pressure of the air by plugging up the horn port and see the intermod products change frequency.


watch your drive voltage on this device, DC included. too much voltage will damage the ceramic and at the very least crack it, but in pro-audio repair, i have actually seen them burnt by internal arcing.
 
Last edited:

audioguru

Well-Known Member
Most Helpful Member
15kHz, 17khz and 21kHz are audio frequencies that are clearly and loudly heard by kids and teenagers. You might make them upset at your "high frequency noises".
 

ozarkshermit

New Member
I live so far out in the country that I see kids about twice a year - that will be no problem - -

Maybe it will make me a little more looney though . . .(if that's possible)

Ken
 

dougy83

Well-Known Member
The circuit you posted is an emitter-follower, and even though you have a supply of 12-15v, the voltage across the piezo won't exceed the 5v you are powering your microcontroller from. As someone said earlier, you'll need to translate from the microcontroller output voltage to the driver supply voltage.

An example circuit attached is shown that will give almost the full supply voltage swing to the piezo transducer.
 

Attachments

audioguru

Well-Known Member
Most Helpful Member
The circuit you posted is an emitter-follower, and even though you have a supply of 12-15v, the voltage across the piezo won't exceed the 5v you are powering your microcontroller from. As someone said earlier, you'll need to translate from the microcontroller output voltage to the driver supply voltage.

An example circuit attached is shown that will give almost the full supply voltage swing to the piezo transducer.
Good point!
I forgot that the 4.5V inputs (with the loading of the transistors) form the micro-controller causes only about 7.4V p-p across the speaker which is only 2.6V RMS and the power into an 8 ohm speaker is only about 0.9W.
The dog would need to wear the tweeter on his ears to hear it.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top