Greetings!

I want to power a piezo at a precise frequency, without using the microcontroller.

I thought that a ceramic resonator would be perfect, as it's already designed to resonate at a given frequency. Yet, when I try to connect my piezo to the output pin of the resonator.... nothing happens. My multimeter shows 0V, and even with a led instead of a piezo, nothing happens.

Any ideas why?

Thanks in advance,

Christian

Hi Christian,

A ceramic resonator still needs an external circuit to actually form an oscillator. It's called a resonator, but think of it like a tuning fork, where you can strike it and it will vibrate at a fairly predictable frequency, but you still need some external energy input to drive it, adding energy to keep it going. The circuit that's needed is basically an inverter between the two sides of the resonator. (However, a joke among electrical engineers is that whenever you want a circuit to oscillate, it won't, and when you want it to be stable, it will oscillate...)

Another concern here is that the frequency of many ceramic resonator is in the several-MHz range, at which point you wouldn't hear anything at all coming from the piezo! Is your ceramic resonator in the audible range of frequencies?

Mike

Hello Mike,

Thank you for your well explained answer.

No, the ceramic resonator isn't in the audible range. That's why I tried to see what was happening by using my multimeter or a LED.

I actually tried to built an audible resonator with an op-amp (LM358) to see if everything was alright. It worked, but the LM358 doesn't seem to like high frequencies.

I will get an inverter to use with the ceramic resonator and post my results here.

Thank you again for the support!

Christian

I did get an inverter DM74LS240N and it seems to be working, although I have experienced some weird "reset" of the microcontroller when my oscillator circuit is plugged-in.

The next step planned to check if the oscillator is working fine is to use a flip flop counter CD74HC4020E and use it to reduce the frequency at a more easily readable one, perhaps in the single digit khz, at which point it should be audible.

I have yet to succeed the last step, but I'm working on it.

Christian

Ok, so I've got some results...

I used 2 flip flop counters, so it should reduce the frequency 268 435 456 times (2^28). If my math is correct, for a 2MHz signal, I should obtain a switch in the final output at every 134s.

The problem is that I get a switch at each 2s (approx.) Worse: I obtain the same time with a 7.370MHz resonator. Not only does my signal seem WAY faster than predicted, but it seems to always be the same.

Could someone explain me why?

Thanks in advance!

Christian

Hi Christian,

Sounds like the oscillator is actually running much faster than you intended -- perhaps it isn't really oscillating around the ceramic resonator's frequency, but just around the propogation delay of the inverter itself. (Your 2s implies about 134MHz at the oscillator! But this is much faster than the other parts of your circuit are supposed to work, so who knows.) Just as a test: if you add some capacitance to ground at the input of your inverter, does the 2s slow down? For example, just put a wire in that row of your breadboard, and hold the insulation with your hand.

Let's take a look at this link: Ceramic Oscillator Principles. There's a lot of math to getting the phase offset right so that you get an oscillation where you want it, but basically take a look at Figure 2.2 and Figure 3.2 with the table -- it suggests some resistor and capacitor values to try. Can you give that a shot and let us know what happens?

Mike

Hello Mike,

Once again, thank you for the answer.

Here is the datasheet for the ceramic resonator, if it can help. The recommended resistance from both the datasheet and your link is 1M ohm, which appears to kill the resonance. When I put a single wire, or a low resistance of about 100 ohm, it oscillates, but with the 1M ohm, nothing happens.

As you suggested, I tried to change the capacitance to ground, and yes, a simple wire changes the frequency. In fact, just by touching the input wire it changes. In both case it accelerates, which leave me to think that the very montage itself has too much capacitance and ruins everything.

Hi Frozenlock,

I'm now suspicious about the inverter. Yours is a "LS" series, which does not use MOSFETs, but rather uses BJTs internally, which means there is a non-zero current requirement at the input. (Low-power TTL.) If you want to stick with that inverter, take a look at Figure 5.1 of the "Ceramic Resonator Principles" link I posted earlier. That shows that with TTL-based inverters, the requirements for the resistor and capacitor are dramatically different.

The easiest thing to try would be to get a CMOS inverter instead of the TTL ones.

Mike

It's working now, thanks to your advice! (It was the inverter)

Hi Frozenlock,

Excellent... fun stuff!

Mike