40KHz Oscillator Doesn't

On a sunny day (Thu, 23 Nov 2006 09:39:04 +0000) it happened Phil Cross wrote in :

Those resonators are mostly capacitive, so you would get at most 90 degrees, and a huge attenuation with 4M7. I have used these resonators in a LC circuit setup. You can also drive these from a low impedance power amp that makes it to that frequency.

You probably need to make the op amp circuit more unstable, so that it almost oscillates on its own at the frequency you want, as said in other replies you cant assume the transducer behaves in the way you want, but it most likly has a a dip in its impedance at resonance, my gues is it doesnt make for a very good inductor or capacitor at resonance that you get from a ceramic or quartz device. so putting it in the negative feedback path would have the wrong effect, you need positive feedback.

A possible circuit might be to have positive and negative feedback with the transducer in the lower part of the negative feedback divider, with the negative feedback canceling the positive feedback the circuit is stable untill the transducer robs the circuit of negative feedback at resonance therefore leaving more positive feedback.

Colin =^.^=

That site looks to be a useful primer, and the author uses LTspice for simulation.

--
Tony Williams.

Here's an application note with lots of info and several example circuits.

-- Joe Legris

colin a écrit :

Well, how do you expect to have a different loop gain between both circuits? There's no low side impedance on the feedback network/divider.

Hint: opamps have (almost) infinite input impedance and your transducer has a finite one.

--
Thanks,
Fred.

There's your problem. The op-amp has a very high input impedance, and the resonator has a low impedance and lots of stray capacitance. Try putting a 47K resistor from input to ground so the resonator works into a more reasonable impedance.

Also you may not be getting the phase-shift you expect. You may have to a dd a RC phase shift network to get the phase right.

47K is probably still way too much. To be effective it has to be of the transducer impedance order of magnitude, which surely is much lower than 47K.

--
Thanks,
Fred.

hbt feed back to the +

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5

Thank you for all the responses. I am now going to put a wet towel around my head, study them and the links in them, and see what I come up with.

Thanks again, Phil.

Tell me - does the wet towel work as good as foil? ;-)

(anybody else see "Total Recall"?) ;-)

Cheers! RIch