RLC oscillator?

Dogenigt · Posted: Mon May 23, 2011 8:07 am Post subject: RLC oscillator?

Hi everyone,

I've been reading up on RLC oscillators, which basically consists of an inductor, which is like an electric coil, a capacitor and a resistor.
http://en.wikipedia.org/wiki/RLC_circuit

What I want, is to make an oscillator with a variable pitch. The nice thing about this specific circuit, is that no IC's are needed and it's so darn simple. And the best thing: it can't produce sine waves, instead of this annoying square wave, every IC produces!

The wikipedia says the following about variable tuned circuits:
"Adjustable tuning is commonly achieved with a parallel plate variable capacitor which allows the value of C to be changed and tune to stations on different frequencies."

But I have no good idea about how to make this into a schematic...
I want to connect a female jack to this, and this parallel plate variable capacitor to change to pitch. Anyone have experience or proper understanding in this field?
Help is so appreciated!

Best regards,
Paleorama

Dogenigt · Posted: Mon May 23, 2011 8:18 am Post subject:

Here's a good article on building OSC's:
http://how-to.wikia.com/wiki/How_to_build_an_oscillator_circuit

I am still a bit confused, for the simple reason that I am still pretty green in the field.

blue hell · Posted: Mon May 23, 2011 9:30 am Post subject:

LRC is not very suitable for audio range oscillators as you'd need large L and C values.

A tunable C is always small valued, like max 500 pF, and a 1 mH coil is still small, and yet no real fun to make anymore.

With such values you'd get an LC product of 5 * 10^-13, the square root of that is 707 * 10^-9, and the inverse of that then is 1.4 * 10^6 which divided by 2 * PI yields a frequency of about 225 kHz. (and that would be the minimum frequency value obtainable with the component values chosen).

It also will need some active component(s), just like any other oscillator, to make it actually keep oscillating (as in the Colpitts/Hatrley/Clapp examples).

A workable option would be to build two of such oscillators, one at a fixed frequency and one on a variable frequency. When you then mix the signals in a multiplier one of the output signals would be the difference of the two frequencies - and that one could very well be audible. (this idea is used in a Theremin).

Dogenigt · Posted: Mon May 23, 2011 7:27 pm Post subject:

billfusion · Joined: May 19, 2011 Posts: 10 Location: San Diego

Want to try fun pats find an old style car radio (pre diital) they use a vari inductor bank for tuning
or try a VXO a variable xtal osicaltor

JovianPyx · Posted: Tue May 24, 2011 9:30 pm Post subject:

That looks like a Hartley oscillator. Those output signals _probably_ would be referenced to the negative side of the "battery" or power supply. I remember as a kid making one of those using a center tapped audio output transformer for the inductor, the center tapped side was used as shown in the schematic and the other winding of the transformer (secondary) was connected directly to a small speaker. This was called a "code practice oscillator" that was commonly used in those days with a morse code key to practice sending morse code. Fun stuff, but as Jan said, RLC circuits are really not very useful in audio. There are other circuits that you can build as oscillators for producing sinewaves, many resonant filters (like state variable filters) can be adjusted to "self oscillate" and when they do, they make nice sinewaves. These circuits use (usually) OTAs and can be voltage controlled to allow playing music with them. Many vintage synths have the ability to use filters to generate sinewaves (as well as being used in the "standard" way a filter is used).

Dogenigt · Posted: Wed May 25, 2011 3:19 pm Post subject:

Thanks for your help and response!
Would it be possible to sort of trigger the oscillation in an inductor with a square-wave oscillator, like with a 555, to generate a sine wave signal in the chosen input frequency?

Got inspired by this:

JovianPyx · Posted: Wed May 25, 2011 3:48 pm Post subject:

Not sure how practical it would be (and hopefully I'm answering the question you asked), but yes - if you have an LC circuit (which is called a tank circuit) where the capacitor is in parallel with the inductor, you can energize the LC circuit with an impulse from something like a 555 timer and maybe a MOSFET switch so that the energizing system doesn't load the LC circuit. Once energized, the LC circuit will resonate for a period of time (this is where I'm not sure how practical). The load resistance would determine how long the circuit will ring. Load resistance can be reduced by using a preamp like a JFET or MOSFET to "listen" to the LC circuit and pass it to a mixer. You could then add a resistor like a pot in parallel with the LC to control how quickly the waveform fades to zero.

The waveform will be a damped sine and would mimic a synth circuit of a sinewave oscillator fed into a VCA controlled by an AR envelope generator that has an exponential decay. It would have a rather natural decay and natural sound. I've emulated this in digital hardware with a high Q state variable filter in bandpass mode and energizing the filter with an impulse. It sounds nice IMO.

The problem musically with the LC circuit is that it will be fixed tuned. However, if you had a bank of them, you could get different tones with different size inductors and capacitors. It would be a challenge to tune it so that it's in tune with a piano (for example). You would have to "cut and try" different capactors a lot to get a bank of them to play a musically tuned scale - but I'm not sure you even want that. The digital system I designed to do this is quite well tuned and will play melodies in tune with other instruments.