1.It's said that the source for oscillations is a noise signal,generated in resistors.Can anyone please explain me ,how it occurs?
2.In a oscillator what actually do we do,i mean can anyone please explain me the working of an oscillator(say RC phase shift oscillator)?
Heyy frndz please help me out.

 

1. "Johnson–Nyquist noise... is the electronic noise generated by the thermal agitation of the charge carriers (usually the electrons) inside an electrical conductor..."

So basically it's just the sound of electrons moving around, and it's a physical property of electrical conductors. In an oscillator the amplifier amplifies the naturally-occuring noise of the resistors.

2. A sine wave oscillator requires an amplifier with positive feedback at a particular frequency. Noise is broadband, so the RC network is used to isolate a frequency for feeback.

Notice the feedback path goes to the inverting input, which is normally negative feedback. We want positive feedback. The RC network introduces frequency-dependent phase shift. Since the feedback path is going to the inverting input, the frequency of interest needs to be shifted by 180 degrees in order to result in positive feedback. The phase shift of the RC network varies by frequency, so there will be only one frequency where the exact amount of required phase shift occurs.

So even though you've got a broadband noise signal, positive feedback occurs at only one frequency, and you get oscillation.

I find Wien bridge oscillators better. Instead of phase shift they use a bandpass filter to isolate a frequency for feedback.

 

Thank you so much,that was a wonderful reply.
But it again brings some doubts to mind that of course for a particular frequency oscillations do occur because at that frequency only the phase shift is 180 degrees in the feedback network for RC phase shift oscillator ,for which the total phase shift becomes 360 degrees and hence satisfying barkhausen's criteria.
My question to you is that what happens to the frequencies other than the frequency of oscillation....?
and does the noise signal have a fixed frequency?
AND why is noise broadband in nature????

 

My question to you is that what happens to the frequencies other than the frequency of oscillation....?
and does the noise signal have a fixed frequency?
AND why is noise broadband in nature????

The other frequencies are also shifted by the RC network, just at different amounts, such that they are not at the proper phase to result in positive feedback.

Noise doesn't have a fixed frequency, it has all frequencies (theoretically).

Why is it broadband? I have no idea, that's beyond me. I'm sure there's a nice scientific explanation of the physics behind it, I just don't know what it is.

 

thanks again...
noise is continuously generated in conductors so when that amount of noise is again fed back positively then the amplitude of output must rise every cycle,how cum it stays constant??

 

http://www.electronics-tutorials.ws/oscillator/hartley.html

please help me through this page......

it talks about tank circuits ..how are these tank circuits different from an RC circuit as in case of wein bridge?
initially it sayz that we need continuous feedback to meet the losses in the tank circuit which produce the oscillations,but den it sayz that the source of oscillations are noise signals and that these tank circuits are just frequency selecting networks...
can anyone clarify......please

 

The LC 'tank' circuit accomplishes the same thing as the RC circuit in the Wien Bridge - frequency selecting - just in a different way.

LC tank circuits use the property of inductor-capacitor resonance to isolate a frequency.

RC bandpass filters use cascaded low-pass and high-pass RC filters to isolate a frequency.

 

initially it sayz that we need continuous feedback to meet the losses in the tank circuit which produce the oscillations,but den it sayz that the source of oscillations are noise signals and that these tank circuits are just frequency selecting networks...
can anyone clarify......please

I think that means that if you give an LC resonant circuit a momentary noise signal, it will oscillate briefly but decay. The amplifier provides the "continuous feedback" needed to sustain the oscillations. The source of the oscillation is noise, from which the tank isolates one frequency for oscillation.

 

incase of an hartley oscillator 180 degree phase shift is provided by the amplifier and further 180 by the LC circuit,please explain how the Lc ircuit provides 180 degree phase shift....

 

No, the Hartley oscillator doesn't work based on phase shift, it works on inductor-capacitor resonance which is reinforced and sustained by an amp. I provided a link that explains it. Also, the link you posted has a page just before it that explains resonance.

 

thanks again...
noise is continuously generated in conductors so when that amount of noise is again fed back positively then the amplitude of output must rise every cycle,how cum it stays constant??

see only the noise corrsponding to the frequency of oscillation gets a positive feedback which also means a gain of 1 for the others it doesnt give a gain of 1 that is either the other frequencies have going down effect on the gain,, or a going up effect on the gain which is followed again by a going down effect.