Maker Pro
Maker Pro

Crystal Oscillator Circuit.

cjdelphi

Oct 26, 2011
1,166
Joined
Oct 26, 2011
Messages
1,166
Yeah I can follow the instructions on a circuit and build one no worries, but i'm still a little confused on the theory.

the LC tank circuit, simply an inductor (to store a charge) and a capacitor, am I right in assuming, to get the oscillation going, the capacitor and inductor simply throw back and forth the electrons, like a tank of water sloshing back and forth... right?.

how's a crystal come into the equation? random noise is produced and fed to the crystal and some random noise hopefully matches the frequency of the crystal?

This is a simple JFET pierce crystal oscillator. We can use a wide frequency range of crystal using this circuit without circuit modification. Here is the circuit :

JFET-Pierce-Crystal-Oscillator.gif


This circuit uses Q, 2N3823, that is maintained thus insuring good stability of frequency because the crystal does not loaded by the JFET. [Source: National Semiconductor Application Note]

Page Source: http://www.simplecircuitdiagram.com/2010/01/25/jfet-pierce-crystal-oscillator/


ok, so fine, random noise from the Capacitor and inductor is produced, but how exactly does it go from a randomish oscillation to a stable oscillation with the Crystal in the circuit?

in short, how the hell does this circuit actually work :) - Thanks...
 

davenn

Moderator
Sep 5, 2009
14,081
Joined
Sep 5, 2009
Messages
14,081
Hi Olivier

thanks for posting that link .... as a radio tech I'm ashamed to say i could not explain the operation of a Pierce osc. mainly probably cuz most osc's I use in radio work are colpitts osc's.

That link still didnt totally explain to me how the circuit works. I guess what I would like to see is current flow through the cct for each 1/2 cycle of the oscillation.
I want to see how the osc is occurring.

maybe Res or Steve can shed some light on that :) hey lads ??? haha

my real confusion lies in my lack of real understanding of the operation of a FET compared to a bipolar transistor

cheers
Dave
 
Last edited:

cjdelphi

Oct 26, 2011
1,166
Joined
Oct 26, 2011
Messages
1,166
lol, exactly Daven, I've read as much info as i can including colpitt oscillators , which is why I searched high and low for the simplest crystal oscillator circuit i can find, and the simplest one is the one (I believe anyway) i posted..

Now, what i've read is, random noise is generated from the inductor and capacitor, a simple oscillator is the driving force behind it, now if the frequency of the crystal happens to match the frequency of the oscillation the circuit produces....

the signal which comes back from the crystal, gets amplified by the transistor, and then fed back into the crystal.... and so on..

but I don't understand how and why the circuit tunes itself on the crystal, why would the inductor and capacitor stop oscillating the way IT wishes to and starts to oscillate at the frequency of the crystal?
 

davenn

Moderator
Sep 5, 2009
14,081
Joined
Sep 5, 2009
Messages
14,081
but I don't understand how and why the circuit tunes itself on the crystal, why would the inductor and capacitor stop oscillating the way IT wishes to and starts to oscillate at the frequency of the crystal?

Thats really the function of the crystal. It alone is the primary reason for that circuit to oscillate. And that oscillation will occur at the resonant freq of the crystal.
That is the norm for any osc cct that is crystal controlled. Its why they produce excellent stability with low phase noise.
I dont know if you can replace that crystal in a Pierce osc with an L/C tank cct and still get a free running osc as you can in a Colpitts osc cct

BTW welcome to the forums :)
good to have you aboard

cheers
Dave
 

junas

Nov 4, 2011
3
Joined
Nov 4, 2011
Messages
3
Hi,
cjdelphi the circuit does not tune itself on the crystal,u design the circuit i.e you find appropriate values of resistors and capacitors using jfet dc biasing conditions and moreover these values that u determine have to satisfy the condition for oscillations.
You would know the natural frequency of the oscillator when u buy it,so
f=1/(2*pi*sqrt(LC))
where C=C*Cm/(C+Cm)
Cm is mounting capacitance which will be told to u when u buy da crystal and 'C' is the capacitance u use.
 

foTONICS

Sep 30, 2011
332
Joined
Sep 30, 2011
Messages
332
as you go higher in frequency LC tuned circuits started acting funny. you can operate a crystal at much higher frequencies with a far superior stability. and above that you want overtone crystals (just a different form of crystal) which operates on a harmonic of the frequency you want but that is not so important. just that crystals are far more stable. more expensive though
 

foTONICS

Sep 30, 2011
332
Joined
Sep 30, 2011
Messages
332
my real confusion lies in my lack of real understanding of the operation of a FET compared to a bipolar transistor

I believe it has something to do with the high input resistance of the FET so as to not load down the crystal which can cause it to drift. i think drift is right, im still a student so be lenient.
 

davenn

Moderator
Sep 5, 2009
14,081
Joined
Sep 5, 2009
Messages
14,081
I believe it has something to do with the high input resistance of the FET so as to not load down the crystal which can cause it to drift. i think drift is right, im still a student so be lenient.

Hi foTONICS

I was hoping some one way back then was going to give me a quick lesson in FETs lol

Now what I do understand is that the hi impedance would not so much cause drift but does not "load" the crystal down. Loading the circuit would literally drag the crystal freq .. usually down. Drift implies that the freq would be moving around.

Dave
 
Top