# Pierce oscillator

Discussion in 'General Electronics Chat' started by bennyk@37.com, Dec 7, 2013.

1. ### bennyk@37.com Thread Starter New Member

Feb 27, 2013
6
0
hello

could anyone explain to me how the pierce oscillator works. i have been working on a simulation on LT spice and am wondering if anyone could explain how it works. the LTspice simulation includes a transformer and transistor

i can show a picture if necessary

2. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
Pierce doesn't need a transformer. Just a couple of capacitors and a crystal...

Even better, post the .asc

3. ### LvW Active Member

Jun 13, 2013
674
100
At first, there are many different versions for a PIERCE oscillator (e.g. with/without transformer). In the following a give you a very general explanation on the operating principle of all PIERCE alternatives:

As you probably know, a circuit oscillates if there is one frequency for which the loop gain is (slightly) larger than unity with a phase shift of 360 deg (0 deg). This can be achieved using different circuit topologies.
The basis of the PIERCE principle is as follows:

* As an active element we use an inverting voltage-controlled voltage source (high-resistive output in reality): BJT or FET or CMOS inverter.
* In addition we need a passive feedback network that introduces additional 180 deg phase shift into the feedback loop. So we have 360 deg in total (because we have an inverting active source).
* Exploiting the internal output resistance of the active device an external lowpass PI-topology is used consisting of two grounded capacitors with a series inductance L in between (sometimes replaced by a crystal).
* This 3rd-order network produces the required additional 180 deg phase shift.

Last edited: Dec 8, 2013
4. ### bennyk@37.com Thread Starter New Member

Feb 27, 2013
6
0
here is the circuit. I am not sure what the transformer and transistor are for? could someone explain the operation of this circuit please?

thanks

File size:
15.6 KB
Views:
155

Oct 2, 2009
5,451
1,071
6. ### bennyk@37.com Thread Starter New Member

Feb 27, 2013
6
0
thanks for the reply, so if everything to the right of the transformer was removed then the circuit should act as an oscillator no problem? why is the modular needed do you think, what is its purpose

7. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
Did you click on the Wikipedia links I provided? They explain both the oscillator and the modulator.

8. ### bennyk@37.com Thread Starter New Member

Feb 27, 2013
6
0
i was under the impression that a mixer was used to multiply two frequencies. this seems just to have the output of the oscillator going into a rectifier? i cant see where the multiplication occurs?

also would i be right in saying the transistor is there to amplify the signal from the crystal?

9. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
V1 is other input to the modulator.

No, any oscillator needs an amplifier, so JI is the heart of the oscillator. Look up Barkhausen Criterion.

10. ### bennyk@37.com Thread Starter New Member

Feb 27, 2013
6
0
i thought that V1 is powering the crystal oscillator? so J1 isn't an amplifier?

the basic circuit of the pierce oscillator given above shows only a crystal, two capacitors and an inverter? this circuit is different and so im not sure how to understand its operation

11. ### LvW Active Member

Jun 13, 2013
674
100
In some circuits the crystal is replaced by an inductor, see my post#3.
Remember: The crystal can also be used as a large-Q inductor.
I both cases, the feedback loop is closed through a 3-rd order classical passive lowpass (including the transistor output resistor) causing a phase shift of -180deg.

Last edited: Dec 10, 2013
12. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
Let me try it again.

Any oscillator (be it Pierce, Colpitts, Hartley, or whatever) needs an amplifier at its core. The amplifier can be a tube, transistor, fet, jfet, opamp or logic inverter. The schematic Benny posted happens to use a JFet (J1) for its amplifier.

In looking at the circuit again, V1 is the dc supply voltage. The three-winding transformer is configured as center-tapped secondary winding which drives the two diodes used as a full-wave bridge rectifier to make a DC output, possibly acting as an RF voltmeter (where the inductor labelled OUT is the coil of the voltmeter)?

The transformer primary would have a very strange current waveform, consisting of short pulses (one per cycle of the oscillator. The current would be somewhat effected by the Q of the crystal. Could this thing be some sort of crystal tester? Where did the circuit come from? It is drawn using the LTSpice graphics editor.

13. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
I took Benny's circuit, and put it into LTSpice, and played with the component values to make it work. I chose the motional parameters for the crystal as though it were a 4MHz crystal. Here are the steady-state waveforms, and a separate sim run showing start up. I am also posting the LTSpice .asc file if someone wants to play with it...

File size:
229 KB
Views:
84
File size:
120.7 KB
Views:
71
File size:
1.4 KB
Views:
60
14. ### alfacliff Well-Known Member

Dec 13, 2013
2,449
429
actually the two diodes are a frequency multiplyer times two multiplyer. if it were mixing, there would be another signal coming in. a voltage doubler type frequency multiplyer.
cliff