Maker Pro
Maker Pro

Working on getting a Pierce Oscillator to Oscillate

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Hello,

I am fairly new to electronics, and I am trying to build a very simple crystal oscillator circuit, to use with some radio experimentation that I intend to do.

The simplest crystal oscillator seems to be the Pierce oscillator, and that is what I am trying to build. I've tried many different variations on the standard circuits, but I can't seem to get any of them to work.

The main circuit that I am trying to build is this type:
https://www.electronicspoint.com/crystal-oscillator-circuit-t240691.html

But as I said, I have tried several different variations, with different values.

The other problem is that I don't have an oscilloscope, I have hacked a PC sound card into an oscilloscope using the microphone port. It seems to be working ok, but I don't know how accurate it it, or if it is capable of measuring the 1MHZ crystal that I am using.

I am also trying to build the circuit on a breadboard, and I have read that is not a good idea for crystals, particulaly higher frequency ones, since they contain stray capacitance, could that be a problem?

I have built the circuit above using the following values:
Inductor: 3.3mH
Capacitor: 0.001uf, (1nF)
Crystal: 1MHZ
Resistor: 10M
Transistor: 2N5485 JFET

I am also using 2AA batteries, so 3V. Is there anything obviously wrong here?

Thanks for any help!


What I would like to know is if my circuit *should* be working with the component values that I have chosen, and if the problem might be hacked scope.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
your PC scope is useless at RF frequencies. How do you know it is not oscillating?

I am not an expert on RF circuits, but your inductor seems way too high.

As for breadboardiing an RF oscillator, I have made 100Mhz oscillators work on a breadboard.

bob
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
That's what I was thinking too. I read somewhere that a PC scope can only handle about 44 Khz or so?? It may well be working, I just didn't want to drop a fair amount of cash on a piece of equipment unless I knew I needed it.
When I turn on the power to the circuit, the pc scope does immediatly begin showing a regular wave on the screen, very different from what it shows when the power is off. It's not at all a sine wave, but it is definitely a wave, so I thought that maybe it was my component values that were off?

I don't have any other inductors here to test with, but I could buy a few, they are cheap, thanks for the advice!

It's good to know it's at least possible to breadboard a 1Mhz crystal, I don't want to solder anything until I know it's at least functional.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
The scope is probably aliasing, which means it is showing a wavefofm that is a complex interaction between the sampling frequency / phase and that of the input. which means maybe it is oscillating. Do you have a multimeter with a frequency function. That would tell you something.

Bob
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
At 1Mhz, there should be a healthy carrier on a medium wave radio at 1MHz or 300m.

3V supply may be a bit low, you could try 6V.
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Unfortunately my multimeter doesn't seem to have a frequency function, just current volts and resistance.

I tried several times with a small radio, putting the aerial right up to the circuit and slowly moving through the frequences, but couldn't hear anything but normal stations and static, that's what I tried before I hacked the PC scope. I suspect it isn't oscillating, since it would produce radio waves at 1Mhz if it was.

I wil try 6V then and see if that helps.
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
I tried with 6V, and there was no sine wave, but I still get my strong, regular wave. I took a screenshot of it, if it helps:

viewfile.php
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
One option is to get a 32 KHz crystal that (if it oscillates) will generate a signal you can see on your PC scope (although you may not be able to see the real shape of the waveform.

As for some of the comments above, the inductor has to be a high value to prevent the low impedance from the power supply acting to damp the oscillations.

I believe this circuit starts oscillating by amplifying noise (thermal or other). The crystal acts as a filter to increase the gain at it's resonant frequency.

I should build one just to see how well it works.
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Steve, I think I might do that, just to test the circuit. Would the values of the components change if I get a crystal at a different frequency?

If the 3.3 mH inductor is the correct value, it seems my hacked scope is the issue then... I think I will just go and buy a simple one, I'm sure it's the kind of thing I would use a lot.

Thanks for your help!
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
Yeah, I'd try the same values. If I get some time today I'll build one with whatever I can find and report back to you :)

edit: it may pay to increase the value of the capacitor for a really low frequency crystal.
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
That's vey kind of you, thank you! I ordered a few 32 khz crystals from eBay, so I'l try those once they get here. I have a few other caps I can try as well.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Unfortunately my multimeter doesn't seem to have a frequency function, just current volts and resistance.

I tried several times with a small radio, putting the aerial right up to the circuit and slowly moving through the frequences, but couldn't hear anything but normal stations and static, that's what I tried before I hacked the PC scope. I suspect it isn't oscillating, since it would produce radio waves at 1Mhz if it was.

I wil try 6V then and see if that helps.
What you would hear if you tuned the frequency it is oscillating at is the radio going very quiet. It will be near 1000 on the AM dial.

Bob
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I tried to build one of these last night using some components I had on hand.

It would not oscillate with either the 10MHz or 23kHz crystals I happened to have on hand.

There are so many things it could have been (I didn't have an RF choke anywhere near 2mH so I had to use an inductor not rated for RF, or the construction, or the random jfet I chose).

I also looked up a few references which indicated that this design is not overly suited to low frequency crystals. I don't have those on hand any more though...
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Hello,

I tried the radio again, and I got nothing at all, the sound didn't change anywhere. I also tried removing the crystal, and the waveform I was getting didn't change, which indicated that the crystal is not doing anything.

Steve, 10Mhz is quite similiar to what I am using, so I think it may be that the circuit just isn't suitable?
I've tried so many others, but didn't get any oscillations either... Is there some other simple configuration you might know of that I could try? I've got a bunch of capacators and resistors of different values as well as some bipolar transistors I could use, and some chips with nand gates that can be wired as inverters, right?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
This may be useful. Note that a circuit given DOES NOT oscillate at 32KHz.

They do, however, recommend the CMOS inverter circuit for the 32kHz crystal.

Page 67 of this book also has useful information.
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Thank you very much, I had a good look at those links. I will try the CMOS inverter one when I get the 32khz xtals I ordered.

I found this circuit:
http://www.hqew.net/files/Images/Ar.../pierce-quartz-crystal-oscillator-circuit.gif

For a 2Mhz crystal, which I will try tomorrow, since I already have most of the components. The style of transistor is different from the ones I have, but they are both JFET's, so they should be compatible, right?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Here's Pierce Osc. theory. Note the value of L is much lower than what's been discussed here. It doesn't appear to be a choke or function as one either. I like to think of RF Choke function as the inverse of a bypass cap. Neither are an element related to resonance or impedance matching.

http://pr.erau.edu/~lyallj/ee412/pierce_exp.html

Chris
 

Zerosum

Dec 31, 2012
15
Joined
Dec 31, 2012
Messages
15
Thanks a lot CDRIVE, I will build that one later, and see how well it works. I am convinced I am doing something obviously wrong, it can't be this hard to build an oscillator, can it? They are relatively simple circuits...
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I did some research on this oscillator and I'm surprised that cdrive's link gives that explanation for the inductor.

Many Pierce oscillators do not actually have an inductor or a tuned circuit other than the crystal (I'll refer to the basic inverter oscillator which has a crystal and a coupe of capacitors).

I was somewhat surprised at the *huge* value for the choke, but it was replicates in several independent publications (including, I believe, an application note from one of the major semiconductor manufacturers of old).

As I said, I'm surprised, but I'm willing to defer to that explanation.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Steve, since I posted that link I've also found similar information that you have, including very large inductors that make no sense to be anything other than a choke. I've tried to spice the Pierce to the point of my head exploding. I'm also questioning the information in my link. I think I'll stick with Colpitts.

Chris
 
Top