Crystal Controlled Oscillator Circuit

[waggy55]

Hello,

I would really appreciate any help on this circuit that I've tried to design (I've attached a copy).

I have biased the +ve input to half the rail voltage, and the output and consequently the -ve input are both also biased to half the rail voltage.

I was hoping the the Xtal would cause voltage fluctuations at the -ve input around half the rail voltage, which would then cause the output to fluctuate either side of half the rail voltage. (I hope that made sense).

However....my circuit doesn't do this. Instead, the two inputs and the output all remain constant at half the rail voltage.

Would anyone be able to point out any glaring (/stupid) mistakes that I've made? Or whether the circuit is just not possible?

Thanks a lot in advance for any help you could give!

 

[BrownOut]

An oscillator required regenerative feedback. Where is yours? Why don't you use an architecture that works, instead if just hooking up some stuff and hoping?

 

[waggy55]

Thanks for your "help". As you may have guessed, I'm fairly new to implementing electronic design. I am aware of Colpitt's oscillator and other designs. However, I have limited components available. Because I started my design like this, for my own sense of completion I would like to finish the circuit using an op-amp so I can learn why it hasn't worked so far. I don't want to just have a completed circuit design, I want to understand why it works.

 

[BrownOut]

Your opamp is too frequency limited for you crystal. Old opamps only work at very low frequencies. I strongly suggest, as you are new to electronics, that you build an oscillator that works rather than try to make a whole new design. After that, you can explore other types and devices. If you don't know everything there is to know about a device, you can flounder for a long, long time.

 

[waggy55]

Ah right, ok. Thanks a lot. If that is the case, I'll move onto a proven design. I have learnt a fair bit about op-amps over the past week though!lol Thanks for your advice.

 

[BrownOut]

No it won't work ( see above post )

 

[waggy55]

Sorry, I'd edited my previous post, as I was writing it while you'd submitted a new one. Thanks though, I'll move on to a proven design.

 

[BrownOut]

At the frequency you're trying to achieve, component layout and lead inductance is going to bite you in the butt. If you really want Mhz, you're going to have to study high frequency layout methods. Don't even think about using a breadboard.

 

[waggy55]

Ah right. I'm working on stripboard too, so that's not going to be great with respect to high frequency. I might just try creating a square wave generator using an op amp, or an op Amp–based relaxation oscillator, as it's only to create a clock signal to test the function of a PAL.

 

[BrownOut]

You might get away using a stripboard @10Mhz. Just lay it out VERY carfully, make the leads short. Pay attention to how stray capaticance might be created. If it doesn't work, you might have to play with the layout till it does.

 

[waggy55]

Ok, I'll make sure I'm careful. Thanks a lot for your advice. I was hoping it wasn't working (I've tried numberous designs) because the op-amp would have rolled-off too much at that frequency - apparently the -3dB point is at 1MHz. Thanks again.