Colpitts crystal oscillator in LTSPICE

The crystal has to be in its inductive region for a Colpitts to work.

Cheers

Phil Hobbs

Oh, and you can make simulated oscillators start faster by putting a current source in parallel with the inductor. Make it 1 uA or something, and drop it to zero near the start of the simulation. That's the SPICE equivalent of plucking a guitar string.

Try it with 1 nA when you're done, to check for startup problems.

Cheers

Phil Hobbs

On Sat, 25 Jul 2015 10:56:05 -0400, bitrex Gave us:

Oscillators (of this type) can have what are known (referred to) as "hard start" issues.

Usually a little inductance and or capacitance on the transistor leg provides enough lead or lag to allow the oscillations to get going.

Here's a programmable crystal Spice model...

For Fo the parallel resonance, set Series=0, for series resonance, set Series=1

(You can often get all these parameters from the crystal datasheet.)

As Phil pointed out most feedback-based oscillators (which queue up in a linear operating region) need a kick start.

Start with Q low, otherwise you will have timestep issues, and won't see any oscillation. Then tighten up Q as you make max timestep smaller.

My rule-of-thumb for oscillators is max timestep = 1/Fo/32 ...Jim Thompson

Thanks, I'll give that a try.

Thank you. Usually most oscillators I design get going simply by setting the option "Start external supply voltages at 0 volts" or whatever - that gives them enough kick. It doesn't seem to be happening in this case.

Does the crystal stabilized Colpitts need any other modifications from the standard version? One can just swap the inductor for a crystal and it should work the same, right?

Select "skip initial operating point" in the transient analysis, or it will never start. Or goose it as Phil suggests.

XOs are terrible to sim in time domain. The sims take forever and there's no reasonable way to measure the frequency to PPM resolution. Once I get one running, I cut over to AC loop analysis to fine-tune things.

Post your netlist so people can play with it.

That's another way of accomplishing a kick start.

If it doesn't start, the gain/phase criteria isn't met.

I don't speak buzzword (Colpitts, Pierce, etc) oscillators because they're rarely used with microchips... most of my clocks are simply a crystal around an 'HCU04 or equivalent, and my PLL VCO's are AGC'd differential configuration.... starting with my MC1648 from the mid '60's. ...Jim Thompson

Nope. With an inductor there's only one place it can oscillate. With a crystal and any external inductance, there are two, one below the series resonance, and one above the parallel resonance.

Also of course the rate of change of reactance with frequency is doubled for a crystal, because the L & C are fighting each other.

The reason your usual procedure isn't working here is that the circuit constants are too extreme--you aren't getting any initial amplitude on the series LC. Try the current source.

Cheers

Phil Hobbs

Filed away. Thanks!

Cheers

Phil Hobbs

I'm inherently lazy... so I will spend days developing Spice gimmicks to save a few aggravating minutes time doing tweak, tweak, tweak... and Spice models... that's how I got into the Spice behavioral modeling game ;-)

Here's one of several "tanks" I have built into my symbol libraries...

L^@REFDES %pin1 %pin2 {@Zo/(6.283185*@Fo*@Q)} C^@REFDES %pin1 %pin2 {@Q/(6.283185*@Fo*@Zo)} R^@REFDES %pin1 %pin2 @Zo

(PSpice can do math inside a symbol without needing a subcircuit call)

In subcircuit jargon this would be:

.SUBCKT TANK pin1 pin2 PARAMS: Fo=1Meg Zo=10K Q=10

I also have tanks with inductive and capacitive taps... probably should do a tuned transformer with everything selectable... to emulate IF strips ;-) ...Jim Thompson

Have you checked the thing in open loop to make sure it has a chance of starting?

Is the transistor biased correctly?

Why a Darlington? That seems to not just be asking for trouble, but sending a footman with an engraved invitation.

You say you're just using a cap to simulate the crystal -- don't you mean a cap and an inductor, or (better) two caps and an inductor?

Like John said -- post your circuit, maybe there's something obvious we can help you with.

Those values would seem to indicate a 32.768kHz time-keeping kind of crystal. It might be even the darlington input Z and gain are too low?

Perhaps try to get it to sim-osc with an op-amp first and then go descrete?

piglet

For that matter, why not choose a rail-rail op-amp in a SOT-23 or smaller that has known-good large-signal behavior?

Just sayin' -- you can always do better with discretes, but it may take lots of work to get there.

I used to ramp the supplies to get oscillators to start (and the simulator converge).

Does that really tell much, given component tolerance?

Here's a Dropbox link to the files:

I'm actually attempting to use the Darlington buffer inside a LM13700 section to build a Colpitts crystal oscillator, which will then be 2 quadrant multiplied by an external signal within the transconductance amp itself...

I'd guess that C2 and C3 are way too big.

Lowering C2 and C3 to 470p, making the emitter resistor 4.7k, keeping the feedback cap the same and setting the crystal resonant frequency to

The waveform is nothing to write home about, however.

Don't quit now; keep going.

And maybe make the upper one a bit smaller than the lower one. But I never understood Colpitts oscillators.

I did recently design one using a coaxial ceramic resonator, at 600 MHz. It works fine even if I don't understand it.