crystal oscillator frequency

"Johnny Chang" a écrit dans le message de news: snipped-for-privacy@m34g2000hsf.googlegroups.com...

Yes if it is used in parallel resonance mode. You can then theoritically tune it down to the serial resonance frequency (a variation of a couple of KHz for a MHz range crystal) by increasing the loading capacitor. No (or very little) if it is used at its serial resonance. But 99.9% of the CMos-based crustal oscillators are parallel mode.

No problem.

Friendly yours, Robert

Yes but it cannot change much, 500ppm is considered a large deviation for AT and SC cut crystals.

You can also use a series inductor to tune it above the marked frequency. 'Pulling' crystals like this is a technique often used by radio amateurs to cover part of a band with a single crystal.

Leon

-- Leon Heller Amateur radio call-sign G1HSM Yaesu FT-817ND transceiver Suzuki SV1000S motorcycle snipped-for-privacy@btinternet.com

You can also shift it by putting a capacitor in series with the crystal. This leads to a better oscillator in most cases. The circuit looks like this:

Y1 C3 !-! From amplifying device -------+-----!!----! !----+--- To amplifying device ! !-! ! =3D=3D=3D C1 =3D=3D=3D C2 ! ! GND GND

The ratio of C1/C2 is constant so that amplifier needs a nearly constant gain. Changing C3 changes the total capacitance seen by the crystal.

"No problem" may be the wrong answer here. How stable do you need the resulting oscillator to be? When you pull a crystal, you do suffer some reduction in stablity because the capacitor's value matters more.

You have to be really really careful about that, though...if the inductor is too large, the tank will resonate with the crystal being capacitive. In general there are two frequencies where a given crystal will have a given capacitive reactance. Also the crystal doesn't have nearly as high a rate of change of reactance with frequency out there, so the resulting Q is much lower.

Cheers,

Phil Hobbs

A series inductor tunes it DOWN in frequency. A series capacitor tunes it up.

One group that explored the obtainable range describe their stuff here:

Cheers,

Joop