Question about quartz oscillators

[Flux]

Can you change the frequency of the quartz oscillator by pulsing it with a strong magnetic field? Or are they shielded too well?

Can an applied pulsed magnetic field alter the frequency to electrons so that they tunnel into to oscillator (by electromagnetic inductance) and the pulses change its frequency?

Can you put any other frequency through an appliance/computer other than 60hz/50hz? Will it destroy the machine?

Does the BIOS have full control of the cpu/gpu clock frequency? Or is it hardwired in by the manufacturer? Are there any ways to change the frequency of the device without manually overclocking it yourself?

 

[Frank]

Quartz oscillators are pretty dual-minded: feed them a current and they contract and expand, make them contract and expand and they will create a modulated current. And they do have a build in (harmonic) speed at which they oscillate in either case.

Magnetic fields don't work on the crystal, but they could on the case and current. Then again, you would need to oscillate that field as well, and when it isn't in lockstep with the crystal, nothing much will happen.

 

[_xxx_]

Asked another way around, what do you want to achieve?

 

[Flux]

Quartz oscillators are pretty dual-minded: feed them a current and they contract and expand, make them contract and expand and they will create a modulated current. And they do have a build in (harmonic) speed at which they oscillate in either case.

Magnetic fields don't work on the crystal, but they could on the case and current. Then again, you would need to oscillate that field as well, and when it isn't in lockstep with the crystal, nothing much will happen.

if you pulsed a current into the box at 333Mhz(wave frequency not pulse rate) and
pulsed it at 1Ghz, what would happen?

Theoratically by inductance they should be attracted to the 333Mhz Bus/oscillilator crystal and thus increase its clock.

Will the ciurcuts/transistors only work on a 60/50Hz waveform?

 

[Frank]

The whole idea of a quartz crystal is, that when you feed it a DC current, it will expand and contract, thereby modulating the current. (You need some tiny capacitors as well, but that doesn't matter.) And the rate at which they contract and expand is simply due to their physical dimensions. If you make it bigger, it will oscillate slower. The rate is fixed by it's size and geometry. That's what makes it such a good clock.

If you feed it a wave, I don't think it's going to oscillate, or at most slower. And if it doesn't oscillate, the wave will more or less pass through, although the amplitude will change.

If you want to be able to select a different CPU/bus speed, you would be best off with taking the crystal out, and replacing it with an oscillator that uses capacitance. You can adjust the speed of such a clock to whatever you like.

But the MB and BIOS have to be able to use that as well, for various stuff. And in that case it's almost certainly simpler and cheaper just to buy a MB that already has one of those.

IIRC, Asus and ABit made those a few years ago, and I'm sure you can still buy such over/underclock friendly motherboards, that allow you to specify just about any speed you fancy.

 

[Mize]

Actually many quartz oscillators, especially SAW-type, can be tuned based on drive circuitry. The peak self-resonance of a SAW device, for example, is a function of IDT spacing, surface doping and the RF drive circuit. These are commonly used as picogram-level detectors for explosives and more...a thin film that preferentially adsorbs the target is place on the wave-propagating surface and mass adsorption changes the self-resonance (deltaf/f equals deltam/m). So if you change the drive frequency you will change a self-resonance, but I can't say it'd be predictable without knowing exactly what you're trying to accomplish.