Testing a crystal.

You need a special probe to keep from loading down the circuit, and if it is a typical 32.768 kHz watch crystal the waveform isn't clean even when it is working properly. It is a tuning fork type of crystal, not the typical cut used at higher frequencies. A high input impedance FET amplifier with a fraction of a pF coupling capacitor is typically used to prevent loading, but you can often see a waveform at the drive side of the crystal. They use an unbuffered inverter and a resistor to bias it into the linear range, but the AC voltage in the circuit pushes it out of the linear range.

The easiest test that I can think of is to breadboard a CMOS oscillator circuit. Use one inverter out of a hex inverter like this:

View in a fixed pitch font such as Courier

|\\ +--| >0---+----> OUT | |/ | | | +--\\/\\/\\--+ | 1 Mohm | | | | \\ | / 2.7 kohms | \\ | / | | | Xtal | +---|[]|--+ _|_ _|_ 55pf ___ ___ 60pf _|_ _|_ \\ / \\ /

If the circuit oscillates at the crystal frequency, then the crystal is likely good. If it doesn't oscillate at all, or oscillates at the wrong frequency, then it's likely bad.

Testing crystals,ceramic resonators and filters Signal generator (RF) and scope. Connect the grounds together and then connect the sig gen to the scope via the crystal. Sweep the oscillator and resonant response at fundamental and overtones will show on scope trace. Use a f counter as well if f is required. If testing while in circuit (unpowered) use less then 1V sig gen o/p . The response may be muted comnpared to free component especially for ceramic resonators.

from my tips files below

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

...and the device number for the gate needs to have a "U" in it (for "unbuffered").