NASA Office of Logic Design

A scientific study of the problems of digital engineering for space flight systems,
with a view to their practical solution.

Some Characteristics of Crystal Clock Oscillators During the Turn-On Transient

Here are some examples of oscillator response for the start up transient.   It seems that there are as many "signatures" during the transient as the number of oscillators that I can find to test.  Additionally, test conditions such as power ramp rate, time between power cycles, temperature, all play a role in the characteristics of the turn-on transient.  For instance, while many of the oscillators that I have tested have shown the tendency to stay at zero until oscillation starts, we see the opposite in the 8 MHz oscillator that was tested with the results shown in Figure 1.  Some oscillators will start to oscillate early with the amplitude following V_CC closely.  Others will wait a considerable amount of time before starting to oscillate.

Even a single physical oscillator can have wildly different and repeatable responses as a function of the test conditions.   Thus, one can assume that various anomalous outputs, such as runt pulses, can occur.  For the 200 kHz flight oscillator that was tested with the results shown in Figure 2, depending on the ramp rate of the power supply the glitch can be made to grow or disappear.  Additionally, the initial pulses were not well formed and did not have the correct period, which also has been seen in other oscillators.

Figure 1.  Oscillator starts high and then several low-going runt pulses are output.

Figure 2.  The oscillator outputs runt pulses.  Amplitude and width of the runt
pulse is a function of the test conditions such as temperature, ramp rate of V_CC, etc.

Some oscillators start functioning almost immediately, at extremely low voltages.  The charts in Figures 3A and 3B are two views of the same start-up transient, with the trigger of the oscilloscope set at 1V.  I didn't try any lower trigger settings.  Figure 3A shows the waveform at around the 1V level while Figure 3B shows how the amplitude of the oscillator's output rises with the power supply.

Figures 3A and 3B.  Oscillator functioning lower then V_CC = 1V.

Home - NASA Office of Logic Design
Last Revised: December 02, 2003
Digital Engineering Institute
Web Grunt: Richard Katz