Analog trigger modes have been around ever since oscilloscopes were
introduced.
Besides letting you stabilize repeating waveforms in an
oscilloscope-like display, analog trigger mode in data acquisition is especially important to study sudden
(single-shot) change of an analog signal
What are the traditional analog trigger modes?
In traditional data acquisition devices, one can find the following
common analog trigger modes:
Rising edge trigger mode, which is also called positive slope
trigger: Trigger happens when the signal passes a trigger
level while it is rising
Falling edge trigger mode, which is also called negative slope
trigger: Trigger happens when the signal passes a trigger
level while it is falling
Above Level trigger mode: Trigger happens simply when the
signal is above a trigger
level.
Below Level trigger mode: Trigger happens simply when the
signal is below a trigger
level
To assist better study of the trigger events, features like pre-trigger
length, post-trigger length and hysteresis are added.
Problems with traditional trigger modes
Case 1: We have a slow moving sine wave, swing from rail
to rail, but once a while there is a glitch riding on it. Now we
need to catch the glitch and study it.
Since the waveform is swinging from rail to rail, none of the
traditional analog trigger modes can catch it.
Case 2: We have a sine wave of 1Hz, swing from rail to
rail, but once a while its frequency changes to 10Hz. We need to
trigger on the change
Again, none of the traditional analog trigger modes can catch it.
In calculus (a branch of mathematics) the derivative is a measure of how a function changes as its input changes. Loosely speaking, a derivative can be thought of as how much one quantity is changing in response to changes in some other quantity; for example, the derivative of the position of a moving object with respect to time is the object's instantaneous velocity.
By monitoring the first-order derivative of an analog signal, we
can trigger on the sudden change of a waveform!
Although the first order of derivation is described as df/dt, in
practice of data acquisition, since dt, which is the inverse of sample
rate, is always the same during the session, we can simply substitute
df/dt with df and monitor the change of df continuously.
If the signal is relatively clean or the change rate is not
too small, a simple two-point difference is enough. This method
is demonstrated in the youtube
video below.
If the signal is noisy and the change rate is small, linear-regression
instead of two-point difference is necessary to warranty
a real trigger. In this case, a FPGA or a high-power DSP will be
the perfect candidate for the job when high-speed operation is
required.
For easily implementation, we choose two-point difference in the
following solution to trigger on a relatively strong glitch.