0
\$\begingroup\$

I am trying to understands the basics of an analog oscilloscope trigger. As stated here,

Instead of collecting a single snapshot of the input signal, an oscilloscope repeatedly updates the display with newly collected snapshots. Each snapshot is called a trace.

The trigger is used to not overlap the successive traces of the signal on the screen.

Each trace begins when a trigger event occurs: that is, when the signal has a specific slope (positive or negative) and a specific voltage level. If I understood correctly, the trace begins each time a trigger event occurs. However, this would imply that the oscilloscope can show in its display just one period of the signal. Instead, there is plenty of images of oscilloscope displays, using the trigger, with multiple periods of the signal.

For example, from this page:

enter image description here

It seems that in all these examples the trigger is ignored for an arbitrary number of periods after the trigger event.

How is it possible? Isn't this a contradiction?


Edit:

The trigger is used to not overlap the successive traces of the signal on the screen.

Sorry, I was meaning exactly the opposite. One of the answers points this out.

\$\endgroup\$

2 Answers 2

2
\$\begingroup\$

Once triggered the trace runs to completion (at whatever speed you set the timebase to). Any trigger events occurring during the trace are ignored. After a trace has completed the next trigger event that occurs will start the next trace.

The trigger is used to not overlap the successive traces of the signal on the screen.

Overlap can be interpreted in somewhat different ways. In this case we use the trigger so that each successive trace is drawn in the same place as its predecessor. So if we have a stable input signal we see an apparently static trace on the scope, even though it's actually being continuously redrawn from different samples of the input signal.

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Most scopes have a user-defined holdoff setting to delay longer than the minimum amount before arming the trigger circuitry to capture the next trace. \$\endgroup\$ Commented Mar 9, 2021 at 20:17
2
\$\begingroup\$

The difference in quality of scopes is often measured by how good the trigger circuit works to prevent aliasing on the screen with the signal in chop mode or how fast it can retrigger in alternate mode and how clean it can separate signal from noise or noise from signal. So the Channel chopper would have a PLL that ensures chopping is a blur compared to sweep rate triggered with an offset frequency to the signal chopper. (Details...)

DSO’s are different in acquisition but have similar trigger requirements.

But there are a variety of signals that have noise that you may wish to trigger on instead of the steady signal, then you decide LF reject or HF reject AC or DC couple, select the source, the time delay on waiting for next trace or delay for expanded sweep. Other triggers include TV which AC couples then clamps to negative sync pulse to restore DC then you can adjust sweep for a whole horizontal signal or a vertical signal or alternate sweeps etc.

TEK used to make the best triggers. HP not bad, Rigol don’t know.

The nice thing about analog scopes is they often allowed you to viewed the filtered trigger signal so you got an extra trace for free. ;)

I used to get creative and use the external X sawtooth ramp signal to inject into an HDD servo then view the data signal as the heads moved across +/10% of track back and forth to see how the signal integrity changed (Adjacent track interference, (ATI) with Position error signal (PES) during write) to validate specs for design reviews or second sourcing “R/W heads”. One could do the same thing with an FM signal on a RX or a radar with Doppler so the X trigger was synchronous with the FM or servo position error at some sweep rate it could manage.

To get even fancier, I would select channel 2 for trigger then use Vertical channel out on the rear BNC selected by trigger source then adjust the gain and position for DC offset to test Power supplies with a swept current using the signal to control load current with a power transistor. then check the stability by inverting the Ch2 polarity for next bunch of sweeps. (A-B mode)

Sorry for too much detail, some may recollect the same.

\$\endgroup\$
3
  • \$\begingroup\$ Thanks for your post, it includes many tweaks and tips due to a massive experience. However, I struggle to find here an answer to my question. \$\endgroup\$
    – BowPark
    Commented Mar 9, 2021 at 21:27
  • \$\begingroup\$ The sweep time has nothing to do with the trigger event. It only determines how much the signal is expanded or compressed. The only thing you cannot do is retrigger on the same cycle or even the next cycle in analog scopes because of the inherent blank time for the 25kV ish flyback transformer to reset the flux and thus create a really high voltage sawtooth. \$\endgroup\$ Commented Mar 9, 2021 at 21:30
  • \$\begingroup\$ This allows starting the trace in exactly the same position. which is adjustable by X offset. But the time must be short but often measured by experts for quality and BW of scope. maybe 10% of the fastest sweep rate is allowed for blank time with no chance to retrigger during that time. Of course slow pulses can be displayed consecutively and DSO's have an advantage while they capture before and after and put the triggered signal in the middle of the trace if you want. \$\endgroup\$ Commented Mar 9, 2021 at 21:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.