I just learned that one must set the Acquire feature of a scope to "Peak Detect" to see the possible glitches of pulses.

But in the following document it also mentions that:

"This mode is effective when at the horizontal scale setting of 5 ms/division or slower." Here is the source: http://www.tek.com/manual-topic/acquire

How reliable is this information? Do one really needs to set the horizontal scale to 5 ms/division or more to make use of the peak detect mode? Why is that?

  • \$\begingroup\$ First of all, every scope is a little bit different in implementing this, so you can't necessarily take one manufacturers statement and apply it to all others. On a lot of scopes you don't see any difference with peak detect mode when you are in the fastest sampling mode, only when you are in slower sampling modes. \$\endgroup\$ – PlasmaHH Nov 24 '16 at 10:28
  • \$\begingroup\$ Oh i'm trying to get what you mean. Do you mean if the sampling rate is already set to max. in a scope, there is no need to set the Acquire to Peak Detect to see glitches? \$\endgroup\$ – floppy380 Nov 24 '16 at 10:33

Since there is a no continuous sampling above Peak Mode, glitches may be missed with , capture, processing time, display and repeat missing events

Peak Detect. Use Peak Detect acquisition mode to detect glitches as narrow as 10 ns and to limit the possibility of aliasing. This mode is effective when at the horizontal scale setting of 5 ms/division or slower .

Note: If you set the horizontal scale (seconds/division) setting to 2.5 ms/div or faster, the acquisition mode changes to Sample because the sample rate is fast enough that Peak Detect is not necessary. The oscilloscope does not display a message to tell you that the mode was changed to Sample

However , this scope can miss glitches <10ns unless they are continuous rep. rate sync'd to fundamental or harmonic of sweep speed with sampling rate up to max rate.

Although, if there are no triggers, a glitch can be reliably captured on manual trigger mode at max sample rate up to 1/2 of sample time (best case)

Displayed below is an 8ns glitch at -10.8V with 1ns rise time, that would not be captured in slow "Peak mode" on your DSO. (not slow poke)

enter image description here

Note the interpolation by straight lines on this old LeCroy.

Whereas your DSO would produce nice textbook waveforms on single shot capture with 50 Ohm termination only.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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