Author Topic: Waveform templates for scope reviews  (Read 1519 times)

0 Members and 1 Guest are viewing this topic.

Offline Wuerstchenhund

  • Super Contributor
  • ***
  • Posts: 2847
  • Country: gb
  • Only sparsely active but available via PM
Waveform templates for scope reviews
« on: July 18, 2014, 06:54:31 pm »
Hi All!

I got two 'new' scopes and am preparing to do reviews of them. The only limitation is that my primary signal source is still a Siglent SDG1020 20MHz AWG (didn't find time yet to expand my home lab).

However, often enough people are looking for specific tests that the reviewer doesn't think of when doing the review. So aside from a demonstration of functionality and the obligatory waveform/s tests, is there anything else you would like to see in a review?

I also thought that if someone is looking for a specific test (i.e. glitch capturing with a specific signal) then he could provide the signal as waveform template (.csv format) which I can upload to my Signet SDG1020 and do the test with. Those with newer Siglent scopes (i.e. SDS1000CNL/CML) can also just capture a waveform and save it as data file for Siglent AWGs.

Thinking further, if we collected these files we could build a standardized set of tests for review, which could make it easier to compare basic performance parameters of different scopes.

Any comments/ideas/rants?

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10665
  • Country: us
  • DavidH
Re: Waveform templates for scope reviews
« Reply #1 on: July 19, 2014, 11:31:49 pm »
The two tests that I like to do involve high speed pulses which your 20 MHz AWG is not going to handle at all.

I would test for glitch response at slow horizontal speeds with pulses below 100 nanoseconds using both the minimum and maximum record lengths.  An old 2230 for instance can catch a 100 nanosecond pulse 100% of the time at any sweep speed.  An old 2232 does the same with a 10 nanosecond pulse.  More modern DSOs should only be limited by their vertical bandwidth but some fail spectacularly.

The other test I like to do is to check for aliasing on fast edges which will reveal non-linearity and jitter in the digitizer.  Modern real time oscilloscopes seem to be more susceptible to this than old equivalent time oscilloscopes because their real time sampling rates are relatively low.  Aliasing is revealed by "wobbulation" in the sin(x)/x reconstruction.  This test may also be done with a fast sine wave source operating at close to the oscilloscope's rated vertical bandwidth independent of the current Nyquist bandwidth.

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo