EEVblog Electronics Community Forum

Products => Test Equipment => Topic started by: tangofan on December 23, 2014, 10:09:55 pm

Title: Measuring a non-periodic event with an oscilloscope?
Post by: tangofan on December 23, 2014, 10:09:55 pm
Hi all,

I am newly interested in electronics and am planning to buy a low-cost oscilloscope, like the Rigol 1054Z. One thing that I'd like to do is to measure "non-periodic" events, e.g. the signal after I press, switch or release a button in my circuit. And I should note that the signal in the circuit itself might be periodic before the event that I want to measure, say AC or DC with PWM.

Is this a standard feature on all oscilloscopes (including the 1054Z)? If not, does this feature have a name, so I can check out the specification of the oscilloscope to see, if it is there?

And finally do I need any other special equipment to measure this (other than a 2nd probe to capture the trigger signal)?

When googling for "oscilloscope non-periodic" I came across this article (  [url=http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php]http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php (http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php)[/url] ), which suggested that it was no problem, but I'd like to verify this, before I go out and buy something that does not meet my needs.

Thank you in advance for your help and advice,
Bernhard

Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Galaxyrise on December 23, 2014, 10:20:07 pm
Let's take one of the examples you cite: "measure the signal after I press a button".  You would do that by having a probe connected to channel 1 and the signal you want to measure and another probe on channel 2 and the button.  You'd set the scope to trigger off channel 2 when it crosses a certain threshold.  Then you'd do you measurement on channel 1. 
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: tggzzz on December 23, 2014, 10:25:44 pm
There are (at least) two fundamental ways in which non-periodic signals can be visualised on a scope. Which is more relevant depends on what you are trying to see.

The key phrases are
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Lightages on December 23, 2014, 11:20:26 pm
I suggest you watch the video in this thread. It will help with many questions regarding oscilloscopes.
https://www.eevblog.com/forum/beginners/oscilloscope-training-class-%28long%29/ (https://www.eevblog.com/forum/beginners/oscilloscope-training-class-%28long%29/)
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: T3sl4co1l on December 24, 2014, 01:44:08 am
Simple answer: if it's not in Equivalent Time (ET) sampling, it's doing it real time, and a SINGLE or NORM trigger will capture the event, when set up correctly of course.

AFAIK, all mainstream scopes these days are real time (100MHz BW with 1Gsa/s or so sample rate).  Older scopes (usually), with ET sampling, are still fine for single shot events, but only up to the sample rate where ET kicks in.

Analog scopes can also be used for single events, but unless you have a good eye, you'll want some storage mechanism... a storage CRT, or a mounted camera (put the camera in front of the CRT with the shroud, open the shutter, expose the event, then button it up and send it to the photo lab..), or a digital camera with long exposure.  This also helps if you have a slow but repeating signal.

Here's an example of a xenon flash tube light pulse (or rather, several) I recorded on a Tek 475 with 15 second exposure:

(http://seventransistorlabs.com/Images/Photoflash_Discharge2_sm.jpg)

3 x 50us of trace, out of 15,000,000us of exposure time, isn't too bad for a reasonable quality image.

Curiously, the earlier digital scopes did real time sampling on broadly similar time scales (I think the HP 54600 was around 10us/div fastest real time).

Tim
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: miguelvp on December 24, 2014, 02:53:36 am
You can set the trigger to pulse mode so if it's periodic at say 100ms but when you push the button it causes the signal to be 150 ms, you can set the mode to trigger when the pulse is greater than 110ms.

You can trigger on positive or negative, less than or greater than, the 1054z has pulse triggering btw.

Of course the values can be changed for your needs.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Wuerstchenhund on December 24, 2014, 05:58:08 am
I am newly interested in electronics and am planning to buy a low-cost oscilloscope, like the Rigol 1054Z. One thing that I'd like to do is to measure "non-periodic" events, e.g. the signal after I press, switch or release a button in my circuit. And I should note that the signal in the circuit itself might be periodic before the event that I want to measure, say AC or DC with PWM.

Is this a standard feature on all oscilloscopes (including the 1054Z)?

It depends on what you mean with "measure". Any scope (even some old analog banger, but I guess that's not what you want anyways) should be able to trigger to external events (i.e. a button press). That's pretty much standard on scopes. However, if you want to trigger on specific signal properties (i.e. runts, specific signal properties, digital sequence) then you'd need a scope with advanced trigger capabilities which usually aren't available in the low end class.

This assumes that "measuring" means you want to get the even on the screen. If you want to get the parameters of a signal deviation then a low end scope will only provide you with a very basic set of measurements. If you want to dig deeper you'd need a more advanced scope.

Quote
When googling for "oscilloscope non-periodic" I came across this article (  [url=http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php]http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php (http://www.radio-electronics.com/info/t_and_m/oscilloscope/oscilloscope-trigger.php)[/url] ), which suggested that it was no problem, but I'd like to verify this, before I go out and buy something that does not meet my needs.

This article is very generic, and not everything described there is available in a low end scope.

I suggest you think about what exactly you need (not just now, but also in the near future), as you might well find that a low end scope doesn't fit your needs. In which case I'd also consider buying a more advanced second hand scope.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Wuerstchenhund on December 24, 2014, 06:07:15 am
I suggest you watch the video in this thread. It will help with many questions regarding oscilloscopes.
https://www.eevblog.com/forum/beginners/oscilloscope-training-class-%28long%29/ (https://www.eevblog.com/forum/beginners/oscilloscope-training-class-%28long%29/)

I only skimmed through the video, and while the presentation style is good it seems to discuss analog scopes only. It's a nice history lesson but of little relevance to anyone wanting to buy a new scope in 2014.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: tangofan on December 29, 2014, 11:14:43 pm

It depends on what you mean with "measure". Any scope (even some old analog banger, but I guess that's not what you want anyways) should be able to trigger to external events (i.e. a button press). That's pretty much standard on scopes. However, if you want to trigger on specific signal properties (i.e. runts, specific signal properties, digital sequence) then you'd need a scope with advanced trigger capabilities which usually aren't available in the low end class.

By "measure" I'm referring to displaying the signal on the screen.

Quote
I suggest you think about what exactly you need (not just now, but also in the near future), as you might well find that a low end scope doesn't fit your needs. In which case I'd also consider buying a more advanced second hand scope.

As a beginner it's quite hard to know exactly what I'll need in the future, but one thing that I DO know that I would need is to trigger on specific signal properties. Example: If I run a DC PWM signal through my circuit and then press a button to cut the signal off, I'd like to trigger a measurement/capture based on the fact that the signal is no longer there (e.g. 0V level for more than x Microseconds).

And ideally it would be able to capture measurements on an ongoing basis, so when the trigger is "fired", it would still have some values right before the trigger event (Not sure, if this is possible though).

So if this is only possible with an advanced scope, what is the technical term to look for in the feature list (other than "advanced trigger capabilities", which probably is a bit too generic)?

Thanks & best regards,
Bernhard
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Someone on December 29, 2014, 11:28:40 pm
While you may be getting lost in the huge range of terms used for measurement tools, for instance the "Tektronix Oscilloscope Fundamentals"
http://circuitslab.case.edu/manuals/Oscilloscope_Fundamentals_-_Tektronix.pdf (http://circuitslab.case.edu/manuals/Oscilloscope_Fundamentals_-_Tektronix.pdf)

You are probably better served by reading the "Triggering Fundamentals"
www.tek.com/dl/55W_17291_6_0.pdf (http://www.tek.com/dl/55W_17291_6_0.pdf)
Trigger delay is such a basic function its only mentioned in passing through a few paragraphs. While it is a given these days in realtime sampling scopes, the older engineers who grew up with analog scopes were quite entertained with negative trigger delays when they first encountered them.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: miguelvp on December 30, 2014, 12:51:58 am
As a beginner it's quite hard to know exactly what I'll need in the future, but one thing that I DO know that I would need is to trigger on specific signal properties. Example: If I run a DC PWM signal through my circuit and then press a button to cut the signal off, I'd like to trigger a measurement/capture based on the fact that the signal is no longer there (e.g. 0V level for more than x Microseconds).

And ideally it would be able to capture measurements on an ongoing basis, so when the trigger is "fired", it would still have some values right before the trigger event (Not sure, if this is possible though).

So if this is only possible with an advanced scope, what is the technical term to look for in the feature list (other than "advanced trigger capabilities", which probably is a bit too generic)?

Thanks & best regards,
Bernhard

Let me reword what I said before.

By default the scope trigger is on type "edge" change it to "pulse" and change the parameters of the pulse to trigger when the negative pulse (0 V or anything less than the max value of the signal) is greater than x micro seconds.

The 1054Z supports pulse triggering.

Depending on your time scale and trigger position you'll be able to see the signal before it triggered.

Edit: also in the Acquire menu you can change the memory depth to match how much data you want, on the default "auto" it might not be too much before the trigger.

I have a DS2000 series Rigol but I'm pretty sure the 1054Z supports all of this as well.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: Wuerstchenhund on December 31, 2014, 11:54:37 am
By "measure" I'm referring to displaying the signal on the screen.

Ok, that's easy then.

Quote
As a beginner it's quite hard to know exactly what I'll need in the future, but one thing that I DO know that I would need is to trigger on specific signal properties. Example: If I run a DC PWM signal through my circuit and then press a button to cut the signal off, I'd like to trigger a measurement/capture based on the fact that the signal is no longer there (e.g. 0V level for more than x Microseconds).

And ideally it would be able to capture measurements on an ongoing basis, so when the trigger is "fired", it would still have some values right before the trigger event (Not sure, if this is possible though).

You should be able to do that with any decent low-end scope.
Title: Re: Measuring a non-periodic event with an oscilloscope?
Post by: tangofan on January 01, 2015, 06:21:03 am
Thank you very much to all of you who responded to my question and so generously shared your knowledge. You've helped me a lot.

A very Happy New Year,
Bernhard