AnsweredAssumed Answered

Infinium 90000A edge then edge trigger

Question asked by rraman64 on Aug 22, 2016
Latest reply on Aug 26, 2016 by rraman64

Like • Show 0 Likes0
Comment • 8

Hai all,

I have specific task of using the edge then edge trigger and store the wave forms for offline analysis.

Have two time-correlated random pulse trains into ch 1 & 2. I arm with ch1 and trigger with edge then edge trigger mode on event 0. When I use quick save using multipurpose trigger task, I was expecting each set/file to have pulse in both channels. How ever, most of the sets/files have only channel 1 pulse while channel 2 is baseline (thresholds are well above noise).

Is there any thing I miss in using this trigger mode to capture & save only when both chennels have one pulse each?

Thanks in advance

-rajaraman

Fungible Aug 24, 2016 9:53 AM

Correct Answer

rraman64,

I think your solution is in Sequence trigger. The title of this discussion says you have a 90000A. This should have Sequence trigger.

The setup below will trigger on ch1 falling edge, then ch2 falling edge. If no trigger occurs in 100ns, it resets and starts looking for ch1 trigger again. This should trigger on Case 2 only.

Sorry I did not pick up on this sooner.

See the reply in context

No one else had this question

Outcomes

8 Replies

Fungible Aug 22, 2016 12:46 PM
Mark Correct
Correct Answer
This is a complex trigger mode. Which means a number of things might be going wrong. Which might mean we need more information about your task and setup.

Sounds like your "edge then edge" trigger mode is set up as- trigger on edge (pos or neg) of channel 1, delay by 0 events, then trigger on edge (pos or neg) of channel 2. Is this correct?

If it is correct, then it is odd that some of the files have no pulse in the channel 2 file. Because if there was no pulse in channel 2, it should not have triggered. You note the threshold is above the noise. Which means it did not trigger on noise. What is the time base setting? Is the pulse the trigger saw on channel 2 so narrow you can't see it?

What file format are you saving the data in (.wfm, csv, etc)?
Forget about file saving for the moment, if you place the scope in "trig'ed" mode (not "auto"), clear the screen and wait for a trigger, does a pulse appear in channel 2? Wondering if its an error in the data storage as opposed to the trigger error.

"Is there anything I miss in using this trigger mode to capture & save only when both channels have one pulse each?"
Not sure "edge then edge" is a mode that will guarantee triggering with just one pulse on each channel. It will only guarantee you got an edge (might be more then one) on one channel fallowed by an edge (might be more than one) on the other channel.
Like • Show 0 Likes0
Actions
- rraman64 @ Fungible on Aug 22, 2016 11:36 PM
  Mark Correct
  Correct Answer
  Thank you for Prompt response.
  Let me explain what I am looking for in more detail.
  
  We have pair of pulses (start - ch1 and stop - ch2) from our experiment which are produced randumly.
  The time differences between each pair of pulses is what we are interested in (with need to apply contraints on the pulse heights).
  So when a start pulse occur, there will always be a stop pulse which may occur almost simultaneoulsy or at a later time extending to inifinity! But the problablity of the time diference decays as exponential and hence we should be able to see lots of pairs within "hold off time of 100ns". At the same time, our experiments are so tuned that probability of seeing two pairs within 100 ns is negligible.
  
  So within 100 ns, only 4 possibilities exists.
  1) no pulses in both channels,
  2) A pulse each in both channels,
  3) a pulse in ch1 and not in ch2,
  4) A pulse in ch2 and not in ch1.
  I want to capture only case-2.
  
  by "arming" on ch1, I am eliminating case 1 & 4
  by demanidng "edge then edge" with delay by 0 event on ch2 and triggering with ch1, I am supposed to capture only case-2.
  But in reallity I end up with case-2 and case-3. See captured images of these two cases.
  Unfortunately case -2 is only about 10% and I am ending up in storing & processing more than 90% of unwanted signals
  
  With respect to saving, I have tried all formats.
  
  My feeling is either "edge then edge" or "hold off time" is not doing what it is promised to do.
  Like • Show 0 Likes0
  Actions
  - Fungible @ rraman64 on Aug 23, 2016 12:58 PM
    Mark Correct
    Correct Answer
    Thanks for the detailed explanation of your objectives and the screen shots were a big help.
    
    I believe the problem is in your understanding of “Delay by”. You have defined “Delay by event = 0”. When the system sees “0 events” it disables the delay. So the scope did nothing with respect to channel 2. It armed on a falling edge of channel 1 and then triggered on the next falling edge of channel 1. This will result in capturing Case 2 and 3, which is what you are seeing.
    
    If you changed the number of events to 1, you would still capture Case 2 and 3.
    If you changed the trigger to “arm on falling edge ch1, delay by 0 events and trigger on falling edge of ch2”, you would capture Case 3 and 4.
    
    I don’t think “Edge then Edge” trigger mode can be made to capture just Case 2. A “Hold Time” trigger would work except for the fact the maximum hold time you can set is 25ns. You want 100ns. So I am not sure you can do this with hardware trigger.
    
    Do you have an InfiniiScan license in you scope? It would appear under the “Trigger” menu or in the “About Infiniium” menu.
    
    How often do the events occur? Are they free running and at what rate?
    Like • Show 0 Likes0
    Actions
    - rraman64 @ Fungible on Aug 24, 2016 12:06 AM
      Mark Correct
      Correct Answer
      Thanks a lot for the valuable information.
      
      No. I dont have Infiniscan liscence.
      Events (qualified start pulse) are typically separated by about 0.2 to 1 microsecond. this was intentional to avoid multiple event registrations in each sweep.
      
      Ok I am abondaning the "edge then edge"
      I just started exploring "setup & hold" triggering as you suggested.
      Looks promising as I do see only case-2. I will play around the controls to see its applicability for my requirements.
      
      Alternately I can make use of quick measuements to extract information I want.
      But is there a way to save quick measurement results of each event?
      I only see option to save cumulative results.
      may be I should go for low level programming?
      I have liscence for matlab user functions. can it be used?
      
      thanks again for opening another door to explore.
      Like • Show 0 Likes0
      Actions
      - Fungible @ rraman64 on Aug 24, 2016 9:53 AM
        Unmark Correct
        Correct Answer
        rraman64,
        
        I think your solution is in Sequence trigger. The title of this discussion says you have a 90000A. This should have Sequence trigger.
        The setup below will trigger on ch1 falling edge, then ch2 falling edge. If no trigger occurs in 100ns, it resets and starts looking for ch1 trigger again. This should trigger on Case 2 only.
        Sorry I did not pick up on this sooner.
        1 person found this helpful
        Like • Show 1 Like1
        Actions
        rraman64 @ Fungible on Aug 24, 2016 10:49 PM
        Mark Correct
        Correct Answer
        Haa. This is great!
        Though I did explore sequential trigger, i got carried away by "AND qualifier" than the sequence-reset option sat at top.
        Now it works nicely.
        I will process the saved data and see.
        
        I am still interested in saving the quick measurement data as array for each event instead of final mean data.
        Should I post this as new reqeuest?
        
        Thanks a lot again.
        Like • Show 0 Likes0
        Actions
        Fungible @ rraman64 on Aug 25, 2016 1:43 PM
        Mark Correct
        Correct Answer
        You can post the question under a new heading if you want a broader viewing (and perhaps more comments) for the question.
        
        The saved quick measure data includes a column labeled “current”. That number is the measurement for the last data capture. Sounds like that is the number you want.
        Below is a quick measure save after clearing the display (that clears measurement data) and one capture. Note the current, mean, min and max are all the same. On the next capture the current will have the value for that capture and mean, min, max, range and std dev will be update based on the new value in the current column.
        
        Like • Show 0 Likes0
        Actions
        rraman64 @ Fungible on Aug 26, 2016 9:31 AM
        Mark Correct
        Correct Answer
        Thanks a lot
        I will post this requirement as new one
        Like • Show 0 Likes0
        Actions

Infinium 90000A edge then edge trigger

Outcomes

Related Content

Recommended Content