AnsweredAssumed Answered

Saving more than one channel help

Question asked by mulligan252 on May 25, 2014
Latest reply on May 30, 2014 by mulligan252

Like • Show 0 Likes0
Comment • 8

I wish to remotely capture and save three channels on the oscilloscope (Infinnium 9000 series) at the same time . What is the best way/ commands to capture and save multiple channels on the oscilloscope?

I'm not sure how the scope saves all channels at once. Is there a separate buffer for each channel or is there just one buffer but the scope somehow determines which data is from which channel? I am trying to implement the same thing in my code as selecting the options on the GUI "multipurpose on trigger" and "save all waveforms".

Thanks alot for any help in advance.

No one else has this question

Outcomes

8 Replies

algoss May 27, 2014 4:36 AM
Mark Correct
Correct Answer
Here are a few hints:

- If 3 channels are turned on, and you trigger the scope, 3 channels get captured.
- If you save channel 1, then channel 2, then channel 3, they will all have the same timestamps, relative to the trigger.

So it's really no different than capturing 1 channel, just 3 times as much data to save.

Al

Disclaimer: For more reliable response, you should consider calling your local Agilent Technical Call Center. The Agilent Forums are monitored on an "as available" basis, and aren't necessarily the fastest way to get technical questions answered.
Like • Show 0 Likes0
Actions
- mulligan252 @ algoss on May 27, 2014 5:47 AM
  Mark Correct
  Correct Answer
  When each channel is captured, are they stored in separate buffers tho? You mentioned that there is only one buffer. I would like to be able to call :WAVeform:Data? and retrieve all the data for the 3 channels at once.
  
  We are running tests to figure out the best and fastest way to retrieve waveform data from the scope remotely. Either by using a python script running on the scope or by sending commands remotely.
  
  If I select "Multipurpose on trigger" and "save all waveforms" via the screen interface, the scope stores each channel data in series in a file such as .H5. What exactly is the scope doing here? There doesn't seem to be the overhead that there is when using :DIGitize. We then transfer them files via a python script to the destination.
  
  In you opinion, should it be faster acquiring data by sending commands remotely instead of using the script. Which commands can I send in order to implement the same action that the scope is performing when I select "Multipurpose on trigger" and "save all waveforms" via the screen interface? Thanks
  Like • Show 0 Likes0
  Actions
  - algoss @ mulligan252 on May 27, 2014 6:16 AM
    Mark Correct
    Correct Answer
    Q - When each channel is captured, are they stored in separate buffers tho? You mentioned that there is only one buffer. I would like to be able to call :WAVeform:Data? and retrieve all the data for the 3 channels at once.
    
    A - There is one buffer per channel. Once the scope has triggered, you need a sequence like:
    :WAV:SOUR CHAN1
    :WAV:DATA?
    :WAV:SOUR CHAN2
    :WAV:DATA?
    :WAV:SOUR CHAN3
    :WAV:DATA?
    
    Q - We are running tests to figure out the best and fastest way to retrieve waveform data from the scope remotely. Either by using a python script running on the scope or by sending commands remotely.
    A - Do you want to save the data into files or get the data from the scope buffer for processing? The commands for each are very different. Running Python on the scope, or running it remotely, assuming GBit LAN and relatively little other traffic on the LAN will run at about the same speed. They both go through the same 'socket' on the scope.
    
    Q - If I select "Multipurpose on trigger" and "save all waveforms" via the screen interface, the scope stores each channel data in series in a file such as .H5. What exactly is the scope doing here? There doesn't seem to be the overhead that there is when using :DIGitize. We then transfer them files via a python script to the destination.
    
    A - Again, do you want to save the data in files and transfer those files, or just get the data for processing and save then? Sucking the data from the scope will be faster than saving into files and transferring the files. The transferred files will be larger than the raw data that is in them. Converting the data on the scope is relatively slow. HDF5 is the slowest.
    
    Q - In you opinion, should it be faster acquiring data by sending commands remotely instead of using the script. Which commands can I send in order to implement the same action that the scope is performing when I select "Multipurpose on trigger" and "save all waveforms" via the screen interface? Thanks
    
    A - Running the scope remotely and transferring the data as it is captured will be faster then asking the scope to save it all, then transferring it later. The file conversion and save time will kill you...
    
    Al
    Like • Show 0 Likes0
    Actions
    - mulligan252 @ algoss on May 27, 2014 7:11 AM
      Mark Correct
      Correct Answer
      I just want to suck the data from the scope as fast as possible. 250KB per channel/signal at present. You told me the fastest way to do this was by using :DIG and then WAVEFORM:DATA?. I disagree, because we are loosing waveforms when the trigger reaches 10 Hz in that case.
      
      If I however get the scope to save the data as files (HDF5 in this case) and transfer later , we are not loosing any data, even when the trigger is at 10 Hz!!!!. I thought also that this HDF5 format would be the fastest as it is the smallest of the available file formats containing only the voltage data with some header information.
      
      I have tried the following after every trigger, and again it is too slow and we are loosing data:
      :WAV:SOUR CHAN1
      :WAV:DATA?
      :WAV:SOUR CHAN2
      :WAV:DATA?
      :WAV:SOUR CHAN3
      :WAV:DATA?
      
      The interesting thing now is why is it when the scope is creating and saving the files that it is much faster than trying to pull the data directly from the scope buffers? We are using a dedicated GBit LAN so network speed is not an issue.
      
      This :DIGitize command is what takes too long. Is it suffice to poll the register using :ADER? to see when the buffer(s) is full and then transfer using :WAVEFORM:DATA?. By doing this, It seems much faster and I am not loosing any data. Is this simply what the scope is doing behind the scenes when it is saving the files with "multipurpose on trigger" ?
      
      Also, are there commands to store each channel data locally on the scope for example after every trigger? There are only 4 waveform memories it seems but is there more memory I can use? Many thanks
      Like • Show 0 Likes0
      Actions
      - algoss @ mulligan252 on May 27, 2014 7:20 AM
        Mark Correct
        Correct Answer
        You have apparently found a solution that works for you, and don't like my answers much, so have at it... There is, of course, no guarantee that your solution will provide the same level of performance, or work at all, at some future date after a new version of scope SW has been loaded, or on a different model of scope.
        
        Q - Also, are there commands to store each channel data locally on the scope for example after every trigger?
        
        A - What do you mean "locally"? There are the 4 waveform memories, discussed below, and there are files. There is also segmented mode, discussed in a previous thread.
        
        Q - There are only 4 waveform memories it seems but is there more memory I can use?
        
        A - I don't understand this question. The 4 "waveform memories" have nothing to do with the waveform buffers, They are used for loading file data back into the scope for math or comparison with live data.
        
        Al
        Like • Show 0 Likes0
        Actions
        mulligan252 @ algoss on May 27, 2014 7:40 AM
        Mark Correct
        Correct Answer
        I am not saying that I don't like you answers. I am still unclear as to why it is faster for the scope to make the files and store them in real time in comparison to trying to acquire remotely the data directly from the buffers. It doesn't make sense. I would like to understand exactly what is the scope doing under the hood when it is saving the files and I can try and implement the same in my code?
        Like • Show 0 Likes0
        Actions
        algoss @ mulligan252 on May 28, 2014 5:41 AM
        Mark Correct
        Correct Answer
        When you use something like "Multipurpose" to capture and store data, you are using a built-in macro that was written into the SW. It doesn't have to parse any commands, do any interaction with the LAN, or anything else. Data is captured and transferred to the disk, that's it. If that's all you need, you should take advantage of it. In general, a small, well-defined task can be written to be faster than a more general task that requires interaction with, for instance, a user.
        
        Most customers that I talk to are interested in the fastest way to get deep traces, 10 MPoints or more, off the scope for processing or storage. They can't risk filling up the disk in the scope, and they may want to do some processing, or data reduction, on the fly.
        
        Al
        Like • Show 0 Likes0
        Actions
        mulligan252 @ algoss on May 30, 2014 4:00 AM
        Mark Correct
        Correct Answer
        Ok, thanks for your help.
        Like • Show 0 Likes0
        Actions

Saving more than one channel help

Outcomes

Related Content

Recommended Content