Waveform Files

There are two classes of waveforms : file waveforms and derived waveforms. File waveforms are created by analysis programs on CICADA or on the cluster; they can have multiple versions. Examples are: MB-IP-SL, FM-TOTNE. Derived waveforms are algorithms which calculate the data dynamically from file waveforms, e.g., M-PTOT.

There is a single waveform file for each shot. Every shot starting with 8486 with minimal plasma current has a waveform file. Most waveforms are written to the file within 5 minutes after the shot by analysis tasks that run automatically for various diagnostics. However the cluster analysis tasks can be run at any time after the shot. Also new versions of a given waveform can be written months or years later to correct for bad calibration data or other problems.

The standard utility to look at waveform data is CUPLOT which is described in section  4.1. Waveforms have names that are up to 12 characters and begin with dd-, where dd is the diagnostic identifier, or d-, where d is a class of derived waveform. For instance, MB-IP-SL is generated by the MB (magnetics) diagnostic; it is the plasma current and is the slow sweep waveform (covering the duration of the plasma shot). When a waveform is written with the same name as an existing waveform, a new version of the waveform is created. The default for CUPLOT is to return the last entered and presumably best version. Each waveform can have a maximum of 2048 points and typically has 800 points. This is a table of some basic waveforms:

Waveform titles are text strings up to 72 characters long that provide more information about the waveform. To get a complete list of waveforms and their titles for a particular shot (remember not all waveforms are available for every shot), use CUPLOT . For example, to get a list for shot 76778:

    $ CUPLOT SH 76778 TT ,   Type all waveform names & titles
    $ CUPLOT SH 76778 TT MB  Type MB waveform names & titles

 
• Find Waveform Files