Data Output Manual

From Nuclear Physics Group Documentation Pages
Jump to navigationJump to search

Data Recording System

Data during a cooldown is recorded using a series of LabView modules run persistently on the central DAQ computer in Demeritt 103. General use of this module is beyond the scope of this particular chapter, which is instead intended to explain how to verify that data is being properly recorded during a cooldown and check this output for online and offline analysis.

Throughout the span of the primary LabView program, any variable of general interest to the group is combined with a variable name and appropriate unit and send to a SubVI whose only function is to save the data. While the main program is running, this SubVI should always be visible on one of the screens as a red square with a large green light and a box containing the time of the last recorded data point. If you cannot locate the SubVI, stop the main program with the shortcut CTRL+. and then restart it by pressing the white arrow on the toolbar near the upper left of the program. If the red VI is not visible, it is quite likely data is not being saved. Contact David Ruth or another LabView expert for help resolving this problem.

Once you have found the red VI, you can check it quickly throughout the duration of your shift to determine that data is properly saving. The large green light is known as the heartbeat, and should change from off to on or vice versa every 20 seconds. The light changing state indicates that the program has just saved a data point in the main data output. Further, the box below contains the date and time of the last saved data point. If everything is running correctly, any time you check, the date and time in the box should be within the last 20 seconds. If it is not, some data has been skipped, and it is necessary to immediately resolve the problem before trying to record more important cooldown data. If the light is not flashing, or the output box contains a time further in the past than a minute or so, shift workers should check the output file as detailed in the next session to see where the problem began.

Throughout the shift it is a good idea to visually confirm that the red VI contains the appropriate information every so often, especially before attempting to record any critical data!

Output File Format

Every time the data recording SubVI logs a point, it records it to a data file in the Box Sync folder on the lab computer. When there is a large volume of data being recorded, the actual files synced to the UNH Box system may be a bit behind if you attempt to check on another computer, but on the main DAQ Computer, you can find the data output files in the filepath Box Sync/UNH Poltarg/dnp_system/autolog/data_record_[date].csv, where [date] should be replaced by the current date. A new data file should be generated every day at midnight. This file can be opened in whatever program you like, but it is likely easiest to view it as a spreadsheet in LibreOffice Calc. Inside this file, every data point saved by the main program represents a single row, while every variable has its own column, with its name delineated at the top of the column. Opening the data file while the program is trying to save to it may cause issues, so I would recommend making a temporary copy if you want to inspect the current day's data. No matter how many variables are saved to the file, the final title column will be marked NMR data. Starting with that column, the next 401 title-less columns will comprise an average of those 20 seconds of NMR data.

To check that a specific variable is saving, simply find the most recent row and ensure that a number that makes sense is recorded in the last few entries of that specific column. To check the NMR data, you will need to take a few more steps. First, to find a specific type of NMR data, check the NMR Status column. While a baseline is being saved, this column will report a value of "Baseline". While TE averaging is being done, it will report a value of "TE". While a TE calibration has been selected, it will report a value of "Polarization". Based on the online calculated value of the Polarization column, you can find the maximum signal for a specific polarization run, by simply looking at the stretch where the status is "Polarization" and finding the highest signal Polarization value marked.

To reconstruct a plot of the signal for any given row, you will need the 401 NMR Data columns, as well as the Central Freq and Freq Span columns. A plot of the signal should be obtained using the 401 Data columns as the Y-Axis, where the X-Axis can be obtained by the frequency information. The X-Axis for a specific row is 401 frequency points where the minimum value is Central Freq - Freq Span, and each successive point steps up by 2*Freq Span/401 to a maximum value of Central Freq + Freq Span. In general the X Axis should not change unless you manually change the central frequency or span in the main LabView program, so you can use the same one for a whole polarization run under most circumstances.

You can plot this however you like, however, an easy way to get a plot is just to transpose the 401 data columns from a horizontal row to a vertical column using "Paste Special" features in a new LibreOffice file, and then, set up the column next to it as the X Axis with the values detailed above, and plot with LibreOffice's built in plot feature. It is not a bad idea before starting a polarization run to briefly check that your TE came out looking like you expect in the data file.

What To Do If It Isn't Saving

If the red SubVI isn't visible or isn't displaying the correct information, or if you find gaps or incorrect data in the data file while inspecting it, it is likely that there is a critical software issue that needs to be resolved. It is possible this issue could be resolved by a simple reboot of the DAQ computer, however, due to the large number of instruments that need to connect to the computer, it is possible this reboot could cause further issues when you attempt to restart the program. If you encounter these issues or are unable to determine why the data isn't saving properly, contact a LabView expert for help. If you are unable to immediately reach an expert, it is best if you periodically record what data you can see manually-- take screenshots of the main VI's front panel every so often and save them with the time-stamp in the title, and record in the logbook any critical or interesting numbers you see until the data recording issue is resolved.