Hi Andy,
Good to see you got the program running, at least on one file. Interesting that, for the top and bottom detectors, the measured standard deviations are lower than for the theoretical ones derived from the Poisson distribution with the same mean. The measured histograms are very uneven to the eye too. It looks as if some bins are “stealing” counts from other bins – it’s conceivable that this might be related to the ADC “glitch”, but more likely something to do with “dead times” or other software/timing issues. The coincidence data is more as expected, but the measurements on this seem to be biased a bit towards the low value side – (abnormal skew??)
BTW, when I started this software development, I discovered I had not put an SD card into my own PicoMuon – I must have decided to trust to the attached RPi for all data recording! There is an SD card in it now, following my swapping of the two scintillator units. I intend to extend the program to read RPi files. I see there are no headers in the RPi files (tut-tut!). I’ll have a look at my own SD files soon.
The size limit of files the program can handle is dependent on a) how much memory your computer has b) maximum size that the Java virtual machine is told to allocate. The latter is configurable up to the limit of the computer (I think). How big is your 6-day file? Did you get an “out of memory” error when you tried to load it? I don’t have a “progress bar” to show the file loading state – I shall need to think about doing that for long files as large files will take time to load.
The reason for the “Didn’t find UKRAA: tag” error is almost certainly to do with splitting your 6-day file into pieces. The first piece will have the header information, but subsequent ones won’t. The header information is required for “sanity checking” and to provide information that I feel it is useful to have on the upper panel. One way around this is to have a facility which will split the larger file into smaller pieces each with a copy of the header information.
There are questions about handling large data sets – e.g. does it make sense to load several days of data in, given that we are looking for changes over shorter periods (I suppose). Also, it might be handy to select a subset of data from files. I am giving these issues some thought.
Cheers, John