search

cta-observatory / dragonboard_testbench

Collection of programs to look at test data from the LST prototype camera.
1 stars 2 forks source link

Spike Treatment #46

Open MHoerbe opened 8 years ago

MHoerbe commented 8 years ago

Got some first spike data. Spikes were collected for sample_ids [5:36].

The spikes have been divided into (all) spikes, single, double and triple-or-higher spikes. Despite some interesting correlations it seems like the spikes' width is the most considerable criterion.

Furthermore, the cell_id plots could be subdivided into each pixel-channel tuple in the future. Yet, they show correlating behaviour even without that split.

hist_adc 20 hist_adc 20_double hist_adc 20_single hist_adc 20_triple_or_higher hist_adc 40 hist_cell_id_adc 20 hist_cell_id_adc 20_double hist_cell_id_adc 20_triple_or_higher hist_pixel_adc 20 hist_pixel_adc 20_double hist_pixel_adc 20_single hist_pixel_adc 20_triple_or_higher hist_sample_id_adc 20 hist_sample_id_adc 20_double hist_sample_id_adc 20_triple_or_higher

MHoerbe commented 8 years ago

Further analysis of spike properties and distribution reveals some dependence on the length of the spikes

hist10_adc hist10_delta_t hist10_pixel hist10_sample_id hist10_sample_id_1of2 hist10_sample_id_2of2

MHoerbe commented 8 years ago

For the record: the adc count of the first histogram (1-spikes) is scaled like (20k, 40k, ... 120k)

maxnoe commented 8 years ago

Just a small remark. I do not think we can call 10-spikes spikes anymore. But very interesting indeed.

MHoerbe commented 8 years ago

Here the same plots based on sample_ids 10 to 35 (where the 'pattern' should not be visible anymore). hist10_sample1036_adc hist10_sample1036_cell_id_1of2 hist10_sample1036_cell_id_2of2 hist10_sample1036_delta_t hist10_sample1036_pixel hist10_sample1036_sample_id