Bob sent this slack on may 14th
Quick note Luisa - I caught an issue in the culling code which may or may not have been included in the SDC version. In short, in the event filter (the goodEventMet() function) does not include a filter for the event time. Apparently, there are events with good ebins, quality flags, etc. that have a fill value for the 'event_times' data item. I'm not sure how this happens, but clearly it does. This does not have an impact on the actual cull - since times that fall outside the pointing are not used in the cull. HOWEVER, it does impact the results for U45, since it uses the mean of the filtered times with the SPICE kernel as part of the Earth event removal. It results in a SPICE error and basically zeros out the entire pointing. I'm working on some simple fixes now.
Bob sent this slack on may 14th
Quick note Luisa - I caught an issue in the culling code which may or may not have been included in the SDC version. In short, in the event filter (the goodEventMet() function) does not include a filter for the event time. Apparently, there are events with good ebins, quality flags, etc. that have a fill value for the 'event_times' data item. I'm not sure how this happens, but clearly it does. This does not have an impact on the actual cull - since times that fall outside the pointing are not used in the cull. HOWEVER, it does impact the results for U45, since it uses the mean of the filtered times with the SPICE kernel as part of the Earth event removal. It results in a SPICE error and basically zeros out the entire pointing. I'm working on some simple fixes now.