|
Size: 2015
Comment:
|
Size: 3591
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 8: | Line 8: |
| associated with {{{TIMESRC}}} equal 2 and {{{DATASRC}}} equal 3, 5 or 6. | |
| Line 11: | Line 13: |
| 4. '''ENV mode events''': events with {{{ENVSRC}}} with non-NULL values. | 4. '''ENV mode events''': events with {{{ENV}}} with non-NULL values. |
| Line 31: | Line 33: |
=== Statistics for two TMs operating at a given eROday === The following histogram shows the (mean) percentage of removed data in each TM when only two cameras are in operation at the time. The mean percentage is obtained for the eROdays where the TMs were functioning. Dark colours show CXFWPOS=8, i.e. with optical filter applied, while light colours show CXFWPOS=13, i.e. with calibration source. {{attachment:twoTMs_inoperation_FWPOS8_13.png|Histogram two telescopes | width=500}} It might be that TM6 shows almost double of removed/corrupted data than before since there was a period (~6 eROdays) where TM6 suffered and enhancement of ... (TBD with Norbert). The next histogram shows the same information as above, except that the bars show the percentage of each kind of removed/corrupted data: {{attachment:twoTMs_inoperation_FWPOS8_13_indv.png|Histogram single telescope | width=500}} Again, most of the removed data has `SUBSEC` values above 95 or are duplicates. === Remarks === * Maybe future patches to the CE software will avoid these events being transmitted. * There are no statistics, but many of these removed/corrupted events might fall into different categories. The percentages presented in the above histograms depend on the order in which `evprep` removes them. * The filtering all but NULL-valued ENV is temporary. Eventually, we expect genuine events with non-NULL env, indeed Konrad has requested a test of this mode soon. |
Removed events
eSASS task EVPREP currently removes these events. Future eSASS version will only flag them and exclude them from the quality good time intervals (QUALGTI).
Duplicated events: events that are received with identical frame time and have the same RAWX and RAWY coordinates.
95 subsec events: events with SUBSEC values above 95.
associated with TIMESRC equal 2 and DATASRC equal 3, 5 or 6.
Data/Time events: events with DATASRC or TIMESRC not equal zero.
ENV mode events: events with ENV with non-NULL values.
Undefined PHA events: events with PHA undefined (NULL).
General statistics of removed events
69 eROdays have been processed with evprep. These eROdays lie within the 43068-43193 range (~26.08.19-16.09.19). The eROdays are chosen based on the CXFWPOS value in their respective HK-2 file: FWPOS=13 or FWPOS=8, for the whole duration of the eROday. eROdays with mixed CXFWPOS values are discarded for the moment.
Statistics for a single TM operating at a given eROday
The following histogram shows the (mean) percentage of removed data in each TM when only one camera is in operation at the time. The mean percentage is obtained for the eROdays where a given TM was functioning. Dark colours show CXFWPOS=8, i.e. with optical filter applied, while light colours show CXFWPOS=13, i.e. with calibration source.
TM2, TM4 have not operated alone so far. And TM5 did not operate alone with the filter.
The next histogram shows the same information as above, except that the bars show the percentage of each kind of removed/corrupted data:
Indeed, most of the removed data has SUBSEC values above 95 or are duplicates.
Statistics for two TMs operating at a given eROday
The following histogram shows the (mean) percentage of removed data in each TM when only two cameras are in operation at the time. The mean percentage is obtained for the eROdays where the TMs were functioning. Dark colours show CXFWPOS=8, i.e. with optical filter applied, while light colours show CXFWPOS=13, i.e. with calibration source.
It might be that TM6 shows almost double of removed/corrupted data than before since there was a period (~6 eROdays) where TM6 suffered and enhancement of ... (TBD with Norbert).
The next histogram shows the same information as above, except that the bars show the percentage of each kind of removed/corrupted data:
Again, most of the removed data has SUBSEC values above 95 or are duplicates.
Remarks
- Maybe future patches to the CE software will avoid these events being transmitted.
There are no statistics, but many of these removed/corrupted events might fall into different categories. The percentages presented in the above histograms depend on the order in which evprep removes them.
- The filtering all but NULL-valued ENV is temporary. Eventually, we expect genuine events with non-NULL env, indeed Konrad has requested a test of this mode soon.
Individual camera analysis