ASTROSAT CZTI

Data Quality Report

Orbit 20220505_A09_142T02_9000005112_level2

Creation date: 2022-05-11T12:36:15.843274
Switch to: Orbit-wise | Merged OBSID-wise
web.py, SVN revision $Rev$

Source files: Data quality report (detailed, human readable) | Data quality report (concise, machine readable)

Observation information


Basic statistics of files

Param Original file Final file
FilenamemodeM0/AS1A09_142T02_9000005112cztM0_level2_bc.evt modeM0/AS1A09_142T02_9000005112cztM0_level2_quad_clean.evt
Size (bytes)784,578,240294,024,960
Size748.2 MB 280.4 MB
Events in quadrant A4,652,9971,936,367
Events in quadrant B6,226,9081,910,967
Events in quadrant C5,268,6711,910,108
Events in quadrant D6,283,1011,809,376

Level1 data integrity


Mode SS
QuadrantBADHDUFLAGTotal packetsDiscarded packets
A03980
B03980
C03980
D03980

Mode M0
QuadrantBADHDUFLAGTotal packetsDiscarded packets
A0405270
B0478470
C0416160
D0463660

Mode M9
QuadrantBADHDUFLAGTotal packetsDiscarded packets
A0980
B0980
C0980
D0980

Data saturation


QuadrantTotal secondsSaturated secondsSaturation percentage
A19441230.118307%
B194426713.451291%
C194421710.879539%
D194425102.623187%

Noise dominated fraction

Noise dominated data is calculated using 1-second bins in cleaned event files. If a bin has >2000 counts, and if more than 50% of those come from <1% of pixels, then it is considered to be noise-dominated and hence unusable.

Quadrant # 1 sec bins Bins with >0 counts Bins with >2000 counts High rate bins dominated by noise Noise dominated (total time) Noise dominated (detector-on time) Marked lightcurve
A 29212 19447 0 0 0.00% 0.00%
B 29212 19441 32 32 0.11% 0.16%
C 29212 19442 8 8 0.03% 0.04%
D 29212 19443 256 256 0.88% 1.32%

Top noisy pixels

Top three noisy pixels from each quadrant. If the there are fewer than three noisy pixels in the level2.evt file, extra rows are filled as -1

Pixel properties Quadrant properties
Quadrant DetID PixID Counts Sigma Mean Median Sigma
A 10 83 427098 1812.56 1074 1059 235.0
A 13 6 41478 171.96 1074 1059 235.0
A 8 6 18015 72.14 1074 1059 235.0
B 5 172 1533925 7446.01 1056 1035 205.9
B 5 255 342080 1656.63 1056 1035 205.9
B 3 112 90387 434.03 1056 1035 205.9
C 15 214 765570 3145.4 1051 1044 243.1
C 0 10 343563 1409.19 1051 1044 243.1
C 1 23 21571 84.45 1051 1044 243.1
D 8 230 607933 2394.22 1053 1023 253.5
D 12 6 392662 1544.99 1053 1023 253.5
D 13 249 317321 1247.77 1053 1023 253.5

Detector plane histogram (DPH)

Histogram calculated using DETX and DETY for each event in the final _common_clean file

Quadrant A Quadrant B
Quadrant D Quadrant C

Count rate plots and images

Plot type Count rate plots Images
Comparison with Poisson distribution
Blue bars denote a histogram of data divided into 1 sec bins.
Red curve is a Poisson curve with rate = median count rate of data.
Quadrant-wise count rates
Data is divided into 100 sec bins
Module-wise count rates for Quadrant A
Data is divided into 100 sec bins
Module-wise count rates for Quadrant B
Data is divided into 100 sec bins
Module-wise count rates for Quadrant C
Data is divided into 100 sec bins
Module-wise count rates for Quadrant D
Data is divided into 100 sec bins

Housekeeping plots

Parameter Plot
CZT HV Monitor
D_VDD
Temperature 1
Veto HV Monitor
Veto LLD
Alpha Counter
_CPM_Rate
CZT Counter
+2.5 Volts monitor
+5 Volts monitor
_ROLL_ROT
_Roll_DEC
_Roll_RA
Veto Counter