Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200711/mfreyberg_timejump.pdf Дата изменения: Tue May 13 17:50:45 2014 Дата индексирования: Wed Apr 13 13:37:13 2016 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п

SAS-7.1: status rep ort on EPIC-pn time jumps
Marcus Kirsch + Michael Freyb erg ESAC + MPE Garching

· History + theory · SAS-7.1: implementation and testing

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

1


EPIC-pn frames

· Intrinsic EPIC-pn time resolution: 40 ns (25 MHz oscillator) · PNAUX1 time resolution: 20.48 µs (512 в intrinsic) · Event time computation by OAL: · Do es not need frame time values of · Accessed CCF values: T = T0 + FTCOARSE + FTFINE в 20.48 µs + Tcal EPN TIMECORR ####.CCF[FRAMETIME]

EPN TIMECORR ####.CCF[TIMECORR]

· Event time: center of integration time of that CCD · Frames without events (after EPEA) are not transmitted ( gaps) · Frame times are not multiples of 512 oscillator clo cks: jitter by 1 FTFINE unit
Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

2


EPIC-pn time jumps
· normal reset after 32400s (but: telemetry outage ...) · Difference of 2 event times has to b e integer multiple of frame time · Deviations indicate p ossible time jumps

· = time difference in units of frame time · = deviation from integer numb er
Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

3


EPIC-pn frames

· Large intervals require high precision knowledge (and stability) of frame times · Only (but very !) relevant for time anomaly detection: · Event time difference divided by frame time should b e (close to) an integer · Most typical time anomaly: wrong increments of FTCOARSE: full seconds shift · Once event times are randomized: no way to detect residual time anomalies

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

4


EPIC-pn time jumps: SAS-7.1 testing: example FF mo de

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

5


EPIC-pn time jumps: SAS-7.1 testing: versions

Time jumps after SAS correction

40
FF EFF SW LW Timing Burst all

remaining time jumps [%]

30

20

10

0
oCCF+oalgo
Michael Freyb erg MPE Garching

2
nCCF+oalgo nCCF+19

4 Version
nCCF+20 nCCF+21 nCCF+22

6
nCCF+23 nCCF+24
6

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007


EPIC-pn time jumps: SAS-7.1 testing: status
56 / 2971 21 / 25 / 10 / 3/ 2/ 934 381 709 278 63 FF 1.88 %

EFF 2.25 % SW 6.56 % LW 1.41 % T B 1.08 % 3.17 %

117 / 5336

sum 2.19 %

0

200 400 600 800 1000 1200 1400 revolution

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

7


EPIC-pn time jumps: SAS-7.1 testing: summary
Mode FF EFF SW LW Ti Bu SUM oldCCF+oldalgo 5.8 20.8 37.2 39.0 12.0 52.9 15.6 newCCF+oldalgo 0.7 20.0 12.8 1.3 1.7 6.0 5.0 newCCF+19 29.0 58.33 59.09 28.57 18.18 25.00 35.81 newCCF+20 3.4 3.6 6.4 1.3 1.3 2.9 3.3

Mode FF EFF SW LW Ti Bu SUM

newCCF+21 1.9 2.8 8.0 1.6 2.4 4.6 2.5

newCCF+22 1.9 2.3 6.6 1.4 1.1 3.2 2.2

newCCF+23 newCCF+24 2.25 2.26 2.77 2.5 6.15 6.8 1.42 1.4 1.60 1.6 4.17 5.41 2.49 2.5

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

8


EPIC-pn frame times: 25 MHz oscillator

Total difference of the effective frequency of internal EPIC-pn oscillator (combined with external OBT oscillator effects) from the nominal value of fpn = 25 MHz as function of time and instrument mode. External events like eclipse seasons or RGS switch offs are indicated; dotted vertical lines show CDMU re-syncs. The long-term increase of the frequency is probably due to external temperature effects or to ageing the EPIC-pn oscillator. There is no apparent mode dependency. ftp://ftp.xray.mpe.mpg.de/people/mjf/public/mpe report 288 159.pdf
Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

9


EPIC-pn frame times: 25 MHz oscillator

Evolution of EPIC-pn quadrant box temperature for quadrant Q1 (which includes the target CCD) for FF mode, averaged over an exposure. The temperature sensor has only a coarse sampling (1 step = 0.88 C), for shorter exposures only one value during an exposure is realized. The ordinate has been intentionally inverted, higher temperatures are toward the bottom. When comparing with oscillator frequency plot a similarity in terms of general trend is suggestive: a decreasing quadrant box temperature increases the frequency of the oscillator (which is not temperature stabilised). ftp://ftp.xray.mpe.mpg.de/people/mjf/public/mpe report 288 159.pdf
Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

10


Current situation: SAS-7.1

· acceptance threshold (margin) of time jump vs. jitter: environment variable SAS JUMP TOLERANCE, in FTFINE units · iterating + testing + fine tuning: to make sure that all existent time jumps are detected and prop erly corrected and that no new time jumps are introduced by the software · Strictly sp eaking: frame times are not constant in [ms] but in [clo cks] · However, SAS assumes [ms], but absolute effect small (other effects ?)

Michael Freyb erg MPE Garching

XMM-Newton EPIC Calibration Meeting, Mallorca, 06/07 Nov 2007

11