Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200403/abbey.pdf Дата изменения: Wed May 14 13:20:05 2014 Дата индексирования: Wed Apr 13 13:11:37 2016 Кодировка: Поисковые слова: вторая космическая скорость

MOS Timing Mode Spectral Features and Possible Improvements
Tony Abbey, Leicester University

Calibration Ops Meeting Vilspa 23-24 Mar 2004

1


Overview of current concerns
· Timing mode has poor DNL because:
­ Timing mode sequences free run ­ Design decision was therefore not to use Gatti because communication with the EMCR control processor would interrupt the timing f lo w

· Next slide explains Gatti and DNL
Calibration Ops Meeting Vilspa 23-24 Mar 2004 2


DNL and Gatti
DNL ­ Differential Noise Linearity is caused by different digital values of the ADC having differing channel widths due to internal accuracy and noise. Gatti ­ sliding scale baseline offset technique which applies a precise analogue offset before ADC conversion and removes it digitally from the result. In the case of the MOS, this offset is changed every frame in imaging modes. Gatti causes the DNL to be smeared out over the width of the offset range (256 DN for MOS), and become invisible.

Calibration Ops Meeting Vilspa 23-24 Mar 2004

3


PSR0540 spectrum
PSR0540 re v 041 timing spe ctrum patte rn 0 e v e nts
80 70 60 50 40 30 20 10 0 200 300 400 500 600 700 800

Calibration Ops Meeting Vilspa 23-24 Mar 2004

4


3C273 spectra with/without DNL
3C273 r e v 38 2 timin g sp e c tr u m p at te r n 0 e v e n t s
250 200 150 100 50 0 200 300 400 500 600 700 800

3C273 re v 735 timing spe ctrum patte rn 0 e v e nts
250 200 150 100 50 0 200

300

400

500

600

700

800

3C 273 rev 382 S W spectrum pattern 0 events
200 150 100 50 0 200

300

400

500

600

700

800

Calibration Ops Meeting Vilspa 23-24 Mar 2004

5


Capella LW rev 0043 without/with Gatti

Calibration Ops Meeting Vilspa 23-24 Mar 2004

6


1700-37 rev 219 comparison
1700-37 re v 219 timing spe ctrum patte rn 0 e v e nts
1000 800 600 400 200 0 500 7 00 900 1 1 00 1300 1 5 00 17 0 0 1 9 00 21 0 0 2300 2 50 0

1700-37 re v 219 SW spe ctrum patte rn 0 e v e nts
400 350 300 250 200 150 100 50 0 500

700

900

1100

1300

1500

1700

1900

2100

2300

2500

Calibration Ops Meeting Vilspa 23-24 Mar 2004

7


Binning
170 7- 37 T im in g Mo d e MO S 1 02 19 B i n n ed X 16 wi th st art p o in t - 8DN 10 00 9 00 8 00

7 00

6 00 Coun ts

5 00

4 00 C ount s B i nned

3 00

2 00

1 00

0 500 700 900 1 100 1 300 1 500 C h a n n els 17 00 1 900 21 00 2 300 25 00

Calibration Ops Meeting Vilspa 23-24 Mar 2004

8


Binning2
H er X 1 T iming Mode MO S 1 02 07_0 1341 20101 Binne d X1 6 w it h s t art poin t -8 DN 25 00 20 00

15 00 Coun ts 10 00 C ount s B i nned 5 00 0 0 200 40 0 600 800 Cha nnels 100 0 12 00 1 400

Calibration Ops Meeting Vilspa 23-24 Mar 2004

9


Alternative ­ bring back Gatti
· Gatti DAC could be incremented by EMCR every timing frame · Slight asyncronism in setting and subtracting could result in a few events getting 1DN error (negligible effect compared with energy resolution) · Discuss with Saclay if software mod is needed
Calibration Ops Meeting Vilspa 23-24 Mar 2004 10


Timing diagram

Calibration Ops Meeting Vilspa 23-24 Mar 2004

11


Timing mode sequence in the EMCR/EDU
I n t e rf a c e : L B R C/ H S t a rt Observation L BR A/ C s t a rt s equenc e
EOI HEADER DATA EOL DATA EOL RECOGNIZED TIMING EVENTS

L BR A/ C

S DD

HB R

EOL

DATA EOL TIME

Y=1024

EOL DATA EOL

St o p Observation s to p s equenc e

EOL TIME

TRAILER

Calibration Ops Meeting Vilspa 23-24 Mar 2004

12