F OS FIL TERGRA TING WHEEL REPEA T ABILITY MEASUREMENTS - PDF document

F OS FIL TER�GRA TING WHEEL REPEA T ABILITY MEASUREMENTS ANURADHA K ORA TKAR and STEPHAN MAR TIN Sp ac e T elesc op e Scienc e Institute� ���� San Martin Drive� Baltimor e MD ����� Abstract� The p ost�COST AR F OS Filter�Grating Wheel �F GW� rep eatabilit y w as tested for b oth the red and blue detectors� The test consisted of obtaining rep eated F OS images in an unin terrupted sequence through � di�eren t gratings on eac h detector� The observ ations consisted of exp osures using G���H� G���H and G���H with F OS�RD� and G���H� G���H and G���H with F OS�BL� All other observing parameters w ere unc hanged during the test� The standard deviations of the p ositions of sp ectra with resp ect to eac h other are found to b e indep enden t of the c hosen grating� The rms deviations of the p osition of the sp ectra in the F OS X and Y directions are of the order of ��� dio de and � ybases� resp ectiv ely � Deviations as large as ���� dio des and �� ybases are also seen� These v alues are similar to those obtained from pre�launc h calibration activities� The non�rep eatabilit y of the F GW could b e a ma jor source of instrumen tal inaccuracy in the w a v elength calibration of sp ectra obtained without con temp oraneous calibration lamp sp ectra� It should b e noted that the absolute w a v elength calibration for a t ypical F OS sp ectrum is not only a�ected b y in ternal instrumen tal e�ects� suc h as� F GW p osition and ap erture lo cation� but is also a�ected b y the target acquisition accuracy � F OS � W a v elength Calibration Key w ords� �� In tro duction Pre��igh t measuremen ts of the F OS �lter�grating wheel �F GW� rep eatabilit y sho w ed that the F GW rep eatabilit y w as of the order of �� microns ����� ybase units in Y and ��� dio des in X� in b oth the X and Y directions of the F OS blue detector� and of the order of �� microns ����� ybase units in Y and ��� dio des in X� in b oth the X and Y directions of the F OS red detector �CAL�F OS���� CAL�F OS���� CAL�F OS����� This non�rep eatabilit y of the F GW leads to w a v elength calibration errors� Unin terrupted rep eated observ ations with di�eren t gratings w ere obtained in the �rst part of the calibration program ���� to c hec k the F OS F GW rep eatabilit y � The red side detector measures used the G���H� G���H and G���H gratings while the blue side detector measures used the G���H� G���H and G���H gratings� Observ ations for eac h grating are listed in T ables I I�VI I� Eac h observ ation w as in the F OS IMA GE mo de� and sp ectra of the in ternal Pt�Cr�Ne w a v elength calibration lamp using the ��� ap erture w ere obtained in a con tin uous random sequence of appro ximately �� �� exp osures in eac h grating� The F GW w as the only instrumen t comp onen t that w as mo v ed during the en tire observing run for eac h detector whic h lasted �� min utes for the red side detector� and �� min utes for the blue side detector� The exp osure times for eac h single exp osure are giv en in T able I� Other exp osure details suc h as n um b er of dio des used for eac h grating� and the w a v elength range co v ered are also giv en in T able I�

A� K ORA TKAR AND S� MAR TIN T ABLE I Exp osure Characteristics for eac h Grating Con�g� Grating Exp osure Dio de W a v elength Num b er of Emission � Time �s� Range Range � A� Lines Used F OS�RD G���H ��� ��� � ��� ���� � ���� �� F OS�RD G���H �� ��� � ��� ���� � ���� �� F OS�RD G���H �� ��� � ��� ���� � ���� �� F OS�BL G���H �� ��� � ��� ���� � ���� �� F OS�BL G���H �� ��� � ��� ���� � ���� �� F OS�BL G���H � ��� � ��� ���� � ���� �� �� Data Analysis Since eac h IMA GE mo de observ ation used ��� dio des� with NXSTEPS��� O VER� SCAN��� YSTEPS���� and YSP A CE��� there w ere �� sp ectra with ��� pixels eac h� stored as �� groups in the original data �le� The �� sp ectra w ere then con v erted in to t w o�dimensional images with ��� � �� pixels eac h using the STSD AS task ok � r apid lo F or eac h grating the �rst image w as c hosen as the reference image� The ��� � �� pixel images w ere then cross�correlated t w o�dimensionally with the reference frame for eac h grating� The STSD AS task is used to determine the cross�correlation cr ossc or �see Bracew ell ���� for details�� Since the cen tering option is used in the cr ossc or task� the maxim um p o w er is exp ected at the ����� ��� pixel p osition in the cross� correlated image for this analysis� The shift of an image with resp ect to the reference image is th us the p osition of maxim um p o w er in the cross�correlated image� The cross� correlated images are of high signal�to�noise� since man y �see T able I� emission lines with signi�cance lev els � � ab o v e the bac kground are used in the cross�correlation � analysis� The p osition of maxim um p o w er in the cross�correlated image is giv en in T ables I I�VI I �columns � and ��� The o�sets with resp ect to the reference cen tral pixel �columns � and �� are giv en also in the tables� The measuring uncertain t y of the cross correlation p eak is ���� pixels in X �where � pixels � � dio de� and � ���� � pixels in Y �where �� ybases � � pixel�� The standard deviations �in units of pixels� for eac h grating are giv en in the tables� �� Results Only the deviations from the reference pixel are imp ortan t for determining the F GW rep eatabilit y � b ecause the absolute v alue of the reference pixel is arbitrary � These deviations are of the same order for all gratings� The � � rms deviations of the p osition of the sp ectra are ���� pixels in the F OS Calibr ating HST� Post Servicing Mission