Home

VIMOS User Manual

image

Contents

1. 2d09S9 9 W o qe rv juo SIosn poinqrnsrp 4 91 aLS SOW ALS SON SOWIA joudoedg eo sourSO NIA Sxseur jroads o poou mis 8 oujoeurojoud jo 1 SON syseu arivo ALS SOW ALS SOW SOINIA p epue s eo sour SOWIA Ayroods ou Jejs prepuejs onoods jo 54 GOIN e 91 IWS SON se owes qmeosiN SONIA soun amsodxo jo uoneoyroads ou SOJA Sore pue sy poyoeny arivo 15 MAT ALS NAT SOWIA 10ydoads e9 mir SONIA orneurojoud gt 48 jo uonvA1osqo NAI gravo aus NAI se owes eo IC SOINIA sown ounsodxo jo uoreoy roads ou NAI sore pue seg poyoeny e ny SOINIA Surge ou ye adoa poy wey 19311199 HT wonsy eo Sur SOWIA DIOU 21 JO SUOTEAJoSqO SuIseuly arivo 15 ALS DINI SOWIA WOJOY Te SW SOWIA sprepurys oLnewojoyd jo Surge AIS AMS OWI SOWIA Teo SUIT SOINIA sprey Adwo 3unrod site Surseuy SNOLIVA IHSHO NOLLV ISI IVO HINLLHOIN alIvo eums SON je uoouog pur ory se que Arg eo sour SOINIA eun amsodxa jroads sore 199198 SON arivo dINV TAVAM SOW HAVM SOW SONWIA ory eo sour SOWA eun eunsodxo Ay1oads sore U9919S
2. 34 Characteristics of the VIMOS filters 123 Basic characteristics of the old VIMOS CCDs 135 Detector Readout Modes o deb da ee Rud a a 135 Old detectors read out noise and conversion factors 136 xvi VIMOS User Manual VLT MAN ESO 14610 3509 Chapter 1 Introduction 1 1 Scope The VIMOS User s Manual is intended to provide information on the following topics e Overall description of the instrument Sect 2 e Observing with VIMOS Sect 3 e Observing and calibration templates Sect 4 1 2 More Information on VIMOS All VIMOS Manuals are available on the VIMOS instrument WEB pages together with the most updated informations on the instrument http www eso org sci facilities paranal instruments vimos Information and software tools for the preparation of service and visitor mode observations with VIMOS are available at http www eso org sci observing phase2 http www eso org sci observing phase2 SMGuidelines html In particular information concerning mask preparation with VMMPS and guide star selection with Guide cam is given in the following additional pages http www eso org sci observing phase2 S MGuidelines VMMPS VIMOS bm http www eso org sci observing phase2 S MGuidelines Guidecam VIMOS VIMOS html Visiting astronomers will find further instructions on the Pa
3. 1 bc rere 0 8 0 6 0 4 0 2 La Es 550 600 500 Wi nm 550 Wi nm 600 Figure A 1 U B V filter transmission curves 125 T T I T 1 tE S a vw d as E ES E vw dE 4 2 n 4 C a 5 T 1 T T z tE ca eil 96 0 4 4 0 2 1 1 6 Wavelength nm t Norm T Norm T NN UN HT 4 EpL IP en A OA O dy KE EIE L 1 800 900 1000 800 900 1000 Wavelength nm Figure A 2 R Lz filter transmission curves 126 Norm T A E SD AS ESEJE MIRI REDS Norm T TTTTTTT RES 400 500 600 700 400 500 600 700 Wl nm Wi nm Norm T Norm T Libani pura dii AAA E 500 600 700 800 900 100800 600 700 800 900 1000 Wi nm Wi nm Figure A 3 OS blue OS red transmission curves it 600 800 Wi nm 1 1000 Norm T o TT TT AA 1 1 500 600 700 Figure A 4 GG435 GG475 transmission curves 800 500 Wavelength nm 1 700 127 128 Appendix B VIMOS Grisms This appendix contains the efficiency curves of all standard grisms available for VIMOS and the approximate wavelength range for a slit which is located in the field centre Whe
4. ci ee e e A A ee a Gas 19 2 Detector syste ss usos os ee E EE Ed how mox A et 4 19 253 More ls 52 222 ee eS ee Gee ew 21 26 Calibration Uas ams sas D doe e aaa upon a Sada ee 21 2 7 AFC Active Flexure Compensation 22 2 8 Known shortcomings of 22 2 8 1 masks vignetting in IMG and MOS 22 26 2 IBUTADEME 222222252 958021 amp xo AA X GAS 22 2 8 3 SOLVED Non optimal focus for some 22 2 8 4 SOLVED Defocusing at some adapter angles 23 2 8 5 SOLVED IFU focusing issues 23 200 Spectrom EE cama ae a ra a 23 2 8 7 Temperature dependent bias level lt os ea ccosa ca coro ceea to er ta ea 24 2 8 8 Temperature dependent image scale 2 2 22 2 22 2 2 4 2 4 4 24 Observing with VIMOS 25 34 Interactions With the telescope lt c lt s sa ga gacama e048 Aa 25 3 1 1 Guide stars and Guidecam software tool a 25 212 ego seno x oboe x Roy A 26 3 13 Telescope and Instrument Focus 26 3 2 Atmosphene Dispersion 2222224 ROG WX E Bard a Sea wade 26 5 9 Prngmgand Hexut s o 222 EGRE GS Va E Yom a d 26 EP maeme IMG Lr VP T 27 3sbl P
5. You will be asked to confirm the ID for this object taken as the highest current ID 1 92 VIMOS User Manual VLT MAN ESO 14610 3509 Confirm ALPHA VIMOS 08 00 04 627 DELTA VIMOS 10 42 56 76 ALPHA SKV DELTA SKY XIMAGE 797 8 7 vIMAGE 1600 7 0 50 0 5 595 5 4 gt AIMAGE O BIMAGE O THETA_IMAGE O Ki 2 The new object appears at the bottom of your object list the catalog window Note that the task is called New Ref because this is the only case we consider useful but in principle once the object is in the catalog you could change its flag to any allowed value Compulsory objects You can flag a few objects as being the top priority These objects will be allocated slits immediately after reference objects have been defined on the mask i e VMMPS will make sure that they get slits allocated if not blocked by a reference object If two compulsory object block each other VMMPS will allocate one in this mask and the other in the next mask of the series if applicable To flag an object as compulsory select the button Comp after having clicked on the object in the image display The FLAG of the object will be set to Compulsory in the catalog If you added a column FLAG in your input catalog this flag can be used to set high priorities Sometimes a large number of C compulsory targets in the input catalogue results in crashing VM
6. ge 0 6 G 1 1 8 WI m crons Figure B 3 Efficiency curve of the high resolution grisms HR blue FW HR _blue92P The vertical lines mark the desired blaze and the central wavelengths The spectral range refers to the slit in the center of the field 133 160 jp AWMSHU HRA nita 80 zl Desired Bleze a e EN 2 Min Mee Peak GM E no 40r j P Centre Y 20 19 d 4 AA aa 03 0 0 8 1 158 WI mizrons Figure B 4 Efficiency curve of the high resolution grism HR_orange The vertical lines mark the desired blaze and the central wavelengths The spectral range refers to the slit in the center of the field The cutoff wavelength is in most cases given by the order separation filters the red CCD limit or the limit of the VIMOS optics in the blue VIMOS Efficiency Wavelenght nm Figure B 5 Efficiency curve of the high resolution grisms HR red VPHG The spectral range and efficiency refers to the slit in the center of the field The cutoff wavelength is in most cases given by the order separation filters the red CCD limit or the limit of the VIMOS optics in the blue 134 Appendix C Old detector system until May 2010 The old detector mosaic of VIMOS consists of four 2kx4k EEV 44 82 CCDs thinned and back side illumi nated single layer coating It will be in use until May 2010 if the VIMOS detector upgr
7. 4 LAIA Version estas a E SO Ue Ee eee e 4 14 14 Version 76 1 a a sa alas a ea Sh se ee xm m aa a wad ae ad 4 1 3 Acknowledgments 0 046 626 09 a eed ee ee 4 2 Instrument Characteristics 5 ZA 25 dade de doe HEAR 3 211 Istrumentmodes oa a a 9e om o wo Roo 7 ES A A es ee we Ge ge a a 7 2 2 1 IMG mode orientation of the field of view 8 222 mage MATAN oa dsd a al dea 8 223 maging Piers o sc s eb a 8 224 LETO POS iaa a a do o EE eh 8 2 3 Multi Object Spectroscopy 12 Ge WEE EENHEETEN 12 2 32 Special resolution us luos ais eR ad Roa eas mo ROAD a 12 20 3 Multiples characteristics ds usu doe ea Sh be ara dx Sa ea ee 12 2 3 4 MOS mode orientation of the field of view 13 24 Integral Field Unit cas sala o a A AA 13 vil viii VIMOS User Manual VLT MAN ESO 14610 3509 2 4 1 mode orientation of the field 14 2 4 2 Fiber arrangement and spatial characteristics 14 7 459 Special NRG at ve E A A AAA Uum 17 2AA Spectral f solOHOH uunc e a A a o 17 205 Multiples Characienshos zs som uox aa a SA SG eae 17 23
8. Another consideration is the ratio of the total MOS integration time to the pre image integration time Observations where one pre image can provide masks for multiple MOS follow ups do not pay much penalty in pre image integration time and may in some circumstances benefit from the extra check of a pre image On the other hand an observation that will consist of just one relatively short MOS follow up for which the pre image is in any case unlikely to reveal faint sources would be well served by the PILMOS option 7 24 PILMOS features to be aware of e PILMOS does not work if the reference objects have entries in the ID column that are not integers and or if their is another format inconsistency e g in a number of rows some of the fields have merged It is the user s responsibility to check that their catalogs are properly formatted after starting PILMOS and entering the pointing coordinates into it no image is displayed and a dialog box pops up with the error message Can t find host IP address for wvgarc Normally it can be solved by restarting the tool This is not related to the PILMOS functionality but rather an issue of the regular Guidecam tool as well e if the magnitude column containing the string mag MAG as column description is the last column of the catalog and is the only magnitude column PILMOS does not display the targets in the fake PILMOS finding charts However note that the VIMOS catalogue gener
9. VIMOS instrument components 6 WEBpage 1 VIMOS Upgrade Project new CCD 6 old CCD 5 visiting astronomers general informations 1 observing with VIMOS 25 WEB page 1 zero points 8
10. scope Offset Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IMG Instrument Mode SEQ HEADER no F Header type This template is for acquisition of photometric data zero points in Imaging mode The acquisition template to use with this template is VIMOS_img_acq_Preset The coordinates to define in the acquisition template are those of the standard star field Offsets from these coordinates to position the star field in one or the four quadrants of the instrument will be applied automatically by the template The parameters of this template are Exposure time seconds This is the integration time in seconds per exposure Filter The available filters are U B V R L z To produce a valid calibration OB the filter in the first observation template following the acquisition template should be the same as the one in the acquisition template Number of exposures per telescope offset This parameter defines how many ex posures are taken at each offset position 1 or 4 1 or 4 quadrants This parameter allows to define if the standard star image will be taken in 1 or 4 quadrants If the 1 quadrant option is selected the standard star field will be positioned in only one quadrant by applying an appropriate offset from the coordinates of the star field defined in the target pa
11. 3h 45 lt 6 lt 5 e PA 90 N S slits Hour Angle within 2h any observable These restrictions could be lifted with a waiver but it should be very well justified Notice that the default value for pre imaging is set to PA 90 All service and visitor mode users who have particular constraints for the orientation of their fields should request a waiver 3 5 4 VIMOS Mask Preparation Software VMMPS The use of VMMPS Chapter 6 is compulsory for the creation of the so called ADP Aperture Definition in Pixels files which are subsequently loaded into p2pp when preparing Observation Blocks ADP files will be used in Paranal to manufacture the masks ADP files should never be edited by hand Here are some general features of VMMPS for a quick reference guide Consult the VMMPS web page for instructions on how to retrieve the package and the documentation e VMMPS allows to define a mask in two situations From a contributed catalogue consisting of list of targets with accurate relative astrometry better than 0 3 rms Input catalog is a list of RA DEC coordinates This catalogue needs to have with the same relative astrometric accuracy coordinates of bright objects stars in the field of view These objects stars after detection on the pre images will be used to register the target coordinates to the VIMOS coordinate reference frame Some of these bright stars will be used as reference for acquisition VIMOS
12. Step 0 most crucial Prepare you catalog s and image s You have prepared carefully your catalog s and have the corresponding VIMOS pre images ready You thought carefully of the objects in your catalog only objects you do not mind having a slit allocated to them are in there VIMOS pre image and contributed catalogs have common objects you did not modify your VIMOS pre image Step 1 Chose the right task to start with According to your input catalog you go to Step 2 if you have a VIMOS pre image and contributed catalog to Step 3 1f you have a VIMOS catalog Step 2 From VIMOS pre image and contributed catalog create a VIMOS catalog Skip this step if you have already a VIMOS catalog Step 3 Starting VMMPS for mask definition Step 4 optional re arranging your VIMOS catalog Step 5 First pass defining special objects Have you been careful and defined at least 3 reference objects per quadrant Step 6 The automatic slit assignment your first ADP draft Spend some time thinking of your parameters Tip select only the wavelength range you really need O Step 7 optional Fine tuning last corrections Only go through this step if you are sure to know what you are doing Step 8 Writing your final ADP And remember never to edit this file by hand Make finding charts after pressing Plot Slit button
13. The user can specify the desired exposure level Offered only in visitor mode The parameters of this template are e Exposure Level keep constant Filter The available filters are U B V R L z ADU This is the desired exposure level which the template will attempt to e Number of exposures This parameter defines how many exposures are taken The telescope should be offseted between exposures List of offsets These parameters are used to specify offsets in aresecond in SKY coordinates The offsets should be large enough so that halos of bright stars do not affect the data we attempt to use only empty fields buth the VIMOS FoV is large so sometime this is not possible The offsets should be at least 15 in size e Rotator angle The rotator angle at which the frames will be obtained For normal use this is the angle on the sky e Rotator angle mode Whether the rotator angle refers to angles on sky or to absolute adapter rotator value These last option is for tests only and should not be used by visitors 58 VIMOS User Manual VLT MAN ESO 14610 3509 4 7 2 VIMOS img cal Photom VIMOS img cal Photom tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no UB VRIzFree NODEFAULT Filter SEQ CAL FLAG no 14 1 1 or 4 quadrants SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele
14. bottom panels Note that position angle increases as rotator angle Figure 2 4 shows the instrument FoV keeping fixed the detectors with x axes increasing to the right and y axes increasing to the top Top right panels refer to the situation with the rotator at 0 and bottom right panels refer to the situation with rotator at 90 The corresponding positions of the detectors on sky are shown in the left panels The gap between the quadrants is approximately 2 The pointing coordinates correspond approximately to the center of the gap 2 2 2 Image Quality For the most updated information please have a look at the Quality Control and Data Flow Operations Web pages The focal plane of VIMOS is not strictly flat because of the complexity of the optical elements involved This results in slight focus variations across the field which translates in a variable PSF from the center to the periphery The sky concentration effects are in any case small and negligible for flat field and photometric calibrations 2 2 3 Imaging Filters VIMOS provides positions for 10 broadband filters in any of the four channels Presently installed filters inside the instruments are U Bessel BVRI Gunn 2 and some order separation filters Only the U BVRIz filters are available for imaging programmes The special U is not a standard Bessel filter see Figure A 1 With the Bessel J and Gunn 2 filters the CCDs show no fringes The complete list of filters
15. somewhat To do so you must have the column s on which you want to constrain your objects in the catalog and you must have made up your mind which range you want to allow for this columns If this is fulfilled go in the catalog window to the pull down menu Options and select Set Search Columns The following window will pop up Search Columns for diska visitor VMMPS_tests fornax_contributed_ Column name x RADIUS Label for min value Min X RADIUS Label for max value Max X RADIUS At this point you should chose the column s on which you want to select your catalog Click on the col umn name in the right sub panel then click on the left arrow in the central bar the selected column name s should appear in the left sub panel Ignore the lower part of that window you can use it for indicative pur poses but you are not expected to fill in any values at this point Once all the columns you wish to select on are in the left sub panel click OK The pop up window will disappear and your catalog window will get updated as shown in the example below 1 E Edit Options Search Options o 1 03 38 16 8885 353710685 Min Radius 00 777 Max Radius 5 45 Mn R MAG R_MAG 22 5 Max Objects 10000 Select Area Set From Image ALPHA_VIMOS DELTA_VIMOS X_IMAGE CT COLOR MAG FLAG 91300 35 37 02 82000 00137 97800 35 36 19 13000
16. the first set of offsets should be 0 0 If there are less offset values than the defined number of offset positions the template will return at the beginning of the list of offsets and apply them again IMPORTANT We stress that in contrast with other ESO instruments the offsets defined by the user in P2PP refer to the apparent movement of the target on the detector independently of the choice of Offset coordinates SKY or DETECTOR and not to the offsets applied to the telecope For example if 5 telescopes offsets are defined number of telescope offsets together with the follow ing series of offsets apparent movement of the target on the detector RA or X 0 10 5 DECorY 2 7 the offsets applied to the telescope will be RAorX 0 10 5 0 10 DECorY 2 7 2 7 2 If SKY option for coordinates is selected then after the preset the first offset will be applied to place the target 2 arcsec North with respect to its previous position on the detector by moving the telescope 2 arcsec to the South 0 offset in RA At this position the first exposure will be taken After that the target will be moved further 10 arcsec West and 7 arcsec North with respect to the position of the previous exposure etc until the number of exposures is reached If DETECTOR option for coordinates is selected then the directions of the offsets on sky depends on the actual rotator angle If we assume that the orientation is such that the positi
17. 0 Guide Start DEC TEL ROT OFFANGLE no 180 180 0 Rotator on sky TEL TARG ADDVELALPHA yes 15 15 0 Additional Velocity RA TEL TARG ADDVELDELTA yes 15 15 0 Additional Velocity DEC TEL TARG ALPHA no 0 TEL TARG DELTA no 0 TEL TARG EPOCH no 2000 3000 2000 Epoch TEL TARG EQUINOX no 2000 3000 2000 Equinox TEL TARG OFFSETALPHA no 3600000 3600000 0 Alpha offset TEL TARG OFFSETDELTA no 3600000 3600000 0 Delta offset TEL TARG PMA no 10 10 0 Proper Motion Alpha TEL TARG PMD no 10 10 0 Proper Motion Delta Fixed values Parameter Hidden Value Label INS MODE no IFU Instrument Mode SEQ PRESET no T Preset Telescope T F This template is for the acquisition of a field to be observed with the Integral Field Unit Unlike the MOS acquisition template the IFU acquisition template requires to define the filter grism combination that will be used in the following observation template This is to optimize the setup of the instrument at acquisition time The coordinates of the field at the center of the IFU need to be entered in the Target Package at the bottom left of the main P2PP window The parameters for this template are e Filter This should be the same filter as the one used in the first observation template following this acquisition template In order to produce a valid OB the filter in the first observation template following this acquisition template should be the same The available filters a
18. 00438 12800 35 40 54 51000 00506 50200 35 40 54 29000 00587 94300 35 39 56 86000 00732 85400 35 41 15 73000 00905 89000 35 35 32 46000 01087 72400 35 36 37 86000 01136 25300 35 36 53 53000 01405 04000 35 33 23 86000 01929 30100 35 33 50 36000 02003 14700 35 36 05 94000 900091 E C0 CO CIC O0 CO CQ C0 CO s HEIDI LE NO lOO HDI HP PPS ID TA TA TA CO CA A TA CA Un CD N Modify Flags NewRef Comp Forb Automatic Slit Assignment AN Here we have chosen to select according to the R magnitude say we changed our mind and decided to allow slits only for objects brighter than 22 5 despite the fact that our VIMOS catalog contains objects down 88 VIMOS User Manual VLT MAN ESO 14610 3509 to R 24 Notice the two new fields in the upper part of the window saying Min R MAG and Max R_MAG Enter there the values on which you want to select the catalog and hit return or click on Search below the data field The catalog display will be updated as well as the overlay on the image They will now show only objects fulfilling the new selection criteria And only these objects will be used for assigning slits in the next step VIMOS User Manual VLT MAN ESO 14610 3509 89 6 2 6 Step 5 First pass defining special objects Alright if you made it to here you are ready to assign some slits to some objects You are doing well so far In VMMPS you will f
19. 11 PRO CCD MASK Y 2 2 1 4704649e 13 PRO CCD MASK Y 2 3 6 7603913e 17 PRO CCD MASK Y 3 0 8 9077175e 11 PRO CCD MASK Y 3 1 7 827672800000001e 15 PRO CCD MASK Y 3 2 3 7850445e 17 PRO CCD MASK Y 3 3 1 8410463e 20 1 2 EE UILQUY 3T ERA o 12558 797 NS REF1 TYPE SQUARE INS REF1 ID 22595 INS REF1 OBJ RA 243 9472351074219 INS REF1 OBJ DEC 6 144344329833984 S S 1 2 na N VIMOS User Manual VLT MAN ESO 14610 3509 INS REF1 DIMX 30 INS REF1 DIMY 30 INS SLIT1 TYPE RECTANGLE INS SLIT1 ID 27467 INS SLIT1 OBJ RA 244 0057983398438 INS SLIT1 OBJ DEC 6 09309196472168 INS SLIT1 X 108 INS SLITI Y 2175415 INS SLIT1 DIMX 58 INS SLIT1 DIMY 4 88 INS SLIT2 TYPE RECTANGLE INS SLIT2 ID 27313 INS SLIT2 OBJ RA 243 9036865234375 INS SLIT2 OBJ DEC 6 095680713653564 INS SLIT2 X 161 INS SLIT2 Y 394 478 INS SLIT2 DIMX 44 INS SLIT2 DIMY 4 88 INS SLIT3 TYPE RECTANGLE INS SLIT3 ID 26930 INS SLIT3 OBJ RA 243 9173889160156 INS SLIT3 OBJ DEC 6 101338863372803 INS SLIT3 X 263 INS SLIT3 Y 633 422 INS SLIT3 DIMX 38 INS SLIT3 DIMY 4 88 INS
20. 2 3 2 6 3 4 3 5 3 6 3 7 4 6 release for P76 Phase 76 0 Mar 21 2005 27 3 5 7 3 6 6 release for P76 Phase 3 7 7 5 0 Nov 26 2004 all release for P75 phase II and P76 Phase I 1 3 Jun 14 2004 all release for P74 phase II and P75 Phase I 1 2 Jan 12 2004 all release for P73 phase II and P74 Phase I 1 1 July 14 2003 all release for P72 phase II and P73 Phase I 1 0 Feb 11 2003 all first release for P71 phase and P72 Phase I draft Oct 24 2002 all Draft delivered by VIRMOS Consortium Contents 1 Introduction 1 iU AO eae kk wees ad dai add 1 1 2 More Information on VIMOS ssa 2 6a o casaca a aa aaa 1 1 3 Contact Information 2 Las ida do da ee eee ee 1 14 Version notes changedrac A s stages ali era 2 LAT VERSION ORO PDC 2 2 20200522202 a dd er d EI ee A 2 14 23 Version 92 0 222222225825 a eR ka a 2 124 Version SO uos loulou eme donee obese Pob NUTUS GE Eum ded 3 LAY VERSION SUD 2 052 a Ree eedem etd cs e ee denied 3 LAO VerSIOm SSD uu uus udo moo obe mue Ree e ese ou OU a a 3 LA VERSION BID ie Ft WA d A a 3 148 Version 6 0 2 24 4404 ee de Pade A de DOR ede Re CUR 3 14 9 WVerson85 D 222222522255 SOROR ca RR cian REALE RE A a a e a 4 VERSION SAO M e td ad ere 4 Lat Version 80D ss uu om e 4 a oum O 4 4 1 512 Version
21. Guide Star DEC Coordinates of the guide star when the SETUPFILE option of the Get Guide Star From field is selected e Guide Star PPOS Position of the Guide Probe when the SETUPFILE option of the Get Guide Star From field is selected e Optional additional velocity moving targets in alpha and delta Additional Velocity RA and Additional Velocity DEC This velocity should be expressed in arcsec sec These parame ters are mostly intended for imaging or possibly IFU programmes of moving targets If no additional velocity is needed these parameters should be left to their default values 0 42 VIMOS User Manual VLT MAN ESO 14610 3509 4 5 2 VIMOS img FlatSky VIMOS img FlatSky tsf To be specified Parameter Hidden Range Default Label INS FILT NAME no U B VRIzFree NODEFAULT Filter TEL GS1 PPOS no POS NEG POS Selected probe position POS NEG TEL ROT OFFANGLE no 180 180 0 Rotator angle TEL ROTATIO no SKY ABSOLUTE SKY Rotator angle mode SKY AB SOLUTE TEL TARG ADDVELALPHA yes 15 15 0 Additional Velocity RA TEL TARG ADDVELDELTA yes 15 15 0 Additional Velocity DEC TEL TARG ALPHA no 0 TEL TARG DELTA no 0 TEL TARG EPOCH no 2000 3000 2000 Epoch TEL TARG EQUINOX no 2000 3000 2000 Equinox TEL TARG OFFSETALPHA no 3600000 3600000 0 Alpha offset TEL TARG OFFSETDELTA no 3600000 3600000 0 Delta offset TEL TARG PMA no 10 10 0 Pro
22. IFU field of view is available To produce a valid OB the mask shutter option should be the same in every template of the OB This parameter is ignored when MR or HR grisms are used 66 VIMOS User Manual VLT MAN ESO 14610 3509 Chapter 5 Guidecam Tutorial This chapter contains the information about the Guidecam software tool which was in the past present as a separate document entitled Vimos Guidecam User Manual VLI MAN ESO 14610 3512 written by S Brillant Note that users wanting to take advantage of the new pre imagless MOS mode should use the new PILMOS application instead of the Guidecam application The PILMOS application is essentially an enhanced version of the Guidecam application that allows field and guide star selection in exactly the same way as described in this chapter but also has some additional functionality for creating VIMOS catalogues and simulated pre images from contributed catalogues See the instructions in Chapter 7 5 1 Getting started with Guidecam Using Guidecam is fairly simple e Start Guidecam by issuing the command guidecamLite after installation This opens up a skycat GUI with a specific Guide Probe pull down menu Figure 5 1 shows the main Guidecam panel Select the Settings option of the Guide Probe menu This opens up the Guide Probe Settings panel which is represented in Figure 5 2 e Enter the coordinates of the target in the Telescope section of the panel Note that t
23. If there is vignetting it is possible to select another star and to repeat the operation until a suitable guide star is found The guide probe can be positioned on a given guide star in 2 positions called positive and negative POS and NEG This is referenced to as Alternate position By selecting the corresponding button inthe Guide Probe Settings panel figure 5 2 and pressing the Apply button the guide probe will be positioned on the same star on its alternate position This option may allow to reduce or even remove the vignetting that may be present in the other position 70 VIMOS User Manual VLT MAN ESO 14610 3509 K USNO at ESO File Edit Options Catalogs Archives ImageServers H uu uuUZX Figure 5 3 GuideCam the USNO at ESO panel e Once a suitable guide star has been identified you can take note of its position by pressing the Get Guide Star Coords ofthe Guide Probe Settings panel This popups an information panel con taining the RA and DEC of the selected guide star and indicating whether the position of the guide probe is positive or negative alternate position These coordinates as well as the guide probe position should be noted and further entered manually into P2PP in the corresponding fields Guide Star RA Guide Star DEC and Guide Probe Position of the VIMOS img acq Preset template Finally the Get Guide Star From should be set to SETUPFILE see VIMOS Template Refer ence Guide 5 5 Generat
24. OS red Filter GG475 GG435 Free NODE FAULT INS GRIS NAME no LR blue LRred blue Grism HR red HR orange MR NODEFAULT INS SLIT NAID no 0 8 1 0 1 2 1 5 1 8 2 2 5 5 0 1 0 Slit Width SEQ CAL FLAG no 14 1 or 4 quadrants SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no MOS Instrument Mode SEQ HEADER no F Header type This template is used for the observation of a standard star in mode MOS This template can only be used following the VIMOS_mos_acq_Standard acquisition template and not using the VIMOS_mos_acq_Mask This template can only use the specifically designed standard mask from the repository The parameters of this template are e Exposure time seconds This is the integration time in seconds per exposure e Filter The available filters are U B V R I z OS blue OS red GG435 GG475 e Grism The available grisms are LR blue LR red MR HR blue HR orange HR red e Number of exposures per telescop posures are taken for each offset position offset This parameter defines how many ex e 1 or 4 quadrants This determines whether the standard star will be positioned in one or four of the quadrants of the instrument As many exposures as definedin Number of exposures per telescope offset
25. SLIT4 TYPE RECTANGLE INS SLIT4 ID 26695 INS SLIT4 OBJ RA 243 9308166503906 INS SLIT4 OBJ DEC 6 104647159576416 INS SLIT4 X 312 INS SLIT4 Y 867 445 INS SLIT4 DIMX 56 INS SLIT4 DIMY 4 88 INS SLIT5 TYPE RECTANGLE INS SLIT5 ID 26710 INS SLIT5 OBJ RA 243 9892578125 INS SLIT5 0BJ DEC 6 10461950302124 INS SLIT5 X 312 INS SLIT5 Y 1886 62 INS SLIT5 DIMX 56 INS SLIT5 DIMY 4 88 INS SLIT38 TYPE CURVE INS SLIT38 ID 1294 INS SLIT38 0BJ RA 54 63366317749023 INS SLIT38 0BJ DEC 35 58836364746094 INS SLILII38 X 1934 2 VIMOS User Manual VLT MAN ESO 14610 3509 INS SLIT38 Y 1748 86 INS SLIT38 DIMX 72 6812 INS SLIT38 DIMY 72 68 INS SLIT38 BEZIER DY 3 41463 INS SLIT38 BEZIER AX 0 INS SLIT38 BEZIER BX 0 INS SLIT38 BEZIER CX 72 6812 INS SLIT38 BEZIER XX 1897 86 INS SLIT38 BEZIER AY 0 INS SLIT38 BEZIER BY 0 INS SLIT38 BEZIER CY 37 0329 INS SLIT38 BEZIER YY 1734 33 INS SLIT46 TYPE RECTANGLE INS SLIT46 ID 26017 INS SLIT46 0BJ RA 243 9325103759766 INS SLIT46 0BJ DEC 6 108108043670654 INS SLIT46 X 381 9 INS SLIT46 Y 897 INS SLIT46 DIMX 18 78 INS SLIT46 DIMY 9 113 114 VIMOS User Manual VLT MAN ESO 14610 3509 6 5 User check list You can use this list to go step by step through the process and tick off each step one by one
26. User Manual VLT MAN ESO 14610 3509 31 ED Skycat version 2 6 5 steht gef VIMOS MPS o Zoom 2243 259 0 Y 24280 327 0 07 59 57 125 d 10 40 07 01 65535 2116 Low 5000 High 15000 Auto Set Cut Levels 2 Grid i Scale Z 2 S zi n Figure 3 2 VMMPS Graphical User Interface Example of ADP file superimposed to field of view From a catalog of target positions in pixel coordinates extracted from a VIMOS image e VMMPS maximizes the number of slits that can be assigned to the targets in the input catalogue in one or more masks depending on user s choice Upon realization of a first mask definition VMMPS allows manual interaction to unselect targets define curved slits tilted slits etc However note that manual modification of the mask definition should be done only by experienced users and with care e VMMPS allows to define some objects as compulsory e VMMPS requires that some bright stars are defined as reference targets for acquisition purposes Figure 3 2 shows the VMMPS Graphical User Interface 3 5 5 Finding charts Finding charts for service mode OBs must be prepared with VMMPS One finding chart for each quadrant must be attached showing all targets with overplotted slits see Fig 3 2 In addition the first finding chart i
27. VLT MAN ESO 14610 3509 3 5 2 PILMOS Pre imageless MOS observations As of June 2012 Pre ImageLess Multi Object Spectroscopy PILMOS is being offered as an option to VI MOS users However currently there are restrictions for MOS observations in PILMOS mode Constraints that must be met are detailed in Section 7 2 1 In addition we encourage users to take note of the expected PILMOS performance as described in reference 1 of Chapter 7 The implementation of the PILMOS op tion for mask preparation utilises tools and procedures that were already in place for pre image based mask preparation consequently Section 7 3 describes the procedure with particular emphasis on the new PILMOS specific aspects Assuming that the constraints detailed in Section 7 2 1 can be met then PILMOS may be particularly appropriate if e There is a lack of suitable stars for cross correlation in the field e Sources are in any case too faint to be identified on VIMOS pre images e Each pre image field has a relatively low total MOS exposure time associated with it e astrometry accurate to 0 2 is available e Wide slits gt 1 are to be used Nevertheless since PILMOS is a new mode we do not yet recommend this strategy to inexperienced VIMOS users 3 5 3 Rotator on sky Because of the atmospheric dispersion effects Sect 3 2 the orientation on sky for MOS and pre imaging observations is restricted to e PA 0 E W slits Hour Angle within
28. all the objects of your contributed catalog It will still have all the columns that you defined in the contributed catalog plus two columns X IMAGE Y IMAGE giving in pixels the object coordinates in the reference system of the pre image The advantage is that these latter coordinates can now be transformed into mm in the focal plane i e mm coordinates on the mask by using the CCD to Mask transformation present in the header of the pre image VIMOS User Manual VLT MAN ESO 14610 3509 85 6 2 4 Step 3 Starting VMMPS for mask definition At the beginning of this step make sure that your display is blank safest way by selecting Clear in the File pull down menu of skycat or be sure that the image displayed is indeed the pre image that you want to use for this step You have your VIMOS catalog ready Then go ahead and selected in the main window from the pull down menu VIMOS MPS the task Assign slits from a Vimos catalog produce ADP You will be asked to select your VIMOS catalog and the associated VIMOS pre image The latter will appear in the display and all objects in your catalog will be marked as white circles The VIMOS catalog will be shown in a separate window Your screen will look like this Search Options Object Name Equinox 2000 0 03 38 16 835 b 35 37 06 85 Min Radius 0 0 Max Radius 5 45 Max Objects 10000 Select Area Set From Image Search Resu
29. and can be used right away by VMMPS provided that 1t has the right format see below 3 PILMOS case Chapter 7 describes the pre imagless MOS mode that is available since June 2012 VMMPS handles this case in the same way as the Contributed pre image catalog case PILMOS essentially provides an alternative way to arrive at Step 3 6 2 4 below We nevertheless encourage PILMOS users to read through the full procedure in this chapter as well as Chapter 7 in order to gain an understanding of the requirements and assumptions of both strategies We recommend Sextractor Bertin amp Arnouts 1996 A amp AS 117 393 without getting any royalties for it but because it serves well the purpose and you will recognize some similar structures parameters VIMOS User Manual VLT MAN ESO 14610 3509 75 6 2 Step by step guide 6 2 1 Step 0 most crucial Getting ready preparing your catalogs This step is done outside VMMPS But as mentioned above it is the most important one What you need e for each quadrant one VIMOS pre image processed by the ESO pipeline and e EITHER for each quadrant a VIMOS catalog e ORacontributed catalog it can be one catalog including objects on all 4 quadrants or be four catalogs one per quadrant together with for each quadrant a VIMOS pre image catalog The format of the catalogs catalogs accepted by VMMPS have plain ASCII format The file name should not use the extension dat as this causes
30. be defined in detector coordinates e g X Y or in RA DEC All offsets are to be defined in arcseconds The offsets refer to the target so that the target is offset by the requested amount by moving the telescope in the direction opposite to the offset values Examples are provided further down in the description of the templates e Only some combinations of filters and grisms are allowed See Sect 2 3 and Table 2 5 4 4 List of templates Table 4 4 lists the VIMOS templates The description of the templates given in the following pages consist in a table giving all the parameters keywords of each template whether hidden or visible their range of values and the label used in P2PP The hidden parameters have default values which cannot be changed by the users in P2PP After each table a description of some of the important parameters follow using the P2PP label as heading In other words if the preset coordinates are 0 0 then the first pointing of the telescope will center the field at R_1 RA 0 offsetRA_1 DEC_1 DEC_0 offsetDEC_1 where offsetRA DEC_1 are the first couple of offset keyword values and after N offsets the telescope will be pointing to RA N RA_N 1 offsetRA_N DEC_N DEC_N 1 offsetDEC_N 39 VLT MAN ESO 14610 3509 VIMOS User Manual
31. by selecting File gt Make Finding Chart menu Step 9 Make finding chart for each quadrant For each quadrant the finding chart has to be prepared with VMMPS using the plotSlit button and selecting the File gt Make Finding Chart option YOU ARE DONE Chapter 7 PILMOS Tutorial 7 1 Introduction Until 2011 VIMOS MOS observations have required mandatory pre image observations to be carried out usually several months before the actual MOS exposures in order to ensure accurate positioning of the slits on the MOS masks This conservative approach was adopted because there are significant technical difficulties involved in accurately predicting slit positions without a pre image However recent upgrades to VIMOS have improved its stability and a detailed investigation into the accuracy to which slit positions could be predicted concluded that for many realistic observing programmes it would indeed be a valid approach As of June 2012 Pre ImageLess Multi Object Spectroscopy PILMOS is being offered as an option to VIMOS users Currently there are restrictions for MOS observations in PILMOS mode In addition we encourage users to take note of the expected PILMOS performance as described in 1 The implementation of the PILMOS option for mask preparation utilises tools and procedures that were already in place for pre image based mask preparation consequently section 7 3 describes the procedure with particular emphasis on the n
32. c cue eoe UR EEG eo IFU mode attached night time calibrations TEU mode calibration pap coi 0 4 A IPU mode Pipeline cc chee Oa a e do Se e mode Exposure Time 3 7 VIMOS overheads and OB duration 0 0 0 ess 4 User Guide to the VIMOS Templates 4 1 4 2 4 3 4 4 4 5 4 6 4 7 Introduction 2s es ENEE 4 Sar RUE OE A AO d E RCE ue Template names E ee rd RO bee know Reminder sos mom mrs e EN Rey ge EE er A Listor temples sapos sow A wd BP Cen templates s cs ama 4455244 a 4 4 5 1 4 5 2 4 5 3 4 5 4 4 5 5 VIMOS Ima acg Preset uu es eee ED o we Rom RU VIMOS img deg Plaisky asas wa ra aa Ae a VIMOS mos acq Mask 2 22 2222 a EE VIMOS mos acq Standard 2222 4 54 ta ea mp3 e a Aa VIMOS ifu acg Preset oc coa no a e a ee E E EENHEETEN 4 6 1 4 6 2 4 6 3 VIMOS Img obs e a VIMOS mos obs Offset o eee s eee VIMOS iOS OM su ous be dow aa a a a ba da e Calibration 2 2 222252 4 5 2 4 7 1 4 7 2 4 7 3 4 7 4 4 7 5 4 7 6 4 7 7 4 7 8 ue
33. calculator avail able on the ESO WEB pages http www eso org observing etc 3 6 Integral Field Unit IFU 3 6 1 Acquisition accuracy In mode acquisition is blind i e limited by the accuracy of the telescope pointing 3 accuracy In addition to this the pointing accuracy degrades for targets with declination closer to the South Celestial Pole At 70 the uncertainty introduced by the current WCS model for the is also 3 increasing rapidly further south We are working to improve the current model If you plan to observe very southern targets please be aware that we cannot currently guarantee a very accurate pointing i e the target might fall partly out of the IFU FOV 3 6 2 Dithering jitter pattern With the large field of view of the VIMOS IFU techniques equivalent to shift and add classically used in imaging are a possible mode of operations Dithering jitter pattern is a list of differential pointing positions around a central position Several short images with exposure times set to have sky background limited exposures are taken at each position of a dithering pattern which are combined to reconstruct an image with the total required exposure time This allows to eliminate the sky contribution cosmic rays fringing cosmetic defects on the detectors and to produce very accurate flat fielding for correction of pixel to pixel variations The dithering pattern depends on the mean size of the obse
34. close to meridian we have historically requested that the slits be oriented N S forcing the rotator angle to be set at 90 This rotator angle is the default one to be used for pre imaging and the same rotator angle will be used during spectroscopic follow up The normal VIMOS operational scenario has been to carry out MOS observations as close as possible to Meridian within 2 hrs Starting during Period 90 for pre image based MOS observations and starting P92 also for PILMOS observations we have relaxed this constraint as follow for targets at declinations in the range 45 lt lt 5 we permit observations with slits oriented E W with rotator angle set at 0 and with HA within 3 hrs from meridian Notice that increasing the allowed HA range permits observations at higher airmasses which increases the FWHM of the images leading also to increased slit losses For a full report see Sanchez Janssen et al 2012 in the VIMOS web More information and simulations can be found on the VIMOS web Figure 3 1 shows a worst case scenario of MOS observations with LR blue spectral coverage 370 670 nm for a field with declination 20 For details see also Cuby et al Proc SPIE 3355 p36 47 1998 For a different view on the optimal orientation see Szokoly G P ASA 443 703 2005 3 3 Fringing and Flexures The amount of uncorrected flexures has a number of operational consequences e The accuracy of the slit positions
35. determined from pre imaging depends on the rotator position This is for the time being one of the reasons why slit widths narrower than 0 6 are not allowed Accuracy of sky subtraction flat fielding accuracy of sky subtraction flat fielding may be limited by flexures With dithering jittering sequences median running sky frames can be built in imaging or spectroscopy allowing accurate sky subtraction and or flat fielding of the data e In spectroscopy arcs and flats are taken during daytime and at constant rotator angle Depending on the variation in rotator angle between daytime calibrations and night time observations some flat field residuals and wavelength calibration offsets may be introduced due to the flexures In order to mini mize eliminate these effects night time calibrations can be taken with an adequate template see Sect 4 VIMOS User Manual VLT MAN ESO 14610 3509 27 T T Site Paranal o 2 00 arcsec 4 1 Dec 20 pd E fg ks Tstart 2 hrs Tend 2 hrs e 8 a 45micron 37micron Y 4 67micron pos E Gg E ge 5 E t E d SS E o ai o a EM e E Es po o E 341 4 er m 1 1 1i 1 d 10 5 0 5 10 RA arcminutes Figure 3 1 Example of atmospheric dispersion at Paranal for a field at declination 20 observed with the LR_blue grism from 2 hrs from Meridia
36. directly in the boxes If this results in a larger range than offered by default the software will not be able to allocate any slit However if you are say only interested in the very blue part of the spectra you could type in e g lower cut 3700 A and upper cut 4500 Sometimes the algorithm fails with an ungraceful core dump See Section 6 3 98 VIMOS User Manual VLT MAN ESO 14610 3509 A The software will then place the slits in a band at the top of the detector dispersion direction is y with red at top and this will make sure that the full blue end of all spectra lands on the detector A word of caution the values you type in here will only constrain the positioning of the slits by the software but of course they will not influence the physical length of the spectrae for that you would need new blocking filters Thus in case of multiplexity in the dispersion direction handle these parameters with care in order to avoid contamination along the dispersion direction If you have no clue of what this is all about you should probably not try the dirty trick The VIMOS instrument web pages give a good overview of the wavelength range you can expect for each grism as a function of the slit position on the mask For details see http www eso org sci facilities paranal instruments vimos inst mos html VIMOS User Manual VLT MAN ESO 14610 3509 99 6 2 8 Step 7 optional Fine tuning last corrections This step is op
37. during all clear nights Table 3 1 indicates the nature frequency and accuracy of VIMOS calibrations expected to be taken as part of the calibration plan Only calibration frames required for the reduction of science data are reported More calibrations will be taken by the operation staff for the purpose of maintaining and monitoring the instrument configuration and performance e g mask to CCD transformations etc These calibrations are not reported here Calibration monitoring including photometric zero points can be found on the pages of the Quality Control group http www eso org observing dfo quality 3 4 4 IMG mode pipeline See http www eso org qc pipeline status html for the information concerning the latest status of the ESO pipeline 3 4 5 IMG mode Exposure Time Calculator The approximate limiting magnitudes in direct imaging mode can be obtained from the exposure time calcu lator available on the ESO WEB pages http www eso org observing etc 3 5 Multi Object Spectroscopy MOS 3 5 1 Pre imaging Pre imaging must be used to prepare MOS observations when e no suitably accurate astrometric catalogue of the targets is available e or when curved or tilted slits are desired on specific targets In addition if a long total MOS exposure time is expected per pre image field then there is not much to be gained from skipping pre imaging Pre imaging is carried out in service mode see Sect 3 4 1 30 VIMOS User Manual
38. filter Only standard filter grism combinations are allowed as indicated in Table 2 5 Other combina tions should be requested with a waiver e Grism The available grisms are LR blue LR red MR HR blue HR orange HR red In order to produce a valid OB the filter and the grism in the first observation template following the acquisition template should be the same as the one selected in the acquisition template e IFU Magnification This parameter defines the sampling and correspondingly the field of view of the IFU The 2 available samplings are 0 67 arcsec fiber and 0 33 arcsec fiber To produce a valid OB the magnification should be the same in every template of the OB e IFU Shutter This parameter is only meaningful when using the LR grisms In LR spectroscopy there are 4 rows of fibers along the dispersion direction whereas in MR and HR modes only one slit of fibers is used hence the variation of field of view between LR and MR and HR modes In case the 56 VIMOS User Manual VLT MAN ESO 14610 3509 object of interest is of limited size in LR mode and if only the central 1 4 x 1 4 of the field of view is needed it is possible with the IFU Shutter option to mask the outer part of the IFU always masked in MR and HR modes This allows to avoid second order overlap due to sky and or background objects between the 4 adjacent slits along the dispersion direction True means that the shutter is in place and masks t
39. or B filter might be affected to a degree still to be determined New shutters As part of the VIMOS upgrade project a new set of bi stable shutters i e they do not draw current in the open nor in the closed position has been installed The exact properties of these shutters i e shutter delay still need to be determined The shutter one for each CCD camera guarantees uniform illumination of the CCD to the 1 level or better for exposure times as short as 1 sec the shortest possible exposure time 2 5 2 More information More updated information on detector characteristics cosmetics and performance on sky at http www eso org observing dfo quality 2 6 Calibration Units VIMOS has three calibration units illuminating a calibration screen at the back of the Nasmyth shutter closing the Nasmyth tunnel Each unit has the following lamps e Flatfield lamps 10 halogen lamp for imaging screen flats in VRIz and MOS spectroscopic screen flats in LR red and blue 50 halogen lamp for imaging screen flats in U B MOS screen flats MR HR red and blue and all IFU screen flats e Arc lamps He Ar Ne In the web an updated atlas of spectral lines for the different grisms can be found http www eso org sci facilities develop detectors optdet instruments html 22 VIMOS User Manual VLT MAN ESO 14610 3509 2 7 AFC Active Flexure Compensation Since 1st August 2010 VIMOS co
40. pixels along the x direction of the detector i e along the slit e Any other columns whose names do not match the above ones Notice only that the names X_IMAGE and Y_IMAGE are forbidden in the contributed catalogue they are used computed later If these are present the software will kindly reject your catalogue until you remove them well perhaps not too kindly All columns present in the contributed input catalogue will be carried over to the output catalogue Some columns will be used to generate new columns in the output catalogue PILMOS Output The PILMOS application will produce the following output files and write them in the pilmos directory e VIMOS catalogues One catalogue for each quadrant is created The catalogues ave the same format as produced by Step 1 of VMMPS by the cross correlation of the contributed and VIMOS Pre image catalogues see Section 6 2 1 However it will not contain any A IMAGE IMAGE THETA IMAGE or X RADIUS columns The filename will be lt filein gt lt quad gt vm cat where filein is the root name of your contributed catalogue and quad is the quadrant identifier B 1 A 2 A 3 or B 4 A PILMOS source chart for each quadrant This is a FITS file containing an image with the targets in the contributed catalogue shown at the appropriate detector co ordinates the outline of the guide probe vignette and the fixed vignetting in the detector corners A header that contains all of
41. should have these same modes 135 136 C 2 Controllers The 4 CCDs are controlled by two FIERA controllers The detectors of channel 2 and 3 are controlled by FIERA A VIMOSA the detectors of channel 1 and 4 by FIERA B VIMOSB Read out noise and conversion factor Table C 3 indicates the readout noise and gain values for the 4 chips Note that these values are subject to change and are indicated here only for reference Exact values to be found in the image headers under the following keywords HIERARCH ESO DET GAIN and HIERARCH ESO DET OUTI RON the number x in OUTx could change depending on the output amplifier used Table C 3 Detector read out noise and conversion factors For reference only subject to change Chip Read out Mode RON e rms K e7 adu CCD 1 225kpxs low gain IMG mode 5 3 1 70 CCD 2 225 kpx s7 low gain 5 2 1 86 CCD 3 225kpxs low gain 5 3 1 95 CCD 4 225 kpx s71 low gain 4 9 1 80 CCD 1 225kpxs high gain MOS and modes 4 3 0 51 CCD 2 225kpxs high gain 4 6 0 56 CCD 3 225kpxs high gain 4 7 0 57 CCD 4 225kpxs high gain 4 3 0 52 RON and conversion factor monitoring is available through the QC1 WEB pages C 4 Fringing These old detectors suffer from strong fringing in the red With the z Gunn and Bessel J filters the fringing amplitude is around 7 in imaging and in spectroscopy the amplitude of fringing for the
42. the essential keywords that would have been in a pre image of your field This is essentially a simplistic fake pre image based upon your contributed catalogue with the WCS ap propriate for the co ordinates entered into Guidecam and with an indication of the guide probe vignette for the chosen guide star The filename will be pilmos quad sim N fits where quad is the quadrant identifier B 1 A 2 A 3 or B 4 and N is a counter that allows differentiation between the output of multiple uses of the PILMOS application These are the files required by Step 3 of VMMPS Assign slits from a VIMOS catalogue produce ADP of VMMPS see Section 6 2 4 7 3 2 Step 2 VMMPS If the PILMOS application has been used successfully to generate the files described in Section 7 3 then the Step 2 of VMMPS Cross correlate Contributed and VIMOS Pre Image catalogues which requires measured source centroids from a pre image can be skipped Instead it is possible to go straight to VMMPS Step 3 Assign slits from a VIMOS catalogue produce ADP which can now be executed exactly as described in Sec tion 6 2 4 The PILMOS source chart is essentially a fake pre image that should be used in this step Sources will be identified by VMMPS and can be selected for slit allocation in the same way as for a real pre image ADPs will be produced as described in subsequent sections of 1 and header information from the PILMOS 120 VIMOS User Manual VLT MAN ESO 14610 3509 sourc
43. to be many bright targets for cross correlation All objects shall have very good 0 2 relative astrometric coordinates RA DEC on single astrometric system Further all objects should have good 2 absolute astrometry The mandatory columns in the contributed catalogue are in upper case and in the following order ID defining an identifier which must be unique and integer number RA defining the right ascension of the object DEC defining the declination of the object RA and DEC entries are allowed in 2 formats decimal degrees hh mm ss s including semi colons The optional columns whose names are reserved are CROSS is for each object either 1 if that object shall be used for the cross correlations or O if it shall not e g if it is too faint too elongated etc Note that this is not relevant for PILMOS VIMOS User Manual VLT MAN ESO 14610 3509 119 FLAG shall be one of upper or lowercase C compulsory R reference F forbidden S selectable no blank entry is supported any other flag will be replaced by S Though not mandatory in the sense that the PILMOS code will not complain if it is not provided PILMOS users are advised to select reference stars based upon their astrometric source data e A_ARCSEC the semi major axis in arcsec e B_ARCSEC the semi minor axis in arcsec e PA the position angle E of N in degrees These three are not used by the PILMOS code to compute the object size in
44. wavelength range second in order to define the multiplexity LR grism allow up to 4 slits in dispersion direction the MR grism up to 2 The selection of the grism will automatically set up the minimum and maximum wavelengths in the widgets below such that your spectra have the maximum length on the detector around the zero deviation wavelength of the grism The slit width arcsec this will define the slit width for all objects except for reference objects defined by a fixed 4 box and curved slits for which you have defined the slit width separately e Sky Region arcsec defines the minimum sky region that VMMPS will allocated on each side of your object in the slit The pipeline requires at least 1 8 arcsec More will reduce the number of position able slits but increase your sky subtraction capabilities later Low High Cut A define the wavelength range that you need i e the common length of the spectra As mentioned under Grism the value set are giving you the maximum spectral length 500 pixels in LR mode 2000 pix in MR mode 4000 pix in HR mode centered around the zero deviation wavelength of the grism The software will guarantee that length for all spectra On the other hand this will constrain the physical region on the mask in which it can place slits E g if you pick the full range in HR mode only sources in a narrow band in the middle of the detector will be picked It is thus a good idea to reduce the range t
45. will be taken at each telescope position 1 or 4 e Slit Width Define which of the slits available in the masks will be used The width of the available slits are 0 8 1 0 1 2 1 5 1 8 2 0 2 5 5 0 Once the user select the slit width the rest is totally automatic the star will be offset from the acquisition box to the selected slit in the first quadrant and then moved to the same slit in the other quadrants 62 VIMOS User Manual VLT MAN ESO 14610 3509 4 7 6 VIMOS ifu cal Specphot VIMOS ifu cal Specphot tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no U B V RI z OS blue OS red Filter GG475 GG435 Free NODE FAULT INS GRIS NAME no LR blue LRred blue Grism HR_red HR_orange MR NODEFAULT INS IFUE NAME no 0 33 0 67 0 67 IFU Magnification arcsec fiber INS IFUS MODE no IFU Shutter SEQ CAL FLAG no 14 1 or 4 quadrants SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IFU Instrument Mode SEQ HEADER no F Header type This template is used for the observation of a spectrophotometric standard star in IFU mode It is to be used in a OB with the VIMOS fu acq Preset acquisit
46. will consider as executed within specification an OB producing vignetted frames implicitly assuming that the vignetting was already verified by the user If no guide star is provided by the user i e the option CATALOGUE of the Get Guide Star From P2PP parameter of the VIMOS ifu acq Preset template is selected the guide star will be selected by the Telescope Operator at execution time If the user provides a guide star by selecting the option SETUPFILE of the Get Guide Star From P2PP parameter of the VIMOS_img_acq_Preset template efforts will be made to use this guide star for imaging pre imaging and for the subsequent MOS observations although this cannot be entirely guaranteed Observing conditions seeing Moon etc may indeed force to select a different guide star than the one selected by the user In addition it may happen that a guide star suitable for pre imaging observations will not be usable for the spectroscopic follow up if e g the seeing has changed ESO reserves the right to charge to the time allocated to the run the time spent in identifying and setting up the alternative guide star The use of the guidecam tool is mandatory also for the preparation of the finding charts to be delivered as part of their Phase II material The finding charts should clearly indicate the guide star to be used 72 VIMOS User Manual VLT MAN ESO 14610 3509 Chapter 6 VMMPS Tutorial 6 1 The VMMPS philosophy This chapter contai
47. written in the ADP file are in format hh mm ss ddd for RA and dd mm ss ddd for DEC Instead for all slits automatically assigned the coordinates are written in decimal degrees Then when the FITS header is written the coordinate keywords HIERARCH ESO INS SLIT7 OBJ RA HIERARCH ESO INS SLIT7 OBJ DEC get truncated just before the first and therefore have incorrect values This may create problems for some data reduction pipelines e g 106 VIMOS User Manual VLT MAN ESO 14610 3509 6 3 5 Other problems e There are reports of a possible bug the image is displayed but a pop up message appears stating ER ROR Fits file VIMOS image after which one can not proceed with the cross correlation If that happens the user can try to load the image again using the Open sub menu of the File menu in the skycat main window If that does not work please contact ESO s User Support at usd helpWeso org VIMOS User Manual VLT MAN ESO 14610 3509 107 6 4 Examples of catalogs 6 4 1 A contributed catalog The following shows an example of a contributed catalog Note that all mandatory columns ID RA DEC are in as well as a couple of optional reserved columns X IMAGE Y IMAGE and a few user columns VL_color V MAG All these will be carried over to the output of the Cross correlation task i e stored in the resulting VIMOS catalog that will serve as input to VMMPS ID RA DEC V
48. 38 24 581 6 35 37 5949 Equinox 2000 618818 55532 4 Bitpix 32 Inmatch 43 Low T8839 7 0 551 arcsec High 55532 4 Auto Set Cut Levels mat tors EE Scale 21219121 Matching tolerance arcsec Sigma clipping for residuals Number of iterations rms 1 166 arcsec The vmmps cross window summarizes your input top part and awaits your tuning of the cross correlation parameters At the beginning the bottom part is empty The cross correlation parameters are e the matching tolerance in arcsec what is the maximum difference in position that you want to allow between VIMOS pre image and contributed catalog to still consider two objects to be the same If your astrometry in the contributed catalog is perfect pick a small value you are sure that if VIMOS pre image and contributed catalog position do not agree by say 2 the objects cannot be the same Remember however that the VIMOS pre image suffers from optical distortion differential atmospheric refraction etc i e the position in the VIMOS pre image catalog will not be perfect If you suspect the absolute astrometry of your contributed catalog to have a systematic offset use a larger value e g 5 This should take care of it and al
49. 47 Ou 10 3681 0 0007 9 1127 769 61 476 21 994 6 201 89 3 8 5845 0 0002 O 1146 020 73 446 10 094 8 440 36 7 9 5569 0 0003 11 797 370 124 978 4 882 3 866 87 6 10 2443 0 0003 12 456 249 66 226 35 585 7 872 3 8 10 2253 0 0010 13 993 617 51 635 19 266 10 830 16 2 99 0000 99 0000 14 2036 250 59 659 58 085 8 955 UN 10 4253 0 0012 15 281 472 76 981 10 187 7 927 OQ 10 6445 0 0008 16 412 076 102 149 4 210 3 252 89 7 11 2830 0 0006 17 882 229 477 262 35 568 3 459 89 9 8 0405 0 0001 l9 179 795 304820 1 146 402901 8 0 17 1447 0 0356 19 1205 669 89 548 4 645 2 893 6 0 11 3859 0 0007 20 1527 264 133 530 3 779 3 458 75 23 10 6818 0 0003 21 871 428 246 348 9 820 4 277 89 1 9 4710 0 0002 22 1458 521 103 164 2a LIZA 32 078 2 240 13 4509 0 0025 23 370 528 123 550 2 785 2 643 55 8 11 4401 0 0005 24 472 827 106 763 2 265 2 148 53 4 12 7780 0 0014 25 934 416 137 586 Suez 2 9052 52 2 11 0887 0 0004 26 385 186 121 72 24494 25315 FDO 12 1251 0 0008 27 1517 8302 11362 22236 2 197 552526 12 4142 0 0010 28 926 796 142 882 3 188 2 830 6359 11 0968 0 0004 29 832 140 129 522 2 786 2 713 17 4 11 4509 0 0005 30 975 537 113 508 2 163 2 068 10 8 13 0277 0 0017 110 VIMOS User Manual VLT MAN ESO 14610 3509 6 4 4 Example of an ADP file Your final ADP file should look something like this It starts with a few PAF INS keywords defining the file itself In then adds the crucial telescope pointing parameters before describing for which instrument c
50. 5 3 Minor changes after MHI feedback 3 6 1 92 1 May 23 2013 Sec 1 4 1 2 1 Version for P92 Phase II 2 2 2 Fig 2 3 24 241 Fig 2 6 2 8 5 3 4 1 3 5 3 3 6 1 6 2 1 6 2 3 6 2 8 7 3 1 92 0 Feb 25 2013 Tables 2 6 2 7 2 9 Version for P92 Phase I Corrected wavelength cov and 2 10 erage of new HR blue grism 91 1 Nov 24 2012 6 2 6 Modified recomendations for reference objects MRE 2012 09 14 email Version for P91 Phase 91 0 3 2012 Version for 91 Phase 90 1 May 29 2012 Added PILMOS chapter Released for P90 Phase II 90 0 Feb 24 2012 released for P90 Phase 89 0 Dec 28 2011 2 3 2 4 and 2 8 released for P89 Phase II Added new HR blue grism info Added subsections in 2 8 88 0 Sept 15 2010 several Released for P88 Phase I and II 87 1 Dec 12 2010 several Released for P87 Phase II Section 2 2 4 with char acteristics of band passes with new detectors Sev eral places AFC update Instrument shortcomings section added 87 0 Aug 28 2010 all Modified for new detectors and AFC system To be used in Phase I P87 86 1 Mar 03 2010 Ch 2 3 Incorporated some of the feedback from IOT Added QE graph of new CCDs 86 0 Feb 27 2010 Ch 2 3 Phase I P86 Info about new CCDs and upgrade project New policy for waivers for OB duration in SM 85 2 Nov 30 2009 Chapter 4 Further descriptions of the SkyFlat templates Re lease for P85 Phase II 85 1 Nov 29 2009 Chapter 4 Added aut
51. 5 860nm 645 875nm VIMOS User Manual VLT MAN ESO 14610 3509 19 2 5 Detectors Quantum Efficiency of new and old CCDs 100 QE x 300 400 500 600 700 800 900 1000 1100 A nm Figure 2 9 QE curves of the new thick line and old dashed line VIMOS CCDs Notice that the QE of the new CCDs correspond to a single detector Tom 2 5 1 Detector system Since August 1st 2010 VIMOS has four new red sensitive detectors They are 2 CCD 44 82 1 D42 back illuminated non inverted mode operation deep depletion devices The new devices have exactly the same format as the old ones that is 2kx4k Their main difference compared to the old ones is an increased quantum efficiency in the red and reduced fringing The response of the new detectors is comparable to the old ones in the range 400 600 nm where it starts to increase reaching a factor of two improvement at approximately 850 nm At the U band there is a loss of 30 Please check the VIMOS news pages and or contact ESO s User Support Department for further news Table 2 11 indicates the basic characteristics of the VIMOS CCDs Figure 2 9 shows the quantum effi ciency curve of one of the CCDs Detector read out modes and format The CCDs have two read out ports but only one is used This increases the read out time but as a benefit the gain offset and read out noise are constant along the entire CCD In spectroscopic mode the full CCDs 2048x4096 pixels are rea
52. 6 12 26 38 26 38 Bppss 28 13 2812 28 18 2825 Vemos 2788 27 79 2791 2785 Rpgss 28 00 27 92 28 02 27 99 I 27 70 27 57 27 16 27 63 ZGunn S E Table 2 4 Calculated isophotal fluxes and wavelengths Filter AU FU Expected zero point AA nm Wm yum mag nm nm U 378 0 4 89E 08 26 25 370 9 42 1 Bpgss 4312 6 51E 08 28 10 431 3 96 0 Veess 5441 3 63E 08 27 80 547 5 86 7 Rggss 639 8 2 19E 08 27 98 646 3 133 I 814 2 1 06E 08 27 60 828 7 186 ZGunn 900 2 0 83E 08 26 79 935 5 178 1 For Vega as defined in Tokunaga and Vacca PASP 117 421 2005 2 Wavelength at which half the transmission is at shorter wavelengths filter profile only 3 Band width as defined by the distance between the half transmissions filter profile only 11 12 VIMOS User Manual VLT MAN ESO 14610 3509 2 3 Multi Object Spectroscopy MOS A variety of grisms with different wavelength ranges and dispersions is available in MOS mode Order sep arating filters are available to eliminate the overlap for a given slit between orders 1 the order used for science and order 2 always present on the detector This is done by restricting the wavelength range of a particular grism to less than one octave in wavelength The Oth order contamination is present only in multi layered mode i e when observing with the LR and the MR grisms Table 2 5 indicates the recommended and allowed grism filter combinations Any other combinations of grisms and fil
53. EL TARG PMA no 10 10 0 Proper Motion Alpha TEL TARG PMD no 10 10 0 Proper Motion Delta Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WIN1 BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no MOS Instrument Mode SEQ HEADER no F Header type SEQ PRESET no T Preset Telescope T F TEL AG GUIDESTAR no CATALOGUE Get Guide Star from TEL ROT OFFANGLE no 90 Rotator on sky This template is used for the acquisition in mode MOS using the specially designed set of calibrations masks provided by Paranal This set of masks contains a series of slits of different width that can be used for all calibrations that requires the use of a standard star 1 e spectrophotometric radial velocity or metallicity standard Since the template is using a pre defined set of masks no ADP need to be provided by the user No reference stars are used for the acquisition IMPORTANT the coordinates of the standard star should be accurate to better than 1 including proper motion effects the star has to be visible within a 10 x10 acquisition box VIMOS User Manual VLT MAN ESO 14610 3509 47 Note the coordinates of the standard star should be set by the user in the target sub window in P2PP Note the rotator angle is fixed for this template at 90 corresponding to the set of slits in the masks oriented along the N S direction The parameters of the template are e Exposure ti
54. HR grisms reaches in 900nm 40 for slits of 1 This strong fringing was responsible for most of the flat fielding and sky subtraction woes in the red with these old detectors Appendix D Abbreviations and Acronyms The following abbreviations and acronyms are used in this manual ACQ Acquisition ADP Aperture Definition in Pixels ADU Analogue to Digital Unit AFC Active Flexure Compensation A R Adapter Rotator AT Acquisition Template BOB Broker of Observation Blocks CCD Charge Coupled Device DEC Declination ESO European Southern Observatory ETC Exposure Time Calculator FEU Filter Exchange Unit FIERA Fast Imager Electronic Readout Assembly FOMPU Folding Mirror Positioning Unit FPAL Focal Plane Adaptation Lens FWHM Full Width Half Maximum GEU Grism Exchange Unit GUI Grafical User Interface HR High Resolution IFU Integral Field Unit IOT Instrument Operation Team IMG Imaging Mode MEU Mask Exchange Unit MOS Multi Object Spectroscopy Mode OB Observation Block OS Order Sorting PRL Pupil Relay Lens PSF Point Spread Function PTV Peak To Valley P2PP Phase 2 Proposal Preparation RA Right Ascension RMS Root Mean Square RON Read Out Noise TIO Telescope Instrument Operator TSF Template Signature File VIMOS Visible Multi Object Spectrograph VMMPS Vimos Mask Manifacturing Preparation Software 137 138 VLT cm kpx min nm px um Very Large Telescope ngstrom Electron Centimeter Hour KiloP
55. Haute Provence Saint Michel Observatoire France Laboratoire d Astrophysique de l Observatoire Midi Pyr n es Toulouse France Osservatorio Astronomico di Capodimonte Napoli Italy Istituto di Fisica Cosmica e Tecnologie Relative Milano Italy Osservatorio Astronomico di Brera Milano Italy Istituto di Radioastronomia Bologna Italy Osservatorio Astronomico di Bologna Bologna Italy Chapter 2 Instrument Characteristics 24 FOLDING MIRROR PUPIL RELAY LENS NR co N VIMOS Figure 2 1 VIMOS Optical layout 2 1 Overview VIMOS is the Visible Multi Object Spectrograph for the Very Large Telescope VLT of the European Southern Observatory ESO This instrument has been built to provide the ESO community with a wide field spectrograph with high throughput and multiplex dedicated to deep surveys VIMOS operates in the 0 37 1 0 um domain in three main observing modes direct imaging IMG multi slit spectroscopy MOS and integral field spectroscopy IFU Because of the large field of view VIMOS optical path is split into four channels The four quadrants of the instrument are operated in parallel i e it is not possible to mix modes or filters or grisms between quadrants The field of view of the four channels is 4 x 7 x 8 in IMG and MOS and up to 54 x 54 in IFU mode in low resolution spectroscopy It provides an image scale of 0 205 pixel in MOS and IMG and 0 67 or 0 33 per res
56. ILT NAME no U B V RI z OS blue OS red Filter GG475 GG435 Free NODE FAULT INS MASKI ID yes 100000 499999 99 99 Mask ID INS MASK1 NO yes 0 15 99 99 Cabinet Slot No INS MASK2 ID yes 100000 499999 99 99 Mask ID INS MASK2 NO yes 0 15 99 99 Cabinet Slot No INS MASK3 ID yes 100000 499999 99 99 Mask ID INS MASK3 NO yes 0 15 99 99 Cabinet Slot No INS MASKA4 ID yes 100000 499999 99 99 Mask ID INS MASK4 NO yes 0 15 99 99 Cabinet Slot No TEL TARG ADDVELALPHA yes 15 15 0 Additional Velocity RA TEL TARG ADDVELDELTA yes 15 15 0 Additional Velocity DEC TEL TARG ALPHA no 0 TEL TARG DELTA no 0 TEL TARG EPOCH no 2000 3000 2000 Epoch TEL TARG EQUINOX no 2000 3000 2000 Equinox TEL TARG PMA no 10 10 0 Proper Motion Alpha TEL TARG PMD no 10 10 0 Proper Motion Delta Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WIN1 BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no MOS Instrument Mode SEQ PRESET no T Preset Telescope T F This template is used for the acquisition of a field to be observed in MOS mode Note that the acquisition image covers only the central 2048 x 2400 pixels of each of the 4 CCDs whereas the spectroscopic images use the full CCD size 2048 x 4096 pixels Preparing a MOS acquisition template requires that masks have been prepared with VMMPS The output of VMMPS are Aperture Definition Files in
57. I_color V_MAG 1 3 21 46 54 37 13 43 9 242751 23526 2 3i2132349 0T7 37 417 32 2 1 948 LO 3 EK e 1 774 23393 4 3121296417 37 07 29 6 1 599 23227 5 352156027 37 510 22 5 1 927 235 33 6 E EE 26 3 1 063 2326 7 3 21 58 51 37 11 34 0 1 198 23 44 8 3 21 58 65 37 13 54 4 2272 22 83 9 ILL EE 153929 2333 10 3 222 03 02 37 208 21 6 1 924 22 41 2 2 222220331 37208 208 7 0 8833 23 415 1 2 32223503569 37 11 45 7 0 9918 23 31 13 3 22203 98 37207201 0 1 462 23 08 14 3 22 04 55 37 10 33 4 1 101 23439 15 S222 2055 18 37209213 1 393 22 9 16 3 22 06 29 37 20 40 4 1 095 23 09 17 222290840353 SN 1 592 23519 18 SAA LCE 0 9589 22 95 19 3 22 08 47 37 13 56 4 1 662 2205 20 3 22 10 81 37 14 42 0 1 694 23 258 21 34222 11 01 37zl10 31 6 1 664 23553 22 52224312192 3TUL7 405 3 1 725 23 41 23 3 22 112 29 E 1 103 23 54 24 ILLA 202 22 79 25 35222154672 ST TELS 6 5 1 203 2303 26 Se 22 914 66 37 552995 2 0 8767 23332 27 ILL O IN SS 266 1 283 22439 28 3 22x015 99 37 10253 06 1 441 2354 32 29 3 22 16 20 37 15 47 4 22099 30 3 225 19 935 97 17 3549 1 712 23 24 108 VIMOS User Manual VLT MAN ESO 14610 3509 6 4 2 VIMOS pre image catalog The VIMOS pre image catalog was obtained by running our favorite source detection algorithm on the original VIMOS pre image processed by the ESO pipeline The output was just plain x y detector coordinates we added manually a unique ID That list together with
58. In addition the guide star should be chosen with the Guidecam software see Chapter 5 and be on the same astrometric system as the contributed catalogue This may mean applying an offset to the contributed catalogue in order to make it consistent with the guide star catalogue selected in Guidecam 7 2 3 Other Considerations PILMOS is not intended to replace completely the conventional pre image approach there will be many observations for which PILMOS will not be appropriate The most obvious is any observation for which a good astrometric catalogue of sources is not available a priori and the pre image is intended to be used in order to obtain accurate astrometry The accurate placing of curved or tilted slits on extended sources is also likely to require a pre image On the other hand if you have good astrometry for faint targets but few good stars for astrometric calibration of the pre image PILMOS may be a good option Moreover if your observations can be arranged such that targets are not close lt 200pix to the detector edges and in particular the edges of the total FoV then slit losses will be minimised Also if wide slits gt 1 5 can be used then there is little to be gained from taking pre images The expected performance of PILMOS as compared to conventional pre imaging is presented in detail in 1 Users who are concerned about the implications of source centroiding accuracy offered by the two methods should consult that document
59. LEASE iia A qe Rh ES RUE ne 28 SCIENCES IMARIOS mo monde vox m gaba bog E Ress Edd M GL e xS 28 3 4 3 IMG mode calibration plan 29 344 IMG mode pipeline 6 264046 a ded a x muxo 29 3 4 5 IMG mode Exposure Time Calculator 29 3 5 Multi Object Spectroscopy 5 29 A A be e do deu 29 3 5 2 PILMOS Pre imageless MOS observations 30 35 2 BODDOPGUSEY lora a Eee a 30 3 5 4 VIMOS Mask Preparation Software VMMPS 30 3 25 FOIE chans cocoa el A 2 31 90 Visitors using MOS sas A gor osos 9 XE DOS E ox Ro E hes 32 35 MOS mode BUD ocu m no e ER 32 3 5 8 MOS mode attached night time calibrations 32 3 5 9 MOS mode calibationplan 32 35 10 MOS mode pipeline ce c soe a El m RU a 32 VIMOS User Manual VLT MAN ESO 14610 3509 3 6 3 5 11 MOS mode Exposure Time Calculator Integral Field Dei IBID caga wo am Rw ae CA 4 ee RE SRE 3 6 1 3 6 2 3 6 3 3 6 4 3 6 5 3 6 6 3 6 7 ACCURACY 26 2 4 sso om ga a a aa Dithering jitter pattern 2 2 2 2 2 2 2 2 0 mode ensm Dnlersetup
60. MAN ESO 14610 3509 3 1 2 Cumulative offsets For small telescope offsets the telescope may keep the same guide star However observations involving cumulative offsets greater than about 30 often result in significant vignetting of the guide probe or even in the loss of the guiding star For these reasons all dither positions for offset patterns used in the OBs must fit inside a 30 x 30 size box If larger offsets are necessary users need to prepare separate OBs and use appropriate guide star for each of them 3 1 3 Telescope and Instrument Focus The telescope focus is automatically set by the active optics system No intervention is required by the ob server Defocusing of the telescope is not possible during the observations The instrument focus is corrected automatically for the different thickness of the various filters and for temperature variations autofocus 3 2 Atmospheric Dispersion VIMOS is not equipped with Atmospheric Dispersion Corrector in none of its modes The impact of atmo spheric dispersion will be mostly noticed in MOS mode In imaging some image elongation may be noticed at high airmasses in the U or B band In IFU the effect will mostly be noticed with the LR_blue grism and with the 0 33 fiber sampling In MOS mode the effects can lead to severe slit losses if some precautions are not taken To minimize atmospheric dispersion effects at high airmass and assuming that the observations are performed
61. MPS during slit assignment See Sec 6 3 2 for details Forbidden objects You can exclude individual objects from the automatic slit assignment task To do so select the button Forb after having clicked on the object in the image display VIMOS User Manual VLT MAN ESO 14610 3509 93 If the FLAG of an object is set of F Forbidden that object will not be considered for slit assignment This has the same effect as removing the object from the catalog The cases in which you might need this are rare it application useful when you use the same catalog for different purposes E g imagine that you use a single catalog to prepare masks for two different grisms and that you have a few objects for which blue spectra are useless These objects could be flagged manually forbidden when preparing the masks for the blue grisms and save you the time to duplicate the catalog Curved slits You have the possibility to define curved or tilted slits To do so click on the button Curved This will pop up the Curved Slit window essentially a Pick Object skycat window with several special features Slit Width arcsec 10 PIT Rotate Angle degree 0 Straighten Rotate Image Statistics Image X 12683 Image Y 12374 08 00 11 180 d 10 44 11 29 Once you clicked on the object in the image it appears in the Curved Slit window and you can define the curved slit To do this clic
62. Manual VLT MAN ESO 14610 3509 4 6 3 VIMOS ifu obs Offset 55 VIMOS ifu obs Offset To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no U B V RI z OS blue OS red Filter 06475 GG435 Free NODE FAULT INS GRIS NAME no LR blue blue Grism HR red HR orange MR NODEFAULT INS IFUE NAME no 0 33 0 67 0 67 Magnification arcsec fiber INS IFUS MODE no TF F IFU Shutter SEQ CATG no SCIENCE TECHNICAL SCI Observation Category ENCE SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset SEQ NOFF no 1 1000000 NODEFAULT Number of Telescope Offsets SEQ OFFSETI LIST no 0 List of offsets arcsec in RA SEQ OFFSET2 LIST no 0 List of offsets arcsec in DEC SEQ RETURN no Return to Origin T F Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WIN1 BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IFU Instrument Mode SEQ HEADER no F Header type SEQ OFFSET COORDS no SKY Offset Coordinate This template is to be used for observation with the Integral Field Unit The parameters of this template are e Exposure time in seconds This is the integration time in seconds per exposure e Filter The available filters are U R 7 OS blue OS red GG435 GG475 and no
63. OSL 0 INS SHU1 POSH 769 INS SHU2 POSL 769 VIMOS User Manual VLT MAN ESO 14610 3509 INS SHU2 POSH 1549 INS SHU3 POSL 1549 INS SHU3 POSH 2249 INS SHU4 POSL 2249 INS SHU4 POSH 2440 PRO CCD MASK DAYTIM 2004 03 25T19 21 18 334 PRO CCD MASK TEMP 14 490387 PRO CCD MASK X0 131 872451572 PRO CCD MASK XX 0 11921553 PRO CCD MASK XY 0 00010888833 PRO CCD MASK X_0_0 0 18869123 PRO CCD MASK X_0_1 0 00054856849 PRO CCD MASK X_0_2 4 812277 07 PRO CCD MASK X 0 3 1 3219545e 10 PRO CCD MASK X 1 0 0 00022477421 PRO CCD MASK X 1 1 5 4715579e 07 PRO CCD MASK X 1 2 5 5136728e 10 PRO CCD MASK X 1 3 1 8931227e 13 PRO CCD MASK X 2 0 8 3582309e 08 PRO CCD MASK X 2 1 3 428154e 10 PRO CCD MASK X 2 2 4 1386207e 13 PRO CCD MASK X 2 3 1 420323e 16 PRO CCD MASK X 3 0 6 977006e 12 PRO CCD MASK X 3 1 5 0065001e 14 PRO CCD MASK X 3 2 8 499529799999999 17 PRO CCD MASK X 3 3 2 9409612e 20 PRO CCD MASK YO 126 018852108 PRO CCD MASK YY 0 11921683 PRO CCD MASK YX 7 9496414e 05 PRO CCD MASK Y 0 0 0 20919197 PRO CCD MASK Y O 1 0 00026838485 PRO CCD MASK Y 0 2 6 4661712e 08 PRO CCD MASK Y 0 3 1 2921672e 11 PRO CCD MASK Y 1 0 0 00039152516 PRO CCD MASK Y 1 1 1 644576e 07 PRO CCD MASK Y 1 2 7 9799032e 11 PRO CCD MASK Y 1 3 6 9466793e 14 PRO CCD MASK Y 2 0 2 9800619e 07 PRO CCD MASK Y 2 1 4 8490222e
64. PL FILE KEEP T D 4TE 9BS 2002 1 122 28 637 TE en AIPA POS MAM 2POSH 2110 VIO CLD MA SID SY IPI MZ 11 RITUAL AS PRO CCD MA SK TEMP 40259544 PROSCCD MA SEXO 132 3336671 02 Figure 4 3 P2PP VIMOS ADP browser window 46 VIMOS User Manual VLT MAN ESO 14610 3509 4 5 4 VIMOS mos Standard VIMOS mos Standard tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS ADM1 yes ADM File 1 INS ADM2 yes ADM File 2 INS ADM3 yes ADM File 3 INS ADM4 yes ADM File 4 INS ADP1 yes ADP File 1 INS ADP2 yes ADP File 2 INS ADP3 yes ADP File 3 INS ADP4 yes ADP File 4 INS FILT NAME no U B V RI z OS blue OS red Filter GG475 GG435 Free NODE FAULT INS MASK1 ID yes 100000 499999 99 99 Mask ID INS MASK1 NO yes 0 15 99 99 Cabinet Slot No INS MASK2 ID yes 100000 499999 99 99 Mask ID INS MASK2 NO yes 0 15 99 99 Cabinet Slot No INS MASK3 ID yes 100000 499999 99 99 Mask ID INS MASK3 NO yes 0 15 99 99 Cabinet Slot No INS MASK4 ID yes 100000 499999 99 99 Mask ID INS MASK4 NO yes 0 15 99 99 Cabinet Slot No TEL TARG ADDVELALPHA yes 15 15 0 Additional Velocity RA TEL TARG ADDVELDELTA yes 15 15 0 Additional Velocity DEC TEL TARG ALPHA no 0 TEL TARG DELTA no 0 TEL TARG EPOCH no 2000 3000 2000 Epoch TEL TARG EQUINOX no 2000 3000 2000 Equinox T
65. Pixels ADP files Information concerning the target coor dinates and equinox are taken from the ADP files and are automatically loaded in the P2PP target package when including the first ADP file in the template Figure 4 2 shows the P2PP window with an example of a MOS Acquisition The parameters specific to this template are 44 VIMOS User Manual VLT MAN ESO 14610 3509 K UbsBlock Name VIMO SCHO il OL sal 7 ny obs Ofisst hs NF sst Ro E SO Enzo Ra AS LETTRITTAS Po x Storm Er zditlonc zdltlonz ve 22 to DEC PDF lle fenas useri Made ADF lle ADF Flle AUF ile 4 fernaz use o vmC4M1adr __ l Figure 4 2 P2PP Example of a MOS OB Exposure time in seconds This is the integration time for the acquisition image that will be taken through the mask to check the position of the reference stars and of the science target A typical value is 60s depending on the brightness of the reference stars defined with VMMPS Filter This filter is the one to be used for the acquisition image It is highly recommended to choose the same filter as the one used for the pre imaging R Optional additional velocity moving targets in alpha and delta Additional Velocity RA and Additional Velocity DEC This velocity should be expressed in arcsec sec These parameters are mostly intended for imaging and possi
66. S follow up observations after the CCD upgrade i e after 31 August 2010 The use of pre imaging taken between 31 August 2010 and 24 May 2011 can result in a systematic overall offset affecting all quadrants the same way but making field acquisition difficult specially in crowded fields e Added information about new keywords giving coordinates of actual IFU FoV center e Added warnings about IFU pointing in declinations close to the south celestial pole 1 43 Version 92 0 For Phase I P92 Updated the wavelength coverage of the new HR blue grism can be found in tables 2 6 2 7 2 9 and 2 10 VIMOS User Manual VLT MAN ESO 14610 3509 3 1 4 4 Version 90 1 This is the version of the VIMOS manual for P90 phase I and phase II The new HR blue grisms were installed The peak overall efficiency has been measured to be approximately 25 compared with 15 previously See MOS curves at http www eso org sci facilities paranal instruments vimos img VIMOS newHR blueHolog moseff png During P89 we have tested the new Pre Imageless MOS PILMOS mode This new mode will offer to take multi object spectroscopic observations with 4 VIMOS masks without the previously mandatory pre imaging The fake pre images will be simulated based on the users contributed catalogue to be used in input for mask creation with VMMPS We expect to make this mode available fully from P91 If requested via waiver it will also be available for P90 Phase 2 prep
67. SOW arivo dAVTLV 14 SOW dNVTSOW SONIA 21409991257 qvo sour SOWIA eun ounsodxo 4j12eds sey 195 SOIN arIvo eums NAI 8 40 pur ory se IUES QHeo Ae q eo HIT SONIA eun amsodxa jo uoneoyroads ou sore pur sp uS NAI arivo dINV TAVAM NAI AAVA NAT SOWIA oTy Teo njr SOWIA eum eunsodxo fj12eds sore NAI arivo dAVTIVTA NAI dIAV T HI SOWIA Je quo Ss eo njr SOINIA eum ounsodxo 4j12eds sey NAI arivo dAVTIVTA INVTONTSONIA 184099125 eo Sur SOJNIA Sep 1 8 arivo MAYVA NV DIAS SONWIA eg qe272eds SONIA SOW 10 NAT UI oure1 eg arivo 5 5 0945 SOWIA eg erd SONIA SOW 10 NAI UI serg arivo Nava OVNI AAYA DINI SONIA eg eo Sur SOWIA OW ur eg arivo SVId OVNI SVId DINI SONIA eg eo 3ur SOWIA OW ur serg SNOLLVA IHS8O NOLLV IST IVO HNILAVO HONHIOS IOHf8O SOW STO SOW SOWIA 195 O so sour SOINIA pour SOW uruoneA msqO HONHIOS L 2790 NAI S80 NAT SOWIA ISHO sqo njr SOWIA AT 90 SONHIOS Loargo AY FOV STO OWT SONIA SHO sqo Sur SOWIA OWI wonearesqQ HONHIOS S O DINI SOWIA IO sqo SUIT SOWIA epourpjx uruoneA jsqO SNOLLIVASIHSHO HONHIOS NOILISIQOOV LOHfHO IT IS ADVIAI IIIS SOWSOWIA 1049945 boe sour SOWIA Sur jroads uoneA1esqo prepuejs sow 10 uonIsinhoy NOLLISINOOV LOHfHO LTIS HOVIAI ITIS SOW SOWIA prepuejs
68. a list of only z 6 emission line galaxies will probably not allow you to do a nice cross correlation with the VIMOS pre image catalog you might want to add a few V 12 16 mag stars Run your favorite source detection algorithm on the original pipeline processed VIMOS pre image Make sure that you retain only point source with S N gt 20 or so and store all this in a VIMOS pre image catalog Make sure that you prepared the three above ingredients according to Step 0 and move now to Step 2 VIMOS User Manual VLT MAN ESO 14610 3509 81 Your catalog comes from a VIMOS pre image the VIMOS catalog case e Jump to Step 3 If you are sure that you have a VIMOS catalog for each quadrant fulfilling the requirements of Step 0 just identify the corresponding associated VIMOS pre image for each of the 4 quadrants You are ready to jump right into the mask definition move to Step 3 82 VIMOS User Manual VLT MAN ESO 14610 3509 6 2 3 Step 2 Transforming your contributed catalog into a VIMOS catalog This is a necessary step but quite painless if you followed the instructions of Step 0 From the pull down menu VIMOS MPS see Step 1 select the task Cross correlate Contributed and VIMOS pre image catalogs You will see popping up the following VIMOS Cross Correlation sequence window here under KDE Linux K VIMOS Cross Correlation sequence Select Contributed catalog ascii Sele
69. adapter rotator angles This has resulted in a defocus which affects image quality and in spectroscopic modes resolution The degradation can reach 10 20 larger FWHM in IMG or a 20 30 decreased resolution R in MOS or IFU at absolute A R angles of 116 50 degrees Starting 1st April 2012 we correct focus for A R angle and the above situation has been much alleviated 2 8 5 SOLVED focusing issues It has been found during the March 2012 focus improvement intervention that the focusing range of the VI MOS camera did not allow for proper focusing of the IFU units for all temperatures in the normal range for Paranal 5 20 The defocusing is particularly bad when temperatures go above 10 degrees An interven tion to improve this was successfully carried out in Q4 2012 This defocusing was there since the start of operations 2 8 6 Spectrum rotation The blocking mechanism of the grism units the mechanism that fix the grism in place is not perfect and allow for some spectrum rotation as a function of A R angle The peak to valley effect depends of the particular grism used but for a whole rotation of the A R the spectrum can rotate by 10 pix over 4000 pix A redesign of the grism blocking mechanism is under study 24 VIMOS User Manual VLT MAN ESO 14610 3509 2 8 7 Temperature dependent bias level Although this is a common problem for Quadrant 2 the temperature sensitivity is particularly large We thus strongly re
70. ade goes according to plan Table C 1 indicates the basic characteristics of the VIMOS CCDs Figure 2 9 shows the quantum efficiency curve of one of the CCDs Table C 1 Basic characteristics of the old VIMOS CCDs Parameter Value Number of pixels 2048 x 4096 Pixel size um 15 Dark current at 120 C px h T Deviation from Linearity Full well lt 0 5 Charge transfer efficiency gt 0 999999 C 1 Detector read out modes The CCDs have two read out ports but only one is used This increases the read out time but as a benefit the gain offset and read out noise are constant along the entire CCD In spectroscopic mode the full CCDs 2048x4096 pixels are read out in imaging mode readout is windowed to 2048 2400 pixels In fact in both spectroscopic mode and in imaging mode the CCDs are vignetted by some hardware In spectroscopic mode the illuminated area is about 1970x4096 pixels and in imaging mode the illuminated area is about 19702300 pixels The actual values depend on the individual CCDs The default readout modes are indicated in Table C 2 Table C 2 Old Detector System Readout Modes and Characteristics VIMOS Mode Size Speed Binning Gain Port Windowing Readout time Imaging 4x2kx2k 225kpxs 1x1 Low l port Not supported Im Spectroscopy 4x2kx4k 225kpx s 1x1 High 1 port Not supported 1m 1 new detector system
71. ailable any more From October 2005 a set of four new and more efficient HR red VPHG grisms is installed and available in the 4 channels 3 The value quoted is the precision with which we can center a reference fiber in each quadrant The reference fiber is attached to the frames that hold the masks place Thus the motion of the masks is larger than this by an amount yet to be determined but close to 1 5 pix peak to valley 4 Some restrictions apply in operations currently the number of user masks that can be mounted at the instrument cabinet is 8 as 2 slots are used by calibration masks Please refer to the VIMOS web pages for updates channel is equipped with a new deep depletion E2V 44 820 1 D42 device henceforth new CCDs with exactly the same format but with higher throughput in the red and reduced fringing The 4k pixels are arranged along the dispersion direction to maximize the spectral coverage VIMOS is installed on the Nasmyth B focus of UT3 Melipal The VIMOS mechanical and optical characteristics are summarized in Table 2 1 A schematic optical and opto mechanical layout is shown in Figs 2 1 and 2 2 e The focal plane is divided in 4 identical channels and is equipped with the mask exchange units MEU which host up to 8 pre punched masks for MOS spectroscopy per quadrant the IFU fiber head and the IFU masks e The Focal Plane Adaptation Lens FPAL reduces field curvature from 2m to 4 1m partially corrects as
72. al VLT MAN ESO 14610 3509 4 7 4 VIMOS mos cal NightCalib VIMOS mos cal NightCalib tsf To be specified Parameter Hidden Range Default Label SEQ ARC no Night Arc SEQ FLATFIELD no FT T Night Flat field SEQ NARCS no 1 1000000 7 Number of Arcs SEQ NFLATS no 1 1000000 3 Number of Flats Fixed values Parameter Hidden Value Label SEQ HEADER no F Header type This template is to be used when screen flats or wavelength calibrations are needed during the night right after a science observation This template doesn t contain any setup definition and the data will be taken with exactly the same instrument setup as used in the previous science template This template is to be used in a science OB after a science templates It can not be used alone in an OB after an acquisition template The exposure time parameters are automatically defined by the current instrument setup The parameters of the template are e Night t Flat Field T F Ifsetto T 3 screen flats will be taken e Night t Arc T F If T an arc spectrum will be taken VIMOS User Manual VLT MAN ESO 14610 3509 4 7 5 VIMOS mos cal Standard 61 VIMOS mos cal Standard tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no U B V RI z OS blue
73. aration Starting during Period 90 we have relaxed the constraint that forces MOS slits to be oriented NS as follow for targets at declinations in the range 45 lt lt 5 we permit observations with slits oriented E W with rotator angle set at 0 and with HA within 3 hrs from meridian Note that this is still not allowed with PILMOS Users planning to request slits oriented in this way need to request a pre imaging run For a full report see Sanchez Janssen et al 2012 in the mos section of the VIMOS web pages 1 4 5 Version 89 0 This is the version of the VIMOS manual for P89 and later We expect two major changes in VIMOS affecting users in P89 The first one is the commissioning of new volume phase holographic HR blue grisms with increased efficiency compared to the current ones albeit somewhat reduced resolution The second is a redesign of the focusing mechanism in all quadrants that we hope will be able to keep correct focus at all rotator angles The new focusing mechanism will come with new focusing logic that will allow to adjust focus depending on the grism used These changes will conclude the VIMOS upgrade project 1 4 6 Version 88 0 This is the version of the document for P88 and later VIMOS has been running stably after the intervention Only remaining work to be done are improvements to the focusing units to ensure proper image quality and spectral resolution at any rotator angle and the installation of
74. asks are approximately 27 cm from the hinge into the instrument The quadrants to be mostly affected depends on the rotator angle The areas likely to be vignetted have Y pixels lt 200 in Q 1 and 2 and gt 2200 for Q3 and Q4 To ensure that no MOS slits are vignetted users should not place slits in the area likely to be vignetted 2 8 2 fringing There is interference produced by internal reflexions in the IFU unit which origin is not completely understood For details see Jullo et al 2008 i Lagerholm et al 2012 The fringing is nearly zero at 400nm and progressively increases to longer wavelengths This problem will probably remain for the life of the instrument 2 8 3 SOLVED Non optimal focus for some grisms It has been found that the optimal focus depends on what grism is in the light path This means that spec troscopic observations have not been carried out with the optimal focus This situation has been solved since March 2012 Starting in April 2012 the instrumental focus is corrected depending on which grism is in the light path VIMOS User Manual VLT MAN ESO 14610 3509 23 al 00 13 9 35 7 amp h A 54 06 1083 118 7 Q4 direction of rotation Figure 2 10 Vignetting of the IMG FoV by IFU masks flexures as a function of absolute rotator angle 2 8 4 SOLVED Defocusing at some adapter angles The focusing mechanisms specially that of Q3 suffers from flexures which are particluarly large at some
75. ates to position the field in one or the four quadrants of the instrument will be applied automatically by the template The parameters of the template are e Exposure time seconds This is the integration time in seconds per exposure Filter The available filters are U B V R L z To produce a valid calibration OB the filter in the first observation template following the acquisition template should be the same as the one in the acquisition template Number of exposures per telescop posures are taken at each offset position 1 or 4 offset This parameter defines how many ex 1 or 4 quadrants This parameter allows to define if the astrometric field image will be taken in 1 or 4 quadrants If the 1 quadrant option is selected the astrometric field will be positioned in only one quadrant by applying an appropriate offset from the coordinates of the field defined in the target package of the acquisition template If the 4 quadrant option is selected the astrometric field will be positioned sequentially in the 4 quad rants by applying the appropriate offsets from the coordinates of the field defined in the target package of the acquisition template As many exposures as defined in Number of exposures telescopes offset will be taken at each telescope position 1 or 4 At the end of the template the telescope will offset again to return to its position at the beginning of the template 60 VIMOS User Manu
76. ation is not affected by this it is only the simulated fields which are no more than a crude visual aid anyway that are affected The magnitudes are propagated properly they just don t get recognised as magnitudes and used to scale the flux in the simulated images instead zero flux is used and the images are empty apart from the guide probe vignette The work around is to just add another column to the catalaog that is not a magnitude column VIMOS User Manual VLT MAN ESO 14610 3509 117 7 3 Preparation of MOS Masks without a Pre Image 7 3 1 Step 1 PILMOS Guidecam Application The PILMOS application is very similar to the existing Guidecam application described in Chapter 5 The difference is that it includes the possibility of supplying a contributed catalogue from which a VIMOS cata logue and simple simulated pre image are generated For pre imageless observations the PILMOS application should be used instead of the Guidecam application to choose a guide star for your observations the proce dure is however identical to that described in Chapter 5 The only difference is that in order to take advantage of the pre imageless option then you will need to supply the filename of your contributed catalogue in the appropriate field of the Guide Probe Settings panel and click the PILMOS button The PILMOS application includes instrumental calibration parameters that are continually monitored and will be updated with each release of th
77. ble 2 8 shows the spatial characteristics of the IFU mode 2 4 3 Spectral range Table 2 9 shows the spectral range covered in IFU modes by the various grisms as obtained by the automatic pipeline when reducing the spectrophotometric standard star With low and intermediate resolution grisms the spectral coverage is independent of the position of the IFU pseudo slits With the high resolution grisms the spectral coverage is determined also by the fixed positions of the pseudo slits in the focal plane but it varies slightly from quadrant to quadrant The spectral coverage for each quadrant derived from the automatic reduction of standard star observations with VIMOS pipeline is listed in the Table 2 10 2 4 4 Spectral resolution In IFU mode spectral resolution is fixed by the combination grism fiber size the fiber FWHM is about 3 2 pixels corresponding to 0 67 The spectral resolution is about 1 25 times the spectral resolution obtained in MOS mode with a 1 slit width The spectral resolution for various grism filter combinations is given in Table 2 9 2 4 5 Multiplex characteristics The same multiplex capability described in MOS mode is also used in IFU mode With the low resolution grisms there are 4 pseudo slits of fibers along the dispersion direction With intermediate and high resolution grisms only the central pseudo slit is used in practice by masking the entrance field of the IFU so that the other pseudo slits do not rec
78. bly IFU programmes of moving targets and a priori unlikely to be of much use in MOS mode If no additional velocity is needed these parameters should be left to their default values ADP files 1toADP file 4 These 4 parameters are the names of the ADP files produced by VMMPS The files 1 to 4 correspond to the 4 quadrants of the instrument and should be set in the right order When selecting an ADP file entry in P2PP a browser window opens up allowing to locate the file on the local disk This window is presented in Figure 4 3 The available filters are U B R I z OS blue OS red GG435 GG475 and no filter Free VIMOS User Manual VLT MAN ESO 14610 3509 45 K Sport File verl mt RI Gel w n air Formex ree omi acm 2 022 0202 22 Far ty mijen ai D ren 023 vn azz Far ty resend ai tee WQII a 2624 mA az force veerd vni air aer vnl nda Hales APKOR START ral gamme PAFID 6D 4 90314 2 1 2002 11 18T1 7 45 12001 1 PAF NAME oras ms ari PANDESE TADO Aperture Det iew File 10 Piso Tummps 2 AF CRTE D AYTIM 2002 1 1174521 6 001 RG RAL IT 7 AK HCKNAME l ALCHCRDAY PAF CHCKACHFCKSIIM PAHO REMO OCSCONQUAD IMS FOCUT TEME 144277247 INS ADT TVPE MOST INS ADENFRSION INS ADF UNIT PIXEL IMS AEK Llb I W LJ LL DIN AME MODAN USLIUAD T
79. boe sour SONIA yseur Sur prdads uoreA1ssqo prepuejs souriopuonrsmboy NOILISIQOOV LOHfHO LTTS JOVI LIIS SON SONIA boe sour SONIA UONBAJOSqO oouors sour10puonismboy 8 ou Psg Doe nr SOWLA yoydoods uontsinboy ou yosaig boe njr SOWIA oouorss 10 uonrsmboy seGeuir ou uaz ye SUON Wp uontsinboy 8 ou josoig bag Sur SONIA sqo prepurys Surseurr 107 uonismboy pawa ou PsA b Sur SOWIA 840 aguas Surgeurr1ojy uonIsmboy ou Kysy bor Sur SOINIA jeg As Surgeurriopuonrsmboy SNOLLISINOOV DIV AdAL HL oureu JO Jo oureN yM A 40 4 5 VIMOS User Manual VLT MAN ESO 14610 3509 Acquisition templates 4 5 1 VIMOS img Preset VIMOS img Preset tsf To be specified Parameter Hidden Range Default Label INS FILT NAME no U B VRIzFree NODEFAULT Filter TEL AG GUIDESTAR no SETUPFILE CATALOGUE Get Guide Star from CATALOGUE TEL GS1 ALPHA no 0 240000 0 Guide Star RA TEL GS1 DELTA no 900000 900000 0 Guide Start DEC TEL GS1 PPOS no POS NEG POS Selected probe position POS NEG TEL ROT OFFANGLE no 180 180 90 e on sky TEL TARG ADDVELALPHA yes 15 15 0 Additional Velocit
80. brations are running in the morning We are analyzing the data to determine in which circumstances night calibrations will be required Please see the instrument web pages for the latest news 3 4 Imaging IMG Imaging mode is used for both pre imaging runs and science imaging runs In both cases users are requested to use Guidecam for the selection of guide stars See Sect 3 1 1 concerning Guidecam and Sect 5 6 concerning the policy regarding Guide Stars 28 VIMOS User Manual VLT MAN ESO 14610 3509 3 4 1 Pre imaging Since summer 2012 pre imaging is no longer mandatory prior to a MOS follow up however there are still many types of observation for which it is advisable essential see Sections 3 5 1 and 3 5 2 e Asa consequence of the CCD upgrade all pre images taken before the intervention are not valid anymore for MOS follow up observations after the CCD upgrade i e after 31 August 2010 e The use of pre imaging taken between 31 August 2010 and 24 May 2011 can result in a systematic overall offset affecting all quadrants the same way but making field acquisition difficult specially in crowded fields Pre imaging is carried out in service mode only The following instructions for preparation of pre imaging OBs should be strictly followed Program ID Itis mandatory for Phase I to request a specific pre imaging run in the proposal form and for Phase II to prepare the pre imaging OBs by using the corresponding Program ID Fil
81. ckage of the acquisition template If the 4 quadrant option is selected the standard star field will be positioned sequentially in the 4 quadrants by applying the appropriate offsets from the coordinates of the star field defined in the target package of the acquisition template As many exposures as defined in Number of exposures per telescope offset will be taken at each telescope position 1 or 4 At the end of the template the telescope will offset again to return to its position at the beginning of the template VIMOS User Manual VLT MAN ESO 14610 3509 4 7 3 VIMOS img cal Astrom 59 VIMOS img cal Astrom tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no UB VRIzFree NODEFAULT Filter SEQ CAL FLAG no 14 1 1 or 4 quadrants SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IMG Instrument Mode SEQ HEADER no F Header type This template is for acquisition of astrometric data in Imaging mode The acquisition template to use with this template is VIMOS img acq Preset The coordinates to define in the acquisition template are those of the astrometric field Offsets from these coordin
82. commend to use the overscan region on each image to determine the best bias level to be substracted 2 8 8 Temperature dependent image scale During the VIMOS upgrade project we found that the image scale depended on temperature This is the result of the changing optical configuration needed to keep the instrument in focus at all temperatures that is VIMOS is not telecentric The effect was discovered once we started to use the active flexure compensation mechanism Every day we take an image of a pinhole mask with the instrument at the same rotation angle This image 16 taken after acquiring a reference fiber located in the instrument frame and not in the pinhole mask at the edge of the FoV and positioning it at the same pixel position We found that the position of the central pinhole of the mask depended on temperature Starting on December 2011 we correct the reference pixel position depending on temperature to counterbalance this effect With this we can ensure that the central position of the masks do not depend on temperature but we can not change the effect on scale Between 5 degrees and 15 degrees the effect was approximately 2 pixels With the temperature compensation scheme that we have implemented we can expect a maximum shift of only 0 5 pixels for the slits located at the top or bottom of the FoVs Chapter 3 Observing with VIMOS All observations with VIMOS are done via observing blocks OBs OBs consist of one acquisition
83. ct VIMOS Pre image catalog ascii Select VIMOS Pre image File fits Cross Correlate This window asks you to input in that order the catalogs and the image Just click from top to bottom and follow the instruction Once you have filled the last item e that window will look as shown below e the selected VIMOS pre image will appear in the display K VIMOS Cross Correlation sequence Select Contributed catalog ascii home mkissler VIMOS VMMPS exp n1316 contributed cat Select VIMOS Pre image catalog ascii home mkissler VIMOS VMMPS exp n1316 preimage cat Select VIMOS Pre image File fits home mkissler VIMOS VMMPS exp n1316 Image Dr Gross Correlate and after you have clicked on the button Cross correlate e the above window will disappear and e the window vmmps cross will pop up VIMOS User Manual VLT MAN ESO 14610 3509 83 You will then get something like this K Skycat version 2 6 5 fomaxQ1 fits 1 5 vmmps_cross Zoom Object Markus Target X 1484 0 Contributed cat H home mkissler VIMOS VMMPS Old vmmps test fornax contribu VIMOS Pre image cat home mkissler VIMOS VMMPS Old vmmps test fomax preimag VIMOS Pre image E home mkissler VIMOS VMMPS Old vmmps test fornaxG1 fits Output VIMOS cat S home mkissler VIMOS VMMPS Old vmmps test fornax cantrib Value 148601 03
84. d 550 950nm 550 950nm 550 950nm LR blue OS blue 360 670nm 360 670nm 360 670nm MR GG475 500 1000nm 500 1000nm 500 950nm HR blue EW Free 370 535nm 370 624 nm 370 451 nm HR blue P Free 370 552nm 468 674nm 370 566nm HR red GG475 630 870nm 690 930nm 570 805nm HR orange GG435 520 760nm 577 817nm 463 703 blueNEW The ranges correspond to Q1 for slits positioned at y 50 1220 and 2390 in the acquisition image for the bottom center and top ranges and it is common to all quadrants For other quadrants visit the VIMOS web pages 2 HR blue not available after March 2012 Info given for reference plane location Each IFU mask one for each channel of VIMOS has a set of 4 fibre slits or pseudo slits for a total of 6400 fibers 2 41 IFU mode orientation of the field of view Figure 2 5 shows the orientation of the instrument field of view in IFU mode for the four quadrants On the detectors wavelength dispersion is along the y axis with wavelength increasing with y An actual example for PA 0 is shown in Fig 2 6 The increasing position angle on the sky east of north corresponds to the positive rotator angle Spatial characteristics are described in the Section below As for IMG MOS modes the pointing coordinates in the header correspond to the center of the VIMOS field of view Due to the offset position of IFU with respect to the center the keywords RA and DEC in the head
85. d out in imaging mode readout is windowed to 2048x2400 pixels In fact in both spectroscopic mode and in imaging mode the CCDs are 20 VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 11 Basic characteristics of the new VIMOS CCDs Parameter Value Number of pixels 2048 x 4096 Pixel size um 15 Dark current at 120 C e px h lt 7 Deviation from Linearity Full well lt 0 5 Charge transfer efficiency gt 0 999995 vignetted by some hardware In spectroscopic mode the illuminated area is about 1970x4096 pixels and in imaging mode the illuminated area is about 1970x2300 pixels The actual values depend on the individual CCDs The default readout modes are indicated in Table 2 12 Table 2 12 New Detector System Readout Modes and Characteristics VIMOS Mode Size Speed Binning Gain Port Windowing Readout time Imaging 4x2kx2k 225kpxs 1x1 Low l port Not supported 1m Spectroscopy 4x2kx4k 225kpx s 1x1 High 1 port Not supported 1m old detector system had exactly these same modes Controllers The 4 CCDs are controlled by two FIERA controllers The detectors of channel 2 and 3 are controlled by FIERA A VIMOSA the detectors of channel 1 and 4 by FIERA B VIMOSB Read out noise and conversion factor Table 2 13 indicates the readout noise and gain values for the 4 chips Note that these values are subject to change and are
86. detectors 19 Vigneti 222222222 22 o A b Robe 23 Atmospheric Dispersion Example 27 VMMPS Graphical User Interface assa sudor a de 31 example of pre umaging OB sus Rome ee 41 2 Example of a MOS 24 24 42 9 4 44 P2PP VIMOS ADP browser window 2 45 VIMOS mg obs Offset Offsets in DETECTOR mode 52 VIMOS_img obs_Offset Offsets in SKY mode 52 Guidecam main Panel o ul nes A SO E e E 68 Guidecam Guide Probe Settings 1 69 suidecam USNO at ESO panel cs 2s euo E 70 Guide probe setting panel Contributed catalogue button 117 Hilter transmission CHEVER 124 Hilter transmission NEVES n d ma he doe a a ado a da 125 Order Sorting Filter transmission 126 Order Sorting Filter transmission 127 Efficiency curve of the low resolution grisms ens 130 Efficiency curve of the medium resolution grism 131 Efficiency curve of the high resolution 132 Efficiency curve of the high resolution 133 Efficiency curve of the high reso
87. e application expected to be once per VLT period Hence it is essential that you have the latest version of the PILMOS application A link to the official USD Phase 2 page for downloading the PILMOS tool is available at http www eso org sci facilities paranal instruments vimos tools The tar file includes comprehensive installation instructions E Alternate Position Instrument Selection Figure 7 1 Guide Probe Settings window with Cont Cat button bottom middle and PILMOS button bottom 2nd from right Required Input Figure 7 1 shows the Guide Probe Settings interface from the PILMOS application The required input when setting up PILMOS observations is then e Telescope Co ordinates Nothing is changed here relative to conventional guidecam pre image use These are the co ordinates of the centre of the field located at the point where the four detectors meet 118 VIMOS User Manual VLT MAN ESO 14610 3509 Preset RA HH MM SS sss Preset Dec DD MM SS sss Rot on sky Currently the only allowed orientations are 0 for 45 lt 6 lt 5 and 90 for any observable No waivers will be granted for PILMOS observations for any other PA setup Guide Star Co ordinates Nothing is changed here relative to conventional guidecam pre image use Choose a guide star and its co ordinates will be reflected in the Guide Star and Guide Probe fields Guide Probe Orientation Nothing is chang
88. e center of the field In this example the DETECTOR option is selected and the star not the telescope is moved in X and Y according to the list of offsets 1 1 X Figure 4 5 Illustration of the VIMOS img obs Offset template The black dots represent the position of a star which was originally at the center of the field In this example the SKY option is selected and the telescope is moved in RA and DEC in the opposite direction of the offsets in the list so that the star moves in RA DEC according to the offset list VIMOS User Manual VLT MAN ESO 14610 3509 4 6 2 VIMOS mos obs Offset 53 VIMOS mos obs Offset tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no U B V RI z OS blue OS red Filter GG475 GG435 Free NODE FAULT INS GRIS NAME no LR blue LRred blue Grism HR red HR orange MR NODEFAULT SEQ CATG no SCIENCE TECHNICAL SCI Observation Category ENCE SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset SEQ NOFF no 1 1000000 NODEFAULT Number of Telescope Offsets SEQ OFFSETI LIST no NODEFAULT List of offsets arcsec along the SEQ OFFSET2 LIST no 0 Ds of offsets arcsec perpen dicular to the slit SEQ RETURN no FT T Return to Origin T F Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binnin
89. e chart will be propagated to the ADPs just as it would have been from a real pre image 1 Bristow P et al PILMOS Implementation and Testing available from the VIMOS web pages http www eso org sci facilities paranal instruments vimos inst mos html 2 VIMOS Pipeline User Manual VLT MAN ESO 19500 3355 Issue 6 2 Appendices 121 Appendix A VIMOS Filters Table 1 lists the characteristics of the VIMOS filters The original transmission curves delivered by the consortium are given in the following figures The zero points and band pass characteristics valid for the new set of detectors can be found in Section 2 2 4 Table 1 Characteristics of the VIMOS filters Updated filter characteristics can be found in http www eso org sci facilities paranal instruments vimos inst imaging html Filter Ay nm FWHM nm U 370 50 B_BESS 430 97 0 V_BESS 546 89 0 R_BESS 648 5 130 0 I 830 180 0 z GUNN 950 160 GG475 edge filter high pass gt 475 nm GG435 edge filter high pass gt 435 nm OS red box filter band pass 550 950 OS blue box filter band pass 370 670 123 124 Norm T Norm T Norm T Norm T CL LL id ias prepa epecep pep Ly daa ic ay rel 1 1 1 1 360 380 400 360 Wavelength nm 380 400 o 1 1 1 1 1 HAHAH TTTTTTTTTTTT 0 8 1 10 1 1 0 6 0 4 0 2
90. e exposure time calculator available on the ESO WEB pages http www eso org observing etc VIMOS User Manual VLT MAN ESO 14610 3509 35 3 7 VIMOS overheads and OB duration The current overall efficiency of VIMOS is around 55 The actual efficiency depends strongly on which instrumental mode is used and on which observational mode service vs visitor The way in which some functions are operated may change at any time Consequently we cannot give here an estimate of the over heads that is reliable on the long term The user is referred to the current Call for Proposals and to the VIMOS web pages for the most updated estimate of the overheads http www eso org sci facilities paranal instruments vimos overview html It is important during Phase I to be very explicit in the ESO proposal section Justification of requested observing time and observing conditions specifying the ETC input parameters used calculating the total on target time Then to use a realistic split of that time into OBs of appropriate duration It is possible to use the the Phase 2 Preparation Tool P2PP tutorial account for simulation of example OBs see P2PP User Manual or VIMOS Phase 2 web pages Waiver requests for Service Mode OB duration longer than 1h will be rejected unless a sound scientific argument for longer OBs is presented In addition the need for OBs longer than 1h must be explicitly present and justified also in Phase I proposal Sometimes waivers f
91. e measured with the same accuracy In some cases this may degrade the image quality Magnitude range The default minimum and maximum values are 11 and 13 5 whereas the minimum and maximum permitted values are 10 and 14 With the faintest guide stars some degradation of the active optics correction might take place depending on seeing airmass Moon possible extinction by thin cirruses and position of the guide star in the field of view vignetting beyond 11 arcminutes To the extent this is possible stars brighter than 13 5 magnitude should be chosen Too bright objects may result in poorer correction as well In the USNO at ESO pane the maximum number of objects can be changed to either limit or increase the number of stars to be displayed Note whenever a value is changed and entered a new search takes place and the results are displayed on the images while the updated list of stars appears on the Search Results panel Pressing the Search button after changing more than one value will also start a new search and display the results 5 4 Selecting the Guide Star To select a guide star Click with the left mouse button on one of the white circles identifying the USNO stars This opens a small panel asking Move to the selected guide star Pressing Accept will move the guide probe to the star Vignetting of the VIMOS field of view takes place whenever the guide probe yellow intercepts the VIMOS field of view green rectangle
92. e parameters of the template are e Exposure time seconds This is the integration time in seconds per exposure e Number of exposures This parameter defines how many exposures are taken e Filter The available filters are U B V R I 7 OS blue OS red GG435 GG475 e Grism The available grisms LR blue LR red MR HR blue HR orange red e IFU Magnification This parameter defines the sampling and correspondingly the field of view of the IFU The 2 available samplings are 0 67 arcsec fiber and 0 33 arcsec fiber To produce a valid OB the magnification should be the same in every template of the OB e IFU Shutter This parameter is only meaningful when using the LR grisms In LR spectroscopy there are 4 rows of fibers along the dispersion direction whereas in MR and HR modes only one slit of fibers is used hence the variation of field of view between LR and MR and HR modes In case the object of interest is of limited size in LR mode and if only the central 1 4 x 1 4 of the field of view is needed it is possible with the Shutter option to mask the outer part of the IFU always masked in MR and HR modes This allows to avoid second order overlap due to sky and or background objects between the 4 adjacent slits along the dispersion direction True means that the shutter is in place and masks the external part of the IFU False the default value means that the mask is not in place and the full
93. e present the software will kindly reject your catalog until you removed them All columns present in the contributed input catalog will be carried over to the output catalog Some columns will be used to generate new columns in the output catalog see below Step 2 The example of a contributed catalog is given in Section 6 4 1 Sometimes a large number of C compulsory targets in the input catalogue results in crashing See Sec 6 3 2 VMMPS during slit assignment VIMOS User Manual VLT MAN ESO 14610 3509 77 Preparing your VIMOS pre image catalog The VIMOS pre image catalog accompanies the contributed catalog with the only purpose to transform the RA DEC coordinates into x y coordinates This automatically corrects for any distortion in the VIMOS optics and most importantly for the mask manufacturing for x y the transformation into mm ona physical mask is known we have determined that for you and stored the information in the pre image header You need in the VIMOS pre image catalog a small number e g 50 or more of objects that are also in your contributed catalog In the VIMOS pre image catalog you shall have the x y coordinates of these objects obtained from exactly the same VIMOS pre image as you will input into VMMPS Note that the more good objects are in common between the two catalogs the better the result of the cross correlation i e the transformation from RA DEC to pixels will be Since the objects will be
94. e unvignetted telescope focal plane Beyond this radius there is vignetting by the tertiary mirror of the telescope a circle of 13 5 arcminute radius red corresponding to the maximum search radius for guide stars in blue the rotation axis of the Guide Probe arm the position of the USNO stars selectedin the USNO at ESO panel figure 5 3 The stars are indicated with white circles 5 3 The USNO at ESO panel The USNO at ESO panel figure 5 3 provides for the following entries e Search radii min and max The default value for the maximum radius is 12 arcminutes with a maxi mum permitted value of 13 5 Below 11 arcminutes i e within the green circle the Nasmyth field of VIMOS User Manual VLT MAN ESO 14610 3509 69 Guide Probe Settings E Show State Guide Star Automatic Image Load Catalog Selection RA a Show Probe Dec Show Detector Show Fields 3 Alternate Position Instrument Instrument Selection Currently Selected Vimos amp vIMOS Telescope Guide Probe Preset R 0 00 00 0 0 00 00 0 RA Preset Dec 0 00 00 0 0 00 00 0 Dec Rot on sky deg 30 0 Adapter deg Tum Table deg nn nn nn nnn J2000 Apply Get Guide Star Coonds Close Figure 5 2 Guidecam the Guide Probe Settings panel view is vignetted by the tertiary mirror M3 of the telescope Therefore the aberrations of the telescope cannot b
95. ed current MOS OBs might need to be redone with new pre image Also the ETC for the current detectors will give discrepant results specially redward of 700nm Please look for information in the instrument web pages We apologize for the problems that this project will generate to the users during P85 1 4 10 Version 84 0 VMMPS and Guidecam documents were merged into this manual All the links have been updated to reflect the 2008 changes in the ESO web 1 4 11 Version 80 0 Some changes in the night time calibration plan 1 4 12 Version 79 1 Substantial chances to the description of the instrument field of view Changes by S Bagnulo and M Rejkuba Dr S L Morris is acknowledged for having improved Fig 2 7 An error in Table 5 has been corrected 1 4 13 Version 79 0 VIMOS manual has been ported to the standard format for VLT manuals Changes by 5 Bagnulo 1 4 14 Version 78 1 Version 78 1 contains more detailed and accurate information about the spectral range that can be observed in spectroscopic modes MOS and IFU with the various grisms Table C 3 has been updated Hyperlinks have been added to the text The various sections of the manual have been re arranged All changes by S Bagnulo and M Rejkuba comments received by G Marconi and L Christensen 1 5 Acknowledgments The following institutes have participated to the development of VIMOS e Laboratoire d Astrophysique de Marseille Marseille France Observatoire de
96. ed etc Since these have been our main sources of frustration when we debugged the software These will typically have the effect of not opening some windows not showing the plug in in skycat etc OK so let s assume that you have used your particular version of VMMPS successfully in the past but now you in trouble with that particular set of catalogs images 6 3 1 Cross correlation task Process crashes during cross correlation Check that the number of stars with CROSS 1 is below 500 e The circles surrounding the objects used and discarded during the cross correlation are not properly displayed in the image See section 6 2 3 This does not mean that the cross correlation went wrong you need to go to step 3 to visualize further the results of the transformation of your contributed catalog e You have run the cross correlation but not objects were found in common Bummer Check if your input catalogs have all the mandatory columns there is an overlap of objects between your contributed and your VIMOS pre image catalog you have up loaded the right pre image which is used to make the link between contributed and VIMOS pre image catalog coordinates Your final transformation gives RMS gt 1 hmmm are you sure that your contributed catalog has a good astrometry Especially that it does not suffer from any distortions Also check that your pre image catalog does not contain many spurious detec t
97. ed for the matching and counted in ncross E g you can safely put any faint high z galaxy to CROSS 0 in the contributed catalog since it is unlikely to find a match in the VIMOS pre image catalog We recommend to flag fewer than 500 stars with CROSS 1 in the contributed catalog otherwise the cross correlation procedure can crash On the display you will find marked the objects finally used for the transformation nmat ch green as well as the objects that were found to have a counter part in the first pass but were rejected later red This helps you to visually inspect whether the task did something meaningful The log tells you further for each iteration how many objects were found to match and what the RMS of the transformation in arcsec is for these objects You can readjust the parameters and run the task again and again and again until you are satisfied if you cannot reach that spiritual stage press Quit Once you are happy with the precision of the transformation we recommend a RMS of the order of 0 2 or lower click on Save Vimos Catalog and guess what it will save your contributed catalog as a VIMOS catalog The name of the output is composed of the name of your contributed catalog followed by cat that is lt name of contributed catalog gt _vm cat You can now use this catalog to define slits Step 3 A brief word on the output it has a format complying with the one described in Step 0 It will contain
98. ed here relative to conventional guidecam pre image use By default the guide probe will be set to the NEG position and its silhouette drawn accordingly Clicking Alternate position will set the probe to the POS position its silhouette and the Adaptor and Turn Table fields will be updated to reflect this Cont Cat button This is a new functionality required for PILMOS observations This brings up a file system navigation window that should be used to locate the contributed catalogue file The contributed catalogue should be a text file with the coordinates of spectroscopic targets in exactly the same format as described in Section 6 2 1 of this manual APPLY button This button should be clicked after changes to any of the above fields including after entry of a contributed catalogue file name otherwise the file name will not be passed to the PILMOS simulation PILMOS button Pressing this button will send the field and guide probe co ordinates and the con tributed catalogue to the PILMOS simulation Finding charts should be prepared as described in Chapter 5 5 Format of contributed catalogue The following is reproduced from Chapter 6 except for some exceptional information for PILMOS empha sized The contributed catalogue contains all the objects that you wish to observe Note that for PILMOS few bright objects to be used as reference targets for each quadrant need to be included in the contributed cata logue but there do not need
99. egory NICAL SCIENCE SEQ NEXPO no 1 1000000 1 Number of Exposures per Tele scope Offset SEQ NOFF no 1 1000000 VODEFAULT Number of Telescope Offsets SEQ OFFSET COORDS no SKY DETECTOR NODE Offset Coordinates FAULT SEQ OFFSET1 LIST no 0 List of offsets arcsec in X or SEQ OFFSET2 LIST no 0 List of offsets arcsec in Y or SEQ RETURN no Red to Origin T F Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WINI BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IMG Instrument Mode SEQ HEADER no F Header type This template is used for Imaging mode observation both for Science and Pre imaging The parameters of this template are the following e Exposure time in seconds This is the integration time in seconds per exposure e Filter The available filters are U B V R L z In order to produce a valid OB the filter in the first observation template following the acquisition template should be the same as the one selected in the acquisition template For pre imaging the use of R filter is mandatory e Return to Origin T F This parameter defines whether or not after a dithered set of ob servations the telescope will return to the position at the beginning of template As an example this parameter should be set to T when the aim of the OB is to execute multi color imaging observations with the same pattern of offsets o
100. eive sky light VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 9 Spectral characteristics in IFU mode Grism Default Order Spectral range Spectral Res Dispersion Spectral sorting filter common to all quadrants 0 67 fiber A pixel multiplex LR_blue OS Blue 400 670 nm 220 5 3 4 LR red OS Red 590 915 nm 260 7 3 4 MR GG475 490 1015 nm 720 2 5 1 HR blue NEW Free 370 535 nm 1440 0 71 1 HR blue 4P Free 415 620 nm 2550 0 51 1 HR orange GG435 525 740 nm 2650 0 6 1 HR red GG475 645 870 nm 3100 0 6 1 Note new HR_red VPHG grisms have been installed in October 5 2005 Note New VPH HR blue grisms are in use since March 2012 The old grisms are not available any more Their characteristics are given here for comparison purpose The resolution for the new VPH HR blue grism is given for the undeviated wavelength of 401 6 nm which corresponds to 2100 at 587 5 nm the wavelength for the old grism R value Table 2 10 IFU spectral coverage in the various quadrants Grism 01 Q2 Q3 Q4 LR blue OS blue 390 670 395 675nm 400 680nm 400 680nm LR red OS red 570 915nm 580 930nm 590 930nm 585 930 MR 66475 485 1015 490 1020nm 490 1020nm 490 1015 nm HR blue free 370 537 370 535nm 370 550nm 370 563nm HR orange GG435 525 745nm 525 745nm 525 745nm 525 740 nm red GG475 635 860nm 635 865nm 63
101. en you create the ADP files and verified again when the ADP file is attached to an OB If an inconsistency is found the ADP file will not be accepted by p2pp VIMOS User Manual VLT MAN ESO 14610 3509 103 6 2 10 Loading verifying previously defined mask A simple way to come check the ADP your produced is offered in the VMMPS pull down menu 3rd item Verify ADP file By selecting this task you will be asked for an input ADP file and the corresponding pre image Similar to the slit assignment task it will display the pre image and open a catalog window this time containing only the objects allocated in that given Mask You can then as in Step 7 overlay clear the slits with the buttons Plot Slits and Clear Slits Under no circumstances edit the ADP file by hand 8 ADP file that has been once saved cannot be later modified 104 VIMOS User Manual VLT MAN ESO 14610 3509 6 3 Troubleshooting and known problems are some features that we have encountered for you We are grateful for anyone sending us more fea tures Remember that an expert is the one who has already made all mistakes himself you can save some time by using this list in case of problems We only assume that you have install the software properly Especially checked whether your command skycat points indeed to the right version that your environment variables have been set correctly the right Tcl Tk libraries are being us
102. ent and executed according to the user defined parameter values Unless specified otherwise the information provided in this part applies to both service and visitor modes 4 2 Template names VIMOS templates are divided in groups according to the instrument mode and to the observations to be performed This is reflected in the name of the various templates The template naming scheme is the following VIMOS mode type description gt tsf where mode is the mode of the instrument either img mos ifu type is the type of the template either acq obs or cal description 1s a string identifying the purpose of the template Offset for an observation Preset for a preset NightCal for attached night calibrations etc 4 3 Things to know Reminder e See Sect 2 2 1 for a description of the orientation on sky of VIMOS e The default orientation on sky for MOS and pre imaging templates is 90 see Sect 3 2 Any departure from this value should be requested with a waiver e Offset convention The templates make extensive use of telescope offsets which are entered manually as lists The convention is that offsets are relative to the previous position and that the first exposure is 37 38 VIMOS User Manual VLT MAN ESO 14610 3509 executed after the first offset is applied Thus if the first exposure is desired at the position of the preset then the first offset should be 0 0 In some templates the offsets can
103. ent setup as used in the previous observation template This template is to be used in a science OB after observation templates It can not be used alone in an OB after an acquisition template The exposure time parameters are automatically defined by the instrument setup The parameters of the template are e Night t Flat Field T F Ifsetto T 3 screen flats will be taken e Night t Arc T F If T an arc spectrum will be taken VIMOS User Manual VLT MAN ESO 14610 3509 4 7 8 VIMOS ifu cal TwFlat 65 VIMOS ifu cal TwFlat tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds DPR CATG yes CALIB TEST CALIB INS FILT NAME no U B V RI z OS blue OS red Filter GG475 GG435 Free NODE FAULT INS GRIS NAME no LR blue LRred blue Grism HR_red HR_orange MR NODEFAULT INS IFUE NAME no 0 33 0 67 0 67 IFU Magnification arcsec fiber INS IFUS MODE no TF F IFU Shutter SEQ NEXPO no 1 1000000 1 Nunber of Exposure Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WIN1 BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IFU Instrument Mode SEQ HEADER no F Header type This template is used for the observation of twilight flats in IFU mode Itis only offered in visitor mode It should go together with an attached arc Th
104. er is in place and masks the external part of the IFU False the default value means that the mask is not in place and the full IFU field of view is available To produce a valid OB the mask shutter option should be the same in every template of the OB This parameter is ignored when MR or HR grisms are used VIMOS User Manual VLT MAN ESO 14610 3509 63 e Number of exposures per telescope offset This parameter defines how many ex posures are taken for each offset position e 1 or 4 quadrants This determines whether the standard star field will be positioned in one or four of the quadrants of the As many exposures as defined in Number of exposures per telescope offset will be taken at each telescope position 1 or 4 64 VIMOS User Manual VLT MAN ESO 14610 3509 4 7 7 VIMOS ifu_cal_NightCalib VIMOS ifu cal NightCalib tsf To be specified Parameter Hidden Range Default Label SEQ ARC no Night Arc SEQ FLATFIELD no FT T Night Flat field SEQ NARCS no 1 1000000 7 Number of Arcs SEQ NFLATS no 1 1000000 3 Number of Flats Fixed values Parameter Hidden Value Label SEQ HEADER no F Header type This template is to be used when screen flats or wavelength calibrations are needed during the night right after a science observation This template doesn t contain any setup definition and the data will be taken with exactly the same instrum
105. ere for comparison purpose The spectral resolution for the HR blue grism is for the undeviated wavelength 402 nm Notice that for the old HR blue grism the resolution was measured at 587 5 nm At this wavelength we have for the new VPH HR blue R 1680 The wavelength range for the new grism is given for a slit at position yz1220 in acquisition image in Q1 and is common for all quadrants 2 With HR orange grism the spectral range in quadrant 1 for a standard star is 505 745 nm With HR red grisms the spectral range in quadrant 1 for a standard star is 650 865 nm New HR red VPHG grisms have been installed from October 5 2005 However please note that in the multi layer mode there is some order overlap the second order of one slit may overlap with the first order of the adjacent slit There is also 09 order contamination between different layers of slits For faint objects order overlap is usually not a problem only the sky will contribute significantly to the overlap with a contribution that cannot be distinguished from the first order This however constrains the positioning of the slits in the focal plane stacked slits along the dispersion direction have to be strictly parallel and of the same length to allow the second orders to overlap nicely with the first order of the adjacent slits With the low resolution grisms the use of no filter instead of the default order sorting filters OS blue and OS red will result in longer spectra and
106. ers do not refer to the IFU pointing but rather to the center of the VIMOS FoV Prior to P91 the only way to determine the actual position of the IFU was using the center coordinates together with the IFU offset position coded in the keywords HIERARCH ESO OCS CON WCS IFUCENX and HIERARCH ESO OCS CON WCS IFUCENY and the PA on sky after some spherical trigonometry This was needed because the only other relevant information on the headers was HIERARCH ESO TEL TARG ALPHA and HIERARCH ESO TEL TARG DELTA which contained the coordinates of the target Any offsets from the original pointing would result in an IFU pointing that would differ from these keywords Starting in P91 the relevant header keywords which contain the information of the actual pointing of the are HIERARCH ESO INS IFU RA and HIERARCH ESO INS IFU DEC Notice that the whole ESO archive of data has been retro fitted with these keywords so now it is possible to search for IFU observations simply queryng by target coordinates 2 4 2 Fiber arrangement and spatial characteristics The special arrangement of fibers for VIMOS IFU head is shown in Fig 2 7 The fibers numbering has been arranged following the scheme mask line module fiber i e 4 B 5 67 The arrangement of the fibers into the 4 pseudo slits mask one for each channel is reported in Fig 2 8 The detailed IFU table is distributed to the users together with the data release A
107. ew PILMOS specific aspects and directs the reader to existing documentation for the unchanged aspects of the procedure 7 2 Requirements and Restrictions for PILMOS usage The initial PILMOS release is subject to conservative restrictions some of which we expect to relax in future periods The priority is clearly that the PILMOS option does not lead to slit losses that are greater than would have been experienced via the pre imaging approach 7 21 Restrictions In addition to the normal restrictions for MOS Pre image PILMOS observations must abide by the following constraints e Slit widths gt 1 e No curved or tilted slits Observations that cannot be made within these constraints must use the conventional pre image approach There is currently no plan to support curved or tilted slits in PILMOS mode 7 2 2 Catalogue requirements The catalogue requirements are identical to those currently stipulated in the VMMPS documentation sec tion 6 2 1 of the VIMOS user manual 1 namely 115 116 VIMOS User Manual VLT MAN ESO 14610 3509 e All objects shall have very good lt 0 2 relative astrometric coordinates RA DEC on a single astro metric system Notice that this constraint differs from the one in sec 6 2 1 as in that section we refer to the relative positioning accuracy within a quadrant while here we refer to all sources to be observed in all quadrants e All objects shall have good 2 absolute astrometry
108. ey are there spooky Try defining a few virtual points and click View Fit that should work There are occasions in which the subroutine algorithm used to optimize the slit positioning SPOC does not converge In such cases you will see a core dump message saying Child process killed It does not only happen if there are too many compulsory C targets but also if there are too many reference CR or forbidden F targets One of the possible reasons is that the number of compulsory C targets is too large Try to solve this problem by reducing the number of targets changing the flag from C to S for a subset of them If you still have problems please contact ESO s User Support at usd help eso org for a possible workaround 6 3 3 Problems with checksum in ADP files When attaching the ADP file created with VMMPS to an OB in P2PP I get the message that the file is corrupt What does it mean Answer The P2PP reports that the file is corrupt due to inconsistent CHECKSUM value This may happen due to opening the file with some editors which add some invisible characters e g end of line character or sometimes by transferring files between windows and linux computers depending on the protocol and encoding used for the transfer Additionally recently a bug was identified in the step where VMMPS writes the final ADP file If the button Make 1 final ADP is selected the pop up window is displayed asking w
109. fiber blocks are marked with white rectangles For rotator angle 0 North is to the right East to the top Mask 2 Mask 1 PRIMARIAS Line D Se Hd A 5 Line 343 322 Line B For all modules of the Sr ici 80 80 Forall modules of the 5 4 mask 1 1525 35 24559 524 3 2 Line Aj gt gt Gap ZU GA Line A pone a EC Line A For all modules of the 80 For all modules of thel mask 3 80 mask 4 Lise D 15533 34 To Line B bine C 1 133143 RAS Er Mel Line D 31 3 4 sr 12 1 Mask 3 Mask 4 View from the telescope Figure 2 8 IFU mode fibers numbering on the masks VIMOS User Manual VLT MAN ESO 14610 3509 17 Table 2 8 Integral Field Unit spatial characteristics Spatial sampling Field of view Number Pseudo slits Grism Shutter arcsec fibre of fibres quadrant LR blue LR red OFF 0 67 fibre 54 x 54 80 x 80 4 LR blue LR red ON 0 67 fibre 27 x 27 40 x 40 1 LR blue LR red OFF 0 33 fibre 27 x 27 80 x 80 4 LR blue LR red ON 0 33 fibre 13 x 13 40 x 40 1 MR HR blue HR orange HR red ON 0 67 fibre 27 x 27 40 x 40 1 MR HR blue HR orange HR red ON 0 33 fibre 13 x 13 40 x 40 1 When small field can be accommodated this avoids second order contamination by the sky or background objects from the 3 pseudo slits corresponding to the outer part of the field of view Ta
110. g factor DET WIN1 BINY no 1 Y binning factor INS MODE no MOS Instrument Mode SEQ HEADER no F Header type The parameters of this template are Exposure time in seconds This is the integration time in seconds per exposure Filter The available filters are U R 7 OS blue OS red GG435 GG475 and no filter The standard grism filter configurations are given Table 2 5 Only standard filter grism combinations are allowed as indicated in Table 2 5 Other combina tions should be requested with a waiver Grism The available grisms LR blue LR red MR HR blue HR orange HR red Return to Origin T F This parameter defines whether or not after a dithered set of ob servations the telescope will return at the position at beginning of the template This parameter is in general to be left in the default value T in service mode If another value is desired in service mode a waiver should be requested Number of Telescope Offsets This parameter defines the number of telescope offsets dur ing the template The number of exposures taken at each telescope position is determined by the next parameter Number of exposures per telescope offset The pattern of offset is defined in the List of offsets arcsec along the slit and List of offsets arcsec perpendicular to the slit parameters Number of exposures per telescope offset This parameter defines how many expo sures a
111. g has internally been transformed into sky cat format and is now displayed with the standard skycat catalog window This window shows all objects in given radius 5 45 arcmin by default from the center of your pre image The data section in the middle of the window allows you to look at your catalog check values for given ob jects etc A click on an object in the image display will highlight that object in the catalog an allows you to examine its entries The main purpose of the catalog window is to edit interactively the column FLAG that was added by VMMBS and set to S select able for all objects if not present in your input catalog How and why to edit this column is explained in Step 5 and uses the lower part of the window Modify Flags The other purpose of this window is to allow you all operations permitted on skycat catalogs editing values for selected objects searching for objects etc We do not expect you to need these and therefore recommend to you not to use mis use or ab use of them Except maybe for an object selection explained in Step 4 but this is optional If you are satisfied with what you see move directly to Step 5 VIMOS User Manual VLT MAN ESO 14610 3509 87 6 2 5 Step 4 optional Re arranging your VIMOS catalog This step is optional You can skip it and move directly to Step 5 In case you are not fully satisfied with all objects in your catalog you have the chance to re arrange it
112. he coordinates have to be in J2000 and the format should be as indicated For pre imaging observations preparation the default value of 90 degrees for the rotation angle on sky should not be modified unless a waiver is requested When pressing the Apply button Guidecam will download from the ESO archive the DSS image of the field and open a panel USNO at ESO displaying the USNO star catalog of the field This panel is represented figure 5 3 Allow for a few seconds to a few tens of seconds for the image to be downloaded e At this point on it is possible to position the Guide Probe on the available catalog stars and to evaluate if vignetting occurs 5 2 The main Guidecam GUI The Guidecam GUI figure 5 1 displays e the image of the field e the projected shadow of the Telescope Guide Probe yellow onto the telescope focal plane 67 68 VIMOS User Manual VLT MAN ESO 14610 3509 XA guidecaLile Release 1 25 1 2 File View Graphics Figure 5 1 Guidecam main panel skycat Superimposed to the DSS image of the field is in yellow the projected shadow of the Telescope Guide Probe in green the VIMOS field of view rectangle and in green and red the 11 and 13 5 arcminute radii corresponding respectively to the unvignetted Nasmyth field of view and to the maximum search radius for guide stars the VIMOS field of view green with the arrow pointing to the N a circle of 11 arcminutes radius green corresponding to th
113. he external part of the IFU False the default value means that the mask is not in place and the full IFU field of view is available To produce a valid OB the mask shutter option should be the same in every template of the OB This parameter is ignored when MR or HR grisms are used Return to Origin T F This parameter defines whether or not after a dithered set of ob servations the telescope will return to the reference position This parameter is in general to be left in the default value T in service mode If another value is desired in service mode a waiver should be requested Number of Telescope Offsets This parameter defines the number of telescope offsets dur ing the template The number of exposures taken at each telescope position is determined by the next parameter Number of exposures per telescope offset The pattern of offset is defined in the List of offsets arcsec CORRECT LABEL and List of offsets arcsec CORRECT LABEL parameters Number of exposures per telescope offset This parameter defines how many expo sures are taken for each offset position List of offsets arcsec List of offsets between exposures Offsets are defined relative to the previous position The offsets are defined in RA DEC If the first image is to be taken at the preset coordinates the first set of offsets should be 0 0 If there are less offset values than the defined number of offset positions the template will re
114. hether to overwrite and save the resulting file adp The users should answer yes to this question in order to have the correct CHECKSUM written in the ADP file However at this point it is not possible to make any more modifications to the mask design If the button Make 1 final ADP file is selected again and the file saved again the resulting adp file will have incorrect CHECKSUM saved even if no modifications are made The ADP file created with VMMPS has empty CHECKSUM which then results in an error when I try to submit the OBs Answer Make an ADP file by following the usual procedure and at the end when your reach the step where the button Make 1 final ADP has to be selected a window is popped up saying that the file adp exists and asking whether you want to overwrite it If you say NO the adp file will still be there because it was created before the question was issued but it has empty checksum value ADP files with empty checksum are not accepted when submitting OBs Please make sure that when you create ADP files you answer YES to the question whether to overwrite the file However never do this step twice because this will write an incorrect CHECKSUM 6 3 4 ADP files and FITS headers The coordinates of MOS targets in the FITS headers make no sense They are different from those in the VMMPS ADP files Answer If after the automatic slit adjustments a slit is modified inter actively manually then the coordinates
115. ibrations consists of using a pinhole mask to determine accu rately the transformation between millimeters on mask and coordinates on the CCD we know at which mm we cut the holes in the mask and we measure the x y coordinates of the holes on the image processed by the ESO pipeline This transformation is stored and updated on the instrument workstation Now when your VIMOS pre image gets taken this transformation valid at that particular moment is stored in the header of your VIMOS pre image Thus with the help of that transformation you or rather VMMPS know how to go from the x y coordinates of that image to position in mm on the mask i e position on the focal plane The combination of the transformation in that header and the coordinates in that image allows to determine the precise mm coordinates for the mask Should you use transformation and x y from different images the result would not be correct anymore Thus the importance to derive the coordinates from exactly the image that will be fed to VMMPS and from which the transformation will be taken 80 VIMOS User Manual VLT MAN ESO 14610 3509 6 2 2 Step 1 What is your starting point Are your sure you fulfilled Step 0 If you have not suffered at least a little bit while preparing your catalog if you have not thought twice whether you had the right objects in there there is a good chance that you will suffer a serious headache before Step 8 So take your time re thinking about y
116. indicated here only for reference Exact values to be found in the image headers under the following keywords HIERARCH ESO DET GAIN and HIERARCH ESO DET OUTI RON the number x in OUTx could change depending on the output amplifier used Table 2 13 New detectors read out noise and conversion factors For reference only subject to change Chip Read out Mode RON e rms e7 adu CCD 1 225 kpxs low gain IMG mode 3 9 1 76 CCD 2 225kpxs low gain 3 8 1 79 CCD 3 225kpxs low gain 3 6 1 79 CCD 4 225kpxs low gain 3 6 1 71 CCD 1 225 kpxs high gain MOS IFU 3 2 0 53 CCD 2 225 kpxs high gain 3 1 0 55 CCD 3 225 kpxs gt high gain 3 1 0 56 CCD 4 225 kpxs high gain 3 1 0 54 RON and conversion factor monitoring is available through WEB pages VIMOS User Manual VLT MAN ESO 14610 3509 21 Fringing The new detectors have substantially reduced fringing it is negligibly small in I and z band imaging mode In spectroscopic mode there is some residual fringing that depends on the slit width and grism used for a 1 slit fringing at 900nm is only 1 in the high resolution grating growing in nearly a linear fashion with the inverse of the slit width For the low resolution grating it is not detectable Red leaks With CCDs with much higher sensitivity in the red the issue of red leaks needs to be studied with care Imaging observations with the U
117. ing Finding Charts Guidecam should be used to generate finding charts for VIMOS imaging and pre imaging modes Under the File menu select the Make finding charts option new pop window will appears to be filled with Run ID P I name and OB name s Click on Draw to obtain a preview of a finding chart compliant with the rules defined for phase 2 proposal preparation Then press Save to produce a postscript file Detailed and updated instructions are given at http www eso org observing p2pp P2PP FC cookbook html 5 6 Guide Stars Policy Starting from P72 ESO applies the following policy regarding guide stars e Selection of a Guide Star for VIMOS observations is mandatory for MOS and IMG mode to assess VIMOS User Manual VLT MAN ESO 14610 3509 71 and control the amount and position of vignetting that may be introduced by the Guide Probe When the guide star is selected the coordinate should be properly entered in the acquisition template together with the position of the guide probe and the guide star option have to be selected properly see 4 4 If this is not done properly your OB will not pas validation and will not be scheduled for observation See Phase 2 Web pages for detailed information In case of observation involving offsets it is the responsibility of the user to check with guidecam that the chosen guide star does not produce unacceptable vignetting in any of the frame produced by the execution of the OB ESO
118. ion template The parameters of the template are e Exposure time seconds This is the integration time in seconds per exposure e Filter The available filters U B V R I 7 OS blue OS red GG435 GG475 e Grism The available grisms are LR blue LR red MR HR blue HR orange HR red In order to produce a valid OB the filter and the grism in the first observation template following the acquisition template should be the same as the one selected in the acquisition template e IFU Magnification This parameter defines the sampling and correspondingly the field of view of the IFU The 2 available samplings are 0 67 arcsec fiber and 0 33 arcsec fiber To produce a valid OB the magnification should be the same in every template of the OB e IFU Shutter This parameter is only meaningful when using the LR grisms In LR spectroscopy there are 4 rows of fibers along the dispersion direction whereas in MR and HR modes only one slit of fibers is used hence the variation of field of view between LR and MR and HR modes In case the object of interest is of limited size in LR mode and if only the central 1 4 x 1 4 of the field of view is needed it is possible with the IFU Shutter option to mask the outer part of the IFU always masked in MR and HR modes This allows to avoid second order overlap due to sky and or background objects between the 4 adjacent slits along the dispersion direction True means that the shutt
119. ions around the edges of the images This may confuse the cross correlation software 6 3 2 Automatic slit assignment e VMMPS currently does not check for the presence of reference objects see section 6 2 6 It is required that at least one reference object is defined per ADP file and your OBs will be rejected by the User Support Department if this is not the case You loaded pre image and catalog but hardly any no object appears in the catalog window or is marked in the display check in the catalog window the Max Radius It happened to us that with the columns X WORLD and Y WORLD in the catalog skycat decided to pick its own Radius which produced the funniest effect OK you are not in a mood of laughing right now never mind Try changing by hand the Max Radius to something really big 27007 The classical one you pressed during step 6 the button Assign Slits but of the gazillion of objects in the catalog none was assigned a slit VIMOS User Manual VLT MAN ESO 14610 3509 105 you might have been too ambitious in the common wavelength range Try reducing it check the flags in your catalog they are not all F for some reason are they Curved slits you picked the object and are now clicking madly in the Curved Slit window to determine the fancy curved slit but no little dots appear We have not found out why this happens every now and then but we know that even if you do not see them th
120. irst assign all special objects before the rest of the space will be filled automatically with objects We have 5 types of special objects e reference objects cyan diamonds e new reference objects cyan diamonds e compulsory objects blue triangles e forbidden objects yellow crosses e curved slits green squares They are all illustrated in the example below NGC2243 1153 0 2440 5 SS igh i000 Auto Set Cut Levels y Grid Do 7 amati 2 90 VIMOS User Manual VLT MAN ESO 14610 3509 The way to assign one of these is the same for all types in the image display click on the object you want to assign a special type That object must be in the catalog i e have a white circle around it exception new reference objects The white circle will be highlighted in the catalog window under Modify Flags click on the button corresponding to the type you want to assign to the object e g you just clicked in the image on a object that you want to be a reference so click now on the button Ref curved slit is a special case described in more details below in the catalog window the object will be highlighted You can check whether in the column FLAG the type was changed from S select able to the new type e g R for reference e repeat the above steps as often as you need After this operation you will have updated the FLAG column in y
121. isplay finish by pressing Enter in the Pick Object window The new slit will be created with default values the slit width will be the one that you defined in Step 6 during the automatic slit assignment the slit length defined by the X RADIUS of the new object added to the minimum sky region defined in Step 6 Note that in low resolution slits that overlap in dispersion direction get common slit width in Step 6 to facilitate sky subtraction with second order contamination but this is not automatically done here Also there is no check for slit overlap so be extremely careful You can edit modify the values for the new object A new row with that object will appear at the end of the catalog in main catalog window and the entries for this object can be modified by you To do this click on the row with the new object that gets highlighted then and go to the pull down menu Edit to choose Edit selected object This opens a window that allows you to change any entry for that object typically you want to change only the values of DIMX SLIT slit length and DIMY SLIT slit width For science targets you should keep DIMY_SLIT below 4 as wider slits will probably result in saturated screen flat calibrations to use wider slits you will need a special waiver If in doubt use the values of the other objects in the catalog as reference Once you are done click on Enter verify that the values have been updated in the catal
122. ixel Minute Millimeter Nanometer Pixel Second Micrometer Index active flexure compensation 22 Aperture definition in pixels ADP files 25 30 31 calibration units 21 CCD 5 charge transfer efficiency 20 controller 20 dark current 20 exposure shutter 21 format 19 20 full well 20 gain 19 20 number of pixels 20 pixel size 20 QE curve 19 readout modes 20 RON 19 20 exposure shutter 21 field of view 5 filters 5 8 11 21 25 123 127 allowed combinations with grisms 12 broad band filters 8 123 focus 26 no filter with LR grisms 13 order separation 12 order sorting 12 14 130 131 133 pre imaging 28 resolutions with grisms 17 standard grism filter combinations 12 user supplied 12 fringing 21 grisms 4 7 12 14 25 33 129 allowed filter combinations 12 allowed filter combinations per OB MOS 32 atlas of lines 21 atmospheric dispersion 26 27 attached calibrations 32 blocking mechanism 23 calibration plan IFU 34 exchange unit GEU 7 fringing 21 HR orange 13 HR red 13 low resolution 13 LR 12 LR blue 26 MR 12 MR OS blue 12 multiplex 12 multiplex IFU 17 resolution 13 spectral coverage IFU 18 spectral coverage and position in FoV 14 spectral coverage in different quadrants IFU 18 spectral range 12 13 spectral range IFU 17 spectral resolution 12 spectral resolution IFU 17 spectrum rotation 23 standard grism filter combi
123. k in the little display to define the points through which the curved slit should pass 94 VIMOS User Manual VLT MAN ESO 14610 3509 start with a point centered on your object maybe Once you have defined a few points click on View Fit and the slit will appear Note that at any given x the Ay is kept constant to warranty constant slit width dispersion along the slit You can modify the slit width by using the slit width bar or type in directly the value in the associated box and click View Fit again If you wished a straight slit just click Straighten Rotate after having clicked View Fit If you wish to give the slit a given position angle use the Straighten Rotate Angle bar or type in directly the value in the associated box and click on Rotate Remember that in normal pre imaging North is left x on detector and East is up y on detector thus a positive angle here is measured from north to west 90 which by the way would make very limited sense Curved Slit 1 Slit Width arcsec 10 IP Rotate Angle degree 15 p Image Statistics image X 12683 Image Y 12374 a 080011180 10 44 11 29 Equinox 2000 Once you are happy with you curved slit click Save This will update the FLAG of the object to A and store the fit parameters VIMOS User Manual VLT MAN ESO 14610 3509 95 Unselecting objects You can at any time re
124. low to find matching stars The danger are mis identifications if the RMS of the final transformation is large this might have happened Try then to go back to smaller values e the Sigma clipping for residuals the task will find a number of objects in common between VIMOS pre image and contributed catalog It will compute the position difference in the two catalogs and some objects e g mismatches will show high values Define with this parameter when such an object will be rejected from the final fit 84 VIMOS User Manual VLT MAN ESO 14610 3509 e the Number of iterations the task will compute a first transformation for the objects in the contributed catalog applied it and if the number of iterations is gt 1 restart the cross correlation to compute a new transformation Depending on the number of out layers distortions in the field the final transformation might significantly improve with a few iterations If you see no clear improvement in RMS between two successive passes you probably reached the necessary number of iterations Once you have set the parameters click Run Cross to run the cross correlation You will get the log output of the task display in the lower part of the window The log tells you how many objects were found in your VIMOS pre image nref and contributed ncross catalog respectively Do not forget that you can set a column CROSS in your contributed catalog and only the objects with CROSS 1 will be us
125. lows to avoid second order overlap due to sky and or background objects between the 4 adjacent slits along the dispersion direction True means that the shutter is in place and masks the external part of the IFU False the default value means that the mask is not in place and the full IFU field of view is available To produce a valid OB the mask shutter option should be the same in every template of the OB This parameter is ignored when MR or HR grisms are used Rotator on Sky This parameter defines the orientation of the instrument on sky The IFU has a square geometry with the sides oriented N S and E W With a rotator angle of zero the orientation of the IFU on the sky is with the North up and the East to the left The increasing positive position angle on the sky east of north corresponds to the positive rotator angle Therefore the value of this keyword should be equal to the desired position angle PA on the sky see Fig 2 5 Optional additional offsets in alpha and delta Alpha Offset and Delta offset with respect to the coordinates of the object The coordinates at the center of the IFU are the coordinates specified in the Target Package plus the offsets here specified This can be useful e g to define small offsets between the initial pointings on the same field over several OBs Get Guide Star From Two options are offered for this parameter CATALOGUE This is the default option for which the guide star will be searched th
126. lts 134 VIMOS DELTA VIMOS 76600 ISE 00128 91300 37502 00137 97800 426 195 00438 13400 1 96 11 00507 12800 40 54 00506 50200 40 54 00587 94300 2 39 56 00732 85400 41 15 00905 89000 35 32 01087 72400 36 37 01136 25300 36 53 01405 00700 34 40 01569 Search Plot Fiter Modify Flags Ref NewRef Forb Curved UnSel D Automatic Slit Assignment gt There are reports of a possible bug loading the image in some circumstances See Section 6 3 86 VIMOS User Manual VLT MAN ESO 14610 3509 Let s focus on the catalog window We dragged it a little wider to see all columns That window displaying your catalog might look like this just 1 File Edit Options Search Options Object Mame 9 252520222777 Equinox 2000 525022520000 03 38 16 635 0 777777 5 3 6 5 5 2 00 77 DO al Max Radius 5 45 4 10000 Select Area Set From Image Search Results 134 VIMO DELTA VIM X AG IMA RA ANS E 01062 91200 35 27202 00137 01240 597600 35236213 00438 01452 13400 35 36 11 00507 01488 35 SETZ SE 90 35 355 35 CI Y WO CO Y CO CO Y Y CO Ll OY HS HP ds IS P NO PI I IPSO BO H H UU Uu TA uu TA u u TA u uj Im E Modify Flags NewRef Comp Forb Automatic Slit Assignment Your catalo
127. lution 133 xiii xiv VIMOS User Manual VLT MAN ESO 14610 3509 List of Tables 2 1 2 2 2 3 2 4 2 9 2 6 Pos 2 8 2 2 10 2 11 2AL 542 EI 3 2 did A 1 ed 2 C3 Change ECO es eed BE o a ee VIMOS opto mechanical characteristics 6 VIMOS Imaging characteristics and performance 8 Measured imaging zero points a 11 Calculated isophotal fluxes and wavelengths 11 Standard grism filter combinations 12 MOS spectral resolution and spectral ranges observed using the mask for spectrophotometric SIMA is a a et a AR 13 Spectral coverage AA for three positions of the slit in field of view MOS mode 14 MOS 2 n0 koe dec AUR od dd AE TED dea a E A 17 Spectral characteristics in IEU mode ss c ca cew eau 54244 e a a 18 spectral coverage the various quadrants 18 Basic characteristics of the new VIMOS CCDs scorso REGERE GE 20 New Detector Readout Modes 2 222222 ee 20 New detector read out noise and conversion factors 20 VIMOS Calibration Plan IMG Mode moss o om gym 29 VIMOS Calibration Plan MOS Mode 33 VIMOS Calibration Plan
128. masking shutter IFU shutter is provided in front of the input micro lens array this allows the exclusive use of the central 1 4 of the field when is used with intermediate and high spectral resolution grisms In this configuration only 1 fiber slit per quadrant mask is used for a total of 1600 fibers The templates for IFU mode offer the option to insert a shutter to limit the field to the central part also for low resolution grisms VIMOS User Manual VLT MAN ESO 14610 3509 15 VIMOS IFU ORIENTATION ON SKY Rotator angle PA 0 Rotator angle PA 90 AN AN 1 4 4 3 2 i 4 Lo 4 Figure 2 5 Orientation of the VIMOS IFU field of view on sky for position angles 0 and 90 in IFU mode May 25 2013 at 04 46 32 Figure 2 6 left Image of M25 showing the IFU field of view for position angle 0 after pipeline re construction Notice the quadrant ordering starting from top right counterclockwise Q1 Q2 Q3 and Q4 b right The same field with North up and East to the left from the DSS The total FoV is approximately 1 arcmin for both images 16 VIMOS User Manual VLT MAN ESO 14610 3509 VIMOS field Mask 2 Mask 1 Channel 2 Channel 1 Rotator 0 N Mask 3 Mask 4 Channel 3 Channel 4 IFU head View from the telescope Modules of the line A for mask 4 Modules of the lines B C D for mask Figure 2 7 IFU head details of the fiber modules Low transmission
129. me seconds This is the integration time for the acquisition image that will be taken through the mask to check its positioning e Filter This filter is the one to be used for the acquisition image It is recommended to choose the right combination filter exptime to avoid the saturation of the star in the acquisition image The available filters are U B V R I z OS blue OS red GG435 GG475 and no filter Optional additional velocity moving targets in alpha and delta Additional Velocity RA and Additional Velocity DEC This velocity should be expressed in arcsec sec These parameters are mostly intended for imaging and possibly IFU programs of moving targets and a priori unlikely to be of much use in MOS mode If no additional velocity is needed these parameters should be left to their default values 48 VIMOS User Manual VLT MAN ESO 14610 3509 4 5 5 VIMOS ifu_acq Preset VIMOS ifu acq Preset tsf To be specified Parameter Hidden Range Default Label INS FILT NAME no U B V RI z OS blue OS red Filter 06475 GG435 Free NODE FAULT INS GRIS NAME no LR blue LRred blue Grism HR red HR orange MR NODEFAULT INS IFUE NAME no 0 33 0 67 0 67 IFU Magnification arcsec fiber INS IFUS MODE no T F F IFU Shutter TEL AG GUIDESTAR no SETUPFILE CATALOGUE Get Guide Star from CATALOGUE TEL GS1 ALPHA no 0 240000 0 Guide Star RA TEL GS1 DELTA no 900000 900000
130. mo echo e VIMOS imge cal PROM ga bk oe Rc ke md oe ewe a VIMOS mug cal AStrOM as so oo dore d mox m o9 3 VIMOS mos calNiebtCalib lt cases tags gaga 55 VIMOS Standard 2222 22 22 2 2 2 4 44 VIMOS Specphot 2 242 222 EO RE a XC NEU VIMOS ifucalNighiCalib sms ee a gata gas bea RUA R9 E mE VIMOS ifu ca a ir as a A e 5 Guidecam Tutorial Getting started with Guidecani olor pump da Themam Gude QUI iss s nuu o Rn 9e xm ehe ga ba The USNO AL ESO panel lt a c mica X emo E eene BG Selecting the Guide Sia ak ooo ga RR Ges a En Generating Finding Charts ooo rms ars a aa ee 5 1 5 5 3 5 4 203 5 6 Stars POLEN LT 6 VMMPS Tutorial 1X 33 33 33 33 33 34 34 34 34 35 67 67 67 68 69 70 70 73 x VIMOS User Manual VLT MAN ESO 14610 3509 61 The VMMPS philosophy 6 1 1 Whatdoes VMMPS do 6 1 2 This is not FIMS you have little chance for interaction 6 1 3 The importance of your catalog 6 1 4 different catalogs that you can come With 6 2 Step by step guide css usos ala a AA 6 2 1 Step O most crucial Getting ready preparing your catalogs 6 2 2 Step 1 What is your star
131. n The blue and red trails show the evolution of the blue and red wavelengths during the exposure Atmospheric spectra start with N E orientation and move to N W through a N S orientation while crossing the Meridian The maximum amplitude of the dispersion is 2 at 2 hrs from Meridian Slits clearly have to be oriented N S as a default orientation to minimize the slit losses at this declination assuming that all MOS observations will be carried out within 2 hrs from Meridian Attached flat and arc calibrations are mandatory for all IFU observations During the VIMOS upgrade one attached arc and three flats have also been mandatory for all MOS science observa tions The first part of the upgrade completed by 1st August 2010 included changes of shutters CCDs and the installation of the Active Flexure Compensation system In addition many other mechanical parts are closely monitored A second part of the upgrade finished by April 2011 improved image quality and its uniformity accross the field by recentering the cryostats A third part of the upgrade has been completed during March 2012 and included the exchange of the HR blue grisms by VPH ones and a full upgrade of the focusing hardware and software The last and remaining part of the upgrade took place during Q4 2012 was aimed at improving the focusing range for the IFU mode All these activities required access to the instrument during the day which cannot be done if the MOS cali
132. n P2PP should be a finding chart made with Guidecam showing the guide star and the orientation of the guide probe see Fig 5 1 32 VIMOS User Manual VLT MAN ESO 14610 3509 3 5 6 Visitors using MOS Visitors using the MOS mode are requested to prepare Observation Blocks ahead of their scheduled visitor run and upon completion of their pre imaging observations like in service mode This is to guarantee that visitors will have a sufficient knowledge of the VIMOS Mask Preparation Software VMMPS Sect 3 5 4 prior to their visit at Paranal and to allow mask manufacturing ahead of time Limited additional support is guaranteed for mask preparation on the mountain up to 3 masks per night to be prepared at least 48 hrs in advance The instrument has a capacity of 10 masks per quadrant loaded simultaneously one slot is normally occupied by a mask dedicated to observations of spectrophotometric standard stars and a second one for a pinhole mask used for daily instrument health check However some restrictions may apply for technical reasons in the number of masks which can be hosted simultaneously into the cabinets Please refer to the Call for Proposals and to the instrument WEB pages for the updated status of the instrument situation Note that masks cannot be exchanged at night 3 5 7 MOS mode grism filter setup In MOS mode only one filter grism combination per OB is permitted the only filter exchange allowed is the one between acquisi
133. n reason being that VMMPS was conceived as a tool to define masks for surveys Unlike FIMS it will hardly allow you to allocate slits manually Instead it will automatically optimize the slit assignment for object in your catalog Of course you can pick some or even all objects by hand but it will cost you time and nerves and the allocation will not be optimized in terms of numbers It is long term objective of ESO to harmonize the mask preparation tools between VIMOS and FORS after a few periods of VIMOS observations of experience and of feedback from the users At the heart of vmmps is SPOC the Slit Positioning Optimization Code described by Bottini et al PASP 117 996 2005 Given a catalog of objects SPOC maximizes the number of observable objects in a single exposure and computes the corresponding slit positions It does so by solving a very complex 2D combinato rial problem 73 74 VIMOS User Manual VLT MAN ESO 14610 3509 6 1 3 The importance of your catalog VMMPS fully relies on your catalog to contain exactly the objects that you want to observe In VMMPS you have hardly any freedom to set priorities for your preferred objects Your catalog should contain all and only the objects you really want to observe VMMPS will optimize the allocation of slits for you such as to maximize the number of objects observed from your catalog assuming that you have no favorites In summary VMMPS relies heavily on your catalog Thus instead
134. n the same field using different templates This parameter is in general to be left to its default value T in service mode If another value is desired in service mode a waiver should be requested e Number of Telescope Offsets This parameter defines the number of telescope offsets dur ing the template The number of exposures taken at each telescope position is determined by the next parameter Number of exposures per telescope offset The pattern of off sets is defined in the parameters List of offsets arcsec X or RA and List of offsets arcsec in Y or DEC Offsets are defined relative to the previous position e Number of exposures per telescop sures are taken at each offset position offset This parameter defines how many expo VIMOS User Manual VLT MAN ESO 14610 3509 51 e Offset coordinates This parameter defines the coordinate system used to define telescope offsets If the SKY option is selected telescope offsets are in RA and DEC if DETECTOR is selected telescope offsets are in X and Y on the detector In both cases the offsets are given in arcsec e List of offsets arcsec in X or RAandList of offsets arcsec in Y or DEC List of offsets between exposures Offsets are defined relative to the previous position The off sets are defined in RA DEC or X Y on detector depending on which option is selected for the Offset coordinates field If the first image is to be taken at the preset coordinates
135. nal offsets in alpha and delta added to the coordinates of the object specified in the target panel at the bottom of the main p2pp window The coordinates at the center of the instrument i e the central point between the four quadrants are the coordinates specified in the Target Package plus the offsets here specified Get Guide Star From Two options are offered for this parameter CATALOGUE This is the default option for which the guide star will be searched through a catalog and will be selected at the telescope by the Telescope Operator If this option is selected the subsequent fields Guide Star RA DEC will be ignored VIMOS User Manual VLT MAN ESO 14610 3509 4 gt A es ALS g t pr magir lei i tiuu 2905_ fu be Of bel science v05 mq bs f set 2905 ncs Je f set Alika n fs21 poste time eernnr si Letz a oc ervatizn Cato gore uta tur ur sky Jun Exp usures pet Telas Of anal vele city 24 Addr onal y elccit DEL Guid Star Ra Scart DEL Figure 4 1 P2PP example of a pre imaging OB SETUPFILE If this option is selected the guide star that will be used is the one for which the coordinates are defined in the Guide Star RA DEC fields A dedicated tool Guidecam is delivered by ESO to help the users to determine the best possible guide star This option is mandatory for service mode imaging and pre imaging OBs e Guide Star RA and
136. nations 12 VPH 6 VPH HR red 13 VPH HR red 18 IFU masks 14 fiber arrangement 14 image scale 5 IFU 5 IMG 5 MOS 5 imaging 7 broad band filters 8 123 filters 123 zero points 8 IMG mode 7 isophotal magnitudes 8 11 manuals 1 masks 7 30 31 acquisition 43 ADP files 44 exchange at night 32 exchange unit MEU 6 for spectrophotometric standard observation 13 manufacture in visitor mode 32 139 140 mask to ccd matrix 29 32 number simultaneous 6 pre images for preparation of 7 preparation software VMMPS 25 spectrophotometric 46 observing 25 observing modes 5 direct imaging 5 integral field spectroscopy 5 multi slit spectroscopy 5 overheads 32 33 35 P2PP WEB page 1 p2pp 25 Paranal Science Operations contact information 1 WEB page 1 Phase 35 PILMOS 115 Application Guidecam 117 contributed catalogue 118 GuideProbe Settings 117 requirements 115 restrictions 115 source chart 119 Proposal preparation Phase 28 35 Phase II 28 red leak 21 service mode observations contact information 1 observing with VIMOS 25 shortcomings 22 grism blocking 23 grism induced defocusing 22 IFU fringing 22 IFU masks vignetting 22 rotator angle induced defocusing 23 spectrum rotation 23 temperature dependent bias level 24 temperature dependent image scale 24 spectroscopy grisms response 129 standard instrument characteristics VIMOS 6
137. ns the info from the VMMPS Cookbook by M Kissler Patig and G Marconi VLI MAN ESO 14610 3511 formatted to fit in this manual 6 1 1 What does VMMPS do VMMBS is a software that in the end produces a file that you must attach to any of your VIMOS multi object spectroscopy OBs That file called Aperture Definitions in Pixels ADP file contains the list of all slits positions sizes shapes that will be milled into that particular mask The file contains this information in pixels coordinates and hosts in addition the transformation that it got from the pre image header on how to transform that particular x y values into millimeters on the mask Further that file has as well all the necessary pointing information obtained from the pre imaging Creating these ADP files is your goal As an input you need the VIMOS pre image and a catalog of objects And how to get from there to your desired output product is all explained step by step below NOTE that VMMPS handles separately each quadrant You will have to run VMMPS 4 times to produce one full set of masks to be attached to 1 OB the software helps you somewhat to do that The important point here is that you will have to prepare for VMMPS an input set catalogs s and image for each quadrant 6 1 2 This is not FIMS you have little chance for interaction Many MOS users will be familiar with FIMS the software that helps you defining masks for FORS VMMPS is very different The mai
138. o the strict minimum you need This will allow to select more objects e The Number of Masks you can produce one or more masks for a given field If more than one mask is selected the reference objects will be placed in al masks the other special objects allocated in order of priority The task will fill the first masks with slits then the second with objects that have not been allocated yet and so forth e The Main Name of your ADP this name will be extended by the quadrant number the mask number and the grism name when your ADP files get written on disk Once you have set up all the above parameters you can Assign Slits at the bottom of the window This will produce a log shown in the middle sub panel telling you how many select able stars were found in the catalog the number of special objects the total of different placed slits If you are satisfied with the results click Close and Save and move on to Step 7 If you wish to change parameters again just do so and run Assign Slits again If you think that you screwed something fundamental up hey you really forgot reference objects then go back to Step 5 by clicking Back to Changing FLAGS e Dirty Trick with that title it will be the most read section of the cookbook but beware If you are interested in a redder or bluer spectral region than offered by default by the software you can instead of using the widget bar in Low High Cut type in numbers
139. of attached nighttime calibrations arcs and flats see Sect 3 6 5 is MANDATORY for all IFU OBs This template will provide 1 or 3 flat field images for HR grism 3 screen flats are mandatory for the other cases the user can choose the number as well as l arc taken at the same rotator position of the science target The execution time for these IFU nighttime calibrations is always subtracted from the total allocated time 3 6 5 IFU mode calibration plan Observations of spectro photometric standard stars with the same setting used for science observations are nor mally performed within a few nights from the science observations Table 3 3 indicates the nature frequency and accuracy of VIMOS calibrations expected to be taken as part of the calibration plan Only calibration frames required for the reduction of science data are reported More calibrations will be taken by the oper ation staff for the purpose of maintaining and monitoring the instrument configuration and performance e g mask to CCD transformations etc These calibrations are not reported here Calibration monitoring data can be found on the pages of the Quality Control group http www eso org observing dfo quality 3 6 6 IFU mode pipeline See http www eso org qc pipeline status html for the information concerning the latest status of the ESO pipeline 3 6 7 mode Exposure Time Calculator The approximate limiting magnitudes in IFU mode can be obtained from th
140. of using your brain intensively while running VMMPS rather use your intellectual power and scientific judgment to create a smart input catalog Your catalog is the most important item in the process 6 1 4 The different catalogs that you can come with We distinguish between two different scenarios 1 Contributed pre image catalog case Prior to pre imaging with VIMOS you know already which objects you want to obtain spectroscopy for and have precise astrometry for them E g you were running an imaging survey with a wide field imager and have identified your candidates on a given astrometric system This catalog must have precise relative astrometry for all objects and for a number of reference stars It is important that the astrometry of the science objects and the reference stars are on the same system You will now obtain pre imaging with VIMOS extract from there a short list of objects detected on the VIMOS image the VIMOS pre image catalog and VMMPS will help you to bring your contributed catalog onto the VIMOS coordinate system 2 VIMOS catalog case You only know which objects or type of objects you are after but have no list with precise RA DEC for your objects E g you rely on the VIMOS pre image to determine the objects you will obtain spectroscopy for This catalog is obtained directly from the VIMOS image using your favorite analysis tools It takes coordinates in detector x y as well as any other parameter you want to convey
141. og and click on Close A good practice is then to Plot Slits and visually inspect whether your new slit has the right length i e dimension in x or whether it is overlapping touching other slits Remember to be generous with 100 VIMOS User Manual VLT MAN ESO 14610 3509 space between slits as you have to respect some physical mask stability criteria We caution again from doing to many modifications at this point none should be necessary since it is much easier to screw things up here than to improve them Remember that the task already optimized the al location If you modify anything be sure that you are still respecting all constraints including contamination issues along the dispersion direction calibration issues pipeline requirements If you have any doubts do not modify anything that mask is certainly good enough already The software will not check whether your new slits are overlapping in the spatial or spectral direction with already allocated ones The display now shows all objects in your input catalog marked with white circles and the objects for which slits were allocated in this mask with yellow squares special objects are not marked any longer as such Zoom Object NGC2243 136 0 2431 0 339 07 59 55 412 10 40 06 40 High 15000 Auto Set Cut Levels 4 Grid VIMOS User Manual VLT MAN ESO 14610 3509 101 You can use the but
142. olution The spectral resolution obtained with a 1 slit width with the various grisms is given in Table 2 6 2 3 3 Multiplex characteristics A particular feature of the MOS mode is the possibility of stacking multiple layers of slits along the dispersion direction This is only possible with the low resolution grisms up to four slits along the dispersion direction and the intermediate resolution grisms up to two slits for which the spectra are short enough in number of pixels to allow stacking With the high resolution grisms only one slit can be placed along the dispersion direction VIMOS User Manual VLT MAN ESO 14610 3509 13 Table 2 6 MOS spectral resolution and spectral ranges observed using the mask for spectrophotometric standard stars Grism Default Order Spectral Number Undeviated Spectral Res Dispersion Spectral sorting filter range of pixels wavelength nm 1 slit A pixel multiplex LR_blue OS Blue 370 670 nm 550 480 180 5 3 4 LR red OS Red 550 950 nm 550 750 210 7 3 4 MR GG475 480 1000 nm 2000 700 580 2 9 2 MR OS red 550 970 nm 2000 700 580 2 5 2 HR blue NEW Free 370 535 nm 2324 402 1150 0 71 1 HR blue P Free 415 620 nm 4096 510 2050 0 51 1 HR orange GG435 520 760 nm 4096 631 2150 0 6 1 HR red GG475 630 870 nm 4096 740 2500 0 6 1 New VPH HR blue grisms are in use since March 2012 The old grisms are not available any more Their characteristics are given h
143. olution element in IFU mode depending on the elongator setting used Until June 2010 each of the four channels was equipped with a 2048 x 4096 pixels EEV 44 82 backside illuminated single layer coating CCD Henceforth these will be the old CCD Since August 1st 2010 each 5 6 VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 1 VIMOS opto mechanical characteristics Instrument Location Nasmyth B VLT UT3 Melipal Opto mechanical layout 4 beams each a complete focal reducer F 1 88 output Wavelength coverage 0 37 to 1 microns Detectors 4 x 2048 x 4096 15 um pixel Deep depletion E2V 44 82 new Spectral Resolution R 180 to 2500 1 arcsec slit Filters 10 per channel U BVRIz OS red OS blue GG435 GG475 170mm diameter Grisms 6 per channel LR red LR blue MR HR blue 1 HR orange HR red Flexures Activel compensation positioning accuracy lt 0 3 pix during acquisition Masks 10 masks simultaneously loaded in instrument practical maximum per channel EN Slits of any position and shape width gt 0 6 length lt 30 Multiplex 840 simultaneous slits 10 long at R 200 210 simultaneous slits 10 long at R 2000 2500 Integral Field Unit 54 x 54 field 6400 fibers with 0 67 sampling 27 x 27 field 6400 fibers with 0 33 sampling In March 2012 a new set of HR blue grisms were installed Their peak efficiency is 25 compared to 15 with the old grisms The old grisms are not av
144. omatic IP tables to P85 template descrip tion Added img_SkyFlat templates and description Added further explanation to Chapter 4 Introduc tion Release for P85 Phase II 85 0 Sep 1 2009 1 5 3 4 3 4 7 7 Added VIMOS IFU TwFlat template description Normal update for P85 Phase I 84 4 July 1 2009 3 4 3 4 6 x Further offsets description and other minor fixes for Phase II 84 3 March 26 2009 4 5 1 4 6 1 Minor fixes 84 2 March 26 2009 4 5 1 4 6 1 Fixed wrong description of offsets and created new Figs 4 4 and 4 5 continued on next page vi VIMOS User Manual VLT MAN ESO 14610 3509 Table 1 continued Issue Rev Date Section Parag af Reason Initiation Documents Remarks fected 84 1 March 11 2009 4 5 1 4 6 1 4 6 2 Clarified offsets description Modified reference ob 4 6 3 ject position recommendation 83 0 Aug 31 2008 Pre imaging validity comment and release for P83 Phase I 82 1 Mar 05 2008 release for P82 Phase I and II 81 0 Aug 31 2007 release for P81 Phase I and IT 79 1 Dec 13 2006 Added clarifica tions about instrument FoV release for P79 Phase II Phase II 79 0 Aug 23 2006 front page and release for P79 Phase I fonts 78 1 Jun 06 2006 all release for P78 Phase II 78 0 Mar 01 2006 none release for P78 Phase I 77 2 Dec 01 2005 Appendix release for 77 Phase 77 0 Aug 31 2005 3 6 1 3 7 2 release for 77 Phase 76 1 June 30 2005 2 1 2 2
145. onfig uration the ADP has been created It then gives the most important part the CCD to Mask transformation Only then comes the part describing first the reference objects and then all slits defined for this mask with all curved slits appearing at the very end and an example is appended here PAF PAF TYPE paramfile PAF ID 073 A 0602C 3 18 2004 05 12T15 50 59 001 1 HDR START T I PAF NAME wfil6 vm LR Red MlOl adp D PAF DESC ADP Aperture Definition File Pixel PAF CRTE NAME vmmps 3 22 PAF CRTE DAYTIM 2004 05 12T15 50 59 001 PAF LCHG NAME PAF LCHG DAYTIM PAF CHCK NAME PAF CHCK DAYTIM PAF CHCK CHECKSUM 954768746 PAF HDR END INSTRUME VIMOS OCS CON QUAD 1 S FOCUL TEMP 12 31496 S ADF TYPE MOS S ADF VERSION 3 22 S ADF UNIT PIXEL 5 5 5 ADF ID wfil6 vm LR Red MlOl adp ADF COMMENT MASK1 ID 1 TPL FILE DIRNAME SINS ROOT SINS USER ADF TPL FILE KEEP T DATE OBS 2004 04 20T06 57 05 802 TEL GS1 ALPHA 191457 06000 TEL GS1 DELTA 460255 35599 TEL GS1 PPOS NEG ADA POSANG 90 TEL TARG ALPHA 191427 80880 TEL TARG DELTA 460426 61600 TEL TARG EQUINOX 2000 TEL TARG NAME INS ADF GRISM NAME LR Red INS ADF GRISM SPECTLEN 640 INS ADF SKYREG 1 8 INS REF NO 1 INS ARC NO O INS SLIT NO 46 INS SHU NO 4 INS SHU1 P
146. or OB durations of up to 1 5h could be granted if the following conditions are met 1 it was announced in the Phase I proposal that the waiver was going to be requested 2 overheads were estimated during Phase I with both 1h and 1 5h OBs and the same on target total integration time 3 the seeing constraint is not better than 1 2 4 the target does not come too close to the zenith at Paranal 5 there is no scheduling conflict with other programmes at similar RA and 6 the user accepts that the observatory guarantees only the first 1h of execution time to be executed within constraints 36 VIMOS User Manual VLT MAN ESO 14610 3509 Chapter 4 User Guide to the VIMOS Templates 4 1 Introduction This section describes the VIMOS templates Templates are to be used during phase II for service and visitor programmes Templates are high level subroutines writen in TcL to carry out observation subtasks such as preset telescope or take an image Each template has an associated signature file which defines the data type value ranges and information label for each argument to the template e g filter to use integration time dither pattern etc The observer then use P2PP to assign values to the different arguments Templates are grouped together to build an Observation Block OB which typically consists of one acquisition template and one or more observing or calibration templates At the telescope OBs are passed to the instrum
147. ou catalog You are ready to pass to Step 6 Be reminded that at this point all objects with X RADIUS gt 45 pix will be flagged as forbidden Note that instead of doing this interactively inside VMMPS you could have created earlier in the input catalog the column FLAG and assigned the right types except for curved slits to the objects E g if you knew in advance which ones would be your reference objects you could have created a column FLAG in the input catalog and set it to R for these objects S for all others Once you launch the automatic slit assignment Step 6 the task will allocate the slits in the above order of priority i e first put boxes around the reference objects then try to fit all compulsory objects then try to fit all curved slits and than finally assign normal slits as many other objects as possible the latter is described in Step 6 But first let s have a closer look at each type of special objects Reference objects Reference targets should be stellar and typically have R 15 18 The acquisition exposure time should be checked assuming imaging mode with the VIMOS ETC requiring high S N 50 100 It is important to check that they do not saturate in good conditions seeing 0 5 The spectra of these objects are some time the only spectra available to estimate the image quality If reference targets are fainter which is allowed longer acquisition time will be necessary thu
148. our catalog Read through Step O again it is worth it we believe that it will actually save you time OK enough warnings if you went carefully through Step 0 you are ready to start VMMPS To do so type gt vmmps amp and you should see the following window pop up if you are a regular skycat RTD user this will look very familiar to you but notice the new plug in in the top bar VIMOS MPS or VMMPS for people who like acronyms of acronyms Skycat version 2 6 5 1 File View Graphics Go Data Servers VIMOS MPS Cross correlate Contributed and VIMOS Pre image catalogs Assign slits from a VIMOS catalog produce ADP Verify ADP File Zoom Object What to do next depends on the type of catalog you have prepared You have your own pre VIMOS catalog the contributed catalog case e proceed with Step 2 The VMMPS task of defining slits only accepts VIMOS catalogs and you need to produce one Do not panic VMMPS has a built in task to help you doing so using your contributed catalog together with a VIMOS pre image catalog and a VIMOS pre image What you have is a list of RA and DEC of your objects a contributed catalog and the VIMOS pre image freshly out of the ESO pipeline Make sure that the relative astrometry in your catalog is very good and that your list actually also contains on the same astrometric system a good number of objects that appear on the VIMOS pre image e g
149. per Motion Alpha TEL TARG PMD no 10 10 0 Proper Motion Delta Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode INS MODE no IMG Instrument Mode SEQ PRESET no T Preset Telescope T F This template is used to preset telescope to acquire an empty field to obtain twilight flat field exposures It differ from the normal imaging acquisition in its handling of guiding and active optics which are impossible during twilight Only offered in visitor mode The coordinates of the empty field have to be entered in the Target window at the bottom of the P2PP window e Filter The available filters are U R I z e Rotator angle The rotator angle at which the frames will be obtained For normal use this is the angle on the sky e Rotator angle mode Whether the rotator angle refers to angles on sky or to absolute adapter rotator value These last option is for tests only and should not be used by visitors VIMOS User Manual VLT MAN ESO 14610 3509 4 5 3 VIMOS mos acq Mask 43 VIMOS mos Mask tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS ADM1 yes ADM File 1 INS ADM2 yes ADM File 2 INS ADM3 yes ADM File 3 INS ADM4 yes ADM File 4 INS ADP1 no ADP File 1 INS ADP2 no ADP File 2 INS ADP3 no ADP File 3 INS ADP4 no ADP File 4 INS F
150. ranal Science Operations WEB page and the Paranal Observatory home page http www eso org sci facilities paranal http www eso org sci facilities paranal sciops 1 3 Contact Information In case of specific questions related to Service Mode observations and proposal preparation please contact the ESO User Support Department 2 VIMOS User Manual VLT MAN ESO 14610 3509 usd help eso org For visitor mode observations please contact the Paranal Science Operations Team paranal eso org 1 4 Version notes change track The first edition of this User Manual was delivered by the VIRMOS Consortium P I O Le F vre 1 4 1 Version 94 0 For Phase II P94 e Pre imaging users are requested to indicate whether they need a photometric standard in the same night e MOS users are reminded to use similar wavelength in aquisition and science e Clarification of a core dump bug message in VMMPS e Added list of PILMOS features to be aware of 1 4 2 Version 92 1 For Phase II P92 e Added recommendation to use slit width below 4 to avoid saturating screen flats e Added recommendation to use fewer than 500 stars for cross correlation in VMMPS e Now PILMOS observations can be used for the same orientations on sky as the observations prepared with pre imaging e Added date of validity of pre imaging consequence of the CCD upgrade all pre images taken before the intervention are not valid anymore for MO
151. ration frames required for the reduction of science data are reported More calibrations will be taken by the operation staff for the purpose of maintaining and monitoring the instrument configuration and performance e g mask to CCD transformations etc These calibrations are not reported here Calibration monitoring data can be found on the pages of the Quality Control group http www eso org observing dfo quality 3 5 10 MOS mode pipeline See http www eso org qc pipeline status html for the information concerning the latest status of the ESO pipeline VIMOS User Manual VLT MAN ESO 14610 3509 33 Table 3 2 VIMOS Calibration Plan MOS Mode Calibration Number Frequency When Products Accuracy Comment Bias 5 Daily Day time Master Bias RON Darks 3 Monthly Day time Dark current Screen Flats gt 32 Attached to all MOS Flat 5 All used masks science OBs All used masks Screen Arcs 1 Attached to all Dispersion coeff 0 3 pix science OBs Specphot 4 Weekly Night time Response 10 All grism filter Standard Star quadrants function setups used Internal accuracy not considering instrument flexures see Sect 2 7 During the VIMOS Upgrade Project starting February 1st 2010 all MOS science settings must be calibrated at night with at least 3 flats and 1 arc 3 5 11 MOS mode Exposure Time Calculator The approximate limiting magnitudes in MOS mode can be obtained from the exposure time
152. re U B V R L z OS blue OS red GG435 GG475 and no filter Only standard filter grism combinations are allowed as indicated in Table 2 5 Other combina tions should be requested with a waiver Grism This is the grism that 15 used in the first observation template following this acquisition tem plate In order to produce a valid OB the grism in the first observation template following this acquisition template should be the same The available grisms are LR blue LR red MR HR blue HR orange HR red VIMOS User Manual VLT MAN ESO 14610 3509 49 IFU Magnification This parameter defines the sampling and correspondingly the field of view of the IFU The 2 available samplings are 0 67 arcsec fiber and 0 33 arcsec fiber To produce a valid OB the magnification should be the same in every template of the OB Observations requiring different samplings on the same field should resort to several OBs IFU Shutter This parameter is only meaningful when using the LR grisms In LR spectroscopy there are 4 rows of fibers along the dispersion direction whereas in MR and HR modes only one slit of fibers is used hence the variation of field of view between LR and MR and HR modes In case the object of interest is of limited size in LR mode and if only the central 1 4 x 1 4 of the field of view is needed it is possible with the IFU Shutter option to mask the outer part of the IFU always masked in MR and HR modes This al
153. re it applies e g wavelength range the info presented in this Chapter relates to the old detectors 129 130 Figure B l X Efficiency 100 An So 40 100 eu bu 40 1 Central W LORRED Min Sper Peak Fff i WL microns Nienirrd Rave Centro W LOWBLU units Min Speu Peak Elf 0 4 0 8 4 Wi m icrore Efficiency curve of the low resolution grisms LR_red LR_blue The vertical lines mark the desired blaze and the central wavelengths The spectral range refers to the slit in the center of the field The cutoff wavelength is in most cases given by the order separation filters the red CCD limit or the limit of the VIMOS optics in the blue 131 Desirec Maze INES unis central Wavelength 4 E Min Spec Peak ET Ye BOF amp x E E Ka E 40r 20 0 l A Wl microns Figure B 2 Efficiency curve of the medium resolution grisms MR The vertical lines mark the desired blaze and the central wavelengths The spectral range refers to the slit in the center of the field The cutoff wave length is in most cases given by the order separation filters the red CCD limit or the limit of the VIMOS optics in the blue 132 Efficiency of the HR blue grating Transmission HIGHBLU units Desired Blaze centrel 4 Cn Spec Peak Bf Efficiency 1
154. re taken at each offset position 54 VIMOS User Manual VLT MAN ESO 14610 3509 e List of offsets arcsec along the slitandList of offsets arcsec perpendicular to the slit Offsets are defined relative to the previous position Offsets are defined in DETECTOR coordinates along or perpendicular to the slits Usually only offsets along the slits need to be defined If the first image is to be taken at the preset coordinates the first set of offset should be 0 0 If there are less offset values than the required number of offset position the template will return at the beginning of the list of offsets and apply them again For example if 3 telescope offsets are defined Number of Telescope Offsets together with the following series of offsets apparent movement of the target on the detector of offsets arcsec along the slit 12 0 of offsets arcsec perpendicular to the slit List ct ct List the offsets applied to the telescope will be Offsets arcsec along the slit Offsets arcsec perpendicular to the slit or oo the first exposure is taken at 1 from the pointing position along the slit then the second exposure is taken 2 up along the slit with respect to the previous exposure and the third exposure is taken at the same position as the second Observation Category SCIENCE is for a regular science MOS program Never choose TECH NICAL VIMOS User
155. rough a catalog and will be selected at the telescope by the Telescope Operator If this option is selected the subsequent fields Guide Star RA DEC will be ignored Given that the quasi total Nasmyth field of view is accessible to choose a guide star in IFU mode the users can safely use this option SETUPFILE If this option is selected the guide star that will be used is the one for which the coordinates are defined in the Guide Star RA DEC fields As indicated above it is unlikely that this option is required in IFU mode Note that Guidecam tool is not intended to be used to select guide stars for the IFU mode but only for the IMG and MOS mode of VIMOS Guide Star RA and Guide Star DEC Coordinates of the guide star when the SETUPFILE option of the Get Guide Star From field is selected Optional additional velocity moving targets in alpha and delta Additional Velocity RA and Additional Velocity DEC This velocity should be expressed in arcsec sec If no additional velocity is needed these parameters should be left to their default values 50 VIMOS User Manual VLT MAN ESO 14610 3509 4 6 Science templates 4 6 1 VIMOS img obs Offset VIMOS img obs Offset tsf To be specified Parameter Hidden Range Default Label DET WIN1 UIT1 no 1 100000 NODEFAULT Exposure time seconds INS FILT NAME no UB VRIzFree NODEFAULT Filter SEQ CATG no PRE IMAGE SCIENCE TECH Observation Cat
156. rved objects it should be larger than the mean size of the objects for the method to work efficiently 3 6 3 IFU mode grism filter setup In IFU mode only one filter grism combination per OB is permitted Users who want to observe the same targets with different filter grism combinations are requested to submit separate OBs for different filter grism combinations and to consider the respective overheads 34 VIMOS User Manual VLT MAN ESO 14610 3509 Table 3 3 VIMOS Calibration Plan IFU Mode Calibration Number Frequency When Products Accuracy Comment Bias 5 Daily Day time Master Bias RON Darks 3 Monthly Day time Dark current Screen Flats 3 Daily Day time IFU Flat with Monitoring HR Orange Screen Arc 1 Daily Day time IFU Arc with Monitoring HR Orange Screen Flats I to 3 Attached to all IFU Flat 5 science OBs Screen Arcs 1 Attached to all Dispersion coeff 0 3 pix science OBs Specphot 4 Weekly Night time Response 10 grism filter Standard Star quadrants function setups used Internal accuracy not considering instrument flexures see Sect 2 7 2 With grism HR blue 3 screen flats are always mandatory IMPORTANT NOTE The execution time for all IFU attached night time calibrations screen flats and arcs is subtracted from the total allocated time 3 6 4 mode attached night time calibrations To better take into account the effects of instrument flexure the use
157. s if there is a valid scientific reason in that case a waiver must be specifically requested to ESO before you submit your Phase II information For a full report see Sanchez Janssen et al 2012 in the mos section of the VIMOS web pages and the corresponding 2014 A amp A paper 3 4 2 Science imaging Observation Category SCIENCE It is mandatory to set the Observation Category to SCIENCE in the observation template VIMOS img obs Offset see Sect 4 otherwise data will be treated as pre imaging and will not be processed as regular imaging data Rotator on sky Users are encouraged but not forced to use the default value of the rotator angle 1 e 90 VIMOS User Manual VLT MAN ESO 14610 3509 29 Table 3 1 VIMOS Calibration Plan IMG Mode Calibration Number Frequency When Products Accuracy Comment Bias 5 Daily Day time Master Bias RON Darks 3 Monthly Day time Dark current Sky Flats 5 3 7 days Twilight time Master Flat 2 U BVRIz Photom Std 4 quadrants See text Night time Zero Point 596 U BVRI Photom Std 1 quadrant Weekly Night time Extinction coeff 10 U BVRI high airmass Astrometric fields 4 6 months Night time Distortion map 1 pixel R filter 3 4 3 IMG mode calibration plan Observations of IMG photometric standard stars during the same night as science observations are guaranteed for programs that requires photometric conditions and performed on a best effort basis
158. s limiting the available science time When selected the reference objects gets its FLAG parameter set to These objects are put by the software in each mask produced for that quadrant before any other object gets allocated a slit The reference objects are assigned a 4 x4 box Remember that in VIMOS the masks of the 4 quadrant are absolutely stiff with respect to each other Thus in theory it would be enough to have 1 reference object per quadrant or even only 1 quadrant with reference objects The software only checks for 1 reference object in your catalog and lets you get away with it However we recommend to allocate 3 to 4 reference objects per quadrant in a triangular pattern With some experience you will also put them out of the way of slits or slit rows e g in a triangular pattern away from the field center but avoiding the field corners and edges where optical distortions can be large These reference stars will allow you or the astronomer on duty to verify that indeed no single mask is tilted with respect to the three other ones Make sure that these reference objects are centered in their respective boxes as the acquisition software will use the position of these objects for a first estimate of the offset needed to be VIMOS User Manual VLT MAN ESO 14610 3509 91 applied to the telescope to center the mask If you intend to observe very faint objects which will only become visible after the combination of man
159. second order overlap It is strongly recommended to reduce the multiplex number of slits along the dispersion direction to avoid reduce multiple overlaps The multiplex can be reduced by increasing the distance between slits 2 3 4 MOS mode orientation of the field of view The orientation of the instrument field of view in MOS mode can be inferred from Figs 2 3 and 2 4 taking into account that not tilted slits are oriented along the x axes and that wavelength dispersion is along the y axis with wavelength increasing with y 2 4 Integral Field Unit IFU The Integral Field Unit IFU is a dedicated opto mechanical system aimed at producing spectra of a con tiguous area of up to 54 x 54 arcsec For VIMOS set at PA 0 the center of the FoV of the IFU is located approximately 667 due west and 1 6 due south of the VIMOS center field See Fig 2 3 The sky image is projected onto a micro lens array with a choice of two spatial samplings magnifications 0 33 or 0 67 per resolution element fibre Each 2D input micro lens is coupled to an optical fiber The output of the fibers is re arranged on a linear set of micro lenses to produce an entrance slit to the spectrograph at the mask focal 14 VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 7 Spectral coverage AA for three positions of the slit in field of view MOS mode Grism Order sorting AA AA AA filter CCD center CCD bottom CCD top LR red OS re
160. served are FLAG shall be one off C compulsory R reference F forbidden S select able no blank entry is supported any other flag will be replaced by S select able e X WORLD e Y WORLD the coordinates of the object in the world coordinate system not used by VMMPS A IMAGE the semi major axis in pixels e B IMAGE the semi minor axis in pixels THETA IMAGE the position angle in degrees positive x axis is O positive counter clockwise of the object e X RADIUS the object size in the x spatial direction of the detector A IMAGE IMAGE and THETA IMAGE are used to compute X RADIUS The column X RADIUS is used when one or more of A IMAGE B IMAGE THETA IMAGE are not given otherwise it is overwritten If X RADIUS and any one of A IMAGE IMAGE THETA IMAGE is missing the object size in 2 is set to 10 pixels If X RADIUS gt 45 pixels the object is automatically flagged as forbidden by the software Any other columns can be added with any names The catalog later allows to constrain with any number of criteria on any number of columns the select able objects So typically you might want to add color mag nitude or may other columns that you might want to use as selection criteria to the catalog An example is given in Section 6 4 Sometimes a large number of C compulsory targets in the input catalogue results in crashing VMMPS during slit assignment See Sec 6 32 The reason for this is that the manufac
161. set a FLAG to S select able i e erase the special flag you allocated to a object To do this click on a special object then click on the button UNSEL If you look now in the catalog the FLAG has been reset to 5 96 VIMOS User Manual VLT MAN ESO 14610 3509 6 2 7 Step 6 The automatic slit assignment your first ADP draft Once you have at least defined a few reference objects and modified all other flags of your choice in Step 5 you can let VMMPS assign all the slit automatically the relaxing part Select in the catalog window at the very bottom Automatic Slit Assignment This will open the follow ing pop up window with the central sub panel empty to start with mmps poc 1 LR Blue multiplex4 Here you should define several parameters VIMOS User Manual VLT MAN ESO 14610 3509 97 e The Code Method Maximum or Normal Optimization yeah you think maximum is always best but wait The Maximum optimization mode is more efficient in terms of number of placed slits but favors the smallest X RADIUS objects The Normal Optimization is less efficient in terms of number of slits 10 less but does not bias against large objects If you have only point source go for Maximum If you have extended objects you might want to use the Normal optimization to avoid biases in your survey e The Grism the task needs to know the Grism for two reason First in order to define correctly the
162. template including target information followed by observing templates related to the observing mode The users will fill out the parameter fields keywords of the templates e g grisms filters etc All the preparations are done with the Phase II proposal preparation tool p2pp In addition VIMOS masks have to be prepared with the VIMOS mask manufacturing preparation software VVMPS The strategy behind observing blocks and templates is to prepare the observations well in advance to minimize any interactive steps during the observations optimization and service mode compatibility The execution of the OBs will be mostly automatic and the execution will be done by telescope and instrument operators and staff astronomers Direct interaction at execution time is only required for the target acquisition and to assess the quality of the data The preparation of visitor and service mode observations requires special care and the relevant documen tation should be consulted See Sect 1 for instructions on how to retrieve the corresponding information 3 1 Interactions with the telescope 3 1 1 Guide stars and Guidecam software tool For telescope guiding and active optics correction it is necessary to find a guide star of magnitude in the range 11 13 5 within the Nasmyth field of view The guide star is picked up by the Guide Probe which is attached to the telescope adapter Note that the adapter is independent of the telescope rotator
163. ter used The correction of the distortions is derived from R filter observations only To minimize the slit losses due the distortion of the instrument which do depend on the filter the use of the R filter for pre imaging is mandatory Although pre imaging with filters other than R is not permitted users may request with a waiver to include in the pre image OB templates with other filters Note that no waiver will be granted to design MOS masks based on observations done with any other filter but R Observation Category PRE IMAGE Itis mandatory to set the Observation Category to PRE IMAGE in the observation template VIMOS img obs Offset see Sect 4 Failure to do so will result in delays to process and deliver the pre imaging data and may result in MOS observations not being carried out Standard star requirements Users requesting CLR sky transparency for their pre images are requested to specify in the README whether they need a photometric standard taken in the same night Rotator on sky rotator on sky otherwise called position angle on sky cannot be freely selected by the user This is in order to minimize the slit losses due to atmospheric dispersion Sect 3 2 The default value is 90 and corresponds to an orientation of the field with E to the top N to the left see Fig 2 3 For declinations in the range 45 lt lt 5 this parameter can be set to 0 In special cases it may be possible to deviate from these rule
164. ters should be requested with a waiver Defining in the OBs combinations not allowed and for which a waiver was not requested will result in these OBs being rejected by the User Support Department ESO will not accept requests for user s supplied grisms or filters for VIMOS Table 2 5 Standard grism filter combinations for MOS and IFU modes Grism Filter Default Allowed LR blue OS blue Free LR red OS red Free MR GG475 OS red OS blue HR blue Free HR red GG475 HR orange 66435 The grism filter configuration MR OS blue is not supported by the pipeline Transmission curves are available in Appendix A for the filters and Appendix B for the grisms Table 2 6 shows the spectral range covered in MOS modes by the various grisms as obtained by the automatic pipeline when reducing the spectrophotometric standard star This spectral range is valid for slits close to the center of the field 2 3 1 Spectral range In MOS mode slits can be positioned at any position in the imaging field of view With high resolution grisms the observed spectral range changes slightly according to the position of the slit Table 2 7 shows the spectral ranges covered by the detector for each grism and default order sorting filter for the slits located at three different positions in the field of view With low resolution grisms the spectral coverage is independent of the position of the slit 2 3 2 Spectral res
165. the new VPH high resolution blue grism 1 4 7 Version 87 0 The first part of the VIMOS upgrade project is almost finalized The instrument has been furnished with the new set of red sensitive detectors new bi stable shutters and an active flexure compensation system As usually happens in a project like this we could expect things to initially evolve before the finally stabilize so please visit the VIMOS web pages for the latest news 1 4 8 Version 86 0 As of this writing Feb 2010 the new detectors have been characterized and the upgrade is planned to take place between the months of June and July 2010 If everything goes according to schedule we will have the new detectors and new shutters in normal use during P86 The upgrade project includes the implementation UP of an active flexure compensation mechanism which nevertheless will need to be tested before we can commit to a final date for its installation and commissioning The Service Mode policy for granting a waiver for OB duration longer than 1h has been clarified See Sec 3 7 4 VIMOS User Manual VLT MAN ESO 14610 3509 1 4 9 Version 85 0 This manual is for P85 preparation As of this writing 1 Sep 2009 there is an ongoing project that aims to upgrade the instrument It is foreseen that the detectors will be changed by new deep depletion e2v detectors which will increase the red response and reduce fringing The changes to the instrument will imply that once VIMOS is upgrad
166. the program to crash Use of cat or data should work Some example are given in Section 6 4 The input format is e a header line with column names in upper case There is no preferred order for the columns but see below for mandatory and reserved columns and names The first line is interpreted as the header line Header entries are separated by any number of blanks or tabs A lt return gt marks the end of the header line e the data sorted in columns Columns on a given row are separated by any number of blanks tabs The number of columns and their order is defined by the number of entries and their order in the header No empty field is allowed you will get a friendly warning if some are present The first row after the header line defines the beginning of the data the last row defines the end of data Each row must end with a lt return gt Note that no blank lines are allowed in the catalogs and you will get a friendly warning if any is present 76 VIMOS User Manual VLT MAN ESO 14610 3509 Getting your contributed catalog into the right format Assuming that you are in case 1 you need to prepare the two catalogs described in the this and the next section The contributed catalog contains all the objects that you wish to observe plus or including enough bright objects that can serve for the cross correlation with the VIMOS pre image catalog and as reference stars for aligning the mask on the sky All objec
167. the transformation matrix in the header of the pre image will be transformed into ra dec and cross correlated with the contributed catalog This will in turn allow to determine the best transformation from ra dec of the contributed catalog to x y in pixel coordinates ID X_IMAGE Y_IMAGE 1 76 554 497 854 2 78 566 258 963 3 100 714 83 548 4 445 282 69 524 5 1589 924 51 424 6 767 057 126 705 7 1391 149 121 832 8 229 290 137 767 9 800 180 157 851 10 1010 375 156 611 11 1351 326 177 466 12 980 319 184 649 13 548 496 1327459 14 1534 235 204 140 1 9 1310 024 208 284 16 823 662 233 845 424 209 727 220 367 18 706 188 228 008 19 1797 688 243 397 20 1936 794 250 302 21 1996 330 245 786 22 1046 128 251 701 23 1737 622 249 215 24 2018 013 246 191 25 13512735 266 140 VIMOS User Manual VLT MAN ESO 14610 3509 109 6 4 3 A VIMOS catalog The following shows an example of a VIMOS catalog ID X IMAGE Y_IMAGE A IMAGE B IMAGE THETA IMAGE MAG MAGERR 1 233 036 549 141 66 651 4 305 89 9 7 4464 0 0001 2 891 844 360 569 53 105 7 911 87 4 7 9803 0 0002 3 100 238 84 054 60 828 24 763 T0 0 11 2455 0 0053 4 601 006 07 561 428 910 9 910 70 76 9 9766 0 0027 5 1708 843 51 041 104 712 12 829 19 99 0000 99 0000 6 978 752 48 620 12 126 4 303 12 4 12 7846 0 0047 7 1209 031 74 048 28 632 6 684 2 9 10 1976 0 0008 8 1212 790 59 548 18 403 5 1
168. tigmatism and make the the field curvature concentric with the new exit pupil e The Pupil Relay Lens PRL correct for pupil chromatism in combination with FPAL and is the first element of the collimator e The folding mirror section hosts the four folding mirrors and the active flexure compensation system AFC e The Filter Camera Section the beam section hosts optical analyzers filters and or grisms the camera and the exposure shutter in front of the CCD VIMOS User Manual VLT MAN ESO 14610 3509 7 7 P W o Mirrors AS Ge 1 COSME Main Structure MOS Mask Figure 2 2 VIMOS Opto Mechanical layout e FEU the filter exchange units one for each channel are located below the main body of the instrument below the Filter Camera Section each unit hosts a juke box of 10 filters permanently mounted e GEU the grism exchange units are located below the Filter Camera Section they host a carousel of six grisms permanently mounted 2 1 1 Instrument modes VIMOS offers the following modes for observations e Imaging IMG see Sect 2 2 e Multi Object Spectroscopy MOS see Sect 2 3 e Integral Field Spectroscopy IFU see Sect 2 4 IMG mode uses broadband filters U BVRIz MOS and IFU modes share the set of grisms and order sorting filters given in Table 2 5 In MOS mode a set of four laser punched masks made of INVAR is inserted in the focal plane In mode fibers are arranged along pseudo sli
169. ting point 6 2 3 Step 2 Transforming your contributed catalog into a VIMOS 6 2 4 Step 3 Starting VMMPS for mask definition 6 2 5 Step 4 optional Re arranging your VIMOS catalog 6 2 6 Step 5 First pass defining special objects 6 2 7 Step 6 The automatic slit assignment your first ADP draft 6 2 8 Step 7 optional Fine tuning last corrections 6 2 9 Step 8 Writing your final ADP you have succeeded 6 2 10 Loading verifying previously defined mask 6 3 Troubleshooting and known 6 3 1 Cross correlation task o cs ee ee EEN ss 6 3 2 Automatic slit 6 3 3 Problems with checksum in ADP files 6 3 4 ADP files and FITS headers G3 Other problems a so o Re ee a 64 Examples of catalogs ssa sa a a 64 1 Acontributedeaalos 22 22 2222 44 2 6 4 2 A VIMOS pre image catalog 64 3 A VIMOS pus sp me on emo RO E A 644 Example of an ADP file 65 Usereheck list pa ao dados hades a a 7 PILMOS Tutorial ZE cds a cx ed 7 2 Requirements and Restrictions for PILMOS usage Tal WResiichots 222442554 A 1 4 2 Catalogue requirements 2 lt s so so ERR x pa das 123 Other Consideratlons uo mo om mmm mmm oa a pais 7 24 PILMOS features to be a
170. tion and science template Users who want to observe the same targets with different filter grism combinations are requested to submit separate OBs for different filter grism combinations and to consider the respective overheads Another important thing to keep in mind is that the filter in the acquisition template should overlap in wavelength with the central wavelength of the grism used in the science exposures 3 5 8 MOS mode attached night time calibrations To correct for the effect of instrument flexures it has been mandatory in previous periods to attach nighttime arcs see Sect 3 5 9 for all MOS OBs using LR blue and HR blue grisms During the VIMOS upgrade one attached arc and three flats have been mandatory for all MOS science observations The upgrade has been finished and the new HR_blue grisms have been commissioned We are currently P93 analysing day and nighttime calibrations taken since the upgrade for all grisms to analyse whether nighttime calibrations can be skipped in future periods as of P95 Look at the instrument web pages for the latest information on required calibrations 3 5 9 MOS mode calibration plan Observations of spectro photometric standard stars with the same setting used for science observations are normally performed within a few nights from the science observations Table 3 2 indicates the nature frequency and accuracy of VIMOS calibrations expected to be taken as part of the calibration plan Only calib
171. tional You can skip it and move directly to Step 8 The end of Step 6 triggers your old catalog window to disappear and a new one to appear as well as the display to be updated The new catalog window looks as below and is composed of as many virtual layers as you defined masks for this quadrant You can step from window to window under Modify Slits Choose Mask and use the pull down widget that also indicates on which mask you are currently working field 1Q1M 1 adp_cat 1 Search Options Object Name Equinox 2000 080007778 S 10 44 14 87 Radius 60 Radius 5 45 Max Objects 10000 0 33 2256 72 404 429 774 408 1355 84 1855 49 351 687 1163 35 1729 13 2156 54 9 E 275 173 10 44 54 29054 1028 DONO NO AD XO XO AD AO XO ADO Fiter Modify Slits canoso most Clear Slits Make Final 3 ADPs At this point your ADP has been written and you could proceed without changes to Step 8 Optionally if you have the feeling that you should modify a few slits proceed with this Step but be careful you are about to try to optimize an optimization The window lists all objects that have been allocated slits in this mask You can now do the final careful fine tuning by unselecting some slits Unsel and or adding some slits New Slit following the procedure that should start to be familiar to you by now interacting with the d
172. to which the instrument is attached In IMG and MOS mode it is not always possible to find a guide star such that the Telescope Guide Probe is not vignetting the VIMOS field of view Therefore users are asked to select the guide star for their observations The choice of the guide star is mandatory for pre imaging for imaging and for MOS programs In fact the guide star for MOS is selected at the time of pre imaging guide star coordinates and position of guide probe whether positive or negative is stored in the pre image headers and subsequently propagated by VMMPS see Sect 3 5 4 the VIMOS Mask Preparation Software into the Aperture Definition in Pixel ADP files At the time of the MOS acquisition this information is extracted from the ADP and the same guide star as used for pre imaging will be re used for the MOS observations Only in some exceptional cases the telescope operator may have to select another guide star see Sect 5 6 Users need not select the guiding star for observations carried out in IFU mode The position and the amount of the vignetting introduced in the field of view by the guide probe can be controlled with the help of the Guidecam software that allows one to visualize the geometry of the VIMOS field of view and of the Telescope Guide Probe and allows users to select the guide star most suitable to their observation For a tutorial on the use of the Guidecam tool see Chapter 5 25 26 VIMOS User Manual VLT
173. together with the transmission curves are presented in Appendix A see Ta ble A 1 and Figs 1 to 4 ESO will not accept requests for user s supplied filters for VIMOS 2 2 4 Zero points The measured zero points with the new detectors referred to airmass 1 are given in Table 2 3 We have estimated isophotal magnitudes and effective wavelengths for Vega and are given in Table 2 4 together with the expected zero points VIMOS User Manual VLT MAN ESO 14610 3509 9 VIMOS FoV Orientation ON SKY IMAGING MOS MODE DE e Rotat le PA 0 gt 6 otator angle Figure 2 3 Orientation of the VIMOS field of view on sky for position angles 0 and 90 in IMG and MOS mode Uncertainties due to the poorly defined distortion solution at the very edges of the detectors and the fact that the edges are not straight when projected onto the sky are of the order of 1 5 pixels Distances are not shown to scale refer to the given values 10 VIMOS User Manual VLT MAN ESO 14610 3509 VIMOS FoV ORIENTATION ON SKY ON DETECTOR Rotator angle PA 0 Rotator angle PA 90 E N Q2 QL Q2 Q3 y 1 N a Figure 2 4 A comparison of the orientation of the VIMOS field of view as seen on sky left panels and on detector right panels for position angles 0 and 90 in IMG and MOS mode VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 3 Measured imaging zero points Filter Q1 Q2 Q3 Q4 U 26 18 2
174. ton Plot Slits and Clear Slits to overlay the actual slits and expected spectral range NGCzz43 259 0 24280 gr j 200 Bitpix Low 5000 High 15000 futo Set Cut Levels 21 Grid m Once you have made the last modifications you can move on to the next mask of the series as described above Choose Mask and once you are through with all masks proceed to Step 8 You are almost there 102 VIMOS User Manual VLT MAN ESO 14610 3509 6 2 9 Step 8 Writing your final ADP you have succeeded Once you are happy with all slits in all masks for this quadrant click on Make final X ADPs where X indicates the number of masks you have defined and for which ADP files will be written A last pop up window will appear Choice 1 What do you want to do now Next Quadrant Exit VMMPS and ask you whether you want to restart VMMPS pass e g with the next quadrant exit VMMPS after saving all results go back to Step 7 you forgot something Depending on your choice you will go back to Step 1 go have a coffee or go back to Step 7 respectively In the first two cases you will have created ADP files ready to be attached to OBs CONGRATULATIONS IMPORTANT note that an APD file edited by hand e g changing the slit width with a text editor will not be accepted by p2pp Based on the slitlets position and size a checksum parameter is calculated wh
175. ts in the so called masks IFU masks may be inserted in the instrument focal plane in the same position as occupied by the INVAR masks in the MOS mode The spectral characteristics of the two spectroscopic modes are therefore similar but not identical 2 20 Imaging Direct imaging is used to produce wide field images VIMOS is the largest imager on the VLT using broad band filters This mode has two functions produce science images and produce the mandatory pre images needed to prepare masks see Sect 3 Table 2 2 summarizes the main characteristics of the imaging mode 8 VIMOS User Manual VLT MAN ESO 14610 3509 Table 2 2 VIMOS Imaging characteristics and performance Field of view 4 7 x 8 separated by 2 gaps illuminated CCD area 2048 x 2350 pix slightly vignetted on the corners outside 8 Pixel scale 0 205 pixel Image Quality 80 encircled energy in 0 4 within the whole field Filters U Bessel BVRI Gumn z see section A F ratio 1 88 Limiting Magnitudes in NI U 26 2 B 27 4 V 27 0 R 26 6 1 25 5 z 25 1 1 5c detection for an AO star 2 2 1 IMG mode orientation of the field of view Figure 2 3 shows the orientation of the instrument field of view At 0 rotator angle the x axes of the CCDs are oriented E and the y axes are oriented N top panels Increasing rotator angle moves the compass coun terclockwise At 90 rotator angle the x axes are oriented S and the y axes are oriented E
176. ts shall have very good 0 1 relative astrometric coordinates RA DEC on a single astrometric system Further all objects should have good 4 2 absolute astrometry The mandatory columns in the contributed catalog are in upper case and in the following order e ID defining an identifier which must be unique and integer number e RA the right ascension of the object e DEC the declination of the object RA and DEC entries are allowed in 2 formats decimal degrees hh mm ss s including semi colons The optional columns whose names are reserved are CROSS is for each object either 1 1f that object shall be used for the cross not e g if it is too faint too elongated We recommend not to flag CROSS 1 FLAG shall be one of upper or lower case compulsory R reference F correlations or O 1f it shall more than 500 stars with forbidden S select able no blank entry is supported any other flag will be replaced by S select able e A ARCSEC the semi major axis in arcsec e B ARCSEC the semi minor axis in arcsec PA the position angle E of N in degrees The latter three will be used to compute the object size in pixels along the i e along the slit Any other columns whose names do not match the above ones x direction of the detector Notice only that the names X IMAGE and Y IMAGE are forbidden in the contributed catalog they are used computed later If these ar
177. turing machine can cut slits of max 20 length which translate into 150 pixels and that you are allowed to add 10 50 pixels sky on each side VIMOS User Manual VLT MAN ESO 14610 3509 79 Getting your input VIMOS pre image ready Good news your VIMOS pre image processed by the ESO pipeline is ready as it is to be accepted by VMMPS Bad news if your image is not a VIMOS pre image or if the VIMOS pre image was not processed by the ESO pipeline or if you had the bad idea to modify the pre image even in the slightest detail e g you had 3 dithered pre images and combined them etc VMMPS will not work or in the worse case will seem to work but you will never see your objects through your slits Indeed your x y coordinates will be transformed into mm for the mask manufacturing using the transformation in the image header assuming that we deal with original x y system If you have translated re scaled re binned the image this transformation will not be valid anymore Just make sure that your input to VMMPS is an original untouched VIMOS pre image processed by the ESO pipeline Also make sure that this pre image is the one you used to obtain the coordinates for your VIMOS pre image catalogor VIMOS catalog If you would like to learn a little about the background of all this read further this section If you are just lazy and happy with the information so far go ahead and jump now to Step 1 Some Background One of the VIMOS cal
178. turn at the beginning of the list of offsets and apply them again For example if 5 telescope offsets are defined Number of Telescope Offsets together with the following series of offsets apparent movement of the target on the detector in RA DEC RA 0 10 5 0 5 10 the offsets applied to the telescope will be see also 4 6 1 RA 0 10 5 0 10 DEC 0 5 10 0 5 Observation Category SCIENCE is for a regular science IFU program Never choose TECH NICAL VIMOS User Manual VLT MAN ESO 14610 3509 4 7 Calibration Templates 4 7 1 VIMOS img_cal_FlatSky 57 VIMOS img cal FlatSky tsf To be specified Parameter Hidden Range Default Label INS FILT NAME no UB VRIzFree NODEFAULT Filter SEQ EXPLEVEL no 200 40000 10000 Exposure Level SEQ NEXPO no 1 1000000 1 Number of exposure SEQ OFFSET1 LIST no 0 List of offsets arcsec in X or SEQ OFFSET2 LIST no 0 List of offsets arcsec in Y or TEL ROT OFFANGLE no 180 180 0 Se angle TEL ROTATIO no NONE ABSOLUTE NONE Rotator angle mode SKY AB SOLUTE Fixed values Parameter Hidden Value Label DET READ MODE no Medium Readout mode DET WIN1 BINX no 1 X binning factor DET WIN1 BINY no 1 Y binning factor INS MODE no IMG Instrument Mode SEQ HEADER no F Header type SEQ NOFF no 1 Number of Telescope Offsets This template takes SEQ NEXPO dithered sky flat frames
179. unts with an active flexure compensation system AFC With this system we can position the the image of the FoV onto the detectors to better than 0 3 pixels at the start of the exposure in all quadrants and maintain it there to within 1 5 pixels for the duration of a typical exposure Previously VIMOS had astatic compensation systems installed only in quadrants 2 3 and 4 Image motion due to instrument flexures under gravity were approximately 4 pixels peak to valley for the channels 2 3 and 4 and 6 pixels peak to valley for channel 1 for the whole rotator range With the new AFC we can control flexures to better than 221 5 pixels peak to valley for the range of rotations in typical exposures 2 8 Known shortcomings of the instrument VIMOS problems issues and hints related to data reduction and observations are summarized on the VIMOS QC page http www eso org observing dfo quality VIMOS ServiceMode ServiceMod html problems Here we list some shortcomings that should be be kept in mind when planning the observations 2 8 1 masks vignetting in IMG and MOS mode When using VIMOS in IMG or MOS mode the IFU masks are stowed The mechanism that hold the heavy masks in place have some play and depending on rotator angle they could be seen vigneeting the FoV When stowed the IFU masks are folded into the center of VIMOS that is the area between Q1 Q4 and Q2 Q3 The hinge is near the focal plane while the far end of the m
180. used for cross correlation their position in pixel coordinates shall be as good as possible Therefore make sure that you selected bright S N gt 10 unsaturated if possible point like objects on the VIMOS pre image This comparatively short list of objects shall be written in an ASCII file with the mandatory columns in upper case and in the following order e ID an identifier which must be unique and integer number e X_IMAGE the x coordinate in pixels on the VIMOS pre image e Y IMAGE the y coordinate in pixels on the VIMOS pre image Any other column is allowed but will be ignored and not carried over to the output catalog of the cross correlation see Step 2 The example of a VIMOS pre image catalog is given in Section 6 4 78 VIMOS User Manual VLT MAN ESO 14610 3509 Getting your VIMOS catalog into the right format Assuming that you are in case 2 you only need to prepare a VIMOS catalog The VIMOS catalog can e either be obtained as an output of the cross correlation task in which case you do not have to worry it will have the right format e or be produced directly by the user that s you from the VIMOS pre image The mandatory columns are in upper case and in the following order e ID an identifier which must be unique and integer number e X IMAGE the x coordinate in pixels on the VIMOS pre image e Y IMAGE the y coordinate in pixels on the VIMOS pre image The optional columns whose names are re
181. ve x axis points to the East and positive y axis points to the North see Figure 2 3 then if the target falls at pixel x 0 y 0 after the telescope preset then the first exposure will be taken with the target at pixels x 0 y 10 offset is 2 which is approximately 10 pixels In other words the telescope was moved 10 pixels in the directions of the negative y axis of the detector Then the target will move to the pixel position z 50 and y 45 on the detector which corresponds in this case PA 0 to a movement of the telescope of 10 East and 7 South with respect to the previous position etc until the number of exposures is reached If the rotator angle is different than zero for example PA 45 and the offset defined in P2PP are RAorX 0 2 0 0 2 0 DEC or Y 5 0 5 5 0 5 then Figure 4 4 and Figure 4 5 give the performed offsets in the case of DETECTOR and SKY coordi nates respectively 52 VIMOS User Manual VLT MAN ESO 14610 3509 e Observation Category PRE IMAGE is for a pre imaging program SCIENCE is for a regular science imaging program Never choose TECHNICAL It is critical that users define PRE IMAGE for pre imaging programs Failure to do so will result in slow processing and delivery possibly till the end of the period of the data 1 1 X Figure 4 4 Illustration of the VIMOS img obs Offset template The black dots represent the position of a star which was originally at th
182. ware 7 3 Preparation of MOS Masks without a Pre Image 7 3 1 Step 1 PILMOS Guidecam 32 DEPL VMMPS o sc 6G eem aa Appendices A VIMOS Filters B VIMOS Grisms VIMOS User Manual VLT MAN ESO 14610 3509 xi Old detector system until May 2010 135 Cl Detector Tead 0ULMOdES cs sa sd mo ME EE 504 9 a O 135 C2 Controllers ss xu spams how A a 136 Read out noise and conversion actor 2 4 136 Co BRO ke a EO Re SA O ko de 136 D Abbreviations and Acronyms 137 Index 139 xii VIMOS User Manual VLT MAN ESO 14610 3509 List of Figures 2 1 2 2 2 3 2 4 25 2 6 Dd 2 8 29 2 10 34 E 4 1 4 2 4 3 4 4 4 5 dl 3 2 a 74 A 2 A 3 4 B 1 B 2 B 3 B 4 B 5 VIMOS Optical layout 242 80 Rr Rn E 5 VIMOS 2 22 7 VIMOS FoV orientation relative positioning sky 9 NVIMOS FOV orentallol uoo ac vu oe EE A O ee A 10 VIMOS IFU FoV onentoliDU sumus 4 n PR ada o Se ee 15 VIMOS IPU ottefitatlot 2 2 oc so A A 15 IPU details of the ber modules 22222222 a ae 16 IFU fiber aumibering scheme 1 sass seo a A A AR 16 Quantum efficiency vs A for new
183. x ES EUROPEAN SOUTHERN OBSERVATORY Organisation Europ ene pour des Recherches Astronomiques dans l H misph re Austral Europ ische Organisation f r astronomische Forschung in der s dlichen Hemisph re ESO European Southern Observatory Karl Schwarzschild Str 2 D 85748 Garching bei Miinchen Very Large Telescope Paranal Science Operations VIMOS User Manual Doc No VLT MAN ESO 14610 3509 Issue 94 0 Date 24 05 2014 VIMOS IOT S Mieske F Sel Prepared Edited E 222 e SSC Se eet A S Approved ee EE C Dumas Released ta etn EE Date Signature 11 VIMOS User Manual VLT MAN ESO 14610 3509 This page was intentionally left blank VIMOS User Manual VLT MAN ESO 14610 3509 111 The list of people that have contributed to the preparation of this document at different times includes Stefano Bagnulo Stephane Brillant Paul Bristow Peter Hammersley Gianni Marconi Steffen Mieske Marina Rejkuba Ruben Sanchez Janssen Fernando J Selman VIRMOS Consortium Burkhard Wolff 1v VIMOS User Manual VLT MAN ESO 14610 3509 This page was intentionally left blank VIMOS User Manual VLT MAN ESO 14610 3509 Table 1 Change Record Issue Rev Date Section Parag af Reason Initiation Documents Remarks fected 94 0 May 24 2014 3 4 1 3 5 7 6 3 2 Minor changes after MHI feedback 7 2 1 7 2 4 92 2 May 28 2013 2 4 1 2 8 5 3
184. y exposures then assign a few slits per quadrant to brighter objects which will serve as a sanity check for the observer The presence of those objects in the resulting spectral frame will indicate that indeed everything went fine with the observations New reference objects It can happen that no object in your catalog is suited as reference star In that case you can add reference stars directly from the image This should not be the default mode since the centering algorithm used by skycat is a simple Gaussian fit and might not return as accurate coordinates as your favorite source detection program The whole alignment procedure on sky might become less precise than with a well prepared catalog including reference stars on the same system as your scientific targets The procedure to pick new reference stars is a little different than the one described above since obviously you cannot click on a highlighted object in the image display In order to select new reference stars just click on the button New Ref This will pop up a Pick New Object window see below Move the cursor in the image display onto the object you want to select and click Then select ENTER in the panel below to enter the object in the catalog Pick New Object 1 Area of image to be examined Image Statistics Image X 7973 Image Y 16007 08 00 04 627 d 10 42 56 76 Equinox 2000 om om
185. y RA TEL TARG ADDVELDELTA yes 15 15 0 Additional Velocity DEC TEL TARG ALPHA no 0 TEL TARG DELTA no 0 TEL TARG EPOCH no 2000 3000 2000 Epoch TEL TARG EQUINOX no 2000 3000 2000 Equinox TEL TARG OFFSETALPHA no 3600000 3600000 0 Alpha offset TEL TARG OFFSETDELTA no 3600000 3600000 0 Delta offset TEL TARG PMA no 10 10 0 Proper Motion Alpha TEL TARG PMD no 10 10 0 Proper Motion Delta Fixed values Parameter Hidden Value Label INS MODE no IMG Instrument Mode SEQ PRESET no T Preset Telescope T F This template is used for acquisition of a field in imaging mode for both imaging and pre imaging OBs Figure 4 1 shows the P2PP window with an example of a pre imaging Observing Block OB The coordinates of the target have to be entered in the Target window at the bottom of the P2PP window The parameters for this template are the following Filter This filter should be the same as for the subsequent imaging template The available filters U R I z Rotator on sky This parameter defines the orientation of the instrument on sky For pre imaging templates this parameter should be set either to 0 if declination lies in 45 lt 6 lt 5 or to 90 for any declination unless a waiver is requested The orientation corresponding to 90 is such that the slits will be oriented N S for the spectroscopic follow up Alpha offset andDelta offset These are optional additio

Download Pdf Manuals

image

Related Search

Related Contents

  Considérations sur les démarreurs  PL・SAー 。sPE  Guías de instalación v1.1 Benutzerhandbuch v1.1    Manual del usuario    AEG Protect Home. 600 VA  TiX640, TiX660, TiX1000  Hardware Manual - RTD Embedded Technologies, Inc.  

Copyright © All rights reserved.
Failed to retrieve file