Home

Phone (922)605200 -Fax (922)

1. 18 4 5 GHOSTS A 20 4 6 CALIBRATIONS REQUIRED al nn lb laica 21 4 6 1 Paralellism calibration issues 21 4 6 2 Wavelength calibration ss 22 4 7 TUNING ACCURAC a A iss un aaa ua sete tea in 23 4 8 TUNING SPEED 1 E 23 4 9 TUNING RANGE um ai 23 4 10 OSIRIS TE EON iis td umaku ohay dai 24 5 OBSERVING STRATEGIES 24 5 1 SEEECTING OEE BAND Cati ii atan 25 5 1 1 Continuum subtraction ora nnnn nn nn sers 25 TABLE OF CONTENTS Page 5 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 5 1 2 Off band wavelength and FWHM tuning selection 25 52 DEBLENDING LINES custodiada 26 5 3 ON LINE FWHM 000 4 27 5 4 DECIDING TARGET POSITION AND ORIENTATION 2 28 5 5 REMOVING GHOSTS COSMIC RAYS AND COSMETICS 29 321v Half field masking aa dero er rte ROREM 29 5 5 2 Azimuthal dithering 29 5 53 TF tuning dithering pattern ss 30 5 6 TUNABLE TOMOGRAPHY Gu la ta didas 30 IO Techniques da o o a e 30 5 6 2 applications sise 31 5 7 BAND SYNTHESIS TECHNIQUE benne er ainia 31 E A
2. 31 5 7 2 SOME applications a 32 5 6 SOURCES OF INSTRUMENTAL PHOTOMETRIC ERRORS 32 6 STARTING UP OSIRIS TF S sss ges 33 7 CALIBRATING OSIRIS TE 34 7 1 PARALLBEDISM cec tree tn Tex c re dee ee ub EP te ele t 34 7 1 1 General considerations sse eene 34 7 1 2 OSIRIS TF parallelization procedure ss 34 213 Lackofparallelisii i 43 7 1 4 Troubleshooting ss 44 11 41 The X or Y Sweep d0es Pinamar id 44 71 42 X Y or Z give very different values in consecutive exposures or even mirrored rings 44 7 2 WAVELENGTH 7 45 7 2 1 General considerations ss 45 7 2 2 Calibration using the ICM eene 46 7 2 3 Calibration using astronomical targets 46 7 2 4 Calibration using sky rings ss 47 752 3 Tradeoffs sca OT RERO NES eh A ETE 49 7 2 6 Measuring TF FWHM AA 49 7 3 EBUX CALIBRATION aura yak Aka akui tete ite tatnen rat 49 8 OPERATING OSIRIS TE seems 50 8 1 TESETUP anay E t em ie 50 8 2 STANDARD EXPOSURES rnt ete ees teri ena ete reser dint 50 8 3 SHUFFEED EXPOSURES uti st it elle den tati od eerte ed ode tue ceeds 53 8 4 EAST MODES A aie 5
3. 3 SUI OD TECHNOLOGY DIVISION Draft TFMANUAL June 28 2009 Project Ref PROJECT DESTINATION OSIRIS TITLE TF USER MANUAL INSTITUTO DE ASTROFISICA DE CANARIAS 38200 La Laguna Tenerife ESPA A Phone 922 605200 Fax 922 605210 TF USER MANUAL Page 2 0170 Date June 28 2009 Code Draft File TFMANUAL DOC AUTHOR LIST Name Function Jordi Cepa OSIRIS Principal Investigator APPROVAL CONTROL Control Function Revised by Approved by Change Description 28 06 09 First Draft Page 3 0170 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC SUMMARY This document provides a general description of the basics concepts about Tunable Filters plus brief and simple practical instructions for designing observations for different types of targets and scientific cases and for calibrating and operating OSIRIS Tunable Filters As instrument commissioning and operation progresses it is expected that this document that complements the OSIRIS user manual will be updated accordingly The section on charge shuffling and fast modes is still pending This document is still a draft Page 4 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC AUTHOR LIST ayway sao 2 AP
4. 9 1 2 Day time routines Once selected the required wavelength and Z range for the proposal to be observed during the night check the LOG for the corresponding XY values Were XY not available for this pair Z it is necessary determining them For this purpose e Choose an emission line from the lamps available at the GTC ICM see C 1 with a wavelength as near as possible to the line to be observed Then choose the mode so that the 2 15 as near as possible the 2 required Use the TF calculator to this purpose e Insert the ICM and switch the lamp on and the dome lights off e Insert the 2 5 long slit specific for tuning the TF e With the default XY scan in Z up and down near the Z provided by the TF calculator till the line is located Sec 7 2 2 e Determine X and Y for this wavelength and Z ranges as described in Sec 7 1 2 and note it down in the LOG These values must be logged since XY are very robust and once determined do not need to be determined again The same line and mode will be used for Z calibration Sec 7 2 2 9 1 3 Practical TF operation and calibrations during the night e While pointing to the target insert ICM and switch on the required lamp for Z calibration see 7 2 2 e Set XY for the line from the LOG or determined during the day e Once the telescope is pointing to the target and the rotator angle has been set and stable avoiding the rotator range from 30 to 60 perform a Z calibratio
5. tacon Stem som ous 6080 Corral cher Conta GIDAS Lon SECA i a 3 ta lt E calif SEQUENCES 0008 AREA OPTIONS mess as mue Oceni omua cen owas couz owna con 2 ous pow pem ae cept 0 02 as Dymo 5 I I a 2098 meme x bent AER E Lm LEONARD 5735 z 0 CR El 09 Laund j T tsimpte events Receiving ima 2 saoimage dse cube yinspector v lchamnelsena mmm 4 EEE wea reo 3 07 am en ermina conned simple Event s i g Ima Saoimage dsa Cube aJimspect Be Figure 7 4 CCDs AREA flap of Data Acquisition Window flap of the OSIRIS DAS menu e SEQUENCES of Data Acquisition Window of the OSIRIS DAS menu choose Shutterless exposure type Figure 7 5 just opens shutter when starting and closes it after the last charge shuffle Page 38 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC e Set lines to 75 number of lines to shuffle from exposure to exposure This is enough for not spilling over the detector but keeping a healthy distance between slit im
6. Now OSIRIS is doing the following programming DAS clearing CCD opening shutter exposing shuffling charge while changing TF tuning to the next Z by a step AZ and so on After the last exposure N it does a last shuffle and closes shutter When the sequence is finished the TF 15 kept Swinging awaiting for further synchro signal from the DAS and must be stopped by cliking on stopSweepSwinging banner of TF Inspector panel blue circle of Figure 6 1 If any other operation is tried an error will appear The TF cannot be anything else till the swinging is stopped 7 Stopped the swinging by cliking on stopSweepSwinging banner of TF Inspector panel blue circle of Figure 6 1 8 Display the resulting image and search for the brightest slit image either by using the vertical graph of DS9 For this purpose use zoom to fit frame and graph vert from view menu Change scale to adjust to the maximum intensity by changing the maximum using scale pars of scale menu Then use the pixels near the left edge of CCD2 to give an idea have a look to cursor position in the inset of Figure 7 6 moving the cursor till the Page 39 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC best representative profile is found Determine the slit image of higher counts taking into account both peak value and symmetry of the remaining slit images around it For example in Figure 7 6 it is clear that the
7. S 2 931 5 934 6 934 6 938 7 938 7 942 9 Ke Ko NIN I 942 9 947 1 0 3 34 8 947 1 951 2 SIN Page 62 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC B 2 Blue etalon OSs Page 63 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC C OSIRIS TF WAVELENGTH CALIBRATORS For wavelength calibration either arc lines or emission line objects can be used Arc lines have the advantage that are more evenly distributed are usually more intense and can be used during day time operation Emission line targets usable for wavelength calibration should e Have a good sky coverage Be relatively compact e Have lines of narrow intense and precise wavelengths Compact PN fulfil some of these in general and a list is provided below However PN are concentrated near the Galactic plane and do not have many strong lines evenly or properly distributed They are usually useful for zero redshift targets OSIRIS team considers that for some applications in type 2 QSO at different redshifts could be used they are not very narrow indeed but are compact reasonably intense and more interesting located over all extragalactic sky at a variety of redshifts spanning the whole optical wavelength range redwards 500nm However the exposure times are typically an order
8. separate etalons The coating reflectivity determines the shape and degree of order separation of the instrumental profile This is fully specified by the coating finesse N which has a quadratic dependence on the coating reflectivity The OSIRIS TF was coated to a finesse specification of N 50 red 100 blue which means that the separation between periodic profiles is respectively fifty one hundred times the width of the instrumental profile At such high values the profile is Lorentzian to a good approximation For a given wavelength changes in plate spacing d correspond to different orders of interference m This in turn dictates the resolving power mN according to finesse Figure 4 1 OSIRIS red etalon at the IAC Optical Lab while undergoing calibration tests In general as can be appreciated in Eq 2 2 for a given order small changes in d change slightly the wavelength while for a given wavelength the change of order requires a larger change in d This is important to keep in mind Page 17 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 1 0 0 9 0 8 0 7 0 6 0 5 0 4 Tunable filter Gaussian Transmission 0 3 0 2 0 1 0 0 0 565 0 570 0 575 Wavelength microns Figure 4 2 Spectral response of a TF wrt a Gaussian The TF response can be considered Gaussian with a good approximation above FHWM but is more winged be
9. Code Draft C 2 2000 h m s 04 15 54 57 Selected compact Planetary Nebulae Size arcsec Dec J2000 d m s 48 49 40 8 Comments Too faint 184 0 02 1 05 46 50 01 24 22 02 7 184 6 00 6 05 58 45 34 25 18 44 0 It s the Southern object 226 7 05 6 07 37 18 91 09 38 48 1 228 8 05 3 07 40 22 20 11 32 30 1 049 3 88 1 12 59 27 61 27 38 14 1 061 9 41 3 1640 18 14 38 42 20 1 043 3 11 6 18 27 48 26 14 29 07 4 051 4 09 6 18 49 47 55 20 50 39 4 027 6 09 6 19 16 28 22 09 02 36 6 042 9 06 9 19 34 33 53 05 41 02 5 055 1 01 8 19 40 25 18 18 49 18 8 Too faint 082 5 11 3 19 49 46 57 48 57 40 2 058 3 10 9 20 20 08 76 16 43 53 3 065 0 27 3 21 29 59 42 12 10 26 5 089 8 05 1 21 32 30 95 44 35 47 7 111 8 02 8 23 26 14 83 INSTITUTO DE ASTROFISICA DE CANARIAS 58 10 54 7 38200 La Laguna Tenerife ESPA A Phone 922 605200 Fax 922 605210 TF USER MANUAL Code Draft Page 66 of 70 Date June 28 2009 File TFMANUAL DOC C 3 Selected bright type Il QSO 2000 2000 Flux EW OIII h m s Redsh
10. TF USER MANUAL Code Draft File TFMANUAL DOC 10 2 Calibration images 10 2 1 Bias Bias shall be obtained and applied as usual Same readout speed and binning must be selected See OSIRIS User Manual for the procedure to obtain bias 10 2 2 Flatfields Likely due to the non uniform illumination of the dome dome flats are not useful unless the target to be analyzed is relatively small 1 the flat is used to correct for small scale variations Otherwise sky flats must be then obtained using the same observing parameters approximate TF tuning and same OS This is of course a limitation in the number of wavelengths to be observed each night but please take into account that in TF scaming unless the wavelength range to be covered is very large or is required changing OS the same flatfield can be used 2 and that continuing a scan another day is acceptable as long as the same wavelength calibrator is used see Sec 8 1 10 3 Sky rings The OH atmospheric emission lines are observed through the TF as rings Figure 7 11 due to its centre to edge wavelength variation They are equivalent to the fringing observed in direct broad band imaging and like it are an additive effect They increase the background photon noise in the zones affected by the ring thus reducing the S N in these zones Sky rings not necessarily require correcting as long as the ring is not on the target and the target is not too faint Nevertheless if
11. and in general it is possible to distinguish between two situations e Filters in a collimated beam OSIRIS case beams from different points of the GTC focal plane reach the TF at increasing incident angles with symmetry with respect to the optical centre Then there is a progressively increasing shift to the blue of the central wavelength as the distances r to the optical centre increase according to Eq 4 2 Figure 4 5 However since the beams coming from the same point of the FOV are parallel the FWHM is nearly the same Figure 4 6 This is the case of OSIRIS since OSIRIS TF are located in the pupil of the collimated beam Since this is a pure geometric effect the wavelength variation is completely fixed and predictable because it depends only on the Page 19 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC incident angle that is completely determined by the ratio between the telescope and the instrument collimator focal distances 2 A r A 1 7 952010 43 are Filters in convergent beam 1 near focal plane beams from different points of the focal plane reach the filter with a variety of angles that vary from point to point with symmetry with respect to the optical centre 1 only at the optical centre the beam is symmetric In this case the central wavelength also shifts to the blue but 1t is much more noticeable and severe the FWHM increase lin
12. m dr PP No X a has Z a de O 4 4 0 0L lt lt f t V yO gt 1 580 600 620 640 700 Wavelength nm Figure 2 2 Variation of the transmission profile of a FPF with finesse The profiles were determined for an ideal FPF Equation 2 25 with R 0 68 0 81 and 0 92 4 0 Orders m 10 and m 9 are shown Page 13 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL front elevation side elevation JA pi s e 8 gt capacitor A reflective coating reference capacitor B optical gap a piezo electric transducer Figure 2 3 Front elevation and side elevation of a Queensgate Instruments etalon Note that the thickness of the optical gap is exaggerated In Figure 2 3 we show the structure of a gap scanning etalon manufactured by Queensgate Instruments Ltd now IC Optical Systems In recent years these etalons have undergone considerable improvements It is now possible to move the plates between any two discrete spacings at very high frequencies 200 Hz or better with no hysteresis effects while maintaining 2 2000 parallelism measured at 633 nm The etalon spacing is maintained by three piezo electric transducers as discussed below 2 2 4 Piezo electric transducers Piezo electric materials undergo dimensional changes in an applied electric field Conversely they devel
13. 08 40 41 08 38 38 19 8 0 31309 6573 9 51 2 41 08 42 34 94 36 25 03 1 0 5610 7815 6 388 4 881 08 48 56 58 01 36 47 8 0 3498 6759 2 34 4 60 09 02 46 94 01 20 28 4 0 51326 7575 3 60 7 324 09 04 14 10 00 21 44 9 0 3529 6774 2 79 1 92 09 14 42 33 00 06 37 2 0 5608 7815 6 41 6 295 09 20 14 11 45 31 57 3 0 4022 7019 5 75 0 134 09 21 52 45 51 53 48 1 0 5872 7945 8 52 7 199 092318 06 01 01 44 8 0 3865 6939 4 65 6 82 09 23 56 44 01 20 02 1 0 3796 6909 4 30 4 121 09 43 12 82 02 43 25 8 0 5919 7970 8 37 4 232 09 48 36 05 00 21 04 6 0 3239 6629 0 37 6 99 102746 03 00 32 05 0 0 6136 8079 0 56 9 319 10 39 51 49 64 30 04 2 0 4017 7019 5 175 9 304 10 45 05 39 56 11 18 4 0 4282 7149 7 70 9 62 TF USER MANUAL Page 67 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC 10 48 07 74 00 55 43 4 0 6427 8221 2 52 1 237 115314 36 03 26 58 6 0 5749 7885 7 118 4 238 115718 35 60 03 45 6 0 4906 7465 1 169 3 178 12 26 56 48 01 31 24 3 0 7320 8671 8 73 8 189 1228 45 74 00 50 18 7 0 5712 7885 7 55 4 293 1232 15 81 02 06 10 0 0 4805 7410 1 219 9 593 132323 33 01 59 41 9 0 3503 6759 2 146 8 512 133550 36 01 24 39 3 0 5228 7625 4 38 3 75 133735 02 01 28 15 7 0 3282 6654 0 57 4 40 140740 06 02 17 48 3 0 3092 6553 9 100 5 277 141315 31 014221 0 0 3801 6909 4 139 0 339 14 46 42 29 01 13 03 0 0 7259 8636 7 38 7 118 15 01 17
14. 7 2 TF control panel from the OSIRIS Inspector The banners for setting are marked in green and the banner for setting up the Z swinging is marked in red The following step is programming the OSIRIS DAS could have been done beforehand leaving 1 ready just when pressing expose The following parameters have to be changed Figures 7 3 7 4 and 7 5 6 Sorry for the wizard names Page 37 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC e OPTIONS flap of the Data Acquisition System flap of the OSIRIS DAS menu set a readout speed of 500 kHz to reduce readout time Figure 7 3 Redoran 80 ETE ves sooo s ass tosta sra 4 6 Figure 7 3 OPTIONS flap of the Data Acquisition System flap of the OSIRIS DAS menu Set readout speed at 500 kHz and gain 4 75 e CCDs AREA flap of Data Acquisition Window of the OSIRIS DAS menu select CCD1 CCD2 Output A Do not use split mode Figure 7 4 TG 0850 uma Cons mae montre ame Repo Mrs Shacorfowaion Vn onoontputen DsisAeous ontonponeni
15. a correction is required there are several ways to proceed e Ifthe targets are not very large a superflat with the sky ring pattern can be obtained by combination of dithered and smoothed images even of different pointings i smoothing the dithered images to be corrected using a kernel that wipes small scale structures sources but keeps large scale structures the rings ii obtain the media of the combined smoothed unmatched dithered images using a rejection algorithm the sources are not in the same position due to the dithering but the rings does since their position depend on the wavelength tuned only Even images of different pointings can be used as long as the wavelength tuned is the same 111 Subtracting the resulting superflat from the original images e Running for example SExtractor to remove sources thus creating again a superflat e Using specially devised programs for creating a synthetic ring image by fitting an azimuthal average of the image and subtracting it e Fitting a 2D surface to the ring creating a synthetic ring image and subtracting it 12 In other words if observing a nearby wavelength with the same OS the same flatfield can be used Page 58 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC Needless to say synthetic ring images does not introduce photon noise in the final image as the other methods does Some of these programs are available in the o
16. bright sources are in the FOV and their ghosts could spoil the image of the target there are no need to worry Figure 4 7 Example of ghosts in a tunable filter The cross marks the optical centre The red circles mark the diametric ghosts of the centre of the galaxy and that of an exponential ghost while the green ones marks the exponential ghosts The second exponential ghost does not produce a noticeable diametric ghost Of course ghosts drives the location of the target in the FOV as is dealt with further on in this document in Section 5 Exponential ghosts cannot be removed by dithering Luckily OSIRIS TF does not have this kind of ghosts 4 6 Calibrations required In general as described in 2 2 3 three piezos control the separation between the plates An exact control of the piezos is achieved via the etalon controller CS 100 of ICOS However for the purpose of a user manual the TFs are controlled via the Inspector interface that send specific commands to the CS 100 though Then the user can forget about operating the CS 100 directly although sometimes some comments are included for the sake of completeness 4 6 1 Paralellism calibration The piezo expansion or contraction is achieved by varying the HV supplied The plate separation is non linearly and differentially measured by capacitance bridges in terms of bits Figure 2 3 Each bridge provide then the relative separation with respect to two perpendicular axes whos
17. is left for a further version devoted to shuffle and fast modes Page 54 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 8 4 1 Charge shuffling 8 4 2 Frame transfer 9 OBSERVING WITH OSIRIS TF 9 1 General procedure The recommended procedure for using a TF is described in this section 9 1 1 Preparing observation According to the previous sections depending on the scientific program the line to be observed the type of target its size velocity field or velocity dispersion redshift of Doppler shift accuracy of this shift and the number of targets it is necessary to determine e Position of the target s in FOV e Orientation of the detector on the sky Wavelength to be tuned in the centre of TF Dithering pattern to be used or use of the half field mask e Technique to be used single exposures shuffled images fast photometry TF tomography or band synthesis e FWHM to be used Use calculator for exploring possibilities This might drive reconsidering the technique to be used if FWHM is too narrow e Wavelength range to be scanned for TF tomography or band synthesis e Exposure time e Use of TF OS or SDSS filters for continuum subtraction e The above steps should provide the information required for the OB including the set of Z pairs to be observed Page 55 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC
18. of magnitudes those used for PN 8 Beware that some have peculiar velocity fields For example Eskimo TF USER MANUAL Code Draft Page 64 of 70 Date June 28 2009 File TFMANUAL Selected arc lines Order Exp Line X Y FWHM Lamp 4 FAS os bi bi cal Order Ne 6506 5 643 28 s 34650 34165 14 0 15 Mg Bese portus 34675 34170 20 5 10 Ne 69295 694 44 1 34675 34185 13 8 14 Hg Ar 6965 4 694 44 25 34600 34160 13 9 14 7032 4 694 44 1 34650 34170 13 9 14 7173 9 709 45 2 34675 34170 13 5 14 7245 2 723 45 1 34660 34170 13 2 15 7438 9 738 46 1 13 5 14 Hg Ar 7635 1 754 50 34650 34175 13 5 14 34650 34160 121 14 8231 6 819 52 2 34650 34180 13 1 13 34700 34170 152 11 Hg Ar 8264 5 819 52 12 7 13 Xe 8409 2 838 58 15 34600 34150 13 1 12 Hg Ar 8521 4 838 58 12 3 13 Xe 8819 4 878 59 2 34375 34060 122 15 Xe 89522 893 50 10 34550 34140 12 1 16 Hg Ar 9123 0 902 40 30 34575 34125 125 18 Xe 9162 5 911 40 6 34610 34150 12 3 18 INSTITUTO DE ASTROFISICA DE CANARIAS 38200 La Laguna Tenerife ESPANA Phone 922 605200 Fax 922 605210 V84 PN ID 153 7 01 4 Page 65 of 70 Date June 28 2009 File TFMANUAL TF USER MANUAL
19. of parallelism than Y Using AX 50 and AY 25 with 14 and 8 steps is typically enough However be sure that you have the maximum intensity image more or less centred Page 43 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 7 1 3 Lack of parallelism A lack of parallelism can be seen as e Distorted sky rings e Asymmetric wavelength calibration Z scans that are in opposite directions depending whether there is an excess or lack in X or Y values see figure 7 10 e Lower intensities of slit images in wavelength calibration Z scans e Wavelength shifts The main consequences for the data are e Transmission losses e Wider FWHM and distorted spectral response for a TF in a collimated beam The XY resolutions used for parallelism calibration 50 and 25 bits respectively have been chosen as the most convenient Larger steps are not accurate enough and the XY errors affect wavelength and transmission as shown in the following table approximate values to serve as demonstration only for the red TF errors Red TF Then it is important keeping a good parallelism better than 50 bits in X and 25 in Y Again note that Y is more sensitive Page 44 of 70 TF USER MANUAL Date June 28 2009 Code Draft File TFMANUAL DOC i Fle Edt Frame Zoom Scale Color Region WCS Analysis Zoom Scale Calor Region WCS Analysis Joni z12
20. several ways to alleviate this problem that will be described in the following subsections Anyway please note that in many cases ghosts do not required to be removed since they can be seen only for very bright sources and are diametric so that they can be readily identified 5 5 1 Half field masking Inserting a half field mask that hides detector CCD1 avoids ghosts coming from that part of the field Then no dithering is done and cosmic rays can be removed by specific programs since the pixel size is small enough 0 127 arcsec so that a cosmic ray cannot be confused with an astronomical target even in good seeing conditions binning and seeing depending of course However in this case only half the field can be used then this is not the optimal technique for observing blank fields 5 5 2 Azimuthal dithering pattern When only one relatively small source is of interest the dithering can be done following the circle of equal wavelength Page 30 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 5 5 3 tuning dithering pattern If several relatively small targets spread on the FOV are to be observed it is possible to do a three point dithering where the TF tuning is changed to minimize wavelength variations at the edge of the TF FOV in one direction Then different sections of different images can be combined This is valid only when using tunable tomography 5 6 Tunable tomography
21. the CCD while varying X Y or Z from shuffle to shuffle From now on N will be the number of steps in X Y or Z AX AY and AZ the X Y or Z increments in bits respectively from one step to the following Xi Yi and Zi the initial X Y or Z values respectively and n the step counting from the first one were the line is brightest Page 35 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC This procedure has however some very important points to keep in mind e light must be stable Then several minutes should pass after switching lamp on and exposure times larger than 100ms must be used to get rid off AC electric power variations and shutter accuracy Single intense emission lines must be used No other line unless very weak must be in about 10nm both sides of the spectra e The TF must be tuned to the line 1 e Z must be the one corresponding to the emission line The last requirement implies an iterative procedure to calibrate XY and Z sequentially In detail is as follows 1 Switch the selected lamp on first thing to do to allow it warming up and reducing calibration time and the dome lights off 2 Set rotator at the requested value for observations try avoiding the range 30 to 60 3 Insert the calibration mirror 4 Insert the long slit of 2 5 arcminutes this is a special one for this type of calibrations any other could be used instead as lon
22. to exposure 100 bits of the value predicted by the TF calculator It is advisable using 20 steps of 10 bit each In other words N 20 AZ 10 Zi Z predicted 100 This resolution is equivalent to approximately 0 2nm and it is possible to calibrate to half this i e 0 1nm that is the goal to be achieved Alternatively the first time AZ 10 can be set and afterwards AZ 5 can be used However please note that this finer step does not increase accuracy since makes the intensity variation smoother and wider This is achieved by setting an approximate XY values nearby the range of Z and wavelength chosen by clicking with the cursor on the SetXAxis and SetY Axis of the TF OSIRIS Inspector panel Figure 7 2 and introducing the desired XY values or using the default ones Then open the menu of the setupSweepSwinging by pressing the green bullet that appears when passing the cursor at the end of this banner Set init as Zi step equal to AZ and nsteps to N Then press the banner runSweepSwinging Blue circle of Figure 6 1 A tiny white stripe will begin to oscillate along the banner indicating that the system is waiting for DAS commands to begin syncro between TF tuning and CCD charge shuffling Beans Toolbox Iri 5 E san D wanna enses rosas Figure
23. 2 cost 2 1 1 R 1 8 A where T is the transmission coefficient of each coating plate cavity boundary R is the reflection coefficient d is the plate separation is the refractive index of the medium in the cavity usually air 1 and 015 the angle of incident light Thus the FPF transmits a narrow spectral band at a series of wavelengths given by mA 2 cos 0 2 2 where m is an integer known as the order of interference The peak transmission of each passband is 2 2 E E gt 2 3 R T A where is the absorption and scattering coefficient of the coatings 4 1 T R and the minimum transmission halfway between the resonant wavelengths is Therefore the contrast between the maximum and minimum transmission intensities is 2 26 2 4 T 1 For a contrast greater than 100 the reflection coefficient R of the coatings needs to be greater than or about 0 82 The wavelenght spacing between passbands known as the inter order spacing or free spectral range FSR is about 2 5 m which is obtained from Equation 2 2 by setting consecutive integral values of m Each passband has a bandwidth 6A full width at half peak transmission given by A 1 R E gt 2 6 TF USER MANUAL Code Draft Page 10 of 70 Date June 28 2009 File TFMANUAL DOC derived from Equat
24. 3 SAT Charge SHUI 54 8 42 id ia 54 9 OBSERVING WITH OSIRIS TEFE 54 9 1 GENERAL PROCEDURE csesscsscesecesesssssececececsesensaeeececeesenesssaeeeeeceeseneaeaeeeeeeseseneas 54 9 LL Preparing observation see a 54 912 Day time ses 55 9 1 3 Practical TF operation and calibrations during the night 35 10 POST PROCESSING TF DATA 56 Page 6 0170 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 10 1 GENERAL CONSIDERATIONS oeei n Lu a 56 10 2 CALIBRATION IMAGES yt u aaa tuy Saga taya ut h G 57 1021 Bids esca ERI DNI 57 10 2 2 i ia ue aie ir I e 57 10 3 RINGS rl ea led et bee i ta ba ota vest da Bret ed 57 11 POWERING DOWN gu u oe ve ve seo Fav on eu oed o F uen ea pe Vo eae tete sos 58 12 TROUBEESHOOT neve senten ee eee Re ee een To ee eeu usaq AREE 58 12 1 E CU EREBZES2 ie untu nt die na een 58 12 2 CANNOT BE SHUTDOWN iia 58 12 3 NOT STARTUP tia metre modern eia ee Eee dense 59 ANNEXES NO 60 A IU GO HONO U
25. 324 ts CCD 2 17 ET gt gt ml 5 xxmm FE Cat Mew mame Zoom cab Cobr Regon WCS nal UEUDUECEUEUEUEUEUETS en1 _213609 tt CCD 2 Lo en1 _214451 1t CCD 2121 2 vane El El image rrame1 zoom E mz T Cs j Z T r l i i iN E Figure 7 10 Example of intensity losses and resulting asymmetric slit image intensity profiles obtained for the same Z calibration scan in the following situations top left using Xbest 50 the Z scan is asymmetric and concave below the maximum intensity Top rigth using Xbest 50 the Z scan is asymmetric and concave above the maximum intensity Bottom left using Ybest 25 the Z scan is asymmetric and concave above the maximum intensity Bottom rigth using Ybest 25 the Z scan is asymmetric and concave below the maximum intensity 7 1 4 Troubleshooting 7 1 4 1 The X or Y Sweep does not stop Likely the system is waiting for more triggers Check that the number of X or Y Sweeps of the TF is the same as the number of loops in the DAS To cure it take another exposure with the number of exposures lacking or never more never mind 7 1 4 2 The X Y or Z give very different values in consecutive exposures or even mirrored rings First of all check that the number of Z steps or X Y Sweeps are
26. 5 6 1 Technique Tunable tomography TF scanning consist in obtaining set of images of the same pointing at different consecutive wavelengths Figure 5 5 The characteristic parameters are e Initial and final wavelength or Z e Step in wavelength AA not equivalent to a constant AZ since the 4 2 relation is in general not exactly linear an gt a Figure 5 5 Tunable tomography consists in scanning a wavelength range using the TF For the same telescope pointing a set of images at different wavelengths are taken The step must be carefully chosen since for a given wavelength range to be scanned a step too fine will increase the observing time and overheads required but a step too coarse would introduce larger photometric errors that can be evaluated using Eq 5 1 but now considering that the maximum error will be half the scan step i e in the worst situation an emission line would be located in the middle of a step M _ 1 7 5 3 Please note that further on this document AA is NOT FSR of Eq 2 5 and 2 7 but the scan step or wavelength step between consecutive exposures in tunable tomography Page 31 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC In this observing mode the different images are not combined but analyzed separately Usually aperture photometry of the sources of each image provide pseudo spectra that are used for identifyi
27. 70 6580 6590 6600 6610 6620 6630 6640 Wavelength A Figure 5 3 Comparison of TF and line profiles The following table can be used for selecting to most suitable FWHM according to the expected line width Please note that they are approximate values Flux error Table 5 1 Approximate error fluxes depending on line width 8 and TF FWHM 02 For example the typical velocity field of a spiral galaxy 250 km s at zero redshift would require TF FWHM of at least Inm at for an error smaller than 5 and the minimum 3 Please note that 250km s is the FULL velocity field not the velocity field at FWHM Page 28 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC FWHM recommended for the red TF is of 1 2nm Hence this is not a problem in this case but it might be for OII 372 7nm blue TF or when observing objects at significant redshifts where the line widths are by the factor 1 z Since the range of available FWHM are quite limited if a certain photometric accuracy is required and the needed FWHM cannot be obtained it is possible to synthesize a wider FWHM by summing images of a scan see further on in this document 5 4 Deciding target position and orientation The presence of ghosts and the centre to edge wavelength variation drive target position on the OSIRIS FOV To avoid excessive wavelength variation the target should be as centred a
28. 70 nm blue arm and 650 960 nm red arm They offer monochromatic imaging with an adjustable passband of between 0 6 and 3 nm In addition TF frequency switching can be synchronized with movement of charge charge shuffling or frame transfer on the OSIRIS CCDs techniques that have important applications to many astrophysical problems 2 2 Introduction to Fabry Perot filters FPFs In its simplest form a Fabry Perot filter FPF consists of two plane parallel transparent plates which are coated with films of high reflectivity and low absorption The coated surfaces are separated by a small distance typically to mm to form a cavity which is resonant at specific wavelengths Light entering the cavity undergoes multiple reflections Figure 2 1 with the amplitude and phase of the resultant beams depending on the wavelength At the resonant wavelengths the resultant reflected beam interferes constructively with the light reflected from the first plate cavity boundary and all the incident Page 9 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC energy in the absence of absorption is transmitted At other wavelengths the FPF reflects almost all of the incident energy 2 2 1 Performance of an ideal FPF The general equation for the intensity transmission coefficient of an ideal FPF perfectly flat plates used in a parallel beam as a function of wavelength is 2 1 qu J 1
29. 96 54 55 18 3 0 3386 6699 1 118 0 72 15 06 08 09 02 07 44 2 0 4392 7204 8 98 8 685 15 17 11 47 03 31 00 2 0 613 8076 0 57 0 545 15 43 40 02 49 35 12 6 0 5116 7570 3 51 9 317 164131 73 38 58 40 9 0 5959 7990 8 221 7 597 17 0151 98 38 59 01 4 0 5962 7990 8 32 8 187 171559 79 28 07 16 8 0 5237 7630 4 53 8 220 172603 09 60 21 15 7 0 33340 6674 1 39 5 155 22 51 02 40 00 04 59 9 0 550 7760 5 43 7 455 22 56 12 18 01 05 08 1 0 648 8251 2 28 7 182 231239 40 00 50 05 5 0 479 7405 1 53 0 486 23 58 18 87 00 09 19 5 0 4022 7019 5 143 0 195 23 58 31 16 00 22 26 5 0 628 8151 1 214 0 676 Adapted from Zakamska et al 2003 AJ 126 2125 Redshifts from NED database Assuming a concordance cosmology Ho 71 km s Mpc Q 5 0 8 0 7 Typical line widths of 400 500 km s HB is much fainter but beware of confusing with nearby OINII A4958 9 which is about 3 times fainter Suggested exposure times are of about 10 s for fluxes 107 erg cm g Page 68 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC D SELECTED STANDARDS FOR FLUX CALIBRATION Page 69 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL E SOME USEFUL FORMS E 1 For keeping track of parallelisms and 7 offsets Comments VTrr Ne 6506 5 643 28 34650 34165 29876 14 0 15 N
30. B sun 34675 34170 40946 20 5 10 Ne 6929 5 694 44 1 34675 34185 30398 13 8 14 6965 4 694 44 25 34600 34160 30213 13 9 14 Ne 7032 4 694 44 1 34650 34170 29869 13 9 14 Ne 7173 9 709 45 2 34675 34170 29141 13 5 14 Ne 7245 2 723 45 1 34660 34170 28773 13 2 15 Ne 7438 9 738 46 1 30261 13 5 14 Hg Ar 7635 1 754 50 34650 34175 29304 13 5 14 34650 34160 31798 12 1 14 Xe 8231 6 819 52 2 34650 34180 34559 13 1 13 34700 34170 40083 15 2 11 8264 5 819 52 34417 12 7 13 Xe 8409 2 838 58 15 34600 34150 39424 13 1 12 Hg Ar 8521 4 838 58 38984 12 3 13 Xe 8819 4 878 59 2 34375 34060 40662 12 2 15 Xe 8952 2 893 50 10 34550 34140 40071 12 1 16 INSTITUTO DE ASTROF SICA DE CANARIAS 38200 La Laguna Tenerife ESPA A Phone 922 605200 Fax 922 605210 TF USER MANUAL Code Draft Page 70 0170 Date June 28 2009 File 9123 0 902 40 30 34575 34125 39237 12 5 18 9162 5 911 40 6 34610 34150 39031 12 3 18
31. E 60 B OSIRIS ORDER SORTERS 61 1 REDETALEON OSS re la aka E ua dit 61 B 2 BEUEELDADIONIOSS CS toed Ta qalas 62 C OSIRIS WAVELENGTH CALIBRATORS 63 SELECTED ARCTINES EE e ee ete 64 5 2 SELECTED COMPACT PLANETARY NEBULAE 0 004 nnn nnn nnn 65 C 3 SELECTED BRIGHT TYPE us 66 SELECTED STANDARDS FOR FLUX CALIBRATIONN 68 E SOME USEFUL FORMS AAA A eo peo V Tenes ee Vea o 69 FOR KEEPING TRACK OF PARALLELISMS AND Z OFFSETS 69 Page 7 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC LIST OF ABBREVIATIONS AC Altern Current CCCS Control Cabinet Control System CCD Charge Coupled Device DAS Data Acquisition System FSR Free Spectral Range FWHM Full Width Half Maximum GCS GTC Control System GTC Gran Telescopio Canarias HV High Voltage ICM Instrument Calibration Module IPA Instrument Position Angle LCU Local Control Unit MAD M dulo de Arranque Distribuido MCS Main Control System MOC M dulo Comunicaciones OB Observing Block OSIRIS Optical System for Imaging and low Resolution
32. Integrated Spectroscopy OS Order Sorter PN Planetary Nebula QSO Quasi Stellar Object SA Support Astronomer SDSS Sloan Digital Sky Survey SDSU San Diego State University S N Signal to noise ratio STIS Space Telescope Imaging Spectrograph TBC To Be Confirmed TBD To Be Defined TF Tunable Filter Page 8 0170 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 1 INTRODUCTION 11 Scope This document is aimed at providing the 1 Users with the basic information required for designing a successful observation of their favourite targets using OSIRIS TFs 2 GTC SAs with the basic information required for start up initialize configure use calibrate and power off the OSIRIS TFs Some troubleshooting basics is provided as well IMPORTANT NOTE It is advisable that TF are always operated by GTC SAs or by OSIRIS Instrument Team members only not by visiting astronomers regardless of their experience in etalons 1 2 Contact information The information presented here can be found in the form of slides in the OSIRIS WWW http www iac es project OSIRIS The calculators can be found in the same page as well For specific or further queries please contact OSIRIS PI at jcn iac es 2 TUNABLE FILTERS 2 4 TF concept A key aspect of OSIRIS is the use of tunable filters TFs OSIRIS TFs are a pair of tunable narrow band interference filters Fabry Perot etalons covering 370 6
33. PROVAL CONTROL eer u 2 DOCUMENT CHANGE RECOR 2 HINT NIS PP 3 TABLE OF CONTENT eet so u e eve aab kunt ga 4 LIST OFABBREVIATIONCS acea area e e ui vo Eee rne 7 1 INTRODUCTION EL 8 1 1 SCOPE cates enw a e a ou te Erde e eL 8 1 2 CONTACT INFORMATION 2 02200000000000000 sua Pasa 8 2 TUNA BEB FILTERS 525 gt e ao on au 8 2 1 GON CEP A A 8 22 INTRODUCTION TO FABRY PEROT FILTERS 2 8 2 2 1 Performance of an ideal FPF sss eene 9 2 22 MAMAS a eae e aN etie E i E eddy 11 2 23 Gap scanning etalons ias 12 224 Piezo electric 13 2 2 5 Capacitance micrometry 14 2 3 CHARGE SHUEBLDING css did id 14 2 4 ORDER SORTERS emi oerte ai 14 3 TUNABLE IMAGING VERSUS SPECTROSCOPY 14 4 OSIRIS TF CHARACTERISTICS AND FEATURES 16 4 1 GENERAL A h maqa a 16 4 2 DIMENSION RN I nca tt de E 17 4 3 COATINGS eee do AE T Dee maka aa q ta 17 4 4 WAVELENGTH VARIATION ACROSS THE
34. Red init OK m 599 El 2 5 mG D Ru X LA 902091 we Q CE 1225 Friday 26 06 09 Figure 6 1 TF panel of OSIRIS Inspector Init banner is located at the top left below Initialize Device marked with a red circle The system is in IDLE state green circle ready to operate with HV activated e Now IDLE status the high voltage is applied to TF piezos and the instrument must not be powered off till the TF have been properly shut down see Sec 11 Page 34 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 7 CALIBRATING OSIRIS TF 7 1 Parallelism 7 1 1 General considerations As mentioned above TF parallelization consist in determining the X and Y values that keep plates parallel and depends on Z and A OSIRIS TF Parallelism is very robust and do not vary with time even when switching off and on again the TF controller Hence once the XY values for a certain Z and A range are determined they can be used around these Z and A values from then on Checking parallelism values from time to time are recommendable though Then it is strongly recommended that SAs keep track of the XY values obtained versus Z and noting down rotator position and TF temperature gradient on a daily basis A summary table of these values should be updated accordingly and made readily available in the control room for daily operation The parallelism
35. This is the reason why mirrored images could be observed This does not happens in X or Y Sweep since the loop is exited when the number of triggers are completed Please note that this only affects measures where the CS100 is waiting for triggers coming from the DAS TF calibrations or shuffles not when a given X Y or Z 15 set manually 7 2 Wavelength calibration 72 1 General considerations While parallelization is a Day Time procedure because it is very stable with respect to time temperature changes and even instrument rotation although again parallelism varies with Z and wavelength specially for the blue etalon wavelength calibration is a Nightly procedure since the Z A calibration depends upon many factors as already pointed out in Section 4 and the calibration must be checked during the night even for the same wavelength and order The wavelength calibration consist in establishing the relation between Z values in bits and the wavelength This relation is non linear enough so that a linear approximation can be deemed valid only locally The principles for OSIRIS wavelength calibration are different though In a climatic chamber for minimizing temperature and humidity variation and at the mean yearly temperature of the observatory the relation between Z and wavelength has been derived for every order and through the full wavelength range that each TF can cover This relation 15 fitted by polynomials and programmed in a ja
36. ages so that they can be distinguished easily Set loops to N e Set appropriate exposure time not smaller than 500 ms Expose by clicking the running man icon Elle Tools He UTC 5 15 55 osiris GCS OSIRIS DAS Sequencer Confge image Monitars Alame DetaAcquiciionComronent ShitCorfiguration Monitors WindawConfgurstion Data Acquistion System Spit Tools AcquistionPanel SDSU Controller Sruter Control Log GIDAS Log GEICA i Panels AE Jm Monitors SEQUENCES CCDs AREA OPTIONS L Beans uns ES ENEE Qu s CFR ANSFER Image Salt reso sima RDS iv eroe CLEAR EXPOSURE E ID p COMMAND 4 24 37 99 COMMAND 5 15 03 OSIRIB NCE TFComponant Red stoo8weepSwinging EEES i eo 2e sas oce cojo simple ever receiving a 3 scale rara trecaivina JE wring com near coses z 4 Figure 7 5 SEQUENCES of Data Acquisition Window of the OSIRIS DAS menu The parameters to be set are marked and with the recommended values
37. at Inspector or using python commands After that you can either proceed with the observations or initialize everything again 12 3 TF do not startup Check whether the TF status OSIRIS MCS TFComp on Figure 6 1 is Page 60 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC ANNEXES A REFERENCES Atherton et al 1981 Optical Engineering 20 806 Atherton P D 1995 ASP Conf Ser 71 50 Cepa et al 2009 OSIRIS User Manual code pending Hanuschik R W 2003 A amp A 407 1157 Hicks et al 1984 J Phys E 17 49 Jones R V Richards J C S 1973 J Phys E 6 589 Page 61 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL B OSIRIS ORDER SORTERS Red etalon OSs 33 lt nm 7 6462 6516 5 667 6 677 4 2 677 4 692 7 0 692 7 707 1 9 707 1 721 8 3 799 4 816 8 4 816 8 834 1 Go 99 NIN Y A DIO SI 921095 aJo S I gt AJN AY oO SUIS No 00 Lo DIN 45 45 45 45 I Its oO om ojo DIU Q I oo 00 On oo Q NO gt S Ape ere NO
38. case one additional image must be obtained at the beginning and at the end of the scan in a conventional filter equivalence this would be similar to the zone where the spectral response is varying and the flat zone has not been reached yet and that the photometric accuracy refers to the wiggles generated when adding the images Figure 5 6 Alternatively instead of obtaining this extra images it is possible to correct them by multiplying by the appropriate factor gt 1 that can be easily calculated As a guideline the photometric accuracy that can be achieved as a function of the scan step is summarized in the table below Figure 5 6 Page 32 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC ol en 6640 6660 680 6700 520 6640 6660 6580 800 6 6560 6620 6640 6660 Wavelength A Wavelength A Wavelength A Figure 5 6 Theoretical examples of the band synthesis technique Left a scan step equal to the TF FWHM gives photometric uncertainties of 20 Middle a scan step 2 3 of the TF FWHM gives photometric uncertainties of 5 Right a scan step equal to half the TF FWHM gives photometric uncertainties of 2 The first and last images are either not used or corrected via the corresponding multiplicative factor Of course the images obtained can be used both for Tunable tomography and band synthesis For example in tunable tomography imag
39. ccuracy ard m 7 2 r 2 4 where n is the number of nm of the allowed drift and AX is the wavelength variation in nm from the centre of the TF A to the ring Eq 4 3 AA 1 9520 107 A r 7 3 Page 48 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL with the radius r in arcminutes If the ring radius varies in more than the tolerated value the Z must be changed If r increases the Z must decrease and vice versa The TF tuning must be set so that the radius of the ring is of at least 1000 pix but smaller than 1800 pix the 8 arcminute diameter TF FOV Use CCD2 for measuring the ring using for example DS9 remember that the gap prevents directly fitting a single ring for both detectors An experienced user might use ring radius and the above expressions for fast recalibration without using the ICM However it is advisable to use the ICM to avoid errors from a tired and sleepy user This procedure has the advantage of not requiring change of telescope pointing or switching autoguider off Only OS and TF tuning However please note that the exposure times required are at least of one minute and that a intense single and relatively isolated line must be used but there are not many fulfilling these requirements For a good sky line map the reader 15 referred to Hanuschik 2003 that can found on line at http www eso org observing dfo quality UVES pipeline sky_spect
40. dure a final extra slit image can appear This setup is achieved by clicking on the bullet of the calibrateSweepX banner From the deployable menu that appears choose the option 14 50 Figure 7 7 Figure 7 7 Calibrate X and Y options marked in red on the OSIRIS TF Inspector panel 12 13 Please note that the X or Y sweep contrary to the Z swinging stops by itself after the programmed number of triggers from DAS However in all cases X Y or Z the user must set the new determined X Y or Z when finishing the corresponding X Y or Z calibration sequences because the TF is left in a value different from the best X Y or Z value Check that the last value of X is Xlast Xi NAX in this case the last value is the last Otherwise correct accordingly as described for the Z calibration Page 41 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 14 Look for the brightest slit image as before and calculate Xbest accordingly Figure 7 8 gt 96 xx www una gt SAOImage 459 File Edit View Frame Zoom Scale Color Region WCS Analysis File Jan02_043907 fits CCD 2 L2 559 73 000 1523 000 123 000 Y 1523 000 0 125 Angle 0 000 edit view frame power square root 3000 2000 1000 0 Figure 7 8 Example of a X calibration image of 14 steps of 50 bit Please note that in th
41. e X calibration the slit image intensities are not symmetric This is normal The final spurious slit image can be appreciated in the vertical graph In this case n 5 2 and then Xbest Xi 4 2x50 15 Set Xbest by clicking on the bullet of the setXAxis banner inserting the corresponding value on the deployable menu and clicking on setX Axis banner to set it Check that the X value of the TFComponent Red positionX has been updated Calibrate Z as before since changing X or Y might change Z not viceversa Changing in 50 or Y in 25 bits should not noticeably affect Z Since the approximate Z value should be nearby the previous one a AZ 5 could be used from now on Please note that approximately this step is equivalent to 0 1nm In other words calibrating Z in steps of 5 bits you are calibrating with an accuracy of 0 1nm the one required Set Zbest as before Now calibrate Y using a similar procedure as done with X but now using N 8 AY 25 Yi Ypredicted 100 since is more sensitive than X Again a final extra slit image can appear Check that the last value of Y is Ylast Yi NAY in this case the last value is the last Otherwise correct accordingly as described for the Z calibration Page 42 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 20 Determine and set Y best as with X Figure 7 9 H v ds9 96 File Edit Vie
42. e Z smaller the gap the larger is the FWHM 47 Tuning accuracy The theoretical tuning accuracy is 0 02 nm in wavelength and FWHM as provided by the CS 100 etalon controller in most cases this is the typical value of 1 bit in Z The real accuracy is driven by the wavelength calibration accuracy that can be of the order of 0 1 0 2nm 4 8 Tuning speed The tuning can be changed in an interval between 10 to 100ms depending on the change in Z Really for large Z differences the TF control system automatically move the etalon in steps to avoid out of range failures For fast modes it is advisable limiting the range of Z movement to the minimum hundreds 4 9 Tuning range The tuning range is driven at the smallest plate separations lowest resolution largest TF FWHM by the coatings touching while at the largest separation higher resolution smaller TF FWHM is driven by the order sorter filters that are designed to block orders at a certain TF FWHM since the different orders become closer as m increases FWHM reduces The OS filters are designed to block orders for a minimum FWHM that is of 1 2nm for the red TF and between 0 6 and 1 2nm TBC for the blue TF Then although the plates can be separated more than specified 1 e the FWHM reduced below 1 2nm the OS do not guarantee that other orders do not enter the nominal field of 8 arcminute in diameter Narrower OS FWHM would be required for using TF FWHM below 1 2nm for th
43. e distances are larger than expected since the TF spectral response has more wings than a standard interference filter S Deblending lines Equation 5 1 can also be used for deblending lines Known the redshift one tuning for each line can be observed From the line separation and the FWHMs of the tunings the contribution of the other lines to each tuning can be estimated and corrected simply via simultaneous equations system Relative response al seii 6540 6550 6560 6570 6580 6590 6600 Wavelength A Figure 5 2 can be deblended from NII 658 4nm if the redshift or Doppler shift is known via defining a simultaneous equation system with TF transmissions derived from the TF FWHM and line relative positions Page 27 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 5 3 On line FWHM selection The FWHMs are quite narrow and nearly Gaussian from peak to half transmission Then the line width must be quite narrower than the TF FWHM or otherwise some flux will be lost It can be easily demonstrated that assuming Gaussian line profiles the flux error can be approximated by 1 AY Flux error o Ej 5 2 n where 6 is the TF FWHM and dl is the line width For example observing a line with the same width than the TF will drive to loosing 28 of the flux a more precise calculation gives 3694 o Relative response e e RG 6540 6550 6560 65
44. e dithering pattern combined with TF tuning Sec 5 5 2 and 5 5 3 Page 33 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 6 STARTING UP OSIRIS TF e see CookBook power on the TF were it not powered on before e TF menu of the OSIRIS Inspector click on Init XYZ should be set to some reasonable default values after few seconds and the OSIRIS MCS TFComp info pass from ON to IDLE status Figure 6 1 Inspector v 01 06 01 00 File Tools Help UTC 13 25 42 osiris LOSIRIS MQ Sequencer Configs Image Monitors Alarms Report TFControl EE Split Tools ess gt OSIRIS MCS TFComponent_Red MONITORS 5 0 c4 q gt Panels Calibration Operations initialise Device Monitors iil 9 T AAAL AAA CS TFComp IDLE FComponent_Red positi 34600 0 None calibrateSweepX mu Monitors Move Z Axis FComponent Redposii 341500 None calibrateSweepY ponent Red Displaceme 33450 0 None BE moveAbsolute Beans Toolbox Chart Sweep Table setxAxis TFComponent Red of setz 00 setupSweepTable TFComponent_Red typeSwe SWEEP SET setYAxis runSweepTable TFComponent_Red activateModeV OPERATE Sweep Swinging Shutting Down TF setups Restoring from Abort runSweepSwinging shutdown stopSweepSwinging COMMAND 13 25 03 OSIRIS MCS TF Component
45. e image and the off line or continuum image The on line has line plus continuum photons and the off line only continuum photons to be subtracted from the on line image to give the emission line continuum free image This can be done in two ways 1 Using the TF for line and continuum has the advantage that the spectral response and FWHM is the same and that you can select the continuum as near as desired from your line thus alleviating possible continuum variations with wavelength It is even possible using the technique of shuffled exposures Section 8 3 to on line averaging continuum on both sides blue and red of the line and averaging possible seeing and atmospheric variation Then it is possible to achieve a very good continuum and sky subtraction with a direct pixel to pixel difference between your on line and your continuum image However TF have quite narrow FWHM and the exposure times are quite large 2 Using a medium band and OS or broad band filter SDSS In this case continuum subtraction is not as good and certainly not as direct and must be faced with caution This document is not intended to be a manual for continuum subtraction there are many different techniques that can be used that will not be dealt with here However since now the continuum filter is tens of times wider that TF the exposure times required are reduced accordingly 5 1 2 Off band wavelength and FWHM tuning selection If the TF is to be used fo
46. e red TF However depending on the position of the central wavelength with respect to the OS operating wavelength range the FWHM could be reduced down to 0 6nm for a wavelength centred with respect to the order sorter operating wavelength range This trick must be used with caution and only by experienced and warned observers Page 24 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 410 OSIRIS FOV According to the above the OS guarantees a FOV free or other orders of 8 arcminutes in diameter Although OSIRIS TF were originally specified for minimum FWHM of 1 2nm over the whole wavelength range the blue TF coatings are thicker and then the minimum plate separations are larger resulting in a minimum FWHM that is not driven by the order sorter as in the red On the contrary the OS have been chosen according to the maximum FWHM attainable with the blue TF Order sorters overlap in wavelength and their wavelength range where is guaranteed that they suppress other orders is smaller than their FWHM Appendix B This is because OS are tilted 10 5 degrees with respect to TF and grisms to avoid ghosts due to backwards reflections from the detector the TF is not tilted and then have these reflections as explained above The tilting axis is with enough approximation the detector gap 5 OBSERVING STRATEGIES The observing strategies for a TF are driven by the following instr
47. e value in bits are termed X and Y Page 22 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC Then regulating X and Y is necessary for achieving plate parallelism The price to pay for a slight lack of parallelism is a reduction of the peak transmission and an asymmetry of the spectral response Due to the thick coatings of OSIRIS etalons the parallelism depend both on wavelength and the order This is a weak dependence for the red etalon but more noticeable in the blue The CS 100 has been programmed to compensate for this variation in the blue etalon However the parallelism values are very stable in time and no changes has been detected so far worth periodically checking anyway 4 6 2 Wavelength calibration Once the etalon is parallelized then all piezos act co ordinately keeping parallelism while increasing or decreasing plate separation in closed loop operation of the CS 100 and the plate separation is non linearly quantified in bits by means of the parameter Z in bits Essentially wavelength calibrating an etalon means establishing the dependence between wavelength and Z for the different orders 1 FWHM This dependence is a function of wavelength and order For this reason the calibration should be performed near the wavelength and order of interest Wavelength calibration depends at least of the following e Day to day variation For this reason it is advisable calib
48. emission line maps for targets widespread on the FOV either extended or of small size The main disadvantage is that only one wavelength can be observed at a time The following table and the flux diagram below help deciding the most appropriate mode osition not required survey osition required pre imaging This introduce biases On the fly observations enabled For ex Galaxy clusters Drilling masks overheads F pre imaging avoid IFU mosaic CASE FOR TUNABLE IMAGING Figure 3 4 Tree for deciding the most appropriate mode TF versus spectroscopy Av is the line width or velocity dispersion the target diameter p the density of targets and AFlux the error flux required Page 16 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 4 OSIRIS TF CHARACTERISTICS AND FEATURES 41 General The OSIRIS TF manufactured by IC Optical Systems with plate separations accurately controlled by means of capacitance micrometry has the appearance of a conventional Fabry Perot etalon in that it comprises two highly polished glass plates Figure 4 1 Unlike conventional COS etalons it also incorporates very large piezo electric stacks which determine the plate separation and high performance coatings over half the optical wavelength range The plate separation can be varied between about 3 44m to 10 The highly polished plates are coated for optimal performance over 370 960 nm using two arms
49. es can be combined in for example pairs or terns to increase S N in the case of faint targets depending on line widths and TF FWHM or all added together to serve as pseudo broad band image for target detection 5 7 2 Some applications Scanning the velocity curve of a large spiral galaxy compensating for the centre to edge wavelength variation e Scanning a wide spectral line 1 that of a high redshift galaxy or a QSO 5 8 Sources of instrumental photometric errors Summary Then the sources of photometric errors of instrumental origin again please note that contributions of readout or photon noises must be considered aside are e The FWHM of the TF depending on line width driven by velocity dispersion velocity field peculiar velocities and redshift Can be evaluated using Eq 5 2 or Table 5 1 If required can be corrected using the method of band synthesis Sec 5 7 e The contribution of the line to the off band or to other lines to the on line image Can be evaluated using Eq 5 1 e The wavelength variation across the target Depends on the size and the velocity field Can be evaluated using Eqs 4 3 and 5 1 If required can be corrected using the method of band synthesis Sec 5 7 Dithering that varies the wavelength of pixels of the source from one image to a dithered one Can be evaluated using Eqs 4 3 and 5 1 If required can be corrected using the method of band synthesis Sec 5 7 or choosing a suitabl
50. escope For example an object which is one degree across in the collimated beam imaged with m 50 has N 500 according to Equation 2 13 A more relevant analysis to consider the change in central wavelength of the filter as the ray angle 0 is varied in Equation 2 2 For example a change in ray angle from 1 to 3 produces a change of 0 1 in the central wavelength of the Page 12 of 70 TF USER MANUAL Date June 28 2009 Code Draft File TFMANUAL DOC filter at any given order Therefore at high resolving powers 1000 a FPF may not be truly monochromatic across a desired field of view 2 2 3 etalons In order to manufacture a tunable FPF which can change the central wavelength for a given order it is necessary to be able to adjust either the refractive index of cavity the plate separation d or the angle 0 as can clearly be seen from Equation 2 2 In a gap scanning etalon the plate separation can be controlled to extremely high accuracy Transmission of a Fabry Perot filter 1 0 de O de oc 1 d 3um 9 0 M 1 finesse n T E 478 oc ceu 4 F 347 4 5 E 06 4 4 pa e 04 7 E J n E tes EN iqq S L 4 Pu dx 0 2 E fog Jes E
51. ff line python pipeline devised by the instrument team and described in the OSIRIS User Manual 11 POWERING DOWN ii YOU MUST NEVER EVER POWER OFF OSIRIS IF THE HAVE BEEN INITIALIZED AND NOT SHUTTED DOWN L E if in the IDLE status YOU CAN DAMAGE THE TF e the Inspector TF menu Figure 6 1 click on Halt and wait till the command has been executed e Then click on Shutdown Figure 6 1 and wait till the command has been executed takes three seconds Please note After shutdown the TF cannot be operated again unless you click on Start first to put the TF in ON status Figure 6 1 12 TROUBLESHOOTING 12 1 LCU freezes If LCU PORT 1 the one that controls TFs freezes while using the TF or with the TF initialized and you have to initialize it either by using Crate_27 switchOff or by going to the electronic Cabinet A and using the LCU switch once the system is recovered you must first init the TF and then first halt and after shutdown at Inspector After that you can either proceed with the observations or initializing everything again Resetting LCUs do not send commands to the TF then it is safe initializing LCU PORTI while TF operation but then the TFs would be left in the same status forever unless they are initialized again This is not convenient and in addition there 1s risk of switching off power with the TF in operation or initialized and th
52. fil the following e Narrow lines 1 FWHM smaller than 1 2nm and no peculiar velocity fields e Single lines e Intense lines to reduce exposure time e Of known Doppler or cosmological shift e Emission line region very compact These requirements immediately drive to compact PN as wavelength calibrators Appendix C 2 provides a selection of the compact and bright PN available from La Palma However PN allow mainly calibrating intense lines such as HB and OIII A500 7nm and perhaps ArIII A713 6nm and SIIT A906 9nm Other lines be used as well but must be analyzed with caution OII AA372 7nm and SII AA671 6 673 1nm are doublets and Ha has the nearby NII AA654 8 658 3nm lines then are useful for smaller FWHM Also PN concentrate near the galactic plane Then in general PN are ideal for calibrating TF observations of Galactic targets 7 The redshift of a galaxy is NOT a Doppler shift that of a PN does Page 47 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL EIS Comments Blended with Ha Blended with NII Blended with Ha Doublet NII and Ha within OS Doublet NII and Ha within OS HelA7065 3A within OS Useful for FWHM det SIII No other line within OS Useful for FWHM det SIII No OS available at this moment Wavelength of relatively intense PN lines For extragalactic observations of any line at any redsh
53. further characterizing operation environmental conditions 10 POST PROCESSING TF DATA 10 1 General considerations The TF data reduction procedure is like that of narrow band direct imaging requires debiassing flatfielding combining dithered images if required flux calibration using aperture or PSF photometry of spectrophotometric standard stars and continuum subtraction if required The main differences with narrow band direct imaging are e For removing cosmic rays cosmetics or sky rings only dithered images at the same tuning Z can be combined taking care of the possible wavelength shift of the dithered images specially at the edges of the TF FOV see Sec 5 5 for some hints on alleviating this problem e Depending on the type of sources subtracting continuum images might be unnecessary if a pseudo spectra is obtained by aperture photometry of the sources observed at different contiguous wavelengths Sec 5 7 2 e Images of the same field at different contiguous wavelengths can be combined in the band synthesis technique Sec 5 7 e In the red spectral domain instead of fringing sky rings appear see 10 3 10 An experienced user might use the ring radius to this purpose thus saving the few minutes required for using the ICM and restarting the autoguider As images obtained using different conventional filters would not be combined for this purpose either Page 57 of 70 Date June 28 2009
54. g as it is not narrower 5 Using the tables of appendix B and or the TF calculator set approximate XY values for a given line of Z and A nearby the ones required OSIRIS Tunable Filter Setup Tool Mozilla Firefox Archivo Editar Ver Historial Marcadores Herramier mtas Ayuda dar 13 fie 1110 OSIRIS TFCalculator4JOstrisTEZ html w 6 2 a Li M s visitados e Comenzar a usar Firef OSIRIS Tunable Filter Setup Tool Computes CSS 100 digital counts for tunning the TF TF Red TF Input parameters and FWHM A found 12 55 13 13 13 66 13 88 TF Red TF 14 27 15 17 16 19 17 64 18 37 20 18 Wavelength A 6563 21 99 15 17 A 5 A 30075 digital units Band width 15 5 Change in lambda 0 18 for incr 2 1 Change in lambda 0 19 A for incr Z 1 FSR 428 59 A 4 Inicio 53 Windows Live Messen f TFCalculator TFCookBook Micros Figure 7 1 Basic TF Calculator Given the TF wavelength and bandwidth returns Z nearest FWHM and the slopes 1 Z to the blue AZ 1 and to the red AZ 1 of the predicted Z FSR is provided as well for etalon lovers Java XML script courtesy of J I Gonz lez Serrano based on Guillermo Herrera filtered data obtained at 10 5 C See TF calculator notes Page 36 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 6 Take a set of exposures charge shuffling the CCD and varying Z from exposure
55. ift ao E i Comments 00 25 31 46 10 40 22 2 0 30349 6523 9 71 1 59 00 50 09 81 00 39 00 6 0 72760 8656 8 142 0 281 00 56 21 72 00 32 35 8 0 48400 7430 1 124 0 303 01 14 29 61 00 00 36 7 0 389 6954 4 33 8 39 01 20 32 21 00 55 02 0 0 601 8015 9 49 8 309 01 23 41 47 00 44 35 9 0 399 7004 5 94 0 379 01 34 16 34 00 14 13 6 0 555 7785 6 107 3 457 01 42 37 49 14 41 17 9 0 389 6954 4 42 6 52 01 49 32 53 0048 03 7 0 566 7840 6 69 2 228 01 57 16 92 00 53 04 8 0 422 7119 7 202 0 324 01 59 11 66 14 39 22 5 0 319 6604 0 42 7 301 02 10 47 01 10 01 52 9 0 540 7710 5 208 6 1155 02 17 57 82 01 13 24 4 0 375 6884 3 28 6 37 02 17 58 19 00 13 02 7 0 343627 6729 1 55 5 95 02 18 34 42 00 46 10 3 0 372 6869 3 58 2 66 02 27 01 23 01 07 12 3 0 363 6824 3 69 1 89 02 34 11 77 07 45 38 4 0 309781 6558 9 74 0 159 02 49 46 09 00 10 03 1 0 408001 7049 6 28 2 109 03 14 49 11 01 05 02 3 0 557 7795 6 43 3 132 03 19 50 54 00 58 50 6 0 626 8141 0 89 9 539 03 2029 78 00 31 53 5 0 3835 6929 4 25 2 35 03 32 48 50 00 10 12 3 0 3096 6558 9 39 7 45 07 37 45 88 40 21 46 5 0 6137 8076 0 47 4 180 07 39 10 48 33 33 53 8 0 4459 7239 8 68 9 353 07 5607 16 46 14 11 5 0 5936 7975 8 27 6 53 07 57 07 43 27 36 33 6 0 8189 9102 4 43 6 184 08 01 54 24 44 12 34 0 0 5560 7790 6 119 9 1245 08 12 53 09 40 18 59 9 0 5513 7765 5 79 1 103 08 13 30 42 32 05 06 0 0 3968 6999 5 47 4 78 08 15 07 42 43 04 27 2 0 5097 7560 3 144 2 252
56. ift OIIIJA500 7nm emitted by Typell QSO might be used at a expense of one order of magnitude integration times it is a strong single line QSO are quite very well distributed relatively compact and with a variety of accurately determined redshifts that place almost anywhere beyond 500nm The line is somewhat wide though 300 500 km s A list of selected Type QSO is provided in Appendix C 3 The procedure for Z calibration using any compact astronomical target will consist in placing the object into the 2 5arcsec long slit near the optical centre of the TF in CCD2 and starting a shuffled sequence varying Z as for an ICM calibration lamp The brightest image of the source pinpoints the emission line 7 2 4 Calibration using sky rings The OH group produce relatively strong emission lines specially redwards 700nm These are a nuisance in broad and narrow band imaging as they are for long slit spectroscopy However they happens at precise wavelengths and with definite relative intensities and can be used for calibrating spectra or as in this case TF Since the FOV is fully illuminated by these emission lines rings are produced Known the wavelength of the emission line the radius of the ring that can be obtained using DS9 provides the central wavelength tuned As a rule of thumb for a wavelength drift lower than 0 1nm the variation Ar of the ring radius r should be in the worst case ar lt 0 02 7 1 r For better a
57. in a wider inter order spacing for a given resolution the finesse needs to be increased For a finesse greater than 100 a reflection coefficient R of greater than or about 0 97 is necessary Equation 2 9 However so far we have considered the ideal situation where the plates are flat and parallel and the incoming light is parallel In particular Equations 2 1 2 3 2 5 2 7 and 2 9 refer to this situation using the subscript r to distinguish the results from a real filter In practice plate defects and the angular size of the beam limit the maximum finesse obtainable The effective finesse N is approximately given by Loes ay ES N N N 2 11 where N is the reflective finesse from Equation 2 33 N is the defect finesse due to plate defects and N is the aperture finesse due to the solid angle of the beam The defect finesse 27 N 2 12 where is a length scale related to deviations from flat parallel plates The exact details depend on the type of deviations Atherton et al 1981 A FPF manufactured with N 80 and a reflection coefficient of 0 97 100 performs with a finesse of about 60 The aperture finesse s 2 13 where Q is the solid angle of cone of rays passing through FPF This equation is related to the A dependence on Gin Equation 2 2 In terms of astronomical imaging the effect of aperture finesse is negligible for most objects in the field of view of a tel
58. ion 2 1 The ratio of inter order spacing to bandwidth is called the finesse air is c AR glass T 2 7 E 1 R R air R AR glass 2 7 2 8 air Figure 2 1 Schematic diagram of interference with a Fabry Perot filter The outside surfaces of the glass are coated with anti reflective AR coatings while the inside surfaces are highly reflective usually R gt 0 8 The air cavity in the middle is not shown to scale usually d is about 10 whereas the glass is over 20 mm thick on both sides At resonant wavelengths the first reflection shown with a solid line interferes destructively with light coming from the cavity in the same direction dashed lines The phase difference arises because the first reflection is internal while all the other reflections are external with respect to glass On the other side of the cavity only constructive interference occurs At non resonant wavelengths destructive interference occurs in the cavity and the first reflection dominates For an ideal FPF it is given by 2 9 Page 11 0170 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC Thus we can see that the resolving power of a 15 equal to the product of the order and the finesse mN 2 10 2 2 2 Limitations It is apparent from the above equations that to obtain a higher resolution for a given order or to obta
59. is can damage the device 12 2 TF cannot be shutdown If you cannot halt and shutdown the TF using Inspector do not panic yet Try closing and opening inspector and giving cntrl 6 If this does not work go to the terminal used to start Inspector and type e pythondevelopment with a dot first and a blank space e TunableFilters Page 59 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC python haltTF py python shutdownTF py If this does not work It is likely that either l Common services are down and you do not have communication with the TF In this case the CS most be initialized first Then halt and shutdown the TF via Inspector or python commands as described above before initializing everything again LCU PORT the one that controls the TFs is down This LCU be initialized either by using Crate 27 switchOff or by going to the electronic Cabinet A and using the LCU switch Resetting LCUs do not send commands to the TF then it is safe initializing LCU while TF in operation but then the TFs would be left in the same status forever unless they are initialized again This is not convenient and in addition there is risk of switching off power with the TF in operation or initialized and this can damage the device To avoid that once the system is recovered you must first init the TF and then first halt and after shutdown
60. ked to the corresponding transmission reduction Figure 4 6 Also am etalon must never be placed into the focal plane since coating inhomogeneities are then varying spectral response as a function of target position 1000 2000 3000 4000 5000 6000 Figure 4 4 Image of a Ne spectral lamp illuminating the TF at a central wavelength to the red of the line A ring forms due to the central wavelength variation across the OSIRIS FOV Both CCD detectors are shown without including the gap 300 Ring measured Expected 0 2 A bit 250 200 150 100 Distance to center arcsec Mo 6840 6860 6880 6900 6920 _ 6940 Wavelength Figure 4 5 Distance to the centre of FOV versus wavelength Data represented as blue diamonds obtained from the different ring of Figure 4 4 when varying TF tuning 2 represented on the predicted variation red dots Note the deviation at smaller distances due to the increasing error in radius determination and the slightly different equivalence between wavelength and Z wrt Z Page 20 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC TRANSHISSION 18 100 492 49 595 498 500 50 504 WAVELENGTH Figure 4 6 Change of FWHM and transmission of a filter into a convergent beam The faster the beam the more severe the effect From Atherton et al 1981 4 5 Ghosts One feature of etalons is that they p
61. low FWHM This has to be taken into account when selecting the on and off frequencies With very good approximation the spectral response of a TF given by eq 2 1 can be expressed by 2 l 41 Ed 4 1 where is the wavelength at maximum transmission 4 0 Dimensions The OSIRIS TF are model ET 100 Then the clear aperture is 100 mm diameter mechanical envelopes and installation procedures are provided in the maintenance manual but OSIRIS TF are approximately 170mm diameter by 100 of thickness Its approximate weight is of 8 kg 4 3 Coatings This is a critical aspect of TF performance as shown in section 2 2 For the OSIRIS TF the main difficulty is achieving a relatively constant reflectivity for a wide spectral range from 370 to 670nm for the blue TF and from 650 through 1000nm for the red TF This implies multilayered coatings 1 e thick coatings Then the minimum distance widest FWHM between plates is driven by the minimum distances between the coating surfaces not the Page 18 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC plate surfaces In other words the distance between plates or gap 15 larger Also since the coatings are always somewhat rugged the coatings do not touch uniformly but by small spots across the FOV that progressively increase their size as the plates are pressed together Of course this is an undesirable situation since this could da
62. m Note the cursor position as shown in the inset image above and the extra slit image at the end on the vertical graph In this case 11 5 and then Zbest Zi 10 5x 5 9 Check that the last value of Z is Zlast Zi N 1 AZ 1 the last Z value should be the expected final one minus one step If not it means that some steps has been lost Zi is Zi mAZ with typically m 1 3 Correct Zi accordingly based on the Zlast Page 40 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 10 11 provided by Inspector and counting backwards This should not happen or rarely now but is worth checking Set Zbest in the TF by inserting the Z value in the point deployable menu that appear when cliking on the bullet of moveAbsolute banner blue circle of Figure 7 2 and clicking on the moveAbsolute banner to set it Check that Z value has changed in the TFComponent_Red Displacement Now the TF is theoretically tuned to the line But it might be possible that the parallelism is not correct It will be checked in the next steps using a similar procedure exposing shuffling charge but now changing X or Y instead of Z Take a set of exposures charge shuffling the CCD and varying X from exposure to exposure 350 bits of the tentative initial value It is advisable using 14 steps of 50 bit each In other words N 14 AX 50 Xi X predicted 350 As in Z proce
63. mage the coatings affecting performance and driving to re polish and recoat the plates a major work that would take several months and change the performance and the calibration However the Inspector limits the maximum Z minimum distance achievable so that in principle the TF operator can forget about this feature ET100 OSIRIS Red 94 2 8 620 1000nm ET100 OSIRIS 91 3 390 670nm 10 18 15 8 ES E c s 5 89 5 4 B 3 2 0 04 375 425 475 525 575 625 675 600 650 700 750 800 850 900 950 1000 Wavelength nm Wavelength nm Figure 4 3 Mean transmissions T for the blue left and red right OSIRIS TF The mean reflectivity R 100 T with a very good approximation This results in a mean R 91 for the blue TF and 94 for the red TF The wavelength dependence of the reflectivity R translates into a wavelength dependence of the FWHM range Also please note that the R is well behaved above 425nm for the blue TF and above 650nm for the red TF Hence deviations are expected at lower wavelengths 4 4 Wavelength variation across the FOV The TF as any interference filter changes its response with the incident angle according to the formula Ay 4 2 where is the central wavelength for normal incidence for the incident angle and n the refraction index As a consequence
64. maximum happens between n 11 and 12 It can be either an exact value or a fraction of step that can be easily appreciated yielding the required accuracy of 0 1nm Alternatively a fitting program such as the IRAF routine TFCalibrate cl for example can be used It has been demonstrated that both procedures are equivalent of similar accuracy and speed and that both require experience Please note that the bottom slit image is the first one initial Z value and be sure that you have displayed all slit images 20 in this case Simply count the number n of slit images from the first one to the brightest one both included and the best Z value will be given by Zbest Zi n 1 AZ Please note that the procedure yields an extra final slit image i e 21 images in this case with smaller intensity Figure 7 6 It is produced by a final charge shift and the time elapsed for the shutter to close Just ignore It a ds9 e File Edit View Frame Zoom Scale Color Region WCS Analysis Jan15_211802 fits CCD 2 12 1649 49 000 y 1227 000 33 000 Y 1227 000 0 125 Angle 0 000 view frame color www gt square root histogram r 2000 1000 0 Figure 7 6 Z calibration scan 20 slit images can be seen The first is the bottom one The tuning lies between image 11 and 12 as can be appreciated both from the maximum intensity and symmetry Non symmetric intensities are suspicious of lack of parallelis
65. n and note down TF and telescope temperature TF temperature gradient during the last hour and the offset in Z units wrt calculator e You could use two lines to the blue and red of the required wavelength if desired e Remove ICM and switch lamp off e WHILE starting autoguider tune XYZ for observing an sky calibration ring see 7 2 4 and write down the ring radius and centre as determined by using DS9 this is accurate enough as long as you use a reasonable zoom e Set the required XYZ to observe the target For the red TF the XY dependence wrt and Z is weak This is not the case for the blue TF The correction required will be provided by the TF calculator in the future Meanwhile this procedure is recommended Ideally the wavelength should be within hundred nm and the Z range within few thousand units Page 56 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC e Expose see 8 2 e Every 20 30m check Z calibration observing the sky ring by setting the required XYZ see 7 2 4 Do not forget to reset XYZ for the target again Recalibrate using ICM if required wavelength drift larger than the accuracy required Please log see for example 1 form wavelength shifts together with TF temperature temperature gradients TF cabinet relative humidity and rotator position This cumulative information will be very useful for implementing future temperature corrections and for
66. nd the wavelength at the edge of the target in opposite direction is 654 7nm The photometric errors induced by this line decentring are respectively 7 and 45 from Eq 5 1 if a FWHM of 3 5nm is assumed Then it is possible to choose a TF tuning wavelength that minimizes this variation when the wavelength difference at both edges of the target is the same For example tuning at 657 3nm gives a photometric error equal at both edges of the target and of the order of 25 Of course the velocity field of the target must be taken into account in this procedure since it increase the photometric errors if the velocity field is known it is possible to play around with the target and rotator position to minimize it Were this photometric error too much it is possible to synthesize a wider FWHM by adding TF scans see 5 7 Please do not forget this detail Page 29 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL Figure 5 4 Changing rotator angle is useful for minimizing the wavelength variation across the target 5 5 Removing ghosts cosmic rays and cosmetics As stated above ghosts can be removed using the same dithering that remove cosmic rays and detector cosmetics However dithering move targets on the TF FOV 1 e changes the wavelength at which the target is observed This is specially severe at the edges of the FOV where the wavelength variation is larger Figure 4 5 There are
67. ng emission lines and determining its fluxes and Doppler shifts or redshifts In the case that the TF images are going to be used for continuum subtraction the images of the same scan or several of them added together can be used for this purpose as long as they are separated enough to achieve the required photometric accuracy based on TF FWHM scan step and the number of images Sec 5 1 2 This might require obtaining some additional images at the end and or the beginning of the scan at the end and beginning would allow averaging possible continuum variations 5 6 2 Some applications e Scanning a spectral region for deblending neighbouring lines e Scanning a target looking for systems of high velocity faint or diffuse ionized gas e Scanning blank fields searching for serendipitous emission line targets for example OTELO and HORUS surveys in a certain volume of universe determined by the FOV and the initial and final wavelength for every emission line detected e Scanning the velocity field of galaxy clusters allows determining emission line objects and even the cluster velocity dispersion for example GLACE survey e Scanning a certain emission line of a target of inaccurate redshift 5 7 Band synthesis technique 5 7 1 Technique As before but in this case the final destination is not analyzing images separately but adding them together providing a wider synthetic filter Figure 5 6 The main difference is that in this
68. op an electric field when strained mechanically Under an applied electric field a piezoelectric crystal deforms along all 1ts axes It expands in some directions and contracts in others The dimensional change expansion or contraction of a piezo electric material is a smooth function of the applied electric field The material is sufficiently stiff that piezo electric transducers PZTs can respond on sub microsecond time scales Atherton 1995 The resolution is limited only by the precision with which the electric field can be controlled For this reason PZTs are commonly used for rapid switching and sensing as indeed they are in the Queensgate etalons However all piezo electric materials exhibit hysteresis particularly in the relationship between the voltage applied and the amount of expansion Thus servo control system 15 required to tune the spacing between two plates to high accuracy Page 14 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 2 2 5 Capacitance micrometry In a seminal paper Jones amp Richards 1973 show that capacitance micrometry can be used to detect motions on scales as small as 10 Using this basic method Queensgate Instruments have developed a capacitance bridge system to monitor parallelism and spacing of a Fabry Perot etalon Hicks et al 1984 Information from the capacitance bridge is used to drive PZTs in a closed loop control system to maintain the pa
69. ot require changing pointing Does not require switching autoguider off They require at least 60s exposure Might be tricky to relate ring radius with wavelength More useful for monitoring than calibration Trade off of different TF wavelength calibrators We suggest using ICM for wavelength calibration and perhaps sky ring for checking if the calibration is stable and repeating it using ICM if drifting too much 7 2 6 Measuring TF FWHM Please note that FWHM 1 TF order must not be determined using the ICM since the illumination of the FOV is not representative of that of an astronomical object and larger values are usually obtained FWHM can be safely measured using PN The measures obtained are consistent with those provided by the TF calculator 7 3 Flux calibration Using standard spectrophotometric stars as usual using the same settings than for obtaining the data The STIS Next Generation Spectral Library is recommended In the annexes a list of suitable standards of flat spectra and convenient magnitudes are provided The full list can be accessed at http archive stsci edu prepds stisngsl datalist html Page 50 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 8 OPERATING OSIRIS TF 81 TF setup Set the appropriate XYZ as indicated in the calibration procedure Sec 7 1 2 XY calibration might be required for new Z and wavelength ranges to be done during
70. out any of the above using Output A Output B or both split mode I e every CCD has two possible ADCs Page 52 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 01 05 08 00 Ele Tools Help TE Foo y OOO Tr cites AcquisitonPanel SDSU Controller Shutier Control Lag GIDAS Log GEICA 7 3 0 OSIRIS a Log Y 5 PEE SONC Blu a EEE E E O GS nh IE ramen Acquistion ends OK Cia D alm O DAS State 10 1 al Sem His a E ner E O Lkshoremonitor e ala on 2 id 1075 2099 21481 51 1075 2099 2148 O Cep 4 Output Beans Tuoibux Ocot m orinterinrCarenado O Humidtygensorcabinets Hunliiysensorcabinet O cep z ouput H3 CCD 2 Output B O CCD 1 SPLT 4B O CCD 2 SPLT CCD 1 CCD2 Output A O CCD 1 CCD 2 SPLIT B scneduler OTest v PRESCAN L wNDons s x Frea 298 Bot Comer v Window D A LEFT A LEFT Clea
71. r COMMAND 05725 OSRISIDASIDelsAcquisIton I exposeo ur INFO 05756 Enditeration 56 END 0 57 56 End of Execution Sequence ETE a ara Jr pra ops pos a a 4 E Figure 8 2 CCDs AREA flap of Data Acquisition Window flap of the OSIRIS DAS menu It is possible to choose the CCD to be read out and the Output A B or both to be used Readout speeds of OPTIONS flap Figure 8 3 e Four slow speeds 25 50 100 200 kHz and three fast speeds 500 725 925 kHz are available Recommended ones are 100 or 200 for imaging For quick look acquisition and TF calibration use 500 If aborting and exposure the DAS must be initialized again See troubleshooting section on the OSIRIS USER manual to do that Gain of OPTIONS flap Figure 8 3 e 4 75 This is equivalent to 1 e ADU Page 53 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 01 05 05 00 Tools Help 07030709 runner cos DAS Socuoncor Cong image Monitors Alarme Report Monitors ShitOomiguration WindowConfgursiun DataascuistianComponent Roparshutor Data Accuisiion System s ___ Spit Toots AcauisitionPanel SDSU Controller Shutter Control Log GIDAS Log GEICA i ET E Sel O ans gt EE Panels One AC State Acquisitansystemide 0 EIE Pear O
72. r continuum subtraction the TF tuning of the off band has to be chosen so that no or few emission enters into the continuum filter This is driven by the photometric accuracy required via the following expression derived from Eq 4 1 4 4 1 1 7 Page 26 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL This equation can be interpreted in this case as providing the transmission 7 normalized to unity of your continuum filter tuned at and of FWHM at the wavelength A of the on line tuning For example let assume that the continuum must be chosen with a contribution from the emission line lower that 5 Then for a given FWHM of say 1 8nm the tuning of the continuum must be placed 4nm away from the line tuning Of course wider FWHM require increasing the wavelength difference between on and off line tunings Line ON band OFF band Line ON band OFF band Relative response Relative response 0 0 6540 6550 6560 6570 6580 6590 6600 6610 6620 6630 6640 6540 6550 6560 6570 6580 6590 6600 6610 6620 6630 6640 Wavelength A Wavelength A Figure 5 1 Left A proper tuning of the off line wavelength minimizes contribution from your line to the continuum according to Eq 5 1 Right Increasing the FWHM of the off line tuning will require increasing the wavelength difference between on and off line wavelength tunings Th
73. rallelism and spacing In Figure 3 we show the basic structure of a gap scanning etalon There are two x channel and two y channel capacitors and a fifth reference capacitor which monitors the spacing with respect to a fixed reference capacitor in the circuit The two etalon plates can be kept parallel with an accuracy of 2 200 for many weeks at a time 2 3 Charge shuffling Central to almost all modes of OSIRIS use is charge shuffling Charge shuffling is movement of charge along the CCD between multiple exposures of the same frame before the image is read out For shuffled TF imaging an aperture mask ensures that only a section of the CCD frame is exposed at a time For each exposure the tunable filter is systematically moved to different gap spacings in a process called frequency switching This way a region of sky can be captured at several different wavelengths on a single image Alternatively the TF can be kept at fixed frequency and charge shuffling performed to produce time series exposures The TF plates can be switched anywhere over the physical range at rates in excess of 100 Hz although in most applications these rates rarely exceed 0 1 Hz If a shutter is used this limits the switching rate to about 1 Hz Charge on OSIRIS CCDs can be moved over the full area at rates of 30 50us line it is only when the charge is read out through the amplifiers that this rate is greatly slowed down to the selected readout speed The high cosme
74. rating before every observing run This dependence is barely noticeable in OSIRIS since is masked by temperature variations e Humidity This is likely the most severe although it can be ignored in OSIRIS since the instrument is flushed with dry air e Temperature This is a highly non linear effect where etalon undergoes several phases of different variations ET100 are quite large and take up to three hours to stabilize versus temperature changes However this is not as serious as it seems since implies only calibrating more frequently depending on the history and the temperature gradient It has been demonstrated to be safe operating with TF temperature gradients of at least 0 6 C hour produced by temperature differences between TF and telescope of several degrees as long as calibration is checked every 20 or 30 minutes When the temperature gradient is of the order of 0 1 0 2 C hour the tuning can be considered stable for at least one hour However in a future version of the instrument control the Inspector will take care of this effect at user s request e TF position Depending on instrument rotator angle Up to now extensive and painful tests have been done during several nights at different wavelengths and orders The results are not conclusive sometimes the tuning changes others not without any apparent Please confirm this implementation with OSIRIS responsibles gt However caution must be taken when opening the in
75. re time that you will use now is SIMPLE EXPOSITION The user must have left some margin as indicated in Secs 5 6 and 5 7 Page 51 0170 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC 01 05 08 00 Ele Tools Help EPR NE AcquisitionPanel SDSU Controller Shutter Contral Log GIDAS Log GEICA x 3 0 OSIRIS E Log y Log Y al gt DAS Panel S ONCS a ja F wheel 160 a Acquistion ends OK Cid 0 alm OrAheel DAS State 10 D Focus B amp O LkShorel o li al ei 7 CHARGE TRANSFER SHUTTERLESS FRAME TRANSFER ON W time ms E INFO O05786 End teration 86 18 ves re a soc Para Jr pros opresor pos ja 4 E Figure 8 1 Data Acquisition System flap of the OSIRIS DAS menu The running man for exposing the internal flaps of SEQUENCES CCDs AREA and OPTIONS and the window for setting the exposure time in ms can be seen Readout configurations of CCDs AREA flap self explanatory Figure 8 2 e You can readout only CCDI only CCD2 or both CCD1 CCD2 e You can read
76. roduce ghosts In any astronomical instrument the detector is a source of light any light that is not detected or absorbed is reflected This light reflected by the detector follows the same optical path in opposite direction according to the principle of reversal light in an optical system entering the etalon and reflecting in the most reflective surface 1 that of the reflective coating of the etalon cavity going back and hitting the detector in a place symmetric with respect to the optical centre of etalon This has two important implications for the observer 1 Diametric ghosts are symmetric with respect to the centre of the etalon Figure 4 7 that in the case of OSIRIS is almost the centre of the OSIRIS field in the gap between detectors at the line 1990 But not the centre of the telescope field since OSIRIS is off axis and then is not the pointing centre 2 Diametric ghosts can be easily removed by the classical dithering procedure since moving the image in one direction shifts its ghost in the opposite direction with respect to the TF optical centre Then when piling up the images taking as reference the image of the target all ghosts fall in different pixels and can be removed with average sigma clipping of similar algorithms Page 21 0170 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL 3 Only very bright usually saturated sources generate ghosts Figure 4 7 Hence unless very
77. rum html It is advisable using the sky line at 894 35nm with the OS 878 59 at FWHM 1 21nm The recommended parallelism is of X 34550 Y 34140 and it is recommended tuning the wavelength at 898 2nm for obtaining a ring radius of 1000 pixels Figure 7 11 Sky ring at 894 35nm with the OS 878 59 with FWHM 1 21nm tuning the TF at 898 2nm for obtaining a ring radius of 1100 pixels The exposure time is of 120 5 TF USER MANUAL Code Draft Page 49 of 70 Date June 28 2009 File TFMANUAL DOC 7 2 5 Trade offs Calibrator ICM lamp Advantages Short exposure times No change of telescope pointing required No need for acquiring target on a slit Disadvantages Takes some minutes inserting ICM mirror During calibration might be lost Suitable lines not available in some spectral regions autoguiding Provide good zero redshift calibrations Can be used as flux calibrators for the same price Longer exposure times than ICM lamps Need for acquiring PN on a long slit Need for change pointing Some lines are faint Some lines are not single Available near Galactic plane TypellQSO Calibration be done almost anywhere on the sky Calibration can be done almost anywhere the spectra beyond 500nm Longer exposure times than ICM lamps Need for acquiring QSO on a long slit Need for change pointing The line is fainter than PN The line is broad Sky rings Does n
78. s Ora DM Acquistion ende OK Ci 0 alm 2 DAS idle State E aie aids Monitors 7 JB RI SEQUENCES CCDs AREA OPTIONS shutter Ucehurettonitur Beans Tuoibux ER Ba Fast hus AA O Hurnidiysensorcabineta i al i C Apply Dark Gain pars anniy reordering raw cata Readout Rato KH 600 scneduler OTest ICONMANDI 05725 1 exp0se0 INFO 05756 Enditeration 56 TEND 5578 End of Execution Sequence eere a ara aree pra ops po a 4 Figure 8 3 OPTIONS flap of the Data Acquisition System flap of the OSIRIS DAS menu It allows setting readout speed and gain and taking bias by ticking on Apply Dark option Important notes 1 Do not abort an exposure unless necessary since the DAS will hang up this problem should be solved by now 2 Do not move slits while reading out the CCD Otherwise electronic noise appears in the images However you CAN move any other mechanism wheels collimator TF while reading out 8 3 Shuffled exposures This section is left for a further version devoted to shuffle and fast modes 8 4 Fast modes This section
79. s possible but it cannot be placed in the centre because aside of falling into the gap between detectors a mirrored ghost image of the source will overlap with the real image This might be acceptable in some cases for example if the user is interested in radial dependences only However in general the target should be placed near the optical centre of the TF but fully off it Near the TF centre but on CCD2 the rightmost is a convenient place If the object is elongated the wavelength variation can be minimized by turning the GTC rotator to align the major axis of the target perpendicular to the radial direction for the TF optical centre Figure 5 4 However since the target is not at the centre of the TF the TF must be tuned to the red of the line so that the target is observed at the wavelength of the line required This can be evaluated using Eq 4 3 For example a target of a diameter of 2 arcminutes should be placed somewhat more than arcminute from the centre Then the wavelength corresponding to the zero redshift at this position is not 656 3nm but 656 8nm Even with the above strategy and unless the target is very small the wavelength dependence across the FOV will produce that some parts of the target are observed at different wavelengths The induced photometric error can be evaluated using Eq 4 3 and 5 1 For example for the same example above the wavelength of the edge of the target near the TF centre is 656 8nm a
80. strument for changing filters or masks shortly before observations See environmental conditions in user manual Page 23 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC relation with wavelength order temperature or whatsoever However a range of rotator values where this wavelength variation if present is more severe has been identified from 30 to 60 It is advisable avoiding it whenever possible Otherwise calibrating to compensate TF temperature variations should be enough to compensate for that effect as well The possible origin of this variation with orientation could be attributed for example to relative mechanical shift of the plates of the reference capacitor that thus change its capacity e TF history if plates collide the TF calibration might change This is unlikely to happen since the Z range has been limited to safe values However caution must be taken when reaching values around or above Z 40000 for the red for the blue is still TBD There are several ways for parallelizing and wavelength calibrating a TF The practical procedure used for OSIRIS is described in section 7 As a rule of thumbs when Z increases gap decreases a little bit by hundredths the central wavelength shifts to the blue 1 when observing a ring Figure 4 3 the ring shrinks When Z varies several thousandths 1 a FSR the mode changes and the wavelength as well The larger th
81. the day before observations set using a table of know XY values versus Z and wavelength Calibrate Z and determine the offset with respect to the calculator Write down the TF temperature and the temperature gradient in C hour LKShore 2 number 2 The graph displays the temperature versus time Change scale accordingly You are then ready to expose There are two possible cases of TF operation e Standard tunable imaging shuffling or fast modes where exposures at different wavelengths are taken Wavelength calibration is very critical in this case and special care should be taken e Tunable tomography or band synthesis where a set of exposures at consecutive nearby wavelengths are taken Wavelength calibration is not so critical in this As long as the same reference wavelength for calibration ICM or PN is used although an offset might exists it would affect equally all images In other words images from one day to another will be at the same wavelength although all them might have an offset the same offset for all of them 8 2 Standard exposures Go to Data Acquisition System flap of the DAS menu On CCDs AREA flap in the middle of the screen set the desired readout configuration On OPTIONS flap set the readout speed and gain desired On the SEQUENCES flap set the exposure time in ms To take an exposure just click on the icon of the running man The only exposu
82. the same as the number of DAS loops If this is the case then what has happened is that pick up noise on the DAS CS100 line is interpreted as DAS trigger signals by the TF and the Xi Yi or Zi is not the programmed one but several steps have been added This usually does not happen in X or Y Sweeps but is more common in Z Swinging However please note that this problem is supposed to be already cured Page 45 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC Check the final X Y or Z value by using the low level python commands stop the Z Swinging first on the same terminal from which Inspector was started pythondevelopment with a dot first and a blank space e cd TunableFilters e python sendCommandTF py X Y or Z for getting the current X Y or Z values respectively The correct values should be one step less than the final Z and the final X or Y in the case of X or Y The reason is that when the final Z is exposed in the Swinging procedure the CS100 drives the Z to reversal values Then if the Z Swinging has ended successfully the TF will be left one step less in Z than the final value when stopped If the value of Z is more than one step backwards from the final value then the starting value has been shift forward by the same number of extra steps backwards This allows inferring the correct Zbest from the defective Z Swinging image unless the number of extra steps is too large
83. tic quality of OSIRIS CCD allows moving charge up and down many times before significant signal degradation occurs In this way it is possible to form discrete images taken at different frequencies where each area of the detector may have been shuffled into view many times to average out temporal effects in the atmosphere 2 4 Order sorters A Fabry Perot Filter clearly gives a periodic series of narrow passbands To use a FPF with a single passband it is necessary to suppress the transmission from all the other bands that are potentially detectable This is done by using conventional filters called order sorters because they are used to select the required FPF order 3 TUNABLE IMAGING VERSUS SPECTROSCOPY For a complex instrument such as OSIRIS with a wide variety of observing modes and submodes one of the concerns of an applicant or a SA is whether the chosen mode is the most appropriate for the observing program to be developed Since Tunable Imaging is a relatively new and not widespread mode most confusions arise between the convenience of the use of this mode versus spectroscopy Page 15 of 70 Date June 28 2009 Code Draft File TFMANUAL DOC TF USER MANUAL In brief the main advantages of TF versus spectroscopy is the ability to flux calibrate the emission a tricky issue in MOS and even in long slit spectroscopy slit slicing the image differential refraction centring errors and of obtaining 2D
84. umental effects 1 The spectral response of the TF more peaked and with more wings than a Gaussian or a squared 5 layer interference filter Figure 4 2 2 The diametric ghosts 3 The centre to edge wavelength variation and affects the way to design an observation depending on the characteristics of the sources Photometric accuracy requested Please note that from now on we will refer to photometric accuracy of instrumental origin not due to readout or photon noise Possible neighbouring lines to the one studied e Velocity field or line width of the target e Size of the target e Redshift of the target playing with the possible variables e TF central wavelength for your On and Off images e TF FWHM e Position of the source in the FOV Page 25 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC e Rotator position These mark the difference with respect to direct imaging with conventional filters and of obtaining useful data instead of wasting observing time obtaining useless data In other words in tunable imaging the design of an observation is extremely important as the observing procedure is In the following sections will derive recommended observing strategies depending on the type of sources and the scientific aims by playing with the previous parameters 51 Selecting off band 5 1 1 Continuum subtraction In line imaging two images are usually required the on lin
85. va XML calculator Temperature humidity and rotator variations does not change the shape of the Z A relation but only the offset between the value predicted by the calculator and the measured one Extensive and detailed measures show that this offset can be considered constant along Z and wavelength with an accuracy of 5 in Z 1 e better than 0 1nm Anyway and for securing calibration it is advisable determining the offset for Z and A values as near as possible the requested ones Then wavelength calibrating the OSIRIS TF consist in determining this Page 46 of 70 Date June 28 2009 TF USER MANUAL Code Draft File TFMANUAL DOC offset This can be done either using a calibration lamp of the ICM or using astronomical targets 7 2 2 Calibration using the ICM The calibration procedure using the ICM has already been described within the parallelization procedure of the previous section It is worth reminding that an accurate Z can be obtained only after parallelization 1 e determining the best XY values for the given range of Z and wavelength 7 2 3 Calibration using astronomical targets However since single and intense lines must be used and given that it is difficult to have an extensive set of lamps of different species and that lamps of some elements are not available in pencil beam format 1 Helium it is sometimes advisable using astronomical objects for wavelength calibration This objects must ful
86. variation changes are more noticeable near high Z values and at the reddest A In general XY should be checked in ranges of say 5000 in Z and 50nm in wavelength more frequently beyond 850nm were coatings vary most A table with the parallelism values determined so far is included in Appendix C 1 712 OSIRIS TF parallelization procedure The easiest way to parallelize an etalon is visually observing the rings obtained using viz a fluorescent light and hand tuning XY using the CS 100 Since the parallelism is very robust it will be kept after switching off the CS 100 and mounting the TF on the instrument However a procedure for parallelization with the TF mounted in the instrument is necessary specially bearing in mind that there is a parallelism variation with Z and A Several procedures have been devised to this purpose and here a suitable one is described both for parallelism and wavelength calibration This parallelisation procedure for the TF is a task to be done during Day time The basis consist in maximizing the intensity of the light in the optical centre of the TF when tuned for observing an emission line coming from a calibration lamp when varying X and Y This is the same procedure to be employed for wavelength calibration but then varying Z A lack of parallelism XY or a lack or tuning Z will reduce the intensity measured This procedure is achieved by inserting a wide centred long slit and charge shuffling the charge on
87. w Frame Zoom Scale Color Region WCS Analysis File Jan02_044114 fits CCD 2 L2 3534 65 000 1583 000 115 000 Y 1683 000 0 125 Angle 0 000 view frame una power square root 3000 2000 1000 0 Figure 7 9 Example of a Y calibration image of 8 steps of 25 bit Please note that in the Y calibration the slit image intensities are not symmetric This is normal The final spurious slit image can almost be appreciated in the vertical graph In this case n 5 and then Ybest Yi 4x25 21 Calibrate Z again since changing X or Y might change Z 22 If X and Y were 50 and 25 respectively of the initial tentative values you have finished If not repeat the procedure since changing X Y might change Y X for achieving parallelism Once finished you have XYZ for the required Z range and wavelength Of course if XY were already known you can proceed directly to calibrate Z by setting XY and ignoring the previous steps accordingly Feel free to repeat any calibration step if you are not confident with the result Best wasting some time in calibrating than losing much more time throwing your data to the trash after trying painfully to reduce them This procedure looks more complicated than it really is It is more difficult writing the procedure than understanding or executing it An experienced user can do the previous calibration sequence in few minutes NOTE X is less sensitive to lack

Phone (922)605200 -Fax (922)

Contents

Download Pdf Manuals

Related Search

Related Contents