AMBER User Manual
Contents
1. As stated in VLT I User Manual the user has the possibility to use a guide star for the Coude systems different from the target Refer to this manual for the limitations of this option 5 Fringe Search Search the optical path length OPL offset of the tracking delay lines yielding fringes on AMBER actual zero OPD by OPD scans at different offsets When fringes are found the atmospheric piston is calculated and the OPL offsets corresponding to zero OPD are applied 6 If FINITO is used the above step is performed by FINITO and not by AMBER 7 Observations Start to record data of interest with suitable DIT In P96 it is foreseen to only use DITs of 25ms or 50 ms for standard absolute phase observations and DITs of 100 ms for differential phase observations The longer DIT allows a larger wavelength range in MR K MR H or HR K observations 8 If FINITO is used longer DITs are available AMBER User Manual VLT MAN ESO 15830 3522 15 7 2 Computing time overheads for added bands NEWS As of P93 the execution time of MR and HR OBs is reduced from 25 to 20 minutes This implies that the execution of an OB is 20 minutes in each of the three spectral settings Hence the default CAL SCI CAL sequence requires 60min 3 OBs regardless of spectral setting Users interested in several spectral positions should add 15 minutes for each additional spectral band per OB Similarly users interested in repeating the same spectral band to obtain mo2. contrast by a factor exp 0 where oa is the fringe phase standard deviation over the frame acquisition time Therefore since the jitter varies with time this attenuation factor is unfortunately not stable and there is a high probably that data taken several minutes later on the calibrator will not allow to cancel the term and will result in final biased visibilities Since FINITO measures fringe phase several times during one AMBER frame acquisition typical integration time is of the order of 1ms it provides a way to compute a contemporaneous estimation of the attenuation factor Therefore an a posteriori frame to frame correction is possible 8 4 2 Application The commissioning of the VLT I Reflective Memory Network Recorder RMNrec in February 2008 has made possible to store the real time FINITO OPDC Optical Path Difference Controller Ma chine and Delay Lines data into proper FITS files Preliminary results published by Lebouquin et al SPIE 2008 7013 p33 Sch ller et al eds have shown encouraging perspectives for AMBER data post processing using FINITO data These results have been confirmed by technical tests which have shown the possibility for a very significant increase in visibility precision and have motivated the decision to include FINITO data within AMBER data However the reader should be warned that performant corrections can only be reached if FINITO performs well cophasing i e if the source is bright and not too res
3. defined Template Signature File Unit telescope 8m VLT I Main Array array of 4 UTs VLT INterferometric Commissioning Instrument VLT I Sub Array array of ATs Very Large Telescope Very Large Telescope Interferometer Visitor mode 000 AMBER User Manual VLT MAN ESO 15830 3522 23 Ck12 4 dkio N Choa amp dk23 O gt 0 2 10 20 30 40 el 0 2 LO 03 0 0 4 ae x 0 2 10 20 30 40 Pixels Figure 5 Example of carrying wave c k solid green line and d k dashed red line AMBER User Manual VLT MAN ESO 15830 3522 24 ACQ OBS Std Figure 6 Standard observation mode Std
4. each individual aperture i e 250mas for the ATs in K and 60 mas for the UTs in K For most observations this will not have consequences but can be limiting to observations of objects that consists of several components e g binaries stars with disk and or winds etc that have a spatial extension equal or superior than the interferometric FoV While such observations are not impossible the observer will have to take into account this incoherent flux contribution in his data analysis 6 1 5 Complex fields When observing complex fields within a few arcseconds it is necessary that MACAO STRAP behaves very well in order to disentangle the desired object from others see VLT I Users Manual for seeing and limiting magnitude of STRAP MACAO However for fields with several objects within 1 to 3 arcseconds it is not guaranteed that MACAO will perform properly It is therefore recommended to use a guide star in this situation For fields with objects with separations less than an arcsecond MACAO will resolve the objects down to 0 1 0 15 arcsec For separations smaller than 0 3 arcsec it cannot be guaranteed that the proper target has been injected into the fiber These acquisitions have to follow a non standard extensive procedure and require the presence of the PI in Visitor Mode 6 1 6 Bright objects In Low Resolution observations LR of very bright objects mag lt 0 the detector can saturate even when using Neutral density filters during excell
5. limited Detailed information on the wavelenght ranges can be found in this table The closure phase error in low resolution is dominated by some not understood systematics We believe it is not possible to reach a better precision even by stacking frames e Medium resolution H band data suffer from systematics in the phases not understood at the moment Fast optical path difference fluctuations due to vibrations and atmosphere lead to e a decreased instrument contrast e a degraded instrumental contrast stability and therefore a degraded final precision and accuracy of calibrated visibility For that purpose when FINITO is used as a fringe tracker the fringe tracking data is now recorded to allow post processing of AMBER visibilities see Sect 8 4 5 AMBER in P96 In P96 the modes offered are the High Resolution K band HR K Medium Resolution K band MR K and H band MR H and the low resolution K and H bands LR HK with a spectral resolution A AA of approximately 12000 1500 and 35 respectively In LR HK mode the K band will be acquired simultaneously with the H band The laboratory field stabilizing instrument IRIS is always used during an observation and it uses 25 of the K band flux Additionally FINITO uses 70 of the H band flux Since P84 FINITO is part of the standard mode in HR K MR K and MR H see here Any proposal asking not to use FINITO in these modes should properly explain the reason why and requi
6. the instrument discussing the instrument layout A reference to instrument specifics to be kept in mind while planning the observations or when reducing the data can be found in Sect 4 In Sect 5 additional information pertinent to observing with AMBER in P96 are presented Sect 6 and 7 provide the basic information needed to prepare an observing program Finally Sect 8 presents the current calibration plan for AMBER data and a description of FINITO fringe track data 1 4 Contact Information This document is evolving continually and is updated according to changes in Paranal operations of AMBER or on request by the AMBER user community All questions and suggestions should be chan neled through the ESO User Support Department email usd help eso org and home page http www eso org sci observing phase2 USD MIDI html The AMBER home page is located at the following URL http ww eso org sci facilities paranal instruments amber inst Any AMBER user should visit the instrument home page on a regular basis in order to be informed about the current instrument status and any late development 2 Context 2 1 Is AMBER the right instrument for your program AMBER is designed to deliver very high angular resolution information on celestial sources The instrument provides in a decoupling of the single telescopic diffraction limit from the angular reso lution limit owing to the instrument s conjunction with the VLT I telescope delay line ar
7. EUROPEAN SOUTHERN OBSERVATORY Organisation Europ ene pour des Recherches Astronomiques dans 1 H misphere Austral Europ ische Organisation fiir astronomische Forschung in der s dlichen Hemisphare ESO European Southern Observatory Karl Schwarzschild Str 2 D 85748 Garching bei M nchen O Very Large Telescope Paranal Science Operations AMBER User Manual Doc No VLT MAN ESO 15830 3522 Issue 96 Date 27 02 2015 W J de Wit 27 02 2015 ENEE EE ee doh oe E EE ee eee oor eet Date Signature C Dumas Approved s dig EEN Date Signature A Kaufer WN AON ect dress EE NE BONA HON E Swe ee ee a Date Signature AMBER User Manual VLT MAN ESO 15830 3522 This page was intentionally left blank AMBER User Manual VLT MAN ESO 15830 3522 iii Change Record Issue Rev Date Section Parag affected Remarks 83 0 2008 09 03 Various sections FINITO use 1 Limiting Magnitudes 83 1 2008 09 12 1 Limiting Magnitudes 84 0 2009 02 07 most of the doc removed all VLTI specific parts 2 6 performance table 85 0 2009 08 26 8 a separate section for Calibration plan cold darks 86 0 2010 02 27 7 2 1 simple OB is now 25 minutes in LR 2 6 updated performances in HR K 6 1 CAL SCI CAL sequence is default 87 0 2010 08 29 8 FINITO tracking info now recorded 89 0 2011 08 29 7 2 OB duration modified LR 20min MR HR 25min 8 Amber self coherencing described Minor improv
8. RARCH ESO DEL FT SENSOR is set to FINITO if FINITO was used NONE otherwise FINITO beams are denoted 0 1 2 The correspondance can be quite confusing bu the following table gives the standard configuration Input AMBER FINITO channel 1 1 2 3 2 5 3 1 Other FINITO important parameters can be found with keywords starting with HIERARCH ESO ISS FNT Timing issue Each AMBER frame has a time stamp in MJD column TIME RMNREC data use microseconds since the date HIERARCH ESO PCR ACQ START in the main header The two are not synchronized perfectly because e Unlike FINITO AMBER is not on the reflective memory network RMN of the VLTI it means fine time alignement is required in post processing to aligne FINITO data on the AMBER data e AMBER frames are tagged in MJD with a time accuracy of le 8 days or 0 87 milliseconds This is an ESO standard and cannot be changed easily While converting UT times to MJD It is a good idea to check the UT date to MJD formula using the Header the values of MJD OBS and DATE OBS AMBER User Manual VLT MAN ESO 15830 3522 18 8 5 2 OPDC1 OPDC2 These two tables are for the channel 1 and 2 of FINITO which corresponds to the optical combination of FNTO FNT1 and FNTO FNT2 respectively that is AMBER beam2 beam3 and AMBER beaml beam2 e TIME in micro seconds since HIERARCH ESO PCR ACQ START e rtOffset real time offset pure accumulated tracking of FINITO in met
9. apply i e the grading will be based on the first cal sci cal sequence only e Owing to the AMBER intervention of Jan 2013 the AMBER limiting magnitudes have been adjusted to somewhat fainter magnitude e Since P88 the generated data files contain in addition to the AMBER data also the FINITO fringe track data indeed when FINITO is used to stabilize the AMBER fringes These data allow a posteriori calibration and frame selection see Sect 8 The corresponding header and data keywords are described in Sect 8 4 e Since P91 AMBER employes group delay tracking i e sel coherencing in both visitor and service mode when FINITO is not used Sect 2 6 2 e The on sky sequence of observations delivering a calibrated measurement is re quired to be of the type calibrator science calibrator CAL SCI CAL because of the intrinsic instability of the AMBER transfer function Programs interested in differential measurements only will be allowed to execute a CAL SCTI sequence AMBER User Manual VLT MAN ESO 15830 3522 2 Before initiating an AMBER proposal the user is kindly advised to consult the AMBER web pages for any late information not included in this manual The AMBER pages can be found at this URL http www eso org sci facilities paranal instruments amber 1 3 Contents of the manual Sect 2 presents a quick overview of interferometry and details on what AMBER can deliver Sect 3 provides a technical description of
10. ase products suffer from systematics not understood at the moment Usually the use of the fringe tracker biases the calibrated visibility The main source of bias when using the fringe tracker is when a jump of one fringe does not correspond to a jump of one fringe in the science channel FINITO operates in the H band hence AMBER H band data collected using FINITO in cophasing are much less biased than medium K data The precision and accuracy of visibilities can be significantly increased by using a posteriori visibility calibration using FINITO recorded data Technical tests in low spectral resolution AMBER User Manual VLT MAN ESO 15830 3522 7 mode with an excellent fringe tracking performance cophasing have shown that precisions as good as 1 on squared visibilities could be reached on bright targets See 8 for a more detailed explanation 2 6 2 Self coherencing AMBER employes the ability to track the fringes in order to maintain induced optical path fluc tuations piston well within the coherence length This was offered in visitor mode since P89 and in both visitor mode and service mode since P91 Self coherencing also known as group delay tracking is always employed when FINITO is not used At each frame acquisition a quick look software extracts the main observables from the data the fringes amplitude signal to noise and piston The determined optical path correction is sent as an offset to the delay lines Since t
11. at least 1 magnitude brighter AMBER User Manual VLT MAN ESO 15830 3522 6 Table 1 AMBER characteristics Description Specification Number of beams Three Spectral coverage JHK 1 2 5 um Spectral resolution in K R 35 R 1500 R 12000 Spectral resolution in J amp H same as in K Instrument contrast 0 8 Optical throughput 2 in K 1 in J and H Detector size 1024 x 1024 detector array Detector read out noise 11 377 Detector quantum efficiency 0 8 Observables Visibilities V FS A Differential visibilities V f A V F Ao Differential phase Closure phase ITA EIER f Ao than the limiting magnitudes and with a standard number of frames taken Better performances can be obtained in better conditions or by stacking more frames should be specifically asked see foot notes 11234 for exceptions NG means not guaranteed KA w mode FINITO calibrated V diff CP low HK not used 10 NG bel coherencing 5 NG 3 cophasing 7 NG 3 medium K coherencing 5 Qo de cophasing 5 de Pe medium H any mode 5 202 4 high K cophasing 5 Ta 528 The closure phase error in low resolution is dominated by systematics namelly a strong depen dency of the closure phase with the piston fringes phase shift or OPD shift We believe is is not possible to reach a better precision even by stacking frames The medium H band ph
12. ation 2 4 4 Closure phase The closure phase the sum of the phases of the 3 baselines inside a triangle is independent from any atmospheric and instrumental phase offsets It is therefore a very robust quantity in terms of calibration stability 2 4 5 Image reconstruction Image reconstruction consists in computing an approximation of the object brightness distribution out of the Fourier components measured by the interferometer In order to get a meaningful image it is important to measure the maximum number of spatial frequencies in the u v plane This iterative procedure can be carried with several software tools that have been specifically developed for optical interferometry and take into account among other things the sparcity of the coverage and the lack of phase information However the superiority of image reconstruction to model fitting can only appear with a significant paving of the u v plane In this manual we do no address the question of model fitting or image reconstruction We focus on the description of AMBER operation and its interferometric observables extraction 2 5 AMBER characteristics The main characteristics of AMBER are summarized in Table 1 For offered modes see the AMBER web pages 2 6 AMBER performances 2 6 1 AMBER accuracies The following table shows the typical observables accuracies in good conditions seeing of 0 8 with the UTs 0 6 with the ATs coherence time of 4ms or better for targets
13. c fringes obtained on a celestial source with a telescope baseline B and light wavelength A yield the amplitude and phase of a Fourier transform component of the source brightness distribution at the spatial frequency f B X If the full Fourier transform is sufficiently sampled i e the spatial power spectrum of the source s brightness distribution is sampled at many different spatial frequencies then an inverse Fourier transform yields a model independent reconstruction of the source brightness distribution at the wavelength and an angular resolution A Bmax There are two ways to collect and sample the Fourier transform of spatial information in order to assess the geometry of the source 1 obtain data on different baselines triplets 2 rely of the natural super synthesis by earth rotation and 3 the fact that AMBER records data simultaneously in many spectral channels Currently most AMBER observations do not aim for image reconstruction since it requires a very AMBER User Manual VLT MAN ESO 15830 3522 4 large quantity of data hence a great amount of observing time Not all science programs have goals which need imaging because the information provided by one AMBER single observation is already rich There are different observables which can be grouped as follow e the visibility amplitude is related to the object s projected size along the projected baseline vector The morphology of the object can therefore be retrieved throu
14. chitecture It delivers a far higher angular resolution than any other ESO instrument It also has some strong limitations which one should be aware of in order to make sure that AMBER is the right instrument for a given research program AMBER does not return a mirror image of a luminous source on the sky Instead it combines the light coming from three telescopes either the auxiliary or unit telescopes which creates fringes between each telescope pair see Fig 1 Each of the three fringe system is characterized by its contrast also called visibility and phase These quantities are related to the brightness distribution of the celestial object In addition AMBER disperses the combined light and thus delivers spectrally dispersed data at very high angular resolution i e spectro interferometry If your target has a characteristic size in the range 2 30 milli arcsecond and it is brighter than K 9 then AMBER can probably bring you information any other ESO instrument cannot 2 2 AMBER and other ESO instruments AMBER yields information at angular resolution scales between A D and A D B being the telescopes separations ranging from 16m to 130m and D the diameters of the telescopes 8m for the UTs and 1 8m for the ATs A single mode instrument like AMBER has no direct access to structures larger then A D One might need in certain cases information at small spatial frequencies i e larger scales in order to complete the data collected
15. e correlated flux 2 P P Vj for all baselines and M k is a matrix with the values of the carrying waves cij k and dij k The matrix M k is the so called pixel to visibility matrix P2VM The AMBER internal calibration process consists into measuring this P2VM The P2VM calibration procedure occurs every time the instrumental setup is changed The P2VM is automatically included in the standard templates and thus requires no input or configuration by the observer AMBER User Manual VLT MAN ESO 15830 3522 11 4 Instrument limitations and problems The following caveats and limitations should be taken into consideration e Mechanical vibrations When using UTs the VLT I makes use of system that actively reduces the effect of the mechanical vibrations of the telescopes This system called Manhattan2 helps to reduce the optical path variations of the beams but its use makes sense only when the FINITO is used The reason is that the piston introduced by telescope vibrations is much smaller than that introduced by atmospheric turbulence Residual vibrations may still exist however and it prevents a stable transfer function Absolute data calibration is therefore much better when using the ATs Spectral range When FINITO is used for any given spectral mode the full spectral range can be read out on the detector When FINITO cannot be used and the DIT is required to be short then the spectral range which can be read out from the detector is
16. e ee EA 5 201 AMBER accuracies 4 4 a4 a Ge ee ea bk ARR EN ERR PE ek RE 5 26 2 Self coherencing 2 62245 4e 224 42 DRY a aa ee Se ee 7 26 3 Polarization control oo e sa a oroi Sa A LN AE EN AA 7 2 6 4 Performances issues prior to P86 7 3 AMBER overview 7 3 1 AMBER principle 454647486 ek Ako a a a e a 7 3 2 AMBER layout 0 064 4056 ee ens caes a a e aod 8 Sal Warm OPIS ooo a e ct ae ew E EE as 8 eee APOCO colo a a RoR ee i E da e ENT A ra Bo 10 Sito Detector 22 554 66 86 A AAA e eee 10 Bee e E GT E eh ee ae ea 10 38 3 From images to visibilities gt s c s aas sabs 285844 soaa SG RY ER ea es 10 4 Instrument limitations and problems 11 5 AMBER in P96 11 5 1 Service and Visitor Modes s lt bc ee da tya 12 6 Preparing the observations 12 6 1 Proposal guidelines 2 0 20 8254 822 bee eH RRR ee ee ee 12 6 1 1 Guaranteed time observation Objects ss p a sacs eaa osaakaan seun 12 6 1 2 Time critical combination of triplets oaoa 12 Gio Calibrator Stars 2 456 ge pe ae ee AER RA AAA AR 12 Gila Pedal View lt 2 2 eo a Aaa a 13 AMBER User Manual VLT MAN ESO 15830 3522 6 120 Complex Nelda 2 24454 2 a a o ane a a G ai ee ee gi 6 1 6 Bright objects sas sos gaa a a a a as 6 2 Choice of the AMBER configuration 6 2 1 Instrument set up s e ca dba be eS ee ea ee ada a ee 8 G22 Observing modes s s i ocs a ee ble eae Roe eo oe PA Oe ee 7 Introducing Observat
17. e the ESOFORM package for details 6 1 3 Calibrator Stars The user should use appropriate calibrator stars in terms of target proximity magnitude and apparent diameter It should be provided by the user with the submission of the Phase2 material To help the user to select a calibrator a tool called CalVin is provided by ESO see here http www eso org observing etc Three calibration sequences are offered in service mode e SCI CAL first science then calibrator is reserved to program only requesting differential quan tities This is the default mode for the HR setting as the HR setting does not allow any reliable absolute visibility calibration AMBER User Manual VLT MAN ESO 15830 3522 13 e CAL SCI CAL science bracketed by calibration is mandatory for any program requesting absolute calibration and therefore is the default mode e As of P93 a long calibration sequence of CAL SCI CAL SCI CAL is offered Restrictions apply only LR seeing lt 1 2 and THN conditions Regular rules regarding successful execution of containers with long execution times apply i e the grading will be based on the first cal sci cal sequence only see this webpage under topic OB execution times must be below 1 hour A waiver is mandatory Details on calibrating the science target can be found in Section 8 6 1 4 Field of View AMBER is a single mode instrument and therefore the field of view FoV is limited to the Airy disk of
18. ed just the ratio between the visibility at a given wavelength and a reference channel Another possible application of the differential visibility if the study of objects with angular char acteristic of the order of A2 BAA AA is the wavelength range the visibility will vary inside the recorded band due to the super synthesis effect This is for example a powerful tool to detect and characterize binary with separation a A BAA 2 4 3 Differential phase Because the instrument delivers spectrally dispersed fringe information one can measure variations of the phase with the wavelength The principle is exactly the same as in astrometry except that the reference is the source itself at a given wavelength The most remarkable aspect of this phase variation is that it yields angular information on objects which can be much smaller than the interferometer resolution limit These features come from the possibility to measure phase variations much smaller than 27 ie 14 When the object is non resolved the phase variation f f Ao yields the variation with wavelength of the object photocenter e A e Ao This photocenter variation is a powerful tool to constrain the morphology and the kinematics of objects where spectral features result from large scale relatively to the scale of the source spatial features Note that if this is attempted over large wavelength ranges the atmospheric effects have to be corrected in the data interpret
19. editor Society of Photo Optical Instrumentation Engineers SPTE Conference Series volume 5491 page 528 October 2004 First result with AMBER FINITO on the VLTI the high precision angular diameter of V3879 Sagittarii J B Le Bouquin B Bauvir P Haguenauer M Schller F Rantakyr and S Menardi A amp A 481 553557 April 2008 Fringe tracking performance monitoring FINITO at VLTI A M rand F Patru J P Berger I Percheron and S Poupar In Society of Photo Optical Instrumentation Engineers SPIE Conference Series volume 8445 July 2012 Perspectives for the AMBER Beam Combiner A M rand S Stefl P Bourget A Ramirez F Patru P Haguenauer and S Brillant In Society of Photo Optical Instrumentation Engineers SPIE Conference Series volume 7734 July 2010 Glossary Constraint Set CS List of requirements for the conditions of the observation that is given inside an OB OBs are only executed under this set of minimum conditions Observation Block OB An Observation Block is the smallest schedulable entity for the VLT It consists of a sequence of Templates Usually one Observation Block include one target acquisition and one or several templates for exposures Observation Description OD A sequence of templates used to specify the observing sequences within one or more OBs Proposal Preparation and Submission Phase I The Phase I begins right after the Call for Proposal CfP and ends at the deadl
20. ements to the document 90 0 2012 02 24 1 Simple update No new features 91 0 2012 08 26 8 4 FINITO RMNREC keywords explained 92 0 2013 02 25 Various sections Streamlining and upating 92 1 2013 03 12 8 5 Background on RMNREC data 93 0 2013 08 30 7 2 LR MR HR 20 minute OB duration 6 13 long calibration sequence 6 1 3 specifics cal sci for HR 94 0 2014 02 28 2 6 3 polarization control 96 0 2015 02 28 No changes AMBER User Manual VLT MAN ESO 15830 3522 This page was intentionally left blank lv AMBER User Manual VLT MAN ESO 15830 3522 v Contents 1 INTRODUCTION 1 UM o e yaa E A eae Ge EM ats g ae a ee GA 1 12 AMBER newssectioni sc sto podiet aa eg e e E a EE DS 1 1 3 Contents of the manual s lt a sa eatr a a ne a ee A 2 LA Contact Information 01 e fab A eb ea a oe A ee 2 2 Context 2 2 1 Is AMBER the right instrument for your program o 2 2 2 AMBER and other ESO instruments 0 000002 ee ee eee 2 Zo Optical interferometry DaAsiCS 2 ace TN eee ee a i BE e a A a 3 2 4 AMBER observables ee 4 241 Absoluta visibility VPA AN ewe EAR ER rada a OS 4 24 2 Differential visibility V J A V CAG lt lt lt ARA 4 24 3 Differential phase coso rie EN e EE LN e AA a H 5 Uae Close PASE saw ss hara A A A we eRe Re ES 5 2 4 5 Image reconstruction 2 5 25 AMBER characteristice ccs sopce aoe e oe ee a bok bee Pak AO Pe ds 5 2 6 AMBER performances lt s sose w sawa s bee eee ee EN e
21. ent weather conditions The user should consult the webpages for the latest information on the magnitude limits If possible the user should try to use the MR spectral configuration if the scientific goals still can be achieved in this mode 6 2 Choice of the AMBER configuration 6 2 1 Instrument set up The instrument set up is defined by the spectral configuration of the instrument and the 3T config uration Any change of the spectral configuration requires an internal calibration i e spectral calibration and P2VM calibration This is automatically taken care of by the internal calibration plan and no action or setups are needed from the user Note that only one spectral configuration is allowed in one OB Any change of the neutral densities the ADC the position of the fiber heads i e all elements located before the spatial filters does not require internal calibrations They can be used or not depending on the source characteristics AMBER User Manual VLT MAN ESO 15830 3522 14 6 2 2 Observing modes Without FINITO only fixed DITs of 25 50 or 100 ms ATs only are offered With FINITO longer DITs are available In MR K and HR K the choice of short DIT restrict the width of the central band The details on the exact wavelength ranges DITs and central wavelengths available can be found on the AMBER Instrument webpage http www eso org instruments amber index html 7 Introducing Observation Blocks OBs For general VLT instrume
22. ers e fringeFlag obsolete e offValid obsolete e opdcState state machine controller state 0 idle 1 fringe search 2 on hold 3 group delay jump 4 group delay tracking 5 on hold 6 phase jump 7 phase tracking e uwrapPhase unwrapped phase This is actually in radians not meters files before April 2013 were wrongly labeled e fullOffset offset between the zero OPD prediction and actual position including instrument offset refraction and so on This is in meters During the states 2 and 5 no fringe tracking commands are sent as the controller waits for the signal to noise ratio to rise above a given level known as the close level In states 3 and 6 the controller decided that the offset between the target and current position is too large and needs to be corrected via a jump 8 5 3 FNT1 FNT2 As for the OPDC tables these two tables are for the channel 1 and 2 of FINITO which corresponds to the optical combination of FNTO FNT1 and FNTO FNT2 respectively that is AMBER beam2 beam3 and AMBER beam1 beam2 e TIME in micro seconds since HIERARCH ESO PCR ACQ START e Coher group delay in radians not meters files before April 2013 were wrongly labeled e CoherFlag Obsolete e Phase fringes phase as measured by FINITO in radians not meters files before April 2013 were wrongly labeled e PhaseFlag Obsolete AMBER User Manual VLT MAN ESO 15830 3522 19 e SNR Signal to noise ratio of the fr
23. function C Pinst CPmeasurea C Poxpected CAL the calibrated phase closure is estimated by CP C Pmeasurea CU Poet Ve asia V pocted C AL the calibrated visibility is estimated Other quantities can be calibrated for example the chromatic phase dispersion The chromatic phase dispersion is a function of the air path between each pair of telescopes With many CAL at different DL stroke one can compute a polynomial fit to the differential phase and extrapolate the polynomials coefficients as a function of air path difference All calibrator stars observation DPR CATG CAL are made public by ESO so users can retrieve all calibrators taken in a given night in order to refine their estimation of the transfer function Sequence CAL SCI CAL should be used if absolute products will be used this is the most common case Some particular programs only require differential interpretation users should use the SCI CAL sequence for this special programs 8 4 FINITO fringe tracking information 8 4 1 Principle Even during the shortest AMBER integration times 25 ms and with FINITO operating correctly the random optical path fluctuations jitter have sufficient amplitude to lead to 1 a contrast decrease therefore a bias and 2 an unstability of the visibility transfer function Both linked phenomenon contribute to a significant decrease of AMBER performances During one AMBER frame acquisition residual fringe motion reduces the fringe
24. gh a modelling of the brightness distribution Visibility will not be sensitive to non centro symmetric brightness distributions e The phase is not directly measurable by AMBER However differential phase and closure phase the phase of the so called bispectrum are measurable The closure phase and the differential phase are powerful tools to investigate asymmetry in the source geometry It is important to note that the wavelength dispersion gives spatial information of two different kinds On the one hand there is the spectral information which allows to study the characteristic size of emission line regions absorption line regions e g with respect to the continuum emission On the other hand the wavelength plays a role because different wavelengths have different spatial resolutions B A In other words the spectral dispersion helps to fill up the spatial frequency space called also u v plane after the usual variables for the spatial frequencies One should constantly keep in mind these two complementary roles of the wavelength dispersion 2 4 AMBER observables We introduce here the observables which can be extracted from AMBER data The instrument delivers the following quantities for spectral resolutions of 35 1500 and 12000 and a spectral coverage involving the K H and J bands see web pages e the absolute visibility in each spectral channel e the differential visibility i e the ratio between the visibility in each spec
25. he AMBER observables depend on the self coherencing performances technical validations show a sig nificant improvement of the data quality While these numbers are to be taken with caution good conditions relatively bright sources the instrumental transfer function level may increase by several 10 and its stability improved with the largest effect in the low spectral resolution mode Closure phase accuracy is also significantly increased This operating mode can be desactivated upon request in the README Proposals requiring performances better than these should state how they are going to be obtained special calibration large data set etc 2 6 3 Polarization control Birefringence is the optical property of a material having a refractive index that depends on the polarization As the pathlength of the two polarizations in a birefringent di electric is different a phase delay is introduced AMBER s monomode fibers are birefringent Overlaying the fringe systems of the two polarizations introduces therefore a loss in fringe contrast Previous standard AMBER operation would therefore detect the fringes of only one of the two polarizations With the installation of the birefringent Niobate plates it is now feasible to correct for the polarization effect introduced by the mono mode fibres As a result fringe systems can be overlapped and all the light that reaches the detector can be used This is the reason for the 0 5 increase in the K ba
26. he dispersed image on the detector chip DET 3 2 4 Calibration unit The Calibration and Alignment Unit OPM CAU emulates the VLT I for test and calibration pur poses The matrix calibration system OPM MCS is set of plane parallel plates which can be introduced in the beam sent by the OPM CAU in order to introduce the A 4 delays in one beam necessary to calibrate the instrument To increase the instrument contrast one polarization is eliminated by a polarization filter OPM POL located after the dichroics 3 3 From images to visibilities The raw data produced by AMBER are images of the coherent overlap of the 3 beams dispersed by a prism LR or grisms MR and HR One get in addition 3 photometric outputs corresponding to each beam An image of the detector is displayed in Fig 1 The fringes are processed for each wavelength individually The first action consists in separating the 3 fringes pattern apart During the calibration the fringes corresponding to each baseline is recorded The interference term of the base ij is for the pixel k mij k 21 PiP ci k Viz cos Bi k dig Kk Vig sin D 5 k 1 The quantities c k and d j k are called the carrying waves and are displayed in Fig 5 The interferogram subtracted from photometry can write corr k as icorr k Y milk 2 j gt t M k x C 3 where C is a vector of the values Rij Tij corresponding respectively to the real and imaginary part of th
27. ine for CfP During this period the potential users are invited to prepare and submit scientific proposals For more information http www eso org observing proposals index html AMBER User Manual VLT MAN ESO 15830 3522 21 Phase II Proposal Preparation P2PP Once proposals have been approved by the ESO Observation Program Committee OPC users are notified and the Phase II begins In this phase users are requested to prepare their accepted proposals in the form of OBs and to sub mit them by Internet in case of Service mode The software tool used to build OBs is called the P2PP tool It is distributed by ESO and can be installed on the personal computer of the user See http www eso org observing p2pp Service Mode SM In Service Mode opposite of the Visitor Mode the observations are carried out by the ESO Paranal Science Operation staff PSO alone Observations can be done at any time during the period depending on the CS given by the user OBs are put into a queue schedule in OT which later send OBs to the instrument Template A template is a sequence of operations to be executed by the instrument The observation software of an instrument dispatches commands written in templates not only to instrument modules that control its motors and the detector but also to the telescopes and VLT I sub systems Template signature file TSF File which contains template input parameters Visitor Mode VM The classic observatio
28. ing information before submitting a proposal as well as the ESOFORM user manual The ESOFORM package can be downloaded from http www eso org observing proposals Considering a target which has a scientific interest the first thing to do is to determine whether this target can be observed with AMBER or not Please note that the limiting magnitudes for AMBER observations depend on the seeing and sky transparency and that appropriate weather conditions have to requested in the Phase 1 proposal The details of the current magnitude limits can be found at the AMBER instrument webpage http www eso org instruments amber inst 6 1 Proposal guidelines For general information about the VLT I facility please refer to the VLT I User Manual 6 1 1 Guaranteed time observation objects Check any scientific target against the list of guaranteed time observation GTO objects This guaranteed time period covers the full P96 Make sure the target has not been reserved already The list of GTO objects can be downloaded from http www eso org sci observing teles alloc gto html 6 1 2 Time critical combination of triplets For successful observations in either service or visitor mode it is very important that special schedul ing constraints such as the combination of different triplets within a certain time range or other time critical aspects are entered in Box 13 Scheduling Requirements The proposal should also be marked as time critical se
29. inges e MOD modulation in radians not meters files before April 2013 were wrongly labeled e FNTX1 resp 2 for FNT2 FNTXO photometric channels for 1 2 and 0 raw data in ADU e FNTX1A resp 2 for FNT2 FNTX1B resp 2 for FNT2 interferometric channels raw data ADU AMBER User Manual VLT MAN ESO 15830 3522 20 9 Bibliography 1 Observing with the VLT Interferometer Les Houches Eurowinter School Feb 3 8 2002 Ed 10 itors Guy Perrin and Fabien Malbet EAS publication Series vol 6 2003 EDP Sciences Paris The Very Large Telescope Interferometer Challenges for the Future Astrophysics and Space Science vol 286 editors Paulo J V Garcia Andreas Glindemann Thomas Henning Fabien Malbet November 2003 ISBN 1 4020 1518 6 Observing with the VLT Interferometer Wittkowski et al March 2005 The Messenger 119 p14 17 reference documents templates calibration plan maintenance manual science technical op eration plan especially VLT MAN ESO 15000 4552 the VLT I User Manual The VLTI real time reflective memory data streaming and recording system R Abuter D Popovic E Pozna J Sahlmann and F Eisenhauer In Society of Photo Optical Instrumen tation Engineers SPIE Conference Series volume 7013 July 2008 The VLTI fringe sensors FINITO and PRIMA FSU M Gai S Menardi S Cesare B Bauvir D Bonino L Corcione M Dimmler G Massone F Reynaud and A Wallander In W A Traub
30. ion Blocks OBs 1 1 Standard observation 0BS Std 2 6 5 se saeara ti ga a eee AN 7 2 Computing time overheads for added bands e 8 Calibration Plan St Data products e 4 44544 p eika es be Pew Se ea ee eee ae be ak G2 Dork Tames A E 6 dE ke e Oe A AR a EN we Oe a io Calbralsl BS A A bee oe RE a EEA we ee eR Ae ee HS 8 4 FINITO fringe tracking information SAT PECE a ma ak Hoh SO le od eae EE S42 Application 2445 2 2 ese Se be DR ea ee A a ae 8 5 AMBER FINITO RMNREC data description ST General os ros socs eee wb eae A a e Se VIPUL ORDES co SR A ee ee ok ee Rt ao bese BS ogee A Goo AI a Re ee hee eo a OS e 9 Bibliography 10 Glossary 11 Acronyms and Abbreviations vi 13 13 13 13 14 14 14 15 15 15 15 15 16 16 16 17 17 18 18 20 20 22 AMBER User Manual VLT MAN ESO 15830 3522 1 1 INTRODUCTION AMBER Astronomical Multi BEam combineR combines interferometrically the near IR light com ing from two or three telescopes of the VLT I It measures simultaneously a variety of interferometric quantities the fringe visibility differential with respect to wavelength visibility differential phase closure phase and differential closure phase These observables measure spatial details of a celestial source at a very high angular resolution the highest available from any ESO instruments AMBER can reach an angular resolution of the order of 1 milli arcsecond 1mas 0 001 and a spectral reso lutio
31. ions is not possible without moving the grism wheel every time This is why cold darks are taken in the morning It is recommended to use cold darks and sky for the actual data reduction Warm darks are currently kept for consistency with the previous observation procedure 8 3 Calibrator stars Calibrator stars are stars with known angular diameters yielding to the highest possible visibility knowing that e fringes SNR should be comparable between SCI and CAL e CAL should be as close as possible to SCI ideally lt 25deg and similar airmass e CAL should be observable one hour before AND one hour after the SCI target This is to ensure that it can be observed after or before the SCI if the later has been observed at the limit of its LST constraint In the case of bracketed observations i e CAL SCI CAL and impossibility to find a calibrator observable before and after a second calibrator should be used AMBER User Manual VLT MAN ESO 15830 3522 16 Considering that the choice of calibrator can be tailored to the actual specificities of the scientific goal the users are responsible for the choice of their calibrators and the creation of the subsequent OBs ESO offers the CalVin tool to chose the calibrator stars The observation of calibrator stars are used to measure the transfer function of the instrument namely e visibility transfer function Ves by Vos Ve easurea sCu Met e phase closure transfer
32. n mode The user is on site to supervise his her program execution AMBER User Manual VLT MAN ESO 15830 3522 22 11 Acronyms and Abbreviations AD AMBER AO AT CfP CP CS DI DIT DDL DL DRS ESO ETC FINITO FT IRIS LR LST MR MIDI MIR NDIT NIR OD OB OT OPC OPD OPL Phase I P2PP QC REF SM SNR STRAP TBC TBD TSF UT VIMA VINCI VISA VLT VLT I VM MACAO Applicable document Astronomical Multi BEam Recombiner Adaptive optics Auxiliary telescope 1 8m Call for proposals Closure Phase Constrain set Differential Interferometry Detector Integration Time Differential Delay line Delay line Data Reduction Software European Southern Observatory Exposure Time Calculator VLT I fringe tracker Fringe tracker InfraRed Image Stabiliser Low Resolution Local Sideral Time Multiple Application Curvature Adaptive Optics Medium Resolution MID infrared Interferometric instrument Mid InfraRed 5 20 microns Number of individual Detector Integration Near InfraRed 1 5 microns Observation Description Observation Block Observation Toolkit Observation Program Committee Optical path difference Optical path length Proposal Preparation and Submission Phase II Proposal Preparation Quality Control Reference documents Service Mode Signal to noise ratio System for Tip tilt Removal with Avalanche Photo diodes To be confirmed To be
33. n of R 35 in H and K band simultaneously R 1500 in H or K independently or R 12000 in K band 1 1 Scope This document summarizes the modes possibilities and limitations of AMBER as offered to the ESO community for P96 running from Oct 1st 2013 to March 31st 2014 Only the modes for P96 that are supported by ESO are discussed in this document Bold face font is used to emphasize any important issue regarding AMBER in P96 and they should be considered carefully by the reader This instrument manual should be used in conjunction with the P96 VLT I user manual avalaible from the manual webpages 1 2 AMBER news section At the start of this issue we would like to highlight the following items e For P96 there are no changes with respect to P94 no AMBER operations in P95 e For P94 the AMBER limiting magnitudes have again been increased by 0 5 for those modes that do not involve FINITO This increase is owing to an improved polarization control Please consult the limiting magnitude table for the current limiting magnitudes Please also read the limitations on polarization conrol in Sect 2 6 3 e Since P93 AMBER can be used in a container of cal sci cal sci cal which will take 100 minutes of total execution time It can be used in low spectral resolution and for seeing lt 1 2 and thin cloud coverage condition A waiver needs to be requested Regular rules regarding successful execution of containers with long execution times
34. nd related limiting magnitudes as of P94 see the table on the webpage Note however that the alignment procedure of the Niobate plates which is performed daily in order to overlap the two fringe systems cannot be performed perfectly and will lead to a reduction of 5 in fringe contrast The usage of the niobate plates is recommended for faint targets but not recommended for low contrast but bright objects In case of doubt or for more information contact the instrument scientist 2 6 4 Performances issues prior to P86 Note that previous to P85 AMBER showed spurious fringing in HR K which was difficult to calibrate and led to degraded performances in this mode A solution has been implemented to fix the problem the performance of this mode are now similar to performances in MR K 3 AMBER overview 3 1 AMBER principle Fig 2 summarizes the key elements of AMBER s conceptual design A set of collimated and parallel beams are focused by a common optical element in a common Airy pattern which contains the AMBER User Manual VLT MAN ESO 15830 3522 8 Figure 2 Basic concept of AMBER 1 multi axial beam combiner 2 cylindrical optics 3 anamorphosed focal image with fringes 4 long slit spectrograph 5 dispersed fringes on 2D detector 6 spatial filter with single mode optical fibers 7 photometric beams fringes 1 in Fig 2 The spacing between the beams is selected for the Fourier transform of the fringe
35. nts an Observation Block OB is a logical unit specifying the telescope instrument and detector parameters and actions needed to obtain a single observation It is the smallest schedulable entity which means that the execution of an OB is normally not interrupted as soon as the target has been acquired An OB is executed only once when identical observation sequences are required e g repeated observations using the same instrument setting but different targets or at different times a series of OBs must be constructed Because an OB can contain only one target science and associated calibration stars cf Sect 8 should be provided as two different OBs Thus each science object OB should be accompanied by a calibrator OB These OBs should be identical in instrument setup having only different target coordinates Moreover with single telescope instruments any OB can be performed during the night In the case of interferometric instrument the instant of observation define the location of the observation in the u v plan 7 1 Standard observation OBS_Std A standard observation with AMBER in P96can be split in the several sub tasks see Fig 6 1 Configuration Setup of the desired spectral resolution wavelength range and DIT 2 Internal calibration of the chosen instrument configuration P2VM see sec 3 2 4 3 Acquisition Slew telescopes to target position on sky and slew the delay lines to the expected zero OPD position 4
36. olved Also it is important to note that the latest version of the amdlib pipeline Inttp www eso org observing etc AMBER User Manual VLT MAN ESO 15830 3522 17 3 0 does not include the post processing This correction is therefore left to the observer Further testing will be carried in the future to better constrain the observational specifications requested to obtain good results 8 5 AMBER FINITO RMNREC data description 8 5 1 General When the FINITO fringe tracker 6 is used with AMBER 7 real time data are recorded along the raw AMBER frames These additional data can be used to refine the data reduction of AMBER 8 9 They are generated by continuously recording the content of the Reflective Memory Network RMN 5 and can be found as binary extension in the AMBER FITS files FNT1 FN T2 OPDC1 and OPDC2 FNT extensions refer to the raw FINITO data whereas the OPDC tables contain data regarding the active control of the optical path delay as the name suggests OPDC stands for Optical Path Delay Controller In main header The main header contains a lot of information and allows in particular to reconstruct the configu ration of the VLTI af the time of the observations e telescope s stations configuration is retrieved HIERARCH ESO ISS CONF STATIONi for i in 1 2 3 e AMBER configuration its three beams corresponds to VLTI input channels HIERARCH ESO ISS CONF INPUTi for i in 1 2 3 e HIE
37. ormation in Robbe Dubois et al 2007 A amp A 2007 464 13 and Petrov et al 2007 A amp A 464 1 3 2 1 Warm optics The three spatial filter inputs one for each spectral band are separated by dichroic plates For example the K band spatial filter OPM SFK is fed by dichroic which reflect wavelengths higher than 2 ym and transmit the H and J bands After the fiber outputs a symmetric cascade of dichroics combines the different bands again but the output pupil in each band has a shape proportional to the central wavelength of the band Therefore the Airy disk and the fringes have the same size for all central wavelength This allows the same spectrograph achromatic optics to be used for all bands and the same sampling of all the central AMBER User Manual VLT MAN ESO 15830 3522 9 Figure 4 Photography of AMBER at Paranal AMBER User Manual VLT MAN ESO 15830 3522 10 wavelengths to be operated Then the beams enter the cylindrical optics anamorphoser OPM ANS before entering the spec trograph SPG through a periscope used to align the beam produced by the warm optics and the spectrograph 3 2 2 Spectrograph The spectrograph has an image plane cold stop a wheel with cold pupil masks for 2 or 3 telescopes The separation between the interferometric and photometric beams is performed in a pupil plane inside the spectrograph after the image plane cold stop 3 2 3 Detector After dispersion the spectrograph chamber sends t
38. pattern to show separated fringe peaks non homothetic mapping The Airy disk needs to be sampled by many pixels in the baseline direction an average of 4 pixels in the narrowest fringe i e at least 12 pixels in the baseline direction while in the other direction only one pixel is sufficient Each spectral channel is thus concentrated in a single column of pixels 3 in Fig 2 by cylindrical optics 2 in Fig 2 The fringes are dispersed by a standard long slit spectrograph 4 in Fig 2 on a two dimensional detector 5 in Fig 2 The spectrograph must be cooled down to about 60 C with a cold slit in the image plane and a cold pupil stop In practice it is simply cooled down to liquid nitrogen temperature High accuracy measurements require spatially filtered optical beams The single way to achieve such filtering with decent light transmission is to use single mode optical fibers 6 in Fig 2 The flux transmitted by each filter must be monitored in real time in each spectral channel This explains why a fraction of each beam is extracted and sent directly to the detector 7 in Fig 2 Before entering the fibers the beams should be cleaned from the differential atmospheric refraction in the H and J bands or in some cases from one polarization 3 2 AMBER layout Fig 3 shows the global implementation of AMBER with the additional features needed by the actual operation of the instrument The user can find more detailed inf
39. re a waiver FINITO fringe tracking information will be recorded with AMBER data see Sect 8 As of P91 AMBER self coherencing is operational in visitor and service mode AMBER has the capability to send a correction to the delay line in order to better maintain the fringes within coherence length by measuring the optical path difference from its fringes This will result in an increased and more stable instrumental contrast and a less dispersed closure phase See the AMBER instrument webpage http www eso org instruments amber inst for the most recent information on the exact wavelength ranges and section 6 2 1 for the configuration options for the spectrograph AMBER User Manual VLT MAN ESO 15830 3522 12 5 1 Service and Visitor Modes For P96 AMBER is offered in service mode and in visitor mode see Sect 10 for the definition of these modes During an observing period the unique contact point at ESO for the user will be the User Support Department email usd help eso org and homepage http www eso org sci observing phase2 USD MIDI html The visitor mode is more likely to be offered for proposals requiring non standard observation proce dures The OPC will decide whether a proposal should be observed in SM or VM As for any other instrument ESO reserves the right to transfer visitor programs to service and vice versa 6 Preparing the observations Proposals should be submitted through the ESOFORM Carefully read the follow
40. re frames should add 15 min per OB A maximum of 2 additional bands per observation i e per OB is allowed 8 Calibration Plan 8 1 Data products The observatory shall provide the following calibrations to science SCI or calibrator stars CAL data 1 daily darks obtained with the same DITs as the data Two different types of darks are provided see Sect 8 2 2 daily sky obtained with the same DITs as data taken right after the on target data 3 daily Pixel to Visibility matrix P2VM for all observations All pairs SCI CAL or triplet CAL SCI CAL should be taken with the same P2VM taken prior to the sequence The validity of the P2VM is 6 hours 4 at period change or any instrument intervention bad pixel and flat field maps 8 2 Dark frames We provide two different types of dark frames cold and warm darks Cold darks are taken by closing the spectrograph with a cold metallic patch so the detector sees an element at the temperature of the cryostat Conversely warm darks are taken by closing shutters outside of the cryostat hence there will be a residual of thermal emission especially at longest wavelengths Warm darks are taken right before the observations whereas cold darks are taken the following morning The reason why the cold darks cannot be taken simultaneously is because the cold patch is on the same wheel as the spectrograph slit Hence taking cold darks for every observat
41. tral channel and the visibility in a reference spectral channel average of several other channels for example e the differential phase i e the difference between the phase in each spectral channel and the phase in a reference channel e the closure phase is the phase of the bispectrum computed in each spectral channel The bispectrum is the complex product of three visibilities along a closed triangle The closure phase is therefore theoretically equal to the sum of the three phases along the three baselines This quantity is to a great extent independent from atmospheric perturbations 2 4 1 Absolute visibility V f A One visibility measurement for a single baseline can constrain the equivalent size of the source for an assumed morphology Visibility measurements for several spatial frequencies obtained through Earth rotation different wavelengths different telescopes combinations constrain severely the models The visibility should be carefully calibrated see Sect 8 2 4 2 Differential visibility V f V f Ao In some cases one is interested in variations of size of a target as function of the wavelength This is the case when observing a structure which is present in a spectral line whereas the continuum corresponds to an unresolved structure One can then calibrate the measurement in the line by those AMBER User Manual VLT MAN ESO 15830 3522 5 in the continuum and the knowledge of the absolute visibility is not requir
42. with AMBER The best suited ESO instruments that can give access to these data are NAOS CONICA and SINFONI which measure diffraction limited images AMBER User Manual VLT MAN ESO 15830 3522 3 2 156 um HR 2005 02 26 2 183 um 2 135 um MR 2004 12 26 Figure 1 Image of the fringes recorded by AMBER in high top and medium bottom spectral res olution around the Bry emission line of Eta Car The P1 2 3 channels are the photometric channels IF stands for interferometric channel Weigelt et al 2007 A amp A 464 87 in the same wavelength domain as AMBER With NAOS CONICA it is possible to do both imaging and spectroscopy and SINFONT is unique in that it does full field spectroscopy in a 3 by 3 field Further information on these instruments can be found at https www eso org sci facilities paranal instruments naco and http www eso org sci facilities paranal instruments sinfoni The VLT I offers a second interferometric instrument MIDI This instrument operates in the N band 10 ym and combines and disperses the mid infrared light coming from two telescopes again either ATs or UTs Both AMBER and MIDI use the same VLT I infrastructure and many aspects regard ing observation preparation and scheduling are similar More information on MIDI can be found at the following web address http www eso org sci facilities paranal instruments midi 2 3 Optical interferometry basics The contrast and phase of monochromati
Download Pdf Manuals
Related Search