User Guide to 1.5-m Fiber Echelle Spectrograph
Contents
1. Guider CCD From GAM Figure 4 Schematic representation of the mechanical design of the front end module The comparison source module light box is mounted at the telescope jointly with the electronics box which powers the light sources thorium argon lamp and quartz lamp The electronic box connects to FEM with a short cable and to the control room with the ethernet cable and additionally by a cable that goes through the J27 round connector on the telescope and is used for manual control of the lamps The guiding CCD camera is powered from the electronics box through FEM a short cable 12V while its digital signal is sent to the guiding PC in the computer room by a dedicated Gigabit Ethernet cable 1 3 What is inside FEM The main components of FEM are shown in Fig 4 The incoming beam is focused on the concave mirror M2 with two holes in it one for the star and another larger for the sky not used The light which goes through the central hole is transformed from F 7 5 to F 5 by a pair of small lenses and is injected into the fiber with a core diameter 100 micron The co alignment between the hole and the fiber is very important To check it easily an auxiliary optics a small mirror and a strong eyepiece can be lowered in front of M2 Alignment check in Fig 2 The back end of the fiber is illuminated put paper on the slit turn on the light in the coude room and the image of its front end can be seen through t2. the intensity between red and blue wavelengths This lamp serves for flat fielding the spectra The electronics box contains a simple pulse generator for controlling the prism servo motor in the FEM relays for switching the spectral lamps logic and the power supplies 12V and 5V The DC DC converter for the spectral lamp is powered by 12 V The power switch on this box activates all the electronics Please switch off the electronics at the end of the night The control of the lamps motor and LED done by the data taking software In addition there is a small box in the control room for manual operation It has a switch with 3 positions In the central neutral position the prism is retracted starlight enters the spectrograph In the left position the prism is In the Th Ar lamp is on and the red LED is illuminated In the right position the Quartz lamp is on with yellow LED illuminated 1 5 Acquisition and guiding The guiding camera is GC650 from Prosilica It has 659x493 square pixels pixel size 7 4 micron 0 38 arcsec on the sky field of view 3 arcmin diameter The signal depth is 14 bits 4096 counts Minimum exposure time is 0 000001s 10 microseconds The camera is powered by 12 V from FEM Its digital signal is transmitted by a dedicated Ethernet cable to the guiding PC which is rack mounted in the computer room This PC runs under Linux its name is ctioxb fixed IP adress 139 229 12 62 The standard PCguider program r
3. 68 set to center the blaze function on the detector Fig 9 shows that this centering is good The collimator focus is 99 80 The cross disperser angle is set to 1114 These settings must never be changed 7000 J 6500 E 4 6000 F J 5500 J Wavelength A 5000 J 4500 F 4 fi L 0 500 1000 1500 2000 Y pixels Figure 8 Correspondence between wavelength and line number Y for the current spectrograph setting full frame readout The correspondence between the vertical coordinate on the CCD and the wavelength in is plotted in Fig 8 It is well approximated by second order polynomials A 4023 47 1 6599Y 1 9810 Y 1 Y 3 06 0 602L 3 410 L L 4000 2 The full coverage is from 4020 to 7300 The iodine setup Y from 400 to 1600 corresponds to the wavelengths from 4670A to 6650A The fiber image is projected onto the slit with 1 5x magnification The actual slit width is about 40 um larger than the micrometer reading as deduced from the transmission measurements Fig 10 For high resolution work we recommend the micrometer setting of 50 um actual slit width 90 um which gives R 43000 at the expense of 70 transmission For faint stars open the micrometer to 100 wm to transmit the full fiber image The Schmidt camera of the spectrograph has no field flattener so the focal surface has a curvature radius of about 590mm with a convex side pointing to the camera It is not p
4. Cerro Tololo Inter American Observatory User Guide to 1 5 m Fiber Echelle Spectrograph Telescope FEN ER Bem ae ph COD A Tokovinin Version 3 April 29 2010 ech_manual pdf 1 Instrument description The echelle spectrograph was used at the Blanco telescope and now after having been de commissioned it is located in the coud room of the 1 5 m telescope and connected to it by optical fiber The Front End Module FEM matches the fiber to the telescope and provides for auxiliary functions such as guiding and comparison spectra The Back End Module BEM couples the fiber with the spectro graph and permits inserting the iodine cell in the beam 1 1 Interface to the GAM Figure 1 M1 mirror attached to the GAM probe The Guiding and Acquisition Module GAM is normally used for offset guiding It has been modified by adding a piggy back mirror M1 on top of the guide probe Fig 1 Positioning this probe at the center of the field X 268 Y 259 makes M1 to intercept the on axis beam without vignetting and to direct it sideways to the FEM which is attached to the GAM hatch door The optical axis of the beam deflected by M1 is at 78mm below the lower surface of the GAM top plate Normal operation of the probe is not perturbed it captures the field at a fixed offset and if we are lucky to find guide stars there the standard guiding is possible The object would be centered in the GAM camera field if we set Y 96 but then FEM re
5. ceives no light 1 2 FEM and its connections Figure 2 shows the FEM attached to the GAM The space is restricted from all sides To remove the FEM loosen the 4 M4 screws turn the FEM anti clockwise and pull it away leaving the screws in place Be careful not to damage the cables and the optical fiber Installation in the reverse order The FEM is permanently installed on the telescope but any repair or trouble shooting require its removal because accessing it is difficult ra Figure 2 FEM attached to the GAM door views from two sides Light Electronics To control room Ethernet Electronics Switch Main fiber cable Comp light fiber Gigabit Ethernet BEM Coude room PC 139 229 12 62 ccp Guider Temperature controler PC 139 229 12 29 Monitor VNC connection Computer room Fiber Control room Figure 3 Connections of the 1 5 m echelle system The main fiber cable can be detached from the FEM Its exact position is critical it is defined by two pins Be careful not to damage them The optical alignment between the fiber and the hole must be checked each after time the cable was detached Do not detach the fiber cable unless it is really necessary The connections of FEM are shown in Fig 3 The orange fiber cable brings the light of the comparison sources it is permanently fixed inside the FEM to detach it the FEM must be opened Comparison light Bench
6. ece of white paper on the lower BEM lens and switching on the light in the coude room Detach FEM from the telescope without disconnecting On the FEM release the fixing screw on the alignment check optics Fig 2 right move the lever down to un cover the hole and see the round image of the fiber end through the hole In the case the image is not round obstructed by the hole on one side make a very slight correction with the set screws in the fiber holder Be careful not to turn these screws by more than 1 4 of the revolution and always leave them tightened Spectrograph focus Take two Th Ar exposures with the Hartmann mask in the West and East positions Export the images to FITS and run focus pro in IDL to obtain the plot like Fig 11 If the spectrograph is properly focused the FWHM of unsaturated Th Ar lines near the center of the detector should be 1 95 pixels slit setting 50 wm no masks 12
7. fficiency the number of detected photons relative to the number of incident photons outside atmosphere is estimated at 1 with a 60 micron slit transmitting 60 of light Fig 13 A V 5 star gives flux of 116 el pixel s The width of the orders in the cross dispersion direction is about 3 8 pixels so the flux per pixel in the 2 dimensional frames is 3 8 times less than the integrated flux A S N 100 will be reached on a V 7 star in a 10 min exposure Some stars show a reduced efficiency presumably due to clouds poor guiding or bad seeing Efficiency of 1 7 is expected with the fully opened 150 micron slit giving spectral resolution R 25000 For comparison other echelle spectrographs have efficiency 7 Hamilton HARPS 12 HIRES or even 15 FEROS 3 Check list The checks listed below should be made each time the instrument is installed at the telescope or modified Some of these checks may be useful for trouble shooting e Verify all connections Fig 3 11 Electronics functionality Move the prism in and out and hear the motor moving through the microphone placed near the FEM Switch on quartz or Th Ar lamps see the light coming out of the fiber connector on the light box CCD camera functionality The image of the field with two holes big and small should be seen on the PCguider screen with the dome illuminated Make sure this image is sharp Fiber hole co alignment Back illuminate the fiber by placing a pi
8. he hole and centered by tiny motions in the fiber holder Do not forget to retract the auxiliary optics after the alignment check The light which does not pass through the hole is reflected by M2 then directed by a flat mirror M3 to the optics of the guider The 3x reduced image of the field i e of M2 is formed on the CCD camera GC650 from Prosilica The digidat signal is sent via Ethernet cable to the guiding PC see below The light of the comparison lamps is directed into the main fiber by moving a small 2 mm prism which slides behind M2 The prism is pushed by a cam mechanism which in turn is actuated by a small servo motor 1 4 Light box and electronics M2 Quartz lamp Filter a Fiber end E a t Th Ar lamp mi Figure 5 Optical scheme and picture of the Light Box The light box contains the thorium argon spectral lamp and the quartz lamp The SMA connector of the fiber core diameter 0 4 mm receives the image of the thorium cathode formed by a F 18 mm lens it is aligned with the fiber The lamp is powered by 500 V voltage produced by the DC DC converter in the light box while a 50 kOhm 12W resistor in series limits its current to 8ma The life time of this lamp is short 500 hours so it should be switched on only during comparison exposures The quartz lamp 5V 10W is also co aligned with the fiber so its light is very bright The 4 mm thick BG38 color balance filter is installed to equalize
9. on o D T 0 2 F 1 position is 1 00mm so that the CCD is slightly by 0 1mm beyond the best focus at the center of the field positive defocus in Fig 11 to offset partially the field curvature Figure 11 Contour plot of the spectrograph fo cus after final adjustment May 23 2008 The G pa J 7 CCD lines from 400 to 1599 are read out iodine i Og et 4 J i setup so the Y center is at line 1000 The defo _A cusat the center is 0 093 mm CCD behind the 500 500 1000 focus the curvature radius of the focal surface is 0 5m concave to the detector i Ethernet systran fiber ADAM Temperature 12 cell control Ethernet Coude room Computer room Telescope Prism and comparison lamps Figure 12 Elements of the new controller system and their connections 2 3 Iodine cell The iodine cell is a glass cylinder 50 mm diameter 100 mm length wrapped in a heater tape The temperature controller maintains the cell at 55 C there is a sensor below the heater tape When switched on the controller can over shoot and heats the cell well above the desired temperature This must be avoided by watching the temperature If it is over shooting push the Stop button on the controller watch the temperature to go below the set point then push Run again The cell is housed in a small black container which can be inserted manually in the pa
10. ossible to adjust the detector tilt and focus for the whole field simultaneously These parameters are controlled by taking thorium exposures with two masks west and east and finding the relative X displacement between the matching lines in both images automatically with focus pro in IDL The detector tilt was adjusted so that the center of the curved focal surface is near the detector center Figure 11 shows the iso focal lines with the inner contour corresponding to the perfect focus The dewar focus NOAO IRAF V2 ae EXPORT Thu 14 39 52 24 Apr 2008 NOAO IRAF V2 ar Thu 14 42 22 24 Apr 2008 obj025 ccl 28 WR 1546 K3 300 ap 28 beam 28 obj eel 281 TIR 4846 K3 300 ap 28 beam 28 a 30000 4 30000 ff f 20000 zoo00 15000 15000 l 10000 M 5000 A Ma 4 100007 7 0 500 1000 1500 2000 1100 1200 1300 1400 Figure 9 Plots of the extracted spectra of HR 4546 around Ha left the whole order right zoom on Ha The spectrum has not been flat fielded the difference in the response between the two amplifiers left and right halves is apparent Figure 10 Slit transmission versus oak micrometer reading The points are measurements with Quartz lamp the curve is the analytic model t 0 0L 3 1 2 m acos8 BV1 87 where 0 50 100 150 8 is the ratio of slit width to fiber Slit micrometer reading gt diameter Transmissi
11. rallel beam of the BEM being mounted on the hinge This motion will be motorized A new container heater combination has been fabricated in January 2010 but it is not in use now In the near future the in out cell motion will be remotely controlled 2 4 CCD and data acquisition The CCD is a 2048x2048 SITe chip with 24 micron pixels It is housed in the CTIO SITe_6 dewar and works with the Monsoon Orange controller it replaced the old Arcon controller in January 2010 The new controller consists of several inter connected modules Fig 12 We use the readout with two upper amplifiers UL UR The gain and full chip readout time can be optimized by selecting fast mode for bright stars or normal mode for faint stars Table 1 When only 1 2 of the chip is read out iodine cell observations the fast readout takes about 10s The CCD is not yet saturated at the maximum count 65000 ADU to be checked 10 Table 1 Readout modes and parameters Parameter Fast mode Normal mode UL UR UL UR Readout time s 18 27 Gain e ADU 3 3 3 0 0 91 0 81 RON ADU 2 8 3 3 8 0 8 2 RON electrons 9 2 9 9 7 3 6 6 1000 0 T T 100 0 Pt 3 10 05 x Ve 4 F x x ly 1 0 F E xX 0 1 L L L L L 4 5 6 7 8 9 10 V magnitude Figure 13 Detected flux near 550 nm electrons per pixel per second summed across order vs star magnitude The dotted line shows 1 total efficiency 2 5 System efficiency The overall e
12. the spectrograph Left without the iodine cell right with the cell in the out position The Back End Module BEM is mounted on the front plate of the Blanco Echelle Spectrograph by means of the AR90 angular bracket attached to the plate with two 1 4 screws Fig 7 The bracket holds the rail with two carriages The upper carriage holds the 50 mm focus lens and the fiber cable termination The cable must be inserted and fixed by tightening the M3 set screw This assembly is pre aligned to give a parallel and well centered beam The diameter of the emerging collimated beam is about 10mm The lower carriage holds the 75 mm focus lens which re focuses the fiber end image on the slit The magnification is 1 5x so the image diameter is 150 um The slit is 42 mm below the surface of the front plate its orientation is indicated in Fig 7 The X and Y knobs permit to move the fiber image on the slit in respectively perpendicular and parallel directions The image is focused by moving the whole lower carriage along the rail With the fiber illuminated sky through the telescope or quartz lamp the image of the fiber is focused and centered on the slit by looking in the post slit viewer Initially the slit is wide then it is set to the desired width between 60 um and 150 um with the smallest decker 2 2 Spectrograph setting and focus The echelle grating is 31 6 1 mm the cross disperser is number 3 226 1 mm The angle of the echelle is 5
13. uns on this computer Fig 6 After turning on the electronic box hence the camera enter the guiding program by VNC connec tion from a suitable terminal vncviewer 139 229 12 62 1 In the VNC screen open the PCguider from a menu activated by the left mouse button In order to see better star images use automatic VNC ctioxb 1 pcguider IE J CTIO PC GUIDER 5 0 2 ctiob0 conf Options Mode View Window x E Max Value 378 Leaky av 0 E Guide Box Size 25 R Y Beas Background 47 ea x Ref 50 Total Count 434077 a MIA a Y Ref 50 FwHm a83 Increment 1 B Saus A CIS A ROS coer toss to Stdev Em sles REFRESH DISP mm Freer oa Gan i Ofrset o MI overscan Erase CT sow cm Next img is Mome pcguider Data a0 fits Figure 6 Snapshot of the PCguider screen during guiding intensity scaling this is a new program feature To do so use menu in PCguider Options Parameters olut off the default is sigma When the field is illuminated by sky or dome light dark images of the two holes are seen The guiding box should be centered on the small hole normally at X 289 Y 255 You may want to shift the box by 1 pixel to achieve better centering of the star in the hole The normal box size is 29 pixels The North South direction coincides with Y After pointing the telescope the object should be seen on the screen Adjust camera integration time to avoid sat
14. uration max counts lt 4096 In the PCguider menu use Windows Camera Control Integration Time set the new value press Enter The effect is immediate In Fig 6 the integration time is 100 ms Using hand paddle move the telescope to bring the star into the box Most of the image disappears in the hole and we see a ring donut as shown in Fig 6 The guiding loop can now be closed To view the image better adjust the display using its control panel display menu Options Control panel It is helpful to adjust the contrast and brightness and to use zoom for a larger view of star image in the hole For focusing the telescope move the star away from the hole make sure that the image is not satu rated in the camera max count lt 4096 and in the display With bad settings of brightness contrast the central part of the image looks flat with right settings it is peaked The position of the guider arm must be checked because the GAM control software systematically goes wrong so the displayed coordinates of the guide arm do not match its actual position The correct position should be X 268 Y 259 Z 41 Put a bright star away from the hole defocus the telescope and verify that the image is ring like not cut on one side If this is not the case re initialize the GAM control program and set the GAM probe again where it should be 2 The spectrograph 2 1 Back end module Figure 7 Back End Module at
Download Pdf Manuals
Related Search