Home

Bayfordbury Single-object Integral Field Spectrograph (BASIS)

1. 55 15 Appendix E Original Initial Plan 65 Bayfordbury Single object Integral Field Spectrograph BASIS 5 1 INITIAL PLAN 1 1 A brief overview of the project The Bayfordbury Observatory University of Hertfordshire UK is acclaimed to be the best teaching observatory in the UK with six optical telescopes four 16 inch and two 14 inch a 4 5m radio dish and a 115m baseline radio interferometer Due to the number of telescopes available it has been possible for the optical telescopes to each host a different instrument Current instruments available are high spec CCDs a fast frame rate camera for planetary or lunar imaging a slit spectrograph SBIG SGS an Ha filter for solar observations and more Recent developments have now enabled a robotic drive for one of the 16 inch telescopes once again pushing Bayfordbury Observatory further into the lead as the UK s best teaching observatory To keep in line with Bayfordbury Observatory s status I will build and commission an Integral Field Unit IFU for one of the 16 inch telescopes It appears that not only will this be the UK s first teaching IFU but also the first IFU hosted by such a class of telescope lt 0 5m The benefit of using an IFU is that it enables the observer to obtain spatially resolved spectra of a cer
2. Bayfordbury Single object Integral Field Spectrograph BASIS 59 14 3 Third Night Date 10 02 2012 Start Time 19 00 End Time 23 00 Site Bayfordbury Observatory Overall Weather Clear Moon Below Horizon 23 00 Sky Brightness 18 78mag arcsec Some galaxies to observe if able NGC 2748 z 0 005 NGC 3079 z 0 0038 NGC 3877 z 0 00016 NGC 4026 z 0 00019 NGC 4125 Stars to do the 2 star alignment with Dubhe Alpheratz Balanced telescope with weights on adjustable bar mount in park position performed two star alignment on Capellar and Betelgeuse alignment failed maybe too close in RA will perform two star alignment on Dubhe and Alpheratz alignment worked alignment check on Betelgeuse showed some offset auto guiding on star near Betelgeuse Moved to point at Jupiter manually centred in TV focus on fibre polish paper with focuser all the way in _ A EN Green paper was used to focus on Moved mirror Jupiter now not in centre of telescope so will have to centre it and re align guide scope then re align system Focuser will be offset by IFU inset to counter adjust focal plane pointed Jupiter in centre of telescope using paper as guide and callipers as measurement half of focuser outer diameter all the way around re aligned guide scope with sticker on screen showing crosshair CCD centre 60 Richards these are the hand control hold mode coordinates RA 02 08 36 DEC 11 54 34 Jupiter is o
3. End due to cloud 23 50 Report ACP could not connect to the telescope so I did manual pointings Went to Jupiter and found the IFU this time in focus Took a 60s exposure please find attached This is completely raw i e no dark subtraction no cooling Literally just took an exposure to see what would happen At this point Jupiter was near the horizon so I did not have time to set everything up for an ideal exposure The attached exposure shows Jupiter falling on the inner 7 cores f o v 25 It took some time to centre it by no means perfect where the only reference was how bright each core spectra was Jupiter then set so I went to the Moon as clouds came in Took a 60s spectrum of the Moon to get all cores illuminated note clouds made the Moon faint hence 60s exposure All in all I m glad to get some spectra from BASIS To do on next commissioning night 1 Properly calibrate ACP 2 Find a solution to the inconsistent pointings and inconsistent offset to the IFU due to flexure in the TV Cannot do any deep sky observations until this is sorted End of Commissioning Reports Bayfordbury Single object Integral Field Spectrograph BASIS 65 15 APPENDIX E ORIGINAL INITIAL PLAN 1 INITIAL PLAN 1 1 A brief overview of the project The Bayfordbury Observatory University of Hertfordshire UK is acclaimed to be the best teaching observatory in the UK with six optical telescopes four 16 inch and two 14
4. 150 I mm diffraction grating with adjustable tilt angle to E L i _ N allow a user defined central y Colimator wavelength 600 lmm diffraction grating The grating can be exchanged by rotating the grating unit by 180 om car ol ST E Figure 3 11 5 Image of the SBIG SGS as viewed from above with the 1 D fibre array connected See labels for descriptions The dotted green line shows the optical path up to the yellow line which is the CCD and the white lines represent the mirrors obscured by other components in this image 34 Richards 4 DATA REDUCTION Before any observations were taken some thought had to be spent on finding a way to reduce the data that would be collect as there is little point in having data you cannot be utilised The following section discusses the various stages of data reduction 41 First order data reduction The obvious first step is to manually look at the data to find any interesting features Because the spectra have been aligned on the CCD it is possible to take a cut in the horizontal direction getting pixel values for each wavelength to extract a fibre s spectrum Very simply this can be done in two ways 1 MaxImDL has a very useful graph tool see Figure 5 5 5 which allows the user to place a horizontal line on an image and the graph window will display the pixel values along that line This graph is essentially the spectrum from that fibre and it can be exported as
5. aed siu luooO X 1 1deuo dures j Suruoirssituuro m Ml MM Y 3urpm3 omny Ml jJ 007X T OFUO uong e sul 398J19JU1 SOS OLAS OVS _ AE jn rT j j juorsnuqnsj ensqy aids _ 1 1 aussi j j j yxmmr miieooj OOTXT E fear ONO Aeue q Em Ae11e JO Sur pour xewoz MN ABLIL SUd OJ01JA 10 SuNLOD YY Ha atemos ped 15 jInpuods 19qqny faq pp wizz up d en1u Joded Surystjog JOAR9 9 AQI s pe q 10zey ABLIV su O1 IJA c fourusew ONO 952 J191ut SDS DIAS 0 Keue dq jJ JUMP ONO JAU ui dnoo Ave SUd OJDIHA 0 Aql 198 Suruo1gui x putr pue n 3 eondo uonov 54 Richards Internal External 002 005 006 XXX Loose Tube 50 125 24 core Optical Fibre Polishing 30um 00A 001 002 18 12um 00A 001 002 16 09m 00A 001 002 15 03m 00A 001 002 13 Olum 00A 001 002 12 DCS MJ018A Object3D Fibre Lapping Film 30um 12um 9um 3um 1 um Paper UHU Plus Endfast 300kg Epoxy Adhesive 33g Fibre Pen Cleaver Cleaver FullCure VeroBlack 3D Printing http www cablemonk ey co uk acatalog Loos e Tube OM2 fibre ca ble html http www cablemonk ey co uk acatalog Fibre Lapping Film html http www sylmasta co m acatalog UHU_Adhe sives html http www tools4it co m fibre optic tools pe
6. The OM2 fibre that was purchased see Appendix C came as a loose bundle of 24 fibres protected in furcation tubing and then again in a rubber tube This was ideal for use in BASIS as the total number of fibres used is 23 19 in the bundle 4 sky fibres and so coming pre protected saved time effort and money The redundant fibre was cut short to be out of the way but there is no reason why it could not be used as a channel for something It could have been an additional sky fibre but the decision was made not to because it would break the symmetrical pattern of spectra on the CCD The attenuation of the OM2 fibre used is shown in Figure 3 6 2 Note that the scale is in dB km The unit dB decibel is a logarithmic value defined in Eq 3 6 Therefore an attenuation of 5dB results in a power efficiency of 10 or 32 When speaking of optical fibres this value tells you how much light is being lost down the fibre The convention of using units of dB km for optical fibre is because the units are very small per metre Looking at Figure 3 6 2 even in the short wavelengths where the attenuation is the greatest 30dB km for a 5m length of fibre the attenuation is only 0 25dB which translates to 94 efficiency Science grade fibre can produce a better efficiency but only by a couple of percent at most when dealing with a 5m length That couple of percent is negligible when dealing with the combined efficiencies of all the other components in BASIS so
7. 1 7 442 Pixels Res Element Z_658nm Wm 2p 1 7 64 Spatial pixels Z_806nm Wm 2y 1 7 91 Fluxes F r Fluido Fie Funba Stellar Magnitude 445nm 551nm 658nm 806nm Novo s 445nm None s S51nm Nas 658nm Nte 5 806nm 10 0 6 61E 13 3 61E 13 2 29E 13 1 23E 13 1 46E 01 1 00E 01 7 38 02 4 89E 02 10 5 4 17E 13 2 28E 13 1 45E 13 7 76E 14 9 23E 02 6 31E 02 4 66E 02 3 09E 02 11 0 2 63E 13 1 44E 13 9 12E 14 4 90E 14 5 83E 02 3 98E 02 2 94E 02 1 95E 02 11 5 1 66E 13 9 08E 14 5 75E 14 3 09E 14 3 68E 02 2 51E 02 1 85E 02 1 23E 02 12 0 1 05E 13 5 73E 14 3 63E 14 1 95E 14 2 3202 1 59E 02 1 17E 02 7 75E 03 12 5 6 61E 14 3 61E 14 2 29E 14 1 23E 14 1 46E 02 1 00E 02 7 38E 03 4 89E 03 13 0 4 17E 14 2 28E 14 1 45E 14 7 76E 15 9 23E 03 6 31E 03 4 66E 03 3 09E 03 13 5 2 63E 14 1 44E 14 9 12E 15 4 90E 15 5 83E 03 3 98E 03 2 94E 03 1 95E 03 14 0 1 66E 14 9 08E 15 5 75E 15 3 09E 15 3 68E 03 2 51E 03 1 85E 03 1 23E 03 14 5 1 05E 14 5 73E 15 3 63E 15 1 95E 15 2 32E 03 1 59E 03 1 17E 03 7 75E 04 15 0 6 61E 15 3 61E 15 2 29E 15 1 23E 15 1 46E 03 1 00E 03 7 38E 04 4 89E 04 15 5 4 17E 15 2 28E 15 1 45E 15 7 76E 16 9 23E 04 6 31E 04 4 66E 04 3 09E 04 16 0 2 63E 15 1 44E 15 9 12E 16 4 90E 16 5 83E 04 3 98E 04 2 94E 04 1 95E 04 16 5 1 66E 15 9 0RE 16 5 75E 16 3 09E 16 3 68E 04 2 51E 04 1 85E 04 1 23E 04 17 0 1 05E 15 5 73E 16 3 63E 16 1 95E 16 2 32E 04 1 59E 04 1 17E 04 7 75E 05 17 5 6 61E 16 3 61E 16 2 29E 16 1 23E 16 1 46E 04 1 00E 04 7 38E 05 4 89E 05 18 0 4 17E 16 2 28E 16 1 45E
8. 14 5 Fifth Night Date 03 03 2012 Start Time 19 00 End Time 00 00 Site Bayfordbury Observatory Overall Weather Clear Moon High gibbous moon 19 00 22 00 00 00 Sky Brightness 18 09mag arcsec Sky Brightness 18 35mag arcsec Sky Brightness 17 89mag arcsec Aligned crosshair scope 1536x1024 768x512 centre on MaximDL 2bin 1024x682 512x341 centre on CCDOPS full low TV aligned Reset telescope and run 2 star align using Dubhe and Alpheratz Align successful check with Alhena pointing still way off hoping ACP will correct this telescope connection failed so will need David Campbell onsite to fix find offset with Jupiter using manual alignment IFU inset of 1 29cm IFU length of 2 27cm focus offset is 0 98 this will be used for the 2 inch focus mount from now on focus offset applied hopefully get spectra from Jupiter via MaximDL Manually centred Jupiter using the spectra as a spatial guide Took 60s exposure 64 Richards Really good result and knowing that Jupiter is 25 in diameter then only illuminating the inner cores means the focus is good Jupiter is too low and it is too cloudy to do any more observations Go to the Moon to get spectra down all fibres Analyze Process Fiter Color Pugon Window Heb ON 7 afin YW mas Be b 4HOO AWAR DAD Fophielp press Fl Vertical cut shows different in fibres but can t be certain on throughput variations etc at this point
9. 2007 Galaxy emission line spectrum Kinney et al 1996 Expected Instrument Efficiency 01 5577 29 0 28 gt S 29 5 0 26 e gt El g tes 8 ao cai 0 24 3 E y D E Nal 5893 E bo 30 5 Hgl 5770 Hel 5791 0 22 2 3 Nal S688 g mL s75S 31 D 2 Sel 5351 31 5 0 18 32 naea ia a AO 0 16 4500 4710 5000 5250 5500 5750 6000 6250 6500 6830 7000 Wavelenth A 4 Figure 3 4 3 Representation of the spectra able to be obtained using the 150 l mm grating The vertical red lines show the bandwidth 2120A The template galaxy spectrum solid blue line has a resolution of 15A The expected efficiency solid green line is a convolution of all efficiencies and will be discussed in 3 11 The dotted blue line is the telluric spectrum at similar resolution 3 5 SBIG ST 7E CCD The SBIG SGS s camera is an ST 7E which uses a Kodak KAF 0402E CCD The main factor that affects the performance of BASIS when it comes to using the ST 7E is the Quantum Efficiency QE of the chip itself QE is defined as the ratio of outgoing electrons against incoming photons where 100 would be for every photon the pixel collects one electron is displaced i e one count The specifications of the CCD are given in Table 3 5 the QE of the KAF 0402E is shown in Figure 3 5 1 and an image of the ST 7E connected to the SBIG SGS is shown in Figure 3 5 2 The QE is included in the convolved efficiency shown in Figure
10. didn t illuminate central core so will try further away 58 Richards offset 7 5 7 5 It went well and stayed clear but unfortunately no galaxy spectra I had to do some more calibration checks on the acquisition and guiding and that lead to more problems arising After temporarily sorting them out I tried to centre a star on the central core but could not find the IFU let alone a core I then observed the Moon and got light down the instrument it works and then Mars ran some offsets and found the IFU and located the offset of the IFU This took the best part of Shrs to do I also wavelength calibrated the spectrograph in situ ready for observations using a 650nm laser diode No galaxy spectra because the pointing on the telescope was all over the place half the night sky out so I need to sort that out before attempted to point at galaxies I could find bright objects by eye and slew the telescope accurately enough but the galaxies I have on the list are in the middle of nowhere and so do not have any naked eye reference stars Note should be taken that Mars is only 4 on sky and so to have light from Mars falling down all areas of the IFU means that the telescope is out of focus The eyepiece and IFU don t have the same Focal Plane This needs to be fixed To do on next commissioning night Align the telescope properly Focus the telescope to the IFU Find offset again Observe standard star Observe galaxy
11. of greater than 1 25arcmin and less than 2 00arcmin 1 25 lt D25 lt 2 00 2026 galaxies 2 B magnitude of less than 13 Mp lt 13 686 galaxies 3 Visually selected from SuperCosmos Read 2008 2arcmin square POSSIIR thumbnail images such that the galaxies would fit well into the larcmin angular diameter of the IFU 641 galaxies list used for CKT images see 7 3 4 RA between 9 hours and 12 hours 9 lt RA lt 12 for the time of year of the commissioning nights February to April 2012 and a Declination of greater than 45 degrees Dec gt 45 to observe galaxies near zenith 94 galaxies 5 Declination between 60 and 70 degrees 60 lt Dec lt 70 Best visually selected from CKT images to get a reference on how well these galaxies are observed from Bayfordbury Observatory and on a Meade LX200 16 inch 17 galaxies 6 Best galaxies for each type see Figure 7 2 1 where the galaxies might be slightly oversized but the first observations would be better with more light so the IFU is centred more around the core of the galaxy The BASIS IFU of larcmin diameter is overlaid for reference Any of the galaxies from the 641 galaxy list would be suitable for BASIS which ones exactly would be due to the RA constraint on the time of year and the users discretion depending on what science is desired The current constraint is to just get any galaxy observations and so bright oversized galaxies would be the best to start with More catalogues could
12. 16 7 76E 17 9 2345 6 31E 05 4 66E 05 3 09E 05 Stellar Magnitude 445nm Sre Yipix 551 nm Sre Vipix 658nm Sre Vipix 806nm Sre Yipix 10 0 7118 1 6083 9 5221 8 4272 8 10 5 4491 2 3838 7 3294 7 2696 0 11 0 2833 8 2422 1 2078 8 1701 0 ILS 1788 0 1 28 2 1311 6 1073 3 12 0 1128 1 964 2 827 6 677 2 12 5 711 8 608 4 522 2 427 3 13 0 449 1 383 9 329 5 269 6 13 5 283 4 242 2 207 9 170 1 14 0 178 8 152 8 131 2 107 3 14 5 112 8 96 4 82 8 67 7 15 0 71 2 60 8 2 2 42 7 15 5 44 9 38 4 32 9 27 0 16 0 28 3 24 2 20 8 17 0 16 5 17 9 15 3 13 1 10 7 17 0 11 3 9 6 8 3 6 8 17 5 7 1 6 1 52 43 15 0 A II A IR OA 445nm Dark Sky Y pix 551 nm Dark Sky Y pix 658nm Dark Sky Y pix 806nm Dark Sky Y pix 3 9 33 29 24 445nm Bright Sky Y pix 551nm Bright Sky Y pix 658nm Bright Sky Y pix 806nm Bright Sky Y pix 391 5 334 6 287 2 235 0 53 Bayfordbury Single object Integral Field Spectrograph BASIS 12 APPENDIX B BASIS TIME LINE CHART Z10Z L0 90 01 Z10Z L0 10 7107 90 90 Z107 AP Z107 0 67 TIJ OT 60 80 LO 90 SO PO 0 CO 10 OP 9 SE PE CE TEJ Ic IT ST HC EC TC UT OT GT BT LT OT ST we eT CI JOqUINN JIM saq jueyiodwy qId4S C10C CO PT s our pe q Woday 3107 20 60 Sur DIUIIIS 11OZ Z 1 61 SutuoIrssrtutuoo 1107 21 80 A quiassy LLOC OL LT SUIDINOG _ ueneognoN mdy gids PEx rna Pt j j 4 99S916 Be W 1Sute8P UNS ONIL J j j AJU9IIIJJO JUSWINYSU j j
13. 2012 w40 Conference Amsterdam NLD Bayfordbury Single object Integral Field Spectrograph BASIS 69 2 REFERENCES Note Original Initial Plan references only Bacon R et al 2001 The SAURON project I The panoramic integral field spectrograph MNRAS 326 1 pp 23 35 Bland Hawthorn J et al 2011 Hexabundles imaging fiber arrays for low light astronomical applications Optics Express 19 3 Dreyer J L E amp Sinnott R W 1988 The Complete New General Catalogue and Index Catalogue of Nebulae and Star Clusters Source Catalog Reference NGC 2000 0 Sky Publishing Corporation and Cambridge University Press Holmes A 2011 Private Correspondence Holmes A amp SBIG 2001 Operating instructions for the SBIG SGS and spectra analysis software http www sbig com images documents products 222 Hook I M et al 2004 The Gemini North Multi Object Spectrograph Performance in Imaging Long Slit and Multi Object Spectroscopic Modes PSAP 116 819 pp 425 40 p3d 2011 Screenshots http p3d sourceforge net index php page screens Pasquini L et al 2002 Installation and commissioning of FLAMES the VLT Multifibre Facility The Messenger No 110 pp 1 9 Paturel G et al 2003 HYPERLEDA I Identification and designation of galaxies Astronomy and Astrophysics 412 pp 45 55 Sandin C et al 2010 p3d a general data reduction tool for fiber fed integral field spectrographs A amp A 515 1d A35 Sch
14. 3 4 3 KAF 0402E QE 100 Quantum Efficiency QE e 4600 1850 100 350 5600 5850 6100 6350 6600 6850 Wavelength A Figure 3 5 2 ATC 2010 ST 7E connected to the SBIG SGS at stage 6 in Figure 3 4 1 18 Richards KAF 0402E Specifications Architecture Full Frame CCD Enhanced Response Total Number of Pixels 784 x 520 Number of Active Pixels 765 x 510 Pixel Size 9um x 9um Average QE 65 Dark Current at 0 C l e Read Noise 15 e Read out 16 bit Full Frame Download 1 second Table 3 5 Kodak 2003 KAF 0402E specifications 3 6 Optical fibre cable The fibre cable used in BASIS is a COTS OM2 50 125 250 fibre used by the communications industry Polymicro 2008 There was a long discussion as to what fibre should be used to achieve the best performance possible but it was decided that the OM2 fibre was as good as any science grade fibre when working with lengths of lt 3m The length is the main factor in performance statistics This is because of two main properties of the fibres known as attenuation and Focal Ratio Degradation FRD Attenuation is the name given to the amount of light lost in the fibre over the length specified No fibre is perfect and so when light propagates down the fibre some light is not reflected by the cladding and so escapes from the core see Figure 3 6 1 There are times where the light is absorbed by the composite materials of the fibre itself leading
15. A a oA a a u ua u un uu ayna 8 2 1 The case for integral field spectroscopy and its teaching oococcccncnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnos 8 2 2 CRC INCLINE FUN TANG uuu u aA tater Casella real a ela tla ea ein E a ata area hota es 9 3 Bayfordbury Single object Integral Field Spectrograph BASIS 14 3 1 Choosing the site 14 3 2 Existing equipmentl AA AAA 15 3 3 Meade TARO BG i Biase is scat umum laste asi aaa uapa kus aa A 15 3 4 Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS 16 3 5 SDICOSI 7 CED u cuaca e ee ss dee es dede e ds ROO 17 3 6 Optical ibre cable iii AAA ARA AAA ARAS 18 3 7 Integral Field Unit UE US o dcceuad EE EE iene aa payaa qapa ahay 19 3 8 I D fibre array un veer este Pete Sn Ce Se Sd DC gr ns DRC PE PT On CEPT POT SU DICE TEE a SU DICE TEE FOr NTSC EET E 25 3 9 Manufacturing of integration parts amp wavelength calibration check 26 3 10 Control units and software acquisition and guiding 29 3 11 BASIS instrument summary A A A A Ad 30 A DA cee ee ee ae 34 4 1 First order da
16. Fibre 50um core 125um cladding 250um buffer Fibre Aperture 50um 2 6 arc seconds on sky Fibre Bundle 19 fibres bundled in a layered circular pattern Bundle Aperture 54 2 arc seconds leading edge of first fibre core to trailing edge of last core Fill Factor 4 Sky Fibres 4 sky fibres symmetrically displaced by 5mm 4 3 arc minutes from the bundle Efficiency 0 68 0 98 from attenuation for 5m fibre length 0 7 from IFU construction Spectrograph Model Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS Slit 1 D array of 23 fibres with 50um cores Slit length 2 9mm 23 fibres long without the buffer Grating 150 lines per millimetre Camera Make SBIG ST 7E Camera CCD Kodak KAF 0402E CCD Pixel Size 9um x 9um CCD Array Size 765 x 510 pixels CCD Average QE 65 Spectral Resolution T Spectral Bandwidth 2120A Spectral Limits Chosen 4710A to 6830A Central Wavelength Adjustable Overall Instrument Integral Field Unit 19 fibre bundle and 4 displaced sky fibres IFU On sky Aperture 54 arc seconds 1 arc minute Resolution 7A max of 75kms Bandwidth 2120A adjustable central wavelength Efficiency Expected 22 Realistic 5 Signal to Noise 10 for a 20min exposure of a 13mag arcsec source 0 15 error on BPT diagram Bayfordbury Single object Integral Field Spectrograph BASIS 31 LOSS colored 4 n Ha Velocity Field of NGC 4026 NGC 4026 Example w e Ds e PS e e e bi e L y
17. Field Spectrograph BASIS 11 Hexabundle 009 s Velocity Field Figure 2 2 1 Croom et al 2011 Ha velocity field of spiral galaxy 6dFGS gJ195722 2 55081 The red circle overlaid on the image on the left shows the outer circumference of the IFU Hexabundle shown on the right Each spaxel shown on the right is an individual fibre and has its own corresponding velocity shown by the scale on the far right in units of kms BPT diagrams On top of dealing with line positions and line widths the next piece of information that can be extracted from spectral line data is the strengths of lines It follows that the greater the line strength or line height the more of that ionisation is present Fundamental physics allows the understanding of comparing different line strengths to each other This has long been done and is still one of the greatest ways to truly probe the environment and composition of the emission source Baldwin Phillips amp Terlevich 1981 pioneered the way in a simple graphical classification system the product of which are BPT diagrams They compare different line strength ratios to find out which category an emission spectrum would fall into This is because there are a variety of different ionisation sources The main ones known to date are Active Galactic Nuclei AGN Essentially these are as the name suggests the nuclei of galaxies that are incredibly active extremely energetic pheno
18. High gibbous moon 22 00 Sky Brightness 18 78mag arcsec IFU was dismounted for Open Night Complete recalibration needed BASIS still works though spectra are off centre vertical Wavelength set to 2 4775micron on SGS micrometer see Second Commission Night Notes Mounted SBIG STL 6303E to TV Got TV in rough focus Too cloudy to do anything Wrote instructions for ACP calibration Run setup scripts LimitingMagnitude vbs CalibrateGuider vbs Train Corrector vbs ACP Preferences PinPoint Exp interval 20s set by LimitingMagnitude vbs Max pointing err 0 08arcmin Sarcsec seeing Note may fail Skip targets if pointing update fails tick Weather WeatherWatcher Weather ACP Weather vbs parks when weather says anything but clear set up but needs activating each time via connect Observatory Draw limiting angle 40deg from zero horizon maybe more MaximDL CameraControl Exposure Autosave Collect 20 10min darks Collect 50 biases Collect 20 twilight flats can average a later date via ACP run AutoFlat vbs for sky flats write script for targets top 10 then top 50 check user manual for script guide and tags apply offset via force RA and Dec Report Re mounted all equipment as it had been taken off due to an Open Night and maintenance Too cloudy to do any observations To do on next commissioning night 1 Everything Bayfordbury Single object Integral Field Spectrograph BASIS 63
19. Integral Field Spectrograph BASIS 67 There is an uncertainty in the efficiencies set for the IFU and slit interfaces and have the possibility of swinging either way The main constraint on these interfaces would be the selected tolerances This will depend on a number of factors CNC Machining tolerances including cost factors the alignment of the fibres to the micro lens array the alignment of the fibres at the slit with respect to the optical axis of the spectrograph optics the thermal expansion and the variable expansions between different materials the quality of end face polishing and the induced stress on the fibres when housed glued in the interfaces The components needed to build this instrument are listed in 1 2 under the header Sourcing and those to be purchased are listed in Appendix C The micro lens array is the most critical component and so is the one that will hold up progress if it is delayed The Corning50 125CPS fibre has already been sourced A stand alone camera will be mounted to the guide scope of the LX200 which will do the acquisition and guiding This is necessary as the SBIG SGS self guiding feature only works when using the slit at the focal plane of the LX200 not when using a pseudo slit comprising of fibre optics To reduce the data given by the ST 7E I intend to use pre written software called p3d Sandin et al 2010 see Figure 4 p3d is a general data reduction tool for fibre fed Integral Fie
20. Springer Verlag Berlin Heidelberg New York 1 3 XID p 2069 Dearden S J 1999 A Comparative Review of some http users erols cominjastro faas pages paper001 htm Denny R B 2012 ACP Observatory Control Software DC 3 Dreams http acp dc3 com index2 html Diffraction Limited 2012 MaxIm DL Version 5 http www cyanogen com maxim_main php Dreyer J L E amp Sinnott R W 1988 The Complete New General Catalogue and Index Catalogue of Nebulae and Star Clusters Source Catalog Reference NGC 2000 0 Sky Publishing Corporation and Cambridge University Press DSS 2010 Digitized Sky Survey STScI NASA Healpixed by CDS http alasky u strasbg fr Dss Color Dunlop J S 2011 The Cosmic History of Star Formation Science 333 no 6039 pp 178 81 Fogarty L 2012 Private Correspondence Galactica 2012 LX200 ACF 16 F 10 UHTC C PE EQUATORIAL http www galactica pt ver 305 1 1 3 19 meade Ix200 acf 16 f10 uhtc c pe equatorial php GSMT 2002 Section 4 7 1 Wide Field Multi Object Multi Fiber Optical Spectrograph MOMFOS http www gsmt noao edu book ch4 4_7_1 html Hill G J et al 2008 The Hobby Eberly Telescope Dark Energy Experiment HETDEX Description and Early Pilot Survey Results Panoramic Views of Galaxy Formation and Evolution ASP Conference Series 399 p 115 Ho L C Filippenko A V amp Sargent W L W 1993 A Reevaluation of the Excitation Mechanism of LINERs The Astrophysical Journal 417 p 63
21. a csv file for further analysis 2 The fit file of the exposure can be converted into a ascii file which can then be loaded into Microsoft Excel or something similar Graphs of the spectra can then be made using the inbuilt graphing tool In either case wavelength calibration and sky subtraction can be performed though spectral fitting will need to be done by a more advanced software package Line ratios can be crudely taken from the un fitted spectra though the error bars would be larger by how much would depend on the quality of the spectra 1 e SNR 4 2 Data reduction software There are a few open source data reduction software packages for the purposes of fibre fed integral field units which can be easily obtained the main ones being IRAF Tody 1986 R3D S nchez 2006 and the more recent p3d Sandin et al 2010 All of these can perform full data reduction to the point of creating a data cube of the galaxy which can then be used for full analysis The main steps of data reduction which are automated within software packages are 1 Subtraction of the dark frame includes bias 2 Removal of cosmic rays which can be done using an integrated software routine for example MIDAS Crane amp Banse 1982 using the filter median script If multiple exposures of the same target are available 23 it is possible to take their median and compare it pixel by pixel to each individual exposure If the difference
22. and Science Limits Due to the nature of instrument building the dates assigned to the following tasks are deadlines meaning that as long as they are completed before this date the project is on track though 1 will endeavour to complete the tasks ahead of these dates to allow for inevitable delays The dates shown are in Week Number w where Week Number wl is the week starting 03 10 2011 A visual spreadsheet representation is provided in Appendix B Sourcing w03 Micro lens array w04 Razor blades w04 Fibre cleaver w04 Optical glue and index matching gel w04 Polishing paper w04 Zemax modelling of 1D array w05 1D array housing CNC w05 127um drill bit w05 Micro lens array AR Coating Siltint w07 Fibre to Micro lens array coupling interface CNC w08 LX200 focal plate interface w08 1D array to SBIG SGS interface CNC w09 Rubber conduit w09 p3d software Assembly wl0 IFU wll Fibre feed w12 Slit to SBIG SGS interface Commissioning w15 Installation onto LX200 w15 Auto guiding w18 Commissioning Science Limits w22 Instrument efficiency w22 True SNR against mag arcsec2 Report Deadlines 27 10 2011 w04 Initial Plan 08 12 2011 w10 Poster Presentation 09 02 2012 w19 Sample Chapter amp Contents Page 29 03 2012 w26 Project Report amp Viva in May SPIE Important Dates 19 13 201 1 w12 Abstract Submission 24 02 2012 w21 Author Acceptance Notification 04 06 2012 w36 Manuscripts Due 01 06 07
23. and coupling efficiency The 50um core size of the OM2 fibre results in a poor fill factor where ideal cases would be gt 50 If the fibres had their buffer removed which is easily achieved with a sharp razor blade by scraping off the soft buffer the closest the fibres could get to each other is a core to core separation pitch of 125um For a bundle of 19 fibres taking five fibres width to be the diameter of the bundle the fill factor is given by Eq 3 7 1 iaa sum of the aperture areas 19m 50um _ sion iS spread area of all apertures Tm 5x125um re wa Eq 3 7 1 Knowing that the IFU would be a 19 fibre bundle the next challenge is exactly how one makes a fibre bundle which in itself is a difficult and delicate task The communications industry has dedicated machines to do this which do it very accurately but the cost of getting a small bundle made would run into thousands of pounds Although outside the budget of this project it would become a viable option if BASIS proves to be successful The following list describes the various stages of the development of the IFU fibre bundle This process took most of the time in the project effectively pushing the initial time line so far that it was in effect scrapped with last minute shuffling to be able to do the commissioning and get some on sky data In all this process took 4 months including the time for the 1 D array Bayfordbury Single object Integral Field Spectrogra
24. around though this was later found not be the case as discussion in 5 4 Figure 3 7 6 Microscope image of the new fibre bundle in the substrate after polishing Description is similar to that in Figure 3 7 5 Bayfordbury Single object Integral Field Spectrograph BASIS 25 Even though the IFU took up the majority of time in this project the process of getting to the result shown in Figure 3 7 6 ruled out various alternate methods and discovered problems not originally thought of all of which resulted in a decent and easily replicable IFU instruction manual to follow report Richards 2012 This last version took only a few hours to build not including the curing time and shows the consistency in making the IFU using this method Thinking about it some more a fibre connector possible an FC style connector would work just as well over all and possibly use four separate connectors for the sky fibres This will be left for future work discussed in 6 as time does not permit changes to be made within the deadline of this project 3 8 1 D fibre array With the IFU being on one end of the fibres the other end is the 1 D fibre array slit connected to the SBIG SGS To fit all the 23 fibres along the slit their buffers had to be removed which meant handling the fibres very delicately The conventional method of a 1 D fibre array is to use a V Groove a series of v shaped channels in a substrate that allow the fibres to
25. be possible to get data cubes on galaxies The exact throughput and efficiency of BASIS is yet to be determined but light can get down the instrument and enough light to get spectra If it turns out that the efficiency is closer to the 5 used in the calculations for SNR then there should not be a problem when it comes to observing a galaxy It would have been great to get galaxy observations within the time period of this project but knowing that the concept does work it is just a matter of time before the first velocity field is created Figure 5 5 1 A 60s exposure of Jupiter Here fibre numbers 1 to 7 have been illuminated where fibre number is the central spectrum shown here Wavelength is in the x axis short to long wavelengths from left to right respectively 4700A to 6800A 40 Richards 500 0 Fibre 1 450 0 Pixel Count we o 250 0 200 0 150 0 100 0 0 0 100 0 200 0 300 0 400 0 500 0 600 0 700 0 Pixel Location Figure 5 5 2 Spectrum of fibre number 1 from the 60s exposure of Jupiter Pixel location and counts have been assigned to the x and y axis accordingly thought the x axis 1s the wavelength axis from the CCD ranging from 4700A to 6800A Note the O telluric absorption feature at pixel number 695 Figure 5 5 3 A 60s exposure of the Moon Here all fibres have been illuminated including the sky fibres where fibre number is the central spectrum shown here Wavelength is in the
26. be reduced via a similar selection criteria e g NGC Dreyer amp Sinnott 1988 HyperLEDA Paturel et al 2003 and NED Schmitz et al 2011 though at the moment the RC3 catalogue works very well 44 Richards Spiral Edge on NGC 4026 Spiral Face on NGC 4100 Elliptical NGC 4125 J2000RA h m s 11 59 25 6 J2000RA h m s 11 59 25 6 J2000RA h m s 11 59 25 6 J2000Dec d m s 50 57 43 J2000Dec d m s 50 57 43 J2000Dec d m s 50 57 43 Mp 11 61 Mg 11 64 My 10 80 Figure 7 2 1 Best first targets of BASIS of different galaxy types Spiral Edge on Spiral Face on and Elliptical The BASIS IFU of larcmin diameter is overlaid for reference The images are taken from the DSS archive DSS 2010 via the Aladin reference tool Bonnarel et al 2000 where NGC 4125 happened to land at the border of two tiled exposures leaving the borderline shown The RA and Dec is in J2000 and the Mb is from the RC3 catalogue 7 3 Target galaxy imaging To get a reference on the observing conditions BASIS would have to work to the list of 641 galaxies as described in Step 3 of the selection criteria in 7 2 was submitted to the CKT for observations The CKT is also a Meade LX200 16 inch telescope and uses an SBIG STL 6303E camera for photometric observations It also is a fully automated and robotic telescope running the ACP Observatory Control software meaning that it would work through the 641 galaxies when it c
27. continue More cloud coming in 7 End of night Report All parts of BASIS fit on the telescope and are in apparent working order Centre alignment on the TV CCD can be achieved though offset to the IFU still needs to be found Cloud cover stopped observations To do on next commissioning night Calibrate the wavelength on the SBIG SGS with 650nm laser Find offset to IFU Observe standard star HIP 31681 Observe 56 Richards 14 2 Second Night Date 03 02 2012 Start Time 19 00 End Time 02 00 Site Bayfordbury Observatory Overall Weather Clear Moon High gibbous moon 19 00 23 00 02 00 Sky Brightness 18 11mag arcsec Sky Brightness 18 09mag arcsec Sky Brightness 18 74mag arcsec total of 765pix 212nm 683 650 33 33 212 0 1557 0 1557 765 119 0802 765 119 0802 646pix align 650nm laser 2 4775microns on micrometer testl not long enough exposure 60s test2 long enough exposure not centred NC 120s moved TV 4pix left test3 NC 4p1x not enough moved TV 8pix left test4 NC no spectra NS as was not doing the auto guiding control correctly this one was 12pix left on TV moved TV correctly back to 4pix left test5 NC NS go back to centre on AG to get reference point once more test6 NC NS not sure why NS maybe previous was scattered light from moon as is close by went 5pix right on TV offset 5 0 test7 NC NS offset 0 5 test8 NC NS offset
28. detrimental to the instrument and needed to be corrected for before any automated or deep sky observations with BASIS can be done The decision was then made to use an off axis guider OAG but in this case having the IFU in the off axis side mount position where the guide camera normally goes The primary mount in this case holds the STL 6303 removing the flexure problems in the TV set up A new 14 inch mount shown in Figure 3 9 5 was needed to attach the IFU as no OAG with a 2 inch side mount was available With this in place the acquisition and the guiding using the tracking chip within the STL 6303 should be consistent and accurate within five arc seconds The five arc second 5 tolerance on the acquisition and guiding is a tight one but with the fibre aperture being 2 6 on sky the average seeing at Bayfordbury being 3 and with the fibre pitch being 12 9 on sky 5 is within the overlap tolerance between fibres The LX200 alone cannot achieve this level of accuracy so the telescope control software ACP Denny 2012 was installed alongside the image processing software MaxImDL V5 Diffraction Limited 2012 This combination has the ability to perform plate solving the process of using stars within the field of view to locate where on the sky it is pointing and correct for any error in the pointing accordingly 30 Richards ACP can be used to train the telescope to a pointing accuracy of lt 5 when tens
29. focal length f 5 4 was side mounted to the LX200 with an SBIG STL 6303 camera 9um square pixels with an array of 3072 x 2048 pixels hereon TV will be used to describe the TeleVue and STL 6303 Using an eyepiece on the LX200 the TV was centre aligned as much as possible within 10 pixels and then the cross hair finder scope was also centred as best as possible Because the LX200 had no tracking or auto guiding on this process had to be done quickly as the self tracking in the telescope s mount cannot be relied upon over a few tens of seconds To get within the 10 pixel alignment tolerance the process took 30mins within re centring the target in the eyepiece every 30 seconds It was rather tedious but was necessary for accurate acquisition The IFU is by no means perfect and therefore it cannot be assumed that the fibre number in the bundle is on the optical axis of the telescope Due to this the offset between the TV centre and the centre of the IFU needed to be known Jupiter was used in this situation because it has an angular diameter of 25 meaning that if centred in the IFU only fibre numbers 1 to 7 would be illuminated see 5 Using a star to find the offset would be incredibly difficult due to the fill factor of the IFU After centre aligning the TV it was discovered that due to the weight of the STL 6303 2kg there was considerable flexure in the TV 2 when pointing at different on sky locations This inconsistency is
30. focussing the LX200 works well and will be used in all future cases were the focus is needed The focus for the SBIG SGS was found in the lab as described in 3 9 and was glued in place so that it could not change F e 2 are N Fibre polish paper s Primary Mirror Lock Coarse Focuser that moves the LX200 s Primary Mirror in and out eres CP y k gt ES Figure 5 1 Picture of the back of the LX200 where the fibre polish paper has been stuck to the back of the empty 2 inch focuser Jupiter was used to find the focus using the coarse focus by moving the primary mirror in and out See labels for descriptions 38 Richards 5 22 Wavelength solution A 650nm laser diode was used for the wavelength calibration in setting the central wavelength It was calculated by using the galaxy template given in Figure 3 4 3 to be such that the 650nm laser would need to land on pixel number 646 in the 765pix axis Figure 5 2 shows one of the calibration images where the laser line has been set its required pixel value Using the laser diode to set the wavelength is a good method and will be used for future calibration The wavelength calibration for use in data reduction will be done by taking an exposure of an arc lamp similar to that done for Figure 3 9 8 though this time with the instrument on the telescope No such exposure had been taken yet due to limited time and the priority to get so
31. http 1fs wikidot com 52 Richards 11 APPENDIX A BASIS SNR CALCULATOR BASIS SNR Calculator 445nm SNR Brig Sinm SNR Brigh 658nm SNR Brigh 806nm SNR Brigh x SS 58 5 34 24 x 305 e EEC 3432 IC 25 E 8 is IS E CI 11 37 60 00 60 00 BASIS SNR Dark Night BASIS SNR Bright Night 50 00 50 00 40 00 40 00 445nm 450m Z 30 00 30 00 551nm 551nm 658nm 658nm 20 00 806nm 20 00 060m 10 00 10 00 0 00 r r r Y T 0 00 r r r r Y 10 0 105 110 115 120 125 130 135 140 145 15 0 155 160 165 17 0 175 180 10 0 105 110 115 120 125 130 135 140 145 15 0 15 5 16 0 165 17 0 17 5 18 0 Apparent Magnitude Apparent Magnitude ST 7E Specs Dual CCD Kodak KAF 0402ME T1 TC 237 Pixels Pixel Size microns CCD Size mm Total Pixels Exposure s Cooling K A D Converter A D Gain Back Focus cm 764 x 510 9x9 6 9 x 4 6 390 000 0 12 to 3600 228 0 1 16 Bits 2 3e ADU 2 3 Detector Specs Chip ReadNoise Dark QE 445nm QE 55Inm QE 658nm QE 806nm DT RN KAF 0402ME 0 54 0 71 0 74 0 48 1200 225 mag aresec 2 2 mag arcsec 2 mag arcsec 2 sl s s 1 Night 445nm 551nm 658nm 806m 445nm SSInm 658nm 806nm 20 20 20 1 46E 05 1 00E 05 7 38E 06 4 89E 06 15 15 15 1 46E 03 1 00E 03 7 38E 04 4 89E 04 LX200 Effective Area m h ergsec 6 63E 27 Total Throughput c sec 1 3 00E 18 Exposure Time s Z 445nm Wm 2u 1 7 18 Resolution lambda l Z _ 51nm Wm 2u
32. inch a 4 5m radio dish and a 115m baseline radio interferometer Due to the number of telescopes available it has been possible for the optical telescopes to each host a different instrument Current instruments available are high spec CCDs a fast frame rate camera for planetary or lunar imaging a slit spectrograph SBIG SGS an Ha filter for solar observations and more Recent developments have now enabled a robotic drive for one of the 16 inch telescopes once again pushing Bayfordbury Observatory further into the lead as the UK s best teaching observatory To keep in line with Bayfordbury Observatory s status I will build and commission an Integral Field Unit IFU instrument for one of the 16 inch telescopes It appears that not only will this be the UK s first teaching IFU but also the first IFU hosted by such a class of telescope The benefit of using an IFU is that it enables the observer to obtain spatially resolved spectra of a certain target in a single observation In the field of Astronomy IFUs are primarily used to observe galaxies nebulae and Ha features and can come in a variety of array sizes depending on the specifications of the host telescope and science goals Current world leading IFUs include FLAMES Pasquini et al 2002 based at VLT GMOS Hook et al 2004 at Gemini North South SAURON Bacon et al 2001 at WHT and SPIRAL Sharp amp SPIRAL Team 2006 at AAT All of these are monolithic lenslet arrays that cha
33. is and correcting for a factor of 1 z If the redshift is unknown then this is how it is obtained Multiplying the observed Doppler shift by the speed of light c gives a measure of the line of sight velocity v of the stars or gas within the galaxy see Eq 2 2 2 Appeal v Doppler shift sl E ctri C Eq 2 2 1 PE cs fart ctri Eq 2 2 2 If the velocity is negative then it is said to be blue shifted and if it is positive then it is said to be red shifted normally shown by their respective colours on a velocity field An example of such a velocity field showing the galaxy kinematics derived from the Ha line position is given in Figure 2 2 1 which has been taken from Croom et al 2011 In the same way that single fibre spectroscopy is the integration of all the spectra from the individual sample points known as spatial pixels or spaxels in an IFS each spaxel is still the integration of the all the spectra from the emission sources within its aperture The line width of a particular spectral line contains information of the velocities of the emission sources within the observation aperture meaning that the smaller the spaxel the narrower the line will become Even though it is impossible to deconvolve spectra the line widths do give information about the peculiar motions of the gas stars within the aperture The wider the line the greater spread of velocities 2 2 2 Bayfordbury Single object Integral
34. jet cleaner and then submersed in a 1M Caustic solution for 24hrs after which end stage cleaning is performed 1 e wiping off the remnants of the matrix Due to the 3D printer not being able to print holes as small as 130um but being ok with holes of 260um diameter and greater it was decided to move away from using individual holes for individual fibres and instead use a larger hole to encompass all the fibres The hexagonal holes printed in the Stage 2 test turned out to be rather un hexagonal so trying to get the fibres to fit to a hexagonal hole will not work The next step forward was to make a circular hole large enough to fit all the 19 fibres This was relatively straight forward but it became difficult trying to thread 19 fibres that have had their buffer removed through a tight hole Bayfordbury Single object Integral Field Spectrograph BASIS 23 Stage 4 Figure 3 7 3 Fibre positioning substrate with a 1 2mm hole in the middle and four 260um holes displaced by 5mm from the centre of the middle hole The substrate is 20mm 1 x 20mm w x 2mm d When a fibre s buffer is removed the unprotected glass is very fragile and needs little motivation to snap Forcing fibres into a tight hole with lubrication puts too much stress on them and they snap even for someone who is experienced with handling optical fibres This resulted in the decision to leave the buffer on and have the bundle twice the size with the same num
35. lens array were used then the f 2 beam would be more conserved through the fibre but the problem comes when using the SBIG SGS as it is built to accept light beams with f numbers between f 6 3 and f 10 The higher the f number the more efficient the spectrograph will be though too high and the focal length would be so long the spectrograph would be metres long All of these factors come together when building a telescope instrument and finding the correct proportions is a difficult task As the LX200 and the SBIG SGS are fixed components then the IFU has to work to these parameters The decision was made to omit the micro lens array and have bare fibres at the focal plane This means the fibres are coupled with the f 10 beam from the LX200 meaning that some FRD would take place As mentioned in 3 6 the longer the fibre the more FRD but for a 5m length of fibre the FRD is quite small Without doing a full lab set up to measure the FRD in the OM2 fibre which takes months to perform it is safe to assume that the f 10 would be reduced to at most f 7 across the 5m length where in fact is it likely to be less than this The FRD to f 7 tells you the output cone angle of the OM2 which is within the acceptance of the SBIG SGS of f 6 3 to f 10 This validates the use of the OM2 and also omitting the micro lens array The decision to omit the micro lens array means the IFU would have bare fibres at the focal plane The trade off here is fill factor
36. on a number of factors 3D printer tolerances including cost factors the alignment of the fibres at the slit with respect to the optical axis of the spectrograph optics the thermal expansion and the variable expansions between different materials the quality of end face polishing and the induced stress on the fibres when housed glued in the interfaces The components needed to build this instrument are listed in 1 2 under the header Sourcing and those to be purchased are listed in Appendix C A stand alone camera will be mounted to the guide scope of the LX200 which will do the acquisition and guiding This is necessary as the SBIG SGS self guiding feature only works when using the slit at the focal plane of the LX200 not when using a pseudo slit comprising of fibre optics p3d Calar Alto 3 5m PMAS PPAK 2kx4k CCD Reduction Panel File View Options Help Master bias runta6_0086ib _nbias fits Ls A 0 2 ee Disp mask runt46_00734b_dnask Fits 1 Dil amp 2 1 Trace mask run146_00730b_trace fit Fiber flat run146_00868b_imemb_Flatf fits A amp Object unt46_00729b _oextr its Display of extracted objects 70 6562 114 z 1312 3 i Log yes object slot 3 run146_00725b_oextr fits 3 right click to change the o Figure 1 1 3 p3d 2011 A screenshot of the p3d Figure 1 1 4 Example of a suitable target selected from the software using the Potsdam Mul
37. on campus or near campus observatory for teaching whether it be a single small aperture telescope on the department s rooftop or multiple small aperture telescopes at a nearby isolated site Therefore creating an IFS that is well within an institution s budget lt 500 and utilises existing equipment is an attractive solution to the gap of integral field spectroscopy teaching found in observational astronomy practicals With a practical on integral field spectroscopy students can learn not only the fundamentals of astrophotonic instrumentation including the spectrograph but also and maybe more importantly the physics of galaxies or other extended sources e g planetary nebula discussed in 3 11 Bayfordbury Single object Integral Field Spectrograph BASIS 9 Focal Spectrograph Spectrograph Lenslets Datacube Lenslets Fibres Slicer Figure 2 1 1 Westmoquette et al 2009 amp Allington Smith 2006 Schematic of different types of IFSs all resulting in a data cube for a particular source There are benefits and drawbacks for each style so all types can be found in operation today There is another attractive outcome of creating an IFS for a small aperture telescope which is the possibility of doing a first glance survey of hundreds or even thousands of galaxies This of course all depends on the performance of the instrument and the requirements are outlined in 2 2 The type of galaxies that can be o
38. overall instrument efficiency Total Throughput of 0 05 a 20min exposure of a 12mag source gives a possible SNR of 10 see Appendix A If multiple exposures were stacked to produce a 2hr exposure it would be possible to get a Signal to Noise Ratio SNR of 12 for 12mag and SNR of 10 for 13mag source The overall instrument efficiency Total Throughput of 5 is a harsh estimation The reason for setting it at this level is that instruments always perform worse than originally predicted The SBIG SGS has five optical surfaces between the slit and the CCD Most of the components of the SBIG SGS have good coatings and come from Melles Griot Holmes 2011 giving a 0 90 efficiency per surface The ST 7E uses a Kodak KAF 0402ME CCD chip that has an average optical quantum efficiency QE of 0 65 Therefore the efficiency of the SBIG SGS equates to 0 38 say 0 35 including misalignment of the spectrograph optical For the rest of the instrument there are a number of efficiency factors to take into account 0 90 due to the IFU interface 0 90 due to the slit interface 0 95 due to the Corning50 125CPS s attenuation at 500nm for 2m length 0 70 due to focal ratio degradation FRD of Corning50 125CPS at 500nm for 2m length 0 92 due to Fresnel reflection losses at both ends of the fibre and 0 90 due to the Meade LX200 16 inch Accounting for all of these factors the expected instrument efficiency Total Throughput would be 17 Bayfordbury Single object
39. plane of 40mm 1 4 inch BASIS is well within this tolerance with the largest separation of fibres sky fibre to sky fibre being 10mm 14 inch It can be assumed then that the focal plane is uniform in focus across the entire IFU There is a screw lock on the back of the primary mirror that enables the user to adjust coarse focus and which can be tightened to minimise mirror lag Figure 3 3 1 Galactica 2012 Meade LX200 16 inch Schmidt Cassegrain optical telescope with an equatorial mount 16 Richards 3 4 Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS The spectrograph chosen to be used for BASIS is the Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS see Figure 3 4 1 Again this is a widely used spectrograph one that most institutions around the world have with over 500 sold globally Holmes 2012 The capability of being able to adjust the central wavelength on the CCD by using a micrometer to change the grating angle is one of the main reasons for choosing the SBIG SGS Not only can the user change the grating angles but also change the grating all together from low resolution 150 lines per millimetre to a high resolution 600 lines per millimetre 600 l mm SBIG have recently released an even higher grating that has 1800 I mm though only the 150 l mm and 600 l mm are available on the SGS used for BASIS There is a trade off with using higher resolution gratings and that is
40. saving a factor of 10 on cost as well as the ease of buying the fibre you can buy communication grade fibre at a hardware store warrants the use of the OM2 fibre 3 7 Integral Field Unit IFU Fibre fed integral field spectrographs give rise to a number of options when designing an IFU The first choice is whether or not to use a micro lens array The advantage of using a micro lens array is that you can greatly increase the fill factor and light coupling efficiency with the optical fibres where the fibres are positioned at the back face of the array The disadvantage is that having the micro lenses in the optical path lowers the f number The f number describes the acceptance cone angle where a high f number f 10 has a small acceptance angle and a long focal length and a low f number f 2 has a large acceptance angle and a short focal length At the Cassegrain focal plane where the IFU is positioned the LX200 s light beam is f 10 If a standard micro lens array were used the beam would be reduced to f 2 which is at the limit of the OM2 s acceptance This takes us back to FRD where the degradation is the reduction in f number the result is shown in Figure 3 6 1 and Figure 3 7 1 20 Richards FOCA RATH A puput Til ME FEAL MATI Figure 3 7 1 GSMT 2002 Focal Ratio Degradation in optical fibres The larger the input f number the more FRD is present more deviation from a perfect fibre If the micro
41. sit along them to ensure equal spacing and height Equal spacing is desirable as it will make the automated data reduction of the raw CCD spectra easier Equal height is even more desirable as any deviation in height will be reflected in the wavelength direction on the CCD To account for any deviation in height a calibration emission source can be used arc lamp and the varying line positions on the CCD per spectrum will indicate any variations in the height of the fibres at the slit There were no facilities in house to build a v groove and it was again out of this project s budget to get one made commercially thousands of pounds The decision made was to print a housing for the 1 D array using the 3D printer see Figure 3 8 1 for the drawing where instead of using v groove channels to align the fibres the fibres would be placed side by side along a flat surface with walls of width equal to the total number of fibres width A cap would then be placed on top and the fibres glued in place using the same UHU glue as before Another desirable when building a 1 D array is to have the fibres organised in a patter relating to their position with the IFU Figure 3 8 2 shows such a pattern and is the one used here This pattern is one that optimises the physical visualisation of the IFU when looking at raw CCD spectra It also makes sure that the fibres that are receiving the least amount of light from a target the outer right of the bundle and the
42. sky fibres are not next to the central core at the CCD It removes the possibility of a false emission line from light leaking into adjacent spectra To number the fibres in the IFU individual fibres were illuminated whilst looking at the IFU through a microscope making sure the other fibres were covered One by one the fibres were located in the IFU and a piece of tape with its position number was placed at the end of the fibre This is a task that requires patience as locating a fibre position can sometimes be tricky Figure 3 8 1 Drawing of the 1 D array housing It is 50mm long and an overall 3mm tall The distance from the bottom of the housing to the inside surface is 1 438mm to take into account the radius of the fibres when aligning them to the centre of the housing The caps SOOSCSCOOCOCCOCSSGGOCOSE ECCS are both 1mm in height and 10mm long and fit into their adjacent sections The width of the larger cap is 5 908mm Figure 3 8 2 Schematic of fibre number mapping from the IFU top to the 1 D array bottom and the small cap is 3mm to account for 23 fibres width with and without their buffer on respectively 26 Richards Re eo owe s pat L Ar Sua T D Wi EAS Lt E y gt yi og 00 e Lay are ih Pie See kde sos dd ay Mot Figure 3 8 3 Image of the 1 D array during the curing Figure 3 8 4 Microscope image of the end face of the process after the fibres had bee
43. the solid white arrow The images have been cropped from their original size but there is no difference in pixel location The dotted white arrows show additional residual ghost features from bright foreground stars The red arrow shows the location of SN 2012a The ghost is due to an effect called Residual Bulk Image RBI which is the residual charge on the CCD substrate due to saturated pixels falling back into the pixels after read out The RBI seems to be inherent on all KAF CCD chips and poses a real problem for photometry and of course creating super nova false positives To deplete the RBI it takes 25mins to let all most of the charge fall into the pixels from the substrate You can get around this by performing an infrared flash on the CCD which equalises the RBI across the entire CCD and can then be corrected with the bias frame This is something that might prove difficult with the CKT but a solution must be found before accurate photometry can be performed There is no risk of the RBI being a problem with the SBIG ST 7E because the photon count will be too low from observations and therefore never saturate the CCD It is an amusing coincidence that the exposure that caused the RBI to create the ghost in the exposure of NGC 3344 was of NGC 3239 which was only being observed because it does have a supernova SN2012a see Figure 7 4 3 red arrow so the CKT was getting another measurement for its light curve 48 Richards
44. to complete the tasks ahead of these dates to allow for inevitable delays The dates shown are in Week Number w where Week Number 1 wl is the week starting 03 10 2011 A visual spreadsheet representation is provided in Appendix B though this is the original Gantt chart Time Line Chart It holds then that all parts below need to be in before wk9 noted by the suffix 1 for latest Sourcing see Appendix C w09 1 Optical Fibre 24 core w09 1 Polishing paper w09 1 Glue w09 1 Fibre cleaver w09 1 3D printing of IFU IFU Mount Focus Mount 1 D array housing SBIG SGS Mount w09 p3d software Assembly Commissioning w10 IFU w15 Installation onto LX200 wll Fibre feed w15 Auto guiding w12 Slit to SBIG SGS interface w18 Commissioning Science Limits Report Deadlines w22 Instrument efficiency 27 10 201 1 w04 Initial Plan w22 True SNR against mag arcsec 08 12 2011 w10 Poster Presentation 09 02 2012 w19 Sample Chapter amp Contents Page 29 03 2012 w26 Project Report amp Viva in May SPIE Important Dates 19 13 201 1 w12 Abstract Submission 24 02 2012 w21 Author Acceptance Notification 04 06 2012 w36 Manuscripts Due 01 06 07 2012 w40 Conference Amsterdam NLD 1 3 Initial Plan comments The above initial plan is somewhat different to the original one drafted at the start of this project given in Appendix E The reason for the change is one that is inherent to all instrumentation projects the unpredic
45. 0 o O IS N me CEE J Figure 3 11 1 Example Ha velocity field of NGC 4026 a suitable target for BASIS with the fibre bundle IFU overlaid to scale The scale of the velocity field is in units of kms and 75kms is representative of BASIS s maximum resolution log Olll 5007 HB 15 10 05 00 05 log NII 6583 Ha Figure 3 11 2 Example of NGC 4026 placed on a BPT diagram red with error bars relating to that expected from BASIS signal to noise ratio of 10 The signal to noise ratio SNR of 10 for a 20min exposure of a 13mag arcsec source was equated using the BASIS SNR calculator displayed in Appendix A Here an overall instrument efficiency of 5 was used instead of the expected 22 a convolution of all individual expected efficiencies The reason why 5 is used for this calculation is that instruments tend to always be under budget when it comes to efficiency but even when accounting for this a SNR of 10 is ideal for the purposes of BASIS Stacking exposures together can increase the SNR and the brightest sources will be observed first to minimise the errors on the initial data Target selection and observations are discussed in 6 With this level of performance BASIS could be used to carry out a full survey of hundreds possibly thousands of galaxies and gather useful data to aid our understanding of galaxy composition and evolution Just what type of galaxies will be discussed in 7 but in s
46. 000 6500 Rest Wavelength A 0 61 os MD 0 47 1 19 Seyfert NGC 1358 o 1 0 72 gt rom 1 30 og pal 0 32 H Relative Intensity f 5000 5500 6000 6500 Rest Wavelength A 0 61 7 g Ly 0 47 fs o 1 19 Ha N I 0 72 i g pal 0 32 o 11 LINER NGC 1052 01 1 30 Relative Intensity f t o 5000 5500 6000 6500 Rest Wavelength A Table 2 2 2 Standard emission line flux ratios for the classification of the source of ionisation in an emission line spectrum extracted from Kewley et al 2006 An example spectrum for each classification is given for visual interpretation of the spectral analysis The spectra shown here have been extracted and modified for comparison from Ho et al 1993 14 Richards 3 BAYFORDBURY SINGLE OBJECT INTEGRAL FIELD SPECTROGRAPH BASIS The main drive for building BASIS was the attempt to create an IFS that could be used by institutions and amateur astronomers around the world To do this a number of constraints had to be put on the instrument These included to use as much Commercial Of The Shelf COTS products as possible to be installed at an average site i e not at the top of a mountain to use existing components where possible to be within a reasonable budget lt 500 see Appendix C and to be as easily replicated as possible with construction methods available to the average user The following sections
47. 1 D array comprising of the 19 fibre IFU and 4 sky fibres displaced from the main bundle to minimise contamination see Figure 1 1 2 The fibre feed will be 5m long and protected in a flexible and lightweight rubber conduit Mounts for both the IFU onto the telescope and the 1D array onto the SBIG SGS will be made using a 3D printer Object3D FullCure VeroBlack The plate scale of the LX200 is 51 6 mm meaning that each IFU element of 50um core diameter has an aperture of 2 6 Therefore the 19 fibre bundle will have a field of view of 1 e Figure 1 1 1 Holmes amp SBIG 2001 SBIG SGS Figure 1 1 2 Schematic of the fibre bundle configuration with sky fibres four fibres displaced from main bundle actually 5x the distance shown The red dots are representative of the 50um flbre cores Note that the hexagonal shape is ideal though most likely not possible for BASIS 6 Richards The SBIG SGS s camera is an ST 7E which uses a Kodak KAF 0402ME chip with 9um pixels in a 764x510 array Holmes amp SBIG 2001 Using the 150 lines mm grating and with the 1 D fibre array giving a 50um slit it is possible to obtain a resolution of R 750 at 500nm 7A This means that with taking into account the specifications of the SBIG SGS and applying an overall instrument efficiency Total Throughput of 0 05 a 20min exposure of a 13mag arcsec source gives a possible SNR of 10 see Appendix A If multipl
48. 303E to be placed at the primary position of the off axis guider mount meaning that the flexure problems with the camera mounted on the TeleVue would no longer be present Thought would have to be taken on how to make sure sky fibres can be used in the off axis position but considering that a standard off axis guider has a CCD of 5mm x 5mm in size then it should not be too difficult to incorporate The use of FC connectors with this proximity is a problem due to the overall diameter of an FC connector being 9mm Separate sky exposure could be taken but that increases the time overheads dramatically and would never be exact There may be a way to include the sky fibres in the same FC connector as the fibre bundle but it would be the separation of the sky fibres to the bundle would be at most 1 arc minute 1 D Array 1 Extra room is needed in the 1 D array to ensure no fibres are pushed upwards as they are packed in to allow room for consecutive fibres A V groove would be ideal but the cost just does not seem worth it in this case SBIG SGS No changes needed It appears to be working as expected Bayfordbury Single object Integral Field Spectrograph BASIS 43 7 OBSERVATIONS As it was not possible to get observations of standard stars and galaxies in the time of this project the following section details the work in creating a target list of standard stars and galaxies for future observations 7 1 Standard stars To be able to get
49. 5 0 test9 NC NS offset 0 5 test10 NC NS not sure what to do except go to an object about 30 diameter and bright Quick observation of Orion s nebula by centring the bright centre to the centre of the TV m42 centred on guide camera to bright region m4 _ testl NS 120s take longer exposure of m42 Scrap that going to Mars as is brighter and saves time now 22 50 Bayfordbury Single object Integral Field Spectrograph BASIS 5 7 Quickly take 25s exposure of Moon Mars centred on TV and nearby star is being used for auto guiding near bottom just to right mars_testl NC NS checked eyepiece and it was not centred on eyepiece Moved telescope until Mars was centred on eyepiece best by eye TV with nearby star just underneath and to the right of Mars mars_test2 NC S will perform crosshair offsets of 5pix again until centre found offset 0 5 mars_test3 NC S carry on with offsets offset 5 0 mars_test4 NC S carry on with offsets offset 0 5 mars_test5 NC S first time illuminating Ist ring finish crosshair offset before testing either side of this position offset 5 0 mars_test6 NC S Ist ring visible still so assume in between these two offsets will go for the corner of the offset square offset 5 5 mars_test7 NC S getting closer as lines are brighter will halve the offset in both directions offset 2 5 2 5 if AG allows half pixels mars_test8 NC S
50. 8 FUTURE WORK As this project is something that could actually gather new scientific data and one that could be replicated by others the decision was made to submit it to the SPIE Astronomical Telescopes and Instrumentation 2012 conference in July 2012 titled BASIS Bayfordbury single object integral field spectrograph The relevant deadlines were marked in the Project Line shown in 1 2 which were met and the project was accepted for the conference in form of a poster SPIE requests that a manuscript be published in the conference proceedings which is to be of equivalent standard to a research paper being published in a journal For the purposes of this paper and also for continued help throughout the project a small BASIS Team was created that includes the individual people noted in the start of the acknowledgements Additional to the SPIE conference a poster presentation has also been accepted to the UK Germany National Astronomy Meeting 2012 NAM 2012 conference in Manchester UK March 2012 and also to the Scientific Committee on Antarctic Research Astronomy amp Astrophysics from Antarctica 2012 SCAR AAA 2012 conference in Portland Oregon USA July 2012 The poster for the NAM and SCAR AAA conferences will be a more generic topic of Integral field spectroscopy on small aperture telescopes In addition to conferences there will be a push for a press release once a galaxy has been observed and its data cube analysed Ta
51. Bayfordbury Single object Integral Field Spectrograph BASIS Samuel Richards School of Physics Astronomy and Mathematics University of Hertfordshire Hatfield ALIO 9AB UK SN 08169597 AST4 BSc Hons Astrophysics with a Research Year 29 March 2012 ACKNOWLEDGMENTS I would like to extend my deepest gratitude to Prof William Martin and Prof Hugh Jones for their continuous guidance and help throughout this project It has been an honour to have them as joint supervisors and I will cherish the wisdom given in this time In addition I would like to also thank Prof Joss Bland Hawthorn USyd Dr Julia Bryant USyd Dr Sergio Leon Saval USyd Dr Lisa Fogarty USyd Dr Jon Lawrence AAO and Dr Michael Goodwin AAO for their help in the start up of this project whilst I was in Sydney working as part of the SAMI Group For their comments and guidance I would like to thank Prof Elias Brinks Dr Mark Sarzi Dr Daniel Smith Dr Mark Gallaway and Mr David Campbell I take this opportunity to highlight the assistance of Mr David Campbell in the commissioning phase of the project a true asset of Bayfordbury Observatory I would also like to thank the Science and Technology Research Institute and the Centre for Atmospheric amp Instrumentation Research at the University of Hertfordshire for the use of their Optics Laboratory and 3D Printer respectively Funding for this project came from the University of Hertfordshire s School of Phy
52. E Nal 5893 if NIL 6548 30 5 gt y Hel 5770 Hgl 5791 a SIL6716 4 on 673 Nal 5688 z xe NII 5755 v log Fluz erg slem A 31 Sel 5351 4710 5250 5750 6250 6830 Wavelenth 4 4 Figure 2 2 0 Template Sc galaxy and telluric spectra Data for templates was taken from Kinney et al 1996 amp Sanchez et al 2007 respectively Emission line labels are given to denote the atom and ionisation state that creates them 2 2 1 Velocity Fields The easiest property of a galaxy obtained via spectral analysis would be the galaxy s kinematics in the form of velocity fields using the Ha line relative to its control position rest frame wavelength 6563A As stars and gas move within but primarily rotate with the galaxy they cause a Doppler shift on the spectra observed If you can obtain spectra from different parts of the galaxy i e from an IFS then you can get Doppler shifts for the spectra from each sample point Knowing the position of the rest frame wavelength after fitting a Gaussian profile to the line the simple subtraction of the observed line position A away from the control line position A lt and then dividing by the control line position yields the Doppler shift see Eq 2 2 1 Care must be taken to make sure the overall Redshift z of the galaxy has been accounted for This is done by looking at the integrated spectrum of the entire galaxy finding where the position of the emission line
53. F sine On the fibres that is coming out of the fibre cores The colouring around the fibre cores comes from the fibre buffer In the communications industry the fibres are coloured to make channel recognition and matching easier The last problem to tackle in building the IFU was the residual lubricant on the fibres A completely new unit was made but this time Tetrahydrofuran THF was used to remove the silicone lubricant It was rather persistent hence the use of THF If the fibres are left too long tens of seconds in the THF then the buffer is also stripped so caution is needed not to exposure the fibres too long Dipping them in and out of the solution a few times seemed to work the best followed by a quick wash with distilled water and an air blow dry The fibres were then lubricated with the UHU glue before threading in the central hole of the fibre positioning substrate The same procedure of glueing cutting and polishing from Stage 4 was performed and the result is shown in Figure 3 7 6 Again the organisation of the fibres in the bundle is really good though the polishing might be a bit better The polishing was made difficult due to the proximity of the IFU Mount VeroBlack material the glue not being the best type of glue for this application and the fibre buffer appearing to be softer than that of science grade fibre previously worked on This time the central fibres did not seem to slide in and out when moving the unit
54. Holmes A 2011 Private Correspondence Holmes A 2012 Private Correspondence Holmes A amp SBIG 2001 Operating instructions for the SBIG SGS and spectra analysis software http www sbig com images documents products 222 Hook I M et al 2004 The Gemini North Multi Object Spectrograph Performance in Imaging Long Slit and Multi Object Spectroscopic Modes PSAP 116 819 pp 425 40 Kauffmann G et al 2003 The host galaxies of active galactic nuclei Monthly Notices of the Royal Astronomical Society 346 4 pp 1055 77 Kewley L J et al 2001 Theoretical Modeling of Starburst Galaxies The Astrophysical Journal 556 1 pp 121 40 Bayfordbury Single object Integral Field Spectrograph BASIS S51 Kewley L J Groves B Kauffmann G amp Heckman T 2006 The host galaxies and classification of active galactic nuclei Monthly Notices of the Royal Astronomical Society 372 3 pp 961 76 Kinney A L et al 1996 Template ultraviolet to near infrared spectra of star forming galaxies and their application to K corrections The Astrophysical Journal 467 pp 38 60 Kodak 2003 KAF 0402E ME Device Performance Specification Document http www eso org sci facilities paranal instruments mascot doc KAF 0402 ELongSpec pdf 1 p 17 Lahav O amp Suto Y 2004 Measuring our Universe from Galaxy Redshift Surveys Living Rev Relativity 7 http www livingreviews org Irr 2004 8 Law D Weijmans A M amp Wr
55. Observational evidence for AGN feedback in early type galaxies Monthly Notices of the Royal Astronomical Society 382 4 pp 1415 31 Schawinski K et al 2010 Galaxy Zoo The Fundamentally Different Co Evolution of Supermassive Black Holes and Their Early and Late Type Host Galaxies The Astrophysical Journal 711 1 pp 284 302 Schmidt M amp Green R F 1983 Quasar evolution derived from the Palomar bright quasar survey and other complete quasar surveys The Astrophysical Journal 269 Part 1 pp 352 74 Schmitz M et al 2011 The NASA IPAC Extragalactic Database NED Enhanced Content and New Functionality American Astronomical Society AAS Meeting 217 344 08 Bulletin of the American Astronomical Society 43 Sharp R amp SPIRAL Team 2006 The SPIRAL IFU integral field spectroscopy at the AAT AAO Newsletter No 110 p 24 SUSS 2011 Image courtesy of SUSS Hexagonal grid micro lens array http www suss microoptics com shop microlens arrays fused silica circular lenses hexagonal grid microlens array nr 13 9910 101 000 html TASS 1997 Bright Northern BVRI Standards http stupendous rit edu tass refs skiff_photom html Tody D 1986 The IRAF Data Reduction and Analysis System Society of PhotoOptical Instrumentation Engineers SPIE Conference Series p 733 Westmoquette M S et al 2009 The integral field spectroscopy IFS wiki arXiv 0905 3054 This article accompanies the opening of the IFS wiki site
56. The sky fibres might be compromised in this shift though if sky subtraction is desired then offset sky exposures could be taken additional to the galaxy observations though this would increase the time overheads As the sky brightness at Bayfordbury Observatory does not often exceed 17mag arcsec its contribution to observed spectra would be negligible and therefore the sky fibres are rendered mute However if long exposures more than an hour are desired then they would need to be used to make sure there is no contribution due to the sky background Proposed changes to the instrument for future use will be discussed in 6 Figure 3 11 3 Figure 3 11 4 and Figure 3 11 5 show BASIS installed on the telescope when in the design phase of using the 2 inch Focus Mount Meade LX200 16 inch with an equatorial mount SBIGSGS f L ie Z4 Ls 1 Figure 3 11 3 Picture of BASIS on the LX200 from inside the telescope dome The orange cable is the fibre optic cable and the labels show the position of the IFU and the SBIG SGS Bayfordbury Single object Integral Field Spectrograph BASIS 33 SBIG STL 6303E TeleVue 102 102mm diameter refractor je VOG S b Cross hair finder scope f Meade LX200 16 inch 2 IFU with 2 inch Focus Mount 1 D Fibre array positioned at the slit using the SGS Mount JU 1 D array housing was glued P lt in place when focus was a
57. alignment failed as I used stars with not enough RA separation 2nd alignment completed using stars Dubhe and Alpheratz 2 3 Slewed to Jupiter not in centre of telescope so alignment did not really work 4 Whilst at Jupiter I fixed the focus problems that last seem to be responsible for the 1000 times less efficiency from last time I did this by focussing on a piece of paper similar to what was done for SAMI in the rotation calibration Amusingly you could see Jupiter s cloud bands on the paper projection Re centred and re aligned everything all over including the calibration of centring the guide scope Re ran the Two Star Align procedure Went to another star to check alignment and yet again still off by 20arcmin Calculated telescope offset and went to another star it again was off centre Applied the calculated offset and after was still 10arcmin off centre Pointing accuracy seems to change depending on the location of sky so simple offsets cannot be done Ended the night as I need to seriously come up with some fix for this issue before I can observe anything I think David Campbell and I came up with the solution so should be better by next time To do on next commissioning night 1 Obtain MaxImDL Licence 2 Install ACP software 3 Re calibrate whole set up 62 Richards 14 4 Fourth Night Date 01 03 2012 Start Time 19 00 End Time 22 00 Site Bayfordbury Observatory Overall Weather Cloudy Moon
58. and do so in substrates gt 1mm A series of tests were then performed on the 3D printer to find its realised resolution Figure 3 7 3 shows the array that were printed to perform this test where as many parameters were varied as possible including the hole size hole shape hole orientation substrate thickness and manufacturing consistency 22 Richards Stage 3 Figure 3 7 2 Drawing of the 3D printer parameter test array Each row is half the thickness of the previous starting with 2mm 1mm then 0 5mm A zoomed image of one substrate is given in the lower right Here the holes in the first column are circular the second are square and the third are triangular The first row has 130um holes spaced 250um apart The second row has 260um holes spaced 400um apart The third row has 400um circular holes a hexagon of edge to edge diameter of 1 275mm five fibre diameters and the same hexagon but rotated by 30 to get a vertex on the leading edge The whole set was replicated five times giving the array in the main image All substrates are 10mm 1 x 10mm w The outcome of this parameter test revealed a consistent printing resolution of gt 200um which means printing holes of 130um in size is not possible The shape of the hole did not really make a difference but when threading fibres through the triangular holes were the easiest The orientation of the hexagons produced no difference The substrate thickness did have a big dif
59. and the ghost lines up well with the bright foreground star BD 17 2217 see Figure 7 4 3 UGC5840_H slpha_600s_B2_001 fts 358 x 4424 NGC 3344 Clear Filter 300s NGC 3344 Ha Filter 600s Date 27 01 2012 Date 03 02 2012 Time 17 09 Time 22 37 Figure 7 4 1 300s Clear Filter and 600s Ha Filter exposures of NGC 3344 showing the super nova feature located by the white arrows A Rainbow colour map has been applied to bring out the features where dark blue is low flux and bright red is high flux The colour map is normalised to each exposure respectively The dimensions at the bottom of the exposure are the size of the image here in arc minutes Bayfordbury Single object Integral Field Spectrograph BASIS 47 Figure 7 4 2 2 5s Open Filter exposure of NGC 3344 taken by the 2 5m NOT showing no super nova feature expected position located by the white arrow A Rainbow colour map has been applied to bring out the features where dark blue is low flux and bright red is high flux The dimensions at the bottom of the exposure are the size of the image here in arc minutes Note the rotational difference from Figure 7 4 1 given by the compass bearing in the lower right NGC 3239 Clear Filter 300s NGC 3344 Ha Filter 600s Date 27 01 2012 Date 03 02 2012 Time 17 09 Time 22 37 Figure 7 4 3 600s Clear Filter exposure of NGC 3239 and 600s Ho Filter exposure of NGC 3344 taken by the CKT showing the ghost located by
60. are to be used FC connectors are standard commercial fibre optic connectors that have a plastic ferrule with a 127um hole to thread the fibre down The diameter of the ferrule is 2 5mm and its length is 15mm It is then possible to use a 1 3mm drill bit to enlarge the hole such that the 19 fibre bundle can be threaded down FC connectors also have a key lock that fixes rotation which means that the rotation would be consistent if the IFU is taken off the telescope for observatory open nights maintenance and then put back on for observations It would also mean that the exact position of the fibres projected onto the sky could be known consistently Sky fibres would possibly have their own connector so there would be a total of five connectors The FC connectors would also make the whole IFU unit much more robust meaning that it would last a lot longer when being handled 2 Use a one part epoxy for the glueing to remove the effect of bubble caused by the mixing process in two part epoxy The exact glue is unknown but UV curing glue would work well 3 When glueing the fibres in the bundle be more diligent with the application of the glue to make sure that when the glue cures that no fibres can slide in and out of the bundle IFU Mount 1 To change to mount to a 14 inch mount such that the IFU could be placed in the off axis position of an off axis guider mount This would allow the acquisition and guiding camera at the moment the SBIG STL 6
61. area summing up all fibre core areas against the total spread area of the fibres The implications of this will be discussed in 3 7 lt 4 Length of optical fibre gt Figure 3 6 1 Schematic of an optical fibre cable The grey is the fibre buffer 250um diameter the black is the fibre cladding 125um diameter and the white channel is the fibre core 50um Note not to scale The green cones a and b are the acceptance and exit cones respectively with the acceptance cone drawn over b to show FRD The green arrows show the light path note the greater angle of reflect through the length of the fibre due to FRD The yellow arrows show the light escaping from the core going into the cladding and eventually finding its way out of the fibre completely This adds to the attenuation greatly Bayfordbury Single object Integral Field Spectrograph BASIS 19 100 N T I s oe suma usmc isasapicasaqicusaqi y BOI 7 EE a N sa w a Ln e ja l N MN NER s P PEE ee Wavelength nm Figure 3 6 2 Adapted from Polymicro 2008 OM2 optical fibre attenuation The blue area is the data from Polymicro 2008 and the yellow area is the estimated data Data for sub 500nm is very rare as there is no need to go this far down in communications Note the scale is in dB km so for a 5m length of fibre these values are greatly reduced Power in s O Somer Eq 3 6
62. atterns and so only gets 26 clear time see Table 3 1 Its location just outside of Greater London in the countryside means that light pollution is a factor but not as much as most UK observatories that are inside cities Table 3 1 gives descriptive figures of Bayfordbury With all factors considering including existing equipment Bayfordbury is a great site to host BASIS and to test its capabilities Bayfordbury Observatory Location Hertfordshire UK Longitude 0 094399 Latitude 51 774891 Altitude 65m Percentage of clear time 26 2 Average seeing 3 Table 3 1 Bayfordbury Observatory site data taken from the Ordnance Survey UK and Bayfordbury 2012 Figure 3 1 Bayfordbury Observatory monthly average percentages for clear nights Bayfordbury Single object Integral Field Spectrograph BASIS 15 3 2 Existing equipment A big part of keeping the cost down is the use of existing equipment The main components of BASIS are the telescope it is mounted on the IFU and the spectrograph Two of those are pre existing at Bayfordbury the telescope Meade LX200 16 inch and the spectrograph Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS see 3 4 for a full detail of the SBIG SGS For the IFU the method of construction was chosen be done on an Object3D printer using the FullCure VeroBlack material The 3D printer was already in house at the University of Hertfordshire and has the great benef
63. ber of fibres The trade off here is that the fill factor goes down to 4 This is a very low fill factor but for a quick prototype which BASIS currently is it will suffice Keeping the buffer on made the whole process much easier and the substrate that was printed is shown in Figure 3 7 3 The substrate was then cleaned left in the 1M Caustic solution for 24hrs and cleaned again Fibres with their buffer on were threaded through the 260um holes to clear out the residual matrix material The central hole was expanded to 1 3mm using a drill bit this size of drill bit has a thread of 10mm long The 19 fibres were then inserted into the hole After eventually arriving at a fibre positioning substrate that works the next stage is glueing the fibres in place and polishing the entire unit down until a clean flat surface is achieved Before glueing the fibres were first threaded through the IFU Mount see 3 9 which the fibre positioning substrate was connected to All fibres in furcation tubes come with a lubricant added during manufacture so this had to be removed before glueing A tissue was used at this stage to remove the silicone gel lubricant The glue used was UHU Plus Endfest 300kg Epoxy Adhesive 33g a two part room temperature curing epoxy This glue was chosen for its curing properties cheapness and ease of purchase though in hindsight registered optical glue may have been the better choice As the fibres still have their buffer o
64. bserved with a small aperture telescope are near by redshift z 10 and bright apparent surface brightness 14 Most of this class of galaxies have not been observed with an IFS yet so any data that can be obtained would contribute to our understanding of galaxy composition and evolution All the current IFSs around the world have a purpose and most of them are probing into a set of parameter space that is specific to that instrument Combined they are unravelling some of the greatest mysteries about galaxy composition and evolution so for a new parameter space to be added to this knowledge is only beneficial 2 2 Scientific rationale The spectral energy distribution of a star tends to peak in the optical and as the luminosity of a galaxy primarily comes from the stars it contains the host galaxy normally has strong spectral features in the optical Analysis of these spectral features gives insight into galaxies stellar and gas kinematics star forming regions and rates what the primary ionisation source is via placement onto a BPT diagram see 2 2 2 and much more Spectral line analysis Is a widely used tool and there are many ways in which you can extract information to find answers to the above properties Figure 2 2 0 shows a template optical emission spectrum from an Sc type galaxy with the average telluric spectrum overlaid By looking at such a spectrum the variations in emission line strengths and widths are easily visible whi
65. ch is why they are used to probe the properties of the source of emission The labels for each line are also given and knowing which lines are which is key to the analysis Fundamental physics dictates the positions of these lines most of which are due to the ionisation of that particular atom Roman numerals are given after the atomic symbol to denote the level of ionisation where T is the neutral atom no ionisation II is the first level of ionisation one electron removed HT is the second level of ionisation two electrons removed and so on The Greek letters a B y etc denote the level of transition in the Balmer Series of atomic Hydrogen from principal quantum number n 3 to n 2 3 2 4 2 5 2 etc respectively The line widths and strengths are a product of the environment and makeup of the location of ionisation Therefore by knowing the observed features we can know something about the ionisation source and its environment Spectral classification of galaxies depends greatly on these data so when spectral data is obtained with spatial information via an IFS much more can be known about the galaxy than by just using a single aperture instrument 10 Richards 28 3055085282 Telluric Spectrum S nchez et al 2007 Galaxy emission line spectrum Kinney et al 1996 Ho 6563 28 5 01 5577 29 s OIIL 5007 7 HG 4861 z 01 6300 NI1 6584 OTI 4959 30 Hel 5461 OL 6364 E
66. cover these constraints and describe the development and method of construction for each part 3 1 Choosing the site When it came to choosing the site to install BASIS there was not much of a choice though this was not by any means a disadvantage Due to BASIS being built within the University of Hertfordshire the Bayfordbury Observatory which is a part of the university was the default choice If there were a number of sites to choose from Bayfordbury would still have been chosen This is because Bayfordbury is acclaimed to be the best teaching observatory in the UK with four 16 inch and two 14 inch optical telescopes a 4 5m radio dish and a 115m baseline radio interferometer With this amount of telescopes at hand it is possible to assign each telescope with a specific instrument purpose Bayfordbury continues to push its capabilities and lead at the forefront of using small aperture telescope technology with recent the commissioning of the UK s first fully remote telescope the UK s best Lucky Camera sensitive high frame rate camera that can obtain high resolution imaging and the UK s longest baseline radio interferometer within a university s observatory It is only right then that this project follows in these footsteps and so implementing the World s first IFS on a small aperture telescope at Bayfordbury seems fitting As with all astronomical observatories in the UK Bayfordbury suffers from the unpredictable weather p
67. ctrum Kewley et al 2001 that has a resolution of 15A It shows then that even if the theorised 7A is an overestimation the spectral lines needed for the science goals can still be resolved though realistically a 15A resolution is the limit before some lines become too convolved to be resolved Small modifications had to be made to the SBIG SGS for use in BASIS mainly the removal of the slit mask the calibration LED the entrance aperture glass mask and other unneeded components to the self guiding system that were obstructing instalment How the fibre slit is attached to the SBIG SGS will be covered in 3 8 Figure 3 4 1 Holmes amp SBIG 2001 SBIG SGS The Figure 3 4 2 Holmes 2011 SBIG SGS Zemax numbers are respective to those in Figure 3 4 2 though in Optical Ray trace design without the first mirror this case there is a mirror similar to stage 5 that is in Light path starts 1 Entrance slit 1 D array of fibres between stages 1 and stage 2 Here the letter a indicates 50um slit 2 Collimator 3 Adjustable grating 4 the micrometer that adjusts the grating angle and therefore Focuser 5 Mirror 6 CCD the central wavelength on the CCD Bayfordbury Single object Integral Field Spectrograph BASIS 1 7 BASIS template spectrum A in 765pix direction giving 21204 bandwidth 28 Lower wavelength boundary 4710 4 Upper wavelength boundary 6830 4 Telluric Spectrum Sanchez et al
68. decision was made just to get some data no matter how poor it might be Surprisingly it came out rather well and Figure 5 5 2 and Figure 5 5 4 show the spectra obtained from the central fibre in each observation extracted by converting each exposure into a ascii file and plotting with Microsoft Excel As the light from these targets is essentially reflected sunlight they appear the same though it is interesting that in both cases it is possible to see an absorption feature at pixel number 695 which corresponds to the O atmospheric absorption line also seen in Figure 5 5 6 Figure 5 5 5 shows a vertical cut displaying clear separation of each fibre s spectrum and crudely indicates that their respective throughputs are somewhat variable It is not possible to quantify the throughput from these exposures due to the conditions of observation variable cloud cover uneven illumination due to gibbous moon etc The most important claim to take from these observations is that the spectrum of Jupiter proves that integral field spectroscopy on small aperture telescopes is possible and that spatially resolving an object can be done The telluric absorption feature and the overall spectral profile of both Jupiter and the Moon correspond well with what is expected of these objects see Figure 5 5 6 At the end of the day if accurate and consistent acquisition can be achieved something that is proving difficult to do on small aperture telescopes it would
69. e exposures were stacked to produce a 2hr exposure it would be possible to get a Signal to Noise Ratio SNR of 12 for 13mag arcsec source The overall instrument efficiency Total Throughput of 5 is a harsh estimation The reason for setting it at this level is that instruments always perform worse than originally predicted The SBIG SGS has five optical surfaces between the slit and the CCD Most of the components of the SBIG SGS have good coatings and come from Melles Griot Holmes 2011 giving a 0 90 efficiency per surface The ST 7E uses a Kodak KAF 0402ME CCD chip that has an average optical quantum efficiency QE of 0 65 resulting in the efficiency of the SBIG SGS equating to 0 38 The overall size of the telescope spectrograph dome permits the use of a short length of fibre in this case 5m This means the optical performance of the communication fibre used in BASIS is at worst 15 loss in the blue and at best 7 loss in the red Communication grade fibre was chosen to keep cost to a minimum Therefore working with the following efficiencies LX200 0 90 IFU 0 90 Fibre 0 70 including Transmission and Fresnel reflection losses and Focal Ratio Degradation SBIG SGS 0 38 the expected efficiency of BASIS is 0 22 There is uncertainty in the efficiencies set for the IFU and slit interfaces which have the possibility of swinging either way The main constraint on these interfaces would be the selected tolerances This will depend
70. ectra will be slightly blue shifted at the CCD Residual bubbles from the mixing of the UHU epoxy are also clearly visible adding to the conclusion that glue better suited to this purpose should be use possibly a one part UV curing glue All the fibres are active and the array seems to be in working order 3 9 Manufacturing of integration parts amp wavelength calibration check There are a few other parts needed to be built before the IFU can be attached to the telescope and the 1 D array connected to the SBIG SGS hereon SGS Mount Starting at the telescope the IFU needed a mount and also a focuser mount Figure 3 9 1 and Figure 3 9 2 show the drawings of the IFU Mount and 3 inch Focus Mount These were both printed on the 3D printer as 1t was easier to do this than to build from metal Both of them came out as intended and there was no difficulty in attaching the IFU to the IFU Mount for glueing polishing see Figure 3 7 4 This was also the case for the new IFU when attaching it to the 3 inch Focus Mount using M3 nuts and bolts M3 refers to the size and is classed as Metric 3mm diameter of thread The focuser that was available for use at the time of commissioning had a 2 inch mount so one that would fit to the IFU using the M3 holes was milled from aluminium adapted from a scrap piece found in a left overs draw at the University of Hertfordshire see Figure 3 9 4 This was used for the first batch of commissioning but after p
71. entred around the nucleus these regions are known as LINERs They may resemble Seyfert II galaxies but are lower in luminosity and have a distinguishable stronger OI optical line strength Kewley et al 2006 Low lonisation Emission line Regions LIERs These are distinguishable from LINERs by their high dependence on the Hydrogen Balmer Line series and a stronger Helium line at 3970A rest frame It is suggested that this is because LIERs are more associated with star forming regions than AGN Buttenshaw 2011 12 Richards HII Regions These are regions of the galaxy in which Star Formation SF is active Massive young stars are born which peak their emission in the ultraviolet resulting in the ionisation of the surrounding Hydrogen cloud hence HII The greater the star formation rate SFR the stronger the Hydrogen Balmer Line series If SF is strong then when other emission line strengths are compared to the Balmer Series their respective ratios are lower The age of the galaxy does play a part in this relationship as the older the galaxy the lower the SFR and more generations of stars meaning metal rich stars which have a lower effective temperature and therefore lower ionisation ability Dunlop 2011 Wolf Rayet WR Stars Stars of mass gt 20 30Mo go through intense mass loss on the main sequence and by the time they have reached the Red Giant Branch RGB have shed their Hydrogen envelope exposing a Helium ric
72. er project or many projects albeit seeing the extent of the Ha emission in a galaxy is a useful reference for BASIS and will be used when BASIS observes said galaxies Bayfordbury Single object Integral Field Spectrograph BASIS 45 UGC4821_Clear_3005_B2_001 fts 1 79 x 2 211 NGC 3486 Clear Filter 300s Figure 7 3 300s Clear Filter and 600s Ho Filter exposures of various galaxies A Rainbow colour map has been applied to bring out the features where dark blue is low flux and bright red is high flux The colour map is normalised to each exposure respectively The dimensions at the bottom of the exposure are the size of the image here in arc minutes 46 Richards 7 4 super nova watch Due to the CKT taking exposures of nearby galaxies there is a chance that the Ha follow up will reveal a super nova Even though the chances are extremely small it was thought that the Ha follow up of NGC 3344 did indeed show a super nova feature Figure 7 4 1 shows the location of the feature which is also visible in Figure 7 3 It s prominence in the Ha follow up and no respective feature in the Clear exposure leads one to believe that this must be a super nova These exposures were only examined by eye on the 23 02 2012 and seeing that they were taken six days apart strengthened the possibility that it could be a super nova The 24 02 2012 was cloudy and so no follow up could be done at Bayfordbury so time was requested
73. f gentle persuasion and upgrading The resulting instrument is one that is pleasing and paves the way to a full working instrument after a few necessary upgrades are carried out including changing the design of the IFU to fit within an FC connector and be able to be mounted on the off axis position of an off axis guider mount Once this is done the instrument will be ready to observe the proposed target galaxies in 7 2 taking advantage of the ACP Observatory Control Software This project also acts as the pre cursor of many other related projects including the pipeline needed to reduce and analyse the integral field spectra gathered by BASIS and the mapping of HII regions using the galaxy images taken by the CKT It becomes evident then that upon the success of this instrument larger involvement in the form of a BASIS Team to help with such data reduction is needed It is exciting to think about the role of BASIS in the years to come 50 Richards 10 REFERENCES Allington Smith J 2006 Basic principles of integral field spectroscopy New Astronomy Reviews 50 4 5 pp 244 51 ATC 2010 CCD kamery SBIG http www atc astro eu Sbig Bacon R et al 2001 The SAURON project I The panoramic integral field spectrograph MNRAS 326 1 pp 23 35 Baldwin J A Phillips M M amp Terlevich R 1981 Classification parameters for the emission line spectra of extragalactic objects Publications of The Astronomical Society of the Paci
74. ference where the smaller holes on the 2mm thick substrate had closed up roughly half way down the hole The 0 5mm substrate was too thin and therefore too flexible so would not be able to hold the fibres properly The support matrix mentioned earlier had a large role in this test too as the matrix filled all of the holes An optic fibre is not strong enough to push it out and no metal wire that thin in diameter could refrain from bending With multiple sets of the substrates it was possible to do an investigation into using Caustic solution Sodium Hydroxide NaOH to corrode away the matrix without harming the substrate Solutions of 0 1M 0 1M 0 1 Moles where IM of NaOH is 40g Litre 0 2M 0 5M and 1M were made up and each column of substrates from the array went into each solution respectively with one column of substrates omitted for control The results of this test took a couple of weeks as it was repeated with a replicated substrate array After 1hr in the solution little had happened to any of the substrates and with checks at 1hr 2hrs 4hrs 6hrs 8hrs 24hrs and 36hrs only from the 24hr check did a real difference in the matrix become evident which was only for the 1M solution too After 36hrs all of the substrates matrixes were partially dissolved to allow the remainder to be pushed through with threading an optical fibre To this end all future parts printed had their matrix cleaned as much as possible by hand or the water
75. fic 93 pp 5 19 Bayfordbury 2012 All Sky Camera http star herts ac uk allsky Bland Hawthorn J et al 2011 Hexabundles imaging fiber arrays for low light astronomical applications Optics Express 19 3 Bonnarel F et al 2000 The ALADIN interactive sky atlas A reference tool for identification of astronomical sources Astronomy and Astrophysics Supplement 143 http aladin u strasbg fr pp 33 40 Buttenshaw N 2011 Telling LIERs from LINERs Final Year Projects PAM University of Hertfordshire Hatfield Hertfordshire AL10 9AB Cappellari M et al 2011 The ATLAS3D project I A volume limited sample of 260 nearby early type galaxies science goals and selection criteria Monthly Notices of the Royal Astronomical Society 413 2 pp 813 36 CDS 2012 Centre de Donn es astronomiques de Strasbourg VizieR catalogue service http vizier u strasbg fr cgi bin VizieR Crane P amp Banse K 1982 The Munich Image Data Analysis System Memorie della Societa Astronomica Italiana 53 1 pp 19 29 Croom S M et al 2011 The Sydney AAO Multi object Integral field spectrograph SAMI Monthly Notices of the Royal Astronomical Society arXiv 1112 3367 p accepted Dec 2011 Dahari O amp De Robertis M M 1988 A statistical study of properties of Seyfert and starburst galaxies Astrophysical Journal Supplement Series 67 pp 249 77 de Vaucouleurs G et al 1991 Third Reference Catalogue of Bright Galaxies
76. h core There are two main types of WR stars categorised by the composition of their core either WN Nitrogen rich cores or WC Carbon and Oxygen rich cores If emission spectra has a strong broad Helium emission line as well as respective strong broad emission lines of Nitrogen for WN and Carbon and Oxygen for WC then it is safe to assume that the ionisation source is from WR stars All of the above ionisation sources exhibit different spectral features so spectral line analysis can differentiate between them Spatially resolved spectral data from an IFS can lead to analysis of the different regions of the galaxy instead of assuming the whole galaxy is ionised by one source This is why integral field spectroscopy can remove the biases found in single aperture spectroscopy Figure 2 2 2 shows BPT diagrams comparing different line strength ratios and the classification regions for each ionisation source First order interpretation of the three main regions of the BPT diagrams is as follows Star Formation Produces strong Hydrogen Balmer Series giving rise to low ratios in comparison to any other line strengths Bottom left on BPT diagram AGN Strong high ionisation of all atoms leading to a high ratio of OIII HB near equal ratios of NU Ha and SH Ha and a fair ratio of OI Ha Top on BPT diagram LINERs Low ionisation meaning a low or near equal ratio of OIII HB and closer to equal ratios of NII Ha SII Ha and OI Ha Righ
77. h two rings and the greater the fill factor the ratio of spaxel on sky area to surface area of observed source the better but at minimum a few percent These lines are the ones that can be observed using a small aperture telescope and are therefore the lines that can be used for not only teaching the physics of galaxies but also contributing to current knowledge of galaxy composition and evolution Bayfordbury Single object Integral Field Spectrograph BASIS 13 log O II 5007 HB 15 10 0 5 00 1 0 0 8 0 6 0 4 0 2 0 0 0 2 20 15 10 05 0 0 log NII 6583 Ha log SII 4 6717 6731 Ho log OI 6300 Ho Figure 2 2 2 Schawinski et al 2010 BPT diagrams comparing different line strength flux ratios The dotted curve Kauffmann et al 2003 Ka03 represents the theoretical maximum cut off for ionisation due to star formation The solid curve Kewley et al 2001 Ke01 represents the empirical cut off for ionisation due to star formation The solid line Schawinski et al 2007 S07 represents the empirical separation between ionisation from AGN above the line and ionisation from LINERs below the line Primary source Standard line ratio formula for classification Example spectra for each classification of ionisation Kewley et al 2006 Ho et al 1993 ea 0 61 log 1 30 Ha Ha N 11 O111 0 72 Star Formation Hp og SUD 0 32 Ha l 1 30 Relative Intensity f 5000 5500 6
78. ight S 2012 The Mapping Nearby Galaxies at APO MaNGA http dunlap utoronto ca research surveys Osterbrock D E 1991 Active galactic nuclei Reports on Progress in Physics 54 4 p 579 p3d 2011 Screenshots http p3d sourceforge net index php page screens Pasquini L et al 2002 Installation and commissioning of FLAMES the VLT Multifibre Facility The Messenger No 110 pp 1 9 Paturel G et al 2003 HYPERLEDA I Identification and designation of galaxies Astronomy and Astrophysics 412 pp 45 55 Polymicro 2008 Low OH and High OH Optical Fiber http www polymicro com catalog 2_15 htm Read M 2008 SuperCOSMOS Sky Surveys SSS Batch form pixel data http www wfau roe ac uk sss batchfile html p extracted 2012 Rees M J 1984 Black Hole Models for Active Galactic Nuclei Annual Review of Astronomy and Astrophysics 22 pp 471 506 Richards 2012 Contact to request BASIS Instruction Manual samuelnathanrichards gmail com Sanchez S F 2006 Techniques for reducing fiber fed and integral field spectroscopy data An implementation on R3D Astronomische Nachrichten 327 9 p 850 Sanchez S F Cardiel N Verheijen M amp Benitez N 2007 Integral Field Spectroscopy of the core of Abell 2218 ESO Astrophysics Symposia pp 193 98 Sandin C et al 2010 p3d a general data reduction tool for fiber fed integral field spectrographs A amp A 515 1d A35 Schawinski K et al 2007
79. ing instruments that have contributed so much to this field already e g 2dfGRS and SDSS Lahav amp Suto 2004 amp Bland Hawthorn et al 2011 These biases arise because when a single fibre is placed over a target galaxy or other then the spectrum that is observed is the convolution of the light from the entire galaxy This is bad because the emission observed from a certain part of a galaxy is different to that from another If one area e g the core of the galaxy is much brighter than another area e g extended regions then the information about the emission from the extended region is hidden behind the bright spectrum of the core There is another problem when using single fibre spectroscopy which is the exact placement of the fibre on sky and which part of the galaxy or extended region it is observing if the galaxy is larger than the aperture of the fibre This results in spectra being assigned to the galaxy that is not a full representation of the galaxy it is observing Both of these biases in addition to some others are greatly reduced or even removed when observing a source with an integral field spectrograph An integral field spectrograph IFS that has really shown the true abilities of integral field spectroscopy is SAURON Bacon et al 2001 on the 4 2m William Herschel Telescope WHT and in particular the recent project ATLAS 3D Cappellari et al 2011 carried out using SAURON collecting data on 260 early type galaxies I
80. ircular attachment plate has M3 sized holes that allow for rotation adjustment The extra height at the entrance of the SGS Mount s slide is in case the rubber protection over the fibres needs to go inside as well Figure 3 9 4 2 Side image of the milled 2 inch Focus Mount The M3 holes in the 2 inch Focus Mount were pre existing so new M3 holes in the IFU Mount were drilled to match these Figure 3 9 2 Drawing of the 3 inch Focus Mount The outer diameter of the Focus Mount is 3 inch to fit onto the telescope s 3 inch focuser and the overall length of the Focus Mount is 76mm to back set the IFU for purposes of focusing and protection The IFU Mount attaches to the back side of the Focus Mount using M3 nuts and bolts in the provided holes Ed Figure 3 9 4 1 Front image of the milled 2 inch Focus Mount The IFU Mount is connected to the back via new M3 holes drilled in the diagonals of the IFU Mount plate Figure 3 9 5 Drawing of the 1 inch Focus Mount The IFU insert into the back using the M3 holes for attachment The square shaft accounts for the length of the respective IFU shaft and the circular shaft is the 14 inch mount extending for 20mm 28 Richards Figure 3 9 6 First spectra through the entire instrument Figure 3 9 7 1s exposure after fixing the focus rotation illuminated with a near white LED source This is a Is and grating angle to adjust the wavelength Spectra are ex
81. is greater than a specified sigma value then that pixel is replaced by the corresponding pixel in the median image There are problems with this method when using data from world leading instruments but for the purposes of BASIS it should work well 3 Performing aperture tracing the process of taking a cut perpendicular to the dispersion and fitting a Gaussian profile to the light peaks This provides knowledge of where on the CCD the spectra are which will be used in the following steps 4 Wavelength calibration using an arc frame an exposure of an arc lamp that accurately provides known emission lines preferably as narrow as possible The pixel locations are then calibrated with the wavelengths so that in any of the observations the wavelength positions are well defined in each spectrum Bayfordbury Single object Integral Field Spectrograph BASIS 35 Correcting for throughput variations by taking a flat field traditionally a twilight flat Any variation in pixel counts between spectra is directly proportional to the variation in performance of their respective fibres This can then be corrected similarly to a standard flat field used in photometric imaging Subtraction of sky background contribution This is the primary purpose of having sky fibres outside the fibre bundle in the IFU At this point the sky fibres spectra have gone through the above steps and then a median sky spectrum is created If any contribution is ob
82. ising of the 37 element IFU and 4 sky fibres coupled to what would be redundant micro lenses in the array structure see Figure 3 The fibre feed will be 2m long and protected in a flexible and lightweight rubber conduit for protection Mounts for both the IFU onto the telescope and the 1D array onto the SBIG SGS will be made using a Computer Numerical Control CNC machine The plate scale of the LX200 is 51 6 mm meaning that each IFU element of 250um diameter has an aperture of 12 9 Therefore the 37 element hexagonal grid micro lens array will have a major axis aperture of 1 5 sul s YiY sYrYs s A a A P W 1 s J 6 33 m sl al asX 32 1 2 30 31 s NS CODCOD COSOOCC COSCO GX aG G OX s 05 G3G9 Gs 2323252 G G3G3G3G3sSIYS Figure 3 Top Schematic of the micro lens array configuration with sky fibres S1 S4 noting that the empty space within the perimeter contains redundant micro lenses the same size as those shown with the perimeter being the edge of the 3mm x 5mm array Bottom Configuration of 1D array with mapped fibres The SBIG SGS s camera is an ST 7E which uses a Kodak KAF 0402ME chip with 9um pixels in a 764x510 array Holmes amp SBIG 2001 Using the 600lines mm grating and with the 1D fibre array giving a 50um slit 1t is possible to obtain a resolution of R 750 at 500nm This means that with taking into account the specifications of the SBIG SGS and applying an
83. it of being able to do one offs essential for product development It takes a while to print some parts dependant on the height of the part half an hour per millimetre of height depth This time is offset against the labour hours that would go into building it by hand so in the end it is much more efficient to print parts This ability to print one offs meant that changes could be made very easily but also meant that changes happened frequently From the initial design of some parts to what the final print looked like is quite different The process of development of individual parts is detailed in the following respective sections 3 3 Meade LX200 16 inch The telescope chosen to host BASIS was the Meade LX200 16 inch Schmidt Cassegrain optical telescope see Figure 3 3 1 one of the most sold high end small aperture telescopes around the world This means that not only does it fit BASIS as being a telescope that can achieve deep sky research but also that it makes BASIS that little bit more replicable for other institutions and amateur astronomers It has a 3 inch opening for mounts at the focal plane though most instruments have 2 inch or even 14 inch mounts that fit into either an adapter or usually a focuser that acts as the adapter This is because the focal plane is not a perfect plane and is curved meaning the further away from the optical axis the more unfocused it will be To this end most instruments are designed to a working focal
84. its of manufacturing in many areas The first difficulty comes with trying to make 127um holes in a substrate No matter the material this proved more problematic than originally thought It is possible to buy 127um drill bits but they only have a thread that extends for l1mm which means the substrate has to be less than Imm thick not enough to hold a fibre in position Threading fibres through two or even three substrates that had larger holes in them and then interlocking the substrates to fix the position of the fibres was the next method This would work in theory but getting it to work practically was very tricky and so this method was dropped Using a chrome mask to create the holes would work but is very expensive thousands of pounds therefore not deemed to be viable option either Sights were then focused on using an in house 3D printer Object3D This 3D printer uses the polymer material FullCure VeroBlack 3D printers work by layering down a liquefied photosensitive polymer and immediately exposing it to UV light hardening the substrate This whole process is called Digital Light Processing DLP There is a soft polymer support matrix that fills the gaps in the product design to give support during the printing process The matrix is removed afterwards and the finished product is left This all sounds very straight forward but the unknown here is whether or not the 3D printer could print holes as small as 127um
85. k and where relevant the complications that arose respectively The very first step was to install mount all the different pieces of equipment SBIG STL 6303E onto the TeleVue 102 securing a safe position for the SBIG SGS at the base of the telescope mount attaching the IFU onto the 2 inch focuser with the 2 inch Focus Mount 5 1 Focus By manually pointing the telescope at a star HIP 31681 the TV was the first piece of equipment to be focussed The TV has a manual focuser so using short exposure time the focus was found very quickly Attempting to focus the LX200 was more involved It was assumed that the focus of a 2 inch eyepiece would be at the back plate of the eyepiece but after taking an observation of Mars with the IFU 1t was found that this was an incorrect assumption A piece of fibre polishing paper semi transparent was attached to the back of the focuser with no eyepiece see Figure 5 1 and focus was found using the coarse focus on the LX200 by moving the primary mirror unlocking the mirror for focus and locking it back up after focus had been found The focuser was set all the way in and after equating the in set of the IFU to the 2 inch Focus Mount to be 9 8mm the 2 inch focuser was moved back to this distance This seemed to work perfectly as further observations of Jupiter revealed that the size of Jupiter on the IFU was as expected 25 arc seconds only illuminating the central seven fibres in the bundle This method of
86. l bandwidth of BASIS is larger than originally theorised in Figure 3 4 3 with a bandwidth of 330nm instead of 212nm respectively Note the reason for the low a line in Figure 3 9 8 could be because the lamp was still warming up This is a large increase in bandwidth and so when observing the extra bandwidth will be given to the blue end of the spectrum to see if it 1s possible to observe down to the OII13727 emission line If what is being seen here is correct then this opens up more science goals that BASIS could achieve if it is somehow wrong then nothing will have changed 3 10 Control units and software acquisition and guiding When reading through the commissioning reports given in Appendix D it becomes evident that there were real difficulties with the acquisition and guiding the process of locating a target with the telescope such that it consistently lands in the centre of the IFU and the process of keeping said target in the middle of the IFU over any length of exposure respectively 2m telescopes have systems for this that cost millions of dollars to build due to their importance If the acquisition is not consistent then no reliable science can be achieved and the same goes for not keeping the target in the same position throughout the exposure Because the IFU is located at the Cassegrain focus of the telescope it is not possible to use a camera at the same location Therefore a TeleVue 102 refractor 102mm diameter 540mm
87. ld Spectrographs IFSs It has a user friendly interface and includes a vast range of options for case specific data reduction There will be an inevitable uncertainty on the expected performance of this instrument so decisions on exact science goals and the use of the instrument will be left for a future project My aim will be to carry out various observations to discover the limits of the instrument so viable science goals can be drawn To do this I will reduce various catalogues NGC Dreyer amp Sinnott 1988 HyperLEDA Paturel et al 2003 and NED Schmitz et al 2011 to find suitable targets see Figure 5 p3d Calar Alto 3 5m PMAS PPAK 2kx4k CCD Reduction Panel ons Help laster bias rund46_00661b_mbias fits race mask run146_00730b_trace fits Disp mask run145_00731b_dmask fits Fiber flat run146_00668b_imemb_Flatf Fits Object Fun146_00729b_oextr fits Display of extracted objects 70 6562 11A 47 4312 1042 97 952 mn Log yes object slot 3 run146_00725b_oextr fits right click to change the orientation Figure 4 p3d 2011 A screenshot of the p3d software using the Potsdam Multi Aperture Spectrophotometer PMAS instrument as an example Figure 5 Example of a suitable target selected from the NGC catalogue with the BASIS IFU overlaid to scale 68 Richards 1 2 Project Time Line For this project there will be four main stages Sourcing Assembly Commissioning
88. lks seminars at research institutions will be sought after with the first one being for the Science and Technology Research Institute STRI at the University of Hertfordshire on the 23 03 2012 Public outreach of BASIS and the science of integral field spectroscopy has already started with involvement at the Public Open Nights at the Bayfordbury Observatory The project will inevitably be continued beyond the deadline of this report and the instrument developed further to incorporate the changes as proposed in 6 The target for the SPIE paper is to have at least 50 galaxies observed which does not leave much time to resolve all of the current complications though the success here will come from time and effort which have been applied since the start of this project Sights will be set to utilise the SAMI_tools package in the data reduction of such galaxy observations Bridging the divide between amateur astronomy and scientific research is something I have a passion for For that reason I would like to see this instrument and its science used by as many people as possible A document providing full instructions on how to build such an instrument and to carry out the science will be provided in due course when an instrument in full working order has been completed Richards 2012 Once in full completion an article will be submitted to astronomy magazine such as Astronomy Now so a wider audience including amateur astronomers can read abou
89. me observational data Figure 5 2 A Is exposure of the 650nm laser diode showing that it is set at the 646pixel location as desired The variations in line strengths are primarily due to non uniformity in illumination 5 3 Alignment position rotation Because the IFU is at the Cassegrain focus of the telescope there is no room for a camera to do the acquisition and guiding As previously explained in 3 10 the SBIG STL 6303E camera was then mounted to a TeleVue 102 refractor that itself was side mounted onto the LX200 The centre of the TV was aligned with the centre of the LX200 as best as possible by eye roughly within 2arcmins which had to be done quickly because the telescope drive cannot track that well by itself It was then found that this centring was inconsistent due to flexure in the TV which can be read about in the Commission Reports given in Appendix D The SBIG STL 6303E is too heavy to be used as the alignment camera There are no other options in camera for use on the TV so a solution needs to be found that can permit accurate and consistent acquisition The proposed idea is to use an off axis guider mount with the SBIG STL 6303E in the primary mount position and the IFU in the off axis mount position This will correct the problem of flexure and will remove the need to align two different telescopes The rotation of the IFU is set as best as possible by a spirit level but when the acquisition is sorted the exact rotati
90. mena that closely match the luminosity of their host galaxy They are the result of gas accreting onto a Super Massive Black Hole SMBH As the gas accretes onto the SMBH it heats up due to the conversion of kinetic and potential energies 1 e angular momentum gravitational etc Rees 1984 The heating up disassociates electrons from atoms ionising them producing a peak in the ultraviolet optical wavebands Osterbrock 1991 There are quite a few sub classes of AGN and further more sub classes of those sub classes The first biggest category classification is between types Quasars and Seyfert The separation is due to the luminosity of the AGN and host galaxy and boarders at an absolute My 23 Schmidt amp Green 1983 If the galaxy and AGN are brighter than M 23 it is classified as a Quasar or QSO and if fainter a Seyfert Seyfert galaxies make up most of the known AGN They can be divided into two sub classes called Seyfert I and Seyfert II The divide this time comes from the emission line width If the Full Width Half Maximum shows peculiar velocities of 3000kms then it is a Seyfert I galaxy if 350kms then it is a Seyfert II galaxy Dahari amp De Robertis 1988 Low lonisation Nuclear Emission line Regions LINERs When spectra only really show emission lines up to the first ionisation level of any atom and are relatively strong it is safe to assume that the environment is in a state of low ionisation When c
91. mitz M et al 2011 The NASA IPAC Extragalactic Database NED Enhanced Content and New Functionality American Astronomical Society AAS Meeting 217 344 08 Bulletin of the American Astronomical Society 43 Sharp R amp SPIRAL Team 2006 The SPIRAL IFU integral field spectroscopy at the AAT AAO Newsletter No 110 p 24 SUSS 2011 Image courtesy of SUSS Hexagonal grid micro lens array http www suss microoptics com shop microlens arrays fused silica circular lenses hexagonal grid microlens array nr 13 9910 101 000 html End of Original Initial Plan End
92. n aligned and the caps 1 D array after polishing The yellow dots are the placed on The end face is to the left and there is a clamp 1lluminated cores of the fibres Note fibre number 22 left holding down the smaller cap to ensure there was no of array on image is higher than the others and the movement during curing remnant bubbles left in the glue after curing To position the fibres in the 1 D array the first task was to strip the buffer off back to 3cm and then insert the fibres one by one into the housing starting with fibre number 22 and working the way through to fibre number 23 as shown in Figure 3 8 2 This was difficult and required a lot of patience and calming of nerves When fibres have had their buffer removed they are brittle and so it is a demanding task to thread 23 such fibres by eye each of 125um diameters into a space 3mm wide and 130um tall without breaking them After the fibres had all been aligned in the 1 D array the caps were placed on top and the fibres were glued in place see Figure 3 8 3 The glue was left to cure and then the 1 D array was polished under the same conditions as Stage 4 from 3 7 The result shown in Figure 3 8 4 was again one that was better than expected though could still be improved by allowing extra room in the width of the 1 D array compromising on equal spacing Fibre number 22 fibre on the far left of Figure 3 8 4 1s sitting slightly higher than the other fibres meaning that 1ts sp
93. n cleavers pen cleaver prod_22 html In house CAIR 13 APPENDIX C BASIS PURCHASED PRODUCTS LIST Product Product Item ae Source Number Description 20 4 30um 4 12um 4 9um 4 3um 4 lum Price P amp P Total each 60 00 45 12 3 95 12 73 15 54 5 50 21 04 2 70 6 00 6 00 Estimated Total 100 00 Total 238 89 1 6 exchange US 382 25 Contingency Total 400 00 The contingency total allows for changes to the above list Changes are likely as the project develops due to obstacles that need to be overcome The interface components for the spectrograph and the telescope will be constructed using the Object3D Printer in house at the University of Hertfordshire Bayfordbury Single object Integral Field Spectrograph BASIS 55 14 APPENDIX D COMMISSION REPORTS 14 1 First Night Date 03 02 2012 Start Time 19 00 End Time 21 00 Site Bayfordbury Observatory Overall Weather Cloudy Moon High gibbous moon Installed BASIS onto the telescope Installed the SBIG 6303E camera onto the TeleVue 102 TV hereon Ensured telescope mount fixings were tight Focussed TV Centre aligned HIP 31681 on the TV to check accuracy of centre alignment within a couple of pixels lt amea Oviplay W ty Merc Track Filter A O DSS Window Help OG OS 80S eeseea an serao Gre Focur TS fut TAA 200 SW 00 or Hm ea Shady 6 Cloud cover too much to
94. n sky via Stellarium RA 02 07 29 DEC 11 47 37 Will point back to Betelgeuse to see alignment if any pointing offset Betelgeuse is out of centre handset says 05 55 53 07 24 41 moved to centre Betelgeuse is off TV centre but re telescope centre with callipers change TV centre after centring handset says 05 56 09 07 24 42 need to re align telescope via two star align align v3 Dubhe Almaak Dubhe centred Almaak centred telescope two star align successful go to Betelgeuse for align check Betelgeuse BG is off centre by a lot handset says BG is 05 55 51 07 24 25 after moving BG to centre of TV handset says 05 58 51 07 03 29 offset equals 00 03 00 00 20 56 go to Regulus to check offset Regulus not in centre handset says 10 09 02 11 54 01 apply previous offset 10 12 02 11 33 05 go to these coordinates via handset Regulus is way off after applying offset v1 when centred Regulus the handset says Bayfordbury Single object Integral Field Spectrograph BASIS 61 10 11 48 11 45 50 Regulus centre aligned offset v2 End night 23 14 after discussion with David Campbell about running automated system on LX200 Need MaxImDL licence and manually moved dome It got to 10degC tonight After having all the pointing issues last time I started the night by re balancing the telescope completed and then proceeded to align the telescope via a Two Star Align standard procedure 1 1st
95. n they are protected so most glues should not present stress issues to the fibres The glue was mixed and poured over the substrate into the well of the IFU Mount see Figure 3 7 4 and left for 24hrs to allow the glue to fully cure After the 24hr curing period the fibres were cut short to the base using a pair of pliers The whole unit was then polished using sand paper on a rotary grinder P1200 grade extremely coarse for fibre optics but this was only to even out the glue IFU Mount Once the coarse polish was finished a series of different graded fibre polishing papers were used to get the desired finish starting at 30um paper then 12um 9um 3um and finally lum paper The end finish is as best it can be when polishing by hand The result is shown in Figure 3 7 5 and turned out much better than expected The organisation of the fibres within the bundle follows near concentric rings very close to the shape of a Hexabundle Even though this was a great result there were some complications with the central fibres sliding in and out of the array due to residual lubricant 24 Richards Stage 5 Figure 3 7 4 Fibre positioning substrate Figure 3 7 5 Microscope image of the connected to the IFU Mount and glued in place The 19 fibre bundle is in the middle and the four sky fibres surround it There are fibre bundle in the substrate after polishing The bright white dots are due to the light small dioplels o
96. ntegral field spectroscopy will only become more important in the years to come and with this the next generation of astronomers will be the ones to analyse the vast swathes of data that will be produced by large scale surveys These large scale surveys tens of thousands of galaxies are only possible by taking advantage of recent advancements in astrophotonics and their implementation on world class 4 meter telescopes such as SAMI Croom et al 2011 There is one area that has not yet been exploited which is the main premise for this project the continued advancements in commercially available small aperture telescopes and their respective instruments that have the potential to achieve initial integral field spectrographic observations of hundreds or even thousands of galaxies This opens the door for the science provided by integral field spectroscopy to be more attainable by institutions around the world In this attainability it is possible for the next generation of astronomers to get hands on experience of the instrumentation and science of IFSs There are currently in the order of thirty IFSs around the world ranging from lenslet arrays to fibre feeds with or without lenslets to image slicers Westmoquette et al 2009 see Figure 2 1 1 though all but one are on 2 m telescopes making hands on astronomy teaching a difficult task due to time constraints and practicality It 1s because of this that most institutions around the world have an
97. of pointings are performed across the observable sky ACP was designed to be a full Observatory Control software meaning that it can remotely run the telescope in a fully automated robotic mode running from a scheduler once the hardware to allow this has been installed One telescope and dome unit at Bayfordbury Observatory the CKT another 16 inch Meade LX200 has already been upgraded to permit this Following the lessons learnt with upgrading the CKT will mean that BASIS could become fully automated and be run from a scheduler This would be required if a full survey is accepted 3 11 BASIS instrument summary As an overall instrument BASIS has turned out to be a neat little instrument that has the potential to obtain the science goals as proposed in 2 2 2 After going through the instrument part by part 3 2 to 3 10 Table 3 11 summarises all of the different parameters and goals of BASIS some of which are currently not known and so are shown in italics Figure 3 11 1 and Figure 3 11 2 show a simulated velocity field and BPT diagrams respectively and take into account the specifications of BASIS when equating the error bars BASIS Specifications Telescope Model Meade LX200 Schmidt Cassegrain Primary Mirror 16 inches 406 4mm Focal Length 4064mm Focal Ratio 10 Focal Plate Scale 51 6 arc seconds per millimetre Integral Field Unit Model Optical fibre bundle with displaced sky fibres Optical Fibre OM2 Communication
98. of the future work Realistically to diligently modify the p3d code into full working order as the BASIS data reduction software Is an entire project on its own There is another software package not yet mentioned that would also work well and that is SAMI_tools Fogarty 2012 This is the package written for the SAMI instrument that as previously mentioned uses fibre bundles Hexabundles for their IFUs and so will closely match the data produced by BASIS An example of a SAMI data cube is given in Figure 4 2 2 where the display is that explained in the final stage of the data reduction procedure above Previous involvement on the SAMI instrument might make this a more favourable route to take 4 33 Pipeline The end goal would be to create a pipeline that takes in raw CCD spectra performs the necessary calibration and line fittings and outputs a data cube along with tabulated values for the galaxy s parameters Again it might be best to work alongside the SAMI team in using modifying their pipeline Bayfordbury Single object Integral Field Spectrograph BASIS 37 5 COMMISSIONING The commissioning of BASIS started in early February 2012 and the full reports can be found in Appendix D This was rather later than initially hoped but due to complications with the construction of BASIS explained in 3 the first working version of BASIS was that with the 2 inch Focus Mount The following explain the steps taken for each commissioning tas
99. on can be found by running on sky North South East West offsets with the IFU 5 4 Throughput Performing a flat field will give the fibre to fibre variations and when coupled with a standard star observation the overall throughput of the instrument will be found This has not been done yet as no standard star can be observed currently due to acquisition problems and it was also found recently that the central fibres are again moving in and out of the bundle At the end of Stage 5 in 3 7 it was mention that the central fibres had indeed worked loose Either this is due to the lubricant still being present despite using THF or the glue not being applied properly The latter would be the dominant cause but in any case it means a new IFU needs to be made This is why no flat field has been taken as the results would be in error beyond analysis Care will be taken when making the next version that the glue is applied sufficiently Bayfordbury Single object Integral Field Spectrograph BASIS 39 5 5 Initial observations Regardless of all the problems it was possible to get some initial observations the best being the 60s exposure of Jupiter Figure 5 5 1 and the 60s exposure of the Moon Figure 5 5 3 on the Fifth Commission Night Both of these observations were taken with brute force 1 e pointing the telescope manually and not applying any cooling to the CCD or taking any dark frames It was very close to the end of the night and the
100. on the 2 5m Nordic Optical Telescope NOT located in La Palma and with thanks to Prof Jesper Sollerman of Stockholm University an expert In supernovae and gamma ray bursts The resulting exposure from the NOT is given in Figure 7 4 2 which shows no such super nova feature as seen in Figure 7 4 1 This means one of two things 1 The super nova had already finished and had become too dim to see with the NOT or 2 The super nova feature is in fact an instrument ghost The former does not seem likely due to brightness of the feature in the CKT s Ha follow image and the NOT is a much more sensitive instrument This means it is most likely that the feature is caused by an instrument ghost a remnant from a previous exposure or an effect caused by the optics of the instrument set up An investigation was then carried out into what could have caused the ghost given by the following steps 1 Check other Ha follow up exposures for similar ghosts Ghosts detected but not near any of the galaxies Seem to be related to bright stars 2 Check multiple exposure of the clear filter to see if it is a filter effect Archival images from the CKT show similar ghosts still related to bright stars 3 Check previous and following exposure to NGC 3344 Ha follow up to see if ghost persists No ghost found 4 Check the CKT s observing log for the 03 02 2012 Found that the CKT observed another target between performing the Ho follow ups Target was NGC 3239
101. ope its overall field of view is arc minute 2 6 arc seconds per core There are four sky fibres displaced from the fibre bundle providing night sky spectra for data reduction and calibration purposes The fibre run has an overall length of 5m which is terminated at the other end with a 1 D fibre array positioned at the slit of the spectrograph The SBIG SGS permits a wavelength range of 4710 6830A and a resolution of TA Working with an overall instrument efficiency of 5 means that it is possible to achieve a signal to noise ratio of 10 for a 20min exposure of a 13mag arcsec source BASIS s performance not only allows for teaching the principles of experimental and observational integral field spectroscopy but also can bridge the divide between amateur astronomy and research science providing a first pass survey of 102 10 nearby galaxies in 100 nights Future observations and analysis of 50 galaxies will act as a proof of concept for the survey mode Table of Contents Bayfordbury Single object Integral Field Spectrograph BASIS 0 000 l ACKNOWLEDGMENTS p S uay iS ON gas estes a et aa ade es ea ead 2 ABSTRACT 00 ESA OA AA AO OA Salone tannin nda Glosa 3 y nititPlan uuu u uytu akuy s tu tas ae ee taa ee ae ee ee rere ae 5 1 1 A brief 6Verview ot the PLOT Caos roe sate soa acca rea kaa aca aan aca aan ae ack ar ela tanita 5 1 2 Project Lod Ba eee renee epee uu nds 7 1 3 inal Plan COMME uu unn E ES an p BO a ae ea Rees 7 2
102. ould as they were submitted at a low priority The exposures are done with a clear filter and for 300 seconds After checking the images it was decided to run follow up 600s exposures using the Ha filter to see the distribution across the galaxy After a couple of months 250 galaxies have been observed with 25 having Ha follow ups Figure 7 3 shows some examples of observed galaxies with their Ha counterparts The observations of all the galaxies came out rather well and show that this level of deep sky observation is possible from Bayfordbury Observatory using a Meade LX200 16 inch telescope Figure 7 3 clearly shows the Ha contribution for each galaxy where the colour map has been set to Rainbow black dark blue is low flux and bright red is high flux to bring out the features NGC 3344 and NGC 3846 s Ha counterparts reveal star forming regions within the spiral arms of their respective galaxies and NGC 2768 seems to exhibit weak extended Ha emission The scale of the Rainbow colour map is normalised to each exposure so the range can be seen by how dark the background is where the darker means the overall intensity is greater As time goes on the CKT will work through the rest of the galaxies on the list of 641 galaxies and will also perform Ha follow ups This catalogue of galaxy images is a great reference tool for Bayfordbury and having the Ha follow ups can lead to some interesting investigations that will be left for anoth
103. ph BASIS 21 Stage 1 Stage 2 There are various ways to make a fibre bundle the best is the Hexabundle used in SAMI Bland Hawthorn et al 2011 amp Croom et al 2011 Unfortunately there were no spare Hexabundles for use in this project so the first attempt at making a bundle was using heat shrink around a loose bundle of fibres This was done at the University of Sydney in July 2011 It was somewhat successful though could not give a circular bundle as the heat shrink undergoes a non uniform contraction The result is given in Figure 3 7 2 Here there are 61 fibres and the bundle has had a light polish to remove most dust imperfections The fibres were cleaved with a pair of pliers hence why some fibres have snapped end faces This was the best one of about ten made the others were a lot worse Reproducing this method is not viable and the need to have more fibres than required to produce a working 19 fibre bundle somewhere in the array is wasteful When done with only 19 fibres the bundle was misshapen beyond use Using heat shrink is then not an option Live Figure 3 7 2 Microscope image of heat shrink fibre bundle after polishing After trying the heat shrink method an investigation into making a fibre positioning substrate was undertaken It was this stage that took the longest 2months It was quickly discovered that making a fibre positioning substrate 1s not a trivial task and one that pushes the lim
104. posure and the bright blue part of the LED s spectrum clearly resolved Note how the top fibre is slightly blue is falling off just to the right of the CCD Slight rotation is shifted corresponding to fibre number 22 seen in the non level spectra with respect to the CCD The haziness and overlapping of spectra indicates that it is out of focus Sednen Lamp Spectrum ced a lt Figure 3 9 9 Sodium Lamp spectrum using the Andor lab dia e een single fibre spectrograph The x axis here is in wavelength Figure 3 9 8 Sodium Lamp spectrum from fibre number 80g from 200nm to 1100nm Line markers a and b 1 when exposed for 60s correspond to those in Figure 3 9 7 and have a reading of 585nm and 815nm respectively After inserting the 1 D array into the SGS Mount and illuminating the fibres with a near white LED source a ls exposure using the ST 7E was taken to see if anything could be detected the result of which is shown in Figure 3 9 6 Again this was better than expected with the fibres near focus only a slight rotation correction needed and a small shift in wavelength using the micrometer to get the bright blue part of the LED over to the left hand side note that the wavelength ascends from left to right on all spectrograph CCD exposures Making the required adjustments the result is shown in Figure 3 9 7 another Is exposure This time the spectra are clearly resolved and the peak of the LED s spec
105. proper flux calibration out of BASIS it would need to observe bright standard stars A full list of Bright Northern BVRI Standard Stars can be found at TASS 1997 120 stars and this catalogue will be used for flux calibrating BASIS Which standard star to observe will depend on the time of year and the conditions that night 1 e if the moon is up then find one that is near zenith but furthest from the moon The V mag My of these stars range from 5 2 to 10 5 with 60 of My lt 7 matching well with the performance of BASIS Even though it is no longer used as a standard star for flux calibration observing Vega would give a good idea of the performance of BASIS as it is a well studied star and has a My 0 7 2 Target galaxies It has been mentioned a few times now that the performance of BASIS would mean that to get a SNR of 10 needed for spectral analysis it would require a 20min exposure of a 13mag arcsec galaxy with an angular diameter of larcmin To find galaxies that would fit this bill the Third Reference of Bright Galaxies RC3 de Vaucouleurs et al 1991 was reduced using the online catalogue tool VizieR CDS 2012 The RC3 includes most parameters of the galaxy which meant that parameter constraints could be applied After downloading the entire catalogue the order of constraints was 1 Declination greater than 0 degrees Dec gt 0 V magnitude less than 15 My lt 15 Isophotal diameter up to a magnitude of 25
106. roblems with acquisition see 3 10 a 14 4 inch mount was needed so one was printed using the 3D printer see Figure 3 9 5 The SGS Mount was a bit more involved but it printed as required see Figure 3 9 3 The 1 D array housing slides inside the SGS Mount which attaches to the SGS via M3 nuts and bolts The reason for the oddly shaped M3 holes on the SGS Mount is to allow for rotation when connected It was discovered when removing the slit mask from the SBIG SGS at the start that the slit had a slight tilt to it 2 to allow better coupling with the diffraction grating Unsure on how much the 1 D array would need to be tilted if any the option was to allow for rotation The ability to slide the 1 D array in and out of the SGS Mount allowed for focussing the array with respect to the spectrograph The exact focal plane was unknown so again having the option to allow for adjustment was needed Bayfordbury Single object Integral Field Spectrograph BASIS 27 Figure 3 9 1 Drawing of the IFU Mount Relating to Figure 3 7 4 the fibre positioning substrate fits into the inset well at the top 20mm x 20mm The overall length from end to end is 60mm with the attachment plate exactly half way The holes are M3 holes and are used in the attachment of the IFU Mount to the Focus Mount Figure 3 9 3 Drawing of the SGS Mount The overall length is 62mm with a diameter of 12 5mm to fit the entrance aperture hole in the SBIG SGS The c
107. se similar science goals galactic kinematics stellar mass and population host halo mass and merger history As BASIS uses an IFU it would pursue similar science goals though sights will first be set on finding the instrument s limits to see which science goals are possible BASIS s IFU is a 37 element hexagonal grid micro lens array see Figure 1 out of which coupled optical fibres feed an SBIG SGS Originally the IFU was to be a 1x19 Hexabundle Bland Hawthorn et al 2011 but it was not possible to source one in time The other reason for changing to a micro lens array over the Hexabundle is that a single aperture of the micro lens array is a factor of 2 5 greater than that of a single aperture of the Hexabundle meaning that more light is collected per aperture if the micro lens array were to be used Considering the Meade LX200 LX200 from here on that will host BASIS has a 16 inch primary mirror the larger single aperture is the better choice Figure 2 Holmes amp SBIG 2001 SBIG SGS 66 Richards The SBIG SGS Figure 2 has a slit length of 6mm which is long for such a compact spectrograph This length is what constricts how many fibres can be used in this system as the fibres are positioned along the slit in a 1D array and therefore constricts how many elements can be used in the IFU The fibre to be used is Corning50 125CPS 50um core and 125um cladding which gives rise to a 5 125mm long 1D array compr
108. served it is then directly subtracted from each of the fibre bundle s spectra Flux calibration using a standard star This is done in near enough the same way as in photometric imaging where in this case a standard star is observed using one or multiple fibres in the IFU depending on the size of the IFU The standard star s spectrum is then integrated and compared to a calibrated measurement to get flux units for the pixel values When convolved with the throughput variations the response factor for each fibre can be found The data cube can now be created and is usually visualised with an overlay of the IFU s fibre positions such that selecting a fibre will reveal its spectrum as shown in Figure 4 2 2 The resulting data cube can then be used in the analysis of the spectra for purposes such as velocity fields and line ratios for BPT diagram placement psd Calar Alto 3 5m PMAS PPAK 2kx4k CCD Reduction Panel Master bias unta 00661 biass 0 s K a Trace mask Jruntd6_o07a0btrace fitg 2 s l e a Disp nask runas o0mib nasinqa Ln O A Fiber flat runtd6_0066ebincnb flat fits D A Mimo Display of extracted objects Comparison images 0 Tessa er 1312 1042 7 g5 m Log ues object slot 3 run146_00725b_oextr fits 3 right click to change the orientation Figure 4 2 1 A copy of Figure 1 1 3 for ease of reference p3d 2011 A screenshot of
109. sics Astronomy amp Mathematics Final Year Student Projects budget I appreciate all the support from my family and friends throughout the duration of this project in particular Caroline Richards and Nick Buttenshaw for their time in proofing this report ABSTRACT The art of integral field spectroscopy is one that has come on in leaps and bounds over the last decade and is really pushing our understanding of galaxy formation and evolution Of the 30 such instruments around the world all but one are on 2 meter class telescopes It is now possible to exploit recent advancements in small aperture telescopes lt 0 5m to enable an integral field spectrograph with a performance that allows taxonomy via optical emission line analysis HB to SII Here is presented a cheap and easily replicable integral field spectrograph for such use on small aperture telescopes in this case a Meade LX200 16 inch Schmidt Cassegrain LX200 a telescope that many institutions around the world have as part of their respective observatories The spectrograph used for this instrument is a Santa Barbara Instrument Group Self Guiding Spectrograph SBIG SGS which is again an instrument that many institutions around the world already use The integral field unit IFU is positioned at the Cassegrain focus of the telescope and consists of a 19 optical fibre bundle with each fibre having a 50um diameter core and a pitch of 250um Coupled with the f 10 beam from the telesc
110. t BASIS Bayfordbury Single object Integral Field Spectrograph BASIS 49 9 CONCLUSIONS Integral field spectroscopy on small aperture telescopes is a difficult task but is one that with the advent of BASIS has been proven to be possible The use of fibre optics coupled with the existing equipment of the Meade LX200 16 inch telescope and the SBIG SGS enables a relatively simple design which can be replicated for use at most observatories around the world that would be otherwise classed as a teaching observatory As this class of observatories accommodate undergraduate and or postgraduate teaching then the institution also has the option to offer the practical teaching of integral field spectroscopy something that has been very restricted before However given the expected performance of BASIS it is not limited to just being used as a teaching instrument but truly bridges the divide between amateur astronomy and scientific research The goal to obtain observations and analysis of at least 50 galaxies will act as the proof of concept for a full scale survey of 10 nearby galaxies over 100 clear nights translating to a period of a couple of years at Bayfordbury Observatory As an instrument the hoped for costing of lt 500 was achieved when utilising the existing telescope 20 000 and spectrograph 5 000 Even with this low cost the prototype version the one reported here still managed to do everything it was asked to do after a bit o
111. t on BPT diagram Each of these category definitions need to pass certain mathematical criteria to be classified and such formulae and definitions can be found in Table 2 2 2 which also show an example spectrum of each classification The flux value is obtained after fitting a Gaussian profile to each emission line It is accepted that meeting two of the criteria is enough to be classified Baldwin et al 1981 Which criteria are used is normally confined by the wavelength parameters of the observation i e the bandwidth of the spectrograph It holds though that the more criteria that are met the greater the accuracy of the classification The empirical cut offs shown in Figure 2 2 2 are a good way to get a first order classification The line ratios OI HB and NII Ho are the most commonly used ratios for classification as they are normally the brightest lines and fall within an optical bandwidth obtainable by most spectrographs To be able to observe these lines rest frame wavelengths HBA4861 OIIM4959 5007 NIM6548 6584 Ho 6563 a resolution of sInm is needed along with the ability to either obtain all of them in one observation bandwidth of 210nm or by patching together two observations bandwidth 40nm centred around OIIA4959 and HoA6563 respectively A signal to noise ratio of few at minimum is required to be able to get flux values The greater the number of spaxels the better with a realistic minimum being 19 hexagon wit
112. ta redUCHON uyu u ai a aa Saa uapa atau u Bata aa aksi as 34 4 2 Data reduction SOLOW ALG u l IA t ASA cee eee een 34 4 3 A ana aS IR Ia NR ANS 36 5 A O a 37 5 1 FOCUS A og uy uama S aa aa 37 5 2 Wavelength Sol ON u u uu alana TO at 38 5 3 Alignment position rotation a snssssnsseansnnnnnsnsssnnnnnssssssrsenneennnensssessssssasssaa 38 5 4 SEEN OU DN NG a e cana II O E u asss 38 5 5 It A uA Ee A RG uaS 39 6 Proposed Instrument Upg rades 42 T Q CODSOp OI S yD uu u au aaa aus umanaka aaa atas sa 43 7 1 Sandar d S ATS u u u Eat a Situa amaya ua SSS Sk ned ed 43 7 2 TE AV OE TANTES arun tata ath i aaah el Minoan Guna Konstan u 43 7 3 Target galaxy imaging A O 44 7 4 super nova WA di A E AS EA EA A A AAA A 46 S UU eS 48 9 CONCISO oia 49 10 Referentes AAA AE AS 50 11 Appendix A BASIS SNR Calculator 52 12 Appendix B BASIS Time line Chart 53 13 Appendix C BASIS Purchased Products List 54 14 Appendix D Commission Reports
113. tability of difficulties and problems that arise as the project develops This meant the instrument itself has changed and so the plan had to change The main differences were the decision to no longer use a fibre fed micro lens array as the IFU to keep the buffer on the fibres such that the IFU doubles in diameter but fill factor is lost and to only use 19 fibres in the bundle instead of the proposed 37 fibres Full details of these changes will be discussed within their respective sections mainly 3 3 The dates for the Report deadlines were met so too were the SPIE Important Dates which resulted in a future paper of this project being accepted for publication in July 2012 at the SPIE Astronomical Telescopes and Instrumentation conference 8 Richards 2 INTRODUCTION In attempting to answer the question Why do galaxies look and behave the way they do one must perform complex analysis of multi wavelength data from many sources It is because of this complexity that integral field spectroscopy the ability to simultaneously obtain a 2 D image of a source and the spectra from each spatial pixel was born and now flourishes 2 1 The case for integral field spectroscopy and its teaching The art of integral field spectroscopy is one that has come on in leaps and bounds over the last decade and is really pushing our understanding of galaxy formation and evolution It also overcomes the single aperture biases found in some of the world lead
114. tain target in a single observation In the field of Astronomy IFUs are primarily used to observe galaxies nebulae and Ho features and can come in a variety of array sizes depending on the specifications of the host telescope and science goals Current world leading IFUs include FLAMES Pasquini et al 2002 based at VLT GMOS Hook et al 2004 at Gemini North South SAURON Bacon et al 2001 at WHT and SPIRAL Sharp amp SPIRAL Team 2006 at AAT All of these are monolithic lenslet arrays that chase similar science goals galactic kinematics stellar mass and population host halo mass and merger history As BASIS uses an IFU it would pursue similar science goals though sights will first be set on finding the instrument s limits to see which science goals are possible BASIS s IFU is a 19 fibre bundle out of which coupled optical fibres feed an SBIG SGS Originally the IFU was to be a 1x19 Hexabundle Bland Hawthorn et al 2011 but it was not possible to source one in time so the closest match is built The SBIG SGS Figure 1 1 1 has a slit length of 6mm which is long for such a compact spectrograph This length is what constricts how many fibres can be used in this system as the fibres are positioned along the slit in a 1 D array and therefore constricts how many elements can be used in the IFU The fibre to be used is OM2 communication fibre SOum core 125um cladding and 250um buffer which gives rise to a 2 875mm long
115. the bandwidth due to the higher dispersion from a greater line density The decision to use the 150 I mm grating over the 600 I mm was made due to the science goals As mentioned in 2 2 more science can be achieved with a bandwidth that spans from the HBA4861 to SHA6731 emission lines This 2000A bandwidth is only achievable when using the 150 1 mm The trade off is lower resolution but with a 50um slit this is 7A which is valid for the proposed science goals in 2 2 It would be possible to switch to the 600 I mm grating to achieve a higher resolution 2A but it would mean having to take two exposures adjusting the central wavelength in between due to the shorter bandwidth 400A If the user only wanted to obtain one line or one set of lines 1 e HaA6563 then switching to the 600 l mm is valid but because the dispersion is greater less light per unit area the exposure would need to be longer to achieve the same signal to noise ratio A representation of the 150 I mm grating is given in Figure 3 4 3 which also contains expected performance that includes the expected performance of the SBIG SGS without the CCD 59 five optical surfaces each with 90 efficiency In Figure 3 4 3 the vertical red lines show the bandwidth 2120A which contains all desired emission lines for the science goals with some leeway either side in case of misjudgement with the central wavelength dispersion The solid blue line is a template emission line spe
116. the p3d software using the Potsdam Multi Aperture Spectrophotometer PMAS instrument as an example Here Is displayed the data cube so U 71 0 B t 0 001546 0 1878 404 Single P 300 m A minmax linear i e F ie 49420112129 X 19 5451 045 553544 91 A Minimum 0 Maximum 100 M Auto Scale Single Pixel R1 1 12 E Divide Std x 96 procedure 126 pgplot procedures registered DPUSER gt bufferl readfits Users fogarty Data SAMI Data Commissioning Jul2011 Reductions_Jani2 110703 ced2 cube 6390 6850 tp 03jul20029red fita DPUSER gt Mouse left change colormap Keys d measure distances g Caussfit I unlock position s c x source continuum delete spectral point r remove spec y Figure 4 2 2 Fogarty 2012 A screen shot of SAMI tools When the cursor is placed over one of the spaxels fibres then the spectrum for that fibre is displayed underneath 36 Richards Out of the three data reduction software packages mentioned above the one that would be the most suttable for BASIS is the p3d package see Figure 4 2 1 as mentioned in the initial plan 1 1 This is because with little effort maybe quite involved for an undergraduate it can be modified to accommodate any form of fibre fed IFU having a user interface that appears logical and friendly Due to time constraints the modification of the p3d has not been achieved and so it would have to be part
117. ti Aperture RC3 catalogue with the BASIS IFU overlaid to scale Spectrophotometer PMAS instrument as an example Bayfordbury Single object Integral Field Spectrograph BASIS 1 To reduce the data given by the ST 7E I intend to use pre written software called p3d Sandin et al 2010 see Figure 1 1 3 p3d is a general data reduction tool for fibre fed Integral Field Spectrographs IFSs It has a user friendly interface and includes a vast range of options for case specific data reduction Initial data reduction will be done manually by visually inspecting the spectra There will be an inevitable uncertainty on the expected performance of this instrument so decisions on exact science goals and the use of the instrument will be left for a future project My aim will be to carry out various observations to discover the limits of the instrument so viable science goals can be drawn To do this I will reduce various catalogues RC3 de Vaucouleurs et al 1991 NGC Dreyer amp Sinnott 1988 HyperLEDA Paturel et al 2003 and NED Schmitz et al 2011 to find suitable targets see Figure 1 1 4 12 Project Time Line For this project there will be four main stages Sourcing Assembly Commissioning and Science Limits Due to the nature of instrument building the dates assigned to the following tasks are deadlines meaning that as long as they are completed before this date the project is on track though I will endeavour
118. to a greater amount of light loss FRD is the inherent property of fibres that roots itself in the fact that fibres are by no means perfect light guides It is the property that describes the change in cone angle from the accepted light at the start of the fibre to the exited light at the end of the fibre see Figure 3 6 1 There are a number of factors that contribute to this mainly how clean and polished the end faces are external stress on the fibre 1 e glues clamps etc and the bend radius of the fibre minimum radius to prevent this is 10cm Optical fibres with a core size of 50 um class as multi mode fibre meaning that they can accept multiple modes of light rather than single mode fibre core size 8um that can only accept one mode of light A crude way to think of this is that multi mode fibres can accommodate multiple pathways that the light can travel on whereas the single mode fibre can only accommodate one pathway a straight line down the optical axis of the fibre and therefore has no acceptance cone but an acceptance column a cone with an extremely small acceptance angle The reason why 5Oum core fibre was chosen was due to the science goals creating a 50um slit at the SBIG SGS to achieve a resolution of 7A As most instrumental parameters mean that if the result from deciding a parameter is good there will be one that is forced to be bad the trade off here is the fill factor in the IFU This is the ratio of total collecting
119. trum is in roughly the correct location This exposure proves that the whole instrument is working and rather well at that Note how the top fibre is slightly blue shifted corresponding to fibre number 22 mentioned at the end of 3 8 The bottom fibre is slightly displaced in the spatial axis here which means that the corresponding fibre number 23 is not touching the adjacent fibre number 21 at the slit It appears that in the process of moving the other fibres down to allow room for the last fibre has caused the end fibre number 22 to jump up in the 1 D array There is little that can be done for it now but even though this is the case all of the spectra are still resolvable and able to be wavelength calibrated so all should be ok Bayfordbury Single object Integral Field Spectrograph BASIS 29 To check the wavelength calibration the IFU was illuminated using a Sodium lamp and compared to the same spectrum on an Andor single fibre lab spectrograph with a resolution of 1nm The results are shown in Figure 3 9 8 and Figure 3 9 9 This exposure was taken after adjusting the micrometer to capture these specific lines Most of these lines are outside of the wavelength range of that specified in Figure 3 4 3 but this was just a test to have a comparison The emission line markers a and b identify the corresponding lines in either exposure From the Andor spectrograph they have readings of 585nm and 815nm respectively which means that the spectra
120. ummary they are nearby z 10 such that their angular diameter matches well with the angular diameter of the IFU 1 and have a surface brightness across this diameter of 13mag arcsec to get the SNR needed for emission line analysis 32 Richards Because BASIS is designed for a small aperture telescope and uses existing equipment that many institutions around the world already have its function can extend into the teaching of integral field spectroscopy for undergraduates even postgraduates With large integral field spectroscopy surveys on the horizon including HETDEX Hill et al 2008 MaNGA Law et al 2012 and SAMI Croom et al 2011 it would seem fitting to start teaching integral field spectroscopy on an instrument that can produce the fundamentals of galaxy science and not take up time on what would be over subscribed large aperture telescopes The exact form of this teaching would be left to the discretion of the institution but practicals could be written for students to follow A good idea would be to have the student produce a velocity field of a galaxy something that would work well in addition to the standard undergraduate practical of using a small radio dish to get the rotation curve of the Milky Way by observing the HI 21cm line features of the spiral arms Before any of this is possible the biggest hurdle to overcome is the acquisition and guiding which should be solved with the recent change to an off axis IFU
121. x axis short to long wavelengths from left to right respectively 4700A to 6800A Fibre 1 Pixel Count 0 100 200 300 400 500 600 700 Pixel Location Figure 5 5 4 Spectrum of fibre number 1 from the 60s exposure of the moon Same description as Figure 5 5 2 4700A to 6800A Again note the O telluric absorption feature at pixel number 695 Bayfordbury Single object Integral Field Spectrograph BASIS 41 Figure 5 5 5 A vertical cut through the 60s Moon exposure as shown by the MaxImDL Graph tool The Graph window displays the intensity profile of the cut showing each fibre s spectrum 160000 140000 Counts 120000 100000 T w T T T 400 450 500 550 600 650 Wavelength nm File 1 JUPITE 1 Paged Figure 5 5 6 Dearden 1999 A 10s 100 x 1s exposure of Jupiter using an MS 125 spectrograph with an InstaSpec MkIV CCD Dotted black lines show the respective bandwidth of BASIS and the same O absorption feature is also labelled 42 Richards 6 PROPOSED INSTRUMENT UPGRADES As a new IFU needs to be built and other instrument changes are needed it is a good time to think about any improvements that can be made to the instrumentation to not only better the data but also the calibration and handling of the Instrument The proposed upgrades are IFU 1 Instead of using the fibre positioning substrate method for making the fibre bundle FC connectors

Bayfordbury Single-object Integral Field Spectrograph (BASIS)

Contents

Download Pdf Manuals

Related Search

Related Contents