Home

View/Open - Calhoun: The NPS

Contents

1. DAL M 1 Art tb tet an 44 q rite rey er vr Vect tf tai e 4 e ve 9 ves M Ry 2 rad UH MN MIN yes tery 4 44 074 4 Deis peut George E LE Kun H Hi dul Ree LIC 47 ah acies rd HE EHE UM ELM 24544 Uyay ote pig gts MI hat Mia is V esito teat s 4 4 314a tuat 27 HANH At n E certs 225 5 Mem pe 3 40 t few ta a rd 44147 CH STETTEN PEE um wa un ist recti prse APA tte DEM PAN AnI ei 5 1 po t Le M Toe DA Por vitet ba z oz qaum PR elt Nut HU ALPHA Lose ce pond Abed PERF EURER tante B ELS nen T E gnati f US sy sa es wr TEE FoU a PT utet Velen 438 40 69407 oe Ponda ba ape Lee met 472 4 p TM wp portr 15 a ne w 2 Ko Ra M peser fa 1
2. 91797 1 2 231 Mr P HERO EHE YO KEPPRA ET jap si iy vd galt s META mov Ady k t b PETES 9085 197 M bv 4 31 DELIS anta d A iyd dst H na an i i id er 3 TU eg dede S ai bak eat Ye XP Pa f L2 Dar AOR AND 92 8 3593 bal die ren D yv NA HR EHEN FETT Miren ke DOT T RINDE NAS rw 23 B y P wd A ha 1 58 js 196 Ra echa i dA afi D 25 4 A PETS Mii fe 25244520746 A 106 SS 0 Mz aes 27 PR St eb ape er TRATE NPS ARCHIVE 1997 12 VEM T TOE OTARA 4 4 6 et E r S DICTI de eram O 1 n ve DIETE NO a n Ml rias Vivis Ad Munoa XR Me NOM 1 5S uted AF REST FAR den SISTI 26 b iuri gi BR erc eC RE r Lern p y lt t 157254 TUTTI ME Uo wine ate ed mr al Medals tele anata
3. PE ahs Sy up Atte eo pue Rus n M un first DAS 47 y va e 4 PD MA punire y lw qi ay it PPS piv By gie w 9 9 qu t Li gt 4 osplasad por LI MNT EI TI e aa ETE E PLE 4 arr vege SITE Pits amp Grate E APO a s gt 6 a Pal m KU roti yore rh eri NM Peas campo ip 5 WD EAT e amp P nga ary vn p ana 1 wer dy iw vim M S mr E 0 oj erp ye nee FY vv H P 4 4 a Sette Py oO A TT nan Lda LEITET at 4 ri u rh e vy er ft 49 Me f MUEVE TET wry veh peu 2 97 442 ut m An serep ORE DN scare on y ug eret A Eng amp ort ch ey A dite vo t UE RH am WU gern 4 33 ENTIA ea Ree CH 44 40424 54 Ma a tu e lt Yu i594 2 ut 2 uo SP d a 4 4 xa 74 19
4. eg al ow tg PALA POOL 7844497 Moser rra pe Te Ai po n ona viget pnm ds Y fast 1 4446 a e Trotz a Aw rt imita t JU de ML Ee RUE HE od ia inte gende ie at aA ai DEIO guter o DD gt Du od PTS y pata 11 pos y AP erg TTO Vista 2422 be 274 pee 1 16974747 2 JURE ers 1 eto 4 eose IPP rA en nod terre purr BI fep 2457 d dy 4 av Yee en p IN pete A o vp ren 70534202747 a wt 4 795 H ine 15 h UN y d ELE 5 ug fan Ver Ba ta eire MS VALETE RR E 408 t mts ecm a da ro Vr ag 2 HV 15 a opie E vote Say 4505 pe e mE C mr E DM tte 116 e fes du eil us PREA 4 PA biu nran
5. jie 5 AU en deci LLLI 4 HPN d php 1109 TP CE ELEI P LET KA ie sete 4 2361 n i f Jur kee tefl Ley 1 E ui np et de CETT LT 1 IEEE 1 te rabo e de tr TD reU Un L i 1 7 IM i f liat tpl 1 t 4 deep A 92449 1 f ote 4 oh dered Mr HIFI y panel OON T tra f hd t rab y 1624 a 4 4 AS wk lipa nt ue 1 sw aL d e Fedre cade bel at arto 1 LL 4 t r 7 p MPO Sardo ed adera Ar tel E DALLAS OR Kar ERE Y Fayard Late Pr Ea Po be inel d At i gt x p Hal thee RAPE SQUID Um AN t 1 pare vi 4 C e D Ur 4 Hf fed 2 t hh eti t e v par vue ab a d wa
6. pea w UN 1 04 49 nds ves e att T mos oa 4 K she
7. nipp ipl Ag A KO H PTS AA Par dr A E bee DOE hn 25 1 bate M Op La 5 bebe e p Mu i DL as et A fx rd Pateat 4 ta i MA t fi Ab iii fg Gh ASH 1 1 i 4 1 Im TOT bal y tiii INI A e of 4 aS aT AU O TN Qr i P efe doo eiu mA etl GE O TIO e si Apte 46 Op DEDE util EK h 4 pet 1 re in lvl if peed 1 cop r Bathe hi hi PAZ LUPA Je lfl lir p b K etn 4 vir r La 4 t abs dedito rita h tom ata set Um UA w EN mre KEH st t agn D eit a Boma nahen vf IO H ac aee LO tot Cr ACORN eter ei ge 7i mU y boronh b P ar 454 Yale 1 1 4 my aa s gd on s D TE NE 1 eg i D 4 rns Da t ra T O TEEN sib PON 14 65 mm t y CH DIN i 4 4 1 ae ist A A 5
8. s ange B OU A e PI TP de 9 10960 9 4 de An muse tin EAS RAM to mun p ord ieu i Re rt en en 57 s i r am pen S ren se gu it A puree 97 4 174 CLT PLE AL ud v PL Y mr nomm gt IUS ed KIESA are 3 TS AH UI 2149 u dedo T t 4 i elu ran to yo a8 a ra ies A 19 EHRT 4447204 ae t aZ 4 18 7745 94 wethepe e MI CD aa 060 479627 pry t tt 4447 14 A US A A 2 S UL P pun ty 4 eto nt shed et e v m 7 ig V tran s et e 05 ra H 849284 Qe 9 4 s lt are Lie ee 240456 7 2 aa pat 91 yi JA Peg T 469 ott 2 27 aqo su wp 7161 nu 44 A FT n E p e AQ re 244 opti 424 4 PP I qb 411 4144 rt RER f X PANI puta o 84724 01 4 SEIT LUE port inte bd ipm ni
9. Treue ik al Bd 5 Drm KI E tab te 499 9 Are sql ala ISA PN PRETI EN eds tad af ward OPER hn ied aad 2 2 2 4 9 4 ER rs 01216 1 70 pA ML TRI Met AAA f gt 7 Xm aha tdt 01 00 9 53 9 vA EI les aye V Tia E 42 1 98 yit 582 Y tura hos nr Nas nn T sated UTE 24 6 Lm 334132 AA ON Re o wa Sahara thie or Rey na Ele ved isos AA LJ 0430 rut y 4 Rete diit 2235 in 487 NIMM oy TIN 4 ERU 22 A nonum 2 gi Hess vhi RE D nana yen CERTE p por a IA AR4 8 1 y OB etate z a 57 pies JH Rie eer Wert E Eher PP SEVERE ir ee JL LAB 91607 JM ETC Ko nO 4 1961499 as ARAM 4 UIT ERI 3 4 3210 431 y x 4 hoe nei REA a Les Ay 1 de Per De
10. d 2 LITTLE ih d rbv ie 91 ote 94261 gerne up es u s a P m oid atis P m Nd ES 54 17 4 best YA Pte ET 7 fen 72 r y mn Es f igh ro 7 444 Ine e n TI w HET rool Easoko ia MILI 9 af EL E d potop we 44 a kes te 1 aba plt y 1 oo D fen MA Jj UE 5454 AM ix Tu 2 a taty bie my a gt aoi ery ites i v d late T ELI KD yr ra q Pareto 4 4 BM 26 t h 4 we green 5 m Ne um a P gestae qe pP opt Ye 1 1 Vrat e mm D Ae TUN e 99 ctete Ih E ILI das 19 MEE nu PAP tue iib CS 1 y rra I This arte tyta Ap fee Fay Pe mm 2235 4 eeu EXT e i a prop gic S dh om tg ak 4 AUREUM EL nr JE ERE fro t 71919 beh eM at Te Ate ds 14799479 a ew M u v
11. E xw d UNIV Ab s pt slas 14 24 EN 1 amp pem LN ah 1 9 faso 1742 Dd n 22 99 59725 BUNT MN ymo nt D 44 tos 950 vi 9 7 t snm m erqza devo 2 n 2 la MISI Ps 3 394 n 494 822 E et vpi IF f 4 3 3 Dede wu 4 PP re Te te do u OTE war rots E ui Mx u Pr LE 1476 om 09 n n v sh own sivit At ar NR 7 9 PES dd VLA Aci 1 By 1 Mt RR LI Sm te ne wir Aty m 7501 se yg si a 5 us 14 4 t ph fh sayy tovt e usn ld ara i O Vy vit b Ipan my UNA alge y 2 En O 294 95992 y e Him AA were MAS t 17 aa 56667 C MEE e nr ah 2p uf quum teste at y wre y PYA pent rage SEE op nse
12. STO EC ELLE NER otare Cyan ek L D waren y 44 10 oe a EIE Ptr i ater rite Dis 744 en M M a gt Mo EPA pref t 4 Ayr Ni TER M 1 44 1 en Pee 21 2 loo M wt OFF a AT O 0 1 Jew 62444 MEL GARA a od P ur y z sas D gr 1 6 ny site aly E 4 TT el bake GENIAL 4 1 4 t S 1 ser uo v dren YT Per dl Vo rS psec UM n d v dt ater m IT tu P Na tu v 4 die E GANT en ur nat Maen 1 tent BE EN S ni ers P b M USE diniya pos dd aha yes EM UR ka CY M V pi TN nu neve nd s try de MEN ou i afe Au LE te L Se t
13. jE Je 4 ie 7 pip bo l via g 56 K 4 e 4 DOM Mat 2 wl iA e 001 ja Ca yapa ist AT ei tol E 411 At fla fh J 4 f hath t vic a 8117 n las Lem 1 y dt 4 E fee tye ie k be amp oat FeO a MC 8 22 Se 4 Li bh faa Lay bh x deg dor e aha Who meade ie sa ger y emba t pos 1 1h CP E ALIR pigwe yadta So aye 4 es aby La Mel 244 aft piat wad Vahl i ZI TE UN EL a r rer ee ka He pots ACN Wu Tera re Fe blat rotot oth PUT LI 4 ESL 4 ha 11 UR Ea na end 4 te ML last Fr gt a t Bal Mi LAT H PA aea dorm p sts sre 7 ee kont E 2 di 4 05 we d auf te ee dE ns ar ih b 25517 tod Fox or hanl teil i at pi rda PA te i E 2 AN te t inet a old h fb we te M y 8a Mine he 444
14. p uf d Uo V ba QA HC PRA Caefar y IS atto tone aret USA Tg Urt M vin t dee 9 9749104 y Hv asah CAG tht os t E i 9 bj avi wur de eni MC Pero PhS YR yt 8 747 do wn KO et E Po gt T 4 44 15 IH rs Uh as Ja yis Veturia 116 sas aom r Y pes d a p x s we Fe laad ro e us tun ih NEUEN TEN ete 477 VM mE NITE en ix pte i t de ito n k pera QU oy ts Jere E Quot there Hia TAL yen MEE uber deter ne the tate ty A VIR eine SNe ee grutas s Serv LO AINE ede VERS dL RETE eren SO e 49 uite Fr ey dee PRA A nu cd In A ats V t P vot al dahin MI E IRL Dear oo 1 Fe apn wo ieee Mele ee 12 i 7 Ya et eee HH sin TUI
15. puc 94 TERA Ah M NOE yuan A EA M j 1 SKS m y nera a ade INE 1 vi gt y sy ehr 27 T 4 4 zu aga 714 Mp 1 P 1 e s d it Al i EO 474 4 t Du k A EE Ev da tue Eide a Ie ph tnt ve ny x pe 4 MS v4 PT 4474 Arap 479000649 ran E Vert PARET 4 d i TH HE ES 4 4 e pt cut eta pi at 1 fr LE eee ra A AL 4 23 up le Her rete E s PTS Coe ae ge te 6 y re MET AAA a pre qom RIA E pd 44 DAN E Dn ysi Sets 4046 A rera n 8414 MEM 2547 1 65 14 2 14 119 N d n M 4 J ET 4 S AR er 47194 thue 142 veh ani eine PIN at m ye 6 16 424 CS O Hy WEI YOT e o Ier I Nee 4 4 47 2 T 1 4 4 UN gras qao E NES TR 47 EAT d vt EBD E pe rata be spt tune m ee 1 49 Su PA Ha 4 44524 74 10 44
16. 474 5 es CER Tu n et 940 ott 214 807 AEN M ye mentiti M tes n lid p cC iei d e piste x 2 4 epa sadi grat gh nn 55 2 qui trn 54192222629 Boy ge a 4 pte Mega ot E Vi eq Pt 25966174684 oa OS E RUE o a EPG CHART TA ER EHRE RR Eder E EA 4 i ROE dete at PY 6 te 44444685 134 x Vase ee Ea weu ath SALES 1279 irae 455 H x Naas 2 2 pute i viuit 13 Im LL m S U oos ug M TES fe Uu REA bronce A een wu io y ott v d aad ety AN d B ft 4 et UH ut st DAC FILII Maia EPUM af Je ee pee BEIM HER vig pem 2 g j amp kaqa a Sri ay LOS TEES seta ote aer 31097 wr bet e Mi b 70097 a se P hth Coda ba bah a deed ITU Mal ia a va et rtl r an erige sagi
17. ES MT iar S DTE 4 L y Wr TA VOIE A P oy an 00 fu jus H 1 tyres go ts js a Ze RA E ur 4 1 n rit E 4 tat PAAA ten 2 2 4 i CUNT prance tent ise DL vt fue por du T m 70429047 45 PM Y vate DEM ne MEI 91484 SUA vA Cry RARA SR RE gei ne N ia N T t 6 E eye mad er wur ILE S We De 714005958 ca Macc LA E un Pup WU opu US poset tat 4 Er TEN Bo bade 79 4441 te j LIBR I gt i TGRADU HOOL CONT mas eO Y 93942 01901 NAVAL POSTGRADUATE SCHOOL Monterey California THESIS DESIGN CONSTRUCTION AND OPERATION OF THE NAVAL POSTGRADUATE SCHOOL S ULTRAVIOLET IMAGING SPECTROMETER NUVIS by Todd Hooks December 1997 Thesis Advisor David D Cleary Approved for public release distribution is unlimited REPORT DOCUMENTATION PAGE Public 1 reporting burden for this collection of informauon 15 estimated to average hour per response including the time for reviewing instruction searching existing data sources gathering and maintaining the data nee
18. i tty 50 x Se fo 6 4 8921449 373 S 2 4 i n Vor jk 14 le T eos ta a E 5 y gt 4 d E I i M er ut gt LL P Pee aed a 4 Fou m 44 we 1 gt u 4 tarsi 2 TENES x qof PN 1327401 252 wi at 24 24715 t rtd um Fed Yar tade n ELI 9 9 9 1 a5 Sw 2473 4 j la ees 11 Py 172 mt lt y 1 Su 1 1 7 5 A Pats 40451
19. upper left corner the connectors are the OS 2 mouse port and the parallel printer port Starting in the lower left are DB 9 and com2 DB 25 both RS 232 serial ports The next connector was the VGA connector seen with the monitor cable attached Following this was a mini 68 SCSI 3 connector of the Adaptec 2940 UW The IMAQ card connectors were next followed by the PIO 24 and DAC 02 connectors can be seen there were no available expansion bays left This has caused some inconvenience by preventing the installation of a LAN card which would have been used for data transfers Figure 28 Computer Connectors Since each hypercube can take up to 109 Mbytes data storage was critical This was the reason for the 9 1 Gbyte hard drive the largest available drive that would fit in the computer But even with this large drive there would not be enough storage for extended field use The solution was to use an external storage device 46 envisioned that for field use data would be taken during day and stored on hard drive During the night the data would then be transferred to an external storage device After review of the available external SCSI tape and disk drives I purchased an 8 Gbyte HP SureStore DAT SCSI tape drive and an Iomega Gbyte SCSI drive The tape drive would be used to store all the data while the JAZ would be used to transport a limited amount of data from computer to computer No
20. SNOISNINIO 0313 2345 ISIMYIHLO STINA SALON L ALLLNYND VIN 31 95 16 8 8 31v0 p34 sSy ON OMA Hd 1 AN Mara Ajquias sy 103314 Bugejoy TILIL Figure 4 Scanning Mirror Assembly 11 path The dimensions were calculated in same manner as baffles and scanning mirror size except that the optical path lengths for the center and edges were calculated in detail The result was a window with the optical axis offset from the geometrical center The minimum height of the window was calculated to be 88 mm The minimum width was determined to be 68 mm 40 mm from the optical axis toward the telescope mirror and 28 mm from the optical axis away from the telescope mirror Five millimeters were added to both the height and the width The final dimensions were 93 mm by 73 mm 43mm and 30 mm with respect to the optical axis Immediately outside the window was a 3 2 mm by 3 2 TITLE mirror Back EI JOB Code PROJECT ENG J Code DWG No Mirror Back DATE 9 5 97 SCALE 51 Rev 2 MATERIAL 6061 Aluminum QUANTITY 1 SECTION A A N N 4 n mo N N AS ADA oor m EE zi 0 100 1007 0 100 4 40 0 260 DP Figure 5 Scanning Mirror Holder ENTRANCE WINDOW AND FILTER 1 Entr
21. 35 2 Image Intensifier 38 FRAME CAPTURE GAIDEN T eee ETE 38 DIGITAL UO CARD o RERO D w 41 D DIGITAL TO ANALOG CONVERTER 4 COMPUTER rc BIS ees s ss 46 CABLES AND WIRING HARNESS ss m 49 1 AD 49 2 TT E 51 a 53 eae ee 30227 5 B BASIC AYU EEE 722 53 DEANDDYNAMIE LINK LIBRARIES 0 DTE 55 Vili TDA NAMIC EINKTIBRARYEORTDE 61 EI 20210 EEG 15 rc 75 Figure SC LIST OF FIGURES Optical Path in NUVIS 1 filter window 2 scanning mirror housing 3 anning mirror 4 baffle 5 baffle 6 telescope mirror 7 telescope mirror housing 8 slit 9 diffraction grating 10 diffraction grating holder 11 image intensified camera After MacMannis 1997 3 Platine 7 A 9 SCAMMM Mirror AssembIy 11 Eeue o S nune WOR Holder o e edet NU Cera e ERR rex as 12 Mato O Piran c Ay dow Drawing 200 22 222 90 1 15 Eeue 7 Solar I radiance at Earth s Surface 2 15 Ime 5 Transmission Curve tor UG 5 Filter 16 Transmission Curve for 16 10 Imag
22. buffer passed in pBuffer error imgSequenceSetup sessionID numRemainder void buffer skipArray TRUE FALSE return O j Function SequenceClose Parameters a pointer to a pointer to interfaceID Return returns zero LONG WINAPI SequenceClose LONG numArguments PULONG ppBuffer LONG error PULONG ppBuffer 0 ULONG interfaceID pInterfaceID close this interface free all resources error 1mgClose interfaceID TRUE return Q fEnd Nuvis cpp The following file Nuvis def contains the functions defined in Nuvis cpp 0 0 0 313 3 3 3191313 919 90313 392393332933193393339333331333933339933 69 Nuvis def The DEF file for the Nuvis DLL DLL 9319191919191919191313191919 91919 91919 9191929 91919191913191919191919 91919191919191919 19 9 9 LIBRARY nuvis CODE PRELOAD MOVEABLE DISCARDABLE DATA PRELOAD SINGLE EXPORTS The names of the DLL functions Test GrabOpen GrabImage GrabClose SnapOpen Snaplmage SnapClose SequenceOpen Sequencelmage 70 SequenceClose End Nuvis def 71 T2 LIST
23. Bos 4 DOE tS ig tA pon PEN UN ef 9 steil Lg 11 9 5 n T er I ira ree AN Dur TERRI d RA p Le 1 LN insit rei Marita a ILLE Y Lun hn ELLES e fv 4 CH BC H an i x T RETRO NP abe 136 at neh Dol ope we LI oe bad i N a F Heim vn at 4 eri 4 Pi wri ee 2 qn D Da VAPEUR y Umi A 14 4 cR TM des E iu Sn dye hares qth promisit 476 t MNA 932 Segun Puis eA T eno tta ea ey MOTE 29 un PLAT ud CUTS attis q 3126 ut Ima tp IUS SN p 41406 3 TAA fa M gt u ieee mta ase au ns ee AD IRE a m 4 y atra le HH vo 152 q qa te 4 La AN a i quq O ens bats re Deu ui Hen ag A ez Ar mt sno Vibe ba wt AS Tea
24. Keithley Metrabyte 1991 D DIGITAL TO ANALOG CONVERTER In order to remotely control the image intensifier gain and in conjunction with the shutter control control the CCD camera exposure a DAC was needed The first choice for this card was the National Instrument card discussed above But with the selection of the PIO 24 as the digital card the addition of the National Instrument PC AO 2DC card with 16 I O lines and two 12 bit DAC taxed the available addresses on the computer bus As a result I again chose Keithley Metrabyte card the DAC 02 The DAC 02 was an ISA EISA card with the same type of installation as described above for the PIO 24 except there was no option for an IRQ setting The base address for the DAC 02 card was Hex 390 This address corresponds to the lower byte of channel zero 44 The DAC 02 provided a two channel 12 bit DAC with output ranges of 0 10 V 0 5 V 5 10 V and a 4 20 mA current loop The output was selected by jumpering the reference voltages on the cards DB 25 connector as shown in Figure 25 Table 4 below shows the jumper pins for the desired output ranges s s de 14 10 V Reference 03 15 5 V Reference 04 16 DAC 1 Reference Input 05 17 DAC 1 Bipolar Output P 06 18 DAC 1 Unipolar Output ud 07 19 DAC 1 4 to 20 mA Output Ie 08 20 10 V Reference 09 21 5 V Heference 10 22 DAC 0 Reference Input 1 23 DAC 0 Bipolar Output 12 24 DAC 0 Uni
25. PS aber ET st 427 re EX M et 1 9 i K TAN megio oroad 4 ED 7 fa ar dp A A OPUS dy 1111114 parte 794442971784 18742 P qe M Co TAS Cu Re M 1 t n pi Tus TER Ar un 2 22 L ghee A NOE VUE ER 42 OE 14 ts ee Br ope v HE en y MAP TEC PEO E 3 20 249 94 Boa pista Pd wwe EM 2 4 Shia Mead bel EOS iiio i 9 ASE 2x ET DUE Ta teg EE e e Em I eU e ge PU 82525 m 7 2 421162744474 7 NT ANE i ag mi e See 4 276225 que 35 1 4 04 lt mes e se j H td T EH u j h D 616214 etit 446 4175 5 pt 1 4 Mot ane 191 pi ban 2 TuS An 4 UL To eid fp p ease 42 42 4 AS A it sigh shares tn Ai 79 gig AS 5 NN Pid sees 49 te Y 4 4 IO ELI 92445011 uni inf Los 74 vr eum
26. and the telescope mirror The baffles were located 317 5 mm and 381 mm from the rear edge of base plate and centered along optical axis as shown in Figure 1 The size of baffles are calculated by using the IFOVs Since the baffles were not intended to be the limiting aperture 10 mm was added to each dimension For the placed at 317 5 mm the horizontal and vertical dimensions are 61 mm and 76 mm respectively For the baffle set at 381 mm the horizontal and vertical dimensions were 61 mm and 83 mm respectively The edge facing the telescope mirror was tapered at 30 degrees The edge facing the scanning mirror had no tapper which created a knife edge to reduce reflection Figure 2 is the drawing for the baffle nearest the telescope E SCANNING MIRROR ASSEMBLY The scanning mirror assembly consists of a number of separate components 1 the scanning mirror 2 the scanning mirror housing 3 the stepping motor and 4 the absolute encoder The stepping motor and absolute encoder are covered in Chapter 3 1 Scanning Mirror The size of the scanning mirror was calculated using a process that was similar to that used for the baffles but unlike the baffle calculations the angle of rotation of the scanning mirror had to be taken into account The zero angle for the scanning mirror was at 45 degrees from the telescope mirror optical axis seen in Figure 1 The scanning arc was 10 degrees 5 degrees to either side of the zero
27. np diy oh 3 1 n 5 y pistan Pia nen rna arn 2 als xj a 4 5 21 unt Um LI ry she wills he wy t 5 60 band 277 W Pn e 4 1434 e 24 42 7 hel e BA 16222 Kr i6 MU bison 1 rte Be a DIM 1 14 ft 200 0 3 2 26 De A Us 4 4 AX rol ht Geared VT I estelas ote y spe 4 pi 2 1 2 7 He 241 19 got ef ri en 4 LI uy enn nre aat h y i y j4 Mi r MA m bi 1 POE HZ i ot r1 T hos H 2 4 i y h gt MESSA inet 4 DU TE rover i ns eleg wA f 52 Ue list y da 3165 ts A 2 7 Loa r is l 3 nw p Kd A t 127 Pe k vt PL ds 4 FR Fiat S a V p on 6
28. pee Eu RC I MPO EM e NC ECT HE 21 dde queso 4 ef X 7 ry 446456 147 24 nin 1 5 in Vb arg d reme et dote Ergo 4 9 Monster rm siete Wat ont pet Los pat D 24624498246 te ans UL UL ye HI E pM 1241124 2585 eta rimas st 41947 47474427747 ver III LOC etait Peta qa 4 te n ton ow 2 41 ae q PP gamas 16 2214429 LOG 647 wt Ra pim E M LARA HEUT AT prr 97 fan l 12 4 En p P tn ju 4 77 ment 1743748 gies P Pg peo eye 934 a pte t CLOS Re TS HOME En YA 194 pautas mem p 4 7 t ye 2 To DELL 1 u f uni er t u MER A ata wate IW qe MSti heo ea t 15951 4 s d m p ultri ap V tp versos ater 23 TOME ater ey 2221 141284475484 24444 UA ERIN gt 9 IH eT x TETE mn die ag d 0o g Dre m a ht 9 4 M ac A
29. was A 15 at 632 8 nm SORL also provided the telescope mirror housing With the mirror mounted in the housing the optical axis was 3 5 inches above the top of the base plate Figure 1 shows the telescope mirror along with the other optical components SLIT The spectrometer slit dimensions were determined by the focal length of the diffraction grating and the size of the camera s CCD pixels as projected on the image intensifier input window The width of the slit was arbitrarily decided to be three CCD pixels which was approximately 60 um this point in NUSIS s design there was way to determine whether a 60 um slit would produce the desired image on intensifier photocathode or simply limit the intensity of the light A quick look at the geometry showed that the width of the slit would be the limiting factor in determining the sizes of the baffles scanning mirror and entrance window Therefore it was decided to make the slit 90 um wide or 4 5 pixels on the image intensifier Next the height of the slit had to be determined The diffraction grating provided a flat field focus in the 300 400 nm spectral range over a distance of 25 mm Since the height of the camera s CCD as projected on the photocathode was 17 7 mm anything outside this would be lost With a maximum usable height of 17 7 mm we chose the slit height to be 15 mm which allows for slight misalignments A search found no commercially available slits that met both th
30. 454 See iiec ia 7t NT amp n um DEL ia t LS LI LI L 92 53 a 47 meas ache AY PR nao D T Dus VPE i 141 so A t rtt 2S y Ag 44 poo 3 HOC C x 660 1 at on 1 a L P m s gt 254 1 A se t ty JA m T 5 an 41 s 4 1 t 0 E 9 M 1 2 wee D 1 5 d h J z 2 netos 4 n te lt 5 1 4 E e 24 2 A S 1 1 T s 4 t Far 15 4 te n 4 7 4 4 4 B 2 L 4 et 7 i j x E TOME Dou Wl EP j c n ry m Ae mu oP i rd t 44 d 24 E fon 4 e 2 i 2 Le f 2 TCR D w t lt N 4 47 j t M y q s EM ET ME Yo y eae na d t 4 TT f ii yare dogo a 4 2 ate 454 yy 22 15 6 II PED In
31. GA d Ed er eh MES ee N 7 NS T 225493 EAS N RT HE irme ces wo TO 5 ST a cn s ebat Qui aed ner die AL 3 PRU um rur ry x ne yy Me rtr Tons Finis pe onard us pre kuya arva Te P R 4 utr 5 ad promette pue bed Ue atu ASA s 2 iti CRM ui 4 iG SWR RA p PM M Y poU aestate spletcar in Ma rrt Mos ot echt gta 47 au ix t PRETI Mah i SENI A Teen zen s kor n um pos ctae 2191 oy hg ew AR 64149986402 OA PO TES reeht el F Pf y pi Dee MATE Magis HM cp te w w 5 ba ok Les Dap TS 4 d 7 A dvds dhe vie od php oranges dod ane ES Arete cae se be La yey e t atte sr n path ls e ni gat ee mah on 2 any 44 gt LR elt ied
32. ati 4 A 44 27 Hp SR PSP MENS a313 194 gt H elas AN XE A ga PA RAR LU TOPAY hfe in 3 2245 7 DILE TUS 2 7 st as e iX etn d 447 54 5 rt at AR TN 1 4 ee 4 Y qt M LER LI Du 4105 5 4 or k ER at 57 at aks 235 nr 14 ti unit Mf S 4 E
33. be more difficult finding a portable computer than I had expected I finally found a case shown in Figure 26 distributed by a company called Case Depot NES 7 j kil 5 Ex A 0 Figure 26 NUVIS Control Computer 46 The case included a 10 inch Sony monitor and keyboard that folded up covering the face of the computer The case had two 5 25 inch drive bays six expansion slots and a 230 Watt power supply Along with the computer I purchased a hardened shipping container shown in Figure 27 UM ME LL 4 aa cm cM S fo x 57 EN m ee 4 3 e 7 Figure 27 Computer Cargo Case I chose to install an ASUS PSST2P4 motherboard with a Pentium P5 200 processor in the portable case This motherboard had an AT footprint that would fit in the case and it was the best rated Pentium motherboard on the market The motherboard would accept up to 512 Mbyte of RAM but due to the physical limitations of the case only 128 Mbyte would fit The hard drive was a 9 1 Gbyte Ultra Wide SCSI Quantum drive The SCSI controller was a Adaptec 2940 UW The card was an ATI Graphics Xpression with 2 Mbyte video RAM A standard 3 5 inch drive and a 20x Toshiba CD ROM were installed in the drive bays The computer was assembled and tested n about week 47 Figure 28 shows connectors for control computer Starting in
34. eft D System pe ond kasa c fos ey TEES aPptwri y p Az S une amp doy 4442 1909 nma as pos SMe t v 1 22 ree E DIUI 0 NE DI on i p o 4 ta 4 77 nan erya tat nm SAN y ET v Put eed eM P ASP UR aeris vv un dM MUTA HE Hi 2 m 4 eu veg 4 men pod 2 5 Vie bna H Peay 1 61 Paar He yt FHN hs 1 Ni PLI Pone sep be Pip Eoi f SA raa 2 AS nie d 4 iod Ud RT d A HE nA od p LES k cry 1029 96 VA Set Lan Pr LB 7 42 Ju 11 6 151 t te Ware tatty pt LI SEO AD PA ay 10 Sete a 2 we wos Aera gates sat eren pass mt mm weg atem teo pet m 2774 p DINE vl 7424 sqr yo 6 3 op rere Read y tz oye e Paar Ty 11941196 AT uoi p OU Sara Be yr ttal e esi oM eh T te
35. handling has been incorporated in the following functions All functions operate Hinclude windows h define _NIWIN include lt iostream h gt include niimaq h include nitypes h Declare the DLL finctions prototypes long Test int int amp DLLEntryPoint The entry point of the DLL BOOL WINAPI DIIEntryPoint HINSTANCE hDLL DWORD dwReason LPVOID Reserved switch dwReason case DLL_PROCESS_ATTACH 61 break case DLL PROCESS DETACH break return TRUE Test Function to determining if IDL calling convention and this function are operating as expected LONG WINAPI Test LONG numArguments PULONG ppBuffer PULONG pnumFrames ppBuffer 0 ULONG numFrames pnumFrames PBYTE pSkip PBYTE ppBuffer 1 PBYTE pBufferArray PBYTE ppBufferf 2 UINT for i 0 1 lt numFrames i Doki i pSkip j for i 0 1 lt numFrames 1 pBufferArray 15 pBufferArray return 0 62 Function GrabOpen Parameters pointer to a pointer to sessionID pointer to pointer to interfaceID Return returns Zero LONG WINAPI GrabOpen LONG numArguments PULONG ppBuffer ULONG interfaceID ULONG sessionID LONG PULONG pSessionID ppBu
36. mirror holder needed to be balanced The motor was replaced with a AM23 210 3 motor with an incremental encoder option providing 210 oz in torque The new motor had a 40 percent torque increase and it added an incremental encoder Without the incremental encoder step indexing was determined from the step pulses ordered by the stepping motor controller With the incremental encoder the step indexing came from the actual number of steps that the motor turned thus creating a close loop control This eliminated step slippage caused by the high torque load of the scanning mirror when scanning vertically B STEPPING MOTOR CONTROLLER DRIVER The stepping motor controller driver is shown Figure 19 and like the stepping motor was purchased from AMS The model was a MAX 410 with incremental encoder option It was a stand alone driver that interfaces to a PC or dumb terminal through a RS 422 to RS 232 proprietary adapter part number SIN 8A An RJ 45 connector with an 8 lead cable 29 connected adapter to controller The controller driver used standard 110V AC line power The controller input and outputs include a five line connector for the stepping motor a RJ 45 connector for connecting additional controllers in series an incremental encoder input option and five digital input output lines or triggers J2 PARTY LINE SERIAL OUTPUT TO NEXT AXIS OR TERMINATE 31 PARTY LINE SERIAL JS AUXILIARY READY INDICATOR LIGHT INPUT FROM SIN
37. to be custom made therefore simulations will be needed to provide a manufacture with the required transmission curve Even though the Visual Basic interface provides a good control for NUVIS I recommend that a robust multi windowed application be either developed in house or be contracted out The only way to take full advantage of the hardware capabilities is through the C or language At a later date the current camera should be replaced with a 10 bit or greater grayscale camera This will also require the replacement of the frame capture card Therefore a camera with a digital output and a frame capture card with a digital input should be pursued The need for this was a result of experience With only an 8 bit grayscale the dynamic range was narrow making it very difficult to take 1mages without saturating the CCD A 10 bit or greater gray scale will help alleviate this problem 60 APPENDIX DYNAMIC LINK LIBRARY FOR IDL The following file is the code file for NUVIS DLL NUVIS DLL is a Dynamic Link Library that provides an interface rapping for IDL allowing IDL to calling the controlling functions for National Instruments IMAQ frame capture card Filename Nuvis cpp LT Todd A Hooks Date 11 2 1997 Purpose Provide an interface rapping for IDL and National Instruments Frame Capture Card Remarks Little to no error
38. was responsible for the off axis parabolic telescope mirror slit and the diffraction grating The telescope mirror slit and diffraction grating will be covered briefly but for detailed information see MacMannis 1997 The telescope mirror was a mirror we already had on hand It therefore became the starting point for the entire design layout size and optical component selection My work really began from here OPTICAL COMPONENTS Figure 1 shows a line drawing of the final design of the imaging spectrometer NUVIS The optical components marked 1 through 11 are identified in the figure caption and described in detail below All of these components are securely attached to the base plate Figure 1 Optical Path in NUVIS 1 filter window 2 scanning mirror housing 3 scanning mirror 4 baffle 5 baffle 6 telescope mirror 7 telescope mirror housing 8 slit 9 diffraction grating 10 diffraction grating holder 11 image intensified camera After MacMannis 1997 TELESCOPE MIRROR The telescope mirror was an off axis parabolic mirror manufactured by Space Optics Research Labs SORL sales order number SN4338 The mirror was 50 8 mm in diameter had a focal length of 152 4 mm with an off axis distance of 63 5 mm The surface was coated with aluminum and was overcoated with MgF providing a reflectivity greater than 9596 for wavelengths above 280 nm The scratch dig was 60 40 and the surface accuracy
39. 5 A 78 m tiere TET SEITDEM 1 E p o on 2 4 144 i ae E Pide NEE 7 TAU DE RA du ebrei ENT 5 4 M n LEE ies TIS sy t LIMON y Ia re b nra 4 7 RAE 4 Por AE Apr io EAT 94 17 4 ya 4 b ETE 1971 toner EN ee enirn qam VII Nain L a 1 v am POS du EY Ha La EM U tot oia t t PN mm LT vp 1 a Mu rr 6 aei oe Lie b ah 9 Prat tao Ta tn he bee us puta e nar d aq ti DOE sce it Ido d ret aly te rm Pn Ciis ad d Sta 444 AE ispast en ee regen Ny o gt ei EE Pues pur e 4 ion r r 1 SUM Li ih A E aera ie i s m A t qoc 10 190019 1 a PA Ub 4 p ee aa ura vole edet 5 we P a oed Se n H ne gens 103 m dire va e Ore MEL ET ener ra He 2
40. 8A OR PREVIOUS AXIS FAULT INDICATOR LIGHT J6 MOTOR CONNECTOR J4 ENCODER INPUT ENCODER STATUS LIGHTS J3 LIMITS HOME AND GO INPUTS FRONT MOUNTING BRACKET ON OFF SWITCHI ON LINE INDICATOR LIGHT AC POWER weur Figure 19 Max 410 Stepping Motor Controller Driver From Advanced Micro Systems 1996 The MAX 410 controller used basic ASCII serial commands The software included simulates a dumb terminal During testing a Wyse dumb terminal was used The controller had the capability of providing 40V at 4 amps It can step at 200 51 200 steps per revolution in increments of multiples of two 200 400 800 etc As 30 discussed above the controller provides internal indexing that was based number of steps sent or if an incremental encoder was used then on the actual number of steps detected by the encoder The controller provides full control over acceleration deceleration ramp up speed slow down speed jog speed running current usage and holding current usage just to mention a some of the controller s capabilities In our application the resolution was set to 25 600 steps per revolution which corresponds to 0 25 mradians per step After the completion of a command one of the trigger outputs on the controller changes state This output was used to trigger the National Instruments frame capture card discussed below to capture a image ABSOLUTE ENCODER I needed some method of d
41. A n A 4 51219 X EHE 910144105 n 543 3 ra ETAT KA 040 1 Rap a 143 O89 Old wd Pau ae dee 4 RI TI TONAN Sd a en en nd Ad Pike DH PATENS TES V Hes N auda A vett x AKT LTO oi E energie a U han s Korea be a at Ada on 24 DOE 4 h e Ade Kee wb steely 4 Masha I D e 45 8404 947 fw uL stas er Hoy m d iw DR EY 1 95 FAN Qed gt a METIN ri Patt dae Seat IR EAT ra 22 4 2 DAL LE AY dri 4 mx 410 1 ae 2 54 4 CORN n pea 444 34 E aM va Ns PT D 1 15 2 941145 i 4 S ts ey 2 333 a Dae A ais 2395536120 IIA PAR LM L m Mta xm O irs Veh Gon VAG PER PO RT t ate tat Sg l lt s K 370 et paca Y des d
42. Ft Belvoir VA 22060 6218 DOE E ee 2500 Naval Postgraduate School 411 Dyer Rd Monterey CA 93943 5000 Bx BENI IDE GEI c FORMEN TC Physics Department PH CL Naval Postgraduate School Monterey CA 93943 5000 a iuum coe DOE EL C O Michael A Hooks 3 Box 398 Kittanning PA 16201 Dr Ponal W Physics Department PH WA Naval Postgraduate School Monterey CA 93943 5000 Dr William B Meier Chairman Physics Department PH WB Naval Postgraduate School Monterey CA 93943 5000 2 5 10 99 22527 200 mue Ip c orn 4 ed ah TRH V n LIBRAR z 4 AE D E CETA 4 I I MI I anie SM on yours EL fat n EI I ts n Peet 3 2768 00 o AT PA TO A UT s 1494 1 na phye Tika at Q s Purse s a GA 1 UA pa my kU 1 dull pf TH hr yat atv Pol is i
43. Lowtran 7 The two filters that appeared to be the most promising were Schott glass UG 5 and UG 11 filters After contacting Schott directly I obtained a data printout of the transmittance for both filters seen in Figures 8 and 9 The mirror reflectance was greater than 95 percent for wavelengths above 280 nm for each mirror therefore used a constant reflectance of 95 percent The image intensifier supplier had provided the sensitivity data with the intensifier seen in Figure 10 Ziemer amp Associates Inc 1997 Solar Irradiance 200 300 400 500 600 700 800 900 1000 Wavelength nm Figure 7 Solar Irradiance at Earth s Surface From Lotran 7 15 UG 5 Transmittance curve 0 9 0 8 Transmittance e gt e gt o ho 0 1 200 300 400 500 600 700 800 900 1000 Wavelength nm Figure 8 Transmission Curve for UG 5 Filter From Schott UG 11 Transmittance curve 0 8 0 7 Transmittance e e c o gt ol o 0 1 0 200 300 400 560 600 700 800 900 1000 Wavelength nm Figure 9 Transmission Curve for UG 11 Filter From Schott image intensifer sensitivity 70 60 Sensitivity mA W 1 200 300 400 500 600 700 800 900 1900 Wavelength Figure 10 Image Intensifier Sensitivity After Ziemer amp Associates 1997 With all the required data I wrote a simple program that would read the data from text files interpolate the data elements so tha
44. OF REFERENCES Advanced Micro Systems MAX 410 Technical Reference Guide 1996 Appleman Dan Visual Basic 5 0 Programmer s Guide to the WIN32 API Ziff Davis Press Emeryville California 1997 Clausing Howard P A Clausing Inc private conversations 1997 Delft Electronische Producten DEP Second Generation Image Intensifiers 1994 Electro Optical Services Inc letter 1997 Giligan Larry Electro Optical Services Inc private conversations 1997 Johnson E O Design development and testing of an ultraviolet hyperspectral imager Master s Thesis Naval Postgraduate School Monterey California 1996 Keithley Metrabyte Corporation PIO 24 User s Guide 1991 Keithley Metrabyte Corporation DAC 02 User s Guide 1994 MacMannis A R The design of the Naval Postgraduate School s Ultraviolet Imaging Spectrometer NUVIS Master s Thesis Naval Postgraduate School Monterey California 01997 Pulnix Operations and Maintenance Manual of the TM 754 TM 765 High Resolution Camera 1996 Schott Glass Technologies Inc Optical Glass Filters Walden B S An analysis of middle ultraviolet dayglow spectra Master s Thesis Naval Postgraduate School Monterey California 1991 Ziemer amp Associates Inc letter 1997 15 74 INITIAL DISTRIBUTION LIST Defense Technical Information Center 8725 John J Kingman Rd STE 0944
45. Oriel holder that allows rotation of the grating about its center and about an axis passing through the center For further details on the diffraction grating and its holder see MacMannis 1997 H IMAGE INTENSIFIED CAMERA 1 Requirement for an Image Intensifier The earth s atmosphere is relatively transparent to visible and infrared IR radiation but it is an effective absorber of UV radiation Only a small fraction of UV radiation incident on the earth s atmosphere is transmitted to the surface Therefore in order to observe the UV spectrum at or near the earth s surface a detector that was extremely sensitive to UV light was needed It was also important that the detector be fast which required a detector that did not require long integration times The best detector for this application was an image intensifier 20 2 Image Intensifier Optical Parameters The image intensifier used was dual MCP image intensifier manufactured by Delft Electronische Producten DEP a Netherlands based company The image intensifier retains positional information and relative intensity therefore our spectral and spatial input was related to the output The intensifier serves two purposes in our application 1 it provides a usable photon to photon gain of about 10 and 2 it produces a shift between the input and the output wavelengths shifting the UV light to the visible spectrum to match the camera s peak sensitivity At the image intensifier inpu
46. SA or EISA slot That was all that was required for setup The card s control turned out to be more difficult In C and the I O commands allow direct access to the computer bus but Visual Basic has no such commands I had two options attempt to find I O commands from Keithley Metrabyte or write drivers myself Ichose the latter because it gave me full control and allow me to write commands that mirrored C and 42 Since PIO 24 was functionally equivalent to 8255 PPI it requires setup command to configure the three I O ports Table 3 shows the configuration bytes for the PIO 24 The base address of the PIO 24 was the address of port A Port s B and C addresses follow sequentially The configuration byte address was base plus three PA PB PCUppe PCLower 07 ns ps ps n2 i OUT OUT 5 YE Byte m m TTA The PIO 24 has a DB 37 connector with the pin out shown in Figure 24 Connection and use of this card was straight forward and no surprises were encountered 43 DIC COM 9 i 5V E 016 12V Er E 2 DIC COM i PA5 E DIG COM PAG amp 7 DIC COM PCO PBO n ES 2 8 3 E E 14 5 7 PCS 4 gt PCS 5 5 PB6 4 PB 3 a as EI INTERRUPT ENABLE 2 INTERRUPT INPUT V 1 Figure 24 PIO 24 Connector From
47. The solution was to write the functions in Assembly language and then call them from FORTRAN With the ability to write Assembly inline within C code I felt that this was the simplest solutions Therefore I started writing a DLL with the Assembly functions that were equivalent to the existing C and I O functions After several unsuccessful attempts I looked for help I found the help I needed in a book called Visual Basic 5 0 Programmer s Guide to the WIN32 Appleman 1997 This book included CD ROM that solved my problems it contained a DLL APIGID32 DLL that among other things contained functions written in Assembly language This DLL gave VB the same VO capability as E IDL AND DYNAMIC LINK LIBRARIES While I was working with Visual C and Visual Basic an opportunity arose to use NUVIS to observe twelve solid rocket motor firings at NPS Rocket Range With a deadline looming we needed a software interface immediately The VB program was not completed 55 so an alternate option was to use IDL to take data The idea of using IDL had been toyed with early on but no frame capture card on the market had controlling functions that were callable from IDL The problem was that IDL used a UNIX calling routine In this UNIX calling form two parameters are passed one by value the other by reference The best way to explain how it works is to give an example Suppose I want to pass two variables and four ar
48. al cameras The disadvantages are that we are limited to 8 bits of grayscale 640 by 480 lines of resolution and a more limited control of the 1mage exposure Having decided to use a video camera I next focused my attention to the types available Larry Giligan private conversations 1997 of Electro Optic Services the company which coupled the camera and intensifier and installed the intensifier power supply recommended that I purchase a camera with a 2 3 inch CCD instead of the more common Y inch CCD He explained that the optical coupling of a 25 mm image intensifier to the 2 3 inch was easier and more efficient than that of the smaller inch CCD The 2 3 inch CCD requirement narrowed my search to only a handful of manufactures The requirement for on chip frame integration was the final deciding factor Only one manufacture had a camera that met all these requirements Pulnix Inc The Pulnix TM 745e as mention above has a 768 by 494 2 3 inch CCD that measures 10 mm by 9 3 mm and has a pixel size of 1 lum by 13 um As shown in Figure 21 the camera had three connectors a BNC and proprietary 6 pin and 12 pin The 6 pin connector and its pinouts are shown in Figure 22 It was used to control the shutter speed through a 3 bit TTL compatible signal The shutter speeds available are listed in Table 1 A digital I O card discussed below was used to control both the camera shutter speed and the frame integration 36 Figure 21 C
49. al requirements were add The first was that all edges be slightly rounded to prevent chipping and that the corners have a 3 2 mm curvature The 3 2 mm curvature was to help prevent chipping but it was primarily added to make the machining of the filter holder easier Image intensifer sensitivity with UG 11 Sensitivity mA micron 1 200 300 400 500 600 700 800 900 1000 Wavelength Figure 12 Sensitivity w th UG 11 Filter Figure 13 shows the entrance window with the UG 5 filter the missing screws on the top and bottom of the filter holder are used to hold the filter cover Figure 13 Entrance Window 19 DIFFRACTION GRATING The diffraction grating was the single most difficult optical component to obtain This had to due with two factors 1 the wavelengths of interest and 2 the grating needed to have a flat field focus MacMannis 1997 identified a supplier of a grating that would suit our needs It was a flat field imaging spherical diffraction grating from Instruments S A Inc ISA As discussed above the grating provide a flat field focus in the wavelength range of 300 400 nm over a length of 25 mm It had a ruling density of 1200 grooves mm The grating was manufactured using a Zerodeur substrate coated with aluminum and overcoated with a UV protective coating of MgF to protect the aluminum The effective reflectivity over this bandwidth was greater than 95 percent The grating was mounted in an
50. amera Connectors From Pulnix 1996 6 PIN Connector 1 D2 4 412V DC 2 GND 5 DO O 3 Video 6 D1 3 NOTE Do D1 D2 are shutter control inputs Figure 22 Camera P6 Six Pin Connector From Pulnix 1996 a 1 E From T 1996 The 12 pin connector shown in Figure 23 supplied the camera power 12 V as well as optional horizontal and vertical sync frame integration control bit and option video output The video was carried via a RG 59 75 ohm coaxial cable connected to the RS 170 interface using a standard BNC connectors Frame integration was accomplished by providing a logic level low to the integration pin As long as the logic level was low charge will continue to accumulate on the 2 PIN Connector 1 1 GND 7 Vp in 2 12V DC 8 GND 3 GND 9 Hp In 4 Video 10 5 GNO 11 Integration Control 6 Vini 12 GND Figure 23 Camera P12 12 Pin Connector From Pulnix 1996 DR Image Intensifier As mentioned above the CCD camera and the image intensifier were sent to Electro Optical Service Inc to be optically coupled and have a image intensifier power supply added Electro Optical Services returned the components as a single unit image intensified camera The image intensifier power supply required 12 5 V It also input fora O 10 V reference to control the MCP voltage 0V corresponding to O V across the MCP and 10 V cor
51. ameters pointer to a pointer to sessionID a pointer to a pointer to numFrames a pointer to a pointer to skip a pointer to a pointer to buffer Return returns zero LONG WINAPI Sequencelmage LONG numArguments PULONG ppBuffer CONST ULONG MAX FRAMES 50 CONST ULONG FRAMESIZE 307200 LONG error 0 PULONG pSessionID ppBuffer 0 ULONG sessionID pSessionID PULONG pnumFrames ppBuffer 1 UONG numFrames pnumFrames PULONG pSkip ppBuffer 2 ULONG skip 2 pSkip ULONG skipArray 350 0 67 ULONG buffer 350 0 PB YTE pBuffer PBYTE ppBuffer 3 UINT i J UINT numLoop numRemainder if numFrames 350 return 1 numLoop numFrames MAX FRAMES numRemainder numFramesg oMAX FRAMES for i 2 0 i numFrames 1 skipArray i skip if numLoop 0 for i 0 1 lt numLoop 1 for 0 J lt FRAMES j buffer j ULONG pBuffer pBuffer FRAMESIZE j configure the session for a sequence with numFrames using the buffer passed in pBuffer error imgSequenceSetup sessionID MAX FRAMES void buffer skipArray TRUE FALSE if numRemainder gt 0 for i 0 i lt numRemainder buffer i ULONG pBuffer 68 pBuffer FRAMESIZE configure session a sequence with numFrames using
52. ance Window The entrance window shown in Figure 6 had the tightest dimensions in the optical mm lip designed to hold the filter 12 2105 41224 AE PXI WIEN e H7101 SRN N d RAJATI AJAR RHO TMN 042 0 969 4006 0 0002 V ZS 400 4 1468 0260 009 0 5134204 4339 SC 0 0060 EV6 gt 0E6 0 310H NYHL GU CALINYNO 1397504 SCL 0 LA 16 65 31 9 QLapIS oN Hd epo DOHA 9po5 AN BOP 3204 JUBISAA GPS suu V NOILO3S ing Entrance Window Draw Figure 6 13 2 Filter During the optical design a single MicroChannel Plate MCP image intensifier was used to observe the spectrum of a mercury lamp While using this image intensifier it became apparent that its sensitivity to visible light would present difficulties if any stray light were present optical filter needed to be added in addition to making NUVIS light tight and using baffles to reduce internal reflections The placement of the filter required some careful consideration The filters with the best optical properties were interference filters Interference filters require that incident light be collimated and be normal to the surface This meant that any interference filter used would have to be mounted within the telescope It would ha
53. and 2 channel DAC The driver software and examples included were excellent and unlike Keithley Metrabyte 4 National Instrument cards are Plug n Play making installation a breeze The only drawback was that the National Instrument card did not provide access to the computer s 12 V bus For this reason and this reason alone I decided to use a card from Keithley Metrabyte It might seen strange that I would make my decision solely on the voltage that a card can supply but this was easily explained NUVIS was a field instrument therefore the fewer the external support components required the better If I had used the National Instruments card I would also need an additional 12 V power supply for the camera image intensifier and the absolute encoder The Keithley Metrabyte card that I chose for the digital was the PIO 24 It was an ISA EISA card that was functionally equivalent to the 8255 PPI providing 24 bits of digital I O in three ports A B and C Its I O lines provided a high of 20 mA per bit compared to the normal 12 mA per bit The installation required that you find an available bus address and then set the dip switches on the card to that address The PIO 24 address was Hex 300 This process was repeated for the IRQ setting if the card s interrupts are to be used I had no reason nor saw future reason to use interrupts therefore the interrupt setting was disabled After setting up the card it was inserted into a free I
54. and to allow clearance for the mirror brackets 10 mm was added to the width and 6 mm was added to the height requiring a mirror that was 90 mm by 90 mm The actual projection of the slit at the position of the scanning mirror was an ellipse therefore the corners were cut at a 45 degree angle 5 mm from the corner along each side In addition to help prevent chipping and to make handling easier mm by Imm bevel was placed around the entire finished side edge of the finished side of the mirror Figure 3 is a drawing of the scanning mirror The company that manufactured the scanning mirror was P A Clausing The scanning mirror substrate was made of Zerodeur to minimize thermal expansion The surface was coated with aluminum and overcoated with a proprietary UV coating called UV280 The UV280 protects the surface while still allowing greater than 95 percent reflectance for wavelength above 280 nm UV280 was choosen over MgF based on the recommendation of the mirror manufacture Clausing 1997 UV280 was more durable and has a higher resistance to moisture than MgF making it the better choice Our requirement for the mirror flatness was to have a deviation of no more than 5 10 at 632 8 nm across the diagonal which required a minimum mirror thickness of 12 mm The finished mirror exceeded the requirements with a flatness of 20 at 488 nm across the diagonal and scratch dig of 20 10 This makes the telescope mirror the limiting reflective su
55. angle that provided a total Field Of View FOV of 20 degrees f O XX YVININY SET OF LOX X Tom 90 Sd S3HONI 6313508458 38 mW3H10 9631Nf SC V 08 068 L 4 Qc L S31ON m 7 A 2 2 7 MH33WVHO 0 2 2 Z 2 7 L 7 uinutumiy TIVI E E TE dm NOILO3S 1650 31vQ Hd ONS jezAjeuy AN L 311 Figure 2 Drawing The scanning mirror had to be large enough to fill the entire IFOV when offset 50 degrees from the telescope optical axis I decided that the mirror width would be determined by the longest optical path By this I mean that the optical path from the slit to the edge of the scanning mirror farthest from the telescope mirror would be used to determine the scanning mirror size to avoid vignetting The reason for this was that if the actual optical paths had been used the optical center of the scanning mirror would not have corresponded to the geometrical center and the scanning axis The end result was that the mirror was wider than actually needed The benefits of a smaller scanning mirror were more than offset by having the optical geometric and scanning axis coincide Given these considerations the minimum mirror dimensions were 80 mm wide by 84 mm high To ensure that the mirror does not cause vignetting
56. at field imaging diffraction grating an 1mage intensified camera assembly and the support controlling electric and electronic hardware and software This is part of a continuing project to build test and use this sensor in support of military and government agencies gt TABLE OF CONTENTS A C n REED INI POP cse 5 3 TELESCOPE MIRROR a 3 A o 1 INSTANTANEOUS FIELDS OF 5 D 5 E SCANNING MIRROR ASSEMBLY 6 6 2 Scanning Mirror Housing AA Pc 10 ENTRANCE WINDOW AND FILTER 2 1 Sirio 12 2 c ta Lice Se l4 G DIFFRACTION GRA TINGSET 20 li IMAGE INTENSIFIED CAMERA en een 20 tor an Image Intensifier 2 20 2 ImaseIntensitier Optical Parameters 22 22 2 21 COUP INIT co 23 4 piicalgearameters of the om E 23 ENCES RU oC 24 vli IIESEFECTRICAL MECITANICAL COMPONENTS 2551 7 222 2 2 2 A S PISIS VIQR ee ta a 2 STEPPING MOTOR CONTROLLER DRIVER 29 C ABSOLUTE ENCODER u N el IS GONIEONEN TS 35 A IMACE INTENSIFIED CCD CAMERA coace Penn 35 ll CGD A ISO
57. ble 1 Table 2 Table 3 Table 4 Table 5 Table 6 LIST OF TABLES Camera Shutter Speed Control From Pulnix 1996 937 Image Intensifier DB 9 Pin Out From Electro Optical Services 1997 38 PIO 24 Configuration Byte From Keithley Metrabyte 1991 43 DAC 02 Output Configuration From Keithley Metrabyte 1994 45 Ne en 31 Interna Harness LE lise 2227 52 Xi 4 ACKNOWLEDGMENT The author would like to acknowledge the financial support of HYMSMO for making NUVIS possible XV I INTRODUCTION Imaging spectrometry 15 a new and rapidly growing form of spectrometry Unlike traditional spectrometry where only spectral datais obtained imaging spectrometry produces two dimension spacial image with the corresponding spectral data Numerous multi spectral and several hyper spectral instruments have been constructed to investigate the visible through infrared regions of the spectrum but the ultraviolet UV region of the spectrum has all but been ignored Naval Postgraduate School NPS UltraViolet Imaging Spectrometer NUVIS a hyperspectral imaging spectrometer was designed to investigate and map the emission absorption and reflection of solar UV radiation by the earth s surface gas clouds and military targets The ability to combine both the spectral and spacial data allows large areas to be imaged and analyzed Once a c
58. d and having a frame capture card alone The problem was that in this type of card the captured image was held in video memory To save the image to a disk the image must first be copied from video memory to system memory byte by byte did not have the lower functions needed to quickly and efficiently transfer the images from the video memory to system memory Once in system memory provided ample speed and ease to do what ever I wanted The result was that when transferring images from video memory to system memory the frame rate dropped from 30 fps to less than 5 fps This problem coupled with the complexities of Visual C lead me to look for another solution What I found was a new card from National Instruments IMAQ PCI 1408 that supported numerous programming languages including Visual and Visual Basic The card was capable of a full 30 fps with 8 bits of grayscale and it was fully documented with excellent manuals and numerous example programs The IMAQ was a PCI BusMaster card that has two external connectors BusMaster means that the card can take control of the PCI bus and conduct data transfers without processor interaction One of the cards connectors was a standard BNC for RS 170 video 40 The other connector was DB 15 that had three digital I O lines that could be used as either digital I O or triggers horizontal and vertical sync input and input for four video sources The only draw back to this card was the d
59. ded and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of this collection of information including suggestions for reducing this burden to Washington headquarters Services Directorate for Information Operations and Reports 1215 Jefferson Davis Highway Suite 1204 Arlington VA 22202 4303 and to the Office of Management and Budget Paperwork Reduction Project 0704 0158 Washington DC 20503 AGENCY USE ONLY Leave blank 2 REPORT DATE 3 REPORT TYPE AND DATES COVERED December 1997 Master s Thesis 4 TITLE AND SUBTITLE 5 FUNDING NUMBERS DESIGN CONSTRUCTION AND OPERATION OF THE NAVAL POSTGRADUATE SCHOOL S ULTRAVIOLET IMAGING SPECTROMETER NUVIS 6 AUTHOR S Todd A Hooks 7 PERFORMING ORGANIZATION NAME S AND ADDRESS ES 8 PERFORMING ORGANIZATION Naval Postgraduate School REPORT NUMBER Monterey CA 93943 5000 9 SPONSORING MONITORING AGENCY NAME S AND ADDRESS ES 10 SPONSORING MONITORING AGENCY REPORT NUMBER 11 SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the official policy or position of the Department of Defense or the US Government 12a DISTRIBUTION AVAILABILITY STATEMENT Approved for public release distribution is unlimited 13 ABSTRACT maximum 200 words 12b DISTRIBUTION CODE Hyperspectral imaging spectrometers produce image comprised of
60. e Intensifier Sensitivity i s oe ws de ee Eeue with B O5 Filter 2 2222 02 TREES 18 iZ sensitivity with JO I Filter 2 22 22 25502 19 Entrance Window Meer 2 ea 19 14 P 20AF Phosphor Emission Curve 2 22 uen TM 745e CCD Se a a a a a a e 22 lop ol NONIS mE seus eo as 29 Sde View ol 220000022 25 Figure 18 Graphical Representation of Hypercube 2 Figure 19 MAX 410 Stepping Motor Controller Driver From Advanced Micro Systems A lnc MEUS Pot USES nen Mente oe ane aed 30 20 2 Absolute Encoder From US Digital 32 Figure 21 Camera Connectors From Pulnix 1996 57 Figure 22 Camera P6 Six Pin Connector From Pulnix 1996 517 Figure 23 Camera P12 12 Pin Connector From Pulnix 1996 38 Figure 24 PIO 24 Connector From Keithley Metrabyte 1991 44 Figure 25 DAC 02 Connector From Keithley Metrabyte 1994 45 2 60 NU Computer 46 Cargo aser cian TEN o TERR ES 47 29 CompulerConneclors He 22222222002 77 55 a 48 El NU VISI 49 Eure 30 NUVIS Cool Software Interface oe 54 xli Ta
61. e width and the height required It was decided to fabricate slit using razor blades See MacMannis 1997 for further details on the slit A holder had to be constructed for the slit The original holder was a post holder with the slit at the top As construction continued it was found that this type of holder did not allow for a light tight enclosure new holder was designed that would fasten to both the image intensifier housing and one of the internal light tight walls The holder had two purposes 1 had to hold the slit and 2 It had to provide a light tight seal around both the slit and the image intensifier An added benefit was that by fixing the relative position of the slit and the image intensifier a degree of freedom was removed therefore making optical alignment simpler C INSTANTANEOUS FIELDS OF VIEW Using the dimensions of the slit and the focal point of the telescope mirror the Instantaneous Fields Of View IFOV were be calculated The Equations Slit Height Telescope Mirror Focal Length Slit Width Telescope Mirror Focal Length were used to calculation the IFOVs Where is the horizontal IFOV and 0 is the vertical IFOV From these equations I determined to be 0 59 milliradians and to be 98 5 milliradians The IFOVs are then used to determine the size of the baffles scanning mirror and entrance window D BAFFLES In order to reduce stray light two baffles were placed between the scanning mirror
62. erfaceID LONG EITOT PULONG pSessionID ppBuffer 0 PULONG pInterfaceID ppBuffer 1 sessionID pSessionID interfaceID pInterfaceID stop grab acquisition error lt imgSessionStopAcquisition sessionID close this interface free all resources error imgClose interfaceID TRUE return O 64 Function SnapOpen Parameters pointer to a pointer to sessionID pointer to a pointer to interfaceID Return returns Zero LONG WINAPI SnapOpen LONG numArguments PULONG ppBuffer ULONG interfaceID ULONG sessionID LONG PULONG pSessionID ppBuffer 0 PULONG pInterfacelD ppBuffer 1 Open an interface and a session error imgInterfaceOpen img0 amp interfacelD error imgSessionOpen interfaceID amp sessionID pSessionID sessionID pInterfacelD interfaceID return Q j Function Snaplmage Parameters a pointer to a pointer to sessionID a pointer to a pointer to a image buffer Return returns Zero LONG WINAPI SnapImage LONG numArguments PULONG ppBuffer ULONG sessionID LONG PULONG pSessionID ppBuffer 0 PBYTE pBuffer PBYTE ppBuffer 1 sessionID pSessionID 65 grab c
63. etermining the scanning mirror position at power up 1 determined that the easiest most reliable and most accurate way to do this was to use an absolute optical encoder There are a large number of incremental encoder manufactures but very few that manufacture absolute encoders My search lead me to a company called US Digital Corp US Digital s absolute encoder called the A2 has a resolution from 2 to 65 536 steps per revolution and allows an origin to be set using nonvolatile memory The encoder was available in several variants such as sleeve shafted and bearing shafted The model I chose was the A2 S K 250 a kit form of the encoder that had no shaft but would except an external 6 4 mm shaft that was 15 2 20 3 mm long The kit included the encoder and an optical encoder disk that had a 6 4 mm center hole with a set screw Figure 20 shows a picture of the disassembled encoder 31 20 je Figure 20 A2 Absolute Encoder From US Digital The control and operation of the encoder was through a RJ 75 6 pin connector that connected to a proprietary RS 232 adapter The adapter also required a 9V external power source that was provided During operation of NUVIS it was discovered that the adapter needed a power supply between 6 18 V The significance of this will be explained below A test platform was built using a 102 mm by 102 mm piece of aluminum angle iron On one side a Frame Size 23 stepping motor mount wa
64. ffer 0 PULONG pInterfaceID ppBuffer 1 Open an interface and a session error imgInterfaceOpen img0 amp interfacelD error imgSessionOpen interfaceID amp sessionID configure the session for a grab but do not start the acquisition yet error imgGrabSetup sessionID FALSE start the acquisition error imgSessionStartAcquisition sessionID pSessionID sessionID pInterfaceID interfaceID retum 0 Function GetGrab Parameters a pointer to a pointer to sessionID a pointer to a pointer to a image buffer Return returns Zero LONG WINAPI GrabImage LONG numArguments PULONG ppBuffer ULONG sessionID LONG error 63 PULONG pSessionID ppBuffer 0 PBYTE pBuffer PB YTE ppBuffer 1 sessionID pSessionID grab a copy of the acquisition buffer into my own user buffer error imgGrab sessionID amp pBuffer TRUE return 0 Function GrabClose Parameters a pointer to a pointer to sessionID pointer to pointer to interfaceID Return returns zero LONG WINAPI GrabClose LONG numArguments ppBuffer ULONG sessionID ULONG int
65. haracteristic signature has been determined data analysis allows specific objects to be identified and located One of NUVIS first applications will be to analysis the SO gas and ash emissions from volcanos Military applications include battle field target identification and location through the use of Unmanned Aerial Vehicles NUVIS started in 1990 with an instrument called MUSTANG Middle Ultraviolet SpecTrograph for Analysis of Nitrogen Gases designed to analyze the earth s ionosphere MUSTANG utilized an Ebert Fastie spectrometer with a diode detection array more information see Walden 1991 Johnson 1996 attempted to modify MUSTANG to produce an imaging spectrometer The modified instrument was named Dual Use UltraViolet Imaging Spectrometer DUUVIS Johnson finished DUUVIS but did not have time to conduct tests do determine the quality of the images During January February 1997 tests were conducted on DUUVIS by MacMannis 1997 and myself We discovered that DUUVIS could not be modified to produce acceptable images due to optical limitations At about the same time funding was obtained from Hyperspectral Measurement and Signals Intelligence Support for Military Operations HYMSMO to build a new instrument from scratch From this point the design of NUVIS branched into two areas I became responsible for all electronics the scanning mirror baffles filters software and the position of these components Andy MacMannis
66. he vertical spacial dimension and the corresponding horizontal spectral components In order to obtain a two dimensional image with the corresponding spectral component multiple frames that were horizontally separated by one horizontal IFOV were combined to produce a hypercube Figure 18 shows the graphical representation of a hypercube SPATIAL Height SPATIAL Width SPECTRAL Figure 18 Graphical Representation of a Hypercube After Johnson 1996 27 The construction of a hypercube was accomplished by moving scanning mirror half the horizontal IFOV As discussed above the horizontal IFOV was 59 mradians This required that the mirror be moved approximately 3 mradians per frame Converting the radians to steps it was found that the scanning mirror must scan at about 21 000 steps per revolution There were two options available to moving the scanning mirror servomotor or a stepping motor After reviewing both types of motors and their respective controllers I decided to use astepping motor and micro stepping motor controller driver The stepping motor had several advantages over the servomotor 1 stepping motors and micro stepping controller drivers are more common than the servomotor and its controller 2 servomotors and servomotor controllers with the required resolution and require torque were not available and 3 servomotor controls were not as accurate as current micro stepping motor controllers The stepping mo
67. ied Unclassified Unclassified NSN 7540 01 280 5500 Standard Form 298 Rev 2 89 19 SECURITY CLASSI UL Prescribed by ANSI Std 239 18 11 Approved for public release distribution is unlimited DESIGN CONSTRUCTION AND OPERATION OF THE NAVAL POSTGRADUATE SCHOOL S ULTRAVIOLET IMAGING SPECTROMETER Todd A Hooks Lieutenant United States Navy B S The Virginia Military Institute 1989 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN PHYSICS from the NAVAL POSTGRADUATE SCHOOL ABSTRACT Hyperspectral imaging spectrometers produce an image comprised of the standard two dimensional spatial scene and the corresponding spectra of each scene Hyperspectral imaging is a relatively new and fast growing field with both commercial and military applications Commercial applications vary from vegetation identification and mapping surface geological identification and mapping to atmospheric composition and mapping Military applications include target identification and classification airborne chemical identification and mapping and rocket plume identification This thesis describes the design and operation of the NPS Ultraviolet Imaging Spectrometer NUVIS NUVIS is hyperspectral imaging spectrometer designed to investigate the ultraviolet region of the spectrum NUVIS is comprised of a scanning mirror telescope assembly using an off axis parabolic mirror a slit a fl
68. ifficulties capturing integrated frames The IMAQ only had horizontal and vertical inputs no outputs With no output it had to take its timing from the video signal Therefore without any way to control the sync timing signals there was no easy or efficient way to time the capture of an integrated frame There were two ways around this One was to use an external controllable sync timing source to provide the timing signals to both the camera and the IMAQ card The other fix was to use software integration Since the sensitivity of the camera and intensifier were found to be better than expected integration should not be needed But if integration 15 required the software solution will be employed With the issue of the frame capture card settled I still had not mastered the visual side of Visual My solution was to abandon Visual altogether and use Visual Basic This created a trade off Visual Basic made the visual design much easier but the language was not as powerful as especially its inability to handle pointers Despite the trade off Visual Basic proved to be a good choice DIGITAL VO CARD In order to control the camera shutter speed and frame integration Ineeded four TTL logic bits I was familiar with Keithley Metrabyte cards but after having such success with the National Instruments IMAQ card I decided to look at what National Instruments offered What I found was the PC AO 2DC a combined 16 bit digital I O card
69. ion of this thesis NUVIS still requires some work Some ofthe work required was the result of experience gained during the rocket firings other items just weren t completed due to the lack oftime I will continue to work on NUVIS for nearly two months after completion of this thesis therefore the recommendations that are schedule for completion are annotated As discussed above NUVIS was originally designed to scan horizontally During the rocket firings NUVIS was mounted vertically on a tripod By changing the orientation the effects of gravity on the scanning mirror could no longer be ignored The result was that the stepping motor was unable to move or position the mirror accurately The solution to this problem was to balance the scanning mirror and to replace the stepping motor with a more powerful motor that included an incremental encoder As discussed above a new stepping 59 motor with incremental encoder was installed but balancing scanning mirror and scanning mirror holder has yet to be completed NUVIS construction was focused on looking at the UV region from 300 400 nm By coincidence the manufacture of the diffraction grating also makes gratings that are physical and optical replacements providing different bandwidths A new grating with a 200 400 nm bandwidth will be ordered and in December 1997 In order to use the increased bandwidth of the new grating a new filter will be needed This filter most likely will have
70. ixels remotely controllable electronic shutter speed on chip frame integration at least 8 bit grayscale and standard RS 170 video interface These requirements need some explanation NUVIS was designed with atotal horizontal FOV of 20 degrees This means that the scanning mirror must scan through an arc of 10 degrees Using the 25 600 steps per revolution a 10 degree corresponds to 711 frames for the total horizontal FOV This meant that ahigh speed camera was essential in order to take data in areasonable amount of time Even though speed was critical I felt that low light conditions may require frame integration I wanted a camera that used a digital interface to transfer the video data frame by frame This type of camera proved inadequate because most single frame digital interface cameras could not capture and transfer at more than a few fps Cameras that did have ahigh speed digital interface required a special computer interface that when combined with the camera made the system cost unresonable The solution was to use a video camera with a standard RS 170 interface that would provide up to 30 fps 35 A video camera had several advantages and disadvantages The advantages were that video cameras were relatively inexpensive and were readily available from dozens of manufactures Also the frame capture computer cards for RS 170 interface video cameras were about one third the price of the digital interface card needed for digit
71. ll NUVIS functions and view video from one dialog base window As new components were added to NUVIS I planned to add the corresponding controls and code to the program I first ran into trouble with the PIO 24 and DAC 02 cards Visual Basic although a powerful language was not as powerful as Visual The designers had decided that VB would not be allowed direct access to the computer bus Keithley Metrabyte had included functions for VB that allow I O but these functions were very cumbersome to use and their complexity was an over kill for this application Without I O controlling functions I would not be able to control the camera shutter speed and integration as well as the image 54 intensifier gain I thought that by writing a Dynamic Link Library DLL in Visual using the functions I would be able to call the functions from VB After several unsuccessful attempts I decided I needed some help I began to search for a book that address VO and VB After a week of no luck I broke down and called Microsoft The technician went through a long explanation of why VB had no I O functions and also why my attempts to write a DLL were unsuccessful The bottom line Microsoft did not want VB programers to have direct access to I O and the VB calling procedure prevented the use of VO functions DLL With this new information I decided to change my approach Years ago while programming in FORTRAN 77 I needed VO functions
72. nd integration control Port C Upper and Port A are only connected for future use Note the digital and analog low level grounds that were tied to together This was because the PIO 24 12 V supply was referenced to digital ground therefore the 0 10 V signal from the DAC 02 also had to be referenced to the digital ground 50 Table 5 External Wiring Diagram 2 Wiring Harness The internal wiring harness pin out is shown in Table 6 Spare indicates an unused pin NUVIS 1 was the input DB 25 NUVIS 3 was an output DB 25 for future expansion The 12V powers the image intensifier camera and the encoder adapter 12 SEI Use Spare az gt gt on on ga ge BE O O Jg gt 2 4 5 5 Spare Spare A7 Spare 9 Shutter DO PCI Shutter D1 PC2 Shutter D2 z Integrate 4 Spare C5 Spare C6 Spare u gt C7 Spare U A 5V Spare 12V Power ON 12V Spare Interrupt Enable Spare wk 8 Interrupt Spare 19 10 V Gain Control A 20 GND Ties analog and digital GND in 1 gt lt ag ND Ties analog and digital GND Table 6 Internal Wiring Harness 52 V SOFTWARE VISUAL When I started the design of NUVIS I fully expected to use Visual for soft
73. nt and more durable The fiber optic bundle or tapper was 25 mm in diameter on the intensifier side and 13 7 mm in diameter on the camera side I decided that the camera CCD would be completely enclose withing the circle of the fiber optic tapper i e the rectangle was within the circle This allows for full utilization of the CCD while sacrificing some of area of the image intensifier This had the effect of reducing the bandwidth by about 15 nm for a total bandwidth of 85 nm 4 Optical Parameters of the CCD Camera The CCD Camera used was a Pulnix TM 745e video camera with a Sony 2 3 inch CCD having 768 by 494 pixels As mention above the CCD measures 10 mm by 9 3 mm with pixels that were 11 um by 13 um and the peak sensitivity was between 500 nm and 650 nm as shown in Figure 15 In order to calculate the slit width I needed the pixel size as projected on the input window of the image intensifier This was done by starting with the diagonal of the CCD which was 13 7 mm The diagonal was then dividing by the diameter of the image intensifier 25 mm The result was then multiplied by the pixel dimensions The resultant projected pixel on the photocathode had the dimensions of 20 1 by 23 8 25 ENCLOSURE With all of the optical components and their holders identified the construction of the enclosure could begin The first thing that was done was to machine two joining sides of the base plate making a square corner In this co
74. o 4 xa 14 att 347 446729442747 arai dy 15 9 rtp cuo DU ta wu vc i VA ANOS 8 9 mi 2 tg hed ole E s N s A A hao 4 DS EN era hi a DE ado parta o ela VN X lt 24026 41 2 P A EY i 1 O gl ih s i 445422 568125216 25 1 44 Stic sais 25 S 4 e BECK 44 0 LSU T S MSS IAM dant sen ERAS AM sede NU 82 4 V atakay 274 nz a Tak un e Ke 1 5 0 25 4 OS riw J dad PET 2 4 A PAE 266 a PR APA i y ri 2 ds h Lino AG elas pne 4 4 4 1 4 45 5 ads e Y jae 1 iei abi tordo S L4 94 IIS RU OCT ER 1 RET lt 6 1 1 lt 15 ls 23074348 1 PP TTL E Mi Le p ACA LE s eut t 4 2 PA id 3 8497 5 LN Shy A DEW M P hs 1 IPIE TN gi
75. o access all the upper level functions of the IMAQ While using the new DLL we found that IDL was not fast enough to keep up with the data rate flow IDL was receiving data at only about 4 fps too slow to take data at a rocket firing that would only last about 8 seconds The solution was to write functions in the DLL that allowed IDL to simply pass the number of frames to capture and the pointer to the first element of the array that was to receive the data After writing the new functions IDL was able to record at 30 fps The Appendix contains the printout of the NUVIS DLL that was developed for IDL 2 v VI SUMMARY AND RECOMMENDATIONS When my thesis started I had no idea of the difficulties that I would encounter As the design and construction preceded I would solve one problem only to be confronted by several more As a result the time it took to build NUVIS exceeded the estimates But regardless of the difficulties NUVIS was operational in time for twelve solid rocket motor firings that were held at the NPS Rocket Lab during the week of November 3 1997 When the time came to begin writing this thesis I realized that NUVIS was a living instrument It would continue to undergo changes and modifications as experience is gained during its use As a result I felt that this thesis would best serve as a sort of user s manual containing detailed descriptions of the components operation and the pitfalls to avoid At the complet
76. opy of acquisition buffer into my own user buffer error imgSnap sessionID amp pBuffer return 0 Function SnapClose Parameters a pointer to a pointer to interfaceID Return returns Zero LONG WINAPI SnapClose LONG numArguments PULONG ppButffer LONG PULONG pInterfaceID ppBuffer 0 ULONG interfaceID interfaceID pInterfaceID close this interface free all resources error imgClose interfaceID TRUE return 0 j Function SequenceOpen Parameters a pointer to a pointer to sessionID a pointer to a pointer to interfaceID Return returns Zero LONG WINAPI SequenceOpen LONG numArguments PULONG ppBuffer LONG error ULONG sessionID 0 ULONG interfaceID 0 66 PULONG pSessionID ppBuffer 0 PULONG pInterfaceID ppBuffer 1 an interface and a session error imglInterfaceOpen img0 amp interfaceID error imgSessionOpen interfaceID amp sessionID pSessionID sessionID pInterfaceID interfacelD return 0 Function Sequencelmage Par
77. polar Output 48V 13 25 DAC O 4 to 20 mA Output Figure 25 DAC 02 Connector From Keithley Metrabyte 1994 Oto5 V 2 to 22 15 to 16 Oto 10 V 201022 24 14 to 16 18 5 to 45 V 21 to 22 23 15 to 16 17 10 to 10V 20 to 22 23 14 to 16 Table 4 UE Output Configuration From Keithley Metrabyte 1994 Control of the DAC 02 used the same functions as the PIO 24 In this case each DAC channel used two addresses a lower and upper address The four most significant bits 45 MSB on lower address were four least significant bits LSB in the 12 bit word The eight MSBs of the 12 bit word come from the second address The output of the channel does not change until the card receive two bytes one on the lower address and one on the upper address of the corresponding channel All four addresses for the two channels were sequential E COMPUTER The final electronic component was the control computer The control computer should have been easy but as with everything else involved with this instrument it was not What made it so difficult was the fact that the computer needed to be portable rugged and capable of accepting the PCI IMAQ 1408 card the PIO 24 ISA EISA card and the DAC 02 ISA EISA card The need for I O cards made using a laptop impossible but a standard computer was neither portable nor rugged The only alternative was a portable style computer case that was common in the early 19805 It turn out to
78. problems were encountered with the computer or its components during assembly testing or extended operation F CABLES AND WIRING HARNESS Cable The external cabling for NUVIS was somewhat involved Atthe computer there were two RS 232 serial ports DB 9 and DB 25 the IMAQ BNC connector the PIO 24 DB 37 connector and the DAC 02 DB 25 connector shown in Figure 29 At NUVIS there was a BNC and three DB 25 connectors These are designated NUVIS NUVIS 1 NUVIS 2 and NUVIS 2 Figure 29 NUVIS Connectors 49 The computer RS 232 serial ports com and com2 are connected to 410 interface adapter and the A2 encoder interface adapter mounted on NUVIS NUVIS 2 As discussed above the MAX 410 connector was connected directly to the computer com2 DB 25 serial port From this adapter an RJ 45 connector with 8 lead cable connects the adapter to the MAX 410 The other communications port serial port DB 9 was connected to NUVIS via a DB 9 computer to DB 25 NUVIS 2 cable The encoder adapter resided inside NUVIS on the other side of the DB 25 The heart of the external cabling was the cable that connected the PIO 24 and DAC 02 to the first DB 25 connector on NUVIS called NUVIS 1 Table 5 shows the connections for this cable pin for pin A third DB 25 connector NUVIS 3 provides external access to digital I O and Mone sources Notice that Port C Lower was used for controlling the camera speed a
79. rays to a function The first parameter can be anything I want but the second parameter was a pointer to an array of pointers The second parameter points to the first element of an array whose elements are pointers that point to the data that was passed In this example the first two array elements point to the two variables while the next four array elements point to the first elements of the four arrays that were passed This Unix type of argument passing was difficult and confusing to work with and it was not support by the frame capture card manufactures As a result IDL was initially dismissed as the language for acquiring the data although it will be the language used to analysis the data With a time crunch and me being the only person familiar with VB IDL was again considered Steve Finny and David D Cleary of NPS Physics Department were experienced with programming IDL enough so that they felt they could build the GUI in several days The problem was still getting IDL to call the National Instruments IMAQ functions for the frame capture card The difference now was that I had gained experience making DLLs during my efforts to develop I O functions for VB I first wrote a test function in a DLL that could be called by IDL After this success I began to incorporate the actual IMAQ functions combining as many as possible in order 56 to minimize required IDL calls The result was a DLL that was callable from IDL allowing IDL t
80. responding to 1600 V across the MCP Electrical connections were made through a DB 9 connector on the image power supply housing The pin out is listed in Table 2 The reference voltage was provided by a Digital to Analog Converter DAC computer card discussed below DB 9Pin Function Description ec HERES VV Table 2 Image Intensifier DB 9 Pin Out From Electro Optical Services 1997 B FRAME CAPTURE CARD In order to get the data from the analog RS 170 video format to a digital format that could be analyzed a video frame capture card was required B amp W frame capture cards vary 38 greatly in both quality and cost but high quality does not always mean high cost Frame capture cards are relatively new and as a result the better cards have not had the time to break out from the poorer quality cards An import factor that I discovered when considering a frame capture card was that the quality of the controlling software and documentation are just as important as the quality of the hardware itself The frame capture cards can be classified into two basic classes those with built in display drivers and those without At first the cards with the built in VGA display drivers appeared to be the better choice These cards share the video memory Part of the video memory was used to capture and hold the image while the remainder was used by the computer display This arrangement allows for very fast and efficien
81. rface in the optical path Finished Side ee Figure 3 Scanning Mirror 2 Scanning Mirror Housing The scanning mirror housing seen in Figure 4 was a 127 mm diameter aluminum tube with wall thickness of 4 8 mm The height of the housing was 152 4 mm with a 4 8 mm plate filling the top and another 4 8 mm plate 28 6 mm from the bottom allowing room to mount the absolute encoder Pressed into the center of each of the plates was a bearing that has an internal diameter of 6 4 mm for the scanning mirror shaft There are two sections cut out of the side of the scanning mirror housing tube each section was cut such that it produces a square aperture 90 mm by 90 mm The two openings were centered on the optical axis and were 90 degrees apart A scanning mirror holder seen in Figure 5 was made of 19 mm honey comb aluminum At the top and bottom of the holder was a 6 4 mm steel shaft that slides through the upper and lower bearings where the stepping motor and absolute encoder attached The holder was designed such that when the scanning mirror was in the holder its finished surface bisects the center of the scanning shaft Along the sides of the holder was a 6 mm removable aluminum bar that over lapped the scanning mirror by approximately 2 mm The holder design allows the scanning mirror unlimited rotation while preventing any torque from being placed on the mirror 10 YWININV I 0 200 X 39Nv u3101 SSHINI NI
82. rner the telescope mirror was mounted Using the telescope mirror the location of each component was determined It was my job to place the individual components determine the dimensions of each component that had to be made and draw sketches of these components I didn t have the time to convert these drawings to blueprints that could be handed to a machinist Therefore help was needed and the decision to hire someone was made Jay Adeff a member of NPS Physics Department was hired to help with the physical design of the case slit holder and scanning mirror assembly Jay took my dimensions and sketches and turned them into professional blue prints Jay s assistance was invaluable without which NUVIS would not have been completed With completed blue prints we needed a machinist Glen Harrell a member of the NPS Space Systems Academic Group was hired to do the machining of NUVIS Once the machining was completed the parts were anodized black Figures 16 and 17 show the completed NUVIS As can be seen extensive weight relieving was used over the entire exterior of NUVIS The final dimensions of NUVIS were 62 cm by 39 cm by 37 cm by 28 cm by 26 cm by 19 cm high The final mass was about 14 kg 24 gt 4 gt LI dli Figure 16 Top View of NUVIS 25 26 ELECTRICAL MECHANICAL COMPONENTS A STEPPING MOTOR The instantaneous field of view produce a frame that included t
83. s machined while on the opposing side holes were drilled and tapped for the encoder The stepping motor and encoder were attached with the encoder using the motor shaft This test stand allowed me to test and experiment with both the encoder and the stepping motor and its controller driver The test procedure I used was simple but it would allow me to fully test both components I first set both the encoder and the stepping motor origins so that they coincided Then I wrote a short program using the stepping motor controller driver built in language that would generate a random number of steps in either the clockwise or counter clockwise directions I then allowed this program to run for over 24 hours At the end of the test I 32 compared stepping motor index with the encoder index After nearly one billion steps the encoder and stepping motor index still coincided exactly I felt that the repeatability of the two components was more than adequate 33 IV ELECTRICAL COMPONENTS A IMAGE INTENSIFIED CCD CAMERA 1 CCD Camera The search for a CCD camera was exhaustive After nearly two months of research into cameras manufactured by over two dozen companies I decided on the Pulnix TM 745e This decision was somewhat arbitrary Most manufactures had cameras that were indistinguishable from one another The requirements that I finally decided on were a black amp white camera with a 2 3 in CCD that had more than 640 by 480 p
84. t each data matrix would have the same number of elements and then multiply all the matrixes together The software that I used was MATLAB by MathWorks The results of my simulations proved surprising Both the UG 5 and UG 11 turned out to be much better than had been originally thought The image intensifier sensitivity with the UG 5 filter seen in Figure 11 provided a sharp cutoff at 300 440 nm with only a slight sensitivity appearing at 650 750 nm The image intensifier sensitivity with UG 11 filter seen in Figure 12 had an even sharper cutoffs at 300 400nm but the sensitivity was only half of that found when using UG 5 decided to sacrifice out of band contamination due to a wider bandwidth of the UG 5 for the higher sensitivity Despite my 17 decision to use UG 5 I felt both filters were worth purchasing in case I experienced difficulties with the extended bandpass of the UG 5 filter Image intensifer sensitivity with UG 5 25 Sensitivity mA micron 1 o 0 200 300 400 500 700 800 900 1000 600 Wavelength nm Figure 11 Sensitivity with UG 5 Filter The filter physical dimensions were determined by adding the window dimensions tolip size The result was a filter 100 mm by 80 mm that served as a band pass filter and as a window It had to be sufficiently thick to prevent easy breakage therefore the thickest variant of theses filters 3 mm was used Two additional physic
85. t image display While one image was captured the previously captured image was displayed all without the need to move the images into system memory This arrangement prevented loading down the PCI bus and therefore freed the computer to work on other tasks The only time that data would be required to be transferred over the computer s PCI bus was during actual recording With little experience I decided to purchase a frame capture card with built in driver from the same distributor JKN Electronics Inc where we purchased the camera This card had the added benefit that it an interface that allowed direct control of the Pulnix TM 745e camera shutter speed and frame integration I installed the card we purchased in a computer loaded the drivers and example program and connected the camera Everything worked great from the display to the camera control Believing one problem had been solved my attention turned to other components 39 About a month latter it was time to begin programming This was when I discovered documentation and software were as important if not more important than the hardware The documentation that came with the card was sparse and what there was written for an expert Visual programmer After two months of encountering one problem after another with this card I ran up against one final obstacle I had discovered that there was a trade off between having driver on the frame capture car
86. t was the photocathode made of modified S 20 semiconductor material The standard S 20 semiconductor material has been modified to shift the peak sensitivity toward the UV and minimize the sensitivity in the visible and infrared regions The photocathode was sensitive to light between 200 700 nm as seen in Figure 10 At the image intensifier output was a P 20AF phosphor screen The screen produces output light that peaks between 500 630 nm as shown in Figure 14 corresponding to the camera s peak sensitivity of 500 650 nm as shown in Figure 15 2 P 20AF Phosphor Emission Characteristics 0 9 0 8 0 7 0 6 0 5 0 4 0 3 Relative Spectral Emission On 01 0 480 500 520 540 560 580 600 620 640 660 680 Wavelength Figure 14 P 20AF Phosphor Emission Curve From DEP 1994 TM 745e CCD Sensttivity 0 9 08 Du 0 6 05 04 Relative Sensitivity Urs 0 2 0 1 400 500 600 700 800 900 1000 1100 1200 Wavelength nm Figure 15 TM 745e CCD Sensitivity After Pulnix 1995 22 9 Fiber Optic Coupling The image intensifier input and output windows were 25 mm in diameter where as the camera s CCD was a rectangle 10 mm by 9 3 mm Some form of coupling was needed between the two components There are two choices for coupling optical components such as this lens coupling and fiber optic coupling Fiberoptic coupling was chosen over lens coupling because it was physically smaller more efficie
87. the standard two dimensional spatial scene and the corresponding spectra of each scene Hyperspectral imaging is a relatively new and fast growing field with both commercial and military applications Commercial applications vary from vegetation identification and mapping surface geological identification and mapping to atmospheric composition and mapping Military applications include target identification and classification airborne chemical identification and mapping and rocket plume identification This thesis describes the design and operation of the NPS Ultraviolet Imaging Spectrometer NUVIS NUVIS is a hyperspectral imaging spectrometer designed to investigate the ultraviolet region of the spectrum NUVIS is comprised of a scanning mirror telescope assembly using an off axis parabolic mirror a slit a flat field imaging diffraction grating an image intensified camera assembly and the support controlling electric and electronic hardware and software This is part a continuing project to build test and use this sensor in support of military and government agencies 14 SUBJECT TERMS Hyperspectral Imaging Ultraviolet Imaging Spectrometer NUVIS Support to Military Operations Support to Government Agencies 15 NUMBER OF PAGES 96 16 PRICE CODE 20 LIMITATION OF ABSTRACT 17 SECURITY CLASSI 18 SECURITY CLASSI FICATION OF REPORT FICATION OF THIS PAGE FICATION OF ABSTRACT Unclassif
88. tor part number AM23 150 2 and the stepping motor controller driver were purchased from Advanced Micro Systems AMS The motor was a Frame Size 23 bipolar 4 lead two phase series wound hybrid stepping motor It has a natural 1 8 degree per step which equates to 200 steps per revolution with 150 oz in holding torque and 3 3 oz in sec rotor inertia It operates at a maximum of 10 volts and draws a maximum current of 2 8 amps It has a 6 4 mm diameter shaft with a length of 20 5 inch length The motor was chosen by first estimating torque needed to turn the scanning mirror 9 mradians per second Nine mradians per second corresponds to 30 IFOV per second This was driven by the camera s 30 fps maximum frame rate The torque value was then tripled 28 to take into account scanning mirror holder that had not yet been designed Given this value I chose the a stepping motor that had a 15 percent higher torque value than needed NUVIS was initially designed to scan horizontally so my calculations for the scanning mirror torque were for horizontal scanning 1 gravity was not a factor Later when given the opportunity to image the plumes of solid rocket motors at NPS Rocket Range we needed to scan vertically The motor torque was insufficient to step and hold the scanning mirror during microstepping at 25 600 steps per revolution The stepping motor needed to be replaced with a higher torque motor and the scanning mirror and the scanning
89. ve been inconvenient to have the filter mounted internally making it nearly impossible to change the filter while in the field Despite this CO eT ens I researched the availability of both bandpass and short pass interference filters I found several filters that met the optical requirements but none that were manufactured in the sizes needed The only way to get interference filters large enough was to have them custom made but the cost was not worth the benefits This forced a change in tactics I decided to place the filter outside the telescope and if possible use the filter as the input window This had two advantages 1 it eliminated the need for a quartz window and 2 it made it easy to change the filter The down side was finding an absorption filter that met both the optical requirements and the even larger size requirements of the entrance window My search failed to find bandpass or shortpass filters that had the optical properties 14 needed There were several filters that came close to what was required and were available in custom sizes all for a reasonable cost I decided to simulate the spectrum of the output of the image intensifier using these filters To do this I needed the solar irradiance at the earth s surface the transmittance curves of the filters the reflectivity curves of the mirrors and the sensitivity curve of the image intensifier The solar irradiance data at the surface seen in Figure 7 was obtained using
90. via ort roe XE Ae s 4 1 v L4 E 1 2 pot 4 IRR RS e qu FED OD oro YT proie ghe fto ine atia M s SL PEOR rs acd 72255 444794 is poten Ua edt pide e uy ey rii t raes A NESE 1 74682471341 cat te d E He ee AT OY 16 proe ET EL Vere md seg n qo yo 1 oo gt 46 4 0 AT ae 174704443 eae reel nerds no e TET ap v 0 EHE D pra tuy vea wt vg u bn sf eh tie TEE 42 5 pag Lic Bebe bes tr na t geen 19998 4 4 en er avy E TA oF DR 57084 e M eri NT E pe i 92 wir t tmt pta yt nn phe a arde RO i seo OS i at s HR PH pi awali TEST Eh fedus me a mee Ph Mare m t prt 24 9 4 TERT bl
91. ware development I was thoroughly familiar with ANSI standard and with felt that with the aid of several good programing books and tutorials I would be able to learn what I needed to design the Graphical User Interface GUI that I envisioned for NUVIS control As discussed above this assumption was wrong I did find several references that did a good job of teaching the visual side of the programming but these books all but ignored the integration of the actual code After several months of very slow progress I decided that there had to be a better way to get the job done What I found to use instead of Visual was Visual Basic VB B VISUAL BASIC A new version of Visual Basic Version 5 0 had just been released It had a performance was on par with earlier versions of but the visual side of programming was much simpler Visual Basic is a powerful compiler based programing language that allows the development of Windows programs using a simpler development environment than Visual Within a month of receiving the newly released version of VB I had developed a windows program shown in Figure 30 that allowed me to view live video capture and save and capture and save sequences of images 35 Nuvis Control 3 i 573 3 n SEN est t ONE E 3 E E d oS 1121 Figure 30 NUVIS Control Software Interface The program gives the user control of a

Download Pdf Manuals

Related Search

Related Contents

  平成27年度公立大学法人宮城大学 グリーン購入の推進  440bx mainboard  Progress Lighting P2783-77 Instructions / Assembly  User Guide - Colorbeam    filtersystempoint 2-spa-170412-01.indd  Regolamento di scalo  Single Channel Mini DVR with Built-In Camera  Samsung NP-NC213S User Manual (FreeDos)  

Copyright © All rights reserved. Failed to retrieve file