UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Design and construction of a vacuum spectrograph Lubzinski, James Francis 1950

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1950_A8 L8 D3.pdf [ 53.97MB ]
Metadata
JSON: 831-1.0085423.json
JSON-LD: 831-1.0085423-ld.json
RDF/XML (Pretty): 831-1.0085423-rdf.xml
RDF/JSON: 831-1.0085423-rdf.json
Turtle: 831-1.0085423-turtle.txt
N-Triples: 831-1.0085423-rdf-ntriples.txt
Original Record: 831-1.0085423-source.json
Full Text
831-1.0085423-fulltext.txt
Citation
831-1.0085423.ris

Full Text

DESIGN AND CONSTRUCTION OF A VACUUM SPECTROGRAPH by JAMES FRANCIS LUBZINSKI A Thesis submitted i n partial fulfilment of the requirements for the Degree of MASTER OF ARTS in the DEPARTMENT of P H Y S I C S <0' THE UNIVERSITY OF BRITISH COLUMBIA October, 1950 THE UNIVERSITY OF BRITISH COLUMBIA V A N C O U V E R . C A N A D A D E P A R T M E N T O F P H Y S I C S October 12, 1950. £r. L. W. Dunlap, Librarian, University of Bri t i s h Columbia. i)ear Dr. Dunlap: This letter w i l l certify that the thesis of Mr. James F. Lubzinski has been carefully studied by the under-signed, and that the thesis meets the required standards and an abstract has been approved by the Department. Yours sincerely, G. M. Shrum Head of the Department GMS:lc A. M. Crooker Professor of Physics DESIGN AND CONSTRUCTION OF A VACUUM SPECTROGRAPH by James Francis Lubzinskl ABSTRACT A vacuum spectrograph i s designed to cover from the visible to the far ultraviolet, ^he general path equation i s expanded as a power series to the fourth power. Conditions for minimizing this path are given to the third power, 'I'here are two different settings in the visible r e g i o n — a grating of 576 lines per millimetre i s used and the s l i t i s set at nor-mal incidence of 20°; in the far ultraviolet, a grating of 1152 lines per millimetre i s used and the s l i t i s set at glancing incidence of 80°. The housing and vacuum system d i f f i c u l t i e s are ex-plained and figures are given showing how some of these are overcome. The optical parts were designed to f a c i l i t a t e in focussing, stress being made on designs which are free from machining d i f f i c u l t i e s . Although the grating and plate holders are different from those previously used, the s l i t design i s such that i t i s completely free from mechanical d i f f i c u l t i e s usually encountered. A plate diaphragm is used to increase the number of exposures. The complete set of drawings used in the construction of this apparatus are given in the appendix. A B S T R A C T A vacuum spectrograph is designed to cover from the visible to the far ultraviolet. The general path equation is expanded as a power series to the fourth power. Conditions for minimizing this path are given to the third power. There are two different settings in the visible region -a grating of 576 lines per millimetre is used and the s l i t is set at normal incidence of 20(->; in the far- ultraviolet a grating of II52 lines per millimetre i s used and the s l i t i s set at glancing incidence of 80°. The housing and vacuum system d i f f i c u l t i e s are explained and figures are given showing how some of these are overcome. The optical parts were designed to f a c i l i t a t e i n focussing, stress being made on designs which are free from machining d i f f i c u l t i e s . Although the grating and plate holders are different from those previously used, the s l i t design is such that i t is completely free from mechanical d i f f i c -u lties usually encountered. A plate diaphragm is used to increase the number of exposures. The complete set of drawings used i n the construction of this apparatus are given in the appendix. Jrage i ACKNOWLEDGEMENTS The author wishes to express his indebtedness to Doctor A.M. Crooker, under whose supervision this work was carried out. Indebtedness to Mr. S.C. Argyle for his assistance in the early stages of design is also hereby acknowledged. Thanks too are due Mr. A.J. Fraser and Mr. W. Mayer for technical assistance; to Mr. G.G. Lockhart for pattern making and welding; and to Mr. E.M. Price for assistance with el e c t r i c a l circuits. Gratitude is hereby expressed to Mr. T. Mouat, Mr. T. Dauphinee and Mr. T. Collins for information of a general nature. The author wishes to express his gratitude to Dr. S. Maddigan for the interest he had taken in this project and for obtaining various parts by means of a grant from the British Columbia Research Council., In conclusion the author wishes to acknowledge with his sincere gratitude the grant-in-aid he received from the National Research Council of Canada. J.F. Lubzinski The University of British Columbia October 2, 1950. Page i i INTRODUCTION The objective, to measure precisely certain wave lengths i n the far-ultra-violet, brought about the need to build a vacuum spectrograph which is capable of producing results at least comparable in accuracy to that of previous investigators. Before deciding on any design in particular not only were the instruments of previous workers investigated but also the f a c i l i t i e s at our disposal were studied. Y/ith these findings i n mind the plans in some cases had to deviate from the more standard types of design. The reasons for using a concave diffraction grating vacuum spectro-graph are rather obvious. To determine wave length to a greater degree of accuracy, the index of refraction for a i r must also be known to a greater degree. With a vacuum spectrograph this can either be determined or entirely neglected. The second important factor is the high absorption of short wave length by a i r which reduces the intensity. Furthermore, in order to eliminate the determination of index of refraction of any sub-stance, i t is necessary to have the entire light path i n free space. This fact automatically eliminates the use of prisms and source windows. This spectrograph was designed with two alternative positions for a source, thus giving scope for a wide range helpful in setting up the instrument by making i t possible to start in the visible region. C O N T E N T S Page No. Acknowledgements i Introduction i i List of Plates iv List of Figures v CHAPTER I The Concave Grating Theory . . . 1 Focussing Test CHAPTER II Spectrograph Design 12 CHAPTER III Design of the Spectrograph Housing 16 CHAPTER IV The Vacuum System 25 CHAPTER V Detail. Description of Some of the Principal Parts 55 The Spectrograph S l i t The Grating Holder The Plate Holder The Plate Diaphragm Bibliography . v i List of Working Drawings Appendix A' Working Drawings for Spectrograph Appendix B Page iv L I S T OF P L A T E S Plate Following No. Page No. The Vacuum Spectrograph showing . opening for loading plate and setting up.apparatus.in general ... 15 II The Vacuum Pump showing sylphon connections and anti-vibration mountings . 22 III The Spectrograph S l i t removed from i t s housing ... J2 IV The Spectrograph Grating Holder in operating position 57 V The Spectrograph Plate Holder shown from the side from which plates are inserted ... 39 VI The Plate Diaphragm 41 L I S T O F F I G U R E S Page v Chapter Figure No. No. Page No. II III IV 1 (a) (b) (<0 (d) (e) 2 (a) 3 (a) (b) (c) (d) (e) (f) (g) 4 (a) (b) (o) (d) (e) (?) (g) 00 (i) 0) 00 5 (a) (b) (c) (d) (e) (f) (g) 00 (i) (j) Plot of the function "h" Plot of the function " f " Plot of the function 'Lf^  " Spectrograph plan Spectrograph elevation S l i t connection Operating position of the Spectrograph The main internal member Levelling gear and recoil mechanism Under carriage plan Pumping system diagram The trap valve The trap valve activator The interlock system Wiring diagram of interlock system Control panel Pressure gauge circuit diagram Pressure gauge calibration Fore-pump mounting Shock absorbing system Rate of pumping The S l i t Mechanism Schematic diagram of s l i t mechanism S l i t controls S l i t Housing The component parts of the s l i t S l i t calibration Component parts of the grating holder Plate holder adjusting block Component parts of an adjusting block Diaphragm l i f t i n g mechanism 2 7 8 10 10 15 16 17 18 19 20 21 22 23 25 25 26 27 28 29 30 31 31 32 35 34 35 36 37 37 39 4o 41 42 Page 1 C H A P T E R I THE CONCAVE GRATING THEORY Since the wave lengths to be measured are very short, and also in order to get high dispersion, large angles of incidence w i l l have to be used. To be able to work in this region a complete understanding of the concave grating theory is essential. Various references were consulted and checked for the length and width of grating and the size of s l i t to be used. Of the many articles read and evaluated, Zernke's was found to be mathematically complete and straightforward. Therefore his theory, being the most applicable to this work, is here developed and expanded to the degree deemed necessary for the size of angles used i n this spectrograph. Fermat1s principle of least time is applied to a general path of light of any order M; that i s , the conditions that must be applied to make a general path an extremum. Take the origin of a cartesian system at the center of the grating so that the radius of curvature is at x «- R and the ruled lines are parallel with ZTso that a plane parallel with X:and Z cuts the sphere of radius R at the ruled lines. The co-ordinates of the three points being the Page 2 source A(x,y,z), a point P on the grating P( , , ), and the image B(x',y*,z'). The center of the circle is then (R,0,0). The optical path length for the central or zero order image is equal to AP plus: PB. From grating theory the increase in path between rulings is (d sine <P B?n^) for image of theiwth order is»»A. Hence increase i n path length in centimeters is Nm where N is number of lines per centimeter. Hence a general path is V: = AP 4- PB-. + Hm.A/> where the rulings are equally spaced by P. y Figure 1 (a) v/= f ( x - * ) 2 + ( y - ^ 2 * (*-.) 2.)* That i s Page 3. Since we are interested in finding the minimum path for a fixed grating then the function "V" must be expanded in powers of the variables of the grating p and q, the height of the s l i t z , and the image z'. Let : x: = r cosine 9 y » r sine ^ x1 a r' cosine y 1 • r 1 sine ^  1 and since P. lies on a circle of radius R then the condition. (R-$) 2 \ p? 4. q2 B g2 may be used; but since % is small then % may* be neglected whereby we have S = 2 ^ ( P 2 * q 2) making this substitution into the general path equation we have the following v H r h f - c o s ^ ( P V ) + _ - 2 s i n ^ P - 2 - a 7 * *• r5 R r? . ' E 2 j * r ' / i - ( p V ) + -!| - 2 ^ - f - 2-^7 r'* R r 1 2 r 1 rt«w which when expanded in ascending powers of p, q, z, and z' i s V = r + r' - (sirifr + s i n ? 1 + Nm«A)P 2 2 1 1 x 1 (1 + .1 f cos 9 t cos y McoBg> _cos <y> \ _2 2 r *" r 1 P R Ft ' + 1 (1 + 1 C O B y _ cos ^ ' \ 2 2 r r'~ R ~ R + | ( | 2 ) f | ( ^ ) - *£ 1 /sinfr cos 2 y sin g»' cos 2 9>x sin9> cos y _ s i n ^ cosy'  ¥ 2 r2 r' 2 rR r'R •f i (siny> 4. sinjP ' _ cosy sin9» - cos p 1 sing> 1 ) pq 2 2 "72 riz r R r i R Page 4 • 1 / s i n ^ N ? . 1 ^sin g>' . . ? - ) Pq Z - ) PqZ;« _ 1^  1 + 1 + r cos2g> + r'cos 2p ' cos g> _ cos » ' _ 5 sin 29* _ gs 8 r2 r ' 3 - R 2 R 2 rR - r'R r2 + g s i n 2 g cosy + £• s i n 2 ^ 1 cosg' + ^ s i n V 4 ^ sin^g. ' ) p 4 rR. r'R - r'5 r i 5 .-. f 1 + r c o s 2 ^ ^ r'cos^p' „ cos9> _ cos g>' ) q4 8r> • 8r'2- -8R2 - -8R2 - -4rR • -4r'R- . 8r5 8r«? - (__L_ 4. 1 + r' cos2<p 1 f rcos2g> _ cos s> _ cos ' •4r5 4 r ' ? ~ H 4R^ 2rR 2r'R -„ 2. sj-n29> _ 2 s i n 2 ^ ' + ^ s i n 2 ^ cosg> .j. =5 s i n 2 p ' cos ' ) p 2 q 2 4 r5 4 r'? • 4 rR 4 • r'R 4r3 4. rR ' ' v 4 r ' 3 4' r'R ^ 4 r 3 4. r.5 4 r ' 3 4 r«? ' P f ( _ L _ _ "I °2£*L - 2 )p 2q Z :4 . M 1 ^2££l -2 9 i n ^ )p2q: 2 r 5 2 rR 5 • r5 2r '3 2? r'R 2 r? 2r3 2- Tffi 2 r 5 2' r'R-Applying Fermat's principle, the points A' and B' l i e on conjugate rays i f the function of the distance is a minimum with respect to the variables p and q. That is ^ v - & v B 0 Since here we are interested €>p dq Page 5 in the action of the grating - the variation of the path given by the function "V" with respect to the variables p and q of the grating - so the object and image points A and B are taken in the plane of symmetry XOY (Z = Z' =0). Therefore the two partial differential equations are : = (- s i n * - sinff«' + NmA) + ( 1 +1, + dp r r 1 cos2{? ^ cos.2^ 1 _ cosg> _ COS $>X ) p . r r' - K R - . f. 3 (sin^> cos 2^ 4. siny 1 cos2,ff> _ sinff> cosy - sin y ' cos^ p2 2 ' r . 2 r ' 2 R R R ' R ' ^ _1 ^sin g> sin y> ' _ cos.y sin g> - cos y 1 sin y' ) q2 2 r 2 r ' 2 - rR r'R -- 1. (_L_ 4- 1 + rcos2g> +. r' cos 2p 1 _ cos#> ..cosy' _ ^ a i n 2 ^ _ gsin 2 _ r5 r ' 3 R 2 - R 2 - rR r'R- J V r g sin2g> cosfo, 5 sin2g> 'cosy' 4. 5 sin^g> + ^ sin2*" & ' ^ p j rR r'R • - ' -r5 r ' 3 -_ / 1 + 1 + cos2g>' ^ r cos2p> _ cosg> _ cos ff»' •2r? 2r '3 2 R2 2 . R2 -rR r'R-- ^ sin2g» - 1 sin 2?'' + j> s i n 2 y cos 9> + ^  sinj?>' cos g>') 2 2 . r 5 2 r ' 5 2 • rR 2 r l R ',' H q 0 • 1 1 _ cos 9 ' r R _ C O S ? ' \ " R , •sin^» . s i n ^ ' „ cosp>sin«p - cos0»'sin^' • ) qp ' r 2 r ' 2 . rR r'R ' 1 4. 1 + r, cos2<p ± r'cos 2^' _ cos$> _ COS <f> ' •2r? 2r '3 2R2 2R2 - rR r'R ' 1 * 1 + r ' 2 1 0 cos 1 4. r: c o s ^ - cos9> - COS j& ' "2r3 2r '3 2R2 2R2 rR r'R - 1 s i n 2 ft „ £ s i n 2 y ' + ^5 s i n ^ c o s y + ]5 sin 2y' cos g>' \ p 2 a 2 r5 2' r ' 3 2 rR 2 r'R * 4 Page 6 To further simplify.the problem consider a ray where P is at the origin (x - 0) = -sin 9 - s i n ^ 1 +• Nm X 3p *1 « 0 <)q Thus the f i r s t condition is that s i n f f s i n ? ' b HmJ\ which gives the relation between the angle of incidence and the angle of diffraction (the well-known diffraction grating formula). If the above condition is to be f u l f i l l e d , focussing ( a l l rays from A pass through to B') is only possible by varying r and r'. Therefore the condition for a l l points along p (horizontal focussing) which satisfy = 0 is found by letting the coefficient of p vanish cos2ff> _ C O S J P ^ cos^y ' _ cos g>x _ Q. r ~ R: r' R . that i s , i f r s R cos g> r' - R cos 9 1 the conditions of the ^Rowland cir c l e , as shown by the diagram, and likewise the conditions for a l l points along q (vertical focussing') which satisfy — « Q are found by letting the dq coefficient of q.vanish. (1+1- - C Q B ^ - cos 9>' ) a = n r r' R R For a stigmatic image, both the horizontal and the vertical conditions, coefficients of p and q, must be simultaneously satisfied. This is only possible i f . r - °< and ^= 0; known as the Wadsworth collimator mount. Page 7 Since we are interested in resolving distances in the horizontal direction i t is not a serious disadvantage i f the image of a point source is a vertical line. However, i f the defocussing i s too great, a serious loss of intensity may result. To examine this matter the simplification of taking the points A and B on the plane of symmetry w i l l not suffice z z since the f i r s t degree terms in q, r and ~rr are missing. Adding these two terms to the above equation we have: ^ r. r 1* . R\ R- . r r 1" Simplifying this equation, since each point of the s l i t produces a vertical image, the inverse is here considered - or that each*point of the image is produced by a part of the length of the s l i t . Thu3, for the relation between the s l i t , the grating rulings and the image length, we have r z- = hq z? r 1 where h n cos g> (sin ^  tanjp + sin$> 1 tan<p') as h denotes the vertical spread here i t serves as a measure of astigmatism. The following figure shows the relation between the angles of incidence and diffraction and the degree of astigmatism. Page 8 From the foregoing diagram i t is seen that for large angles, where h>I, to keep vertical astigmatism to a minimum, h => I The angle of d i f f -raction should not be greater than 45°; hence from f i r s t order consider-ations, i t appears that the s l i t length should be about the same length as the rulings on the grating. In consideration of the high power terms in p and q, i f f- 0 e)p and T~ r 0 then the normal to the wave front, V a const is at an angle to the line AB which was determined by the foregoing conditions. The vertical displacement is given by z a r i^X The horizontal displacement i s given, by t s - L SI cos 9 1 dp Figure 1 (c) Since the vertical displacements here are much smaller than those discussed before, seeing'that a l l factors are of a higher order w i l l therefore be negligible. The terms which w i l l produce shifts in the horizontal direction are proportional to p2 and q 2 and hence lop-sided comas. The coefficient of the f i r s t is called the horizontal coma as in analogy to the vertical Ba-ge 9 coma, the coefficient of B e c o n d . The horizontal coma is given by j/lt ~ r > 3. ^ s i n p cos 2^ + sin» 1 cos2g> ' _ sin cos 9> - siny'cosj»' \p2 cos- 1 2 ~~ r 2 r ' 2 • rR • r ' R - $r' noflp> (sing» _ sin g>' ) (cos g> _ I , 2cos ' r ' r' r R and vanishes for the entire spectrom i f the s l i t is placed on the Rowland circl e ; r » R cos q> . Before proceeding to investigate the vertical coma directly, since the variables q, z, and a' are related, 9 0 a l l the terms of the 2nd order must be considered together. ^Jt' - r' I / s i n ^ ^ s i n ^ ' _ cosg>s in g> - cos? ' s i n p ' ) q2 ^ ' 2 ~ 2COSP1 L r 2 r ' 2 . rR r'R - ( 2qz.- z 2) - B i n f (2qz» - z^ 2 )] r ' 2 - J Take the case, z:^hq , where the s l i t length is longer than that which is required by that length of grating rulings; hence z may be eliminated. In order that the condition for eliminating the horizontal coma be preserved substitute the values r <=» R- cos** and r' «• R cosp 1 A^2 a ( s i n ^ ' t a n 2 ^ ' f sin2?» 'tan2£> sin f> - s i n ^ 'tan^> 'sin2j»-sin^)q2 - — - — (ta n p 1 +- sin p 'tans>' sins>- i s i n 2 p )qz! Rcos 1 So the f i r s t term is that which defines the vertical coma and when written as A t » -=- q< 2R f 2 — 'ir one can see that this may be reduced by masking the grating but, due to loss of intensity, the degree of masking Page 10 has to be considered from the plot of the function f f = sin^t 'tan 2p 1 +• sin^cp 'tan 2^ sin9>- s i n ^ ' tanS* * sin 2* 1 - sin^u 40 20 9 -2 -7 :3 . . ,* * * * i * / . - • z •'/ j t • 4 ' i i t I I ; ; , » "OS .z;oi' "' ... <>•'. 9> ' -40 ~20 O . ZO 40 Plot of the function "f" Figure T (d) Hence i t is easily seen that the angle of diffraction should be kept small. In the case where z:<^hg here the s l i t is the restricting element and each point of the image receives light from the whole s l i t length: thus q may be neglected. t = f r ' ( s i n 9) z2 _ 2cos#' r ' 2 2 cosy . r 2 The f i r s t term being proportional to z 2, this equation may be written as: i* X z 2 - J ^ - s i n y s i n ^ ' ( s - n ^ + s i n J ? I ) 2 z , 2 m? Rh 2 cos GO 40 2o 9 , '3 ~y 9 • 40 -20 ZO 4-0 Figure 1 '(e) : Plot of the function " f 1 " So for large angles of incidence a short s l i t should be used. The error proportional to zz' is the t i l t error and is zero i n the middle of a spectral line. Page 11 FOCUSSING TEST Since the image is astigmatic, a horizontal s l i t i s used in front of the focal plane and short spectral lines are then produced in the focal plane. Because of different angles of inclination the light rays must come from a definite part of the s l i t - the short spectral lines are then automatically stigmatic. If the source s l i t image is then reduced to about 1/^0 vertically and, being the same'in the horiz-ontal poeition, the errors are then magnified to about ^0 times ( a 3 i n the case where either the grating or the s l i t i s not perpendicular to the plane of the Rowland circle) so the reduction is helpful in lining up this apparatus. Since the vertical coma which produces a horizontal shift pro-portional to z2 appears in the focal plane i n the same manner as curvature of the s l i t , therefore the vertical coma may be reduced by the right s l i t curvature. Now as the horizontal s l i t is moved perpendicular to the plane (Z 1-) then the t i l t error proportional to zz' appears as the inclination of the spectral lines. Page 12 C H A P T E R I I SPECTROGRAPH DESIGN To cover a wide range, from the visible to the far ultra-violet, two gratings were available. An ultra-violet grating of I I 5 2 lines per millimetre and, in the visible region, a grating of 576 lines per m i l l i -metre. To make the most use of these, i.e. to get a high dispersion, two source or s l i t positions are used. In these two cases the angle of incidence is i n the ultra-violet at glancing incidence, and in the visible at normal incidence. The two grating equations are m\ = b(sin<p^ - sinj* 1.) for glancing incidence = b(sin^>2 * 3^-n 9*x) for normal incidence. The major problem i n selecting these two angles of incidence is to determine the minimum angle of diffraction that w i l l be required in order that some spectral lines w i l l occur i n both settings. The selection of this minimum diffraction w i l l determine the size of apparatus necessary. Page 15 Taking 80° as the largest permissible angle of incidence, and using the H 5 2 lines per millimetre and 2000 mm diameter grating, we consider m A for five different angles of diffraction: 9>x = 40 mA = 2970 A". 50 1900 60 1025 70 592 80 0 m?i increases as .1 decreases. Then taking an angle 20° from the normal and using the 576 lines per millimetre and 2000 mm diameter grating, m i s given for the same five angles: 9X • 4o m^= 5200 k "- 50 7350 . 60 9100 70 IO58O 80 11150 m yy increases as 9> increases. The spectral lines considered are the H, D and He lines which occur in the vacuum grating region with 'their higher orders in the region 2400-5000 H. D He 1641 I656.5 1215.7 1215.1 IO85.O 1025.6 921.6 Page 14 The reciprocal dispersion, or plate factor, in A per millimetre dH - b cos & where ds » Rd? ds mR . _ bd mR2 where d •=» R cos<P is the distance from the pole of the grating, is given for the above selection of diffraction angles. When incident angle i s 80° dispersion is II52 lines/mm 576 lines/mm grating grating 9= 4o 6.650 15.500 50 5.580 11.160 60 4.540 8.680 70 2.969 5-95.8 80 0.0 0.0 When incident angle is 20° dispersion is * II52 lines/mm 576 lines/mm grating grating g,- 40 6.65O lJ.JOO 50 5.58O 11.160 60 4.540 8.680 70 2.969 5.958 80 I.507 5.OI5 In the following diagram are marked : the two angles, the two s l i t pole lengths, the plate length and the angle which was selected between the common normal to the grating and the center line of the apparatus. Figure 2 (a) P L A T E I The Vacuum Spectrograph showing opening for loading plate and setting up apparatus in general Rage 16 0 H A P T E R. I l l DESIGN OF THE SPECTROGRAPH HOUSING T.b illustrate the main points to be considered in designing the housing for the vacuum spectrograph, the following diagram is a schematic il l u s t r a t i o n showing the three principal parts : the s l i t , the grating and the plate. Figure 5 (a) : Spectrograph plan Page 17 Since i t is desirable to get as much as possible of the perimeter of the Rowland circle on the photographic plates that are to.be placed inside the vacuum, the plane of the Rowland circle is therefore placed in the center of the tubular housing. A tubular housing is used' to avoid rein-forcing which would be necessary to withstand the outside pressure i f a f l a t chamber were used. It is readily seen from the diagram that this factor determines the- volume and hence the size of the pumping equipment that is necessary to maintain a vacuum. The second major problem is to provide access to the vacuum chamber to set up the spectrographic apparatus without impairing the r i g i d i t y of the spectrograph mounting. CENTRE OF GRAVtTY WOO PEN TABLC Figure 5 (*>) : Spectrograph elevation Page 18 Gutting the tank horizontally would provide an excellent mounting base for the spectrograph but the sealing flanges would have to be faced on a milling machine and such an oval seal is very d i f f i c u l t to do. Splitting the tank in any other direction, except in the circular cross section, involves d i f f i c u l t and costly machining of the sealing flanges. Manholes are undesirable, as i t is very d i f f i c u l t to set up apparatus through these unless the manholes are large enough, in which case only more seals are introduced. Conceding that the above is the more logical choice, one finds that the two requirements mentioned at f i r s t are of a conflicting nature. That i s , to get as much of the plate holder outside as oossible does not allow enough cross sectional area of the base for a very rig i d mount. Likewise, the inverse - or using a very large base - does not permit the plate holder to be sufficiently out i n the open. Figure J (c) : S l i t Connection Jrage 19 In deciding where the cut should be made the f i r s t consideration is to avoid interference with the two side tubes which support the s l i t s and the sourcest these should obviously be rig i d with the main part of the system. The ring is therefore cut just in front of these, so the size of base was calculated by finding where the center of gravity is for the beam and for the head and then by letting this be the central position of the metal base. In this way, the load on the wooden table is then only in a downward direction and hence eliminates any side stress in the wooden table below. Figure 5 (<0 : Operating position of the Spectrograph Page The wooden table is so designed that the points of load rest on the solid frame and not on the panel work of the surface. A l l joints are by mortise and tenon and each set of four legs is laced together with panels making three pedestals. These are then set in concrete so that a l l the legs w i l l be firm and evenly loaded. The steel base had to be broken just below the tank, solely for the purpose of ma chin in f- the flanges, as otherwise i t would be impossible to turn this piece on a lathe with the base extending in the forward direction. Figure 5 ( e) : The main internal member The main internal member is an aluminum casting mounted on a single point of support at the edge of the head as a cantilever, and held down at the back of the head by two levelling screws which f a c i l i t a t e levelling the beam. Since both the plate holder and the grating are mounted on this beam i t is therefore r i g i d l y fixed to the spectrograph head by three Allen head cap bolts at the back and two such bolts at the base. These bolts do not hold the load but act as lock bolts to prevent shifting of the beam. At both these points of contact lead pads are used to form a proper f i t and to act as damper for vibrations. Page 21 The removable part of the shell is moved back by an adjustable track system. In order to be able to line up the movable tank with the fixed Figure 5 (f) : Levelling gear and recoil mechanism head, two degrees of freedom at right angles to the direction of travel are necessary. The vertical movement is accomplished by raising and lowering the track by means of six levelling screws. The horizontal movement of the tank is obtained by mounting i t on the two axles by a single bearing or pinion at the mid point of each axle, both these being in line with each other and the line passing through them beine; at right angles to the face of the sealing flanges. This permits the tank to rock i n an arc of a circl e , but since the movement necessary is very small the tangential motion is sufficiently horizontal. The position of the tank is then rigidly fixed by means of two set screws i n the front axle which are on Page 22 each side of the afore-mentioned bearing. The rear axle, having no such set screws, is free to swivel about the mounting bearing and allows the tank to ride as i f on a three point suspension. 7 — L MouHttNG BEARINGS J— \ 1 i i KLJ Figure 5 (g) : Under carriage plan P L A T E I I The Vacuum Pump showing sylphon connections and anti-vibration mountings Page 23 C H A P T E R I V THE VACUUM SYSTEM A very high pumping speed is desirable for this type of apparatus where the vacuum system has to be opened rather frequently to adjust and load new plates. Besides using f a i r l y large pumps, a by-pass system, using two modified steam valves and a trap valve, has been incorporated to eliminate the need of shutting off the pumping system when a vacuum seal has to be broken. FORE - POMP STEA VALVE SYLPHON Dl FFUbloN P BAFFLt TRAPV. D Figure 4 (a) : Pumping system diagram Page 24 The schematic diagram on the previous page shows the elements of the of the two channels. The direct fore-pump connection or the by-pass has f i r s t a flexible sylphon link so that the" vibratory motion of the fore-pump is not transmitted to the spectrograph and, second, a modified 260 crane steam valve. The other path starts off with a trap valve: ' the reason for such a valve w i l l be explained later. Next there is'a water baffle to condense o i l vapor that might tend to enter the spectrograph chamber. Directly below this is the 2^0 Cenco diffusion pump; the path then goes through a similar flexible connection and valve as the by-pass to the/fore-pump . With the above system, i t is not only possible to leave" the pumping system operating when the tank is open but also possible to cut in either the fore-pump or the diffusion pump at w i l l . It should be" understood that when the fore-pump alone is hooked up, the valve between the fore-pump and the diffusion pump is shut off as there would be back pressure on the diffusion pump. Also i t should be understood that when the diffusion pump is hooked in, this valve is opened before the trap valve, otherwise back pressure in the diffusion pump would build up very fast. This eliminates the time needed for cooling off and heating up again when shutting down and starting up. I f a l l the ai r had to pass through the diffusion pump this would be necessary. Thus, after reloading the plates, i t is easily pumped down by means of the by-pass to where the diffusion pump can take over. The trap valve was designed to stay closed or close i t s e l f , due to it s weight, i f conditions require i t , even though i t has to be opened manually: a worm gear i3 used so that any desired, opening may be set. Figure 4 (b) : The trap valve Page 26 Since the engaging pin which locks together the rotating stem and the external bell-crank is held in position by a D 0 powered solenoid, the D C contact points are held together by an A 0 powered relay. This relay does not close u n t i l the pressure inside the vacuum chamber is low enough for the diffusion pump to take over: this does not allow the whole volume of a i r to pass through the diffusion pump which would break down some of the o i l . The A C relay that feeds power to the heating c o i l of the diffusion pump being a double poll also feeds power to the previously mentioned relay; but this relay does not close, even though the diffusion pump switch is thrown, unt i l the baffle water i s turned on. This inter-locks the pump and the valve with the water supply and acts as a safety Figure 4 (d) : The interlock system Page 27 Figure 4 (e) : Wiring diagram of interlock system Page 28 measure, shutting off the spectrograph from the pumping system, which stops o i l vapours from entering i f the water pressure f a i l s and stops the rush of a i r through the diffusion pump should a large leak develop. Figure 4 (f) : Control panel 3 Section one of the above panel contains the ttytee valves: fore-pump valve, by-pass valve and the diffusion pump valve;' The diffusion pump valve has green and red light indicators for open and shut positions, while the other two have rising stems which indicate their positions. The second section has the fore-pump switch, a pressure indicator which turns on the green pilot light on the panel and makes the circuit for the diffusion pump valve. Below this is the inlet to the tank i t s e l f . The third section is the diffusion pump control. The top valve is the baffle water, the second is for cooling the boiler in cases where the pump has to be turned off quickly, the last tap is for blowing out the water from the boiler cooling c o i l which has to be free from water before starting the Page 29 heating c o i l . The ammeter measures the current through the filament. The red.pilot light i s the indication whether the heating c o i l i s on the high or low range as set by the lower l e f t switch, or i f the heater c o i l has been turned on without turning on the baffle water. In the latter case, not only i s the pilot light red but, also, the heater does not go on t i l l the water has been turned on. I f the baffle water is not turned on, neither w i l l the diffusion pump valve open. The fourth section is the pressure gauge. The two top knobs are the zero setting and the sensitivity; the center is the voltage control; while the bottom two are the off and on switch and the high and low switch. The top meter is the supply voltage while the bottom meter is the pressure gauge. The following are a circuit diagram and calibration charts for the high and low ranges. Figure 4 (g) : Pressure gauge cir c u i t diagram Page 30 Figure 4 (h) : Pressure gauge calibration Page 51 The f i f t h and last section is for adjusting the plate diaphragm which w i l l be discussed later. The fore-pump mounting had to be slightly different from that used i n other cases, due to the fact that the openings are- in a horizontal direct-ion. There is a net torque of about 95 foot-pounds tending to t i p the pump over. This i s overcome by setting the pump back a sufficient distance from the front shock absorbers,, in this way producing a counter acting torque. This is analytically shown in the diagram below. < 2<a. > Figure 4 (i) : Fore-pump mounting To absorb as .much of the vibration as possible a double shock system was used. The major vibration i s absorbed in the rubber suspension, but it s natural frequency which is transmitted is absorbed by a layer of kem-pack. To prevent the pump being dragged across the floor, a socket f i t is employed as illustrated below. B mutt be $ho.ditoi~ thou U Figure 4 (j) •: Shock absorbing system Page 52 220 zoo C\ .— ICRl ^ 146 loo 1 I V \ 4 0 2 o \ • • I 1 r < 2 / £ /<? 2 0 2 2 2 ^ 2G> 2 8 /V MINUTES Figure 4 (k) : Rate of pumping P L A T E I I I The Spectrograph S l i t removed from i t s housing Page 55 C H A P T E R V DETAIL DESCRIPTION OF SOME OF THE PRINCIPAL PARTS Some of the spectrograph parts had to be specifically designed because they could not be obtained; in other cases different requirements were necessary, and in some cases i t was f e l t that better results could be obtained i f differently built equipment were used. THE SPECTROGRAPH SLIT^ There are two distinct problems in developing a s l i t for this type of spectrograph. The most important point is the age-old d i f f i c u l t y of keep-ing the two sides of the s l i t parallel at a l l times when setting the width of the s l i t . Solution of this d i f f i c u l t y was attempted by tryinc to I Firmre 5 (a) ; The S l i t Mechanism Page 34 design a s l i t , the accuracy of which would be independent of the accuracy of machining. The type of mechanism that w i l l probably give absolute accuracy and s t i l l be relatively easy to produce is shown in detail in the photograph on the previous page, and is explained by the schematic diagram below. Figure 5 (t) : Schematic diagram of s l i t mechanism Pivots A and A1 are fixed to the outer frame, while pivots B and B' are fixed to the inner cam which is; free to rotate with respect to the outer frame. The two semicircles C and C1 are held together by means of a garter spring D which exerts a force in the direction as indicated by arrows: . on these two semicircles are mounted the shutters E and E'. The knife edge X:and X1 is adjusted to be perpendicular to the plane of the Rowland circle and is at a very small angle to the line Y and Y1 which just passes the two pivots A'and A'. Now, as the inner cam with pivots B and B1 rotates with respect to A and A1, the semicircles 0 and C' move parallel to the line Y and Y', while the s l i t opens as the sine of the angle XOY. One can easily see that the accuracy is independent of the accuracy of the bearing on which the inner cam rotates and on the position Pare 55 of any of the pivots, but depends only on the accuracy of the one surface between the two semicircles, since this i s a plane B u r f a c e i t is the easiest operation t h a t can be performed; moreover, the fact i s these surfaces increase i n accuracy with use. The second major problem is to be able to set the s l i t exactly on the Rowland c i r c l e . One readily sees the advantage of having a l l controls independent of each other - the varying of one should not in any way alter the setting of any other. In plate III is shown the main bearing by which the adjustments making the s l i t perpendicular to and on the perimeter of the Rowland circle are made. Figure 5 ( c) : S l i t Controls This main bearing is moved back and forth along a chord of the circl e by means of the thumb screw M and held under tension by returning spring M'. Perpendicularity is adjusted by means of thumb screw N and spring N' which rotates the s l i t mechanism on the same bearing as above. In this case, since each movement depends on the accuracy of the surface on which the seat of the other thumb screw rides, these adjustments can not be said to be independent of each other but, when making adjustment with both these controls at the same time (that i s when both settings approach the correct Page j6 p o s i t i o n a t the sane time) the degree o f movement i n each o f the two d i r e c t i o n s becomes smaller and approaches zero as the p o s i t i o n i s approached thus the e r r o r t h a t may enter i n t o one s e t t i n g due to the movement o f the other a l s o approaches zero. Therefore i t i s not necessary to have these two c o n t r o l s independent as once they are set there i s no more need to vary them. The adjustment of the s l i t width which i s accomplished by means of thumb screw 0 and h e l d under t e n s i o n by s p r i n g 0' i s e n t i r e l y independent of the other two adjustments. This i s very necessary as the s l i t i s adjusted very f r e q u e n t l y . The e n t i r e s l i t mechanism i s b u i l t out of r o l l e d brass i n order to avoid pores. More important than t h i s , however, i s the f a c t t h a t to get a smooth surface the metal must not have any sand i n i t that would damage the machining t o o l . F i g u r e 5(d) : S l i t Housing Page 37 Figure ^ (e) : The component parts of the s l i t loo to 60 0 fit <S) d SO 40 -*> zo 20 /o •i Slit Wnth m MM Figure 5 (f) : S l i t calibration P L A T E I V The Spectrograph Grating Holder in operating position Page 58 THE GRATING HOLDER The main purpose of the grating holder is to be able to line the grating so that i t is at the focal point which makes the image and object distance a minimum. Four degrees of freedom are necessary - two trans-lational and two rotational. The grating center has to be raised up to the plane of the Rowland circle and i t has to be moved along a radius so that i t is set on the perimeter. Rotation in the horizontal plane is necessary to face the grating to the center of the circle,while rotation in the vertical plane is necessary to bring the ruled lines perpendicular to the plane. Actually more adjustments are necessary than just those to accomplish the above movements. In the cases of rotation the central point of the grating has to be brought to the axis of rotation. Just as in the case of the s l i t there is the advantage of having independent control. In the following discussion reference w i l l be made to plate IV showing the grating holder as in operating condition and the following figure where the component parts are shown dismantled. The three thumb screws A raise the main base in such a way that the central point of the grating is in the plane of the Rowland c i r c l e . More-over, since i t is a three point mounting, i t is possible to level the base so that the axis of rotation of the horizontal motion is perpendicular to the plane. Thumb screw B moves the grating along a radius of the circle, hence is locked in position by means of lock screws B' . Thumb screw 0.' rotates the grating in the plane of the circle and is locked in position by lock screw C 1. Thumb screw D rotates the grating in a vertical plane and is locked in position by lock nut D'. A l l the above movements are Page 59 returned by compression springs which, eliminate a great deal of back lash. A l l lock nuts are also spring loaded so that,when loosened,tension of the spring s t i l l holds the various ports in position. The thumb screw adjusts the grating so that i t is directly above the axis of rotation in the horizontal plane and the three cap nuts line up the gratinr- with the movements of the mounting. Figure 5 (g) : Component parts of the grating holder P L A T E V The Spectrograph Plate Holder shown from the side from which plates are inserted Page 40 THE PLATE HOLDER To line the plate so that i t is at the image point of the optical system Requires only one degree of freedom. It is necessary to line the plate in a horizontal direction only - i.e. in the plane of the Rowland ci r c l e . Vertical movement is unessential as the photographic plates are sufficiently wide to allow a shift of image on i t . The piste holder, ae shown in plate V, is made by bending two aluminum m i l s to a one metre radius. They are then laced together at each end. 3y means of the lacing bars the holder is moved along a mean radius on a dual track system* this is to avoid distortion of the holder r a i l s . Figure 5 (h) : Plate holder adjusting blocks Page 41 The following figure shows one of the adjustable tracks dismantled. The base A is mounted on three points and held down by a central bolt. In i t is milled a single "V" track. The square bar B rides in this groove and is held down by Gap G which was milled at the same time as the base. Firure 5 (i) : Component parts of an adjusting block P L A T E V I The Plate Diaphragm Page 42 THE PLATE DIAPHRAGM I n order to increase the number of exposures without having t o break the vacuum s e a l a p l a t e diaphragm, i s moved v e r t i c a l l y across the p l a t e . This mask covers most of the p l a t e except f o r a narrow s t r i p of about one m i l l i m e t r e i n the center and, when moved i t s f u l l l e n g t h , w i l l alow about f i v e exposures on the same p l a t e . The diaphragm i s mounted i n f r o n t of the p l a t e by means of p a r a l l e l linkages which are moved together by two t o r s i o n bars powered by a synchronous motor. The generator i s mounted under the l a s t panel. A r e d u c t i o n gear system i s used so that at l e a s t one t u r n i s necessary t o move the diaphragm one place: the amount of movement i s i n d i c a t e d by a vo l t m e t e r . Fipttre -5 (j) : Diaphragm l i f t i n g mechanism Page v i B I B L ' I O G R A P H Y ANDERSON, R.O. and MACK, J.E. On Compensating for the Aberrations of the Concave Grating, J. Opt. Soc. Am., Nov. 1954, 24, 11, 292. BELL, E.E., NOBLE, R.H. and NIELSEN, H.H. A Recording Vacuum Grating Spectrometer for the Infra-Red, Rev. Sci. Instr., Jan. 1947, 18, 1, 48. BEUTLER, H.G. The Theory of the Concave Grating, J. Opt. Soc. Am., May 1945, 55, 511. BOYCE, J.C. Spectroscopy in the Vacuum Ultraviolet, Rev. Mod. Phys., Jan. 194l, 15, 1 , 1 . ESTERMANN, I. and FONER, S.N. Vacuum Valve, Rev. Sci.Instr., Jan. 1947, 18, 1, 64. KAYSER, Heinrich, G.J. Handbuch der Spectroscopic, Vol . 1 , 533-559> 1905, Leipzig, S. .Hirzel (1901..) KURIE, Franz N.D. Vacuum Systems, Seals, and Valves, Rev. Sci. Instr.., Aug. 1948, 19 , 8, 485. MACK, J.E., STEHN, J.R. and EDLEN, Bengt. On the Concave Grating Spectrograph, Especially at Large Angles of Incidence, J. Opt. Soc. Am., May 1932, 22, 245. MACK, J.E. and STEHN, J.R. On the Concave Grating, Especially at Large Angles of Incidence: Addendum, J. Opt. Soc. Am., May 1955 , 25, 184." MACK, J.E. and ANDERSON, E.E. A 21-Foot Multiple Range Grazing Incidence Spectrograph. Rev. Sci. Instr. Feb. 1944,15,2,28. WILSON, Robert R. A Vacuum-Tight Sliding Seal, Rev. Sci.Instr. Feb. 1941, 12, 2, 91. ZERNIKE, F. von (Groningen), Die Abbildungsfehler Des Konkav-gitters und Ihre Hebung, Verhandelingen, Op.25 Mei 1955  Aangebaden Aan Prof. Dr. P. Zeeman, 1955, 'S-Gravenhage, Martinus Nijhoff. APPENDIX A . LIST OF WORKING DRAWINGS 1. Spectrograph assembly drawing. 2. Spectrograph head. 5. Spectrograph housing 4. S l i t tube; vacuum connections; sylphon fitting s ; beam support. 5. Beam. 6. Levelling screws; wheels and axle. 7. Mount and track. 8. Table. 9. Panel. 10. Recoil mechanism; cover plate; track levelling blocks; lead pads. 11. Trap valve mechanism; bell cranks. 12. Diffusion pump locking pin. l j . Assembly drawing of s l i t . 14. S l i t housing. 15. S l i t cover and source connection. 16. Detail s l i t mechanism. 17. Assembly drawing of grating holder. 18. Detail base plate; sides and bearing. 19. Detail of screws. 20. Assembly and detail of plate adjusting blocks. 21. Assembly of plate diaphragm. 22. Detail of diaphragm bearing. 25. Torsion bars. A P P E N D I X B WORKING DRAWINGS USED IN THE CONSTRUCTION OF THE SPECTROGRAPH Dull s/$ kolas aquQlt/ spaced on a 2.1 o'lama-ta.r circ/e. FRONT VIEW SIDE VIEW CROSS SECTION W« Id M Q c / ) i * i c d washa.t to tha bcctrmq p\q tu. then dull and * e a m - for a 3/+ ba-anrty rr ^ PART P T H E C O N T R O L C O V E R I i l l Ik 3 " =n7T -^TT k\ I I.\i\l • i A T E MO PAF E C T PART B M O V E S T H E S L I T ( I N T H E PLAIN) A C R O S S T H E R O W L A N D C I R C L E 1 B Y M E A N S OF T H U M B S C R E W I-l A N D R O T A T E S T H E S L I T A B O U T A N A X I S I N T H E P L A I N O F T H E R O W L A N D C I R C L E B Y M E A N S OF T H U M B S C R E W 1-2 V iiWiMlil St t Iff ^ DRILL FOR 3/8 BOLT 6 H O L E S E Q U A L L Y S P A C E D ON A 4 2 D I A M E T E R C I R C L E PAR T E R O T A T E S IN T H E P f t A N E OF T H E ROWLAND CIRCLE BY C A P S C R E W N) A N D S P R I N G M AND F I X E D B Y S P R I N G L O A D L O C K N U T F P L A N SPACE 6 - 1 / 8 N F F L A T H E A D BOLTS ON A 4 - 5 " DIAMETER C I R C L E DRILL a T A P 3 HOLE S EQUALY SPACED ON A * 3 " DIAMETER CIRCLE F O R A 1 / 8 - N F B O L T SIDE VIEW mm m © T 1 h p o <y I J . -Ii H I 89 END VIEW • U J J U U L I • 2 8 — I U u U L J Hj •9 p F 1 1 ! l O 1 i A ro 4 5 " * I1 — —>• 1 OF 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0085423/manifest

Comment

Related Items