UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The Triumf Second Arm Spectrometer System (SASP) Chakhalyan, Jacques A. 1995

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_1995-0170.pdf [ 5.58MB ]
Metadata
JSON: 831-1.0086807.json
JSON-LD: 831-1.0086807-ld.json
RDF/XML (Pretty): 831-1.0086807-rdf.xml
RDF/JSON: 831-1.0086807-rdf.json
Turtle: 831-1.0086807-turtle.txt
N-Triples: 831-1.0086807-rdf-ntriples.txt
Original Record: 831-1.0086807-source.json
Full Text
831-1.0086807-fulltext.txt
Citation
831-1.0086807.ris

Full Text

THE TRIUMF SECOND ARM SPECTROMETER SYSTEM (SASP) by Jacques A. Chakhalyan B.Sc, Diploma, M.Sc, St. Petersburg Technical University, 1988 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN THE FACULTY OF GRADUATE STUDIES (Department of Physics) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1995 © Jacques A. Chakhalyan, 1995 In p resen t ing th is thesis in partial fu l f i lment o f t h e r e q u i r e m e n t s f o r an advanced d e g r e e at the Univers i ty o f Brit ish C o l u m b i a , I agree that t h e Library shall make it f reely available f o r re ference and s tudy. I fu r ther agree that pe rmiss ion f o r ex tens ive c o p y i n g o f th is thesis f o r scholar ly p u r p o s e s may b e g r a n t e d b y t h e head o f m y d e p a r t m e n t or by his o r her representat ives. It is u n d e r s t o o d that c o p y i n g o r pub l i ca t i on o f this thesis fo r f inancial gain shall n o t be a l l o w e d w i t h o u t m y w r i t t e n permiss ion . D e p a r t m e n t The Univers i ty o f Brit ish C o l u m b i a Vancouver , Canada Date 2- Q. O ¥• G 3 DE-6 (2/88) Abstract The Second A r m SPectrometer (SASP) is a 130 ton 769 M e V / c QQClamshe l l magnetic spectrometer which pivots at the T2 location on the beam line 4 B , and operates simultaneously in a coincidence mode with the existing M R S spectrometer. The new spectrometer was installed during the 1993-1994 and has undergone a series of commissioning runs as a single arm spectrometer. In this thesis two periods of commissioning, A u g . 13th to Sept. 23th and Nov. 18th to Dec.23th are described. The presented materiel is mainly con-centrated on the first period which was devoted to the electronics, locating the S A S P focal plane, and resolution studies. The second period is presented as a stage of development of different triggers and optics studies. The results obtained are described and analyzed. Al though S A S P was carefully designed to be operated over a wide range of angles, there is a problem with trying to operate at small angles. The extended field of the first Q l quadrupole is the major source of this problem. In the last Chapter the minimum possible angle for safe operation of S A S P is determined. The calculations are based on the developed model of the extended Q l magnetic field. ii Contents Abstract ii Table of Contents iii List of Tables vi List of Figures vii Acknowledgments xi Dedication xii 1 Introduction 1 2 Beam Extraction and 4B Line. 5 2.1 Main Characteristics of the 4B Line 5 3 SASP Detectors. 11 3.1 Front End Drift Chamber (FEC) 11 3.2 Vertical Drift Chamber (VDC) 15 3.3 Trigger Paddles 16 3.4 The S l Scintillator 17 4 SASP Optics. 18 4.1 General Consideration of SASP Optics 18 4.2 SASP optics 26 4.3 Initial attempt at the experimental determination of SASP optics. 30 5 SASP Detector Read-Out. 54 5.1 SASP Trigger 54 5.2 SASP Data Stream 59 i n 6 Interconnection between Data Acquisition System and N O V A . 64 6.1 V D A C S - Data Acquisition System at SASP 64 6.2 Data Information Decoding at Chambers 71 7 Auxiliary Software. 84 8 Small Angle Operation. 87 9 Summary 104 References 106 Appendix A 109 Optimization of the Optical Aberration Corrections (Theory). . . . 109 Appendix B 111 Optimization of the Optical Aberration Corrections (Practical Ap-proach) I l l Appendix C 120 Listings of the user subroutines for aberration corrections 120 Appendix D 137 Brief summary of parameters of the C A M A C modules used at SASP. 137 Appendix E 139 A sample of a twotran source code 139 Appendix F 142 References to $RAW vector in NOVA 142 Appendix G 144 i v T i 2 6 . Effect of high-order aberrations 144 Appendix H 146 DATA ANALYSIS ON UNIX W I T H NOVA 146 9.1 A little bit of UNIX 146 9.2 L O G I N and L O G O U T 146 9.3 C H A N G I N G PASSWORD 146 9.4 FILES and DIRECTORIES 147 9.5 C R E A T I N G and D E L E T I N G D I R E C T O R Y . 147 9.6 D E L E T I N G file (files) 147 9.7 M O V I N G FILES (the way to rename files or directories). . . . 148 9.8 C O P Y I N G FILES 148 9.9 LISTING C O N T E N T OF F I L E 148 9.10 EDITING F I L E 148 9.11 PRINTING file 150 9.12 G E T T I N G H E L P 150 10 N O V A on U N I X . 150 10.1 RESTORING T H E B I N A R Y D U M P 150 10.2 C O P Y I N G T H E DATA FILES INTO T H E DISK 151 10.3 R U N N I N G NOVA 151 10.4 L I N K I N G NOVA 152 10.5 STARTING NOVA 152 10.6 G E T T I N G DASSHELP ON A L P H A 154 v L i s t of Tables 1 Some High Resolution Spectrometers for Intermediate Energy Physics 3 2 Design Specifications for SASP 4 3 Technical Characteristics of the 4B line 10 4 Characteristics of F E C 11 5 Characteristics of V D C 16 6 Calculated characteristics of the SASP Quadrupoles. A l l num-bers are calculated for the central momentum and at the pole tip 27 7 Measured characteristics of the SASP Quadrupoles when ex-cited so that the quadrupole field equals the designed value. A l l numbers are given for the central momentum and are ex-trapolated to the pole tips 27 8 Coil and steel details of the SASP quadrupoles Q l and Q2. . . 28 9 Angular (9) dependence of the effective length of the modeled Q l extended field and the distance Z of bend for a 310 MeV proton beam 92 10 Angular (9) dependence of the extended field for a 310 MeV proton beam 92 11 Angular (9) dependence of the extended Q l field 92 12 Some aberration coefficients included in resolution optimization. 136 13 References to $RAW in NOVA.( cont'd ) 143 v i List of Figures 1 General requirements of the 4B-Line tunes.(A.Point-to-point fo-cussing. B . The effect of R12=0. C. Dispersed focus.) 7 2 Particle identification plot (PID) involving R F timing informa-tion used for determination of the particle's identity. L R F T I M E denotes T O F . L A P D S U M denotes A E . . . . : 8 3 General layout of the 4B line 9 4 The SASP detector stack 12 5 A 3-dimensional view showing the position of the planes at F E C ( A U T O C A D drawing) 14 6 VDC's schematic cross section 15 7 General SASP lay-out. Coordinate systems are shown 20 8 The SASP Dipole along with the quadrupoles Q l and Q2. . . 28 9 Measurement of the R ; J parameters in pictures. Illustrations of the techniques used for measuring of the Ki-j parameters. . . . 31 10 Geometrical relations used in locating the focal surface at SASP. 32 11 The S X F T H scatterplots and S X F histograms with correct (his-tograms on the right side) and incorrect (histograms on the left side) value of S F O C A L 34 12 Focal plane position of the SASP S F O C A L F with respect to the SXF coordinate. A l l numbers are given in 50 ^m units. The solid line is the curve of equation 9 35 13 R n dependence on the momentum deviation 6 for SASP. The solid line is a curve of fit, see reference [30] 37 14 Measured values of the R i6 coefficient versus momentum devi-ation 6 for SASP 39 15 Measured values of the R33 parameter versus momentum devi-ation 8 for SASP 41 vn 16 Scatterplot SXOYO used to determine the optical centers of F E C at SASP. LXOPOS denotes the X 0 coordinate of F E C and LYOPOS denotes the Y 0 coordinate of F E C 42 17 Histograms of <j> angle differences. The top histogram shows the difference between <j> a s determined from F E C and V D C l -VDC2 coordinates (Method A) . The bottom histogram shows the difference between <f> as determined from F E C and V D C l coordinates (Method B) 48 18 The scatterplot (top picture) of the <j> angle defined via the F E C and V D C l coordinates. The scatterplot (bottom picture) of the <j> angle defined via the F E C and V D C l -VDC2 coordinates . . 49 19 The measured R 2 i , R22, R26 and R 3 4 T R A N S P O R T matrix pa-rameters as a function of momentum deviation 6 for SASP. Solid lines are the fit curves 52 20 Cont'd. The measured R 4 3 and R 4 4 T R A N S P O R T matrix pa-rameters as a function of the momentum deviation £ for SASP. Solid lines are the fit curves 53 21 The functional part of the electronics which forms the E V E N T signal. The labels in italic font denote the corresponding N O V A variables 61 22 The functional part of the electronics which forms the L A T C H signal. The labels in italic font denote the corresponding NOVA variables 62 23 The functional part of the electronics which forms the Slow Auxiliary Trigger signal. The labels in italic font denote the corresponding NOVA variables. 63 24 Data Flow and Control over V D A C S - N O V A 65 25 Wire arrangement for the F E C . The renumbered F E C wires are labeled by circled numbers 77 viii 26 Passage of a particle through V D C 78 27 The V pattern reconstruction at the V D C chambers. The coor-dinates in brackets are the wire number and corresponding drift time 79 28 Spectrum of excited levels of 1 2 C from the (p,7r+) reaction as it is observed at the focal plane without cuts on the V D C mean time applied 79 29 Spectrum of excited levels of 1 2 C from the (p,7r+) reaction as it is observed at the focal plane with cuts on the V D C mean time applied 80 30 The histogram of the V D C mean time spectrum. The vertical dashed lines are cuts on the V D C mean time 80 31 Diagnostic spectra used during SASP commissioning in 1993-1994. The position of the planes in all histograms is the same as for the H S B A D ( 'BAD P L A N E ' ) 81 32 The real on-line (top pictures) and schematically depicted (bot-tom pictures) scatterplots of clusters vs. hits for the F E C planes. 82 33 The real on-line (top picture) and schematically depicted (bot-tom picture) scatterplots of clusters vs. hits for the V D C X I plane. The scatterplots for the rest U I , X2, U2 planes appear similar 83 34 Position of a beam dump with respect to the first quadrupole. 93 35 Calculated field map of the first quadrupole Qi when operat-ing at 1000 A, generated by TOSCA. Shown is a typical non-deflected proton beam trajectory for a SASP angle setting, 6 of 15°. The expected deflection , 7 is 3.5° 94 36 Influence of the extended field on a beam 95 37 4B beamline from T2 to the Beam Dump 96 ix 38 Envelope calculated by T R A N S P O R T for the carbon target. No multiple scattering is included. 0=20° and P=817.79 MeV/c. 97 39 Envelope calculated by T R A N S P O R T for the carbon target. Multiple scattering is included. 0=20° and P=817.79 MeV/c. . 98 40 Envelope calculated by T R A N S P O R T for the lead target. Mul-tiple scattering is included. 0=20° and P=822.99 MeV/c. . . . 99 41 Envelope calculated by T R A N S P O R T for the carbon target. Multiple scattering is included. 0=20° and P=817.79 MeV/c. . 100 42 Envelope calculated by T R A N S P O R T for the lead target. Mul-tiple scattering is included. 0=20° and P=822.99 MeV/c. . . . 101 43 Calculated dependence of entrance angle 0 vs. exit angle 7 for 310 MeV proton 102 44 Geometrical relations used in the determination of the critical angle 103 45 Influence of T126 aberration term on a focal plane 145 x Acknowledgments In writing this thesis I have constantly appreciated my debt for the huge amount of help that I have had from my supervisors Garth Jones and Patrick Walden, their insight into problems taught me much more about Physics and research style than any textbooks. Without their gracious guidance and en-couragement this work would not have been done. There are others whom also deserve special thanks, particularly David Hutcheon, Stanly Yen and Alan Ling for having generous hearts and the pa-tience to answer on all my questions. Many words of thanks must find their way to Rudy Abegg, Ed Auld, Mona Benjamintz, John Campbell, Randy Churchman, Willie Falk, Wayne Felski (deceased), Andrew Green, Peter Green, Elie Korkmaz, Andy Miller, Allena Opper and Munasinghe Punyasena. To all of you who have helped in large and small ways, I would like to say thank you very much. No words can describe the value of support from my family and my parents. I would like to express my deepest appreciation to my lovely wife, Julia, and my son, Daniel, for their patience and huge moral support. My very special thanks to Igor Strakovsky for his contribution to my out-look on Physics and on life. Finally, I would like to thank my friend Dave Joffe for endless hours of discussions. xi Dedication To my family and to my parents with love. xi i 1 Introduction Spectrometers have been widely used in nuclear physics experiments requiring high resolution to resolve details of the nuclear structure and for measurements requiring high-precision triggers to detect reactions with low cross-sections. At T R I U M F energies solid state devices can not stop particles and thus the only practical way of achieving resolutions in the 1 0 - 4 level is through the use of focusing magnetic spectrometers. Although a large variety of designs have been successfully used .for high resolution magnetic spectrometers, not many address the combination of large angular and momentum acceptance and very high resolution. Instead, most spectrometers (Saclay, S L A C , DESY etc.) are rather modest in terms of ac-ceptance and resolution, or sacrifice resolution for solid angle. The list of spectrometers used in nuclear physics around the world is shown in Table 1. The notations are as follows: A6nb - is the angular acceptance in the non-bend plane of the spectrometers, see Fig. 7. A90 - is the angle acceptance in the bend plane of the spectrometers, see Fig. 7. 6(9S) - is the scattering angle resolution of the spectrometer at the target. Q or D - are the abbreviations for Dipole (D) and for Quadrupole (Q). For instance, QQD means a Quadrupole-Quadrupole-Dipole magnet system. The main advantage of the SASP spectrometer at T R I U M F is its combined high energy resolution, large momentum acceptance and geometrical solid an-gle, as indicated in Table 1 and reference [2]. It is therefore particularly useful for those experiments involving the investigation of closely spaced energy levels in reactions where the differential cross-sections are small. S A S P is a Second A r m SPectrometer designed either for operating as a single arm high resolution large solid angle magnetic spectrometer or for use in 1 a Dual Arm Spectrometer System (DASS) in conjunction with the already existing MRS spectrometer, providing an opportunity to register with high resolution the coincident two charged particles in a final state [1]. The SASP specifications are listed in Table 2. The scope of this thesis is to describe some aspects of the commissioning of the new spectrometer facility during August 1993 - February 1994. The main emphasis is to summarize and to describe in detail the main components of SASP spectrometer system. The is logically subdivided into several sections in the following order: 1. A brief description of the cyclotrotron and 4B proton channel. 2. A general description of detectors currently available at SASP. Technical specifications together with operational characteristics are provided. 3. A consideration of optical design of SASP and a comparison to measured properties are considered. The Rf, terms extracted from measured data are compared with the theoretically predicted values. 4. A detailed description of SASP electronics and the associated Data Ac-quisition system (NOVA <-• V D A C S *-* T W O T R A N ) . The information on supplied software is also provided. 5. An analysis of the beam T R A N S P O R T for small angle SASP operation. The mock-up model of the Q l fringe field is given. Based on this model, a critical angle 0C7.,ttCa/, for which the beam deflected by the SASP Q l quad fringe field is not significant enough to require a steering dipole, is derived. 2 Table 1: Some High Resolution Spectrometers for Intermediate Energy Physics. Type P r max T o t a l r max R e s o l . An 0b R e s o l . c e n t . r a y bend 1 ( x l t T 4 ) ( i s r ) 1 6 ( 6 S ) (GeV/c) Pmt'n A P / P 8nb (mr) NIKHEF QDD 0.6 150° 1.1 ~2 5.6 1/1 2-4 QDQ 0.75 90° 1.1 ^5 16. 1/1 2-4 SACLAY D 0.6 153° 1.4 ~4 6.5 6/1 2-4 D 0.9 169.7° 1.1 1.5 4.8 4/1 2-4 BATES DD 1 90° 1.06 0 .4 -0 .5 5.4 6/1 <1 SLAC 8 GeV D 1.6 90° 1.1 5 2. 2/1 ~ 0 . 3 1.6 GeV/c QQDDQ 8 29.6° 1.08 5 0.75 4/1 ~ 0 . 3 LAMPF HRS QDMQ 1.52-2 150° 1.05 0.4 2.6 1/1 0.8 SATURNE SPES I QSD 1.9 97° 1.04 0.5 3.8 2/1 SPES IV QQDDSQ+ 4. 0° 1.01 3 2.4 3/1 MAM I B hadron QDD 0.67 110° 1.2 <1 28.0 0 .7 /1 1-3 CEBAF QQDnQ 47 45° 1.1 0.5 8.0 2 .2 /1 <0.5 4 GeV TRIUMF SASP QQD 0.716 90° 1.28 ~2 13.5 2 .5 /1 2 MRS QD 1.4 60° 1.12 1.0 2.0 0 .6 /1 ~3 RCNP QD 0.965 65° 1.3 ~2 >20 0 .6 /1 3 Table 2: Design Specifications for SASP. Spectrometer type QQD Configuration Vertical Focusing (bend plane) Point-to-Point Maximum Central momentum 716 M e V / c Design Central momentum 660 M e V / c Solid angle at 0 % A P / P 13.5 msr Resolution with 2mr 0.02% A P / P multiple scattering Radial linear magnification -0.6149 D / M 4.7 cm/% Optical length at 660 M e V / c 7.02 m Angular acceptance (Bend plane) ±103 mr Angular acceptance (Non-Bend) ±42 mr Focal plane tilt 45° Total bend angle 90° Angular resolution (No FEC) 2 mr Maximum target spot size Vertical 10 cm Horizontal 4 cm Current angular range 22°-135° 4 2 Beam Extraction and 4B Line. 2 .1 Main Characteristics of the 4 B Line. SASP is positioned along proton channel 4B and therefore, in order to op-erate successfully, the beam characteristics of the 4B line are an important consideration. The cyclotron is designed to accelerate proton beams which are both un-polarized (up to 500 ^ A ) and polarized in any arbitrary direction (up to 3.5 fiA with 70-75% polarization). The primary proton beam is produced by the T R I U M F sector-focused cy-clotron. The protons, accelerated as H _ , ions , are extracted by stripping the two electrons in a carbon or aluminum stripper foil. The radial distance of the foil from the center of the cyclotron determines the energy of the extracted beam, which is variable from about 183-518 MeV. These beams are typically extracted, in three directions, one into the proton hall, one into the meson hall and one into the vault for production of radioisotopes. A general lay-out of the 4B line is depicted in Fig. 3. A few remarks about the proton beam are worth mentioning here. First, to change the beam intensity the height of the stripper foil must be changed, in order to vary the amount of beam intercepted. Second, for effective control of the beam, careful consideration has to be given to the kind of stripper foil in use, because the radial width of the strippers determines the resolution of the spectrometer when it is not in dispersed matching mode (this operation will be described later in the text). Third, for trigger and time-of-flight applications the cyclotron radio-frequency (RF) is available for use in experiments. This frequency is 23 MHz, which means that the beam arrives at the target in beam buckets separated by 43 ns. A proton bunch can be up to 5 ns in duration. This bunch width can be varied down to 2 ns. Since the R F period is 43 ns, timing relative to the R F signal is of modulo 43 ns. This 23 MHz R F signal, available 5 for use in the experiment, has an arbitrary, but fixed phase with respect to the timing of the proton bunches. This signal provides a reference time with respect to which the time at which particles are detected at the experiment can be measured. This facilitates the determination of particle identification by Time-Of-Flight (TOF), see Fig. 2. The two distinct groups of particles can be clearly seen in this 3d-plot, thus allowing to separate the background (peaks on the right side) from the events of interest (the peaks on the left side). Briefly, the main characteristics of the BL4 line can be summarized as follows: 1. A maximum intensity of 1/xa based on present shielding limits. 2. Beam energies between 200 and 500 MeV can be extracted. 3. The 4B line makes use of 2 super-conducting solenoids and 2 benders. The beam can be polarized in any arbitrary direction at any energy. 4. The 4B line also makes use of the twister, a set of six quads that can rotate the dispersion plane of the 4B line from horizontal to vertical. The SASP spectrometer can be dispersion matched continuously (0 to 20 cm/% (AP/P ) ) . There are several additional requirements that the 4B line has to fulfill. They arise from the specific experimental conditions desired, and involve such characteristics as beam dispersion, focusing requirements and magnification. Al l of these requirements are illustrated pictorially in Figures 1 and in Table 3. Normally a small beamspot size on the target is desired. In addition, it is desirable to have a beam dispersion adjustable within wide limits, typically from 0 to -20 cm/%. The 4B channel must be focus-to-focus tuned. For more details see [3]. 6 Figure 1: General requirements of the 4B-Line tunes.(A.Point-to-point fo-cussing. B. The effect of R12=0. C. Dispersed focus.) 7 'PID' — Figure 2: Particle identification plot (PID) involving R F timing information used for determination of the particle's identity. L R F T I M E denotes T O F . L A P D S U M denotes A E . 8 9 Table 3: Technical Characteristics of the 4B line. Maximum Momentum 1090 MeV/c Dispersion for Momentum Matching 0-20 cm/% Emittance Horizontal 8.0 7T mm • mr Vertical 1.07T mm • mr Resolution >0.15 % A P / P Polarization any direction 70% Intensity <1 LlA Beam Energies 200 -r- 500 MeV 10 3 SASP Detectors. Being a general purpose magnetic spectrometer, SASP contains a number of particle detectors, each of which will be described in detail below. The general layout of detectors currently available at SASP is shown in Fig. 4. 3.1 Front End Drift Chamber (FEC). The main purpose of the F E C is to allow complete reconstruction of a par-ticle trajectory within the spectrometer with an aim to optimize momentum resolution. The F E C can also be used to define in the software, the aperture of the spectrometer and hence its solid angle. The main characteristics of the chamber is given in Table 4. Table 4: Characteristics of F E C Type of chamber Drift-Planar Active area 8 x 16 cm 2 Number of wires ( total ) 208 Gas mixture pure isobutane Thickness of windows 25 /um aluminized mylar Thickness of sense wire (X plane) 12.5 jim. Au plated W Thickness of field wire (X plane) 50.0 fim Au plated BeCu Thickness of guard wire (X plane) 50.0 (im Au plated BeCu Thickness of sense wire (Y plane) 12.5 /zm Au plated W Thickness of field wire (Y plane) 50.0 jira Au plated BeCu Thickness of guard wire (Y plane) 50.0 ^m Au plated BeCu Thickness of gas enclosure ~ 6.2 cm Operation pressure 1/3 atmosphere pressure 11 5S1 CM Figure 4: The SASP detector stack. 12 T h e F E C is a drift t ype chamber cons i s t ing of four w i r e p lanes , two i n the v e r t i c a l ( X ) and two i n the h o r i z o n t a l ( Y ) d i rec t ions . It is s c h e m a t i c a l l y de-p i c t e d i n F i g u r e 5. T h e X and Y planes are denoted as anode planes because the sense wires (the readout wires) are at g r o u n d p o t e n t i a l , i .e. , pos i t i ve po-t en t i a l w i t h respect to the H i g h V o l t a g e ( H V ) . T h e f ie ld wires on the anode p lane are at negat ive H V . T h e v e r t i c a l d i r ec t i on is denoted as X i n order to be consistent w i t h the T R A N S P O R T code [13] conven t ion . T h e X p lane is a lways i n the b e n d p lane . T h e t o t a l n u m b e r of wires i n the F E C is 208 (sense, f ie ld a n d g u a r d wi res ) . E a c h X p l ane conta ins 32 sense wires , 33 f ie ld wires a n d 3 g u a r d wires . E a c h Y p lane consists of 16 sense wires , 17 f ie ld wires and 3 g u a r d wires . W i r e spac ing is 5 m m f rom sense w i r e to sense w i r e and 2.5 m m f r o m sense w i r e to field wi re . T h e wi re pa t t e rn is as fol lows: . ..-f-s-f-s-f-s-f-s-. . . , where ' f denotes a f ie ld wi re and 's ' denotes a sense wi re . Be tween the anode planes a n d outs ide ca thode planes the ex t r eme up-s t ream and d o w n s t r e a m anode p lane are at negat ive H V and consist of 6.4 /im a l u m i n i z e d m y l a r . T h e anode to ca thode p lane spac ing is 3.175 m m . T h e anode planes are l abe led X o , Y o , X 0 and Y'0. T o remove the l e f t / r i g h t a m b i g u i t y a round the sense wires of such chambers the p r i m e d p lane ( X 0 or YQ) is offset by one ha l f the wi re spac ing w i t h respect to u n p r i m e d one ( X o or Y o ) . C o m p a r i n g w i r e number s w h i c h fire between the two planes , p r i m e d and u n p r i m e d , removes the a m b i g u i t y . F u r t h e r detai ls on me thods for u t i l i z i n g the F E C chamber i n f o r m a t i o n for software t rack r econs t ruc t ion , and other uses are p r o v i d e d i n C h a p t e r 6.2 Gas for the chamber is s u p p l i e d f rom the P r o t o n A r e a G a s H a n d l i n g S y s t e m (PAGHS)[31]. F E C operates at 250 to r r . T h e gas used i n the F E C is pu re i sobutane . H i g h Vo l t age ( H V ) for the chambers is supp l i ed by a 4032A L e C r o y h i g h vol tage supp ly p r o v i d i n g t y p i c a l l y , 1250 V at a few fj,A of leakage cur ren t . 13 Figure 5: A 3-dimensional view showing the position of the planes at FEC (AUTOCAD drawing). 14 3.2 Vertical Drift Chamber (VDC) . T h e m a i n par t of the S A S P detector sys t em is a pa i r of v e r t i c a l drif t chambers , i n s t a l l ed above the focal surface and t i l t e d at 45° w i t h respect to the reference cent ra l m o m e n t u m tra jectory. In order to recons t ruc t a t r ack , the p a r t i c l e mus t pass t h rough at least three drif t cel ls . Hence , there is a m i n i m u m t rack angle tha t the V D C s can accept . T h e V D C s ce l l geomet ry and t i l t angle are designed to detect a l l the desired rays . E a c h V D C is m a d e up of two anode w i r e planes , an X p lane and a U p lane , where the U p lane consists of wires s t rung at an angle of 30° w i t h respect to the X d i r ec t i on , see F i g . 6. Window Cathode O 0 0 0 0 0 0 c ) o o o Plane Cathode 0 0 0 0 0 0 0 c 3 o o o p i a n e Cathode Window F i g u r e 6: V D C s schemat ic cross sec t ion . T h e X p lane has 1057 wires i n a l l : 304 sense wires , 606 f ie ld wi res , and 147 gua rd wires . T h e U p lane contains 902 wires i n a l l , 288 sense wires , 574 f ie ld wires , and 40 gua rd wires . T h e d iameters of the sense and field wires are p r o v i d e d i n T a b l e 5. T h e gua rd w i r e d iameters are as fol lows: one of 300 fim A u - p l a t e d A l , twen ty of 178 / / m A u - p l a t e d A l and the r e m a i n d e r of 50 / jm A u - p l a t e d B e C u . T h e wi re spac ing f r o m sense w i r e to sense w i r e is about 6 15 mm and the space separation from sense wire to field wire is about 2 mm. The wire pattern is ...-s-f-f-s-f-f-s-... . The cathode planes, at negative H V , consist of 25 Lim sheets of aluminized mylar. The sense wires are at ground and the field wires are at negative H V . The anode to cathode plane spacing is 15.24 mm. The entrance windows of the chambers are 25 Lim mylar aluminized on one side only. The two outer cathodes are aluminized on one side and the central plane is aluminized on both sizes. The structural characteristics of the VDCs are summarized in Table 5. Table 5: Characteristics of V D C Type of chamber Drift-Planar Active area 182.4 X35 .0 cm 2 Gas mixture 50% isobutane -f 50% argon Thickness of windows 25.0 Lim aluminized mylar Thickness of sense wire (X plane) 20.0 firn Au plated W Thickness of field wire (X plane) 50.0 Lim Au plated BeCu Total thickness of chamber 11.2 cm High Voltage (HV) for the V D C s chambers is supplied by two B E R T R A N high voltage supplies providing, typically, - 9000 V . 3.3 Trigger Paddles. There are five scintillator paddles labeled as PD0, PD1, PD2, PD3 and PD4 located right above VDC2 chamber, see Figure 4. They play a major role in the definition of event triggers and can also be used for limiting an active area of the focal plane. Each paddle consists of a scintillator plastic (24" x l 8 " x 1 /8" BC412 scintillator material) attached through light guides to a photomultiplier tube (RCA C31000/8575R). The method of using these counters as a part of the trigger is discussed further in the section on the trigger electronics and 16 software analysis of data (see Chapter 5). 3.4 The SI Scintillator. The last element of the SASP detector system consists of a scintillator (Si) which is an essential element of the trigger system. The SI counter consists of a plastic scintillator (24" x84" x l / 8 " BC412 scintillator material), coupled to PMT's (RCA C31000/8575R) at each end. 17 4 SASP Optics. 4.1 General Consideration of SASP Optics. During the last two decades, important improvements have been made in magnet spectrometer design techniques with the development of more and more powerful and sophisticated computation tools in the area of ion-optics (TRANSPORT [13], R A Y T R A C E [12], R E V M O C [15], ZGOUBI [14] etc.) and also in the area of calculations of magnetic fields from steel and coil con-figurations (POISSON, TOSCA) . Although these codes allow convergence to the 'best' design by trial and error, the design of magnet spectrometers always starts from an intuitive first-guess based upon previous experience and extrap-olation from existing magnet systems [8]. The following discussion provides some simple considerations of the spectrometer optics essential for an un-derstanding of SASP, an understanding based on the first- and second-order matrix formulation of beam transport developed by K . Brown [13]. Any magnet system, no matter how complex, may be thought of as a set of magnet elements described by some transport matrix. These elements are placed along an assumed reference trajectory. From within the region in phase space occupied by the particles a single point is chosen to be the reference tra-jectory. It is roughly at the center of the distribution of the particles which is transmitted through the spectrometer. Its path through the magnetic system is known as the reference trajectory. Its momentum is labeled as the reference momentum[ll]. The coordinate system used is defined as follows (see Figure 7). The two coordinates perpendicular to the initial reference trajectory are normally re-ferred to as 'horizontal' and 'vertical'. One of the coordinates lies in the vertical plane (bend-plane). The other coordinate describes the non-bend plane. These transverse coordinates at SASP are labeled as X and Y , respectively, whereas the longitudinal coordinate is denoted as Z. A l l SASP magnet elements are 18 assumed to be s y m m e t r i c about the i r m idp lanes . It is c u s t o m a r y to refer to th is p lane as Z - X plane . 19 Figure 7: General SASP lay-out. Coordinate systems are shown. • 20 T h e passage of a pa r t i c l e t h rough a magne t sy s t em m a y be desc r ibed to first order i n te rms of a t ranspor t m a t r i x R [10]. A t any p o s i t i o n , th is charge pa r t i c l e is specified by a vector X , w h i c h is g iven by : X = 6 y <t> l where, at the p o s i t i o n Z a long the t ra jec tory : x - is the v e r t i c a l d i sp lacement of the ray. 0 - is the angle tha t th is ray makes i n the ve r t i ca l p lane ( t a n - 1 - ^ - ) . y - is the h o r i z o n t a l d i sp lacement of the ray. - i s the ho r i zon t a l angle of the ray ( t a n - 1 ^ ) . 1 - is the difference between an ac tua l ray p a t h w i t h respect to the reference t ra jec tory . 8 - the dev i a t i on i n m o m e n t u m , A P / P of the ray f rom tha t of the reference t ra jectory . A s i n d i c a t e d i n the def in i t ions , a l l of the above quan t i t i e s are g iven w i t h respect to a reference t ra jec tory . T h u s i f a magne t sys t em is represented b y the R m a t r i x and a ray by the vector X then the fo l lowing m a t r i x equa t ion represents the t raversa l of the ray t h r o u g h the magne t sys t em: RX, (1) where X/ is the i n i t i a l value of X. Fo r the S A S P spec t rometer , the X vec tor at the focal p lane is g iven by : 21 % focal 0focal V focal 4>'focal Ifocal \ fifocal ) R12 0.0 0.0 0.0 # 1 6 #21 R22 0.0 0.0 0.0 # 2 6 0.0 0.0 R33 # 3 4 0.0 0.0 0.0 0.0 # 4 3 R44 0.0 0.0 01 yi h h \ 0.0 0.0 0.0 0.0 0.0 # 6 6 / Here the ent rance ray X/, is tha t at the target p o s i t i o n , and X/ o c a ; denotes the coordinates of the ray at the focal p lane . It is clear tha t m a n y elements i n the R t j m a t r i x are equal to zero. T h i s is a ref lect ion of a few facts. F i r s t , i f the m o m e n t u m of the pa r t i c l e does not change w h i l e the pa r t i c l e passes t h r o u g h the magne t sys t em, as is genera l ly the case i n a spec t rometer , the o n l y non-zero cS t e r m is Re6, and i t is equa l to un i ty . Secondly , to first order , x and 6 are decoupled f rom y and <f>. T h i s means tha t the x / o c a / and 6focal coordina tes of the exi t ray are dependent o n l y on X; and 6] and not on yj and <f>j. T h e same s ta tement applies also to y / O C o/ and <j>focah w h i c h are independen t of X] and 61. Fo r these reasons, the values of R13, R23, R 1 4 , R24 and the i r t ransposes are equal to zero. T h i r d l y , because S A S P is des igned to have po in t - t o -po in t focus ing, the t e r m R12 = 0. E v e r y m a g n e t i c sys tem is fundamen ta l l y p lagued by h igh order aber ra t ions . It is on ly by severely l i m i t i n g A x , A y , A0, A</> and 8 to a close a p p r o x i m a t i o n of the cen t ra l (reference) ray that first order opt ics shows some v a l i d i t y . E v e n then there are h igh-order effects w h i c h are diff icul t to ignore , such as T126 w h i c h t i l t s the focal p lane , i .e. , T i 2 6 -8 acts l ike a non-zero R i 2 coefficient (for fur ther d iscuss ion of h igh-order aberra t ions see A p p e n d i x G ) . T h e a i m of the S A S P magne t spec t rometer is to measure the m o m e n t u m of a charged scat tered pa r t i c l e . B e i n g a vec tor quan t i ty , the m o m e n t u m of a pa r t i c l e is charac te r ized by b o t h d i r ec t ion and m a g n i t u d e . In order to measure m o m e n t u m w i t h a magne t i c spec t rometer , a m i n i m u m 22 of three position sensitive detectors are required, not all being at one end of spectrometer. For SASP there is the front end chamber (FEC) and a set of position sensitive devices at its exit (VDCs). For any particle ray, xj, 61 and 8j are all unknown. To solve for 81, three equations are required. From the first-order transport equation the coordinates of a particle trajectory at the SASP position detectors can be written as: XFEC = XI + L F E C • ®i (2) 9focal = R21 • + R22 • 81 + R26 • $1 where, LFEC is the distance from the target to F E C and 81 is : 81 = P l ~ P c e n t r a l • 100% (3) •* central Knowledge of 0 / O C a/ , x / o c a / and XPEC allows the determination of 81 directly using the system of equations 2. Since the magnitude of the central momentum P ' centra l is known from the calibration of SASP [27], the magnitude of the momentum can then be obtained from equation 3. The calibration of SASP was done by using two techniques [27]. 1. Using a two-body reaction such as p 1 2 C elastic scattering. With the pro-ton beam energy fixed, the momentum of a scattered particle is strictly dependent on scattering angle. Knowledge of the scattering angle (SASP angle setting) fixes P c e n t r a / - Therefore, if the B field is measured then the field can be related to P C e n < r a / as P c e n t r a / = KSASP • BSASP, where the SASP calibration is KSASP- It is a constant for all P C e n t r a / and BSASP, and was found to be numerically equal to 50.309 M e V / c k G - 1 . 2. Using a p+A—>s+m+R three body reaction and the MRS spectrometer, where s, m and R are the SASP particle, the MRS particle and recoil 23 nucleus respectively. MRS was calibrated. The momentum and direction of the MRS particle is known. If SASP is placed at the correct direction, then this fixes the momentum of the SASP particle. RSASP was set using SASPSETTING, a program predicting the power supply currents required for Ql ,Q2 and the dipole based on B vs. I curves and field maps. Given a required P c e n t r a < > SASPSETTING calculates both current and ^SASP- The current is set. Upon examining the three body reaction, the locus of particles at the focal plane of MRS are plotted against the locus of particles at the SASP focal plane. At the SASP focal plane position for which ^central is to be determined, the corresponding MRS focal plane position is found from which PMRS is determined (since MRS is calibrated). From three-body kinematics, the correct ^central lor SASP is then obtained. Although both methods of operation (with and without F E C ) are available at SASP, when the high intensity beam is needed, operation without F E C is preferable because the F E C maximum useful rate is limited at ~ 2 MHz. Hence at some sacrifice to the resolution, SASP can operate without using the F E C . This may be desirable, for example, where the cross section of the reaction is small and to obtain a reasonable data rate the beam current has to be large. In such a case, since the F E C counting rate would exceed the limit (~ 2 MHz), the F E C H V must be turned off. To the first approximation R12 = 0 (focal plane condition) and X / is small, presumably ~ 0 . Thus, x / o c a ; ~ R i 6 • Si (even with x; 7^  0, R i 6 • Sj dominates x / o c a / lor all reasonable values of x/) and therefore X / o c a / measures Si to first order. In this case the last two equations of 2 must be used to obtain Si and 9i (x; is assumed to be zero). Even in cases where X / is large, as long as the beam is dispersion matched to SASP (this method is further discussed later on), x; = 0 is a reasonable approximation. However, using the relation x / o c a / ~ R J 6 • Si to determine Si is crude and higher-order (non-linear) parameterization is 24 required to improve the resolution, i.e., 6j = A o + X / o c a / ( A 1 + x / o c a / ( A 2 + - - - ) ) -The beam resolution of the T R I U M F cyclotron (~0.15% A P / P ) is much larger than the optimum resolution of SASP(~0.02% A P / P ) . Thus in order to achieve the resolution of which SASP is capable the technique of Disper-sion Matching must be employed. This simple and elegant method allows a spectrometer to operate with an energy resolution better than the energy spread in the beam. Denoting R n as the Magnification M , Ri6 as the Dispersion D s a s p , Ss as the momentum deviation for spectrometer and Sb as the relative momentum of the incident beam, equation 2 can be rewritten in the form: Xfocal = M • xi + Dsasp • Ss (4) and Ss can be related to Sb via 5 _ _ _ _ ^ . ^ . _ _ _ ^ . _ _ . ^ t (5) ps pb ps dpb ps dpb In order to dispersion match, the beam at the SASP target must be at a momentum focus (i.e. R12 = 0 and R i 6 = D(> 7^0) such that x/ ~ Df, • Sb where D;, is the Dispersion of the beam. Combining 4 and 5 plus preceding expression Xfocai may be expressed as a function of Sb as follows: Xfocal = (M • Db + Dsasp • ^y1 • — ) • Sb (6) apt, Ps From this equation the idea of the Dispersion Matching technique can be demonstrated. Since it is desirable to make x / o c a ; independent of the beam spread, the term in brackets in equation 6 must be equal to zero. This gives the required dispersion, D^, of the beam, which is n Dsasp pb dps M ps dpb When the beam dispersion is set to this value, the spectrometer is said to be dispersion matched. Theoretically, when this beam condition is set, the experimental resolution no longer depends on the beam resolution. 25 4.2 SASP optics. S A S P consists of three optical elements, two quadrupoles followed by a dipole with a wedge-shaped pole gap which together with quadrupoles Q l and Q2 provide focusing in the axial direction. Since S A S P has a large geometrical ac-ceptance, it was absolutely necessary to set the second-order term T \ i 2 to zero in order to retain the spectrometer resolution. To do this, the entrance/exit pole curves along with quadrupole multipoles were specifically designed to make the second order term T n 2 zero [9]. The reason for this can be clearly seen from the following equation: Xfocal = Rll • X] + R12 • 0 + ... + 7 l i 2 • Xi • 0 + ... (8) In S A S P R12 is equal to zero (point-to-point focus) and, because of the large acceptance, the 6 and x j are large. Therefore it is highly desirable to have the T112 term small for x / 0 c a / to be independent of 0. The quad multipoles and the entrance and exit dipole curves also correct other aberrations such as the T126 term (which tilts the focal plane) but fixing T112 = 0 is the most important correction. Both quads are mounted on tracks that allow positioning of the elements along the platform, see F ig . 8. The first quad has an aperture of 20 cm and focuses in the vertical (bend) plane. The second quad provides focusing in the horizontal (non-bend) direction and has the same 20 cm aperture. Table 6 shows the technical characteristics of the quads as they were designed, the values listed are those at 10 cm from the center of the beam line, at the pole tips. The actual, measured magnetic field and its multipole components agree to the design specifications within errors (except for the sextupole component of the second quadrupole [4]) and are given in Tables 7 and 8. These components were measured at radius of 9.5 cm and were later extrapolated to the pole tips. As it is shown in the R A Y T R A C E study [28], this disagreement leads to an 26 increase in the F W H M of the ray bundle from 0.4673 to 0.6209 m m at the focal plane. Table 6: Calculated characteristics of the S A S P Quadrupoles. A l l numbers are calculated for the central momentum and at the pole t ip. Mult ipole components Q l F ie ld (T) Q2 F ie ld (T) quadrupole 0.846635 -0.850683 sextupole 0.031286 ± 0.2 -0.012548 ± 0.017 octopole -0.055274 ± 0.04 0.021891 ± 0.0048 decapole -0.007545 ± 0.05 -0.000419 ± 0.0068 dodecapole 0.0 ± 0.027 0.016523 ± 0.0016 Table 7: Measured characteristics of the S A S P Quadrupoles when excited so that the quadrupole field equals the designed value. A l l numbers are given for the central momentum and are extrapolated to the pole tips. Mul t ipole components Q l Fie ld (T) Q2 Fie ld (T) quadrupole 0.846635 -0.850683 sextupole 0.028518 -0.012536 octopole -0.046905 0.018852 decapole -0.002172 -0.001885 dodecapole 0.023907 0.026110 First-order calculations using the R A Y T R A C E code yield the following transfer matr ix R for the S A S P spectrometer [4]: 27 Table 8: Coil and steel details of the SASP quadrupoles Q l and Q2. Current (A) Bore diameter (cm) Mean pole tip field (kG) Q i 893 21.222 9.4 Q2 938 21.186 9.4 Figure 8: The SASP Dipole along with the quadrupoles Q l and Q2. 28 0.59892 0.0000 0.0000 0.0000 0.0000 2.81323 3.93637 -1.67322 0.0000 0.0000 0.0000 5.78558 0.0000 0.0000 -4.00192 -0.20450 0.0000 0.0000 0.0000 0.0000 -12.63950 -0.88546 0.0000 0.0000 \ 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 / The quality of a spectrometer is, to a considerable degree, given by the quality of the dispersive elements, i.e., the SASP dipole. The dipole is designed to bend charged reaction products by 90° in the vertical direction. The dipole is of a clamshell type and has a tapered magnet gap of 10 cm at the inside radius and 15 cm at the outside radius with the maximum available field of 1.5 T at the central radius. Hence the two polefaces are planes inclined with respect to each other at an opening angle of 3.41°. Detailed description of the dipole design may be found in reference [9]. The design philosophy of SASP is to correct for the most important chro-matic and geometrical aberrations by introducing in the quadrupoles Q l and Q2 sextupolar, octupolar, and higher order components of the magnetic field, using the entrance/exit dipole curves, see [4], and correcting for the smaller aberrations in software. In the case of SASP, large angular and momentum acceptances were required, along with the determination of the complete kine-matics of every event (momentum and angles) to a high degree of accuracy. This implies the requirement of a full determination of the trajectories in the focal surface. In principle, these aberrations may be corrected by software. However, the really good resolution may be obtained only if the aberrations are reasonably small up to a large order. SASP has a second quad instead of only a single one as is the case with MRS. This allows more aberrations to be corrected in the hardware. Additionally, the advantages of having such a 29 vertical bending configuration are: 1. To first order, momentum and scattering angle measurements are decou-pled. 2. Floor occupancy and radial extension are much less than those in a horizontal configuration. 4.3 Initial attempt at the experimental determination of SASP optics. This section provides a detailed description of how the optical characteristics of the SASP spectrometer have been obtained and analyzed. The following is a step by step description of the procedure followed during the commissioning shifts in 1993-1994 [16], [17], [18]. In order to make the subject clear, Fig. 9 is provided as a pictorial illustration of methods used. Locating the focal plane. Detailed mapping of the focal plane was carried out using the 2 0 8Pb(p,p/) reaction at a beam energy of 200 MeV with the SASP set at 30° degrees with respect to the beam direction. The elastically scattered proton had a momentum of 643.95 MeV/c while the proton from the first (£^=2.6145 MeV) had a momentum of 639.32 MeV/c . Both types of protons were used for this analysis. In order to map out the focal plane, the magnetic settings of SASP were varied so that the protons of interest scanned the entire focal plane. The typical current delivered on a target was about 0.5 - 1.0 nA. Rays from the Pb elastic peak (or Ist exited state) were projected onto a planar focal surface defined to be a certain distance (SFOCAL) below the X] V D C plane, see Fig. 10. S F O C A L was adjusted until the position of a peak on this focal surface was independent of the trajectory angle. This is shown in Fig. 11. Shown are the scatterplots of the coordinate (SXF) along the focal plane versus the trajectory angle in the VDCs chambers. The figure shows results 30 31 for correct and incorrect values of SFOCAL. If SFOCAL is correct, then the loci should appear as vertical bands. Such an observation suggests that every trajectory corresponding to a particular value of momentum intersects the focal plane at the same point and is independent of the 6 angle. Figure 10: Geometrical relations used in locating the focal surface at SASP. In the SASP data acquisition system (NOVA), this SFOCAL parameter is defined so that it should have a negat ive sign and be expressed in 50 Lim 32 uni t s . T h e negat ive s ign means tha t the focal p lane is be low the X i p lane of the V D C l . S ince the focal surface is not a p lane , S F O C A L is not constant but depends on S X F . F i g u r e 11 is an i l l u s t r a t i o n of the p rocedure descr ibed above. A n a t t e m p t was m a d e to de te rmine the shape of the focal p lane . F o r th is purpose a f i t t i ng of the measured d a t a w i t h 4 t / l order p o l y n o m i a l s was ca r r i ed out , l ead ing to the fo l lowing a n a l y t i c a l express ion for the focal p l ane p o s i t i o n as a func t ion of S X F : SFOCAL = A0 + Ai • SXF + A2 • SXF2 + A3 • SXF3 + A4 • SXF4 (9) where, A o = - 4 6 2 0 . 7 , A ^ O . 7 4 4 6 5 , A 2 = - 7 . 3 1 9 2 9 - 1 0 - 5 , A 3 =5 .173753- l ( r 9 , and A 4 = - l . 5 7 3 5 6 4 - 1 0 - 1 3 . T h i s is t aken to be the reference focal surface a n d a l l references to the focal surface i n the r ema inde r of th is work refer to th is fit . A s u m m a r i z e d resul t is dep ic ted i n F i g . 12. R n - B e n d plane magnification. In order to e x p e r i m e n t a l l y de t e rmine the value of R n the fo l lowing m e t h o d was used. A target cons is t ing of two horizontal n y l o n wires w i t h a separa t ion of 12.79 m m was set up at the target ladder . A T V c a m e r a was pos i t i oned so tha t i t v i ewed this target , and the pos i t i on of these two wires was then m a r k e d on the T V screen i n the S A S P c o u n t i n g r o o m . T h e n , w i t h the b e a m v i s i b l e , u s ing a Z n S screen subs t i t u t ed for the target , the b e a m was steered to the m a r k e d pos i t ions of the wires and the sett ings of the s teer ing magne ts no ted . T o adjust the pos i t i on of the p ro ton beamspo t on the Z n S target t h u m b -wheel 54, w h i c h steers the b e a m ve r t i ca l ly , was used. A f t e r f i nd ing the values of t h u m b w h e e l 54 needed to pu t the beamspo t at the pos i t ions of the top and b o t t o m h o r i z o n t a l wires , the Z n S screen was r emoved a n d the rea l w i r e target was pu t back . T h e inc iden t b e a m was adjusted to i m p i n g e on one bar at a t i m e so tha t two da t a runs ( top-wire and b o t t o m - w i r e ) were co l lec ted for a p a r t i c u l a r m o -33 'OOG SOHO 9008 $T<" 1SG0 UOOf SP, chads Alt * 1115 SJ-JUHailliH.95 EKHIki- IS, I 116 n-JW-1991 liliOI.U i i i i i i 11 11 11 i i i 411 SHI Ml SXFK S u i = 1283 Kit * : SID KMO m i SXFX Sui- 21163, Ml F i g u r e 11: T h e S X F T H scat terplots and S X F h i s tograms w i t h correct (his-tograms o n the r ight side) and incor rec t (h is tograms on the left side) value of S F O C A L . 34 - 1 0 0 0 - 2 0 0 0 < ^ - 3 0 0 0 u. 00 - 4 0 0 0 - 5 0 0 0 h - 6 0 0 0 V 5 0 0 0 1 0 0 0 0 1 5 0 0 0 2 0 0 0 0 2 5 0 0 0 3 0 0 0 0 SXF Figure 12: Focal plane position of the SASP S F O C A L F with respect to the SXF coordinate. A l l numbers are given in 50 fim units. The solid line is the curve of equation 9. 35 m e n t u m of p r o t o n b e a m . W h e n the sett ings of t h u m b w h e e l 54 are k n o w n i t is possible to beg in m e a s u r i n g a p o s i t i o n of e i ther an elast ic peak or a first ex i t ed state of 1 2 C at the focal p lane . T h e R n = <9X/ O C Q ; /<9X/ pa rame te r character-izes the magn i f i ca t ion of an image i n the bend-p lane w i t h respect to the target w h i c h is cons idered as the object . He re dXj is the separa t ion be tween the wi re bars , 12.79 m m . T h e cor responding po in t s i n the focal p lane are e i ther pos i t ions of e las t ic peak or first ex i t ed state for 1 2 C at the focal p lane . T h e values of R n are then s i m p l y g iven by : ( p o s i t i o n of peak f rom wi re 1 — pos i t i on of peak f r o m wi re 2 ) / i n = y= — • 0.05 12.79 m m ^ (10) where the factor 0.05 is the convers ion coefficient f rom 50 fim un i t s to m m uni t s and 12.79 m m is the dis tance between the wi re bars . T h e factor of y/2 is r equ i red i n th is f o r m u l a because the V D C chambers are i n c l i n e d b y 45° w i t h respect to the reference t ra jectory . F i g u r e 13 presents the measured values of R n as a func t ion of m o m e n t u m dev i a t i on 8. T h e expec ted value for 6=0 is -0.59892. Ri 6-Dispersion of the spectrometer. T h e d i spers ion , Ri6, of the spec t rometer is defined as dXf/d6. O n e possi-ble way of m e a s u r i n g the Ri6 pa ramete r of the S A S P t ranspor t m a t r i x is to e m p l o y ine las t i c p ro ton sca t te r ing f r o m a 1 2 C target . In th is m e t h o d A X / is t aken as the difference between the m e a n peak p o s i t i o n of the g r o u n d a n d first ex i t ed states of 1 2 C and the cor responding difference i n m o m e n t u m is t aken for the two peaks . In order to de te rmine the m o m e n t a of these peaks f r o m the conserva t ion of e n e r g y - m o m e n t u m i n th is r eac t ion one can wr i t e : Pbtam ~\~ Plarg  = Pscatt ~\~ Precoil (H) 2 • M 2 + M 2 2 c + 2 • Mi2C • ( E b e a m - E s c a t t ) -— 2 • (Ebeam ' Escatt — Pbeam ' Pscatt ' COS 9) = M 2 £ c o i / (12) 36 -0.9 --15 -10 -5 0 5 10 15 20 5]%) Figure 13: R n dependence on the momentum deviation 6 for SASP. The solid line is a curve of fit, see reference [30]. 37 where, P&e am, Ptar5, Pscatt and Precou are the 4-momenta (lab. system) of: an incoming beam (Ebeam, P beam), a target ( 0 , M t Q r 5 e t ) , a scattering particle (Escait,~P scatt) and a recoil particle ( E r e c o i / , P recou) respectively. A l l masses are in M e V units. Equation 12 is obtained by squaring the equation 11. On differentiating M r e c 0 1 ; wi th respect to Fscatt, one obtains the dependence of the change in momentum on change in recoil mass (here the p 1 2 C reaction is assumed). A p ^ - M i 2 C • AMrecoii ^13^ (Mi2C + Ebeam)-fiscatt ~ Pbeam " COS 9 Pscatt where f 3 s c a t t = ° ca (14) ^scatt Therefore if a shift in a position at the focal plane ( A X / ) of the scattered particle is known then one can find the dispersion of the spectrometer as follows: * » = 1ST < 1 6 ) A P A6 = =^f^ (16) where Po is the momentum of the elastically scattered particle. In order to compare this value of Ri6 with the design value (for central momentum), Ri6 = 2.81323, the experimental data were analyzed as indicated above and summary of measured points is depicted in F ig . 14. R 3 3 - Non-bend plane magnification. In principle the method for finding R 3 3 is similar to that for R n , apart from the fact that instead of horizontal wires one uses vertical ones separated by 5mm. The horizontal steering of the beam is controlled by thumbwheel 55. The formula for R 3 3 is similar to that for R n and refers instead to the Y planes: 38 39 (position of peak from wire 1 — position of peak from wire 2) R33 = — • U.05 5.0 m m (17) where the factor 0.05 converts the coefficient from 50 fj.m units to m m units. Following the method mentioned earlier a full set of spectra was collected, allowing for the determination of R 3 3 on the deviation of particle momentum from the central trajectory, as shown in F i g . 15. The T R A N S P O R T value is -4.00192. Determination of the optical centers of all chambers. The position of the optical centers of the chambers was found by delivering an achromatically tuned 200 M e V beam to the C H 2 target with S A S P at 30°. Then by looking at the scatterplot Xo vs. Yo coordinates at the F E C the geo-metrical centroids were determined and it was assumed that to a considerable degree this geometrical center should coincide to the optical one for the F E C planes as shown in F i g . 16. In order to determine the optical centroids of the V D C chambers the fol-lowing method was used. The beam was shifted 1 cm above and below the optical center. A new set of cuts on Xo and Yo coordinates were defined that allowed the definition of a narrow cone around the entrance optic axis. The spectrometer was tuned for the central momentum of 637 M e V / c that exactly matched the momentum of elastically scattered protons from the 1 2 C(p ,p ' ) re-action for 200 M e V incident energy. Therefore, the cut on the F E C Xo and Yo coordinates effectively determines the position of the 'central ray'. Final ly, by looking at the position of the peaks in cut histograms of X l 5 X 2 , Y i and Y 2 coordinates one can choose those centroids as the optical centers of the corresponding V D C chambers. The values of the optical centers for the various chambers, obtained in this fashion, were determined (50/xm units in N O V A ) as follows: LX0_CENT. =1860.200 40 41 Run # BXDYD Rax-I GOD I 239 H to o O 999 H 3 9 1H 9 6 ' 50. 16-JUN-1994 15t 3Tt 34,42 i i i i I -i i_ " • 51"* 57" I" 5"," * •••*•"* i " * " f ' " • ZZ L» 7 « " • • • * " • " " • - - • • ..-..--iB a-». - . . . . . . •™. * § §• —< • » • § • • • • mim mm •••• • • v | •—--——-- -•—-••-—--_•-• I••• | 1 - - • -i 1 1 r n 1 1 r T 1 1 r an 1 1 r — r 10BD 2D 40 3020 LXOPOS Sum= 16888. 4000 Figure 16: Scatterplot SXOYO used to determine the optical centers of F E C at S A S P . L X O P O S denotes the X 0 coordinate of F E C and L Y O P O S denotes the Y 0 coordinate of F E C . 42 LYO_CENT =773.200 LX1_CENT =13153.7998 LY1_CENT =-4186.250 LX2_CENT =13184.700 LY2_CENT =-4406.3799 Further in the text, the same variables will be labeled as follows: Reconstructing the non-bend-plane angle at the target. There are three principal methods for extracting the <f> information at the SASP spectrometer. What follows is a detailed treatment and comparison of these methods. First, all necessary formulae are defined to provide a clear understanding of the differences between these methods. A l l numerical multipliers are the ap-propriate conversion coefficients relating the SASP units (50 /zm) to T R A N S -PORT units (cm, mrad). Further, the variables with superscript transp. (for instance, Ztransp) refer to T R A N S P O R T code [13] notation. In all three methods of calculating <f>, the basic assumption is that the beam spot in the non-bend plane is intrinsically very small and located at the target center. The 4B twister rotates the extremely small vertical phase space of the cyclotron (l7r mm-mrad) into the horizontal phase space at the SASP target at 4BT2. Consequently small vertical spot sizes of <1 mm can be achieved. Thus assuming Y j = 0 for the trajectory origin, <f> is simply proportional to YQ, the F E C Y coordinate, and is given by: X center \rcenter ~ycenter ~\rcenter Reenter 0 5 1 0 5-^-1 1 1 1 J A 2 and Yc2> <t> = (18) 43 where hpEC is the distance from F E C to the target, or trajectory origin. However this equation assumes that Y 0 and LFEC a r e in the same units and <j> is in radians. These are not in standard T R A N S P O R T units. Y 0 is in 50 Lim units from the Data Acquisition (NOVA) Drift routine and has fur-ther to be corrected for the F E C center offset Y=E N T E R ) LF.EC is in meters, the T R A N S P O R T standard, and so, to obtain <f> in units of mrad : , y y g n t e r ) Q.Q05.1000 ^ = ~ W ioo—' m r a d ( 1 9 ) where the factor 0.005 is the conversion coefficient from 50 Lim to cm, the factor 100 is.the conversion coefficient from m to cm and 1000 is the conversion coefficient from rad to mrad. During an experiment, when the F E C count rate exceeds ~2 MHz, the F E C H V must be turned off preventing <f> from being determined by the method described. The following two methods describe ways of reconstructing the 4> angle from the V D C information only. Some preliminary definitions are required because V D C s are tilted 45° and <f> must be determined in coordinates which are perpendicular to the central ray. First, we introduce the V D C s coordinates with respect to the optical center of SASP: Xi = Xr«w - XfnteT (20) y . yravj -ycenter (^1) where the subscript refers to either X l 5 X 2 , Y j or Y 2 coordinate at V D C . Second, from the basic set of equations: </>VDC = # 4 3 • Yi + # 4 4 • <f>i (22) YVDC — # 3 3 • Yi + # 3 4 • (j>i (23) and assuming Y / is equal to zero, one can derive two estimates of 4>i as follows: fa = method A (24) # 4 4 y <t>i = method B (25) # 3 4 44 First the method B wi l l be developed. It has to be pointed out that transport matrix elements to focal surface are known, not to V D C . From F i g . 10 the geometrical relations between the T R A N S P O R T and S A S P coordinates at V D C l and V D C 2 (where Z[ and Z 2 are equal to zero) are given by: Z2=X'2-sm45° (26) X2 = X^-cos45° (27) Y2 = Y2' (28) 2 i = * J - s i n 4 5 ° (29) Xx = X[ • cos 45° (30) YX=Y{ (31) Second, each particle has a different path length D ^ between the V D C X and V D C 2 chambers, which are separated by the distance L V D C = 3 9 . 4 0 cm ( S V D -C D I S T in N O V A ) , where Bzi and Dz2 are given by the following equations: SFOCAL v . . Dzi = TZ— + X[ • sin 45° cos 45° (32) _ LVDC + SFOCAL . L)z2 = — V -X2 • sm 45 (33) cos 45° Th i rd , by making use of the T R A N S P O R T formalism one can write the following T R A N S P O R T matr ix equation (which can be symbolically written as ( V D C l ) = ( d r i f t from focal surface to V D C l ) x ( S A S P ) x ( t a r g e t ) ) : \ fa I 1 Dzi 0 1 # 3 3 # 3 4 # 4 3 # 4 4 \fYl\ After mult iplication of the matrices we end up with the following equations: Vi = ( # 3 3 + # 4 3 • Dzi) • Yj + (R34 + R44 • Dzi) • <t>i (34) <i>\ = # 4 3 • Yi + R44 • <t>! (35) 45 From equation 34, assuming Y/=0, the equation for <f>j can be written as: <f>i = (36) R34 + R44 • Dzi Thus the method B is developed. Method A is straightforward and follows from the <f> angle definition: = D l i : Ynzi ^ = ^ (38) K44 The <f>i angle for the method A is thus defined. In order to convert the <j>j angle (rad) to T R A N S P O R T units the following conversions of units must be made: ftran, = f . 10QQ I N M R A D (39) < r a n s Dzx • 0-00005 • 100 Dzi = 1 0 0 Q cm/50/xm (40) y i t r o n * = Yt • 0.005 in cm (41) LVDC = SVDCDIST • 0.005 cm/50/xm (42) SFOCALtrans = SFOCAL • 0.005 cm/50 /zm (43) where the factor 0.00005 is the conversion coefficient from 50 / i m to m, the factor 100 is the conversion coefficient from m to cm, the factor 1000 is the conversion coefficient from rad to mrad and the factor 0.005 is a conversion coefficient from 50 fxm to cm. By the definition of the T R A N S P O R T coordi-nates the T R A N S P O R T Y points in direction of decreasing angles and hence # r a n » = -(f)!. 46 In order to compare these two methods (A and B), the following simple estimates can be done. Taking values of R 3 3 = -4.00192, R 4 3 = -12.63950, R 3 4 = -0.2045, R 4 4 = -0.88546, S F O C A L = 0.0317 cm, and AY/=0.2 cm one can estimate the error of A(f>j = 3.7 mrad (method B). This requires use of the equations 34 and 36. The same error estimate for the method A can be done as follows. Using equation 38: yields an error of <j>] = 2.855 mrad (method A) , which is better than that obtained by the method B (2.9 vs 3.7 mrad). Figure 17 confirms the error estimate given above. Scatterplots in Fig. 18 illustrate the <f> angles at the target calculated with the formulae derived above. The straight line of unit slope indicates the quality of the R A Y T R A C E calculations in comparison with the real data. A slight offset from the diagonal indicates the errors in the determination of the optical centers of VDCs and F E C . Note that everywhere in the <f> reconstruction we have made use of the original values of the transport matrix parameters from the R A Y T R A C E design. Reconstructing the bend-plane angle at the target. The basic set of equations 2 is used to reconstruct the 6 angle where R 1 2 is assumed to be zero. Note that all longitudinal distances in T R A N S P O R T must be in 10 m units in order to convert mrad to cm in R i 2 terms as required by the T R A N S P O R T convention. This can be seen from the relation: L\fa = &<j>VDC\ # 4 4 (44) MVDCI = # 4 3 • A F 7 (45) LFEC = DpEcim) • 100 (cm/m) = 0.1 • DFEC (cm/mrad) (46) 1000 (mrad/rad) 47 Run # 101 20-JUN-1994 1 & 21 i 4 3 . 3 8 Figure 17: Histograms of <f> angle differences. The top histogram shows the difference between <f> as determined from F E C and V D C 1 - V D C 2 coordinates (Method A ) . The bottom histogram shows the difference between <j> as deter-mined from F E C and V D C l coordinates (Method B ) . 48 Run H 1 0 1 2 I - J U N - 1 3 3 4 1 3 J 4 3 E 5 I . 4 2 HEPHI3_1 ttaK = 177. Q 4 _J I I I ' 1 ' » • I i i i i. O D —j 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 [— - S O - 2 3 2 33 at S P H I 3 S u n - I 1 5 8 7 , Run tt 1 0 1 2 I - J U N - 1 9 3 4 1 3 : 4 6 : 0 1 . 3 5 H£PHI2_1 l tax= I D 7 . at -I 1 1 « 1 1 1 1 1 1 I i i • i I i i i L -OD — | 1 i 1 1 1 1 • 1 1 1 1 1 1 1 1 1 1 1 1 \ -- 6 0 - 2 9 2 33 3 4 SPHI2 Suni= I 1 • I 6, Figure 18: The scatterplot (top picture) of the <f> angle defined via the F E C and V D C l coordinates. The scatterplot (bottom picture) of the <f) angle defined via the F E C and V D C l -VDC2 coordinates 49 where DFEC is the distance from the target to F E C in meter units. To sim-plify the notation, the following definitions must be introduced. Firs t , all the coordinates are defined with respect to the center of the chambers ( F E C and V D C s ) and in T R A N S P O R T units, for instance, X1 = (XRRAUJ - X f n l e r ) - 0 . 0 0 5 (cm/50>m). Second, the N O V A ' s S F O C A L and S V D C D I S T parameters (for definitions see F i g . 10) are also defined in 50 fim units, S F O C A L - 0 . 0 0 5 (cm/50//m) and L V D C = S V D C D I S T - 0 . 0 0 5 (cm/50/xm). Th i rd , from Fig . 10, the path of a scattered particle to V D C l ( D ^ i ) and to V D C 2 ( D ^ 2 ) is given by: D SFOCAL ^ cos 45° N LVDC±SFOCAL DZ2 = — + X2 • sin 45° (48) cos 45° B y transporting an event to V D C l and to V D C 2 one can write the following matrix equations in a symbolic form: • ( V D C l ) = ( d r i f t to V D C l ) x ( S A S P ) x ( t a r g e t ) ) • (VDC2)=(dr i f t to V D C 2 ) x ( S A S P ) x (target)) or explicitly, after multiplication of the matrixes: Xx = (Rn + R21 -Dzi)-Xj + Dzi • #22 • #i + (Rie + DZi • # 2 6 ) • 6 (49) X2 = (Ru + R2i •DZ2)-XI + DZ2 • R22 • 0j + (R16 + DZ2 • #26) • 6 (50) XFEC = Xi + LFEC • 61 (51) Further simplification is possible under the assumption of the achromaticity of the proton beam. The F E C information is supposed to be available. For achromatic tunes X ; ~ 0. Hence from the second equation of 2 the momentum deviation 8 can be defined as: 8 = (52) R- -16 50 O b v i o u s l y , OvDC = 9 focal = Xi — X\ (53) (Z2-Z1)-V2 and f rom the t h i r d equa t ion of sys t em 3: 0 9focal — i?26 • 8 1 " P F r o m the first equa t ion of 3 another 9i def in i t ion is p r o v i d e d : (54) 0i = (55) T h e 6i angle ob t a ined f r o m the equa t ion 55 gives ' t r ue ' value of 6i, whereas 6i f rom the equa t ion 54 gives the angle ex t r ac t ed f r o m the V D C i n f o r m a t i o n . T h e c o m p a r i s o n of 6i f r o m 55 vs. 0j f rom 54 al lows the d e t e r m i n a t i o n or adjus tment of the R22 pa ramete r . F o r m o r e on this see [30]. In p rac t i ce i t tu rns out tha t the T R A N S P O R T m a t r i x requires o n l y m i n o r cor rec t ions to the theore t i ca l one, i f at a l l . A s u m m a r y of the rest of the measured t r anspor t m a t r i x coefficients versus m o m e n t u m dev i a t i on 8 is g iven i n F i g . 19 and F i g . 20. T h e case where there is no F E C i n f o r m a t i o n ava i lab le requires spec ia l con-s idera t ion . If there is no F E C o p t i o n then there is no choice bu t to assume tha t X / = 0 and tha t a l l the formulae tha t were der ived for the case of a c h r o m a t i c b e a m are v a l i d . However , i n the case of the dispersed b e a m the error i n the m o m e n t u m dev i a t i on can be as large as: where the n u m e r i c a l values of R n = 0.59892, R 1 6 = 2.81323 a n d A X / = 2.0 c m were used. A X / is the b e a m spot height a n d some dispers ions w i l l lower th is error. T h e recons t ruc ted 6j can be ob ta ined f r o m the equat ions 49 and A8 = Rn • AX! = 0.43 (56) 50: XVDCI — Ri6 • 6 (57) (58) 51 Figure 19: The measured R 2 1 , R 2 2 , R 2 6 and R 3 4 T R A N S P O R T matrix param-eters as a function of momentum deviation S for SASP. Solid lines are the fit curves. 52 B m 0* -10.5 -11 -11.5 -12 -12.5 C O -13 -13.5 -14 h -14.5 -15 -10 10 8(%) 15 20 -0.3 -0.4 -0.5 -0.6 5 -0-7 -0.8 --0.9 --1 -1.1 -15 10 8(%) 15 20 Figure 20: Cont'd. The measured R 4 3 and R 4 4 T R A N S P O R T matrix param-eters as a function of the momentum deviation 8 for SASP. Solid lines are the fit curves. 53 5 SASP Detector Read-Out. 5.1 SASP Trigger. In this section the SASP electronics is considered together with its interface to the V D A C S and N O V A data acquisition system. The implemented trigger system is based on the following global concept: it is desired to have a trigger system which will store the chamber information in a buffer as fast as possible so that the chamber information does not have to be stored in long delay cables while the trigger electronics makes a decision to either accept or to reject the event. As a result, the University of Alberta group and T R I U M F have developed over the years the concept of a fast trigger which would latch the chamber information immediately, thereby avoiding cable delays, followed by a slower trigger which would decide to either keep the event and pass it on via the C A M A C branch to the data acquisition system, or clear the registers and await the next event. As a direct consequence of this concept, the trigger electronics has to be positioned beside the chamber. Thus the fast electronics must be operated via remote programmable units such as LeCroy E C L line logic units with the result that one can never see directly the signals which a running experiment sends to the trigger electronics. This way of doing things is different from the usual method of running all the detector analog signals to a counting room, forming the trigger in the counting room (where one can observe all the steps in the trigger electronics via an oscilloscope) and then latching the chamber information. Meanwhile, the chamber and counter signals must wait for the trigger generation while traveling down lines of delay cables. Specifically to the SASP electronics, the main building blocks of the elec-tronic scheme consist of the following units (for more information consult the LeCroy Catalog [21]): 54 4434 (ECL) - is a 32 channel 24 bit scaler. 4413 (NIM) - is an updating discriminator with 16 inputs. 4415A (ECL) - is a non-updating discriminator with sixteen inputs via a front-panel 34-pin connector. 2229 (ECL) - is an 8 input T D C having one common start for all channels. 2249 - is an A D C with 12 analog inputs of current-integrating type. For convenience, what follows is a discussion in detail of a few examples of typical trigger situations, since at the hardware level it is possible to build rather flexible trigger requirements combining severalfold coincidences. Typi-cally, a particle going through SASP hits several detectors and as a result gen-erates a number of signals in the system. A particle passing through the spec-trometer produces signals from the F E C (if it is ON), VDCs (VDC1,VDC2), paddles (PD0-PD4) and the Si scintillator counter. For more details on a particular detector at the SASP see Chapter 4. The signal from the paddle counter is split in 3 ways, and is fed to: • an A D C - Model 2249A and as shown in Fig 21, • an updating discriminator - Model 4413 (labeled as LT -low threshold) • a non-updating discriminator -Model 4416 (labeled as HT - high threshold) After that the signals from the 4413/4416 discriminators feed: • a T D C - Model 2229 (not shown in Fig. 21). • a Programmable logic unit - Model 2365 (labeled as PADDLES LT). 55 • a Majority logic unit - Model 4532 (labeled as PADDLES>1). • a Programmable logic unit - Model 2365 (labeled as PADDLES HT). Combining these signals one can get the following signals, that attribute to the particular particle's identification (ID): PION - is formed by the ' A N D ' combination of P A D D L E S LT and PADDLES HT(veto) signals. PROTON - is formed by the PADDLES LT signal. DIPROTON - is formed by the 'OR' combination of PADDLES>1 and PADDLES HT signals. These signals are an essential part of the trigger definition and on the physical level they are defined by the A E / A x losses. Very flexible trigger definitions can be generated by their appropriate combination in the Programmable logic unit(2365) (labeled as E V E N T ) with other signals such as S l signal, R F signal and X I signal from the V D C l chamber plane and possibly in the future, if it is required, with a C signal from a Cherenkov counter and/or a signal from an A C T I V E collimator. Such an E V E N T signal is defined as the L E V E N T variable in the NOVA data acquisition system. Note that only the top SASP detectors are used in the definition of an E V E N T signal. Besides this E V E N T trigger signal, there is also a Slow Auxiliary Trigger (SAT), see Fig. 23, which is formed by: • an OR combination of signals from the F E C wires which fired. • the signal formed by the 2365 logic unit, fed into a mean-timer(708) (labeled as M T in Fig. 23) and then into the module 4508 labeled as F R O N T E N D E V E N T . 56 • in order to control the timing of the SAT signal with respect to the E V E N T signal, this signal is fed into the P U L S E W I D T H G E N E R A T O R and a set of delays and controls from inside NOVA by two parameters labeled in Fig. 23 as T R M I N and T R M A X . • finally, the SAT signal is formed by the P U L S E W I D T H generator signal and a few other signals in the unit 4508 which is labeled as SLOW A U X T R I G G E R in Fig. 23. A simplified picture of a typical event signal in the electronics can look like this (although one has to understand that a particular path is dependent on the experimental requirements and the trigger definition): • a particle generates signals from the F E C X or/and Y planes and the slow auxiliary trigger signal is formed. • depending on the type of event (pion, proton or diproton) the signals from the paddles (PD0-PD4) and the Si counter make up an E V E N T signal which sets the latch (part of scheme labeled as L A T C H in Fig. 22) to prevent any other triggers from altering the TDCs and ADCs. • being activated by their gate signals, the ADCs and TDCs start digitiz-ing. • when the SAT signal is present the data acquisition computer (VAX/11 - VMS) reads the data stream. Otherwise, with SAT absent, a FAST Clear signal is generated to reset the T D C s and A D C s to their initial states, ready to receive next event. • the latch is than cleared either by the Fast Clear or the Computer Clear. Some of the E C L modules can be controlled from inside the N O V A data analysis system. In NOVA, the E C L arrays have the following notation LERxxx , where xxx is a LeCroy module number. For instance, for a delay module 57 4518/128 there is an ar ray L E R 4 5 1 8 A ( 1 6 ) a n d a va r i ab le L C N 4 5 1 8 A . T h e ab-b r e v i a t i o n is as fol lows: L E R means a S A S P E C L var iab le i n N O V A , 4518 means the L e C r o y ca ta log n u m b e r of th is m o d u l e , and (16) is a n u m b e r of i npu t s f r o m 0 to 15. A n o t h e r set of parameters tha t are c losely re la ted to L E R 4 5 1 8 A is L C N 4 5 1 8 A , e.g., L C N 4 5 1 8 A = 5711. In th is n o t a t i o n , 5 is a b r anch n u m b e r , 7 is a crate n u m b e r and 11 is a s t a t ion n u m b e r . A l l changes i n delays can be i n t r o d u c e d e i ther v i a t y p i n g i n a c o m m a n d l ine (e.g., L E R 4 5 1 8 A ( 0 ) = n , where the delay i n ns is 8 + 8 n , 0 < n < 15), or by m a k -ing changes i n the N O V A A S C I I file. T h e U S R E X E T R I G / C A L C c o m m a n d calcula tes a l l the e lectronics E C L arrays f rom N O V A names , e.g. L T R M I N , L T R M A X etc. T h e last s tep is t o d o w n l o a d the c a l c u l a t e d L E R x x ar rays i n t o the C A M A C modu le s w i t h the U S R E X E T R I G / S E T c o m m a n d . T h e t r igger def in i t ion can be set f rom ins ide N O V A , for e x a m p l e , as fol lows: LEVENT = (BIT5|BIT7)*((LEPADL&LEX1)&LES1) LSLOWAUX =1 LFETRIG = ( (BIT0IBIT1) |BIT7)*(LFTX0|LFTY0) and can be ver i f ied by t y p i n g i n the N O V A c o m m a n d l ine - ' l o o k t r i g ' . T h e response is as fol lows: LEVENT 1*4 Var -160 (BIT5|BIT7)*((LEPADL&LEX1)&LES1) LSLOWAUX 1*4 Param 1 LFETRIG 1*4 Var -131 ( (BITO IBIT1) |BIT7)*(LFTXO|LFTYO) LPADL 1*4 Var 2 LPION LPADHT 1*4 Var 2 LPION LTHRESH 1*4 Param 30 LTHRESHHI 1*4 Param 350 58 LRFDEL 1*4 Param 30 Some explanation is required to understand the trigger definition via NOVA variables. The variable L E V E N T defines the particular trigger which is im-plemented for each particular experiment. In example given above, the top paddles PD0-PD5 along with the X l plane of V D C and the top Si counter are included in the trigger. To download this trigger to the C A M A C modules the user must type in the NOVA prompt - 'trigset'. The response should be : TRIGGER CONDITIONS CALCULATED AND DOWNLOADED TO E C L . Another possible combinations of trigger definitions available for user are: 1. no f r o n t end i n t r i g g e r . LSL0WAUX=1 2. f r o n t end i n t r i g g e r . LSL0WAUX=((BIT5IBIT7)IBIT4)*(LFEGATE) 3. Paddles and SI counter i n EVENT. LEVENT=(BIT5IBIT7)*(LEPADL&LES1) 4. Paddles , SI and RF i n EVENT. LEVENT=(BIT5IBIT7)*((LEPADL&LES1)&LERF) 5. Paddles , S I , XI and RF i n EVENT. LEVENT=(BIT5|BIT7)*(LEPADL&LES1)&(LEX1&LERF)) 5.2 SASP Data Stream. The data stream at SASP and its structure is controlled by the ( T W O T R A N ) file, xxx.TWO , where xxx is any name allowed by the V D A C S data acquisi-tion system. For a detailed description of the T W O T R A N commands and how they can be used in C A M A C read-out see references [22] and [23]. A typical example of twotran code is given in Appendix E . The structure of the data stream is determined by the order in which commands appear in the part of the twotran code called S U B R O U T I N E 59 R L A M and S U B R O U T I N E R S C A L . There are two types of events scaler and real events. R S C A L services a scaler event and R L A M services a real event. For DASS/SASP operation there is a set of precompiled binary files to perform these functions. They can be found on the PH1DAC computer in the directories [EXPTB.DASSjDASS.exe, [EXPTB.SASP]SASPB5.exe, [EX-PTB.MRSjMRSB7.exe. The labels B5 and B7 in the names of executables stand for B R A N C H 5 and B R A N C H 7 which are used by the SASP and MRS electronics respectively. After receiving a L A M signal the computer reads: • D C R - a Digital Coincidence Register • TDCs - in the appropriate CSSCAN command, the parameter R = 8 indicates how often the read command is applied to the T D C , in other words R is the number of addresses (registers) for each module. The number is equal to 8 because 8 channel TDCs are in use at SASP. The CSSCAN command must be called for each module read. • A D C - in this case, the parameter R=12 in the appropriate C S S C A N command indicates how often the read command is applied to the A D C . The number is equal to 12 because 12 channel ADCs are in use at SASP. • Drift Chamber Data - is read by the CSQSTOP twotran command via the 4290 LeCroy drift chamber time digitizing system, in this case the parameter R = 110 indicates that the chamber information has a variable length. 60 Figure 21: The functional part of the electronics which forms the E V E N T signal. The labels in italic font denote the corresponding N O V A variables. 61 a s F i g u r e 22: T h e func t iona l par t of the e lect ronics w h i c h forms the L A T C H s igna l . T h e labels i n i t a l i c font denote the cor respond ing N O V A var iables . 62 Figure 23: The functional part of the electronics which forms the Slow A u x i l -iary Trigger signal. The labels in italic font denote the corresponding N O V A variables. 63 6 Interconnection between Data Acquisition System and NOVA. 6.1 V D A C S - Data Acquisition System at SASP. This chapter is not intended as a substitute for basic documentation like ref-erences [22] and [24]. It is simply a brief outline of some absolutely essential information on V D A C S and its connection to NOVA. V D A C S is a standard T R I U M F Vax Data Acquisit ion System and as its name suggests, it runs on the VAX-11 workstation under the V M S operating system. The heart of V D A C S is a front end processor, the CES 2180 Starburst module, essentially based on the PDP-11 computer. Being a computer itself this Starburst module actually manages the process of communication between C A M A C modules and C A M A C crate controllers, thereby creating an effective buffer and leaving the functions of monitoring and on-line data analysis for the VAX-11 computer. At the same time, to effectively use this Starburst module, one must supply a set of instructions written in a high-level language T W O T R A N [23]. The user must write the code in T W O T R A N and then compile it as with a regular code but using the T W O T R A N compiler on the VAX-11 (for greater details see the reference [23]). V D A C S consists of a number of processes: C A M A C process - transfers data from the Starburst to the V A X computer. The data is stored in the memory pseudo-device GRA1:. S E R V E R process - is a process that allows direct control over the C A M A C process by specifying a set of commands. M E S S E N G E R process - is a process for delivering various messages to the users. Nova is a system designed to aid in the analysis (either on-line or off-line) of experimental nuclear physics data. It receives as input (either from a data 64 DETECTORS CAMAC HARDWARE E X P E R I M E N T A L D A T A 2180 Sturburst B U F F E R E D D A T A user V A X 11/VMS C N O V A user C A M A C C O M M A N D S IVDACS C O N ' I K O T . C O M M A N D S / N user TAPE Figure 24: Data Flow and Control over VDACS-NOVA. 65 file - e.g. magnetic tape or disk, or from VDACS) an 'event' consisting of raw data variables. These are combined in user specified ways to produce calculated variables. Finally, histograms of either raw or calculated variables are produced. From the user's point of view, the interconnection between NOVA and V D A C S is not very complex. The only important point is that N O V A reads the data stream from the device called 'GRAl:' into a vector $RAW of a variable length. The order of reading is specified in the twotran code and in particular by the part called R S C A L and R L A M . While N O V A is read-ing the data stream it is filling out the set of variables listed in Appendix F. There is a one to one correspondence between the order in which twotran reads the data and the $RAW content (which represents one event for the user). The content of this $RAW vector can be seen by typing in the V D A C S command line 'SERVER>show event2/GRAl: ' . In N O V A , it is also possi-ble to assign some descriptive names to this data stream, e.g. LAPDO = $RAW(LSTART+ADC3+0) where: LSTART is a global offset in data stream and the fixed portion of the SASP part of the event begins at position LSTART, ADC3 is an offset that specifies an A D C with N = 3 in the SASP electronics diagram, and 0 is a subaddress or an input number of this A D C . Having defined these variables the user may carry on any calculations and create any new vectors and variables allowed in NOVA, for reference see [24]. At the same time NOVA itself not only collects data, books them in his-tograms and analyzes the data but it also controls the E C L Trigger electronics via a set of variables. The great simplification for experimenters is that the E C L programming from within N O V A is done in terms of 'physical' parameters that frees the user from caring about interconnections and influences of one of the parameters on others. However, for an experienced user who wishes to program the E C L electronics directly in terms of the E C L parameters, NOVA can either read or write the E C L data in C A M A C modules: 66 • from the disk file or • calculate all necessary parameters from those specified in 'physical' terms. The following is a short description of the most important of these param-eters and their meaning. Again for more details refer to the T R I U M F Design Note [25]. It is worth noting that the E C L - N O V A interface is in constant evolution and some details will undoubtedly change in the future. LECLFILE - is a variable that controls the way the E C L arrays are sent to the C A M A C modules. If it is equal to zero then it is done via automatic calculation, otherwise it is done via the data file. LECLOGFILE - is a variable, when specified as non-zero, opens a log file for recording. LECLRAW - is a vector that provides a permanent record of the C A M A C addresses and data copied into the modules. In short, this is the content of the E C L electronics tables during the run. At this point the set of E C L parameters and their 'physical' meaning are outlined. Here we closely follow the T R I U M F Design Note [25] and interested readers are strongly advised to use this reference for greater detail. First of all, the group of parameters that manage the chamber information are described. LXOLO, LXOHI - the acceptance of the XO plane at the F E C chamber. Unless otherwise specified all chamber wire parameters are given in 50 /zm units. By default all cuts are 'wide open'. The rest of the information outside of these limits will be ignored. LYOLO, LYOHI - the same as above but for the YO plane. LX1LO, LX1HI - these two parameters specify the acceptance of the X I plane at the V D C l chamber and disable some part in the 0708 alternat-ing routers. The next group of parameters deals with the paddle signals and the trigger implementation. LTHRESH - is a parameter that defines a threshold value for the paddle counter discriminators (PD0-PD4). The value must be specified in mV and this value is common for all paddle scintillators and the S l counter, see Appendix D for more information on the 4413 discriminator. The default value of the threshold is 30 mV. LPDMIN, LPDMAX - these parameters allow the user to select the num-ber of paddles to be included in the trigger definition. By default, L P D M I N = 0 and L P D M A X = 4, meaning that all paddles are included. LFTXO, LFTYO, LFTNOTO, LFTXM, LFTYM, LFTNOTM, LFTFET - is a long group specifically intended for managing the signals from the Front-End detectors (FEC) to the 4508 Logic module. In the hardware there is a correspondence between the input number at the 4508 Logic module and a bit number assigned to those LFTxx parameters, e.g.: LFTX0=BIT0 L F T F E V = B I T 7 LFETRIG - is an output from the Front-End 4508 module. By assigning to this variable different combinations of logical expressions , one can define the output signal. The form of such an expression is: LFETRIG=BITO*(ANY A L L O W E D L O G I C A L EXPRESSION) which means the output BITO of the 4508 module will be set to 1 when-ever the logical expression in the bracket is ' T R U E ' . Here one has to note that the only allowed L O G I C A L E X P R E S S I O N must be combined in terms of input parameters like LFTYO, L F T F E T etc. As of the writing of this thesis the number of the inputs was predefined. 68 LEPADL, LES1, LEACTCOL, LEX1MULT, LEX1, LERF - with the corresponding bits set to the BIT3, BIT2, BIT4, BIT5, BIT1 and BITO. LEVENT - is a very important parameter that basically defines the nature of the trigger. Technically this is a logical ' A N D ' of those signals comprising the trigger combined in one allowed logical expression. For instance, LEVENT=(BIT5|BIT7)*((LEPADL&LES1)&(LEX1&LERF)) where (BIT5—BIT7) means the 'OR' of outputs 5 (BIT5) and 7 (BIT7) in the 2365 octal matrix unit labeled as E V E N T in the electronics scheme. The next large group implements a set of parameters for the Slow Auxiliary Trigger (SAT). LEVIN, LMRS/MSASP, LFEGATE, LFEGATE2, LEVIN2, LAUXxx - these variables are assigned to the inputs of the second 4508 look-up module and are used to define the Slow A U X Trigger. The output of the Slow Auxiliary Trigger is defined in a similar way as for L E V E N T by attributing the module output to the parameter L S L O W A U X : LSL0WAUX=(BIT1IBIT7)*LFEGATE Last, but not least, a large group of predefined E C L variables controls the timing parameters and delays at the SASP trigger electronics: LTRMIN, LTRMAX - these two parameters define the timing between the Front End Trigger and the other slow coincidence signals in the elec-tronics and determine a start time for the L F E G A T E signal. Unless otherwise stated all timing parameters are given in ns. The L T R M I N and L T R M A X parameters cannot be adjusted continuously but exhibit a 'granularity' of 8 ns. Also there are minimum and maximum widths of 69 these signals that are 10 and 94 ns correspondingly. In order to extend the time up to 132 ns one can use another set of parameters LTRMIN2 and L T R M A X 2 . L R F D E L - this parameter designates the delay of the R F signal in the trigger ( L E V E N T ) gate. L X 1 D E L , LS1DEL, L P D E L - these values are the delays of the X I , S l and paddle signals in the trigger ( L E V E N T ) gate. The range is 2-32 ns with a step of 2 ns. L E V D E L , L E V D E L 2 , L F E G D E L , L F E G D E L 2 - these are the delays L E V E N , L E V E N 2 (24-384 ns) L F E G D E L and L F E G D E L 2 (8-128 ns) in the Slow Auxiliary Trigger. L A U X 1 D E L , L A U X 2 D E L , L A U X 3 D E L - these parameters define the delays of the L A U X 1 , L A U X 2 and L A U X 3 signals in the Slow Auxil-iary Trigger. 70 6.2 Data Information Decoding at Chambers. The process of decoding the drift chamber information is rather complex. A detailed description of all parameters and their meanings is given in great detail in the reports [25] and particular in [26]. In this thesis, a short introduction to the subject is presented instead, with main emphasis on the practical aspects and the ways in which the chamber information relates to the V D A C S - N O V A interface [26]. Drift chamber decoding is handled in NOVA with a F O R T R A N subroutine invoked as a USER F U N C T I O N (USR1). The following definition is required to do this: LDRIFTVEC = USR1(LDRIFTVEC, LDRIFTLEN, LWIRETBL, $RAW, LDROFFSET) It it is worth noting that exact names have to be used. A l l information ex-tracted from the wire chamber is returned to N O V A in a single data vector LDRIFTVEC(39) . The length of the vector is appropriate for the SASP set up including 4 V D C s and 4 FEC's (XO, YO, X M and Y M ) . Note that in N O V A all indices always start at 0, so this vector is really 40 words long. The following is the structure of LDRIFTVEC(xx) . 0 - Raw T D C time for the wire number labeled as LWIRE. If negative it is an E R R O R CODE. 1 - Raw T D C time for the wire adjacent to LWIRE. For F E C this is not usually wire LWIRE+1. 2 - The mean time for all V D C planes in which a good track was found. 3 - The mean time for all F E C planes in which a good track was found. 4 - Bit mask of 'Bad planes'. This is an inclusive 'OR ' of the next two words. 5 - Bit mask of planes with no track found (Missing). 6 - Bit mask of planes with two or more tracks found (Multiples). 71 7 - Bit mask of planes in which the mean time is 'Skewed' (greater than a certain amount away from the average for the corresponding plane type). 8 - The position of the track for the first plane. 9 - The meantime calculated from the track for the first plane. 10 - 256*(number of Clusters) + (number of Hits) for the plane. For F E C it refers to the XO plane only. 11 - 256*(number of Clusters) + (number of Hits) for the second F E C plane. For a V D C , this word contains the angle of the track. Words 8-11 are repeated for each subsequent chamber in the system (that is, words 12-15 are the same values for the second chamber, words 16-19 for the third chamber etc.). In order to make use of these returned values, either in other NOVA expressions or as variables to be histogrammed, the user must define other variables and set their values from L D R I F T V E C . For example, to histogram the meantime for all V D C planes, the user would have to enter NOVA commands such as: INTEGER*4 LVDCMEANT LVDCMEANT=LDRIFTVEC(2) DEF1/XDATA=LVDCMEANT In addition, a few vectors must be defined in NOVA which makes an inter-face to the 4290 chamber read-out system[21], namely: LWIRETABLE(83) - a vector that determines the correspondence between the wire numbers read out by the chamber electronics and a particular physical wire plane. $RAW - a vector that contains the raw event information (read from C A -MAC/disk tape). In order to specify the position of the drift chamber data in the SASP data stream in N O V A , the user must declare the length 72 of the fixed part of data stream(LFIXLEN) and offset(LDROFFSET) such as: LFIXLEN=32 LDROFFSET=LSTART+LFIXLEN, This is the position in $RAW where the Drift chamber block starts. The first data words in the event are $RAW(0), $RAW(1), etc.. Thus, SRAW(LDROFFSET) should be the byte count (inserted by V D A C S ) for the variable length block read from the chamber read-out system 4290. $RAW(LDROFFSET) will be the first data word read from the 4290 system (which will be the number of wires if everything is work-ing). This vector obviously has no fixed dimension, because the chamber information constitutes a word of variable length. The size of the vector is generated by V D A C S automatically. LPLANES(59) - a vector that specifies the particular geometry of the wires in the chamber planes. This vector is five-words in length. In order to set the parameters one has to use the appropriate bit structure for these words. At the time when the on-line analysis is started, values are read from this vector, massaged appropriately, and written into the corresponding E C L tables. The detailed meanings of the bits of the words is described in the T R I U M F Design Note [25]. LPRMTBL(13) - a vector that is one of the most important from the user point of view. This vector contains a set of parameters that control the process of decoding. It is not set directly by the user, but is filled from the E C L variables defined in N O V A . If any of these variables is not defined in NOVA, an appropriate default value is taken. Also, it defines most of the first level cuts on the chamber events. By defining these cuts appropriately the user can effectively control the process of 73 on-line selection of events. Again, as was the case earlier, the detailed table which describes the meaning of the parameters is given in [26]. Electrons released due to ionization from the passing particle drift toward the F E C anode wires. For simplicity, we will name as hit the number of wires fired and as cluster the number of good tracks found. A valid event pattern for the XO F E C plane is one hit - one cluster. For the XO' F E C plane the 'right' pattern is one hit - zero cluster as stipulated by the way the USR1 drift routine decodes the chamber information, see Fig. 32. In other words, for the XO' plane the USR1 decoding routine relies on the information obtained for the XO plane. The observed pattern is a result of reconstruction of the tracks for F E C . If the drift routine finds only one track for F E C then it simply takes for the coordinate of the track the wire number. For the case whre the tracks are found in both planes of F E C , the drift routine attempts to reconstruct the track position by using the drift times and by interpolating this position to the F E C cathode plane. To simplify this algorithm the number order is changed as is shown in Fig. 25. The F E C drift chamber information is in the form of wire number and drift time only for the wires that were struck. It is important to note that the T D C for each wire is reset each time that wire is hit, and all the drift chamber T D C s are operated in common STOP mode, where the stop time is defined by the first level trigger - typically focal plane scintillators. A smaller drift time results in a larger T D C value, and a larger drift time in a smaller T D C value. The absolute address for a piece of the data stream is converted to a wire number for each plane. After this, these wire numbers and the T D C values are transformed into relative positions. When a particle passes through the V D C planes the USR1 drift decoding routine starts searching for the wires which fired. A valid hit requires that at least three adjacent wires are struck. More wires per event are allowed although only the wires with the minimum drift time (maximum T D C value) 74 and its two neighbors are used in the calculation of position as shown in Fig. 33. By numbering the wires which fired as, for example, 1, 2, 3 and 4, one can construct a V pattern by creating a scatterplot of the wire number vs. corresponding drift time to the wire as shonw in Figures 26 and 27. The time labeled as V D C mean time corresponds to the variable defined in NOVA as L V D C M E A N T I M E . It is worth noting that the drift decoder calculates a 'meantime', which is the T D C value for a hypothetical wire which lies right at the intersection of the particle track with the plane of the wires. This value is obtained by software interpolation of the T D C values of the wires that are actually hit, and is not the hardware mean of two timing pulses as is the case for F E C . Thus the name 'meantime' is a misnomer, since it is not the mean of any raw quantities. Also, it is important to note that the T D C value is inversely related to drift time and that the meantime is a T D C value. Thus the meantime, which corresponds to the minimum drift time is equal to the maximum T D C value. In an experiment the histogram involving this V D C mean time may appear as depicted in Fig. 30 and may be used for effective event selection. Looking at the V D C mean time histogram, one observes a peak of 15 ns width with long tails produced by the other beam bunches. Also, an additional structure is observed to the left and to the right of the peak with a time separation of about 43 ns. Thus, by setting a cut around this peak as it is shown in Fig. 30, the focal plane spectrum ( H S X F K C A in NOVA) can be cleaned up significantly, thus reducing the background level from accidentals, as can be seen by comparing see Figures 28 and 29. The diagnostic of the drift chambers is based on information from the vector LDRIFTVEC(39) containing of a number of diagnostic words. This information is used to create several diagnostic histograms. The meaning of these diagnostic histograms is summarized as follows: 75 LDRIFTVEC(4) - a bit mask of ' B A D P L A N E ' , that is, either a plane with no track found or more than one track found. These parameters can be set separately and are labeled in NOVA as L B A D P L A N E with the H S B A D histogram attributed. LDRIFTVEC(5) - a bit mask of 'MISSING P L A N E ' . The corresponding NOVA variable is named L M I S S P L A N E with the HSMISS histogram attributed. LDRIFTVEC(6) - a bit mask of ' M U L T I P L E T R A C K ' . The corresponding NOVA variable is named L M U L T P L A N E with the H S M U L T histogram attributed. LDRIFTVEC(7) - a mask of ' P L A N E S W I T H S K E W E D M E A N T I M E ' . The corresponding NOVA variable is named L S K E W P L A N E with the H S S K E W histogram attributed. The order of the chamber planes in these histograms is as follows: V D C - X I , UI , X2 and U2, F E C - XO, YO, XO', YO', see Fig. 31. 76 Interpolated Track Position FEC Phmes } 14 ' 15 • t C a t h o d P l a n e 16 « ' R e v e r s e d ' P l a n e s Interpolated Trade Position 14 • 1.' CattlOd P l a n e ' N o r m a l ' P l a n e s Par t i c le T r a c k Figure 25: Wire arrangement for the F E C . The renumbered F E C wires are labeled by circled numbers. 77 Figure 26: Passage of a particle through V D C . 78 F i g u r e 27: T h e V pa t t e rn r econs t ruc t ion at the V D C chambers . T h e coord i -nates i n brackets are the wi re n u m b e r and cor respond ing drif t t i m e . R u n tt 1 5 9 3 Q - n R Y - 1 9 9 4 1 3 : 0 7 : 3 5 . 1 3 H S X F K C H M a x - 9 1 . 1 D D - \ ' ' ' • 1 1 1 1 1 1 . . . . 1 . , , . L SXFK Sum= 32686. [ x l D ' 3 ) F i g u r e 28: S p e c t r u m of exc i t ed levels of 1 2 C f r o m the (p,7r+) r eac t ion as i t is observed at the focal p lane w i t h o u t cuts on the V D C m e a n t i m e a p p l i e d . 79 Run *t 1 59 30- I1FIY-1 9 9 4 1 3 : 4 . 2 : 5 9 . 0 6 HSXFKCB H n - 99. 1DD H ' ' Y—' 1 ' ' ' • 1 ' ' ' ' 1 ' ' ' ' L F i g u r e 29: S p e c t r u m of exc i t ed levels of 1 2 C f rom the (p,7r+) r eac t ion as i t is observed at the focal p lane w i t h cuts on the V D C m e a n t i m e ap p l i ed . ( 3 4 9 . DDDDI [ 3 B 5 . 000Q) S u n : 5 3 3 8 5 . Xc : 3 6 5 . 4 5 3 FWHM: 1 5 . 5 0 6 Run # 1 59 3 0 - r i F I Y - l 994 1 3 : 1 0 : 2 2 . 8 6 HSVDCMT Max- 3565. 4D D D —I 1 1—-1 ' 1 1 1 1 ^ 1 • 1 1 i I i i i I LVDCffERNTIME Sum= . 31566. F i g u r e 30: T h e h i s t o g r a m of the V D C m e a n t i m e s p e c t r u m . T h e v e r t i c a l dashed l ines are cuts on the V D C m e a n t i m e . 80 Run * 159 30-ffir-193< 13:46:52.10 KSBHO in-ema. 41333 1 2 ( 6 8 L B f i O P L f l N E S u i = 2 6 5 8 0 9 . Run fl 159 30-rBT-1991 13: *3: *D. 75 HSIM.1 nil. IBB7. L M U L 7 P L H N E S u i = 6 7 4 2 3 . Run II 159 30-tBY-1S9< 13:48:35.41 KSUSS Rn -62O0O. 828111 I i i i i I i i i i I i i i i I I 2 4 6 8 10 L f l l S S P L A N E S u i = 2 5 6 3 8 6 . Run » 159 30-nRri99« 13:56:23.42 HSSKEU nil.36635. nioc I i i i i i i i i i i PLANE WITH SKEWED MEANTIME 1 1 1 1 I 1 1 1 | R i i i | i i i i | . i i I 2 4 6 8 10 L S K E U P L f l N E S u » = 5 0 2 0 4 . Figure 31: Diagnostic spectra used during SASP commissioning in 1993-1994. The position of the planes in all histograms is the same as for the H S B A D ('BAD P L A N E ' ) . 81 c-n 1 1 1 1 1 1 1 1 129 N ;j7 w D J U D X 10 J I ' L_L XOPLANE-FEC 1 1 1 1 I 1 : 33 L X O H I T S J I = 76370. ::<!fl'' 1 2 4 6 LXOPHIT Sui= 76401, 4 3 1 1 J \j 1 0 FEC XO Plane 4 3 Vi si. M fe 2 M 1 0 FEC XO' Hum • 0 1 2 3 HITS 0 1 2 3 HITS F i g u r e 32: T h e rea l on- l ine (top p ic tures) and s chema t i ca l l y dep ic t ed ( b o t t o m pic tures) scat terplots of clusters vs . h i t s for the F E C planes. 8 2 5 D 0 4 2 9 -c<T~ 3 5 7 -i C D X _ 2 6 6 ^ 2 1 4 H _ J > < 1 4 3 -_ J 71 -_1 I I I I I I I I L_ _ l I I L _ X I P L A N E - V D C l - i — i — i — i — | — i — i 1 — i — | — i — r 1 7 3 3 L X 1 H I T S u m = - i — | — i — i — i — i — | — i — i — i -5 0 6 7 6 8 5 9 4 . I x 1 D" 1) — | 1 1 1 r -8 3 1 0 0 4 3 cn UQ E—| 2 ZD 1 o VDC XI Plane | | m mm O 1 2 3 4 5 H I T S F i g u r e 33: T h e rea l on- l ine (top p ic tu re ) and s chema t i ca l l y dep ic t ed ( b o t t o m p ic tu re ) sca t terplots of clusters vs. h i t s for the V D C X I p lane . T h e scat ter-plo ts for the rest U l , X 2 , U 2 planes appear s i m i l a r . 83 7 Auxiliary Software. In addition to the track reconstruction software already described, there are a number of various computer programs available for the users. They can be subdivided into two main groups. • Software for dealing with the SASP hardware, like setting a dipole field, changing high voltage values, etc. • Another type that is intended for calculating some kinematics parameters required by the on-line data acquisition programs. It is worth noting that the NOVA on-line data analysis system is itself available as a code of this kind. An example of a NOVA session in the UNIX environment is given in Appendix H . The following is a brief description of the first type of programs in the first group. S A S P S E T T I N G . This program calculates the current settings for the magnetic elements required for operating the SASP dipole and quadrupoles for a given momentum. The code begins to operate up on the typing of the command SASPSETTING. The user is then asked to provide information regarding the type of particle and the energy or momentum of the particle at the central momentum of SASP. The program then calculates the current (in Amps) and the magnetic field (in kilogauss) for each of the three magnets. The values for the current are interpolated from the measured B versus I data for the magnets using a cubic spline algorithm, and an attempt to correct for the changing effective lengths of the quadrupoles with excitation is provided. If the calculated current lies outside the range that has been measured, a warning is given. S A S P M . The S A S P M is a program for remote control of the SASP power supplies. It allows the user to set the SASP magnet to any desired current. In order to invoke the code the user must type in the command line S A S P M / (keywords). The keywords and their meaning are explained below: 84 S A S P M / - queries the power supply for the present values. It displays the values of the currents of the dipole and quadrupoles for the user. In short, this command displays the current status of the SASP magnets. S A S P M / D val - sets the SASP dipole current to the value val (real or in-teger value). Occasionally, the program refuses to change the SASP settings. If this happens, the command should simply be repeated until it does work. S A S P M / Q l val - sets the Q x current to the value val (real or integer value). Again, if it doesn't work the first time, simply repeat the command again. SASPM/Q2 val - the same as above except that it applies to the Q2 magnet. S A S P M / O N , S A S P M / O F F - switches all the SASP power supplies ON or OFF, when called. S A S P M / R E - resets all the SASP power supplies interlocks, as would be required, for example, if the interlock status on any magnet is ' B A D ' . If it does not reset then an expert should be consulted. S A S P / H - this last command provides a help screen. A word of caution:' S A S P M / should N O T be run when the data ac-quisition software (VDACS) is running.' VSETSASP. This piece of software gives the computer control of the LeCroy HV4032A system which sets the detector (photomultipliers and F E C ) High Voltages. The V S E T S A S P code may then be run by issuing commands like: VSETSASP 16/M 2300/V 12/C There are a number of arguments allowed from which the user may select. mainframe/M - selects which mainframe the user wishes to interact with. For SASP the allowed number is 16, thus the user must enter 16/M. 85 mainframe/S - lists the current values of the voltages for all channels of the selected mainframe, e.g., 16/S lists all 32 voltages set at the SASP mainframe. mainframe/D - similar to the /S key, except that this argument yields the demand voltages i.e., the voltage settings in the LeCroy HV4032A mem-ory. The H V could be off, but these are the voltages to which each channel would be set to if the H V was turned on. mainframe/N, mainframe/F - turns the High Voltages ON and OFF , re-spectively. voltage/V - sets the voltage value for a unit in the given mainframe e.g., 2000/V is a 2000 volt setting. channel/C - sets the channel to which the voltage specification from the most recent / V command applies e.g., 12/C specifies channel 12. channel/Z - resets the voltage of the channel to zero (set 'zero' voltages). channel/R - restores the voltage of the channel selected to the value stored in the memory of the mainframe (i.e., what was read out with a main-frame/D command). In contrast to the F E C , H V remote control for the V D C s H V must be set manually. The H V is set by a double 10-turn pot (NIM Module) placed in the Parity electronics room next to the Parity counting room. The current choice for the best value for all V D C chambers is 9500 Volts [16], which is 9.5 in the 10 turn pot. 86 8 Small Angle Operation. Although, as indicated in Table 2, SASP was carefully designed to be operated over a wide range of angles, there is a problem with trying to operate at small angles. The extended field of the first Q l quadrupole is the major source of this problem. After interacting with a target the proton beam passes through the beam pipe to the beam dump. On its way the beam is exposed to the extended magnetic field of the front Q l quadrupole causing a deviation from the original trajectory. This is shown schematically in Figs. 34, 35 and Fig. 36. Fig 35 shows that the strength of the extended field is comparable to that inside the Q l quadrupole and that it can have a significant effect on the primary beam trajectory. As a result of the resulting deflection, (depending on experimental arrangement), there is the possibility that the beam may hit the beam pipe, causing a sufficient increase in radiation that the safety system will shut off the cyclotron. The main objective of this chapter, then, is to determine the minimum pos-sible (critical) angle, 6cr, for the safe operation of SASP. In order to check the influence of the extended magnetic field on the proton beam a 3D computer model of the Q l quadrupole was developed. A field map of the horizontal optic axis plane was generated by the code T O S C A . This is shown in Fig. 35. Next, the horizontal field map was incorporated into the R A Y T R A C E code and beam particle trajectories were traced through the field for various angle settings, 6, of the SASP spectrometer. This trajectory integration through the Q l extended field enabled the beam particle deflection angle, 7, to be calcu-lated for a given 6. This result is shown in Fig. 43. As expected, the deflection angle, 7, decreases as the SASP angle setting increases. Since Runge-Kutta trajectory integration through the Q l field map via R A Y T R A C E is an awk-ward way in practice of assessing the disruption to the beam line optics by the Q l extended field, a simplified model was developed from the 7 vs. 6 dependence of Fig. 43 so that approximate results could then be obtained 87 by merely using T R A N S P O R T code. A complete description of the model is given below. The motion of charged particle in the magnetic field is governed by the following equations: Pbeam = 0.3 • Bext • (59) L = 7 • p (60) yielding Pbeam = °-3 B ^ L ^ 7 0.3 • Bext • L , 7 = WJEL— (62) •* beam or for non-constant field 7 = 4^- • I Bextdl (63) 0.3 beam where Pteam is the proton beam momentum, Bext is the extended Q l field, p is the radius of curvature of the particle's track, 6 is the SASP angle setting and 7 is the exit angle. The 7 vs. 6 plot together with equation 63 thus provides information on the dependence of the extended field, ^Bext dl(0), on the angle 9. Now one must associate Jfiext dl(#) with a field gradient setting. In doing this we make use of the field map produced by T O S C A from which we obtain the field gradient G.R by measuring the slope from the plot and measuring the length of the field L # . In addition we use the fact that for a quadrupole the magnetic field setting scales with the momentum of the particle and is is numerically given by 0.038482 K G / M e V / c for SASP. For the standard SASP setting: G • L 0.8466 kG/cm • 30.0 cm rt , , ^ = „„ ' w . = 0.038482 k G / M e V / c (64) p 660 M e V / c 1 1 y ' Thus with measurement from the plot G ,R-LR and the calculated JiBext dl(9) for 310 MeV proton, we can create a ratio which is independent of momentum 88 and is a universal function of 9. !iBextdl{6) f(0) (65) GR • LR Hence given any old G L and scattering angle 9 we can calculate / ; B e x t dl(0) = (G-L)-i(9) and from equation 63 one can get 7. The calculated i(9) function is given in Table 10. In T R A N S P O R T a quadrupole is defined in terms of an effective length L, a field at the pole tip B (or gradient G inside the quadrupole), and a radius of the entrance aperture A . The next step is to associate this extended field with the strength of the quadrupole.. From the plot displayed in Fig. 35 one dx G G0 0.588 kG/cm can find a slope | ^ in the extended field region, which we call G 0 . assuming this ratio % is constant, it is equal to: = 0.5882 (66) G 1.0 K G / c m Therefore, from equation 66 the gradient setting is G 0 = 0.5882-G and for L = 34.3891 cm (experimentally measured effective length of Ql ) and with the help of equation 64 one can get G and therefore G 0 . Later these G 0 s will be used in T R A N S P O R T calculations (see Table 11). The next parameter which is essential for the description of the quadrupole is the length of the quad. Obviously, the length is a function of the angle 9 , as is evident in Fig. 44. Simple geometry implies the following trigonometrical relations: if 9 < 9cr - if the entrance angle 9 is less than a critical value 9cr, (for the definitions of these quantities see Fig. 44) then one can write: W t a n e - = 8 5 ? / ^ ( 6 ? ) LQi = a (68) cos 9 v ' where L = 34.3891 cm and W = 30 cm. 89 if 9 > 9cr - if the angle 9 exceeds the critical one 9^, then the situation is somewhat more complicated. First, we define some variables. At the point A and B, shown in Fig. 44, the following expressions apply: XA = -(85 - L/2) - tan 5 (69) ZA = 85 - L/2 (70) XB = -W (71) ZB = tan0/VK (72) Thus one can simply write the effective length of Qi along the BL4B direction as: At this point, the mock up of Qi is completed. For numerical values calculated on the basis of the above formulae we refer to Table 9. When all required quantities for the modeled Q l extended field at BL4B are defined, they may be used in any raytrace program, for example, the T R A N S P O R T [13] code. As an example we will consider two targets typical for the SASP experiments, namely lead (22.5 mg/cm 2) and carbon (50 mg/cm 2). First we have to optimize the beam dump line to have point-to-point focusing. The result of the T R A N S P O R T calculations is given in Fig. 38. The next step is to introduce the multiple scattering into T R A N S P O R T calculation in order to see the effect of it on the beam envelope. The multiple scattering angles of 1.482 mrad (carbon) and of 2.792 mrad (lead) were introduced into the T R A N S P O R T code and the results are depicted in Fig. 40 for lead and in Fig. 39 for carbon. From these figures it is clearly seen that the beam does not intersect the beam pipe anywhere. At this point we can include the (73) 90 modeled Q l extended field in the T R A N S P O R T calculation to observe the effect of deflection. In order to do this we have to determine the starting angle 0 from the condition that deflection is less than the radius of the beam pipe (5cm), that is: (ZQI3 - ZQI) • tan 7 < 5 cm (74) where Z Q I 3 is the distance from the target to the Q13 entrance (6.34 m) and ZQI is the distance from target to the modeled Q l effective center. It turned out that angle of about 20° is a good starting point. New envelopes are plotted in Fig. 41 (carbon target) and in Fig. 42 (lead target). As expected the envelope became worse and the extended Q l field does affect the beam. Looking at Fig. 42 we can notice that the beam envelope does intersect the beam pipe of 10 cm diameter and therefore we cannot run heavy targets like lead in the 4B line at a SASP angle of 20°. In order to make a conclusion regarding the carbon target we have to de-termine an offset of rays after the extended Q l field. Unfortunately, by the nature of the T R A N S P O R T code, we cannot run a ray which has some ini-tial angle with respect to the optical axis. Instead, we can take a ray with a given slope as -tan 7 and transport this ray through the Q1-Q13-Q14 system, starting at the center of modeled Q l . In doing this we can take the T R A N S -PORT matrices for the elements of 4B-Line, namely a quadrupole matrix and a drift-space matrix [13]. As a result of such ray tracing we can find the ray which will have the maximum offset. It turned out that this maximum offset is equal to 10.435-7 at Q14 position for the x coordinate. By comparing this result with the beam envelope calculated by T R A N S P O R T (see Fig. 41) we can conclude that the offset of the ray must be less then 2.5 cm. This implies the following limit for 7 < 0.4174 mrad and hence 6 >22°. This angle 9 of 22° is the desired critical angle, 8CR, for a 50 mg/cm 2 carbon target and for the proton beam of 310 MeV. The beam is assumed to be achromatically tuned. 91 Table 9: Angular (9) dependence of the effective length of the modeled Q l extended field and the distance Z of bend for a 310 MeV proton beam. 9 deg 5 1 0 1 5 2 0 2 5 L Q I m 0 . 3 4 5 2 0 5 0 . 3 4 9 1 9 6 0 . 3 5 6 0 2 2 0 . 3 6 5 9 6 1 0 . 3 7 9 4 4 1 Z Q I m 0 . 9 3 2 1 9 2 0 . 9 4 2 4 5 9 0 . 9 1 8 5 5 5 0 . 9 0 5 9 6 4 0 . 8 9 1 9 3 2 Table 10: Angular (9) dependence of the extended field for a 310 MeV proton beam. P, MeV/c @ 310 MeV .(G-L) kG i{9) //B dl kG-m 7 mrad. 9 deg. 823.29 31.6820 0.137420 3.079 112.22 5.0 823.23 31.6795 0.086131 2.962 107.94 10.0 823.13 31.6757 0.038797 1.3342 48.63 15.0 822.99 31.67030 0.006910 0.23784 8.644 20. 822.81 31.6634 0.002337 0.07694 2.805 25.0 Table 11: Angular (9) dependence of the extended Q l field. 9 deg. (G-L) @ 310 MeV kG G k G / m Go k G / m 5.0 31.6820 92.1280 54.1021 10.0 31.6795 92.1208 54.0981 15.0 31.6757 92.1097 54.0916 20.0 31.6703 92.0940 54.0824 25.0 31.6634 92.0739 54.0705 92 X(cm) Figure 35: Calculated field map of the first quadrupole Q a when operating at 1000 A , generated by T O S C A . Shown is a typical non-deflected proton beam trajectory for a SASP angle setting, 9 of 15°. The expected deflection , 7 is 3.5°. 94 X \ Magnetic Field * \ Beam Figure 36: Influence of the extended field on a beam. 95 CM OO L U O 13 cq CO w in T-H Ot ou CO J Figure 37: 4B beamline from T2 to the Beam Dump. 96 UJ3 X ^ A Figure 38: Envelope calculated by T R A N S P O R T for the carbon target. No multiple scattering is included. 0=20° and P=817.79 MeV/c. 97 UJD X 3 A Figure 39: Envelope calculated by T R A N S P O R T for the carbon target. Mul-tiple scattering is included. 6=20° and P=817.79 MeV/c. 98 3 X 3 A Figure 40: Envelope calculated by T R A N S P O R T for the lead target, scattering is included. 0=20° and P=822.99 MeV/c. 99 Multiple UID X 3 A Figure 41: Envelope calculated by T R A N S P O R T for the carbon target. Mul-tiple scattering is included. 6=20° and P=817.79 MeV/c. 100 5 10 15 20 25 6 de% Figure 43: Calculated dependence of entrance angle 6 vs. exit angle 7 for 310 MeV proton. 102 1 0 3 9 Summary This chapter briefly outlines the results obtained and described in this thesis. In Chapter 2 the main characteristics of the 4BL line are discussed. The tuning of the 4BL line along with some practical recommendations for obtain-ing the best possible resolution of the proton beam are given. Chapter 3 concerns the detector facility available at SASP. General charac-teristics and some technical specifications of chambers (FECs and VDCs) are provided. In a section on the trigger paddles PD0-PD5 and the S i counter, the main operational parameters and geometrical sizes are specified. Chapter 4 is the main one in context of this thesis. It concerns the spec-trometer optics (design philosophy) and some practical aspects of the SASP optics. The roles of the quadrupoles and the dipole are considered together with their influence on high order aberrations. The experimental methods used during the first attempt to measure the SASP optics are given in de-tail. The transport matrix R ,j calculated from the R A Y T R A C E code and the measured one are compared. The result of the measurements allows one to conclude that the calculated and measured matrixes are mainly in good agree-ment. The disagreement between the measured and calculated R n parameters was resolved recently. The new experimentally obtained R n is very closed to that calculated by R A Y T R A C E . Triggering of events and possible variations of the triggers together with other aspects of the SASP electronics are discussed in Chapter 5. The interconnection between the data acquisition system (VDACS) and the NOVA data analysis code is considered in Chapter 6. In particular, the decoding of drift chamber information for both F E C and V D C s is considered in detail. The software, which is available for the experimenter, is described in Chap-ter 7. In particular, the emphasis is on practical aspects of use of these pro-grams. 104 Chapter 8 is dedicated to small angle operation of SASP. The effect of beam deflection as a result of interaction with the extended Q l magnetic field is considered. A model that represent this field is devised. Based on this model, the critical angle for safe SASP operation is predicted. This angle is, which depends also on the thickness of the target, about 22° for a 50 mm/cm 2 carbon target. 105 References [1] P.L. Walden, M . J . Iqbal. Proceedings of the DASS/SASP (Dual Arm Spec-trometer System/Second Arm Spectrometer) Workshop. TRI-86-1, De-cember 1986. [2] [3] [4; [5 [e: [7 [8 [9 [10 [11 [12; [13 E .G. Auld. DASS/SASP report of the conceptual design, ibid. Stanley Yen. Beam Une J^B. Optics Problems. T R I U M F (unpublished), 1993. Stanley Yen. Optics of the Q-Q-Clamshell Second Arm Spectrometer. TRI-U M F , 1987. Stanley Yen. SASP as a Neutron Detector. TRI-DN. Stanley Yen. SASP. Book 8. pg.84. Stanley Yen and Munasinge Punyasena. SASP. Book 8. pg.84. Alan Otter. Proposed DASS/SASP Clamshell Dipole. TRI-DN-87-1, 1987. Fraser A . Duncan. Monte Carlo Simulation and Aspects of the Magneto-static Design of the TRIUMF Second Arm Spectrometer.UBC MSc. The-sis, 1988, unpublished. B. de Raad, A . Minten and E. Kei l . Lectures on Beam Optics. C E R N 66-21, 1966. A. P. Banford The Transport of Charged Particle Beams., London, Spon Limited, 1966. S. Kowalski, H.A.Enge. RAYTRACE. MIT, 1986. Karl L. Brown, Sam K . Howry. TRANSPORT. A Computer Program for Designing Charged Particle Beam Transport System. SLAC Report No.91, July 1970. 106 [14] F. Meot, S. Valero. ZGOUBI User's Guide. S A T U R N E Note LNS/GT/93 -12, 1993. [15] C.J . Kost, P. Reeve. REVMOC. A Monte Carlo Beam Transport Program. TRI-DN-82-28, 1988. [16] SASP Commissioning Logbook No.1. T R I U M F (unpublished), 1993-1994. [17] SASP Commissioning Logbook No.2. T R I U M F (unpublished), 1993-1994. [18] SASP Commissioning Logbook No.3. T R I U M F (unpublished), 1993-1994. [19] Collection of various documents collectively known as the 'grey binder'. T R I U M F (unpublished). [20] TRIUMF Medium Resolution Spectrometer (MRS) Manual. T R I U M F (unpublished), 1986. [21] 1992 Research Instrumentation Catalog. LeCroy Research System Divi-sion, 1992. [22] D. Diel. Vdacs Data Acquisition Primer. TRI-CD-91-01, 1991. [23] Twotran Reference Manual. T R I U M F , Data Acquisition Software Group, 1989. [24] P.W. Green. NOVA (Version 1.300). TRI-DNA-91-1, 1991. [25] P.W. Green. ECL. T R I U M F Preliminary Draft (unpublished), 1992. [26] P.W. Green, A . Ling, C A . Miller. MRS/SASP/DASS NOVA Interface. T R I U M F Preliminary Draft (unpublished), 1993. [27] P.L. Walden. Private communication. [28] Stanley Yen. Private communication. 107 [29] E. Byckling, K . Kajantie. Relativistic Kinematics John Wiley Sz Sons, 1973. [30] M . Punyasena, S. Yen, M . Hartig, You Ke. Empirical Determination of the Focal Surface and First Order Transfer Matrix Elements of the SASP., T D N , 1994. [31] John Campbell. Private communication. 108 Appendix A Optimization of the Optical Aberration Corrections (Theory). Although most of the aberrations were corrected in the hardware, some res-olution optimization was still required. In order to simplify this procedure a software code to first determine the aberration coefficients and then to per-form an automatic optimization of these optical aberrations was developed [28]. Here we will describe the theory that lies behind this optimization pro-cedure. Suppose that one wishes to correct the focal plane position (SXFK in N O V A notation) for a number of optical aberrations in the magnet spectrometer ( like (x/0 2), (x/</>2), (x/x#) and so on). The kih optical aberration is denoted by so that in the above coefficients - $ 1 = 0 2 , $ 2 = ^ 2 > $ 3 = xj • 0. For any \ t h event, instead of writing as a function of the track informa-tion - $k{xi,8i,yi,<f>i), we introduce the new function $k(ti), where the vector ti = (xi,6i,yi, (]){). The major objection of this procedure is to minimize the width of the monoenergetic line at the focal plane by applying a set of aber-ration corrections, i.e. the program automatically selects a set of aberration corrections Cj,- • - , C M , such that for a given sample of N events the quantity D which by definition is: N D = Y, (SXFK, + d • + C 2 • $2(ti) + • • • + CM • $M(U))2 (75) t=i is minimized. Equation 75 can be rewritten in a more compact form as: N M  2 D = £ (SXFK, + J2Cj- (76) «'=i >=i where the index i runs over the number of events (usually a large number) and j runs over the number of aberration functions (usually a small number). Thus if C i is the peak centroid and $i=\ then the remaining terms Yl^Li 0?" $j{U) in equation 76 constitute a correction of the ray position about the peak 109 centroid. In order to minimize D a set of coefficients C, must be selected and this implies that condition: dD dCk 0 for k = 1 to M 3D N M = £ 2 • (SXFK,+£ d • *,•(<;)) • 9K$) «'=1 3=1 dCk N M or, J2(SXFKi + £ C , - • *k(U) = 0 3=1 (77) (78) (79) «=i In order to be able to solve this set of linear equations ( 79) for we will rewrite the system in the following way: N M N •(EsxFKi • **(<:•)) = Y,°r (5>;#) • **(£)) 3 = 1 i=l (80) »=i The formalism may be simplified even more by introducing the following new notations: N (81) i=i N bk = -J2SXFK,-^k(U) (82) Thus the system of linear equations ( 79) may be reduced to the simple form: (83) M bk = ^2 akj • Cj for K = 1 to M 3 = 1 or, in matrix form a n 0,12 • • • aiM aii 0,22 • • • o2M \ «M1 OM2 • • • OMM J \ CM J \ OM ) 1 ci ^ c2 ( b l \ b2 110 Appendix B Optimization of the Optical Aberration Corrections(Practical Ap-proach). This appendix deals with a very practical approach to the optical aberration correction and its realization in SASP software. The following NOVA segments and the associated F O R T R A N user subrou-tines allow the user to automatically optimize the optical aberration correc-tions. The theory is given in Appendix A. The variables that the user needs to set are: Condition L O P T _ X F K _ W I N -defines a region of X F K containing a peak which the user wishes to sharpen up. e.g. L0PT_XFK=XFK>2000 & XFKO000 . Condition L O P T _ C O N D - defines the condition under which the event is considered for optimization, e.g. LOPT_CQND=LPID&LMWOK&LTH_WIN&LPH_WIN&LOPT_XFK_WIN. The expla-nation of this condiution is given in Chapter on trigger at SASP. The integer*4 variable L O P T I M I Z E , controls the optimization action to be taken: L O P T I M I Z E = 0 - N O R M A L DATA T A K I N G takes data, correcting for aberrations using current values of coefficients in C O E F . L O P T I M I Z E = l - A C C U M U L A T E S DATA for N E W aberration correc-tions. LOPTIMIZE=2 - D E T E R M I N E S N E W set of aberration corrections. The integer*4 array K E Y , controls which aberrations are 'active' and which are 'inactive', e.g. K E Y ( i ) = l means aberration i is active, K E Y ( i ) = 0 means it is inactive. KEY(O) must always be set to 1, since this sets the centroid for the peak. K E Y ( l ) to KEY(17) are set to 1 if aberration corrections involving the 111 F E C are desired. KEY(18) to KEY(30) are set to 1 if aberration corrections involving the V D C s only are desired. Thus, the normal mode of usage is as follows: 1. Set KEY(0) = 1. Set either K E Y ( l ) to KEY(17)=1 for aberration correc-tions using F E C information or set KEY(18) to KEY(30)=1 for aber-ration corrections using V D C information. The unused K E Y ( i ) values should be set to zero. 2. Set the condition LOPT_XFK_WIN to define a window around some peak in X F K that you want to sharpen up. 3. Set the condition L0PT_C0ND as desired. 4. Type P A U S E or E N D R U N to halt data acquisition. Set L O P T I M I Z E = l to initiate the optimization process. Then type following commands: ZERO S X F K , ZERO S X F K O P T and C O N T or B E G R U N to resume NOVA analysis. On the next processed event satisfying condition L0PT_C0ND, the running sums AKJSUM(k, j ) and BKSUM(k) are zeroed. LOP-TIMIZE will automatically be set to 11 to indicate that optimization process is in progress. Subsequent processed events which satisfy con-dition L0PT_C0ND will increment the running sums A K J S U M ( k j ) and BKSUM(k) . The events used for resolution optimization are displayed in spectrum S X F K O P T , which can be monitored to check the statistics used for optimization. 5. When spectrum S X F K O P T has a sufficient number of events in the peak of interest to attempt optimization, then within N O V A , first type P A U S E or E N D R U N to stop analysis then set LOPTIMIZE=2. Zero the S X F K C spectrum by typing: ZERO S X F K C and then C O N T or B E G R U N to resume NOVA analysis. The next processed event will trigger the calculation of a new set of optimized aberration correction 112 coefficients COEF(j). L O P T I M I Z E will be automatically set to 0 to disable any further calculation of running sums. Enter C O N T to resume NOVA acquisition. A l l subsequent events will be processed with the new coefficients COEF(j). The corrected focal plane coordinate is stored in variable X F K C , and the 1-D histogram of X F K C should exhibit sharper peaks than those of the uncorrected X F K . In the current version of the code ( September 1994 ), there are 18 aberration coefficients in the present list (0 to 17). The numerical values of the aberrations are calculated in the user subroutine USR2, not in NOVA. As a consequence, if any aberrations other than the ones already programmed are desired, it will be necessary to change user subroutine USR2 and relink NOVA. The user can add more aberrations if desired, to a maximum of 31 in total (0 to 30). If more are added, the user will have to increase the value of MAX_ABER (the maximum aberration number) in subroutine USR2 from the current 17. The user does, however, have a way of disabling some of the aberrations. The integer*4 array K E Y is used to control whether a particular aberration is 'active' or 'inactive'. KEY( i ) = 1 makes the \ t h aberration active and K E Y ( i ) = 0 makes the \ i h aberration inactive. An "inactive" aberration is not used in the computation of the corrected coordinate X F K C , nor is it used when the user requests the calculation of a new set of aberration correction coefficients. Since K E Y can be changed at will from NOVA, the user can select which aberrations are to be used. If the number of additional aberration coefficients is desired: 6. Insert the following sequence into N O V A , to permit correction of up to 31 optical and kinematic aberrations (either make this a command file and execute it in NOVA, or embed this into a *.ASC dump file using a text editor). 113 ! Def ine a window on XFK, set on a peak !that we want to sharpen up CONDITION LOPT_XFK_WIN ! Def ine a v a r i a b l e LOPTIMIZE, to i n d i c a t e ! the o p t i m i z a t i o n a c t i o n ! d e s i r e d . INTEGER*4 LOPTIMIZE ! Def ine AKJSUM and BKSUM, !the cumulat ive sums of the A ( k , j ) and B ( k ) . REAL*4 AKJSUM(960),BKSUM(3l) REAL*4 ABER(30) ! MAX.ABER i s maximum a b e r r a t i o n number ! to be c a l c u l a t e d INTEGER*4 MAX_ABER PARAMETER MAX.ABER MAX_ABER=17 ! i n i t i a l i z e the a b e r r a t i o n ! c o r r e c t i o n c o e f f i c i e n t s to zero REAL*4 C0EF(30) PARAMETER COEF C0EF(0)=0. C0EF(1)=0. C0EF(2)=0. C0EF(3)=0. C0EF(4)=0. 114 C0EF(5)=0. C0EF(6)=0. C0EF(7)=0. C0EF(8)=0. C0EF(9)=0. C0EF(10)=0. C0EF(11)=0. C0EF(12)=0. C0EF(13)=0. C0EF(14)=0. C0EF(15)=0. C0EF(16)=0. C0EF(17)=0. C0EF(18)=0. C0EF(19)=0. C0EF(20)=0. C0EF(21)=0. C0EF(22)=0. C0EF(23)=0. C0EF(24)=0. C0EF(25)=0. C0EF(26)=0. C0EF(27)=0. C0EF(28)=0. C0EF(29)=0. C0EF(30)=0. ! Introduce keys to denote whether ! a particular aberration is "active" 115 ! or no t . KEY(n)=l i f a b e r r a t i o n i s a c t i v e , ! =0 i f i n a c t i v e . ! I n a c t i v e a b e r r a t i o n s are not used to ! c o r r e c t XFKC and not used ! i n the l e a s t squares f i t t i n g procedure ! The user can a c t i v a t e or i n a c t i v e any ! one of the c o e f f i c i e n t s ! by s e t t i n g the corresponding element of KEY to z e r o . INTEGER*4 KEY(25) PARAMETER KEY KEY(0)=1 KEY(1)=1 KEY(2)=1 KEY(3)=1 KEY(4)=1 KEY(5)=1 KEY(6)=1 KEY(7)=1 KEY(8)=1 KEY(9)=1 KEY(11)=1 KEY(12)=1 KEY(13)=1 KEY(14)=1 KEY(15)=1 KEY(16)=1 KEY(17)=1 KEY(18)=0 116 KEY(19)=0 KEY(20)=0 KEY(21)=0 KEY(22)=0 KEY(23)=0 KEY(24)=0 KEY(25)=0 KEY(26)=0 KEY(27)=0 KEY(28)=0 KEY(29)=0 KEY(30)=0 subrout ine c a l l to c a l c u l a t e the o p t i c a l a b e r r a t i o n s and determine the a b e r r a t i o n - c o r r e c t e d XF XFKC=USR2(XFKC,XFK,XI_TR,TH_TGT_TR,PH_TGT_TR,MAX.ABER,KEY,COEF,ABER) subrout ine c a l l to determine running sums of a b e r r a t i o n s AKJSUMSUM=USR3(AKJSUM,BKSUM,XFK,LOPTIMIZE,MAX.ABER,KEY,ABER) subrout ine c a l l to determine new a b e r r a t i o n c o r r e c t i o n c o e f f i c i e n t s C0EF=USR4(COEF,LOPTIMIZE,MAX.ABER,KEY,AKJSUM,BKSUM) 7. Introduce into NOVA OPSEQ the following segment: 117 IF (LOPTIMIZE=0)THEN ! calculate XFKC by adding the ! aberration corrections ! to XFK EVAL XFKC END IF i IF ((L0PTIMIZE=1 | L0PTIMIZE=11) & LOPT_XFK_WIN)THEN ! evaluate running sums for ! aberration corrections EVAL AKJSUM END IF IF(LDPTIMIZE=2)THEN ! evaluate new aberration ! correction coefficients EVAL COEF END IF 8. Fortran user subroutines which must be compiled and linked into N O V A are USR2, USR3, USR4, D L S A R G , F A C T O R and SUBST. The last 3 are used to solve the system of linear equations. They are included in code instead of using the IMSL routines because from the NOVA manual, it wasn't clear how, or if, it must be linked to both an object library containing its own user subroutines and a second object library containing the IMSL routines. If the linking of IMSL routines can be solved, then the last 3 subroutines can be eliminated. The corresponding IMSL routine is also called D L S A R G , so no other changes need to be made. Note that IMSL is not available on P H l D A C ( data acquisition V a x / l l - V M S ), so the user will need to compile and link on the cluster 118 d then copy the executable file over to PH1DAC. 119 Appendix C Listings of the user subroutines for aberration corrections. The following is a listing of the user subroutines. Q $ :fc $ stc ite $ :4c $ $ $ $ $ $ $ % :1c $ sfc sf: c * * C * SUBROUTINE USR2 * C * * c C subrout ine to determine a b e r r a t i o n c o r r e c t e d XF C The k i n e m a t i c a l l y c o r r e c t e d f o c a l p lane coord inate C XFK, i s c o r r e c t e d f o r o p t i c a l a b e r r a t i o n s , and the f i n a l C r e s u l t re turned i n XFKC. The c a l c u a t e d va lues of the C a b e r r a t i o n terms are a l so re turned i n array ABER C SUBROUTINE USR2(XFKC,XFK,XI_TR,TH_TGT_TR,PH_TGT_TR, 1 MAX.ABER,KEY,COEF,ABER) C IMPLICIT NONE REAL*4 XFKC,XFK,XI_TR,TH_TGT_TR,PH_TGT.TR REAL*4 C0EF(0:30) ,ABER(0:30) INTEGER*4 KEY(0:30) , I ,MAX.ABER C C C F i r s t eva luate the o p t i c a l a b e r r a t i o n s . C C a b e r r a t i o n 0 = c e n t r o i d of peak 120 ABER(0)=1. C C a b e r r a t i o n 1 = (x /ph i ) r e s i d u a l k inemat ic c o r r e c t i o n . ABER(1)=PH_TGT_TR C C a b e r r a t i o n 2 = (x/phi**2) r e s i d u a l q u a d r a t i c k inemat ic c o r r e c t i o n C + o p t i c a l a b e r r a t i o n ABER(2)=PH_TGT_TR**2 C C a b e r r a t i o n 3 = ( x / x i ) r e s i d u a l f i r s t order d i s p e r s i o n mismatch ABER(3)=XI_TR C C a b e r r a t i o n 4 = (x /x i**2) r e s i d u a l second order d i s p e r s i o n mismatch ABER(4)=XI_TR**2 C C a b e r r a t i o n 5 = (x / th) f i r s t order focus ABER(5)=TH_TGT_TR C C a b e r r a t i o n 6 = (x/th**2) ABER(6)=TH_TGT_TR**2 C C a b e r r a t i o n 7 = ( x / x . t h ) ABER(7)=XI_TR*TH_TGT_TR C C a b e r r a t i o n 8 = (x/th**3) ABER(8)=TH_TGT_TR**3 C C a b e r r a t i o n 9 = (x / th .ph**2) ABER(9)=TH_TGT_TR*PH_TGT_TR**2 121 c C a b e r r a t i o n 10 = (x /x . th**2) ABER(10)=XI_TR*TH_TGT_TR**2 C C a b e r r a t i o n 11 = (x /x**2 . th) ABER(11)=XI_TR**2 * TH_TGT_TR C C a b e r r a t i o n 12 = (x/th**4) ABER(12)=TH_TGT_TR**4 C C a b e r r a t i o n 13 = (x/th**2.ph**2) ABER(13)=TH_TGT_TR**2 * PH_TGT_TR**2 C C a b e r r a t i o n 14 = (x/ph**4) ABER(14)=PH_TGT_TR**4 C C a b e r r a t i o n 15 = (x /x . th**3) ABER(15)=XI_TR * TH_TGT_TR**3 C C a b e r r a t i o n 16 = (x/th**5) ABER(16)=TH_TGT_TR**5 C C a b e r r a t i o n 17 = (x/th**6) ABER(17)=TH_TGT_TR**6 C C There are 18 a b e r r a t i o n s i n the present l i s t (0 to 17). C The user can add more a b e r r a t i o n s i f d e s i r e d , to a maximum C of 31 i n t o t a l (0 to 30) . MAX_ABER=17 122 c C XFK i s a l ready c o r r e c t e d f o r k inemat ic s h i f t s C Add the c o r r e c t i o n s f o r the o p t i c a l a b e r r a t i o n s of the spectrometer . C Note that we do not c o r r e c t f o r "aberrat ion" 0 , which i s C not r e a l l y an o p t i c a l a b e r r a t i o n , but i s used i n the l e a s t C squares f i t t i n g r o u t i n e to f i t the c e n t r o i d of the peak. XFKC=XFK DO I=1,MAX_ABER IF(KEY ( I ) .NE.0 )THEN XFKC=XFKC+C0EF(I)*ABER(I) END IF ENDDO RETURN END C C C Q $$$$$$$$ C * * C * SUBROUTINE USR3 * C * * Q $ $ £ $ $ $ $ $ 3fc $ $ $ sfc $ 34c $ $ • • c C T h i s subrout ine accumulates the running sums necessary C to c a l c u l a t e a new set of o p t i c a l a b e r r a t i o n c o r r e c t i o n s . C SUBROUTINE USR3(AKJSUM,BKSUM,XFK,LOPTIMIZE,MAX.ABER,KEY,ABER) IMPLICIT NONE 123 REAL*4 AKJSUM(0:30,0:30),BKSUM(0:30),ABER(0:30),ABER_PRODUCT REAL*4 XFK INTEGER*4 KEY(0:30) ,MAX.ABER,LOPTIMIZE,I ,J ,K C C INPUT PARAMETERS: c C XFK k i n e m a t i c a l l y c o r r e c t e d f o c a l p lane p o s i t i o n C LOPTIMIZE =1 i n d i c a t e s that t h i s i s the f i r s t event C f o r c a l c u l a t i o n of a new set of a b e r r a t i o n c o r r e c t i o n s C =11 i n d i c a t e s tha t accumulat ion of running sums C f o r c a l c u l a t i o n of new a b e r r a t i o n c o r r e c t i o n s i s C i n e f f e c t , but t h i s i s not the f i r s t event C MAX.ABER maximum a b e r r a t i o n number index C KEY( i ) =1 i f the i t h a b e r r a t i o n i s a c t i v e , =0 i f i n a c t i v e C ABER(i) the i t h a b e r r a t i o n , e . g . ABER(i)=THETA**2 f o r the C o p t i c a l a b e r r a t i o n (x / the ta**2) , e t c . C C OUTPUT PARAMETERS: c C AKJSUM(k,j) running sum of ABER(K)*ABER(J) C BKSUM(k) running sum of -XFK*ABER(K) C LOPTIMIZE =11 on l e a v i n g t h i s s u b r o u t i n e , to i n d i c a t e that C accumulat ion of running sums f o r a new set C of a b e r r a t i o n c o e f f i c i e n t s i s i n e f f e c t . C C C T h i s subrout ine should be c a l l e d only i f L0PTIMIZE=1 C ( t h i s i s f i r s t event f o r o p t i m i z a t i o n ) or L0PTIMIZE=11 (accumulat ion C of running sums under way). Any other va lue i n d i c a t e s an e r r o r 124 C i n the NOVA OPSEQ. C IF(LOPTIMIZE.NE. l .AND. LOPTIMIZE.NE.11)THEN TYPE * , ***ERROR***' SUBROUTINE USR3 CALLED WITH LOPTIMIZE = \ LOPTIMIZE BUT SHOULD BE CALLED ONLY FOR LOPTIMIZE=l OR 1 1 . ' THIS SUBROUTINE CALCULATES RUNNING SUMS FOR' CALCULATION OF NEW ABERRATION COEFFICIENTS.' PROBABLE ERROR IN NOVA OPERATION SEQUENCE.' THIS SUBROUTINE IS NOT EXECUTED' TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * RETURN END IF C C I f t h i s i s the f i r s t event f o r LOPTIMIZE=l, t h i s means we are C j u s t s t a r t i n g a new round of accumulat ion of runn ing sums, so C we have to zero a l l the running sums C IF(LOPTIMIZE.EQ.1)THEN DO 1=0,30 DO J=0,30 AKJSUM(I,J)=0.0 ENDDO BKSUM(I)=0.0 ENDDO L0PTIMIZE=11 END IF C 125 c C Now c a l c u l a t e the running sums of A ( k , j ) and B(k) C DO K=0 ,MAX.ABER IF (KEY (K) . E Q . D T H E N DO J=K,MAX.ABER IF(KEY(J) . E Q . D T H E N ABER_PRODUCT=ABER(J)*ABER(K) AKJSUM(K,J)=AKJSUM(K,J)+ABER.PRODUCT AKJSUM(J,K)=AKJSUM(J,K)+ABER.PRODUCT END IF ENDDO BKSUM(K)=BKSUM(K)-XFK*ABER(K) END IF ENDDO RETURN END C C c c c c c ******************* c * * C * SUBROUTINE USR4 * c * * c ******************* c 126 C This subroutine i s used to calculate a new set of aberration C correction coefficients from the running sums that have been C accumulated up to this time. C SUBROUTINE USR4(COEF,LOPTIMIZE,MAX.ABER,KEY,AKJSUM,BKSUM) C C input parameters: c C LOPTIMIZE =2 on entry to denote calculat ion of new coeff ic ients C MAX.ABER maximum aberration number index C KEY KEY(i)=l i f the i t h aberration i s in use C =0 i f the i t h aberration is not in use C C AKJSUM AKJSUM(k,j) = accumulated running sum of c ABER(k)*ABER(j) C BKSUM BKSUM(k) = accumulated running sum of -XFK*ABER(K) C C output parameters: c C LOPTIMIZE =0 on exit to disable calculat ion of new aberration C correction coeff icients; a l l subsequent events C w i l l use the current set of coeff icients to correct C for aberrations C COEF COEF(i) i s the correction coefficient for the i t h c aberration. C C The system of l inear equations i s solved in double precis ion c arithmetic, for greater numerical accuracy. C 127 IMPLICIT NONE REAL*4 COEF(0:30),AKJSUM(0:30,0:30),BKSUM(0:30) REAL*8 C(31) ,A(31 ,31) ,B(31) INTEGER*4 LOPTIMIZE,MAX.ABER,I,J,INEW,JNEW,N_ACTIVE_ABER INTEGER*4 KEY(0:30),IP0INTER(31) C C We should have L0PTIMIZE=2 on e n t r y , otherwise t h e r e i s l i k e l y C an e r r o r i n the NOVA opera t ion sequence. IF(LOPTIMIZE.NE.2)THEN TYPE * : ***ERR0R***' SUBROUTINE USR4 CALLED WITH LOPTIMIZE='.LOPTIMIZE BUT SHOULD HAVE L0PTIMIZE=2' THIS SUBROUTINE CALCULATES A NEW SET OF' OPTICAL ABERRATION CORRECTION COEFFICIENTS' PROBABLY ERROR IN NOVA OPERATIONAL SEQUENCE' NEW ABERRATION COEFFICIENTS NOT CALCULATED' TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * TYPE * RETURN END IF C C C The matr ix AKJSUM conta ins e n t r i e s f o r a c t i v e and n o n - a c t i v e C a b e r r a t i o n s , so we must f i r s t e x t r a c t the "act ive" rows and columns C and put them i n t o matr ix A. S i m i l a r l y , e x t r a c t the "act ive" e n t r i e s C of BKSUM and put them i n t o vec tor B. C Note that the i n d i c e s f o r A , B , C s t a r t from 1, not z e r o . INEW=0 DO I=0,MAX_ABER 128 IF(KEY(I) .NE.0)THEN INEW=INEW+1 ! We need a p o i n t e r to t e l l us which row (column) of the ! o r i g i n a l v e c t o r the INEW row (column) of the new ! vec tors and matr ices A , B and C correspond t o . ! Note that INEW has va lues 1 to 31. IPOINTER(INEW)=I JNEW=0 DO J=0,MAX.ABER IF(KEY(J) .NE.0)THEN JNEW=JNEW+1 A(INEW,JNEW)=AKJ SUM(I,J) END IF ENDDO B(INEW)=BKSUM(I) END IF ENDDO N_ACTIVE_ABER=INEW ! no. of a c t i v e a b e r r a t i o n s C C C C a l l IMSL r o u t i n e DLSARG (double p r e c i s i o n v e r s i o n of LSARG) C to s o l v e the system of l i n e a r equat ions A*C=B f o r the unknown C v e c t o r C. CALL DLSARG(N_ACTIVE_ABER,A,31,B,1,C) C C C F i r s t set a l l a b e r r a t i o n c o e f f i c i e n t s to z e r o . DO 1=0,30 C0EF(I)=0.0 129 ENDDO C C Dump the e n t r i e s of vec tor C back i n t o the c o r r e c t e n t r i e s of array COEF DO INEW=1,N_ACTIVE_ABER I=IPOINTER(INEW) ! corresponding index i n COEF COEF(I)=C(INEW) ENDDO C LOPTIMIZE=0 RETURN END C C C 3fC J^ C f^c 3ft sfc 5(c 3|c f^c 3fC 3fC C * * C * DLSARG * C * * c ************** c C T h i s i s a user r o u t i n e to so lve a system of l i n e a r equat ions C A*X=B C T h i s i s a s u b s t i t u t e f o r the r o u t i n e of the same name i n the c IMSL l i b r a r y . I t has been taken s t r a i g h t from Conte and deBoor, C Elementary Numerical A n a l y s i s : An A l g o r i t h m i c Approach, c 1972, pages 129 and 132. C I t has been v e r i f i e d to g ive i d e n t i c a l answers to s e v e r a l C t e s t cases of d i f f e r e n t dimensions . C What a nightmare of spaghe t t i programming! (What do you expect 1 3 0 C from 1972 F o r t r a n ? ) c SUBROUTINE DLSARG(N,A,LDA,B,IPATH,X) C IMPLICIT NONE REAL*8 A(31,31) ,B(31) ,X(31) ,W(31,31) ,D(31) INTEGER*4 I , J , N,LDA,IPATH,IPIVOT(31) . IFLAG C C F i r s t use Gauss e l i m i n a t i o n wi th p a r t i a l p i v o t i n g to C c a l c u l a t e a t r i a n g u l a r f a c t o r i z a t i o n of the matr ix A. CALL FACTOR(A,H,IPIVOT,D,N,IFLAG) C TYPE * , ' ' TYPE * , 'MATRIX W IS ' DO 1=1,N TYPE 50 , (W(I ,J ) ,J=1 ,N) 50 F0RMAT(5X,10F10.4) ENDDO TYPE * , ' ' TYPE * , ' V E C T O R IPIVOT IS' TYPE 51 , ( IPIV0T(J) ,J=1 ,N) 51 FORMAT(5X,1015) IF(IFLAG.EQ.2)THEN TYPE * , ' ' TYPE * , ' * * E R R 0 R * * ' TYPE *, 'ILL-CONDITIONED SYSTEM OF LINEAR EQUATIONS' TYPE * , ' — SOLUTION NOT POSSIBLE' DO 1=1,31 X(I)=0. 131 ENDDO RETURN END IF c c S u b s t i t u t e to so lve f o r v e c t o r X CALL SUBST(W,B,X,IPIVOT,N) RETURN END c c ?fc 3fc 3ft 3fC 3fc sfc 3fC c * * c * FACTOR * C * * C * * * * * * * * * * * * C c T h i s subrout ine i s c a l l e d by DLSARG to c a l c u l a t e a t r i a n g u l a r C f a c t o r i z a t i o n of the matr ix A. c SUBROUTINE FACTOR(A,W,IPIVOT,D,N,IFLAG) IMPLICIT NONE REAL*8 A(31 ,31) ,W(31 ,31) ,D(3 l ) REAL*8 ROWMAX,COLMAX,RATIO,AWIKOV INTEGER*4 I F L A G , K , N , N M 1 , J , K P 1 , I P , I P K , I , I P I V O T ( 3 1 ) IFLAG=1 C INITIALIZE W, IPIVOT, D DO 10 1=1,N IPIV0T(I)=I ROWMAX=0. 132 DO 9 J=1,N W ( I , J ) = A ( I , J ) 9 RDWMAX=MAX(ROWMAX,ABS(W(I,J))) IF(ROWMAX.EQ.O.O)GOTO 999 10 D(I)=ROWMAX C C GAUSS ELIMINATION WITH PARTIAL PIVOTING NM1=N-1 IF(NM1.EQ.0)RETURN DO 20 K=1,NM1 J=K KP1=K+1 IP=IPIVOT(K) COLMAX=ABS(W(IP,K))/D(IP) DO 11 I=KP1,N IP=IPIVOT(I) AWIKOV=ABS(W(IP,K))/D(IP) IF(AWIKOV.LE.COLMAX)GOTO 11 COLMAX=AWIKOV J=I 11 CONTINUE IF(COLMAX.EQ.O.)GOTO 999 C IPK=IPIVOT(J) IPIVOT(J)=IPIVOT(K) IPIVOT(K)=IPK DO 20 I=KP1,N IP=IPIVOT(I) W(IP ,K)=W(IP ,K) /W(IPK,K) 133 RATIO=-W(IP,K) DO 20 J=KP1,N 20 W(IP,J)=RATIO*W(IPK,J)+W(IP,J) IF(W(IP,N) .EQ.0 . )G0T0 999 RETURN 999 IFLAG=2 RETURN END C C C * * C * SUBST * c * * c *********** c C T h i s subrout ine i s c a l l e d by DLSARG to so lve f o r the C unknown v e c t o r X i n the matr ix equat ion A*X=B, once C the t r i a n g u l a r f a c t o r i z a t i o n of A i s a v a i l a b l e . SUBROUTINE SUBST(W,B,X,IPIVOT,N) IMPLICIT NONE REAL*8 W(31,31) ,B(31) ,X(31) ,SUM INTEGER*4 J ,N,IPIVOT(31) ,IP,K,KM1,NP1MK,KP1 C IF(N.GT.1)G0T0 10 X(1)=B(1)/W(1,1) RETURN 10 IP=IPIV0T(1) X(1)=B(IP) 134 DO 15 K=2,N IP=IPIVOT(K) KM1=K-1 SUM=0. DO 14 J =1,KM1 14 SUM=W(IP,J)*X(J)+SUM 15 X(K)=B(IP)-SUM IF(W(IP ,N) .EQ.0 . )THEN TYPE 999,IP,N,W(IP,N) 999 FORMAT(' DIVISION BY 0 ERROR IMMINENT'/ 1 ' W ( ' , 1 3 , ' , ' , 1 3 , ' ) = ' .1PE10.3) END IF X(N)=X(N)/W(IP,N) K=N DO 20 NP1MK=2,N KP1=K K=K-1 IP=IPIVOT(K) SUM=0. DO 19 J=KP1,N 19 SUM=W(IP,J)*X(J)+SUM IF(W(IP ,K) .EQ.0 . )THEN TYPE 999 , IP ,K,W(IP ,K) END IF 20 X(K)=(X(K)-SUM)/W(IP,K) RETURN END 135 Table 12: Some aberration coefficients included in resolution optimization. Number Definition Physical meaning ABER(O) 1. A B E R ( l ) dx/d<f> X residual kinematic correction ABER(2) d x/d<f>2 residual quadratic kinematic correction ABER(3) dx/dxi first order dispersion mismatch ABER(4) dx/dxj second order dispersion mismatch ABER(5) dx/dO first order focus ABER(6) dx/392 second order focus ABER(7) d2x/dxde ABER(8) dx/de3 ABER(9) d2x/dedcf>2 ABER(IO) d2x/dxde2 A B E R ( l l ) d2x/dx2dd ABER(12) dx/de4 ABER(13) d2x/882d(f>2 ABER(14) dx/dcf)4 ABER(15) d2x/dxd03 ABER(16) dx/dO5 ABER(17) dx/de6 136 Appendix D Brief summary of parameters of the C A M A C modules used at SASP. 4434 (ECL) - Each channel can operate at up to 20 MHz requiring input pulse duration of >10 ns. The maximum peak rate for a single channel is 30 MHz and double peak resolution for this model is ~30 ns. 4413 (NIM-ECL) - This model has an adjustable threshold level from -15 mV to -1.0 V which is common to all channels. The maximum input pulse rate is about 150 MHz while a typical double pulse resolution is 5 ns. This model has 2 separate E C L level outputs for each channel, the pulse width being continiously continuously variable from 3.5 ns to 100 ns. 4415A (ECL) - This unit can accomodate negative or positive input signals, with the discrimination threshold for each being adjustable from -30 mV to -600 mV for negative polarity signals and 30 to 600 mV (magnitude). This threshold level is common to all channels. The maximum input repetition rate for this 4415A model should not exceed 9 MHz with an output signal duration set to 100 ns. The model has 2 E C L outputs per each channel, with an output duration, that can be varied from less then 100 nsec to more than 1 LIS depending on the setting of the timing capacitor , see [21]. 2229 (ECL) - This modes has 8 independent channels. There is one Fast Clear input for all channels. The minimum width of the common start signal must be more then 5 ns whereas the minimum width for the fast clear input is 50 ns. Having an 11-bit binary output this model allows a full scale time range up to 1 LIS (500 ps/chan), or, for the best time resolution (50ps/chan), a 100 ns range. 137 2249 - The input level for the 2249 model may vary from -50 mV to +50 mV. Also, in this model there is one fast clear and one gate input both of which are common for all channels. A gate input must be a NIM level signal with a duration between 20 and 200 nsec and a voltage level whose magnitude is not less then -600 mV. The minimum duration of the fast clear signal is 50 ns, with a subsequent delay of about 2 fis for to completely recover after a fast clear. 138 Appendix E A sample of a twotran source code. SUBROUTINE RSCAL INTEGER*4 117440 /7968/ INTEGER*4 L l / l / F lags to keep t r a c k of whether there are th ings other than s c a l e r events happening. SCALF i s set . F A L S E , by s c a l e r s , .TRUE, by LAM events . I f i t i s . F A L S E , on entry to s c a l e r events ( i n d i c a t i n g two success ive s c a l e r events ) , the b u f f e r i s f l u s h e d . T h i s f o r c e s updat ing of s c a l e r page. LOGICAL SCALF LOGICAL OLDSCF COMMON / C E V 1 2 / SCALF i OLDSCF = SCALF SCALF = . F A L S E . ! Set up s c a l e r s read CFSA ('/.B=5 , '/.C=7,%N=4, '/.A=0, '/.F=16 , '/.DATA=I 17440) ! Read C=7 32-channel s c a l e r CFQIGNORE (°/.B=5, °/.C=7, '/.N=4, '/.A=0, °/.F=2, '/.R=32) ! Now read IBP s c a l e r s from Crate 5 CFSCAN C/.B=5, y,C=2, */.N=8, */.A=0, '/.F=0, */.R=12) CFSCAN(*/,B=5 , °/,C=2, '/,N=9, */.A=0, '/.F=0, '/.R=12) CFSCAN(°/.B=5, y.C=2, C/.N=10, '/.A=0, */.F=0, y.R=12) ! Mark b u f f e r to be analyzed r e g a r d l e s s MUSTPROCESS IF (OLDSCF) GOTO 10 FLUSH read 32 s c a l e r s read 12 UP IBP read 12 DOWN IBP read 12 OFF IBP 139 10 RETURN END i * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * SUBROUTINE RLAM INTEGER*4 LI 111 INTEGER*4 L2 111 INTEGER*4 L3 / 3 / INTEGER*4 L4 / 4 / ! T h i s i d e n t i f i e s var ious types of events (to a l low f o r mixed ! MRS / SASP / DASS experiments l a t e r on ! Only IFSASP i s used so f a r INTEGER*2 IFMRS/1/ INTEGER*2 IFSASP/2 / INTEGER*2 IFDASS/3/ LOGICAL SCALF COMMON / C E V 1 2 / SCALF SCALF = .TRUE. ! SASP event STORE IFSASP ! Response to LAM from TDC i n s l o t 12 ! D i s a b l e 4299 c l e a r CFSA ('/.B=5, y.C=4,%N=19,C/.A=0, '/.F=22, */.DATA=L4) ! Read DCR (16 b i t s ) CSSA C/.B=5, ' / .C=4,„N=13, °/.A=0, '/.F=2) ! Read T D C s CSSCAN C/.B=5, y.C=7, y,N=9, '/,A=0, '/,F=0,'/,R=8) CSSCAN C/.B=5, y„C=7, JJN=12, °/,A=0, '/,F=0, y.R=8) ! Read ADC's CSSCAN C/.B=5, y.C=7, y.N=2, '/.A=0, '/.F=0, y,R=12) 140 CSSCANC/.B=5 , y.C=7, y.N=3, '/.A=0, e/.F=0, y.R=12) ! C l e a r TDC, ADC (I assume C l e a r on the Output r e g i s t e r does t h i s , so why ! waste the time) CSSA C/.B=5 , y.c=7, y.N=2, y.A=o, y.F=9) CSSA (y.B=5, y.c=7, y.N=3, y.A=o, y,F=9) CSSA (y.B=5, y.c=7, y.N=9, y.A=o, y.F=9) CSSA C/.B=5, y.c=7, y.N=12, y.A=o, y.F=9) ! next l i n e added to c l e a r der (feb 24,94) CSSA (°/.B=5, y,C=4, °/.N=13, y.A=o, y.F=9) ! Send c l e a r pu l se on output r e g i s t e r CFSA C/,B=5,y„C=4,y.N=19,y,A=0,y.F=20,y.DATA=L2) CFSA (y.B=5,y.C=4,y.N=19,y.A=0,y.F=22,y.DATA=L2) ! D r i f t chamber readout v i a 4299 CSQSTOP C/.B=5 ,y.C=4,'/.N=20 ,y.A=l,'/.F=2 , '/.R=l 10) ! ! enable 0/P REG 4299 c l e a r , and c l e a r i t CFSA (y.B=5,y.C=4,y.N=19,°/.A=0, C/.F=20,y.DATA=L4) CSSA C/.B=5 , '/.C=4,y.N=20,*/.A=0,'/.F=l 1) RETURN END 141 Appendix F References to the $RAW vector in N O V A . L A F E T $RAW(LSTART+ADC2) L A F E V $RAW((LSTART+ADC2)+1) LAPDO $RAW(LSTART+ADC3) LAPD1 $RAW((LSTART+ADC3)+1) L A P D 2 $RAW((LSTART+ADC3)+2) L A P D 3 $RAW((LSTART+ADC3)+3) LAPD4 $RAW((LSTART+ADC3)+4) LAS1H $RAW((LSTART+ADC3)+6) LAS1L $RAW((LSTART+ADC3)+5) LDCRO $RAW(LSTART) LDCR1 $RAW(LSTART+2) L D R I F T V E C USR1(LDRIFTVEC,. . . ,$RAW,. . . ) L P D O R $RAW((LSTART+TDCl2)+2) L P D O R T D C $RAW((LSTART+TDC12)+2) L R F T I M E $RAW((LSTART+TDC12)+5) LS1HTDC $RAW((LSTART+TDCl2)+6) LSI LTD C $RAW((LSTART+TDCl2)+3) LSIOR $RAW((LSTART+TDC12)+1) LTAUX1 $RAW(LSTART+18) LTAUX2 $RAW(LSTART+19) L T A U X 3 $RAW(LSTART+20) 142 Table 13: References to $RAW in NOVA.( cont'd ) L T E V $RAW((LSTART+TDC12)+1) LTEV2 $RAW((LSTART+TDC12)+5) L T E V E N T $RAW((LSTART+TDC9)+3) L T F E G $RAW((LSTART+TDCl2)+3) LTFEG2 $RAW((LSTART+TDCl2)+4) L T F E L A T C H $RAW((LSTART+TDC9)+4) L T F E L A T C H B A R $RAW((LSTART+TDC9)+4) L T O F $RAW((LSTART+TDC9)+7) L T R F $RAW((LSTART+TDCl2)+7) LTS1 $RAW((LSTART+TDC9)+1) LTX1 $RAW-((LSTART+TDC9)+2) L W O R D C N T $RAW(LDROFFSET) /2 RATES USR0(RATES,TOTS,OLDS,. . . ,$RAW,.. . ) 143 Appendix G T i 2 6 - Effect of high-order aberrations. Normally, a number of high-order aberrations are present in any magnetic system, even in very simple ones. There are high-order effects which are diffi-cult to ignore, such as T ^ which tilts the focal plane i.e., T126 -8 acts like a non-zero R 1 2 . In order to make it zero a drift length L is used: 1 L 0 \ / Ru T126 • 8 Ri6 \ / Rn + L • R21 T126 • 8 + L • R22 Rie + L • R26 0 1 0 R21 R22 R26 R21 R22 R26 V 0 0 1 / \ 0 0 1 / \ 0 0 1 \ thus, set T126 -8 +L • R 2 2 =0 the drift length is given as: L = - Z ^ i (84) n.22 The transverse (X) dimension is R 1 6 -8 which is expected position for a focus of rays with momentum which is illustrated in Fig. 45. From the Fig. 45 the following relation may be obtained: tan a = —Rw • ^ 22^126 (85) if T i26=0 then o:=90° (it is usually ~70 and it is practically very difficult to get a=45°) as it is desired in a first-order system. In any simple magnetic system a is closer to 0° ,even where tight restrictions on A x , Ay, A8, L\<f> and 8 exist. Thus high-order aberrations is a plaque on all all more or less realistic magnetic systems SASP has special problems because it does not have small A x , A y and 8. Thus high-order aberrations would be especially severe for SASP if curve Dipole entrance/exit surfaces, quadrupole multipoles did not remove the worse terms. 144 X Figure 45: Influence of T i 2 6 aberration term on a focal plane. 145 Appendix H DATA ANALYSIS ON UNIX WITH NOVA. (Practical Approach). 9.1 A little bit of U N I X . WARNING: UNIX is CASE SENSITIVE !!!!!!!!! The alph04 is a D E C A L P H A computer running the OSF/1 operation system which is a clone of UNIX. Assume that the is at any computer (the only requirement is to have a network connection to alph04). Notation is as follows: < > - labels the correspondence between UNIX <—> V A X . < means 'read my comments'. 'xxxxxxxx' - is what the user must type in command line. 9.2 L O G I N and L O G O U T . Enter command: 'telnet alph04.triumf.ca' or 'telnet alph04'. Massages will appear such as: T r y i n g . . . Connected t o ALPH04.TRIUMF.CA, a ALPHA running DSF . OSF/1 ( a l p h 0 4 . t r i u m f . c a ) ( t typ4) l o g i n : < enter your l o g i n name Password: < enter your password Be watchful the upper and lower case letters have a different meaning. In order to logout type 'exit' or press Ctr+D. The first method of logging out is preferable. 9.3 C H A N G I N G P A S S W O R D . If for some reason, you are unhappy with your current password type 'passwd' and follow the on-screen instructions. 146 9.4 FILES and D I R E C T O R I E S . Is < > DIR • Is - means list all files. • Is -1 - means long listing including full info on files. • Is *.for - means list files which end as .for (e.g. nova.for). cd < > SET DEF [xxxx] 'cd /usrl/users/jacques/nova_for_unix' means change directory from the current one to /usrl/users/jacques/nova_for_unix. The 'cd' command without any arguments returns you to the directory where you logged in. The 'pwd' command means - 'where am I ? ' (like W H E R E command on V M S ) . 9.5 C R E A T I N G and D E L E T I N G D I R E C T O R Y . • 'mkdir mydirectoryname' - means create directory with name mydirec-toryname. • 'rmdir mydirectoryname' - means delete directory with name mydirec-toryname • 'rm -rf mydirectoryname' - means delete directory, including all subdirectories, with name mydirectoryname 9.6 D E L E T I N G file (files). rm < > DELETE ( W A R N I N G : U N I X H A S N O B A C K U P F I L E S ! ! ! ! ! ) You cannot restore deleted file or files thus, think twice before removing your files or directories. 147 • 'rm myfile' - means delete file with name myfile • 'rm *.old' - means remove all files with .old extension. • 'rm - i myfile' - means asks user for permission to delete every file which is going to be deleted (highly recommended!) 9.7 M O V I N G FILES (the way to rename files or direc-tories). • 'mv old.name new.name' - means rename file with old.name to new.name. • 'mv old.name.directory new.name.directory' - means rename, directory with old.name to new.name.directory. 9.8 C O P Y I N G FILES. cp < > COPY • 'cp filel file2' - means copy filel to file2 • 'cp directory.name newdirectory.name' - means copy the whole content of the directory with name directory.name to the directory with newdi-rectory.name. 9.9 LISTING C O N T E N T OF F I L E . more < > TYPE The 'more filename' command means look up the file with name filename. 9.10 E D I T I N G F I L E . The editor is a personal choice. I recommend E M A C S if you don't have X -Windows terminal. In case if you do have X-Terminal then try: EMACS(I like it), X E D I T or NEDIT (type either 'xedit' or 'nedit'). 148 In order to start the E M A C S editor, type in the command line 'emacs myfile'. BASIC E M A C S Commands: • Ctrl + a - (press 'Ct r l ' key and 'a') go to the beginning of line • Ctrl + e - go to the end of line use arrow key to go -up, down, left, right • Esc + > - (press 'Esc' key and '>') go to the end of file • Esc + < - (press 'Esc' key and '<') go to the beginning of file backspace or delete keys - to delete characters • Ctrl + x + s - to save changes. • Ctrl + x + c - exit without changes • Ctrl + x + u - UNDO the changes. • Ctrl 4- x 4 i - insert some other file • Ctrl 4- k - delete up to the end of line Ctrl 4 SPACE - set mark for block marking and use arrow keys to mark block. • Ctrl 4- w - take block in buffer • Ctrl 4 y - restore block from buffer to the cursor location • Ctrl 4- s - forward search • Ctrl 4 r - backward search • Ctrl 4g - undo last command 149 9.11 P R I N T I N G file. In order to print to the specific printer the user can type: 'setenv P R I N T E R printer.name' that means redirect print-out to the printer with name printer.name (lsr_119, IsrJih, lsr_207, IsrJf etc.). Or the user can use a regular hprint com-mand 'hprint filename print.options' or 'hprint filename print.options printer.name'. 9.12 G E T T I N G H E L P . To get help for any command or programs type 'man command.name' that asks for manual pages for command.name. 10 NOVA on UNIX. 10.1 R E S T O R I N G T H E B I N A R Y D U M P . There is NO way to get binary dumps written by V D A C S on UNIX. So dump your files as asci and transfer them to the alpha with a command: 'ftp alph04' which starts file transfer program. After this, you will see some messages: E r i c h . T r i u m f . C A Mul t iNet FTP user process 3.2(106) Connect ion opened (Assuming 8 - b i t connect ions) < a l p h 0 4 . t r i u m f . c a FTP server (OSF/1 V e r s i o n 5.60) ready . ALPH04.TRIUMF.CA> ALPH04.TRIUMF.CA>'user your .user .name' <Password r e q u i r e d f o r Jacques . Password: 'your .password' <User Jacques logged i n . ALPH04.TRIUMF.CA>'send f i l e l . f i l e 2 , f i l e 3 ' < — t r a n s f e r f i l e s ALPH04. TRIUMF. CA> f q u i t ' < — e x i t f t p 150 10.2 C O P Y I N G T H E D A T A FILES I N T O T H E DISK. Insert the tape into the tape drive labeled as 'TZ13 /dev/nrmtlh ' at Room #204 Chem. Annex.: Using the known filenames on the tape, choose the file names you wish to dump and type: 'cp /dev/nrmtlh headerl.dat'. The user will see small file with a name headerl.dat which is a header. The user can look up this file and see the name of the next file on the tape. It is useful if you don't know what the next file on the tape is. Then the user must to type: 'cp /dev/nrmtlh datafilel.dat' and you will see the big file with name 'datafilel.dat'. Again by typing: 'cp /dev/nrmtlh header2.dat' you can dump the next header on the tape. The next data file is obtained by typing in: 'cp /dev/nrmtlh datafile2.dat'. Thus by continuing to do this the desired data can be dumped onto the user's disk. 10.3 R U N N I N G N O V A . Now that you have the data files you want on disk as well as the dump files that go with the data files one must construct a N O V A module to run them. The first step is to enter: • 'setenv NOVADIR /data4/pewg/nova' which sets NOVA location. • 'source $NOVADIR/novasetup' which sets all required paths. • 'setenv C E R N . L E V E L 94a' which sets the C E R N lib directory. • 'setenv P R I N T E R lsr_119' which sets the printer to the laser printer at room 119. This invokes the proper NOVA path links to the source files and libraries. You can check the links are in with 'env NOVADIR ' command. The response should be '/data4/pewg/nova' and the links are in place. Alternatively one can insert the following lines into one's .login file: 151 setenv NOVADIR /data4/pewg/nova source $NDVADIR/novasetup setenv CERN.LEVEL 94a setenv PRINTER l sr_119 Or simply copy my .login file by typing in: 'cp /usrl/users/jacques/nova_for_unix/login .login ' and the same links are set up automatically at login time. Now one is in a position to build an executable NOVA module. Copy over A L L of USRxx files to your offline working directory by using ftp command (see above). SECOND issue the command: 'f77 -c filel.for file2.for file3.for ' that compiles your USRxx Fortran files. Make a new library by typing: 'ar r libyourlib.a filel.o file2.o file3.o' or 'ar r libyourlib.a 'Is *.o'. This command adds all compiled files to library with name 'libyourlib.a'. Note that, the name of the library should look like 'libxxx.a'. 10 . 4 L I N K I N G N O V A . In order to create your personal version of NOVA type: 'setenv N O V A U S R yourlib' which tells the linker that you have your own library. Note you have to skip the first 'libxx' in the library name and '.a'. The 'novalink' command creates your personal NOVA. The response if all's well is: L i n k i n g u s i n g User l i b r a r y l i b y o u r l i b 10 . 5 S T A R T I N G N O V A . You are finally there. NOVA should begin with a >novastart 'your nova name' That means - choose your process if there is more than one NOVA running. >nova < begins NOVA N0VA>IM0DE NOVA < t e l l s nova i t ' s going to look at a data f i l e 152 NOVAXDPEN 'your data f i l e ' <— which you have dumped to d i s k N0VA>L0AD 'your dump f i l e ' <— which you have dumped to d i s k NOVA is running and ready to go (NOVA is case insensitive !). NDVA>EA < s t a r t s a n a l y z i n g your data f i l e N0VA>STAT < confirms that i t i s a n a l y z i n g N0VA>DA < stops the data a n a l y z i n g . Can be r e s t a r t e d wi th another EA command. N0VA>DSP 'name' < d i s p l a y s p lo t (h i s togram) l a b e l e d 'name'. Nova plots out on hardcopy as defined in command 'setenv P R I N T E R lsr.name'. So you have your plot output channeled to a specific laser printer N0VA>SET TERM='your t e r m i n a l type ' <— set the c o r r e c t t e r m i n a l type ATARI f o r A t a r i t e r m i n a l s VT640 f o r VT640 t e r m i n a l s DECW f o r DEC WINDOW t e r m i n a l s N0VA>DSP/VT640 'your p l o t name' <— d i s p l a y s your d e s i r e d p l o t i n a form tha t can be t r a n s f e r to a l a s e r p r i n t e r . N0VA>HARDC0PY/REM0TE <-- sends d i s p l a y e d p l o t to des ignated p r i n t e r You should now be running NOVA and capable of almost everything that is available on the familiar online real-time NOVA used on PH1DAC in the counting room. Some additional notes are worth pointing out: When running on a DECwindows or Xll-Terminals terminal an extra command must be inserted to tell N O V A where the graphics window is. This command has the form 'setenv DISPLAY 'DECwindow node':0.0 ' where 'DECwindow node' is the node of the DECwindow terminal. This is given on a sticker attached to the terminal. This command can be issued anytime before one tries to run 153 NOVA. Alternatively, if one is running from the same DECwindow terminal all the time, one can stick the line 'setenv DISPLAY 'DECwindow node':0.0' into one's .login file. 10.6 G E T T I N G D A S S H E L P O N A L P H A . To accsess DASSHELP on alph04 type: 'lynx http://alph04.triumf.ca:9090' or if you are useing X-Terminal or DECWindows then you can in addition to ' lynx. . . ' type: 'mosaic http://alph04.triumf.ca:9090' 154 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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.831.1-0086807/manifest

Comment

Related Items