Home

User and programmers guide to the neutron ray-tracing

1. 108 DTU Collimator Source 4r PSD Figure 8 1 A sketch of the test instrument for the component V_sample Figure 8 2 Scattering from a V sample measured by a spherical PSD The sphere has been transformed onto a plane and the intensity is plotted as the third dimension The instrument consists of a narrow source a 60 collimator a V sample shaped as a hollow cylinder with height 15 mm inner diameter 16 mm and outer diameter 24 mm at a distance of 1 m from the source The sample is in turn surrounded by an unphysical 4n PSD monitor with 50 x 100 pixels and a radius of 10 m The set up is shown in figure 8 1 8 2 1 Scattering from the V sample test instrument In figure 8 2 we present the radial distribution of the scattering from an evenly illuminated V sample as seen by a spherical PSD It is interesting to note that the variation in the scattering intensity is as large as 10 This is an effect of anisotropic attenuation of the beam in the cylindrical sample DTU 109 8 3 The triple axis spectrometer TAS1 With this instrument definition we have tried to create a very detailed model of the conventional cold source triple axis spectrometer TAS1 at the now closed neutron source DR3 of Ris National Laboratory Except for the cold source itself all components used have quite realistic properties Furthermore the overall geometry of the instrument has been adapted from the detailed technical drawings of the rea
2. e var is a string giving the name of the variable on the X axis e g E Tmin is the lower limit for the X axis Tmax is the upper limit for the X axis e m is the number of elements in the detector arrays e amp N 0 is a pointer to the first element in the array of N values for the detector component or NULL in which case no error bars will be computed e amp p 0 is a pointer to the first element in the array of p values for the detector component e amp p2 0 is a pointer to the first element in the array of p2 values for the detector component or NULL in which case no error bars will be computed e filename is a string giving the name of the file in which to store the data Two dimensional detectors monitors The results of a two dimensional detector monitor are written to a file using the following macro DETECTOR_OUT_2D t xlabel ylabel tmin max Ymin Ymax M N amp N 0 0 amp pl0 0 amp p20 0 filename Here e is a string giving a descriptive title e g PSD monitor e zlabel is a string giving a descriptive label for the X axis in a plot e g X position cm 92 DTU e ylabel is a string giving a descriptive label for the Y axis of a plot e g Y position cm Tmin is the lower limit for the X axis Tmax is the upper limit for the X axis Ymin is the lower limit for the Y axis Ymax is the upper limit for the Y axis e m is the nu
3. SETTING PARAMETERS s1 S2 This declares the definition and setting parameters of the component These parameters can be accessed from all sections of the component see below as well as in EXTEND sections of the instrument definition see section 6 3 Setting parameters are translated into C variables usually of type double in the gen erated simulation program so they are usually numbers Definition parameters are trans lated into define macro definitions and so can have any type including strings arrays and function pointers However because of the use of define definition parameters suffer from the usual problems with C macro definitions Also it is not possible to use a general C expression for the value of a definition parameter in the instrument definition only constants and variable names may be used For this reason setting parameters should be used whenever possible Outside the INITIALIZE section of components changing setting parameter values only affects the current section DTU 87 There are a few cases where the use of definition parameters instead of setting parame ters makes sense If the parameter is not numeric nor a character string i e an array for example a setting parameter cannot be used Also because of the use of define the C compiler can treat definition parameters as constants when the simulation is compiled For example if the array sizes of a multidetector are definition parameters
4. Continuous neutron source with a specified Gaussian diver gence A simple continuous circular neutron source with flat en ergy wavelength spectrum General continuous neutron source with tunable shape spec trum and divergence Source like component that reads neutron events from an ascii binary virtual source file recorded by Virtual_output Detector like component that writes neutron state for use in Virtual_input Table 7 1 Source and source related components of the McStas library 102 DTU MCSTAS optics Description Arm Beamstop Bender Collimator_linear Collimator_radial DiskChopper FermiChopper Filter_gen Guide Guide_channeled Guide_gravity Guide_wavy Mirror Monochromator_curve Monochromator_flat Pol_bender Pol_guide_vmirror Pol_mirror Pol_simpleBfield Selector Slit V_selector Vitess_ChopperFermi Arm optical bench Rectangular circular beam stop A curved neutron guide shown straight in mcdisplay A simple analytical Soller collimator A radial Soller collimator Disk chopper Fermi Chopper with rotating frame calculations This components may either set the flux or change it filter like using an external data file Straight neutron guide Straight neutron guide with channels bender section Straight neutron guide with gravity Can be channeled and focusing Straight neutron guide with gaussian waviness Single mirror plate l Doubly bent
5. DTU 13 are useful and the component can be thoroughly tested before being used in a complex simulation with many different components The McStas system is based on ANSI C making it both efficient and portable The meta language allows the user to embed arbitrary C code in the specifications Flexibility is thus ensured since the full power of the C language is available if needed 1 4 Overview The McStas system documentation consists of the following major parts A short list of new features introduced in this McStas release appears in chapter 2 Chapter 3 explains how to obtain compile and install the McStas compiler associ ated files and supportive software Chapter 4 concerns Monte Carlo techniques and simulation strategies in general Chapter 5 includes a brief introduction to the McStas system section 5 1 as well a section 5 2 on running the compiler to produce simulations Section 5 3 explains how to run the generated simulations Running McStas on parallel computers require special attention and is discussed in section 5 6 A number of front end programs are used to run the simulations and to aid in the data collection and analysis of the results These user interfaces are described in section 5 4 The McStas meta language is described in chapter 6 This chapter also describes a set of library functions and definitions that aid in the writing of simulations See appendix B for more details The McStas component libra
6. INITIALIZE ht C initializations AF NEXUS 4 upu XML compress zip DTU 75 It is possible to set the type of NeXus file with a string argument containing words 4 5 or XML Optionally if the string also contains the compress or zip word the NeXus file will use compression for Data Sets We recommend the syntax NEXUS 5 compress which is the default You may choose the name of the output file with the f filename option from the instrument executable or mcrun see Sections 5 3 5 4 2 and Table 5 2 Then the output format is chosen as usual with the format NeXus option when launching the simulation All output files are stored in the output filename as well as the instrument description itself Other formats are still available When run on a dis tributed system e g MPI detectors are gathered but list of events see e g component Virtual_output are stored as one data set per node 6 3 5 The TRACE section As asummary the usual grammar for component instances within the instrument TRACE section is COMPONENT name comp parameters AT RELATIVE reference PREVIOUS ABSOLUTE ROTATED RELATIVE reference PREVIOUS ABSOLUTE The TRACE keyword starts a section giving the list of components that constitute the instrument Components are declared like this COMPONENT name comp pi 1 p2 2 This declares a component named name that is an instance of the
7. Source code package for building McStas on at least Linux and Windows XP This package should compile on most Unix platforms with an ANSI c compiler Refer to section 3 4 19 3 2 Licensing The conditions on the use of McStas can be read in the files LICENSE and LICENSE LIB in the distribution Essentially McStas may be used and modified freely and copies of the McStas source code may be distributed to others New or modified component and instrument files may be shared by the user community and the core team will be happy to include user contributions in the package 3 3 Installation on windows As of release 1 10 of McStas the preferred way to install on Microsoft Windows is using a self extracting exe file The archive includes all software needed to run McStas including perl a c compiler PDL PGPLOT a vrml viewer and Scilab 4 0 Use PGPLOT or install Matlab if possible since support for Scilab will eventually end Installation of all the provided support tools is needed to get a fully functional McStas The option not to install the tools is included for people who want to upgrade from a working previous installation of McStas The safe and fully tested configuration installation is to install all tools leaving all in stallation defaults untouched Specifically you may experience problems if you install to non standard locations Simply follow the guidance given by the installer pressing next all the way T
8. e Any number of LINK sections may be given each one contains HTML code that will be put in a list item in the link section of the description web page This usually consists of an lt A HREF gt lt A gt pointer to some other source of information e Optionally an ZINSTRUMENT_SITE section followed by a single word is used to sort instruments by origin location in the Neutron Site menu in mcgui e After VEND no more comment text is read by McDoc 100 DTU Chapter 7 The component library Abstract This chapter presents an abstract of existing components As a complement to this chapter and the detailed description in the McStas component manual you may use the mcdoc s command to obtain the on line component documentation and refer to the McStas web page 2 where all components are documented using the McDoc system 7 1 A short overview of the McStas component library The table in this section gives a quick overview of available McStas components provided with the distribution The location of this library is detailed in section 5 2 2 All of them are believed to be reliable and some amount of systematic tests have been carried out However no absolute guaranty can be given concerning their accuracy The contrib directory of the library contains components that were submitted by McStas users but where responsibility has not yet been taken by the McStas core team Additionally the obsolete directory of the lib
9. hy This gives C declarations that may be referred to in the rest of the instrument definition A typical use is to declare global variables or small functions that are used elsewhere in the in strument The Zinclude file keyword may be used to import a specific component definition or a part of an instrument Variables defined here are global and may con flict with internal McStas variables specially symbols like x y z sx sy sz vx vy vz t and generally all names starting with mc should be avoided If you can not compile the instrument this may be the reason The DECLARE section is optional 6 3 3 The INITIALIZE section INITIALIZE at C initializations hy This gives code that is executed when the simulation starts This section is optional In strument setting parameters may be modified in this section e g doing tests or automatic settings 6 3 4 The NEXUS extension The NeXus format 47 requires to link the simulation to additional libraries HDF and NeXus which must have been pre installed Preferably McStas should have been installed with the configure with nexus on Unix Linux systems To activate the NeXus output the compilation of the instrument must be done with flag DUSE_NEXUS 1NeXus The resulting executable is no longer portable The default NeXus format is NeXus 5 with compression However that format may be changed with the optional keyword NEXUS to follow the INITIALIZE section namely
10. name comp See section 5 2 2 for details on which directories are searched This facility is often used to refer to existing component definitions in standard component libraries It is also possible to write component definitions in the main file before the 76 DTU instrument definitions or to explicitly read definitions from other files using Zinclude not within embedded C blocks The physical position of a component is specified using an AT modifier following the component declaration AT x y z RELATIVE name This places the component at position x y z in the coordinate system of the previ ously declared component name Placement may also be absolute not relative to any component by writing AT x y z ABSOLUTE Any C expression may be used for x y and z The AT modifier is required Rotation is achieved similarly by writing ROTATED z y 2 RELATIVE name This will result in a coordinate system that is rotated first the angle in degrees around the x axis then around the y axis and finally around the z axis Rotation may also be specified using ABSOLUTE rather than RELATIVE If no rotation is specified the default is 0 0 0 using the same relative or absolute specification used in the AT modifier We strongly recommend to apply all rotations of an instrument description on Arm class components only acting as goniometers and position the optics on top of these This usually makes it much easier to o
11. 44 DTU USR1 Request information status USR2 HUP Request information and performs an intermediate saving of all monitors status and save This triggers the execution of all SAVE sections see chapter 6 INT TERM Save and exit before end status Table 5 4 Signals supported by McStas simulations e Divide simulation into parts using a file for saving or generating neutron events In this way a guide may be simulated only once saving the neutron events at the guide exit as a file which is being read quickly by the second simulation part Use the Virtual_input and Virtual_output components for this technique e Use source optimizers like the components Source_adapt or Source_Optimizer Such component may sometimes not be very efficient when no neutron importance sam pling can be achieved or may even sometimes alter the simulation results Be careful and always check results with a shorter non optimized computation e Complex components usually take into account additional small effects in a simula tion but are much longer to execute Thus simple components should be preferred whenever possible at least in the beginning of a simulation project e The SPLIT keyword may artificially repeat events reaching specified positions in the instrument This is very efficient but requires to cast random numbers in the course of the remaining propagation e g at samples crystals See section 6 4 6 for details A g
12. The following predefined constants are useful for conversion between units Name Value Conversion from Conversion to DEG2RAD 27 360 Degrees Radians RAD2DEG 360 27 Radians Degrees MIN2RAD 27 360 60 Minutes of arc Radians RAD2MIN 360 60 27 Radians Minutes of arc V2K 10 my h Velocity m s k vector A7 K2V 107 A my k vector T Velocity m s VS2E mn 2e Velocity squared m s72 Neutron energy meV SE2V 2e mn Square root of neutron en Velocity m s ergy meV FWHM2RMS 1 8 log 2 Full width half maximum Root mean square standard deviation RMS2FWHM 8 log 2 Root mean square stan Full width half maxi dard deviation mum MNEUTRON 1 67492 10 2 kg Neutron mass my HBAR 1 05459 10734 Js Planck constant A PI 3 14159265 T FLT MAX 3 40282347E 38F a big float value 126 DTU Appendix C The McStas terminology This is a short explanation of phrases and terms which have a specific meaning within McStas We have tried to keep the list as short as possible with the risk that the reader may occasionally miss an explanation In this case you are more than welcome to contact the McStas core team e Arm A generic McStas component which defines a frame of reference for other components e Component One unit e g optical element in a neutron spectrometer These are considered as Types of elements to be instantiated in an Instrument description e Component Ins
13. determine the two angles 6 Wz is chosen from a uniform distribution in 0 27 giving 272 as for the circle The probability distribution of 9 should be Wg sin 0 for 0 7 whence by eq A 2 0 cos z Exponential decay In a simple time of flight source the neutron flux decays expo nentially after the initial activation at t 0 We thus want to pick an initial neutron emission time from the normalised distribution Y t exp t 7 7 Use of Eq A 2 gives x 1 exp t r For convenience we now use the random variable x 1 x with the same distributions as x giving the simple expression t T In 21 Normal distributions The important normal distribution can not be reached as a simple transformation of a uniform distribution In stead we rely on a specific algorithm for selecting random numbers with this distribution A 2 Random generators Even though there is the possibility to use the system random generator as well as the ini tial McStas version 1 1 random generator the default algorithm is the so called Mersenne Twister by Makoto Matsumoto and Takuji Nishimura See http www math sci hiroshima u ac jp m mat MT emt html for original source It is considered today to be by far the best random generator which means that both its period is extremely large 219937 1 and cross correlations are negligible i e distributions are homogeneous and independent up to 623 dimensions I
14. mcdoc tools The first example generates an index html catalog file using the available components and instruments both locally and in the McStas library The library catalog of components is opened using the BROWSER environment variable e g netscape konqueror nautilus MSIE mozilla If the BROWSER is not defined the help is displayed as text in the current terminal This latter output may be forced with the t or text option Alternatively if a component or instrument comp is specified as in the second example it will be searched within the library and an HTML help will be created for all available components matching comp The last example will list the name and description of all McStas tools Additionally the options web manual and comp will open the McStas web site page the User Manual this document and the Component Manual all requiring BROWSER to be defined Finally the help option will display the command help as usual See section 6 7 for more details about the McDoc usage and header format To use the mcdoc front end the program Perl should be available 5 4 7 Translating McStas components for Vitess mcstas2vitess Any McStas component may be translated for usage with Vitess starting from Vitess version 2 3 The syntax is simply mcstas2vitess Compo comp This will create a Vitess module of the given component Let us assume the component Compo shall be translated The tool first creates a sm
15. resulting in additional menus and tools On the other hand the nw option will force Matlab Scilab to run in the current terminal which is usually faster If you experience slow data sets plotting or even failure you might try the binary stor age format either from the mcgui Simulation Configuration options menu or specifying e g the p Scilab_binary option from the command line To use the mcplot front end the programs Perl and either Scilab or Matlab are required 3 7 5 4 5 Plotting resolution functions mcresplot a 05 o 05 w imev o mev w meV a a1 o 1 2 o Q 87 Q 87 Q 87 Q 87 Bragg Gaussian Half Widths File Output res Instrument simulation parameters AQ 0 3133 a Date Thu May 15 10 00 08 2003 AQ 0 9263 X along lt Q gt in plane None AQ 0 3838 a Y perp to X in plane Z upwards Aw 2 2128 meV Resolution matrix o 56 464535 0 050060985 0 14036386 77 8074575 0 050060985 52 049324 0 090785419 0 0049608603 0 14036386 0 090785419 37 642463 0 022637548 77 8074575 0 0049608603 0 022637548 1 1323679 Covariance matrix 10 os ol 0 37970629 0 0001159 0 00015882804 2 6180041 0 0001159542 0 01921267 4 6335128e 05 0 000716238112 0 0001586882804 4 6335128e 05 0 02656624 0 0016259803 2 6180041 0 00071623812 0 0016259803 18 933766 Figure 5 10 Output from mcresplot with PGPLOT backend Use P C and G keys to write hardcopy files The mcresplot
16. 0 0002 eee ee B 2 Reading a data file into a vector matrix Table input read_table lib B 3 MonitornD Library 0 0 0 0 000002 eee ee B 4 Adaptive importance sampling Library B 5 Vitess import export Library bp tce Soke we Ae ee A ee B 6 Constants for unit conversion etc 2 a The McStas terminology Bibliography Index and keywords 127 129 131 DTU Preface and acknowledgements This document contains information on the Monte Carlo neutron ray tracing program Mc Stas version 2 0 building on the initial release in October 1998 of version 1 0 as presented in Ref 1 The reader of this document is supposed to have some knowledge of neutron scattering whereas only little knowledge about simulation techniques is required In a few places we also assume familiarity with the use of the C programming language and UNIX Linux If you don t want to read this manual in full go directly to the brief introduction in chapter 5 1 It is a pleasure to thank Prof Kurt N Clausen PSI for his continuous support to this project and for having initiated McStas in the first place Essential support has also been given by Prof Robert McGreevy ISIS We have also benefited from discussions with many other people in the neutron scattering community too numerous to mention here In case of errors questions or suggestions do not hesitate to contact the authors at mcstas support mcstas org or con
17. 74 AT 74 COMPONENT 73 132 COPY 79 95 DECLARE 72 86 DEFINE COMPONENT 84 INSTRUMENT 71 DEFINITION PARAMETERS 84 END 75 92 EXTEND 78 95 116 FINALLY 75 91 GROUP 78 79 116 INITIALIZE 72 87 ITERATE 82 JUMP 82 MCDISPLAY 91 116 NEXUS 72 OUTPUT PARAMETERS 85 86 POLARISATION PARAMETERS 85 PREVIOUS 74 RELATIVE 74 ROTATED 74 SAVE 75 88 SETTING PARAMETERS 84 SHARE 87 115 SPLIT 83 STATE PARAMETERS 85 TRACE 73 87 WHEN 81 82 Library 115 adapt_tree lib 123 Components 13 33 56 74 98 contrib 98 102 data 103 122 misc 102 monitors 101 obsolete 98 optics 100 samples 101 DTU share 67 69 103 115 sources 99 Instruments 103 mestas r see Library Run time monitor_nd lib 122 read_table lib Read_Table 69 120 Run time 12 34 67 69 103 115 ABSORB 87 115 ALLOW_BACKPROP 88 115 DETECTOR_CUSTOM_HEADER 90 DETECTOR_OUT 88 MC_GETPAR 86 116 NAME_CURRENT_COMP 116 POS_A_COMP 116 POS_A_CURRENT_COMP 116 PROP_DT 115 PROP_GRAV_DT 115 PROP ZO 87 115 RESTORE_NEUTRON 116 ROT_A_COMP 116 ROT_A_CURRENT_COMP 116 SCATTER 79 88 115 SCATTERED 79 116 STORE_NEUTRON 116 Shared see Library Components share vitess lib 56 123 MCNP 37 40 mestas hosts 61 Monte Carlo method 22 113 Accuracy 26 Adaptive sampling 25 Direction focusing 25 Random number Mersenne Twister 114 Stratified sampling 25 MPI 45 47 61 MYSELF 82 Neutron state
18. Assuming as before that N follows a normal distribution we reach o N p Np Further assuming that the individual weights p follow a Gaussian distribution which in some cases is far from the truth we have N o p o X2 pi No p and reach o I N P 07 pi 4 6 The statistical variance of the p s is estimated by o p gt gt p Np N 1 The resulting variance then reads N 277 2129 Q O 5 E P 4 7 3 For almost any positive value of N this is very well approximated by the simple expression AUODDD 4 8 As a consistency check we note that for all p equal this reduces to eq 4 4 In order to compute the intensities and uncertainties the monitor detector components in McStas will keep track of N X p I gt gt pl and My X p a 1This is not correct in a situation where the detector counts a large fraction of the neutron rays in the simulation but we will neglect that for now DTU 27 4 3 Weight factor transformations during a Monte Carlo choice When a Monte Carlo choice must be performed e g when the initial energy and direction of the neutron ray is decided at the source it is important to adjust the neutron weight so that the combined effect of neutron weight change and Monte Carlo probability of making this particular choice equals the actual physical properties we like to model Let us follow up on the simple example of transmission The probabil
19. ESS see 5 and the European Union through nmi3 E learning and nmi3 MCNSI7 FP7 see the home pages 6 7 8 DTU Chapter 1 Introduction to McStas Efficient design and optimization of neutron spectrometers are formidable challenges which are efficiently treated by Monte Carlo simulation techniques When McStas version 1 0 was released in October 1998 except for the NISP MCLib program 8 no exist ing package offered a general framework for the neutron scattering community to tackle the problems currently faced at reactor and spallation sources The McStas project was designed to provide such a framework McStas is a fast and versatile software tool for neutron ray tracing simulations It is based on a meta language specially designed for neutron simulation Specifications are written in this language by users and automatically translated into efficient simulation codes in ANSI C The present version supports both continuous and pulsed source instru ments and includes a library of standard components with in total around 130 compo nents These enable to simulate all kinds of neutron scattering instruments diffractome ters spectrometers reflectometers small angle back scattering for both continuous and pulsed sources The core McStas package is written in ISO C with various tools based on Perl and Python and is freely available for download from the McStas website 2 The package is actively being developed and
20. Physics Department Technical University of Denmark Kongens Lyngby Den mark Emmanuel Farhi lt farhi ill fr gt Institut Laue Langevin Grenoble France Erik Knudsen lt erkn fysik dtu dk gt Physics Department Technical University of Denmark Kongens Lyngby Den mark Kim Lefmann lt lefmann fys ku dk gt Niels Bohr Institute University of Copenhagen Denmark as well as authors who left the project Peter Christiansen lt pchristi hep 1lu se gt Materials Research Department Ris National Laboratory Roskilde Denmark Present address University of Lund Lund Sweden Klaus Lieutenant lt klaus lieutenant helmholtz berlin de gt Institut Laue Langevin Grenoble France Present address Helmotlz Zentrum Berlin Germany Kristian Nielsen lt kristian nielsen mail tele dk gt Materials Research Department Ris National Laboratory Roskilde Denmark Presently associated with MySQL AB Sweden ISBN 978 87 550 3679 6 ISSN 0106 2840 Information Service Department Riso DTU 2012 Contents Preface and acknowledgements 1 Introduction to McStas 1 1 1 2 1 3 1 4 Development of Monte Carlo neutron simulation Scientific background 1 2 1 The goals of McStas The design of McStas Overview cso et A Be see sh OF eaten 2 New features in McStas 2 0 2 1 2 2 2 3 2 4 2 5 2 6 Werrel s t es oreta a e SLE el Be Runtime corea oia ek a a aa Componen
21. and rotation relative to another component An example is given in section 6 3 9 and some additional examples of instrument definitions can be found on the McStas web page 2 and in the example directory As a summary the usual grammar for instrument descriptions is DEFINE INSTRUMENT name parameters DECLARE C_code INITIALIZE C_code NEXUS TRACE components FINALLY C_code END 6 3 1 The instrument definition head DEFINE INSTRUMENT name a4 d2 This marks the beginning of the definition It also gives the name of the instrument and the list of instrument parameters Instrument parameters describe the configuration of the instrument and usually correspond to setting parameters of the components see section 6 5 A motor position is a typical example of an instrument parameter The input parameters of the instrument constitute the input that the user or possibly a front end program must supply when the generated simulation is started By default the parameters will be floating point numbers and will have the C type double double precision floating point The type of each parameter may optionally be declared to be int for the C integer type or char for the C string type The name string may be used as a synonym for char and floating point parameters may be explicitly declared using the name double The following example illustrates all possibilities DEFINE INSTRUMENT test d1 double d2 int i char s1 string s2 H
22. and units 68 NEXT 82 Optimization 35 41 Parallel computing 61 Parameters Definition 73 84 Instruments 36 50 71 DTU Optimization 42 50 Optional default value 36 71 85 Protecting see Keyword OUTPUT Scans 50 Setting 73 84 PREVIOUS 82 Removed neutron events 75 88 116 ROTATED 82 Signal handler 38 INT signal 75 TERM signal 75 USR2 signal 75 Simulation optimization 35 Testing the distribution 16 Tools 14 28 HDF 60 72 HTML 37 60 IDL 38 57 59 Matlab 30 37 52 53 55 57 59 mcconvert 57 mcdisplay 52 mcdoc 56 96 98 mcformat 57 61 megui 29 32 45 52 54 mecplot 30 38 46 48 51 54 58 mcresplot 55 mcrun 32 50 mestas2vitess 56 123 NeXus 45 60 72 Octave 57 Perl libraries 45 53 54 56 PGPLOT 30 37 45 52 55 58 Scilab 30 37 38 52 53 55 57 59 VRML 60 VRML OpenGL 37 38 52 57 Tripoli 37 40 Vitess 37 40 56 133 Bibliographic Data Sheet DTU Title and author s User and Programmers Guide to the Neutron Ray Tracing Package McStas Version 2 0 Peter Willendrup Emmanuel Farhi Erik Knudsen and Kim Lefmann ISBN ISSN ISBN 978 87 550 3679 6 0106 2840 Dept or group Date Materials Research Department November 2012 Groups own reg number s Project contract No Pages Tables Illustrations References 134 12 27 49 Abstract Max 2000 char The software package McStas is a tool for carrying out Monte C
23. as found in the McStas library sorted by neutron site When selecting one of these a local copy of the instrument description is transferred to the active directory so that users have modifi cation rights and loaded One may then view its source Edit and use it directly for simulations trace 3D View The Tools menu gathers minor tools Tools Plot current other results Plot current simulation results and other results Tools Online plotting of results installs a DSA key to be used for ssh clustering and MPI see Section 5 6 DTU 49 McStas Configuration options j Plotting options PGPLOT original McStas _ Use binary files faster _j 3 pane view with PGPLOT trace Clustering None single CPU Editor options Advanced built in editor W Surround strings with quotes Compilation options W Optimize g 02 Optimization options Precision Figure 5 7 The configuration options dialog in mcgui Tools Dataset convert merge Opens a GUI to the mcformat tool in order to convert datasets to other formats merge scattered dataset e g from successive or grid simulations and assemble scan sets This tool does not handle raw event files Tools Shortcut keys displays the shortcut keys used for running and editing instru ments Tools Install DSA key installs a DSA key to be used for ssh clustering and MPI see Section 5 6 The Histogrammer In addition to these tools the Neut
24. files for Isotropic SSqw component Data q w S q w Table 7 8 Data files of the McStas library MCSTAS examples Description instr This directory contains example instruments accessible throught the mcgui Neutron site menu Table 7 9 Instrument example files of the McStas library 106 DTU Chapter 8 Instrument examples In this section we present a few typical instruments We then give a longer description of three selected instruments We present the McStas versions of the Ris standard triple axis spectrometer TAS1 8 3 and the ISIS time of flight spectrometer PRISMA 8 4 But first we present an example of a component test instrument the instrument to test the component V_sample 8 2 These instrument files are included in the McStas distribution in the examples directory Most of the instrument examples there in may be executed automatically throught the McStas self test procedure see section 3 5 The list of instrument examples has been extended considerably see the Neutron Site menu in mcgui See page 5 1 8 1 A quick tour of instrument examples 8 1 1 Neutron site Brookhaven The former Brookhaven reactor hosted the H8 triple axis spectrometer This latter was modelled in order to cross check the NISP Vitess Restrax McStas and IDEAS neutron propagation Monte Carlo codes Results were published in Neutron News 13 No 4 24 29 2002 8 1 2 Neutron
25. have been defined and should be handled in the extension of the original INITIALIZE section On the other hand if you do not derive a component as a whole from a parent you may still use specific parts from any component DEFINE COMPONENT name DECLARE COPY parent1 98 DTU INITIALIZE COPY parent2 EXTEND At C code to be concatenated to the parent2 INITIALIZE ht TRACE COPY parent This mechanism may lighten the component code but a special care should be taken in mixing bits from different sources specially concerning variables This may result in difficulties to compile components 6 7 McDoc the McStas library documentation tool McStas includes a facility called McDoc to help maintain good documentation of com ponents and instruments In the source code comments may be written that follow a particular format understood by McDoc The McDoc facility will read these comments and automatically produce output documentation in various forms By using the source code itself as the source of documentation the documentation is much more likely to be a faithful and up to date description of how the component instrument actually works Two forms of documentation can be generated One is the component entry dialog in the mcgui front end see section 5 4 1 The other is a collection of web pages documenting the components and instruments handled via the mcdoc front end see section 5 4 6 and the complete documentation for all avai
26. instrument in trace mode when using PGPLOT e Clustering option None MPI ssh e Choice of editor to use when editing instrument files e Automatic quotation of strings when inserting in the built in editor e Possibility to not optimize when compiling the generated c code This is very handy when setting up an instrument model which requires regular compila tions e Adjustment of final precision when doing parameter optimization To save the chosen settings for your next McStas run use Save Configuration in the File menu File Save configuration saves user settings from Configuration options and Run dia logue to disk File Quit exits the graphical user interface front end The Simulation menu has the following features Simulation Read old simulation prompts for the name of a file from a previous run of a McStas simulation usually called mcstas sim The file will be read and any detector data plotted using the mcplot front end The parameters used in the sim ulation will also be made the defaults for the next simulation run This function is also available using the Read button to the right of the name of the current simulation data Simulation Run simulation opens the run dialog window explained further below Simulation Plot results plots using mcplot the results of the last simulation run or spawns a load dialogue to load a set of results The Neutron Site menu contains a list of template example instruments
27. integer S string Lambda D 2 36 Output to dir i force Browse Neutron count 1000000 _ gravity BEWARE Random seed Simulate steps 0 _ Plot results Format PGPLOT Clustering None single CPU Number of nodes 2 Figure 5 8 The run dialog in mcgui The run dialog is used to run simulations It allows the entry of instrument parameters as well as the specifications of options for running the simulation see section 5 3 for details It also allows to run the mcdisplay section 5 4 3 and mcplot section 5 4 4 front ends together with the simulation The meaning of the different fields is as follows Run Instrument parameters allows the setting of the values for the input parameters of the instrument The type of each instrument parameter is given in parenthe sis after each name Floating point numbers are denoted by D for the C type double I denotes integer parameters and S denotes strings For parameter scans and optimizations enter the minimum and maximum values to scan optimize separated by a comma e g 1 10 and do not forget to set the Scanpoints to more than 1 Run Output to allows the entry of a directory for storage of the resulting data files in like the dir option If no name is given the results are stored in the current directory to be overwritten by the next simulation Run Force Forces McStas to overwrite existing data files Neutro
28. language Files referenced with include are read when the instrument is translated into C by the McStas compiler and must contain valid McStas meta language input and possibly C code Files referenced with include are read when the C compiler generates an executable from the generated C code and must contain valid C Embedded C code is used in several instances in the McStas meta language Such code is copied by the McStas compiler into the generated simulation C program Embedded C code is written by putting it between the special symbols and as follows at Embedded C code ht The Z and AF must appear on a line by themselves do not add comments after Additionally if a include statement is found within an embedded C code block the specified file will be included from the share directory of the standard component library or from the current directory and any user specified search directories as a C library just like the usual include but only once For instance if many components require to read data from a file they may all ask for Zinclude read_table lib without duplicating the code of this library If the file has no extension both h and c files will be searched and included otherwise only the specified file will be imported The McStas run time shared library is included by default equivalent to Zinclude mcstas r in the DECLARE section For an
29. of specifications of the form name value separated by spaces or line breaks Multiple p options may be given together with direct parameter specifications on the command line If a parameter is assigned multiple times later assignments override previous ones The N count or numpoints count option may be used to perform a series of count simulations while varying one or more parameters within specified intervals Such a series of simulations is called a scan To specify an interval for a parameter X it should be assigned two values separated by a comma For example the command mcrun sim instr N4 X 2 8 Y 1 would run the simulation defined in sim instr four times with X having the values 2 4 6 and 8 respectively After running the simulation the results will be written to the file mcstas dat by default This file contains one line for each simulation run giving the values of the scanned input variables along with the integrated intensity and estimated error in all monitors Additionally a file mcstas sci when using Scilab format is written that can be read by the mcplot front end to plot the results on the screen or in a Postscript file see section 5 4 4 DTU e When performing a scan the f file and file file options make mcrun write the output to the files file dat and file sim instead of the default names e When performing a scan the d dir and dir dir options make mcrun put all output in a newly created directory di
30. parameters gt If you don t want to spread the simulation run it as usual sim mpi lt instrument parameters gt 5 6 3 McRun script with MPI support mpich The mcrun script has been adapted to use MPICH implementation of MPI Two new options have been added e mpi lt number gt tells mcrun to use MPI and to spread the simulation over lt number gt nodes DTU 67 e machines lt file gt defines a text file where the nodes which are to be used for parallel computation are listed by default mcrun will look at HOME mcstas hosts and MCSTAS tools perl mcstas hosts When used on a single SMP machine multi core cpu this option may be omitted When available the MPI option will show up in the mcgui Run dialog Specify the number of nodes required Suppose you have four machines named node1 to node4 A typical machine list file machines 1list looks like nodei node2 node3 node4 You can then spread a simulation sim instr using mcrun mcrun c mpi 4 machines machines list sim instr lt instrument parameters gt Warning when using mcrun with MPI be sure to recompile your simulation with MPI support see c flag of mcrun a simulation compiled without MPI support cannot be used with MPI whereas a simulation compiled with MPI support can be used without MPI 5 6 4 McStas MPI Performance Theoretically a computation which lasts T seconds on a single computer should lasts at least seconds when it
31. possible to identify clearly the purely elastic structure contribution from the quasi elastic and inelastic one dynamics What are the contributions from the sample geometry the container the sample environment and generally the instrument itself And last but not least how does multiple scattering affect the signal Most of the time the physicist will elude these questions using rough approximations or applying analytical corrections 42 Monte Carlo techniques also provide means to evaluate some of these quantities Technicalities of Monte Carlo simulation techniques are explained in detail in Chapter 4 1 2 1 The goals of McStas Initially the McStas project had four main objectives that determined its design DTU 11 Correctness It is essential to minimize the potential for bugs in computer simulations If a word processing program contains bugs it will produce bad looking output or may even crash This is a nuisance but at least you know that something is wrong However if a simulation contains bugs it produces wrong results and unless the results are far off you may not know about it Complex simulations involve hundreds or even thousands of lines of formulae making debugging a major issue Thus the system should be designed from the start to help minimize the potential for bugs to be introduced in the first place and provide good tools for testing to maximize the chances of finding existing bugs Flexibility When you c
32. site Tools This category currently contains a special Histogrammer that can read any event file from Vitess MCNP tripoli and McStas in order to produce histograms of any type This is a very powerful tool to analyse sparse huge events data sets 8 1 3 Neutron site ILL Cold and thermal guide models are given in the ILL neutron site Descriptions are very accurate based on actual drawings including all elements with curvature and gaps Sim ulated capture fluxes were found in very good agreement with measurements Additionally the IN12 triple axis instrument has been detailed in its year 2005 con figuration It is located at the end of the H15 cold guide The sample is a Vanadium rod DTU 107 The IN6 instrument is a hybrid time of flight spectrometer with 3 monochromators and a Fermi chopper The sample is an isotropic scatterer liquid It makes use of the SPLIT keyword to enhance statistics The ILL_TOF_Env example is a simple neutron pulse e g from a chopper system coming out of IN4 IN5 or IN6 illuminating a sample with a container around and a cylinder shaped sample environment 8 1 4 Neutron site tests This large set of examples shows simple instruments using particular components samples polarized beams detectors They may be used as starting point for further more complex models 8 1 5 Neutron site ISIS You will find here examples of ISIS instruments some of them using the ISIS moderator component obtaine
33. supported by DTU Physics Institut Laue Langevin ILL Paul Scherrer Institute and the Niels Bohr Institute NBI The system is well tested and is supplied with several examples and with an extensive documentation Besides this manual a separate component manual exists The release at hand McStas 2 0 is a major upgrade from the last 1 12c release meaning partial loss of backward compatibility especially in terms of uniform naming of component input parameters Porting your existing personal instrument files and components should be trivial but if you experience problems feel free to contact mcstas users mcstas org or the authors 1 1 Development of Monte Carlo neutron simulation The very early implementations of the method for neutron instruments used home made computer programs see e g papers by J R D Copley D F R Mildner J M Carpenter J Cook more general packages have been designed providing models for most parts of the simulations These present existing packages are NISP 9 ResTrax 10 McStas DTU 9 1 2 2 11 Vitess 12 13 IDEAS 14 and IB Instrument Builder 15 Supplementing the Monte Carlo based methods various analytical phase space simulation methods exist including Neutron Acceptance Diagram Shading NADS 16 Their usage usually covers all types of neutron spectrometers most of the time through a user friendly graphical interface without requiring programming skills The neutron ray tracing Mon
34. the arrays can be statically allocated in the component DECLARE section If setting parameters were used it would be necessary to allocate the arrays dynamically using e g malloc Setting parameters may optionally be declared to be of type int char and string just as in the instrument definition see section 6 3 OUTPUT PARAMETERS 51 52 This declares a list of C identifiers variables functions that are output parameters i e global for the component Output parameters are used to hold values that are computed by the component itself rather than being passed as input This could for example be a count of neutrons in a detector or a constant that is precomputed to speed up computation Using OUTPUT PARAMETERS is strongly recommended for DECLARE and internal global component variables and functions in order to prevent that instances of the same com ponent use the same variable names Moreover see section 6 5 2 below these may be accessed from any other instrument part e g using the MC_GETPAR C macro On the other hand the variables from the SHARE sections should not be defined as OUTPUT parameters The OUTPUT PARAMETERS section is optional STATE PARAMETERS 2 y Z Uz Uy Uz t 81 82 P This declares the parameters that define the state of the incoming neutron The task of the component code is to assign new values to these parameters based on the old values and the values of the definition and setting parameter
35. use the zoom factor option This magnifies the display of each component along the selected axis only e g a Soller collimator is magnified perpendicular to the neutron beam but not along it This option may produce rather strange visual effects as the neutron passes between components with different coordinate magnifications but it is occasionally useful When debugging it is often the case that one is interested only in neutrons that reach a particular component in the instrument For example if there is a problem with the sample one may prefer not to see the neutrons that are absorbed in the monochromator shielding For these cases the inspect comp option is useful With this option only neutrons that reach the component named comp are shown in the graphics display As of McStas 1 10 the PGPLOT version has a special mode for time of flight ap plications Using the new commandline options TOF T and tmax TMAX chopper acceptance diagrams can be generated from the statistical information from the simulated neutron rays As the use in non interactive please use with a limited number of neutron rays n ncount For export of graphics combine with e g gif The mcdisplay front end will then require the Perl the PGPLOT and the PGPerl packages to be installed It may be necessary to set the PGPLOT_DIR and PGPLOT_DEV environment variable consult the documentation for PGPLOT on the local system in case of difficulty Matlab and Sci
36. warning messages is an indication that either some components overlap or some neutrons are getting outside of the simulation for instance this usually happens after a monochromator as the non reflected beam is indeed lost A DTU 77 special warning appears when no neutron ray has reached some part of the simulation This is usually the sign of either overlapping components or a very low intensity For experienced users we recommend as well the usage of the WHEN and EXTEND key words as well as other syntax extensions presented in section 6 4 below 6 3 6 The SAVE section SAVE ht C code to execute each time a temporary save is required ht This gives code that will be executed when the simulation is requested to save data for instance when receiving a USR2 signal on Unix systems or using the Progress_bar component with intermediate savings It is also executed when the simulation ends This section is optional 6 3 7 The FINALLY section FINALLY ht C code to execute at end of simulation ht This gives code that will be executed when the simulation has ended When existing the SAVE section is first executed The FINALLY section is optional A simulation may be requested to end before all neutrons have been traced when receiving a TERM or INT signal on Unix systems or with Control C causing code in FINALLY to be evaluated 6 3 8 The end of the instrument definition The end of the instrument definiti
37. which for simple neutron beam configurations may be modelled with analytical methods This is valid for individual elements such as guides 10 DTU 26 27 choppers 28 29 Fermi choppers 30 31 velocity selectors 32 monochromators 33 36 and detectors 37 39 In the case of a limited number of optical elements the so called acceptance diagram theory 17 27 40 may be used within which the neutron beam distributions are considered to be homogeneous triangular or Gaussian However real neutron instruments are constituted of a large number of optical elements and this brings additional complexity by introducing strong correlations between neutron beam parame ters like divergence and position which is the basis of the acceptance diagram method but also wavelength and time The usual analytical methods such as phase space theory then reach their limit of validity in the description of the resulting effects In order to cope with this difficulty Monte Carlo MC methods for a general review see Ref 41 may be applied to the simulation of neutron instruments The use of prob ability is common place in the description of microscopic physical processes Integrating these events absorption scattering reflection over the neutron trajectories results in an estimation of measurable quantities characterizing the neutron instrument Moreover using variance reduction importance sampling where possible reduces the computati
38. wrong results with some components An instrument consists of a list of components through which the neutron ray passes one after the other The order of components is thus significant since McStas does not automatically check which component is the next to interact with the neutron ray at a given point in the simulation Note that in case of a negative propagation time from one component to the next the neutron ray is by default absorbed as this is often an indication of unphysical conditions The instrument is given a global absolute coordinate system In addition every component in the instrument has its own local coordinate system that can be given any desired position and orientation though the position and orientation must remain fixed for the duration of a single simulation By convention the z axis points in the direction of the beam the x axis is perpendicular to the beam in the horizontal plane pointing left as seen from the source and the y axis points upwards see figure 6 1 Nothing in the McStas metalanguage enforces this convention but if every component used different conventions the user would be faced with a severe headache It is therefore necessary that this convention is followed by users implementing new components In the instrument definitions units of length e g component positions are given in meters and units of angles e g rotations are given in degrees The state of the neutron is given by its position x y
39. z in meters its velocity vz vy vz in meters per second the time t in seconds and the three spin parameters Sx Sy Sz and finally the neutron weight p described in 4 DTU 71 6 2 Syntactical conventions Comments follow the normal C syntax C style comments may also be used Keywords are not case sensitive for example DEFINE define and dEfInE are all equivalent However by convention we always write keywords in uppercase to distinguish them from identifiers and C language keywords In contrast McStas identifiers names like C identifiers and keywords are case sensitive another good reason to use a consistent case convention for keywords All McStas keywords are reserved and thus should not be used as C variable names The list of these reserved keywords is shown in table 6 1 It is possible and usual to split the input instrument definition across several different files For example if a component is not explicitly defined in the instrument McStas will search for a file containing the component definition in the standard component library as well as in the current directory and any user specified search directories see section 5 2 2 It is also possible to explicitly include another file using a line of the form finclude file Beware of possible confusion with the C language include statement especially when it is used in C code embedded within the McStas meta
40. 00 00000080008 8 8 1 3 Neutron site ILL 70 71 72 74 74 75 75 75 76 78 78 78 78 80 81 81 82 84 85 86 87 87 89 90 90 90 91 94 94 96 96 98 98 98 99 101 101 107 107 107 107 107 8 1 4 Neutron site tests aoaaa ee 8 1 5 Neutron site ISIS lel toi aop na 0 0000 0000000 8 1 6 Neutron site RIS reant i sy e oaa ee SLT Noron ste PSU aa aE R en A a A er ee 8 1 8 Neutron site Tutorial s o s acna a a e a a A A L E 8 1 9 Neutron site ESS eae E aE a a T E a a D a 8 2 A test instrument for the component V sample 8 2 1 Scattering from the V sample test instrument 8 3 The triple axis spectrometer TAS oaa aaa a 8 3 1 Simulated and measured resolution of TAS 8 4 The time of flight spectrometer PRISMA 8 4 1 Simple spectra from the PRISMA instrument Random numbers in McStas A 1 Transformation of random numbers 0 00004 A 2 Random generators aooaa ee Libraries and constants B 1 Run time calls and functions mcstas r 00 B 1 1 Neutron propagation 0000008 B 1 2 Coordinate and component variable retrieval B 1 8 Coordinate transformations 2 08 B 1 4 Mathematical routines 2 2 0 0 0 0 02000008 B 1 5 Output from detectors 0 0 0 0 020000 ae B 1 6 Ray geometry intersections 004 B 1 7 Random numbers 0
41. 10 20 23 1999 See http www mcstas org See http trac mccode org report See http ts 2 isis rl ac uk See http www ess europe de See http neutron eu net en See http mensi risoe dk See http strider lansce lanl gov NISP Welcome html P A Seeger L L Daemen T G Thelliez and R P Hjelm Physica B 283 433 2000 J Saroun and J Kulda Physica B 234 1102 1997 See website http omega ujf cas cz restrax P Willendrup E Farhi and K Lefmann Physica B 350 735 2004 D Wechsler G Zsigmond F Streffer and F Mezei Neutron News 25 11 2000 See http www hmi de projects ess vitess W T Lee and X L Wang Neutron News 13 30 2002 See website http www sns gov ideas See http neutrons dev ornl gov eqsans software ib http code google com p jnads J R D Copley J Neut Research 1 21 1993 E Farhi T Hansen A Wildes R Ghosh and K Lefmann Appl Phys A 74 1471 2002 C Schanzer P Boni U Filges and T Hils Nucl Instr Meth A 529 63 2004 DTU 129 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 130 G Zsigmond K Lieutenant and S Manoshin et al Nucl Instr Meth A 529 218 2004 K Lieutenant J Phys Condens Matter 17 S167 2005 T E Mason K N Clausen G Aeppli D F McMorrow and J K Kjems Can J Phys 73 697 702 1995 K N Cla
42. 44 45 46 47 48 49 D F R Mildner C A Pellizari and J M Carpenter Acta Cryst A 33 954 1977 Grimmett G R and Stirzakerand D R Probability and Random Processes 2nd Edition Clarendon Press Oxford 1992 J A Nelder and R Mead Computer Journal 7 308 313 1965 W H Press S A Teukolsky W T Vetterling and B P Flannery Numerical Recipes 2nd Edition Cambridge University Press 2002 See http www neutron anl gov nexus A Abrahamsen N B Christensen and E Lauridsen McStas simulations of the TAS1 spectrometer Student s report Niels Bohr Institute University of Copenhagen 1998 W H Press S A Teukolsky W T Vetterling and B P Flannery Numerical Recipes in C Cambridge University Press 1986 DTU 131 Index Bugs 12 58 66 73 81 83 Can not compile 69 72 96 Code generation options 33 Comments 69 Components 13 73 Coordinate system 68 Data formats 37 40 46 57 60 72 88 Customization 90 Embedded C code 69 72 78 Environment variable MCSTAS 33 MCSTAS_FORMAT 52 PGPLOT_DEV 45 53 Environment variable BROWSER 56 98 EDITOR 45 MCSTAS 98 115 122 MCSTAS_CC 16 MCSTAS_CFLAGS 16 MCSTAS_FORMAT 16 30 37 54 58 PGPLOT_DEV 54 PGPLOT_DIR 45 53 54 Gravitation 68 Grid computing 57 Grid computing 47 61 Installing 14 16 52 Instruments 71 77 ITERATE 82 Kernel 12 67 Keyword 69 include 33 69 78 115 ABSOLUTE
43. 60 5 5 Data formats Analyzing and visualizing the simulation results 61 5 5 1 McStas and PGPLOT format 61 5 5 2 Matlab Scilab and IDL formats 0 62 5 5 3 HTML VRML and XML formats sys a eo we eS 63 5 5 4 N Xus formats osas dass orea aoa ea be ee we bad 63 5 6 Using computer Grids and Clusters 2 0 ee 64 5 6 1 Distribute mcrun simulations on grids multi cores and clusters SSH TC ra Maacaga tls lak atts e N ead a Ste RS ant aes 65 5 6 2 Parallel computing MPI 4 2 sc0 8 dee ea ee Ae ele 66 5 6 3 McRun script with MPI support mpich 67 5 6 4 McStas MPI Performance 2 4 20 ese eae Aree Doar ty Gok a A 68 5 6 5 MPI and Grid Bugs and limitations 6 The McStas kernel and meta language 6 1 6 2 6 3 6 4 6 5 6 6 6 7 Notational conventions Syntactical conventions Writing instrument definitions oo aa 2 02000 00048 6 3 1 The instrument definition head 00084 6 3 2 The DECLARE section 0 000000 eee eee ee 6 3 3 The INITIALIZE s 6 3 4 The NEXUS extensi 6 3 5 The TRACE section 6 3 6 The SAVE section CCHIONN A bse Yak hk ee BA Be at Ane ee Be OW fo Gk i oe ng a epee See ee ae eee ee i 6 3 7 The FINALLY section 2 0 0 0 020 ee ee ee 6 3 8 The end of the instrument definition 0 0 6 3 9 Code for the instrument vanadium_example instr Writing instrument definitions co
44. 8 3 A sketch of the TAS1 instrument 8 3 1 Simulated and measured resolution of TAS1 In order to test the McStas package on a qualitative level we have performed a very detailed comparison of a simulation with a standard experiment from TAS1 The mea surement series constitutes a complete alignment of the spectrometer using the direct beam and scattering from V and Al203 samples at an incoming energy of 20 0 meV using the second order scattering from the monochromator In these simulations we have tried to reproduce every alignment scan with respect to position and width of the peaks whereas we have not tried to compare absolute intensities Below we show a few comparisons of the simulations and the measurements Figure 8 4 shows a scan of 26 on the collimated direct beam in two axis mode A 1 mm slit is placed on the sample position Both the measured width and non Gaussian peak shape are well reproduced by the McStas simulations In contrast a simulated 26 scan in triple axis mode on a V sample showed a surprising offset from 0 degrees However a simulation with a PSD on the sample position showed that the beam center was 1 5 mm off from the center of the sample and this was important since the beam was no wider than the sample itself A subsequent centering of the beam resulted in a nice agreement between simulation and measurements For a comparison on a slightly different instrument analyser detector collimator inserted see Figur
45. Apart from the meta language McStas also includes a number of C library functions and definitions that are useful for neutron ray tracing simulations The definitions avail able for component developers are listed in appendix B The list includes functions for e Computing the intersection between a flight path and various objects such as planes cylinders boxes and spheres e Functions for generating random numbers with various distributions e Functions for reading or writing information from to data files e Convenient conversion factors between relevant units etc The McStas meta language was designed to be readable with a verbose syntax and ex plicit mentioning of otherwise implicit information The recommended way to get started with the meta language is to start by looking at the examples supplied with McStas modifying them as necessary for the application at hand 70 DTU Z gt Figure 6 1 conventions for the orientations of the axes in simulations 6 1 Notational conventions Simulations generated by McStas use a semi classical description of the neutron rays to compute the neutron trajectory through the instrument and its interaction with the differ ent components The effect of gravity is taken into account either in particular components e g Guide_gravity or more generally when setting an execution flag g to perform gravitation computation This latter setting is only an approximation and may produce
46. DEFINITION PARAMETERS SETTING PARAMETERS xmin 0 01 xmax 0 01 ymin 0 01 ymax 0 01 radius 0 cut 0 xwidth 0 yheight 0 OUTPUT PARAMETERS Neutron parameters x y Z VxX vy vzZ t sx sy sz p INITIALIZE At if xwidth gt 0 xmax xwidth 2 xmin xmax if yheight gt 0 ymax yheight 2 ymin ymax if xmin 0 amp amp xmax 0 amp amp ymin 0 amp amp ymax 0 amp amp radius 0 fprintf stderr Slit 4s Error give geometry n NAME_CURRENT_COMP exit 1 ht TRACE Ae PROP_ZO if radius 0 amp amp x lt xmin x gt xmax y lt ymin y gt ymax radius 0 amp amp x x y y gt radius radius ABSORB else if p lt cut ABSORB else SCATTER hh MCDISPLAY Ae magnify xy if radius 0 double xw yh xw xmax xmin 2 0 yh ymax ymin 2 0 multiline 3 xmin xw double ymax 0 0 double xmin double ymax 0 0 double xmin ymaxtyh 0 0 multiline 3 xmaxt xw double ymax 0 0 double xmax double ymax 0 0 double xmax ymaxtyh 0 0 multiline 3 xmin xw double ymin 0 0 double xmin double ymin 0 0 double xmin ymin yh 0 0 multiline 3 xmaxt xw double ymin 0 0 double xmax double ymin 0 0 double xmax ymin yh 0 0 else circle xy 0 0 0 radius ht END DTU 97 6 6 Extending component definitions Suppose you are interested by one component of the McStas library but you would like to cus
47. Duplication of component instances COPY Often one has a set of similar component instances in an instrument These could be e g a set of identical monochromator blades or a set of detectors or guide elements To gether with JUMPs see below there is a way to copy a component instance duplicating parameter set as well as any EXTEND GROUP JUMP and WHEN keyword Position 82 DTU AT and rotation ROTATED specification must be explicitely entered in order to avoid component overlapping The syntax for instance copy is COMPONENT name COPY instance name where instance_name is the name of a preceeding component instance in the instrument It may be PREVIOUS as well If you would like to change only some of the parameters in the instance copy you may write e g COMPONENT name COPY instance_name parl 0 par2 1 which will override the original instance parameter values This possibility to over ride parameters is very useful in case of describing e g sample environments using the Isotropic_Sqw and PowderN components which allow concentric geometry first in stance must have concentric 1 and the second concentric 0 In case EXTEND GROUP JUMP and WHEN keywords are defined for the copied instance these will over ride the settings from the copied instance In the case where there are many duplicated components all originating from the same instance there is a mechanism for automating copied instance names COM
48. ENT d2 Detector FINALLY At DTU 89 printf Detector counts d1 d d2 d n MC_GETPAR d1 counts MC_GETPAR d2 counts ht This way McStas takes care to rename transparently the two counts OUTPUT parameters so that they are distinct and can be accessed from elsewhere in the instrument EXTEND FINALLY SAVE or from other components This particular example is outdated since McStas monitors will themselves output their contents 6 5 3 The SHARE section SHARE ht C code shared declarations variables definitions functions These should not be OUTPUT parameters hy The SHARE section has the same role as DECLARE except that when using more than one instance of the component it is inserted only once in the simulation code No occurrence of the items to be shared should be in the OUTPUT parameter list not to have McStas rename the identifiers This is particularly useful when using many instances of the same component for instance guide elements If the declarations were in the DECLARE section McStas would duplicate it for each instance making the simulation code longer A typical example is to have shared variables functions type and structure definitions that may be used from the component TRACE section For an example of SHARE see the samples Single_crystal component The Zinclude file keyword may be used to import a shared library The SHARE section is optional 6 5 4 The INITIALI
49. Grid Bugs and limitations e Some header of output files might contain minor errors e The computation split does not take into account the speed or the load of nodes the overall time of a distributed computation is forced by the slowest node for optimal performance the cluster should be homogeneous e Interacting with a running simulation USR1 and USR2 signals is disabled with DTU 69 Chapter 6 The McStas kernel and meta language Instrument definitions are written in a special McStas meta language which is translated automatically by the McStas compiler into a C program which is in turn compiled to an executable that performs the simulation The meta language is custom designed for neutron scattering and serves two main purposes i to specify the interaction of a single neutron ray with a single optical component and ii to build a simulation by constructing a complete instrument from individual components For maximum flexibility and efficiency the meta language is based on C Instrument geometry propagation of neutrons between the different components parameters data input output etc is handled in the meta language and by the McStas compiler Complex calculations are written in C embedded in the meta language description of the compo nents However it is possible to set up an instrument from existing components and run a simulation without writing a single line of C code working entirely in the meta language
50. O C standard so it should be easy to compile it using any ISO C or C compiler E g a typical Unix style command would be cc 0 o name out name c lm The McStas team recommends these compiler alternatives for the Intel and AMD hard ware architectures A gcc which is a very portable open source ISO C compatible c compiler available for most platforms For Linux it is usually part of your distribution for Windows the McStas distribution package includes a version of gcc in the Dev CPP sub package and for Mac OS X gcc is part of the Xcode tools package available on the installation medium B icc or the Intel c compiler is available for Linux Mac OS and Windows systems and is a commercial software product Generally simulations run with the Intel compiler are a factor of 2 faster than the identical simulation run using gcc To use icc with McStas on Linux or Mac OS X set the environment variables DTU 37 MCSTAS_CC icc MCSTAS_CFLAGS g 02 wd177 266 1011 181 To use icc with MPI on Unix system see Section 5 6 installations it seems that editing the mpicc shell script and setting the CC variable to icc is the only requirement On Windows the Intel c compiler is icl not icc and has a dependency for Microsoft Visual C If you have both these softwares available running McStas with the Intel compiler should be possible currently untested by the McStas developer team The 0 option typical
51. OP_ propagation the neutron rays which have negative propagation times are removed automatically e ABSORB This macro issues an order to the overall McStas simulator to interrupt the simulation of the current neutron history and to start a new one 118 DTU PROP _ZO Propagates the neutron to the z 0 plane by adjusting x y z and t accordingly from knowledge of the neutron velocity vz vy vz If the propagation time is negative the neutron ray is absorbed except if a ALLOW_BACKPROP preceeds it For components that are centered along the z axis use the _intersect functions to determine intersection time s and then a PROP_DT call PROP_DT dt Propagates the neutron through the time interval dt adjusting x y z and t accordingly from knowledge of the neutron velocity This macro auto matically calls PROP_GRAV_DT when the gravitation option has been set for the whole simulation PROP_GRAV_DT dt Az Ay Az Like PROP_DT but it also includes gravity using the acceleration Az Ay Az In addition to adjusting x y z and t also vax vy vz is modified ALLOW_BACKPROP Indicates that the next propagation routine will not re move the neutron ray even if negative propagation times are found Further propa gations are not affected SCATTER This macro is used to denote a scattering event inside a component It should be used e g to indicate that a component has interacted with the neutron ray e g scattered or det
52. PONENT COPY root_name COPY instance_name will concatenate a unique number to root_name avoiding name conflicts As a side effect referring to this component instance for e g further positioning is not straight forward as the name is determined by McStas and does not depend completely on the user s choice even though the PREVIOUS keyword may still be used We thus recommend to use this naming mechanism only for components which should not be referred to in the instrument This automatic naming may be used anywhere in the TRACE section of the instru ment so that all components which do not need further referring may be labelled as COPY Origin As an example we show how to build a guide made of equivalent elements Only the first instance of the Guide component is defined whereas following instances are copies of that definition The instance name of Guide components is set automatically COMPONENT CG_In Arm AT COMPONENT CG_1 Guide_gravity 1 L n m 1 AT 0 0 0 RELATIVE PREVIOUS COMPONENT COPY CG_1 COPY CG_1 AT 0 0 L nt td RELATIVE PREVIOUS ROTATED 0 L nt d R 180 PI 0 RELATIVE PREVIOUS COMPONENT COPY CG_1 COPY CG_1 AT 0 0 L nt td RELATIVE PREVIOUS ROTATED 0 L n d R 180 PI 0 RELATIVE PREVIOUS COMPONENT CG_Out Arm AT 0 0 L n RELATIVE PREVIOUS DTU 83 6 4 4 Conditional components WHEN One of the most useful features of the extended McStas syntax is the conditional WHEN m
53. RMAT Sets the file format for result data files from monitors This enables to have simulation files in one format e g HTML and monitor files in an other format e g VRML mpi NB_CPU Distributes the simulation over NB_CPU node re quires MPI to be installed Speedup has been demon strated to be linear in number of nodes when the sim ulation task is ncount is sufficiently large multi NB_CPU grid NB_CPU Distributes the simulation over NB_CPU node re quires SSH to be installed Speedup has been demon strated to be linear in number of nodes when the sim ulation task is ncount is sufficiently large machines MACHINES Specify a list of distant machines nodes to be used for MPI and grid clustering Default is to use local SMP cluster optim Run in optimization mode to find best parameters in order to maximize all monitor integral values Param eters to be varied are given just like scans min max optim COMP Same as optim but for specified monitors This op tion may be used more than once optim prec ACCURACY Sets accuracy criteria to end parameter optimization default is 1073 test Run McStas self test c force compile Force to recompile the instrument Table 5 2 Additional options accepted by McStas simulations 42 DTU McStas PGPLOT Scilab Scilab_binary Matlab Matlab_binary Octave Octave_binary IDL IDL_b
54. T Positions filename Gui support for array AT specifiers More complete polarisation support including numerically defined magnetic fields and advanced sample components Perl or python plotting alternative to PGPLOT Larger variety of sample components DTU Chapter 3 Installing McStas Please REWRITE The information in this chapter is also available as a separate html ps pdf document in the install_docs folder of your McStas installation package 3 1 Getting McStas The McStas package is available in various distribution packages from the project website at http www mcstas org download DTU McStas 2 0 1686 Win32 exe Windows self extracting executable including essential support tools Refer to section 3 3 McStas 2 0 dmg Mac OS X disk image for PPC and Intel machines Please follow the instructions in the README file in the disk image mcstas 2 0 1386 deb Binary Debian GNU Linux packages for 32 bit Intel AMD processors currently built on Debian stable Tested to work on Ubuntu and Debian systems Refer to section 3 4 1 mcstas 2 0 amd64 deb Binary Debian GNU Linux packages for 64 bit Intel AMD processors currently built on Debian stable Tested to work on Ubuntu and Debian systems Refer to section 3 4 1 mcstas 2 0 i1686 unknown Linux tar gz Binary package for Linux systems currently built on Debian stable Should work on most Linux setups Refer to section 3 4 mcstas 2 0 src tar gz
55. TIVE PREVIOUS COMPONENT CG_2 Guide_gravity 1 L n m 1 AT 0 0 0 RELATIVE PREVIOUS ROTATED 0 L ntd R 180 PI 0 RELATIVE PREVIOUS JUMP CG_2_Position ITERATE n COMPONENT CG_Out Arm AT 0 0 L n RELATIVE PREVIOUS DTU 85 Similarly to the WHEN modifier see section 6 4 4 JUMP only applies within the TRACE section of the instrument definition Other sections INITIALIZE SAVE MCDISPLAY FINALLY are executed independently of the jump As a side effect the 3D view of the instrument mcdisplay will show components as if there was no jump This means that in the following example the very long guide 3D view only shows a single guide element It is not recommended to use the JUMP inside GROUPs as the JUMP condition counter applies to the component instance within its group We would like to emphasize the potential errors originating from such jumps Indeed imbricating many jumps may lead to situations were it is difficult to understand the flow of the simulation We thus recommend the usage of JUMPs only for experienced and cautious users 6 4 6 Enhancing statistics reaching components SPLIT The following method applies when the incoming neutron event distribution is considered to be representative of the real beam but neutrons are lost in the course of propagation with low efficiency processes absorption etc Then one may think that it s a pity to have so few events reaching the interesting part of the inst
56. User and programmers guide to the neutron ray tracing package McStas version 2 0 P Willendrup E Farhi E Knudsen U Filges K Lefmann Physics Department Technical University of Denmark 2800 Kongens Lyngby Denmark November 2012 McStas DTU n Abstract The software package McStas is a tool for carrying out Monte Carlo ray tracing simulations of neutron scattering instruments with high complexity and precision The simulations can compute all aspects of the performance of instruments and can thus be used to optimize the use of existing equipment design new instrumentation and carry out virtual experiments for e g training experimental planning or data analysis McStas is based on a unique design where an automatic compilation process translates high level textual instrument descriptions into efficient ANSI C code This design makes it simple to set up typical simulations and also gives essentially unlimited freedom to handle more unusual cases This report constitutes the reference manual for McStas and together with the manual for the McStas components it contains documentation of most aspects of the program It covers the various ways to compile and run simulations a descrip tion of the meta language used to define simulations and some example simulations performed with the program This report documents McStas version 2 0 released November 2012 The authors are Peter Kj r Willendrup lt pkwi fysik dtu dk gt
57. ZE section INITIALIZE ht C code initialization ht This gives C code that will be executed once at the start of the simulation usually to ini tialize any variables declared in the DECLARE section This section is optional Component setting parameters may be modified in this section affecting the rest of the component 6 5 5 The TRACE section TRACE ht C code to compute neutron interaction with component ht This performs the actual computation of the interaction between the neutron ray and the component The C code should perform the appropriate calculations and assign the 90 DTU resulting new neutron state to the state parameters Most components will require prop agation routines to reach the component entrance area Special macros PROP_ZO and PROP_DT are provided to automate this process see section B 1 The C code may also execute the special macro ABSORB to indicate that the neutron has been absorbed in the component and the simulation of that neutron will be aborted On the other hand if the neutron event should be allowed be backpropagated the special macro ALLOW_BACKPROP should precede the call to the PROP_ call inside the component When the neutron state is changed or detected for instance if the component simulates multiple events as multiple reflections in a guide the special macro SCATTER should be called This does not affect the results of the simulation in any way but it allows the fr
58. a divergence collimator before the monitor Alternatively write a new component that produce the required figure of merit The optimization search interval constrains the evolution of parameters It should be chosen carefully In particular it is safer for it to indeed contain a high signal domain and be preferably symmetric with respect to that maximum Using custom optimization routines The user should write a function script or a program that e inputs the simulation parameters which are usually numerical values such as TT in the prisma2 instrument from the examples directory of the package e builds a command line from these parameters e executes that command and waits until the end of the computation e reads the relevant data from the monitors e outputs a simulation quality measurement from this data usually the integrated counts or some peak width For instance for the prisma2 instrument we could write a function for Matlab see section 5 5 for details about the Matlab data format in order to study the effects of the TT parameter 46 DTU function y instr_value p TT p 1 p may be a vector matrix containing many parameters syscmd mcrun prisma2 instr n1e5 TT num2str TT gt PHA 22 PHA1 3 PHA2 2 PHA3 1 PHA4 0 PHA5 1 gt PHA6 2 PHA7 3 TTA 44 format Matlab binary system syscmd path path execute simulation and rehash files s mcstas get the simulation data and the monito
59. al behaviour of such studies may be obtained using the classical analytical approaches but accurate estimates for the flux resolution and generally the optimum parameter set benefit considerably from MC methods Mathematically the Monte Carlo method is an application of the law of large numbers 41 44 Let f u be a finite continuous integrable function of parameter u for which an integral estimate is desirable The discrete statisti cal mean value of f computed as a series in the uniformly sampled interval a lt u lt b converges to the mathematical mean value of f over the same interval n b im gt ra flaau 4 1 i 1 a lt u lt b In the case were the u values are regularly sampled we come to the well known midpoint integration rule In the case were the u values are randomly but uniformly DTU 25 sampled this is the Monte Carlo integration technique As random generators are not perfect we rather talk about quasi Monte Carlo technique We encourage the reader to consult James 41 for a detailed review on the Monte Carlo method 4 2 The neutron weight A totally realistic semi classical simulation will require that each neutron is at any time either present or lost In many instruments only a very small fraction of the initial neutrons will ever be detected and simulations of this kind will therefore waste much time in dealing with neutrons that never hit the relevant detector or monitor An important way of speedin
60. all instrument called McStas_Compo instr consisting of DTU 59 1 component Vitess_input 2 component Compo 3 component Vitess_output This file is parsed to generate the C file McStas_Compo c The last step is to compile the C file and build the executable Vitess module McStas_compo For both steps McStas is used as for any other instrument McStas_compo has to be moved to the directory MODULES that contains all VITESS executables Additionally a fileMcStas_compo tcl is created that contains most of what is needed to get a GUI window in VITESS To obtain that the content of this file has to added into vitess tcl To make it accessible within the given GUI structure of VITESS it is necessary to add the name compo NO capital letters to one of the folders in proc makeModuleSets beginning of vitess tcl The component Virtual_input transfers all neutron parameters to the McStas definition of the co ordinate system and the units used Of course Virtual_output transfers it back This means that Compo works with the McStas definition of co ordinate system and units while it is used as a VITESS module Be careful with axis labelling The original parameters of the component and its position have to be given The origin of this shift is the centre of the end of the previous module as it is always the case in VITESS It is important to notice that as VITESS uses the standard output stream stdout to se
61. ally over the whole 47 solid angle and all interesting directions are known to be contained within a certain solid angle interval AQ only these solid angles are used for the Monte Carlo choice of scattering direction This implies fuc AQ 1 However if the physical events are distributed uniformly over the unit sphere we would have P AQ AQ 47 according to Eq 4 9 One thus ensures that the mean simulated intensity is unchanged during a correct direction focusing while a too narrow focusing will result in a lower i e wrong intensity since we cut neutrons rays that should have reached the final detector 4 4 Adaptive and Stratified sampling Another strategy to improve sampling in simulations is adaptive importance sampling also called variance reduction technique where McStas during the simulations will de termine the most interesting directions and gradually change the focusing according to that Implementation of this idea is found in the Source_adapt and Source_Optimizer components 28 DTU Figure 4 1 Illustration of the effect of direction focusing in McStas Weights of neutrons emitted into a certain solid angle are scaled down by the full unit sphere area An other class of efficiency improvement technique is the so called stratified sampling It consists in partitioning the event distributions in representative sub spaces which are then all sampled individualy The advantage is that we are then sure tha
62. ample Res_monitor TOF_Res_monitor Sample like component for resolution function calculation Sample like component for resolution function calculation in TOF instruments Monitor for resolution function calculations Monitor for resolution function calculations for TOF instru ments Table 7 3 Sample components of the McStas library MCSTAS monitors Description DivPos_monitor Divergence_monitor E_monitor L_monitor Monitor Monitor_nD PSD_monitor PSD_monitor_4PI PreMonitor_nD TOFLambda_monitor TOF_monitor Divergence position monitor acceptance diagram Horizontal vertical divergence monitor 2D Energy sensitive monitor Wavelength sensitive monitor Simple single detector monitor General monitor that can output 0 1 2D signals Intensity or signal vs something and vs something Position sensitive monitor Spherical position sensitive detector This component is a PreMonitor that is to be used with one Monitor_nD in order to record some neutron parameter corre lations Time of flight vs wavelength monitor Rectangular Time of flight monitor Table 7 4 Selected Monitor components of the McStas library 104 DTU MCSTAS misc Description Progress_bar Beam_spy Set_pol Vitess_input Vitess_output Displays status of a running simulation May also trigger in termediate SAVE A monitor that displays mean statistics no output file se
63. and files Finally sending a kill 13277 which is equivalent to kill TERM 13277 will end the simulation before the initial ncount preset A typical usage example would be for instance to save data during a simulation plot or analyze it and decide to interrupt the simulation earlier if the desired statistics has been achieved This may be done automatically using the Progress_bar component Whenever simulation data is generated before end or the simulation is interrupted the ratio field of the monitored data will provide the level of achievement of the computa tion for instance 3 33e 05 1e 06 Intensities are then usually to be scaled accordingly by the user Additionally any system error will result in similar messages giving indication about the occurrence of the error component and section Whenever possible the simulation will try to save the data before ending Most errors appear when using a newly written component in the INITIALIZE TRACE or FINALLY sections Memory errors usually show up when C pointers have not been allocated unallocated before usage whereas mathe matical errors are found when for instance dividing by zero 5 3 4 Optimizing simulation speed There are various ways to speed up simulations e Optimize the compilation of the instrument as explained in section 5 2 4 e Execute the simulation in parallel on a computer grid or a cluster with MPI or ssh grid as explained in section 5 6
64. anual Generally most component parameter names have been uniformized 2 3 1 Components e Single_crystal comp now provides a lattice curvature option M Schulz FRM2 e PowderN comp now handles strain E Farhi R Rogge e Monitor_nD was wrong in the determination of the flux per cm2 and steradians e Guide_anyshape allows to model any reflecting geometry from a set of vertices and polygons e Lens models refractive lenses and prisms E Farhi C Monzat ILL e Lens_simple for refractive lenses with a simple analytical model H Frielinghaus e PSD_Detector can now work in event mode E Farhi e Multilayer_Sample to model a set of refractive layers for e g reflectometry Re quires GNU Scientific Library to be previously installed R Dalgliesh ISIS 2 3 2 New example instruments and Data files e Updated ILL instrument models H16 guide IN22 IN12 D16 e New ILL instrument models IN1 IN8 IN20 D2B D4 IN5 e Added HZB NEAT ToF spectrometer model E Farhi R Lechner e Adde ISIS_CRISP which uses new multilayer sample R Dalgliesh ISIS e Sort test instruments in various categories e Update of many data files 2 4 Documentation e Manual and component manual fully updated 16 DTU 2 5 Tools installation 2 5 1 New tool features Support for per user mcstas_config perl file located in HOME mcstas This folder is also the default location of the host list for use with MPI or gridding simply na
65. arlo ray tracing simulations of neutron scattering instruments with high complexity and precision The simulations can compute all aspects of the performance of instruments and can thus be used to optimize the use of existing equipment design new instrumentation and carry out virtual exper iments for e g training experimental planning or data analysis McStas is based on a unique design where an automatic compilation process translates high level textual in strument descriptions into efficient ANSI C code This design makes it simple to set up typical simulations and also gives essentially unlimited freedom to handle more unusual cases This report constitutes the reference manual for McStas and together with the manual for the McStas components it contains documentation of most aspects of the program It covers the various ways to compile and run simulations a description of the meta language used to define simulations and some example simulations performed with the program Descriptors Neutron Instrumentation Monte Carlo Simulation Software Available on request from Information Service Department Riso DTU Afdelingen for Informationsservice Riso DTU P O Box 49 DK 4000 Roskilde Denmark Phone 45 4677 4004 Telefax 45 4677 4013
66. ation handling for the whole gravitation neutron propagation within the instrument May produce wrong results if the used components do no comply with this option format FORMAT Sets the file format for result simulation and data files N STEPS Divide simulation into STEPS varying parameters within given ranges min max param value Set the value of an instrument parameter rather than having to min mar prompt for each one Scans ranges are specified as min max Table 5 1 Options accepted by McStas simulations For options specific to MPI and parallel computing see section 5 6 scilab gt exec mcstas sci 1 s get_mcstas scilab gt exec mcstas sci 1 s get_mcstas plot and for IDL idl gt s mcstas idl gt s mcstas plot See section 5 4 4 for another way of plotting simulation results using the mcplot front end When choosing the HTML format the simulation results are saved as a web page whereas the monitor data files are saved as VRML files displayed within the web page 5 3 3 Interacting with a running simulation Once the simulation has started it is possible under Unix Linux and Mac OS X systems to interact with the on going simulation This feature is not available when using MPI DTU 41 f file Write all data into a single file file Avoid when using file file binary formats a Do not put any headers in the data files data only format_data FO
67. back end Clicking the left mouse button on a plot produces a full window version of that plot The P key saves a postscript file containing the current plot that can be sent to the printer to obtain a hardcopy the C key produces color postscript The Q key quits the program or CTRL C in the controlling terminal may be used as normal To use the mcplot front end with PGPLOT the programs Perl PGPLOT PgPerl and PDL must all be properly installed on the system It may be necessary to set the PGPLOT_DIR and PGPLOT_DEV environment variable consult the documentation for PG PLOT on the local system in case of difficulty Matlab and Scilab back ends_ A dedicated McStas Mcplot Dialog or menu attached to the plotting window is available and provides many operations duplication export colormaps The corresponding mcplot Matlab and Scilab functions may be called from these language prompt with the same method as in section 5 3 e g matlab gt s mcplot DTU 57 matlab gt help mcplot scilab gt s mcplot matlab or scilab gt s mcplot mcstas m matlab or scilab gt mcplot s A full parameter scan simulation result or simply one of its scan steps may be displayed using the Scan step menu item When the nw option is specified a separate Matlab or Scilab window will appear instead of being launched in the current terminal This will then enable Java support under Matlab and Tk support under Scilab
68. c compiler Visualisation Scilab N Visualisation Browser Visualisation IDL Executable binary Figure 5 1 An illustration of the structure of McStas 5 1 Brief introduction to the graphical user interface This section gives an ultra brief overview of how to use McStas once it has been prop erly installed It is intended for those who do not read manuals if they can avoid it For details on the different steps see the following sections This section uses the Samples_vanadium instr file supplied in the examples directory of the McStas distri bution To start the graphical user interface of McStas run the mcgui command which will open a window with a number of menus see figure 5 2 To load an instrument select Fle Simulation Neutron site Help McDoc Instrument file lt N Undefined site Edit New Simulation results FZ_Juelich Read Status Ok Brookhaven 6 atory 1997 2006 in 2003 2006 l rights rese tests Plotters Scilat m Clustering meth Risoe PSI templateDIFF templateTAS vanadium_example Figure 5 2 The graphical user interface mcgui 32 DTU Tutorial from the Neutron site menu and open the file Samples_vanadium Next check that the current plotting backend setting select Choose backend from the Sim ulation menu corresponds to your system setup The default setting can be adjusted as exp
69. ch SPLIT loop which usually means that they must be reset inside the SPLITed section and assigned used further on A usage example of the SPLIT keyword can be found in the Neutron site ILL ILL_H15_IN6 instrument from the mcgui to enhance statistics for neutrons scattering on monochromators and sample 86 DTU 6 5 Writing component definitions The purpose of a McStas component is to model the interaction of a neutron with a physical component of a real instrument Given the state of the incoming neutron ray the component definition calculates the state of the neutron ray when it leaves the component The calculation of the effect of the component on the neutron is performed by a block of embedded C code One example of a component definition is given in section 6 5 10 and all component definitions can be found on the McStas web page 2 and are described in the McStas component manual There exists a large number of functions and constants available in order to write efficient components See appendix B for e neutron propagation functions e geometric intersection time computations e mathematical functions e random number generation e physical constants e coordinate retrieval and operations e file generation routines for monitors e data file reading 6 5 1 The component definition header DEFINE COMPONENT name This marks the beginning of the definition and defines the name of the component DEFINITION PARAMETERS dj dz
70. cies are pdl g77 libx11 dev xserver xorg dey libxt dev on Ubuntu make install pgplot amp amp configure The installation of McStas in step 5 by default installs in the usr local directory which on most systems requires superuser root privileges 3 4 4 Specifying non standard options To install in a different location than usr local use the prefix option to configure in step 2 For example configure prefix home joe will install the McStas programs in home joe bin and the library files needed by McStas in home joe lib mestas On 64 bits systems you may have to use configure with pic before installing PGPLOT with make install pgplot To enable NeXus format in mcformat you need the NeXus and HDF libraries and have to use configure with nexus 22 DTU To specify a non standard C compiler e g gcec 4 2 or icc that support OpenMP you may use e g configure with cc gcc 4 2 To enable a non standard C compiler to be used with MPI you may have to edit your mpicc shell script to set e g CC icc or redefine the cc to point to your prefered compiler e g In s usr bin gec 4 2 usr bin cc In case configure makes an incorrect guess some environment variables can be set to override the defaults e The CC environment variable may be set to the name of the C compiler to use this must be an ANSI C compiler This will also be used for the automatic compilation of McStas simulat
71. component definition named comp The parameter list gives the setting and definition parameters for the component The expressions 1 2 define the values of the parameters For setting parameters arbitrary ANSI C expressions may be used while for definition parameters only constant numbers strings names of instrument parameters or names of C identifiers are allowed see section 6 5 1 for details of the difference between definition and setting parameters To assign the value of a general expression to a definition parameter it is necessary to declare a variable in the DECLARE section assign the value to the variable in the INITIALIZE section and use the variable as the value for the parameter The McStas program takes care to rename parameters appropriately in the output so that no conflicts occur between different component definitions or between component and instrument definitions It is thus possible and usual to use a component definition multiple times in an instrument description Beware about variable type conversion when setting numerical parameter values as in pi 12 1000 In this example the parameter p1 will be set to O as the division of the two integers is indeed 0 To avoid that use explicitely floating type numbers as in p1 12 0 1000 The McStas compiler will automatically search for a file containing a definition of the component if it has not been declared previously The definition is searched for in a file called
72. concatenate all data from tezt file filename into Table which is as well initialized in the process The block number changes when the numerical data changes its size or a comment is encoutered lines starting by If the data could not be read then Table data is NULL and Table rows 0 You may then try to read it using Table_Read_Offset_Binary Return value is the number of elements read e Table_Read_Offset amp T able filename block amp offset Nrows does the same as Table_Read except that it starts at offset offset 0 means begining of file and reads Nrows lines 0 for all The offset is returned as the final offset reached after reading the nNrows lines DTU 123 e Table Read _Offset_Binary amp Table filename type block amp offset Nrows Ncolumns does the same as Table_Read_Offset but also specifies the type of the file may be float or double the number Nrows of rows to read each of them having neotumns elements No text header should be present in the file e Table Rebin amp T able rebins all Table rows with increasing evenly spaced first column index 0 e g before using Table_Value Linear interpolation is performed for all other columns The number of bins for the rebinned table is determined from the smallest first column step e Table_Info T able print information about the table Table e Table Index Table m n reads the Table m n element e Table_Value T able x n loo
73. contain text header lines These files are platform dependent little or big endian 124 DTU The filename is first searched into the current directory and all user additional loca tions specified using the I option see the Running McStas chapter in the User Manual and if not found in the data sub directory of the MCSTAS library location This way you do not need to have local copies of the McStas Library Data files see table 7 8 A usage example for this library part may be t_Table Table declare a t_Table structure char file BeO trm a file name double x y Table_Read kTable file 1 initialize and read the first numerical block Table_Info Table display table informations x Table_Index Table 2 5 read the 3rd row 6th column element of the table Indexes start at zero in C y Table_Value Table 1 45 1 look for value 1 45 in 1st column x axis and extract 2nd column value of that row Table_Free amp Table free allocated memory for table Additionally if the block number 3rd argument of Table_Read is 0 all blocks will be concatenated The Table_Value function assumes that the x axis is the first column index 0 Other functions are used the same way with a few additional parameters e g specifying an offset for reading files or reading binary data This other example for text files shows how to read many data blocks t_Table Table declare a t_Table structur
74. crun p on Windows provides a convenient command line interface for running simulations with the same automatic compilation features available in the mcgui front end It also provides a facility for running a series of simulations while varying an input parameter The command mcrun sim args will compile the instrument definition sim instr if necessary into an executable simu lation sim out It will then run sim out passing the argument list args The possible arguments are the same as those accepted by the simulations themselves as described in section 5 3 with the following extensions DTU 53 54 Component definition Slit Rectangular slit Author KL HMR Date June 16 1997 Lower x bound pmax m Upper x bound ymin m Lower y bound ymax m Upper y bound DESCRIPTION A simple rectangular slit No transmission around the slit is allowed AT 1 CE RELATIVE a Fi EE Cancel Figure 5 9 Component parameter entry dialog e The c or force compile option may be used to force the recompilation of the instrument definition regardless of file dates This may be needed in case any component definitions are changed in which case mcrun does not automatically recompile or if a new version of McStas has been installed The p file or param file option may be used to specify a file containing assign ment of values to the input parameters of the instrument definition The file should consist
75. ction from guide walls e Passage of a continuous beam through a chopper e Scattering from all types of samples 4 2 1 Statistical errors of non integer counts In a typical simulation the result will consist of a count of neutrons histories rays with different weights The sum of these weights is an estimate of the mean number of 26 DTU neutrons hitting the monitor or detector per second in a real experiment One may write the counting result as Soi Np 4 3 where N is the number of rays hitting the detector and the horizontal bar denotes av eraging By performing the weight transformations the statistical mean value of I is unchanged However N will in general be enhanced and this will improve the accuracy of the simulation To give an estimate of the statistical error we proceed as follows Let us first for simplicity assume that all the counted neutron weights are almost equal p p and that we observe a large number of neutrons N gt 10 Then N almost follows a normal distribution with the uncertainty o N VN Hence the statistical uncertainty of the observed intensity becomes ol vVNp 1 VN 4 4 as is used in real neutron experiments where p 1 For a better approximation we return to Eq 4 3 Allowing variations in both N and p we calculate the variance of the resulting intensity assuming that the two variables are statistically independent a I 07 N p N 0 p 4 5
76. d from MCNP computations 8 1 6 Neutron site Ris This section contains former Ris instruments which constitue the default test series for the McStas distribution 8 1 7 Neutron site PSI The Focus time of flight instrument is with the ILL IN6 and IN12 models the most advanced example It as been built from drawings including the guide and all optics elements Simulated capture fluxes were found in good agreement with measurements 8 1 8 Neutron site Tutorial A typical diffractometer including a powder sample and a Rescal type triple axis models are in this category Instrument parameters enable models to cope with most existing instruments of these classes The TAS model includes a basic in plane UB matrix trans formation It may be used to estimate the resolution functions of a TAS configuration 8 1 9 Neutron site ESS This section contains design studies for the future long pulsed European Spallation Source ESS The only contents is currently the ESS_IN5_reprate instrument for simulating an IN5 TYPE cold chopper multi frame spectrometer for the long pulsed ESS Also serves as example instrument for Tunneling sample comp 8 2 A test instrument for the component V_sample This is one of many test instruments written with the purpose of testing the individual components We have picked this instrument both to present an example test instrument and because it despite its simplicity has produced quite non trivial results
77. degrees DECLARE Ae double coll_div 60 hh Here comes the TRACE section where the actual instrument is defined TRACE The Arm class component defines reference points in 3D space Every component instance must have a unique name Here arm is used This Arm component is set to define the origin of our global coordinate system AT 0 0 0 ABSOLUTE COMPONENT arm Arm AT 0 0 0 ABSOLUTE Next we need some neutrons Let s place a neutron source Refer to documentation of Source_simple to understand the different input parameters The source component is placed RELATIVE to the arm component meaning that modifying the position or orientation of the arm will also affect the source component and other components after that one COMPONENT source Source_simple radius 0 015 dist 1 focus_xw 0 024 focus_yh 0 015 E0 5 dE 0 2 DTU AT 0 0 0 RELATIVE arm Here we have a soller placed to improve beam divergence The component is placed at a distance RELATIVE to a previous component COMPONENT collimator Collimator_linear length 0 2 divergence coll_div xwidth 0 04 yheight 0 06 AT 0 O 0 4 RELATIVE arm We also need something to shoot at here a sample made from vanadium an isotropic scatterer Options are available to restrict the solid angle in which ne
78. e signal and statistics fields of the structure to obtain useful information 5 5 3 HTML VRML and XML formats Both HTML and XML formats are available The former may be viewed using any web browser Netscape Mozilla Firefox Internet Explorer Nautilus Konqueror showing data sets as VRML objects XML data sets may be browsed for instance using Internet Explorer Windows and Mac OS or Firefox GXMLViewer and KXMLEditor under Linux The XML format is NeXus like but not fully compatible However McStas may generate genuine NeXus HDF format see bellow 5 5 4 NeXus format The NeXus format 47 is a platform independent HDF binary data file To have McStas use it 1 the HDF and NeXus libraries must have been installed DTU 63 2 the McStas installation should be done with NeXus bindings e g on Unix Linux systems configure with nexus make make install 3 the compilation of instruments must be done with the DHUSE_NEXUS 1NeXus flag see Section 6 3 4 This is automated with the mcrun tool Section 5 4 2 All results are saved in a single file containing groups of data To view such files install and use HDF View or alternatively HDFExplorer This Java viewer can show content of all detectors including metadata attributes Basic detector images may also be generated 5 6 Using computer Grids and Clusters Parallelizing a computation is in general possible when dependencies between each com puta
79. e 8 5 The result of a 20 scan on an Al2O3 powder sample in two axis mode is shown in Figure 8 6 Both for the scan in focusing mode and for the one in defocusing mode not shown the agreement between simulation and experiment is excellent As a final result we present a scan of the energy transfer Ea hw on a V sample The data are shown in Figure 8 7 8 4 The time of flight spectrometer PRISMA In order to test the time of flight aspect of McStas we have in collaboration with Mark Hagen now at SNS written a simple simulation of a time of flight instrument loosely based on the ISIS spectrometer PRISMA The simulation was used to investigate the effect of using a RIT A style analyser instead of the normal PRISMA backend We have used the simple time of flight source Tof_source The neutrons pass through a beam channel and scatter off from a vanadium sample pass through a collimator on to the analyser The RITA style analyser consists of seven analyser crystals that can be DTU 111 12 2T scan on 1 mm slit 10 intensity O 0 0 5 2T deg Figure 8 4 TAS1 Scans of 20 in the direct beam with 1 mm slit on the sample position measurements o 30 open open 700 600 500 J Q fo T intensity Q T i Q D T 100 3 Figure 8 5 TAS1 Corrected 26 scan on a V sample Collimations simulations x measur
80. e array long n double y Table Table_Read_Array file dat amp n initialize and read the all numerical block n Table_Info_Array Table display informations for all blocks also returns n y Table_Index Table 0 2 5 read in 1st block the 3rd row 6th column element ONLY use Table i with i lt n Table_Free_Array Table free allocated memory for Table You may look into for instance the source files for Monochromator_curved or Virtual_input for other implementation examples B 3 Monitor nD Library This library gathers a few functions used by a set of monitors e g Monitor_nD Res_monitor Virtual_output etc It may monitor any kind of data create the data files and may dis play many geometries for mcdisplay Refer to these components for implementation examples and ask the authors for more details DTU 125 B 4 Adaptive importance sampling Library This library is currently only used by the components Source_adapt and Adapt_check It performs adaptive importance sampling of neutrons for simulation efficiency optimiza tion Refer to these components for implementation examples and ask the authors for more details B 5 Vitess import export Library This library is used by the components Vitess_input and Vitess_output as well as the mcstas2vitess utility Refer to these components for implementation examples and ask the authors for more details B 6 Constants for unit conversion etc
81. e file code c A single das may be used for both input and output filename to represent standard input and standard output respectively 5 2 1 Code generation options By default the output files from the McStas compiler are in ISO C with some extensions currently the only extension is the creation of new directories which is not possible in pure ISO C The use of extensions may be disabled with the p or portable option With this option the output is strictly ISO C compliant at the cost of some slight reduction in capabilities The t or trace option puts special trace code in the output This code makes it possible to get a complete trace of the path of every neutron ray through the instrument as well as the position and orientation of every component This option is mainly used with the mcdisplay front end as described in section 5 4 3 The code generation options can also be controlled by using preprocessor macros in the C compiler without the need to re run the McStas compiler If the preprocessor macro MC_PORTABLE is defined the same result is obtained as with the portable option of the McStas compiler The effect of the trace option may be obtained by defining the MC_TRACE_ENABLED macro Most Unix like C compilers allow preprocessor macros to be defined using the D option e g cc DMC_TRACE_ENABLED DMC_PORTABLE Finally the verbose option will list the components and libraries beeing included in the instr
82. ected This does not affect the simulation see however Beamstop and it is mainly used by the MCDISPLAY section and the GROUP modifier See also the SCATTERED variable below B 1 2 Coordinate and component variable retrieval DTU MC_GETPAR comp outpar This may be used in e g the FINALLY section of an instrument definition to reference the output parameters of a component NAME_CURRENT_COMP gives the name of the current component as a string POS_A_CURRENT_COMIP gives the absolute position of the current compo nent A component of the vector is referred to as POS_A CURRENT_COMP i where iis y or Zz ROT_A CURRENT COMP and ROT _R CURRENT_COMP give the ori entation of the current component as rotation matrices absolute orientation and the orientation relative to the previous component respectively A component of a rotation matrix is referred to as ROT_A CURRENT_COMP m n where m and n are 0 1 or 2 standing for x y and z coordinates respectively POS _A COMP comp gives the absolute position of the component with the name comp Note that comp is not given as a string A component of the vector is referred to as POS A COMP comp i where i is x y or z ROT_A_COMP comp and ROT R COMP comp give the orientation of the component comp as rotation matrices absolute orientation and the orientation rela tive to its previous component respectively Note that comp is not given as a string 119 A component of a rotatio
83. efore starting computa tion The default is to send directly the executatble from the master node which only works on homogeneous systems computation This method shows similar efficiency as MPI but without MPI installation It is especially suited on multi core machines but may also be used on any set of distant machines grids as well as clusters For Windows master machines we recommend the installation of the PuTTY SSH client The overhead is proportional to the number of nodes and the amount of data files to transfer per simulation It is usually larger than the pure MPI method We thus recommend to launch long runs on fewer nodes rather than many short runs on many nodes Requirements and limitation SSH grids 1 A master machine with an SSH client and McStas installation 2 A set of machines homogeneous or heterogeneous with SSH servers 3 On heterogeneous grids a C compiler must also be installed on all slave nodes 4 ssh access from the master node where McStas is installed to the slaves through e g DSA keys without a password These keys should be generated using the command ssh keygen Run e g ssh keygen t dsa on master node enter no passphrase and add resulting ssh id_dsa pub to ssh authorized_keys on all the slave nodes The key generation and registering mechanism may be done automatically for the local machine from the Help menu Install DSA key item of mcgui 5 The machine names listed in the file mcstas h
84. elf the simulation accuracy and the plotter check You should now be able to use McStas For some examples to try see the examples directory Start mcgui mcgui pl on Windows and select one of the examples in the Neutron Sites menu 24 DTU Chapter 4 Monte Carlo Techniques and simulation strategy This chapter explains the simulation strategy and the Monte Carlo techniques used in Mc Stas We first explain the concept of the neutron weight factor and discuss the statistical errors in dealing with sums of neutron weights Secondly we give an expression for how the weight factor transforms under a Monte Carlo choice and specialize this to the con cept of direction focusing Finally we present a way of generating random numbers with arbitrary distributions More details are available in the Appendix concerning random numbers 4 1 Neutron spectrometer simulations 4 1 1 Monte Carlo ray tracing simulations The behaviour of a neutron scattering instrument can in principle be described by a complex integral over all relevant parameters like initial neutron energy and divergence scattering vector and position in the sample etc However in most relevant cases these integrals are not solvable analytically and we hence turn to Monte Carlo methods The neutron ray tracing Monte Carlo method has been used widely for guide studies 17 19 instrument optimisation and design 20 21 Most of the time the conclusions and gen er
85. ements o 112 0 2TA deg ws simulations scaled to the same intensity Collimations open open 30 28 67 DTU 800 700 600 a fo intensity A Q So ow Ca 200 32 32 5 33 33 5 34 34 5 35 35 5 2T deg Figure 8 6 TAS1 26 scans on Al2O3 in two axis focusing mode Collimations open 30 28 67 x measurements o simulations A constant background is added to the simulated data 140 120 100 intensity oO op oO T A oO T 207 Figure 8 7 TAS1 Scans of the analyser energy on a V sample Collimations open 30 28 67 x measurements o simulations DTU 113 rotated independently around a vertical axis After the analysers we have placed a PSD and a time of flight detector To illustrate some of the things that can be done in a simulation as opposed to a real life experiment this example instrument further discriminates between the scattering off each individual analyser crystal when the neutron hits the detector The analyser component is modified so that a global variable neu_color registers which crystal scatters the neutron ray The detector component is then modified to construct seven different time of flight histograms one for each crystal see the source code for the instrument for details One way to think of this is that the analyser blades paint a color on each neutron which is t
86. en implemented for monitor components to output customized meta data The macro DETECTOR_CUSTOM_HEADER t DTU 93 defines a string to be written during the next DETECTOR _OUT call as a field custom This string may additionally use the symbol PRE which is replaced by the format specific comment character e g for McStas PGPLOT for Octave Matlab for Scilab and for IDL and for other formats The argument t to the macro may be a static string e g My own additional information or the name of a character array variable containing the meta data After the detector monitor file being written the custom meta data output is deactivated This way each monitor file may have its own meta data definition by repeating the DETECTOR_CUSTOM_HEADER call You may either do that inside the component SAVE section or within an instrument description in an EXTEND code preceeding the monitor e g following an Arm component 6 5 7 The FINALLY section FINALLY ht C code to execute at end of simulation ht This gives code that will be executed when the simulation has ended This might be used to free memory and print out final results from components e g the simulated intensity in a detector This section also triggers the SAVE section to be executed 6 5 8 The MCDISPLAY section MCDISPLAY At C code to draw a sketch of the component ht This gives C code that draws a sketch of t
87. ence when condition is true The reference may be an instance name as well as PREVIOUS PREVIOUS n MY SELF NEXT and NEXT n where n is the index gap to the target either backward PREVIOUS or forward NEXT so that PREVIOUS 1 is PREVIOUS and NEXT 1 is NEXT MYSELF means that the component will be iterated as long as the condition is true This may be a way to handle multiple scattering if the component has been designed for that The jump arrives directly inside the target component in the local coordinate system i e without applying the AT and ROTATED keywords In order to control better the target positions it is required that except for looping MYSELF the target component type should be an Arm There is a more general way to iterate components which consists in repeating the loop for a given number of times JUMP reference ITERATE number_of_times This method is specially suited for very long curved guides of similar components but in order to take into account rotation and translation between guide sections the iterations are performed between Arm s In the following example for a curved guide made on n 500 elements of length L on a curvature radius R with gaps d between elements we simply write COMPONENT CG_In Arm AT COMPONENT CG_1 Guide_gravity 1 L n m 1 AT 0 0 0 RELATIVE PREVIOUS COMPONENT CG_2_Position Arm AT 0 0 L nt td RELATIVE PREVIOUS ROTATED 0 L nt d R 180 PI 0 RELA
88. eneral comment about optimization is that it should be used cautiously checking that the results are not significantly affected 5 3 5 Optimizing instrument parameters Often the user may wish to optimize the parameters of a simulation i e the best geometry of a given component for example the optimal curvature of a monochromator The choice of the optimization routine of the simulation quality value to optimize the initial parameter guess and the simulation length all have a large influence on the results The user is advised to be cautious when interpreting the optimization results Using iFit for optimization One of the authors of McStas has developed a very flexible and general data analysis and fitting package called iFit based on Matlab Matlab itself is not required as a stand alone distributable binary of iF it exists iFit contains wrapper functionality for compiling and running McStas simulations as object functions and allows to select many different optimizers including swarms and other non gradient methods Please see the iFit documentation for more information Our experience is that iF it together with McStas is a more robust optimization solution than the McStas built in Simplex solution DTU 45 Using the Simplex method The McStas package comes with a Simplex optimization method to find best instrument parameters in order to maximize all or some specified monitor integrated values It uses the Downhill Simplex M
89. ent the param eters used and of course the signal the estimated error on the signal and the number of events used in each bin Additionally all data files indicate their first moment mean value and second moment half width in the statistics field In order for the user to choose the data format we recommend to set it using the format FORMAT or alternatively via the MCSTAS FORMAT environment variable which will also make the front end programs able to import and plot data and instrument con sistently see Section 5 3 The available format list is shown in Table 5 3 Note that the neutron event counts in detectors are typically not very meaningful except as a way to measure the performance of the simulation Use the simulated intensity instead whenever analysing simulation data 5 5 1 McStas and PGPLOT format The McStas original format which is equivalent to the PGPLOT format is simply columns of ASCII text that most programs should be able to read One dimensional histogram monitors time of flight energy write one line for each histogram bin Each line contains a number identifying the bin i e the time of flight followed by three numbers the simulated intensity an estimate of the statistical error as explained in section 4 2 1 and the number of neutron events for this bin Two dimensional histogram monitors position sensitive detectors output M lines of N numbers representing neutron intensities where M and N are t
90. eparate nodes In this case you should use MPICH or the SSH grid There are 3 methods for using MPI e Basic usage requires to compile and run the simulation by hand mpicc mpirun This should be used when running LAM MPI e A much simpler way is to use mcrun c mpi NB_CPU which will recompile and run MPICH e The McGUI interface supports MPICH from within the Run Dialog The MPI support is especially suited on clusters As an alternative the SSH grid presented above section 5 6 1 is very flexible and requires lighter configuration Requirements and limitation MPI To use MPI you will need 1 A master machine with an SSH client server and McStas installation 2 A set of unix machines of the same architecture binary compatible with SSH servers 3 ssh access from the master node where McStas is installed to the slaves through e g DSA keys without a password These keys should be generated using the command ssh keygen Run e g ssh keygen t dsa on master node enter no passphrase and add resulting ssh id_dsa pub to ssh authorized_keys on all the slave nodes The key generation and registering mechanism may be done automatically for the local machine from the Help menu Install DSA key item of mcgui 66 DTU 4 The machine names listed in the file mcstas hosts in your home directory or in the MCSTAS tools perl1 mcstas hosts on the master node one node per line The machines lt file gt option enables to specify t
91. er 3 2 5 3 Various DTU A number of minor bugs ironed out both in components runtime code and tools From release 1 12 McStas is GPL 2 only The debate on the internet about the future GPL 3 license suggests that this license might have implications on the de rived work hence have implications on what and how our users use their McStas simulations for To protect user freedom we will stick with GPL 2 17 2 5 4 Warnings WARNING The dash shell which is used as bin sh on some Linux system Including Ubuntu 7 04 makes the Cancel and Update buttons fail in mcgui Solutions are a b 2 6 If your system is a Debian or Ubuntu please dpkg reconfigure dash and say no to install dash as bin sh If you run another Linux with bin sh beeing dash please install bash and manually change the bin sh link to point at bash Future extensions The following features are planned for the oncoming releases of McStas not an ordered list 18 Increased validation and testing Extend test cases to all most components One instrument pr component Prob ably not in examples Updates to mcresplot to support the Matlab and Scilab backends Global changes of components relating to polarisation visualisation Visualisation of neutron spins in magnetic fields for all graphical backends Array AT specifiers for components i e COMPONENT MyComp Comp AT Xarray Yarray Zarray and A
92. er to achieve renewal of its instruments and guides A significant contribution to both the component library and the McStas kernel itself was early performed at the ILL and included in the package ILL later became a part of the core McStas team Similarly the PSI has applied McStas extensively for instrument design and upgrades provided important component additions and contributed several systematic comparative studies of the European instrument Monte Carlo codes Hence PSI has also become a part of the core McStas team Since year 2001 Ris was no longer a neutron source and the authors from that site have moved on to positions at University of Copenhagen NBI and Technical University of Denmark DTU Physics hence these two partners have joined the core McStas team It is further envisioned that the future ESS Data Management and Software Centre ESS DMSC will contribute to the project in the future 1 2 Scientific background What makes scientists happy Probably collect good quality data pushing the instru ments to their limits and fit that data to physical models Among available measurement techniques neutron scattering provides a large variety of spectrometers to probe structure and dynamics of all kinds of materials Neutron scattering instruments are built as a series of neutron optics elements Each of these elements modifies the beam characteristics e g divergence wavelength spread spa tial and time distributions in a way
93. ere d1 and d2 will be floating point parameters of C type double i will be an integer parameter of C type int and s1 and s2 will be string parameters of C type char The parameters of an instrument may be given default values Parameters with default values are called optional parameters and need not be given an explicit value when the instrument simulation is executed When executed without any parameter value in the command line see section 5 3 the instrument asks for all parameter values but pressing the Return key selects the default value if any When used with at least one parameter value in the command line all non specified parameters will have their value set to the default one if any A parameter is given a default value using the syntax param value For example DEFINE INSTRUMENT test di 1 string s2 hello 74 DTU Here d1 and d2 are optional parameters and if no value are given explicitly 1 and hello will be used Optional parameters can greatly increase the convenience for users of instruments for which some parameters are seldom changed or of unclear signification to the user Also if all instrument parameters have default values then the simple command mcdisplay test instr will show the instrument view without requesting any other input which is usually a good starting point to study the instrument design 6 3 2 The DECLARE section DECLARE ht C declarations of global variables etc
94. et the McStas compiler generate as much code as possible automatically letting the compiler handle some of the things that would otherwise be the task of the user programmer Correctness is improved by having a well tested compiler generate code that would otherwise need to be specially written and debugged by the user for every instrument or component Efficiency is also improved by letting the compiler optimize the generated code in ways that would be time consuming or difficult for humans to do Furthermore the compiler can generate several different simulations from the same specification for example to optimize the simulations in different ways to generate a simulation that graphically displays neutron trajectories and possibly other things in the future that were not even considered when the original instrument specification was written The design of McStas makes it well suited for doing what if types of simulations Once an instrument has been defined questions such as what if a slit was inserted what if a focusing monochromator was used instead of a flat one what if the sample was offset 2 mm from the center of the axis and so on are easy to answer Within minutes the instrument definition can be modified and a new simulation program generated It also makes it simple to debug new components A test instrument definition may be written containing a neutron source the component to be tested and whatever monitors
95. ethod in Multidimensions 45 46 which is a geometric optimiza tion method somewhat similar to genetic algorithms It is not as fast as the gradient method but is much more robust It is well suited for problems with up to about 10 20 parameters to optimize Higher dimensionalities are not guarantied to converge to a meaningful solution When using mcrun section 5 4 2 the optimization mode is set by using the optim option or a list of monitors to maximize with as many optim COMP as required The optimization accuracy criterion may be changed with the optim prec accuracy option From mcgui section 5 4 1 one should choose the Optimization execution mode instead of the Simulation or Trace mode Then specify the instrument parameters to optimize by indicating their variation range param min max e g Lambda 1 4 just like parameter scans Optionally the starting guess value might be given with the syntax param min guess max The optimization accuracy criterion is controlled using the Pre cision entry box in the configuration options See Figure 5 7 Finally run the simulation The optimum set of parameters is then printed at the end of the simulation process You may ask to maximize only given monitors instead of all by selecting their component names in the lower lists in the Run Dialog up to 3 If you would like to maximize the flux at a given monitor with some divergence constrains you should for instance simply add
96. example of include see the monitors Monitor_nD component See also section 6 4 for insertion of full instruments in instruments instrument concatenation If the instrument description compilation fails check that the keywords syntax is cor rect that no semi colon sign is missing e g in C blocks and after an ABSORB macro and there are no name conflicts between instrument and component instances variables 72 DTU Keyword Scope Meaning ABSOLUTE I Indicates that the AT and ROTATED keywords are in the ab solute coordinate system AT I Indicates the position of a component in an instrument defini tion COPY I C copy duplicate an instance or a component definition DECLARE I C Declares C internal variables DEFINE I C Starts an INSTRUMENT or COMPONENT definition DEFINITION C Defines component parameters that are constants define END IL C Ends the instrument or component definition SPLIT I Enhance incoming statistics by event repetition EXTEND I Extends a component TRACE section plug in FINALLY I C Embeds C code to execute when simulation ends GROUP I Defines an exclusive group of components include I C Imports an instrument part a component or a piece of C code when within embedded C JUMP I Iterative loops and conditional jumps INITIALIZE L C Embeds C code to be executed when starting ITERATE I Defines iteration counter for JUMP MCDISPLAY C Embeds C code to display compo
97. eyword is a condition for GROUP This means that when the WHEN is false the component instance is not active in the GROUP it belongs to A usage example of the WHEN keyword can be found in the Neutron site ILL ILL_TOF_Env instrument from the mcgui to monitor neutrons de pending on their fate WARNING Combining WHEN EXTEND and GROUP can result in unexpected behaviour please use with caution Let for instance a GROUP of components all have the same WHEN condition i e if the WHEN condition is false none of the elements SCATTER meaning that all neutrons will be ABSORBed As a solution to this problem we propose to include an EXTENDed Arm component in the GROUP but with the opposite WHEN condition and a SCATTER keyword in the EXTEND section This means that when none of the other GROUP elements are present the Arm will be present and SCATTER 84 DTU 6 4 5 Component loops and non sequential propagation JUMP There are situations for which one would like to repeat a given component many times or under a given condition The JUMP modifier is meant for that and should be placed after the positioning GROUP and EXTEND This breaks the sequential propagation along components in the instrument description There may be more than one JUMP per component instance The jump may depend on a condition COMPONENT name comp p1 1 p2 2 AT JUMP reference WHEN condition in which case the instrument TRACE will jump to the refer
98. f height h and radius r centered at the origin sphere_intersect amp t amp t2 Y Z Ur Vy Vz T Similar to box_intersect but using a sphere of radius r B 1 7 Random numbers 122 rand01 Returns a random number distributed uniformly between 0 and 1 randnorm Returns a random number from a normal distribution centered around 0 and with o 1 The algorithm used to sample the normal distribution is explained in Ref 49 ch 7 randpm1 Returns a random number distributed uniformly between 1 and 1 randtriangle Returns a random number from a triangular distribution between 1 and 1 randvec_target_circle amp vz amp vy Kuz amp dQ aimz aimy aim rf Generates a random vector vz Vy vz of the same length as aim aim aim which is targeted at a disk centered at aim aim aim with radius r in meters and perpendicular to the aim vector All directions that intersect the circle are chosen with equal probability The solid angle of the circle as seen from the position of the neutron is returned in dQ This routine was previously called randvec_target_sphere which still works DTU e randvec_target_rect_angular amp v amp v amp vz amp dQ aim z aim aimz h w Rot does the same as randvec_target_circle but targetting at a rectangle with angu lar dimensions h and w in radians not in degrees as other angles The rota tion matrix Rot is the coordinate system orientation in
99. f the Simulate option A window will pop up showing a sketch of the instrument Depending on your chosen plotting backend the presented graphics will resemble one of those shown in figures 5 4 5 6 ator i D qd DI o ogg 3 Oe gt a PGPLOT Window T ls CY J 5 LU O CH 3 em TOF diagram letDisk out T T z s m 0 l wu sixy x Figure 5 4 Left Output from mcdisplay with PGPLOT backend The left mouse button starts a new neutron ray the middle button zooms and the right button resets the zoom The Q key quits the program Right The new PGPLOT time of flight option See section 5 4 3 for details fvanadium _example PlotinstrumentaD First component Last component target neutrons to trace 1 et Figure 5 5 Output from mcdisplay with Scilab backend Display can be adjusted using the dialog box right For a slightly longer gentle introduction to McStas see the McStas tutorial available from 2 and as of version 2 0 built into the mcgui help menu For more technical details 34 DTU home fys pkwi Beta0 mcstas 1 7 Beta0 examples vanadium_example xim Reset vie Per 2 Tece e Exit_ P 3 source PSD_4pi Redraw A 4 1 iak Figure 5 6 Output from mcdisplay with Matlab backend Display can be adjusted using the window buttons read on from section 5 2 5 1 1 New releases of McStas Release
100. front end is used to plot the resolution function particularly for triple axis spectrometers as calculated by the Res sample component or TOF_res_sample for time of flight instruments It requires to have a Res monitor component further in the instrument description at the detector position This front end has been included in the release since it may be useful despite its somewhat rough user interface 58 DTU The mcresplot front end is launched with the command mcresplot file Here file is the name of a file output from a simulation using the Res_monitor component This front end currently only works with the PGPLOT plotter but ports for Matlab and Scilab may be written in the future The front end will open a window displaying projections of the 4 dimensional resolution function R Q w measured at a particular choice of Q and w see the component manual The covariance matrix of the resolution function the resolution along each projection axis and the resulting resolution matrix are also shown as well as the instrument name and parameters used for the simulation To use the mcresplot front end the programs Perl PGPLOT PgPerl and PDL must all be properly installed on the system 5 4 6 Creating and viewing the library component instrument help and Manuals mcdoc McStas provides an easy way to generate automatically an HTML help page about a given component or instrument or the whole McStas library mcdoc mcdoc comp instr
101. g up calculations is to introduce a neutron weight factor for each simulated neutron ray and to adjust this weight according to the path of the ray If e g the reflectivity of a certain optical component is 10 and only reflected neutrons ray are considered later in the simulations the neutron weight will be multiplied by 0 10 when passing this component but every neutron is allowed to reflect in the component In contrast the totally realistic simulation of the component would require in average ten incoming neutrons for each reflected one Let the initial neutron weight be po and let us denote the weight multiplication factor in the j th component by rj The resulting weight factor for the neutron ray after passage of the n components in the instrument becomes the product of all contributions n P Pn bo e 4 2 j l Each adjustement factor should be 0 lt m lt 1 except in special circumstances so that total flux can only decrease through the simulation see section For convenience the value of p is updated within each component during the simulation Simulation by weight adjustment is performed whenever possible This includes e Transmission through filters and windows e Transmission through Soller blade collimators and velocity selectors in the approx imation which does not take each blade into account e Reflection from monochromator and analyser crystals with finite reflectivity and mosaicity e Refle
102. gt Scilab script conversion tool mcformat Conversion tool for text files and MPI grids mcformatgui GUI for mcformat mcdaemon Instrument results on line plotting When used with the h flag all tools display a specific help SEE ALSO mcstas mcdoc mcplot mcrun mcgui mcresplot mcstas2vitess DOC Please visit http www mcstas org DTU 47 5 4 1 The graphical user interface mcgui The front end mcgui provides a graphical user interface that interfaces the various parts of the McStas package It may be started with the single command mcgui The mcgui mcgui pl on Windows program may optionally be given the name of the instrument file to use Dependencies To run the mcgui front end the programs Perl and Perl Tk must be properly installed on the system Additionally to use the McStas PGPLOT back end the software packages PGPLOT PgPerl and PDL are required It may be necessary to set the PGPLOT_DIR and PGPLOT_DEV environment variable consult the documentation for PGPLOT on the local system in case of difficulty The menus When the front end is started the main window is opened see figure 5 2 This window displays the output from compiling and running simulations and contains a few menus and buttons for easy navigation The main purpose of the front end is to edit and compile instrument definitions run the simulations and visualize the results The File menu has the following features File Open instrument selects the na
103. he component in the plots produced by the mcdisplay front end see section 5 4 3 The section can contain arbitrary C code and may refer to the parameters of the component but usually it will consist of a short sequence of the special commands described below that are available only in the MCDISPLAY section When drawing components all distances and positions are in meters and specified in the local coordinate system of the component The MCDISPLAY section is optional If it is omitted mcdisplay will use a default symbol a small circle for drawing the component The magnify command This command if present must be the first in the section It OO 66 99 takes a single argument a string containing zero or more of the letters x y and z It causes the drawing to be enlarged along the specified axis in case mcdisplay is called with the zoom option For example magnify xy The line command The line command takes the following form line x1 Yli 21 T2 Y2 z2 It draws a line between the points x1 y1 21 and 2 yo 22 94 DTU The dashed_line command The dashed_line command takes the following form dashed_line z7 Y1 21 2 Y2 22 N It draws a dashed line between the points x1 y1 21 and 22 y2 z2 with n equidistant spaces The multiline command The multiline command takes the following form multiline n 71 Yi 215 Ens Yn gt Zn It draws a series of lines through the n p
104. he following steps elements DTU The structure of McStas is illustrated in Figure 5 1 To use McStas an instrument definition file describing the instrument to be simu lated must be written Alternatively an example instrument file can be obtained from the examples directory in the distribution or from another source The input files instrument and component files are written in the McStas meta language and are edited either by using your favourite editor or by using the built in editor of the graphical user interface mcgui Next the instrument and component files are compiled using the McStas compiler relying on built in features from the FLEX and Bison facilities to produce a C program The resulting C program can then be compiled with a C compiler and run in combi nation with various front end programs for example to present the intensity at the detector as a motor position is varied The output data may be analyzed and visualized in the same way as regular ex periments by using the data handling and visualisation tools in McStas based on Perl Python in combination with chaco matplotlib Matlab GNUPlot or PG PLOT Further data output formats including IDL NeXus and XML are available see section 5 5 31 Graphical User Interface Perl Visualisation PGPLOT pgperl PDL Input Meta language Output data multi format File Visualisation Matlab Compilation mcstas
105. he generated C file or if the C file is newer than the executable The executable is assumed to have a out suffix in the filename NB If components are changed automatic compilation is not performed Instead use the File Compile menu item in mcgui The editor window The editor window provides a simple editor for creating and modifying instrument defini tions Apart from the usual editor functions the Insert menu provides some functions that aid in the construction of the instrument definitions Editor Insert Instrument template inserts the text for a simple instrument skeleton in the editor window Editor Insert Component opens up a dialog window with a list of all the compo nents available for use in McStas Selecting a component will display a description Double clicking will open up a dialog window allowing the entry of the values of all the parameters for the component figure 5 9 See section 6 3 for details of the meaning of the different fields The dialog will also pick up those of the users own components that are present in the current directory when mcgui is started See section 6 7 for how to write components to integrate well with this facility Editor Insert Type These menu entries give quick access to the entry dialog for the various component types available i e Sources Optics Samples Monitors Misc Contrib and Obsolete 5 4 2 Running simulations on the commandline mcrun The mcrun front end m
106. he hosts file to use If it does not exist only the current machine will be used for multi processor core machines 5 Without ssh keys passwords will be prompted many times To avoid this we recom mend to use only the local machine for multi cores cpu i e do not use a machine hosts file 6 Signals are not supported while simulating with MPI since asynchronous events cannot be easily transmitted to all nodes This means it is not possible to cancel an on going computation However simulation scans can be interrupted as soon as the on going computation step ends 7 MPI must be correctly configured if using ssh you have to set ssh keys to avoid use of passwords if using rsh you have to set a rhosts file On non local accounts this procedure may fail and ssh always require passwords MPI Basic usage To enable parallel computation compile McStas output C source file with mpicc with the flag DUSE_MPI and run it using the wrapper of your MPI implementation mpirun for mpich or lammpi generate a C source file sim c mcstas sim instr generate an executable with MPI support sim mpi mpicc DUSE_MPI o sim mpi sim c execute with parallel processing over lt N gt computers here you have to list the computers you want to use in a file machines list using mpich implementation refer to MPI documentation for a complete description mpirun machinefile machines list n lt N gt sim mpi lt instrument
107. he number of bins in the DTU 61 two dimensions The two dimensional monitors also store the error estimates and event counts as additional matrices Single point monitors output the neutron intensity the estimated error and the neu tron event count as numbers on the terminal The results from a series of simulations may be combined in a data file using the mcrun front end as explained in section 5 4 2 When using one and two dimensional monitors the integrated intensities are written to terminal as for the single point monitor type supplementing file output of the full one or two dimensional intensity distribution Both one and two dimensional monitor output by default start with a header of comment lines all beginning with the character This header gives such information as the name of the instrument used in the simulation the values of any instrument parameters the name of the monitor component for this data file etc The headers may be disabled using the data only option in case the file must be read by a program that cannot handle the headers In addition to the files written for each one and two dimensional monitor component another file by default named mcstas sim is also created This file is in a special McStas ASCII format It contains all available information about the instrument definition used for the simulation the parameters and options used to run the simulation and the monitor components present in the
108. hen observed in the detector An illustration of the instrument is shown in figure 8 8 Test results are shown in Appendix 8 4 1 Moderator Sample Collimator Monitor Slit Slit Slit PSD Detector 7 blade analyser Figure 8 8 A sketch of the PRISMA instrument 8 4 1 Simple spectra from the PRISMA instrument A plot from the detector in the PRISMA simulation is shown in Figure 8 9 These results were obtained with each analyser blade rotated one degree relative to the previous one The separation of the spectra of the different analyser blades is caused by different energy of scattered neutrons and different flight path length from source to detector We have not performed any quantitative analysis of the data 114 DTU x10 9 PRISMA with RITA backend Intensity D wo T i 3500 6000 6500 7000 7500 time usec Figure 8 9 Test result from PRISMA instrument using colored neutrons Each graph shows the neutrons scattered from one analyser blade DTU 115 Appendix A Random numbers in McStas A 1 Transformation of random numbers In order to perform the Monte Carlo choices one needs to be able to pick a random number from a given distribution However most random number generators only give a uniform distribution over a certain interval We thus need to be able to transform between probability distributions and we here give a shor
109. iles use the I switch for the McStas compiler mcstas I home joe components I home joe neutron include name instr Multiple I options can be given as shown 5 2 3 Embedding the generated simulations in other programs By default McStas will generate a stand alone C program which is what is needed in most cases However for advanced usage such as embedding the generated simulation in another program or even including two or more simulations in the same program a stand alone program is not appropriate For such usage the McStas compiler provides the following options e no main This option makes McStas omit the main function in the generated simulation program The user must then arrange for the function mcstas_main to be called in some way e no runtime Normally the generated simulation program contains all the run time C code necessary for declaring functions variables etc used during the simulation This option makes McStas omit the run time code from the generated simulation program and the user must then explicitly link with the file mcstas r c as well as other shared libraries from the McStas distribution Users that need these options are encouraged to contact the authors for further help 5 2 4 Running the C compiler After the source code for the simulation program has been generated with the McStas compiler it must be compiled with the C compiler to produce an executable The gener ated C code obeys the IS
110. inary XML HTML VRML NeXus Tripoli MCNP Vitess McStas events sim SC pro xml html wrl NXS Original format for PGPLOT plotter may be used with f and d options Scilab format may be used with f and d options Scilab format with external binary files may be used with d option a option implicitly set Matlab format may be used with f and d options Matlab format with external binary files may be used with d option a option implicitly set Octave format may be used with f and d options Octave format with external binary files may be used with d option a option implicitly set IDL format Must be used with f option IDL format with external binary files may be used with d option a option implicitly set XML format NeXus like may be used with f and d options HTML format generates a web page may be used with f and d options Data files are saved as VRML objects OpenGL Virtual Reality file format for data files Simulation files are not saved properly and the HTML format should be used prefer ably NeXus data files HDF All simulation results are stored in a unique compressed binary file This format requires to have NeXus installed Tripoli 4 neutron events file format Use Virtual_tripoli4_input Virtual_tripoli4_output com ponents MCNP PTRAC neutron events file format Use Virtual_mcnp_input Virtual_mcnp_output components Vite
111. ings and then run the simulation Default instrument parameter values see section 6 3 if any will be indicated and entered when hitting the Return key This way of running McStas will only give data for one instrument setting which is normally sufficient for e g time of flight SANS or powder instruments but not for e g continuous beam reflectometers or triple axis spectrometers where a scan over various instrument settings is required Often the simulation will be run using one of several available front ends as described in the next section These front ends help manage output from the potentially many detectors in the instruments as well as running the simulation for each data point in a scan The generated simulations accept a number of options and arguments The full list can be obtained using the help option name out help The values of instrument parameters may be specified as arguments using the syntax name val For example Samples_vanadium out ROT 90 The number of neutron histories to simulate may be set using the ncount or n option for example ncount 2e5 The initial seed for the random number generator is by default chosen based on the current time so that it is different for each run However for debugging purposes it is sometimes convenient to use the same seed for several runs so that the same sequence of random numbers is used each time To achieve this the random seed may be set using the seed or s op
112. instrument It is read by the mcplot front end see section 5 4 4 This file stores the results from single monitors but by default contains only pointers in the form of file names to data for one and two dimensional monitors By storing data in separate files reading the data with programs that do not know the special McStas file format is simplified The file option may be used to store all data inside the mcstas sim file instead of in separate files 5 5 2 Matlab Scilab and IDL formats With these formats McStas automatically writes scripts containing the data as a structure as well as in line import and plot functions for the selected language Usage examples are given in section 5 3 Thus it is not necessary to write a load routine for each format as the script is itself a program that knows how to handle the data Alternatively using mcplot with Matlab and Scilab plotters provide additional functionalities from menus and dialogs see section 5 4 4 When imported through the data generated script see section 5 3 or using mcplot see section 5 4 4 a single variable may be created into the Matlab Scilab or IDL base workspace This variable is a structure constituting a data tree containing many fields some being themselves structures Field names are the initial names from the instrument components files transformed into valid variable names e g containing only letters digits and the character except for the firs
113. invent the code for simulating the individual components in the instrument A third user may write an intricate simulation of a complex component e g a detailed description of a rotating velocity selector and expect other users to easily benefit from his her work and so on McStas attempts to make it possible to work at any combination of layers in isolation by separating the layers as much as possible in the design of the system and in the meta language in which simulations are written The usage of a special meta language and an automatic compiler has several advan tages over writing a big monolithic program or a set of library functions in C Fortran or another general purpose programming language The meta language is more powerful specifications are much simpler to write and easier to read when the syntax of the speci fication language reflects the problem domain For example the geometry of instruments would be much more complex if it were specified in C code with static arrays and pointers The compiler can also take care of the low level details of interfacing the various parts of the specification with the underlying C implementation language and each other This way users do not need to know about McStas internals to write new component or instru ment definitions and even if those internals change in later versions of McStas existing definitions can be used without modification The McStas system also utilizes the meta language to l
114. ions in mcgui and mcrun e CFLAGS may be set to any options needed by the compiler eg for optimization or ANSI C conformance Also used by mcgui mcrun e PERL may be set to the path of the Perl interpreter to use To use these options set the variables before running configure Eg setenv PERL pub bin perl5 configure It may be necessary to remove configure s cache of old choices first rm f config cache If you experience any problems or have some questions or ideas concerning McStas please contact peter willendup risoe dk or the McStas mailing list at mcstas users mcstas org 3 5 Finishing and Testing the McStas distribution Once installed you may check and tune the guessed configuration stored within file e MCSTAS tools perl mccode_config perl on Windows systems e MCSTAS tools perl mccode_config perl on Unix Linux systems where MCSTAS is the location for the McStas library You may on Linux systems ask for a reconfiguration e g after installing MPI Matlab with the commands e g cd MCSTAS tools perl1 sudo mcstas_reconfigure On Windows systems the reconfiguration is performed with the mcconfig pl command The examples directory of the distribution contains a set of instrument examples These are used for the McStas self test procedure which is executed with mcrun test mcrun pl on Windows DTU 23 This test takes a few minutes to complete and ends with a short report on the installation its
115. is distributed over p computers In practice there will be overhead time due to the split and merge operations e the split is immediate constant time cost O 1 e the merge is at worst linear against the number of computers linear time cost O p when saving an event list logarithmic time cost O log p when not saving an event list The efficiency of McStas using MPI has been tested on large clusters up to 500 nodes The computation time decreases in the same proportion as the number of nodes showing an ideal efficiency However a small overhead may appear depending on the cluster internal network load which may be estimated at most of about 10 20 s This overhead comes from the spread and the fusion of the computations For instance spreading a computation implies often an rsh or and ssh session to be opened on every node To reach the best efficiency the computation time should not be lower than 30 seconds or the overhead time may become significant compared to total time 68 DTU McStas computation time using MPI r r Execution time s wee min iy Total time e Computing time templateTOF simulation with 10 initial events o 8 lt _ o 10 10 Number of CPUs Figure 5 11 McStas MPI execution time as a function of computing nodes with tem plateTOF instrument and 1e8 initial neutron events Tests performed on Lonestar TACC US Teragrid 2008 5 6 5 MPI and
116. ity of transmitting the real neutron is P but we make the Monte Carlo choice of transmitting the neutron ray each time fmc 1 This must be reflected on the choice of weight multiplier 7 P Of course one could simulate without weight factor transformation in our notation written as fmc P m 1 To generalize weight factor transformations are given by the master equation fuct P 4 9 This probability rule is general and holds also if e g it is decided to transmit only half of the rays fmc 0 5 An important different example is elastic scattering from a powder sample where the Monte Carlo choices are the particular powder line to scatter from the scattering position within the sample and the final neutron direction within the Debye Scherrer cone This weight transformation is much more complex than described above but still boils down to obeying the master transformation rule 4 3 1 Direction focusing An important application of weight transformation is direction focusing Assume that the sample scatters the neutron rays in many directions In general only neutron rays in some of these directions will stand any chance of being detected These directions we call the interesting directions The idea in focusing is to avoid wasting computation time on neutrons scattered in the other directions This trick is an instance of what in Monte Carlo terminology is known as importance sampling If e g a sample scatters isotropic
117. ks for the closest x value in the first column index 0 and extracts in this row the n th element starting from 0 The first column is thus the x axis for the data e Table_Free amp T able free allocated memory blocks e Table Value2d Table X Y Uses 2D linear interpolation on a Table from X Y coordinates and returns the corresponding value Available functions to read an array of vectors matrices in a text file are e Table Read_Array File amp n read and split file into as many blocks as necessary and return a t_Table array Each block contains a single vector matrix This only works for text files The number of blocks is put into n e Table_Free_Array amp Table free the Table array e Table_Info_Array amp T able display information about all data blocks The format of text files is free Lines starting by characters are considered to be comments and stored in Table header Data blocks are vectors and matrices Block numbers are counted starting from 1 and changing when a comment is found or the column number changes For instance the file MCSTAS data BeO trm Transmission of a Berylium filter looks like Pree H BeO transmission as measured on IN12 Thickness 0 05 m k Angs 1 Transmission 0 1 wavevector multiply 0500 0 74441 0750 0 76727 1000 0 80680 Binary files should be of type float i e REAL 32 and double i e REAL 64 and should not
118. l spectrometer The TAS 1 simulation was the first detailed work performed with the McStas package For further details see reference 48 At the spectrometer the channel from the cold source to the monochromator is asym metric since the first part of the channel is shared with other instruments In the instru ment definition this is represented by three slits For the cold source we use a flat energy distribution component Source flat focusing on the third slit The real monochromator consist of seven blades vertically focusing on the sample The angle of curvature is constant so that the focusing is perfect at 5 0 meV 20 0 meV for 2nd order reflections for a 1x1 cm sample This is modeled directly in the instrument definition using seven Monochromator components The mosaicity of the pyrolytic graphite crystals is nominally 30 FWHM in both directions However the simulations indicated that the horisontal mosaicities of both monochromator and analyser were more likely 45 This was used for all mosaicities in the final instrument definition The monochromator scattering angle in effect determining the incoming neutron en ergy is for the real spectrometer fixed by four holes in the shielding corresponding to the energies 3 6 5 0 7 2 and 13 7 meV for first order neutrons In the instrument definition we have adapted the angle corresponding to 5 0 meV in order to test the simulations against measurements performed on the spec
119. lab back ends_ A 3D view of the instrument and various operations zoom export print trace neutrons is available from dedicated Graphical User Interfaces The inspect option may be used see previous paragraph as well as the first and last options to specify a region of interest Note The Scilab plotter for Windows does not allow to rotate interactively the 3D view and you are to use the alpha and theta choices in the pop up Dialog The mcdisplay front end will then require the Perl and either Scilab or Matlab to be installed VRML OpenGL back ends When using the pVRML option the instrument is shown in Virtual Reality using OpenGL You may then walk aside instrument or go inside elements following neutron trajectories As all neutron trajectories are stored into a VRML file you better limit the number of stored trajectories below 1000 otherwise file size and processing time becomes significant The inspect option is not available in VRML format display 56 DTU 5 4 4 Plotting the results of a simulation mcplot The front end mcplot mcplot pl on Windows is a program that produces plots of all the monitors in a simulation and it is thus useful to get a quick overview of the simulation results In the simplest case the front end is run simply by typing mcplot This will plot any simulation data stored in the current directory which is where simula tions store their results by default If the dir or file o
120. lable McStas components and instruments may be found at the McStas webpage 2 as well as in the McStas library see 7 1 All available McStas documentation is accessible from the mcgui Help menu Note that McDoc compliant comments in the source code are no substitute for a good reference manual entry The mathematical equations describing the physics and algorithms of the component should still be written up carefully for inclusion in the component manual The McDoc comments are useful for describing the general behaviour of the component the meaning and units of the input parameters etc The format of the comments in the library source code The format of the comments understood by McDoc is mostly straight forward and is designed to be easily readable both by humans and by automatic tools McDoc has been written to be quite tolerant in terms of how the comments may be formatted and broken across lines A good way to get a feeling for the format is to study some of the examples in the existing components and instruments Below a few notes are listed on the requirements for the comment headers The comment syntax uses IDENTIFICATION DESCRIPTION PARAMETERS EXAMPLE ALINKS and END keywords to mark different sections of the documentation Keywords may be abbreviated except for ZEXAMPLE e g as IDENT or I Additionally optional keys VALIDATION and BUGS may be found to list validation status and possible bugs in the compone
121. lained in Chapter 3 e by editing the tools perl mcstas_config perl setup file of your installation e by setting the MCSTAS_FORMAT environment variable Next select Run simulation from the Simulation menu McStas will translate the definition into an executable program and pop up a dialog window Type a value for the ROT parameter e g 90 check the Plot results option and select Start The simulation will run and when it finishes after a while the results will be plotted in a window Depending on your chosen plotting backend the presented graphics will resemble one of those shown in figure 5 3 When using the Scilab or Matlab backends full 3D view 4PI PSD monitor PSD_4pi vanadium psd XO 0 214041 dX 103 684 YO 2 63483 dY 52 337 vanadium_psd vanadium psd 4PI PSD monitor 104 9 0 90 0 00 Lattitude deg 0 90 Longitude deg 10 2 9 gg i0 Longitude deg Lattitude deg 40 60 80 50 o Longitude deg Figure 5 3 Output from mcplot with PGPLOT Scilab and Matlab backends of plots and different display possibilities are available Use the attached McStas window menus to control these Features are quite self explanatory For other options execute mcplot help mcplot pl help on windows to get help DTU 33 Lattitude deg 10 2 To visualize or debug the simulation graphically repeat the steps but check the Trace option instead o
122. language by the user and the McStas compiler automatically translates them into efficient simulation programs written in ISO C In realizing the design of McStas the task was separated into four conceptual layers 1 Graphical user interface and scripting layer presentation of the calculations graph ical or otherwise aka the tool layer 2 Modeling of the overall instrument geometry mainly consisting of the type and position of the individual components 3 Modeling the physical processes of neutron scattering i e the calculation of the fate of a neutron that passes through the individual components of the instrument absorption scattering at a particular angle etc 12 DTU 4 Accurate calculation using Monte Carlo techniques of instrument properties such as resolution function from the result of ray tracing of a large number of neutrons This includes estimating the accuracy of the calculation If you don t want to read this manual in full go directly to the brief introduction in chapter 5 1 Though obviously interrelated these four layers can be treated independently and this is reflected in the overall system architecture of McStas The user will in many situations be interested in knowing the details only in some of the layers For example one user may merely look at some results prepared by others without worrying about the details of the calculation Another user may simulate a new instrument without having to re
123. le example of the component Slit is given PITT TEC ooo SSCP PLPC LCCCCCCCCCCCOO COTS STS SCS S SLL CSOSOSOSO SS SS k kkk McStas neutron ray tracing package Copyright 1997 2002 All rights reserved Risoe National Laboratory Roskilde Denmark Institut Laue Langevin Grenoble France Component Slit 1 Written by Kim Lefmann and Henrik M Roennow Date June 16 1997 Version Revision 1 25 Origin Risoe Release McStas 1 6 Rectangular circular slit with optional insignificance cut D A simple rectangular or circular slit You may either specify the radius circular shape or the rectangular bounds No transmission around the slit is allowed If cutting option is used low weight neutron rays are ABSORBED Example Slit xmin 0 01 xmax 0 01 ymin 0 01 ymax 0 01 Slit radius 0 01 cut 1e 10 UP INPUT PARAMETERS radius Radius of slit in the z 0 plane centered at Origo m xmin Lower x bound m xmax Upper x bound m ymin Lower y bound m ymax Upper y bound m xwidth Width of slit Overrides xmin xmax m yheight Height of slit Overrides ymin ymax m Optional parameters cut Lower limit for allowed weight 1 HH HH HF HH KF F HK F FF HH HH FH KF HF KF HF HH HF HF KF HF HF KF HK KF 0 AE FRR OR RR ORR RA RA Ik A 3 21 24 21 24 3 21 24 3 2k 24 2 21 2 21 24 1 24 2k 2K 22k 2 2 2k kkk kkk kk kk kk kk kk kk DEFINE COMPONENT Slit 96 DTU
124. ler optimizations are notoriously buggy the options given above for PGCC on Linux and the ISO C compiler for HPUX have been known to generate incorrect code in some compiler versions McStas actually puts an effort into making the task of the C com piler easier by in lining code and using variables in an efficient way As a result McStas simulations generally run quite fast often fast enough that further optimizations are not worthwhile Also optimizations are highly time and memory consuming during compila tion and thus may fail when dealing with large instrument descriptions e g more that 100 elements The compilation process is simplified when using components of the library making use of shared libraries see SHARE keyword in chapter 6 Refer to section 5 3 4 for other optimization methods 38 DTU 5 3 Running the simulations Once the simulation program has been generated by the McStas compiler and an exe cutable has been obtained with the C compiler the simulation can be run in various ways Simple McStas options In this section the most common simulation parameters are discussed For a full list please consult tables The simplest way is to run it directly from the command line or shell name out Note the leading which is needed if the current directory is not in the path searched by the shell When used in this way the simulation will prompt for the values of any instrument parameters such as angular sett
125. ll it please perform the following tasks 1 Download the package from http www mcstas org download 2 As root issue the command apt get install perl perl tk gcc libc6 dev pdl bash 3 Optionally as root issue the commands apt get install openssh client openssh server apt get install mpi default bin mpi default dev to benefit from MPI and SSH grid parallelization 4 As root issue the command dpkg reconfigure dash and say no to install dash as bin sh 5 Optionally install FreeWRL from http freewrl sourceforge net download html or WhiteDune with apt get install whitedune 6 To view OFF PLY geometry files you may use Geomview or Meshlab that can be installed with apt get install geomview meshlab 7 If you have acces to Matlab R install it prior to McStas so that the configuration can detect it 8 As root issue the command dpkg i mcstas 2 0 amd64 deb or dpkg i mcstas 2 0 i386 deb on 32 bit systems This later step may be replaced by a configure make make install procedure after extraction of the Mctas tarball see below Updating your operating system to a new release may in some cases require you to re install McStas following the procedure above We hope to make a so called apt repository available in the future which will ensure automatic upgrade of McStas in case of a new release 3 4 2 Other Linux Unix systems To get a fully functional McStas installation on Unix systems a few support ap
126. ly enables the optimization phase of the compiler which can make quite a difference in speed of McStas generated simulations The o name out sets the name of the generated executable The 1m options is needed on many systems to link in the math runtime library like the cos and sin functions Monte Carlo simulations are computationally intensive and it is often desirable to have them run as fast as possible Some success can be obtained by adjusting the compiler optimization options Here are some example platform and compiler combinations that have been found to perform well up to date information will be available on the McStas WWW home page 2 e Intel x86 PC with Linux and GCC using options gcc 03 e Intel x86 with Linux and EGCS GCC derivate using options egcc 06 e Intel x86 with Linux and PGCC pentium optimized GCC derivate using options gcc 06 mstack align double e HPPA machines running HPUX with the optional ISO C compiler using the op tions Aa 0all W1 a archive the Aa option is necessary to enable the ISO C standard e SGI machines running Irix with the options Ofast 032 w Optimization flags will typically result in a speed improvement by a factor about 3 but the compilation of the instrument may be 5 times slower A warning is in place here it is tempting to spend far more time fiddling with com piler options and benchmarking than is actually saved in computation times Even worse compi
127. mats including the original McStas PGPLOT and the Scilab and Matlab formats All formats except the McStas PGPLOT may eventually support binary files which are much smaller and faster to import but are platform dependent The simulation data file extensions are appended automatically depending on the format if the file names do not contain any Binary files are particularly recommended for the IDL format e g format IDL_binary and the Matlab and Scilab format when handling large detectors e g more than 50x50 bins For example Samples_vanadium out ROT 90 format Scilab_binary or more generally for bash Bourne shell users export MCSTAS_FORMAT Matlab Samples_vanadium out ROT 90 It is also possible to create and read Vitess MCNP PTRAC and Tripoli4 batch neu tron event files using components e Vitess_input and Vitess_output e Virtual_tripoli4_input and Virtual_tripoli4_output e Virtual_mcnp_input and Virtual_mcnp_output Additionally adding the raw keyword to the FORMAT will produce raw N p p data sets instead of N p a see Section 4 2 1 The former representation is fully additive and thus enables to add results from separate simulations e g when using a computer Grid which is automated in the mcformat tool Other acceptable format modifiers are transpose to transpose data matrices and append to concatenate data to existing files 5 3 2 Basic import and plot of results The previous example will resul
128. mber of elements in the detector arrays along the X axis e n is the number of elements in the detector arrays along the Y axis e amp N 0 0 is a pointer to the first element in the array of N values for the detector component e amp p 0 0 is a pointer to the first element in the array of p values for the detector component e amp p2 0 0 is a pointer to the first element in the array of p2 values for the detector component e filename is a string giving the name of the file in which to store the data Note that for a two dimensional detector array the first dimension is along the X axis and the second dimension is along the Y axis This means that element iz i can be obtained as pliz n ty if p is a pointer to the first element Three dimensional detectors monitors The results of a three dimensional detector monitor are written to a file using the following macro DETECTOR_OUT_3D t zlabel ylabel zlabel xvar yvar zvar min max Ymin gt Ymax mins max gt s M N J amp N O O 0 amp p 0 0 0 amp p2 0 0 0 filename The meaning of parameters is the same as those used in the 1D and 2D versions of DETECTOR_OUT The available data format currently saves the 3D arrays as 2D with the 3rd dimension specified in the type field of the data header Customizing detectors monitors Users may want to have additional information than the default one written by the DETECTOR_OUT macros A mechanism has be
129. me of an instrument file to be used File Edit current opens a simple editor window with McStas syntax highlighting for editing the current instrument definition This function is also available from the Edit button to the right of the name of the instrument definition in the main window File Spawn editor This starts the editor defined in the environment variable VISUAL or EDITOR on the current instrument file It is also possible to start an external editor manually in any case mcgui will recompile instrument definitions as necessary based on the modification dates of the files on the disk File Compile instrument forces a recompile of the instrument definition regardless of file dates This is for example useful to pick up changes in component definitions which the front end will not notice automatically This might also be required when choosing MPI and NeXus options See section 3 for how to override default C compiler options File Save log file saves the text in the window showing output of compilations and simulations into a file File Clear output erases all text in the window showing output of compilations and simulations File Preferences Opens the choose backend dialog shown in figure 5 7 Several settings can be chosen here 48 DTU e Selection of the desired PGPLOT Scilab Matlab HTML VRML output format and possibility to save binary files when applicable improved disk I O e One or three pane view of your
130. me the file hosts mcgui Save Configuration for saving chosen settings on the Configuration options and Run dialogue Possibilty to run MPI or grid simulations by default from mcgui When scanning parameters mcrun now terminates with a relevant error message if one or more scan steps failed intensities explicitly set to 0 in those cases When running parameter optimisations a logfile default name is mcoptim_XXXX dat where XXXX is a pseudo random string is created during the optimisation updated at each optim step We now provide syntax highlighting setup files for vim and gedit editors Rudimentary support for GNUPLOT when plotting with mcplot Data file format is standard McStas PGPLOT 2 5 2 Platform support Mac OS X 10 3 Panther ppc 10 4 Tiger pcc intel 10 5 Leopard ppc intel Windows XP Windows Vista Now with a recent perl version 5 10 plus various fixes New feature on Windows Simulations always run in the background freeing mcgui for other work Any Linux reference platforms are Ubuntu 8 04 and earlier and Debian 4 0 and earlier We have also tested Fedora 8 OpenSuSE 10 3 and CentOS 4 releases recently FreeBSD FreeBSD release 6 3 and its cousin DesktopBSD 1 6 recently tested SUN Solaris 10 Intel tested Sparc probably OK Plus probably any UNIX POSIX type environment with a bit of effort Details about the installation and the available tools are given in chapt
131. mplex arrangements and syntax 6 4 1 Embedding instruments in instruments TRACE 6 4 2 Groups and component extensions GROUP EXTEND 6 4 3 Duplication of component instances COPY 6 4 4 Conditional components WHEN 6 4 5 Component loops and non sequential propagation JUMP 6 4 6 Enhancing statistics reaching components SPLIT Writing component definitions 20 0200 0008 6 5 1 The component definition header 204 6 5 2 The DECLARE section oaoa a 6 5 3 The SHARE section 6 5 4 The INITIALIZE s 6 5 5 The TRACE section 6 5 6 The SAVE section CCION oae ye ae aD Kei eres BOE Ghd Bees os 6 5 7 The FINALLY section a oa a ee 6 5 8 The MCDISPLAY section aooaa aa ee 6 5 9 The end of the component definition 6 5 10 A component example Slit 2 2 0008 Extending component definitions 0 0 0000 pee ee 6 6 1 Extending from the instrument definition 6 6 2 Component heritage and duplication McDoc the McStas library documentation tool 7 The component library Abstract 7 1 A short overview of the McStas component library 8 Instrument examples 8 1 A quick tour of instrument examples 2 2 0200 DTU 8 1 1 Neutron site Brookhaven 0 0 0 0000800008 8 1 2 Neutron site Tools 0 0 0 0 00 0
132. mponents in an instrument definition without duplicating files using the EXTEND modifier EXTEND At C code executed after the component TRACE section ht The embedded C code is appended to the component TRACE section and all its internal variables as well as all the DECLARE instrument variables except instrument parameters may be used To use instrument parameters you should copy them into global variables in the DECLARE instrument section and refer to these latter This component declaration modifier is of course optional You will find numerous usage examples and in particular in the Sources section of the Component manual DTU 81 In some peculiar configurations it is necessary to position one or more groups of com ponents nested in parallel or overlapping One example is a multiple crystal monochro mator One would then like the neutron ray to interact with one of the components of the group and then continue In order to handle such arrangements without removing neutrons groups are defined by the GROUP modifier after the AT ROTATED positioning GROUP name to all involved component declarations All components of the same named group are tested one after the other until one of them interacts uses the SCATTER macro The selected component acts on the neutron ray and the rest of the group is skipped Such groups are thus exclusive only one of the elements is active Within a GROUP all EXTEND sections of the g
133. multiple crystal slabs with anisotropic Gaussian mosaic Flat Monochromator crystal with anisotropic Gaussian mosaic Polarising bender Guide with semi transparent polarising mirror Polarising mirror Numerical precession in analytical B fields A velocity selector helical lamella type such as V_selector component Rectangular circular slit Velocity selector Curved Fermi chopper longer execution time than FermiChop per From the Vitess package Table 7 2 Optics components of the McStas library DTU 103 MCSTAS samples Description Inelastic_Incoherent Isotropic_Sqw Phonon_simple Powder1 PowderN Sans_spheres Single_crystal V_sample Inelastic incoherent sample with quasielastic and elastic contri butions A general S q w scatterer with multiple scattering for liquids powders glasses polymers May be concentrically arranged Coherent incoherent elastic inelastic scattering Single crystal sample with acoustic isotropic phonons simple General powder sample with a single scattering vector General powder sample with N scattering vectors using a data file Can assume concentric shape i e can be used to model sample enviroment Simple sample for Small Angle Neutron Scattering hard spheres Mosaic single crystal with multiple scattering vectors using a data file Vanadium sample or other incoherent scatterer quasielastic broadening Optional Res_sample TOF_Res_s
134. n count sets the number of neutron rays to simulate the ncount option DTU 51 Run Run Run Run Run Run Run Run Run Run Run Run 52 Gravity Activates gravitation handling Not all components full support the use of gravitation but all transport in free space using the PROP_DT PROP_Z0 etc macros will include propagation with gravity Only local internal component propagation without the PROP routines will be gravity less As a conclusion it is considered safe and to high precision correct to apply the gravitation setting if one takes care to use the Guide_gravity component and other gravity supporting guide types in combination with non gravity components that are small in size i e samples lenses etc Random seed Set seed to selects between using a random seed different in each simulation for the random number generator or using a fixed seed to reproduce results for debugging Simulate Trace 3D Optimize selects between several modes of running the simulation e Simulate perform a normal simulation or a scan when steps is set to non zero value e Trace 3D view View the instrument in 3D tracing individual neutrons through the instrument e Optimize find the optimum value of the simulation parameters in the given ranges see section 5 3 5 e Backgrounding bg Simulate or Optimize in the background steps optim sets the number of simulation to run when
135. n matrice is referred to as ROT_A COMP comp m n where m and n are 0 1 or 2 e INDEX_CURRENT_COMP is the number index of the current component starting from 1 e POS_A_COMP_INDEX indez is the absolute position of component index POS A COMP INDEX INDEX_CURRENT_COMP is the same as POS_A_CURRENT_COMP You may use POS_A_COMP_INDEX INDEX_CURRENT_COMP 1 to make for instance your component access the position of the next component this is usefull for automatic targeting A component of the vector is referred to as POS_A COMP_INDEX indezx i where i is x y or z e POS_R_COMP_INDEX works the same as above but with relative coordinates e STORE_NEUTRON indez x y z vx vy vz t S sy SZ p stores the current neu tron state in the trace history table in local coordinate system index is usually IN DEX_CURRENT_COMP This is automatically done when entering each component of an instrument e RESTORE NEUTRON indez Y Z vx vy vz t sx SY SZ p restores the neu tron state to the one at the input of the component index To ignore a component effect use RESTORE NEUTRON INDEX CURRENT_COMP L Y Z VT vy UZ t S SY SZ p at the end of its TRACE section or in its EXTEND section These neutron states are in the local component coordinate systems e SCATTERED is a variable set to 0 when entering a component which is incre mented each time a SCATTER event occurs This may be used in the EXTEND sections t
136. nd neutron events all information printed to screen must use the error stream stderr so that all printf and fprintf stdout occurrences should be changed man ually into fprintf stderr To use the mcstas2vitess front end the program Perl should be available 5 4 8 Translating McStas results files between Matlab and Scilab for mats If you have been running a McStas simulation with Scilab output but finally plan to look at the results with Matlab or the contrary you may use mcconvert _file m sci to simply translate one file format to the other This works only for text files of course The binary files need not be translated 5 4 9 Translating and merging McStas results files all text formats If you have been running a McStas simulation with a given text format output but finally plan to look at the results with an other plotter e g you ran a simulation with PGPLOT output and want to view it using Matlab you may use mcformat file dir d target_dir format TARGET_FORMAT to translate files into format TARGET_FORMAT When given a directory the translation works recursively The conversion works only for text files Some problems may be found when translating from HTML VRML data files or from binary files 60 DTU The merge option may be used to merge similar files e g obtained from grid systems just as if a longer run was achieved The scan option may be used to reconstruct the scan data from a se
137. nent geometry NEXUS I Defines NeXus output type 4 5 XML compression OUTPUT C Defines internal variables to be public and protected symbols usually all global variables and functions of DECLARE PARAMETERS C Defines a class of component parameter DEFINITION SET TING POLARISATION C Defines neutron polarisation coordinates PREVIOUS C Refers to a previous component position orientation RELATIVE I Indicates that the AT and ROTATED keywords are relative to an other component REMOVABLE I Indicates that this component will be removed when the instru ment is inserted into an other one using the include keyword ROTATED I Indicates the orientation of a component in an instrument def inition SAVE I C Embedded C code to execute when saving data SETTING C Defines component parameters that are variables SHARE C Declares global functions and variables to be shared TRACE I C Defines the instrument as a the component sequence WHEN I Condition for component activation and JUMP Table 6 1 Reserved McStas keywords Scope is I for instrument and C for component definitions DTU 73 6 3 Writing instrument definitions The purpose of the instrument definition is to specify a sequence of components along with their position and parameters which together make up an instrument Each component is given its own local coordinate system the position and orientation of which may be specified by its translation
138. nstrument TRACE section is SPLIT COMPONENT name comp parameters WHEN condition AT RELATIVE reference PREVIOUS ABSOLUTE 80 DTU ROTATED RELATIVE reference PREVIOUS ABSOLUTE GROUP group_name EXTEND C_code JUMP reference PREVIOUS MYSELF NEXT ITERATE number_of_times WHEN condition 6 4 1 Embedding instruments in instruments TRACE The include insertion mechanism may be used within the TRACE section in order to concatenate instruments together This way each DECLARE INITIALIZE SAVE and FINALLY C blocks as well as instrument parameters from each part are concatenated The TRACE section is made of inserted COMPONENTS from each part In principle it is then possible to write an instrument as DEFINE concatenated TRACE include parti instr include part2 instr END where each inserted instrument is a valid full instrument In order to avoid some com ponents to be duplicated e g Sources from each part a special syntax in the TRACE section REMOVABLE COMPONENT a marks the component a as removable when inserted In principle inserted instruments may themselves use 4include 6 4 2 Groups and component extensions GROUP EXTEND It is sometimes desirable to slightly modify an existing component of the McStas library One would usually make a copy of the component and extend the code of its TRACE section McStas provides an easy way to change the behaviour of existing co
139. nt e In the IDENTIFICATION section author or written by for backwards compat ibility with old comments denote author date version and origin are also supported Any number of Modified by entries may be used to give the revision DTU 99 history The author date etc entries must all appear on a single line of their own Everything else in the identification section is part of a short description of the component e In the 4PARAMETERS section descriptions have the form name unit text or name text unit These may span multiple lines but subsequent lines must be indented by at least four spaces Note that square brackets should be used for units Normal parentheses are also supported for backwards compatibility but nested parentheses do not work well e The 4DESCRIPTION section contains text in free format The text may contain HTML tags like lt IMG gt to include pictures and lt A gt lt A gt for links to other web pages but see also the ZLINK section In the generated web documentation pages the text is set in lt PRE gt lt PRE gt so that the line breaks in the source will be obeyed e The ZEXAMPLE lines in instrument headers indicate an example parameter set or command that may be run to test the instrument A following Detector lt name gt _I lt value gt indicates what value should be obtained for a given monitor More than one example line may be specified in instruments
140. nt variable For instance calling mcdisplay pScilab Samples_vanadium out ROT 90 or csh tcsh syntax setenv MCSTAS_FORMAT Scilab mcdisplay Samples_vanadium out ROT 90 will output graphics using Scilab The mcdisplay front end can also be run from the mcgui front end Examples of plotter appearence for mcdisplay is shown in figures 5 4 5 6 McStas PGPLOT back end This will view the instrument from above A multi display that shows the instrument from three directions simultaneously can be shown using the multi option mcdisplay multi sim out args DTU 55 Click the left mouse button in the graphics window or hit the space key to see the display of successive neutron trajectories The P key saves a postscript file containing the current display that can be sent to the printer to obtain a hardcopy the C key produces color postscript To stop the simulation prematurely type Q or use control C as normal in the window in which mcdisplay was started To see details in the instrument it is possible to zoom in on a part of the instrument using the middle mouse button or the Z key on systems with a one or two button mouse The right mouse button or the X key resets the zoom Note that after zooming the aspect ratio of the plot may have changed and thus the angles as seen on the display may not match the actual angles Another way to see details while maintaining an overview of the instrument is to
141. ntional detector Integrated intensities error etc is also reported as for DETECTOR_OUT_OD 121 DETECTOR _OUT 3D Used to output the results from a three dimentional detector Arguments are the same as in DETECTOR_OUT_2D but with an addi tional z axis Resulting data files are treated as 2D data but the 3rd dimension is specified in the type field Integrated intensities error etc is also reported as for DETECTOR_OUT_OD mcinfo_simulation FILE f mcformat char pre char name is used to append the simulation parameters into file f see for instance Res_monitor Internal variable mcformat should be used as specified Please contact the authors for further information B 1 6 Ray geometry intersections inside_rectangle amp z amp y x rw yh Return 1 if rw 2 lt x lt aw 2 AND yh 2 lt y lt yh 2 Else return 0 box_intersect amp t amp t2 x Y Z Ur Vy Uz d dy dz Calculates the 0 1 or 2 intersections between the neutron path and a box of dimensions dz dy and dz centered at the origin for a neutron with the parameters x y z Uz Vy vz The times of intersection are returned in the variables t and t2 with t lt tg In the case of less than two intersections t and possibly t2 are set to zero The function returns true if the neutron intersects the box false otherwise cylinder_intersect amp t amp t2 x Y Z Ur Vy Vz r h Similar to box_intersect but using a cylinder o
142. o determine whether the component interacted with the current neutron ray e extend_list n amp arr amp len elemsize Given an array arr with len elements each of size elemsize make sure that the array is big enough to hold at least n elements by extending arr and len if necessary Typically used when reading a list of numbers from a data file when the length of the file is not known in advance e mcset_ncount n Sets the number of neutron histories to simulate to n e mcget_ncount Returns the number of neutron histories to simulate usually set by option n e mcget_run_num Returns the number of neutron histories that have been simu lated until now B 1 3 Coordinate transformations e coords_set z y z returns a Coord structure like POS_A CURRENT_COMP with xz y and z members 120 DTU coords _get P amp x amp y amp z copies the x y and z members of the Coord structure P into x y z variables coords_add a b coords_sub a b coords _neg a enable to operate on coordi nates and return the resulting Coord structure rot_set_rotation Rotation t dz dy dz Get transformation matrix for rotation first gx around x axis then y around y and last around z t should be a Rotation 3 3 double matrix rot_mul Rotation t1 Rotation t2 Rotation t3 performs t3 1 2 rot_copy Rotation dest Rotation src performs dest src for Rotation arrays t rot_transpose Rotation src Rotation des
143. o use grid and cluster computing you will need an SSH client McStas is configured to use PuTTY For MPI parallelisation on Windows we advice you to install MPICH2 from Argonne National Laboratory including development libraries before installing McStas Also your mpiexec exe must be on the PATH You may have to customize the mpicc bat script from the McStas distribution with the proper C compiler and MPI library path If you experience any problems or have some questions or ideas concerning McStas please contact peter willendup risoe dk or the McStas mailing list at mcstas users mcstas org 3 4 Installation on Unix systems Our current reference Unix class platform is Ubuntu Linux which is based on Debian GNU Linux Some testing is done on other Unix variants including Fedora Core SuSE and FreeBSD WARNING The dash shell which is used as bin sh on some Linux system Including Ubuntu 8 04 makes the Cancel and Update buttons fail in mcgui Possible solutions are e If your system is a Debian or Ubuntu please run the command dpkg reconfigure dash and say no to install dash as bin sh See section 3 4 1 e If your bin sh is dash please install bash and manually change the bin sh link to point at bash 20 DTU 3 4 1 Debian class systems As of release 2 0 we provide a Debian binary package 32 bit package for Intel AMD We have tested that the package works properly on Ubuntu and Debian systems To insta
144. odifier This optional keyword comes before the AT ROTATED positioning It basically enables the component only when a given condition is true non null COMPONENT name comp pi 1 p2 2 WHEN condition The condition has the same scope as the EXTEND modifier i e may use component internal variables as well as all the DECLARE instrument variables except instrument pa rameters To use instrument parameters you should copy them into global variables in the DECLARE instrument section and refer to these latter It is evaluated before the component TRACE section Usage examples could be to have specific monitors only sensitive to selected processes or to have components which are only present under given circumstances e g removable guide or radial collimator or to select a sample among a set of choices In the following example an EXTEND block sets a condition when a scattering event is encountered and the following monitor is then activated COMPONENT Sample V_sample AT EXTEND ht if SCATTERED flag 1 else flag 0 ht COMPONENT MyMon Monitor WHEN flag 1 Ale 255 The WHEN keyword only applies to the TRACE section and related EXTEND blocks of instruments components Other sections INITIALIZE SAVE MCDISPLAY FI NALLY are executed independently of the condition As a side effect the 3D view of the instrument mcdisplay will show all components as if all conditions were true Also the WHEN k
145. oints 1 y1 21 2 Y2 22 Zn Yn Zn It thus accepts a variable number of arguments depending on the value of n This exposes one of the nasty quirks of C since no type checking is performed by the C compiler It is thus very important that all arguments to multiline except n are valid numbers of type double A common mistake is to write multiline 3 x y 0 which will silently produce garbage output This must instead be written as multiline 3 double x double y 0 0 The rectangle command The rectangle command takes the following form rectangle plane x y z width height Here plane should be either xy xz or yz The command draws a rectangle in the specified plane with the center at x y z and the size width x height Depending on plane the width and height are defined as plane width height xy x y XZ x Z yz y Z The box command The box command takes the following form box z y z xwidth yheight zlength The command draws a box with the center at x y z and the size rwidth x yheight x zlength The circle command The circle command takes the following form circle plane Y Z r Here plane should be either xy xz or yz The command draws a circle in the specified plane with the center at x y z and the radius r DTU 95 6 5 9 The end of the component definition END This marks the end of the component definition 6 5 10 A component example Slit A simp
146. ommit yourself to using a tool for an important project you need to know if the tool will satisfy not only your present but also your future requirements The tool must not have fundamental limitations that restrict its potential usage Thus the McStas systems needs to be flexible enough to simulate different kinds of instruments as well as many different kind of optical components and it must also be extensible so that future as yet unforeseen needs can be satisfied Power Simple things should be simple complex things should be possible New ideas should be easy to try out and the time from thought to action should be as short as possible If you are faced with the prospect of programming for two weeks before getting any results on a new idea you will most likely drop it Ideally if you have a good idea at lunch time the simulation should be running in the afternoon Efficiency Monte Carlo simulations are computationally intensive hardware capacities are finite albeit impressive and humans are impatient Thus the system must assist in producing simulations that run as fast as possible without placing unreasonable burdens on the user in order to achieve this 1 3 The design of McStas In order to meet these ambitious goals it was decided that McStas should be based on its own meta language also known as domain specific language specially designed for simulating neutron scattering instruments Simulations are written in this
147. on time and gives better accuracy Early implementations of the MC method for neutron instruments used home made computer programs see 42 43 but more recently general packages have been designed providing models for most optical components of neutron spectrometers The most widely used packages are NISP 9 ResTrax 10 McStas 1 2 Vitess 12 and IDEAS 14 which allow a wide range of neutron scattering instruments to be simulated The neutron ray tracing Monte Carlo method has been used widely for guide stud ies 17 19 instrument optimisation and design 20 21 Most of the time the conclu sions and general behaviour of such studies may be obtained using the classical analytical approaches but accurate estimates for the flux resolution and generally the optimum parameter set benefit considerably from MC methods see 4 Neutron instrument resolution in q and and flux are often limitations in the exper iments This then motivates instrument responsibles to improve the resolution flux and overall efficiency at the spectrometer positions and even to design new machines Using both analytical and numerical methods optimal configurations may be found But achieving a satisfactory experiment on the best neutron spectrometer is not all Once collected the data analysis process raises some questions concerning the signal what is the background signal What proportion of coherent and incoherent scattering has been measured Is
148. on must be explicitly marked using the keyword END 6 3 9 Code for the instrument vanadium_example instr A commented instrument definition taken from the examples directory is here shown as an example of the use of McStas DOO OOO OO OO GO OO OO A A RK A a ak 2k 2k 4 2 2k McStas neutron ray tracing package Copyright C 1997 2008 All rights reserved Risoe National Laboratory Roskilde Denmark Institut Laue Langevin Grenoble France Instrument Samples_vanadium Identification Written by Kristian Nielsen and Kim Lefmann 78 DTU Date 1998 Origin Risoe Release McStas CVS_080624 Version Revision 1 3 ZINSTRUMENT_SITE Tests_samples A test instrument using a vanadium cylinder Description This instrument shows the vanadium sample scattering anisotropy This is an effect of attenuation of the beam in the cylindrical sample Example ROT 0 Detector PSD_4pi_I 7 97023e 05 Parameters INPUT PARAMETERS ROT deg Rotation angle of the PSD monitor Link The McStas User manual The McStas tutorial End BAO AOA IORI RR aK Kk Kak kk kk The line below defines the name of our instrument Here we have a single input parameter ROT DEFINE INSTRUMENT Samples_vanadium ROT 0 The DECLARE section allows us to declare variables in c syntax Here coll_div collimator divergence is set to 60
149. on to the graphical user interface 32 5 1 1 New releases of McStas aaa 00002 eee ee eee 35 5 2 Running the instrument compiler 2020 0000 5 35 5 2 1 Code generation options 2 2 ee ee 36 5 2 2 Specifying the location of files 2 20004 36 5 2 3 Embedding the generated simulations in other programs 37 5 2 4 Running the C compiler 0 2 02 0 0000 37 5 3 Running the simulations oaoa 39 5 3 1 Choosing an output data file format 40 5 3 2 Basic import and plot of results aooaa 40 5 3 3 Interacting with a running simulation 41 5 3 4 Optimizing simulation speed oaoa 44 5 3 5 Optimizing instrument parameters sooo oo e e a 45 5 4 Using simulation front ends ooo a 47 5 4 1 The graphical user interface mcgui ooa a 48 5 4 2 Running simulations on the commandline merun 53 5 4 3 Graphical display of simulations mcdisplay 55 5 4 4 Plotting the results of a simulation mcplot 57 5 4 5 Plotting resolution functions mcresplot 58 5 4 6 Creating and viewing the library component instrument help and Manuals medoc 3 4a back WekeaieS a 59 5 4 7 Translating McStas components for Vitess mestas2vitess 59 5 4 8 Translating McStas results files between Matlab and Scilab formats 60 5 4 9 Translating and merging McStas results files all text formats
150. ont end programs to visualize the scattering events properly and to handle component GROUPs in an instrument definition see section 6 3 5 It basically increments the SCATTERED counter The SCATTER macro should be called with the state parameters set to the proper values for the scattering event so that neutron events are displayed correctly For an example of SCATTER see the optics Guide component 6 5 6 The SAVE section SAVE At C code to execute in order to save data hy This gives code that will be executed when the simulation ends or is requested to save data for instance when receiving a USR2 signal on Unix systems see section 5 3 or when triggered by the Progress_bar flag_save 1 component This might be used by monitors and detectors in order to write results An extension depending on the selected output format see table 5 3 and section 5 3 is automatically appended to file names if these latter do not contain extension In order to work properly with the common output file format used in McStas all monitor detector components should use standard macros for writing data in the SAVE or FINALLY section as explained below In the following we use N 50 p to denote the count of detected neutron events p p to denote the sum of the weights of detected neutrons and p2 50 p to denote the sum of the squares of the weights as explained in section 4 2 1 As a default all monitors using the standard macros
151. osts in your home directory or in the MCSTAS tools per1 mcstas hosts on the master node one node per line The machines lt file gt option enables to specify the hosts file to use If it does not exist only the current machine will be used for multi processor core machines DTU 65 6 Without ssh keys passwords will be prompted many times To avoid this we recom mend to use only the local machine for multi cores cpu i e do not use a machine hosts file 7 If your simulation instrument requires data files Powders Sqw source description these must be copied at the same level as the instrument definition They are sent to all slave nodes before starting each computation Take care to limit as much as possible the required data file volume in order to avoid large data transfers 8 Interrupting or sending Signals may fail during computations However simulation scans can be interrupted as soon as the on going computation step ends 9 With heterogeneous systems we recommend to use the mcrun force compile command rather than McGUI which may skip the required simulation compilation on slaves 5 6 2 Parallel computing MPI The MPI support requires that MPICH recommended or alternatively LAM MPI or OpenMPI is installed on a set of nodes This usually also requires properly setup ssh connections and keys as indicated in the ssh grid system Section 5 6 1 Some OpenMPI implementations do not recognise multi cores as s
152. parallel using MPI We recommend this method on clusters see section 5 6 2 Last but not least you may run simulations manually on a number of machines Once the distributed simulation have been completed you may merge their results using mcformat see section 5 4 9 in order to obtain a set of files just as if it had been executed on a single machine All of these methods can be used when available from mcgui 64 DTU 5 6 1 Distribute mcrun simulations on grids multi cores and clusters SSH grid This method distributes simulations on a set of machines using ssh connections using a command such as mcrun grid 4 Each of the scan steps is split and executed on distant slave machines sending the executable with scp executing single simulations and then retrieving individual results on the master machine These are then merged using mcformat The mcrun script has been adapted to use transparently SSH grids The syntax is e grid lt number gt tells merun to use the grid over lt number gt nodes e machines lt file gt defines a text file where the nodes which are to be used for parallel computation are listed by default mcrun will look at HOME mcstas hosts and MCSTAS tools perl mcstas hosts When used on a single SMP machine multi core cpu this option may be omitted e force compile this option is required on heterogeneous systems The C code is sent to all slaves and simulation is compiled on each node b
153. performing a pa rameter scan or the number of iterations to perform in optimization mode Plot results if checked the mcplot front end will be run after the simulation has finished and the plot dialog will appear see below Format quick selection of output format Binary mode may be checked from the Simulation Configuration options dialog box Clustering method selects the mechanism to be used for running on grids and clusters See section 5 6 on parallel computing for more informations Number of nodes sets the number of nodes to use for MPI ssh clustering Inspect component Trace mode will trace only neutron trajectories that reach a given component e g sample or detector First component Trace mode seletcs the first component to plot default is first in order to define a region of interest Last component Trace mode seletcs the last component to plot default is first in order to define a region of interest Maximize monitor Optimization mode seletcs up to three monitors which inte gral value should be maximized varying instrument parameters If no monitor is selected the sum of all monitors is optimized DTU Run Start runs the simulation Run Cancel aborts the dialog Most of the settings on the run dialog can be saved for your next McStas run using Save configuration in the File menu Before running the simulation the instrument definition is automatically compiled if it is newer than t
154. plications are required Essentially you will need a C compiler Perl and Perl Tk as well as a plotter such as Matlab Scilab or PGPLOT Using Scilab is not recommended and support will eventually end In the installer package we supply a method to install PGPLOT and related perl modules see step 5 below DTU 21 On Debian and Ubuntu systems the needed packages to install are perl tk pdl gcc libc6 dev On Ubuntu you need to enable the universe package distribution in the file etc apt sources 1list We also recommend to install octaga vrml viewer from http www octaga com download_octaga html Additionally MPICH OpenMP gcc 4 2 or icc or pgcc openssh Octave Gnuplot HDF and NeXus libraries may be installed to enhance McStas clustering method and data formats 3 4 3 Configuration and installation McStas uses autoconf to detect the system configuration and creates the proper Makefiles needed for compilation On Unix like systems you should be able to compile and or install McStas using the following steps 1 Unpack the sources to somewhere convenient and change to the source directory gunzip c lt package gt tar gz tar xf cd mcstas 2 0 2 Configure McStas configure or configure with nexus with cc gcc 4 2 3 Build McStas only in case of the mcstas 2 0 srce tar gz package make 4 Install McStas as superuser make install 5 Optionally build install PGPLOT as superuser build dependen
155. ptions have been used see section 5 3 the name of the file or directory should be passed to meplot e g mcplot dir or mcplot file It is also possible to plot one single text not binary data file from a given monitor passing its name to mcplot The drawing back end program may be selected among PGPLOT VRML Matlab and Scilab using either the pPLOTTER option e g mcplot pScilab file or using the current MCSTAS_FORMAT environment variable Moreover the drawing back end program will also be set depending on the file extension see Table 5 3 It should be emphasised that mcplot may only display simulation results with the format that was chosen during the computation Indeed if you request data in a given format from a simulation you will only be able to display results using that same drawing back end Anyway the mcconvert utility may convert a McStas data file between Matlab and Scilab formats see section 5 4 8 also handling binary data sets Alternatively the mcformat utility will convert any McStas result into data format see section 5 4 9 but restricting to text data sets In this case we recommend to generate data sets using PGPLOT McStas format and translate into any other format using mcformat The mcplot front end can also be run from the mcgui front end The initial display shows plots for each detector in the simulation Examples of plotter appearence for mcplot is shown in figures 5 4 5 3 McStas PGPLOT
156. r Additionally the directory will have sub directories 1 2 3 containing all data files output from the different simulations When the d option is not used no data files are written from the individual simu lations in order to save disk space e The mcrun test command will test your McStas installation accuracy and plot ter The h option will list valid options The mcrun front end requires a working installa tion of Perl to run 5 4 3 Graphical display of simulations mcdisplay The front end mcdisplay mcdisplay pl on Windows is a graphical visualization tool very useful for debugging It presents a schematic drawing of the instrument definition showing the position of the components and the paths of the simulated neutrons through the instrument It is thus very useful for debugging a simulation for example to spot components in the wrong position or to find out where neutrons are getting lost See figures 5 4 5 6 To use the mcdisplay front end with a simulation run it as follows mcdisplay sim args where sim is the name of either the instrument source sim instr or the simulation pro gram sim out generated with McStas and args are the normal command line argu ments for the simulation as explained above The h option will list valid options The drawing back end program may be selected among PGPLOT VRML Matlab and Scilab using either the pPLOTTER option or using the current MCSTAS_FORMAT environme
157. r data s s prisma2 m_mcstas detector prisma2_tof signal eval s we could also use the statistics field y Mean value of the simulation Then a numerical optimization should be available such as those provided with Matlab Scilab IDL and Perl PDL high level languages In this example we may wish to maximize the instr_value function value The fminsearch function of Matlab is a minimization method that s why we have a minus sign for y value and matlab gt TT fminsearch instr_value 25 will determine the best value of TT starting from 25 estimate in order to minimize function instr_value and thus maximize the mean detector counts 5 4 Using simulation front ends McStas includes a number of front end programs that extend the functionality of the simulations A front end program is an interface between the user and the simulations running the simulations and presenting the output in various ways to the user The list of available McStas front end programs may be obtained from the mcdoc tools command McStas Tools mcstas Main instrument compiler mcrun Instrument build and execution utility mcgui Graphical User Interface instrument builder mcdoc Component library documentation generator viewer mcplot Simulation result viewer mcdisplay Instrument geometry viewer mcresplot Instrument resolution function viewer mcstas2vitess McStas to Vitess component translation utility mcconvert Matlab lt
158. rary gathers components that were re named or considered to be outdated These component are kept for backwards compati bility and they still all work as before The mcdoc front end section 5 4 6 enables to display both the catalog of the McStas library e g using mcdoc as well as the documentation of specific components e g with mcdoc text name mcdoc file comp The first line will search for all components matching the name and display their help section as text where as the second example will display the help corresponding to the file comp component using your BROWSER setting or as text if unset The help option will display the command help as usual DTU 101 MCSTAS sources Description Adapt_check ESS_moderator_long ESS_moderator_short Moderator Monitor_Optimizer Source_Maxwell_3 Source_Optimizer Source_adapt Source_div Source_simple Source_gen Virtual_input Virtual_output Optimization specifier for the Source_adapt component A parametrised pulsed source for modelling ESS long pulses A parametrised pulsed source for modelling ESS short pulses A simple pulsed source for time of flight To be used after the Source_Optimizer component Continuous source with up to three Maxwellian distributions Optimizes the neutron flux passing through the Source_Optimizer in order to have the maximum flux at the Monitor_Optimizer position Continuous neutron source with adaptive importance sampling
159. rient pieces of the instrument and avoid positioning errors The position of a component is actually the origin of its local coordinate system Usually this is used as the input window position e g for guide like components or the center position for cylindrical spherical components The PREVIOUS keyword is a generic name to refer to the previous component in the simulation Moreover the PREVIOUS n keyword will refer to the n th previous component starting from the current component so that PREVIOUS is equivalent to PREVIOUS 1 This keyword should be used after the RELATIVE keyword but not for the first component instance of the instrument description AT x y z RELATIVE PREVIOUS ROTATED z dy z RELATIVE PREVIOUS 2 Invalid PREVIOUS references will be assumed to be absolute placement The order and position of components in the TRACE section does not allow components to overlap except for particular cases see the GROUP keyword below Indeed many components of the McStas library start by propagating the neutron event to the beginning of the component itself Anyway when the corresponding propagation time is found to be negative i e the neutron ray is already after or aside the component and has thus passed the active position the neutron event is ABSORBed resulting in a zero intensity and event counts after a given position The number of such removed neutrons is indicated at the end of the simulation Getting such
160. ron site Tools Histogrammer instr example instrument may read McStas Vitess MCNP and Tripoli event files in order to generate histograms of any type The Help menu has the following features through use of mcdoc and a web browser To customize the used web browser set the BROWSER environment variable If BROWSER is not set mcgui uses netscape mozilla firefox on Unix Linux and the default browser on Windows Help McStas User manual calls mcdoc manual brings up the local pdf version of this manual using a web browser Help McStas Component manual calls mcdoc comp brings up the local pdf ver sion of the component manual using a web browser Help Component library index displays the component documentation using the com ponent index html index file Help McStas web page calls mcdoc web brings up the McStas website in a web browser Help Tutorial opens the McStas tutorial for a quick start 50 DTU Help Current instrument info generates a description web page of the current edited instrument Help Test McStas installation launches a self test procedure to check that the McStas package is installed properly generates accurate results and may use the plotter to display the results Help Generate component index re generates locally the component index html The run dialog Run Simulation h8 test instr l Instrument source h8_test instr HTML docs Instrument parameters D floating point I
161. roup are executed In order to discriminate components that are active from those that are skipped one may use the SCATTERED flag which is set to zero when entering each component or group and incremented when the neutron is SCATTERed as in the following example COMPONENT name0 comp pi e1 p2 2 AT 0 0 0 ABSOLUTE COMPONENT namel comp AT ROTATED GROUP GroupName EXTEND nt if SCATTERED printf I scatter else printf I do not scatter ht COMPONENT name2 comp AT ROTATED GROUP GroupName Components namel and name2 are at the same position If the first one intercepts the neutron and has a SCATTER within its TRACE section the SCATTERED variable be comes true the code extension will result in printing I scatter and the second component will be skipped Thus we recommend to make use of the SCATTER keyword each time a component uses the neutron scatters detects within component definitions see section 6 5 Also the components to be grouped should be consecutive in the TRACE section of the instrument and the GROUPed section should not contain components which are not part of the group A usage example of the GROUP keyword can be found in the Neutron site ILL ILL_H15_IN6 instrument from the mcgui to model 3 monochroma tors Combining EXTEND GROUP and WHEN can result in unexpected behaviour Please read the related warning at the end of section 6 4 4 6 4 3
162. rument usually close to the end of the instrument description If some components make extensive use of random numbers MC choices they shuffle this way the distributions so that identical incoming events will not produce the same outgoing event In this case you may use the SPLIT keyword with the syntax SPLIT r COMPONENT name comp where the optional number r specifies the number of repetitions for each event Default is r 10 Each neutron event reaching component name will be repeated r times with a weight divided by r so that in practice the number of events for the remaining part of the simulation down to the END will potentially have more statistics This is only true if following components and preferably component name use random numbers You may use this method as many times as you wish in the same instrument e g at the monochromator and sample position This keyword can also be used within a GROUP The efficiency is roughly r raised to the number of occurrences in the instrument so that enhancing two components with the default r 10 will produce at the end an enhancement effect of 100 in the number of events The execution time will usually get slightly longer This technique is known as the stratified sampling see Appendix 4 If the instrument makes use of global variables e g in conjunction with a WHEN or User Variable monitoring see Monitor_nD you should take care that these variables are set properly for ea
163. ry contains a collection of well tested as well as user contributed beam components that can be used in simulations The McStas com ponent library is documented in a separate manual and on the McStas web page 2 but a short overview of these components is given in chapter 7 of the Manual A collection of example instrument definitions is described in chapter 8 of the Man ual As of this release of McStas support for simulating neutron polarisation is strongly improved e g by allowing nested magnetic fields tabulated magnetic fields in numerical inputfiles and by close to full support of polarisation in all components As this is the first stable release with these new features functionality is likely to change To reflect this the documentation is still only available in the appendix of the Component manual A list of library calls that may be used in component definitions appears in appendix B and an explanation of the McStas terminology can be found in appendix C of the Manual Plans for future extensions are presented on the McStas web page 2 as well as in section 2 6 14 DTU Chapter 2 New features in McStas 2 0 UPDATE This version of McStas implements both new features as well as many bug corrections Bugs are reported and traced using the McStas Trac Ticket system 3 We will not present here an extensive list of corrections and we let the reader refer to this bug reporting service for details Only importan
164. s Reads a PTRAC neutron events file from MCNP Writes a PTRAC neutron events file for MCNP Reads a Batch neutron events file from Tripoli 4 Writes a Batch neutron events file for Tripoli 4 Table 7 6 Contributed components of the McStas library DTU 105 MCSTAS share Description adapt_tree lib mcstas r monitor_nd lib read_table lib vitess lib Handles a simulation optimisation space for adatative impor tance sampling Used by the Source_adapt component Main Run time library always included Handles multiple monitor types Used by Monitor_nD Res_monitor Enables to read a data table text binary to be used within an instrument or a component Enables to read write Vitess event binary files Used by Vitess_input and Vitess_output Table 7 7 Shared libraries of the McStas library See Appendix B for details MCSTAS data Description lau Laue pattern file as issued from Crystallographica For use with Single crystal PowderN and Isotropic_Sqw Data h k 1 Mult d space 2Theta F squared laz Powder pattern file as obtained from Lazy ICSD For use with PowderN Isotropic Sqw and possibly Single crystal trm transmission file typically for monochromator crystals and fil ters Data k Angs 1 Transmission 0 1 rfl reflectivity file typically for mirrors and monochromator crys tals Data k Angs 1 Reflectivity 0 1 sqw S q w
165. s Note that s and s2 are obsolete and stand for sy and sy members of the polarisation vector To access sz use POLARISATION PARAMETERS sx Sy Sz This line is necessary only if the component handles polarisation of neutrons and thus modifies the spin vector For an instrument to handle polarisation correctly it is only required that one of the components contains this line Optional component parameters Just as for instrument parameters the definition and setting parameters of a component may be given a default value Parameters with default values are called optional parame ters and need not be given an explicit value when the component is used in an instrument definition A parameter is given a default value using the syntax param value For example SETTING PARAMETERS radius height pack 1 88 DTU Here pack is an optional parameter and if no value is given explicitly 1 will be used In contrast if no value is given for radius or height an error message will result Optional parameters can greatly increase the convenience for users of components with many parameters that have natural default values which are seldom changed Optional parameters are also useful to preserve backwards compatibility with old instrument defi nitions when a component is updated New parameters can be added with default values that correspond to the old behaviour and existing instrument definitions can be used with the new componen
166. s of new versions of a software package can today be carried out more or less continuously However users do not update their software on a daily basis and as a compromise we have adopted the following policy of McStas e The versions 2 0 x will possibly contain bug fixes and minor new functionality A new manual will however not be released and the modifications are documented on the McStas web page The extensions of the forthcoming version 2 0 x are also listed on the web and new versions may be released quite frequently when it is requested by the user community 5 2 Running the instrument compiler This section describes how to run the McStas compiler manually Often it will be more convenient to use the front end program mcgui section 5 4 1 or mcrun section 5 4 2 These front ends will compile and run the simulations automatically The compiler for the McStas instrument definition is invoked by typing a command of the form mcstas name instr This will read the instrument definition name instr which is written in the McStas meta language The compiler will translate the instrument definition into a Monte Carlo simu lation program provided in ISO C The output is by default written to a file in the current directory with the same name as the instrument file but with extension c rather than instr This can be overridden using the o option as follows DTU 35 mcstas o code c name instr EY which gives the output in th
167. ss neutron events file format Use Vitess_input Vitess_output components McStas event files in text or binary format Use Virtual_input Virtual_output components Table 5 3 Available formats supported by McStas simulations Format modifiers include binary raw transpose append DTU 43 parallelization McStas attaches a signal handler to the simulation process In order to send a signal to the process the process id pid must be known Users may look at their running processes with the Unix ps command or alternatively process managers like top and gtop If a file out simulation obtained from McStas is running the process status command should output a line resembling lt user gt 182777140 99 23 52 pts 2 00 00 13 file out where user is your Unix login The pid is there 13277 Once known it is possible to send one of the signals listed in Table 5 4 using the kill unix command or the functionalities of your process manager e g kill USR2 13277 This will result in a message showing status here 33 achieved as well as the position in the instrument of the current neutron McStas pid 13277 Signal 12 detected SIGUSR2 Save simulation Simulation file file instr Breakpoint MyDetector Trace 33 37 333654 0 1000000 0 Date Wed May 7 00 00 52 2003 McStas Saving data and resume simulation continue followed by the list of detector outputs integrated counts
168. sult the McStas home page 2 A special bug request reporting service is available 3 If you appreciate this software please subscribe to the mcstas users mcstas org email list send us a smiley message and contribute to the package We also encourage you to refer to this software when publishing results with the following citations e K Lefmann and K Nielsen Neutron News 10 3 20 1999 e P Willendrup E Farhi and K Lefmann Physica B 350 2004 735 e P Willendrup E Farhi E Knudsen U Filges and K Lefmann Journal of Neutron Research 17 expected 2013 McStas 2 0 contributors Several people outside the core developer team have been contributing to McStas 2 0 e UPDATE THIS LIST Thank you guys This is what McStas is all about Third party software included in McStas are e UPDATE THIS LIST e perl Math Amoeba from John A R Williams J A R Williams aston ac uk DTU 7 e perl Tk Codetext from Hans Jeuken haje toneel demon nl e scilab Plotlib from St phane Mottelet mottelet utc fr e and optionally PGPLOT from Tim Pearson tjp astro caltech edu The McStas project has been supported by the European Union through XENNI Cool Neutrons FP4 SCANS FP5 nmi3 MCNSI FP6 McStas was supported directly from the construction project for the ISIS second target station TS2 EU see 4 Currently McStas is supported through Danish in kind work packages toward the Euro pean Spallation Source
169. t performs dest src rot_apply Rotation t Coords a returns a Coord structure which is t a B 1 4 Mathematical routines NORM z y z Normalizes the vector x y z to have length 1 scalar_prod daz dy az br by bz Returns the scalar product of the two vectors az ay az and bz by bz vec_prod amp az amp ay amp a bz by bz Cx Cy Cz Sets ax dy az equal to the vector prod uct br by bz X Cx Cy z rotate amp 1 amp y amp 2 Vz Vy Uz P dx yz Set x y z to the result of rotating the vector Vz Vy Uz the angle y in radians around the vector az dy dz normal_vec amp n amp n amp nz x y z Computes a unit vector nz Ny nz normal to the vector x y z solve_2nd_order t A B C Solves the 2 order equation At Bt C 0 and returns the smallest positive solution into pointer t B 1 5 Output from detectors Details about using these functions are given in the McStas User Manual DTU DETECTOR OUT_OD Used to output the results from a single detector The name of the detector is output together with the simulated intensity and es timated statistical error The output is produced in a format that can be read by McStas front end programs DETECTOR OUT_1D Used to output the results from a one dimensional detector Integrated intensities error etc is also reported as for DETECTOR_OUT_OD DETECTOR_OUT_2D Used to output the results from a two dime
170. t changes are indicated here Of course we can not guarantee that the software is bullet proof but we do our best to correct bugs when they are reported 2 1 Kernel The following changes concern the Kernel i e the McStas meta language and program See the dedicated chapter in the User manual for more details e The STATE PARAMETERS keywords has been removed from components code We now assume that the neutron state parameters are x y Z VX Vy VZ SX SY SZ t McStas 1 X components will need to be adapted by commenting this line The PREVIOUS and MYSELF keywords can be used in component instance parameters and positioning AT ROTATED so that one can make use of MC_GETPAR PREVIOUS parameter directly in the TRACE section 2 2 Run time The number of neutron counts can now exceed 2 10 being stored as a long long e The writing of files has been improved The DETECTOR_OUT functions now return a mcdetector structure which holds all written information Simulations also write a content sim file aside the mcstas sim one which holds the integrated counts for each monitor e A bug has been fixed in the rectangular focusing routine for large solid angles Reported by M Skoulatos L Udby and J Jacobsen DTU 15 2 3 Components and Library We here list the new and updated components found in the McStas lib directory which are detailed in the Component manual also mentioned in the Component Overview of the User M
171. t character which may only be a letter When using Matlab an m_ is appended to avoid name clash with existing Matlab com mands variables In this tree you will find the monitor names which fields contain the monitored data The usual structure is s instrument simulation comp_name file name For instance reading the data from a test instrument using Matlab format will look like For instance in most case the simulation location is mcstas m which turns into field m_mcstas 62 DTU matlab gt s mcstas or mcplot mcstas m from the terminal matlab gt s s Format Matlab with text headers URL http neutron risoe dk Editor farhi on pcfarhi Creator test test instr McStas 1 9 Nov 15 2005 simulation Date 1 0529e 09 File mcstas test 1x1 struct EndDate 1 0529e 09 class root matlab gt s test m_mcstas monitor1 monitor1_y_kz ans Date 1 0529e 09 File gt monitor1l y_kz type array_2d 20 10 ratio gt 1e 06 1e 06 signal Min 0 Max 5 54051e 10 Mean 6 73026e 13 statistics X0 0 438302 dX 0 0201232 Y0 51019 6 dY 20557 1 matlab gt eval s test m_mcstas monitor1 monitor1_y_kz matlab gt dX ans 0 0201232 The latter example accesses the data from the monitor1 y_kz file written by the monitor1 component in the test instrument during the mcstas simulation You may evaluate directly th
172. t each sub space is well represented in the final integrals This means that instead of shooting N events we define D partitions and shoot r N D events in each partition In conjunction with adaptive sampling we may define partitions so that they represent interesting distribu tions e g from events scattered on a monochromator or a sample The sum of partitions should equal the total space integrated by the Monte Carlo method and each partition must be sampled randomly In the case of McStas an ad hoc implementation of adaptive stratified is used when re peating events such as in the Virtual sources Virtual_input Vitess_input Virtual_mcnp_input Virtual_tripoli4 input and when using the SPLIT keyword in the TRACE section on in strument descriptions We emphasize here that the number of repetitions r should not exceed the dimensionality of the Monte Carlo integration space which is d 10 for neu tron events and the dimensionality of the partition spaces i e the number of random generators following the stratified sampling location in the instrument 4 5 Accuracy of Monte Carlo simulations When running a Monte Carlo the meaningfull quantities are obtained by integrating random events into a single value e g flux or onto an histogram grid The theory 41 shows that the accuracy of these estimates is a function of the space dimension d and the number of events N For large numbers N the central limit theorem provides an es
173. t explanation on how to do this Assume that we pick a random number x from a distribution x We are now interested in the shape of the distribution U y of the transformed y f x assuming f x is monotonous All random numbers lying in the interval x x dz are transformed to lie within the interval y y f a dz so the resulting distribution must be U y o x f 2 If the random number generator selects numbers uniformly in the interval 0 1 we have x 1 inside the interval zero outside and we reach Fy A 1 By indefinite integration we reach i wonsi usr A 2 which is the essential formula for random number transformation since we in general know W y and like to determine the relation y f x Let us illustrate with a few examples of transformations relevant for the McStas components The circle For finding a random point within the circle of radius R one would like to choose the polar angle from a uniform distribution in 0 27 giving Vg 1 27 and the radius from the normalised distribution Y 2r R For the radial part eg A 2 becomes y 27 x whence is found simply by multiplying a random number x with 27 For the radial part the left side of eq A 2 gives f U r dr f 2r R dr r R which from A 2 should equal z Hence we reach the wanted transformation r Rx 116 DTU The sphere For finding a random point on the surface of the unit sphere we need to
174. t in a mcstas m file that may be read directly from Matlab using the sim file function matlab gt s mcstas matlab gt s mcstas plot The first line returns the simulation data as a single structure variable whereas the second one will additionally plot each detector separately This also equivalently stands for Scilab using the get_sim file function the exec call is required in order to compile the code 40 DTU s seed Set the initial seed for the random number generator This may seed seed be useful for testing to make each run use the same random number sequence n count Set the number of neutron histories to simulate The default is ncount count 1 000 000 1e6 d dir Create a new directory dir and put all data files in that direc dir dir tory h Show a short help message with the options accepted available help formats and the names of the parameters of the instrument i Show extensive information on the simulation and the instru info ment definition it was generated from t Makes the simulation output the state of every neutron as it trace passes through every component Requires that the t or trace option is also given to the McStas compiler when the simulation is generated no output files Disables the writing of data files output to the terminal such as detector intensities will still be written g Toggles the gravitation approxim
175. t is also extremely fast DTU 117 Appendix B Libraries and constants The McStas Library contains a number of built in functions and conversion constants which are useful when constructing components These are stored in the share directory of the MCSTAS library Within these functions the Run time part is available for all component instrument descriptions The other parts are dynamic that is they are not pre loaded but only imported once when a component requests it using the include McStas keyword For instance within a component C code block usually SHARE or DECLARE finclude read_table lib will include the read_table lib h file and the read_table lib c unless the no runtime option is used with mcstas Similarly finclude read_table lib h will only include the read_table lib h The library embedding is done only once for all components like the SHARE section For an example of implementation see Res_monitor In this Appendix we present a short list of both each of the library contents and the run time features B 1 Run time calls and functions mcstas r Here we list a number of preprogrammed macros which may ease the task of writing component and instrument definitions B 1 1 Neutron propagation Propagation routines perform all necessary operations to transport neutron rays from one point to an other Except when using the special ALLOW_BACKPROP call prior to exectuting any PR
176. t of direc tories which vary by instrument parameters For instance you ran a scan but finally realised you should have prolonged it Then simply simulate the missing bits and ap ply mcformat d scan_data format PGPLOT scan stepO stepN The resulting scan data is compatible with mcplot only when generating PGPLOT McStas format You may conjugate this option with the merge in order to add merge similar data sets before re building the scan The data files are analyzed by searching keywords inside data files e g Source for the source instrument description file If some filenames or component names match these keywords e g using a file Source psd the extracted metadata information may be wrong even though the data itself will be correct 5 5 Data formats Analyzing and visualizing the simulation results To analyze simulation results one uses the same tools as for analyzing experimental data i e programs such as IDL Matlab and Scilab The output files from simulations are usually simple text files containing headers and data blocks If data blocks are empty they may be accessed referring to an external file indicated in the header This file may also be a binary file except with the original McStas PGPLOT format which does not contain any header except if simulation is launched with a option but are in turn smaller in size and faster to import Each data file contains informations about the simulation the instrum
177. t without changes However optional parameters should not be used in cases where no general natural default value exists For example the length of a guide or the size of a slit should not be given default values This would prevent the error messages that should be given in the common case of a user forgetting to set an important parameter 6 5 2 The DECLARE section DECLARE ht C code declarations variables definitions functions These are usually OUTPUT parameters to avoid name conflicts hy This gives C declarations of global variables functions etc that are used by the com ponent code This may for instance be used to declare a neutron counter for a detector component This section is optional Note that any variables declared in a DECLARE section are global Thus a name conflict may occur if two instances of a component are used in the same instrument To avoid this variables declared in the DECLARE section should be OUTPUT parameters of the component because McStas will then rename variables to avoid conflicts For example a simple detector might be defined as follows DEFINE COMPONENT Detector OUTPUT PARAMETERS counts DECLARE ht int counts hy The idea is that the counts variable counts the number of neutrons detected In the instrument definition the counts parameter may be referenced using the MC_GETPAR C macro as in the following example instrument fragment COMPONENT di Detector COMPON
178. tance A named Component of a given Type inserted in an In strument description e Definition parameter An input parameter for a component For example the radius of a sample component or the divergence of a collimator e Input parameter For a component either a definition parameter or a setting parameter These parameters are supplied by the user to define the characteristics of the particular instance of the component definition For an instrument a parameter that can be changed at simulation run time e Instrument An assembly of McStas components defining a neutron spectrometer e Kernel The McStas language definition and the associated compiler e McStas Monte Carlo Simulation of Triple Axis Spectrometers the name of this package Pronounciation ranges from mez tas to mac stas and m c stas e Output parameter An output parameter for a component For example the counts in a monitor An output parameter may be accessed from the instrument in which the component is used using MC_GETPAR e Run time C code contained in the files mcstas r c and mcstas r h included in the McStas distribution that declare functions and variables used by the generated simulations DTU 127 e Setting parameter Similar to a definition parameter but with the restriction that the value of the parameter must be a number 128 DTU Bibliography 10 11 12 13 14 15 16 17 18 19 K Lefmann and K Nielsen Neutron News
179. te Carlo method has been used widely for e g guide stud ies 17 19 instrument optimization and design 20 21 Most of the time the conclusions and general behaviour of such studies may be obtained using the classical analytical ap proaches but accurate estimates for the flux the resolutions and generally the optimum parameter set benefit advantageously from MC methods Recently the concept of virtual experiments i e full simulations of a complete neutron experiment has been suggested asa major assett for neutron ray tracing simulations The goal is that simulations should be of benefit to not only instrument builders but also to users for training experiment planning diagnostics and data analysis In the late 90 ies at Riso National Laboratory simulation tools were urgently needed not only to better utilize existing instruments e g RITA 1 and RITA 2 22 24 but also to plan completely new instruments for new sources e g the Spallation Neutron Source SNS 25 and the planned European Spallation Source ESS 5 Writing programs in C or Fortran for each of the different cases involves a huge effort with debugging presenting particularly difficult problems A higher level tool specially designed for simulating neutron instruments was needed As there was no existing simulation software that would fulfil our needs the McStas project was initiated In addition the ILL required an efficient and general simulation package in ord
180. the absolute frame usually ROT_A_CURRENT_COMP e randvec_target_rect amp vz amp vy amp vz amp dQ aimyz aimy aim height width Rot is the same as randvec_target_rect_angular but height and width dimensions are given in meters This function is useful to e g target at a guide entry window or analyzer blade B 2 Reading a data file into a vector matrix Table input read_table lib The read_table lib provides functionalities for reading text and binary data files To use this library add a Zinclude read_table lib in your component definition DE CLARE or SHARE section Tables are structures of type t_Table see read_table lib h file for details t_Table structure most important members double data Use Table_Index Table i j to extract i j element long rows number of rows long columns number of columns char header the header with comments char filename file name or title double min_x minimum value of 1st column vector double max_x maximum value of 1st column vector Available functions to read a single vector matrix are e Table Init amp Table rows columns returns an allocated Table structure Use rows columns 0 not to allocate memory and return an empty table Calls to Table_Init are optional since initialization is being performed by other functions already e Table_Read amp Table filename block reads numerical block number block 0 to
181. timate of the relative error as 1 V N However the exact expression depends on the random distributions DTU 29 Records Accurarcy 10 10 104 2 5 10 1 10 0 25 107 0 05 Table 4 1 Accuracy estimate as a function of the number of statistical events used to estimate an integral with McStas McStas uses a space with d 10 parameters to describe neutrons position velocity spin time We show in Table 4 1 a rough estimate of the accurarcy on integrals as a function of the number of records reaching the integration point This stands both for integrated flux as well as for histogram bins for which the number of events per bin should be used for N 30 DTU Chapter 5 Running McStas DESCRIBE NEW Python TOOLS This chapter describes usage of the McStas simulation package Refer to Chapter 3 for installation instructions In case of problems regarding installation or usage the McStas mailing list 2 or the authors should be contacted Important note for Windows users It is a known problem that some of the McStas tools do not support filenames directories with spaces We are working on a more general approach to this problem which will hopefully be solved in a further release We recommend to use ActiveState Perl 5 10 Note that as of McStas 1 10 all needed support tools for Windows are bundled with McStas in a single installer file Performing a simulation using McStas can be divided into t
182. tion By default McStas simulations write their results into several data files in the current directory overwriting any previous files stored there The dir dir or ddir option causes the files to be placed instead in a newly created directory dir to prevent overwriting previous results an error message is given if the directory already exists Alternatively all output may be written to a single file file using the file file or f file option which should probably be avoided when saving in binary format see below If the file is given as NULL the file name is automatically built from the instrument name and a time stamp The default file name is mcstas followed by appropriate extension The complete list of options and arguments accepted by McStas simulations appears in Tables 5 1 and 5 2 DTU 39 5 3 1 Choosing an output data file format Data files contain header lines with information about the simulation from which they originate In case the data must be analyzed with programs that cannot read files with such headers they may be turned off using the data only or a option The format of the output files from McStas simulations is described in more detail in section 5 5 It may be chosen either with format FORMAT for each simulation or globally by setting the MCSTAS_FORMAT environment variable The available format list is obtained using the name out help option and shown in Table 5 3 McStas can presently generate many for
183. tion are not too strong The situation of McStas is ideal since each neutron ray can be simulated without interfering with other simulated neutron rays Therefore each neutron ray can be simulated independently on a set of computers When computing N neutron rays with p computers each computer will simulate neutrons As a result there will be p x N neutrons simulated As a result McStas generates two kinds of data sets e intensity measurements internally represented by three values po p1 p2 where po p p2 are additive Therefore the final value of po is the sum of all local value of po computed on each node The same rule applies for p and p2 The evaluation of the intensity errors is performed using the final po p and pz arrays see Section 4 2 1 e event lists the merge of events is done by concatenation McStas provides three methods in order to distribute computations on many comput ers e when using a set of nodes grid cluster or multi cores it is possible to distribute simulations on a list of computers and multi core machines see section 5 6 1 Re sults are automatically merged after completion This method is very efficient and only requires SSH server to be installed configured on slave machines In order to use an heterogeneous system a C compiler should be optionally installed on slave machines e when using an homogeneous computer cluster each simulation including scan steps may be computed in
184. tomize it a little There are different ways to extend an existing component 6 6 1 Extending from the instrument definition If you only want to add something on top of the component existing behaviour the simplest is to work from the instrument definition TRACE section using the EXTEND modifier see section 6 4 2 You do not need to write a new component definition but only add a piece of code to execute 6 6 2 Component heritage and duplication There is a heritage mechanism to create children of existing components These are exact duplicates of the parent component but one may override extend original definitions of any section The syntax for a full component child is DEFINE COMPONENT child_name COPY parent_name This single line will copy all parts of the parent into the child except for the documentation header As for normal component definitions you may add other parameters DECLARE TRACE sections Each of them will replace or extend be concatenated to with the COPY EXTEND keywords see example below the corresponding parent definition In practice you could copy a component and only rewrite some of it as in the following example DEFINE COMPONENT child_name COPY parent_name SETTING PARAMETERS newparl newpar2 INITIALIZE COPY parent_name EXTEND At C code to be concatenated to the parent_name INITIALIZE 3 SAVE At C code to replace the parent_name SAVE At where two additional parameters
185. trometer The width of the exit channel from the monochromator may be narrowed down from initially 40 mm to 20 mm by an insert piece In the simulations we have chosen the 20 mm option and modeled the channel with two slits to match the experimental set up In the test experiments we used two standard samples An AlgO3 powder sample and a vanadium sample The instrument definitions use either of these samples of the correct size Both samples are chosen to focus on the opening aperture of collimator 2 the one between the sample and the analyser Two slits one before and one after the sample are in the instrument definition set to the opening values which were used in the experiments The analyser of the spectrometer is flat and made from pyrolytic graphite It is placed between an entry and an exit channel the latter leading to a single detector All this has been copied into the instrument definition On the spectrometer Soller collimators may be inserted at three positions Between monochromator and sample between sample and analyser and between analyser and detector In our instrument definition we have used 30 28 and 67 collimators on these three positions respectively An illustration of the TAS1 instrument is shown in figure 8 3 Test results and data from the real spectrometer are shown in Appendix 8 3 1 110 DTU monochromator Analyser 3 Detector Focusing Source Collimator Figure
186. ts and Library 2 3 1 Components 2 3 2 New example instruments and Data files oa a a a aaa Documentation aooaa a Tools installation ooo aaa 2 5 1 New tool features 2 5 2 Platform support 2030 VARIOUS 43 8 eis ok ae MR eG 2 5 4 Warnings Future extensions 3 Installing McStas 3 1 3 2 3 3 3 4 3 5 DTU Getting Mestas hini ir a E aE a i E E Licensing heei 6k ee A Installation on windows Installation on Unix systems 3 4 1 Debian class systems 3 4 2 Other Linux Unix systems 3 4 3 Configuration and installation 3 4 4 Specifying non standard options Finishing and Testing the McStas distribution 10 11 12 14 15 15 15 16 16 16 16 17 17 17 17 18 18 19 19 20 20 20 21 21 22 22 23 4 Monte Carlo Techniques and simulation strategy 25 4 1 Neutron spectrometer simulations 2 0 0 0 a a 25 4 1 1 Monte Carlo ray tracing simulations 25 4 2 The neutron weight 0 0 0 0 0 eee ee 26 4 2 1 Statistical errors of non integer counts 4 26 4 3 Weight factor transformations during a Monte Carlo choice 28 4 3 1 Direction focusing 0 0 000002 ee ee eee 28 4 4 Adaptive and Stratified sampling o oo a a 28 4 5 Accuracy of Monte Carlo simulations o soo a a 29 5 Running McStas 31 5 1 Brief introducti
187. ts polarisation vector Read neutron state parameters from a VITESS neutron file Write neutron state parameters to a VITESS neutron file Table 7 5 Miscellaneous components of the McStas library MCSTAS contrib Description Al_window Collimator_ROC Exact_radial_coll FermiChopper_ILL Filter_graphite Filter_powder Guide_curved Guide_honeycomb Guide_tapering He3_cell ISIS_moderator Monochromator_2foc multi_pipe PSD_Detector PSD_monitor_rad SiC SNS_source Source_multi_surfaces Virtual_mcnp_input Virtual_mcnp_output Virtual_tripoli4_input Virtual _tripoli4_output Aluminium transmission window Radial Oscillating Collimator ROC Radial collimator Fermi Chopper with rotating frame and SM coating straight Pyrolytic graphite filter analytical model Box shaped powder filter based on Single crystal unstable Non focusing continuous curved guide shown curved Neutron guide with gravity and honeycomb geometry Can be channeled and or focusing Rectangular tapered guide parabolic elliptic sections Polarised 3He cell ISIS Target 1 and 2 moderator models based on MCNP calcu lations Doubly bent monochromator with multiple slabs a multi pipe slit for SANS Realistic detector model with gas effects A banana PSD monitor SiC multilayer sample for reflectivity simulations SNS moderator models based on MCNP calculations An array of sources described from spectrum table
188. ument 5 2 2 Specifying the location of files The McStas compiler needs to be able to find various files during compilation some explicitly requested by the user such as component definitions and files referenced by include and some used internally to generate the simulation executable McStas looks for these files in three places first in the current directory then in a list of directories given by the user and finally in a special McStas directory Usually the user will not need to worry about this as McStas will automatically find the required files But if users build their own component library in a separate directory or if McStas is installed in an unusual way it will be necessary to tell the compiler where to look for the files The location of the special McStas directory is set when McStas is compiled It de faults to usr local lib mcstas on Unix like systems and C mcstas lib on Windows systems but it can be changed to something else see section 3 for details The location can be overridden by setting the environment variable MCSTAS setenv MCSTAS home joe mcstas for csh tcsh users or export MCSTAS home joe mcstas 36 DTU for bash Bourne shell users For Windows Users you should define the MCSTAS from the menu Start Settings Control Panel System Advanced Environment Variables by creat ing MCSTAS with the value C mcstas lib To make McStas search additional directories for component definitions and include f
189. usen D F McMorrow K Lefmann G Aeppli T E Mason A Schr der M Issikii M Nohara and H Takagi Physica B 241 243 50 55 1998 K Lefmann D F McMorrow H M R nnow K Nielsen K N Clausen B Lake and G Aeppli Physica B 283 343 354 2000 See http www sns gov H Maier Leibnitz and T Springer Reactor Sci Technol 17 217 1963 D F R Mildner Nucl Instr Meth A290 189 1990 R A Lowde J Nucl Energy Part A Reactor Sci 11 69 1960 J R D Copley Nucl Instr Meth A510 318 2003 E Fermi J Marshall and L Marshall Phys Rev 72 193 1947 J Peters Nucl Instr Meth A540 419 2005 C D Clark E W J Mitchell D W Palmer and I H Wilson J Sci Instrum 43 1 1966 A K Freund Nucl Instr Meth 213 495 1983 V F Sears Acta Cryst A53 35 1997 G Shirane S M Shapiro and J M Tranquada Neutron Scattering with a Triple Axis Spectrometer Cambridge University Press 2002 L Alianelli J Appl Cryst 37 732 2004 V Radeka IEEE Trans Nucl Sci NS 21 51 1974 V Peskov G Charpak W Dominik and F Sauli Nucl Instr and Meth A277 547 1989 G Manzin B Guerard F A F Fraga and L M S Margato Nucl Instr Meth A535 102 2004 L D Cussen J Appl Cryst 36 1204 2003 F James Rep Prog Phys 43 1145 1980 J R D Copley P Verkerk A A van Well and H Fredrikze Comput Phys Commun 40 337 1986 DTU 43
190. utrons are emitted no need to simulate neutrons that we know for sure will not reach the rest of instrument Other options for smart targeting are available refer to component documentation for info COMPONENT target V_sample thickness 0 004 radius 0 012 yheight 0 015 focus_r 0 pack 1 target_x 0 target_y 0 target_z 1 AT 0 0 1 RELATIVE arm Here a secondary arm or reference point placed on the sample position The ROT parameter above defines rotation of this arm and components relative to the arm COMPONENT arm2 Arm AT 0 0 0 RELATIVE target ROTATED 0 ROT O relative arm For data output let us place a detector This detector is not very realistic since it has a spher ical shape with a 10 m radius but has the advantage that EVERYTHING emitted from the sample will be picked up Notice that this component changes orientation with the ROT input parameter of the instrument COMPONENT PSD_4pi PSD_monitor_4PI radius 10 nx 101 ny 51 filename vanadium psd AT 0 0 0 RELATIVE arm2 END 6 4 Writing instrument definitions complex arrangements and syntax In this section we describe some additional ways to build instruments using groups code extension conditions loops and duplication of components As asummary the nearly complete grammar definition for component instances within the i
191. will display the integral p of the monitor bins as well as the 2 moment o and the number of statistical events N This will result in a line such as Detector CompNamel p CompNameERR o CompName_N N filename For 1D 2D and 3D monitors detectors the data histogram store in the files is given per bin when the signal is the neutron intensity i e most of the cases Most monitors define binning for an n axis value as the sum of events falling into the n n 1 range i e the bins are not centered but left aligned Using the Monitor nD component it is possible to monitor other signals using the signal variable_name in the options parameter refer to that component documentation DTU 91 Single detectors monitors The results of a single detector monitor are written using the following macro DETECTOR_OUT_OD t N p p2 Here is a string giving a short descriptive title for the results e g Single monitor One dimensional detectors monitors The results of a one dimensional detector monitor are written using the following macro DETECTOR_OUT_1D t zlabel ylabel tvar tmin Tmax M amp N 0 amp pl0 amp p2 0 filename Here e is a string giving a descriptive title e g Energy monitor e zlabel is a string giving a descriptive label for the X axis in a plot e g Energy meV e ylabel is a string giving a descriptive label for the Y axis of a plot e g Intensity

User and programmers guide to the neutron ray-tracing

Contents

Download Pdf Manuals

Related Search

Related Contents