Roger V. SERVRANCKX
(*) Work sponsored by the Department Of Energy and by the National Science and Engineering Research Council of Canada
The program DIMAD studies particle behaviour in circular machines and in beam lines.
The trajectories of the particles are computed according to the second order matrix formalism(1). It does not provide synchrotron motion analysis but can simulate it. The program provides the user with the possibility of defining arbitrary elements to tailor the program to specific uses. The present version of DIMAD is not fully debugged. Please inform one of the following persons about any anomalies observed:
David Douglas at CEBAF 804/249 7512
Lindsay Schachinger at LBL 415 486 6590
Roger Servranckx at Saskatoon 306 966 6054
In 1970 the first author had the good fortune of discovering the program OSECO (Optique du SECond Ordre) written by J.L.LACLARE at SACLAY. Basically OSECO was a second order Tracking program. It was based on the second order matrix formalism of TRANSPORT and was originally written for a CDC computer. Its usefulness in the simulation of the extraction procedure of the Beam Stretcher ALIS and later of EROS led to the desire for a program that would have more analysis power. The first attempt to develop a new program resulted in the program DEPART which was written as a pure differential equation ray tracing program but it soon became clear that DEPART was very awkward to use because of the cumbersome way in which bending magnets were defined in the code. An evolutionary process then took place over a period of several years finally resulting in the present program called DIMAD.
Many people contributed in various ways to the development of DIMAT. Dr Leon Katz, while he was Director of the Linear Accelerator Laboratory at Saskatoon, provided strong support for the work. Sheila Flory, Dean Jones, Edward Pokraka, Jim Morrison, and Jean Mary Miketinac provided programming support at different times during the initial program development. Ideas were borrowed freely from the program OSECO and Jean Louis Laclare helped formulate some of the early developments. Karl L Brown of SLAC became influential during the later development phases. He helped formulate the more recent contributions to the program (geometric aberrations, linear analysis of motion around arbitrary reference orbits, and magnet misalignment simulations). It is for these reasons that he has become a coauthor of the present manual.
The authors wish to thank the many DIMAT users of other laboratories for their comments and assistance in locating the many programming errors that have occurred during the evolution of DIMAD.
In 1984, it became clear that tracking codes should operate in a canonical environment ,should provide options for symplectic tracking and should conform to the input STANDARD(2).
Adapting the input code of MAD(3), Lindsay Schachinger transformed the program so it would enjoy a common input with MAD, thereby conforming to the input STANDARD.
With ideas developed originally by E.Forest(4), David Douglas introduced the symplectic tracking options and the canonical variables.
One exception to the standard format is the units conventions.
Input to dimad can be in either transport units (indicated by the keyword "utransport"), or in standard units (indicated by the keyword "ustandard"). For more information on units, see the next section. The second difference between dimad and the standard is the addition of several keywords for dimad. The added keywords are "quadsext", "gkick", and "mtwiss." These elements are described more fully later. Also, in dimad, the solenoid can have a quadrupole field. Elements which are described in references 2, but which are not implemented in dimad are separator and rbend.
The job title entered on a line following one with the keyword "title." This should be followed with a units keyword. If no units keyword is found, the units are assumed to be the standard units.
A ";" is used to separate statements on the same line. Keywords are uniquely specified by the first four letters, and only those four must be entered. At most 8 letters in a keyword are checked.
The keyword NOECHO can be used to suppress transmission of the input data stream to the output files. The keyword ECHO reinstates the stream of the input data to the output files.
pname = valuewhere pname is any parameter name. Parameters can then be used in element definitions. Value can also be an arithmetic expression involving other parameters. Throughout the element definitions, a parameter value can also be an arithmetic expression. Note that in dimad the relationships between parameters are lost, but during the machine definition phase they are treated correctly.
Examples:
lslot=100
lb=lslot/8
lh = sqrt(lslot)
Note: the value of PI if needed must be defined as an input
parameter. the value halfturn must be understood as either PI
radians or as 180 degrees depending on the units chosen.
label: type [,pkeyw=value,......]where label is the name of the element, type is an element type (see below) and pkeyw is a parameter keyword appropriate for the element type (see below). value again can be a parameter name, or an expression involving parameters. Examples:
b: sbend, l=lb, angle=lb/rho
d0: drift
drift
l is the length.
sbend
l is the length.
angle is the bend angle.
k1 If the standard convention is being used, k1 is given
by the field expansion in the UNITS section. If
transport conventions are being used, k1 is n =
-rho(dBo/dx)/Bo.
e1 is the entrance edge angle.
e2 is the exit edge angle.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/2 is assumed.
k2 If the standard convention is being used, k2 is given
by the field expansion in the UNITS section. If
transport conventions are being used, k2 is
beta=rho**2(d2Bo/dx2)/(2Bo).
h1 is the entrance pole face curvature.
h2 is the exit pole face curvature.
hgap is the entrance half gap size. If hgapx is not given
a value, it defaults to the value of hgap. During
fitting, however, both hgap and hgapx must be varied
together.
fint is the entrance fringe field integral, which defaults
to 0.5. If fintx is not given a value, it defaults
to the value of fint. During fitting, however, both
fint and fintx must be varied together.
hgapx is the exit half gap size. See hgap.
fintx is the exit fringe field integral.
See fint.
rbend
The rbend is a parallel faced dipole magnet. Its parameters are the
same as those of the sbend. Parameters e1 and e2 are not provided by
the user and are set by the program to half the value of the bend
angle.
quadrupole
l is the length.
k1 is the strength.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/4 is assumed.
aperture is the magnet aperture for the Hardware operation.
sextupole
l is the length.
k2 is the strength.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/6 is assumed.
aperture is the magnet aperture for the Hardware operation.
quadsext
l is the length.
k1 is the quadrupole strength.
k2 is the sextupole strength.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/4 is assumed.
aperture is the magnet aperture for the Hardware operation.
octupole
l is the strength.
k3 is the strength.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/8 is assumed.
aperture is the magnet aperture for the Hardware operation.
multipole
l is the length. If the length is zero, the strengths
are interpreted as integrated strengths.
k0 - k20 are the strengths.
t0 - t20 are the tilt angles. If tn is entered without a
value, halfturn/2(n+1) is assumed.
scalefac is a dimensionless strength factor, used to scale all
the strengths together.
tilt is the overall tilt angle.
aperture is the magnet aperture for the Hardware operation.
NOTE1 : only the components with non zero amplitude are stored! If
zero components need be kept for the purpose of
generating errors via the ERROR definition then enter
components with small amplitudes.
NOTE2 : when a quadrupole or a sextupole component is present the matrix of this component is computed for half the length of the multipole. This does not change the value of the total second order matrix. During tracking operations the particles are tracked through half the element as quadrupole or sextupole then the higher order multipole kicks are applied and the particles are tracked through the second half of the quadrupole or sextupole component. This feature is important in computing misalignment effects with multipole components present.
solenoid
l is the length.
ks is the solenoid strength. ks=0.5*Bs/Brho.
k1 is the quadrupole strength.
tilt is the tilt angle. If tilt is entered with no value,
halfturn/4 is assumed.
aperture is the magnet aperture for the Hardware operation.
rfcavity
l is the length.
volt is the cavity voltage.(kV for Utransport)
lag is the phase lag of the cavity with respect to a
nominal particle (0,0,0,0,0,0) at the start of the
machine.(Degrees for Utransport)
freq is the frequency of the cavity.(Hz for Utransport) In
program versions dated July 14 1989 or later the
frequency is defined via the harmonic number. This
has the advantage to match the frequency with full
accuracy to the length of the machine. When the
frequency must be not matched a non integer value for
the harmonic number is entered. Note that the keyword
has remained FREQ (until further notice). The
particle type (electrons or protons) is selected via
the constant definition operation.
energy is the energy.(GeV)
roll
This element performs a rotation of the coordinate system about the
longitudinal axis.
angle is the rotation angle. A positive angle means the new
coordinate system is rotated clockwise about the s-
axis with respect to the old system.
zrot
This element performs a rotation of the coordinate system about the
vertical axis. The angle must be small.
angle is the rotation angle. A positive angle means the new
coordinate system is rotated clockwise about the
local z-axis with respect to the old system.
hkick, vkick
These elements are translated by the program into general kicks
(gkick).
kick a horizontal (vertical) kick of size kick,
angle about the longitudinal axis.
gkick
This element is a general kick.
l is the length.
dx is the change in x.
dxp is the change in x'.
dy is the change in y.
dyp is the change in y'.
dl is the change in path length.
dp is the change in dp/p.
angle is the angle through which the coordinates are
rotated about the longitudinal axis.
dz is the longitudinal displacement.
v is the extrance-exit parameter of the kick. v is
positive for an entrance kick, and negative for an
exit kick. The absolute value of v is used to force
the kick to be applied every abs(v) turns. The
default value of v is 1.
t is the momentum dependence parameter. The kicks dx'
and dy' can be thought of as misalignment errors or
as angle kicks of orbit correctors. In the first case
(t=0) they are momentum independent. When t=1 the
kicks dx' and dy' vary inversely with momentum.
When t is set to a negative integer value -n the kick
is applied every turn and the momentum of a particle
with initial momentum p will oscillate around the
nominal momentum p0 with amplitude (p-p0) and a
period equal to n turns. More than one such kick may
be put in the line (all identical though) the phase
of the cosine oscillation is proportional to the
pathlength of the reference trajectory.
hmon, vmon, monitor
These elements are horizontal, vertical, and horizontal and vertical
monitors, respectively.
l is the monitor length.
xserr,
yserr,
xrerr,
yrerr are the x and y systematic and random errors.
NOTE: the errors are not used in this form presently. Errors are
introduced via the misalignement operations.
marker
A marker is a drift element of zero length. It has no parameters.
ecollimator, rcollimator
An ecollimator is elliptic, and an rcollimator is rectangular. The
particles are checked at the entrance and at the exit of the
collimator.
l is the length.
xsize
ysize are the x and y collimator apertures. The default
apertures are 1 meter.
arbitelm
This is the arbitrary element. Its parameters are used in the user-
supplied routine TRAFCT, which contains the transfer function
describing the effect of arbitrary elements on the individual
particles. All arbitrary elements use the same subroutine. Distinct
arbitrary elements can only be recognized by the program through the
use of one parameter as a flag.
l is the length.
p1 - p20 are the parameters.
mtwiss
l is the length.
mux,
betax,
alphax,
muy,
betay,
alphay are the twiss parameters for this transfer matrix.
betax and betay have default values of 1.
matrix
This element is a general transfer matrix.
rij
tijk are the matrix elements. i,j, and k range from 1 to
6, but j is always less than or equal to k.
label: line=(member1, member2, member3,.....)denotes a beamline called label. The members can be elements, other beamlines, sequences of members, or any of the above preceeded by a repetition count and/or a minus sign for reflection.
Examples are
df:line = (dq, oo, b, oo qf)
fdstar:line = (qf, sf, b, sd, qd)
arc:line = (df, 64*(fdstar,df))
Beamlines can also have formal arguments. An example is
fdstar(sf,sd):line = (qf, sf, b, sd, qd)where sf and sd are not defined elements, but variables. So
super:line = (fdstar(sd1,sf1),df,fdstar(sd2,sf2))is a line in which the elements sd1, sd2, sf1, and sf2 are substituted for the variables sd and sf in the original definition.
use, beamlinename.This causes the beamline beamlinename to be the current machine for dimad.
Next comes the statement
dimatwhich passes control to dimad, after translating the machine into the correct data structures for dimad. Any dimad command can then be issued.
A ';' will cause dimad to stop and return control to the input interface. Now the user can define a new machine and then go back to dimad and do a new calculation, or stop execution with the command stop. The use command causes the old machine to be replaced by a new one. This new machine can be a previously defined beamline. For debugging purposes, the dump command from MAD has been retained. This command produces a dump of the MAD-type data structure describing the machine. After a ";" and return to MAD control, one can specify the use of a new line , keeping the previously defined (and perhaps modified by dimad) elements.To do so one uses the commands :
use,newlinename newbeamThe last command "newbeam" passes control back to dimad. Observe that without newbeam all the element parameters are redefined to their initial input values.
A new MAD command is introduced : EXPLODE . Its purpose is to provide an explicit description of the beamline used.
The lines following the title may have 72 characters.
Any line containing one of the characters "!","*","(","@" in the first column is treated as a comment line.
Each array terminates with a ',' or a ';'. In the first case another operation is expected,in the second case control is returned to the interface program.If the user desires to stop the run at this point, the line following the ';' must contain the MAD command 'STOP'.
In all tracking operations the particle coordinates are checked at the entrance of some element. Particles are lost when the square of the radial excursion is greater than the expulsion factor. It is set at the default value of 1. Its value can be changed via the constant definition operation.
ADIABATIC VARIATIONS
BEAM MATRIX TRACKING
CONSTANT DEFINITION
DETAILED CHROMATIC ANALYSIS
GENERATION OF PARTICLES
GEOMETRIC ABERRATIONS
HARDWARE VALUES LISTING OF MACHINE
INTERACTIVE MANIPULATION OF LATTICE
LEAST SQUARE FIT
LINE GEOMETRIC ABERRATIONS
MACHINE AND BEAM PARAMETERS COMPUTATIONS
MATRIX COMPUTATION
MODIFICATION OF ELEMENT DATA
MOVEMENT ANALYSIS
OUTPUT CONTROL
PARTICLE DISTRIBUTION ANALYSIS
PRINT SELECTION
PROGRAM GENERATION
RMATRIX COMPUTATION (6X6)
SEISMIC PERTURBATION SIMULATION
SET FIT POINT
SET LIMITS TO VARIABLES
SET SYMPLECTIC OPTION ON
SHO VALUES OF CONSTANTS
SIMPLE FITTING
SPACE CHARGE COMPUTATION
TRACKING OF PARTICLES
OPERATIONS ASSOCIATED WITH MISALIGNMENTS AND ERRORS
ALIGNMENT FITTING
BASELINE DEFINITION
BLOCK MISALIGNMENT
CORRECTOR DATA DEFINITION
ERRORS DATA DEFINITION
MISALIGNMENT DATA DEFINITION
REFERENCE ORBIT DISPLAY
SEED
SET CORRECTOR VALUES
SET ERRORS OF ELEMENTS
SET MISALIGNMENT OF ELEMENTS
SHO MISALIGNMENTS
SHO ERRORS
SYNCHROTRON RADIATION DATA DEFINITION
ADIABATIC VARIATIONS
Input format:
ADIAbatic variations of some parameters(up to 80 char)
name pkeyw nopt p1 p2 val1 val2 val3 val4
........
name pkeyw nopt p1 p2 val1 val2 val3 val4
99,
Parameters:
name name of element having a parameter to be varied
pkeyw keyword of parameter to be varied (i.e. k1 for a quad)
nopt option number
1 means variation will be linear according to the
following rule: the parameter remains constant at
value val1 until turn p1 then varies linearly to
achieve the value val2 at turn p2. In this case only
two parameter p's and only two values vali are present
in the input format
2 means the variation will be sinusoidal between turn
p1 and turn p2.The variation is done according to the
formula :
value = val1 + val2*sin((2pi*turn/val3)+val4)
where turn is the current turn at which value is
applied. Outside turns p1 and p2 the original value is
applied.
Input format:
BEAM matrix tracking computations..(up to 80 char)
sigx rxx' rxy rxy' rxl rxp
sigx' rx'y rx'y' rx'l rx'p
sigy ryy' ryl ryp
sigy' ry'l ry'p
sigl rlp
sigp
mprint [list]
or
0
betax alphax etax etapx epsilonx
betay alphay etay etapy epsilony
sigl sigp
mprint [list]
or
0
0 0 0 0 epsilonx
0 0 0 0 epsilony
sigl sigp
mprint [mlist]
Parameters:
sig* sigma extension of the beam.(as defined in reference(1)
and reference (5))
rij correlation cosines as defined in reference (1).
betax,alphax,etax,etapx,betay,alphay,etay,etapy : initial values of
twiss parameters used to define an uncoupled beam.
epsilonx,epsilony emittances in x and y of the input beam.
NOTE: when betax etc are zero the values are obtained from a
previous movement analysis calculation made within a
MATRix operation or a Fit operation that generates a
matrix calculation.
mprint -2 no computation is done. The operation serves only to
define a beam as needed in the operations BEAM tracing
and DETAiled analysis with parameter nvh=1.
-1 print final result only.
0 print all intermediate and final results.
n n>0 used with list. There are n intervals in which
printing will occur.
mprint+1000:when 1000 is added to the value of mprint, the printing
occurs in the same fashion as above but a table of beam
envelopes is printed instead of the full beam matrix.
list contains the beginning and end of all intervals in
which printing is done. List is a set of pairs of
numbers.They are positions in the order list of machine
elements) List may contain up to mxlist numbers (set at
40 initially)
Use the operation SHO Constant to examine the constants.
Input format:
CONStant definition .....(up to 80 Characters)
n(1),val(1),....n(p),val(p)
Parameters :
n(i) is the order number of the ith constant to be redefined
according to the following order : pi,velocity of
light,particle mass,particle radius,particle
charge,reference energy used in taylor series
expansions, least square fit initial tolerance,factor
for maximum function calls in least square fit ,the
expulsion factor and the particle type (0 for electrons
and 1 for protons, the default is 0). Presently when
the particle type is changed the particle mass and
radius is NOT changed. The particle type selection
only affects the rfcavity functions: the particle
velocity is taken into account to compute the phase of
the particle relative to the RF.
val(i) new value of the constant n(i).
For each energy ei,particles are generated around the point Po (xo xo' yo yo') to compute the elements Rij of the matrix describing the linear motion around the trajectory defined by the point Po.
Input format:
DETAiled chromatic analysis...(up to 80 characters)
NH NV NHV
xo xo' yo yo'
dx dx' dy dy'
betax alphax etax etapx
betay alphay etay etapy
Nener Ncoef
e e ...e
1 2 nener
MLOCAT [LIST]
Parameters:
NH 1 only xx' motion is traced and computed. Betax, alphax,
and nux are computed.
0 xx' motion alone is not analyzed.
NV 1 only yy' motion is traced and computed. Betay, alphay,
and nuy are computed.
0 yy' motion alone is not analyzed.
NHV 1 coupled motion xx', yy' is traced. The full beam
matrix is computed.
2 coupled motion is computed.A short print is provided
with x xp y yp betax alphax betay alphay nux nuy for
all energies requested
3 in this case three energies have to be defined :
delta0, delta0 - eps, delta0 + eps . This enables
computing the basic machine parameters associated with
energy delta0. Choose eps comfortably close to zero for
accurate computation of the eta functions. A short
print of the machine parameters is provided. The
values of eta and etap are affected by a scale factor
ETAFAC which can be set via the constant definition
operation. Its default value is 1.0. When set to
1.0e03(say)) the printed values are in mm and mrad.
4 The beam matrix values (as defined in the BEAM
operation) computed for one energy are printed in a
convenient table format. The values of the matrix
values for the beam sigmas (not the correlation
coefficients) are affected by the scale factor SIGFAC
which can be set via the constant defintion operation.
5 same as three with the code added in the first column
to facilitate some plotting work
NOTE: the input beam must have been defined previously in a BEAM
MATRIX TRACKING operation.
0 coupled xx', yy' motion is not analyzed.
When NH, NV, and NHV are all zero, the program prints the
centroid positions only.
dx dx' dy dy'
increments at which the off orbit particles are placed
to compute the sx, cx, sy, cy functions (see reference
(1)).
betax, alphax, etax, etapx, betay, alphay, etay,etapy
initial values used in the twiss function computations.
Nener number of energies for which the analysis is done
(maximum 15).
Ncoef number of coefficients used in the Taylor series
expansion as a function of momentum (max 6).
e ...e momentum values in the form (p - p )/p
1 nener 0 0
MLOCAT indicates the number of intervals in which printing is
to occur. If MLOCAT = -1, then printing occurs at end
of lattice only and no number is in list.
LIST a set of pairs of numbers each of which indicate the
beginning and end position (in the order list of
machine elements) of each of mlocat intervals in which
printing takes place. List may contain up to mxlist
numbers (set at 40 initially)
Input format:
GENEration of particles
nopt sigma1 .... sigma6 scale npart
x0,x'0,y0,y'0,al0,del0
Parameters
nopt : 1 the particles are randomly generated on the surface of
a six dimensional ellipsoid (defined previously by a
BEAM operation. In this case the values of sigmai are
not operational (but for computational efficiency they
should be set to 1)
3 the particles generated have coordinates that satisfy
a six dimensional gaussian distribution.
Note: the same value for the option parameter must be used in the
PARTicle analysis operation (if used) following the
tracking of such particles.
sigmai : the number of sigmas above which the gaussian
distribution is truncated for each of the six variables
x,x',y,y',al,delta. The beam is defined by a previous
BEAM operation which is assumed to define the one sigma
distribution.
scale : scales the beam size by the given factor.
npart : number of particles to be generated. Maximum number mxpart
(initially set at 1000)
x0,x'0,y0,y'0,al0,del0 : centroid coordinates around which the
beam is generated.
Input format:
GEOMetric aberrations .......(up to 80 characters)
betax,alphax,betay,alphay
xco,xpco,yco,ypco,ener
ncase,nturn,njob
nplot,nprint
epsx , epsy
1 1
....................
epsx ,epsy
ncase ncase
anplprt
Parameters:
betax, alphax, betay, alphay
input values of the twiss parameters at the entrance of
the lattice. When betax=0 the twiss parameter values
are obtained from a previously run movement analysis
with nanal not 0. The values corresponding to the first
energy are used. This includes the parameters xco to
ener.
xco, xpco, yco, ypco, ener
coordinates and momentum of the closed orbit around
which the aberrations are to be computed. When betax=0
these parameters are obtained from a previous movement
analysis operation. Values corresponding to the first
energy are used.
ncase number of cases analyzed (maximum 10)
nturn number of turns for tracing (maximum 100)
njob 1 coupled motion analysis is wanted
2 uncoupled motion analysis is wanted
nplot 1 plotting of the resulting particles. The operation
always accumulates the particles at every nplot turns
but plots the accumulation at the end of the job. It
also computes its own plotting windows.
-1 no plotting.
nprint -2 no printing.
-1 printing at end of lattice only.
0 printing after every element.
n printing after every n turns. Normally nprint should
be set = nturn.
epsx, epsy ncase values for the chosen nominal
i i emittances in x and y using the unit
mm-mrad (E-06 m-rad)
anplprt parameter selecting the fast fourier transform options.
When 0 no fourier transform is performed. 1 the fourier
transform components are printed. 10 the amplitude of
the fourier transform is printer-plotted. 100 an
analysis of the peaks is provided. A combination of
those values is allowed eg: 111 all three are done.
It is advised to trace for at least 500
turns,preferably 1000.The number of turns should have
as many low valued factors as possible to benefit from
the speed of the fast fourier transform.
Only the first case of the geometric aberration run is
fourier analysed.
NOTE: presently this operation works only with Transport units. The run must have started with the command UTRANSPORT.
Input format:
HARDware layout and element parameters..(up to 80 char)
E s x y z theta phi psi conv mprint [list]
Parameters:
E momentum (GeV/c) used for computation of field values.
s x y z coordinates of starting point in some absolute
reference coordinate system. The coordinate s is the
length of arc along the reference trajectory.To justify
the choice of the angles theta, phi, and psi, the axis
z should coincide more or less with the longitudinal
axis of the beam. The angles theta, phi, and psi
describe the motion needed to bring the absolute system
of reference in coincidence with the local system of
coordinates. The local system of coordinates is the
system used by the program. Its z axis is tangent to
the reference trajectory. The x axis (uniquely defined
by the bends) is in the midplane of symmetry and points
outwards of the bend. The y axis completes the local
right handed system of reference. To bring the
absolute system in coincidence with the local reference
system, one executes the following rotations (strictly
in the order indicated):
A rotation theta around the y axis (positive when
the z axis turns towards the x axis)
A rotation phi around the x axis (positive when the
z axis turns towards the y axis: i.e. points upwards
for a bend deflecting the beam to the right)
A rotation psi (called sometimes the roll) around
the z axis (positive when the x axis turns towards
the y axis)
conv conversion factor to enable the printout in various
practical units. For feet, the conversion factor is,
for example 0.3048 (the length of a foot in meters).
The program recognizes yards, feet, inches, cm, mm,
microns. However, any conversion factor is accepted
even if not recognized.
mprint -2 no printing of results.
-1 printing final result only.
0 print all intermediate and final results.
n n>0 used with list, there are n intervals in which
printing will occur.
list contains the beginning and the end of all intervals in
which printing is done. List is a set of pairs of
numbers. List may contain up to mxlist numbers (set at
40 initially)
Input format:
INTEractive control of ..(up to 80 char)
niopt nivar
name keyword (repeated nivar times)
Parameters:
niopt option parameter not used now .
nivar number of parameters to be varied (maximum 8)
At run time follow the instructions of the program. This CANNOT work
if at implementation of the program the output channel 9 has NOT
been assigned to the terminal.
Input format:
LEASt square fit of .....(up to 80 char)
nstep nit nvar ncond
betax,alphax,etax,etapx,betay,alphay,etay,etapy
name(i) pkeyw(i) del(i) for i = 1 to nvar
nval(j) valf(j) weight(j) for j = 1 to ncond
nasp
repeat the following nasp times
name1 npas
name(k) pkeyw(k) coef(k) for k = 1 to npas
Parameters:
nstep number of steps taken to approach final fit.
nit number of iterations used in final step of fit.
nvar number of independent variables(max:20).
ncond number of conditions to be met(max:20).
betax...etapy
initial values needed for the function computation.When
betax value is entered as zero, then the program uses
the betax...etapy values computed in the last matrix
operation preceding the present operation.
name(i) name of element with an independent parameter to be
varied.
pkeyw(i) variable element parameter keyword
del(i) this parameter is not used in the new minimizer
implementation, but was kept in the input to avoid a
major change in the input format.
nval(j): reference number of output value to be fitted. Numbers
1 to 20 are for the values of the stable motion
analysis of the total matrix in the same order as
mentioned in paragraph 2.8. of the SIMP operation.
Numbers 21 to 30 refer to betax alphax etax etapx nux
betay alphay etay etapy nuy at the end of the machine.
These values are computed from the initial values
present in the second line of the input format.
Numbers 31 to 40 refer to the values betax nuy computed
at the first fit point defined by the preceding SET Fit
point operation.
Numbers 1031 to 1040 refer to the difference between
the values betax nuy computed at the first and second
fit point defined by the preceding SET Fit point
operation.(ie.:v2-v1)
Numbers 41 to 61 refer to the beam values sigx ...
sigp and the rij at the end of the machine. See
operation BEAM for the meaning of these parameters and
the order in which they appear. These values can only
be fitted if a BEAM operation defining the beam values
at the begining of the machine has preceded the fitting
operation.
Numbers 71 to 91 refer to the same beam values computed
at the first fit point defined by the operation SET Fit
point.
Numbers 1071 to 1091 refer to the differences of the
same beam values computed at the first and second fit
point defined by the operation SET Fit
point.(ie.:v2-v1)
Numbers 93 to 98 fit the average chromatic errors for
betax, alphax, betay, alphay nux, nuy as computed in
the detailed chromatic analysis operation. A fit on
these elements can only be done after a previous
detailed chromatic analysis operation is done which
serves to define the parameters needed for the
computation.
Selected numbers 110 to 666 specify matrix elements in
the following fashion:
ij0 represents the first order matrix element R(i,j).
ijk represents the second order matrix element
T(i,j,k) (as in the TRANSPORT notation).
valf(j) value to be achieved.
weight(j) weight attached to the value(j) in the fit function.
nasp number of associated parameters. If nasp = 0, then the
following data is not to be entered.
name1 name of the basic parameter to which the associated
parameters are connected. It must be present in the
list of basic parameters.
npas number of parameters to be associated to name1 (max:6).
name(k) name of one element having a parameter associated to
name1.
pkeyw(k) keyword of the parameter of name(k) associated with
name1.
coef(k) coefficient with which the BASE parameter (that of
name1) is to be multiplied to obtain the value of the
parameter of name(k).
Input format:
LINEgeometric aberrations....(up to 80 characters)
betax,alphax,betay,alphay
xco,xpco,yco,ypco,ener
ncase,npart,ncoup
nplot,nprint,mlocat,[list]
epsx , epsy
1 1
....................
epsx ,epsy
ncase ncase
Parameters:
betax,alphax,betay,alphay
input values of the twiss parameters at the entrance of
the line.
xco,xpco,yco,ypco,ener
coordinates and the momentum of the trajectory around
which the aberrations are to be computed.
ncase number of cases analyzed(maximum 10).
npart number of particles to be traced(maximum 300).
ncoup not used presently, but a value must be inserted.
nplot 1 plot the resulting particles at the end of the job. It
computes its own plotting windows.
-1 no plotting.
nprint -2 no printing.
-1 printing at end of the line only.
0 printing after every element
mlocat number of intervals in which printing is to occur; used
in conjunction with list. This is not yet implemented.
list intervals in which printing occurs. List may contain
up to mxlist numbers (set at 40 initially)
epsx , epsy ncase values for the chosen nominal
i i emittances in x and y using the unit
mm-mrad (E-06 m-rad)
Input Format:
MACHine and beam parameters....(up to 80 char)
E1 E2 dE NLUM DNU NINT NBUNCH
betax alphax etax etapx
betay alphay etay etapy MPRINT (LIST)
Note: if E1 is zero then nlum is assumed 0 and the input twiss
parameters values are those obtained in a previous matrix analysis.
If E1 is non zero but betax is zero, the first line of parameters must
be given and the initial twiss parameters values will be those of the
preceding matrix analysis.
Parameters:
E1 start momentum for beam data and luminosity
computations.
E2 end momentum
dE momentum step
NLUM 0 no beam size related computations are done.
1 synchrotron integrals and basic beam size computations
are done.
2 Full luminosity computations are made.
DNU dnu value used for optimum luminosity computation.
NINT number of interaction regions
NBUNCH number of bunches
betax...etapy
function values at starting point of lattice.
mprint -2 no printing of results
-1 print final result only.
0 print all intermediary and final results.
n n>0 is used with list. There are n intervals in which
printing will occur.
list beginning and end positions of each interval in which
printing is done. List is a set of pairs of numbers.
List may contain up to mxlist numbers (set at 40
initially)
MATRix computations...........(up to 80 char)
NORDER MPRINT [LIST]
Parameters:
NORDER 1 first order matrix only is printed.
2 second order terms are also printed.
<0 computation is done to order abs(NORDER). MPRINT must
be >0 and the program will print matrices of the beam
line situated between and including the element pairs
defined in LIST.
When norder is -1 or -2 the format of the output is
identical to the input format of the Gxxxxxxx element.
When norder is -11 or -12 th computation is to order 1
and 2 and the format of the output the standard program
output for matrices
MPRINT -2 no printing of matrix.
-1 print matrix at end of machine only.
0 print all intermediary matrices plus final matrix.
n where n>0, used with LIST and indicates the number of
intervals in which printing is to occur.
LIST set of pairs of numbers which indicate the beginning
and end position (in the order list of machine
elements) of each of mlocat intervals in which printing
takes place. List may contain up to mxlist numbers
(set at 40 initially)
Input Format:
MODIfication of input parameters...(up to 80 char)
n
name pkeyw value
name pkeyw value
. . .
name pkeyw value
Parameters:
n number of varies to be made (= number of 'name pkeyw
value' entries which follow).
name name of the machine element which is to be varied.
pkeyw keyword of the parameter in the given element which is
to be changed.
value value the parameter is to be changed to.
Input Format:
MOVEment analysis ..... (up to 80 char)
nprint nturn nanal nit nener ncoef dist
x x' y y' l e .....e
1 nener
naplt delmin delmax dnumin dnumax dbmin dbmax ncol nline
Parameters:
nprint print action for the closed orbit information
0 action after every element
n action after every n turn
nturn number of turns over which analysis is performed. It
enables user to study higher order resonances.
nanal 0 no stability analysis is done, only the closed orbit is
computed.
1 stability analysis is performed (both stable and
unstable).
2 gives information about an order two resonance. (NTURN
must then be equal to 2.)
3 gives information about an order three resonance.
(NTURN must then be equal to 3.)
NOTE: in both cases where NANAL is equal to 2 or 3 the
resonance motion analyzed is supposed to occur in the
horizontal phase plane. If the user wants to study
resonance in the vertical plane, the machine should be
set up so that its planes are exchanged.
In the versions subsequent to April 1 1988, the
coordinates of the particles close to the unstable
fixed point of the first momentum are stored for
subsequent use in a tracking operation. See the demo3
input file for its use
nit number of iterations used. ABS(nit) iterations are
performed. If nit is negative only the results of the
last iteration are printed.
nener number of momenta for which the analysis is performed.
(max 15)
ncoef number of coefficients to be used in the Taylor
expansion of the parameters nu, beta, eta, and etap
versus momentum. The reference momentum in the
expansion is fixed at 0.01(1%). (max 6) The reference
momentum can be changed by the CONStant definition
operation.
dist indicates the 'distance' (in phase space) from the
estimated position of the closed orbit at which the
particles needed for the computation are initially
generated. A safe choice is 0.001 or some lower value
depending on the size of the phase space occupied by
the beam.
x,x',y,y',l estimate of the coordinates of the closed orbit.
e ....e Momenta(dp/p) for which the analysis is
1 nener performed.
naplt 0 no plot of the Taylor expansion is required
1 a plot is required
delmin delmax
min max of dp/p for the plot.
dnumin dnumax
min max for dnu (the first momentum serves as the
reference to compute the tune difference dnu).
dbmin dbmax min max for relative difference in betas.
ncol nline number of columns and lines desired for plot.
Input format:
OUTPut control
nopt
Parameters
nopt 0 all output is supressed (except error messages)
1 The main results of the computation only are printed
2 The main results and the input data are printed
3 All output is printed
4 Used for short printing of tracking results when such
printing can be used for input to plotting programs.
14 Same as 4 . The fifth coordinate will then be the phase
relative to an RF cavity instead of the path length.
Notes : not all output has been affected by this option in the
present version of the program In the operation section
of the input data, this option should only appear
between two operation arrays and not inside one such
array. This operation cannot appear in the Standard
format input section. In this section the command
NOECHO can be used to suppress printout and the NOECHO
can be reversed by use of the command ECHO.
Input format:
PARTicle distribution analysis
nopt
Parameters
nopt must be equal to the parameter chosen in the particle
generation When equal to 1 the values for the beam
sizes are independent of the choice of the scale
parameter of the GENEration operation. This facilitates
comparison between similar beams with different scale
factors(useful in non linearity studies)
Input format:
PRINt selection
keyword
name(i) as many as needed
99
end,
The different keywords are : interval, name, type
Interval : allows to define up to MXLIST intervals (default 40)
The intervals are defined by pairs of names of elements
present in the currently used beamline. The names must
be in ascending order of position and must be unique.
Preferably one shoulduse markers. 99 is the flag that
terminates the sequence of names. 99 is followed by
another keyword or by end,.
name : followed by names of elements at which printing is to
occur. The maximum number is 10. The names may contain
the wild character * . AB* means all names starting
with ab will produce printing.
type : the types are : drift, bend, quadrupole, multipole,
gkick, collimator, rfcavity, sextupole, solenoid,
monitor, quadsext, matrix, mtwiss, arbitrary. Two
types may be defined.
Input format: PROGram generation nopt nint a(1) b(1) ..... a(nint) b(nint) Parameters nopt option number to choose a statement separator 0 no separator character is used 1 ; is the statement separator character (useful for C language) 2 : is the statement separator character. nint the number of intervals in which the generation is to occur. a(i),b(i) the begining and end of the ith interval in which the program simulating tracking of a particle in the segment a(i),b(i) is done
This enables to determine the first order behaviour of beamlines affected by errors and misalignements. The program also provides the entrance and exit orbit displacements. They are needed to correctly use the matrix.
Input format:
RMATrix.......................(up to 80 characters)
x0 xp0 y0 yp0 l0 delta0
dx dxp dy dyp dl ddelta
norder mprint
Parameters:
x0 xp0 y0 yp0 l0 delta0 initial coordinates of reference orbit
When delta0 = 1 then the coordinates are the
coordinates of the closed orbit computed in a previous
movement analysis. The values corresponding to the
first energy are used.
dx dxp dy dyp dl ddelta increments used to generate the six
particles surrounding the reference orbit.
norder order of the computation: 1 or 11 . The order of the
computation is 1. when norder = 11 the output is in the
standard input format.
mprint number of intervals wanted
nlist mprint pairs of numbers defining the intervals for
which the matrix will be computed.
SHO Values of the basic constantsno parameters are used for this operation.
Input format:
SIMPle fitting ........(up to 80 char)
nstep nit nvar
name pkeyw del (i = 1 to nvar)
i i i
nval valf (i = 1 to nvar)
i i
nasp
repeat the following nasp times
name pkeyw npas
name(k) pkeyw(k) mult(k) add(k) k = 1 to npas Parameters:
nstep number of steps to reach the final stage.
nit number of iterations performed in the last step in
order to refine the variable values.
nvar number of variables and conditions (max:10).
name name of element containing a variable.
pkeyw keyword of the parameter to be varied in the element.
del increment by which the variable is to be varied. This
number is divided by 5 in every iteration of the last
step.
nval order number of value to be achieved. The order is
given by the following list: compf nux etax etapx
alphax betax dmux/ddelta chromx dalphax/ddelta
dbetax/ddelta muy nuy etay etapy alphay betay
dmuy/ddelta chromy dalphay/ddelta dbetay/ddelta, where
compf stands for the compaction factor in x. These
values are those computed in the stable motion analysis
of the matrix of the complete machine.Note that the
momentum dependence of eta cannot be fitted
valf the values to be achieved in the final step.
npas total number of parameters to be associated to the
parameter pkeyw of name1.
name(k) name of the element which has a parameter to be
associated with name 1.
pkeyw(k) parameter keyword to be associated.
mult(k),add(k)
multiplicative and additive constants which define the
value of the associated parameter according to the
following formula
parvalue(k) = mult(k)*parval + add(k)
where parval is the value of the parameter used in the
element name1 and to which pkeyw(k) is associated.
This operation only affects the tracking of particles and all operations that use tracking.
Input format:
SEISmic simulation............(up to 80 characters)
xlambs axseis phixs
ylambs ayseis phiys
beginname endname
Parameters:
xlambs,ylambs Wave length(in m) for x and y waves
axseis,ayseis Amplitudes of the x and y waves
phixs,phiys x and y phase shift of each wave.
beginname,endname name of elements where the wave is to start and
where it is to stop. Use unique names, markers
preferably.
Input format:
SET Fit point.................(up to 80 characters)
n Position1 (Position2)
Parameters:
n Number of fit points defined (maximum 2)
Positioni Name (must be unique in machine list) of the element
after which the fitted values are applied. It is
recommended to use a marker for this purpose.
Input format:
SET Limits on variables ......(up to 80 characters)
name keyword upper lower weight distance
........................................
name keyword upper lower weight distance
99,
Parameters:
Name Name of the element for which some parameter will be
affected by boundaries.
Keyword Keyword of the parameter to be affected by the
boundaries.
Upper Lower The upper and lower boundaries affecting the
parameter.
Weight The weight that is applied to the boundary constraint.
Distance A distance parameter that serves to indicate some
flexibility in the boundary constraint. That distance
is progressively reduced as the iterations step
increases.
NOTE: In the present version of the program , this option cannot be followed by any fitting which changes matrices. Any fitting not changing matrices is allowed (eg: in alignment fitting when steering only is involved)
Input format:
SET Symplectic option on......(up to 80 characters)
Option Energy
Parameters:
Option Determines the mode of tracking. This affects only the
operations based on tracking and not those based on
matrix analysis (MATRIX, BEAM MATRIX, MACHINE
FUNCTIONS)
0 non symplectic ray trace is done using the canonical
matrices.
1 Fast version of ray trace is done with the variables
x,x',y,y',al,delta and using the canonical matrices.
2 Fast version of ray trace is done with the variables
x,px,y,py,-tau,DE/E and using the canonical matrices.
3 Slow version of ray trace is done with the variables
x,x',y,y',al,delta and using the canonical matrices.
4 Slow version of ray trace is done with the variables
x,px,y,py,-tau,DE/E and using the canonical matrices.
Note: Options 3 and 4 are not for the general user. They have been
used and maintained for debugging purposes only.
Energy Energy of nominal particle (in GeV)
Input format:
SPACe charge computation...(up to 80 char)
option dpmax dkmax,
Parameters:
option 0 sets the computation off, 1 sets it on.
dpmax The maximum dp/p present in the beam
dkmax The maximum relative change all quadrupole settings that
will produce the required maximum space charged tune
shifts as determined by the theory of linear tuneshift
produced by space charge. The link between dkmax and
the maximum space charge tune shift is generally given
by the natural chromaticity of the machine.
Input format:
TRACking of particles .....(up to 80 char)
NPLOT NPRINT NPART NTURN
particle data ( x x' y y' l dp - for all particles)
MLOCAT LIST NGRAPH XMIN XMAX XPMIN XPMAX YMIN YMAX
YPMIN YPMAX NCOL NLINE (ALMIN ALMAX DELMIN DELMAX)
Parameters:
NPLOT 0 plot action after every element.
-1 no plot action.
n action occurs after n turns (used in conjunction with
MLOCAT and LIST) at MLOCAT locations specified by LIST
elements.
NPRINT 0 print action after every element.
-1 printing at end only.
-2 no printing occurs.
n action occurs after n turns (used in conjunction with
MLOCAT and LIST) at MLOCAT locations specified by LIST
elements.
NOTE: when nplot=-1 and nprint=-2 then mlocat and list
do not appear.MLOCAT and LIST are the same for plot and
print.
NPART number of particles traced.
<0 abs(npart) particles are added to particles already
present from previous operation.
0 existing particles kept - none added.
>0 previously used particles deleted. NPART new particles
introduced.
NTURN number of turns to be traced.
x x' y y' l dp
particle data for NPART particles.
MLOCAT indicates the number of intervals in which printing is
to occur. If MLOCAT is equal to 0, then printing
occurs at end of lattice only. When MLOCAT is 0, no
number is in list.
LIST set of pairs of numbers, each of which indicates the
begining and end position (in the order list of
machine elements) of each of MLOCAT intervals in which
printing takes place. List may contain up to mxlist
numbers (set at 40 initially)
NGRAPH 1 plot x,x' plane
2 plot y,y' plane
3 plot x,y plane
4 plot all planes
11,12,13,14
as above but the graphs are accumulated and the plot is
printed at the end.
15,16
accumulates the al,del or the Phi,del plots where al is
the pathlength coordinate of the particles, Phi is the
phaseshift with respect to the frequency of the
cavities (they must be present for this graph to be
meaningful , the cavities need not be in phase with the
total length of the machine).del is the sixth
coordinate of the particles Note that 15 will present a
correct plot of the longitudinal phase-space only if
the frequency of the cavity and the length of the
machine match perfectly (8 digits usually!!). Using the
value 16 garantees a plot which uses the RF phase
instead of path length differences and is always
readable.
17
is equivalent to 14 as regards the xx',yy' and xy plots
and at the same time will produce an E phi plot
identical to that of produced by the ngraph value of
16.
XMIN,XMAX,XPMIN,XPMAX,YMIN,YMAX,YPMIN,YPMAX
limits for the plotting windows.
ALMIN,ALMAX,DELMIN,DELMAX
limits for the plotting windows for the cases NGRAPH=15
or 16 The are not present for the other values of
NGRAPH For NGRAPH=16 AL is to be interpreted as PHI.
NCOL,NLINE number of columns and number of lines to be used in
plot matrix.
Input format:
ALIGnment fitting .....(maximum 80 characters)
nstep nit nvar ncond nfit nopter
betax alphax etax etapx
betay alphay etay etapy
x x' y y'
0 0 0 0
dx dx' dy dy'
nener ener ... ener
1 nener
Origin
name keywd del
1 1 1
................
name keywd del
npar npar npar
When nfit equals 1 or 2 the following group applies
CORR
name pos opt param del
1 1 1 1 1
..........................
name pos opt param del
ncor ncor ncor ncor ncor
NOTE: ncor+npar=nval
mon pos val# value weight error
1 1 1 1 1 1
..........................
mon pos val# value weight error
ncond ncond ncond ncond ncond ncond
End of the group for nfit 1 or 2
If nfit equals 3 the following group applies :
CORR
mcorr
name(i) opt(i) param(i) for i=1 to mcorr
nmon nskip
name(i) val#(i) value(i) weight(i) error(i) for i=1 to nmon
end of group for nfit 3
nasp
repeat the following nasp times
name keywd npas
name(k) keywd(k) mult(k) add(k) k = 1 to npas
Parameters:
nstep number of steps taken to approach final fit.
nit number of iterations used in final step. During these
iterations the stepsizes del of the parameters are
reduced by a constant factor(5).
nvar number of variables used (maximum 12).
ncond number of conditions fitted.
nfit selects the fitting procedure.
1 Newton's method is used. In this case ncond = nvar.
2 A least square fit is used.
nopter error option parameter for the reading of the monitors,
this is a noise error of the reading.
0 the monitors have no errors
1 the monitor error is the value given in the error
parameter of the monitor(see below) multiplied randomly
by + and - signs.
2 The monitor error is a uniform random distribution with
a sigma equal to the error value.
3 The monitor error is a gaussian distribution cutoff at
two sigmas
4 The monitor error is a gaussian distribution cutoff at
six sigmas
11,12,13,14 the random error is the same as for 1,2,3 or 4
with a fast random generation of the random sequence.
This random could, in some cases, be affected by
unwanted correlations. In case of doubt, check the
STATISTICAL validity of your results with a family of
runs using the options 1,2,3 or 4.
The initial seed used is the same as that defined by the operation
SEED. The generation of the random errors for the monitors is,
INDEPENDENT of that of the misalignments and of the field errors.
betax,alphax,betay,alphay
input parameters used in the computing the beam line
function values
x ,x',y, y' initial values of nominal orbit.
0 0 0 0
dx dx' dy dy'
increments used in the computation of the cx sx cy sy
functions needed to generate the transfer matrices
around the nominal orbit
nener number of momenta traced (maximum 3).
ener values of the momenta (p-p0)/p0.
origin position used as current origin to position the correctors
used later. This position is be specified by the name
of an element.
name keywd
i i names of the elements having parameters to be varied.
npar<=nvar such elements can be used
del not used in the present version, but must be present in
the input.
CORR flag to signal the correctors are going to be used
mcor for fit 3 : number of corrector names. The
program picks the first ncor available correctors
whose name are any of name(i). Remember that ncor =
nvar-npar
name(i) name of corrector
pos relative position (origin + pos is the absolute
i position of the corrector) i
opt option defining the type of corrector(see SETCorrector
i operation
param parameter number of parameter to be varied
i
del increment used in the fitting routine to vary the
parameter
i
mon(i) Monitor name as present in machine list
nmon for nfit 3 : names of distinct monitors. The program
picks ncond monitors whose name fits name(i) AFTER
SKIPPING nskip monitors!
pos(i) Relative position of the monitor with respect to the
origin point.
Val#(i) value number
=(iener-1)*4+1 x value as read by monitor
=(iener-1)*4+2 y value as read by monitor
=(iener-1)*4+3 sigx value as read by monitor
=(iener-1)*4+4 sigy value as read by monitor
value(i) values read are those corresponding to momentum iener
(1 to 3 maximum).
weight(i) used in conjuction with the least square fit. This
parameter enables the user to put more weight on
certain values to be fitted. The higher the weight(i)
the stronger the constraint to fit the value(i).
error(i) used in conjunction with the parameter nopter. If
nopter is zero no error affects the monitors. If nopter
is 1 the monitor mon(i) is affected by the error
error(i).
nasp: number of associated parameters
name1 Name of element to which some parameters are to be
associated.
keywd parameter keyword of element name1 to which some
parameters are to be associated.
npas total number of parameters to be associated to the
parameter par# of name1.
name(k) name of the element which has a parameter to be
associated with name 1.
par#(k) keyword of parameter to be associated.
mult(k),add(k)
multiplicative and additive constants which define the
value of the associated parameter according to the
following formula
parvalue(k) = mult(k)*parval + add(k)
where parval is the value of the parameter used in the
element name1 and to which par#(k) is associated.
Input format:
BASEline definition
npen Kxxxxxxx Kxxxxxxx .... Kxxxxxxx
nsub sigma
Kyyy Kyyy ..... Kyyy
sigma1 sigma2
Parameters
npen Number of penetration points
Kxxxxxxx a set of npen Elements of the KICK type. There MUST be
one such kick at the beginning and at the end of the
lattice.
nsub Number of subdivision points.
sigma the displacement sigma to be used in the generation of
the coordinate displacements of the npen penetration
points.
Kyyy nsub KICK elements defining the intermediate points.
They may not coincide with any of the basepoints.
sigma1 angular sigma (in radians) of the systematic aim error
sigma2 angular sigma (in radians) of the random aim error
Input format:
BLOCk misalignement...........(maximum 80 characters)
Name1 name2 dx dy dz dzr ddel
.....................
Name1 name2 dx dy dz dzr ddel
99,
Parameters
name1 name2 name of two gkick elements whose names uniquely define
the beamline interval to be misaligned as a block.
dx dy dz dzr ddel one sigma values of the random generation of the
x,y z offsets of the roll around the longitudinal axis
and the relative field offset.
NOTE : only 10 distinct intervals are allowed
Input format:
CORRector data definition ....(maximum 80 characters)
Name i1 f1.....in fn,
.....................
Name i1 f1 ....in fn,
99,
Note the comma ending each line defining a corrector i1 f1 and in
fn are pairs of numbers defining the intervals in which the
elements 'name' are to serve as correctors. MAXCOR(600) distinct
elements can be used as correctors.
Input format:
ERROrs data definition ....(maximum 80 Characters)
Name Parameter value .... parameter value;
...............
Name Parameter value .... parameter value;
99,
Note the semicolon ending each line defining errors for one
element and note the line containing 99, to end the input.
Parameter and value are the parameter keyword of the element called
name that is affected by the error. Value is
the value of the
error.
Input format:
MISAlignment data definition....(maximum 80 characters)
Name dm1 dm2 dm3 dm4 dz dzr ddel option
.............
Name dm1 dm2 dm3 dm4 dz dzr ddel option
99,
Note the comma ending each line defining the misalignments The list
is terminated with 99, as in the first element list.
Parameters:For all values of the option parameter, the parameters dz and dzr are the values of the longitudinal displacement and the rotation angle (in radians!!) around the longitudinal axis. The values dm1,dm2,dm3 and dm4 are assumed to be small (either in displaments or angles). The program uses approximate formulae to set up the misaligned element.
The parameter option can take the three values 1,2 or 3 which determines the nature of the misalignment
option = 1 The element is misaligned around the tangent to the
central trajectory at the entrance of the elements.
In this case the parameters dm1,dm2,dm3 and dm4 are
respectively dx,dxr,dy,dyr where dx and dy are the
displacements along the axes x and y and dxr ,dyr
are rotation angles (in radians) around the axes x
and y respectively.
option = 2 The element is misaligned around the chord defined
by the two extreme points of the central trajectory.
In this case the parameters dm1,dm2,dm3 and dm4 are
dx1,dx2,dy1,dy2 where dx1,dy1 are the displacements
at the entrance of the element along the axes x and
y. The parameters dx2,dy2 are the displacements at
the exit of the element along the axes x and y
option = 3 The element is misaligned around the tangent to the
central trajectory at the midpoint of the element.
The parameters dm1,dm2,dm3 and dm4 have the same
meaning as in the case of the option value 1.
option = 4 This does not apply to dipole elements (Bends). In
this case the parameters dm1 and dm3 represent
displacements in x and y respectively. The particle
is also subjected at the entrance and the exit to an
angle kick of dm2/2 and dm/4 in x and y respectively
(same sign at exit as at entrance).This enables to
simulate baseline excursion in a similar way as that
defined under baseline operation. Parameters dz dzr
are in effect but not ddel.This should be mainly
used on monitors.
Input format:
REFErence ........(maximum 80 characters)
nprint sizex sizey
x x' y y' al delta
0 0 0 0
npos pos .... pos
1 npos
Parameters:
nprint controls display
1 a printout is provided
2 a printer-plot is provided
11 or 12 same as above , the orbit is s 4-dimensional closed
orbit defined by the variables x,x',y,y'
21 or 22 same as above , the orbit is a six-dimensional
closed orbit on the variables x,x',y,y',al,delta.
sizex,sizey always needed. They define the boundaries between
which the coodinates x and y are plotted.
x0...delta six coordinates of the input particle traced to
determine the orbit.
npos number of positions selected for interval computation maximum
4.
pos(i) the rms values are computed individually in all the
intervals defined by the values 0,pos(i),end of lattice
Input format:
SEED........(maximum 80 characters)
ns,
Parameters:
ns 0 a seed is generated by the program.
not 0 ns must be positive. The program insures ns is an odd
number and prints the number used as the seed.
SET Corrector values
Name pos opt p1...p4,
....................
Name pos opt p1...p4,
99,
Parameters:
Name Name of corrector element whose value is to be set
pos position of corrector
opt option number defining the kind of corrector
p1...p4 the four parameters used to define the corrector.
if opt = 0 p1 is dx,p2 is dy,p3 is dy' and p4 is ddel The
corrector element is displaced uniformly by dx and dy.
It is preceded and followed by a momentum dependent
kick of dy' (this simulates crudely the effect of
backleg windings providing a Bx induction) The energy
of the particle is changed by ddel (This simulates the
effect of backleg windings providing a change in By)
opt = 1 the parameters have the same definition as above. The
displacement dx and dy are imposed at the entrance of
the magnet. The exit point of the magnet is assumed
fixed. The operation of dy' and ddel remain the same as
above
opt = 2 As for option 1 the parameters keep their definition.
This time the entrance is fixed and the displacements
dx and dy are imposed at the exit of the magnet. In
both cases 1 and 2 a momentum independent slope is
computed and imposed on the magnet.
opt = 3 In this option the corrector acts as a pure dipole
steering magnet. Parameter 1 is dxp and parameter 2 is
dyp. Parameter 3 and 4 are not used. The angle kicks
dxp and dyp are inversely proportional to the momentum
of the partical(ie: dxp and dyp are divided by 1+delta
where delta is the relative momentum of the particle
traced.
Input format:
SET Errors ....(maximum 80 characters)
nopt
nerr
name nint nb nf ....nb nf ,
1 1 nint nint
...........
name nint nb nf ....nb nf ,
1 1 nint nint
Parameters:
The meaning of the parameters are the same as those of the SET
MIS.... operation. See next operation.
A special value of nopt : 5 , is used to read the errors
sequentially from the fortran input file f007. In this case
however one needs to know the order of the element parameters
internal to Dimad. Please contact Lindsay Schachinger to obtain the
information needed for correct use of this option.
Input format:
SET Misalignment .....(maximum 80 characters)
nopt
nmis
name nint nb nf ..... nb nf
1 1 nint nint
.............
name nint nb nf ..... nb nf ,
1 1 nint nint
This operation defines the random generation mode, the elements that
should be actually misaligned, and the intervals location in the beam
line where they must be misaligned.
Parameters:
nopt choice option for the random generators
0 the elements are misaligned by the fixed values given
in the MISA... operation. No randomness is introduced.
1 The misalignment values are obtained by multiplying the
values given in the MISA... operation by +1 or -1
randomly generated.
2 A uniform distribution is generated having the rms
values defined by the MISA... operation.
3 A Gaussian distribution truncated above two standard
deviations is generated with the rms value defined by
the MISA... operation.
4 A Gaussian distribution truncated above six standard
deviations is generated with the rms value defined by
the MISA... operation.
11,12,13,14 the random error is the same as for 1,2,3 or 4
with a fast random generation of the random sequence.
This random could, in some cases, be affected by
unwanted correlations. In case of doubt, check the
STATISTICAL validity of your results with a family of
runs using the options 1,2,3 or 4.
nmis number of misaligned element type (names)
name name of the family of misaligned element
nint number of intervals in which element is misaligned
0 all element with that name are misaligned. In this
case no interval range is given.
-1 no element with the name are misaligned. In this case
no interval range is given.
nb nf beginning and end of range of misaligned elements.
These numbers correspond to the order number in the
machine list.
SHO Correctors
option ....
Parameters
option determines the information to be printed out.
1 The rms, maxima and minima of the values are displayed
2 name
This option prints the values 1 and 2 of the
correctors with the label name . These values are
multiplied by the scale factor SIGFAC as defined in
Constant definition.
3 Under this option the values of the correctors are
printed out in the format accepted as input by the SET
Correctors operation. This can be used to set up an
input file for a misaligned and corrected machine which
can be then studied without executing the alignement
correction procedures.
SHOERRORS (not implemented yet)
SHO Misalignment
nrange ...............
Parameters
nrange 0 then all misalignements are printed
>0 then nrange intervals ni mi are used. The misalignments
are printed in these intervals
-1 then the operation sets up arrays containing all the
misalignement data. Tracking execution proceeds faster.
-2 name1 name1j1 name1k1 ... name1j5 name1k5,
name2 name2j1 name2k1 ... name2j5 name2k5,
....
99,
MISFAC
Up to five names namei can be given and up to 5 intervals
nameij nameik. The names defining the intervals must be
unique (use markers). The operation adds to the elements
namei a random misalignment as defined in the misalignment
data multiplied by the factor MISFAC but using a different
random sequence from that used in the main misalignement
procedure. This operation must be preceded by a SHO MIS
operation with option -1 to be successful.
-10 nameThis operation provides information of the average lateral displacement of the element labeled name using the scale factor SIGFAC. nrange number of intervals in which the printout will occur. if nrange = -1, the the operation sets up arrays containing all the misalignement data. Tracking execution will then proceed faster(more space is needed) ni mi beginning and end of interval in which printout occurs.
SYNChrotron radiation.....(maximum 80 characters)
Energy option randomoption,
Parameters
Energy Initial nominal energy in GeV
option 0 no synchrotron radiation effect is simulated.
1 synchrotron radiation loss is computed in every dipole.
Particles lose that energy at the entrance and exit of
the magnet.
2 synchrotron emittance growth is simulated randomly for
each particle. This growth is due to the spread in the
energy loss of the particles. This effect is simulated
only in the dipoles.
3 both the radiation loss and the emitttance growth are
simulate in the dipoles only.
4 synchrotron radiation is simulated in the dipoles as in
option 2 to which is added a synchrotron quantum
fluctuation in the quadrupoles.
5 synchrotron radiation is simulated in the dipoles as in
option 3 to which is added a synchrotron quantum
fluctuation in the quadrupoles.
randomoption: choice of the random generator, applies to above
options 2 and 3 only.
1 the random generator is binary + and - randomly
affecting the energy spread creating the emittance
growth.
2 the random generator is uniform
3 the random generator is gaussian : note that here the
execution time will be considerably greater than with
choice 2 for the options 2 and 3 above which enable the
emittance growth calculations.
11,12,13 the fast random generator is used to produce the
sequence of random numbers. This sequence may be
affected by some unwanted correlations. In case of
doubt use the options 1,2 or 3.