Examples

Example runs and post-processing files are included along with the distribution of elegant. These are drawn from the author's research and all concern various aspects of the Argonne Positron Accumulator Ring (PAR) and its injection and ejection lines (LTP and PTB, respectively).

The examples are intended to demonstrate program capabilities with minimal work on the user's part. Each demo is invoked using a command (a C-shell script) that can both run elegant and post-process the output. After running the demo, the output can be viewed again without rerunning elegant by invoking the command with the word review added to the command line. Including the word hardcopy on the command line results in the graphs being sent to the default printer, which is assumed to accept Postscript.

The post-processing is typically handled by a lower-level script that is called from the demo script. These lower-level scripts are good models for the creation of customized scripts for user applications.

1. par10h* -- These files provide a demonstration of Twiss parameter computation, tracking, element variation, and map analysis. The lattice is defined with kick elements, which are used for all tracking. After computation of the Twiss parameters for the PAR[6], a series of particles are tracked with different initial x coordinates. Finally, the tunes and Twiss parameters are computed by tracking; they are very close to the analytical values. The post-processing commands make phase-space plots and plots of FFTs of the motion, showing that the motion becomes chaotic at the stability limit. To execute this demo, type the command par10h.

2. par_sympl* -- These files provide a demonstration of the symplecticity of tracking with elegant kick elements. A single large-amplitude particle is tracked for ${\rm 2^{14}}$ turns. The invariant ${\rm J_x}$ is then computed and plotted as a function of turn number. To execute this demo, type the command par_sympl. The post-processing takes quite some time because of the very large number of points.

3. par_chrom* -- These files provide a demonstration of computing chromaticity and other parameters as a function of momentum offset using map analysis. The lattice is the same as par10h.lte, except all of the elements are implemented using second-order matrices. Hence, the chromaticity from tracking should be nearly identical to the analytical results computed by the twiss_output command, which it is. To run this demonstration, enter par_chrom. The reader may wish to try this demo again using ksbend, csbend, or nibend elements in place of the sbend elements, and kquad (ksext) elements in place of the quad (sext) elements.

4. par_damp* -- These files provide a demonstration of damping partition calculation using single turn tracking with synchrotron radiation. The expected value of the longitudinal damping partition for PAR is ${\rm J_\delta = 1.758}$. The user may edit the lattice file, par_damp.lte, to invoke a different element for the dipole magnet. In particular, definitions for numerically integrated dipoles with extended fringe-fields are present. To execute this demo, type the command par_damp.

5. par_dynap* -- These files provide a demonstration of dynamic aperture runs for a series of randomized machines. Also exhibited here are orbit, tune, and chromaticity correction. The post-processing commands make a plot of the dynamic apertures with the physical aperture superimposed. (The orbcorr_plots script can also be used to plot orbit correction information.) To execute this demo, type the command par_dynap. The lattice has been stripped down so that only a few of the more significant multipoles are present. Also, fictitious extra sextupoles have been added to compensate the lack of second-order edge terms in the bending magnets (these would result in nonsymplectic tracking if included). Still, the running time is many hours.

6. ejoptk* -- These files provide a demonstration of the optimization of a multi-turn ejection bump for PAR, using a time-dependent kicker waveform (formed from two cubic splines). After optimization, the lattice is tracked with a realistic beam distribution to verify good transmission and show the centroid position vs z over three turns. To execute this demo, type the command ejoptk.

7. ltp_te* -- These files provide a demonstration of transport line simulation. The Linac-to-PAR transport line is simulated with errors and trajectory correction to predict the transmission losses and the steering error at the exit of the septum. The trajectory correction uses tracking of a beam distribution, which is slower than tracking the centroid, but which produces better results in the presence of the large momentum spread. The reader may wish to verify this by turning off this feature and running the simulation again. To execute this demo, type the command ltp_te. The running time for this demo is quite long.