Particle and Radiation Distribution
With the parameters IDMPFLD and IDMPPAR set to a non-zero value the field and particle distribution is written to binary files for each slice at the end of the undulator. The file format is similar to that of the particle and field output file (see above) with a few differences. The particle phase is absolute and not relative to the radiation phase. The radiation field has the same format as in the radiation output files (see above), representing the intensity if real and imaginary parts are summed in square, but instead listing real and imaginary part separately, the field dump has the complex value listed instead (pairs of real and imaginary part). The radiation field escaping the last slice is added to the output so that the radiation field, escaping first, is the last record in t. Note that there is only one record in z for both files. Harmonics are written in individual files and have a modified extension by adding the harmonic number to the file name.
These files can be used to simulate the effect of a bunch compressor or a monochromator and then reused as an input for a consecutive run of GENESIS 1.3. To indicate GENESIS 1.3 to skip the internal generation of the radiation field or particle distribution the file names of these files are given as the input values for FIELDFILE and PARTFILE.
There are some restrictions of using these files.
While the first point is obvious the second needs some explanation. Because the radiation slips through several electron slices the time-window of the radiation field must be larger than the time-window of the simulated electron beam. It is extended by the slippage length NSLP, which is the number of radiation slices which slips through a single electron beam slice over the entire undulator length. The length is the undulator length divided by XLAMD and ZSEP and rounded to the lower integer. The radiation initially seen by the first beam slice is the NSLP+1 record of the radiation input file. The records between the first and NSLPth slice are copied in the time-record. During the integration these slices are pushed into the location of the simulated electron slice. Generally speaking the time window of the radiation field is shifted forwards with respect to the time window of the electron beam during the integration. In the case that the second GENESIS 1.3 run uses a longer undulator than the first run, which dumped the radiation field and beam distribution, the number of beam slices has to reduced to match the file length of the input radiation field.
Note that Genesis does not check for consistency. The field dump is associated to a certain grid size and radiation wavelength. If a field is imported for a simulation, where these values differs, the energy of the field is not conserved.