The OPAL Discussion Forum

[Simon Friederich <sifriede AT uni-mainz.de>]

Dear OPAL-Team,

I am trying to simulate a DC electron source. I implemented a 2DDynamics fieldmap into OPAL, that I've exported from CST. Furthermore I've set the frequency to 1 Hz as shown in the DC Gun example.

Unfortunately the longitudinal phase spaces look unexpected. Here I've plotted the h5-output for the gun simulation as well as a pure drift simulation of a 100keV electron beam (same bunch sizes and parameters but the energy is fixed) as comparison.

One can see a strange saw-tooth like pattern convoluted with another almost trident-like one.

This is my emitted gaussian distribution (nothing special, I'd say):

distr_gauss:
    DISTRIBUTION,
    TYPE = GAUSS,
    SIGMAR = 0.5e-3, CUTOFFR = 3,
    CUTOFFLONG = 3;

distr_debug:
    distr_gauss,
    EMITTED=true,
    SIGMAT = 10.0e-12,
    NBIN = 0,  // This I've varied
    EMISSIONSTEPS=10,  // this I've varied too
    EMISSIONMODEL = NONE;

Maybe you see something obvious, that I did wrong? (I've also put my input file in the attachement, the fieldmaps are quite large, if you want them I'll add a seafile link)

I've compared the Ez-field of the reference particle. It looks exactly like the one I've exported from CST. (My next idea was to dumb all E-fields and plot this, but this could be a lot of work and might not bring up the source of the problem). And I've also imported a 3DDynamics field, but there is also no difference to be seen (the source has a cylindrical symmetry, so both should work).

Best regards and thank you very much as always in advance

Simon

-- 
-----------------------------------
Dr. Simon Friederich
Helmholtz-Institut Mainz
Kollaboration B/ACID
Johannes Gutenberg-Universität Mainz
Staudingerweg 18
55128 Mainz, Deutschland

Tel.: +49 (0)6131 39-23160
E-Mail: sifriede AT uni-mainz.de
----------------------------------------

TITLE, STRING="DCGun-test";

OPTION, 
        VERSION = 24000,
        //SEED = 123456789, // fix seed for particle generation and 
comparison reason
        ECHO = false,
        INFO = false,
        STATDUMPFREQ = 1, // after how many time steps we dump statistical 
data
        ENABLEHDF5 = true,
        PSDUMPFREQ = 10, // Affects the stdout information **BUNCH**
        AUTOPHASE = 1;

//----------------------------------------------------------------------------
// Beam parameter
REAL freq = 1.3e9;
// REAL Qbunch = 7.7e-12; // 10mA
// REAL Qbunch = 0.77e-12; // 1mA
REAL Qbunch = 0.77e-15; //1 uA
REAL current = Qbunch*freq;

// Simulation parameter
REAL nelectrons = 100000;
REAL dt_s = 10e-12; // time step in s
REAL zstop = 0.5; // in m

//----------------------------------------------------------------------------
// Distribution
distr_gauss: 
        DISTRIBUTION, 
        TYPE = GAUSS,
        SIGMAR = 0.5e-3, CUTOFFR = 3,
        CUTOFFLONG = 3;
        //WRITETOFILE=true;

distr_gauss_injected: 
        distr_gauss,
        EMITTED=false,
        //DEBUG SIGMAZ = 0.015960;
        SIGMAZ = 0.001;// [SIGMAZ]=m; 
        
distr_gauss_emitted: 
        distr_gauss,
        EMITTED=true,
        NBIN = 10,
        EMISSIONSTEPS=20, // prior=20, Refine number of simulation steps 
around emission
        // SIGMAT = 100.0e-12; // [SIMGAT]=s; SIGMAT is used for EMITTED distr
        SIGMAT = 100.0e-12; // Debug [SIMGAT]=s; SIGMAT is used for EMITTED 
distr
        
distr_gauss_emitted_100kV: 
        distr_gauss_emitted,
        Ekin=1e5; //[Ekin] = eV;thermionic energy;  not yet known why BEAM, 
ENERGY=EMASS+1e-4 does not work the same way
        //[default: 1eV] Thermal energy added to beam during emission.

distr_debug:
        distr_gauss,
        EMITTED=true,
        SIGMAT = 10.0e-12,
        NBIN = 0, 
        EMISSIONSTEPS=10, 
        EMISSIONMODEL = NONE;
        //, EKIN=5;
        // EMISSIONMODEL = ASTRA, EMISSIONSTEPS=300, NBIN = 10, EMITTED = 
TRUE;

//----------------------------------------------------------------------------
distribution_to_be_used: distr_debug;
//----------------------------------------------------------------------------
// Fieldsolver 
fieldsolver_default:
        FIELDSOLVER,
        FSTYPE = FFT, // Specify type of field solver ("none" does not 
respect space charge)
        MX = 20, // Number of grid points in x specifying rectangluar grid
        MY = 20, 
        MT = 50,
        PARFFTX = true, // dimension x is distributed among the processors
        PARFFTY = true,
        PARFFTT = true, // changed that!
        BCFFTX = open, // Boundary condition in x,y options: [open]
        BCFFTY = open,
        BCFFTT = open, // Boundary condition in z options: [open, periodic]
        BBOXINCR = 10, // Enlargement of the bounding box in %
        GREENSF = INTEGRATED;

//----------------------------------------------------------------------------
// Source
cathode: SOURCE, Z = 0.0; // not yet known if really necessary
gun: RFCavity, 
        Z = 0,// Z=-0.0015,
        VOLT=2.805, // peak rf field amplitude in MVpm
        FMAPFN="cst2opal_100keV_xz_2DDynamic.t7", // 2DDynamic
        // 3DDynamic
        // FMAPFN="cst2opal_100keV_xyz_3DDynamic.t7", // 3DDynamic
        APVETO=TRUE,
        FREQ=1e-6; // Zero frequency = DC gun

//----------------------------------------------------------------------------
// Beamline
mybeamline: LINE = (cathode, gun);


//----------------------------------------------------------------------------
// Beam
mybeam: BEAM, 
        PARTICLE = ELECTRON, 
        //ENERGY = EMASS+1e-4, // in MeV // But for gun no starting energy 
needed
        ENERGY = EMASS+1e-10, // in GeV
        BCURRENT = current, // in A
        NPART = nelectrons,
        BFREQ = freq*1e-6; // [BFREQ] = MHz

//----------------------------------------------------------------------------
// Simulation
TRACK,
        LINE = mybeamline, 
        BEAM = mybeam, 
        MAXSTEPS=100000, 
        DT=dt_s, // in s
        ZSTOP=zstop; // in m

RUN,
        METHOD = "OPAL-T", 
        BEAM = mybeam, 
        FIELDSOLVER=fieldsolver_default, 
        DISTRIBUTION=distribution_to_be_used;

ENDTRACK;

STOP;