New shape of the target

[mimboch]

Hello, I'm trying to simulate the following shape of the jet gas target, with circular ends and a flat section in the middle . I would be very grateful if anyone could give me a hand. Thank you in advance.

[Status-Mirror]

Hey @mimboch,

I've adapted the basic target demo using expressions from the maths parser to obtain the following target:

which was created with this input deck:

begin:control
    nx = 800
    ny = 120
    t_end = 1.0e-15
    x_min = -20e-6
    x_max = 20e-6
    y_min = -3.0e-6
    y_max = 3.0e-6
    stdout_frequency = 100
    npart = 10 * nx * ny
end:control

begin:boundaries
    bc_x_min = open
    bc_x_max = open
    bc_y_min = open
    bc_y_max = open
end:boundaries

begin:constant
    targ_length = 25e-6
    targ_rod_thickness = 600e-9
    targ_ball_radius = 2e-6
    targ_dens = 1.0e24
end:constant

begin:species
    name = Electron
    mass = 1.0
    charge = -1.0
    frac = 0.8
    temp_ev = 1000

    # Rod part
    density = if (abs(y) lt 0.5*targ_rod_thickness \
                  and abs(x) lt (targ_length/2 - targ_ball_radius), \
                  targ_dens, 0)

    # Balls
    density = if ((abs(x)-targ_length/2 + targ_ball_radius)^2 + \
                   y^2 lt targ_ball_radius^2, targ_dens, density(Electron))
end:species

begin:species
    name = Carbon
    mass = 22033.0
    charge = 6.0
    frac = 0.2
    density = density(Electron) / 6
    temp_ev = 1000
end:species

begin:output
    dt_snapshot = t_end
    number_density = always
end:output

Hope this helps, Stuart

[mimboch]

Thank you so much for your assistance .

[mimboch]

Hello, I should have a smooth transition between the circles and the flat section, as shown in the attached picture. I am trying to achieve this with Expo functions, but it didn’t work. Could you please help me again? Thank you in advance.

[Status-Mirror]

Can you be more specific about the curve-shape? For an exponential, I would need to know where you want it to start from, which point on the circle you want it to end, and the exponential scale-length $k$, for a functional form $\exp(x/k)$.

If the older ball-rod target above doesn't work for you, then your results must be sensitive to the nature of this curve. If I just give you a random curve, it will probably have the same problem as the simple ball-rod.

[mimboch]

Hello, Thank you for your reply; I really appreciate it. Yes, you are absolutely right, but it is not necessarily the goal is to have these curves. For example, I could use a decreasing exponential function from 2 (center of the circle) to 5 µm on the left side and an increasing exponential function from 19 to 23 µm on the right side, with an exponential scale length k=2μm.