beam size

[Samira.kasaei]

Hello Michael,

I tried to calculate beam size for my ring but variation of the beam size is not small along the ring. Could you please take a look to my files?

Thanks alot and best regards,
Samira

[michael_borland]

Samira,

You need to ensure that you generate a matched beam to start. You have

Code: Select all

&bunched_beam
        n_particles_per_bunch = 1000,
        one_random_bunch=0,
        emit_x = 26e-9,
       emit_y = 26e-9,
        distribution_type[0] = 3*"gaussian",
        distribution_cutoff[0] = 3*3,
       ! symmetrize = 1,
        enforce_rms_values[0] = 1,1,1,
&end

&twiss_output
        filename = "%s.twi",
        final_values_only = 0
&end

The beam distribution has no defined beta functions, so some defaults will be used. What you want are the periodic beta functions from twiss_output. You can get the desired result by changing the order of twiss_output and bunched_beam, and adding the use_twiss_command_values=1 parameter to bunched_beam:

Code: Select all

&twiss_output
        filename = "%s.twi",
        final_values_only = 0
&end

&bunched_beam
        n_particles_per_bunch = 1000,
        one_random_bunch=0,
        emit_x = 26e-9,
       emit_y = 26e-9,
        distribution_type[0] = 3*"gaussian",
        distribution_cutoff[0] = 3*3,
       ! symmetrize = 1,
        enforce_rms_values[0] = 1,1,1,
        use_twiss_command_values = 1
&end

--Michael

[Samira.kasaei]

Dear Michael,

Now it is working. However when I add some skew quads in my lattice I got the same vertical beam size. What is wrong in your opinion?

Many thanks and best regards,
Samira

[michael_borland]

Samira,

Please post your files and I'll take a look.

--Michael

[Samira.kasaei]

Dear Michael,

Please find the attached.

Best regards,
Samira

[michael_borland]

Samira,

I think the problem is partly that your skew quadrupoles are relatively weak. For example, if I multiply the skew strengths by 100, there's a clear effect on Sx and Sy in the .sig file.

Also, please note that &twiss_output does not give trustworthy results in the presence of strong coupling. You can use &coupled_twiss_output instead, as well as &moments_output.

--Michael