Achromatic but final phase space smearing and quite strange

[Biaobin]

Hello Michael,

I am using two chicane compressors to compress the bunch, and using the RFCA cavity to dechirp the beam at last. But the final longitudinal phase space is quite strange. All the elements are used first order matrix, I checked the dispersion line and R16 R26 in the *.mpr file, the lattice is actually achromatic. Is there some wrong settings in the *.ele or *.lte files ?

The phase space figure(the initial and final one) is as following:

The .ele and .lte files are also attached.

Thanks in advance and wish you a happy Thanksgiving day.

Biaobin

[michael_borland]

Biaobin,

The problem is that the beam emittance is too large to preserve such a short bunch length. The dependence of the path-length on transverse angle results in growth of the bunch length downstream of the last chicane.

I checked this by using a MAGNIFY element to vary the emittance of the beam. As the emittance decreases, the bunch length is maintained more closely to the value at the exit of the final chicane.

You need a normalized emittance of about 1 um to preserve the bunch length.

StVsEmittance.png (7.34 KiB) Viewed 4468 times

--Michael

[Biaobin]

Hi Michael,

What I still can't understand is that, if I follow the steps as:
1. get initial particle distribution pha0 as

Code: Select all

elegant2astra *.w1

(I am more familiar with the Astra data structure)
2. get the total transfer matrix R in *.mpr file
3. using Matlab to get the final phase space as: pha1 = R * pha0

The final longitudinal phase space looks like this:

Since Elegant is also using the first order transfer matrix method, why are the results different?

Best,
Biaobin

[michael_borland]

Biaobin,

I can see the cause of the confusion. The RFCA element always computes the exact time of flight through the structure, even if it is using a linear matrix for the basic model. You can work around this by replacing your RFCA elements with non-zero length with drift-rfca-drift sequences. Of course, the results are not correct in that case, but it should agree with your expectations for a fully linear simulation.

--Michael

[Biaobin]

Hi Michael,

Thank you so much, it works. Sometimes I was wondering if I can look into the source code which I have downloaded to find out how does the elements work since I cannot quite understand the manual. I just find out that the manual actually has told me that for RFCA "First-order matrix RF cavity with exact phase dependence". Taking this RFCA element for example, which source file should I look at ?

Best,
Biaobin

[michael_borland]

Biaobin,

Look in simple_rfca.c.

--Michael

[Biaobin]

Hi Michael,

Thank you! I am looking at the code.

Have a good day!
Biaobin