Study of TMCI

[arahim]

Is simulation of TMCI is possible in Elegant.

[lindberg]

Yes. Short-range transverse wakefields (impedances) are modeled using the TRWAKE or (ZTRANSVERSE) element. One published example that uses ZTRANSVERSE to determine the transverse instability threshold is described in Section IV of PRSTAB 19, 124402 (2016); there are many other examples [e.g., end of Sec. IV in PRAB 21, 024402 (2018)], although typically without much detail of how to implement things in elegant.

[arahim]

I am using ztransverse for the simulation of TMCI. I have given the command in the following way..
zz_rw_trans: ztransverse, &
zxreal="rw_cdr2014v1.impe1=f+ReZxH", zximag="rw_cdr2014v1.impe1=f+ImZxH", &
zyreal="rw_cdr2014v1.impe1=f+ReZyH", zyimag="rw_cdr2014v1.impe1=f+ImZyH", &
n_bins=4096, bin_size=4.2e-12, smoothing=0, &
ramp_passes=0, bunched_beam_mode = 1

I am getting an error of "you must give an input file and freqcolumn " but in the impedance file frequency is already mentioned in f column.

Here should I use Z or effective Z in both the transverse planes.

[michael_borland]

ZTRANSVERSE doesn't use the <filename>=<x>+<y> syntax. Instead, you give the filename with the INPUTFILE option, then name the various columns. For example,

zz_rw_trans: ztransverse, inputfile="rw_cdr2014v1.impe1",freqcolumn="f",&
zxreal="ReZxH", zximag="ImZxH", &
zyreal="ReZyH", zyimag="ImZyH", &
n_bins=4096, bin_size=4.2e-12, smoothing=0, &
ramp_passes=0, bunched_beam_mode = 1

--Michael

[arahim]

Thanks a lot..it works..

[arahim]

I am tracking 10 ps (one sigma) bunch for microwave threshold in a storage ring. The input for longitudinal impedance is as follows:
zz: zlongit,broad_band=1,freq=30e9,Rs=42e3,Q=1,bin_size=4.2e-12,n_bins=0
I have given the option for auto resizing the number of bins. It works for low beam current but for high beam current (1.8mA in my case), tracking is stopped with some warning. I have attached the doc file containing two snaps of warnings. In first snap, current is 1.5mA, tracking is successfully performed. In second snap current is 1.8mA, tracking stopped after few turns. In second warning, it is suggested to choose the option for auto scaling for number of bins but that I have already done by giving the option n_bins=0. Looking for your comment.

[michael_borland]

The problem is that at a certain current, bunch lengthening or instability occurs (probably instability), which results in the total length of the bunch being too great for the span (nBins*binSize). Unfortunately, the number of bins is chosen once at the beginning. I suggest setting the number of bins manually to a larger value.

--Michael

[arahim]

I want to incorporate amplitude dependent tune shift (ADTS) in TMCI simulation. During calculation of ADTS by elegant, particle is being lost in few hundred of turns but in Accelerator tool box particle is surviving after 1000 turns even with higher initial launching amplitude. What is wrong with my elegant files. I have attached the elegant files for your reference. Thanks in advance.

Abdurrahim

[michael_borland]

Abdurrahim,

The main problem is that you've included an SREFFECTS element (synchrotron radiation) but the rf cavity (RFCA element) is not configured. More fundamentally, you shouldn't include either of these elements if you are interested in ADTS. Further, you are using matrix-based tracking (quad, sben, sext elements) when symplectic tracking is more appropriate (kquad, csben, ksext).

See attached files.

--Michael

[arahim]

Thanks a lot. It worked..