Elegant users forum

Dear Dr. Borland,

Recently we are trying to estimate CSR-induced microbunching gain in recirculating cooling ring of JLab MEIC by using ELEGANT. Before that, I am trying to reproduce some of the results you published in LINAC02 (http://accelconf.web.cern.ch/accelconf/ ... /MO202.PDF) under the parameters of 'model chicane'. However, we met some difficulties and problems on this simulation:

First, we generate initial 6-D phase space distribution by ELEGANT (with 24M simulating particles); second, we replace the longitudinal phase space distribution (t,p) with that generated by flattop distribution (a script you provided Chris before.) Then, I use smoothDist6s to generated desired the ripple wavelengths. However, it is strange that at some (short) wavelengths, the smoothDist6s-generated modulated distribution does not output correct/desired wavelength(s). Also, at some other wavelengths, the generated distribution has multiple (lower than desired frequencies, not harmonic) spectral peaks. I am not sure if there is any other special setting I should make when using smoothDist6s?

ps. We use Pelegant 25.2.2 and compiled from source codes.

I appreciate your help in this situation. Please let me know if you need more information.

Best regards,
Cheng-Ying

Cheng-Ying,

Sorry for the late response. The forum is not sending me notifications of new posts reliably.

My suspicion is that your wavelengths are too short and run into a built-in parameter in the smoothDist6 script. Look for occurences of hard-coded numbers like "1000" and "10000", and trying increasing these values. If this doesn't solve the problem, let me know.

--Michael

Dear Dr. Borland,

In my experience when trying to reproduce Fig. 2 (the last case) in MO202, I found the wavelengths shorter than 7 um would give incorrect initial (modulated) distribution. Let me list below how these 'desired' wavelengths drift:
2 um -> 17.88 um
3 um -> 18.08 um
4 um -> 35.76 um
5 um -> 12.83 um
6 um -> 8.99 um
these drifted wavelengths are derived from sddsnaf for a range of amplitudes.

Further, I did not really realize what you mean 'hard-coded numbers.' Could you detail this for me?
Thank you for the help!

Best regards,
Cheng-Ying

Cheng-Ying,

The hard-coded numbers I'm referring to are in the script file and are used to determine how many bins to divide the beam into. If the modulation wavelength is too short compared to the total bunch length, then these hard-coded numbers may not be correct to ensure that the modulation is performed correctly (or at all).

You can easily modify smoothDist6s. First, copy the script file
% cp `which smoothDist6s` mySmoothDist6s
Then, edit mySmoothDist6s to replace occurrences of 1000 and 10000 with larger values (say 5000 and 50000). Running your modified version should allow imposing shorter modulations.

--Michael