Wrong emittance ratio after tracking full coupling beam.

[Gyeongsu]

Dear, everyone

I put 12 skew quadrupoles at lte file and adjusted tunes such that beam lies in different resonance.

I verified the full coupling happened from twi file as below.

couplingIntegral 6.959754783793667e-03
couplingDelta 9.320972046111820e-05
emittanceRatio 9.998206684736263e-01

I tracked beam with 240 particles and checked horizontal and vertical emittance variation according to the pass number.
however, H and V emittances doesn't become the same. The result looks like that beam become the flat beam after many turns.
I made beam with horizontal emittance 58 pm and vertical emittance 5.8 pm at the beginning.
refer the attachment PPT file.

I tried eliminate skew quadrupoles at lte file. Instread, I put normal quadrupoles' tilt errors at ele file.
I tracked beam with the same condition as the above case using skew quadrupoles.
And I checked horizontal and vertical emittance variation according to the pass number also.
The result looks like that beam become close to the round beam after many turns.
Of course, I cannot know coupling coefficient value in this situation.

Anyway, It seems which skew quadrupoles doesn't affect coupling but quadrupoles' tilt errors do.

Please, attachment file.

Sincerely,
Gyeongsu Jang

[michael_borland]

Gyeongsu,

Please upload files that show the problem. Otherwise, it is very difficult to provide help.

--Michael

[Gyeongsu]

Dear michael,

I am attaching lte file with skew quadrupoles and ele file for running elegant.

You can see S31ACORRECTOR , S32ACORRECTOR, and so on elements In lte file.
In these elements, there are QS31A, QS32A, and so on elements.
These elements are skew quadrupoles.

There are macros such as <latticename>, <rootname>, <passes>, and <nparticle> in ele file.

Sincerely,
Gyeongsu Jang

[michael_borland]

Gyeongsu,

Sorry for the long delay on my reply. I thought I had posted this already.

There were various problems in your simulation setup. The biggest one is that need to use element-by-element synchrotron radiation, not SREFFECTS. Also, the rf cavity needs to be set up correctly.

After doing this, I get reasonable results. Please see attached.

--Michael

[Gyeongsu]

Dear michael,

I appreciate your reply.
I will refer to your comments and attached files.

Sincerely,
Gyeongsu Jang

[Gyeongsu]

Dear Michael,

I appreciate your reply the previous question.

However. the result that we get from your reply reveals that the both of horizontal and vertical emittances become 59 pm the natural emittance of our ring after sufficient turns. Round beam emittance in our lattice is 39 pm.

We tried to remove all skew quadrupoles and put tilt error to normal quadrupoles with gaussian random number.
We set a amplitude of tilt errors as 0.5 mrad.
Then we track beam and verified that the both of horizontal and vertical emittance converge to about 39pm, i.e. round beam emittance.

We attched ppt file with detailed description.
And we attched ele and lte files that tilt errors in normal quadrupoles was put and H and V emittance become 39 pm.
please refer the attached files.

Sincerely,
Gyeongsu Jang

[michael_borland]

Gyeongsu,

I found a number of problems in the files. I still see SREFFECTS elements and don't see SYNCH_RAD=1 and ISR=1 on the CSBENDS, KQUADs, and KSEXTs. Also, several skew quads are powered at the same K1, which excites the zeroth harmonic instead of the coupling harrmonic. This increases the vertical dispersion without coupling emittance. This is confirmed by the output of the moments_output command and tracking.

Note that even if you add errors to the skew quad strengths, this existing zeroth harmonic may dominate. Errors are by defaulted *added* to the values, rather than replacing them.

Here's what I get with all the skew quads at zero

I've fixed these problems and also created an example of driving the coupling harmonic, which is attached. Results are as expected

--Michael

[Gyeongsu]

Dear michael,

I am late to say thanks to you for your reply.
We are doing coupling study by using and referring attached files in your reply.
Thanks a lot.

Sincerely,
Gyeongsu Jang