CEA Codes forum

Dear Didier,

I would like to do following:
- fix quads errors (shift and rotation) ones for all runs in error study
- change other errors for each run independently

For that I put in .dat file commands:

ERROR_CAV_NCPL_STAT 20000 1 0 0 0 0 1.0 1.0 0
ERROR_BEAM_STAT 1, 0, 0, 0, 0, 0, 1, 15, 15, 15, 10, 10, 10, 5
ERROR_QUAD_NCPL_STAT_FILE Error_Quad.txt


File Error_Quad.txt has following view (copied from the TRACEWIN output error file and modified just for quads) and it is placed at the same directory.

QUAD_ERROR dx(mm),dy(mm),drx(°),dry(°),drz(°),dG(%),L(mm),dG3(%),dG4(%),dG5(%),dG6(%)
CAV_ERROR dx(mm),dy(mm),drx(°),dry(°),drz(°),dE(%),dPhase(°),L(mm)
BEND ERROR dx(mm),dy(mm),drx(°),dry(°),drz(°),dg(%),dz(mm)
BEAM_ERROR dx(mm),dy(mm),dφ(°),dxp(mrad),dyp(mrad),de(MeV),dEx(%),dEy(%),dEz(%),mx(%),my(%),mz(%),dib(mA),αxx’_min,αxx’_max,βxx’_min(mm/mrad),βxx’_max(mm/mrad),αyy’_min,αyy’_max,βyy’_min(mm/mrad),βyy’_max(mm/mrad),αzdp_min,αzdp_max,βzdp_min(mm/mrad),βzdp_max(mm/mrad)

QUAD_ERROR [2] 0.110907 0.105804 -0.884976 -0.151432 0.282726 0 96 0 0 0 0
QUAD_ERROR [3] -0.0950807 0.0384054 0.293608 -0.730693 0.314091 0 96 0 0 0 0
QUAD_ERROR [4] 0.0713264 -0.133442 0.822767 -0.88562 0.565643 0 96 0 0 0 0
QUAD_ERROR [5] -0.106505 -0.031903 -0.730138 0.818697 0.0510322 0 96 0 0 0 0

The code starts and then shows Statistic_Errors_1[1][1]= +1.000 Error!!

Problem comes definitely from the quads errors description with file.
ERROR_QUAD_NCPL_STAT (instead of ERROR_QUAD_NCPL_STAT_FILE Error_Quad.txt) works properly.
Just quads errors (with file) without other errors lead to error.
I tried to remove the first lines from Error_Quad.txt, didn't help.

Could you please give me an advice what should I change?

Thank you in advance!

Best regards,
Anna

Dear Anna,

Be carreful. What you want to do is correct but only using the rigth commands:
"ERROR_STAT_FILE" or "ERROR_DYN_FILE"

"QUAD_NCPL_STAT_FILE file" is somthing else completely different. This is described in the explanation of "ERROR_QUAD_NCPL_STAT_FILE" below.

The fifth command uses a file to set errors. This file is located by default in structure file (*.dat) path. It contains lines like the following syntax (be careful to Z unit):

Z0(m) r(0/1/2) dx(mm) dy(mm) φx(°) φy(°) φz(°) dG(%) dz(mm), dG3(%), dG4(%), dG5(%), dG6(%)
Z1(m) r(0/1/2) dx(mm) dy(mm) φx(°) φy(°) φz(°) dG(%) dz(mm), dG3(%), dG4(%), dG5(%), dG6(%)
…
Zn(m) r(0/1/2) dx(mm) dy(mm) φx(°) φy(°) φz(°) dG(%) dz(mm), dG3(%), dG4(%), dG5(%), dG6(%)
An interpolation is performed to define errors applied to an element at a given position. It’s not the case for ‘r’ parameter which is defined without interpolation.

Regards,

Didier

Thank you Didier!

My idea was simply to fix quads errors ones for all runs (in a file) and randomly change other errors...

As far as I understood, if i use ERROR_STAT_FILE for N linacs, I have to create N files (for each run). These files should contain all errors (quadrupoles, cavities, beam). I cannot use this description just for quads (and for cavities ERROR_CAV_NCPL_STAT), right?

Should Z in ERROR_QUAD_NCPL_STAT_FILE be the middle of the quadrupole?

Best regards!
Anna

Dear Anna,
Yes, exactly, it's not possible to use usual error commands mixed with ERROR_STAT_FILE.
No, the positions considered are at the beginning of the elements

Regards,

Didier