Inconsistent beam emittance values from GUI tools

[cemitch]

Dear Didier,

After running TraceWin on a linac input file, I see two different sets of values for the final normalized x-emittance at the exit of the transport. The difference is large (more than a factor of 2). (The discrepancy exists for the other emittances as well as the beam sizes in each dimension, but the differences are much smaller.) I compared values obtained from the GUI using four different workflows:

1) Charts -> Emittances -> Data (to select the final s-position in the output table)
2) Charts -> Output -> enter the final s-position or final element number -> View beam parameters

3) Charts -> Density -> Rms Emittance -> Plot -> Data (to select the final position in the output table)
4) Charts -> PlotWin -> enter the final s-position or final element number -> View beam parameters

The values from 1 and 2 agree with each other, and so do 3 and 4, but the values from 1(2) do not agree with the values from 3(4).
These values do agree at the beginning of the transport, so I am sure that all these procedures should yield consistent normalized emittance values.

I don't understand what the source of this difference could be. Do you have any suggestions?

Thanks,
Chad

[Didier]

Dear Chad,

I don't get the differences on some examples, that you have, so your project have probably some specificities that I don't have. Please send me the files of your project, simplified to the maximum.

Regards,

Didier

[cemitch]

I was able to recover agreement among the four workflows shown previously. This happened after I manually deleted any old output files from within my working directory. It's likely that one or more of the postprocessing tools was accessing output files from an older run with different parameters.

It might be worth having TraceWin delete any old output files sitting within the Calculation Directory, or to create a subdirectory with a unique name on each run, in order to avoid overlap between new and old copies of output files.

Thanks a lot for your time.

[cemitch]

Dear Didier,

My previous post was a bit premature. I still see the emittance discrepancy, when running from a clean working directory, but only when multiparticle tracking using Partran is selected.

Unfortunately, I have a very large lattice, and the effect accumulates over a long portion of the beam line, so I need to provide a significant lattice file in order for you to reproduce this effect. I shortened the lattice as much as possible (attached). There are a very large number of 3D field maps, which I have placed together in "fields.zip".

Thanks,
Chad

[Didier]

Dear Chad,

Using your example, I don't able to see any discrepancy !

Capture.png (1.36 MiB) Viewed 672 times

Regards,

Didier

[cemitch]

Dear Didier,

Thanks; this discrepancy has now been resolved. Although I was running the code using Partran, the default output "Results from" selected in the "Charts" menu was "Envelope". After changing this selection to "Partran", the four workflows described above give consistent results.

In the previous runs, the Envelope and Partran results were being compared to one another.

Chad

[Didier]

Dear Chad,

TraceWin allows you to do a lot of things, perhaps too much today, making it confusing and complicated to understand the whole system.

Regards,

Didier