Gas Scattering lifetime

tiansk · Post by **tiansk** » 11 Apr 2018, 03:42

Hi Michael：
In the Version 33.1.1 of ELEGANT,the elastic_scattering command was added.I used this command to get the elastic scattering lifttime.I have two questions about the command:
1.I add the MARK in different coordinates and different types element,the lifetimes I got are different so much？
2.I wonder the particles lost only on the Physical Aperture,does the Dynamic Aperture affect the tracking results？

The attachment：
bend.gif：scattering angles (xp, yp) for bend element location
quad.gif：scattering angles (xp, yp) for quad element location
Physical Aperture1.jpg：transverse loss coordinates with(tube: maxamp,x_max=0.011,y_max=0.011 and kap: ecol,x_max=4.5e-3,y_max=2.5e-3)
Physical Aperture2.jpg：transverse loss coordinates with(tube: maxamp,x_max=0.011,y_max=0.011)

DA.jpg:
the black line:the Dynamic Aperture found with the command find_aperture
the red line: the elliptical collimator I insert in the lattice file
the scatter plot was the Dynamic Aperture found by particle tracking

some issue about the inelastic_scattering command:

In Manual for elegant,there are two parameters that describe the brehmsstrahlung photon energy(delta_min and delta_max or(k_min and k_max)),but when I set k_max or delta_max，the elegant turns out： unknown variable k_max used in namelist inelastic_scattering。When I check the output losses file，I found the default value of k_max or delta_max may be 1(or -1).

I want to know ：the initial delta between the (k_min,k_max) was not uniform distribution？The deltaLost I got was show in the The attachment DA.jpg. The line is the result of LMA I got with the momentum_aperture command.

In the inelastic_scattering simulation,I found the number of particles only decrease in the first about 10 turns,Should I set the n_passes number large enough in the run_control command??

Could you upload some examples about the gas scattering lifetime simulation about elastic_scattering and inelastic_scattering?

Thanks,
-tiansk

michael_borland · Post by **michael_borland** » 12 Apr 2018, 08:54

Tiansk,

Attached you'll find two examples, one for elastic scattering and another for inelastic scattering. The examples show how to do the tracking and postprocessing to get lifetime and loss distributions.

In answer to your specific questions:
1. You should track for many turns in these simulations to ensure that you see both physical and dynamic aperture effects. If you only track a few turns, you won't see all the effects of nonlinear dynamics.
2. The spacing of points in angle and k is non-uniform, reflecting the decreasing importance of increasingly large scattering events. This gives better statistics.

--Michael

tiansk · Post by **tiansk** » 16 Apr 2018, 22:23

Dear Michael,

I run the example files that you upload I got the deltaLost and LMA file,I plot the LMA and deltaLost in a picture，they matched well. But I run the my files,the deltaLost and LMA didn't match so well，there are some small datas I got，shown in the picture.I think the min(abs(deltaLost))will be used in the lifetime calculation,so I'm using this data to get the lifetime is reliable?

tiansk

tiansk · Post by **tiansk** » 16 Apr 2018, 23:11

Dear Michael,

if I set the s_end = 56.683333334200107，when I only track one cell，I get the deltaLost VS. LMA is well.

michael_borland · Post by **michael_borland** » 17 Apr 2018, 08:38

tiansk,

You shouldn't expect deltaLost to agree with the LMA (deltaPositive and deltaNegative). deltaLost is the delta value when the particle gets lost. In contrast, deltaNegative and deltaPositive are the *initial* delta values of the particle. If the particle is executing synchrotron oscillations and perhaps encountering betatron resonances as the tune varies (due to chromatic effects), then it may be that when it is finally lost, delta is different. The only case where you expect deltaLost to match the LMA is when there is no synchrotron radiation, no acceleration, and no synchrotron oscillations.

--Michael

tiansk · Post by **tiansk** » 17 Apr 2018, 20:43

Dear Michael,

I'm sorry，I write the column name wrong，it‘s not deltaLost column ，but the delta column. I used the delta column data to get the picture!

tiansk

michael_borland · Post by **michael_borland** » 18 Apr 2018, 09:38

tiansk,

The second case looks more reasonable, but I don't know why the first case (without setting s_end) looks the way it does. Can you post your input files?

--Michael

Pradeep · Post by **Pradeep** » 17 Oct 2023, 07:20

Dear Dr. Borland,
Sir, I tried this gas scattering lifetime example, while using Postprocess command as
elasticScatteringAnalysis -tracking run1.eslos -twiss run1.twi -output run1.esanal -pressure PARPressure2015-11-11.sdds -verbose 1

error message: unknown switch: tracking (screenshot attached)

similar error message for inelasticScatteringAnalysis
inelasticScatteringAnalysis -tracking run1.islos -twiss run1.twi -output run1.isanal -pressure PARPressure2015-11-11.sdds -verbose 1

Please help me how this error can be removed?
Thanks and regards,
Pradeep

michael_borland · Post by **michael_borland** » 25 Oct 2023, 12:06

Pradeep,

There's an up-to-date example in the examples file, which you can obtain from our software download page, or at https://ops.aps.anl.gov/cgi-bin/oagLog4 ... les.tar.gz

Look for PAR/elasticScatteringTracking and PAR/inelasticScatteringTracking.

--Michael

Elegant users forum