The OPAL Discussion Forum

[Huiwen Koay <koay AT rcnp.osaka-u.ac.jp>]

Dear Jochem and Matthias,

I finally managed get the estimated bunch size. Thanks for your help.

Anyway, I am still looking for methods to convert Efield input from ascii to h5part. 

Do you have any suggestions?

Thanks again.

Best regards,

Huiwen

On Wed, Jun 26, 2019 at 10:29 PM Snuverink Jochem (PSI) <jochem.snuverink AT psi.ch> wrote:

As Matthias pointed out to me, it should be:

y - longitudinal

Best regards,

Jochem

On 26 Jun 2019, at 15:19, Snuverink Jochem (PSI) <jochem.snuverink AT psi.ch> wrote:

Dear Huiwen,

Matthias and I can reproduce your finding. 

The local frame (OPTION, PSDUMPFRAME=bunch_mean) in the stat file is always:

x - horizontal
y - radial
z - vertical

I suppose your radial momentum (PRINIT attribute of the CYCLOTRON) is not 0?

In that case the local frame is rotated with respect to the initial gaussian input distribution, so RMSx (in the rotated local frame) will no longer be 0.

Best regards,

Jochem

On 26 Jun 2019, at 04:16, Huiwen Koay <koay AT rcnp.osaka-u.ac.jp> wrote:

Dear Matthias,

Thank you for your prompt response.

In your reply, RMSx = sqrt( <x^2> - <x>^2 ). So if I set the initial sigmax=0, I should have the initial RMSx=0. 

However, it is not so. As long as sigmay is not zero, RMSx will not be zero.

It seems to me that 

RMSx = sqrt( <x^2> - <x>^2 ) + f(y)

I am quite confused by this. Hope that you can help in clarifying this matter.

Best regards,

Huiwen

On Wed, Jun 26, 2019 at 3:01 AM Frey Matthias (PSI) <matthias.frey AT psi.ch> wrote:

Dear Huiwen,

Regarding RMSx and RMSy. In OPAL-cycl, x is horizontal and y is longitudinal direction. Sure, when you change sigmax of the initial

distribution you can have a different RMSy due to transversal-longitudinal couplings.

RMS is the root-mean square, e.g. in x-direction it is computed by

RMSx = sqrt( <x^2> - <x>^2 )

where <x> is the centroid and <x^2> the second order moment.

Best,

Matthias

---

Von: opal-request AT lists.psi.ch [opal-request AT lists.psi.ch]" im Auftrag von "Snuverink Jochem (PSI) [jochem.snuverink AT psi.ch]
Gesendet: Dienstag, 25. Juni 2019 17:00
An: Huiwen Koay
Cc: opal
Betreff: Re: [Opal] Questions regarding .stat output and rfmap input

Dear Huiwen,

Concerning questions about OPAL, it is best to use the email list as more users and members of the developer's team will see your message, and you'll get a quicker response.

1) @Matthias, can you answer that question perhaps?

2) I agree there is no reference in the Manual to the h5part format. That should be added and thanks for pointing that out. There is some information here: https://gitlab.psi.ch/H5hut/src/wikis/home

@Achim, can you point Huiwen to how he can convert his ASCII field map to a h5part file?

Best regards,

Jochem

On 25 Jun 2019, at 15:41, Huiwen Koay <koay AT rcnp.osaka-u.ac.jp> wrote:

Dear Jochem Snuverink,

Hi. I have previously sent email to inquire about OPAL input for B fieldmap. Currently I managed to simulate the single particle tracking using opal-cycl. However, when I move onto deeper calculations, such as bunch calculation and central region calculations, I face some troubles. I did send inquiries to some of the members in the mailing lists directly but I receive no reply. Thus, I would like to ask for help again here. I sincerely appreciate if anyone could help in clearing my confusion. 

1) In manual 2.1, the coordinate system of .stat output in opal-t is given. From my interpretation, the local frame in .stat file is similar to that defined in opal-cycl. However, when I changed sigmax in distribution input of type=GAUSS, both RMSx and RMSy changed. The RMSx and RMSy in .stat file seems not to be consistent with the X and Y defined in local frame of opal-cycl. So, in that case, I would like to know the definition of RMSx and RMSy given in the .stat file. 

2) I would like to use an eletric field map in OPAL for central region calculations. However, in type=BANDRF, the input format for rfmap is in .h5part. Regarding this, I would like to ask if you had any reference format of .h5part file so that I can translate my ascii field map to H5part?

Thank you very much for your help.

I am looking forward to your soon reply.

Best regards,

Huiwen