The OPAL Discussion Forum

[Christof Metzger-Kraus <christof.j.kraus AT gmail.com>]

I wouldn't generalize this. There are certainly cases where it makes sense to deviate from this rule of thumb.

Christof 

Adelmann Andreas (PSI) <andreas.adelmann AT psi.ch> schrieb am Do., 19. Dez. 2019, 21:26:

Hi Anastasia IMO the field map should always start at 0.0 and by

using ELEMENTEGDE you can move the cavity very transparently.

 

Cheers A

------
Dr. sc. math. Andreas (Andy) Adelmann
Head a.i. Labor for Scientific Computing and Modelling 
Paul Scherrer Institut OHSA/ CH-5232 Villigen PSI
Phone Office: xx41 56 310 42 33 Fax: xx41 56 310 31 91
Phone Home: xx41 62 891 91 44
-------------------------------------------------------
Friday: ETH HPK G 28   +41 44 633 3076
============================================
The more exotic, the more abstract the knowledge, 
the more profound will be its consequences.
Leon Lederman 
============================================

On 19 Dec 2019, at 20:34, Anastasia Ierides <Anastasia.Ierides AT varian.com> wrote:

Hi Christoff,

 

It looks like that fixed the problem. I guess I was a bit unclear as to the definitions of the physical dimensions of the electric field map and the element edge definitions.

 

Thank you for your help and Happy Holidays!

 

Anastasia

 

From: Christof Metzger-Kraus <christof.j.kraus AT gmail.com> 
Sent: Thursday, December 19, 2019 10:17 AM
To: Anastasia Ierides <Anastasia.Ierides AT varian.com>
Cc: Adelmann Andreas (PSI) <andreas.adelmann AT psi.ch>; opal <opal AT lists.psi.ch>
Subject: Re: [Opal] Extra travel distance added to reference particle data between subsequent attached RF cavities

 

The problem, I think, is that the fields are both shifted in the input file as well as in the field map files. Try to set ELEMEDGE = 0 for all cavities.

 

Cheers

Christof 

 

Anastasia Ierides <Anastasia.Ierides AT varian.com> schrieb am Do., 19. Dez. 2019, 19:03:

Hi Christof,

 

The header for the full geometry is:

 

“1DDynamic 240

1.000000e-02 2.369820e+01 499999

2856

0 2.200000e-01 199

-3.818347e-06

-3.822820e-06

-3.827294e-06

-3.831770e-06

-3.836248e-06”

 

And for the separated fields they are

 

1DDynamic 240

0 2.32400e+00 100000

2856

0 3.1100000e-01 199

-3.818347e-06

-3.820531e-06

-3.822716e-06

-3.824901e-06

-3.827087e-06

-3.829273e-06

 

 

1DDynamic 240

2.324100e+00 6.972300e+00 100000

2856

0 3.1100000e-01 199

-2.915139e-04

-2.914721e-04

-2.914302e-04

-2.913883e-04

-2.913464e-04

-2.913046e-04

 

1DDynamic 240

6.972400e+00 1.22200e+01 100000

2856

0 3.110000e-01 199

9.423340e-07

9.413692e-07

9.404042e-07

9.394390e-07

9.384737e-07

9.375081e-07

 

1DDynamic 240

1.22210e+01 1.746760e+01 100000

2856

0 3.110000e-01 199

-4.077816e-07

-4.074007e-07

-4.070200e-07

-4.066393e-07

-4.062588e-07

-4.058784e-07

 

1DDynamic 240

1.74686e+01 2.37000e+01 100000

2856

0 3.110000e-01 199

1.134056e-06

1.133357e-06

1.132658e-06

1.131959e-06

1.131260e-06

1.130561e-06

 

Thanks,

 

Anastasia

 

 

From: Christof Metzger-Kraus <christof.j.kraus AT gmail.com> 
Sent: Thursday, December 19, 2019 9:53 AM
To: Anastasia Ierides <Anastasia.Ierides AT varian.com>
Cc: Adelmann Andreas (PSI) <andreas.adelmann AT psi.ch>; opal <opal AT lists.psi.ch>
Subject: Re: [Opal] Extra travel distance added to reference particle data between subsequent attached RF cavities

 

Hi Anastasia, 

 

could you send us the header of the field map files (first 10 lines or so)? This should be sufficient for now. 

 

Cheers

Christof 

Anastasia Ierides <Anastasia.Ierides AT varian.com> schrieb am Do., 19. Dez. 2019, 18:42:

Hi Andreas,

Attached you will find my input files for both cases (with the full linac geometry and the separated linac geometry) as well as a plot of the expected (red) normalized electric field felt by the reference particle, versus what is outputted by OPAL (blue), where the geometry seems to be extended to almost twice the distance. I have looked at the initial distribution file (LINAC1D_DIST.dat) and it shows a spatial increase between the separated cavities, even though the cavities are specified to be one right after the other. I have not attached the field map files here as they are a bit large. Will you need the field map files as well?

Thanks,

Anastasia

-----Original Message-----
From: Adelmann Andreas (PSI) <andreas.adelmann AT psi.ch> 
Sent: Wednesday, December 18, 2019 10:49 PM
To: Anastasia Ierides <Anastasia.Ierides AT varian.com>
Cc: opal AT lists.psi.ch
Subject: Re: [Opal] Extra travel distance added to reference particle data between subsequent attached RF cavities

Hi Anastasia can you attach a example, showing your problem?
A

> On 19 Dec 2019, at 02:51, Anastasia <anastasia.ierides AT varian.com> wrote:
> 
> Hello all,
> 
> As a tutorial, I am trying to model a linear accelerator as a beam 
> line of separate RF cavities rather than the whole LINAC geometry as 
> one RFCavity element. I have run the simulation both with a single 
> RFCavity element and with multiple cavities, wherein the dimensions 
> are appropriately defined in the field map files, such that the 
> elements are right next to each other with no space in between.
> 
> In the single element simulation, the refence particle electric field 
> is observed to depict expected behavior in the defined geometry, but 
> in the case of separate elements, there seems to be an additional 
> “space” that is added between the individual elements, extending the 
> overall geometry to almost twice the length. Could this be a mistake 
> on my part in my input files, etc., or is there something in the code 
> that by default forces this type of behavior? For instance, in a 
> situation where elements are too close together, are they forced to have some minimal distance between them, etc.?
> 
> I guess the first question should be, does OPAL allow for the 
> capability to model a whole geometry as separate, yet connected RFCavity elements?
> 
> Any help in this would be greatly appreciated.
> 
> Thanks!