Skip to Content.
Sympa Menu

opal - Re: [Opal] Dynamic timestep in OPAL-t

opal AT lists.psi.ch

Subject: The OPAL Discussion Forum

List archive

Re: [Opal] Dynamic timestep in OPAL-t


Chronological Thread 
  • From: Christof Metzger-Kraus <christof.j.kraus AT gmail.com>
  • To: Simon Friederich <sifriede AT uni-mainz.de>
  • Cc: "opal AT lists.psi.ch" <opal AT lists.psi.ch>
  • Subject: Re: [Opal] Dynamic timestep in OPAL-t
  • Date: Thu, 13 Aug 2020 11:29:52 +0200
  • Authentication-results: localhost; iprev=pass (mail-io1-f52.google.com) smtp.remote-ip=209.85.166.52; spf=pass smtp.mailfrom=gmail.com; dkim=pass header.d=gmail.com header.s=20161025 header.a=rsa-sha256; dmarc=pass header.from=gmail.com

Hi Simon,

unfortunately we don't have adaptive time steps in Opal-T yet. We had a student who implemented adaptive time steps in Opal-cycl (but has been removed). Reimplementing it is (and has been for a long time) on our to-do list (https://gitlab.psi.ch/OPAL/src/-/issues/1). Unfortunately I can't give you any hope that this feature will be implemented any time soon. As long as we don't have more developers contributing code on a regular basis, major changes will be postponed. This is especially true for Opal-T, which I maintain completely in my spare time.

Until adaptive time steps are implemented you have to change the time steps manually using the arrays DT, ZSTOP and MAXSTEPS.

Best regards
Christof

On Thu, Aug 13, 2020 at 10:47 AM Simon Friederich <sifriede AT uni-mainz.de> wrote:
Dear OPAL-Team,

Concerning OPAL-t:

If I understood correctly, the time step at the emission process for the
DISTRIBUTION can be adjusted using EMISSIONSTEPS.

I was wondering, if the time step can be dynamically adjusted (in a
similar manner as for emitted distributions), when the bunch passes a
non-drift element (e.g. magnets, cavities, etc) in order to avoid the
pitfall you've pointed out in section 4.2. (time step is too big leads
to an inaccurate calculation of the trajectory).

More concrete: My beam line is roughly 10m, where a lot of drift space
is necessary. I wanted to use ~50ps for the drifts but ~10ps for the
other elements, if this was possible.

Thank you very much as always in advance!

Best regards
Simon

P.S.:
The new relative positioning of elements with respect to the first beam
line they are assigned to (in OPAL 2.4) works like a charm. Thank you
for this feature!
I also want to thank Achim Gsell for updating the "Compile-OPAL" wiki
entry. The new remarks helped me compiling OPAL 2.4. much quicker!
However there is one thing I'd suggest, if you're also targeting a
"not-so-advanced-linux-user"-group (some beginners don't get to use OPAL
because they are not able to compile it properly I think):
A quick step-by-step tutorial using Ubuntu (as it is one of the mostly
used distributions). If I'm not mistaken one can then skip some of the
toolchain modules by using e.g.  "sudo apt-get install build-essential".
This tutorial should set variables like "${SRC_DIR}" directly (like:
export SRC_DIR=/home/myusername/opal_src;and maybe skip optional aspects
like AMG, trilionos, etc. if you think they are not essential to run OPAL).

--
-----------------------------------
Dr. Simon Friederich
Helmholtz-Institut Mainz
Kollaboration B/ACID
Johannes Gutenberg-Universität Mainz
Staudingerweg 18
55128 Mainz, Deutschland

Tel.: +49 (0)6131 39-23160
E-Mail: sifriede AT uni-mainz.de
----------------------------------------




Archive powered by MHonArc 2.6.19.

Top of Page