RE: [fluka-discuss]: Geometry Error

From: <hayley.smith_at_stfc.ac.uk>
Date: Sun, 6 Oct 2013 11:02:17 +0000

Ah, thank you very much for this Mary, this is really helpful.
I had been reading various sets of lecture notes online but had been struggling with the concept of zones. Your explanation and example have been really useful and helped me to get in the correct mindset.
Thanks once again, much appreciated.
Hayley

________________________________________
From: me_at_marychin.org [me_at_marychin.org]
Sent: 05 October 2013 17:00
To: Smith, Hayley (STFC,RAL,ISIS); fluka-discuss_at_fluka.org
Subject: RE: [fluka-discuss]: Geometry Error

An example might be handy. Say I have a cube bounded by 6 sides, my body definition section would have:

XYP z_lo -10.
XYP z_hi 10.
YZP x_lo -10.
YZP x_hi 10.
XZP y_lo -10.
XZP y_hi 10.

My region definition for the cube would be:

-x_lo +x_hi -y_lo +y_hi -z_lo +z_hi

Now, for the surrounding (analogous to your 'void') region if I think in terms of subtracting regions I would have:

+void -(-x_lo +x_hi -y_lo +y_hi -z_lo +z_hi )

If I think in terms of zones I would have:

+void +x_lo | +void -x_hi | +void +y_lo |+void -y_hi |+void +z_lo |+void -z_hi

no expansion in this case.

:) mary

On 05 October 2013 at 15:52 hayley.smith_at_stfc.ac.uk wrote:
> Dear Mary
>
> Thank you very much for your reply.
>
> I understand that the way I have this defined is making a lot of expansions and can see this in the output files.
> However, I am struggling to see how this can be simplified, as you suggest.
>
> From what I understand, only bodies can be subtracted, rather than regions. So I struggled to simplify in this sense. In the future my geometry will require twenty of these regions defined by planes, so the expansion would be even worse. In your experience what would you suggest as a simplification tactic?
>
> I tried making multiple voids, but this just propagated the expansion problem up to the black body. I think I am still missing some important concepts.
>
> Thanks for any further help you can give,
> Hayley
>
>
> ________________________________________
> From: owner-fluka-discuss_at_mi.infn.it [owner-fluka-discuss_at_mi.infn.it] on behalf of me_at_marychin.org [me_at_marychin.org]
> Sent: 04 October 2013 17:00
> To: Smith, Hayley (STFC,RAL,ISIS); fluka-discuss_at_fluka.org
> Subject: Re: [fluka-discuss]: Geometry Error
>
> Dear Hayley,
>
> The culprit is in the region definition of VOID. Too many expansions; fluka crashes. The debugger works by calling a 'dry' fluka run. So even if you do not use the debugger, fluka still will not run. To solve the problem you can simplify that line of region definition for VOID (which can be easily done using the geometry editor in flair).
>
> :) mary
>
> On 02 October 2013 at 13:49 hayley.smith_at_stfc.ac.uk wrote:
> Dear FLUKA users
>
> I am a relatively new user and have been having some difficulties creating the geometry for my current problem.
>
> I have an assembly of twenty graphite blocks, none of which are cubes. Each block has at least one slanted face. I have been trying to create these by using the generic infinite planes and combining these to describe individual blocks as regions. (From what I have read using PLA is the best tool for this?)
>
> I am able to create, debug and plot the geometry for individual blocks.
> However, when I wish to debug and plot the geometry of multiple blocks at once the debug routine slows down dramatically, and in the case of wanting to debug more than four of these blocks I receive the error “ERROR RC=34304”.
>
> I have been trying for a number of days to figure out my problem but have now run out of ideas. The geometry does not seem complex to me, so I was surprised that I encountered such an error. Am I asking too much of the memory capabilities? (I am working in Flair on FLUPIX with the standard installation settings as advised on the website).
>
> I have already tried to implement some ideas from the thread http://www.fluka.org/web_archive/earchive/new-fluka-discuss/5131.html however I haven’t had any luck, and can’t really see if my problem is even the same as the problem in that thread.
>
> I attach my input file, showing just a few of these blocks for which the debug crashes, in case it is required for further understanding of my problem.
> I am convinced I must be doing something significantly wrong, but am just not able to see it…
>
> Many thanks for any insight you are able to give
>
> Hayley
>
>
> --
> Scanned by iCritical.
>
>
> --
> Scanned by iCritical.
-- 
Scanned by iCritical.
Received on Sun Oct 06 2013 - 14:07:19 CEST

This archive was generated by hypermail 2.3.0 : Sun Oct 06 2013 - 14:07:25 CEST