From: Anna Ferrari <a.ferrari_at_hzdr.de>

Date: Mon, 16 Apr 2012 04:54:34 +0200

Date: Mon, 16 Apr 2012 04:54:34 +0200

Hallo,

I would like to come back just a bit on this argument because the method of

the total tracklength is really very effective!

I did some test and I saw that the convergence is very rapid: with 5x10^7

events -which are processed in few minutes- I could reach a precision of

~1/10^6 if the volume to be measured is of the same order of magnitude of the

sphere test.

I think it can become our standard method...

Last: maybe it is useful to stress the importance of the cosine weight in the

inwards-directed source distribution (easily described via the BEAMPOS card

with SDUM=FLOOD): a sampling from a pure half-isotropic distribution, also

inwards directed, would not work, because it would not describe correctly the

calculation of the average chord of the sphere (and it would not guarantee a

uniform tracklength density inside the volume, condition that is essential to

apply the proportion with the inner volumes we have to measure).

Ciao to everybody,

Anna

Am Fri, 13 Apr 2012 17:16:34 +0200 schrieb Jean-Emmanuel Groetz

<jegroetz_at_univ-fcomte.fr>:

*> Alberto,
*

*>
*

*> Here is the part (as attachment) of the MCNP4C2 documentation related to
*

*> the calculation of volumes. You will find at the end the ray-tracing
*

*> method for stochastic volume.
*

*>
*

*> On the MCNPX forum, I just find a similar trick (from Sergio
*

*> Korochinsky), which follows below :
*

*>
*

*>
*

*> --------------------------------------------------------------------------
*

*> Consider a spherical cell.
*

*>F4 tallies the total track length in the cell.
*

*> Before printing out, MCNP normalizes it with the VOLume of the cell
*

*> and
*

*> the NPS in order to approximate the average flux in the cell.
*

*> Now consider this:
*

*> 1- If the material is VOID, there is no interaction of the particle
*

*> within the cell and the track lengths are actually cords of the cell.
*

*> 2- If the flux at your spherical surface is ISOTROPIC, the track
*

*> lengths
*

*> go from any point on the sphere to any other point on the sphere, and
*

*> therefore every possible cord is sampled.
*

*> 3- Dividing the total track length (sumation of all the cords
*

*> uniformly
*

*> sampled) by NPS gives you the average cord of the sphere, which is
*

*> equal
*

*> to 4*vol/sur = 4*r/3. To get this in the print out you must avoid
*

*> dividing the tally by the volume of the cell (which in any case you
*

*> don't
*

*> know!).
*

*> 4- If you artificially renormalize the tally multiplying it by pi*r^2,
*

*> the tally gives you the volume of the sphere!
*

*> 5- I leave it for you, as an exercise, to demonstrate that if this is
*

*> valid for the sphere then it's also valid for any body included in the
*

*> sphere.
*

*>
*

*> So, to implement the method you have to:
*

*> 1- Use the card VOID and set all the importances to 1 (any internal
*

*> region with imp=0 would create an artificial "absorption" and
*

*> therefore
*

*> cut the cord).
*

*> 2- Create an inward cosine current at the spherical surface source.
*

*> 3- Set the volume of the tally to 1, either with the cards VOL or SD.
*

*> 4- Renormalize the tally by pi*r^2 either with the cards FM or wgt in
*

*> SDEF.
*

*>
*

*>
*

*>
*

*> --
*

*> _______________________________________________
*

*> Jean-Emmanuel Groetz
*

*> Laboratoire Chrono-Environnement UMR CNRS 6249
*

*> Chimie-Physique & Rayonnements
*

*> Université de Franche-Comté
*

*> 16 route de Gray
*

*> 25030 Besançon Cedex
*

*>France
*

*>
*

*> mailto : jegroetz_at_univ-fcomte.fr
*

*> tel : +33 (0) 3 81 66 65 07
*

*> fax : +33 (0) 3 81 66 65 22
*

*> _______________________________________________
*

*>
*

*>
*

*> Le vendredi 13 avril 2012 ? 16:38 +0200, Alberto Fasso' a écrit :
*

*>> Hi Jean-Emmanuel,
*

*>>
*

*>> I don't know how does MCNP calculate the volumes. Do you have any
*

*>>information
*

*>> about it?
*

*>>
*

*>> Alberto
*

*>>
*

*>>
*

*>> On Fri, 13 Apr 2012, Jean-Emmanuel Groetz wrote:
*

*>>
*

*>> > Dear all,
*

*>> >
*

*>> > In addition to previous messages from Mina and Chris, volume calculation
*

*>> > could still remain a problem:
*

*>> > - for FLUKA/FLAIR users with Linux OS, SimpleGeo works very well with
*

*>> > Windows under the VirtualBox software, without any difficulty. The
*

*>> > geometry file could be imported between both OS through shared folders
*

*>> > - once before with a complex geometry (concentric spheres and tori),
*

*>> > MCNP was unable to assess the region volume for an energy deposition
*

*>> > tally. So I had to calculate every region volume...
*

*>> >
*

*>> > The method described by Anna (from Alberto) seems to be neat.
*

*>> >
*

*>> > Jean-Emmanuel
*

*>> >
*

*>> >
*

*>> >> Dear Walker,
*

*>> >> SimpleGeo offers two kinds of volume calcultions. The one is called
*

*>> >> "geometric" and uses the visualized and as such
*

*>> >> discretized geometry. As a consequence it is very fast but might
*

*>> >> fail in case of
*

*>> >> rendering artefacts. There is another one which is called "Quasi
*

*>> >> Monte Carlo" which calculates the volume integral by using stratified
*

*>> >> numerical sequences which act on the analytic description of the
*

*>>geometry.
*

*>> >> Therefore, this method also works in the presence of rendering artefacts.
*

*>> >> If you are looking for the functionality to calculate volumes of
*

*>>arbitrary
*

*>> >> regions directly in your input, like it is provided for example in
*

*>>MCNP(X),
*

*>> >> then I'm afraid that FLUKA itself does not provide such a feature
*

*>> >> - at least not to my present knowledge. Eventually you can do it but you
*

*>> >> will need to do a bit of programming or set up a specific geometry, input
*

*>> >> and post-processing to achieve this.
*

*>> >> Hope that helps
*

*>> >> Chris
*

*>> >>
*

*>>--------------------------------------------------------------------------=
*

*>> > --------------
*

*>> >> Chris Theis
*

*>> >> CERN/DGS-RP - European Organization for Nuclear Research
*

*>> >> 1211 Geneva 23, Switzerland
*

*>> >> Phone: +41 22 767 8069 Office: 892-2A-015
*

*>> >> e-mail: Christian.Theis_at_cern.ch<mailto:Christian.Theis_at_cern.ch>
*

*>> >> www: http://www.cern.ch/theis
*

*>> >>
*

*>>--------------------------------------------------------------------------=
*

*>> > ---------------
*

*>
*

*>>
*

Received on Mon Apr 16 2012 - 09:26:31 CEST

*
This archive was generated by hypermail 2.2.0
: Mon Apr 16 2012 - 09:26:43 CEST
*