Re: [fluka-discuss]: Interactions at surface between air and air

From: Luigi Salvatore Esposito <luigi.salvatore.esposito_at_cern.ch>
Date: Mon, 5 May 2014 12:00:19 +0000

Hi Killian,
maybe it is superfluous, but let me comment that the 80 MeV kinetic energy that you have set is *per nuclear unit mass*,
as it is stated in the FLUKA manual (BEAM card, WHAT(1)).

Luigi


On 5 May 2014, at 13:23, Vittorio Boccone <dr.vittorio.boccone_at_ieee.org<mailto:dr.vittorio.boccone_at_ieee.org>> wrote:

Hi Kilian

So I wonder, what exactly is happening around the surface between the sourounding air and the air in the region. Are there any effects, because FLUAK is perhaps switching between multi- and single-scattering? Or is Fluka adapting path.lengths of the particles, when being near to a regionborder?

Yes FLUKA is doing it, but you can fully control this with the MCS parameters (see the manual for more details).

I have a doubt on what exactly are you plotting on the Y-axis of the 1D graph? Is it a slice of the USRBIN, the central bin or what?

But I think this is not your problem.

Your curves for 5E5 events have large errors; the curves for 1E7 are compatible with the both the 5E5 ones and, if you look carefully, the curves for 1E7 primaries are in the middle of both curves realized with 5E5 primaries.

The position of the peak tells you that the energy is the same e

You can try to run again one of the case with a different random seed and try to see which are the effects. Be sure to average a few bins around the beam.

Best,
V.










On Mon, May 5, 2014 at 12:35 PM, Kilian Baumann <baumann.kilian_at_gmail.com<mailto:baumann.kilian_at_gmail.com>> wrote:
Dear Vottorio,

thank you very much for your answer!
I have attached the input-file. As you can see, I measure the dose along the beam-axis in a waterphantom. In front of this phantom there is a region (1 cm thick, called "ripple") containing air. The sourounding of my set-up is also air. So technically, this region should have no influence at all.
In a second set-up I have just removed this region. The rest remains the same.
When using <1E05 primaries (80 MeV cabon-ions) the depth-doses for these two set-ups do not match within their errors around the bragg-peak. The more primaries I use, the smaller the diefference gets until, when using >1E07 primaries there is almost no difference left. (I have attached the graphs. For a better overview there are only a few error-bars).
So I wonder, what exactly is happening around the surface between the sourounding air and the air in the region. Are there any effects, because FLUAK is perhaps switching between multi- and single-scattering? Or is Fluka adapting path.lengths of the particles, when being near to a regionborder?

Thanking you very much in advance!

Best regards,
Kilian


2014-04-30 23:13 GMT+02:00 Vittorio Boccone <dr.vittorio.boccone_at_ieee.org<mailto:dr.vittorio.boccone_at_ieee.org>>:

Dear Kilian,


When assigning the material of the
region to air, it should be as if there is no region at all (because
the enviroment and the region consist of the same material).
You assign it to air. Was it vacuum before or just the geometrical representation of the phantom was different?
Which sizes are we talking about (meters, cm, mm or um?)

However, the depthdose-curve (for <1E05 primaries) in the water-phantom is significantly different to the depthdose-curve, when there is no
region.
What do you meas about significantly different? Did you check the statistical error of the simulation?

However this effect gets lost, when using >1E07 primaries.
This is a sign that you are very much dominated by the statistical error with less then 10 Millions of particles. This is realistic because the dose in air requires typically a lot of statistics or larger averaging bins/regions.

Your input file would be probably useful to check which are your physics setting and the scoring you are performing.

Best,
V.


On Wed, Apr 30, 2014 at 10:05 PM, Kilian Baumann <baumann.kilian_at_gmail.com<mailto:baumann.kilian_at_gmail.com>> wrote:
Dear Fluka-experts!

I have discovered a (perhaps) strange result, when analysing the
depthdose-curve and fluence, when irradiating a waterphantom with a
region in front of it. My whole setup consists of the beam source (80
MeV carbon-ions), a water phantom and a region between waterphantmom
and source. The enviroment is air. When assigning the material of the
region to air, it should be as if there is no region at all (because
the enviroment and the region consist of the same material). However,
the depthdose-curve (for <1E05 primaries) in the waterphantom is
significantly different to the depthdose-curve, when there is no
region. However this effect gets lost, when using >1E07 primaries.
Furthermore I analysed the fluence at the surface of the region, so to
say at the surface between air and air and indepent on the amount of
primaries one can see a change in fluence at the surface.
My question now is, what FLUKA is exacly calculating near this
surface/region (perhaps some artefacts with single-/multi-scattering)
and if any effects are negiglibly if using a sufficient amount of
primaries.
Thanking you in advance!
Best regards,
Kilian





Luigi

------------------------------------
Luigi Salvatore ESPOSITO
CERN, EN-STI-EET
tel: +41 22 76 79132
fax: +41 22 76 68495
skype: luillo76
Received on Mon May 05 2014 - 15:14:22 CEST

This archive was generated by hypermail 2.3.0 : Mon May 05 2014 - 15:14:23 CEST