Re: Re: [fluka-discuss]: 10 MeV proton+Si, biasing neutron

From: 毕远杰 <biyuanjie_at_tsinghua.org.cn>
Date: Mon, 30 Sep 2013 15:58:55 +0800 (CST)

Dear Mario,

I never use weight window commands before. I read the lecture about biasing in the FLUKA course and still feel confused. It even said 'it requires also more experience and patience to set it up correctly'. So could you explain more to me about how to use it to stabilize the weight? Many thanks.

Best wishes
Yuanjie


> -----原始邮件-----
> 发件人: "Santana, Mario" <msantana_at_slac.stanford.edu>
> 发送时间: 2013年9月28日 星期六
> 收件人: "毕远杰" <biyuanjie_at_tsinghua.org.cn>, fluka-discuss <fluka-discuss_at_fluka.org>
> 抄送:
> 主题: Re: [fluka-discuss]: 10 MeV proton+Si, biasing neutron
>
> Hi Yuanjie,
>
> 1) a factor of 0.005 may be OK, maybe towards the low side. Typically I
> use 0.001 to 0.05, but it will depend on the type of reaction and
> specifics of your simulation. In the past I saw some simulations crashing
> when the factor was too low (e.g. 0.005 for photo nuclear reactions in
> thin metals). According to the manual:
>
> '8) Reduction factors excessively large can result in an abnormal increase
> of the number of secondaries to be loaded on the stack, especially at high
> primary energies. In such cases, FLUKA issues a message that the secondary
> could not be loaded because of a lack of space. The weight adjustment is
> modified accordingly(therefore the results are not affected) but if the
> number of messages exceeds a certain limit, the run is terminated."
>
> So check the output file to make sure the run will not be terminated once
> you shoot for a long simulation
>
> 2) You already have a geometrical biasing step factor of 4. All you could
> do is increase it to 5. If you increase it anyway further FLUKA will
> reduce the factor to the maximum (5). Therefore you don't have much
> potential gain from that side. I'd suggest that you split your shielding
> in more layers. Ideally, you would need to choose small enough layers so
> as to compensate for the neutron attenuation in the shielding with the
> geometric splitting so that the population of neutrons (unweighted) would
> stay constant.
> Please note that when using strong biasing you will have large weight
> dispersions so you may want to consider stabilizing your weight through
> the weight window commands, otherwise the simulation may not be efficient
> and the statistics may get ruined by tracks with very high importance.
>
> Mario
>
> On 9/27/13 12:50 AM, "毕远杰" <biyuanjie_at_tsinghua.org.cn> wrote:
>
> >Dear fluka users
> >
> >I have 10 MeV proton+Si target, and a shielding of 50 cm concrete in the
> >front. Since the low neutron yield( About 3.6e-6 particle/sr in the
> >forward), I need to biasing the neutron.
> >1) I use LAM-BIAS with a factor a 0.005, is that OK?
> >
> >2) I divide the shield to 5 pieces, and use region biasing for it. But
> >still few neutron at the end of the shielding for 1e9 primary. Should I
> >increse the basing factor?
> >
> >BIASING 0.0 4. shied1
> >BIASING 0.0 16. shied2
> >BIASING 0.0 64. shied3
> >BIASING 0.0 256. shied4
> >BIASING 0.0 1024. shied5
> >LAM-BIAS 0.005 SILICON PROTON
> >INEALL
> >LAM-BIAS 0.005 SILICON NEUTRON
> >INEALL
> >
> >Best wishes
> >Yuanjie
> >
>
Received on Mon Sep 30 2013 - 11:19:31 CEST