Re: [fluka-discuss]: Cross section of neutrons

From: Luigi Salvatore Esposito <luigi.salvatore.esposito_at_cern.ch>
Date: Wed, 1 Mar 2017 09:38:00 +0000

Dear Srinidhi,
to complement the answer by Andrew:
1. to understand how the statistical uncertainty is evaluated in FLUKA, refer to these slides<https://indico.cern.ch/event/540415/contributions/2194806/attachments/1285749/1912258/09_Statistics_and_sampling_2015.pdf> from last FLUKA course (in particular page 32 and following)
2. the volume in the USRTRACK seems to be not correct: 1.4cm*1.4cm*0.0055cm = 0.01078 cm3
Best regards, luigi

On 1 Mar 2017, at 05:38, Srinidhi Bheesette <srinidhi.bheesette_at_cern.ch<mailto:srinidhi.bheesette_at_cern.ch>> wrote:

Hi Andrew,

Thanks for the feedback:

1. I am sorry I forgot to mention it my mail. I am using a python script to run these simulation for 500000 times. (I could have specified this in the START card). But I did not understand why I need to run these with 5 independent random number seeds?

2. Is this URSTRACK card right?
<image.png>

I wish to score the energy deposited by protons on the SENSOR region with a dim (1.4cm*1.4cm*0.0055cm)

3. Sure, will post the plot with error bars shortly.

Thanks and hoping to her from you soon.

Cheers,
Shri

(Srinidhi Bheesette)

On 1 March 2017 at 15:26, Andrew Davis <andrew.davis_at_wisc.edu<mailto:andrew.davis_at_wisc.edu>> wrote:
Hi Srinidhi

There are a number of issues assuming that your input deck is representative

1) Your START card only contains a single history to run, from a single run of 1 primary you cannot make any assertions regarding the results of your calculation since you have a large statistical uncertainty. You should likely run several million histories with (at least) 5 independent random number seeds using the already established tools to perform the averaging.
2) You used a USRBIN score (with 1 bin for x-y-z) to determine the energy deposition, you might as well use a USRTRACK score, not an issue really, but I may’ve chosen a USRTRACK instead
3) Your plot doesn’t include any statistical error, its hard judge the accuracy of a Monte Carlo calculation without any mention of the statistical error.

Thanks

Andy

On Feb 28, 2017, at 4:57 PM, Srinidhi Bheesette <srinidhi.bheesette_at_cern.ch> wrote:

Hi all,

I have simulated a pixel detector (just a silicon sensor layer) in FLUKA and bombarding it with point neutron source at different energies from 1-25.5MHz with a step size of 0.5MHz. I run a single simulation.

I then score the energy deposited in the detector (one value for one run) and then plot a graph of the energy deposited on the x-axis the source energy on the y-axis shown below:

<image.png>

The energy seems to increase till 20MeV but then start fluctuating from 20.5 to 25.5MeV and also a spike at 17MeV

Am I missing some cards in my input file (attached) for activating some feature for neutron energies after 20MeV?

Please help.

Cheers,
Shri

(Srinidhi Bheesette)
<sensor-22.5MeV.inp>




__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info
Received on Wed Mar 01 2017 - 12:07:36 CET

This archive was generated by hypermail 2.3.0 : Wed Mar 01 2017 - 12:07:47 CET