RE: [fluka-discuss]: Cards needed for isotopes

From: Joachim Vollaire <joachim.vollaire_at_cern.ch>
Date: Fri, 24 Jun 2016 12:41:49 +0000

Hi Keith

By sequence will not work unfortunately. You can only specify a single extended volumetric source (one cylinder shell / one spherical shell or cartesian one… ) per input. As for the isotope in theory you could define 5 different inputs for the different locations or write your own source routine (source.f) to specify the location of the decay.

At first thought it is what I would do for the source routine. If you have 5 Cartesians volumes (5 boxes) of identical size with same activity, just generate a random number between [0,1[ * 5.
Use the floor function to get the index of the randomly chosen Cartesian volume. Sample 3 random numbers and use the coordinate/dimension of the sampled box (coordinate and dimension given with the SOURCE card or in an external file at initialization….) to select randomly the position within the box. At the end the normalization to use is the total activity in all volumes considered.

Cheers
Joachim


From: owner-fluka-discuss_at_mi.infn.it [mailto:owner-fluka-discuss_at_mi.infn.it] On Behalf Of Mikhail Polkovnikov <?????? ???????????>
Sent: 23 June 2016 18:37
To: Keith Welch <welch_at_jlab.org>; fluka-discuss_at_fluka.org
Subject: Re: [fluka-discuss]: Cards needed for isotopes

On 23.06.2016 16:14, Keith Welch wrote:
Joachim and Mikhail,

Thanks very much for your answers. I think the idea of individual runs (first approach) would be easiest for me. But I'm still confused about the spacial distribution issue. This is maybe a very elementary question, but I am just a novice user. If I have a radionuclide in five different locations, and each location is a simple cartesian volume, how do I designate the BEAMPOS cards to define these five volumes? Simply by sequence?

Thanks again,
Keith


________________________________
From: "Joachim Vollaire" <joachim.vollaire_at_cern.ch><mailto:joachim.vollaire_at_cern.ch>
To: fluka-discuss_at_fluka.org<mailto:fluka-discuss_at_fluka.org>
Sent: Thursday, June 23, 2016 6:15:50 AM
Subject: RE: [fluka-discuss]: Cards needed for isotopes

Hi Keith

1) First approach

If you limit yourself to the built-in source specifications capabilities then you can only request to simulate the decay of a single type of isotope as primary particles. With this approach then you would need for each type of isotope to run a different type of simulations. As I understand from your email to the list that you have 6 isotopes, then you would need 6 different inputs where you would specify for each the source details (concerned isotope, spatial distribution BEAM – BEAMPOS…)…

Then there is some post-processing work to be done as with flair you can plot the results showing the results for a single isotope (which you should normalize to the activity for that specific isotope in your mixture…). I don’t know any straightforward way to merge for example dose rate maps with flair, but you can have a look at my reply a few weeks ago to a person with a similar question for adding different runs.

http://www.fluka.org/web_archive/earchive/new-fluka-discuss/9194.html

Be careful however not to merge outputs corresponding to different isotopes with the FLUKA / flair tools (usbsuw) as first the calculated statistical uncertainty would have no meaning (based on the distribution of score corresponding to different physics case…) and second you would not be able to weight/normalize the different contributions to the activity of each isotope….

2) Second Approach
You use the source routine (SOURCE card) to call the routine each time the code needs to start a primary particle. You need to write your own source.f routine and create a new executable. You could read the information on your isotope at initialization for example in a data file (6 lines, Z = xx A = yy Activity/fraction = xx) and sample based on this distribution. You need to specify the IJHION value in the routine (look at the part of the routine concerning Radioactive Isotope…)…
Either you give each isotope a weight equal to unity and you sample based on the probability function (partial activity fraction) or you sample with the same probability each isotope but you adjust the weight based on the partial activity fraction. At the end the normalization will be done using the WEIPRI (total weight of primary particles generated)… The results will be normalized to one Bq of activity in your system (thus you need to normalize to the total activity considering each isotope…). For the spatial distribution you can use the BEAM / BEAMPOS parameters or write your own logic based on the sample isotope in case

Personally I would go with the first approach as at the end you have more information (individual contribution of each isotope….) which could be re-used if you are changing your mixture activity as the normalization is done at the post-processing level and there is less chance of user error in writing the source routine. Now if you have a (more) complicated mixture with complex a spatial distribution in maybe it is mandatory to use the source routine…

Cheers
Joachim


-----Original Message-----
From: owner-fluka-discuss_at_mi.infn.it<mailto:owner-fluka-discuss_at_mi.infn.it> [mailto:owner-fluka-discuss_at_mi.infn.it] On Behalf Of Mikhail Polkovnikov <?????? ???????????>
Sent: 23 June 2016 09:26
To: Keith Welch <welch_at_jlab.org><mailto:welch_at_jlab.org>; fluka-discuss_at_fluka.org<mailto:fluka-discuss_at_fluka.org>
Subject: Re: [fluka-discuss]: Cards needed for isotopes

On 22.06.2016 04:41, Keith Welch wrote:

Dear Keith,
> 1. Do I need a separate BEAM card for each isotope in the material?
> And/or separate BEAM card for each of the CART-VOL source regions?
> (If so, does each beam card need its own physics, start, etc. cards?)
> 2. Do I understand correctly that I need
You need only one BEAM, HI-PROPE pair of cards if you use only one isotope (nuclide), for example Co-60 will have:
*...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....
BEAM 10000.0 -1.0ISOTOPE
HI-PROPE 27. 60.

> 2 BEAMPOS cards to describe the distribution of the material in each
> CART-VOL region? One that defines the volume and one that defines the
> point around which the volume is centered?
>
Yes.
> 3. Is it even possible to do this in one setup? Do I have to perform
> different models for each single isotope and region and then combine
> them? Or is there a better method, like SOURCE cards?

It's possible, but i think you will have to use preprocessor to define different volume, position, material and region of your sources, e.g. if you have trace amounts of Co-60 dissolved in water as a first source and a chunk of cobalt metal as a second source. Both of these sources have the same primary particles ( gammas from Co-60 decay), but different types of source material, region, volume and activity.


Best regards,

Mikhail


__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info

b��jy����{.n�+��b�w�ka�ɚ�X��X���a��l��0��n��+���F����i�"��

Dear Keith,

My example can be raw and inaccurate!

The simple test with two sources is in attachment. There are two input files (isotope-1.inp, isotope-2.inp) will be created by Flair if in Run page you will start isotope-1 (uses #define S1) and isotope-2 (uses #define S2).

Then you can merge the data from both inputs in one single file.



Best regards,

Mikhail

__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info
Received on Fri Jun 24 2016 - 16:36:49 CEST

This archive was generated by hypermail 2.3.0 : Fri Jun 24 2016 - 16:36:52 CEST