Confirming USRBDX result using other scoring method

Good day Fluka experts,

I scored the differential fluence of thermal neutrons crossing a boundary using the USRBDX card.

I would like to reproduce these results using a different scoring card. What would be the most direct/appropriate way to achieve this?

Thanks for any tips in advance.

Kind regards,
Dalini

Dear Dalini,

Thank you very much for the question! The most direct way would be to compare its results with the ones provided by USRYIELD card. Please be aware that USRBDX calculates angular distributions perpendicular to the boundary surface and USRYIELD calculates the distribution with respect to beam direction defined by BEAMPOS. So for a fair comparison you would need to look at the fluence through a surface perpendicular to the beam direction - I hope your specific geometry permits it.
I hope this helps, please keep us posted if it doesn’t or if other questions arise.

Best wishes,
Alex

1 Like

Hi Alex,

Thanks so much for providing this clear description.
1.) Fortunately, for one of my geometries, I will be able to use USRYIELD because the surface is perpendicular to the beam direction.

2.) However, another geometry which I would like to use has detectors/surfaces whose normal is not parallel to the beam. I checked the USRYIELD card though and it looks like the user can define the direction if the beam direction is not to be used. I will look into how to do this more closely and follow up with any questions.

Thanks again for your help,
Dalini

Dear Dalini,

yes, please keep us updated if this works out for the second geometry. If not, please consider that you can create a USRBDX which is perpendicular to the beam direction and cross-check if the results are the same as with the USRYIELD card. Once you have the confirmation that the results are consistent, you can use USRBDX safely with different geometries.

Best wishes,
Alex

1 Like

Hi Alex, that’s what I thought too. I will let you know how it turns out :slight_smile:

Hey there Alex,

I was on vacation for a couple weeks. I’ve just managed to conduct this task. However, I found that when I use USRYIELD, no neutrons are scored in the interval which I selected. Attached are the results with the USRBDX and USRYIELD results written on the logical units 21 and 31, respectively. I’ve also attached my input file for reference. Would you be able to take a look and help me determine what is off when I implement USRYIELD? Thanks so much in advance for any help.

10MeV_Be_r8_t20_CARDcomp.inp (2.6 KB) 10MeV_Be_r8_t20_CARDcomp_21_sum.lis (75.5 KB) 10MeV_Be_r8_t20_CARDcomp_21_tab.lis (33.2 KB) 10MeV_Be_r8_t20_CARDcomp_31_sum.lis (27.8 KB) 10MeV_Be_r8_t20_CARDcomp_31_tab.lis (14.9 KB)

Kind regards,
Dalini

Dear Dalini,

I see in your input file that the particle type “NEUTRON” is assigned in the USRBDX card, but there is nothing in USRYIELD. Please try inserting “NEUTRON” there and tell us how it goes.

Best wishes,
Alex

Hi Alex, whoops! I was certain I already put that information in. I will attempt to do this again. Thanks!

Hi Alex,

For some reason, although I had NEUTRONS selected in Flair, the input file showed a numerical value for this card. Regardless, I adjusted the card until the card reflected “NEUTRONS” and not a numerical value.

I repeated the calculation and the values are not close. In the USRYIELD result, low energy neutrons were not scored (zero fluence) in many energy intervals. Attached to this email is the updated input and output files. Do let me know if you can identify issues with my USRYIELD input which is leading to this large discrepancy.

10MeV_Be_r8_t20_CARDcomp.inp (2.6 KB) 10MeV_Be_r8_t20_CARDcomp_21_tab.lis (33.2 KB) 10MeV_Be_r8_t20_CARDcomp_31_tab.lis (14.9 KB) .

Kind regards,
Dalini

Thank you Dalini,

one other discrepancy which I see is the "“Kind” entry in USRYIELD, which should be d2 N / dx1 dx2 to score the double differential yield. Maybe this will help?..

Best wishes,
Alex

Hi Alex, acknowledged. I am running the simulation and will follow up.

Hi Alex,

So I re-ran the simulations and made the adjustments you recommended but found that I ran into the same problem. The neutrons scored in the low neutron energy bins do not match when I compare my result to what is obtained in USRBDX - in this case, I am comparing the results listed in tab.lis.

Also, I looked into the results in the sum.lis files. The total response is higher in the USRYIELD case when compared with USRBDX by a factor of two. In the initial simulation, I set the bounds for the polar variable to be 2*pi. I adjusted this to pi to see if my choice of pi was wrong and the response obtained in both cases is equal, within error.

So a few questions:
1.) I am still not sure why I am unable to reproduce the results obtained from the USRBDX card, what else can be wrong here?
2.) I changed the upper bound on the polar angle by a factor of two. Why does this not seem to have an effect on the result? And what is the appropriate choice for the polar angle?
3.) Should I be able to compare the total response in the sum.lis files obtained from USRBDX and USRYIELD?

Please note that I included the area of the boundary in the USRBDX input. I included this factor as a normalization factor in the USRYIELD input.

Thanks in advance for any tips or advice. Please see the input and output files below.

The files with extension _polar were from the simulation with the upper bound on the polar angle set to pi.

The files written with logic numbers 21 and 31 are from USRBDX and USRYIELD, respectively.

10MeV_Be_r8_t20_CARDcomp.inp (2.6 KB) 10MeV_Be_r8_t20_CARDcomp_21_sum.lis (75.5 KB) 10MeV_Be_r8_t20_CARDcomp_31_sum.lis (27.8 KB) 10MeV_Be_r8_t20_CARDcomp_31_tab.lis (14.9 KB) 10MeV_Be_r8_t20_CARDcomp_polar.inp (2.6 KB) 10MeV_Be_r8_t20_CARDcomp_polar_21_sum.lis (75.5 KB) 10MeV_Be_r8_t20_CARDcomp_21_tab.lis (33.2 KB) 10MeV_Be_r8_t20_CARDcomp_polar_31_sum.lis (27.8 KB) 10MeV_Be_r8_t20_CARDcomp_polar_31_tab.lis (14.9 KB)

Dear Dalini,

the problem with your input file is, that you on the USRYIELD card use Polar \theta lab deg as the second quantity, but then you specify the angles in radians. To use radians, you need to select Polar \theta lab.

The maximum angle is 180° (or \pi).

If you run the simulation for a polar angle between 0° and 180°, then the USRBDX (one way current) and USRYIELD scorings will give the same results.

Note: To get the same numerical value, the USRYIELD result has to be multiplied with (integrated on) the width of the second quantity. For Polar angle it has to be multiplied with the solid angle (4 \pi in this case), regardless if the angle was defined in degrees or radians.

Cheers,
David

1 Like

Dear David, thanks for confirming this detail. I will run the simulation and get back to you.

Dear David,

I’ve repeated the simulations and the results are still different. The results have improved, but everything is off by an order of magnitude. I’m not clear on what’s leading to this discrepancy as I’ve followed your advice. Input files and output files are attached.

The files written with logic numbers 21 and 31 are from USRBDX and USRYIELD, respectively.

10MeV_Be_r8_t20_CARDcomp.inp (2.6 KB) 10MeV_Be_r8_t20_CARDcomp_21_sum.lis (75.5 KB) 10MeV_Be_r8_t20_CARDcomp_21_tab.lis (33.2 KB) 10MeV_Be_r8_t20_CARDcomp_31_sum.lis (27.8 KB) 10MeV_Be_r8_t20_CARDcomp_31_tab.lis (14.9 KB)

Best,
Dalini

Dear Dalini,

maybe my explanation wasn’t clear enough, the USRYIELD results are always reported as differential to the second quantity.

In case of Polar angle, they are normalized to solid angle.

If you score on the full range (\theta = 0°… 180°), then you need to multiply the USRYIELD results with 4\pi, to get the same numerical value with the USRBDX scoring.

Cheers,
David

Dear David,

That factor of 4$\pi$ should take care of the difference then! Thank you so much for clarifying this detail.

Best,
Dalini