flux calculation

Questions and discussion about applications, input, output and general user topics
Post Reply
lorenzoloi
Posts: 10
Joined: Sat Apr 10, 2021 11:52 am
Security question 1: No
Security question 2: 7

flux calculation

Post by lorenzoloi » Mon Sep 20, 2021 7:05 pm

Hello to everyone,
I have a 17x17 assembly with 3 kind of fuel definitions and I'm sampling the neutron flux in this different cases. In particular, I have one pin divided axially through the "subz" command (say, pin 1), one pin with Gd inside and with the "subr" division (say pin2) and all the others 287 pins with standard fuel and no divisions (say pin of kind 3).
I have defined 3 detectors in order to sample the neutron flux axially for all the three cases, i.e.:

Code: Select all

det pin1
de thermal_grid
dm fuelPIN1
dz -180 180 100

det pin2
de thermal_grid
dm fuelPIN2
dz -180 180 200

det GroupOfPin3
de thermal_grid
dm fuelPIN3
dz -180 180 200
With these definitions, the flux related to pin3 should represent the average flux in the assembly, while the other 2 fluxes are related to the single pins. Qualitatively the results are correct since they fit quite well the cosine shape. My doubt is related to the order of magnitude of the peak, in fact for the group of pin 3 the peak-flux is 10^(15), while for the single pin1 the peak-flux has the order of magnitude of 10^(13) and for pin 2 the peak flux has the order of magnitude of 10^(12).
As I know, the flux should be indipendent of the dimension of the domain, since the ratio between the number of scores in a volume and a volume itself (i.e. the flux) should be constant as the volume increases.

I'm evaluating the volumes with the -checkvolumes routine.

Do you have any explanation about this change in flux magnitudes?
Is there any relationship between the axial division in "dz" and the flux? since the pin2 has 200 divisions and a lower peak, while pin1 has 100 divisions and a higher one.

Thank you very much

Ana Jambrina
Posts: 682
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: flux calculation

Post by Ana Jambrina » Mon Sep 20, 2021 7:24 pm

The detector gives you the flux integrated over space/energy/direction (and time). To obtain the neutron averaged-flux [n/cm^2 s] you should divide by bin-volume corresponding to the integration spatial-domain: determined by the material detector ‘dm’ fuelPIN1/fuelPIN2/fuelPIN3 and refined and discretized by the axial mesh detector ‘dz’ (detector bin-volume), - and, if necessary, handling the energy dependence.
- Ana

lorenzoloi
Posts: 10
Joined: Sat Apr 10, 2021 11:52 am
Security question 1: No
Security question 2: 7

Re: flux calculation

Post by lorenzoloi » Tue Sep 21, 2021 6:16 pm

Thank you Ana, now it works.

I see that the same results is given if it's added the option 'dv' + volume of the bin in cm^3.
Is there any way to leave the calculation of the bin volume to serpent instead of do it by hand?

Ana Jambrina
Posts: 682
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: flux calculation

Post by Ana Jambrina » Tue Sep 21, 2021 6:22 pm

Yes, instead of post-processing the estimates, you can use the volume detector entry ‘dv’. However, keep in mind that such option is only valid for constant bin-volume, e.g. not applicable for curvilinear mesh detectors where the division is based on equal radii, not equal volume.
Note: the command line option '-checkvolumes' can help you define the volumes.
- Ana

Post Reply