3D Geometry Axial Division

Questions and discussion about applications, input, output and general user topics
Post Reply
Reem
Posts: 12
Joined: Sun Sep 17, 2017 7:56 pm
Security question 1: No
Security question 2: 43

3D Geometry Axial Division

Post by Reem » Tue Apr 10, 2018 12:37 am

Hello,

I am studying the criticality and isotopic composition of a PWR assembly with integrated neutron absorber rods. I tried dividing the rods containing neutron absorber into several axial layers ( + 10 Radial Layers). When taking the K-eff values for each case I've gotten the attached plot.

I cant figure out an explanation for the big shift in the case of 5 layers!

P.S.: Temperatures are taken as average.
Attachments
.png
K-eff Vs Burnup for different cases of axial division
.png (28.26 KiB) Viewed 1974 times
Last edited by Reem on Tue Apr 17, 2018 2:35 am, edited 1 time in total.

User avatar
Jaakko Leppänen
Site Admin
Posts: 2388
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: 3D Geometry Axially Division

Post by Jaakko Leppänen » Tue Apr 10, 2018 2:26 pm

3D burnup calculations are always tricky, because large systems may be subject to numerical instabilities. Have you checked how the axial power distribution behaves as function of burnup?
- Jaakko

Ville Valtavirta
Posts: 470
Joined: Fri Sep 07, 2012 1:43 pm
Security question 1: No
Security question 2: 92

Re: 3D Geometry Axially Division

Post by Ville Valtavirta » Tue Apr 10, 2018 2:51 pm

This might also be due to the fact that assuming a cosine axial shape for the flux, the absorber should deplete first from the middle of the axial height due to the higher flux there.

When you use 1 or 2 axial zones, the difference in flux level between the middle part and the ends cannot be reflected in your nuclide compositions as you don't have separate depletion zones for the middle part and the ends.

If you use a higher number of axial depletion zones you can observe different depletion speeds for the absorber between the ends and the middle, which may be reflected in the reactivity (as there will be less absorber remaining in the high flux areas).

-Ville

Reem
Posts: 12
Joined: Sun Sep 17, 2017 7:56 pm
Security question 1: No
Security question 2: 43

Re: 3D Geometry Axially Division

Post by Reem » Sat Apr 14, 2018 1:38 am

Ville Valtavirta wrote:This might also be due to the fact that assuming a cosine axial shape for the flux, the absorber should deplete first from the middle of the axial height due to the higher flux there.

When you use 1 or 2 axial zones, the difference in flux level between the middle part and the ends cannot be reflected in your nuclide compositions as you don't have separate depletion zones for the middle part and the ends.

If you use a higher number of axial depletion zones you can observe different depletion speeds for the absorber between the ends and the middle, which may be reflected in the reactivity (as there will be less absorber remaining in the high flux areas).

-Ville
Thank you!
Then what could explain the behavior of burn up radial distribution shown in the figure when dividing into two axial layers? ( As I mentioned before, the temperature is averaged).
Attachments
1.png
Burnup distribution at the upper and lower parts of fuel rod ( MWd/kgU Vs Radius) at 40 Mwd/kgU
1.png (35.75 KiB) Viewed 1957 times

Reem
Posts: 12
Joined: Sun Sep 17, 2017 7:56 pm
Security question 1: No
Security question 2: 43

Re: 3D Geometry Axially Division

Post by Reem » Sat Apr 14, 2018 1:53 am

Jaakko Leppänen wrote:3D burnup calculations are always tricky, because large systems may be subject to numerical instabilities. Have you checked how the axial power distribution behaves as function of burnup?
Thank you for your answer.

I havent ( Is it done by setting detectors?)
However, I have checked the radial burnup distribution, and I can tell there is a very big instability.
I am not sure the results this way are reliable ( I am studying self-shielding and Rim's effect), any suggestions to over come this?

Ville Valtavirta
Posts: 470
Joined: Fri Sep 07, 2012 1:43 pm
Security question 1: No
Security question 2: 92

Re: 3D Geometry Axially Division

Post by Ville Valtavirta » Mon Apr 16, 2018 11:15 am

With two axial layers you should get the same radial burnup distribution in both layers (if the system is otherwise axially homogeneous).

If you are simply interested in self-shielding and the rim-effect, have you considered doing 2D calculations?

-Ville

Reem
Posts: 12
Joined: Sun Sep 17, 2017 7:56 pm
Security question 1: No
Security question 2: 43

Re: 3D Geometry Axially Division

Post by Reem » Mon Apr 16, 2018 4:58 pm

Ville Valtavirta wrote:With two axial layers you should get the same radial burnup distribution in both layers (if the system is otherwise axially homogeneous).

If you are simply interested in self-shielding and the rim-effect, have you considered doing 2D calculations?

-Ville
I am already done with the 2D calculations, I wanted to expand the work and study the effect axially for different absorpers.
I tried the IFC files to account for temperature's effect, but I've got results much lower than expected. I only did it for the moderator (with average temperature for the fuel), could that be the problem?

Ville Valtavirta
Posts: 470
Joined: Fri Sep 07, 2012 1:43 pm
Security question 1: No
Security question 2: 92

Re: 3D Geometry Axial Division

Post by Ville Valtavirta » Tue Apr 17, 2018 12:14 pm

It's probably first best to see whether you can reproduce the 2D results (more or less) in a 3D model.

As I don't really know which kinds of results you obtained and which kinds of results you were expecting, I cannot comment on whether that could be due to the multi-physics interface. Fuel temperature does affect the resonance self-shielding and rim-effect.

-Ville

Reem
Posts: 12
Joined: Sun Sep 17, 2017 7:56 pm
Security question 1: No
Security question 2: 43

Re: 3D Geometry Axial Division

Post by Reem » Thu May 03, 2018 7:44 pm

Ville Valtavirta wrote:It's probably first best to see whether you can reproduce the 2D results (more or less) in a 3D model.

As I don't really know which kinds of results you obtained and which kinds of results you were expecting, I cannot comment on whether that could be due to the multi-physics interface. Fuel temperature does affect the resonance self-shielding and rim-effect.

-Ville
I did have matching results. However with IFC files, the criticality is lower. I cant figure out what causes this drop. I have once hexagonal lattice and I am using Point-average interface. Here is the IFC file input:

Code: Select all

1 water 1
./ifcout 5 -191.8 191.8 1
1 0.2 1.0
5
-191.8 -0.74   564 
-95.9  -0.73  568 
0      -0.71   579
95.9   -0.69   590 
191.8  -0.65   594 
Is there any obvious error? If not, It would be great if I can send you serpent input file to help me finding the error.

Regards,
Reem

Ville Valtavirta
Posts: 470
Joined: Fri Sep 07, 2012 1:43 pm
Security question 1: No
Security question 2: 92

Re: 3D Geometry Axial Division

Post by Ville Valtavirta » Mon May 07, 2018 9:07 am

Your exclusion radius in the interface file seems suspiciously low.

Have you used the geometry plotter or the sample card (http://serpent.vtt.fi/mediawiki/index.p ... inition.29) to ensure that the correct temperature and density distribution is being used in Serpent?

-Ville

Post Reply