Analog reaction rate estimates

This category replaces the missing input manual
Post Reply
User avatar
Jaakko Leppänen
Site Admin
Posts: 2387
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Analog reaction rate estimates

Post by Jaakko Leppänen » Wed Apr 04, 2012 10:28 am

Analog reaction rate estimates, essentially based on counting the sampled reaction modes, can be calculated in version 2.1.4 by setting:

Code: Select all

set arr <rrn> <rrp>
where the two parameters are the calculation modes for neutrons and photons, respectively. For neutrons, mode 1 means that only reactions that directly affect neutron balance are included, and mode 2 includes all. For photons the calculation is either on (1) or off (0).

The calculation produces a matlab-readable output file "<input>_arr<bu>.m", where <bu> is the burnup step. The reaction rates are calculated over all materials in the system.
- Jaakko

BorisH
Posts: 2
Joined: Tue Oct 22, 2013 4:02 pm
Security question 1: No
Security question 2: 92
Location: PSI, Switzerland

Re: Analog reaction rate estimates

Post by BorisH » Fri Aug 15, 2014 11:38 am

Hello folks,

It may not be the original intent or the arr card, but it would greatly simplify my life sometimes to have the result in a material-wise form (maybe even just for a selected material). Would it be feasible?

Thanks!

satkinson
Posts: 23
Joined: Thu Apr 09, 2015 6:14 pm
Security question 1: No
Security question 2: 92

Re: Analog reaction rate estimates

Post by satkinson » Tue May 24, 2016 3:42 pm

Hi Jaakko,

Can you just confirm that Serpent does not produce/print endf 27 and that it is the sum of mt 18+ the sum of (102 to 117).

Also I am interested in 201, the total production. Is there any way of calculating this with 51- to 91? There is little information in the reference regarding how many neutrons are produced for each of these reactions, but the number is roughly half of what is expected if I sum 51 to 91.

Cheers

User avatar
Jaakko Leppänen
Site Admin
Posts: 2387
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Analog reaction rate estimates

Post by Jaakko Leppänen » Wed May 25, 2016 9:56 am

The analog reaction reaction rate output is based on the reactions that were actually sampled. In addition there is a number of special reactions that are not sampled, but may be used with detector response functions. The sampled and special reactions are listed for each nuclide at the .out output file.

All reactions with MT number > 200 belong to the second category. I believe MT 27 is some rarely used mode and I'm not Serpent even reads that (you can check the .out file of your run).

The ENDF reaction MT's and other detector response functions are listed at at the Serpent wiki. When I have time, I'll add another column in the table indicating whether or not the reaction is sampled during tracking.
- Jaakko

satkinson
Posts: 23
Joined: Thu Apr 09, 2015 6:14 pm
Security question 1: No
Security question 2: 92

Re: Analog reaction rate estimates

Post by satkinson » Wed Jun 15, 2016 2:05 pm

When you include set inventory or daystep the results no longer normalsie. When Trying to manually normalize the results by using tot_lossrate the results are often 5-10% lower to if normalized by removing these features. Is there a different procedure for normalizing ARR's?

Cheers
satkinson

User avatar
Jaakko Leppänen
Site Admin
Posts: 2387
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Analog reaction rate estimates

Post by Jaakko Leppänen » Thu Jun 16, 2016 12:14 am

You are probably running predictor-corrector calculation? It seems that the analog reaction rates from the corrector step overwrite the previous file from the predictor step, and the rates are not consistent with detector calculations, etc.

To fix the problem, add:

Code: Select all

  if ((long)RDB[DATA_BURN_STEP_PC] == CORRECTOR_STEP) 
    return;
at the beginning of rroutput.c

Let me know if this fixes the problem.
- Jaakko

satkinson
Posts: 23
Joined: Thu Apr 09, 2015 6:14 pm
Security question 1: No
Security question 2: 92

Re: Analog reaction rate estimates

Post by satkinson » Mon Jun 20, 2016 1:18 pm

Hi Jaakko,

This is now working as intended.

Thanks

Satkinson

satkinson
Posts: 23
Joined: Thu Apr 09, 2015 6:14 pm
Security question 1: No
Security question 2: 92

Re: Analog reaction rate estimates

Post by satkinson » Wed Sep 21, 2016 6:55 pm

Hi Jaakko,

When using the ARR's which have multiple region using the same nuclide is their listing in the ARR file as they appear in the geometry. I have a nested section in TRISO particles, so does this indicate that these are all grouped together in that universe. There should be three graphite regions, matrix, particle surroundings (3 layers) and fuel block around the outside, so I was surprised to see four outputs.

The output of the nuclides is;

nuc( 11, :) = ' 6012.09c';
nuc( 12, :) = ' 6012.09c';
nuc( 13, :) = ' 6012.09s';
nuc( 14, :) = ' 6012.09s';

What does the s stand for? For the S versions they have the ARR 1002 and 1004 which are not tabulated on the wiki if you can shed some light on these.

Thanks,

satkinson

User avatar
Jaakko Leppänen
Site Admin
Posts: 2387
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Analog reaction rate estimates

Post by Jaakko Leppänen » Wed Sep 21, 2016 10:40 pm

The 's' means that the nuclide has S(a,b) thermal scattering data associated with it. MT numbers 1002 and 1004 are S(a,b) elastic and inelastic cross sections, respectively. I added the definitions in the Wiki.
- Jaakko

Post Reply