## Serpent 2.1.25 Convergence Test

Questions and discussion about applications, input, output and general user topics

### Serpent 2.1.25 Convergence Test

Dear Jaakko,

By using the card "set his 1", Serpent print an additional output called _his0.m, in which there are 9 parameters printed in that file in the following order:

HIS_IMP_KEFF (4 columns)
HIS_ANA_KEFF (10 columns)
HIS_COL_KEFF (4 columns)
HIS_MEAN_POP_SIZE (4 columns)
HIS_MEAN_POP_WGT (4 columns)
HIS_TRANSPORT_RUNTIME (7 columns)
HIS_TRANSPORT_CPU_USAGE (4 columns)
HIS_ENTR_SPT (13 coumns)
HIS_ENTR_SWG (13 coumns)

It is obvious that the first column for all parameters correspond to cycle number, but what about the rest of columns in each parameter?
Unfortunately, I did not find any explanition either in the forum nor the new wiki.

What is the difference between HIS_ENTR_SPT and HIS_ENTR_SWG? and which one correspond to fission source (Shannon) entropy?

Thanks...
Yousef Alzaben
Ph.D. Student
Karlsruhe Institute of Technology (KIT)
Yousef1986

Posts: 73
Joined: Fri Sep 04, 2015 3:33 pm
Location: Karlsruhe, Germany

### Re: Serpent 2.1.25 Convergence Test

The values are in groups of three: cycle-wise value, cumulative mean and cumulative relative statistical error. The entropies are calculated separately for x-, y- and z-directions, and the total. SPT refers to entropy calculated based on source points and SWT to entropy calculated using source weight. They are equivalent when no variance reduction (such as the uniform fission site method) is used.
- Jaakko

Jaakko Leppänen

Posts: 1927
Joined: Thu Mar 18, 2010 10:43 pm
Location: Espoo, Finland

### Re: Serpent 2.1.25 Convergence Test

Thank you Jaakko.

I understood what you explained earlier that:
- 1st column: Cycle number.
- 2nd column: Cycle-wise value.
- 3rd column: Cumulative mean.
- 4th column: Cumulative relative statistical error.

Out of curiosity, what about HIS_ANA_KEFF (10 columns) and HIS_TRANSPORT_RUNTIME (7 columns), they have more than 4 columns?
Yousef Alzaben
Ph.D. Student
Karlsruhe Institute of Technology (KIT)
Yousef1986

Posts: 73
Joined: Fri Sep 04, 2015 3:33 pm
Location: Karlsruhe, Germany

### Re: Serpent 2.1.25 Convergence Test

The analog estimator of k-eff is calculated using all, prompt neutrons only and delayed neutrons only. The transport running time is (if I remember correctly) for the current transport cycle and cumulative over burnup calculation
- Jaakko

Jaakko Leppänen

Posts: 1927
Joined: Thu Mar 18, 2010 10:43 pm
Location: Espoo, Finland

### Re: Serpent 2.1.25 Convergence Test

Thanks Jaakko not only for your quick response, but also for your effective communication.
Yousef Alzaben
Ph.D. Student
Karlsruhe Institute of Technology (KIT)
Yousef1986

Posts: 73
Joined: Fri Sep 04, 2015 3:33 pm
Location: Karlsruhe, Germany

### Re: Serpent 2.1.25 Convergence Test

Dear Jaakko,

Please, correct me if I'm wrong.

In HIS_ENTR_SPT and HIS_ENTR_SWG output variable of the input_his file, the output order is the following:

<Cycle number> <X Entropy> <Cumul. X> <Error> <Y Entropy> <Cumul. Y> <Error> <Z Entropy> <Cumul. Z> <Error> <Total Entropy> <Cumul. Total> <Error>

It is not fully clear to me. I had some confusion about what is the order of output. Especially if total entropy is the last or it is the first position.

Thanks you for help!
pdarnowski

Posts: 2
Joined: Wed Jun 03, 2015 6:51 pm

### Re: Serpent 2.1.25 Convergence Test

It's actually total, x, y and z. I added a brief description at the Serpent Wiki: http://serpent.vtt.fi/mediawiki/index.p ... ory_output
- Jaakko

Jaakko Leppänen