Page 1 of 1

### Serpent Boundary Conditions

Posted: Mon Jun 10, 2013 7:39 pm
I am trying to model a quarter core for a PWR. I have a lattice constructed of fuel assemblies and water assemblies and they are contained within a square cylinder. I want to have reflective boundary conditions on the positive 'y' and negative 'x' surfaces as these are the locations of the rest of the core. I want to have vacuum boundary conditions on the negative 'y' and positive 'x' surfaces as these are the outer edges of the core. According to the manual the command set bc <cx> <cy> <cz> says the boundary conditions in each direction (y or x) need to be the same. Is there a way for me to specify separate boundary conditions for a single direction? In other words, how can I make the positive x direction a black boundary and the negative x direction a reflective boundary?

### Re: Serpent Boundary Conditions

Posted: Tue Jun 11, 2013 7:45 am
Try:

Code: Select all

``set usym 0 4 0.0 0.0``
This tells the code that universe 0 has 1/4 symmetry with respect to the origin, which should do the same trick. The quadrant that you need to describe in your model is bounded by the positive X-and Y-axes. Also note that the direction of the Y-axis is inverted in the lattice input (the values are read from "south" to "north").

Serpent 1 only allows 1/4 symmetry, but there are more symmetry options in Serpent 2.

### Re: Serpent Boundary Conditions

Posted: Tue Jun 11, 2013 8:10 pm
Thanks for the help. Could you please show me the lattice you are describing that is bounded by the positive x and y axis? If 0 is the fuel assembly and 1 is the water assembly would it be ones of these lattices?

100000000
100000000
110000000
110000000
111000000
111100000
111110000
111111100
111111111

or

111111111
001111111
000011111
000001111
000000111
000000011
000000011
000000001
000000001.

Also, when I use the usym command do I still have to define boundary conditions? Does Serpent know that the outer edges of the lattice are vacuum boundary condition? If not how would I properly define the boundary condition.

Ian

### Re: Serpent Boundary Conditions

Posted: Wed Jun 12, 2013 9:28 am
The easiest way is just to try. Plot the geometry:

Code: Select all

``````set plot 3 500 500
``````
(this will produce a png-format file <input>_geom1.png) It will show the geometry as a whole, symmetry included, as it will be seen by the transport simulation.

There's no need to define boundary conditions.