SFR (Sodium Fast Cooled Reactor)

User Support
1

@paulromano Thanks Mr. Paul for your answer. I will try it.

2

@Rahma I don’t know if it will help for your particular case, but we have a list of example notebooks with many different reactor examples, including an SFR:

3

Hello, paulromano.
I learned the example of the fast reactor, but when I copied the code and ran it in my openmc environment, the keff result was not consistent with the keff in the example, my keff = 1.26646, the number of particles and iterations running were the same as in the example, and the diagram I drew was also very strange, the fuel rod in the middle could not be drawn. Why is that? I was hoping you could answer my question,

1687849900546
1687849900546562Ă—545 3.46 KB

1687849942952
1687849942952569Ă—545 5.47 KB

5

Hello,lichw. I had the same result as you. Deleting or modifying all lattice_id and universe_id maybe help.

SFR_1
SFR_1567Ă—549 5.31 KB

SFR_2
SFR_21067Ă—1043 42.4 KB

but I still have some trouble in runing this example, there is an error “ERROR: Maximum number of lost particles has been reached.”

6

Hello @qq306553577, it seems the gap, clad and sodium cell was added to both the inner fuel universe and outer fuel universe. Defining different cells for the outer fuel should fix the problem.

7

Hello, I ran the same example and stumble upon the same two problems, too. On further inspection of the file geometry.xml, I seemed to have found something worth noticing.

Here is the geometry.xml(modified)(truncated for readability):

<?xml version='1.0' encoding='UTF-8'?>
<geometry>
  <cell id="1" material="13" region="1 -2 -4 5" universe="1"/>
  <cell id="2" material="15" region="2 -3 -4 5" universe="1"/>
  <cell id="3" material="10" region="3 -4 5" universe="1"/>
  <cell id="4" material="13" region="-1 -4 5" universe="1"/>
  <cell id="5" material="14" region="-1 -4 5" universe="2"/>
  <cell id="6" material="10" universe="3"/>
  <cell fill="100" id="7" region="-6 7 -8 -11 9 10 -4 5" universe="4"/>
  <cell id="8" material="10" region="~(-6 7 -8 -11 9 10) -4 5" universe="4"/>
  <cell fill="200" id="9" region="-12 13 -14 -17 15 16 -4 5" universe="5"/>
  <cell id="10" material="10" region="~(-12 13 -14 -17 15 16) -4 5" universe="5"/>
  <cell id="11" material="10" region="-18 19 -20 -23 21 22 -4 5" universe="6"/>
  <cell id="12" material="10" region="~(-18 19 -20 -23 21 22) -4 5" universe="6"/>
  <cell fill="300" id="13" region="-30 31 -32 -33 34 35 -4 5" universe="7"/>
  <cell id="14" material="void" region="~(-30 31 -32 -33 34 35) -4 5" universe="7"/>
  <cell id="1001" material="13" region="1 -2 -4 5" universe="2"/>
  <cell id="1002" material="15" region="2 -3 -4 5" universe="2"/>
  <cell id="1003" material="10" region="3 -4 5" universe="2"/>
  <hex_lattice id="100" n_rings="9" name="inner assembly" orientation="x">
    <pitch>1.24</pitch>
    <outer>3</outer>
    <center>0.0 0.0</center>
    <universes>...</universes>
  </hex_lattice>
  <hex_lattice id="200" n_rings="9" name="outer assembly" orientation="x">
    <pitch>1.24</pitch>
    <outer>3</outer>
    <center>0.0 0.0</center>
    <universes>...</universes>
  </hex_lattice>
  <hex_lattice id="300" n_rings="17" name="core">
    <pitch>21.08</pitch>
    <outer>3</outer>
    <center>0.0 0.0</center>
    <universes>...</universes>
  </hex_lattice>
  <surface .../>
</geometry>

First, let’s look at @lichw’s problem. In his/her diagram, where it’s supposed to be a lattice turned out to have only one rod. I think that’s because of the “fill” property of a cell. the “fill” property only gives out an ID, it doesn’t know whether that’s a universe or a lattice. Therefore, when universe IDs and lattice IDs overlap, problems may arise. I think this ambiguity need to be fixed in the future.

As for @qq306553577’s problem, the answer is already given: assignment of one cell to multiple universes lead to the problem. In the file, it can be seen that universe only exists implicitly as a property of cells or lattices, and only one universe ID is allowed for each property. In this way, some coding complexity may be reduced. Nevertheless, maybe it’s a better idea to issue an warning for the users to see?

8

Dear users,
If you think there were some geometry definition errors on your script, i.e. particle lost as you encounter, I recommend you to turn on the geometry debug mode

openmc.run(geometry_debug=True)

These are some discussions about troubleshooting these kinds of error

2 Likes