MCNP Walkthrough
Now that you can create and navigate directories, build and save text files, and check cluster availability, you can get started running MCNP simulations. (If you don’t know what I’m talking about, go back and start with the DECF walkthrough.)
Remember that you cannot run MCNP on Kepler.
MCNP Input Files
First, we need to construct an MCNP input file. I will describe this more in-depth during discussion.
Below is an example file. I have named it cube.inp. You can create this file yourself by opening it as a text file on DECF (i.e. nano cube.inp).
Simple Uranium-235 Cube
c cell cards for uranium cube (64cm^3) simulation
1 1 -19.1 -1 2 -3 4 -5 6 IMP:N=1
2 0 (1:-2:3:-4:5:-6) IMP:N=0
c end of cell cards for cube simulation
c Beginning of surface
1 px 2 $ uranium cube (w/ side 4cm)
2 px -2
3 py 2
4 py -2
5 pz 2
6 pz -2
c End of surfaces
M1 92235 1 $ pure uranium-235 [ID ZAID Density]
kcode 10000 1 10 50
ksrc 0 0 0
PRINT 50
MCNP Data
MCNP uses a specific set of data libraries. These are not included in the set of files that your computer searches by default. Instead, you have to add them to this path. The command for this is
$ setenv DATAPATH /usr/local/mcnp5_lib-160
MCNP Execution
To run the MCNP input you have created, the command is
$ mcnp5 inp=cube.inp
MCNP Output
MCNP output will be written to a file named outp. (If outp already exists, output will be written to a file named outq, outr, outs, or all the way to out* where the asterisk will be the next unused letter in the alphabet.)
If you go to the bottom of the output file, you should see a table:
problem keff standard deviation 68% confidence 95% confidence 99% confidence
first half 0.30691 0.00063 0.30627 to 0.30756 0.30558 to 0.30825 0.30508 to 0.30875
second half 0.30588 0.00052 0.30535 to 0.30641 0.30479 to 0.30697 0.30438 to 0.30738
final result 0.30640 0.00041 0.30599 to 0.30681 0.30558 to 0.30722 0.30530 to 0.30750
the first and second half values of k(collision/absorption/track length) appear to be the same at the 95 percent confidence level.
***********************************************************************************************************************
The key parameter of interest here is the final result for \(k_{\textit{eff}}\), shown here to be 0.30640. Our uranium cube is clearly not critical.
Other important concepts
Some important things we will note are:
- Comment cards (
c) / signs ($) - Macrobodies
- Why it’s not always best to use fewer surfaces. (recalculation at each boundary)
- Complement operator
# - Elemental vs isotopic ZAID (8000 vs 8016)
- Why are we skipping cycles in KCODE?
- Differences between MCNP/Serpent
- Largely similar (still have surfaces and cells)
- No union operator in Serpent (more general shapes)
KCODE
The KCODE card indicates that we are performing a criticality calculation. The source size is 10,000 particles, we estimate \(k_{\textit{eff}} = 1.0\), we skip 10 cycles before averaging \(k_{\textit{eff}}\), and run a total of 50 cycles (40 are averaged).
KSRC
The initial source location is (0, 0, 0), as specified on the KSRC card.