Efficient MCNP Lattice Source Help: Defining Universes and Tallies in Cell File

[ethnscot]

This is what I hate about MCNP, not a lot of documentation. How do I define all of a universe as a source and a tally? I have a lattice like the below code.
How do I get this code to work with tallies for positions 1,2, and 3 in the lattice; and a source for the 2's. I get the error "sampling efficiency is too low" Since, I have a very, very low efficiency I'm not sure how to fix this.
[CODE title="One That Doesn't Work"]MCNP Cell File
c Created on
901 0 999 imp:p=0 imp:e=0
902 6 -0.001205 -999 901 imp:p=0 imp:e=0
903 0 -902 u=999 lat=1 imp:p=1 imp:e=1
fill=0:2 0:8 0:4
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1
4 2 1 1 1 1 1 1 1 1 1 1 1 1 1
904 0 -901 fill=999 imp:p=1 imp:e=1
42 0 -903 u=1 imp:p=1 imp:e=1
43 0 903 u=1 imp:p=0 imp:e=0
1 2 -0.997 -903 u=2 imp:p=1 imp:e=1
101 0 903 u=2 imp:p=0 imp:e=0
2 2 -0.997 -903 u=3 imp:p=1 imp:e=1
102 0 903 u=3 imp:p=0 imp:e=0
3 6 -0.001205 -903 u=4 imp:p=1 imp:e=1
103 0 903 u=4 imp:p=0 imp:e=0

901 rpp 0 71 0 167 0 119
902 rpp -10.0 10.0 -10.0 10.0 -10.0 10.0
903 rpp -10.1 10.1 -10.1 10.1 -10.1 10.1
999 rpp -5 76 -5 172 -5 124

c water
m2 1000 -0.111902
8000 -0.888098
c Air
m6 6000 -0.000124
7000 -0.755268
8000 -0.231781
18000 -0.012827
mode p e
sdef erg=1 par=2 eff=0.0000001 X=d1 Y=d2 Z=d3 cel=d4
si1 -10.0 10.0
sp1 0 1
si2 -10.0 10.0
sp2 0 1
si3 -10.0 10.0
sp3 0 1
si4 L (3<903<904)
sp4 1
f8:e (2<903<904)
e8 0 0.0000001 5.0
f18:e (2<903<904)
e18 0 0.0000001 5.0
nps 1000000[/CODE]

 

[ethnscot]

I actually figured it out myself. I needed eff to be 1E-10. Now I have ***** lost particle in newcel - zero lattice element hit ***** as an error. Does anyone know how to fix this?

 

[Alex A]

Your x and z are misaligned so you get empty lattice hits. Run mcnp ip Input.txt and do some cross sections through the lattice to see. 'px 1' is helpful.

If you are setting an eff so low you are wasting 99.99% of your computer time that should be a hint something is not right. You seem to want to define your source in one of the other universes, this makes no sense to me. When you have fixed the alignment, locate the position in the real universe and redo the source and I would expect many problems to go away.