Meaning of "Average" Flux Tallies in MCNP

[a1234]

Hello, I've been working with MCNP on and off for a few years now, but just recently realized that I don't entirely understand how tallies are actually calculated in MCNP, and what they signify.

Taking the example of the F2 tally, the user manual (Section 3.3.5.1) states that F2 is the "flux averaged over a surface." I understand that the F2 tally takes the number of particles incident on a surface and divides it by the surface area. This value is multiplied by the source strength/flux multiplier card to obtain the true value of the flux on the surface.

I don't fully understand how this is an "average" flux. Is it simply an average in the sense that it is divided by the total surface area? And if so, how is the standard deviation (which is used to find the relative error) calculated, and what does this value represent physically?

 

[Alex A]

It's an average because different parts of a surface usually have different flux, and this is the total crossings over the total area. I would expect something similar to Poisson statistics, if 100 particles go through a surface then the uncertainty is SQR(100)=10 (to within a certain sigma).

I found F4 tallies to be stranger. They are the total path length of all particles in a cell divided by the volume. Dimensionally it works and it is right, and probably a better way of doing it, but it still feels all kinds of weird.