MCNP Surface Tallies: F1 & F2 on Infinite Cylinders & Planes

[19matthew89]

TL;DR

How to define surface tallies of a cell?

Hi,
I have a question concerning surface tallies like F1 and F2. You have to provide a surface for them. Since, surfaces are defined as infinite (infinitely long cylinders, infinitely extended planes) how can you write the surface tally of a cell? What are the actual tally surfaces for F1 anf F2? The whole (infinite) surfaces or the actual surface of a cell?

As a quick example let's consider a z cylinder

So

1 M# rho_M -2 3 -4

2 CZ 1
3 PZ -10
4 PZ 10

will define the cell 1 which is a cylinder of radius 1 cm and height 20 cm.

May aim is tallying the surface flux F2 through the surface of the cylinder. How can I do it?

Is this

F2:N 2 3 4 T

right?

The manual says that if I use the parenthesis I will get a tally through the union of the surfaces but I'm not interested in the union.

Thanks a lot in advance!

P.S. I know a solution to be sure would be segmenting, but I am hoping for a less cumbersome solution and also then about what F2 (or F1) are actually evaluating.

 

[PSRB191921]

Hi,
you can with a macrobody surface "RCC"
In your case :
c cell
1 M# rho_M -10

c surface
10 RCC 0 0 -10 0 0 20 1

f2:n 10

 

[19matthew89]

Hi,
you can with a macrobody surface "RCC"
In your case :
c cell
1 M# rho_M -10

c surface
10 RCC 0 0 -10 0 0 20 1

f2:n 10

Hi, thanks!

Indeed that could work, but but:
1. my supervisor strongly recommended that I don't use macrobodies. He says that, according to his experience, the code behaves "weirdly" with macrobodies. I don't know what he means with that but I'll try to stick to hi suggestion and not using macrobody structures.
2. I still want to be able to have the tallies on the individual three surfaces defining the cylinder (bottom end, top end and lateral cylindrical surface)
3. Still it's not that clear to me what the tally F2:N 2 3 4 T then does. Clearly you have to assign a surface area via SD card otherwise it cannot compute the tally. Do you confirm then that F2/F1 consider the whole infinite surface for the tally computation? Thanks!

 

[Alex A]

Wow, what an unexpected nightmare of a question. Coool.

TL;DR Summary: How to define surface tallies of a cell?

The manual says that if I use the parenthesis I will get a tally through the union of the surfaces

This is exactly what I expected, trouble is now I check the manuals this isn't what they say. They say this produces an average flux. Do us the favour of double checking, cheers.

The surfaces are not infinite practically because nothing is tracked through the void.

"F2:N 2 3 4 T"
This should be 4 tallies, Surface 2, 3, 4, and the average of the three surfaces.

 

[19matthew89]

Wow, what an unexpected nightmare of a question. Coool.

This is exactly what I expected, trouble is now I check the manuals this isn't what they say. They say this produces an average flux.

Do us the favour of double checking, cheers.

I'm afraid I don't understand your point. It indeed produces the average flux of the union of the surfaces. But that is not useful in defining the surface I'm interested in.

The surfaces are not infinite practically because nothing is tracked through the void.

"F2:N 2 3 4 T"
This should be 4 tallies, Surface 2, 3, 4, and the average of the three surfaces.

Yeah. It is exactly what it does but, from what I understand, it takes into account all the particle crossing the surfaces "at the infinite", namely:
* for the cylinder, the particle crossing the cylinder above the plane defined by surface 4 (@10 cm) and below the plane defined by surface 3 (@-10 cm)
* for the planes 3 and 4, the particles crossing the planes outside the circle defined by the inner part of surface 2

In other words it seems that there is no way of defining just the surface of I'm interested in but using macrobodies.

By the way I "solved" it by segmenting the infinite surfaces and using SD card.
For instance for surface of the circle defined by the intersection of plane 3 and cylinder 2, I wrote

F2:N 3

FS2 -2

SD2 3.141592 1

and then I'll only consider the tally of the first segment.

Finally a last comment about the solution of PSRB191921. I tried that solution out and, as pure simple example, it worked but remember that MCNP does not accept simply this (at least my code was complaining).

c surface
10 RCC 0 0 -10 0 0 20 1

f2:n 10

You have to specify the facets of the macrobodies for the tally F2 to work. So namely you have to write

10 RCC 0 0 -10 0 0 20 1f2:n 10.1 10.2 10.3 T

if you want to have the tallies on the three surfaces defining the cylinder and an average on it.

Cheers!

 

[Alex A]

I assumed union meant as in constructive solid geometry. The manual does use the word "union" and does clarify it means "average" for the F2 tally. You are right!, and you are also right you cannot specify an area this way.

FS is the right way to solve it, but it's not obvious what it means when a surface is subdivided by more than one other surface! The result is tree like.