Geometry error: no intersection found in mcnp

In summary, the geometry error is caused by an overlap of cells and the user is advised to split the cells to avoid the error.
  • #1
khary23
93
6
Hello All,

I have yet another MCNP question. I received the following error "geometry error: no intersection found mcnp" when trying to run a a simulation. I looked at the output and according to it I have an infinite volume in cells 14 and 500. I plotted the geometry and don't see how its infinite. Can someone help me with what I am missing?
 

Attachments

  • RDF_cyl.txt
    4.7 KB · Views: 253
  • RDF_cyl_out.txt
    49.4 KB · Views: 242
Engineering news on Phys.org
  • #2
Cell 14 is outside of surface 3 (a RCC) and outside of surface 10 (a RCC) and so extends to infinity. Also, it has importance of 1 for p and e. The comment says "outside of problem" so that's weird.
Cell 22 is outside surface 10, a RCC, and it has # on a bunch of cells. So it extends to infinity. More weird.
Cell 500 is outside of surface 4, a SO, so it extends to infinity. It has importance 0.

You've got some messy overlaps going on here.

I think possibly you don't need cell 21.
And cell 22 and cell 200 seem to overlap.

I would suggest resisting using the # operator to define cells. Cell 22 is probably a lot simpler if you just figure out what are the inner and outer surfaces.

When you have different radius at different elevations, it is often simpler to split the cells outside those different radii. So you could split cell 22 into three parts, one above the disk, one extending from top to bottom of the disk, and one below the disk. With three parts you can easily get rid of all the # operators. Not sure if it's still true with MCNP 6.1, but older versions run slower if you have a lot of # operators.

Note that you can use the sub-surfaces of the rcc's to define cells. The top of rcc 40 is surface 40.2, the bottom is 40.3. But the sense is weird. +40.2 is outside the top of the rcc, and so it's the positive coord direction. +40.3 is outside the BOTTOM of the rcc, so it's the NEGATIVE coord direction. So the region -40.2 -40.3 is the region between the top and bottom of the rcc.
 
  • #3
DEvens said:
Cell 14 is outside of surface 3 (a RCC) and outside of surface 10 (a RCC) and so extends to infinity. Also, it has importance of 1 for p and e. The comment says "outside of problem" so that's weird.
Cell 22 is outside surface 10, a RCC, and it has # on a bunch of cells. So it extends to infinity. More weird.
Cell 500 is outside of surface 4, a SO, so it extends to infinity. It has importance 0.

You've got some messy overlaps going on here.

I think possibly you don't need cell 21.
And cell 22 and cell 200 seem to overlap.

I would suggest resisting using the # operator to define cells. Cell 22 is probably a lot simpler if you just figure out what are the inner and outer surfaces.

When you have different radius at different elevations, it is often simpler to split the cells outside those different radii. So you could split cell 22 into three parts, one above the disk, one extending from top to bottom of the disk, and one below the disk. With three parts you can easily get rid of all the # operators. Not sure if it's still true with MCNP 6.1, but older versions run slower if you have a lot of # operators.

Note that you can use the sub-surfaces of the rcc's to define cells. The top of rcc 40 is surface 40.2, the bottom is 40.3. But the sense is weird. +40.2 is outside the top of the rcc, and so it's the positive coord direction. +40.3 is outside the BOTTOM of the rcc, so it's the NEGATIVE coord direction. So the region -40.2 -40.3 is the region between the top and bottom of the rcc.

Thank you for the explanation!
 
  • #4
Just let us know with enough anticipation so we can head to the nearest shelter! ;).
 

1. What does the error "no intersection found in mcnp" mean?

The error "no intersection found in mcnp" refers to a problem encountered while using the Monte Carlo N-Particle (MCNP) code for nuclear and radiation transport simulations. It indicates that the code was unable to find an intersection between two or more objects in the geometry model, which is necessary for the simulation to run successfully.

2. What could be causing the "no intersection found in mcnp" error?

There are several possible causes for this error, including incorrect input parameters, invalid geometry definitions, or overlapping objects in the model. It could also be due to a bug in the code or a limitation in the geometry representation.

3. How can I fix the "no intersection found in mcnp" error?

To fix this error, you will need to carefully review your input parameters and geometry definitions to ensure they are correct and do not contain any errors. You may also need to adjust the geometry model to remove any overlapping objects or make other modifications to ensure that the code can find intersections between objects.

4. Is there a way to prevent the "no intersection found in mcnp" error from occurring?

While it is not always possible to prevent this error from occurring, there are some steps you can take to minimize the chances of encountering it. This includes carefully designing and constructing your geometry model, using appropriate input parameters, and performing thorough testing and validation before running the simulation.

5. Can I get more information about the "no intersection found in mcnp" error?

Yes, you can refer to the MCNP user manual or consult with other users or experts in the field for more information about this error. It may also be helpful to check for any updates or patches for the code that could address this issue.

Similar threads

Replies
6
Views
1K
  • Nuclear Engineering
Replies
2
Views
2K
Replies
3
Views
1K
  • Nuclear Engineering
Replies
4
Views
2K
  • Nuclear Engineering
Replies
9
Views
2K
Replies
2
Views
931
  • Nuclear Engineering
Replies
24
Views
4K
  • Nuclear Engineering
Replies
6
Views
2K
  • Nuclear Engineering
Replies
3
Views
2K
Replies
2
Views
2K
Back
Top