Fatal error in function MoveST: Surface crossing missed

Report any suspected bugs and unexpected behavior here
Post Reply
skiptomycue
Posts: 2
Joined: Tue Jun 26, 2018 11:47 am
Security question 1: No
Security question 2: 93

Fatal error in function MoveST: Surface crossing missed

Post by skiptomycue » Tue Jun 26, 2018 12:20 pm

Hi,

This is my first post, I hope I don't mess up.

I am running an input for the RHF reactor @ILL in Grenoble. It has many beams extracting the neutrons and they are all tilted at some angle.
The horizontal ones and the safety rods didn't give me any problem with the standard command vector + rotation angle:


strans 10000 46.01 13.41 0 0 0 53.63333333 1
or
strans 531 rot -26.39 -0.25 0 -4.697 -3.537 0 -2.7139

But when I got to the ones inclined over 2 axis (or horizontal cones, that need a rotation from the starting z axis configuration), I had to use the rotation matrix to preserve my order of rotations:

strans I300 -32.43 45 28.55 -0.81915 0.00001 -0.57358 -0.00001 -1.00000 0.00000 -0.57358 0.00000 0.81915 1

It seems that just in the latter case, with the rotation matrixes, Serpent prints this error:


Inactive cycle 110 / 1000: k-eff = 1.02078
Inactive cycle 111 / 1000: k-eff = 1.02218
Inactive cycle 112 / 1000: k-eff = 1.00657
Inactive cycle 113 / 1000: k-eff = 1.01748
Inactive cycle 114 / 1000: k-eff = 1.02490

***** Mon Jun 25 15:55:11 2018:

- MPI task = 0
- OpenMP thread = 11
- RNG parent seed = 1529934075
- RNG history seed = 8190630888340320507
- RNG history idx = 29963

Fatal error in function MoveST:

Surface crossing missed between [-6.28E+01, 4.68E+01, 1.12E+00] and [-6.33E+01 4.65E+01 1.11E+00] (d = 6.465882E-01, l = 6.465881E-01)

Simulation aborted.

The first time it also happened hours in to the simulation, while from then on it was already during the inactive cycles
I have no error in the geometry of a "no cell", and couldn't detect an "overlap" in any of the plots.
I don't know whether it has to do with the rotation matrixes, but up to now they seem to be the trigger. When I take out those double tilted beams, the error doesn't appear.

Thanks in advance for the help!

Davide



PS: If I recall correctly the image is from an input with an horizontal cone, while the pasted one is from inclined beams, if it is of any help (OpenMP thread seems to be different).


Image

https://ibb.co/ei5eeo

User avatar
Jaakko Leppänen
Site Admin
Posts: 2441
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Fatal error in function MoveST: Surface crossing missed

Post by Jaakko Leppänen » Thu Jun 28, 2018 2:09 pm

This may be due to the rotation, limited numerical precision or some seemingly minor error in the geometry, but since it seems to be a rare event, the easiest solution is to set up the option to ignore geometry errors.
- Jaakko

hopefulundergrad
Posts: 3
Joined: Tue Jul 02, 2019 9:19 pm
Security question 1: No
Security question 2: 19

Re: Fatal error in function MoveST: Surface crossing missed

Post by hopefulundergrad » Tue Jul 02, 2019 9:36 pm

Hi,

also a first-time poster.

I also got a similar error:
Surface crossing missed between [4.16E+01, 1.94E+01, -1.09E+01] and [4.17E+01 1.92E+01 -1. 18E+01] (d = 1.641353E+00, l = 9.679614E-01)

When I set it to ignore geometry errors the program started running again but it started generating a HUGE number of warnings saying variations of :
"Warning message from function MoveST:
Surface crossing missed between [ect,asdf, ]"
with a large variety of numbers in a format similar to the error above.

How will I know if this is a serious error that will affect the outcome or something that I can ignore?

User avatar
Jaakko Leppänen
Site Admin
Posts: 2441
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Fatal error in function MoveST: Surface crossing missed

Post by Jaakko Leppänen » Fri Jul 05, 2019 9:18 am

These errors and warnings are most likely related to overlapping surfaces. I've seen this error at least with super-imposed detectors (ds or dtl) and adf calculations.

If you find such surfaces, try to modify the parameters in such way that there is a small gap between. Also note that overlapping geometry surfaces (i.e. surfaces that make up the cells) is technically a geometry error.
- Jaakko

Post Reply