I am simulating a 10mm x 10mm x 0.25mm peice of silicon at varying distances from a Caesium 137 source. Both source and silicon are in a room of air and a void beyond.
I want to know the electron energy generated within the silicon by interaction with gamma particles, for this I use the F8 tally.
Unfortunately the computation time for 1 particle history takes days, is this normal? I am using threading with a quad core and 8 logical processors.
Dear Jonathan; You have to define your CTME or NPS variables considering the your simulation setup. It is totally up to you to define those 2 variables since one is direct run time (CTME) another is number of tracking in the simulation (NPS). While the first variable directly use the "time", another variable namely NPS depend on performance of your PC since ram or proccessor highly affect the performance of random event generators of any Monte carlo code. This is my suggestion to you that define CTME variable and try for 30 min, firstly. So, you can define it as a latest line of your input as CTME 30. After that, please obtanin the error rate in your output file... (which is just in last tittle with F8 tally mesh pule-height distribution). You have to obtain something like 0.0012 etc. If this value is low, then you can define your CTME around of this run time 30 min something. But as i said, it highly depend on you. You can define it as 1 min or 100 min. Normally it is not possible to have such a runtime with such a simple geometry.. So, try and find the optimum run time for your setup. Good Luck...
By the way, those methods i mentioned is useful for MCNPX or MCNP4C. I am not sure that you can apply them with same commands in MCNP6 since i had not any experience with this version.
- Badges
- Science topic
- Similar topics
- Elementary Particle Physics
- Particle