Identical Result

User Support
1

Hi all,

I used OpenMC to run TBR, flux, and absorption tallies on a dagmc-generated geometry with a custom source. Later, I ran another model with the same geometry, materials, and source, with the heating-local tally, as well as TBR and absorption as a sort of ‘sanity check’ between simulations.

When I went to check that the TBR and absorption results were similar, I discovered that they were exactly the same. They had the same value to as many decimal points as my device would print (over 10), and identical error.

I reran again, with the same outcome. I also reran another model (same source and geometry, different materials) and found that it yielded the exact same result as it had previously (as in this simulation was continuously giving the same results, it was not matching the other one).

The simulations have all been with 7 batches and 40,000 particles, so this occurrence feels extremely unlikely and odd. It feels like an indication that something is not quite right, but I don’t know where to start. What should I do?

Thanks,
Eleanor

2

Hi Eleanor,

to get a statistically different result, add the seed parameter and set it to other than 1. (I am assuming your simulation didn’t specify it and it defaulted to 1)

https://docs.openmc.org/en/stable/io_formats/settings.html

https://docs.openmc.org/en/stable/methods/random_numbers.html

3

Thank you! So then, as I’m initiating my settings object, is there any specific range that the seed needs to be in?

4

Any positive integer should make do. From what I recall, I believe you should use odd numbers since even numbers result in a shorter period (the the number of samples you need before the random number stream starts repeating itself).

1 Like