We are planning to measure beam current using a Faraday cup (FC) and would like to estimate the particle loss due to backscattering before the experiment.
As a preliminary step, I considered a simple case of cylindrical disc of copper (as FC) and let the beam (of size same at our FC position) incident on copper. I am attaching the input file. ex_1.inp (1.9 KB)
From the USRBDX results on bin 22, I obtained a total flux of 64,990,000 particles/(GeV/cm²/pr), which I assume represents the incident beam on the FC. For bin 24, the backscattered flux came out to 191,520 particles/(GeV/cm²/pr), leading to an estimated beam loss due to backscattering of approximately 0.3%.
I tried to convert the flux of 64,990,000 particles/(GeV/cm²/pr) into particles as = 64,990,000 x 0.0001 (beam energy) x 6.1544 (area of copper) = ~40,000 particles/primary.
I’m looking to determine the particle loss in terms of the actual number of electrons, or beam current, rather than flux.
I thought converting flux into number particle would give me the information of incident beam, however I am not getting how to interpret this number = ~40,000 particles/primary.
Could you kindly advise on the approach to calculate this and also on the input file ?
I checked the result using current (I1, LinE, LinΩ). However, the results appear to be the same as those obtained with ϕ. For the reference, I have attached the input file. ex_2.inp (2.0 KB)
I have a few questions regarding this updated input file:
what is the difference between I1 and I2?
What is the unit of the current?
Could you suggest an approach to calculate the number of particles since the current also appears to be in the unit = /cm²/GeV/pr?
The result using current in USRBDX on the 22nd bin is also showing the number = 64,990,000 as an output (which matches the value obtained when selecting ϕ).
could you please tell me how to interpret this result? I want to find the number of particles that scatters back from the copper, could suggest the correct way?
I apologize for not being able to understand the explanation given in the manual :(.
let me define the current, and from its definition I believe you will find the answers to your questions. First of all, let me clarify something: the current in this context differs from electrical current, and would need to manipulate the fluka scoring to get that.
Current (I) and fluence (\phi) both relate to the passage of particles through a surface. The difference between the two is that current simply counts particles crossing the given surface, whereas fluence weighs the crossing by 1/cos\theta where \theta is the angle between the incident particle and the normal vector to the surface.
Focusing on the current, as we said it counts the number of particles crossing a given surface, given as the interface between the two regions, A and B, let’s say. One could then ask whether we should count at all particles going from B to A, or just the ones going from A to B. This is the difference between one-way scoring (I1) and two-way scoring (I2).
Finally, the units of of the current, regardless of whether it is I1 or I2 are 1/cm²/GeV/pr/sr (you almost had it right, but forgot a factor of 1/sr (solid angle selected).
Now going to how to calculate the number of particles backscattering from copper to void.
make sure you define your regions appropriately, so the surface you request your current on is correctly defined. In particular, you currently have the WATER fully embedded in the VOID, meaning that USRBDX will have no way of knowing which of the surfaces you are interested in, could be the particles scattering back, but also on the sides, and also just moving forward. You could for example define a region of void behind the water (e.g. BACKVOID), such that if you see particles crossing from WATER to BACKVOID, you know that they have been backscattered
since you are only interested in particles backscattering, you would probably use a USRBDX I1 crossing, from WATER to BACKVOID
when you scale the results of the scoring and you want to get an integrated number of particles, you multiply the result by the width of the bin (this is quite important, it looks like youwere doing the wrong conversion earlier). In particular, you need to multiply the scoring by 3 numbers: the area (same as the one you put in USRBDX), the energy range (E_max - E_min in your scoring, where i suggest you use 0 as Emin, and put linear energy scoring), and the solid angle range (\Omega_max - \Omega_min, which you should set to 2 \pi rather than \pi since we are talking of a solid angle)
After multiplying the result of USRBDX by these bin widths, the result will indeed be in particles/primary. Alternatively, I advise you look into the _sum.lis files that are created when you process the data in Flair, which already performs these multiplications for you, and gives you results in particles/primary.
When I follow the above steps, I get ~0.2 backscattering particles per incident primary.
I hope this helped, and you get the result you want
Thank you very much for the detailed explanation. Your reply has helped me understand each of the points you mentioned, and I was able to calculate the values given in the sum.lis file accordingly.
Based on my understanding, I interpreted that the integration of the curve obtained from the 1st and 3rd columns in tab.lis should yield the numbers in sum.lis (particles/cm²/pr). After making adjustments according to your suggestions, I now find that the number I obtain for electrons moving in the back direction (from COPPER to BACKVOID) is consistent, whether by integration, multiplying by the bin width, or the value in sum.lis.
However, for electrons moving from BACKVOID to COPPER, the result matches when multiplying by bin width, but differs with integration. Since the result I obtained with sum.lis file, aligns closely with values obtained using Geant4, I believe I can proceed with calculations for the other cases, but I wanted to share this observation with you in case I am overlooking anything.
I am sharing my modified file here for the reference. ex_3.inp (2.0 KB)
for the scoring you mentioned, BACKVOID to COPPER, I get the same from integrating and from tab.lis file. Perhaps can you clarify what you mean by the difference between integrating and multiplying by bin-width? The integrated current should be 1 particle/primary.
Especially if you are intested in overall quantities, independent of angle and energies, the _sum.lis file is a good approach, as it performs operations in the background, ensuring consistency. Nevertheless, I would make sure you resolve this issue, to ensure full control of the analysis.